Contemporary Diabetes

Series Editor: Aristidis Veves

Solomon Tesfaye
Christopher H. Gibbons
Rayaz Ahmed Malik
Aristidis Veves   Editors

Diabetic
Neuropathy
Advances in Pathophysiology and Clinical
Management
Third Edition
Contemporary Diabetes
Series Editor
Aristidis Veves, Beth Israel Deaconess Medical Center
Boston, MA, USA
The Contemporary Diabetes series focuses on the clinical aspects of obesity
and diabetes and provides the practicing health provider with all the latest
information regarding their management. The series also targets both basic
and clinical researchers. The audience includes endocrinologists, internists,
cardiologists, neurologists, nephrologists, podiatrists, ophthalmologists,
family physicians, nurse practitioners, nurse educators, and physician
assistants.
Solomon Tesfaye
Christopher H. Gibbons
Rayaz Ahmed Malik  •  Aristidis Veves
Editors

Diabetic Neuropathy
Advances in Pathophysiology
and Clinical Management

Third Edition
Editors
Solomon Tesfaye Christopher H. Gibbons
Royal Hallamshire Hospital Beth Israel Deaconess Medical Center
Sheffield Teaching Hospitals Boston, MA, USA
and the University of Sheffield
Director of Diabetes Research Aristidis Veves
Sheffield, UK Department of Surgery
Beth Israel Deaconess Medical Center
Rayaz Ahmed Malik Boston, MA, USA
Weill Cornell Medicine
Cornell University
Doha, Qatar

ISSN 2197-7836     ISSN 2197-7844 (electronic)

Contemporary Diabetes
ISBN 978-3-031-15612-0    ISBN 978-3-031-15613-7 (eBook)
https://doi.org/10.1007/978-3-031-15613-7

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2023
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher,
whether the whole or part of the material is concerned, specifically the rights of translation,
reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any
other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, expressed or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Humana imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Sophie, Aida and Sami—thanks for your unfailing support
Contents

Part I Clinical Features and Diagnosis

Introduction����������������������������������������������������������������������������������������������   3
Andrew J. M. Boulton

The Epidemiology of Diabetic Neuropathy ������������������������������������������   5
Christian Stevns Hansen, Laura L. Määttä, Signe Toft Andersen,
and Morten H. Charles

Clinical Features of Diabetes Neuropathies������������������������������������������  37
Gordon Sloan, Qi Pan, Ling Gao, Lixin Guo, and Solomon Tesfaye

Neuropathy in Type 1 and Type 2 Diabetes ������������������������������������������  51
Gulcin Akinci, Dustin Nowacek, and Brian Callaghan

Clinical Diagnosis of Diabetic Peripheral Neuropathy������������������������  67
Bruce A. Perkins and Vera Bril

Diagnostic Techniques for Diabetic Peripheral Neuropathy����������������  93
Long Davalos, Amro Stino, and A. Gordon Smith

Sensory Profiles and Diabetic Neuropathy�������������������������������������������� 113
Juliane Sachau, Manon Sendel, and Ralf Baron
Neurotrophic Factors in the Pathogenesis
and Treatment of Diabetic Neuropathy������������������������������������������������� 127
Nigel A. Calcutt

Treatment Induced Neuropathy of Diabetes ���������������������������������������� 157
Nadia McMillan and Christopher H. Gibbons
Asymmetric Diabetic Neuropathy: Radiculoplexus
Neuropathies, Mononeuropathies, and Cranial Neuropathies������������ 165
Pariwat Thaisetthawatkul and P. James B. Dyck

Motor Neuropathy in Diabetes �������������������������������������������������������������� 183
Karolina Snopek Khan and Henning Andersen
Cardiovascular Autonomic Neuropathy������������������������������������������������ 203
Lynn Ang and Rodica Pop-Busui

vii
viii Contents

Psychosocial Aspects of Diabetic Neuropathy:

From Description to Interventions �������������������������������������������������������� 221
Loretta Vileikyte and Frans Pouwer

Part II Pathophysiology


The Genomics of Diabetic Neuropathy�������������������������������������������������� 239
Abirami Veluchamy, Blair H. Smith, and David L. Bennett

Metabolic Mechanisms in Diabetic Neuropathy ���������������������������������� 253
Mark Yorek
Mechanisms of Nerve Injury in Diabetes:
Dyslipidemia, Bioenergetics, and Oxidative Damage �������������������������� 279
Stephanie A. Eid, Mohamed Noureldein, Masha G. Savelieff,
and Eva L. Feldman

Targeting the Mitochondrion in Diabetic Neuropathy ������������������������ 307
Ahmad Hedayat, Krish Chandrasekaran, Lindsay A. Zilliox,
and James W. Russell
Diabetic Sensory Neurons, Dorsal Root Ganglia,
and Neuropathy���������������������������������������������������������������������������������������� 327
Aparna Areti and Douglas W. Zochodne

Micro- and Macrovascular Disease in Diabetic Neuropathy �������������� 351
Lihong Chen and Aristidis Veves

The Spinal Cord in Diabetic Neuropathy���������������������������������������������� 363
Andrew G. Marshall, Anne Worthington, and Corinne G. Jolivalt

Brain Changes in Diabetes and Cognitive Dysfunction������������������������ 381
Geert Jan Biessels
Lifestyle and Dietary Modifications: Relevance
in the Management of Diabetic Neuropathy������������������������������������������ 397
Jonathan Enders and Douglas E. Wright

Pathophysiology of Neuropathic Pain���������������������������������������������������� 415
Andreas C. Themistocleous and Miroslav Misha Backonja

Central Nervous System Involvement in Painful
Diabetic Neuropathy�������������������������������������������������������������������������������� 427
Dinesh Selvarajah, Joyce Lim, Kevin Teh, Xin Chen,
Jing Wu, and Solomon Tesfaye

Part III Clinical Consequences and Treatments


Characteristics and Treatment of Painful Diabetic Neuropathy �������� 441
Sandra Sif Gylfadottir and Nanna Brix Finnerup
Orthostatic Hypotension and Sudomotor
Dysfunction in Diabetes�������������������������������������������������������������������������� 453
Lauren F. Fanty and Christopher H. Gibbons
Contents ix

Gastrointestinal Neuropathy������������������������������������������������������������������ 471

Karen L. Jones, Chinmay S. Marathe, Tongzhi Wu,
Christopher K. Rayner, and Michael Horowitz

Diabetic Neuropathy: Clinical Management—Genitourinary
Dysfunction in Diabetes�������������������������������������������������������������������������� 491
Vincenza Spallone, Enrico Finazzi Agrò, Roberta Centello,
Claudio Lecis, Luca Orecchia, and Andrea M. Isidori
Index���������������������������������������������������������������������������������������������������������� 531
 Part I
Clinical Features and Diagnosis
 Introduction
Andrew J. M. Boulton
1 Early History of the Diabetic
Neuropathies
Whereas the first ever records of diabetes appear
to be found in ancient Egypt around 1550 BC, the
history of the diabetic neuropathies is relatively
recent. Indeed, the first references to possible
involvement of the nervous system as a compli-
cation of diabetes did not appear until the end of
the eighteenth century [1]. It was Dr John Rollo,
who was a surgeon in the British Royal Artillery,
who wrote two texts on diabetes and in one of
them described a patient with “burning pains in
the palms and hands and the soles of the feet with
weakness and complains of painful and gnawing
sensations in the leg” [2, 3]. However, as pointed
out by Ward [3, 4], reports that such patients with
the symptoms as described above, “had a good
night’s sleep” this would not be typical of painful
diabetic neuropathy. Moreover, Rollo did not
A. J. M. Boulton (*)
Division of Diabetes, Endocrinology and
Gastroenterology, University of Manchester,
Manchester, UK
Manchester Diabetes Centre, Manchester Royal
Infirmary, Manchester, UK
University of Miami, Miami, FL, USA
International Diabetes Federation, Brussels, Belgium
e-mail: ABoulton@med.miami.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_1
acknowledge a direct link between diabetes and
any nervous system disorder.
It was not until 1864 that Marchal de Calvi
correctly described the relationship between dia-
betes and peripheral neuropathies [5]. Indeed, it
was the well-known Parisian physician Jean-­
Martin Charcot who acknowledged the impor-
tance of the observation by Marchal de Calvi:
Charcot also described that in some diabetic
cases, there are signs of neurological disorders
[5]. It was towards the end of the nineteenth cen-
tury that Charcot summarised the clinical fea-
tures of diabetic peripheral neuropathy [6].
Another pivotal observation at the end of the
nineteenth century was made by a surgeon in my
hometown of Nottingham, UK.  It was Dr T
Davies Pryce that described what we now know
as neuropathic foot ulcers and stated that these
were a consequence of peripheral nerve degen-
eration [7].
One of the best descriptions of painful neuro-
pathic symptoms was made in 1894 by Pavy, a
physician at the Middlesex hospital in London
[8], who stated that they were of “of a burning and
unremitting character”. Pavy also observed that
the pains were generally worse during the night.
By the end of the nineteenth century, the neu-
ropathies of diabetes were well recognized:
indeed, it was stated that it was rare to meet a
case of diabetes without some evidence of ner-
vous disturbance [1].
3
4 A. J. M. Boulton
2 The Twentieth Century
It was RT Williamson, a physician at the
Manchester Royal Infirmary, who introduced
physiological measurements such as measure-
ment of vibration perception, into the assess-
ment of neuropathy early in the twentieth century
[9]. Coincidentally, I joined the same hospital as
a physician in 1986—and my group confirmed
that loss of vibration sensation was the most reli-
able predictor of foot ulceration in diabetes [10].
After the discovery of insulin, the understand-
ing of the neuropathies increased, and a series of
important publications appeared from US cen-
tres. First, Jordan, in a series of patients, observed
that neuropathic symptoms did not always match
signs in diabetic patients [11].
A decade later Wayne Rundles from the
University of Michigan published observations
on 125 cases of diabetic neuropathy, and not only
proposed an association between glycaemic con-
trol and neuropathy, but also described the fea-
tures of autonomic dysfunction in diabetes [12].
Again coincidentally, the same institution
remains active in research into the neuropathies
with no fewer than three authors in the book com-
ing from the University of Michigan.
It was Sven-Erik Fagerberg, from
Gothenburg, who in 1959 published extensive
studies on 356 cases of diabetic neuropathy,
and in about half, he obtained and studied sec-
tions of peripheral nerve, observing substantial
abnormalities of the nerve microvasculature
[13]. He proposed the potential of a common
aetiopathogenesis of neuropathy, retinopathy,
and nephropathy. Sadly, the closing years of the
twentieth century may well be remembered for
numerous clinical trials of putative new thera-
pies that aimed to tackle the underlying patho-
genesis of neuropathy, none of which showed
clinical efficacy.
To end this historical review on a more posi-
tive note, the art of clinical medicine remains
important in the diagnosis of neuropathy, and it
was my mentor and renowned neuropathy
researcher John D Ward (1935–2019), who
described a clinical sign of autonomic neuropa-
thy in the diabetic foot—distended dorsal foot
veins [14].
In conclusion, most of our understanding of
the diabetic neuropathies has evolved over the
last 250 years. However, much progress has been
made in the first 20 years of this century and
much of this is covered by the excellent chapters
to be found in this up-to-date review of all aspects
of the diabetic neuropathies. I know each and
every one of the authors and can attest that they
are all experts in their own fields and this will
provide a most useful volume to update knowl-
edge on the many various conditions that are
included under the title of diabetic neuropathies.
References
1. Skljarevski V. Historical aspects of the diabetic neu-
ropathies. In: Veves A, Malik RA, editors. Diabetic
neuropathy: clinical management. 2nd ed. Totowa:
Humana Press; 2001. p. 1–6.
2. Rollo J. Cases of the diabetes mellitus with the results
of the trials of certain acids. London: T.  Gillet for
C. Dilly; 1798.
3. Ward JD. A history of diabetic neuropathy in the 19th
century. In: Hotta N, Greene DA, Ward JD, et al., edi-
tors. Diabetic neuropathy: new concepts and insights.
Amsterdam: Elsevier; 1995. p. 89–95.
4. Ward JD.  Historical aspects of diabetic peripheral
neuropathy. In: Boulton AJM, editor. Diabetic neu-
ropathy. Bridgewater: Aventis; 2001. p. 6–15.
5. De Calvi M. Recherches sur les accidents diabetiques.
Paris: Asselin; 1864.
6. Charcot JM.  Sur un cas de paraplegie diabetique.
Arch Neurol. 1890;19:305–35.
7. Pryce TD.  Perforating ulcers of both feet associ-
ated with diabetes and ataxic symptoms. Lancet.
1887;2:11–2.
8. Pavy FW. The Croonian lecture on new departures in
diabetes. Br Med J. 1894;1:1349–50.
9. Williamson RT.  A clinical lecture on the vibrating
sensation in diseases of the nervous system: deliv-
ered at the Manchester Royal Infirmary. Br Med J.
1907;2:125–7.
10. Young MJ, Breddy JL, Veves A, Boulton AJM.  The
prediction of diabetic neuropathic foot ulceration
using vibration perception thresholds: a prospective
study. Diabetes Care. 1994;17:557–60.
11. Jordan WR. Neuritic manifestations in diabetes mel-
litus. Arch Intern Med. 1936;57:307.
12. Rundles RW.  Diabetic neuropathy: general review
with report of 125 cases. Medicine. 1945;24:111–60.
13. Fagerberg S-E.  Diabetic neuropathy: a clinical and
histological study on the significance of vascular
affections. Acta Med Scand. 1959;164(345):1–97.
14. Ward JD, Boulton AJM, Simms JM, et al. Venous dis-
tension in the diabetic foot – physical sign of arterio-
venous shunting. J R Soc Med. 1983;76:1011–4.
 The Epidemiology of Diabetic
Neuropathy
Christian Stevns Hansen, Laura L. Määttä,
Signe Toft Andersen, and Morten H. Charles
1 Introduction
Diabetic neuropathy is a common and severe
complication of diabetes. It is associated with
substantial socioeconomic and personal conse-
quences. The complication encompasses a het-
erogeneous group of neuropathies, which affect
various parts of the nervous system, leading to a
wide range of different symptoms and signs of
nerve fibre dysfunction. In addition, different
types of neuropathy are independent risk factors
for various end-organ manifestations. For exam-
ple, distal symmetrical polyneuropathy (DSPN)
is an independent risk factor for peripheral arte-
rial disease and peripheral amputations, while
cardiovascular autonomic neuropathy (CAN) is
C. S. Hansen
Steno Diabetes Center Copenhagen, Herlev, Denmark
L. L. Määttä
Danish Pain Research Center, Department of Clinical
Medicine, Aarhus University, Aarhus, Denmark
Steno Diabetes Center Aarhus, Aarhus University
Hospital, Aarhus, Denmark
S. T. Andersen
Steno Diabetes Center Aarhus, Regional Hospital
Gødstrup, Herning, Denmark
M. H. Charles (*)
Steno Diabetes Center Aarhus, Aarhus University
Hospital, Aarhus, Denmark
Department of Public Health, Aarhus University,
Aarhus, Denmark
e-mail: mc@ph.au.dk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_2
independently associated with cardiovascular
morbidity and mortality [1].
While the incidence of major diabetic compli-
cations, such as cardiovascular disease, renal dis-
ease, blindness and amputation, has been
decreasing steadily in most countries for the last
few decades [2], we do not know if this is true for
diabetic neuropathy. Despite its severity, the epi-
demiological data at hand is not as substantial as
for many other diabetic complications. This is
linked to a lack of international consensus on the
diagnostic criteria for this heterogeneous compli-
cation and to the lack of robust, long-term
population-­based studies reporting the incidence
of different neuropathies.
Our epidemiological understanding of dia-
betic neuropathy is mainly based on cross-­
sectional studies performed at different time
points. This allows for point estimates of preva-
lence. In addition, data from large longitudinal
cohort studies and randomised trials have shed
light on the natural history of diabetic neuropa-
thy. However, data on indices rate ratios of dia-
betic neuropathy over time are non-existent.
Thus, it cannot be documented if incidence rates
are improving.
Diabetic neuropathy can be classified into four
groups: symmetric polyneuropathy, autonomic
neuropathy, polyradiculopathies, and mononeu-
ropathies. In this chapter, we will present an in-­
depth analyses of the epidemiology of the two
most prevalent forms of diabetic neuropathy: (1)
5
6 C. S. Hansen et al.

Distal symmetrical polyneuropathy (DSPN) and ropathy in both type 1 (T1D) and type 2 diabetes
(2) cardiovascular autonomic neuropathy (CAN). (T2D) constituting approximately 80–90% of
We also provide a short overview of other forms diabetic neuropathies [1]. Diabetic DSPN is
of diabetic neuropathy. termed typical diabetic neuropathy while the
less common forms are called atypical diabetic
neuropathies [3]. In fact, DSPN is the most
1.1 Strategy for Reviewing common form of neuropathy, irrespective of
Relevant Literature aetiology [4], and diabetes is the most common
cause of DSPN [4].
This chapter is a review of major studies carried Common symptoms of DSPN include neuro-
out in the past four decades. In addition to report- pathic pain, numbness, tingling, prickling, tem-
ing data from landmark studies such as UKPDS, perature insensitivity, and weakness [1, 3].
ACCORD, EURODIAB, ADDITION-Denmark, Symptoms normally start in the toes and feet and
the Search Study, DCCT/EDIC, MONICA/ spread proximally, involving the fingers and
KORA, and others, we also performed PubMed hands in more severe cases. This typical symmet-
searches relating to DSPN and CAN.  However, ric distribution of length-dependent neuropathic
the chapter does not have an exhaustive list of symptoms is described as a stocking-and-glove
studies in the field of diabetic neuropathy research. pattern [1, 3]. Dysesthesias and pain are com-
In addition to a presentation of the prevalence and monly the first symptoms and as DSPN pro-
natural history of diabetic neuropathy we aim to gresses, motor involvement may appear [1, 3, 5].
elucidate risk factors for the complication. DSPN has a major impact on the life and health
For DSPN, relevant literature was obtained of affected people due to invalidating pain, com-
from recent comprehensive reviews on diabetic plicating foot ulcers, falls, and amputations lead-
neuropathy and on the epidemiology of diabetic ing to a decreased quality of life [1, 6, 7].
neuropathy. We performed additional searches on No single decisive definition of DSPN exists.
PubMed using the mesh terms “diabetic neuropa- The American Diabetes Association’s (ADA)
thies”, “prevalence”, “incidence”, and “epidemi- definition of DSPN for clinical practice is “the
ology” from 2013 and onwards. From the presence of symptoms and/or signs of peripheral
prevalence-related search, studies with over 400 nerve dysfunction in people with diabetes after
participants were included. the exclusion of other causes” [1]. ADA recom-
We performed PubMed CAN searches using mends that the assessment for DSPN should
the key words “diabetes”, “autonomic neuropa- include a careful history and assessment of both
thy”, “incidence”, and “prevalence”. Studies per- small (temperature or pinprick sensation) and
formed prior to 1980 were not assessed and we large fibre function (vibration perception and
did not include cohorts with less than 80 partici- 10-g monofilament) [8]. Further evaluation by
pants. However, smaller cohorts from non-­ nerve conduction studies (NCS) or skin biopsies
western populations were included due to a to assess intraepidermal nerve fibre density
paucity of studies reported from these regions. (IENFD) is rarely needed and is primarily
reserved for research purposes. The Toronto
Consensus Panel criteria for DSPN are widely
2 Diabetic Distal Symmetric accepted as the current gold standard for diag-
Polyneuropathy nosing DSPN in a research context [3]. The
Toronto criteria define hierarchical categories of
It is estimated that half of all people with diabe- DSPN graded as possible (symptoms or signs),
tes develop neuropathy during their disease [1]. probable (≥2 following: symptoms, signs, abnor-
Half of these cases may be asymptomatic [1]. mal ankle reflexes), or definite (symptoms and/or
Distal symmetric sensorimotor polyneuropathy signs together with a confirmatory test of DSPN
(DSPN) is by far the most common type of neu- (usually NCS, IENFD)) [3, 9].
The Epidemiology of Diabetic Neuropathy
There is emerging evidence that DSPN can be
present even prior to the diagnosis of diabetes
with a prevalence of 10–30% in people with pre-
diabetes or metabolic syndrome [1, 10–17].
However, such cases of DSPN might not be diag-
nosed unless small fibre impairment is examined:
the natural history of DSPN is hypothesised to
start as small fibre damage and proceed towards
large fibre damage [1, 5, 18, 19].
2.1 Issues to Consider When
Evaluating Epidemiological
Studies of DSPN
Summarising the prevalence and incidence of
DSPN based on existing studies is not straight
forward for several reasons [17]. First, the vary-
ing pattern of nerve fibre damage can give rise to
a heterogeneous presentation of DSPN.  For
example, one patient may have decreased vibra-
tion perception threshold (VPT) due to large fibre
damage, while another patient may have intact
VPT but decreased pinprick or temperature sen-
sation due to small fibre dysfunction. Hence, the
chosen method for testing DSPN has an impact
on prevalence estimates [17, 20]. An example
illustrating this issue is the baseline examination
of young adults with T1D in the Diabetes Control
and Complications Trial (DCCT). The prevalence
of DSPN varied from 0.3% (abnormal reflexes,
sensory signs, and neuropathic symptoms) to
21.8% (abnormal NCS in at least two nerves) in
the standard-care group depending on which test
was used to define DSPN [17, 21]. In recognition
of this issue, the current clinical guidelines for
diagnosing DSPN recommend a combination of
tests and symptoms to detect this complication.
Second, several available tests for evaluat-
ing DSPN are subjective in nature, requiring
people to interpret an external stimulus (vibra-
tion, hot or cold application, etc.) and can thus
be hard to reproduce [17]. Third, the preva-
lence varies between different study-popula-
tions. Hospital-­ based study cohorts can be
expected to have a higher prevalence of DSPN
compared with primary-­ care or community-
based cohorts [17]. The optimal method for
7
estimating the prevalence of DSPN is popula-
tion-based survey [17].
In general, age and diabetes duration are
important factors to consider when comparing
the prevalence of DSPN in different studies, irre-
spective of their setting.
2.2 Prevalence of DSPN
The prevalence of neuropathy in the general pop-
ulation is estimated to be 1–8% and approxi-
mately half of these cases are due to diabetes [4,
6, 22, 23]. The prevalence of DSPN in T1D and
T2D for selected studies and settings is given in
Table  1. These studies reveal a varying preva-
lence of DSPN, from 1.5% [30] to 75.1% [67].
Very similar ranges exist when comparing T1D
and T2D (T1D primary care/community-based:
7–62.5% (median 20.8%) and hospital based:
2.6–55.1% (median 22.1%); and T2D primary
care/community-based: 1.5–63% (median
27.4%) and hospital based: 18.1–75.1% (median
34.4%). The median values are calculated from
the studies included in Table 1 with large varia-
tion in age, diabetes duration, diagnostic tests,
etc.). Furthermore, it is worth noting that studies
defining DSPN by NCS tend to report a higher
prevalence of DSPN compared to other diagnos-
tic tests, underpinning the high sensitivity of
NCS to detect DSPN [29, 31, 44]. Overall, the
prevalence of DSPN increases with diabetes
duration and with increasing age despite the set-
ting or testing method.
In newly diagnosed diabetes, the prevalence
of DSPN tends to be lower [25, 31–34, 41, 45–
47, 65, 68]. A population-based survey of DSPN
in Mauritius reported a prevalence as low as 3.6%
in newly diagnosed T2D (N = 414) where DSPN
was defined using age-specific VPT [34].
Population-based samples from Egypt and India
show a prevalence of 14% (N = 125) and 19.5%
(N = 338), respectively, in people with newly
diagnosed T2D. These studies also used VPT to
detect DPSN but did not apply age-specific refer-
ence values [33, 41]. The ADDITION-Denmark
study, which included individuals with screen-­
detected T2D (N = 1445), reported a prevalence
 8
Table 1  Prevalence of DSPN in type 1 and 2 diabetes
Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 1 (P) USA 363 ≥2 of the following: 34 19.9 28.4 DM duration, HbA1c, smoking status,
EDC [24] symptoms, signs (light HDL-c
touch, pinprick, VPT,
proprioception, muscle
weakness, gait), knee/
ankle reflexes
Type 1 (P) Italy [25] 379 Symptoms and ≥1 sign 28.5 NA NA (15–59 NA
OR ≥2 signs year)
(VPT, ankle/knee
reflexes, foot ulcers, DB,
postural hypotension)
Type 1 (P) Denmark [26] 339 VPT 62.5 13 20 Male sex, age, increased albumin
excretion rate
Type 1 (P) USA [27] 5936 MNISq ≥ 4 11 18 (median) 39 Age, HbA1c, DM duration, lower
socioeconomic status, CVD, smoking,
hypertriglyceridemia, BMI,
retinopathy, reduced eGFR, Charcot
neuroarthropathy, recent severe
hypoglycaemia and/or DKA
Type 1 (P) Scotland [28] 5558 MNNIq ≥ 4 13 20.5 44.7 Age, DM duration, waist-to-hip ratio,
lipids, HbA1c, smoking, low eGFR,
low HDL-c, socioeconomic status, ≥1
other DM complication (nephropathy,
retinopathy, CVD, recent severe
hypoglycaemia or DKA)
Type 2 (P) USA [29] 78 NCS 46.2 NA 62.1 NA
Type 2 (P) Finland [30] 132 (1) Symptoms (1) 1.5 0 Men 54.9 NA
(2) Knee/ankle reflexes (2) 2.3 Women
and ≥1 sign (light touch, (3) 15.2 57.1
pinprick, VPT) OR ≥2
signs
(3) NCS
C. S. Hansen et al.
 Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 2 (P) Finland [31] 132 NCS with or without Baseline: 8.3 10 56 Fasting Glc levels, lower insulin levels
symptoms 10-year at fasting and after oral Glc
follow-up: administration
41.9
Type 2 (P) USA 279 ≥2 of following: 25.8 NA NA (20–74 Age, DM duration, male sex, HbA1c
San Luis Valley Symptoms, ankle year)
Diabetes Study [32] reflexes, temp. sensation
Type 2 (P) Egypt [33] 384 VPT 21.9 NA NA (over HbA1c, age, female sex,
20 year) hypercholesterolemia
Type 2 (P) Mauritius [34] 433 VPT 12.7 6.6 55 DM duration, treatment with insulin or
oral hypoglycaemic agents, height,
fasting Glc, lower 2-h plasma insulin
Type 2 (P) Netherlands [35] 137 (1) Symptoms (1) 18 7.5 68 NA
(2) VPT (2) 53
The Epidemiology of Diabetic Neuropathy

(3) Knee/ankle reflexes (3) 62

(4) Temp. sensation (4) 63
Type 2 (P) Italy [36] 347 ≥2 symptoms and ≥2 of 19 NA NA (≥55 DM duration, fasting and post-prandial
following: “sensation, year) Glc
strength, tendon reflexes”
Type 2 (P) UK [37] 811 NDS ≥ 6 41.6 7.4 65.4 NA
Type 2 (P) Portugal [38] 93 Knee/ankle reflexes and 32.2 10.1 65.4 Age, DM duration, foot skin changes,
≥1 sign (VPT, MI/ischemia
monofilament, pinprick,
temp. sensation,
proprioception)
Type 2 (P) Malaysia [39] 134 NSS and NDS 50.7 8.5 56.5 Age, DM duration
Type 2 (P) Bahrain [40] 1477 ≥2 of following: 36.6 9.5 57.3 Age, DM duration, HbA1c, total
symptom score, sign cholesterol, smoking, BMI, waist
score (“pain”, “touch”, circumference, triglycerides,
temp. sensation, VPT), hypertension
VPT
Type 2 (P) India [41] 1291 VPT 27.8 5.9 +DSPN 60 Age, HbA1c, DM duration
–DSPN 47
(continued)
9
 10
Table 1 (continued)
Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 2 Germany 195 MNSI exam > 2 28 NA 66.8 Age, waist circumference, PAD
predominantly MONICA/KORA
(P) [42]
Type 2 (P) Sweden [43] 156 VPT and/or 34 7.0 61.7 PAD, age, male sex, lower
monofilament HDL-cholesterol
Type 2 (P) Taiwan [44] 133 NCS 29.3 +DSPN 3 +DSPN Age, renal insufficiency, HbA1c,
–DSPN 2 66.2 fasting Glc, systolic BP
–DSPN
54.9
Type 2 (P) UK 3867 (1) Ankle reflexes Baseline: 0 Baseline NA
UKPDS [45] (2) VPT (1) INT 11.8; 53.3
CON 11.4
12-year
follow-up:
(1) INT 35;
CON 37
(2) INT 30.2;
CON 32.8
Type 2 (P) Denmark 1445 MNSIq ≥ 4 13.1 0 NA (–DPN NA
ADDITION-DK 60.8)
Baseline [46]
Type 2 (P) Denmark 1161 (1) Monofilament and/or (1) INT 30.1; 6 INT 59.6, NA
ADDITION-DK 6-yr VPT CON 34.8 CON 59.9
follow-up [47] (2) MNSIq ≥ 7 (2) INT 8.7;
CON 9.3
Type 2 (P) Denmark 452 NCSa and ≥ 1 symptom 27.0 +DSPN 12.9 70.9 Baseline HbA1c, steeper increases in
ADDITION-DK and/or sign (ankle reflex, –DSPN 12.7 HbA1c over time, weight, waist
13-yr follow-up [48] VPT, monofilament) circumference, BMI
Type 2 (P) Brazil [49] 551 (1) NSS = 5–6 OR NSS (1) 6.3 10 65 DM duration
= 3–4 and NDS = 6–8 (2) 14.3
(2) Monofilament
Type 1 + 2 (P) Australia [50] T1 179 Pinprick T1 8.4 T1 12.6 T1 33.8 DM duration, age at DM diagnosis
T2 904 T2 16.8 T2 7.2 T2 63.6
C. S. Hansen et al.
 Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 1 + 2 (P) USA [51] T1 102 ≥2 of following: T1 54 T1 14.5 T1 41 NA
T2 278 symptoms (NSS [52]), T2 45 T2 8.1 T2 63.2
signs (NDS [52]), NCS, (median)
VPT, CDT, DB, VAL,
with ≥1 being NCS, DB
or VAL
Type 1 + 2 USA [53] T1 124 Symptoms T1 30.2 T2 T2 DM duration, hypertension,
(P) T2 2268 T2 37.9 +DSPN 12.0 +DSPN hyperglycaemia, glycosuria
–DSPN 9.6 61.8
–DSPN
61.7
Type 1 + 2 UK [54] T1 213 ≥2 of the following: T1 12.7 NA NA T1: height, retinopathy
(P) T2 864 symptoms, light touch, T2 17.2 T2: age, height, alcohol, HbA1c,
pinprick, ankle reflexes, retinopathy
The Epidemiology of Diabetic Neuropathy

VPT (only for <70 year

olds)
Type 1 + 2 USA T1 1734 MNSI exam >2 T1 7 T1 7.2 T1 18 T1: Age, DM duration, poor glycaemic
(P) SEARCH [55] T2 258 T2 22 T2 7.9 T2 22 control, smoking, diastolic BP, obesity,
increased LDL-c and triglycerides,
lower HDL-c
T2: Age, male sex, DM duration,
smoking, lower HDL-c
Type 1 (H) Europe 3250 ≥2 of the following: 28.5 14.7 32.7 Age, DM duration, HbA1c, height,
EURODIAB Baseline symptoms, ankle/knee background or proliferative diabetic
[56] reflexes, VPT, HRV and/ retinopathy, cigarette smoking,
or postural hypotension reduced HDL-c, CVD, diastolic BP,
severe ketoacidosis, raised fasting
triglyceride, microalbuminuria
(continued)
11
 12
Table 1 (continued)
Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 1 (H) Europe 1172 Same as above Cumulative – – HbA1c, change in HbA1c during
EURODIAB incidence follow-up, DM duration, elevated total
7.3-year follow-up 23.5 and LDL-c, elevated triglycerides,
[57] higher BMI, higher von Willebrand
factor, elevated urinary albumin
excretion rate, hypertension, smoking,
CVD at baseline
Type 1 (H) USA and Canada 1186b NCS and ≥2 of the Baseline: Baseline: INT Baseline: Higher mean HbA1c, age, DM
DCCT/EDIC following: symptoms, INT 7, 6; CONV 6 INT 27; duration, height, macroalbuminuria,
(Summarising signs (light touch, CONV 5 Closeout: INT CONV 27 higher mean pulse rate, β-blocker use,
DPN-findings: Albers proprioception, temp. Closeout: 12; CONV 12 Closeout: sustained albuminuria
et al. [58], Martin sensation, pinprick), INT 9; EDIC year INT 34;
et al. [59], risk factors ankle reflexes CONV 17 13/4: INT 26; CONV 33
[60]) EDIC year CONV 26 EDIC year
13/4: INT 25; 13/4: INT
CONV 35 48; CONV
47
Type 1 (H) Denmark [61] 156 (1) Confirmed: symptoms (1) 2.6 11.3 22 None
or signs (monofilament, (2) 55.1
VPT, pinprick) and NCSa
or ECSc
(2) Subclinical: abnormal
NCS or ECS
Type 2 (H) USA [62] 775 Monofilament 50 11 62 Height, previous foot ulcer, age,
HbA1c, alcohol, smoking, albumin
level adjusted for serum creatinine
Type 2 (H) Tanzania [63] 153 Score based: 28 5.2 44.2 NA
Several signs (VPT,
“sensibility”, hallux
flexion, hair loss, foot
deformities, dry/cracked
skin) OR signs and
symptoms (incl.
claudication)
Type 2 (H) Saudi-Arabia [64] 375 Pinprick and/or vibration 20 8 (median) 50 (median) NA
sensation
C. S. Hansen et al.
 Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 2 (H) Netherlands [65] 73 VPT 50 6.6 (median) 65.9 NA
Type 2 (H) France [66] 427 Ankle reflexes and/or 31.6 10.6 56.9 NA
VPT
Type 2 (H) Sweden [64] 79 Pinprick and/or vibration 19 11.5 (median) 58 (median) NA
sensation
Type 2 (H) Iran [67] 810 ≥1 of following: 75.1 8.2 52.7 Age, proteinuria, DM duration,
pinprick, VPT, light insulin-treatment, retinopathy, IHD
touch, proprioception,
ankle reflexes, muscular
power
Type 2 (H) China [68] 1067 VPT and/or 18.3 7.2 59.8 Retinopathy, HbA1c, DM duration,
monofilament age
Type 2 (H) Denmark 160 VPT Baseline 34.4 INT 5.5 55.1 NA
Steno 2 [69] CON 6.0 (40–65) (INT (targeting hyperglycaemia, risk
The Epidemiology of Diabetic Neuropathy

factors and behaviour) without

significant effect on DSPN)
Type 2 (H) USA 10,251 MNSI exam >2 Baseline: 10 62 INT glycaemic therapy decreased
ACCORD [70] INT 43; DSPN risk
CON 42
5-year
follow-up:
INT 55.6;
CON 58.6
Type 2 (H) China [71] 562 NCS and symptoms and/ 18.1 5.6 56.4 Age, DM duration, hypertension,
or signs (ankle reflex, insulin resistance index, HbA1c,
VPT, monofilament, HbA1c variability, u-ACR
temp. sensation, pinprick,
proprioception
Type 1 + 2 Italy [72] 8757 ≥2 points: VPT, ankle 32.3 12.1 (data only 58.8 NA
(H) reflexes, foot ulcer, other from records of
foot deformity/dry skin/ 1227
infection (1 point per participants)
abnormality per side
(L/R))
(continued)
13
 14
Table 1 (continued)
Participants DM duration Mean age
Study site (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 1 + 2 (H) Germany, Austria, T1 647 ≥2 of following: T1 17.1 T1 10.3 T1 33 T1 + T2: Age, retinopathy,
Switzerland [73] T2 524 symptoms, knee/ankle T2 34.8 T2 9.0 T2 54 macroangiopathy, CAN
reflexes, VPT (medians) (medians) T1: albuminuria, HbA1c
Type 1 + 2 (H) UK [74] T1 2414 NDS ≥ 6 OR NDS = 3–5 T1 22.7 T1 13 T1 45 Age, DM duration
T2 3949 and NSS ≥ 5 T2 32.1 T2 6 T2 63
(medians) (medians)
Type 1 + 2 Iran [75] T1 79 NDS ≥ 6 OR NDS = 3–5 T1 21.5 T1 9.5 T1 28.2 T1: DM duration, education level
(H) T2 521 and NSS ≥ 5 T2 49.3 T2 9.2 T2 57.0 T:2 history of foot ulcer, age, DM
duration, weight, education level, male
sex
Type 1 + 2 Spain [76] T1 348 NDS ≥ 6 OR NDS = 3–5 T1 12.9 T1 13.8 T1 30.5 T1: DM duration
(H + P) T2 2296 and NSS ≥ 5 T2 24.1 T2 9.7 T2 61.3 T2: age, DM duration
BMI body mass index, BP blood pressure, CDT cooling detection threshold, CON conventional treatment, CVD cardiovascular disease, DB deep breathing, DKA diabetic ketoaci-
dosis, DM duration diabetes mellitus duration, DN4 Douleur Neuropathique 4 Questions, ECS electrochemical skin conductance, Glc glucose, H Hospital/out-patient based,
HbA1c hemoglobin A1c, HDL/LDL/total-c high density lipoprotein/low density lipoprotein/total chlolesterol, IHD ischemic heart disease, INT intensive treatment, MI myocar-
dial infarction, MNSI examination Michigan neuropathy screening instrument examination, MNSIq Michigan neuropathy screening instrument questionnaire, NCS nerve conduc-
tion studies, NNS and NDS Neuropathy symptom score and Neuropathy disability score (Young et al. [74]), P primary care/community-based, PAD peripheral arterial disease,
S-LANSS self-report Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) pain scale, u-ACR urin albumin-creatinine ratio, VAL Valsalva’s manoeuvre, VPT
vibration perception threshold
a
NCS measured with handheld DPNCheck©-device
b
Number of participants at EDIC year 13/14, mean age and DM duration reported for these
c
ECS measured with Sudoscan©-device
C. S. Hansen et al.
The Epidemiology of Diabetic Neuropathy
of 13.1% at the time of diabetes diagnosis using
the Michigan Neuropathy Screening Instrument
questionnaire (MNSIq) to define DSPN [46].
Similarly, the prevalence of DSPN in the United
Kingdom Prospective Diabetes Study (UKPDS)
of 3,867 people with newly diagnosed T2D was
11.5%, defined using VPT [45]. Out-patient
clinic-based samples show prevalence’s ranging
from 6.4% [68] to 39% [65] in people with newly
diagnosed T2D.  In summary, estimates of the
prevalence of DSPN in newly diagnosed T2D
range from 4% and 40%, with lower prevalence
(<20%) in larger study-populations.
It is clear from the summary data presented in
Table 1 that the comparison of DSPN prevalence
between studies is challenging. When evaluating
studies of T1D in primary care or community-­
based cohorts, a DSPN prevalence of 7% was
found in young people with T1D in the large
population-based cohort (N = 1734) of the
SEARCH for Diabetes in Youth-study. Here,
DSPN was assessed using the MNSI examination
[55]. In contrast, a smaller Danish study (N =
339) reported a prevalence of 62.5% in young
people with T1D based on VPT [26]. Recently,
DSPN prevalence was estimated using the
MNSIq in two large cohorts of nearly 6000 adults
with T1D in the USA [27] and Scotland [28]
reporting very similar prevalences of 11% and
13%, respectively. However, studies that define
DSPN using symptoms alone might underesti-
mate the true prevalence of DSPN by overlook-
ing the large fraction of asymptomatic cases.
Finally, the population-based Rochester Diabetic
Neuropathy Study cohort found a DSPN preva-
lence of 54% in 102 people with T1D when using
a more complex DSPN definition similar to the
“definite DSPN” category of the Toronto classifi-
cation [51].
Among hospital-based studies of cohorts with
T1D, the European IDDM Prospective
Complications Study (EURODIAB) of over 3000
people showed a prevalence of 28.5% when
defining DSPN using symptoms and signs [56].
Another large multicentre study from the UK
found a prevalence of 22.7% in a cohort with
T1D with a similar DSPN definition [74]. The
DCCT, and its observational follow-up study
15
(EDIC), used a robust definition of DSPN requir-
ing ≥2 of following in addition to abnormal NCS:
symptoms, peripheral sensory signs, and abnor-
mal ankle reflexes. The prevalence in the conven-
tional treatment group was 5% at baseline,
increasing to 35% in the EDIC year 13/14 [58,
59].
In T2D, studies conducted in primary care or a
community-based setting predominantly show a
DSPN prevalence of between 20% and 50%,
again varying by diabetes duration and the
method used for detecting DSPN.  A purely
symptom-based study of more than 2200 people
with T2D in the USA, with a T2D duration of 10
years, reported a DSPN prevalence of 37.9%
[53]. In contrast, symptom-based DSPN preva-
lence early in the course of T2D is as low as 1.5%
[30]. The UKPDS (newly diagnosed T2D) and
ADDITION-Denmark study (screen-detected
T2D) also describe relatively low DSPN preva-
lences of 11.5% and 13.1%, respectively [45, 46].
In contrast, the Rochester Diabetic Neuropathy
Study cohort, with its robust DSPN definition,
found a prevalence of 45% in people with T2D
with a diabetes duration of approximately 8 years
[51]. An even higher prevalence of DSPN was
seen in a smaller primary care-based study from
the Netherlands, which defined DSPN as abnor-
mal temperature sensation with a prevalence of
63% (N = 137) in people with T2D [35]. By con-
trast, a larger sample (N = 904) evaluating DSPN
with pinprick in people with T2D in an Australian
rural area found a prevalence of only 17%,
although age and diabetes duration were similar
to the cohort in the Netherlands [50]. Although
the mean age in most T2D studies is around 60
years, the SEARCH-study, which included 258
young adults with T2D (mean age 22 years),
found a DSPN prevalence of 22%—a prevalence
that is similar to that observed in older T2D pop-
ulations [55].
The prevalence of DSPN in hospital-based
studies of T2D are generally higher than the prev-
alence of DSPN in T1D or in primary care/
community-­ based samples of T2D.  Two large
Asian out-patient surveys report a DSPN preva-
lence of approximately 20% when evaluated
using VPT and/or monofilament [68, 77]. The
16
prevalence in an age- and T2D duration compa-
rable multicentre study from Korea was higher
(33.5%). This is likely explained by the fact that
both symptoms and signs were included their
DSPN definition [78]. Similarly, an earlier
UK-based multicentre study of nearly 4000 peo-
ple with T2D showed a prevalence of 32.1%
when using both symptoms and signs, or several
abnormal signs [74]. The Action to Control
Cardiovascular Risk in Diabetes (ACCORD) trial
of more than 10,000 people defined DSPN using
the MNSI examination and found a baseline
prevalence of 42%. This increased to 58.6% in
the conventional therapy group after 5 years [70].
The high prevalence of DSPN in this cohort is
likely explained by the high cardiovascular risk
profile of the participants. An exceptionally high
prevalence was reported in an Iranian study
including 810 people that defined DSPN using a
clinical exam score where any single impairment,
including small fibre impairment, qualified for a
DSPN diagnosis. The reported prevalence was
75.1% in this middle-aged population with a T2D
duration of 8 years [67].
To conclude, DSPN is a highly prevalent com-
plication of both T1D and T2D. There is a wide
range of reported prevalences, with median prev-
alences around 20–30% for both T1D and
T2D. An increasing prevalence of DSPN is seen
with increasing diabetes duration in both T1D
and T2D. In summary, varying testing methodol-
ogies and differences in population characteris-
tics prevent a simple comparison between DSPN
prevalences. However, it is estimated that an
equal proportion of people with T1D and T2D
are affected by DSPN during the course of their
disease [6, 17].
2.3 Painful Diabetic Neuropathy
Neuropathic pain is defined as “pain caused by a
lesion or disease of the somatosensory nervous
system” [79]. In the case of painful diabetic neu-
ropathy, the damage to the peripheral nervous
system is due to diabetes. As for DSPN, no single
gold standard exists for the diagnosis of painful
diabetic neuropathy, where the pain location
C. S. Hansen et al.
reflects the area of damaged nerve fibres. Similar
to DSPN, the distribution of pain is most often
found in a stocking-and-glove pattern. For this
reason painful diabetic neuropathy will be
referred to as painful DSPN (P-DSPN). Alike
DSPN, neuropathic pain can also be classified as
possible, probable, and definite according to a
hierarchical grading system proposed by the
Neuropathic Pain Special Interest Group of the
International Association for the Study of Pain
(NeuPSIG) [80].
In practice, the diagnosis of P-DSPN relies on
the patient’s subjective description of pain. The
pain is typically described as burning, prickling,
lancinating, tingling or shooting and may occur
together with paresthesias, hyperalgesia, and
allodynia and is often worse at night [1, 3, 81].
Pain may be the first symptom of DSPN that
alerts the patient to seek health care and it has a
major impact on the quality of life of affected
people by interfering with daily activities, sleep,
and enjoyment of life [1, 7, 17, 81–83]. Despite
this, P-DSPN might be underdiagnosed and
undertreated as recent reports have shown that up
to 80% of people with P-DSPN may be undiag-
nosed [84, 85], stressing the importance of accu-
rate screening of DSPN and P-DSPN to
adequately manage the pain and its adverse
effects on quality of life [1].
As for DSPN, the prevalence of P-DSPN var-
ies widely but it is lower than the overall preva-
lence of DSPN. Selected studies (Table 2) report
a prevalence between 5.4% and 42.2% for studies
of T1D and T2D in both hospital- and primary-­
care/community-based settings with a median
prevalence of around 17%. A large community-­
based UK cohort with a predominance of people
with T2D found a P-DSPN prevalence of 13.4%
in T1D and 21.5% in T2D [91]. Interestingly,
26% of the non-DSPN group presented with
P-DSPN suspicious pain, which might indicate
an early stage of nerve fibre damage that is not
detectable by clinical examination. Smaller pri-
mary care/community-based studies from the UK
and Germany that define P-DSPN with both clin-
ical signs and symptoms of pain, report P-DSPN
prevalences between 13.3% and 26.4% in T2D
[82, 86, 87, 90].
Table 2  Prevalence of painful DSPN, and DSPN when stated, in type 1 and 2 diabetes
DM duration Mean age
Study site Participants (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 2, UK [82] 269 Toronto clinical scoring system >5 and 26.4 8 66.7 NA
P-DSPN (P) history suggesting neuropathic pain
Type 2, Germany [86] 195 MNSI exam >2 and neuropathic pain 13.3 NA 66.8 Age, weight, PAD
P-DSPN (P) (MNSIq)
Type 2, Germany [87] 214 MNSI exam >2 and neuropathic pain 21 NA 69 Waist circumference,
P-DSPN (P) (MNSIq) physical activity, PAD
Type 2, Denmark 5514 (1) Possible DSPN: MNSIq ≥ 4 (1) 18 4.6 64.1 (1) Female sex, age, DM
P-DSPN (P) DD2 [88] (2) Possible P-DSPN: bilateral pain in (2) 10 duration, BMI, smoking
the feet and DN4 questionnaire ≥3 (2) Smoking
Type 2, Denmark 5514 (1) Possible DSPN: ≥1 of following: (1) 62.2 . . NA
P-DSPN (P) DD2 [89] Estimates (2) 43.9
The Epidemiology of Diabetic Neuropathy

symptoms, signs (light touch, pinprick,

based sample temp. sensation, VPT), ankle reflexes (3) 22.7
of N = 389 (2) Probable DSPN: ≥2 of the above 3 (4) 5.4
(3) Definite DSPN: NCS or IENFD
and ≥1 of the above 3
(3) Definite P-DSPN: NCS or IENFD
and bilateral pain in feet, signs
Type 2, Korea [78] 3999 DSPN: MNSIq ≥ 3 and monofilament DSPN 33.5 P-DSPN 13.1 P-DSPN P-DSPN: age, female
P-DSPN (H) OR typical DSPN signs and symptoms of which 43.1 Non-P-­DSPN 63.6 gender, fasting Glc,
evaluated by study physician P-DSPN 11.4 Non-P-­ hypertension, previous
P-DSPN: DSPN and VAS ≥ 4 last 48 h DSPN 62.5 cerebrovascular events
or use of pain medication for pain in
legs and feet or legs and hands
Type 2, Taiwan [77] 2837 DSPN: Monofilament and/or VPT DSPN 21.3 10.1 63.9 DSPN: age, insulin
P-DSPN (H) P-DSPN: DSPN and DN4 ≥ 4 of which treatment,
P-DSPN 21.2 microalbuminuria, overt
proteinuria
P-DSPN: age, HbA1c,
lower HDL-c, overt
proteinuria
Type 2, Qatar [84] 1095 DN ≥ 4 34.5 P-DSPN 13.6 P-DSPN VPT, smoking, obesity,
P-DSPN (H) Non-P-­DSPN 57.5 female sex, DM duration
8.2 Non-P-­
DSPN 52.6
(continued)
17
 18
Table 2 (continued)
DM duration Mean age
Study site Participants (N) Neuropathy diagnosis Prevalence (%) (mean, year) (year) Risk factors/markers
Type 1 and 2, UK [90] 350 VPT, NSS, NDS, Pain symptoms score 16.2 NA 63 NA
P-DSPN (P) ≥3 and neuropathic pain in legs ≥1
year
Type 1 + 2, England [91] T1 1338 NSS ≥ 5 and NDS ≥ 3 T1 13.4 T1 17 T1 37.6 Risk for painful
P-DSPN (P) T2 14,206 T2 21.5 T2 4 T2 63.6 symptoms: T2D, female
Overall 21 gender, South Asian
ethnicity
Type 1 + 2, Germany T1 126 DSPN: ≥1 of following: T1 44.3 of NA T1 59.5 T2 DSPN + P-DSPN:
P-DSPN (P) PROTECT study T2 943 monofilament, temperature sensation, which T2 70.0 higher and lower BMI
[85] VPT P-DSPN 54.8
P-DPSN: DSPN and pain/burning in T2 55.3 of
feet which
P-DSPN 61.7
Type 1 + 2, Belgium [7] T1 344 DSPN: Monofilament and/or pinprick T1: T1 16.5 T1 45.9 Nephropathy, obesity,
P-DSPN (H) T2 767 Painful-DSPN: DSPN and DN4 >4 DSPN 25.6 T2 11.0 T2 63.6 low HDL-c, high
P-DSPN 5.8 triglycerides
T2:
DSPN 50.8
P-DSPN 17.9
Type 1 + 2, Libya [92] T1 + T2 450 S-LANSS ≥ 12 42.2 15 50.6 DM duration, smoking
P-DSPN (H) (men), obesity, fasting
Glc
BMI body mass index, BP blood pressure, CDT cooling detection threshold, CON conventional treatment, CVD cardiovascular disease, DB deep breathing, DKA diabetic ketoaci-
dosis, DM duration diabetes mellitus duration, DN4 Douleur Neuropathique 4 Questions, ECS electrochemical skin conductance, Glc glucose, H Hospital/out-patient based,
HbA1c hemoglobin A1c, HDL/LDL/total-c high density lipoprotein/low density lipoprotein/total chlolesterol, IHD ischemic heart disease, INT intensive treatment, MI myocar-
dial infarction, MNSI examination Michigan neuropathy screening instrument examination, MNSIq Michigan neuropathy screening instrument questionnaire, NCS nerve conduc-
tion studies, NNS and NDS Neuropathy symptom score and Neuropathy disability score (Young et al. [74]), P primary care/community-based, PAD peripheral arterial disease,
S-LANSS self-report Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) pain scale, u-ACR urin albumin-creatinine ratio, VAL Valsalva’s manoeuvre, VPT
vibration perception threshold
C. S. Hansen et al.
The Epidemiology of Diabetic Neuropathy 19

Recently, lower P-DSPN prevalences have neurological deterioration is associated with

been reported in the Danish DD2 cohort of increasing age [17] and the risk factor profile of
recently diagnosed T2D.  Confirmed P-DSPN participants may change over time. Furthermore,
was estimated to be present in only 5.4% of the nerve fibre dysfunction may be partly reversible
cohort of more than 5500 individuals when eval- [17, 99–101], further complicating interpretation
uated clinically in a sub-sample of 389 individu- of longitudinal studies of DSPN.
als [89]. By contrast, P-DSPN prevalences from A long-term clinic-based study published in
two hospital-based studies in Qatar and Libya the 1970s followed over 4000 people with T1D
report high prevalences of 34.4% and 42.2%, and T2D for 25 years and found a cumulative
respectively. The latter study was, however, based DSPN incidence of 50% [102–104]. Even though
solely on self-reported and self-evaluated neuro- some aspects of the methodology can be ques-
pathic pain [84, 92]. However, prevalences simi- tioned, the size of the study is noteworthy. DSPN
lar to the ones reported in community-based incidence has since been evaluated in more recent
samples are also found in out-patient settings [7]. study cohorts. These are summarised in the fol-
Reports characterising both DSPN and P-DSPN lowing paragraph.
estimate that approximately 20–40% of people In newly diagnosed diabetes, a small Finnish
with DSPN have neuropathic pain [7, 77, 78, 89]. cohort of people with T2D showed an increase
There is a large disparity between painful from a baseline prevalence of 8.3% to 41.9%
symptoms and clinical signs of DSPN and little is over 10 years when defining DSPN using NCS
known about the natural history of P-DSPN [5, 6, and symptoms [31]. The UKPDS cohort of newly
91, 93, 94]. There are few longitudinal studies of diagnosed T2D evaluated DSPN by VPT and
P-DSPN and most are small. These provide con- found that the prevalence increased from approx-
flicting results, indicating both spontaneous pain imately 11–30% over 12 years, irrespective of
improvement and remission over time [95, 96] as treatment arm [45]. In comparison, the
well as the opposite finding [95, 97]. As recent ADDITION-Denmark cohort of screen-detected
reviews report the duration and severity of diabe- T2D report a cumulative incidence of only 10%
tes, and the severity of neuropathy, as risk factors over 13 years of follow-up [46]. In contrast, the
for P-DSPN, pain remission seems unlikely [6, Bypass Angioplasty Revascularization
83]. In conclusion, approximately 15% of people Investigation 2 Diabetes (BARI 2D) trial of more
with diabetes have P-DSPN but estimates vary than 2,000 people with longstanding T2D and
and may be as high as 40%. confirmed coronary artery disease, showed a
DSPN prevalence of 50% at baseline and a cumu-
lative incidence over 4 years of 66–72% when
2.4 Incidence of DSPN defining DSPN using the MNSI examination
[105]. The ACCORD trial similarly defined
Although reported prevalences of DSPN in T1D DSPN using the MNSI examination. Although
and T2D are similar, the incidence of DSPN is the trial was closed earlier than planned due to
estimated to be higher in T2D (6.1 per 100 higher mortality in the intensive treatment arm,
person-­years) than in T1D (2.8 per 100 person-­ intensive treatment (INT) decreased the risk of
years). This is probably due to differences in the DSPN compared to conventional care (CON).
underlying pathophysiology and populations Baseline prevalences of 42% (CON) and 43%
characteristics of T1D and T2D [1, 6, 17, 98]. (INT) increased to 59% (CON) and 56% (INT)
The slow progression of DSPN complicates after 5 years [70]. Two other smaller longitudinal
investigation of incident DSPN and highlights studies of T2D report an annual incidence of
the importance of testing methodology. Different 6.1% when evaluating DSPN with both symp-
nerve fibre populations may be affected at differ- toms and signs [106] and 20% over 2.5 years
ent rates and minor changes may be undetected when evaluating DSPN by monofilament [62].
due to low sensitivity of tests. In addition, natural More recently, an Indian study of T2D reported a
20
4-year incidence of 28.4% evaluated using VPT
[107].
The landmark studies of DCCT/EDIC provide
detailed information on the incidence of DSPN
[21]. Incidence of confirmed DSPN was 6% and
14% in the intensive and the conventional treat-
ment arms, respectively, at DCCT closeout after
approximately 5 years. The incidence in the
EDIC cohort at year 13/14 was 22% (INT) and
28% (CON), respectively, and the prevalence was
25% (INT) and 35% (CON) [58, 59]. Underlying
NCS abnormalities were also more prevalent in
the CON group and consistently higher than the
prevalence of confirmed or clinical DSPN in both
the INT and CON groups [59].
The observational Epidemiology of Diabetes
Complications (EDC) study of childhood-onset
T1D reported a 15% prevalence of DSPN at the
6-year follow-up in participants free from
DSPN at baseline, corresponding to an inci-
dence of 2.8 per 100 person-years [108]. At 12
years, the incidence per 100 person-years was
1.2 for people with a T1D duration <20 years
and 3.8 for a T1D duration of 20–30 years
[109]. The EURODIAB study for risk factors
for the development of DSPN followed over
1000 T1D patients in multiple centres over
Europe for 7.3 years and found a cumulative
incidence of 23.5% using symptoms and signs
[56, 57]. More recently, a population-­ based
cohort of over 4000 people with newly diag-
nosed T1D from Taiwan reported of a rather
low cumulative incidence of 23.7% over 12
years [110]. However, the participants were fol-
lowed from the time of T1D diagnosis and the
study was based on registries where the detec-
tion of incident DSPN relied on a registered
DSPN diagnosis during follow-up.
In summary, the incidence of DSPN is gener-
ally higher in T2D than T1D. However, there are
exceptions, specifically lower than usual inci-
dence of DSPN in T2D cohorts early in their dis-
ease trajectory. No studies have investigated
incidence rate ratios. Thus, no data is available on
temporal changes in incidence of DSPN. In other
words, it is not possible to assess if the incidence
of DSPN follows the decline as for other diabetic
complications.
C. S. Hansen et al.
2.5 Risk Factors for DSPN
and P-DSPN
A long list of risk factors has been associated
with DSPN.  Risk factors for DPSN in the
reviewed studies are summarised in Tables 1 and
2 and additional studies are referenced to in the
following paragraph.
The duration of diabetes and level of hyper-
glycaemia (HbA1c) are the two most important
risk factors for DSPN [6, 17, 57, 111]. The
importance of hyperglycaemia has been docu-
mented in both T1D and T2D [27, 28, 31, 33, 51,
55, 57, 62, 112] and the effect of strict glycaemic
control has been investigated in several studies
[45, 58, 59, 70, 113, 114]. Optimising glycaemic
control in T1D delays development of DSPN and
slows down the progression of DSPN, whereas
the effect of glucose control is less clear in T2D
[115]. More recently, obesity, including measures
of weight, waist circumference, and BMI, have
emerged as important risk factors of DSPN [15–
17, 46, 48, 111, 116]. Other components of the
metabolic syndrome, such as hypertension, insu-
lin resistance, and dyslipidemia, also seem to
increase the risk of DSPN [6, 16, 17, 57, 111,
116]. Importantly, such factors are modifiable.
Other independent risk factors for DSPN are
smoking, alcohol, increased height (probably as a
proxy for nerve fibre length), and older age [17,
57, 111]. Alcohol consumption is also regarded as
a separate cause of peripheral neuropathy, but the
distinction between alcohol and diabetes as causal
risk factors for DSPN is difficult. The effect of
age on DSPN is controversial since age per se
causes progressive neurological deterioration and
although age has been widely reported as an inde-
pendent risk factor for DSPN, there are also stud-
ies that show the opposite effect [17, 111]. Finally,
recent large national cohorts of T1D in the USA
and Scotland indicate that socioeconomic disad-
vantage may be an additional independent risk
factor for the development of DSPN [27, 28].
Independent risk factors for the development of
P-DSPN have not been studied to the same extent
as risk factors for DSPN. However, any risk factor
for diabetes or DSPN is likely to be a risk factor
for P-DSPN and clear similarities with the known
The Epidemiology of Diabetic Neuropathy
DSPN risk factors were detected in a recent review
of the topic [83]. Specifically, a higher prevalence
of pain has been reported in patients with more
severe DSPN and female sex has been shown to be
a risk factor for P-DSPN [6].
2.6 DSPN Complications
Besides the complicating neuropathic pain,
DSPN increases the risk of foot ulcers, Charcot
neuroarthropathy and amputations [1, 3]. The
lifetime risk for developing a diabetic foot ulcer
during the course of diabetes is 25%, with
approximately 5% requiring amputation [117].
These late complications are also predictors of
mortality [1, 118]. As early identification and
treatment of ulcers can delay or prevent adverse
outcomes, a comprehensive annual foot evalua-
tion is recommended, with more frequent evalua-
tions for patients with sensory loss or previous
ulceration or amputation [1, 8]. It is recom-
mended to educate patients in preventative foot
self-care and apply a multidisciplinary approach
to treating foot ulcers and high-risk feet [8].
DSPN also contributes to increased risk of
falls and fractures through loss of sensation, pro-
prioception, and temperature discrimination, as
well as pain and muscle weakness, all leading to
increased unsteadiness [1]. Additionally, DSPN
negatively affects quality of life and increases the
risk of depression and anxiety [1, 17]. Therefore,
evaluation of the risk of falls and the effect of
diabetic neuropathy on quality of life are recom-
mended [1]. Importantly, as microvascular com-
plications linked to diabetes tend to cluster,
screening, and handling of diabetic nephropathy
and retinopathy must be kept in mind [8, 119].
Lastly, besides the increased mortality associated
with foot ulcers and amputations, DSPN itself
may be a predictor of mortality [1, 118, 120].
3 Diabetic Autonomic
Neuropathy
Autonomic neuropathy is a neurological compli-
cation that covers a wide range of detrimental
changes to parasympathetic or sympathetic neu-
21
rons, or both, in the peripheral nervous system.
This affects various organ systems, eliciting man-
ifestations including orthostatic hypotension,
silent myocardial ischemia, resting tachycardia,
hypoglycaemia unawareness, obstipation, diar-
rhoea, faecal incontinence, urine retention, uri-
nary tract infections, sudomotor dysfunction, and
sexual dysfunction. The most prevalent form of
autonomic neuropathy is cardiovascular auto-
nomic neuropathy (CAN), which will be
addressed in detail below.
3.1 Cardiovascular Autonomic
Neuropathy
Cardiovascular autonomic neuropathy (CAN) is
defined as: “A damage to the autonomic nerve
fibres that innervate the heart and blood vessels,
resulting in abnormalities in heart rate control
and vascular dynamics” [121].
CAN is an overlooked serious diabetic compli-
cation associated with increased risk of morbidity
and mortality [121]. This may be due to the lack
of symptoms from early detrimental changes to
late and severe stages [122]. CAN symptoms are
vague and can manifest as resting tachycardia
[123] and dizziness or faintness when changing
position from lying or sitting to standing [124].
Additionally, the lack of testing capacity for CAN
may be a cause of missing epidemiological data.
Changes in cardiac autonomic activities are
present even in early pre-diabetes [125]. The
severity of CAN increases during lifetime expo-
sure to manifest diabetes [121]. These abnormali-
ties can be reflected in reduced heart rate
variability (HRV) and abnormal responses to car-
diovascular autonomic reflex tests (CARTs); all
signs of CAN [122].
Throughout the latest decades several diag-
nostic tools have been applied to assess the prev-
alence and incidence of CAN.  Adhering to
current international recommendations [1, 122]
we present a summary of studies where CARTs
have been used to diagnose CAN.  Acceptable
methods to assess CARTs include changes in
heart rate and blood pressure in response to active
tests. The recommended active tests include (1)
the measurement of heart rate change when
22
changing position from lying to standing, (2) dur-
ing deep breathing, (3) and the Valsalva manoeu-
vre, and (4) the assessment of blood pressure
change in response to standing (orthostatic hypo-
tension test). Several sets of normative values
have been presented for heart rate CARTs [122].
However, there is no consensus in terms of which
set of normative values to use to diagnose
CAN. This, in turn, affect the estimates of preva-
lence. Furthermore, the number of pathological
tests used to diagnose CAN does vary. Most
researchers categorise early CAN as the presence
of one pathological test and manifest CAN as
more than one pathological test. However, vari-
ous other scoring systems exist, hampering uni-
form estimates of CAN prevalence.
In addition to CARTs, we have included stud-
ies of significance where either a combination of
CARTs and resting HRV was assessed or where
HRV was the only outcome. Early changes in
measures of CAN are seen in indices of
HRV. Adverse changes in HRV have been found
in non-diabetic, pre-diabetic, and diabetic indi-
viduals [73, 126–131], underlining that changes
are not specific to diabetes, but can be seen even
before the disease is diagnosed. In many cases,
lack of normative reference values for HRV
makes comparisons to other studies difficult. It
has also been suggested that HRV measures are
prone to moderate reproducibility [132, 133].
In geographic regions where studies are
scarce, we have included studies where cold pres-
sure tests and handgrip tests have been utilised.
Neither test has been endorsed as a recommended
diagnostic tests for CAN. The handgrip test has a
poor association with CARTs [134]. It is also
worth mentioning the assessment of cardiovascu-
lar autonomic dysfunction by baroreflex sensitiv-
ity. This test describes the relationship between
resting blood pressure and heart rate as a measure
of autonomic function [122, 135]. However, this
method has not been applied in larger epidemio-
logical studies. Advanced cardiac imaging has
been utilised in few studies addressing cardiac
sympathetic function [136, 137], but these tests
are expensive and time consuming and have not
been applied to assess the prevalence of CAN on
a population-based scale.
C. S. Hansen et al.
3.2 Prevalence of CAN
The prevalence of diabetic CAN has not been
investigated as thoroughly as the prevalence of
DSPN. Early studies such as the US Framingham
Heart study performed in the 1980s estimated the
prevalence of CAN using analyses of 2-hour
ECG recordings. CAN measures were found to
be decreased in people with diabetes, however,
prevalence rates were not presented [130]. Later
studies addressing the prevalence of CAN using
HRV and CARTs have been performed. All
selected references addressing the prevalence of
CAN are summarised in Table 3.
3.2.1 Youth
Autonomic dysfunction has not been investigated
thoroughly in young people with diabetes and
only a few studies have applied CARTs to esti-
mate CAN prevalence in youth. It is evident,
however, that autonomic dysfunction is prevalent
in children and adolescent people with T1D and
T2D.  Autonomic dysfunction has been investi-
gated in cohorts of young T1D patients with a
heterogenic range of methods such as pupil size
determination, postural systolic blood pressure
response, hand grip test, and baroreflex [162].
Due to the different assessment modalities, CAN
prevalence was reported varying from 3% to
75%. These studies will not be addressed
further.
Larger studies have applied either HRV or
CARTs to investigate the prevalence of
CAN. Using CARTs and handgrip tests research-
ers found a prevalence of 8% in 221 Chinese chil-
dren with a mean age of 9 years. In the
SEARCH-study performed in the US, CAN was
defined as three or more out of five HRV indices
compared to healthy controls. The mean age of
participants was 18 years. The prevalence was
estimated to be 18% in this cohort of 1646 pri-
marily T1D children and adolescents [159]. A
recent hospital-based Danish study assessed
CAN by CARTs and showed a prevalence rate of
definitive CAN of 9% in a cohort of 156 ran-
domly recruited T1D patients with a mean age of
22 years [61]. Prevalence rates vary substantially
from study to study. The EURODIAB study,
The Epidemiology of Diabetic Neuropathy 23

Table 3  Prevalence of CAN in type 1 and type 2 diabetes

Prevalence (%) Diabetes
(H = Hospital, duration Mean
Diabetes Participants CAN P = primary (mean, age
type Study (N) measures care) year) (year) Risk factors/markers
Type 1 USA [138] 168 CARTs 37.5%(H) 20 29 Age, diabetes
(DB) duration, HbA1c,,
diastolic blood
pressure, LDL, HDL,
triglycerides,
hypertension,
retinopathy
Type 1 Europe [139] 1086 CARTs (LS) 32% (H) 1–7 15–29 age, diabetes
35% 8–14 30–44 duration, HbA1c,,
52% 45–60 45–60 diastolic blood
pressure, HDL,
albuminuria,
smoking, cholesterol/
HDL-cholesterol,
triglycerides
Type 1 Europe [140] NA CARTs NA NA NA NA
(LS, OH)
Type 1 Brazil [141] 151 CARTs (LS, 11.3% 16.3 33.4 Age, diabetes
DB, VM, (19.2%. early duration, HbA1c,
OH) CAN) (H) weight, hypertension,
creatinine, LDL,
eGFR, albuminuria
Type 1 USA [59, 620a R-R 5%a (H) 5.5a 25.5a NA
142] 570b variation, 10%b 11.9b 33.0b
594c valsalva 35%c 26.1 47.0c
ration, OH
Type 1 Denmark 156 CARTs (LS, Early CAN 11.3 22 NA
[61] DB, VM) 28.1% (P)
Definite CAN
9%
Type 1 Egypt [143] 60 CARTs (LS, Early CAN 7.6 15.1 HbA1c
DB, VM, 54.5 (H)
OH, HG, Definite 27.7
cold Severe 18.2
pressure) (H)
Type 1 Korea [144] 95 CARTs (LS, Early CAN 13.1 22.2 NA
DB, VM, 10.5(H)
OH) (H) Definite 2.1
Type 2 Germany 95 CARTs (LS, <1 (H) 0 49.7 NA
[145] DB, VM) (screen-­
(H) detected
dm)
Type 2 Germany HRV Kan ikke få
[146] fat I artikel
Type 2 India [147] 145 CARTs (LS, 44.2(H) 0 (screen 48 NA
DB, VM, detected
OH, HG, dm)
cold
pressure)
(continued)
24 C. S. Hansen et al.

Table 3 (continued)
Prevalence (%) Diabetes
(H = Hospital, duration Mean
Diabetes Participants CAN P = primary (mean, age
type Study (N) measures care) year) (year) Risk factors/markers
Type 2 Pakistan 207 HR, OH, 36.7 (H) NA 54.6 NA
[148] Handgrip,
etc.
Type 2 USA [149] 8135 HRV, QTI 3.5% (H) 10.8 62 HbA1c,, diastolic
blood pressure, BMI,
and triglycerides
Type 2 Italy [150] 557 CARTs (LS, 1.8%(H) <6 58.3 BMI, HbA1c,
DB, VM, (15.3%). months AH-drugs
OH) Early CAN)
(NA)
Type 2 Denmark 469 CARTs (LS, 13.4% (H) 10.0 59.0 Vitamin B12
[151] DB, VM)
Type 2 Korea [152] 3199 CARTs (LS, 23.7 (H) 5.5 59 Age, diabetes
DB, VM) duration,
(H) hypertension, eGFR,
retinopathy,
nephropathy, ApoA-1
Type 2 Denmark 443 CARTs (LS, 15.1 (P) (H) 12.8 70.3 Age, BMI, HbA1c,
[153] DB, VM) weight, diastolic
blood pressure,
Triglycerides
Type 2 Denmark 777 CARTs (LS, 9.0 (P) 6.3 65.0 Diabetes duration,
[153] DB, VM) BMI, HbA1c, weight,
waist circumference,
LDL, Triglycerides,
eGFR, albuminuria
Type 2 Korea [154] 1458 CARTs (LS, 55.7 (H) 8.2 58.1 NA
DB, VM)
(H)
Type 2 India [155] 100 CARTs (LS, Early CAN 25 9.0 53.3 Diabetes duration,
DB, VM, (H) creatinine, eGFR, age
OH) (H) Definite 24
Type 2 USA [156] 4629 HRV 9.7% (H) 4.3 59.7 Age, African-­
(10sek) American ethnicity,
(2of2) BMI, diastolic BP,
triglycerides,
Smoking
Type 1 Germany, T1dm: CARTs (LS, t1dm: 16.8% NA T1dm: NA
and 2 Switzerland, 647 DB, VM) (H) 11–69
Austria [73] T2dm: HRV (8.5% early T2dm:
524 CAN) 16–72
t2dm:
22.1%
(12.2% early
CAN) (H)
Type 1 France [157] T1dm: CARTs (LS, t1dm: 23.0% T1dm: T1dm: Age, Diabetes
and 2 245 DB, VM) t2dm:15.5% 10.0 36.1 duration, HbA1c,
T2dm: (H) T2dm: T2dm: BMI, systolic/
151 6.3 45.5 diastolic blood
pressure, albuminuria
The Epidemiology of Diabetic Neuropathy 25

Table 3 (continued)
Prevalence (%) Diabetes
(H = Hospital, duration Mean
Diabetes Participants CAN P = primary (mean, age
type Study (N) measures care) year) (year) Risk factors/markers
Type 1 USA [158] T1dm: 83 CARTs (LS, t1dm: 10.8% T1dm: T1dm: NA
and 2 T2dm: DB, VM, t2dm:16.2% 24.3 50.9
148 OH) (NA) T2dm: T2dm:
Symptoms 15.3 64.1
Type 1 USA [159] 1646 HRV (≥3 of 12%(t1dm) 8 18 HbA1c, BP,
and 2 5) (H) triglycerides (t1dm),
17%(t2dm) albuminuria
Type 1 Uganda [160] 299 CARTs Early CAN 5.8 50.1 Age, diabetes
and 2 (DB, VM, 29.4(H) duration, retinopathy
OH, Definite 20.4
HRHGa) Severe 2.3
Type 1 China [161] 73 (t1dm) CARTs (LS, 61.6(t1dm) NA NA Age, diabetes
and 2 1974 DB, VM, (H) duration
(t2dm) OH) (H) 62.6(t1dm)
LS heart response to lying to standing position, DB heart rate response to deep breathing, VM heart rate response to
Valsalva manoeuvre, OH orthostatic hypotension test, HG blood pressure response to hand grip, HR resting heart rate,
HRV heart rate variability, SM sudomotor function, HDL high density lipoprotein, LDL low density lipoprotein, eGFR
estimated glomerular filtration rate, BMI body mass index
a
DCCT baseline (control group)
b
DCCT close-out (control group)
c
EDIC 13–14 year follow-up (control group)

where CAN was assessed only by heart rate
response to deep breathing, found a prevalence of
32% in T1D patients in the youngest age group
(on average 22 years of age) [139]. This preva-
lence could be interpreted as a marker of early
CAN and is comparable with the prevalence of
early CAN in the Danish study where a preva-
lence of 28.1% was reported.
Thus, CAN is a significant complication in
children and adolescents. Prevalence rates in
children vary from 5% to 8% and in younger
adults from about 9% to 18% depending on the
methods applied. The prevalence of early CAN
may be about 30%.
3.2.2 Type 1 Diabetes
The majority of CAN prevalence studies have
been performed in T2D patients. However, larger
studies in T1D cohorts or mixed T1D and T2D
patients have also been conducted.
Only few studies of significant size have
been performed prior to 2000. The European
multicentre study DIACAN was conducted in
the early 1990s in several hospital-based diabe-
tes centres. Here patients aged 11–69 years had
a prevalence of 16.8% based on CARTs and
HRV measures.
Later hospital-based studies have reported a
similar prevalence of CAN based on CARTs in
T1D. In Brazil, the prevalence of CAN is 11.3%
in patients with a mean age of 33.4 years [141].
In the DCCT/EDIC study, participants at DCCT
close-out aged approximately 33 years had a
CAN prevalence of 10% when assessed by HRV,
Valsalva heart rate test, and orthostatic hypoten-
sion [59, 142]. In the EURODIAB study a much
higher prevalence of CAN was reported in simi-
lar age groups. Here a prevalence of 35% was
seen in patients in the 30–44-year age range.
CAN was assessed only by heart rate change to
deep breathing, possibly constituting both early
and definite CAN.  In 2018 definite CAN was
reported tin 27.1% of patients in a smaller
Egyptian cohort, where assessments were per-
formed using CARTs, hand grip, and cold pres-
sures test [143]. Similarly, a high prevalence of
23% was found in a French multicentre study of
245 patients with a mean age of 36 years. In 2019
an unprecedented prevalence of 61.2% was
observed in T1D [161] patients in a hospital-­
26
based Chinese cohort. However, CAN was
assessed as a composite of an array of symptoms
and objective measures, which may contribute to
the high estimate.
In conclusion only few significant studies are
available on CAN prevalence in adults with T1D
using recommended CARTs. The prevalence var-
ies from approximately 10%–27%. Regional dif-
ferences may apply, but heterogeneous
assessment methods hamper robust
comparisons.
3.2.3 Type 2 diabetes
A few noteworthy studies of European origin
were performed in the 1990s. In these studies
CAN was assessed by CARTs with [163] or with-
out [145] a handgrip test. CAN was present in
<1% and 8.1% of patients, respectively. The first
study was an investigation of newly diagnosed
patients. With increasing age, a higher prevalence
has been reported by others, where estimates of
definite CAN reached 22.1% [73].
Since 2000, definite CAN has been explored
in several large cohorts. Studies investigating
patients with short diabetes duration indicate a
low prevalence of CAN in early T2D. Using rec-
ommended CARTs and the orthostatic hypoten-
sion test, Italian researchers reported a prevalence
of definite CAN of 1.8% and early CAN of 15%
in patients with a diabetes duration of less than
six months [150]. Hospital-based Indian data of
screen-detected T2D indicate regional differ-
ences. Defining CAN as one or more pathologi-
cal test of six tests, 44.2% were diagnosed with
CAN [147]. Here, no discrimination was applied
between early and definite CAN.
With longer diabetes duration higher CAN
prevalence has been reported. Diabetes duration
of approximately 5 years have been associated
with definite CAN assessed by CARTs (≥2 path-
ological CARTS), with a prevalence between 9%
and 24% (Table 3). Geographical differences are
apparent. Findings range from 9% in the
ADDITION-Denmark study [153] and 15.5% in
a French study [153] to approximately 25% in
studies from Uganda [160] and Korea [152]. In
the GRADE study a relative low CAN prevalence
C. S. Hansen et al.
was found; when using HRV to diagnose CAN,
only 9.7% had CAN [156].
Few studies have examined CAN in patients
with mean diabetes durations of approximately
10–15 years. Danish studies indicate a CAN
prevalence (≥2 pathological CARTS) of about
15%. The ADDITION-Denmark 13-year follow-
­up study showed that 15.1% had definitive CAN
[153] and another study of randomly recruited
patients showed a prevalence of 13.4% [151].
American data seems to mirror findings from
European studies. Using a complex scoring sys-
tem of CARTs and symptoms, CAN prevalence
has been reported to be approximately 16% in
patients with a diabetes duration of 15.3 years
[158]. Only 3.5% of participants in the ACCORD
study had CAN, however, here CAN was assessed
by HRV and QT-interval of ECG recordings
[149]. Higher prevalence has been reported in
Asian studies. Indian data suggests that more
patients have CAN when recommended CARTs
are utilised. In a smaller cohort of 100 patients
24% had definite CAN [155]. When diagnosing
CAN as one or more pathological CARTs, a
CAN prevalence of 55.7% was found in a large
(N = 1458) Korean hospital-based study [154]. In
2019 an unprecedented prevalence of 62.6 % was
observed in T2D patients in a hospital-based
Chinese cohort [161]. However, CAN was
assessed using a composite array of symptoms
and objective measures, which may contribute to
the high estimate. A Pakistani study of 200 T2D
patients showed a prevalence of 36.7% (mean
age 54.6 years), but diabetes duration was not
reported and CAN was assessed by heart rate,
QTc-interval, Handgrip, and orthostatic hypoten-
sion tests [148], making the study difficult to
compare with others.
In conclusion, CAN prevalence in T2D varies
by geography, with lower rates in western coun-
tries. The prevalence in early T2D rages from
<1% to a few percent increasing to 15–22% in
western cohorts and possibly even higher (24–
37%) in non-western populations at later stages
of the disease. No apparent changes in prevalence
have been observed from the 1990s to the
present.
The Epidemiology of Diabetic Neuropathy 27

3.3 Incidence of CAN In the recent ADDITION-Denmark study,

which included patients with screen-detected dia-
A few larger longitudinal study cohorts have betes in primary care, low progression rates of
been established. These studies are either obser- CAN were reported. Participants were screened
vational cohorts or randomised controlled trials. from 2001 and 2006  in primary care. The preva-
For the latter study design valuable information lence of confirmed CAN (≥2 pathological CARTS)
on the natural history of CAN can be derived increased from 9.0% to 15.1% with an annual inci-
from the control groups. dence of 1.8% between the 6- and 13-year follow-
In the observational multi-national up of the cohort [153]. The ADDITION-Denmark
EURODIAB study, which began in 1989, 956 study also evaluated the flow of CAN stages in par-
T1D patients were examined. CAN was diag- ticipants re-examined for CAN (Fig. 1). There was
nosed by either having impaired heart rate or a progressive change in CAN status from 6- to
blood pressures response to standing. In total, 13-year follow-up, although the phenotype of CAN
17% of participants developed CAN in the 7.3-­ was heterogeneous and improvements were seen in
year follow-up period, a mean annual increase of all categories of CAN, suggesting CAN to have an
approximately 2.3%. This was equivalent to an element of reversibility.
incidence rate of 23.4 per 1000 person-years of In summary, only a very few longitudinal stud-
follow-up [140]. In the DCCT/EDIC study a ran- ies with a significant sample-size have been per-
domised placebo-controlled trial with interven- formed to accurately assess CAN incidence.
tion free follow-up, 5% of participants in the Annual progression rates seem similar in both T1D
control group had CAN at baseline (defined by and T2D patients at approximately 2%. No studies
having either abnormal HRV, heart rate response have been performed on population-based cohorts
to Valsalva or orthostatic hypotension). The study that allow for assessment of incidences rate ratios
enrolled patients from 1983 to 1989. In the for- for CAN. Therefore, as for DSPN it is not possible
mer control group 35% of participants had CAN to document a reduction in the incidence of CAN
after 14 years of EDIC follow-up [59, 142]. This over time that can mirror the improvements of pre-
approximates an annual incidence of 2.1%. venting other diabetic complications.

Fig. 1  The flow of participants between clinical groups ries; No CAN (green), early CAN (yellow) or manifest
of CAN: ADDITION-Denmark. Number of participants CAN (red) from the 6- to the 13-year follow-up [153]
and changes in their CAN status among defined catego- (used with permission from Diabetes Care)
28
3.4 Risk Factors for CAN
Several risk factors are associated with diabetic
CAN in cross-sectional investigations, e.g. diabe-
tes duration, higher HbA1c, HDL, triglycerides,
higher blood pressure and BMI.  Nephropathy,
low vitamin B12, DPSN, and retinopathy have
also been found to be associated with CAN
(Table 3). Few studies have addressed risk factors
for progression to CAN. The EURODIAB study
showed that elevated HbA1c, hypertension,
DSPN, and retinopathy predict the risk of devel-
oping CAN [140]. The ADDITION-Denmark
study did not reveal any statistically significant
risk factors for incident CAN.  However, inter-
vention studies have demonstrated that multifac-
torial intervention [164] and intensive
glucose-lowering treatment [21, 165, 166] can
reduce progression to manifest CAN.
3.5 Other Forms of Autonomic
Neuropathy
As mentioned above, other organ systems can be
affected by autonomic neuropathy.
Gastrointestinal autonomic neuropathy is less
common and can be subdivided into gastropare-
ses affecting gastric emptying and diabetic enter-
opathy. Epidemiological data on these
complications are scarce and mainly based on
symptoms. It is estimated that 1–50% [167–169]
of patients with diabetes have symptoms of gas-
troparesis, a complication causing dysmotility of
the stomach. Gastroparesis is associated with
poor glycaemic control, higher mean levels of
HbA1c, and risk of severe hypoglycaemia.
Furthermore, diabetic gastroparesis is associated
with poor drug absorption, malnutrition, and
reduced quality of life due to gastrointestinal
symptoms [170]. Diabetic enteropathy, where
nerves innervation the intestines are affected, is
less well studied. No studies have assessed the
prevalence of diabetic enteropathy using objec-
tive measures.
Autonomic neuropathy may manifest as sex-
ual dysfunction and bladder dysfunction. Sexual
dysfunction in men is typically observed as erec-
C. S. Hansen et al.
tile dysfunction (ED) and is more prevalent in
diabetes patients compared to men without dia-
betes. This may be associated with autonomic
dysfunction. However, a common cause of ED is
atherosclerosis. So far, no studies have published
data to discriminate between these two causes of
ED making it impossible to investigate the preva-
lence of autonomic ED. Similarly, female sexual
dysfunction is more prevalent in people with dia-
betes. Again, no studies have been able to assess
the prevalence of female sexual dysfunction
attributed to autonomic neuropathy.
Urinary incontinence and bladder dysfunction
are linked to the presence of diabetic neuropathy
in both men and women [171]. However, the
prevalence of bladder dysfunction due to auto-
nomic dysfunction has not been investigated.
Autonomic dysfunction of the skin is seen as
both decreased (hypohidrosis) and increased
(hyperhidrosis) sweat production. The preva-
lence of these dysfunctions are unknown.
4 Other Forms of Diabetic
Neuropathy
Diabetic mononeuropathies and radiculopathies
(or polyradiculopathies) with focal presentations
in the innervation area of the affected nerve or
nerve root are atypical forms of diabetic neuropa-
thy and differ from DSPN in onset, course, and
manifestations [1, 3, 10].
Mononeuropathies due to compression occur
more commonly in people with diabetes and
most often involve the median, ulnar, radial, and
common peroneal nerves [10, 172]. However, the
explanation for the predisposition of nerve injury
in diabetic mononeuropathies is considered mul-
tifactorial. Mononeuropathies in people with dia-
betes may not be exclusively due to entrapment
[172]. This is exemplified by studies showing
that over 20% of patients investigated for median
mononeuropathy fulfilled NCS criteria for carpal
tunnel syndrome, even though they were asymp-
tomptomatic [51, 172, 173]. Cranial neuropathies
are more rare, present acutely, and can affect up
to 1% of people with diabetes, primarily involv-
ing the 3rd, 4th, 6th, and 7th cranial nerve [1, 10].
The Epidemiology of Diabetic Neuropathy
Diabetic radiculoplexus neuropathy (DRPN)
or polyradiculoneuropathy presents with cervi-
cal, thoracic, and/or lumbosacral involvement.
Lumbosacral DRPN (diabetic amyotrophy) is the
most common condition and is estimated to have
a lifelong incidence of 1% in people with diabe-
tes [172, 174]. Affected people are typically
middle-­aged or elderly and have T2D. Glycaemic
control, diabetes duration, treatment or severity
of DSPN are unrelated to the risk of DRPN [10,
174]. The presentation is usually abrupt with
severe unilateral thigh pain followed by progres-
sive atrophy, weakness and weight loss. The dis-
order is usually self-limiting and is considered
immune-mediated [10, 172].
Lastly, treatment induced neuropathy (TIN) is
worth mentioning. TIN is considered a rare, per-
haps under-recognised, iatrogenic small fibre
neuropathy, that appears after abrupt lowering of
hyperglycaemia leading to a painful sensory neu-
ropathy often with autonomic manifestations [10,
175]. The disorder is most often seen in T1D and
has a prevalence of up to 10% in patients with
diabetes referred for evaluation of diabetic neu-
ropathy to a tertiary referral centre [175].
References
1. Pop-Busui R, et al. Diabetic neuropathy: a position
statement by the American Diabetes Association.
Diabetes Care. 2017;40:136–54. https://doi.
org/10.2337/dc16-­2042.
2. Gregg EW, et al. Changes in diabetes-related com-
plications in the United States, 1990-2010. N Engl
J Med. 2014;370:1514–23. https://doi.org/10.1056/
NEJMoa1310799.
3. Tesfaye S, et  al. Diabetic neuropathies: update on
definitions, diagnostic criteria, estimation of ­severity,
and treatments. Diabetes Care. 2010;33:2285–93.
https://doi.org/10.2337/dc10-­1303.
4. Callaghan BC, Price RS, Feldman EL.  Distal
symmetric polyneuropathy: a review. JAMA.
2015;314:2172–81. https://doi.org/10.1001/
jama.2015.13611.
5. Feldman EL, Nave KA, Jensen TS, Bennett
DLH.  New horizons in diabetic neuropathy:
mechanisms, bioenergetics, and pain. Neuron.
2017;93:1296–313. https://doi.org/10.1016/j.
neuron.2017.02.005.
6. Feldman EL, et  al. Diabetic neuropathy. Nat
Rev. 2019;5:41. https://doi.org/10.1038/
s41572-­019-­0092-­1.
29
7. Van Acker K, et al. Prevalence and impact on quality
of life of peripheral neuropathy with or without neu-
ropathic pain in type 1 and type 2 diabetic patients
attending hospital outpatients clinics. Diabetes
Metab. 2009;35:206–13. https://doi.org/10.1016/j.
diabet.2008.11.004.
8. American Diabetes Association. Microvascular com-
plications and foot care: standards of medical care in
diabetes-2021. Diabetes Care. 2021;44:S151–s167.
https://doi.org/10.2337/dc21-­S011.
9. Dyck PJ, et al. Diabetic polyneuropathies: update on
research definition, diagnostic criteria and estima-
tion of severity. Diabetes. 2011;27:620–8. https://
doi.org/10.1002/dmrr.1226.
10. Smith AG, Singleton JR.  Diabetic neuropa-
thy. Continuum. 2012;18:60–84. https://doi.
org/10.1212/01.CON.0000411568.34085.3e.
11. Singleton JR, Smith AG, Bromberg MB. Increased
prevalence of impaired glucose tolerance in
patients with painful sensory neuropathy. Diabetes
Care. 2001;24:1448–53. https://doi.org/10.2337/
diacare.24.8.1448.
12. Asghar O, et al. Corneal confocal microscopy detects
neuropathy in subjects with impaired glucose toler-
ance. Diabetes Care. 2014;37(9):2643–6. https://doi.
org/10.2337/dc14-­0279.
13. Bongaerts BWC, et al. Older subjects with diabetes
and prediabetes are frequently unaware of having
distal sensorimotor polyneuropathy: The KORA F4
study. Diabetes Care. 2013;36:1141–6. https://doi.
org/10.2337/dc12-­0744.
14. Im S, Kim SR, Park JH, Kim YS, Park GY. Assessment
of the medial dorsal cutaneous, dorsal sural, and
medial plantar nerves in impaired glucose toler-
ance and diabetic patients with normal sural and
superficial peroneal nerve responses. Diabetes Care.
2012;35:834–9. https://doi.org/10.2337/dc11-­1001.
15. Ziegler D, Rathmann W, Dickhaus T, Meisinger
C, Mielck A. Prevalence of polyneuropathy in pre-­
diabetes and diabetes is associated with abdomi-
nal obesity and macroangiopathy: the MONICA/
KORA Augsburg Surveys S2 and S3. Diabetes
Care. 2008;31:464–9. https://doi.org/10.2337/
dc07-­1796.
16. Callaghan BC, et  al. Metabolic syndrome compo-
nents are associated with symptomatic polyneuropa-
thy independent of glycemic status. Diabetes Care.
2016;39:801–7. https://doi.org/10.2337/dc16-­0081.
17. Ziegler D, Papanas N, Vinik AI, Shaw
JE.  Epidemiology of polyneuropathy in dia-
betes and prediabetes. Handb Clin Neurol.
2014;126:3–22. https://doi.org/10.1016/
b978-­0-­444-­53480-­4.00001-­1.
18. Malik RA, et al. Small fibre neuropathy: role in the
diagnosis of diabetic sensorimotor polyneuropathy.
Diabetes. 2011;27:678–84. https://doi.org/10.1002/
dmrr.1222.
19. Breiner A, Lovblom LE, Perkins BA, Bril V.  Does
the prevailing hypothesis that small-fiber dysfunc-
tion precedes large-fiber dysfunction apply to type 1
30
diabetic patients? Diabetes Care. 2014;37:1418–24.
https://doi.org/10.2337/dc13-­2005.
20. England JD, et  al. Distal symmetric polyneuropa-
thy: a definition for clinical research: report of the
American Academy of Neurology, the American
Association of Electrodiagnostic Medicine, and
the American Academy of Physical Medicine and
Rehabilitation. Neurology. 2005;64:199–207.
https://doi.org/10.1212/01.Wnl.0000149522.32823.
Ea.
21. The Diabetes Control and Complications Trial
Research Group. The effect of intensive diabetes
therapy on the development and progression of neu-
ropathy. Ann Intern Med. 1995;122:561–8.
22. Hanewinckel R, et  al. Prevalence of polyneuropa-
thy in the general middle-aged and elderly popu-
lation. Neurology. 2016;87:1892–8. https://doi.
org/10.1212/wnl.0000000000003293.
23. Hanewinckel R, van Oijen M, Ikram MA, van Doorn
PA.  The epidemiology and risk factors of chronic
polyneuropathy. Eur J Epidemiol. 2016;31:5–20.
https://doi.org/10.1007/s10654-­015-­0094-­6.
24. Maser RE, et  al. Epidemiological correlates of
diabetic neuropathy. Report from Pittsburgh
Epidemiology of Diabetes Complications Study.
Diabetes. 1989;38:1456–61.
25. Veglio M, Sivieri R.  Prevalence of neuropa-
thy in IDDM patients in Piemonte, Italy. The
Neuropathy Study Group of the Italian Society for
the Study of Diabetes, Piemonte Affiliate. Diabetes
Care. 1993;16:456–61. https://doi.org/10.2337/
diacare.16.2.456.
26. Olsen BS, et  al. A 6-year nationwide cohort
study of glycaemic control in young people with
type 1 diabetes. Risk markers for the develop-
ment of retinopathy, nephropathy and neuropathy.
Danish Study Group of Diabetes in Childhood. J
Diabetes Complicat. 2000;14:295–300. https://doi.
org/10.1016/s1056-­8727(00)00078-­7.
27. Mizokami-Stout KR, et  al. The contemporary
prevalence of diabetic neuropathy in type 1 dia-
betes: findings from the T1D exchange. Diabetes
Care. 2020;43:806–12. https://doi.org/10.2337/
dc19-­1583.
28. Jeyam A, et  al. Diabetic neuropathy is a substan-
tial burden in people with type 1 diabetes and is
strongly associated with socioeconomic disad-
vantage: a population-­ representative study from
Scotland. Diabetes Care. 2020;43:734–42. https://
doi.org/10.2337/dc19-­1582.
29. Fujimoto WY, et  al. Prevalence of complications
among second-generation Japanese-American men
with diabetes, impaired glucose tolerance, or normal
glucose tolerance. Diabetes. 1987;36:730–9. https://
doi.org/10.2337/diab.36.6.730.
30. Lehtinen JM, Uusitupa M, Siitonen O, Pyörälä
K.  Prevalence of neuropathy in newly diag-
nosed NIDDM and nondiabetic control subjects.
Diabetes. 1989;38:1307–13. https://doi.org/10.2337/
diab.38.10.1307.
C. S. Hansen et al.
31. Partanen J, et al. Natural history of peripheral neu-
ropathy in patients with non-insulin-dependent
diabetes mellitus. N Engl J Med. 1995;333:89–94.
https://doi.org/10.1056/nejm199507133330203.
32. Franklin GM, Kahn LB, Baxter J, Marshall JA,
Hamman RF.  Sensory neuropathy in non-insulin-­
dependent diabetes mellitus. The San Luis Valley
Diabetes Study. Am J Epidemiol. 1990;131:633–
43. https://doi.org/10.1093/oxfordjournals.aje.
a115547.
33. Herman WH, et  al. Diabetes mellitus in
Egypt: glycaemic control and microvascular
and neuropathic complications. Diabet Med.
1998;15:1045–51. https://doi.org/10.1002/
(sici)1096-­9 136(1998120)15:12<1045::Aid-­
dia696>3.0.Co;2-­l.
34. Shaw JE, et  al. Diabetic neuropathy in Mauritius:
prevalence and risk factors. Diabetes Res Clin
Pract. 1998;42:131–9. https://doi.org/10.1016/
s0168-­8227(98)00100-­4.
35. Verhoeven S, van Ballegooie E, Casparie AF. Impact
of late complications in type 2 diabetes in a Dutch
population. Diabet Med. 1991;8:435–8. https://doi.
org/10.1111/j.1464-­5491.1991.tb01627.x.
36. Beghi E, Monticelli ML. Diabetic polyneuropathy in
the elderly. Prevalence and risk factors in two geo-
graphic areas of Italy. Italian General Practitioner
Study Group (IGPSG). Acta Neurol Scand.
1997;96:223–8.
37. Kumar S, et al. The prevalence of foot ulceration and
its correlates in type 2 diabetic patients: a population-­
based study. Diabet Med. 1994;11:480–4. https://
doi.org/10.1111/j.1464-­5491.1994.tb00310.x.
38. Barbosa AP, Medina JL, Ramos EP, Barros
HP.  Prevalence and risk factors of clinical diabetic
polyneuropathy in a Portuguese primary health care
population. Diabetes Metab. 2001;27:496–502.
39. Mimi O, Teng CL, Chia YC. The prevalence of dia-
betic peripheral neuropathy in an outpatient setting.
Med J Malaysia. 2003;58:533–8.
40. Al-Mahroos F, Al-Roomi K.  Diabetic neuropathy,
foot ulceration, peripheral vascular disease and
potential risk factors among patients with diabe-
tes in Bahrain: a nationwide primary care diabetes
clinic-based study. Ann Saudi Med. 2007;27:25–31.
https://doi.org/10.5144/0256-­4947.2007.25.
41. Pradeepa R, et al. Prevalence and risk factors for dia-
betic neuropathy in an urban south Indian popula-
tion: the Chennai Urban Rural Epidemiology Study
(CURES-55). Diabet Med. 2008;25:407–12. https://
doi.org/10.1111/j.1464-­5491.2008.02397.x.
42. Ziegler D, Rathmann W, Dickhaus T, Meisinger
C, Mielck A. Prevalence of polyneuropathy in pre-
diabetes and diabetes is associated with abdomi-
nal obesity and macroangiopathy: The Monica/
Kora Augsburg surveys S2 and S3. Diabetes Care.
2008;31:464–9. https://doi.org/10.2337/dc07-­1796.
43. Kärvestedt L, et  al. Peripheral sensory neuropathy
associates with micro- or macroangiopathy: results
from a population-based study of type 2 diabetic
The Epidemiology of Diabetic Neuropathy
patients in Sweden. Diabetes Care. 2009;32:317–22.
https://doi.org/10.2337/dc08-­1250.
44. Hsu WC, Yen AM, Liou HH, Wang HC, Chen
TH.  Prevalence and risk factors of somatic and
autonomic neuropathy in prediabetic and diabetic
patients. Neuroepidemiology. 2009;33:344–9.
https://doi.org/10.1159/000254571.
45. UK Prospective Diabetes Study (UKPDS) Group.
Intensive blood-glucose control with sulphonylureas
or insulin compared with conventional treatment and
risk of complications in patients with type 2 diabetes
(UKPDS 33). Lancet. 1998;352:837–53.
46. Andersen ST, et al. Risk factors for incident diabetic
polyneuropathy in a cohort with screen-detected
type 2 diabetes followed for 13 years: ADDITION-­
Denmark. Diabetes Care. 2018;41:1068–75. https://
doi.org/10.2337/dc17-­2062.
47. Charles M, et  al. Prevalence of neuropathy and
peripheral arterial disease and the impact of treat-
ment in people with screen-detected type 2 dia-
betes: The ADDITION-Denmark study. Diabetes
Care. 2011;34(10):2244–9. https://doi.org/10.2337/
dc11-­0903.
48. Andersen ST, et al. Risk-factor trajectories preced-
ing diabetic polyneuropathy: ADDITION-Denmark.
Diabetes Care. 2018;41:1955–62. https://doi.
org/10.2337/dc18-­0392.
49. Reis de Matos M, et  al. Distal symmetric and car-
diovascular autonomic neuropathies in Brazilian
individuals with type 2 diabetes followed in a pri-
mary health care unit: a cross-sectional study. Int J
Environ Res Public Health. 2020;17:3232. https://
doi.org/10.3390/ijerph17093232.
50. Knuiman MW, Welborn TA, McCann VJ, Stanton
KG, Constable IJ.  Prevalence of diabetic com-
plications in relation to risk factors. Diabetes.
1986;35:1332–9. https://doi.org/10.2337/
diab.35.12.1332.
51. Dyck PJ, et al. The prevalence by staged severity of
various types of diabetic neuropathy, retinopathy,
and nephropathy in a population-based cohort: the
Rochester Diabetic Neuropathy Study. Neurology.
1993;43:817–24.
52. Dyck PJ, et  al. Human diabetic endoneurial sorbi-
tol, fructose, and myo-inositol related to sural nerve
morphometry. Ann Neurol. 1980;8:590–6. https://
doi.org/10.1002/ana.410080608.
53. Harris M, Eastman R, Cowie C. Symptoms of sen-
sory neuropathy in adults with NIDDM in the U.S.
population. Diabetes Care. 1993;16:1446–52.
54. Walters DP, Gatling W, Mullee MA, Hill RD.  The
prevalence of diabetic distal sensory neuropathy in
an English community. Diabet Med. 1992;9:349–
53. https://doi.org/10.1111/j.1464-­5491.1992.
tb01795.x.
55. Jaiswal M, et al. Prevalence of and risk factors for
diabetic peripheral neuropathy in youth with type 1
and type 2 diabetes: SEARCH for diabetes in youth
study. Diabetes Care. 2017;40:1226–32. https://doi.
org/10.2337/dc17-­0179.
31
56. Tesfaye S, et  al. Prevalence of diabetic periph-
eral neuropathy and its relation to glycaemic con-
trol and potential risk factors: the EURODIAB
IDDM Complications Study. Diabetologia.
1996;39:1377–84.
57. Tesfaye S, et  al. Vascular risk factors and diabetic
neuropathy. N Engl J Med. 2005;352:341–50.
https://doi.org/10.1056/NEJMoa032782.
58. Albers JW, et  al. Effect of prior intensive insu-
lin treatment during the Diabetes Control and
Complications Trial (DCCT) on peripheral neuropa-
thy in type 1 diabetes during the Epidemiology of
Diabetes Interventions and Complications (EDIC)
study. Diabetes Care. 2010;33:1090–6. https://doi.
org/10.2337/dc09-­1941.
59. Martin CL, Albers JW, Pop-Busui R.  Neuropathy
and related findings in the diabetes control and
complications trial/epidemiology of diabetes inter-
ventions and complications study. Diabetes Care.
2014;37:31–8. https://doi.org/10.2337/dc13-­2114.
60. Braffett BH, et al. Risk factors for diabetic periph-
eral neuropathy and cardiovascular autonomic
neuropathy in the diabetes control and compli-
cations trial/epidemiology of diabetes interven-
tions and complications (DCCT/EDIC) study.
Diabetes. 2020;69:1000–10. https://doi.org/10.2337/
db19-­1046.
61. Christensen MMB, et al. Prevalence of diabetic neu-
ropathy in young adults with type 1 diabetes and the
Association with Insulin Pump Therapy. Diabetes
Technol Therap. 2018. https://doi.org/10.1089/
dia.2018.0249.
62. Adler AI, et al. Risk factors for diabetic peripheral
sensory neuropathy. Results of the Seattle Prospective
Diabetic Foot Study. Diabetes Care. 1997;20:1162–
7. https://doi.org/10.2337/diacare.20.7.1162.
63. Wikblad K, Smide B, Bergström A, Kessi J, Mugusi
F.  Outcome of clinical foot examination in rela-
tion to self-perceived health and glycaemic control
in a group of urban Tanzanian diabetic patients.
Diabetes Res Clin Pract. 1997;37:185–92. https://
doi.org/10.1016/s0168-­8227(97)00072-­7.
64. Nielsen JV.  Peripheral neuropathy, hypertension,
foot ulcers and amputations among Saudi Arabian
patients with type 2 diabetes. Diabetes Res Clin
Pract. 1998;41:63–9. https://doi.org/10.1016/
s0168-­8227(98)00059-­x.
65. de Neeling JN, Beks PJ, Bertelsmann FW, Heine
RJ, Bouter LM.  Peripheral somatic nerve func-
tion in relation to glucose tolerance in an elderly
Caucasian population: the Hoorn study. Diabet Med.
1996;13:960–6. https://doi.org/10.1002/(sici)1096-­
9136(199611)13:11<960::Aid-­dia268>3.0.Co;2-­z.
66. Delcourt C, Vauzelle-Kervroedan F, Cathelineau G,
Papoz L.  Low prevalence of long-term complica-
tions in non-insulin-dependent diabetes mellitus in
France: a multicenter study. CODIAB-INSERM-­
ZENECA Pharma Study Group. J Diabetes
Complicat. 1998;12:88–95. https://doi.org/10.1016/
s1056-­8727(97)98005-­3.
32
67. Janghorbani M, et  al. Peripheral neuropa-
thy in type 2 diabetes mellitus in Isfahan,
Iran: prevalence and risk factors. Acta
Neurol Scand. 2006;114:384–91. https://doi.
org/10.1111/j.1600-­0404.2006.00716.x.
68. Liu F, et al. Screening and prevalence of peripheral
neuropathy in type 2 diabetic outpatients: a random-
ized multicentre survey in 12 city hospitals of China.
Diabetes Metab Res Rev. 2010;26:481–9. https://
doi.org/10.1002/dmrr.1107.
69. Gaede P, et  al. Multifactorial intervention and car-
diovascular disease in patients with type 2 diabe-
tes. N Engl J Med. 2003;348:383–93. https://doi.
org/10.1056/NEJMoa021778.
70. Ismail-Beigi F, et  al. Effect of intensive treatment
of hyperglycaemia on microvascular outcomes in
type 2 diabetes: an analysis of the ACCORD ran-
domised trial. Lancet. 2010;376:419–30. https://doi.
org/10.1016/S0140-­6736(10)60576-­4.
71. Su JB, et  al. HbA1c variability and diabetic
peripheral neuropathy in type 2 diabetic patients.
Cardiovasc Diabetol. 2018;17:47. https://doi.
org/10.1186/s12933-­018-­0693-­0.
72. Fedele D, et  al. A multicenter study on the preva-
lence of diabetic neuropathy in Italy. Diabetes
Care. 1997;20:836–43. https://doi.org/10.2337/
diacare.20.5.836.
73. Ziegler D, et  al. Prevalence and clinical correlates
of cardiovascular autonomic and peripheral dia-
betic neuropathy in patients attending diabetes cen-
ters. The Diacan Multicenter Study Group. Diabete
Metab. 1993;19:143–51.
74. Young MJ, Boulton AJM, Macleod AF, Williams
DRR, Sonksen PH.  A multicentre study of the
prevalence of diabetic peripheral neuropathy
in the United Kingdom hospital clinic popula-
tion. Diabetologia. 1993;36:150–4. https://doi.
org/10.1007/BF00400697.
75. Kiani J, Moghimbeigi A, Azizkhani H, Kosarifard
S.  The prevalence and associated risk factors of
peripheral diabetic neuropathy in Hamedan, Iran.
Arch Iran Med. 2013;16:17–9.
76. Cabezas-Cerrato J.  The prevalence of clinical dia-
betic polyneuropathy in Spain: a study in primary
care and hospital clinic groups. Neuropathy Spanish
Study Group of the Spanish Diabetes Society
(SDS). Diabetologia. 1998;41:1263–9. https://doi.
org/10.1007/s001250051063.
77. Pai YW, Lin CH, Lee IT, Chang MH.  Prevalence
and biochemical risk factors of diabetic periph-
eral neuropathy with or without neuropathic pain
in Taiwanese adults with type 2 diabetes mellitus.
Diabetes Metab Syndr. 2018;12:111–6. https://doi.
org/10.1016/j.dsx.2017.09.013.
78. Kim SS, et al. Prevalence and clinical implications of
painful diabetic peripheral neuropathy in type 2 dia-
betes: results from a nationwide hospital-based study
of diabetic neuropathy in Korea. Diabetes Res Clin
Pract. 2014;103:522–9. https://doi.org/10.1016/j.
diabres.2013.12.003.
C. S. Hansen et al.
79. Jensen TS, et al. A new definition of neuropathic pain.
Pain. 2011;152:2204–5. https://doi.org/10.1016/j.
pain.2011.06.017.
80. Finnerup NB, et  al. Neuropathic pain: an updated
grading system for research and clinical practice.
Pain. 2016;157:1599–606. https://doi.org/10.1097/j.
pain.0000000000000492.
81. Galer BS, Gianas A, Jensen MP.  Painful diabetic
polyneuropathy: epidemiology, pain descrip-
tion, and quality of life. Diabetes Res Clin Pract.
2000;47:123–8.
82. Davies M, Brophy S, Williams R, Taylor A.  The
prevalence, severity, and impact of painful diabetic
peripheral neuropathy in type 2 diabetes. Diabetes
Care. 2006;29:1518–22. https://doi.org/10.2337/
dc05-­2228.
83. Hébert HL, Veluchamy A, Torrance N, Smith
BH.  Risk factors for neuropathic pain in diabe-
tes mellitus. Pain. 2017;158:560–8. https://doi.
org/10.1097/j.pain.0000000000000785.
84. Ponirakis G, et  al. Prevalence and risk factors for
painful diabetic neuropathy in secondary healthcare
in Qatar. J Diabetes Investig. 2019;10:1558–64.
https://doi.org/10.1111/jdi.13037.
85. Ziegler D, et al. Painful and painless neuropathies are
distinct and largely undiagnosed entities in subjects
participating in an educational initiative (PROTECT
study). Diabetes Res Clin Pract. 2018;139:147–54.
https://doi.org/10.1016/j.diabres.2018.02.043.
86. Ziegler D, Rathmann W, Dickhaus T, Meisinger
C, Mielck A.  Neuropathic pain in diabetes, pre-
diabetes and normal glucose tolerance: the
MONICA/KORA Augsburg surveys S2 and
S3. Pain Med. 2009;10:393–400. https://doi.
org/10.1111/j.1526-­4637.2008.00555.x.
87. Ziegler D, Rathmann W, Meisinger C, Dickhaus T,
Mielck A. Prevalence and risk factors of neuropathic
pain in survivors of myocardial infarction with
pre-diabetes and diabetes. The KORA Myocardial
Infarction Registry. Eur J Pain. 2009;13:582–7.
https://doi.org/10.1016/j.ejpain.2008.07.007.
88. Gylfadottir SS, et  al. Diabetic polyneuropathy
and pain, prevalence, and patient characteristics:
a cross-sectional questionnaire study of 5,514
patients with recently diagnosed type 2 diabetes.
Pain. 2020;161:574–83. https://doi.org/10.1097/j.
pain.0000000000001744.
89. Gylfadottir SS, et  al. Diagnosis and prevalence of
diabetic polyneuropathy: a cross-sectional study
of Danish patients with type 2 diabetes. Eur J
Neurol. 2020;27:2575–85. https://doi.org/10.1111/
ene.14469.
90. Daousi C, et  al. Chronic painful peripheral neu-
ropathy in an urban community: a controlled
comparison of people with and without diabe-
tes. Diabet Med. 2004;21:976–82. https://doi.
org/10.1111/j.1464-­5491.2004.01271.x.
91. Abbott CA, Malik RA, van Ross ER, Kulkarni
J, Boulton AJ.  Prevalence and characteristics of
painful diabetic neuropathy in a large community-­
The Epidemiology of Diabetic Neuropathy
based diabetic population in the U.K.  Diabetes
Care. 2011;34:2220–4. https://doi.org/10.2337/
dc11-­1108.
92. Garoushi S, Johnson MI, Tashani OA.  A cross-­
sectional study to estimate the point prevalence of
painful diabetic neuropathy in Eastern Libya. BMC
Public Health. 2019;19:78. https://doi.org/10.1186/
s12889-­018-­6374-­9.
93. Spallone V, Greco C.  Painful and painless dia-
betic neuropathy: one disease or two? Curr Diab
Rep. 2013;13:533–49. https://doi.org/10.1007/
s11892-­013-­0387-­7.
94. Karlsson P, Hincker AM, Jensen TS, Freeman R,
Haroutounian S. Structural, functional, and symptom
relations in painful distal symmetric polyneuropa-
thies: a systematic review. Pain. 2019;160:286–97.
https://doi.org/10.1097/j.pain.0000000000001381.
95. Daousi C, Benbow SJ, Woodward A, MacFarlane
IA.  The natural history of chronic painful periph-
eral neuropathy in a community diabetes popula-
tion. Diabet Med. 2006;23:1021–4. https://doi.
org/10.1111/j.1464-­5491.2006.01904.x.
96. Benbow SJ, Chan AW, Bowsher D, MacFarlane IA,
Williams G.  A prospective study of painful symp-
toms, small-fibre function and peripheral vascu-
lar disease in chronic painful diabetic neuropathy.
Diabet Med. 1994;11:17–21.
97. Boulton AJ, Armstrong WD, Scarpello JH, Ward
JD. The natural history of painful diabetic neuropa-
thy–a 4-year study. Postgrad Med J. 1983;59:556–9.
98. Ang L, Jaiswal M, Martin C, Pop-Busui R. Glucose
control and diabetic neuropathy: lessons from recent
large clinical trials. Curr Diab Rep. 2014;14:528.
https://doi.org/10.1007/s11892-­014-­0528-­7.
99. Smith AG, et  al. Lifestyle intervention for pre-­
diabetic neuropathy. Diabetes Care. 2006;29:1294–
9. https://doi.org/10.2337/dc06-­0224.
100. Singleton JR, et  al. Exercise increases cutaneous
nerve density in diabetic patients without neuropa-
thy. Ann Clin Transl Neurol. 2014;1:844–9. https://
doi.org/10.1002/acn3.125.
101. Malik RA, et  al. Sural nerve pathology in dia-
betic patients with minimal but progressive neu-
ropathy. Diabetologia. 2005;48:578–85. https://doi.
org/10.1007/s00125-­004-­1663-­5.
102. Pirart J. Diabetes mellitus and its degenerative com-
plications: a prospective study of 4,400 patients
observed between 1947 and 1973 (author’s transl).
Diabete Metab. 1977;3:97–107.
103. Pirart J. Diabetes mellitus and its degenerative com-
plications: a prospective study of 4,400 patients
observed between 1947 and 1973 (2nd part) (author’s
transl). Diabete Metab. 1977;3:173–82.
104. Pirart J. Diabetes mellitus and its degenerative com-
plications: a prospective study of 4,400 patients
observed between 1947 and 1973 (3rd and last part).
Diabete Metab. 1977;3:245–56.
105. Pop-Busui R, et al. Impact of glycemic control strat-
egies ontheprogressionofdiabeticperipheral neu-
ropathy in the bypass angioplasty revascularization
33
investigation 2 diabetes (BARI 2D) Cohort. Diabetes
Care. 2013;36:3208–15. https://doi.org/10.2337/
dc13-­0012.
106. Sands ML, Shetterly SM, Franklin GM, Hamman
RF.  Incidence of distal symmetric (sensory) neu-
ropathy in NIDDM.  The San Luis Valley Diabetes
Study. Diabetes Care. 1997;20:322–9. https://doi.
org/10.2337/diacare.20.3.322.
107. Srinivasan S, et  al. Four-year Incident Neuropathy
and its Risk Factors in Subjects with Type 2 Diabetes.
J Assoc Physicians India. 2019;67:34–7.
108. Forrest KY, Maser RE, Pambianco G, Becker
DJ, Orchard TJ.  Hypertension as a risk fac-
tor for diabetic neuropathy: a prospective study.
Diabetes. 1997;46:665–70. https://doi.org/10.2337/
diab.46.4.665.
109. Pambianco G, et  al. The 30-year natural history
of type 1 diabetes complications: the Pittsburgh
Epidemiology of Diabetes Complications Study
experience. Diabetes. 2006;55:1463–9. https://doi.
org/10.2337/db05-­1423.
110. Ou HT, Lee TY, Li CY, Wu JS, Sun ZJ.  Incidence
of diabetes-related complications in Chinese patients
with type 1 diabetes: a population-based longitudinal
cohort study in Taiwan. BMJ Open. 2017;7:e015117.
https://doi.org/10.1136/bmjopen-­2016-­015117.
111. Papanas N, Ziegler D. Risk factors and comorbidities
in diabetic neuropathy: an update 2015. Rev Diabet
Stud. 2015;12:48–62. https://doi.org/10.1900/
RDS.2015.12.48.
112. The Diabetes Control and Complications Trial
Research Group. The effect of intensive treatment
of diabetes on the development and progression of
long-term complications in insulin-dependent dia-
betes mellitus. N Engl J Med. 1993;329:977–86.
https://doi.org/10.1056/nejm199309303291401.
113. Ziegler D, Behler M, Schroers-Teuber M, Roden
M. Near-normoglycaemia and development of neu-
ropathy: a 24-year prospective study from diagno-
sis of type 1 diabetes. BMJ Open. 2015;5:e006559.
https://doi.org/10.1136/bmjopen-­2014-­006559.
114. Patel A, et  al. Intensive blood glucose control and
vascular outcomes in patients with type 2 diabe-
tes. N Engl J Med. 2008;358:2560–72. https://doi.
org/10.1056/NEJMoa0802987.
115. Callaghan BC, Little AA, Feldman EL, Hughes
RA.  Enhanced glucose control for prevent-
ing and treating diabetic neuropathy. Cochrane
Database Syst Rev. 2012;6:CD007543. https://doi.
org/10.1002/14651858.CD007543.pub2.
116. Callaghan BC, et al. Association between metabolic
syndrome components and polyneuropathy in an
obese population. JAMA Neurol. 2016;73:1468–76.
https://doi.org/10.1001/jamaneurol.2016.3745.
117. Markakis K, Bowling FL, Boulton AJ. The diabetic
foot in 2015: an overview. Diabetes Metab Res
Rev. 2016;32(1):169–78. https://doi.org/10.1002/
dmrr.2740.
118. Brownrigg JR, et  al. The association of ulceration
of the foot with cardiovascular and all-cause mor-
34
tality in patients with diabetes: a meta-­ analysis.
Diabetologia. 2012;55:2906–12. https://doi.
org/10.1007/s00125-­012-­2673-­3.
119. Bjerg L, Hulman A, Charles M, Jørgensen ME, Witte
DR.  Clustering of microvascular complications in
Type 1 diabetes mellitus. J Diabetes Complicat.
2018;32:393–9. https://doi.org/10.1016/j.
jdiacomp.2018.01.011.
120. Hsu WC, et  al. Somatic neuropathy is an inde-
pendent predictor of all- and diabetes-related
mortality in type 2 diabetic patients: a population-­
based 5-year follow-up study (KCIS No. 29).
Eur J Neurol. 2012;19:1192–8. https://doi.
org/10.1111/j.1468-­1331.2011.03659.x.
121. Vinik AI, Ziegler D.  Diabetic cardiovas-
cular autonomic neuropathy. Circulation.
2007;115:387–97. https://doi.org/10.1161/
CIRCULATIONAHA.106.634949.
122. Spallone V, et  al. Recommendations for the use
of cardiovascular tests in diagnosing diabetic
autonomic neuropathy. Nutr Metab Cardiovasc
Dis. 2011;21:69–78. https://doi.org/10.1016/j.
numecd.2010.07.005.
123. Ewing DJ, Clarke BF. Diabetic autonomic neuropa-
thy: present insights and future prospects. Diabetes
Care. 1986;9:648–65.
124. The Consensus Committee of the American
Autonomic Society and the American Academy
of Neurology. Consensus statement on the defini-
tion of orthostatic hypotension, pure autonomic
failure, and multiple system atrophy. Neurology.
1996;46:1470.
125. Papanas N, Ziegler D. Prediabetic neuropathy: does
it exist? Curr Diab Rep. 2012;12:376–83. https://doi.
org/10.1007/s11892-­012-­0278-­3.
126. Diakakis GF, et al. Association of impaired glucose
tolerance with increased heart rate and subclinical
inflammation. Hell J Cardiol. 2005;46:394–401.
127. Fiorentini A, Perciaccante A, Paris A, Serra P,
Tubani L.  Circadian rhythm of autonomic activ-
ity in non diabetic offsprings of type 2 diabetic
patients. Cardiovasc Diabetol. 2005;4:15. https://
doi.org/10.1186/1475-­2840-­4-­15.
128. Gerritsen J, et al. Glucose tolerance and other deter-
minants of cardiovascular autonomic function: the
Hoorn Study. Diabetologia. 2000;43:561–70. https://
doi.org/10.1007/s001250051344.
129. Gerritsen J, et  al. Impaired autonomic function is
associated with increased mortality, especially in
subjects with diabetes, hypertension, or a history of
cardiovascular disease: the Hoorn Study. Diabetes
Care. 2001;24:1793–8.
130. Singh JP, et  al. Association of hyperglycemia with
reduced heart rate variability (The Framingham
Heart Study). Am J Cardiol. 2000;86:309–12.
131. Stein PK, et al. The relationship of heart rate and heart
rate variability to non-diabetic fasting glucose levels
and the metabolic syndrome: the Cardiovascular
Health Study. Diabet Med. 2007;24:855–63. https://
doi.org/10.1111/j.1464-­5491.2007.02163.x.
C. S. Hansen et al.
132. Uhlig S, Meylan A, Rudolph U.  Reliability of
short-term measurements of heart rate variability:
findings from a longitudinal study. Biol Psychol.
2020;2020:107905. https://doi.org/10.1016/j.
biopsycho.2020.107905.
133. Hojgaard MV, Holstein-Rathlou NH, Agner E,
Kanters JK. Reproducibility of heart rate variability,
blood pressure variability and baroreceptor sensitiv-
ity during rest and head-up tilt. Blood Press Monit.
2005;10:19–24.
134. Körei AE, et al. Why not to use the handgrip test in
the assessment of cardiovascular autonomic neu-
ropathy among patients with diabetes mellitus? Curr
Vasc Pharmacol. 2017;15:66–73. https://doi.org/10.
2174/1570161114666160822154351.
135. Bernardi L, et  al. Investigation methods for car-
diac autonomic function in human research stud-
ies. Diabetes Metab Res Rev. 2011;27(7):654–64.
https://doi.org/10.1002/dmrr.1224.
136. von Scholten BJ, et  al. Cardiac autonomic func-
tion is associated with the coronary microcircu-
latory function in patients with type 2 diabetes.
Diabetes. 2016;65:3129–38. https://doi.org/10.2337/
db16-­0437.
137. Zobel EH, et  al. Cardiac autonomic function is
associated with myocardial flow reserve in type 1
diabetes. Diabetes. 2019;68:1277–86. https://doi.
org/10.2337/db18-­1313.
138. Maser RE, et  al. Diabetic autonomic neuropathy
and cardiovascular risk. Pittsburgh Epidemiology of
Diabetes Complications Study III. Arch Intern Med.
1990;150:1218–22.
139. Kempler P, et  al. Autonomic neuropathy is asso-
ciated with increased cardiovascular risk fac-
tors: the EURODIAB IDDM Complications
Study. Diabet Med. 2002;19:900–9. https://doi.
org/10.1046/j.1464-­5491.2002.00821.x.
140. Witte DR, et al. Risk factors for cardiac autonomic
neuropathy in type 1 diabetes mellitus. Diabetologia.
2005;48:164–71. https://doi.org/10.1007/
s00125-­004-­1617-­y.
141. Tannus LR, Drummond KR, Clemente EL, da Matta
Mde F, Gomes MB.  Predictors of cardiovascular
autonomic neuropathy in patients with type 1 dia-
betes. Front Endocrinol. 2014;5:191. https://doi.
org/10.3389/fendo.2014.00191.
142. Pop-Busui R, et al. Effects of prior intensive insulin
therapy on cardiac autonomic nervous system func-
tion in type 1 diabetes mellitus: the Diabetes Control
and Complications Trial/Epidemiology of Diabetes
Interventions and Complications study (DCCT/
EDIC). Circulation. 2009;119:2886–93. https://doi.
org/10.1161/CIRCULATIONAHA.108.837369.
143. Metwalley KA, Hamed SA, Farghaly HS.  Cardiac
autonomic function in children with type 1 diabe-
tes. Eur J Pediatr. 2018;177:805–13. https://doi.
org/10.1007/s00431-­018-­3122-­1.
144. Kim HY, Jung HW, Lee YA, Shin CH, Yang
SW.  Cardiac autonomic neuropathy in nonobese
young adults with type 1 diabetes. Ann Pediatr
The Epidemiology of Diabetic Neuropathy
Endocrinol Metabol. 2019;24:180–6. https://doi.
org/10.6065/apem.2019.24.3.180.
145. Ratzmann KP, Raschke M, Gander I, Schimke E.
Prevalence of peripheral and autonomic neuropathy
in newly diagnosed type II (noninsulin-­dependent)
diabetes. J Diabet Complications. 1991;5:1–5.
https://doi.org/10.1016/0891-­6632(91)90002-­7.
146. Ziegler D, Dannehl K, Muhlen H, Spüler M, Gries
FA.  Prevalence of cardiovascular autonomic dys-
function assessed by spectral analysis, vector anal-
ysis, and standard tests of heart rate variation and
blood pressure responses at various stages of dia-
betic neuropathy. Diabet Med. 1992;9:806–14.
147. Jyotsna VP, Sahoo A, Sreenivas V, Deepak
KK.  Prevalence and pattern of cardiac autonomic
dysfunction in newly detected type 2 diabetes mel-
litus. Diabetes Res Clin Pract. 2009;83:83–8. https://
doi.org/10.1016/j.diabres.2008.09.054.
148. Arif ZA, Shaikh IA, Masood N.  Cardiovascular
autonomic neuropathy (CAN) in patients of type
2 diabetes mellitus: a tertiary care hospital based
study. Indian Heart J. 2014;66:751–4. https://doi.
org/10.1016/j.ihj.2014.10.417.
149. Pop-Busui R, et  al. Effects of cardiac autonomic
dysfunction on mortality risk in the action to con-
trol cardiovascular risk in diabetes (ACCORD)
trial. Diabetes Care. 2010;33:1578–84. https://doi.
org/10.2337/dc10-­0125.
150. Zoppini G, et  al. Prevalence of cardiovascular
autonomic neuropathy in a cohort of patients with
newly diagnosed type 2 diabetes: the verona newly
diagnosed type 2 diabetes study (VNDS). Diabetes
Care. 2015;38:1487–93. https://doi.org/10.2337/
dc15-­0081.
151. Hansen CS, et al. Vitamin B12 deficiency is associ-
ated with cardiovascular autonomic neuropathy in
patients with type 2 diabetes. J Diabetes Complicat.
2016;31(1):202–8. https://doi.org/10.1016/j.
jdiacomp.2016.08.025.
152. Chung JO, et  al. Serum apolipoprotein A-1 con-
centrations and the prevalence of cardiovascular
autonomic neuropathy in individuals with type 2
diabetes. J Diabetes Complicat. 2018;32:357–61.
https://doi.org/10.1016/j.jdiacomp.2018.01.006.
153. Andersen ST, et al. Risk factors for the presence and
progression of cardiovascular autonomic neuropathy
in type 2 diabetes: ADDITION-Denmark. Diabetes
Care. 2018;41:2586–94. https://doi.org/10.2337/
dc18-­1411.
154. Ko SH, et  al. Cardiovascular autonomic dysfunc-
tion predicts acute ischaemic stroke in patients
with Type 2 diabetes mellitus: a 7-year follow-up
study. Diabet Med. 2008;25:1171–7. https://doi.
org/10.1111/j.1464-­5491.2008.02567.x.
155. Bhuyan AK, Baro A, Sarma D, Choudhury B.  A
study of cardiac autonomic neuropathy in patients
with type 2 diabetes mellitus: a Northeast India expe-
rience. Indian J Endocrinol Metabol. 2019;23:246–
50. https://doi.org/10.4103/ijem.IJEM_336_18.
35
156. Mather KJ, et  al. Prevalence of microvascular and
macrovascular disease in the glycemia reduction
approaches in diabetes  - a comparative effective-
ness (GRADE) study cohort. Diabetes Res Clin
Pract. 2020;165:108235. https://doi.org/10.1016/j.
diabres.2020.108235.
157. Valensi P, Pariès J, Attali JR.  Cardiac autonomic
neuropathy in diabetic patients: influence of diabetes
duration, obesity, and microangiopathic complica-
tions–the French multicenter study. Metabolism.
2003;52:815–20. https://doi.org/10.1016/
s0026-­0495(03)00095-­7.
158. Low PA, et  al. Autonomic symptoms and diabetic
neuropathy: a population-based study. Diabetes
Care. 2004;27:2942–7.
159. Jaiswal M, et al. Cardiovascular autonomic neuropa-
thy in adolescents and young adults with type 1 and
type 2 diabetes: The SEARCH for Diabetes in Youth
Cohort Study. Pediatr Diabetes. 2018;19:680–9.
https://doi.org/10.1111/pedi.12633.
160. Migisha R, et  al. Prevalence and correlates of car-
diovascular autonomic neuropathy among patients
with diabetes in Uganda: a hospital-based cross-­
sectional study. Glob Heart. 2020;15:21. https://doi.
org/10.5334/gh.765.
161. Pan Q, et  al. Prevalence and diagnosis of dia-
betic cardiovascular autonomic neuropathy in
Beijing, China: a retrospective multicenter clinical
study. Front Neurosci. 2019;13:1144. https://doi.
org/10.3389/fnins.2019.01144.
162. Tang M, Donaghue KC, Cho YH, Craig
ME.  Autonomic neuropathy in young people
with type 1 diabetes: a systematic review. Pediatr
Diabetes. 2013;14:239–48. https://doi.org/10.1111/
pedi.12039.
163. Veglio M, et  al. Autonomic neuropathy in non-­
insulin-­dependent diabetic patients: correlation with
age, sex, duration and metabolic control of diabetes.
Diabete Metab. 1990;16:200–6.
164. Gæde P, et  al. Years of life gained by multifacto-
rial intervention in patients with type 2 diabetes
mellitus and microalbuminuria: 21 years follow-
­up on the Steno-2 randomised trial. Diabetologia.
2016;59:2298–307. https://doi.org/10.1007/
s00125-­016-­4065-­6.
165. Reichard P.  Risk factors for progression of
microvascular complications in the Stockholm
Diabetes Intervention Study (SDIS). Diabetes
Res Clin Pract. 1992;16:151–6. https://doi.
org/10.1016/0168-­8227(92)90087-­8.
166. Tang Y, et  al. Intensive risk factor manage-
ment and cardiovascular autonomic neuropathy
in type 2 diabetes: the ACCORD trial. Diabetes
Care. 2021;44:164–73. https://doi.org/10.2337/
dc20-­1842.
167. Choung RS, et al. Risk of gastroparesis in subjects
with type 1 and 2 diabetes in the general population.
Am J Gastroenterol. 2012;107:82–8. https://doi.
org/10.1038/ajg.2011.310.
36
168. Horowitz M, et al. Relationships between oesopha-
geal transit and solid and liquid gastric emptying in
diabetes mellitus. Eur J Nucl Med. 1991;18:229–34.
https://doi.org/10.1007/bf00186645.
169. Parkman HP, Hasler WL, Fisher RS.  American
Gastroenterological Association technical review
on the diagnosis and treatment of gastroparesis.
Gastroenterology. 2004;127:1592–622. https://doi.
org/10.1053/j.gastro.2004.09.055.
170. Camilleri M.  Clinical practice. Diabetic gastropa-
resis. N Engl J Med. 2007;356:820–9. https://doi.
org/10.1056/NEJMcp062614.
171. Kempler P, et  al. Management strategies for gas-
trointestinal, erectile, bladder, and sudomotor dys-
function in patients with diabetes. Diabetes Metab
Res Rev. 2011;27:665–77. https://doi.org/10.1002/
dmrr.1223.
C. S. Hansen et al.
172. Albers JW, Pop-Busui R.  Diabetic neuropathy:
mechanisms, emerging treatments, and subtypes.
Curr Neurol Neurosci Rep. 2014;14:473. https://doi.
org/10.1007/s11910-­014-­0473-­5.
173. Albers JW, Brown MB, Sima AA, Greene
DA.  Frequency of median mononeuropathy in
patients with mild diabetic neuropathy in the early
diabetes intervention trial (EDIT). Tolrestat Study
Group For Edit (Early Diabetes Intervention Trial).
Muscle Nerve. 1996;19:140–6.
174. Chan YC, Lo YL, Chan ES.  Immunotherapy
for diabetic amyotrophy. Cochrane Database
Syst Rev. 2017;7:Cd006521. https://doi.
org/10.1002/14651858.CD006521.pub4.
175. Gibbons CH.  Treatment-induced neuropathy of
diabetes. Curr Diab Rep. 2017;17:127. https://doi.
org/10.1007/s11892-­017-­0960-­6.
 Clinical Features of Diabetes
Neuropathies
Gordon Sloan, Qi Pan, Ling Gao, Lixin Guo,
and Solomon Tesfaye
1 Introduction
Patients with diabetes may present with a wide
range of neuropathic syndromes with varied clin-
ical presentations, disease aetiologies, prognosis,
and treatments. By far the commonest and most
well understood is diabetic peripheral neuropathy
(DPN). DPN is a chronic and progressive condi-
tion, defined as ‘a symmetrical, length-dependent
sensorimotor polyneuropathy attributable to met-
abolic and microvessel alterations as a result of
chronic hyperglycaemia exposure (diabetes) and
cardiovascular risk covariates’ [1]. This condi-
tion is the precipitant for two of the most trouble-
some chronic complications of diabetes,
painful-DPN, and diabetic foot diseases (i.e., foot
ulceration, Charcot neuroarthropathy, and infec-
tion). The latter is the main precipitant for lower
G. Sloan
Diabetes Research Unit, Sheffield Teaching Hospitals
Trust and the University of Sheffield, Sheffield, UK
Q. Pan · L. Guo
Department of Endocrinology, National Center of
Gerontology, Beijing Hospital, Beijing, China
L. Gao
Department of Endocrinology and Metabolism,
Renmin Hospital of Wuhan University, Wuhan, China
S. Tesfaye (*)
Diabetes Research Unit, Sheffield Teaching Hospitals
and the University of Sheffield, Sheffield, UK
e-mail: solomon.tesfaye@nhs.net
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_3
limb amputation in diabetes. However, a number
of other syndromes may also present in diabetes
which involve the cranial nerves, peripheral
nerves, radicular nerves, nerve plexus, the and
autonomic nervous system. This chapter will
describe the classification of diabetic neuropa-
thies and their clinical features, with the main
emphasis on neuropathies of the lower
extremity.
2 Epidemiology
There is a great variation in the reported preva-
lence of DPN. This is likely secondary to differ-
ing study methods; although regional/population
variation may also contribute. The overall preva-
lence of clinical DPN is estimated at approxi-
mately 30% [2]. Where sensitive diagnostic
techniques are used, such as nerve conduction
studies, the prevalence rises to over 50%. The
prevalence of DPN is dependent upon the dura-
tion of diabetes. In type 1 diabetes mellitus
(T1DM), newly diagnosed patients have a low
prevalence of DPN, but in newly diagnosed type
2 diabetes mellitus (T2DM) the prevalence is
approximately 13% [3]. The most important risk
factor for DPN is HbA1c followed by duration of
diabetes/age [2]. However, other modifiable car-
diovascular risk factors have also been shown to
be independent predictors of DPN [4].
37
38
The prevalence of autonomic neuropathy in
diabetes is also heavily dependent upon the study
methods, particularly the diagnostic technique
and case definition. The prevalence of cardiovas-
cular autonomic neuropathy may be at least 30%
in patients with type 1 diabetes after 20 years dis-
ease duration and 60% in patients with type 2
diabetes after 15 years [5]. The epidemiological
data on other diabetic autonomic neuropathies is
limited, although the cumulative incidence of
diabetic gastroparesis over 10 years was higher in
type 1 diabetes (5%) than in type 2 diabetes (1%)
and in control subjects (1%).
The prevalence of other diabetic neuropathies
is even less well known. Median and ulnar nerve
entrapment may occur in over 30% of patients
with type 1 or 2 diabetes [6, 7]. In a retrospective
study of 8150 patients with diabetes who were
hospitalized in Marsala, Italy, over a 12-year
period, 0.75% had evidence of cranial nerve pal-
sies, with third nerve palsies the most common
(0.35%), followed by seventh (0.21%) and sixth
(0.15%) nerve palsies [8]. The prevalence of cra-
nial nerve palsies in the community, however, is
unknown. A study in Olmsted County, Minnesota
found the prevalence of lumbosacral radiculo-
Polyneuropathy Mononeuropathies
Fig. 1  Clinical presentation of different neuropathic syndromes in diabetes (reproduced from [11])
G. Sloan et al.
plexus neuropathy to be higher in diabetes
(2.79/100,000/year [95% CI 1.94–3.64]) com-
pared to non-diabetes patients (1.27/100,000/
year [95% CI 0.71–1.83]) [9]. Chronic inflamma-
tory demyelinating polyneuropathy has classi-
cally been described to be more common in
diabetes, although epidemiological studies have
failed to show an association [10].
3 Classification
The diabetic neuropathies have a heterogeneous
clinical presentation, natural history, pattern of
neurological impairment, underlying nerve
pathology, risk factors, and disease mechanisms
(Fig.  1). A number of different classification
schemes for diabetic neuropathies have been pro-
posed over the years. The most recent and widely
used classification scheme in the Position
Statement by the American Diabetes Association
stratifies syndromes according to their neurologi-
cal pattern of involvement (Table 1). All of these
neuropathic syndromes may present in patients
without diabetes with other underlying disease
mechanisms. Moreover, patients with diabetes
Plexopathy Autonomic neuropathy
Clinical Features of Diabetes Neuropathies 39

Table 1  Classification of diabetic neuropathies (adapted are more likely to experience entrapment
from Pop-Busui et al. [5])
neuropathies.
Classification of diabetic neuropathies
Generalized symmetric polyneuropathies
 • Diabetic sensorimotor polyneuropathy
   – Mixed sensorimotor neuropathy
4 Clinical Features of Diabetic
   – Predominantly large fibre neuropathy Neuropathies
   – Predominantly small fibre neuropathy
   – Pure small fibre neuropathy 4.1 Diabetic Peripheral
 • Acute painful-distal sensorimotor
polyneuropathies
Neuropathy
   – Treatment induced neuropathy of diabetes
[TIND] In DPN there is a ‘length-related’ pattern of neu-
   – Hyperglycaemia induced ronal dysfunction, with sensory symptoms/signs
 • Autonomic neuropathy starting symmetrically in the toes of both feet and
   – Cardiovascular autonomic neuropathy
    Reduced heart rate variability then extending as the disease advances to involve
    Resting tachycardia the rest of the foot and lower limb in a ‘stocking’
    Orthostatic hypotension distribution (Fig. 2). Once the disease has reached
    Sudden cardiac death approximately the level of the knees it usually is
    Exercise intolerance
    Silent myocardial ischaemia also present in the distal upper-limb, the finger-
   – Gastrointestinal autonomic neuropathy tips. With upper limb involvement, there is a
    Oesophageal dysmotility similar symmetrical pattern of progression proxi-
    Gastroparesis mally in a ‘glove’ distribution. Although the
    Diabetic diarrhoea
    Constipation nerve damage can extend over the entire body
    Faecal incontinence including the trunk, and even the head and face,
   – Urogenital this is exceptional. As the disease advances, overt
    Neurogenic bladder motor manifestations such as wasting of the
    Sexual dysfunction (erectile and female sexual
dysfunction; retrograde ejaculation) small muscles of the hands and limb weakness
   – Sudomotor dysfunction become apparent. The precise temporal course of
    Anhidrosis nerve fibre injury is unclear, it has been proposed
    Hyperhidrosis that sensory small-fibre dysfunction occurs first,
    Gustatory sweating
   – Metabolic (i.e., hypoglycaemia unawareness) followed by sensory and motor large-fibre; how-
   – Pupillary dysfunction ever, there is no clear evidence for this hypothesis
Focal/multifocal neuropathies [11]. Isolated small-fibre neuropathy is a possible
 • Mononeuropathy manifestation of DPN, although this is uncom-
   – Mononeuropathies mon (approximately 2.5%) [12]. When present,
    Cranial nerve [e.g., oculomotor nerve palsy,
facial nerve palsy] pure small-fibre neuropathy involves clinical loss
    Peripheral nerve [e.g., peroneal nerve palsy] of modalities serving pain and thermal sensation
 • Mononeuritis multiplex but preservation of large fibre modalities (such as
 • Radiculoplexus neuropathy vibration, light touch, proprioception, and motor
   – Radiculopathy
function).
    Thoracic
    Abdominal The clinical presentation of DPN is often
    Thoraco-abdominal insidious. Patients may be entirely asymptom-
   – Plexopathy atic; hence the critical importance of diabetic
    Lumbosacral plexopathy
foot screening, as all too often the first presenta-
    Cervical plexopathy
 • Entrapment neuropathies (e.g. carpal tunnel tion of DPN is with a diabetic foot ulcer. The sen-
syndrome, ulnar neuropathy) sory symptoms which develop in DPN are often
divided into ‘positive’ and ‘negative’ symptoms,
the former referring to paresthesias/painful neu-
ropathic symptoms and the latter to symptoms of
40 G. Sloan et al.

Slight sensory loss Moderate sensory loss Extensive sensory loss

Diminished/absent sensation in
toes (light touch/vibration/pin-
prick)
Normal motor examination
except ankle reflex may be
diminished/absent
Significantly impaired/absent
sensation in the foot extending up
the leg – risk of foot ulceration
Absent ankle reflex
Foot deformity
Weakness in toe/ankle
flexion/extension
Absent sensation in the foot and
impaired sensation in the upper
and lower limb (rarely the trunk)
Weakness and/or wasting in the
lower/upper limb
Impaired knee reflex
Gait/balance impairment

Fig. 2  Clinical presentation and progression of sensory changes in diabetic peripheral neuropathy (reproduced from
[11])

sensory loss. Negative symptoms may be sensory abnormalities. Innervation of sweat

described as an awareness of abnormal sensation
such as a ‘numbness’ or ‘deadness’ in their feet.
Patients may describe feeling as if their socks are
always bunched up in their shoes or as if they are
walking on pebbles or cotton wool. Whereas pos-
itive symptoms may be non-painful, such as tin-
gling, pressing, tightness or pins and needles, or
painful, as in painful-DPN.  Sensory symptoms
progress in the aforementioned ‘stocking/glove
distribution’, although different regions of the
body may experience different symptoms. For
example, in advanced disease patients may expe-
rience neuropathic pain in the upper limb whereas
the lower-limb may have only negative symp-
toms. Patients may also notice impaired balance
and gait with advanced lower limb neuropathy.
Once the disease affects the hands patients may
initially feel reduced sensation of finger-prick
blood glucose testing before developing pro-
found difficulties with day-to-day tasks such as
typing, opening tins/jars, and gripping.
The early clinical signs of DPN include a distal
loss of light touch, temperature, and pin-prick
sensation before proprioception and vibration
glands is also impaired resulting in ‘sudomotor
dysfunction’, which leads to skin dryness and
cracking, acting as a potential precipitant for
infection and foot ulceration. Early motor mani-
festations include the absence or reduction in
ankle reflexes bilaterally. Reduce ankle reflexes
may be present in elderly patients without
DPN. Another motor sign is foot deformity, such
as clawing of the toes and prominent metatarsal
heads as a result of denervation of the interphalan-
geal extensor and metacarpal-phalangeal flexor
muscles. More advanced clinical motor findings
include motor weakness, wasting of the small
muscles of the foot and extensor hallucis longus,
high arching of the foot and the absence of knee
reflexes. The presence of foot deformity and sen-
sory loss in the foot predispose the patient to com-
plications such as foot ulceration and Charcot
neuroarthropathy. Similar to foot involvement, the
distal extremity of the limb, the tips of the fingers,
are affected first before spreading proximally in a
‘glove’ distribution. As upper limb involvement
progresses sensory loss becomes more advanced
and there is weakness with wasting of the muscles
Clinical Features of Diabetes Neuropathies
of the hands. However, advanced upper limb
involvement in DPN is rare, and other causative
conditions should be considered, such as entrap-
ment neuropathy or spinal cord pathology.
To complete the examination of the foot it is
important to examine the skin, joints, and vascu-
lar supply of the foot and the footwear of the
patient. The neuropathic foot may be warm with
a bounding arterial pulse, due to peripheral auto-
nomic neuropathy. However, peripheral vascular
disease is a common co-morbidity in DPN. This
may manifest itself as peripheral hair loss, cool/
cold skin, pallor and reduced/absent peripheral
pulses. A thorough assessment of the patients’
footwear is also mandatory. Patients with DPN
commonly wear shoes which are too small as
they ‘feel comfortable’. Furthermore, it is possi-
ble to detect areas of wear from within the shoe.
Trauma resulting from a poor fit, abnormal wear
from internal pressure areas and foreign objects
are common contributors to the pathway to foot
ulceration [13].
4.1.1 Painful Diabetic Peripheral
Neuropathy
Despite patients with painful-DPN often suffering
with severe neuropathic pain. There are generally
few clinical signs or neurological features to dif-
ferentiate these cases from painless-DPN [14]. It
has been reported that the neuropathic deficit in
painful-DPN may be more advanced, but the clin-
ical examination is normally indistinct from pain-
ful- and painless-DPN.  Patients may report
hyperalgesia (increased pain from a stimulus that
normally provokes pain) and allodynia (pain from
an innocuous stimuli), although clinically repro-
ducible brush-evoked allodynia is rare [15].
The symptoms of painful-DPN widely vary
from patient to patient. It can occur at any sever-
ity of neuropathic deficit in DPN. Indeed, severe
sensory loss in the presence of painful neuro-
pathic symptoms might occur, the so-called pain-
ful/painless leg [16]. The distribution of the pain
depends upon the severity of neuronal loss, in
mild cases the pain may be present in only the
toes; whereas advanced cases may have upper
limb neuropathic pain. The most common
41
descriptors of pain are burning (>60%) and elec-
tric/shocks (40–50%) [12, 17, 18]. The preva-
lence of other symptoms is less clear although
sharp, aching, and cold pains may also be pres-
ent. The pain is often very distressing and more
than half of those with painful-DPN grade their
symptoms as severe [19]. The pain may occur
intermittently throughout the day, but more com-
monly it is constant with nocturnal exacerbation,
leading to disruption of sleep in up to three quar-
ters of individuals [20]. Similar to painless-DPN
the course of the disease over time has not been
thoroughly investigated. A questionnaire based
study (n = 105) found that in 72% of patients the
disease had worsened since onset, 12% reported
the pain improved and in 15% it had stayed the
same [17]. Furthermore, longitudinal studies are
necessary to examine the natural history of
painful-DPN.
4.1.2 Co-morbidities of Diabetic
Peripheral Neuropathy
It is important to consider and address other
sequalae of diabetes and DPN and related con-
ditions as part of the holistic assessment of the
patient with DPN. Most pressingly, DPN is the
main precipitant for diabetic foot disease such
as foot ulceration and Charcot Neuroarthropathy,
which can lead to lower limb amputations.
Co-morbid vascular disease, including cardio-
vascular disease, medial arterial calcification,
and peripheral vascular disease, is also common
in patients with DPN [21]. Moreover, there is an
association between DPN, foot ulceration, and
amputation and a high risk of death [22]. The
other microvascular complications of diabetes,
such a autonomic neuropathy, nephropathy, and
retinopathy, may also be present and there are
also potential associations with DPN and
obstructive sleep apnoea and cognitive dysfunc-
tion [21].
Unsurprisingly, patients with DPN can
develop mood disorders, such as depression, and
can have a reduced quality of life [21]. Moreover,
the neuropathic symptoms in painful-DPN can
cast a huge burden on all aspects of patients’
lives, resulting in a limitation in all activities of
42 G. Sloan et al.

daily living, including their employment [12]. Such atypical features might include: weight
Therefore, it is common that these patients have loss; an absence of other complications of diabe-
depression, anxiety, sleep impairment and a pro- tes; asymmetry; rapidly progressing symptoms;
found reduced quality of life. predominant motor signs; predominant upper
limb signs; and evidence of multi-system
4.1.3 Differential Diagnosis ­disorders (i.e., malignancy, autoimmune disease,
of Diabetic Peripheral connective tissue disease, vasculitis, etc.). If a
Neuropathy diagnosis of a peripheral neuropathy/polyneu-
The potential differential diagnosis of DPN is ropathy is made, a careful clinical history and
vast. If there are clinical features present which examination and simple investigations (e.g., thy-
are not typical for DPN, then alternative diagno- roid function tests, vitamin B12, serum/urine
ses should be considered and excluded (Table 2). electrophoresis, full blood count and inflamma-
tory markers) is usually sufficient to exclude
other neuropathies.
Table 2  Differential diagnosis of DPN
Differential diagnosis of DPN
 • Metabolic 4.2 Acute Painful-Distal
   – Hypothyroidism
   – Vitamin deficiency (B12, thiamine, vitamin E,
Sensorimotor
post-gastric surgery, B6, pyridoxine, etc.) Polyneuropathies
   – Uraemia
   – Copper deficiency Acute painful neuropathies are rare transient neu-
   – Liver cirrhosis
   – Critical illness
ropathic syndromes presenting in a peripheral dis-
   – Amyloidosis tribution, which develop over a relatively short
   – Porphyria period of time, i.e. days to weeks. Sensory and
 • Drugs and toxins motor neuropathic deficits are often mild, but neu-
   – Alcohol ropathic pain may be severe. There are two distinct
   – Cytotoxic drugs (e.g., vinca alkaloids, platinum
based agents, bortezomib and thalidomide, etc. syndromes, the first of which occurs within the
   – Antibiotics (e.g., chlorambucil, nitrofurantoin, context of poor glycaemic control, and the second
isoniazid, linezolid, metronidazole) with rapid improvement in glycaemic control
   – Anti-retroviral agents (Treatment induced neuropathy of diabetes; TIND).
   – Immunosuppressant agents (e.g., interferons
and monoclonal antibodies)
 • Infectious 4.2.1 Hyperglycaemia Induced Acute
   – HIV Painful Distal Sensorimotor
   – Leprosy Polyneuropathy
   – Guillain-Barre syndrome
This condition is rare and occurs in patients with
   – Lyme borreliosis
   – Viral hepatitis diabetes and poor glycaemic control. It is poorly
   – Neurosyphilis understood as with improvements in the diagno-
 • Inherited sis and treatment of diabetes it is rarely seen in
   – Charcot-Marie-Tooth disease developed countries. The condition is associated
   – Hereditary sensory neuropathies
   – Fabry disease with burning pain and allodynia, which is worse
   – Other rare inherited neuropathies at night. The condition is also associated with
 • Inflammatory concomitant severe weight loss. There is gener-
   – Chronic inflammatory demyelinating ally a complete resolution of symptoms within 12
polyneuropathy
months, and weight gain with improvement in
   – Sarcoidosis
   – Autoimmune disorders (e.g., connective tissue glycaemic control.
diseases)
   – Vasculitis (e.g., polyarteritis nodosa)
 • Paraneoplastic neuropathy
Clinical Features of Diabetes Neuropathies 43

4.2.2 Treatment Induced Neuropathy Moreover, there may be an association with

of Diabetes
Although TIND is better understood than hyper-
glycaemia induced acute painful distal sensorim-
otor polyneuropathy there is still limited data
relating to the condition. This condition is also
known as ‘insulin neuritis’, although this is a
misnomer as the condition is unrelated to the
therapeutic agent used but rather the speed in
which glucose is lowered. A retrospective study
found that a 2% reduction in HbA1c over 3
months was associated with an approximate 10%
risk of TIND; whereas a 5% reduction is associ-
ated with an over 90% risk of developing the con-
dition [23]. A greater degree of HbA1c reduction
was shown to cause greater severity and more
widespread distribution of neuropathy (Fig.  3).
Therefore, the condition is readily identifiable in
the clinical context of neuropathic symptoms in
the presence of improved glucose control.
TIND and prior weight loss/diabetic anorexia
[23, 25]; although, TIND itself is not believed to
present with weight loss unlike some other dia-
betic neuropathies (e.g., hyperglycaemia induced
acute painful distal sensorimotor polyneuropathy
and radiculoplexus neuropathies).
The symptoms of TIND develop approxi-
mately 2–6 weeks following the improvement in
glucose control [23]. The most common symp-
toms include burning and shooting pain, and par-
aesthesia with hyperalgesia and allodynia in the
distribution of the neuropathy. The symptoms are
in a peripheral neuropathy distribution, similar to
DPN. With mild reductions in HbA1c, the symp-
toms may only involve the toes and feet; how-
ever, with greater reduction in HbA1c the upper
limbs and even abdomen and trunk can be
involved in severe cases. Similar to painful-DPN,
the pain is associated with disturbances in sleep.

Decrease in HbA1C
2-3 points 3-5 points 5+ points
Risk of TIND Development

Greater Decrease in HbA1C

Fig. 3  Clinical phenotype in treatment induced neuropathy of diabetes (reproduced from [24]
44 G. Sloan et al.

The examination findings are often mild but may 4.2.4 Cardiovascular Autonomic
involve sensory loss (reduced/absent pinprick Neuropathy
and/or vibration sensation at the toe/foot) and The presence of CAN holds clinical relevance as
distal toe extensor/flexor weakness and ankle it confers an increased risk of silent ischaemia,
reflex impairment. Neurophysiological findings cardiovascular events, arrhythmias, and cardio-
may reveal mild sensory, axonal neuropathy, and vascular mortality [5, 26]. The condition is read-
intra-epidermal nerve fibre density on skin biopsy ily identifiable in patients with longstanding
may be reduced [23, 25]. The condition is also diabetes, particularly those with DPN with which
frequently associated with autonomic symptoms, it shares risk factors, using sensitive autonomic
most commonly orthostatic intolerance and syn- function tests. However, overt clinical manifesta-
cope/pre-syncope although gastrointestinal tions are less common, but can be disabling.
symptoms and sweating alterations may occur Symptoms may include intolerance of exercise,
with greater reductions in HbA1c [23]. Finally, i.e. breathlessness on exertion, due to impaired
the condition is also associated with an increased heart rate variability, and resting tachycardia.
incidence of other microvascular complications Orthostatic symptoms suggestive of a postural
of diabetes, retinopathy, and nephropathy. drop in blood pressure (>20 mmHg) are gener-
There is scant evidence for the natural historyally associated with advanced disease.
of TIND.  It is classically described to improve Participants may experience light-headedness,
over a period of 12–18 months. A publication pre-syncope or syncope on standing, or some-
assessed 26 individuals with T1DM longitudi- times from lying to sitting. Anyone with sus-
nally over 8 years with comprehensive neurologi- pected orthostatic hypotension due to CAN
cal and clinical testing [25]. Patients were should undergo a careful review of their medica-
grouped into ‘stable glycaemic control’ and tions. Many drugs prescribed in patients with
‘unstable glycaemic control’, the latter of whom advanced diabetes or DPN may cause or worsen
developed a further episode of TIND.  Patients orthostatic hypotension, e.g. antihypertensives,
with stable glycaemic control had improvements SGLT-2 inhibitors, diuretics, and anticholinergic
in intra-epidermal nerve fibre density, neurologi- agents such as tricyclic antidepressants. Cardiac
cal examination and nerve conduction measures ischaemia/infarction may also present atypically
after 8 years; whereas the unstable group had a in patients with CAN due to ‘silent ischaemia’.
deterioration in these parameters. Therefore, ischaemia/infarction may be missed
entirely or present atypically. In the acutely
4.2.3 Autonomic Neuropathy unwell patient with diabetes silent myocardial
Diabetic autonomic neuropathies are sub-­ infarction should be considered and investiga-
classified according to the organ or body system tions such as electrocardiograms or serum tropo-
affected. Abnormalities in autonomic function nin levels may need to be performed to exclude
tests are common in patients with longstanding the condition.
diabetes; however, clinically significant auto-
nomic symptoms are uncommon. Other than spe- 4.2.5 Gastrointestinal Autonomic
cific scenarios, such as TIND, the condition Neuropathy
generally has a gradual onset and is slowly pro- Patients with diabetes commonly experience
gressive. Cardiac autonomic neuropathy is troublesome gastrointestinal symptoms which
believed to be the most common; however, the can be severe, causing great patient morbidity.
epidemiological data is heavily dependent upon These symptoms can be due to autonomic neu-
the studied population, diagnostic criteria, and ropathy, but other factors should be considered
investigative method. Autonomic neuropathies such as autoimmune disorders (e.g., coeliac dis-
are discussed in more detail in Chaps. 12, 26, 27, ease and hypo/hyperthyroidism), medications
and 28. (e.g., metformin and GLP-1 agonists), gastroduo-
denal ulceration, and hyperglycaemia. Autonomic
Clinical Features of Diabetes Neuropathies 45

neuropathy in diabetes can affect any part of the Table 3  Differentiation of entrapment and microvascu-
litic causes of mononeuropathies in diabetes
gastrointestinal tract. The most well described
and common presentation in the upper gastroin- Microvasculitic Entrapment
mononeuropathy neuropathy
testinal tract is gastroparesis. Gastroparesis leads
Onset Sudden Gradual
to a slower transit of food through the stomach
Natural Monophasic, Progressive
which may present with nausea, vomiting, and history maximal deficit at deterioration in
abdominal bloating. Moreover, there may be glu- onset of symptoms symptoms
cose variability with unexplained hypoglycaemia with spontaneous without treatment
resolution
due to dissociation between food absorption and
Location Cranial nerves: CN Upper limb:
insulin administration [5]. Other upper gastroin- III, IV, VI, VII, and median and ulnar
testinal symptoms of autonomic neuropathy may others more rarely nerve
include dysphagia and heart burn. Lower gastro- Peripheral nerves: Lower limb:
intestinal symptoms range from constipation to any nerve; peroneal, peroneal, femoral,
femoral, radial, lateral cutaneous
diarrhoea and faecal incontinence. Indeed, diar- ulnar, and median nerve of the thigh,
rhoea may alternate with constipation. Diabetic more common lateral plantar
diarrhoea is often worse at night and may be nerve
associated with small bowel bacterial Symptoms Acute onset of pain Paraesthesia and
and neurological pain within the
overgrowth.
dysfunction in the territory of the
territory of the nerve with motor
4.2.6 Urogenital Autonomic nerve signs occurring
Neuropathy late
Urinary autonomic dysfunction may lead to hesi-
tancy or increased frequency of micturition. In
serious cases there is urinary retention associated 4.3 Mononeuropathies
with overflow incontinence. Women with diabe-
tes may have reduced sexual desire and increased Diabetic mononeuropathies may occur second-
pain during intercourse due to inadequate vaginal ary to compression (entrapment neuropathies) or
lubrication [27]. Erectile dysfunction is common microvasculitis. It is important to distinguish
in men with diabetes; however, its aetiology is between the two as the former can respond well
often multi-faceted, other contributory factors to treatment, such as surgical release of the nerve
including vascular disease, metabolic control, (Table  3). Moreover, other causes should be
nutrition, endocrine disorders (e.g., hypogonad- actively sought and excluded as the differential
ism), psychogenic factors, and medications [27]. diagnosis of mononeuropathies includes serious
causes such cerebrovascular disease.
4.2.7 Sudomotor Dysfunction
Sudomotor dysfunction can lead to anhidrosis or 4.3.1 Microvasculitic
hyperhidrosis. Anhidrosis is relevant to the Mononeuropathies
patient with DPN where peripheral anhidrosis Microvasculitic diabetic mononeuropathies gen-
can lead to drying and cracking of the skin, which erally may involve the cranial nerves or periph-
can precipitate bacterial infection of the foot. eral nerves. The microvasculitic neuropathies
Hyperhidrosis can be distressing and embarrass- present with a rapid onset nerve palsy, often asso-
ing to patients. It is often associated with eating; ciated with pain [28]. The specific symptoms
in this situation it is known as gustatory depend upon the affected nerve. The most com-
sweating. mon cranial nerve involved is the third. Patients
with this condition present with sudden onset
orbital/frontal pain associated with ptosis and
ophthalmoplegia with pupillary sparing. Fourth
and sixth nerve neuropathies would be associated
46
with ophthalmoplegia and facial nerve palsy
occurs with seventh nerve involvement. The dif-
ferential diagnosis of cranial nerve palsy is vast
and should not be attributed to diabetes without a
careful clinical history, examination, investiga-
tions, and specialist input. Cerebral imaging with
computer tomography or magnetic resonance
imaging (MRI) is generally required to exclude
structural compression of the nerve in question or
cerebrovascular disease.
Peripheral mononeuropathies may involve
any nerve, including the peroneal, sural, sciatic,
femoral, ulnar, and median nerve [28]. Similar to
cranial nerve involvement the presentation is sud-
den. There may be motor and/or sensory signs in
the distribution of the peripheral nerve. The con-
dition requires neurophysiological testing, and
sometimes imaging, to exclude other causes such
as entrapment neuropathy. Both peripheral and
cranial neuropathies have a monophasic course
and resolve spontaneously; although this may
take several months.
Mononeuritis multiplex is a rare syndrome
where there is multifocal, asynchronous periph-
eral neuropathy involving at least two peripheral
nerve trunks. There are some limited reports
which have described the condition in patients
with diabetes [29, 30]. However, specialist input
is advised in any diabetes patient presenting with
mononeuritis multiplex.
4.3.2 Entrapment Mononeuropathies
Patients with diabetes are more at risk of entrap-
ment neuropathies. These neuropathies may be
localized to common sites of compression, e.g.
carpal tunnel, ulnar groove, and fibula head, with
the median and ulnar nerve particularly prone to
compression in diabetes [31]. The overweight/
obese patient with type 2 diabetes appears to be
the most at risk for particularly these entrapment
neuropathies. Less commonly affected nerves
include the radial, peroneal, tibial, and lateral
cutaneous nerve of the thigh [28].
Entrapment neuropathies typically have a
gradual onset and more persisting neuronal dys-
function than mononeuropathies due to micro-
vasculitis. Patients may be asymptomatic or
develop paraesthesia and neuropathic pain in the
G. Sloan et al.
distribution of the nerve. Sensory signs such as a
reduction in sensation in the territory of the
nerve can be detected. Motor signs such as weak-
ness and muscle wasting of the innervated mus-
cles are late signs. Median and ulnar nerve
entrapment neuropathies may occur concur-
rently and can be bilateral. However, if this is the
case other causes should be considered and the
patient referred for careful neurophysiological
testing. Indeed, the diagnosis of entrapment neu-
ropathies often requires nerve conduction stud-
ies; however, this is particularly challenging if
there is concurrent DPN. Without treatment the
nerve dysfunction is typically progressive and
can lead to irreversible neurological impairment
in the affected nerve.
4.4 Radiculoplexus Neuropathies
Diabetic radiculoplexus neuropathies are a group
of asymmetric neuropathies with a common
underlying pathophysiologic mechanism of isch-
aemic injury due to altered immunity and micro-
vasculitis [32]. They were first recognized in the
19th Century and have been known by a number of
names, including: Bruns-Garland syndrome, dia-
betic myelopathy, diabetic amyotrophy, femoral-­
sciatic neuropathy of diabetes, etc. [33]. The
condition is most well recognized to occur in dia-
betes; although a similar presentation may occur
in patients without diabetes [9]. The most common
form is lumbosacral radiculoplexopathy, followed
by truncal radiculopathy (thoracic, abdominal or
thoraco-abdominal) and finally cervical radiculo-
plexopathy [32]. Diabetic radiculoplexus neuropa-
thies begins unilaterally in one region, but may
spread bilaterally or cause other radiculoplexopa-
thies, and the condition may then recur in the same
region. Additionally, the condition can be associ-
ated with autonomic neuropathy.
The condition may occur in patients with T1
or T2DM; however, generally patients have
T2DM, and are in middle or old age [32, 33].
Unlike DPN, the condition seems unrelated to
glycaemic control. The condition generally pres-
ents sub-acutely over a number of days with pain
in the region of the nerve distribution. The pain
Clinical Features of Diabetes Neuropathies
can be extreme and can have a deep or burning
quality. There may be associated contact allo-
dynia. Additionally, although sensory symptoms
are severe, there may be a relative lack of sensory
nerve dysfunction on clinical examination and
neurophysiological testing. The condition is also
frequently associated with weight loss, which can
be severe, raising the concern of malignancy.
Although pain is generally the most prominent
initial symptom, muscle weakness develops later,
which can be profound. The symptom course is
monophasic with invariable improvement; how-
ever, recovery is often incomplete [33]. In par-
ticular, muscular weakness can persist and be
disabling for patients and many patients require
lifelong walking aids, if not wheelchairs, for lum-
bosacral radiculoplexopathy.
Lumbosacral radiculoplexopathy generally
causes symptoms in the hip and thigh, although
the foot and leg may be affected. The quadriceps
are classically profoundly wasted, with hip flex-
ors and abductors also commonly involved. Other
muscles such as the adductors, glutei, and ham-
strings are less commonly impacted. When there
is motor involvement the knee reflex is reduced
or absent.
Cervical radiculoplexus neuropathies are
much less common than other diabetic radiculo-
plexus neuropathies. However, a retrospective
study has systematically reviewed the features of
condition in 85 patients, including the clinical
symptoms and signs [34]. The presentation may
be more acute than lumbosacral disease (within
24 h), with maximal deficits within 1-week. The
pain was described as hurting/aching/pressure,
burning or shooting/lancinating with numbness
and numbness also often present. Weight loss and
contralateral spread were less common than with
cervical radiculoplexus neuropathy. Weakness
was a predominant problem, with involvement in
each one of the three trunks of the brachial plexus
to a similar extent and all three involved in around
a third of patients.
Diabetic truncal radiculopathy presents with
band-like sensory symptoms within a truncal der-
matome, from the back radiating anteriorly [32].
The symptoms are not dissimilar from the initial
presentation of herpes zoster radiculitis; although,
47
without the presence of a vesicular eruption.
Neuropathy symptoms vary and may be similar
to those of diabetic radiculoplexopathy. The
examination findings are generally of sensory
disturbance, such as impairment to light touch/
pinprick sensation and/or allodynia/hyperalgesia,
within the distribution of the nerve involved.
Furthermore, the condition can cause local bulg-
ing of muscles within the myotome due to mus-
cular weakness.
A careful clinical history and examination can
easily differentiate radiculoplexus neuropathies
from other diabetic neuropathies; however, confir-
matory tests are necessary to confirm the diagnosis
and exclude central lesions. Blood tests may reveal
a mild abnormality in inflammatory markers, but a
significant increase may suggest alternative pathol-
ogy. Nerve conduction studies and electromyelo-
gram abnormalities can localize the denervation.
MRI is necessary to exclude spinal cord pathology
(e.g., cord/nerve root compression, malignancy or
demyelination) and full body cross-sectional imag-
ing may be considered to exclude occult malig-
nancy. If performed, the cerebro-spinal fluid protein
level may be significantly elevated.
5 Conclusions
The diabetic neuropathies are a heterogenous
group of disorders which can lead to great mor-
bidity, and even mortality. DPN, the commonest
neuropathic disorder in diabetes, may have a sur-
prisingly varied clinical presentation from patient
to patient. A detailed history and clinical exami-
nation with a few simple blood tests is normally
sufficient to diagnose this condition. However,
although the clinical features of other neuropathic
syndromes affecting the cranial/peripheral nerves
may be easily delineated with careful history and
examination, more detailed investigations, such
as cerebral/spinal imaging and/or nerve conduc-
tion studies are often necessary to exclude more
sinister pathology. Finally, although autonomic
neuropathy is a common occurrence it generally
only presents clinically with advanced disease.
These conditions are difficult to diagnose, requir-
48 G. Sloan et al.

ing exclusion of other causes and detailed investi- characteristics and diagnostic issues. J Diabetes
Investig. 2019;10:1148–57.
gations with specialist input. 12. Truini A, Spallone V, Morganti R, Tamburin S, Zanette
G, Schenone A, De Michelis C, Tugnoli V, Simioni V,
Manganelli F, Dubbioso R, Lauria G, Lombardi R,
References Jann S, De Toni FL, Tesfaye S, Fiorelli M, Spagnoli A,
Cruccu G.  A cross-sectional study investigating fre-
quency and features of definitely diagnosed diabetic
1. Tesfaye S, Boulton AJ, Dyck PJ, Freeman R, Horowitz
painful polyneuropathy. Pain. 2018;159:2658–66.
M, Kempler P, Lauria G, Malik RA, Spallone V,
13. Reiber GE, Vileikyte L, Boyko EJ, del Aguila M,
Vinik A, Bernardi L, Valensi P.  Diabetic neuropa-
Smith DG, Lavery LA, Boulton AJ.  Causal path-
thies: update on definitions, diagnostic criteria, esti-
ways for incident lower-extremity ulcers in patients
mation of severity, and treatments. Diabetes Care.
with diabetes from two settings. Diabetes Care.
2010;33:2285–93.
1999;22:157–62.
2. Ziegler D, Papanas N, Vinik AI, Shaw
14. Shillo P, Sloan G, Greig M, Hunt L, Selvarajah D,
JE. Epidemiology of polyneuropathy in diabetes and
Elliott J, Gandhi R, Wilkinson ID, Tesfaye S. Painful
prediabetes. Handb Clin Neurol. 2014;126:3–22.
and painless diabetic neuropathies: what is the differ-
3. Andersen ST, Witte DR, Dalsgaard EM, Andersen
ence? Curr Diab Rep. 2019;19:32.
H, Nawroth P, Fleming T, Jensen TM, Finnerup NB,
15. Themistocleous AC, Ramirez JD, Shillo PR, Lees
Jensen TS, Lauritzen T, Feldman EL, Callaghan
JG, Selvarajah D, Orengo C, Tesfaye S, Rice AS,
BC, Charles M.  Risk factors for incident diabetic
Bennett DL. The pain in neuropathy study (PiNS): a
polyneuropathy in a cohort with screen-detected
­
cross-sectional observational study determining the
type 2 diabetes followed for 13 years: ADDITION-
somatosensory phenotype of painful and painless dia-
Denmark. Diabetes Care. 2018;41:1068–75.
betic neuropathy. Pain. 2016;157:1132–45.
4. Tesfaye S, Chaturvedi N, Eaton SE, Ward JD,
16. Ward JD.  The diabetic leg. Diabetologia.
Manes C, Ionescu-Tirgoviste C, Witte DR, Fuller
1982;22:141–7.
JH. Vascular risk factors and diabetic neuropathy. N
17. Galer BS, Gianas A, Jensen MP. Painful diabetic poly-
Engl J Med. 2005;352:341–50.
neuropathy: epidemiology, pain description, and qual-
5. Pop-Busui R, Boulton AJ, Feldman EL, Bril
ity of life. Diabetes Res Clin Pract. 2000;47:123–8.
V, Freeman R, Malik RA, Sosenko JM, Ziegler
18. Gylfadottir SS, Christensen DH, Nicolaisen SK,
D.  Diabetic neuropathy: a position statement by
Andersen H, Callaghan BC, Itani M, Khan KS,
the American Diabetes Association. Diabetes Care.
Kristensen AG, Nielsen JS, Sindrup SH, Andersen
2017;40:136–54.
NT, Jensen TS, Thomsen RW, Finnerup NB. Diabetic
6. Rota E, Zavaroni D, Parietti L, Iafelice I, De Mitri P,
polyneuropathy and pain, prevalence, and patient
Terlizzi E, Morelli N, Immovilli P, Guidetti D. Ulnar
characteristics: a cross-sectional questionnaire study
entrapment neuropathy in patients with type 2 diabe-
of 5,514 patients with recently diagnosed type 2 dia-
tes mellitus: an electrodiagnostic study. Diabetes Res
betes. Pain. 2020;161:574–83.
Clin Pract. 2014;104:73–8.
19. Sadosky A, Mardekian J, Parsons B, Hopps M,
7. Dyck PJ, Kratz KM, Karnes JL, Litchy WJ, Klein R,
Bienen EJ, Markman J.  Healthcare utilization and
Pach JM, Wilson DM, O’Brien PC, Melton LJ.  The
costs in diabetes relative to the clinical spectrum of
prevalence by staged severity of various types of dia-
painful diabetic peripheral neuropathy. J Diabetes
betic neuropathy, retinopathy, and nephropathy in
Complications. 2015;29:212–7.
a population-based cohort: the Rochester Diabetic
20. Gore M, Brandenburg NA, Dukes E, Hoffman DL, Tai
Neuropathy Study. Neurology. 1993;43:817–24.
KS, Stacey B. Pain severity in diabetic peripheral neu-
8. Greco D, Gambina F, Pisciotta M, Abrignani M,
ropathy is associated with patient functioning, symp-
Maggio F.  Clinical characteristics and associated
tom levels of anxiety and depression, and sleep. J Pain
comorbidities in diabetic patients with cranial nerve
Symptom Manage. 2005;30:374–85.
palsies. J Endocrinol Invest. 2012;35:146–9.
21. Papanas N, Ziegler D. Risk factors and comorbidities
9. Ng PS, Dyck PJ, Laughlin RS, Thapa P, Pinto MV,
in diabetic neuropathy: an update 2015. Rev Diabet
Dyck PJB.  Lumbosacral radiculoplexus neuropathy:
Stud. 2015;12:48–62.
incidence and the association with diabetes mellitus.
22. Dietrich I, Braga GA, de Melo FG, da Costa Silva
Neurology. 2019;92:1188–94.
AC.  The diabetic foot as a proxy for cardiovascular
10. Rajabally YA, Stettner M, Kieseier BC, Hartung HP,
events and mortality review. Curr Atheroscler Rep.
Malik RA.  CIDP and other inflammatory neuropa-
2017;19:44.
thies in diabetes  - diagnosis and management. Nat
23. Gibbons CH, Freeman R. Treatment-induced neurop-
Rev Neurol. 2017;13:599–611.
athy of diabetes: an acute, iatrogenic complication of
11. Gylfadottir SS, Weeracharoenkul D, Andersen ST,
diabetes. Brain. 2015;138:43–52.
Niruthisard S, Suwanwalaikorn S, Jensen TS. Painful
24. Gibbons CH. Treatment induced neuropathy of diabe-
and non-painful diabetic polyneuropathy: Clinical
tes. Auton Neurosci. 2020;226:102668.
Clinical Features of Diabetes Neuropathies
25. Gibbons CH.  Treatment induced neuropathy of
diabetes-­long term implications in type 1 diabetes. J
Diabetes Complications. 2017;31:715–20.
26. Spallone V. Update on the impact, diagnosis and man-
agement of cardiovascular autonomic neuropathy in
diabetes: what is defined, what is new, and what is
unmet. Diabetes Metab J. 2019;43:3–30.
27. Vinik AI, Maser RE, Mitchell BD, Freeman
R.  Diabetic autonomic neuropathy. Diabetes Care.
2003;26:1553–79.
28. Vinik A, Mehrabyan A, Colen L, Boulton A.  Focal
entrapment neuropathies in diabetes. Diabetes Care.
2004;27:1783–8.
29. Raff MC, Sangalang V, Asbury AK. Ischemic mono-
neuropathy multiplex associated with diabetes melli-
tus. Arch Neurol. 1968;18:487–99.
49
30. Kelkar P, Parry GJ. Mononeuritis multiplex in diabetes
mellitus: evidence for underlying immune pathogen-
esis. J Neurol Neurosurg Psychiatry. 2003;74:803–6.
31. Rota E, Morelli N. Entrapment neuropathies in diabe-
tes mellitus. World J Diabetes. 2016;7:342–53.
32. Laughlin RS, Dyck PJ. Diabetic radiculoplexus neu-
ropathies. Handb Clin Neurol. 2014;126:45–52.
33. Dyck PJ, Windebank AJ.  Diabetic and nondiabetic
lumbosacral radiculoplexus neuropathies: new
insights into pathophysiology and treatment. Muscle
Nerve. 2002;25:477–91.
34. Massie R, Mauermann ML, Staff NP, Amrami KK,
Mandrekar JN, Dyck PJ, Klein CJ.  Diabetic cervi-
cal radiculoplexus neuropathy: a distinct syndrome
expanding the spectrum of diabetic radiculoplexus
neuropathies. Brain. 2012;135:3074–88.
 Neuropathy in Type 1 and Type 2
Diabetes

Gulcin Akinci, Dustin Nowacek,
and Brian Callaghan

1 Introduction have other metabolic syndrome components,

Diabetic peripheral neuropathy (DPN) is charac-
terized by injury to peripheral somatic and/or
autonomic nerves [1]. DPN is also one of the
most common chronic complications of diabetes
affecting up to 50% of people living with diabe-
tes (Table 1) [2, 3]. In addition to a high preva-
lence, DPN causes substantial morbidity.
Specifically, patients with DPN often suffer from
severe pain, reduced quality of life, and frequent
falls [4, 5]. Diabetic foot ulcers and amputation
are also devastating complications of this com-
mon condition. While type 1 diabetes and type 2
diabetes can both lead to DPN, these two condi-
tions differ in substantial ways. Type 1 diabetes is
caused by a lack of insulin, whereas type 2 diabe-
tes is caused by several metabolic abnormalities.
In addition, patients with type 2 diabetes often
whereas type 1 diabetes does not cluster as fre-
quently with these comorbidities. Furthermore,
type 1 diabetes often starts very early in life,
whereas the prevalence of type 2 diabetes
increases substantially as age increases. Given
these major difference between type 1 and type 2
diabetes, it is not surprising that many differences
are also seen in the neuropathy that results from
these two conditions.
Among DPN, distal symmetrical polyneurop-
athy (DSP) is the most common pattern of nerve
injury, thus DSP will be the main focus of this
chapter, but we will also discuss other patterns of
nerve injury (Fig.  1) [6]. Furthermore, we will
cover clinical features, risk factors, and clinical
trials in diabetic neuropathy. We will also high-
light the differences between neuropathy in those
with type 1 and type 2 diabetes.

G. Akinci
Department of Neurology, University of Michigan,
Ann Arbor, MI, USA
Department of Pediatrics, Division of Pediatric
Neurology, Koc University, Istanbul, Turkey
D. Nowacek · B. Callaghan (*)
Department of Neurology, University of Michigan,
Ann Arbor, MI, USA
e-mail: bcallagh@med.umich.edu

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 51

S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_4
52 G. Akinci et al.

Table 1  The incidence of polyneuropathies, mononeuropathies, and radiculopathies

Polyneuropathy Population studied Incidence per 100,000 person years
All causes Netherlands [7] 77.0
Type 1 diabetes USA [8] 2800
Type 2 diabetes USA [9] 6100
Mononeuropathies Population studied Incidence per 100,000 person years
Median neuropathy at the Wrist UK [10, 11] 103 (men 87.8, women 192.8)
(carpal tunnel syndrome) USA [12] 99
Ulnar neuropathy UK [11] Men 25.2, women 18.9
Italy (Siena) [13] 24.7 (men 32.7, women 17.2)
Lateral femoral cutaneous UK [11] Men 10.7, women 13.2
neuropathy (meralgia paresthetica) Netherlands [14] 43
Radial neuropathy UK [11] Men 2.97, women 1.42
Idiopathic facial neuropathy (bell’s UK [15] 20.2
palsy) USA (Rochester, MN) [16] 25 (men 22.8, women 26.9)
Italy (Rome) [17] 53.3
Radiculopathies Population studied Incidence per 100,000 person years
Lumbar USA Military [18] 486 (1,079 in patients over 40)
Cervical USA Military [19] 179 (616 in patients over 40)
USA (Rochester, MN) [20] 83.2 (202.9 in patients 50–54; men 107.3,
women 63.5)
USA United States of America, UK United Kingdom

a c

b d

Fig. 1  Patterns of nerve injury in diabetic neuropathy. (a) distal symmetric polyneuropathy; (b) diabetic radiculopathy
and radiculoplexus neuropathy; (c) multiple mononeuropathies; (d) autonomic neuropathy
Neuropathy in Type 1 and Type 2 Diabetes 53

2 Clinical Features of DPN tective sensation, and progressive numbness.

Other symptoms of late-stage DSP include distal
DPN can involve multiple patterns of nerve motor nerve involvement resulting in muscle
injury; therefore, symptoms can vary signifi- weakness, unsteadiness, imbalance and gait
cantly. These wide ranging neuropathic symp- problems, and impaired proprioception [29].
toms have been recognized in patients with DSP has a similar prevalence in type 1 and
diabetes for more than a hundred years and the type 2 diabetes [30]. Despite similar clinical fea-
earliest classifications of DPN divided patients tures in DSP in those with type 1 and type 2 dia-
into three groups: hyperesthetic, paralytic, and betes, the natural history among subjects with
ataxic forms of neuropathy [21]. The most recent these two major types of diabetes differs signifi-
Position Statement of American Diabetes cantly highlighting the fundamental differences
Association classifies neuropathies based on in disease pathophysiology [31]. Type 1 diabetes
which parts of peripheral nervous system are is caused by absolute insulin deficiency; how-
affected and includes diffuse neuropathies (DSP ever, the pathophysiology of type 2 diabetes is
and autonomic neuropathies), mononeuropa- multifactorial and quite complex involving a con-
thies, and radiculopathies [6]. Most studies have stellation of metabolic insults including insulin
focused on DSP and cardiovascular autonomic resistance [32]. Many clinical presentations of
neuropathy (CAN) because they are the most diabetic neuropathy overlap in patients with type
common manifestations. 1 and type 2 diabetes. However, a few observa-
DSP, a chronic, symmetrical, length-­tional studies give us insight into clinical differ-
dependent sensorimotor polyneuropathy, is the ences in neuropathy presentations between these
most common pattern of DPN [3]. DSP is a sen- major types of diabetes. Symptomatic DSP typi-
sorimotor neuropathy in which sensory ­symptoms cally develops in patients with type 1 diabetes
are predominant at the early stages of the disease after a long disease duration and poorly con-
[6]. DSP is a mixed polyneuropathy that usually trolled hyperglycemia. On the other hand, this
starts with small fiber neuropathy and progresses presentation may be less common in patients
to large fiber neuropathy [22, 23]. Small fibers with type 2 diabetes. Patients with type 2 diabetes
involve predominantly small diameter somatic can develop neuropathy even within a few years
and autonomic nerves, unmyelinated C fibers, of their diabetes diagnosis and some patients may
and/or thinly myelinated A-delta sensory fibers. even exhibit signs and symptoms at the time of
They carry information regarding pain and tem- diabetes diagnosis [33]. Even prior to diabetes
perature and also innervate the autonomic ner- onset, patients with prediabetes may present with
vous system. Large fibers are myelinated fibers painful DSP as the first symptom of carbohydrate
that relay vibratory and proprioceptive informa- intolerance [34, 35].
tion. Symptoms are varied by the class of nerve Multiple studies have investigated differences
fiber affected. The most common early symptoms in the natural history of DSP between those with
related to small fiber involvement include pain type 1 and type 2 diabetes. The Oslo study
and dysesthesia [24]. About one-­third of patients showed that small fiber neuropathy dominates the
with DPN may experience neuropathic pain [25] clinical findings of patients with longstanding
that worsens at rest and may show some relief type 1 diabetes and that the nerve injury is cor-
while walking [26]. Pain can be spontaneous or related with glycemic control [36]. In another
triggered by external stimuli (allodynia and study, subjects with latent autoimmune diabetes
hyperalgesia). Painful neuropathy can be dis- of adulthood (LADA), who had worse glycemic
abling and may persist for years [27]. On some control, were found to have greater small fiber
occasions, pain may diminish after sensory loss involvement than subjects with type 2 diabetes
worsens [28]. As DSP progresses, large nerve despite showing a more favorable BMI and lipid
fibers are affected, resulting in additional symp- profile [37]. Although the frequency and clinical
toms such as tingling without pain, loss of pro- presentation of neuropathy were similar in sub-
54
jects with type 1 and type 2 diabetes (described
as IDDM and NIDDM at that time) in the
Rochester Diabetic Neuropathy Study [38], neu-
ropathy was significantly more severe in type 1
diabetes as staged by the investigators systemati-
cally. However, it should be noted that patients
with type 1 diabetes had a longer duration of dia-
betes, which may account for at least some of this
difference. Loseth et al. [39], on the other hand,
reported that small fiber neuropathy progressed
more rapidly in patients with type 2 diabetes
compared to those with type 1 diabetes after 5
years of follow-up, even though both groups had
similar neuropathy prevalence in an earlier cross-­
sectional study of the same cohort [40]. Also, a
large observational study from England detected
painful neuropathic symptoms more commonly
in subjects with type 2 diabetes than type 1 diabe-
tes, highlighting the presence of painful neuropa-
thy symptoms in relatively early years of diabetes
in type 2 diabetes [25]. Although differences in
fiber type involvement, severity, and pain have
been demonstrated between type 1 and type 2
diabetes, the similarities far outweigh the
differences.
The definition of diabetic autonomic neuropa-
thy (DAN) covers several different clinical pre-
sentations ranging from cardiovascular autonomic
neuropathy (CAN) to diabetic gastroparesis,
hypoglycemia unawareness, and sweating dys-
regulation [6, 41]. Besides a variety of clinical
presentations, the severity of DAN is also hetero-
geneous. Among different presentations of DAN,
CAN is one of the most serious complication and
relatively more comprehensively studied in the
literature. CAN has been shown to be an indepen-
dent risk factor for cardiovascular mortality in
type 1 and type 2 diabetes [6, 42, 43]. Although
the exact prevalence of CAN is still unknown, a
previous study showed that roughly one-fourth of
patients with type 1 and one-third of subjects
with type 2 had signs of CAN when screened
[44]. Early stages of CAN are usually asymptom-
atic but can be still detected by simple screening
tests such as heart rate variability (HRV) tests.
Studies have shown that HRV tests can detect
abnormalities within 2 years in patients with type
1 diabetes and earlier (even at prediabetic stage)
G. Akinci et al.
in subjects with type 2 diabetes [44–46].
Nonspecific symptoms such as fatigue, general
weakness, and dizziness may be present. The
prevalence of CAN reaches up to 50% after 15
years of type 2 diabetes [6]. Although CAN is
unusual in newly diagnosed patients with type 1
diabetes, studies have shown that it develops in
up to 30% of them after 20 years of diagnosis of
diabetes [47, 48].
Gastrointestinal presentations of DAN appear
to be more common in patients with type 1 diabe-
tes but can also be observed in patients with long-
standing type 2 diabetes [49, 50]. These
presentations are typically observed in patients
with poorly controlled glycemia and in the pres-
ence of other microvascular complications of dia-
betes. Diabetic gastroparesis causes delayed
gastric emptying that can trigger brittle diabetes
that in turn may eventually result in poorer glyce-
mic control. In addition, gastroparesis often leads
to symptoms such as nausea, vomiting, early sati-
ety, and abdominal pain. Gastrointestinal dys-
motility is not limited to the upper gastrointestinal
system in diabetes and can affect small bowels
causing constipation and episodes of diarrhea
[51]. Similar to gastroparesis, studies suggest
that diabetic diarrhea may be more common in
patients with type 1 diabetes [52, 53]. Several
conditions should be considered in diabetic
patients presenting with episodes of diarrhea that
include, but are not limited to, bacterial over-
growth, exocrine pancreatic deficiency, coexist-
ing Celiac disease, and metformin use [52, 53].
DAN can also affect genitourinary system.
Clinical presentations may vary from bladder
dysfunction to sexual dysfunction [54]. Erectile
dysfunction (ED) is a relatively common prob-
lem that can develop in males with both type 1
and type 2 diabetes and is associated with a
higher incidence of cardiovascular disease [55,
56]. Incidence of ED is higher in diabetic males
with metabolic syndrome [57]. Diabetic bladder
dysfunction, on the other hand, more commonly
develops in patients with type 1 diabetes [54].
Patients with diabetic bladder may not sense a
full bladder due to afferent nerve involvement or
may suffer from incomplete emptying as a result
of efferent nerve involvement. Clinical conse-
Neuropathy in Type 1 and Type 2 Diabetes
quences may include urinary incontinence, poor
urinary stream, dribbling, and recurrent urinary
tract infections. Urinary incontinence due to
DAN is classically associated with type 1 diabe-
tes and is relatively less common among patients
with type 2 diabetes [58, 54]. Analogous to DSP,
differences in DAN between patients with type 1
and type 2 diabetes exist, but the similarities out-
weigh these differences.
Acute painful neuropathy of poor glycemic
control is a relatively rare clinical presentation
that develops in subjects with poorly controlled
diabetes and is usually associated with severe
weight loss [59]. It is not clear if weight loss is
due to insulin deficiency but neuropathic pain
improves after weight gain following insulin
treatment. Typical patients are poorly controlled
males with type 2 diabetes who are on oral anti-
diabetics (or newly diagnosed) but it can develop
in patients with type 1 diabetes as well. Pain is
usually severe and often leads to depressive
symptoms. Another form of acute painful neu-
ropathy can occur after rapid glycemic control.
This phenomenon had probably been underre-
ported as recent observations suggest a higher
prevalence than previously thought. Recently, the
term “treatment-induced neuropathy of diabetes”
(TIND) has been suggested to describe this con-
dition [60]. Rapid correction of glycemic control
can trigger acute/subacute painful neuropathy
that more commonly happens in poorly con-
trolled patients with type 1 diabetes but it can
also be observed in patients with type 2 diabetes
[61]. Several interesting recent observations
regarding TIND include higher prevalence of eat-
ing disorders among individuals with type 1 dia-
betes that developed TIND and the occurrence of
symptoms mostly in older subjects with type 2
diabetes [60]. The key is to avoid rapid drops in
hyperglycemia and target more gradual glycemic
control in patients experiencing TIND.
Patients with diabetes may present with focal
and multifocal neuropathies such as diabetic
amyotrophy, cranial neuropathy, and limb and
truncal neuropathies, although these clinical pre-
sentations are far less common than DSPs. These
neuropathies are typically observed predomi-
nantly in older patients with type 2 diabetes.
55
Diabetic amyotrophy, a disabling proximal neu-
ropathy affecting lower extremities, can develop
in both type 1 and type 2 diabetes with a rela-
tively higher prevalence in type 2 diabetes [62].
The pathophysiology of diabetic amyotrophy is
not well understood; however, relatively slower
progression after rapid onset of disease suggests
both metabolic and vascular involvement [63].
Likewise, mononeuropathies and other radicu-
lopathies are more commonly reported in type 2
diabetes. Although rare, cranial mononeuropa-
thies (especially oculomotor nerve palsy) can be
observed in patients with diabetes. These patients
are typically relatively older individuals with
longstanding poorly controlled type 2 diabetes
with other complications of diabetes; however,
oculomotor nerve palsy has been reported in type
1 diabetes as well [64]. Typically, diabetes affects
the central portion of the oculomotor nerve with-
out other cranial nerve involvements, causing
deviation of the eye with normal pupil reactions
[65]. However, other causes of oculomotor nerve
palsy such as intracranial tumors, aneurysms, and
brainstem strokes should be considered in differ-
ential diagnosis [66]. Finally, entrapment neu-
ropathies are quite common in people with
diabetes, median neuropathy being the most
common presentation [67].
3 Risk Factors for DPN
The pathogenesis of DPN is complex with a wide
range of risk factors being involved (Fig.  2).
Independent of the patient’s age, diabetes dura-
tion is a risk factor in both types of diabetes type
[68–71]. In addition to traditional risk factors
such as age and disease duration, hyperglycemia
is the most widely studied risk factor; however,
current evidence suggests novel metabolic risk
factors are important in the pathogenesis of DPN
[72].
Current evidence supports a closer relation-
ship between glucose control and DPN in type 1
diabetes than type 2 diabetes [73]. Hyperglycemia
is a robust risk factor for DPN in type 1 diabetes,
and current literature supports the role of good
glycemic control that would reduce the incidence
56
Fig. 2  Signaling pathways underlying nutrient excess and metabolic neuropathy
of DPN [47, 74, 75]. Also, the occurrence and
progression of DPN are more predictable in type
1 diabetes. DPN typically becomes symptomatic
after years of chronic poor glycemic control and
progresses unless glycemic control is achieved
[76].
Previous studies found a significant benefit of
intensive glycemic control to improve DPN
related outcomes in patients with type 1 diabetes.
The landmark Diabetes Control and Complications
Trial (DCCT) study reported that the risk of DPN
and CAN decreased by 64% and 45% after inten-
sive glycemic control [66]. This benefit remained
significant in the observational follow-up of this
cohort [47, 77]. Glycemic improvement was also
associated with lower DPN incidence in the
Pittsburgh Epidemiology of Diabetes
Complications (EDC) study [78]. Similar trends
were observed in the European Diabetes
Prospective Complications (EURODIAB) study
[79] cohort, and the recent SEARCH study con-
firmed this association in youth with type 1 diabe-
tes [80]. In parallel to these evidence, our previous
meta-analysis showed that optimized glycemic
control in type 1 diabetes reduces the develop-
ment of clinical DPN [73].
G. Akinci et al.
On the other hand, there is a weaker correla-
tion between DPN and glycemic status in type 2
diabetes. Type 2 diabetes is usually complicated
with multiple comorbidities that might affect
DPN outcomes and have attenuated the benefit of
glycemic control. Most patients with type 2 dia-
betes are overweight or obese, have dyslipidemia,
hypertension, and a high cardiometabolic risk
profile. In addition, contrary to observations in
type 1 diabetes, DPN can be relatively common
among subjects with newly diagnosed diabetes
[68]. Patients may be diagnosed with DPN at the
same time when they are diagnosed with diabetes
[81]. Signs of DPN can be detected in prediabetic
conditions (in the absence of overt diabetes) such
as impaired fasting glucose/ impaired glucose
tolerance [82–84], obesity [85–90], and meta-
bolic syndrome [85, 91].
The benefit of glycemic control in type 2 dia-
betes is less conclusive. Although benefits of bet-
ter glycemic control have been reported in several
studies in type 2 diabetes [92, 93], several others
have failed to show the benefit of glycemic con-
trol on DPN outcomes [94–97]. Our previous
meta-analysis failed to show a significant benefit
of glycemic control on DPN, although the short-
Neuropathy in Type 1 and Type 2 Diabetes
age of primary outcome studies in type 2 diabetes
constitutes a limitation [73]. Several studies have
also reported an association between glycemic
variability and neuropathy in type 1 and type 2
diabetes [98–100]; however, this association was
not observed in the DCCT analyses [98, 101].
Also, further small-scale studies failed to confirm
this association [102].
Obesity is a major risk factor for type 2 diabe-
tes that plays a fundamental role in the disease
pathogenesis. Obesity contributes to a higher risk
of diabetes complications. The association
between obesity and typical conditions that are
commonly associated with diabetes such as car-
diovascular diseases or nephropathy is robust,
even in the absence of diabetes [103]. Multiple
studies have demonstrated that obesity is consis-
tently associated with neuropathy [85, 88–90,
104]. Based on these studies, obesity has been
proposed as an independent risk factor for dia-
betic neuropathy [105]. Our previous studies
have shown a higher prevalence of DPN in obese
subjects with normoglycemia than in lean con-
trols in two different cohorts [86, 88]. We also
found that CAN, along with DPN, was a common
complication in patients with severe obesity [86].
Higher BMI was associated with reduced HRV
and parasympathetic loss and sympathetic over-
drive in adolescents and youth with type 2 diabe-
tes in the TODAY [106]. Our work and others
also revealed the role of central obesity in DPN
[86, 107]. Therefore, obesity has emerged as the
second most important metabolic risk factor for
neuropathy.
Although obesity is usually attributed to type
2 diabetes, it is being increasingly reported in
type 1 diabetes populations as well [108]. Similar
to type 2 diabetes, several reports indicate asso-
ciations between metabolic abnormalities that
link to obesity and DPN among subjects with
type 1 diabetes. Indexes of obesity have been
reported to be associated with diabetic neuropa-
thy in type 1 diabetes in several studies [79, 80,
109].
Metabolic syndrome is a cluster of conditions
that has been closely associated with insulin
resistance and the risk of type 2 diabetes. The
prevalence of metabolic syndrome is much more
57
common among subjects with type 2 diabetes
than those with type 1 diabetes. Recent studies
found a significant relationship between compo-
nents of metabolic syndrome and DPN [85, 86,
88, 89, 104, 110, 111]. Besides obesity, emerging
data suggest a possible association between other
components of metabolic syndrome and DPN as
well, although this association seems to be less
consistent [72]. In the general population, the
likelihood of DPN is increased in subjects with
metabolic syndrome [112]. DPN prevalence was
also reported to increase in parallel to the increas-
ing number of metabolic syndrome components
[85, 113]. Insulin resistance, the core shared
pathophysiology in type 2 diabetes and metabolic
syndrome, has been identified as a major inde-
pendent risk factor of DPN [114].
Dyslipidemia appears to be a contributing
factor to DPN [72]. Many patients with type 2
diabetes exhibit elevated triglycerides to some
degree. A recent meta-analysis reported a higher
risk of DPN among people with increasing lev-
els of triglycerides [115]. The Utah Diabetic
Neuropathy Study reported that hypertriglyceri-
demia increased the risk of DPN in patients with
type 2 diabetes and triglyceride levels were
associated with loss of small unmyelinated
axons [113]. Hypertriglyceridemia was identi-
fied as an independent risk factor for lower-
extremity amputations in diabetic patients in the
10-year follow-up of the DISTANCE study
cohort [116]. Adolescents and young adults
with type 2 diabetes and CAN had significantly
higher triglyceride levels in the SEARCH study
[45]. A similar association was also reported in
patients with type 2 diabetes from the Anglo-
Danish-Dutch study of Intensive Treatment of
Diabetes in Primary Care (ADDITION)-
Denmark study [117]. The ADDITION study
reported lower HDL cholesterol levels as a risk
factor for diabetic neuropathy [104]. An inde-
pendent significant correlation between low
HDL cholesterol and higher diabetic DPN risk
was also reported in T2D youths within the
SEARCH cohort [80].
Dyslipidemia has previously been proposed as
a risk factor for diabetic neuropathy in subjects
with type 1 diabetes as well [69, 75, 80]. In paral-
58 G. Akinci et al.

lel to these studies, elevated triglycerides were 4 Clinical Trials in DPN

found to be a significant risk factor for DPN in
the recently published report of long-term fol-
low-­up (>23 years) of the DCCT-EDIC cohort
[71]. The study also found a strong independent
association between hypertension and DPN, sup-
porting previously published results from the
Pittsburgh EDC [75], EURODIAB [79], and
SEARCH studies [80].
Finally, hypertension, a major components
of metabolic syndrome, has been previously
found to be associated with diabetic neuropathy
[45]. Although a link between hypertension and
DPN has been reported in patients with type 2
diabetes [118, 119], the association of hyper-
tension with DPN has been more clearly shown
among subjects with type 1 diabetes in multiple
studies [71, 75, 79, 8, 120]. While hyperglyce-
mia remains the most important metabolic risk
factor for DPN, all of the other metabolic syn-
drome components are associated with
DPN.  While the patients with type 2 diabetes
are more likely to have these comorbidities,
they are also associated with DPN in those with
type 1 diabetes.
Several clinical trials have assessed the effect of
glucose control on the development of polyneu-
ropathy in patients with diabetes mellitus. A 2012
Cochrane review identified these studies and ana-
lyzed them in a meta-analysis (Table  2). In this
review, the primary outcome was incidence of
clinical neuropathy in patients without clinical
neuropathy at baseline. Secondary outcomes
comprised of changes in neuropathic symptoms,
nerve conduction studies (NCS) and quantitative
sensory testing, and adverse events. Adverse
events consisted of foot ulcers, amputations, and
hypoglycemic episodes requiring hospitalization.
Outcomes measured after less than 12 months
were not included. Seven clinical trials evaluated
people with type 1 diabetes, eight clinical trials
included people with type 2 diabetes mellitus, and
two trials included patients with both type 1 and
type 2 diabetes. In the trials that studied exclu-
sively type 1 diabetes, most compared different
insulin regimens, and the two studies that included
both types of diabetes also compared different
insulin regimens. Of the eight studies that evalu-
ated type 2 diabetes, three evaluated different

Table 2  Clinical trials investigating the role of enhanced glucose control on neuropathy
Trial Length of study Clinical Other Enhanced glycemic control
Investigator size (years) outcome outcomes superior?
Type 1 diabetes
Holman (1983) 74 2 No QST Yes
Lauritzen (1985) 30 2 No QST No
Dahl-Jorgensen 45 2 No NCS Yes
(1986)
Jakobsen (1988) 24 2 No QST Yes
DCCT (1993) 1441 5 Yes NCS Yes
Reichard (1993) 102 7.5 No NCS, QST Yes
Linn (1996) 49 5 Yes None Yes
Type 2 diabetes
Kawamori (1991) 50 4 No NCS Yes
Ohkubo (1995) 110 6 No NCS, QST Yes
UKPDS (1998) 3867 10 No QST Yes
Tovi (1998) 38 1 Yes None No
Azad (1999) 153 2 Yes None No
Shichiri (2000) 110 8 No NCS, QST Yes
Gaede (2003) 160 8 No QST No
Duckworth (2009) 1791 5.6 Yes None No
ACCORD (2010) 10,251 3.7 Yes None No
Neuropathy in Type 1 and Type 2 Diabetes
insulin regimens and five evaluated other treat-
ment strategies, including more aggressive glyce-
mic goals through the use of diet and exercise,
oral hypoglycemic agents, insulin, or oral hypo-
glycemic agents plus insulin. The primary out-
come was reported in two of the studies involving
type 1 diabetes [118–121], and in four of the stud-
ies involving type 2 diabetes [93, 95, 122, 123].
The largest study in type 1 diabetes was the
DCCT [118, 119], which randomized 1441 par-
ticipants. Primary prevention participants (N =
726) were defined as having insulin dependent
diabetes mellitus of 1–5 years’ duration, no
detectable retinopathy on stereo fundus photog-
raphy, and urinary albumin excretion less than 40
mg/24 h. Secondary prevention participants (N =
715) were defined as having insulin dependent
diabetes mellitus of 1–15 years’ duration, very
mild to moderate non-proliferative retinopathy,
and urinary albumin excretion less than 200
mg/24 h. Almost all participants (N = 1436) had a
baseline neuropathy assessment). Those in the
intensive group received either an external insu-
lin pump, or 3 or more daily insulin injections
guided by frequent blood glucose monitoring.
Participants in the conventional group received 1
or 2 daily insulin injections. The goals of the
intensive group included a pre-prandial blood
glucose between 70 and 120 mg/dL, postprandial
levels less than 180 mg/dL, a weekly 3 AM mea-
surement greater than 65 mg/dL, and a monthly
HgbA1c less than 6.05%. Definite neuropathy
was defined as the presence of two or more of the
following: symptoms, sensory examination find-
ings, and decreased or absent reflexes, and hav-
ing one of these abnormalities was defined as
“possible” neuropathy, and “confirmed” definite
clinical neuropathy also required the finding of
an unequivocal abnormality on nerve conduction
studies or autonomic nervous system testing.
NCS were performed at baseline, at 5 years and
at study end. Participants were followed for a
mean of 6.5 years. In the primary prevention
cohort there was 1.53% per year risk reduction
(95% CI 0.51–2.54) and a relative risk reduction
after 5 years of 53%. In the secondary prevention
cohort there was a 1.97% per year risk reduction
(95% CI 0.90–3.04) and a 52% relative risk
59
reduction after 5 years. In the intensive treatment
groups in both groups combined there was a risk
reduction of 1.74% per year (95% CI 1.00–2.48)
[73, 118, 119].
The other study reporting the primary out-
come of incident neuropathy was Linn et  al.,
which randomized 49 consecutive participants
with newly diagnosed insulin dependent diabetes
mellitus (IDDM) to intensive or conventional
insulin therapy, and evaluated them 5 years after
their clinical diagnosis. Definite neuropathy was
defined as the presence of three of the following:
symptoms, examination signs, abnormal quanti-
tative sensory testing, and peroneal motor nerve
conduction velocity. Intensive therapy included
administration of insulin at least three times daily
by injection with dosage being adjusted based on
monitoring of blood glucose, dietary intake, and
anticipated exercise. Target blood glucose in the
intensive group was defined as self-determined
capillary glucose less than 6.8 mmol/L and less
than 10 mmol/L postprandially, and a hemoglo-
bin A1c less than 6.5. Conventional therapy con-
sisted of 1 or 2 daily injections of insulin
including mixed intermediate and rapid-acting
insulin and variable self-monitoring of blood glu-
cose. The goals of conventional therapy included
the absence of symptoms attributed to glucosuria
or hyperglycemia, and freedom from severe or
frequent hypoglycemia. After 5 years, 1 partici-
pant in the intensive group and 6 participants in
the conventional group developed neuropathy.
This study was much smaller than the DCCT
study, but notably it revealed a 5.45% per year
risk reduction (95% CI 0.95–9.95) and an 86%
relative risk reduction in clinical neuropathy
[121]. The meta-analysis including the DCCT
and Linn studies revealed a statistically signifi-
cant 1.84% per year risk reduction (95% CI 1.11–
2.56) [73]. Therefore, current evidence supports
a large effect of glycemic control on reducing
neuropathy in patients with type 1 diabetes.
While the above two studies were the only
studies that reported clinical neuropathy as an
outcome in type 1 diabetes in the Cochrane
review, a majority of the five other studies that
evaluated the secondary outcomes described
above were also in favor of the intensive treat-
60
ment groups. Multiple studies demonstrate an
annual mean difference (MD) in conduction
velocity of between 0.4 and 0.6 m/s of different
motor nerves in favor of the intensive group, pro-
viding further evidence of improvement in nerve
function in this group. Four studies in type 1 dia-
betes evaluating quantitative vibration testing
showed improvement, with 3 of the studies
reporting enough information to allow for meta-­
analysis, which showed an annual standardized
mean difference (SMD) of 0.32 (95% CI 0.02–
0.43) in favor of intensely treated participants.
Notably, there was a threefold increase in epi-
sodes of serious hypoglycemia in those receiving
intensive treatment. Despite this important
adverse event, the substantial benefit in not only
neuropathy outcomes but also in other health-­
related outcomes (nephropathy and retinopathy)
in patients receiving intensive glycemic control
strongly suggests that the risk/benefit ratio is in
favor of intensive treatment [73].
In the 2012 Cochrane review, four studies
investigated the primary outcome in type 2 diabe-
tes. The largest study was the Accord 2010,
which randomized 10,251 participants in North
America. This study included patients with
HgbA1c greater than 7.5% and cardiovascular
disease or 2 or more cardiovascular risk factors.
Participants were randomized to an intensive tar-
get (HgbA1c less than 6.0%) or a standard target
(7.0–7.9%) and followed for a median 3.7 years.
Clinical neuropathy was defined as a score of
greater than 2 on the Michigan Neuropathy
Screening Instrument (MNSI). 1277 of 2815 par-
ticipants in the intensive arm and 1338 of 2791 in
conventional arm developed neuropathy. Overall
these findings demonstrated a 0.70% per year
risk reduction (95% CI –0.01 to 1.40) and a non-­
significant 5% relative risk reduction in those
receiving intensive therapy [93].
The second largest study was Duckworth
2009, which investigated 1791 military veterans
who had a suboptimal response to therapy for
type 2 diabetes and randomized them to receive
either intensive or standard glucose control for a
median of 5.6 years. Participants in the intensive
therapy group were started on maximal doses,
whereas those in the standard therapy group were
started on half the maximal doses. Clinical neu-
ropathy in this study was defined as a diagnosis
G. Akinci et al.
on a yearly physical exam, with no other details
being provided regarding what constitutes a diag-
nosis of neuropathy. 178 of the 464 participants
developed neuropathy in the intensive group
compared with 199 of 498 participants in the con-
ventional droop. Ultimately this resulted in a
0.29% per year risk reduction (95% CI –0.82 to
1.39) and a 4% relative risk reduction at a median
follow-up of 5.6 years, but these results did not
meet statistical significance [95].
Two other studies evaluated the primary out-
come of incident neuropathy in patients with type
2 diabetes. Azad 1999 investigated 153 partici-
pants over 2 years randomized to conventional
versus enhanced treatment with insulin, which
lowered the hemoglobin A1c by 2.1% lower than
the standard arm, but there was no significant dif-
ference in the incidence of neuropathy between
the treatment groups [122]. Tovi followed 40
elderly type 2 diabetic participants who had sec-
ondary failure of oral diabetic drug therapy with-
out symptoms of hyperglycemia, and these
participants were randomized to insulin versus
oral hypoglycemic agents and followed for 1 year.
Clinical neuropathy was defined as a composite
score utilizing examination and electrodiagnostic
findings, and ultimately no patient's in either group
developed neuropathy [123]. Of these two studies,
there was no statistical significant difference in
favor of either group [73]. Overall the meta-analy-
sis of these four studies (Accord, Duckworth,
Azad, Tovi) revealed a risk reduction of 0.58% per
year (95% CI –0.01 to 1.17) [73]. While glycemic
control did not demonstrate a statistically signifi-
cant decrease in the incidence of neuropathy in
any of these four studies or in the meta-analysis,
the consistent findings indicate a small, but real
effect of glycemic control. However, in compari-
son to the studies in type 1 diabetes, the effect of
glycemic control is much less. These results high-
light the importance of factors other than hyper-
glycemia in the development of neuropathy in type
2 patients. Furthermore, glycemic control is essen-
tial in patients with type 1 diabetes.
While not the primary outcome of this Cochrane
review, the largest study that reported neuropathy
outcomes in patients with type 2 patients was the
UKPDS Study Group 1998, which followed 3867
individuals with newly diagnosed type 2 diabetes
who were randomly assigned to an intensive arm
Neuropathy in Type 1 and Type 2 Diabetes
with a sulphonylurea (chlorpropamide, gliben-
clamide, or glipizide) or insulin, or to a conven-
tional policy with diet. Neuropathy was defined as
a vibration threshold greater than 25 V on a bioth-
esiometer. Participants were followed for as many
as 15 years. They found that there was a modest
risk reduction in favor of intensive treatment, but
this result was only statistically significant at 15
years [94]. Shichiri et al. [124] investigated nerve
conduction study findings in a randomized trial of
intensive glycemic control compared to conven-
tional teratment in patients with type 2 diabetes. A
total of 110 participants were followed for 8 years
with type 2 diabetes (55 with no retinopathy—pri-
mary prevention cohort, and 55 with simple reti-
nopathy—secondary intervention cohort) and
assigned to intensive insulin therapy (three or
more daily injections) or to conventional insulin
therapy (one to two daily injections). There was an
increase of 0.44 (0.09) m/s in the intensive arm
compared to a decline of 0.13 (0.08) m/s in the
conventional arm with a statistically significant
annual MD of 0.56 m/s (95% CI 0.53–0.60) in
favor of the intensive group [124]. Interestingly,
there was a similar effect in conduction velocity
changes seen in type 1 diabetes, even though there
was a much smaller relative risk reduction of clini-
cal neuropathy in participants with type 2 diabetes
compared to type 1 diabetes. These results support
the importance of glycemic control in patients
with type 2 diabetes even if it is not as important as
it is in type 1 diabetes.
Similar to the trials evaluating patients with
type 1 diabetes, those with type 2 diabetes also
suffered more adverse of events in the intensive
group in the three largest studies with the longest
follow-up. Specifically, there was an approxi-
mately threefold higher risk of severe hypoglyce-
mic episodes in those receiving intensive therapy.
In contrast to participants with type 1 diabetes,
the effect of intensive glycemic control and neu-
ropathy was much less impressive, and in this
group the risk of hypoglycemia still remains sub-
stantial so this risk needs to be taken into account
when determining the risk versus benefit ratio of
enhanced glucose control [73]. Importantly, the
ACCORD study demonstrated an increase in
mortality when targeting a HbA1C of less than
6%, which highlights the importance of tradeoffs
between glycemic control and adverse events,
61
which included increased weight gain (RR 1.96,
95% CI 1.81–2.13) and deaths (RR 1.26, 95% CI
1.06–1.51) in those in the intensive group [93].
Since the 2012 Cochrane review, only one
study has evaluated the effect of glycemic control
on neuropathy and this was in patients with type 2
diabetes. Specifically, the ADDITION (Anglo-­
Danish-­ Dutch Study of Intensive Treatment in
People With Screen Detected Diabetes in Primary
Care)-Europe study evaluated if intensive treat-
ment (IT) improved outcomes compared with
routine care (RC). This trial comprised of two
phases, a screening phase and a randomization
phase involving 343 general practices throughout
Denmark; Cambridge, UK; the Netherlands; and
Leicester, UK. Participants were included if they
were 40–69 years (50–69 years in the Netherlands)
without known diabetes. Screening comprised of
various measures depending on what country the
patient was located in, and included oral glucose
tolerance tests, or random glucose measurements
and HgbA1c followed by fasting glucose and oral
glucose tolerance test. All patients with newly
diagnosed type 2 diabetes mellitus were eligible
to participate and randomized between IT and RC
in a 1:1 ratio. Ultimately, 3057 participants took
part in this trial and 196 participants died before
follow-up. Questionnaire data on peripheral neu-
ropathy was collected in 2312 participants.
Targets for the IT arm were HgbA1c less than
7.0%, blood pressure less than or equal to 135/85,
cholesterol less than 5 mmol/L in patients without
ischemic heart disease or less than 4.5 mmol/L in
patients with ischemic heart disease. Treatments
were recommended but therapy decisions were
ultimately deferred to the prescribing clinicians
and patients. After 5.3 years, 1296 subjects in the
IT arm, and 1016 subjects in the RC arm were
evaluated. Ultimately 4.9% participants in the IT
arm screened positive for peripheral neuropathy
based on the MNSI questionnaire compared to
5.9% of participants in the RC group [125].
Therefore, no significant difference was seen in
decreasing the incidence of neuropathy in this
population of patients with screen detected type 2
diabetes with intensive therapy. These results are
very similar to the previous four studies in patients
with type 2 diabetes described above. As a result,
the 2012 Cochrane review conclusions remain
relevant. Namely that glycemic control is more
62
important in patients with type 1 diabetes com-
pared with patients with type 2 diabetes.
While many studies of glycemic control have
been performed to determine its effect on neuropa-
thy outcomes, few studies have addressed the
effects of weight loss on neuropathy outcomes.
The main study addressing this intervention is the
Look Ahead trial, which randomized 5145 over-
weight or obese people with type 2 diabetes to
intensive lifestyle intervention (ILI) (n = 2570) or
to a diabetes support and education (DSE) control
group (n = 2575). Staff responsible for obtaining
data on study outcomes were blinded, but because
of the types of interventions being evaluated,
patients and interventionists were not blinded.
Interventions were terminated 9–11 years after
randomization, but both groups continued to be
followed for primary and secondary outcomes.
Neuropathy evaluations included the MNSI ques-
tionnaire completed at baseline in all participants
and repeated annually, and the MNSI examination.
The baseline MNSI questionnaire scores were 1.9
± 0.04 and 1.8 ± 0.04 in the ILI and DSE groups,
respectively, which was not statistically signifi-
cant. After 1 year when weight loss was maximal,
the ILI group had a statistically significant lower
MNSI questionnaire score compared to the DSE
group (1.7 ± 0.04 and 2.0 ± 0.04, p < 0.001). There
was also a significant association between changes
in the MNSI questionnaire scores and changes in
body weight, HgbA1c, and serum lipids. However,
there was no significant between-group differ-
ences in the proportion of participants with MNSI
examination scores greater than or equal to 2.5.
Overall, this study demonstrated that ILI resulted
in a significant decrease in questionnaire based,
but not examination based neuropathy. Therefore,
weight loss likely reduces neuropathy in those
patients with type 2 diabetes that are also over-
weight or obese. However, the magnitude of effect
is likely small highlighting the need for novel
treatments to prevent neuropathy.
In conclusion, studies comparing different lev-
els of glycemic control on the development of
neuropathy in patients with diabetes reveal clear
differences between type 1 and type 2 diabetes.
Intensive glycemic control reduces the develop-
ment of neuropathy in patients with type 1 diabe-
tes, which emphasizes the importance of
hyperglycemia in this population. In contrast,
G. Akinci et al.
intensive glycemic control has a much smaller
effect on preventing neuropathy in patients with
type 2 diabetes, which highlights the need for
novel therapies addressing other metabolic factors
in this population. Based on the LOOK Ahead
study, weight loss is a promising intervention, but
future studies are needed to verify this finding.
References
1. American Diabetes Association American Academy
of Neurology. Consensus statement: report and rec-
ommendations of the San Antonio conference on dia-
betic neuropathy. Diabetes Care. 1988;11(7):592–7.
2. Dyck PJ, Albers JW, Andersen H, et al. Diabetic poly-
neuropathies: update on research definition, diagnos-
tic criteria and estimation of severity. Diabetes Metab
Res Rev. 2011;27(7):620–8.
3. Feldman EL, Callaghan BC, Pop-Busui R, et  al.
Diabetic neuropathy. Nat Rev Dis Primers.
2019;5(1):41.
4. Morrison S, Colberg SR, Parson HK, Vinik AI. Relation
between risk of falling and postural sway complexity
in diabetes. Gait Posture. 2012;35(4):662–8.
5. Author. Report and recommendations of the
San Antonio conference on diabetic neuropathy.
Neurology. 1988;38(7):1161–5.
6. Pop-Busui R, Boulton AJ, Feldman EL, et  al.
Diabetic neuropathy: a position statement by the
American Diabetes Association. Diabetes Care.
2017;40(1):136–54.
7. Visser NA, Notermans NC, Linssen RS, van den
Berg LH, Vrancken AF. Incidence of polyneu-
ropathy in Utrecht, the Netherlands. Neurology.
2015;84(3):259-264.
8. Forrest KY, Maser RE, Pambianco G, Becker DJ,
Orchard TJ.  Hypertension as a risk factor for dia-
betic neuropathy: a prospective study. Diabetes.
1997;46(4):665–70.
9. Sands ML, Shetterly SM, Franklin GM, Hamman
RF. Incidence of distal symmetric (sensory) neuropa-
thy in NIDDM: the San Luis Valley Diabetes Study.
Diabetes Care. 1997;20(3):322-329.
10. Jenkins PJ, Srikantharajah D, Duckworth AD, Watts
AC, McEachan JE. Carpal tunnel syndrome: the asso-
ciation with occupation at a population level. J Hand
Surg Eur Vol. 2013;38(1):67-72.
11. Latinovic R, Gulliford MC, Hughes RA. Incidence of
common compressive neuropathies in primary care. J
Neurol Neurosurg Psychiatry. 2006;77(2): 263-265.
12. Falkiner S, Myers S. When exactly can carpal tunnel
syndrome be considered work-related? ANZ J Surg.
2002;72(3):204-209.
13. Mondelli M, Fiannini F, Mondelli M, Giannini F,
BalleriniM, Ginanneschi F, Martorelli E. Incidence of
ulner neuropathy at the elbow in the province of Siena
(Italy). J Neurol Sci. 2005;234:5-10.
14. van Slobbe AM, Bohnen AM, Bernsen RM, Koes
BW, Bierma-Zeinstra SM. Incidence rates and deter-
Neuropathy in Type 1 and Type 2 Diabetes
minants in meralgia paresthetica in general practice. J
Neurol. 2004;251(3):294-297.
15. Rowlands S, Hooper R, Hughes R, Burney P. The epi-
demiology and treatment of Bell’s palsy in the UK.
Eur J Neurol. 2002;9(1):63-67.
16. Katusic SK, Beard CM, WiederholtWC, Bergstralh
EJ, Kurland LT. Incidence, clinical features, and prog-
nosis in Bell’s palsy, Rochester, Minnesota, 1968-
1982. Ann Neurol. 1986;20(5): 622-627.
17. Monini S, Lazzarino AI, Iacolucci C, Buffoni A,
BarbaraM. Epidemiology of Bell’s palsy in an Italian
Health District: incidence and case-control study.
Acta Otorhinolaryngol Ital. 2010;30(4):198.
18. Schoenfeld AJ, Laughlin M, Bader JO, Bono CM.
Characterization of the incidence and risk factors for
the development of lumbar radiculopathy. J Spinal
Disord Tech. 2012;25(3):163-167.
19. Schoenfeld AJ, George AA, Bader JO, Caram PM Jr.
Incidence and epidemiology of cervical radiculopathy
in the United Statesmilitary: 2000 to 2009. J Spinal
Disord Tech. 2012;25(1):17-22.
20. Radhakrishnan K, LitchyWJ, O’Fallon WM, Kurland
LT. Epidemiology of cervical radiculopathy: a popu-
lation-based study from Rochester, Minnesota, 1976
through 1990. Brain. 1994;117(pt 2):325-335.
21. Tracy JA, Dyck PJ.  The spectrum of diabetic
neuropathies. Phys Med Rehabil Clin N Am.
2008;19(1):1–26.
22. Breiner A, Lovblom LE, Perkins BA, Bril V. Does
the prevailing hypothesis that small-fiber dys-
function precedes large-fiber dysfunction apply
to type 1 diabetic patients? Diabetes Care.
2014;37(5):1418–24.
23. Feldman EL, Nave KA, Jensen TS, Bennett DLH. New
horizons in diabetic neuropathy: mechanisms, bioen-
ergetics, and pain. Neuron. 2017;93(6):1296–313.
24. American Diabetes A.  Microvascular complications
and foot care: standards of medical care in diabetes-
­2021. Diabetes Care. 2021;44(1):151–67.
25. Abbott CA, Malik RA, van Ross ER, Kulkarni J,
Boulton AJ.  Prevalence and characteristics of pain-
ful diabetic neuropathy in a large community-­based
diabetic population in the UK.  Diabetes Care.
2011;34(10):2220–4.
26. Didangelos T, Doupis J, Veves A.  Painful diabetic
neuropathy: clinical aspects. Handb Clin Neurol.
2014;126:53–61.
27. Boulton AJ, Armstrong WD, Scarpello JH, Ward
JD. The natural history of painful diabetic neuropathy–
a 4-year study. Postgrad Med J. 1983;59(695):556–9.
28. Benbow SJ, Chan AW, Bowsher D, MacFarlane IA,
Williams G.  A prospective study of painful symp-
toms, small-fibre function and peripheral vascular
disease in chronic painful diabetic neuropathy. Diabet
Med. 1994;11(1):17–21.
29. Katoulis EC, Ebdon-Parry M, Lanshammar H,
Vileikyte L, Kulkarni J, Boulton AJ.  Gait abnor-
malities in diabetic neuropathy. Diabetes Care.
1997;20(12):1904–7.
30. Walters DP, Gatling W, Mullee MA, Hill RD.  The
prevalence of diabetic distal sensory neuropathy in an
English community. Diabet Med. 1992;9(4):349–53.
63
31. Skyler JS, Bakris GL, Bonifacio E, et al. Differentiation
of diabetes by pathophysiology, natural history, and
prognosis. Diabetes. 2017;66(2):241–55.
32. Unger RH, Grundy S. Hyperglycaemia as an inducer
as well as a consequence of impaired islet cell function
and insulin resistance: implications for the manage-
ment of diabetes. Diabetologia. 1985;28(3):119–21.
33. Partanen J, Niskanen L, Lehtinen J, Mervaala E,
Siitonen O, Uusitupa M. Natural history of peripheral
neuropathy in patients with non-­ insulin-dependent
diabetes mellitus. N Engl J Med. 1995;333(2):89–94.
34. Sumner CJ, Sheth S, Griffin JW, Cornblath DR,
Polydefkis M.  The spectrum of neuropathy in dia-
betes and impaired glucose tolerance. Neurology.
2003;60(1):108–11.
35. Singleton JR, Smith AG, Bromberg MB. Painful sen-
sory polyneuropathy associated with impaired glu-
cose tolerance. Muscle Nerve. 2001;24(9):1225–8.
36. Sveen KA, Karime B, Jorum E, et  al. Small- and
large-fiber neuropathy after 40 years of type 1 diabe-
tes: associations with glycemic control and advanced
protein glycation: the Oslo Study. Diabetes Care.
2013;36(11):3712–7.
37. Alam U, Jeziorska M, Petropoulos IN, et  al. Latent
autoimmune diabetes of adulthood (LADA) is asso-
ciated with small fibre neuropathy. Diabet Med.
2019;36(9):1118–24.
38. Dyck PJ, Kratz KM, Karnes JL, et al. The prevalence
by staged severity of various types of diabetic neu-
ropathy, retinopathy, and nephropathy in a popula-
tion-based cohort: the Rochester Diabetic Neuropathy
Study. Neurology. 1993;43(4):817–24.
39. Loseth S, Stalberg EV, Lindal S, Olsen E, Jorde R,
Mellgren SI. Small and large fiber neuropathy in those
with type 1 and type 2 diabetes: a 5-year follow-­up
study. J Peripher Nerv Syst. 2016;21(1):15–21.
40. Loseth S, Mellgren SI, Jorde R, Lindal S, Stalberg
E.  Polyneuropathy in type 1 and type 2 diabetes:
comparison of nerve conduction studies, thermal per-
ception thresholds and intraepidermal nerve fibre den-
sities. Diabetes Metab Res Rev. 2010;26(2):100–6.
41. Vinik AI, Casellini C, Parson HK, Colberg SR,
Nevoret ML.  Cardiac autonomic neuropathy in dia-
betes: a predictor of cardiometabolic events. Front
Neurosci. 2018;12:591.
42. Maser RE, Mitchell BD, Vinik AI, Freeman R. The
association between cardiovascular autonomic neu-
ropathy and mortality in individuals with diabetes: a
meta-analysis. Diabetes Care. 2003;26(6):1895–901.
43. Ziegler D, Zentai CP, Perz S, et al. Prediction of mor-
tality using measures of cardiac autonomic dysfunc-
tion in the diabetic and nondiabetic population: the
MONICA/KORA Augsburg Cohort Study. Diabetes
Care. 2008;31(3):556–61.
44. Pfeifer MA, Weinberg CR, Cook DL, et al. Autonomic
neural dysfunction in recently diagnosed diabetic sub-
jects. Diabetes Care. 1984;7(5):447–53.
45. Jaiswal M, Divers J, Urbina EM, et al. Cardiovascular
autonomic neuropathy in adolescents and young
adults with type 1 and type 2 diabetes: The SEARCH
for Diabetes in Youth Cohort Study. Pediatr Diabetes.
2018;19(4):680–9.
64
46. Eleftheriadou A, Williams S, Nevitt S, et al. The preva-
lence of cardiac autonomic neuropathy in prediabetes: a
systematic review. Diabetologia. 2021;64(2):288–303.
47. Martin CL, Albers JW, Pop Busui R. Neuropathy and
related findings in the diabetes control and complica-
tions trial/epidemiology of diabetes interventions and
complications study. Diabetes Care. 2014;37(1):31–8.
48. Pop-Busui R, Low PA, Waberski BH, et  al. Effects
of prior intensive insulin therapy on cardiac auto-
nomic nervous system function in type 1 diabetes
mellitus: the Diabetes Control and Complications
Trial/Epidemiology of Diabetes Interventions and
Complications study (DCCT/EDIC). Circulation.
2009;119(22):2886–93.
49. Choung RS, Locke GR, Schleck CD, Zinsmeister AR,
Melton LJ, Talley NJ. Risk of gastroparesis in subjects
with type 1 and 2 diabetes in the general population.
Am J Gastroenterol. 2012;107(1):82–8.
50. Krishnasamy S, Abell TL.  Diabetic gastroparesis:
principles and current trends in management. Diabetes
Ther. 2018;9(Suppl 1):1–42.
51. Freeman R.  Diabetic autonomic neuropathy. Handb
Clin Neurol. 2014;126:63–79.
52. Valdovinos MA, Camilleri M, Zimmerman
BR.  Chronic diarrhea in diabetes mellitus: mecha-
nisms and an approach to diagnosis and treatment.
Mayo Clin Proc. 1993;68(7):691–702.
53. Lysy J, Israeli E, Goldin E. The prevalence of chronic
diarrhea among diabetic patients. Am J Gastroenterol.
1999;94(8):2165–70.
54. Kempler P, Amarenco G, Freeman R, et al. Management
strategies for gastrointestinal, erectile, bladder, and
sudomotor dysfunction in patients with diabetes.
Diabetes Metab Res Rev. 2011;27(7):665–77.
55. Araujo AB, Hall SA, Ganz P, et al. Does erectile dys-
function contribute to cardiovascular disease risk pre-
diction beyond the Framingham risk score? J Am Coll
Cardiol. 2010;55(4):350–6.
56. Pop-Busui R, Hotaling J, Braffett BH, et  al.
Cardiovascular autonomic neuropathy, erectile dys-
function and lower urinary tract symptoms in men
with type 1 diabetes: findings from the DCCT/EDIC. J
Urol. 2015;193(6):2045–51.
57. Garcia-Cruz E, Leibar-Tamayo A, Romero J, et  al.
Metabolic syndrome in men with low testosterone
levels: relationship with cardiovascular risk factors
and comorbidities and with erectile dysfunction. J Sex
Med. 2013;10(10):2529–38.
58. Sarma AV, Kanaya A, Nyberg LM, et al. Risk factors
for urinary incontinence among women with type 1
diabetes: findings from the epidemiology of diabe-
tes interventions and complications study. Urology.
2009;73(6):1203–9.
59. Albers JW, Pop-Busui R. Diabetic neuropathy: mech-
anisms, emerging treatments, and subtypes. Curr
Neurol Neurosci Rep. 2014;14(8):473.
60. Gibbons CH, Freeman R. Treatment-induced neurop-
athy of diabetes: an acute, iatrogenic complication of
diabetes. Brain. 2015;138(Pt 1):43–52.
61. Gibbons CH. Treatment induced neuropathy of diabe-
tes. Auton Neurosci. 2020;226:102668.
G. Akinci et al.
62. Dyck PJ, Windebank AJ.  Diabetic and nondiabetic
lumbosacral radiculoplexus neuropathies: new
insights into pathophysiology and treatment. Muscle
Nerve. 2002;25(4):477–91.
63. Boulton AJ, Malik RA, Arezzo JC, Sosenko
JM.  Diabetic somatic neuropathies. Diabetes Care.
2004;27(6):1458–86.
64. Krol CG, Klok FA, de Koning EJ. Diplopia as the pre-
senting symptom of type 1 diabetes. Diabetes Care.
2014;37(3):e45–6.
65. Keane JR.  Third nerve palsy: analysis of 1400 per-
sonally-examined inpatients. Can J Neurol Sci.
2010;37(5):662–70.
66. Chou KL, Galetta SL, Liu GT, et  al. Acute ocular
motor mononeuropathies: prospective study of the
roles of neuroimaging and clinical assessment. J
Neurol Sci. 2004;219(1-2):35–9.
67. Vinik A, Mehrabyan A, Colen L, Boulton A.  Focal
entrapment neuropathies in diabetes. Diabetes Care.
2004;27(7):1783–8.
68. Ziegler D, Papanas N, Vinik AI, Shaw
JE. Epidemiology of polyneuropathy in diabetes and
prediabetes. Handb Clin Neurol. 2014;126:3–22.
69. Tesfaye S, Stevens LK, Stephenson JM, et  al.
Prevalence of diabetic peripheral neuropathy and its
relation to glycaemic control and potential risk fac-
tors: the EURODIAB IDDM Complications Study.
Diabetologia. 1996;39(11):1377–84.
70. Adler AI, Boyko EJ, Ahroni JH, Stensel V, Forsberg
RC, Smith DG.  Risk factors for diabetic periph-
eral sensory neuropathy. Results of the seattle
prospective diabetic foot study. Diabetes Care.
1997;20(7):1162–7.
71. Braffett BH, Gubitosi-Klug RA, Albers JW, et al. Risk
factors for diabetic peripheral neuropathy and cardio-
vascular autonomic neuropathy in the diabetes con-
trol and complications trial/epidemiology of diabetes
interventions and complications (DCCT/EDIC) study.
Diabetes. 2020;69(5):1000–10.
72. Callaghan BC, Gallagher G, Fridman V, Feldman
EL. Diabetic neuropathy: what does the future hold?
Diabetologia. 2020;63(5):891–7.
73. Callaghan BC, Little AA, Feldman EL, Hughes
RA.  Enhanced glucose control for preventing and
treating diabetic neuropathy. Cochrane Database Syst
Rev. 2012;6:CD007543.
74. Ziegler D, Behler M, Schroers-Teuber M, Roden
M.  Near-normoglycaemia and development of neu-
ropathy: a 24-year prospective study from diagnosis
of type 1 diabetes. BMJ Open. 2015;5(6):e006559.
75. Maser RE, Steenkiste AR, Dorman JS, et  al.
Epidemiological correlates of diabetic neuropathy.
Report from Pittsburgh Epidemiology of Diabetes
Complications Study. Diabetes. 1989;38(11):1456–61.
76. Callaghan BC, Hur J, Feldman EL.  Diabetic neu-
ropathy: one disease or two? Curr Opin Neurol.
2012;25(5):536–41.
77. Pop-Busui R, Braffett BH, Zinman B, et  al.
Cardiovascular autonomic neuropathy and car-
diovascular outcomes in the diabetes control and
complications trial/epidemiology of diabetes inter-
Neuropathy in Type 1 and Type 2 Diabetes
ventions and complications (DCCT/EDIC) study.
Diabetes Care. 2017;40(1):94–100.
78. Pambianco G, Costacou T, Ellis D, Becker DJ,
Klein R, Orchard TJ.  The 30-year natural history
of type 1 diabetes complications: the Pittsburgh
Epidemiology of Diabetes Complications Study
experience. Diabetes. 2006;55(5):1463–9.
79. Tesfaye S, Chaturvedi N, Eaton SE, et al. Vascular
risk factors and diabetic neuropathy. N Engl J Med.
2005;352(4):341–50.
80. Jaiswal M, Divers J, Dabelea D, et  al. Prevalence
of and risk factors for diabetic peripheral neu-
ropathy in youth with type 1 and type 2 diabetes:
SEARCH for Diabetes in Youth Study. Diabetes
Care. 2017;40(9):1226–32.
81. Akinci G, Savelieff MG, Gallagher G, Callaghan
BC, Feldman EL.  Diabetic neuropathy in children
and youth: new and emerging risk factors. Pediatr
Diabetes. 2020;22(2):132–47.
82. Singleton JR, Smith AG, Bromberg MB. Increased
prevalence of impaired glucose tolerance in patients
with painful sensory neuropathy. Diabetes Care.
2001;24(8):1448–53.
83. Asghar O, Petropoulos IN, Alam U, et  al. Corneal
confocal microscopy detects neuropathy in subjects
with impaired glucose tolerance. Diabetes Care.
2014;37(9):2643–6.
84. Bongaerts BW, Rathmann W, Heier M, et al. Older
subjects with diabetes and prediabetes are fre-
quently unaware of having distal sensorimotor poly-
neuropathy: the KORA F4 study. Diabetes Care.
2013;36(5):1141–6.
85. Callaghan BC, Xia R, Banerjee M, et al. Metabolic
syndrome components are associated with symp-
tomatic polyneuropathy independent of glycemic
status. Diabetes Care. 2016;39(5):801–7.
86. Callaghan BC, Reynolds EL, Banerjee M, et al. The
prevalence and determinants of cognitive deficits
and traditional diabetic complications in the severely
obese. Diabetes Care. 2020;43(3):683–90.
87. Callaghan BC, Reynolds E, Banerjee M, Chant E,
Villegas-Umana E, Feldman EL. Central obesity is
associated with neuropathy in the severely obese.
Mayo Clin Proc. 2020;95(7):1342–53.
88. Callaghan BC, Xia R, Reynolds E, et al. Association
between metabolic syndrome components and poly-
neuropathy in an obese population. JAMA Neurol.
2016;73(12):1468–76.
89. Callaghan BC, Gao L, Li Y, et al. Diabetes and obesity
are the main metabolic drivers of peripheral neurop-
athy. Ann Clin Transl Neurol. 2018;5(4):397–405.
90. Schlesinger S, Herder C, Kannenberg JM, et  al.
General and abdominal obesity and incident dis-
tal sensorimotor polyneuropathy: insights into
inflammatory biomarkers as potential media-
tors in the KORA F4/FF4 cohort. Diabetes Care.
2019;42(2):240–7.
91. Hanewinckel R, Drenthen J, Ligthart S, et  al.
Metabolic syndrome is related to polyneuropa-
thy and impaired peripheral nerve function: a pro-
spective population-based cohort study. J Neurol
Neurosurg Psychiatry. 2016;87(12):1336–42.
65
92. Ohkubo Y, Kishikawa H, Araki E, et  al. Intensive
insulin therapy prevents the progression of diabetic
microvascular complications in Japanese patients
with non-insulin-dependent diabetes mellitus: a ran-
domized prospective 6-year study. Diabetes Res Clin
Pract. 1995;28(2):103–17.
93. Ismail-Beigi F, Craven T, Banerji MA, et  al.
Effect of intensive treatment of hyperglycaemia
on microvascular outcomes in type 2 diabetes: an
analysis of the ACCORD randomised trial. Lancet.
2010;376(9739):419–30.
94. UK Prospective Diabetes Study (UKPDS) Group.
Intensive blood-glucose control with sulpho-
nylureas or insulin compared with conventional
treatment and risk of complications in patients
with type 2 diabetes (UKPDS 33). Lancet.
1998;352(9131):837–53.
95. Duckworth W, Abraira C, Moritz T, et  al.
Glucose control and vascular complications in
veterans with type 2 diabetes. N Engl J Med.
2009;360(2):129–39.
96. Charles M, Ejskjaer N, Witte DR, Borch-Johnsen
K, Lauritzen T, Sandbaek A. Prevalence of neuropa-
thy and peripheral arterial disease and the impact
of treatment in people with screen-detected type 2
diabetes: the ADDITION-Denmark study. Diabetes
Care. 2011;34(10):2244–9.
97. Charles M, Fleischer J, Witte DR, et  al. Impact
of early detection and treatment of diabetes on
the 6-year prevalence of cardiac autonomic neu-
ropathy in people with screen-detected diabetes:
ADDITION-Denmark, a cluster-randomised study.
Diabetologia. 2013;56(1):101–8.
98. Siegelaar SE, Kilpatrick ES, Rigby AS, Atkin SL,
Hoekstra JB, Devries JH.  Glucose variability does
not contribute to the development of peripheral and
autonomic neuropathy in type 1 diabetes: data from
the DCCT. Diabetologia. 2009;52(10):2229–32.
99. Xu F, Zhao LH, Su JB, et  al. The relationship
between glycemic variability and diabetic peripheral
neuropathy in type 2 diabetes with well-controlled
HbA1c. Diabetol Metab Syndr. 2014;6(1):139.
100. Kwai NC, Arnold R, Poynten AM, Krishnan
AV.  Association between glycemic variability and
peripheral nerve dysfunction in type 1 diabetes.
Muscle Nerve. 2016;54(5):967–9.
101. Lachin JM, Bebu I, Bergenstal RM, et al. Association
of glycemic variability in type 1 diabetes with pro-
gression of microvascular outcomes in the diabe-
tes control and complications trial. Diabetes Care.
2017;40(6):777–83.
102. Christensen MMB, Hommel EE, Jorgensen ME,
Fleischer J, Hansen CS.  Glycemic variability and
diabetic neuropathy in young adults with type 1 dia-
betes. Front Endocrinol. 2020;11:644.
103. Ortega FB, Lavie CJ, Blair SN. Obesity and cardio-
vascular disease. Circ Res. 2016;118(11):1752–70.
104. Andersen ST, Witte DR, Dalsgaard EM, et al. Risk
factors for incident diabetic polyneuropathy in a
cohort with screen-detected type 2 diabetes followed
for 13 years: ADDITION-denmark. Diabetes Care.
2018;41(5):1068–75.
66
105. Callaghan BC, Price RS, Feldman EL. Distal symmet-
ric polyneuropathy in 2020. JAMA. 2020;324(1):90–1.
106. Shah AS, El Ghormli L, Vajravelu ME, et al. Heart
rate variability and cardiac autonomic dysfunction:
prevalence, risk factors, and relationship to arterial
stiffness in the treatment options for type 2 diabetes
in adolescents and youth (TODAY) study. Diabetes
Care. 2019;42(11):2143–50.
107. Laitinen T, Lindstrom J, Eriksson J, et  al.
Cardiovascular autonomic dysfunction is associated
with central obesity in persons with impaired glu-
cose tolerance. Diabet Med. 2011;28(6):699–704.
108. Dabelea D, Stafford JM, Mayer-Davis EJ, et  al.
Association of type 1 diabetes vs type 2 diabetes
diagnosed during childhood and adolescence with
complications during teenage years and young adult-
hood. JAMA. 2017;317(8):825–35.
109. De Block CE, De Leeuw IH, Van Gaal LF. Impact
of overweight on chronic microvascular complica-
tions in type 1 diabetic patients. Diabetes Care.
2005;28(7):1649–55.
110. Wiggin TD, Sullivan KA, Pop-Busui R, Amato A,
Sima AA, Feldman EL.  Elevated triglycerides cor-
relate with progression of diabetic neuropathy.
Diabetes. 2009;58(7):1634–40.
111. Han L, Ji L, Chang J, et al. Peripheral neuropathy is
associated with insulin resistance independent of met-
abolic syndrome. Diabetol Metab Syndr. 2015;7:14.
112. Ylitalo KR, Sowers M, Heeringa S. Peripheral vas-
cular disease and peripheral neuropathy in individu-
als with cardiometabolic clustering and obesity:
National Health and Nutrition Examination Survey
2001-2004. Diabetes Care. 2011;34(7):1642–7.
113. Smith AG, Singleton JR.  Obesity and hyperlipid-
emia are risk factors for early diabetic neuropathy. J
Diabetes Complicat. 2013;27(5):436–42.
114. Lee KO, Nam JS, Ahn CW, et al. Insulin resistance is
independently associated with peripheral and auto-
nomic neuropathy in Korean type 2 diabetic patients.
Acta Diabetol. 2012;49(2):97–103.
115. Cai Z, Yang Y, Zhang J.  A systematic review and
meta-analysis of the serum lipid profile in prediction
of diabetic neuropathy. Sci Rep. 2021;11(1):499.
G. Akinci et al.
116. Callaghan BC, Feldman E, Liu J, et al. Triglycerides
and amputation risk in patients with diabetes: ten-­
year follow-up in the DISTANCE study. Diabetes
Care. 2011;34(3):635–40.
117. Andersen ST, Witte DR, Fleischer J, et al. Risk factors
for the presence and progression of cardiovascular
autonomic neuropathy in type 2 diabetes: ADDITION-
Denmark. Diabetes Care. 2018;41(12):2586–94.
118. Author. Effect of intensive diabetes treatment on nerve
conduction in the Diabetes Control and Complications
Trial. Ann Neurol. 1995;38(6):869–80.
119. Diabetes C, Nathan DM, et al. The effect of inten-
sive treatment of diabetes on the development
and progression of long-term complications in
insulin-­dependent diabetes mellitus. N Engl J Med.
1993;329(14):977–86.
120. Ponirakis G, Petropoulos IN, Alam U, et  al.
Hypertension contributes to neuropathy in patients with
type 1 diabetes. Am J Hypertens. 2019;32(8):796–803.
121. Linn T, Ortac K, Laube H, Federlin K.  Intensive
therapy in adult insulin-dependent diabetes mel-
litus is associated with improved insulin sensitivity
and reserve: a randomized, controlled, prospective
study over 5 years in newly diagnosed patients.
Metabolism. 1996;45(12):1508–13.
122. Azad N, Emanuele NV, Abraira C, et  al. The
effects of intensive glycemic control on neuropa-
thy in the VA cooperative study on type II diabe-
tes mellitus (VA CSDM). J Diabetes Complicat.
1999;13(5-6):307–13.
123. Tovi J, Svanborg E, Nilsson BY, Engfeldt P. Diabetic
neuropathy in elderly Type 2 diabetic patients:
effects of insulin treatment. Acta Neurol Scand.
1998;98(5):346–53.
124. Shichiri M, Kishikawa H, Ohkubo Y, Wake N. Long-­
term results of the Kumamoto Study on optimal
diabetes control in type 2 diabetic patients. Diabetes
Care. 2000;23(2):21–9.
125. Sandbaek A, Griffin SJ, Sharp SJ, et  al. Effect of
early multifactorial therapy compared with routine
care on microvascular outcomes at 5 years in people
with screen-detected diabetes: a randomized con-
trolled trial: the ADDITION-Europe Study. Diabetes
Care. 2014;37(7):2015–23.
 Clinical Diagnosis of Diabetic
Peripheral Neuropathy
Bruce A. Perkins and Vera Bril
1 Introduction and Context
The variability in clinical presentation and com-
plexities in measurement of diabetic neuropathy
have not only challenged the clear understanding
of epidemiology, impact, and the clinical trial
development of therapies, but these often compli-
cate the diagnostic process. However, with a clear
understanding of these obstacles and the relevant
clinical tools, clinicians from all fields of spe-
cialty can develop an efficient approach to diag-
nosis. This includes more simplified approaches
suitable to primary care and diabetes management
clinics that screen for and diagnose diabetic neu-
ropathy, to the specialty neurology clinics that
may use more refined diagnostic methods required
for confirmation of neuropathy and the evaluation
of more atypical cases. Certainly, the vast major-
ity of cases of diabetic neuropathy do not require
referral to a neurology specialist for confirmation.
This places a responsibility on the diabetes care
clinician to be able to confidently make a working
diagnosis on clinical evidence and to initiate man-
B. A. Perkins (*)
Division of Endocrinology, Department of Medicine,
Sinai Health System, University of Toronto,
Toronto, ON, Canada
e-mail: Bruce.Perkins@sinaihealth.ca
V. Bril
Division of Neurology, Department of Medicine,
University of Toronto, Toronto, ON, Canada
e-mail: vera.bril@utoronto.ca
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_5
agement, consider other causes of polyneuropa-
thy, while identifying atypical features that
warrant specialist referral. This chapter discusses
these considerations and the approaches to the
clinical diagnosis of diabetic neuropathy, while
chapter “Clinical Features Diabetic Neuropathies”
details clinical features, and chapter “Diagnostic
Techniques for Diabetic Peripheral Neuropathy”
details specialized diagnostic techniques.
1.1 Diabetic Distal Symmetric
Polyneuropathy (Diabetic
DSP): A Brief Overview
The most common form—and the one usually
referred to as “diabetic peripheral neuropathy” or
even more simply as “diabetic neuropathy”—is
the diffuse, symmetrical, slowly-progressive
length-dependent damage to the peripheral and
autonomic nervous system classified technically
by the term “diabetic distal symmetrical polyneu-
ropathy” (diabetic DSP) [1]. It typically remains
asymptomatic for years, may first present clini-
cally with symptoms of abnormal sensation sym-
metrically at the tips of the toes and may over
time spread to the stocking-and-glove distribu-
tion. It involves injury to different anatomical
nerve types that show variable clinical manifesta-
tions between individuals. These can be classi-
fied as small fibers (thinly-myelinated autonomic
B-fibers, and thinly- or un- myelinated autonomic
67
68 B. A. Perkins and V. Bril

and sensory C-fibers) that make up autonomic
and pain and temperature sensory fibers. Damage
to large fibers (myelinated Aα, Aβ, Aδ sensory,
and Aγ motor) are responsible for other sensory
and skeletal muscle impairments [2, 3]. At the
symptomatic stages, some have a large fiber-­
predominant pattern and experience numbness,
tingling, or imbalance owing to sensory ataxia,
while others have small fiber-predominant symp-
toms that may instead present with burning and
stabbing pain, impairment in sensing heat and
cold, or a propensity to clinical autonomic abnor-
malities like postural hypotension and gastropa-
resis [1]. Some experience combined
manifestations of large and small fiber types, and,
regardless of pattern, people with diabetes can be
asymptomatic for extended periods even though
their physical examination or specialized testing
can reveal subtle or even marked impairments in
nerve structure and function. With progression to
foot muscle weakness (motor weakness repre-
sents another large fiber dysfunction), or clinical
autonomic manifestations like dry feet from lim-
ited sweat production (sudomotor dysfunction),
the subsequent risk of ulcer, infection, and ampu-
tation intensify. Complications like these are
feared more than death itself [4], and of great
concern is the recent evidence in certain parts of
the world of a resurgence in the risk of amputa-
tions [5–7].
1.2 Staging of Diabetic DSP
For the purpose of understanding the face value
of symptoms and signs and their measurement
scales, we center the discussion of clinical diag-
nosis in this chapter around the conventional
staging system described in Table 1 [9]. In simple
terms, in practice diabetic DSP (Stage 1b or
higher) is identified through two possible points
in clinical care: Asymptomatic screening, leading
to a clinical diagnostic workup if screening phys-
ical examination results are abnormal, or alterna-
tively through complaints of neuropathic
symptoms on a clinical encounter, leading simi-
larly to a clinical workup involving history, phys-
ical examination, simple laboratory tests, and

Table 1  Conventional diabetic distal symmetric polyneuropathy (diabetic DSP) staging system with notes on diagnos-
tic considerations
Stage Description and considerations
Stage 0 No diabetic DSP
Absence of neuropathic symptoms, signs, and normal NCS.
In clinical practice, such patients would not be evaluated by NCS, and thus Stage 0 would rarely be
confirmed. This is more relevant to clinical research.
May be associated with dysfunction in objective tests other than NCS:
 Small fiber morphology tests: Examples include intra-epidermal nerve fiber density or other
parameters, subtle abnormalities in corneal nerve fiber density
 Small fiber function tests: heart rate variability, sudomotor dysfunction, abnormal vascular reactivity
to heating or other stimuli
 Large Fiber function tests: subtle abnormalities in monofilament sensitivity
 Large Function morphology tests: Subtle abnormalities in sural nerve biopsy
 Measureable subtle dysfunction in these tests could serve as predictive index tests during Stage 0 for
future onset of diabetic DSP. Proof-of-concept exists for the 8-point distal monofilament examination
and for corneal nerve fiber density [8] (CNFL in press)
Stage 1 Asymptomatic diabetic DSP
Stage 1a Absence of neuropathic symptoms, signs, but with abnormal NCS
In clinical practice, such patients would not be evaluated by NCS and thus this stage would rarely be
identified. This is more relevant to clinical research
Stage 1b Abnormality in nerve conduction studies and neuropathic signs, but absence of neuropathic symptoms
In clinical practice, such pateints would be identified by asymptomatic screening as recommended by
guidelines
These subclinical stages 1a and 1b could similarly be associated with abnormalities in other large fiber
morphology and function measures, or in small fiber measures as outlined for Stage 0
Clinical Diagnosis of Diabetic Peripheral Neuropathy 69

Table 1 (continued)
Stage Description and considerations
Stage 2 Clinical diabetic DSP
Stage 2a Criteria for subclinical neuropathy have been met and there are neuropathic symptoms with or without
neuropathic signs
In clinical practice, such people would be identified based on symptomatic complaints leading to a
history and physical examination
At this stage, neuropathy is sufficient for complications including neuropathic pain and sensory ataxia
(imbalance)
Stage 2b Criteria for subclinical neuropathy have been met and there is unequivocal weakness of ankle
dorsiflexion
Beyond this initial motor manifestation change, later stages (undefined by this classification scheme)
include progressive motor dysfunction and the presence of foot complications such as ulceration,
infection, amputation, and the Charcot deformity
DSP distal symmetric polyneuropathy, NCS nerve conduction studies
Table modified on content in Dyck et al. [9]

consideration in a small subset of people for neu- type 2 diabetes, the existence of pre-diabetes may
rologist referral for specialized testing and lead to polyneuropathy and therefore it may be
evaluation. present at diagnosis even without substantial ele-
vations in glycemic exposure [17].

2 Key Clinical Considerations

FOR Diabetic DSP Diagnosis 2.2 The Impaired Protective
Sensation of Diabetic DSP
2.1 Clinical Risk Factors Represents Only One
for Diabetic DSP Are Typically Component Cause of Foot
Present Complications

Elevated cumulative glycemic exposure is the
fundamental risk factor for diabetic DSP onset
[10–12]. As such, age, diabetes duration, and gly-
cemic exposure measured by an inability to
maintain glycated hemoglobin A1c at target lev-
els, generally considered to be 7.0% or less, are
related risk factors [13]. Additionally, cardiovas-
cular metabolic risk factors including presence of
hypertension, abdominal obesity, hypertriglycer-
idemia, and low high-density lipoprotein, as well
as taller height, alcohol abuse, and chronic kid-
ney disease have been identified, in addition to
putative racial and genetic factors that may be
clinically identifiable in future [13–16]. However,
specific threshold levels of risk factors and their
duration have not been proposed. Generally, an
individual with short duration of type 1 diabetes
is extremely unlikely to have diabetic DSP unless
exceedingly high glycemic exposure has occurred
and other risk factors are present. However, in
While diabetic DSP itself is sufficient to cause
pain and imbalance as complications, it repre-
sents only a component cause, neither necessary
nor sufficient, to cause ulcer, infection, the
Charcot deformity, or amputation. Tissue isch-
emia from peripheral vascular disease, and foot
deformity or the minor trauma to skin that incites
ulceration and infection induced by inadequacies
in footwear and general foot care represent the
other component causes. The implication is that
if a clinician is to meaningfully help to prevent
foot complications, the annual foot evaluation
must not focus only on the identification of the
loss of protective sensation representing diabetic
DSP alone. It must include evaluation for arterial
patency (pedal pulses and skin changes), and foot
inspection for presence of abnormalities such as
callouses or deformity that may indicate the pres-
ence of repetitive minor trauma (Fig.  1 and
Table 2).
70 B. A. Perkins and V. Bril

Diabetic Neuropathy alone is

Neuropathy sufficient for pain
(Loss of Protective and imbalance
Sensation) (sensory ataxia)

Component Causes
Peripheral for Foot
Vascular Complications
Disease Amputation
(Pain, Ulcer,
(Impaired Tissue Infection, Charcot
Perfusion) deformity)

General Foot
Care
(Deformity and
Minor Trauma to
Skin)

Fig. 1  A simplified view of the core component causes of
foot complications and amputation. While sufficient for
pain and sensory ataxia, the presence of loss of protective
sensation from diabetic neuropathy alone is not consid-
ered sufficient for advanced foot complications without
the presence of peripheral vascular disease or minor
trauma to skin. The implication is that clinical evaluation
focused only on diabetic neuropathy, without consider-
ation of the other component causes through simple
inspection and palpation for pulses, is insufficient

Table 2  Clinical manifestations according to fiber classification and other key clinical features of diabetic DSP
Fiber classification Large fibers: myelinated Aδ sensory, Aγ motor, Small fibers: thinly-myelinated autonomic
and Aα and Aβ mixed motor and sensory B-fibers, and thinly- or un- myelinated
autonomic and sensory C-fibers
Function Touch, pressure, vibration, proprioception/ Pain sensation, temperature sensation,
balance, somatic motor autonomic
Symptoms Numbness, tingling, unsteadiness Pain, discrimination of hot and cold
Signs Impairments in ankle reflexes, vibration, light Impairments in thermal discrimination,
touch, 10-g monofilament, proprioception, pinprick sensation
sensory ataxia (positive Romberg test) [18]
“Red Flags” that are Asymmetry
inconsistent with Nonlength dependence
diabetic DSP Acute or subacute rather than insidious, chronic onset and progression
Motor predominance [19]
Also consider
Family history of sensory and motor symptoms
Sensory neuropathies causing gait ataxia and proprioceptive dysfunction
Severe dysautonomia [20]
Late stage findings Dry skin (sudomotor neuropathy), callus, fissures, ulcerations, cellulitis
Charcot deformity
Physical examination Inspection for deformity
for other foot  Callus, abrasions, hallux valgus
complication Vascular
component causes  Pallor, cyanosis, capillary refill, pedal pulses (posterior tibial and dorsalis pedis)
Diabetic DSP diabetic distal symmetric polyneuropathy
Adapted from concepts introduced by Pop-Busui et al. [1]
Clinical Diagnosis of Diabetic Peripheral Neuropathy
2.3 Impaired Protective Sensation
Shows Heterogeneity
Between People with Diabetes
Impaired protective sensation may occur asymp-
tomatically (in approximately half of those with
objective evidence of diabetic DSP) [21] or symp-
tomatically. Whether symptomatic or not, the
symptoms and signs may represent small or large
fiber sensory modalities, as well as autonomic and
motor modalities. Severity can be determined by
the extent of sensory modality impairment, and by
the presence and extent of motor weakness in the
muscles of the feet. Furthermore, while autonomic
dysfunction can be observed at early stages by
specialized testing, clinical autonomic manifesta-
tions generally occur at more advanced stages.
Finally, many may present with mixed findings
representing both small- and large-fiber modalities
[1, 11]. The implication of this heterogeneity of
clinical presentation is that the evaluating clinician
must consider these varied attributes, that theoreti-
cally the selection of individual screening tests
may not be sufficient, and that scales that consider
multiple attributes may have advantages over other
investigative tests.
In summary, evaluation of the feet in people
with diabetes represents consideration of risk
factors for diabetic DSP, evaluation of a com-
posite of the varied components of diabetic
DSP, as well as a composite of the features of
neuropathy, peripheral vascular disease, and
foot deformity and foot care. The diagnosis of
diabetic DSP itself requires a process of clinical
assessment, either initiated by asymptomatic
screening or in response to a symptomatic com-
plaint, that considers the results of various clini-
cal history and examination tests. As in any
process of diagnostic medicine, each symptom
or sign incrementally revises the clinician’s esti-
mates of disease probability and reduces clini-
cal undertainty [22].
71
3 Asymptomatic Clinical
Presentation: Screening
for Diabetic DSP
It is estimated that approximately half of people
with objective evidence by physical examination
or specialized testing have no symptoms of neu-
ropathy [1, 21]. Asymptomatic screening for neu-
ropathy has therefore generally been justified
based on “disease principles” criteria for screen-
ing [23]. For example, the epidemiology of dia-
betic DSP is generally understood, its natural
history includes a latent phase, and the target
population for screening—including type 2 diabe-
tes at diagnosis and type 1 diabetes with 5 or more
years of diabetes duration—is well defined [1].
Screening is also supported by “test principles.”
For example, screening test performance charac-
teristics and their interpretation and thresholds are
generally well understood [23]. This justification
exists even if “system principles” are not uni-
formly met. While a diabetes multiprofessional
care infrastructure exists, screening is acceptable
to people with diabetes and carries little harm,
some aspects are not yet well understood. These
include how valid specific tests are when imple-
mented in different settings, the economic evalua-
tion of screening, the lack of neuropathy-­specific,
and the implications of misclassifying neuropathy
as present when it is not (false positive) or falsely
negative. Despite these limitations, the diabetes
community has consistently supported screening
for diabetic DSP [1, 24, 25].
In practice, the annual screening examination
recommended by diabetes organizations can be
accomplished by way of very simple tests for
loss of protective sensation. Examples include
testing pressure sensation with a monofilament,
testing vibration sensation with a tuning fork, or
frankly an even simpler “Touch the Toes”
approach adopted by Diabetes UK in which the
feet are simply exposed, inspected, and touched
72
with the fingertip [24, 26, 27]. Some organiza-
tions, like the American Diabetes Association
through its neuropathy position statement, rec-
ommend a more extensive evaluation. This rec-
ommendation calls for an annual careful
neuropathic history, along with a small fiber
function test (such as temperature or pinprick
sensation), a large fiber test (vibration sensation
using a 128-Hz tuning fork), as well as a simpli-
fied version of the 10 g monofilament test to
assess for feet at risk of ulceration or non-trau-
matic amputation [1].
While we discuss physical examination
maneuvers below under Neuropathic Signs, we
wish to discuss a particular aspect of screening
for impaired protective sensation using common
maneuvers. Most maneuvers have been justified
by their face validity: They make sense, they are
available in clinical practice, they are well rec-
ognized by diverse clinicians, and represent a
key attribute of diabetic DSP. Several common
examples exist that have more extensive research
evidence in that they not only have demon-
strated face validity and reasonable test repro-
ducibility, but also intrinsic diagnostic accuracy
for concurrent validity [28, 29]. Concurrent
validity represents the ability of a test to reason-
ably classify, in a cross-sectional study design,
who in the study sample has neuropathy defined
by a reference standard test (including nerve
conduction studies) and who does not. Several
examples of tests with this level of evidence
exist [28, 29]. However, we highlight a particu-
lar quantitative variation of the 10 g-monofila-
ment examination that has even further evidence
as a screening test for the prediction of future
onset of diabetic DSP. This scoring system has
been created in order to measure more subtle
degrees of abnormality in protective sensation
[8, 24]. First, it is applied only to a distal, non-
callousing site (proximal to the nail bed as
shown in Fig. 2) in order to identify earlier, dis-
tal length-dependent sensory impairment as
compared to the more traditional multi-site test-
ing that includes more proximal sites on the
soles of the feet. Second, rather than simply
grading a 10g monofilament test as a binary
“normal” or “abnormal” result as in the tradi-
B. A. Perkins and V. Bril
tional multi-site testing, in the test variation the
monofilament is applied four times to each great
toe in a random, arrhythmic manner to generate
a quantitative score. For each stimulus, a score
of 0 is assigned if it is not perceived, 0.5 if it is
substantially less than that perceived on the
forehead or sternum, and 1 if it is perceived nor-
mally, resulting in a total score from 0 (com-
pletely insensate) to 8 (completely sensate).
Unlike the binary score, this allowed determina-
tion of two specific thresholds: The concurrent
validity threshold for identifying the presence of
neuropathy in cross-sectional study (this was
determined to be a score of 0–3 out of 8), as well
as a predictive validity threshold not associated
with the current presence of neuropathy but
instead with future onset in a 4-year longitudi-
nal cohort structure (this was determined in a
longitudinal cohort to be a score of 3.5–5.5). A
score of 6–8 is interpreted as ruling out current
neuropathy and extremely low 4-year risk of
onset [8, 24]. In summary, while frank loss of
sensation to the 10g monofilament helps to pre-
dict future ulcer and amputation risk among
those with neuropathy, a quantitative variation
of this examination seeking more subtle changes
in sensation that would otherwise be called
“normal” can be used as a screening tool, such
that asymptomatic individuals (Stage 0 and 1a)
can be reasonably stratified into those at low and
high future risk. This, after all, is the intention
of asymptomatic screening for disease: to iden-
tify at the earliest stages when interventions,
such as improvement in glycemic control, are
more likely to be effective. The identification of
other tests, including objective measures, that
could have sensitive thresholds to identify indi-
viduals at future risk of neuropathy onset repre-
sents an urgent clinical research need [30].
While the more sensitive, quantitative variation
of the monofilament examination provides an
approach to screening for protective sensation
that has been evaluated in meta-analysis [28],
we readily acknowledge that the predictive
validity findings arise form a single-­center lon-
gitudinal cohort and that it is not known with
certainty if there are advantages over other sim-
ple [27] or complex [1] screening approaches.
Clinical Diagnosis of Diabetic Peripheral Neuropathy
Fig. 2  Quantitative variation of the 10 g monofilament
test suitable for diabetic dsp screening for the presence of
neuropathy (impaired protective sensation) or its future
risk. The 10g monofilament is also known as the 10 g
Semmes-Weinstein monofilament. Adapted from Bril
et al. [24] The testing procedure: (1) Show the 10 g mono-
filament to the person with diabetes. (2) Touch it first to
the person’s forehead or sternum so that the sensation is
understood. (3) Instruct the person to say “less” or “same”
every time the monofilament stimulus is perceived. (4)
With the person’s eyes closed, apply the monofilament to
the dorsum of the great toe proximal to the nail bed as
shown in the illustration. Use a smooth motion to touch
4 Neuropathic Symptoms
The common symptoms of diabetic DSP are the
large and small fiber sensory manifestations.
Historically classified as “positive” symptoms,
indicating nerve hyper-function creating a sensa-
tion normally absent, or “negative” symptoms,
that represent hypo-function creating loss of a
sensation normally present, this distinction is not
73
the skin, bend the filament for a full second, then lift from
the skin. (5) Perform this stimulus four times per foot in
an arrhythmic manner so the person does not anticipate
when the stimulus is to be applied. (6) For each of the 8
stimuli, assign a score of 0 if it is not perceived, 0.5 if it is
substantially less than that perceived on the forehead or
sternum, and 1 if it is perceived the same. A score of 3 out
of 8 correct responses means that the presence of neuropa-
thy is likely. A score of 3.5–5 means that the risk of new-­
onset neuropathy in the next 4 years is high. A score of 5.5
or greater indicates that there is a low risk of neuropathy
onset in the next 4 years
clinically informative as there is a subtlety to the
lived experience of polyneuropathy [31, 32].
Some recommend classification by small and
large fiber sensory modalities [1], and we describe
clinical features according to this classification
throughout this chapter (summarized in Table 2).
Regardless of classification, common symptoms
are pain, tingling, numbness, unsteadiness, and a
feeling of weakness. Symptoms typically have a
74
slow, insidious progression, beginning bilaterally
at the tips of the toes moving proximally over
months or years to involve the lower thighs at
which point the hands may become involved.
This “stocking and glove” distribution of sensory
symptoms can even progress to involve the mid-
line of the abdomen, and symptom descriptions
are variable between people. Unequivocally,
symptoms present in the hands that are not
accompanied by clinical features first in the lower
limbs at or below the knee are not consistent with
a diagnosis of diabetic DSP.  Among the small
fiber symptoms, common pain descriptors are
“burning,” “stinging,” “shock-like,” lancinating
pain [33]. Less common descriptors are a
“squeezing,” pressure sensation. Pain can be
present during the daytime or nighttime, and
often have paroxysmal patterns. Small fiber dys-
function also includes impaired temperature dis-
crimination, often described as a loss of ability to
identify hot or cold water with one’s feet in the
bathtub. Large fiber sensory symptoms include
tingling, “pins and needles,” or numbness.
Numbness can represent a lack of sensation, and
is often used variably by people with diabetes to
describe lack of thermal sensitivity, light touch,
lack of pain sensation from minor traumas that
would normally be present, such as the finding of
a stone in one’s shoe or a cut to the skin without
feeling pain or discomfort. People frequently
describe a relative feeling of unsteadiness, which
can be a manifestation of pain, tingling, numb-
ness, or instead from a sensory ataxia in which
either foot proprioception is affected or the stabi-
lizing effect of feeling the ground. Similarly, a
nonspecific feeling of weakness is a very com-
mon descriptor, likely as a consequence of abnor-
mal sensory symptoms, as frequently this is
described even when motor testing is completely
normal. Motor symptoms resulting from muscle
weakness can occur in more advanced diabetic
DSP with impairment of foot dorsiflexion and
plantar flexion that may be described as an inabil-
ity to perform activities that require standing on
the forefoot or on extended toes or heal-walking.
Hyperalgesia and allodynia as manifestations of
abnormal sensory processing are less common.
Hyperalgesia refers to a painful stimulus that
B. A. Perkins and V. Bril
feels exaggerated, while allodynia refers to a nor-
mal stimulus that is perceived as painful such as
severe pain induced by rubbing the toes along the
bedsheets. While uncommon in diabetic DSP
without first experiencing sensory symptoms,
symptoms of autonomic dysfunction such as dry-
ness of skin from sudomotor neuropathy, or the
symptoms of gastroparesis, enteropathy, or clini-
cal cardiac autonomic neuropathy with postural
lightheadedness and syncope may occur. Finally
the symptoms of foot complications like infec-
tion, ulceration and Charcot deformity may be
present symptoms in advanced neuropathy.
5 Neuropathic Signs
A directed neurological examination for diabetic
sensorimotor polyneuropathy (DSP) requires
examination of the lower limbs starting distally at
the toes [33]. As with symptoms and depending
on stage and severity, the signs may be a reflexion
of small fiber dysfunction, large fiber sensory
dysfunction, large fiber motor weakness, pres-
ence of clinical autonomic abnormalities such as
sudomotor neuropathy, and the manifestations of
foot complications (Table 2). Appearance of the
skin and feet can reveal dryness and color
changes, loss of hair, clawing and deformity of
the toes, and ulcerations at late stages. Clinical
sensory function is assessed for small (thermal
sensitivity, pin prick), large (vibration, proprio-
ception), and mixed (light touch) fiber modali-
ties. These sensory tests are performed by first
applying the stimuli to the sternum or forehead so
that the person can appreciate the normal sensa-
tion, and then distally at the first toe, or adjacent
toe, in case of amputation. To determine severity,
the stimuli are then moved proximally to provide
a level at which the sensory function becomes
normal. For vibration, a 128  Hz tuning fork is
sufficient and must be applied on bony surfaces.
Alternatively, small pocket-sized battery-­
operated electronic devices that standardize
vibration can be used [34]. For proprioception,
the distal phalanx is held on the lateral surfaces
and moved upwards and downwards in small
movements. For pinprick, a disposable sterile
Clinical Diagnosis of Diabetic Peripheral Neuropathy
sharp metal tip can be used [35]. For thermal sen-
sitivity, cold sensation can be assessed with a
steel tuning fork cooled under running cold
water, or alternatively, with a simple pen-like
device that has a cool metallic end and a warmer
plastic end for comparison [36]. For light touch,
a cotton wisp may be used. The 10 g monofila-
ment is a nylon thread affixed to a handle that
bends at a standardized pressure of 10 g,
­representing a combination of touch and pressure
sensation. Other physical examination tests such
as 2-point discrimination, or the tactile circum-
ferential discriminator to evaluate a person’s abil-
ity to differentiate stimuli of different diameters,
are not commonly used for assessment of dia-
betic DSP.
Motor function is evaluated by muscle bulk
and power of the small foot muscles, and specific
strength on foot dorsiflexion and plantarflexion
are generally tested. Strength in other lower limb
groups should also be assessed (knee and hip
movements). The deep tendon reflexes need to be
performed with the person fully relaxed and knee
and ankle reflexes should be assessed. Gait and
ability to perform a tandem gait, Romberg test for
sensory ataxia, and also ability to walk on heels
and toes would complete the lower limb exami-
nation. Upper limbs should be examined depend-
ing on lower limb findings [33].
From a diagnostic perspective, the principal
purpose of the physical examination is to deter-
mine, through the process of clinical assessment,
if the various signs are in keeping with the clini-
cal pattern of diabetic DSP through a process,
with each finding, of incrementally revising clini-
cal estimates of disease probability. For example,
as discussed later, signs such as motor predomi-
nance over abnormal sensory tests, asymmetry,
and lack of length-dependence are findings that
substantially decrease the probability of diabetic
DSP. This is of particular importance because it
is known that individual physical examination
signs are known to have very poor inter- and
intra-rater reproducibility, even in the hands of
expert neurologists [37]. This can be partially
overcome by the clinician having a broad, sys-
tematic physical examination including multiple
physical examination maneuvers as described, as
75
well as by the performance of objective confir-
matory tests if diagnostic uncertainty remains
(chapter “Diagnostic Techniques for Diabetic
Peripheral Neuropathy”).
Though research into implementation is lim-
ited, we make brief mention of certain point-of-­
care devices, defined as simple, rapid, lab-quality
diagnostic devices for use by clinicians at the
moment of clinical care for diagnosis. Some are
an extension of the physical examination (battery-­
operated vibrating device in place of a tuning
fork, temperature discrimination devices, or the
“Neuropad” paper to identify sweat function)
[34–36, 38, 39], and, like sensory physical exam-
ination, many are heavily dependent on subjec-
tive patient responses. Objective sudomotor
function assessment by way of electrochemical
skin conductance devices have shown conflicting
evidence in the literature [40–43], but may be
amenable to implementation into practice for
screening [44]. Simplified point-of-care nerve
conduction devices for sural nerve amplitude and
conduction velocity are reproducible and valid,
and in future may have impact on clinical
decision-­making as well as implementation into
practice [45–49].
6 Composite Symptom Scales,
Sign Scales, and Combined
Scales
We have introduced thus for in this chapter the
essential components of clinical evaluation,
mostly representing symptoms and signs that
represent effective clinical practice but that have
not individually been evaluated in investigative
test research for their diagnostic performance.
However, we then introduced some specific
screening maneuvers that have concurrent valid-
ity for identifying the presence of DSP such as
the Ipswich Touch Test adopted by Diabetes UK
[27] and versions of pain, vibration, light touch
sensation [28, 29], as well as the example of a
variation of the 10-g monofilament examination
that may permit prediction of future onset of DSP
[8]. Investigative test research is broad, though,
as these tests can include single symptoms, signs,
76
laboratory parameters, as well as composite scor-
ing systems [50]. As individual symptoms or
signs are likely not sensitive or specific enough to
serve as a biomarker response, for example, to a
therapeutic intervention in a clinical trial, one
strategy is to implement a composite score, such
as a self-reported symptom questionnaire or the
clinical scales that make up a focus of this
Chapter [51]. While these have been developed
primarily as a way to summarize the proportions
and severity of diabetic DSP in trial or cohort
study populations, these composite scales can
serve four key purposes to the clinician [22, 50].
First, they incorporate (and remind the clinician
of the) multiple aspects of the manifestations of
diabetic DSP, rather than focusing on one in iso-
lation. Second, they can provide a diagnostic
threshold for concurrent diagnosis, and, though
they have not yet been evaluated in this way, for
prediction of future onset of diabetic DSP or its
related foot complications. Third, the quantitative
scores allow a measure of diabetic DSP severity
that is typically not conferred by the result of a
single symptom or sign. Finally, many identify
pain and its severity and can therefore be used to
guide therapy for painful diabetic DSP.
7 The Michigan Neuropathy
Screening Instrument (MNSI)
First published in 1994 [52], the Michigan
Neuropathy Screening Instrument (MNSI) has
had major impact on the identification of diabetic
DSP, as well as providing a measure of its sever-
ity, and, potentially also an ability to track pro-
gression and regression in many observational
studies and clinical trials [21, 53–56]. The MNSI
is, in fact, made up of two separate assessments.
First, a questionnaire (symptom scale) compo-
nent, has major value for screening as it is a self-­
administered questionnaire made up of 15 items.
These include questions such as having received
a diagnosis by a physician in the past, or a history
of amputation or ulceration, dry skin, but also
about neuropathic symptoms such as numbness,
pain, temperature sensation, and allodynia.
Furthermore, there are two questions that are
B. A. Perkins and V. Bril
unscored, including generalized weakness and
cramping, that can help the rater understand other
features that may not relate directly or specifi-
cally to DSP and foot complications risk factors.
The second component is a five-part lower
extremity physical examination made up of
examination for abnormal appearance including
presence of callus or deformity, examination for
ulceration, ankle reflexes, and simple 3-level
scores using vibration sensation and the mono-
filament (present, reduced, absent).
7.1 Scale Face Validity
Both the MNSI Questionnaire and MNSI
Examination scales incorporate attributes that relate
specifically to polyneuropathy, however, they test
the presence of other attributes including the pres-
ence of foot deformities. Whether the person has
polyneuropathy or not, the presence of foot abnor-
malities such as callus, fissures, ulceration likely
places them at high risk of foot complications.
Though less specific to diabetic DSP, these features
increase its clinical relevance as an outcome mea-
sure. For polyneuropathy-­ specific research, the
manifestations especially on foot examination may
not be sensitive enough to identify those at earlier
stages of polyneuropathy alone, given that the sen-
sory tests are graded in three simple categories.
7.2 Test Quality and Reliability
The MNSI itself has not been tested formally
using standard methods and metrics for inter- and
intra-rater reproducibility. Studies on variations of
the MNSI show suitably high reliability [57, 58].
7.3 Diagnostic Accuracy:
Concurrent
The original publication, representing analysis of a
small cross-sectional group of type 1 and type 2 par-
ticipants who had diabetic DSP confirmed by nerve
conduction studies, undertook a simple tabulation
suggesting that an MNSI Questionnaire score of 7
Clinical Diagnosis of Diabetic Peripheral Neuropathy
or greater, and an MNSI Examination score exceed-
ing 2, are associated with extremely high predictive
values positive, good sensitivity and extremely high
specificity [52]. However, a diagnostic study design
conducted in a type 1 diabetes observational cohort
that also operationalizing nerve conduction studies
to define ­diabetic DSP cases, suggested that these
particular thresholds were associated with substan-
tially lower diagnostic performance, and suggested
alternate thresholds [51]. First, a threshold score of
4 or greater on the MNSI Questionnaire, such a test
result would carry 40% sensitivity and 92% speci-
ficity. Second, an MNSI Examination result exceed-
ing a score of 2 carried 61% sensitivity and 79%
specificity. However, on inspection of the receiver
operating characteristic curves from this publica-
tion, we observe that an MNSI Questionnaire score
(such as a score of 1 or greater) could reasonably be
chosen to maximize sensitivity to nearly 60%.
7.4 Diagnostic Accuracy:
Predictive
While many longitudinal designs have been con-
ducted, and even some cohort designs (in which
those with neuropathy at baseline have been
excluded), we were unable to find in the literature
analyses that explored whether a certain test
score could predict future onset of diabetic
DSP. While this does not make inherent sense for
the MNSI examination scores, which measure
manifestations of established diabetic DSP (such
as presence of ulcer or complete reduction or
abnormality—rather than subtle impairments—
of vibration or monofilament sensitivity), such a
design could be applied to the MNSI
Questionnaire that has components that are suited
to earlier, pre-diagnostic stages of diabetic DSP.
7.5 Effect on Treatment Decisions,
Impact on Patient Outcomes,
Economic Analysis
The results of MNSI testing have not been studied
on their effect on the treatment decisions made by
clinicians, or whether MNSI testing is associated
77
with improved outcomes relative to current stan-
dards of screening and diagnosis. However, it has
been frequently operationalized as a neuropathy
outcome in large-scale clinical trials including
those evaluating cardiovascular and microvascu-
lar outcomes in type 1 and type 2 diabetes [21,
53–56]. As many of these studies required a mea-
sure of more advanced diabetic foot disease, the
MNSI has provided a successful proof of concept
for applying neuropathy scales in trials.
8 Toronto Clinical Neuropathy
Score (TCNS)
The Toronto Clinical Neuropathy Score (TCNS),
first validated and reported in 2002, was devel-
oped specifically for use in a study of simple
screening measures for DSP that could be used in
the family physician’s office or the diabetes clinic
[29, 59]. Spectrum bias is a very common design
flaw (a source of selection bias) in investigative
test research. It represents inappropriate recruit-
ment of many participants for whom there is no
diagnostic uncertainty, and recruitment of partici-
pants with limited variation in disease characteris-
tic [22]. The original purpose of the TCNS was to
create a quantitative score that ranged across all
diabetic DSP Stages (Table  1) that would allow
for stratified accrual across the full spectrum of
neuropathy to appropriately determine the diag-
nostic performance of simple screening measures
[29]. The scale comprises six symptoms (present/
absent), knee and ankle reflexes (normal, reduced,
or absent scored bilaterally) and sensory tests
(normal/abnormal) as shown in Table 3, Panel A
and ranges from a perfectly normal score of 0
without clinical evidence of DSP to a maximum
abnormal score of 19 points. The symptoms are:
foot pain, numbness, tingling, weakness, unsteadi-
ness, weakness, and presence of upper limb
symptoms. The sensory signs include sensory
tests of pinprick, temperature, light touch, vibra-
tion, and proprioception. The deep tendon reflexes
include the knee and ankle reflexes. Out of a total
score of 19, the grades are defined as follows: 0–5
= no neuropathy; 6–8 = mild neuropathy; 9–11 =
moderate neuropathy; ≥ 12 = severe neuropathy.
78 B. A. Perkins and V. Bril

Table 3  The components of the original Toronto Clinical 8.2 Test Quality and Reliability
Neuropathy Score (TCNS) and the modified TCNS
(mTCNS)
Formal evaluation of reliability reported internal
Sensory test Reflex
Symptom scores scores scores
consistency of attributes, and very good to excel-
A. Scoring of the Toronto clinical neuropathy score lent inter- and intra-rater reproducibility [60].
(TCNS) While in opposition to the findings of poor repro-
Foot pain Pinprick Knee ducibility of individual physical examination
Numbness Temperature reflexes [37], reproducibility is likely greater with investi-
Tingling Light touch Ankle
Weakness Vibration reflexes
gative tests that make up a composite of multiple
Unsteadiness Position sense Reflexes individual tests.
Upper limb Sensory test graded as
symptoms scores graded as  0 =
Symptom scores  0 = normal normal
graded as  1 = abnormal  1 =
8.3 Diagnostic Accuracy:
 0 = absent reduced Concurrent
 1 = present  2 =
absent Diagnostic accuracy has been evaluated exten-
Bilaterally
sively, including relationship with nerve conduc-
Maximum TCNS Score 19 points
tion studies, diabetic DSP classified by signs,
B. Adaptations in scoring to create the modified TCNS
(mTCNS) symptoms, and nerve conduction studies, and
Symptom scores Sensory test Not scored even relative to the results of fiber density on
graded as scores graded as sural nerve biopsies [59, 60]. Furthermore, its
 0 = absent  0 = absent (0) validity has been tested in non-diabetic forms of
 1 = present  1 = present at
without interfering level of the polyneuropathy [61]. In a type 1 diabetes cohort,
with sense of toes TCNS had receiver operating characteristic curve
well-being or  2 = present at area under the curve of 0.86 for identification of
activities level of the Stage 2 diabetic DSP, with an optimal threshold
 2 = present and ankles
interfering with  3 = present at value of 6 or greater identifying DSP with sensi-
well-being but not a level tivity and specificity each approximating 80%
activities proximal to [51]. Additionally, very good overall accuracy
 3 = present and the ankles was achieved even for the identification of sub-
interfering with
well-being and clinical Stage 1a neuropathy, with an area under
activities the curve of 0.74 [62]. However, its diagnostic
Maximum mTCNS score 33 points performance in type 2 diabetes, along with objec-
Adapted from Bril et al. [59, 60] tive tests, was lower [52].

8.1 Scale Face Validity 8.4 Diagnostic Accuracy:

The TCNS has attributes that relate specifically
to manifestations of small and large fiber dys-
function of polyneuropathy, excluding other
components such as foot deformity that raise risk
for foot complications. The broad scoring range
that includes six points for symptoms and 13 for
signs may permit greater sensitivity for earlier
stages, though individual test components are
assigned binary or three-level ordinal scores
(Table 3).
Predictive
A pilot longitudinal study that included baseline
TCNS indicated subtle baseline differences in
scores between those with future incident diabetic
DSP and those who remained neuropathy-­free,
but it did not appear to have diagnostic predictive
validity [63]. Though not conclusive, this implies
that the main role of the TCNS, like other scales,
is optimized for identification rather than on pre-
diction of future neuropathy incidence.
Clinical Diagnosis of Diabetic Peripheral Neuropathy
8.5 Effect on Treatment Decisions,
Impact on Patient Outcomes,
Economic Analysis
Similar to the MNSI, the TCNS has not been
studied for effect on the treatment decisions made
by clinicians or screening and diagnosis out-
comes, but it has frequently been operationalized
as a neuropathy outcome in clinical trials and
cohort studies [44, 64–68]. The trials, however,
are studies of neuropathy therapies as opposed to
the large-scale cardiovascular outcome trials that
the MNSI has been used as a neuropathy out-
come measure.
9 Modified Toronto Clinical
Neuropathy Score (mTCNS)
With the intention of creating a scale that is more
sensitive for earlier stages of neuropathy, the
TCNS has been adapted to the modified TCNS
(mTCNS). The modified scale omits the deep
tendon reflex assessment, which was a somewhat
less reproducible part of the TCNS [60], and that
additionally represented later (motor) dysfunc-
tion on the spectrum of diabetic DSP severity.
The mTCNS also adds greater quantitative levels
to both symptoms and sensory signs to create a
larger measurement range across the earlier DSP
stages, and a more granular scale. The symptoms
of the TCNS are graded according to patient-­
reported outcome principles as absent (0), pres-
ent but not interfering with activities or sense of
well-being (1), present and interfering with sense
of well-being but not with activities (2), and pres-
ent and interfering with activities (3). The signs
of the TCNS are graded as absent (0), present at
the toes (1), present up to the ankles (2), and pres-
ent proximal to the ankles (3) in a stocking distri-
bution. The mTCNS score ranges from 0 to 33, as
detailed in Table 3, Panel B.
9.1 Scale Face Validity
The mTCNS has modifications that, compared to
the TCNS, exclude motor manifestations that con-
79
fers a measure of earlier stages of diabetic DSP,
and expands the scoring of symptoms and sensory
signs to include test components that are assigned
four-level ordinal scores each (Table 3, Panel B).
9.2 Test Quality, Reliability,
and Accuracy
The very good to excellent inter- and intra-rater
reproducibility of the mTCNS was formally eval-
uated and reported along with a new evaluation
of the original TCNS in a multicenter study [60].
Furthermore, as hypothesized, scores shared
lower correlation with the TCNS as they measure
different stages of diabetic DSP (Pearson coeffi-
cient 0.56), but validity for identification of Stage
2 neuropathy remains similar [60]. However, a
specific threshold has not been independently
assessed for the concurrent validity of diabetic
DSP, and also ranges of severity have not been
assigned as for the TCNS.
9.3 Effect on Treatment Decisions,
Impact on Patient Outcomes,
Economic Analysis
Similar to the MNSI and the TCNS, the mTCNS
has not been studied for effect on the treatment
decisions made by clinicians or screening and
diagnosis outcomes, but it has frequently been
operationalized as a neuropathy outcome in clini-
cal trials and cohort studies that by design focus
on earlier stages of diabetic DSP [69, 70].
Furthermore, it is being implemented in contem-
porary studies of populations at risk of earlier
stages of neuropathy (https://clinicaltrials.gov/
ct2/show/NCT04664426).
10 Neuropathy Impairment
Score of Lower Limbs
(NIS-LL)
The NIS-LL is an extensive composite score of
neurological signs and objective tests in the lower
limbs, excluding symptoms, that was designed
80
primarily for implementation in clinical trials
[71]. This physical examination score was devel-
oped by investigators at the Mayo Clinic as a
simplification of an even more extensive score,
termed the NIS-LL+7 that incorporated the
results of seven objective tests including vibra-
tion perception testing, heart rate variability with
deep breathing as a measure of autonomic dys-
function, and five nerve conduction study param-
eters. This extensive version is considered to
have 100% sensitivity for Stage 1a diabetic DSP
[72]. The NIS-LL clinical scale version (exclud-
ing the seven additional objective tests) incorpo-
rates extensive motor, sensory, and reflex physical
examination maneuvers with granular scoring
systems. For example, grading of muscle power
involves eight discrete levels ranging from 0 to 4,
including quarter points. The overall scale is
graded from a perfectly normal score of 0 to a
maximum of 88 points [71].
10.1 Scale Face Validity
The NIS-LL assigns 64 of the 88 points to muscle
strength attributes that are frequently absent
when first clinically identified by clinicians, and
is therefore at face value a measure of later stage
(generally Stage 1b). The potential score for sen-
sory testing is 16/88 points with a maximal loss
of sensory modalities of touch pressure, pinprick,
vibration (using the 165Hz tuning fork), and joint
position. The reflexes at knees and ankles are
considered as present, reduced or absent, and
defined according to age limits adding potential
confusion to non-expert examiners. Both in terms
of the later stage diabetic DSP identification and
the expertise required for examination, the
NIS-LL is not seen as a tool that can be translated
to clinical diagnosis, but rather as a specialize
tool for interval cohort studies and clinical trials.
Given its wide spectrum of motor scores, at face
value it might represent a sensitive measure for
neuropathic change in clinical trials examining
the effect of interventions on more advanced neu-
ropathy stages.
B. A. Perkins and V. Bril
10.2 Test Quality, Reliability,
and Accuracy
The NIS-LL has not been evaluated specifically
for reproducibility or concurrent validity for a
diagnosis of diabetic DSP. However, it was evalu-
ated in a comparative study of neuropathy associ-
ated with impaired glucose tolerance subjects
and in this setting—as with the other scales eval-
uated—its overall diagnostic accuracy was excel-
lent [73].
10.3 Effect on Treatment Decisions,
Impact on Patient Outcomes,
Economic Analysis
Similar to the previous composite scores, the
NIS-LL has not been studied for effect on the
treatment decisions made by clinicians or
screening and diagnosis outcomes as it primar-
ily represents a research tool. However, it has
been extensively applied as a neuropathy out-
come in clinical trials and cohort studies that by
design focus on later stages of diabetic DSP
[74–77].
11 Utah Early Neuropathy Scale
This physical examination scale was developed
specifically to serve the need for measurement
scales that could detect early stages of sensory-­
predominant polyneuropathies such as DSP at
Stage 1b [78]. The scale includes limited scores
for motor function (great toe extension and
ankle reflexes), and greater proportional contri-
bution to sensation for pin and vibration as well
as allodynia. The scale varies from 0 to 42 with
24/42 points for pin sensation, 8/42 for vibra-
tion, and 2/42 for allodynia. To demonstrate an
example of a score for earlier sensory manifes-
tations and how it is operationalized, we include
the scoring for the Utah Early Neuropathy Scale
in Fig. 3.
Clinical Diagnosis of Diabetic Peripheral Neuropathy 81

Utah Early Neuropathy Scale

Motor Examination Right Left

0 normal

2 weak
Great Toe Extension
Total Both Sides (out of 4): Segments for pin sensation reporting

Right Leg Left Leg

Pin Sensation: Right Left
0 normal 6 6

1 for each segment with

reduced sensation 5 5
2 for each segment with
absent sensation 4 4
3 3
Total out of both sides: (out of 24):
2 2
1 1

Allodynia/Hyperesthesia Right Left

0 normal
1 if present In toes or foot

Total for both sides (out of 2):

Large Fieber Sensation Right Left Deep Tendon Reflex Right Left
0 normal 0 normal
1 diminished 1 diminished
2 absent 2 absent
Great toe vibration Ankle
Time s s Total both sides (out of 4):
Great toe joint position

Total both sides (out of 8): Total Score (out of 42)

Fig. 3  Performing the Utah Early Neuropathy Scale
(UENS) Examination. The UENS requires a number 2
(13/4 inch) safety pin and a 128 Hz tuning fork. Pin sensa-
tion is tested by first reviewing normal sharp sensation to
pin on an unaffected portion of the skin. Once this is
established, touch the dorsal surface of the foot and leg
with the pin, working centripetally from the great toe in
1–2  cm increments while asking the subject to respond
when they first feel “any sharpness,” and again more prox-
imally when the pin feels “as sharp as they would expect.”
Repeat to firmly establish these levels. On each side, two
points are scored for each region in which the patient fails
to feel any sharpness. One additional point is scored for
each additional region in which the pin feels less sharp
than expected. Only distal sensory loss is scored. So, for
instance, a person who reported absent pin sensation to
the mid foot dorsum (four points) and reduced sensation
to the low ankle (1 point) bilaterally would score a total of
ten points for this portion of the UENS. Vibration is tested
by first acquainting the subject with vibration (as opposed
to pressure) sensation, then holding the maximally
vibrated tuning fork to the dorsum of the great toe at the
distal interphalangeal joint. Extinction of vibration in less
than 10 s is considered “diminished,” while “absent”
requires that the patient cannot detect the maximally
vibrating tuning fork at the toe. The motor examination is
limited to great toe dorsiflexion. Other aspects are as typi-
cally performed in neurological examination. Figure cour-
tesy of A.  Gordon Smith, co-creator of the Utah Early
Neuropathy Scale. The Scale is further detailed in [78]
82 B. A. Perkins and V. Bril

11.1 Scale Face Validity 12 Neuropathy Symptom Score

(NSS)
The scale does not include symptoms, and there-
fore in the context of diabetic DSP would be ame- First described in 1993 in a descriptive study of
nable for use as a screening tool to identify Stage the prevalence of the hospital-based clinic diabe-
1b disease. Furthermore, the relative weight of tes population in the United Kingdom [85], the
sensory signs is high, including small fiber func- Neuropathy Symptom Score was an interviewer-­
tion, such that variation of scores likely represent based questionnaire meant to evaluate the pres-
earlier-stage neuropathy when interventions and ence and severity of specific polyneuropathy
therapies are most likely to be of benefit. manifestations, with a particular emphasis on
painful diabetic DSP.  In the interview, patients
are asked about their experience of pain or dis-
11.2 Test Quality, Reliability, comfort in the legs. Characteristics are evaluated
and Accuracy such that burning, numbness or tingling in the
legs provides two points, and fatigue, cramping
Reproducibility has not been evaluated cross-­ or aching 1 point. Location provides 2, 1, and 0
sectionally within and between examiners, but the points for feet, calves, and elsewhere, respec-
1-year change in score showed very high consis- tively. Nocturnal symptoms provide 2 further
tency and has been reported as an indication of points, 1 if also daytime, 0 if only daytime. An
excellent reproducibility [78]. Additionally, it cor- additional point was assigned if patients are awo-
related with other scoring systems. For Stage 2 ken at night from symptoms. Finally, alleviating
polyneuropathy confirmed by nerve conduction factors are scored: 2, 1, and 0 if alleviated by
studies, in a study sample with a very high propor- walking, standing, or laying down. With a maxi-
tion of polyneuropathy participants it showed very mum of 9, it is implied that symptoms are present
good overall concurrent validity by area under the if a minimum score of 3 is achieved. Severity was
receiver operating characteristics curce [78]. This graded as mild (score of 3–4), moderate (score of
level of accuracy appeared to be similar to other 5–6), or severe (score of 7–9).
scales [78], and it has been examined in other
causes of polyneuropathy [79, 80]. However, it is
our view that this scale may show advantage in 12.1 Face Validity
study samples that reflect general diabetes practice,
with lower prevalence of advanced polyneuropathy The score, at face value, weighs heavily on pain-
and higher prevalence of early disease [81]. ful diabetic DSP and its consequences, including
night-time symptoms and awakening, and the
alleviating symptoms in the context of such
11.3 Effect on Treatment Decisions, awakening. Consequently, its primary role is in
Impact on Patient Outcomes, the classification of the presence and severity of
Economic Analysis painful diabetic DSP.

Similar to the previous composite scores, the

Utah Early Neuropathy Scale has not been stud-
ied for effect on the treatment decisions made by
clinicians or screening and diagnosis outcomes
as it primarily represents a research tool.
However, it has been extensively applied as a
neuropathy outcome in cohort studies and clini-
cal trials that by design focus on earlier stages of
diabetic DSP [82–84].
12.2 Test Quality and Reliability
While not tested formally for reproducibility,
the NSS has served as a historical benchmark
for the development of updated symptom scores
that correlate highly with it, and that them-
selves have excellent reproducibility in small
studies [86].
Clinical Diagnosis of Diabetic Peripheral Neuropathy
12.3 Diagnostic Accuracy:
Concurrent
During the development of the NSS, studies of its
intrinsic validity were not conducted. However, it
has been evaluated by independent groups pri-
marily in combination with a physical examina-
tion scale termed the Neuropathy Disability Scale
that is discussed below [87, 88].
13 The Revised Neuropathy
Disability Score (NDS)
Developed as a simplification of the NIS-LL and
also first described in 1993 along with the
Neuropathy Symptom Score in a descriptive
study of neuropathy prevalence, the score incor-
porates small fiber sensory attribute, and large
fiber sensory and motor attributes. It involves
examination of the ankle reflex, vibration, pin-­
prick and temperature (cold tuning fork) sensa-
tion at the great toe. The sensory modalities are
scored as either present = 0 or reduced/absent = 1
for each side, and reflexes as normal = 0, present
with reinforcement = 1 or absent = 2 per side.
With a maximum abnormal score of 10, classifi-
cation of severity is graded as mild (score of
3–5), moderate (score of 6–8), and severe (score
of 9 or 10). Generally, a score of 6 or more has
been used to define presence of signs for identifi-
cation of DSP [89, 90].
13.1 Face Validity
This score includes small fiber sensory attributes
that could identify pre-symptomatic or symptom-
atic early-stage DSP, as well as later large-fiber
stages owing to sampling of vibration sensation
and reflexes.
13.2 Test Quality and Reliability
There are no published studies examining intra-
and inter-rater reproducibility of the revised
NDS.
83
13.3 Diagnostic Accuracy:
Concurrent
During the development of the revised NDS,
studies of its intrinsic validity were not con-
ducted. However, it has been partially evaluated
by independent groups [87, 88]. While there are
no cross-sectional studies evaluating the optimal
threshold for the identification of diabetic DSP, a
score of 6 or more has traditionally been used in
studies in which it is the reference standard [89,
90].
13.4 Diagnostic Accuracy:
Predictive
While many longitudinal designs have been con-
ducted, we were unable to find in the literature
analyses that explored whether a certain test
score in those without DSP could predict future
onset of diabetic DSP.  However, among those
followed in a diabetic foot clinic (and likely had
presence of diabetic DSP), the revised NDS was
generally associated with short-term risk of inci-
dent ulceration, though in this study population
the best prediction was associated with abnor-
mality in foot pressure sensation measurement
[91]. In combination with the NSS, the revised
NDS has been used by several research groups
for the identification of higher risk patients for
diabetic DSP, and as the clinical sign and symp-
tom measures to define presence of diabetic DSP
[92, 93].
14 Perspectives
on the Comparison of Scales
Each of the scales chosen for review in this chapter
has been designed with specific research purposes
in mind. At face value they reflect different aspects
of diabetic DSP (from early sensory signs in the
Utah Early Neuropathy Scale, to those that include
advanced foot deformity and ulceration like the
Michigan Neuropathy Screening Instrument
Examination). As summarized in the above sec-
tions and in Table 4, they have different levels of
Table 4  Composite symptom scales, sign scales, and combined scales: general attributes and comparisons
84

Acronym Considerations for face validity Considerations for technical performance

Symptom or Weighted toward Reproducibility Diagnostic Accuracy
signs earlier (sensory) (concurrent validity)
Granularity components or later (motor) For NCS-based
Scale name (range of scores) stages reference standard Other considerations
Michigan MNSI 0–15 Symptoms only Later Not formally Low Can be used as a self-
neuropathy questionnaire assessed administered questionnaire
screening [52] for screening in virtual care
instrument
Michigan MNSI 0–10 Signs Later Not formally Good [51] Operationalized in
neuropathy examination [52] assesseda cardiovascular disease
screening intervention trials [21,
instrument 53–56]
Toronto clinical TCNS [59] 0–19 Both Later Very good to Very good for T1D, Its validation has been
neuropathy score excellent [60] good for T2D extensive, including
agreement with NCS, disease
severity, and sural nerve
biopsy nerve fiber density
[59, 60]
Evaluated in multicenter
study [60]
Modified Toronto mTCNS [60] 0–33 Both Earlier Very good to Correlated with NCS, In a study comparing
clinical excellent [60] not formally assessed multiple scales in the setting
neuropathy score of pre-diabetic neuropathy,
this scale had the highest
numerical sensitivity and
specificity compared to
others [73, 94]
Neuropathy NIS-LL [71] 0–88 Signs Later Not formally Very goodb [75] This represents the signs
impairment score assessed component of the NIS-LL+7,
of lower limbs a scale that includes seven
objective small and large
fiber tests [72]
Diagnostic accuracy
evaluated in pre-­diabetic
neuropathy [73]
B. A. Perkins and V. Bril
Utah early UENS [78] 0–42 Signs Earlier Not formally Very goodb [75] The score incorporates
neuropathy score assessedc multiple levels of sensory
deficit in the lower limbs (see
Fig. 2)
Neuropathy NSS [85] 0–9 Symptoms only Both (pain) Not formally Not formally assessed Interview-based
symptom score assessed on its on its own [87, 88] questionnaire
own Focused on painful
neuropathy and its
consequences
Revised NDS [85] 0–10 Signs Earlier Not formally Not formally assessed Associated with short-term
neuropathy assessed on its on its own [87, 88] risk of ulceration [91]
disability score own
revised
Note that for all scales different diagnostic thresholds have been proposed depending on specific outcomes and specific settings. For example, for screening a more sensitive
threshold is preferred, for diagnosis a more specific threshold is preferred. Many of these scales have been incorporated into reference standard definitions for diabetic DSP that
use confirmatory nerve conduction studies. In those settings, generally a score of only 1 or more is required to support the presence of a symptom or a sign [9]
NCS nerve conduction studies, T1D type 1 diabetes, T2D type 2 diabetes
a
Modified versions have been studied [57, 58]
Clinical Diagnosis of Diabetic Peripheral Neuropathy

b
Study sample had very high proportion with diabetic DSP
c
One-year change was evaluated and reproducibility was high, but generally change over time is a different attribute of ability to detect change in nerve function rather than
reproducibility
85
86
research published on the hierarchy of investiga-
tive test research principles, studies of validity
include different study samples, study sizes, preva-
lence of neuropathy, spectrum of disease, variable
reference standard definitions. Even with harmo-
nization of such procedures, systematic review
and meta-analysis of diagnostic tests are complex
as they require Bayesian techniques like clinical
trial network metanalysis methods [95]. Though
we present their attributes and critical appraisal in
detail, for these methodological reasons we inten-
tionally do not present a quantitative side-by-side
comparison of their diagnostic operating charac-
teristics. One research group has undertaken a
direct comparison of many of these scales in the
setting of pre-diabetic neuropathy and, while they
imply that some are more valid than others (the
modified TCNS), uniformly in that study most
scales had unexpectedly high levels of concurrent
diagnostic accuracy [73]. We challenge clinicians
to consider operationalizing these scales in clinical
setting to suit specific needs: For example, the
self-­administered Michigan Neuropathy Screening
Instrument questionnaire, the Neuropathy
Symptom Scale, or the symptom subscales of
other composite scores like the modified Toronto
Clinical Neuropathy Score, can be applied for vir-
tual diabetes care as has been required during the
severe acute respiratory syndrome coronavirus 2
(SARS-CoV-2) pandemic, while others could be
selected for physical examination screening and
evaluation in the clinical setting.
15 Differential Diagnosis:
A Diagnosis of Diabetic DSP
Requires Consideration
of Other Causes
of Polyneuropathy
A clinical evaluation consistent with the features
of diabetic DSP requires consideration of other
etiologies for distal and symmetric polyneuropa-
thies. First, the presence of atypical features
listed in Table 2 should be identified, as they may
further indicate other causes of peripheral nerve
damage other than polyneuropathy, such as focal,
multifocal, or nonlength dependent forms of neu-
B. A. Perkins and V. Bril
ropathies. These include focal nerve entrapments
and other non-distal and non-symmetric forms
(chapters “The Epidemiology of Diabetic
Neuropathy” and “Treatment Induced Neuropathy
of Diabetes (TIND)”), rare conditions such as
hereditary neuropathy with propensity to pres-
sure palsies (HNPP), and polyneuropathies with-
out sensory predominance such as common
forms of chronic inflammatory demyelinating
polyneuropathy (CIDP) and its variants.
Furthermore, consideration should be made for
factors associated with potentially similar poly-
neuropathy to diabetic DSP such as alcohol,
hypothyroidism, toxins, and drugs. With this in
mind, a simple laboratory panel should be under-
taken: comprehensive metabolic panel to deter-
mine current metabolic risk factors such as
glycemic exposure, presence of hypertriglyceri-
demia, renal dysfunction; thyroid hormone to
indicate hypothyroidism; complete blood cell
count may indicate macrocytosis associated with
alcohol or nutritional deficiencies and infectious
causes; the B12 level itself; erythrocyte sedimen-
tation rate potentially associated with inflamma-
tory causes; and serum protein
immune-electrophoresis to suggest malignancy,
monoclonal gammopathy, or gammopathy asso-
ciated with rare syndromes like Polyneuropathy,
Organomegaly, Endocrinopathies, Monoclonal
protein, and Skin lesions (including hemangi-
oma, hyperpigmentation, hypertrichosis or
Raynaud’s phenomenon as part of the “POEMS”
Syndrome). Testing for infectious causes of poly-
neuropathy should be guided by history, as would
consideration of familial forms of DSP.  It is
important for the clinician to recognize that the
patient with a consistent chronic, insidious,
sensory-­predominant distal and symmetric clini-
cal history and physical examination for diabetic
DSP typically only requires this focused labora-
tory work-up, and that the most common causes
of DSP other than diabetes are B12 deficiency,
alcohol, and hypothyroidism, and diabetic DSP
exacerbated by renal failure. Referral to special-
ist neurology should be strongly considered in
the uncommon case of atypical features (Table 2),
or if rare causes are suspected (Table  5) that
require confirmatory work-up that specialized
Clinical Diagnosis of Diabetic Peripheral Neuropathy 87

Table 5  An non-exhaustive list of diagnoses on the differential of diabetic distal symmetric polyneuropathya
Metabolic and nutritional disorders
 B12 vitamin deficiency (more common)
 Other vitamin deficiencies: pyridoxine, folate, thiamine, vitamin E
 Malabsorption: Bariatric and gastric surgeries, inflammatory bowel disease
 Renal disease (more common)
 Thyroid disease (more common)
 Acromegaly
Inflammatory
 Chronic inflammatory demyelinating polyneuropathy (CIDP) and its variants (may be more common)
Systemic disorders
 Peripheral arterial disease
 Monoclonal gammopathy/paraproteinemia (more common)
 Amyloidosis
 POEMSb
 Sarcoidosis
 Vasculitisa
 Critical illness polyneuropathy
Infectiousa
 HIV
 Hepatitis B
 Lyme
 Syphilis
 Leprosy
Drugs
 Neurologic and mood: phenytoin, amitriptyline, lithium
 Antimicrobials: nitrofurantoin, metronidazole, chloramphenicol, tuberculosis therapies, chloroquine,
hydroxychloroquine
 Cardiovascular: amiodarone, flecainide, hydralazine, nitrous oxide
 Rheumatic: colchicine, gold, leflunomide, methotrexate
 Immunologic: tacrolimus, interferon-α, ipilimumab, nivolumab, pembrolizumab, bortezomib
 Antineoplastic: chemotherapeutic agents, paclitaxel and other taxanes, vinca alkaloids, platinum analogues,
doxorubicin, etoposide, ifosfamide, misonidazole, antinucleosides
Toxins
 Alcoholism (more common, frequently painful neuropathy)
 Heavy metal: lead, arsenic, inorganic mercury, zinc, thallium, gold
 Herbicides: dichlorophenoxyacetic acid, agent orange
 Organophosphate pesticides: parathion, dioxin
 Industrial: acrylamide, polychlorinated biphenyl, vinyl chloride
 Solvents: hexacarbons (glue sniffing), dry cleaning solvents, carbon disulfide, perchloroethylene,
trichloroethylene, triorthocresyl phosphate, ethylene oxide, styrene, toluene, methyl n-butyl ketone, mixed
solvents, and others
Hereditary
 Hereditary neuropathy with propensity to pressure palsies (HNPP)a
 Multifocal acquired demyelinating sensory and motor (MADSAM) neuropathy
 Other hereditary motor, sensory and autonomic neuropathies
a
Some causes listed here are unlikely to present symmetrically. Differential diagnosis lists in the literature frequently
include causes of non-distal and non-symmetrical neuropathies, such as focal entrapment syndromes and other etiolo-
gies discussed in chapters “The Epidemiology of Diabetic Neuropathy” and “Diagnostic Techniques for Diabetic
Peripheral Neuropathy”. We have made an attempt to minimize these conditions on this list, but those indicated with an
asterisk are potential exceptions
b
Polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, skin abnormalities
88
testing might clarify (chapter “Diagnostic
Techniques for Diabetic Peripheral Neuropathy”).
16 Concluding Overview
The clinical diagnosis of diabetic DSP, initiated
either by a screening test or a symptomatic com-
plaint, requires careful consideration of the his-
torical features, symptoms, and signs, where each
finding revises for the clinician an estimate of
disease probability. Composite scores, reviewed
in detail in the chapter, help to remind the clini-
cian of key supportive symptoms and signs for
diagnosis and are commonly used in cohort stud-
ies and clinical trials. Even if the clinical evalua-
tion is consistent with typical features of diabetic
DSP, simple laboratory tests are likely to identify
the presence of other causes. It is essential that
the clinician identify atypical features (listed in
Table 2) or the suspicion of other causes (listed in
Table  5) to seek confirmatory tests and expert
opinion through neurology specialist referral.
References
1. Pop-Busui R, Boulton AJ, Feldman EL, et  al.
Diabetic neuropathy: a position statement by the
American Diabetes Association. Diabetes Care.
2017;40(1):136–54. https://doi.org/10.2337/
dc16-­2042.
2. Manzano GM, Giuliano LM, Nobrega JA.  A brief
historical note on the classification of nerve fibers.
Arq Neuropsiquiatr. 2008;66(1):117–9. https://doi.
org/10.1590/s0004-­282x2008000100033.
3. Carmichael J, Fadavi H, Ishibashi F, Shore AC,
Tavakoli M.  Advances in screening, early diagnosis
and accurate staging of diabetic neuropathy. Front
Endocrinol. 2021;12:671257. https://doi.org/10.3389/
fendo.2021.671257.
4. Wukich DK, Raspovic KM, Suder NC. Patients with
diabetic foot disease fear major lower-­ extremity
amputation more than death. Foot Ankle Spec.
2018;11(1):17–21. (comparative study) (in eng).
https://doi.org/10.1177/1938640017694722.
5. Gregg EW, Li Y, Wang J, et al. Changes in diabetes-­
related complications in the United States, 1990-2010.
N Engl J Med. 2014;370(16):1514–23. (Research
support, U.S.  Gov’t, P.H.S.) (in eng). https://doi.
org/10.1056/NEJMoa1310799.
6. Geiss LS, Li Y, Hora I, Albright A, Rolka D, Gregg
EW.  Resurgence of diabetes-related nontraumatic
B. A. Perkins and V. Bril
lower-extremity amputation in the young and
middle-­aged adult U.S. population. Diabetes Care.
2019;42(1):50–4. (In eng). https://doi.org/10.2337/
dc18-­1380.
7. Hussain MA, Al-Omran M, Salata K, et al. Population-­
based secular trends in lower-extremity amputation
for diabetes and peripheral artery disease. CMAJ.
2019;191(35):955–61. https://doi.org/10.1503/
cmaj.190134.
8. Perkins BA, Orszag A, Ngo M, Ng E, New P, Bril
V.  Prediction of incident diabetic neuropathy using
the monofilament examination: a 4-year prospective
study. Diabetes Care. 2010;33(7):1549–54. https://
doi.org/10.2337/dc09-­1835.
9. Dyck PJ, Albers JW, Andersen H, et al. Diabetic poly-
neuropathies: update on research definition, diagnos-
tic criteria and estimation of severity. Diabetes Metab
Res Rev. 2011;27(7):620–8. https://doi.org/10.1002/
dmrr.1226.
10. Tesfaye S, Stevens LK, Stephenson JM, et  al.
Prevalence of diabetic peripheral neuropathy and its
relation to glycaemic control and potential risk fac-
tors: the EURODIAB IDDM Complications Study.
Diabetologia. 1996;39(11):1377–84. https://doi.
org/10.1007/s001250050586.
11. Feldman EL, Callaghan BC, Pop-Busui
R, et  al. Diabetic neuropathy. Nat Rev Dis
Primers. 2019;5(1):41. https://doi.org/10.1038/
s41572-­019-­0092-­1.
12. Andersen ST, Witte DR, Dalsgaard EM, et  al. Risk
factors for incident diabetic polyneuropathy in a
cohort with screen-detected type 2 diabetes fol-
lowed for 13 years: ADDITION-Denmark. Diabetes
Care. 2018;41(5):1068–75. https://doi.org/10.2337/
dc17-­2062.
13. Tesfaye S, Chaturvedi N, Eaton SE, et  al. Vascular
risk factors and diabetic neuropathy. N Engl J
Med. 2005;352(4):341–50. https://doi.org/10.1056/
NEJMoa032782.
14. Callaghan BC, Gao L, Li Y, et al. Diabetes and obe-
sity are the main metabolic drivers of peripheral neu-
ropathy. Ann Clin Transl Neurol. 2018;5(4):397–405.
https://doi.org/10.1002/acn3.531.
15. Prabodha LBL, Sirisena ND, Dissanayake
VHW.  Susceptible and prognostic genetic fac-
tors associated with diabetic peripheral neu-
ropathy: a comprehensive literature review. Int
J Endocrinol. 2018;2018:8641942. https://doi.
org/10.1155/2018/8641942.
16. Lu B, Hu J, Wen J, et  al. Determination of periph-
eral neuropathy prevalence and associated factors
in Chinese subjects with diabetes and pre-diabetes  -
ShangHai Diabetic neuRopathy Epidemiology and
Molecular Genetics Study (SH-DREAMS). PLoS
One. 2013;8(4):e61053. https://doi.org/10.1371/jour-
nal.pone.0061053.
17. Kirthi V, Perumbalath A, Brown E, et al. Prevalence
of peripheral neuropathy in pre-diabetes: a systematic
review. BMJ Open Diabetes Res Care. 2021;9(1):40.
https://doi.org/10.1136/bmjdrc-­2020-­002040.
Clinical Diagnosis of Diabetic Peripheral Neuropathy
18. Kraiwong R, Vongsirinavarat M, Hiengkaew V, von
Heideken WP.  Effect of sensory impairment on bal-
ance performance and lower limb muscle strength
in older adults with type 2 diabetes. Ann Rehabil
Med. 2019;43(4):497–508. https://doi.org/10.5535/
arm.2019.43.4.497.
19. Callaghan BC, Price RS, Feldman EL. Distal symmet-
ric polyneuropathy in 2020. JAMA. 2020;324(1):90–
1. https://doi.org/10.1001/jama.2020.0700.
20. Maltese G, Tan SV, Bruno E, Brackenridge A, Thomas
S. Peripheral neuropathy in diabetes: it’s not always
what it looks like. Diabet Med. 2018;35(10):1457–9.
https://doi.org/10.1111/dme.13701.
21. Martin CL, Albers JW, Pop-Busui R. Neuropathy and
related findings in the diabetes control and complica-
tions trial/epidemiology of diabetes interventions and
complications study. Diabetes Care. 2014;37(1):31–8.
https://doi.org/10.2337/dc13-­2114.
22. Scott IA, Greenberg PB, Poole PJ.  Cautionary tales
in the clinical interpretation of studies of diagnostic
tests. Intern Med J. 2008;38(2):120–9. https://doi.
org/10.1111/j.1445-­5994.2007.01436.x.
23. Dobrow MJ, Hagens V, Chafe R, Sullivan T,
Rabeneck L.  Consolidated principles for screening
based on a systematic review and consensus process.
CMAJ. 2018;190(14):422–9. https://doi.org/10.1503/
cmaj.171154.
24. Bril V, Breiner A, Perkins BA, Zochodne D,
Diabetes Canada Clinical Practice Guidelines Expert
C.  Neuropathy. Can J Diabetes. 2018;42(1):217–21.
https://doi.org/10.1016/j.jcjd.2017.10.028.
25. Perez-Panero AJ, Ruiz-Munoz M, Cuesta-Vargas
AI, Gonzalez-Sanchez M.  Prevention, assess-
ment, diagnosis and management of diabetic foot
based on clinical practice guidelines: a systematic
review. Medicine. 2019;98(35):e16877. https://doi.
org/10.1097/MD.0000000000016877.
26. Embil JM, Albalawi Z, Bowering K, Trepman E,
Diabetes Canada Clinical Practice Guidelines Expert
C. Foot care. Can J Diabetes. 2018;42(1):S222–7. (In
eng). https://doi.org/10.1016/j.jcjd.2017.10.020.
27. Sharma S, Kerry C, Atkins H, Rayman G.  The
Ipswich Touch Test: a simple and novel method to
screen patients with diabetes at home for increased
risk of foot ulceration. Diabet Med. 2014;31(9):1100–
3. (Randomized Controlled Trial Research Support,
Non-U.S. Gov’t Validation Studies) (In eng). https://
doi.org/10.1111/dme.12450.
28. Kanji JN, Anglin RE, Hunt DL, Panju A.  Does this
patient with diabetes have large-fiber peripheral neu-
ropathy? JAMA. 2010;303(15):1526–32. https://doi.
org/10.1001/jama.2010.428.
29. Perkins BA, Olaleye D, Zinman B, Bril V.  Simple
screening tests for peripheral neuropathy in the diabe-
tes clinic. Diabetes Care. 2001;24(2):250–6. https://
doi.org/10.2337/diacare.24.2.250.
30. Bruce A, LEL P, Lewis EJH, Bril V, Ferdousi M,
Orszag A, Edwards K, Pritchard N, Russell A,
Dehghani C, Pacaud D, Romanchuk K, Mah JK,
Jeziorska M, Marshall A, Shtein RM, Pop-Busui R,
89
Lentz SI, Tavakoli M, Boulton AJM, Efron N, Malik
RA. Corneal confocal microscopy predicts the devel-
opment of diabetic neuropathy: a longitudinal diag-
nostic multinational consortium study. Diabetes Care.
2021;44(9):2107–14.
31. Kirk JK, Hunter JC, Mihalko SL, Danhauer SC,
Shumaker SA.  Perspectives of pain in patients with
type 2 diabetes. Expert Rev Endocrinol Metab.
2019;14(3):215–9. https://doi.org/10.1080/17446651
.2019.1592674.
32. Bai JW, Lovblom LE, Cardinez M, et al. Neuropathy
and presence of emotional distress and depres-
sion in longstanding diabetes: results from the
Canadian study of longevity in type 1 diabetes. J
Diabetes Complicat. 2017;31(8):1318–24. https://doi.
org/10.1016/j.jdiacomp.2017.05.002.
33. Callaghan BC, Cheng HT, Stables CL, Smith AL,
Feldman EL.  Diabetic neuropathy: clinical mani-
festations and current treatments. Lancet Neurol.
2012;11(6):521–34. https://doi.org/10.1016/
S1474-­4422(12)70065-­0.
34. Bracewell N, Game F, Jeffcoate W, Scammell
BE. Clinical evaluation of a new device in the assess-
ment of peripheral sensory neuropathy in diabe-
tes. Diabet Med. 2012;29(12):1553–5. https://doi.
org/10.1111/j.1464-­5491.2012.03729.x.
35. Nather A, Keng Lin W, Aziz Z, Ong C, McFeng
B.  Assessment of sensory neuropathy in patients
with diabetic foot problems. Diabetes Foot Ankle.
2011;2:6367. https://doi.org/10.3402/dfa.v2i0.6367.
36. Gylfadottir SS, Weeracharoenkul D, Andersen
ST, Niruthisard S, Suwanwalaikorn S, Jensen
TS.  Painful and non-painful diabetic polyneuropa-
thy: clinical characteristics and diagnostic issues. J
Diabetes Investig. 2019;10(5):1148–57. https://doi.
org/10.1111/jdi.13105.
37. Dyck PJ, Overland CJ, Low PA, et al. Signs and symp-
toms versus nerve conduction studies to diagnose
diabetic sensorimotor polyneuropathy: Cl vs. NPhys
trial. Muscle Nerve. 2010;42(2):157–64. https://doi.
org/10.1002/mus.21661.
38. Zografou I, Iliadis F, Sambanis C, Didangelos
T. Validation of neuropad in the assessment of periph-
eral diabetic neuropathy in patients with diabetes
mellitus versus the michigan neuropathy screen-
ing instrument, 10g monofilament application and
biothesiometer measurement. Curr Vasc Pharmacol.
2020;18(5):517–22. https://doi.org/10.2174/1570161
117666190723155324.
39. Rodriguez-Sanchez B, Pena-Longobardo LM,
Sinclair AJ.  Cost-effectiveness analysis of the
Neuropad device as a screening tool for early dia-
betic peripheral neuropathy. Eur J Health Econ.
2020;21(3):335–49. https://doi.org/10.1007/
s10198-­019-­01134-­2.
40. Ang L, Jaiswal M, Callaghan B, Raffel D, Pop-­
Busui R.  Sudomotor dysfunction as a measure of
small fiber neuropathy in type 1 diabetes. Auton
Neurosci. 2017;205:87–92. https://doi.org/10.1016/j.
autneu.2017.03.001.
90
41. Novak P.  Electrochemical skin conductance: a sys-
tematic review. Clin Auton Res. 2019;29(1):17–29.
https://doi.org/10.1007/s10286-­017-­0467-­x.
42. Rajan S, Campagnolo M, Callaghan B, Gibbons
CH.  Sudomotor function testing by electrochemical
skin conductance: does it really measure sudomotor
function? Clin Auton Res. 2019;29(1):31–9. https://
doi.org/10.1007/s10286-­018-­0540-­0.
43. Vinik AI, Casellini CM, Parson HK. Electrochemical
skin conductance to measure sudomotor function: the
importance of not misinterpreting the evidence. Clin
Auton Res. 2019;29(1):13–5. https://doi.org/10.1007/
s10286-­018-­0562-­7.
44. Binns-Hall O, Selvarajah D, Sanger D, Walker J,
Scott A, Tesfaye S.  One-stop microvascular screen-
ing service: an effective model for the early detection
of diabetic peripheral neuropathy and the high-risk
foot. Diabet Med. 2018;35(7):887–94. https://doi.
org/10.1111/dme.13630.
45. Shibata Y, Himeno T, Kamiya T, et  al. Validity
and reliability of a point-of-care nerve conduc-
tion device in diabetes patients. J Diabetes Investig.
2019;10(5):1291–8. https://doi.org/10.1111/
jdi.13007.
46. Scarr D, Lovblom LE, Cardinez N, et al. Validity of
a point-of-care nerve conduction device for polyneu-
ropathy identification in older adults with diabetes:
results from the Canadian study of longevity in type
1 diabetes. PLoS One. 2018;13(4):e0196647. https://
doi.org/10.1371/journal.pone.0196647.
47. Lee JA, Halpern EM, Lovblom LE, Yeung E, Bril V,
Perkins BA. Reliability and validity of a point-of-care
sural nerve conduction device for identification of
diabetic neuropathy. PLoS One. 2014;9(1):e86515.
https://doi.org/10.1371/journal.pone.0086515.
48. Perkins BA, Orszag A, Grewal J, Ng E, Ngo M, Bril
V.  Multi-site testing with a point-of-care nerve con-
duction device can be used in an algorithm to diag-
nose diabetic sensorimotor polyneuropathy. Diabetes
Care. 2008;31(3):522–4. https://doi.org/10.2337/
dc07-­1227.
49. Poulose S, Cheriyan E, Poulose A, Cheriyan R,
Vadakkanezath B, Ziemer P.  Usefulness of the
NC-stat DPNCheck nerve conduction test in a com-
munity pharmacy as an educational tool for patients
with diabetes. Can Pharm J. 2015;148(1):17–20.
https://doi.org/10.1177/1715163514561055.
50. Zhou X, Obuchowski NA, McClish DK.  Basic con-
cepts and methods, introduction. In: Statistical meth-
ods in diagnostic medicine. 2nd ed. Hoboken: Wiley;
2011. p. 1–11.
51. Herman WH, Pop-Busui R, Braffett BH, et  al. Use
of the Michigan Neuropathy Screening Instrument
as a measure of distal symmetrical peripheral
neuropathy in type 1 diabetes: results from the
diabetes control and complications trial/epide-
miology of diabetes interventions and complica-
tions. Diabet Med. 2012;29(7):937–44. https://doi.
org/10.1111/j.1464-­5491.2012.03644.x.
B. A. Perkins and V. Bril
52. Feldman EL, Stevens MJ, Thomas PK, Brown MB,
Canal N, Greene DA.  A practical two-step quanti-
tative clinical and electrophysiological assessment
for the diagnosis and staging of diabetic neuropa-
thy. Diabetes Care. 1994;17(11):1281–9. https://doi.
org/10.2337/diacare.17.11.1281.
53. Seferovic JP, Pfeffer MA, Claggett B, et  al. Three-­
question set from Michigan Neuropathy Screening
Instrument adds independent prognostic information
on cardiovascular outcomes: analysis of ALTITUDE
trial. Diabetologia. 2018;61(3):581–8. https://doi.
org/10.1007/s00125-­017-­4485-­y.
54. Mather KJ, Bebu I, Baker C, et  al. Prevalence of
microvascular and macrovascular disease in the gly-
cemia reduction approaches in diabetes - a compara-
tive effectiveness (GRADE) study cohort. Diabetes
Res Clin Pract. 2020;165:108235. https://doi.
org/10.1016/j.diabres.2020.108235.
55. Hansen CS, Jensen TM, Jensen JS, et al. The role of
serum methylglyoxal on diabetic peripheral and car-
diovascular autonomic neuropathy: the ADDITION
Denmark study. Diabet Med. 2015;32(6):778–85.
https://doi.org/10.1111/dme.12753.
56. Pop-Busui R, Lu J, Brooks MM, et al. Impact of gly-
cemic control strategies on the progression of dia-
betic peripheral neuropathy in the bypass angioplasty
revascularization investigation 2 diabetes (BARI 2D)
cohort. Diabetes Care. 2013;36(10):3208–15. https://
doi.org/10.2337/dc13-­0012.
57. Barbosa M, Saavedra A, Severo M, Maier C, Carvalho
D.  Validation and reliability of the Portuguese ver-
sion of the Michigan neuropathy screening instru-
ment. Pain Pract. 2017;17(4):514–21. https://doi.
org/10.1111/papr.12479.
58. Bax G, Fagherazzi C, Piarulli F, Nicolucci A, Fedele
D. Reproducibility of Michigan neuropathy screening
instrument (MNSI). A comparison with tests using the
vibratory and thermal perception thresholds. Diabetes
Care. 1996;19(8):904–5. https://doi.org/10.2337/
diacare.19.8.904.
59. Bril V, Perkins BA. Validation of the Toronto clinical
scoring system for diabetic polyneuropathy. Diabetes
Care. 2002;25(11):2048–52. https://doi.org/10.2337/
diacare.25.11.2048.
60. Bril V, Tomioka S, Buchanan RA, Perkins
BA. Reliability and validity of the modified Toronto
clinical neuropathy score in diabetic sensorimotor
polyneuropathy. Diabet Med. 2009;26(3):240–6.
https://doi.org/10.1111/j.1464-­5491.2009.02667.x.
61. Abraham A, Barnett C, Katzberg HD, Lovblom LE,
Perkins BA, Bril V. Toronto clinical neuropathy score
is valid for a wide spectrum of polyneuropathies. Eur
J Neurol. 2018;25(3):484–90. https://doi.org/10.1111/
ene.13533.
62. Lysy Z, Lovblom LE, Halpern EM, et al. Measurement
of cooling detection thresholds for identification of
diabetic sensorimotor polyneuropathy in type 1 dia-
betes. PLoS One. 2014;9(9):e106995. https://doi.
org/10.1371/journal.pone.0106995.
Clinical Diagnosis of Diabetic Peripheral Neuropathy
63. Lovblom LE, Halpern EM, Wu T, et al. In vivo corneal
confocal microscopy and prediction of future-incident
neuropathy in type 1 diabetes: a preliminary longitu-
dinal analysis. Can J Diabetes. 2015;39(5):390–7.
https://doi.org/10.1016/j.jcjd.2015.02.006.
64. Lewis EJH, Perkins BA, Lovblom LE, Bazinet RP,
Wolever TMS, Bril V. Effect of omega-3 supplemen-
tation on neuropathy in type 1 diabetes: A 12-month
pilot trial. Neurology. 2017;88(24):2294–301. https://
doi.org/10.1212/WNL.0000000000004033.
65. Sekiguchi K, Kohara N, Baba M, et al. Aldose reduc-
tase inhibitor ranirestat significantly improves nerve
conduction velocity in diabetic polyneuropathy: a
randomized double-blind placebo-controlled study
in Japan. J Diabetes Investig. 2019;10(2):466–74.
https://doi.org/10.1111/jdi.12890.
66. Zinman LH, Ngo M, Ng ET, Nwe KT, Gogov S, Bril
V. Low-intensity laser therapy for painful symptoms
of diabetic sensorimotor polyneuropathy: a controlled
trial. Diabetes Care. 2004;27(4):921–4. https://doi.
org/10.2337/diacare.27.4.921.
67. Abraham A, Katzberg HD, Lovblom LE, Perkins
BA, Bril V.  Uric acid levels correlate with sensory
nerve function in healthy subjects. Can J Neurol
Sci. 2019;46(3):337–41. https://doi.org/10.1017/
cjn.2019.9.
68. Dunnigan SK, Ebadi H, Breiner A, et al. The charac-
teristics of chronic inflammatory demyelinating poly-
neuropathy in patients with and without diabetes–an
observational study. PLoS One. 2014;9(2):e89344.
https://doi.org/10.1371/journal.pone.0089344.
69. Ghavami H, Radfar M, Soheily S, Shamsi SA,
Khalkhali HR.  Effect of lifestyle interventions on
diabetic peripheral neuropathy in patients with type
2 diabetes, result of a randomized clinical trial.
Agri. 2018;30(4):165–70. https://doi.org/10.5505/
agri.2018.45477.
70. Wahren J, Foyt H, Daniels M, Arezzo JC.  Long-­
acting C-peptide and neuropathy in type 1 dia-
betes: a 12-month clinical trial. Diabetes Care.
2016;39(4):596–602. https://doi.org/10.2337/
dc15-­2068.
71. Bril V.  NIS-LL: the primary measurement scale for
clinical trial endpoints in diabetic peripheral neuropa-
thy. Eur Neurol. 1999;41(Suppl 1):8–13. https://doi.
org/10.1159/000052074.
72. Dyck PJ, Davies JL, Litchy WJ, O'Brien
PC. Longitudinal assessment of diabetic polyneurop-
athy using a composite score in the rochester diabetic
neuropathy study cohort. Neurology. 1997;49(1):229–
39. https://doi.org/10.1212/wnl.49.1.229.
73. Zilliox LA, Ruby SK, Singh S, Zhan M, Russell
JW.  Clinical neuropathy scales in neuropathy
associated with impaired glucose tolerance. J
Diabetes Complicat. 2015;29(3):372–7. https://doi.
org/10.1016/j.jdiacomp.2015.01.011.
74. Lin X, Yarlas A, Vera-Llonch M, et al. Rate of neu-
ropathic progression in hereditary transthyretin amy-
loidosis with polyneuropathy and other peripheral
neuropathies: a systematic review and meta-analysis.
91
BMC Neurol. 2021;21(1):70. https://doi.org/10.1186/
s12883-­021-­02094-­y.
75. Ziegler D, Edmundson S, Gurieva I, Mankovsky B,
Papanas N, Strokov I. Predictors of response to treat-
ment with actovegin for 6 months in patients with
type 2 diabetes and symptomatic polyneuropathy. J
Diabetes Complicat. 2017;31(7):1181–7. https://doi.
org/10.1016/j.jdiacomp.2017.03.012.
76. Ziegler D, Low PA, Freeman R, Tritschler H, Vinik
AI.  Predictors of improvement and progression of
diabetic polyneuropathy following treatment with
alpha-lipoic acid for 4 years in the NATHAN 1 trial.
J Diabetes Complicat. 2016;30(2):350–6. https://doi.
org/10.1016/j.jdiacomp.2015.10.018.
77. Tesfaye S, Tandan R, Bastyr EJ, et  al. Factors that
impact symptomatic diabetic peripheral neuropathy
in placebo-administered patients from two 1-year
clinical trials. Diabetes Care. 2007;30(10):2626–32.
https://doi.org/10.2337/dc07-­0608.
78. Singleton JR, Bixby B, Russell JW, et  al. The
Utah Early Neuropathy Scale: a sensitive clinical
scale for early sensory predominant neuropathy. J
Peripher Nerv Syst. 2008;13(3):218–27. https://doi.
org/10.1111/j.1529-­8027.2008.00180.x.
79. Boger MS, Hulgan T, Haas DW, et  al. Measures
of small-fiber neuropathy in HIV infection.
Auton Neurosci. 2012;169(1):56–61. https://doi.
org/10.1016/j.autneu.2012.04.001.
80. Andreasson M, Lagali N, Badian RA, et  al.
Parkinson’s disease with restless legs syndrome-
­an in  vivo corneal confocal microscopy study. NPJ
Parkinsons Dis. 2021;7(1):4. https://doi.org/10.1038/
s41531-­020-­00148-­5.
81. Fernandez-Torres R, Ruiz-Munoz M, Perez-­
Panero AJ, Garcia-Romero JC, Gonzalez-Sanchez
M.  Clinician assessment tools for patients with dia-
betic foot disease: a systematic review. J Clin Med.
2020;9(5):487. https://doi.org/10.3390/jcm9051487.
82. Smith AG, Singleton JR.  Obesity and hyperlipid-
emia are risk factors for early diabetic neuropathy. J
Diabetes Complicat. 2013;27(5):436–42. https://doi.
org/10.1016/j.jdiacomp.2013.04.003.
83. Smith AG, Lessard M, Reyna S, Doudova M,
Singleton JR.  The diagnostic utility of Sudoscan
for distal symmetric peripheral neuropathy. J
Diabetes Complicat. 2014;28(4):511–6. https://doi.
org/10.1016/j.jdiacomp.2014.02.013.
84. Gonzalez-Duarte A, Lem M, Diaz-Diaz E, Castillo
C, Cardenas-Soto K.  The efficacy of pregabalin in
the treatment of prediabetic neuropathic pain. Clin J
Pain. 2016;32(11):927–32. https://doi.org/10.1097/
AJP.0000000000000339.
85. Young MJ, Boulton AJ, MacLeod AF, Williams DR,
Sonksen PH.  A multicentre study of the prevalence
of diabetic peripheral neuropathy in the United
Kingdom hospital clinic population. Diabetologia.
1993;36(2):150–4. https://doi.org/10.1007/
BF00400697.
86. Meijer JW, Smit AJ, Sonderen EV, Groothoff
JW, Eisma WH, Links TP.  Symptom scoring sys-
92
tems to diagnose distal polyneuropathy in dia-
betes: the Diabetic Neuropathy Symptom score.
Diabet Med. 2002;19(11):962–5. https://doi.
org/10.1046/j.1464-­5491.2002.00819.x.
87. Asad A, Hameed MA, Khan UA, Ahmed N, Butt
MU. Reliability of the neurological scores for assess-
ment of sensorimotor neuropathy in type 2 diabetics.
J Pak Med Assoc. 2010;60(3):166–70.
88. Jia WP, Shen Q, Bao YQ, Lu JX, Li M, Xiang
KS. Evaluation of the four simple methods in the diag-
nosis of diabetic peripheral neuropathy. Zhonghua Yi
Xue Za Zhi. 2006;86(38):2707–10.
89. Abbott CA, Carrington AL, Ashe H, et  al. The
North-West Diabetes Foot Care Study: inci-
dence of, and risk factors for, new diabetic foot
ulceration in a community-based patient cohort.
Diabet Med. 2002;19(5):377–84. https://doi.
org/10.1046/j.1464-­5491.2002.00698.x.
90. Weintrob N, Amitay I, Lilos P, Shalitin S, Lazar L,
Josefsberg Z.  Bedside neuropathy disability score
compared to quantitative sensory testing for mea-
surement of diabetic neuropathy in children, ado-
lescents, and young adults with type 1 diabetes. J
Diabetes Complicat. 2007;21(1):13–9. https://doi.
org/10.1016/j.jdiacomp.2005.11.002.
B. A. Perkins and V. Bril
91. Pham H, Armstrong DG, Harvey C, Harkless LB,
Giurini JM, Veves A.  Screening techniques to iden-
tify people at high risk for diabetic foot ulcer-
ation: a prospective multicenter trial. Diabetes
Care. 2000;23(5):606–11. https://doi.org/10.2337/
diacare.23.5.606.
92. Dhage S, Ferdousi M, Adam S, et al. Corneal confocal
microscopy identifies small fibre damage and progres-
sion of diabetic neuropathy. Sci Rep. 2021;11(1):1859.
https://doi.org/10.1038/s41598-­021-­81302-­8.
93. Ferdousi M, Kalteniece A, Azmi S, et al. Corneal con-
focal microscopy compared with quantitative sensory
testing and nerve conduction for diagnosing and strat-
ifying the severity of diabetic peripheral neuropathy.
BMJ Open Diabetes Res Care. 2020;8(2):e001801.
https://doi.org/10.1136/bmjdrc-­2020-­001801.
94. Cornblath DR, Chaudhry V, Carter K, et  al.
Total neuropathy score: validation and reliability
study. Neurology. 1999;53(8):1660–4. https://doi.
org/10.1212/wnl.53.8.1660.
95. Menten J, Lesaffre E. A general framework for com-
parative Bayesian meta-analysis of diagnostic studies.
BMC Med Res Methodol. 2015;15:70. https://doi.
org/10.1186/s12874-­015-­0061-­7.
 Diagnostic Techniques for Diabetic
Peripheral Neuropathy

Long Davalos, Amro Stino, and A. Gordon Smith

1 Introduction tions where diagnostic testing plays an important

role. Diagnostic tests can be used to confirm the
The most common peripheral neuropathy associ- suspected diagnosis for clinical or research pur-
ated with both type 1 (T1D) and type 2 diabetes poses and as surrogate measures of disease sever-
(T2D) is a distal symmetric polyneuropathy ity in natural history or therapeutic trials. The
(DPN). DPN is characterized by length-­ clinical spectrum of DPN is reflected in the vari-
dependent sensory loss predominantly affecting ety of diagnostic tests available that evaluate dif-
the distal lower extremities with minimal weak- ferent nerve fiber classes and functions.
ness. There are several phenotypic variants of Appropriate application of these tests is based on
DPN, ranging from painful sensory neuropathy the specific clinical phenotype and context of use
(sometimes in the setting of preferential injury to (clinical diagnosis, epidemiological studies, clin-
small unmyelinated axons), nonpainful neuropa- ical trials, etc.).
thy with sensory symptoms, and asymptomatic Nerve conduction studies (NCS) are the most
DPN, which can present with an insensate foot, widely used diagnostic modality, and assess
often in the setting of foot ulceration. Diagnosis large, myelinated motor, and sensory nerve fibers
of DPN is largely based on a combination of through electrical stimulation of a peripheral
symptoms and signs, with signs conveying nerve with recording of the resulting nerve or
greater diagnostic certainty, as summarized in muscle response. The Toronto consensus criteria
chapter “Clinical Diagnosis of Diabetic require abnormalities of NCS to establish a diag-
Peripheral Neuropathy” by Dr. Bruce Perkins. nosis of “confirmed DPN,” although for clinical
However, there are clinical and research situa- diagnostic purposes, NCS are not necessary for
many patients with signs and symptoms of DPN
(Table 1) [1, 2]. Although NCS objectively quan-
L. Davalos tify large myelinated fiber function, they do not
Department of Neurology and Rehabilitation
Medicine, University of Cincinnati, assess small fibers [3]. A diagnosis of confirmed
Cincinnati, OH, USA small fiber neuropathy (SFN) requires an abnor-
A. Stino mality of a validated measure of small fiber func-
Department of Neurology, University of Michigan tion [4]. Skin biopsy for measurement of
Medicine, Ann Arbor, MI, USA intraepidermal nerve fiber density (IENFD) is the
A. G. Smith (*) best validated measure of small fiber function but
Department of Neurology, Virginia Commonwealth several other techniques are available with vary-
University, Richmond, VA, USA ing degrees of validation supporting their use,
e-mail: Gordon.Smith@vcuhealth.org

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 93

S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_6
94
Table 1  Toronto consensus criteria. Definitions of mini-
mal criteria for typical diabetic polyneuropathy
Diagnostic
category Definition
Possible The presence of any symptom or sign of
DSPN DSPN. Symptoms include decreased
sensation or positive neuropathic
sensory symptoms predominantly in the
toes, feet, or legs. Signs include distal
symmetric decrease in sensation or
unequivocally decreased or absent ankle
reflexes
Probable The presence of a combination of
DSPN symptoms and signs of neuropathy
including any two or more of the
following: neuropathic symptoms,
decreased distal sensation, or
unequivocally decreased or absent ankle
reflexes
Confirmed The presence of an abnormality of NCS
DSPN and a symptom or symptoms or a sign
or signs of neuropathy. If NCS is
normal, a validated measure of SFN
may be used
Subclinical The presence of no signs or symptoms
DSPN of neuropathy are confirmed with
abnormal NCS or a validated measure
of SFN
DSPN distal symmetric polyneuropathy, NCS nerve con-
duction studies, SFN small fiber neuropathy
including corneal confocal microscopy and mea-
sures of sudomotor (sweat) function, which is
subserved by small diameter unmyelinated axons
that are frequently injured in patients with
somatic SFN [5–8]. Small and large fiber func-
tion can be assessed using quantitative sensory
testing for thermal and pain or vibration sensa-
tion respectively. Each of these techniques
requires special equipment and technical exper-
tise and, in the case of IENFD, access to a
­qualified laboratory. While each may be applied
in varying research settings, from a clinical per-
spective, there is interest in development of con-
venient and well-tolerated point of care
techniques that can be used to confirm DPN or
for disease screening purposes. These include
several quantitative sensory testing (QST)
devices (vibrameter, neurometer, and thermoaes-
thesiometer), point of care NCS devices (DPN
Check®, NeuroMetrix, Massachusetts USA), and
Neuropad® (TRIGOcare International GmbH,
L. Davalos et al.
Wiehl Germany), a simple measure of sudomotor
function.
This chapter reviews each of these testing
modalities including their validity and reliability,
and current utility in clinical and research settings.
2 Nerve Conduction Studies
NCS are considered the gold standard for DPN
diagnostic confirmation and are routinely used in
clinical research settings to assess disease pro-
gression or response to therapy. In the clinical
setting, NCS can be used as a quantitative confir-
matory test, especially when a patient presents
with features that are atypical for DPN, such as
acute or subacute onset, non-length dependence,
motor predominance or significant asymmetry of
neuropathic signs or symptoms [9]. While NCS
have a high degree of reproducibility, particularly
nerve conduction velocity, they require signifi-
cant technical skill and are operator dependent.
2.1 Basic Principles and NCS
Abnormalities in DPN
NCS involve electrically stimulating a nerve and
recording the resulting action potential. Sensory
nerve action potentials (SNAPs) are recorded by
placing surface electrodes over a cutaneous nerve
(Fig.  1). Compound muscle action potentials
(CMAPs) are obtained by placing surface elec-
trodes over a muscle supplied by the nerve. The
amplitude of the response is reflective of the
number of axons contributing to the summated
waveform. The time it takes for the response to
occur (distal latency) and the conduction velocity
(CV) reflect the function of peripheral nerve
myelination. Loss of amplitude suggests axonal
loss or dysfunction, whereas slowing of CV or
latency prolongation usually implies demyelin-
ation [10]. F-wave latencies also asses nerve
function and represent the motor conduction time
to and from the spinal cord along the entirety of
the motor nerve axon. Therefore, it assesses the
most proximal nerve segment and the entire
length of a motor nerve.
Diagnostic Techniques for Diabetic Peripheral Neuropathy
Fig. 1  Recording electrodes are placed over the first digit
and an electrical stimulation is applied to the median
nerve at the wrist (top panel). The resulting sensory nerve
action potential (SNAP) is illustrated below. The ampli-
tude of the response is proportionate to the number of
large diameter sensory axons, and conduction velocity to
myelin function
Patients with DPN have reduced SNAP ampli-
tudes, often with mildly slowed motor nerve con-
duction velocity, and mildly prolonged minimal
F-wave latencies [11]. DPN is a length-­dependent
process involving the longest nerves, thus find-
ings are most prominent in the distal lower
extremities. A reduction of 1 μV in the sural nerve
SNAP amplitude associates with a decrease of
approximately 150 fibers/mm2, while a mean
reduction of 1mV of the ulnar, peroneal, and tibial
nerves CMAP amplitudes associates with a den-
sity loss of 200 fibers/mm2 [12]. SNAP ampli-
tudes decline over time, with one study of patients
with DPN associated with T2D experiencing a
5% loss of sural SNAP amplitude per year [13].
Nerve CV also decreases with disease progres-
sion. In the Diabetic Control and Complications
Trial, which assessed T1D patients, the sural sen-
sory and peroneal motor nerve CV in the conven-
95
tionally treated group decreased by 0.56 and 0.54
m/s/year, respectively, over 5 years [14]. In a pro-
spective 8-year study of 45 T1D patients, a 1%
rise in HbA1c was associated with a 1.3 meter per
second decrease in maximal nerve CV [15].
Furthermore, in patients with T2D, a lower rate of
decline was observed in a 10-year natural history
study of 133 patients with newly diagnosed T2D,
in which NCV deteriorated by 0.39 m/s/year in
the sural and 0.3 m/s/year in the peroneal nerves
[13]. In regard to the F-wave, prolongation of the
F-wave latencies has shown to be very sensitive
for DPN as well as highly reproducible [16, 17].
The amplitude reduction of the nerve action
potentials with disease progression is a reflection of
the axonal degeneration and loss of nerve fibers
observed in DPN, which is the hallmark of nerve
injury in this condition [18, 19]. The degree of CV
slowing in DPN is not as prominent as the slowing
seen in primary demyelinating neuropathies such as
acute or chronic inflammatory demyelinating poly-
radiculoneuropathy. These changes could reflect
mild demyelination or a loss of nodal ion channels
without a structural alteration, as is shown in some
studies [20]. The prolonged F-wave latencies may
indicate more widespread nerve involvement. Some
pathological DPN studies indicate distally promi-
nent but diffuse involvement of peripheral nerves
[21, 22], while other studies show patchy fiber
degeneration with fascicular sprouting that begins
proximally and extends distally [18].
2.2 Clinical Utility of NCS in DPN
Diagnosis
The consensus regarding the diagnosis of DPN
has changed throughout the years. The 1998 San
Antonio Conference on Diabetic Neuropathy
[23], and then the American Academy of
Neurology (AAN), American Association of
Electrodiagnostic Medicine (AAEM), and
American Academy of Physical Medicine and
Rehabilitation (AAPM&R) in 2005 [24] pro-
posed criteria for DPN that included abnormali-
ties on NCS.  The Toronto consensus criteria
(2010) defined confirmed DPN symptoms and
signs in addition to abnormal NCS [1].
96
The AAN, AANEM, and AAPM&R recom-
mended a clinical research protocol for NCS in
2005 that proposed the specific studies necessary,
including unilateral examination of sural, ulnar
and median sensory nerves, and peroneal, tibial,
median and ulnar motor nerves with F waves. An
abbreviated protocol was also proposed, more
practical for the clinical setting, which included a
sural sensory and peroneal motor nerve in one
lower extremity. The minimum case definition cri-
terion for electrodiagnostic confirmation was an
abnormality (≥99th or ≤1st percentile) of any
attribute of nerve conduction in two separate
nerves, one of which had to be the sural nerve [24].
In 2011, the Nerve Conduction Criteria Study
reported that the six most sensitive attributes of
NCS for diagnosing subclinical DPN were pero-
neal motor CV, sural sensory amplitude, tibial
motor CV, ulnar motor CV, tibial, and ulnar F-wave
latencies. This study evaluated eight different elec-
trodiagnostic criteria and concluded that utilizing
the “≥1 abnormal attributes (≤1st/≥99th or
≤2.5st/≥97.5th) in ≥2 separate nerves” criteria
was acceptable for clinical practice [25].
These criteria provide general guidance to
practicing clinicians, but they are largely intended
for research settings. While NCS are objective
and reliable, they have some pitfalls that should
be taken into account in the clinical setting. NCS
require a referral to a neurophysiology lab with
trained examiners. The procedure is uncomfort-
able, and is a significant driver of health care
costs attributable to neuropathy evaluation [26].
NCS rarely alter the care of patients with DPN
[27], and routine use for diagnostic confirmation
of typical DPN is not recommended. NCS should
be performed in situations where the clinical fea-
tures are atypical (motor greater than sensory
neuropathy, rapid onset, or asymmetrical presen-
tation), the diagnosis is unclear, or a different eti-
ology is suspected [28].
2.3 Research Applications for NCS
Advantages and Limitations of NCS  Different
DPN cohorts [19, 29, 30] and population-based
studies of healthy and diabetic subjects [31–33]
L. Davalos et al.
show that NCS is an objective and reliable test
for the diagnosis of DPN. In comparison to other
methods such as clinical examination or QST,
NCS has lower coefficient of variation on repeat
testing [34–36]. Furthermore, NCS abnormalities
seem to correlate with nerve morphology [37]
and clinical severity of disease [38]. These attri-
butes make NCS a reliable and standardized tool
to measure nerve function. However, there are
some limitations to consider when utilizing NCS
for diagnostic confirmation. NCS cannot detect
small fiber nerve injury, which may be the earli-
est detectable sign in DPN, particularly in patients
with prominent neuropathic pain [3]. While this
is a relevant limitation for studies focusing on
small fiber predominant neuropathy, most
patients with symptomatic DPN experience pro-
gressive loss both small unmyelinated and large
myelinated axons [39]. Furthermore, the overall
population normative ranges for NCS parameters
may overlap with abnormal values in DPN unless
age, gender, height, and body mass index are
considered [40]. While intraobserver reliability is
high, unless meticulous attention is paid to stan-
dardized protocols and training, interobserver
reliability can be unacceptably low [41, 42]. This
issue is a concern, especially for multicenter ther-
apeutic trials, where even very small interob-
server differences can decrease the statistical
power of a study to detect a meaningful change in
disease course.
In order to use NCS as a quantitative diagnos-
tic method and overcome its limitations, it is nec-
essary to standardize the common reference
values, percentile abnormalities, and the exact
placement of stimulating and recording elec-
trodes among EMG laboratories. Implementing
these measures and encouraging the same clini-
cal neurophysiologist to perform all serial NCS
on a given subject improves the interobserver
reliability and the accuracy of NCS as a diagnos-
tic and research outcome measure tool [42].
Defining DPN Using NCS  The definition of
DPN among studies varies considerably. Some
define DPN only clinically, while others utilize
NCS, QST or a combination. For this reason,
Diagnostic Techniques for Diabetic Peripheral Neuropathy
there is great variability in the reported preva-
lence of DPN (8–50%) [13, 43, 44]. The Toronto
consensus advocates the use of NCS and clinical
abnormalities, and defines confirmed typical
DPN as the presence of an abnormality of NCS
along with symptom(s)/sign(s) of neuropathy [1].
This definition is simple and can be used to estab-
lish the diagnosis of DPN with a high degree of
certainty, although for some epidemiological
studies where NCS are impractical due to access
or cost, clinical criteria alone suffice (signs and
symptoms—“probable DPN”). Composite scores
of neurologic signs and NCS attributes can fur-
ther increase the sensitivity and specificity of the
diagnosis [31, 45]. There is not a formal consen-
sus regarding specific NCS criteria for DPN. An
abnormality of one or more attributes (exceeding
the normal limits between the 1st and 99th per-
centiles; variables, such as age, height, and tem-
perature, should be considered when developing
the reference range and interpreting the results)
in two or more separate nerves is an acceptable
criterion for the electrodiagnostic definition of
DPN [46, 47].
Defining subclinical DPN is more compli-
cated. For research purposes, the Toronto consen-
sus criteria defined subclinical DPN as the
presence of no signs or symptoms of neuropathy
confirmed with abnormal NCS or a validated
measure of SFN [1]. However, it did not specify
the minimum electrodiagnostic criteria needed,
which is crucial for defining this group. The
Nerve Conduction Study evaluated eight differ-
ent criteria (criteria 1–4 evaluated individual
NCS attributes and criteria 5–8 composite sum
scores of NCS attributes) and concluded that all
four composite sum scores of normal deviates
(from percentiles) performed excellently in the
healthy subject and diabetic patient cohorts, par-
ticularly criteria 5 and 6, which included the
composite sum score of two nerve attributes (e.g.,
peroneal CV and sural amplitude). From the indi-
vidual NCS attribute criteria, Criterion 2 (≥1
abnormal attribute in 2 separate nerves) was also
acceptable [25].
97
NCS as an Endpoint Measure in Clinical
Trials and Natural History Studies  Multiple
clinical trials have used NCS as the primary
outcome to assess treatment efficacy and have
failed to meet this primary endpoint [48–51]. In
this regard, there are several findings to suggest
that using NCS as an isolated endpoint measure
is not ideal. NCS cannot detect small fiber
nerve function [52], suffer from high interob-
server variability, and the clinical meaning of
modest changes in nerve conduction velocity
(the most commonly used single attribute as a
study endpoint measures) has not been well
defined. The American Diabetes Association
recommends using validated clinical instru-
ments combined, with electrophysiology; mea-
sures of small fiber damage and repair (e.g.
intraepidermal nerve fiber density), and objec-
tive measures of patient function in the design
of DPN trials [28]. Utilizing composite sum
scores instead of individual attributes can
increase the sensitivity of NCS as an endpoint
measure [25, 45].
Point of Care NCS Devices  One strategy to
overcome the technical challenges of NCS has
been to develop point of care devices capable or
assessing specific NCS parameters (reliably
when used by individuals with minimal training).
Point of care devices typically stimulate the sural
nerve at the level of the ankle while recording the
SNAP amplitude and conduction velocity using a
biosensor that covers a wide surface area in the
proximal lower limb. One device demonstrated
high sensitivity (95%) and acceptable specificity
(71%) for detecting DPN, and demonstrated
good intrarater and interrater reproducibility. The
SNAP amplitude was similar to that obtained
with traditional NCS, however, the device over-
estimated the sural CV [53]. Overall, despite
these limitations, point of care NCS devices are a
promising diagnostic tool that are fast, inexpen-
sive, and easy to use. Further studies are required
to evaluate performance in longitudinal clinical
trials.
98 L. Davalos et al.

3 Skin Biopsy a
with Measurement
of Intraepidermal Nerve
Fiber Density (IENFD)

Skin biopsy for IENFD measurement is a well

validated and reproducible measure of small
unmyelinated axons. Given that NCS do not eval- b
uate this fiber class, there has been a need for a
validated measure of small fiber function.
Langerhans first recognized epidermal nerve
fibers in the 1860s [54], but the existence of epi-
dermal innervation was controversial for much of
the twentieth century since epidermal nerves were
difficult to visualize using standardized histo- c
chemical techniques. Early studies focused on the
density and distribution of Meissner’s corpuscles,
but this technique was of little diagnostic utility
[55]. The availability of an antibody to protein
gene product (PGP) 9.5, a neuronal form of ubiq-
uitin hydrolase, rapidly led to sensitive immuno-
histochemical techniques to visualize nerve fibers
in the skin [56]. This technique has facilitated Fig. 2  Skin biopsies taken from the proximal (a) and dis-
tal thigh (b) and distal leg (c) of a patient with diabetic
widespread study of cutaneous innervation for neuropathy stained with PGP 9.5 demonstrate absent
both clinical and research applications (Fig. 2). IENFD at the distal site with reduced numbers of IENFD
at the distal thigh, where there are frequent axonal
swellings
3.1 Skin Anatomy and Epidermal
Innervation a b

Skin areas are classified as glabrous (hairless) or

nonglabrous. Glabrous skin is located on the
soles, palms, and fingertips, and has more sweat
glands (Fig.  3) and cutaneous sensory organs
dedicated to touch, position, and vibration sensa-
tion. On the other hand, nonglabrous skin has a
greater number of piloerector muscles and hair
follicles, which are supplied by small autonomic
nerves [57]. Both skin areas have abundant epi-
dermal nerves in normal individuals. The epider-
mis is the outermost layer of the skin, which
Fig. 3  A skin biopsy stained with PGP9.5 from a control
contains keratinocytes. The dermis harbors cuta- subject demonstrates a normally innervated sweat gland
neous sensory organs, hair follicles, blood and (a). A biopsy from a patient with diabetic neuropathy
lymphatic vessels, and sweat, apocrine and seba- demonstrates a severely denervated sweat gland (b)
ceous glands. The basement membrane (dermal-­
epidermal junction) is organized as papillae loops. Sensory organs such as Meissner corpus-
(dermal undulations extending into the epider- cles and Merkel cells are located in the apex of
mis) that contain vascular plexus and capillary these papillae.
Diagnostic Techniques for Diabetic Peripheral Neuropathy
Epidermal innervation arises from the net-
work of nerve fibers that run beneath the base-
ment membrane. Intraepidermal fibers (IENFs)
lose their myelin sheath as they cross the dermal-­
epidermal junction and run toward the skin sur-
face between keratinocytes. A single dermal fiber
typically divides into several intraepidermal
branches after crossing the basement membrane
and extends in parallel fashion to near the surface
of the skin. In healthy subjects, IENFs have a
simple morphology, with small bead-like vari-
cosities [58], and the density of epidermal inner-
vation is greater at more proximal sites on the
body than it is at distal locations [59, 60].
3.2 Selection of Biopsy Site,
Processing Technique,
and Quantification of Skin
Nerve Fibers
Skin biopsy is a relatively benign procedure. A
3 mm punch biopsy is obtained following injec-
tion of local anesthetic. An adhesive bandage is
used to cover the biopsy site, which generally
heals through a process of granulation within
7–10 days. There is a very low risk of local infec-
tion. Selection of the biopsy site depends on the
clinical context and intent. For diagnosing length-­
dependent neuropathies, as in DPN, one skin
biopsy sample is taken from the distal part of the
leg (10  cm above the external malleolus) and
another specimen is obtained from the lateral
aspect of the proximal thigh (20  cm below the
anterior iliac spine) [61]. The tissue needs to be
immediately placed in either 2% periodate-­
lysine-­
paraformaldehyde (PLP) or Zamboni
solution (2% paraformaldehyde, picric acid) for
12–24 h at 4 °C, and is then placed in cryoprotec-
tant solution. After that, the samples can be stored
in a freezer or sent to a processing center. The
biopsied tissue is cut into 50 μm sections and
stained with anti PGP 9.5 antibody, which avidly
stains all axons [56]. IENFD can then be ana-
lyzed via two different microscopic techniques.
The most commonly used method is standard
bright field microscopy since it is convenient,
inexpensive, and relatively easy to use. The other
method is confocal imaging, which allows mul-
99
tiple antibodies with different fluorochromes to
be used (multiple structures can be visualized
simultaneously) and allows three-dimensional
reconstruction of skin sections, but is more
expensive and time consuming.
IENFD is determined by counting the number
of PGP9.5-positive fibers crossing the dermal-­
epidermal junction in at least three sections and
dividing this by the length of the epidermal sur-
face (measured with computer software). This
value is then compared with normative age-­
matched ranges [59, 62]. Quantification of linear
IEFND has a high interobserver, intraobserver,
and interlaboratory agreement [63], correlates
significantly with stereological techniques of
skin nerve morphology [64, 65], and with quanti-
fication of nerve fibers per epidermal area [66].
Furthermore, there is no significant variation
with the IEFND calculated from adjacent sec-
tions of the same biopsy or in adjacent biopsies
from the same site [59]. IEFND declines with
age, is lower in males, and is not influenced by
weight or height [67].
3.3 Skin Biopsy Abnormalities
in DPN
Patients with painful neuropathy associated with
diabetes often have preferential injury to small
unmyelinated axons [3]. Abnormalities in distal
leg IEFND have been reported in 47.9% of
patients with early diabetes and 12.5% of predia-
betic individuals. Furthermore, using IEFND
abnormalities as a criterion for confirmed DPN in
early diabetes detects a higher number of patients
with neuropathy (37.5%) in comparison to NCS
(only 14.6%). Similarly, the proportion of
patients with early diabetes with abnormal NCS
is 4.2% compared to 16.7% for IENFD [3]. As
diabetes progresses, studies also show concor-
dant progressive reduction in IENFD that follows
disease course. IENFD declines in patients with
DPN at a rate of between 1 and 3 fibers/mm/year
[68]. In regards of neuropathic severity, IENFD
inversely correlates with the Neurological
Disability Score [69]. One study reported no
­correlation between IENFD and the Neuropathy
Symptom Score, but showed an inverse correla-
100
tion with pain severity [70]. Patients with painful
DPN tend to have more severe IENFD loss com-
pared to those without pain, particularly when
there are little or no objective signs of neuropa-
thy. It is possible that IENFD loss plays an impor-
tant role in the early evolution of neuropathic
pain, while other mechanisms play a more rele-
vant role at more advanced stages [71].
3.4 Clinical Application of Skin
Biopsy
Normative data adjusted for sex and age are avail-
able for IENFD based on the typically performed
brightfield technique [62]. The diagnostic yield of
skin biopsy varies based on the normative values,
imaging technique, and diagnostic gold standard
definition. In a study of 210 patients with SFN,
which included 65 diabetic patients, Z-scores and
5th percentile cutoffs provided the highest speci-
ficity (98 and 95%, respectively) but the lowest
sensitivity (31 and 35%, respectively) compared
with the receiver operating characteristic analysis
(specificity 64%, sensitivity 78%) [72]. Another
smaller study of 58 patients with pure SFN, which
included 19 diabetic patients, showed that a cut-off
IENF density of ≤8.8/mm at the ankle was associ-
ated with a sensitivity of 77.2% and a specificity of
79.6% in diagnosing SFN [73]. In a more recent
study in patients with T1D, IENFD showed a sen-
sitivity of 61% and specificity of 80% for DPN
[74]. The European Federation of the Neurological
Societies and the Peripheral Nerve Society con-
clude that IENFD is a reliable and efficient tech-
nique to confirm SFN with a level A
recommendation [67]. Skin biopsy with assess-
ment for IENFD may be used to confirm small
fiber predominant DPN if NCS are normal,
although a typical clinical DPN may be diagnosed
based on clinical criteria (as is the case with NCS).
3.5 Research Application of Skin
Biopsy
Skin biopsy is a powerful research tool for the
study of DPN and is commonly used as a surro-
gate measure of disease severity in natural his-
L. Davalos et al.
tory studies and clinical trials. Small
unmyelinated axons may be particularly suscep-
tible to injury while being more capable of
regeneration than large myelinated fibers [75].
These characteristics may explain why skin
biopsy is more sensitive to disease progression
than NCS [76, 77].
The relatively uniform and superficial loca-
tion of the IENF in the epidermis allows for the
standardized assessment of nerve injury and
regeneration. Several models take advantage of
the small fibers’ uniquely robust regenerative
capacity, allowing for the accurate measurement
of collateral [78] and regenerative sprouting [52]
in human subjects. Following removal of a con-
ventional 3 mm skin biopsy, the site heals by a
process of granulation and eventually becomes
re-epithelialized. The collagen plug within the
dermis of the healing biopsy site acts as a barrier
to the distal ends of the transected nerves.
Therefore, reinnervation of the epidermis within
the original 3  mm biopsy site can only occur
through sprouting of uninjured fibers surround-
ing the incision line. The degree of collateral
sprouting can be measured by a concentric 4 or
5 mm biopsy centered on the healed 3mm biopsy
site. The most promising technique to induce
cutaneous denervation involves application of
capsaicin either topically as a cream or injected
into the epidermis [79, 80]. When applied topi-
cally in an occlusive bandage, capsaicin pro-
duces a standardized injury that is reproducible
and well-tolerated [52]. Studies in healthy con-
trol subjects and diabetic patients demonstrate
that functional abnormalities in regenerative
sprouting are present among subjects with diabe-
tes with no signs or symptoms of neuropathy
[52], suggesting that regenerative abnormalities
may be an early sign of nerve dysfunction. Thus,
the capsaicin model holds promise as a sensitive
means of detecting early neuropathy and evalu-
ating the efficacy of potential therapeutic inter-
ventions, early in the disease course of
DPN. Given that reinnervation occurs relatively
rapidly, sometimes as early as 10–12 weeks, less
patients can be studied over a short period of
time with this technique, as opposed to NCS,
which may require larger numbers of patients
over years.
Diagnostic Techniques for Diabetic Peripheral Neuropathy
4 Quantitative Sensory Testing
Quantitative sensory testing (QST) is a psycho-
physical evaluation that is particularly valuable
in DPN clinical evaluation and research as it is
the only quantitative testing method that captures
the patient experience of sensory loss or neuro-
pathic pain [81]. Using pre-set automated vibra-
tion and thermal thresholds, QST allows for a
standardized application of stimulus intensity
and quantification sensitivity to stimuli for both
large and small fibers (vibration and thermal per-
ception respectively). A number of different test-
ing platforms are available.
QST provides relatively consistent inter and
intra rater reliability, and is now routinely used as
an endpoint in multicenter clinical trials of DPN
[82]. It has proven itself useful clinically to assess
for asymptomatic DPN [83], to uncover patients at
risk for foot ulceration [83], and to assess disease
severity longitudinally [84]. In one study of 117
patients with painful sensory neuropathy (not lim-
ited to diabetic patients), QST was found to have a
diagnostic sensitivity of 72%, compared to 59%
and 87.5% for quantitative sudomotor axon reflex
testing (QSART) and IENFD, respectively [85].
However, QST does suffer from inter-­assessment
a b
Fig. 4  The Rydel-Seiffer tuning fork is a semiquantita-
tive tool to assess vibration sensation. The tuning fork is
placed on the great toe (a) and the two arms of the tuning
101
variation of up to 150% when using the limits
method, which significantly declined when using
the levels method [86]. In contradistinction to
NCS, QST does not localize abnormal findings to
the peripheral nervous system, as aberrancies in
sensory perception can be traced to any site
between the skin and sensory cortex in the central
nervous system. In addition, QST can often be
bulky, expensive, and available only at referral
centers, limiting its widespread adoption. The
American Academy of Neurology assigned QST a
Class II rating as a diagnostic test, concluding it
was a useful tool in research studies but should not
be the sole criteria for neuropathy diagnosis [87].
4.1 Point of Care QST Devices
Given the challenges of using large, expensive,
and non-portable QST platforms, numerous
mobile point of care devices are available. Some
devices focus purely on one modality. Vibrometers
may be used to assess large myelinated A-beta
fibers. The simplest tool is the graduated Rydel-­
Seiffer tuning fork, which employs a semiquanti-
tative 8 point scale (Fig.  4). A battery-operated
handheld vibrometer (VibraTip™, McCallan
fork are pinched and released. The triangles formed by the
movement allow definition of the number when the sensa-
tion is no longer perceived (b)
102
Medical, Ottawa, Canada) may also be used to
assess vibration perception at the great toe using
a stimulus of 128 Hz, and has been shown to
associate with the neuropathy disability score
[88] and demonstrated agreement with a vibra-
tion perception threshold of >25 V, a value indi-
cating “at risk feet.” [89] The NeuroQuick™
(Schweers, Meerbusch, Germany) is a portable
device that assesses cold perception threshold
[90]. It features an in-built fan that emits cold air
at velocities of varied intensity onto the dorsum
of the foot. The NeuroQuick threshold is defined
as the airflow at which the subject first recognizes
the stimulus. The result correlates with thermal
perception thresholds on standard QST [90].
NeuroQuick is also able to discriminate between
diabetic patients with DPN, diabetic patients
without DPN, and control subjects. It is more
sensitive at detecting DPN than thermal percep-
tion threshold, making it a useful adjunctive test
to diagnose early and asymptomatic DPN [90].
One German study, however, found NeuroQuick
to suffer from low sensitivity in identifying pain-
ful SFN (irrespective of diabetes status), when
compared with standard QST cold detection
threshold technique [91]. The authors concluded
that it was not an adequate stand-alone screen for
cold hypoesthesia in patients with chronic neuro-
pathic pain.
Other devices are available to assess both
large and small fiber function. NerveCheck™
(Phi Med Europe S.L., Barcelona, Spain) is a
portable QST device that performs both vibratory
and thermal testing and has good test-retest reli-
ability and similar diagnostic accuracy to tradi-
tional QST equipment [92]. The Neurometer®
(Neurotron, Baltimore, MD) applies a constant
current via surface electrodes across three cur-
rencies—2000 Hz, 250 Hz, and 5 Hz—to selec-
tively stimulate different nerve fiber types [93].
The current perception threshold (CPT), the min-
imum intensity detectable, is used to ascertain the
presence and severity of DPN as well as the spe-
cific fiber type involved. Each stimulus frequency
selects for a particular fiber subtype, such that
2000  Hz stimulates large myelinated fibers and
5 Hz selects for small unmyelinated fibers. Two
electrodes coated with gel are applied to intact
L. Davalos et al.
skin and current is applied and increased until it
is felt. Once sensed, it is terminated, and
decreased in increments of 0.08 mA and reap-
plied. In such a fashion, the device calculates and
displays the CPT, which ranges from 1 to 25 (this
can be done for each frequency). Values from 6 to
13 are deemed normal, with values of 1–5 (hyper-
esthesia) and 14–25 (hypoesthesia) both being
abnormal and representing sensory neuropathy.
The Neurometer has been extensively studied in
DPN. T1D subjects with essentially normal neu-
ropathy symptom metrics have reduced CPT,
suggesting a potential role as a screening tool for
asymptomatic DPN patients [94]. T2DPN
patients have higher CPT than controls when
stimulated at 2000 Hz, and CCP negatively cor-
relates with motor and sensory nerve conduction
velocities [95]. CPT can distinguish patients with
DPN from control subjects at all three frequen-
cies. However, reproducibility is better in con-
trols than diabetic patients [96]. In addition, one
Korean study cast doubt on the additive value of
CPT beyond conventional DPN evaluations [97].
Among 241 Korean patients with diabetes, CPT
was higher at all frequencies among those with
DPN compared to those without. However, diag-
nostic sensitivity was low. For instance, 87% of
those with reduced sensation to a 10 g monofila-
ment had normal CPT at 2000 Hz. This study cast
doubt on the additive value of CPT beyond stan-
dard clinical measures (128 Hz tuning fork, 10-g
monofilament, and reflex sensitivity).
4.2 Application of QST in Clinical
Trials
While the diagnostic role of QST in clinical set-
tings is limited, and its correlation with pain
symptoms reported by patients is low [98], QST
remains a useful research tool that holds particu-
lar promise for predicting patient response to
neuropathic pain treatments. Numerous trials
have used QST metrics as outcome measures.
The German Research Network on Neuropathic
Pain (DFNS) has gathered QST and other pheno-
typic data on over 1400 patients with various
neuropathic pain states. The DFNS QST protocol
Diagnostic Techniques for Diabetic Peripheral Neuropathy
includes assessment of 13 different mechanical
and thermal tests, permitting a more nuanced
assessment of distinct pain phenotypes. Patients
with peripheral neuropathy most commonly have
sensory loss alone (52%), with 21% having a
mixture of positive and negative sensory signs
[99]. The DFNS protocol can be used to broadly
classify patients into three distinct sensory phe-
notypes: sensory loss (thermal and mechanical
sensory loss), thermal hyperalgesia (preserved
sensory function with heat or cold hyperalgesia),
and mechanical hyperalgesia (loss of thermal
sensory with mechanical hyperalgesia). Patients
may also be classified into those with sensory
loss (the “non-irritable nociceptor phenotype)
versus those with preserved small fiber function
but hyperalgesia (the “irritable nociceptor” pat-
tern) [100].
There is evidence to suggest that specific pain
phenotypes may predict treatment response. The
COMBO-DN study, which compared duloxetine
and pregabalin for painful DPN, demonstrated
that these agents have different effects on specific
aspects of neuropathic pain. For example, patients
with pressing pain and evoked pain who did not
respond to either agent did so when treated with
a combination of the two [101]. Among 29
patients with painful DPN treated with intrave-
nous lidocaine, those with the irritable nociceptor
phenotype were more likely to respond (odds
ratio 8.67, 95% CI 1.4–54.8). These data suggest
careful sensory phenotyping using clinical mea-
sures and QST may have particular utility in neu-
ropathic pain clinical trials in order to establish
predictive models for therapeutic response.
5 Sudomotor Testing
Sweat glands are innervated by postganglionic
cholinergic unmyelinated small fibers, thus
explaining why peripheral neuropathies involv-
ing small fibers also impact sudomotor function.
This may be clinically evident to the patient as
either reduced or increased sweating or it may be
asymptomatic. Sudomotor testing offers another
diagnostic and research tool for DPN. As is the
case with QST, there are a spectrum of tests rang-
103
ing from laboratory based assessments to simple
point of care tools. Sudomotor testing is a sensi-
tive diagnostic tool for small fiber neuropathy
caused by diabetes and other etiologies [102].
5.1 Quantitative Sudomotor Axon
Reflex Testing (QSART)
QSART is a well validated reproducible measure
of postganglionic sudomotor fibers. Acetylcholine
is iontophoresed through a central chamber of a
capsule placed on the skin, which activates nerve
fibers underlying the chamber. When the action
potential reaches a branch point in the nerve, it
propagates orthodromically along the other
branch, stimulating sweat glands. The QSART
device measures sweat production in the area
under the capsule surrounding the central cham-
ber. QSART is typically performed at four
locations.
Among 100 patients with suspected small
fiber neuropathy, the diagnostic utility of QSART
was compared to QST and IENFD. QSART had
higher sensitivity and specificity (82%, 89%)
than QST (88%, 50%) and IENFD (67%, 91%).
QSART and IENFD had positive predictive val-
ues of 90%, but the negative predictive value of
QSART was 80%, while it was 68% for
IENFD.  These results suggest that both IENFD
and QSART both have diagnostic value [103]. A
smaller study focusing on DPN, and which
included patients with both large and small fiber
predominant neuropathies, found IENFD to be
superior to QSART in patients with large fiber
involvement, whereas the diagnostic yield of
both was similar in those with small fiber neu-
ropathy [104]. Limitations of QSART include the
requirement for specialized equipment and the
need for meticulous attention to testing parame-
ters, including medication use and room tempera-
ture. For these reasons, test-retest reliability of
QSART is variable, with some reports suggesting
even marked variability [105, 106]. These factors
also limit the utility of QSART as an endpoint
measure in clinical trials. Nevertheless, in labora-
tories with robust quality control, QSART is a
valuable diagnostic and research tool.
104
5.2 Thermoregulatory Sweat
Testing
Thermoregulatory sweat testing (TST) is per-
formed by the application of a topical dye that
changes color when wet (e.g., cornstarch mixed
with alizarin red). The patient is then placed in a
chamber that is warmed with controlled humid-
ity. TST evaluates the sudomotor system in an
integrated fashion (i.e., not just postganglionic
fibers) across the entire body. The typical finding
in DPN is reduced sweating distally, although
patients with diabetes may have a spectrum of
abnormalities, including regional and global
defects due to focal neuropathies or generalized
autonomic neuropathy, respectively [107]. TST is
a valuable tool in the evaluation of patients with
sweating disorders, but it is not commonly per-
formed in the diagnosis of DPN.
5.3 Neuropad: Indicator
Plaster Test
Neuropad, sometimes referred to as the Indicator
Plaster Test, is a highly sensitive, practical, and
quick measure of sudomotor impairment in DPN
[108]. When the plaster is applied to the plantar
surface of the foot, typically between the first and
second metatarsal heads, the presence of healthy
and normal sweat output produces a color change
of the cobalt II constituent in the plaster from blue
to pink (within 10 min). Lack of color change
within a set time threshold identifies an impairment
in sudomotor function, and by extension, suggests
peripheral neuropathy. Neuropad is sensitive (65–
100%) in the diagnosis of DPN and has a high
negative predictive value (63–100%), but has poor
specificity (32–78.5%) and positive predictive
value (23.3–93.2%) [109]. Meta-analytic data sug-
gests an average sensitivity of 86% and specificity
of 65%, while a multicenter study shows an even
higher sensitivity of 94.9% and negative predictive
value of 98.1% [110]. Thus, it is a valuable and
convenient means of excluding DPN, but not use-
ful in confirming diagnosis. The combination of
monofilament testing with Neuropad is a cost-
effective screening strategy for DPN [111].
L. Davalos et al.
5.4 Electrochemical Sweat
Conductance (ESC)
ESC can be measured using Sudoscan™ (Impeto
Medical, Paris France), which is a novel technol-
ogy that relies upon the concept of electrochemi-
cal conductance to assess for neuropathy [112].
The palms and soles of the examinee are placed
on examining electrodes and a low-voltage con-
stant electrical current is applied. This current
attracts sodium chloride from the sweat on exam-
inee’s soles and palms onto the nickel in the
examining electrodes. By comparing the ratio of
current generated to current provided across the
tested surface areas, electrochemical skin con-
ductance (ESC) is calculated by the device and
serves as a measure of DPN severity. While ESC
requires specialized equipment, the testing is
very quick and well tolerated. Several studies
suggest ESC has good diagnostic performance
for DPN [113] [114]. However, concerns have
been raised regarding variability in ESC values,
lack of age and gender differences (unlike all
other clinical and laboratory based neuropathy
measures), and the degree to which ESC actually
measures small fiber function [115]. Given these
concerns, the clinical value of ESC for DPN
diagnosis remains uncertain.
6 Corneal Confocal Microscopy
The cornea is the most densely innervated part of
the human body, containing myelinated Aδ and
unmyelinated C-fibers, and derives its innerva-
tion from the ophthalmic division of the trigemi-
nal nerve. Corneal confocal microscopy (CCM)
is a noninvasive technique that allows sequential
in  vivo observation of the small unmyelinated
nerve fiber bundle of the cornea, comparable to
that obtained with histopathological examination
[116]. Images are acquired with a confocal
microscope (Heidelberg HRT III with a Rostock
Cornea Module, Heidelberg Engineering,
Heidelberg, Germany). Typically, 5–8 high-­
quality, non-overlapping images from the central
and peripheral-central cornea are chosen, after
which nerve morphology is analyzed to deter-
Diagnostic Techniques for Diabetic Peripheral Neuropathy
mine nerve fiber density (NFD fibers/mm2) and
nerve fiber length (mm/mm2) (Fig.  5) [117].
Normative data are available, which demonstrate
a decline in NFD and NFL with age [118, 119].
Many studies have shown good interobserver
and intraobserver reliability with most parame-
ters when a single image is pre-selected [120–
122]. However, when trying to replicate the
whole process with multiple images, NFL has the
best reproducibility [122]. This reinforces the
need of a standardized protocol for imaging
selection. Given the variability risk of manually
selecting corneal nerve fibers, an automated mea-
surement software has been created. Some stud-
ies report that the automated measurement
underestimates the manual measurement [123],
while others report a good correlation with both
methods [124].
Numerous studies suggest CCM is a sensitive
diagnostic measure of DPN [125–127]. In 2015,
a meta-analysis that evaluated the utility of CCM
in the assessment of DPN concluded that NFD,
NBD and NFL are all significantly reduced in
patients with DPN when compared to healthy
controls and diabetic patients without DPN [128].
However, most positive studies have traditionally
focused on small cohorts of T1D or a combina-
a b
Fig. 5  Corneal confocal microscopy (CCM) from a nor-
mal subject demonstrates normal nerve fiber length and
density (a). CCM from a patient with diabetic neuropathy
105
tion of T1D and T2D patients, and only a few
have adjusted for age [74, 129, 130], which is a
confounder [118]. A large cross sectional study
of T2D patients from the ADDITION trial found
that CCM could not distinguish patients with
DPN from patients without DPN, but NFD was
lower in T2D patients as compared to controls
[131]. These results suggested that CCM might
have a better neuropathy diagnostic value in
patients with DM1 than in DM2. In order to
assess the validity of CCM as a diagnostic tool
for DPN in a larger population, a multinational
cohort of 998 diabetic patients (516 with T1DM
and 482 with T2DM) was studied. The results
showed that NFL, NFD and NBD were signifi-
cantly lower in T1D patients with DPN as com-
pared to T1D patients without DPN, and only
NFL and NBD were significantly lower in T2D
patients with DPN as compared toT2D patients
without DPN [124]. This data supported the find-
ings of prior small cohorts, suggesting that CCM
is a reliable diagnostic tool in both diabetic neu-
ropathy types.
The appeal of CCM as a diagnostic and
research tool is that it assesses the same fiber
class as IENFD in a noninvasive and easily
repeatable fashion. Studies exploring the correla-
demonstrates reduced nerve fiber length and density as
well as proliferation of Langerhan’s Cells (arrowheads)
(b)
106 L. Davalos et al.

tion between CCM and IENFD provide a valu-
able window into the biological meaning of
CCM, particularly since DPN clinically involves
the longest nerves in the body, while CCM mea-
sures some of the shortest nerves. Several small
studies suggest CCM performs comparably well
to IEFND in patients with T1D [74, 129], while
another study did not find a correlation between
CCM and IENFD in T2D patients [132]. A third
study found that CCM’s diagnostic performance
was slightly lower than IENFD for DPN [133].
While larger longitudinal studies comparing
CCM to other neuropathy measures are neces-
sary to determine its clinical meaning and role as
an outcome measure in clinical trials, it remains a
promising research tool.
7 Conclusion
The clinical diagnosis of DPN is based on a com-
bination of symptoms and signs and can be sup-
ported by application of several diagnostic tests
that assess both small and large fiber function
(Table 2). While these tests are not always neces-
sary in settings where patients present with typi-

Table 2  Diagnostic techniques for diabetic peripheral neuropathy

Diagnostic Nerve fiber type
technique target Advantages Disadvantages
Nerve conduction Large diameter Objective, high intra-observer Do not assess small fibers, high
studies myelinated reliability for conduction interobserver variability, require referral
fibers velocity measurement, widely to a neurophysiology lab with trained
available, gold standard for examiners
large fiber neuropathy
Skin biopsy Small diameter Objective, gold standard for Minimally invasive, technically
(IEFND) unmyelinated small fiber neuropathy, demanding requiring an expert
nerve fibers strongest research outcome reference laboratory with robust quality
measure for DPN, high control procedures
intraobserver and interobserver
reliability, can assess early
reinnervation
Quantitative Small and large Fast and easy to perform, Subjective, costly, requires specialized
sensory testing nerve fibers consistent inter and intra rater equipment, abnormality does not
reliability, useful in necessarily localize to the peripheral
phenotyping pain and nervous system
predicting treatment response.
The only technique that
captures the patient experience
Sudomotor Autonomic Fast and easy to perform, Test-retest reliability can vary due to
testing (QSART) small diameter objective, reproducible external factors (e.g. medications), role
unmyelinated limited in research, requires specialized
fibers equipment
Electrochemical Autonomic Fast and easy to perform Requires specialized equipment. High
Sweat small diameter degree of test retest variability, lack of
Conductance unmyelinated age and gender differences (unlike all
fibers other clinical and laboratory based
neuropathy measures), and the degree to
which it actually measures small fiber
function
Corneal confocal Small diameter Noninvasive, reproducible, Not widely available, need for further
microscopy unmyelinated objective, and easily repeatable validation, technique assesses short
nerve fibers length nerve fibers in a disease that is
generally viewed as length dependent
(i.e., involving the longest fibers)
IEFND intraepidermal nerve fiber density, QSART quantitative sudomotor axon reflex testing, PNS peripheral nervous
system
Diagnostic Techniques for Diabetic Peripheral Neuropathy 107

cal clinical features of DPN, they can be very
useful in situations where there are atypical clini-
cal features. Some are also well suited at screen-
ing patients for early or preclinical neuropathy or
to identify those at risk for important health out-
comes (e.g., foot ulceration), and variety of point
of care devices are available (Table  3). Many
measures have particular value as research tools.
While NCS remain the mainstay for the clinical
evaluation of large fiber neuropathy and IENFD
is the current gold standard for small fiber neu-
ropathy, a variety of point of care devices and
tools and newer promising techniques, including
CCM, are now available.

Table 3  Point of care devices for diagnosing diabetic peripheral neuropathy

Diagnostic Nerve fiber type
technique Device target Advantages Disadvantages
Nerve Point of care Large nerve High sensitivity, Overestimates sural nerve
conduction nerve fibers acceptable specificity, conduction velocity
studies conduction good intra and inter rater
studies reliability, fast,
inexpensive, easy to use
Quantitative Vibrometer Large nerve Portable, inexpensive,
sensory fibers easy to perform,
testing associates with
neuropathy disability
score
NeuroQuick Small Portable, discriminates Low sensitivity for painful SFN
myelinated between DPN and compared with standard QST
nerve fibers control subjects, more cold detection threshold; not
(cold detection) sensitive than thermal adequate as a stand-alone
perception threshold measure of impairment in cold
perception
NerveCheck Large and small Portable, good test-retest
nerve fibers reliability, comparable
diagnostic accuracy to
traditional QST
Neurometer Large Discriminates DPN from Reproducibility better in controls
myelinated and control subjects well than diabetic patients. Sensitivity
small may be low
myelinated
nerve fibers
Sudomotor Neuropad Autonomic Sensitive and high Low specificity, not useful at
function small nerve negative predictive value, confirming diagnosis of DPN
fibers fast, easy to perform, and
well tolerated
108 L. Davalos et al.

References 15. Amthor KF, Dahl-Jørgensen K, Berg TJ, Heier MS,

Sandvik L, Aagenaes O, et  al. The effect of 8 years
of strict glycaemic control on peripheral nerve func-
1. Tesfaye S, Boulton AJM, Dyck PJ, Freeman R,
tion in IDDM patients: the Oslo Study. Diabetologia.
Horowitz M, Kempler P, et  al. Diabetic neuropa-
1994;37(6):579–84.
thies: update on definitions, diagnostic criteria, esti-
16. Andersen H, Stålberg E, Falck B.  F-wave latency,
mation of severity, and treatments. Diabetes Care.
the most sensitive nerve conduction parameter
2010;33(10):2285–93.
in patients with diabetes mellitus. Muscle Nerve.
2. Smith AG. Do all neuropathy patients need an EMG at
1997;20(10):1296–302.
least once? Contin Minneap Minn. 2014;20(5):1430–4.
17. Kohara N, Kimura J, Kaji R, Goto Y, Ishii J, Takiguchi
3. Divisova S, Vlckova E, Hnojcikova M, Skorna
M, et  al. F-wave latency serves as the most repro-
M, Nemec M, Dubovy P, et  al. Prediabetes/early
ducible measure in nerve conduction studies of dia-
diabetes-associated neuropathy predominantly
betic polyneuropathy: multicentre analysis in healthy
involves sensory small fibres. J Peripher Nerv Syst.
subjects and patients with diabetic polyneuropathy.
2012;17(3):341–50.
Diabetologia. 2000;43(7):915–21.
4. Singleton JR, Bixby B, Russell JW, Feldman EL,
18. Dyck PJ, Lais A, Karnes JL, O’Brien P, Rizza R. Fiber
Peltier A, Goldstein J, et  al. The Utah early neu-
loss is primary and multifocal in sural nerves in diabetic
ropathy scale: a sensitive clinical scale for early sen-
polyneuropathy. Ann Neurol. 1986;19(5):425–39.
sory predominant neuropathy. J Peripher Nerv Syst.
19. Behse F, Buchthal F, Carlsen F.  Nerve biopsy and
2008;13(3):218–27.
conduction studies in diabetic neuropathy. J Neurol
5. Andersson C, Guttorp P, Särkkä A. Discovering early
Neurosurg Psychiatry. 1977;40(11):1072–82.
diabetic neuropathy from epidermal nerve fiber pat-
20. Misawa S, Kuwabara S, Kanai K, Tamura N, Hiraga
terns. Stat Med. 2016;35(24):4427–42.
A, Nakata M, et  al. Axonal potassium conductance
6. Devigili G, Tugnoli V, Penza P, Camozzi F, Lombardi
and glycemic control in human diabetic nerves.
R, Melli G, et al. The diagnostic criteria for small fibre
Clin Neurophysiol Off J Int Fed Clin Neurophysiol.
neuropathy: from symptoms to neuropathology. Brain
2005;116(5):1181–7.
J Neurol. 2008;131(7):1912–25.
21. Yagihashi S, Yamagishi S-I, Wada R.  Pathology and
7. Lauria G, Lombardi R, Camozzi F, Devigili G. Skin
pathogenetic mechanisms of diabetic neuropathy: cor-
biopsy for the diagnosis of peripheral neuropathy.
relation with clinical signs and symptoms. Diabetes
Histopathology. 2009;54(3):273–85.
Res Clin Pract. 2007;77(1):184–9.
8. Sivaskandarajah GA, Halpern EM, Lovblom LE,
22. Tomlinson DR.  Mitogen-activated protein kinases
Weisman A, Orlov S, Bril V, et al. Structure-function
as glucose transducers for diabetic complications.
relationship between corneal nerves and conventional
Diabetologia. 1999;42(11):1271–81.
small-fiber tests in type 1 diabetes. Diabetes Care.
23. American Diabetes Association American Academy
2013;36(9):2748–55.
of Neurology. Consensus statement: report and rec-
9. Callaghan BC, Price RS, Chen KS, Feldman EL. The
ommendations of the San Antonio conference on dia-
importance of rare subtypes in diagnosis and treat-
betic neuropathy. Diabetes Care. 1988;11(7):592–7.
ment of peripheral neuropathy: a review. JAMA
24. England JD, Gronseth GS, Franklin G, Miller RG,
Neurol. 2015;72(12):1510–8.
Asbury AK, Carter GT, et al. Distal symmetric poly-
10. Gooch CL, Weimer LH. The electrodiagnosis of neu-
neuropathy: a definition for clinical research: report of
ropathy: basic principles and common pitfalls. Neurol
the American Academy of Neurology, the American
Clin. 2007;25(1):1–28.
Association of Electrodiagnostic Medicine, and
11. Weisman A, Bril V, Ngo M, Lovblom LE, Halpern
the American Academy of Physical Medicine and
EM, Orszag A, et  al. Identification and prediction
Rehabilitation. Neurology. 2005;64(2):199–207.
of diabetic sensorimotor polyneuropathy using indi-
25. Dyck PJ, Carter RE, Litchy WJ. Modeling nerve con-
vidual and simple combinations of nerve conduction
duction criteria for diagnosis of diabetic polyneuropa-
study parameters. PLoS One. 2013;8(3):e58783.
thy. Muscle Nerve. 2011;44(3):340–5.
12. Russell JW, Karnes JL, Dyck PJ. Sural nerve myelin-
26. Callaghan B, McCammon R, Kerber K, Xu X, Langa
ated fiber density differences associated with mean-
KM, Feldman E. Tests and expenditures in the initial
ingful changes in clinical and electrophysiologic
evaluation of peripheral neuropathy. Arch Intern Med.
measurements. J Neurol Sci. 1996;135(2):114–7.
2012;172(2):127–32.
13. Partanen J, Niskanen L, Lehtinen J, Mervaala E,
27. Callaghan BC, Kerber KA, Lisabeth LL, Morgenstern
Siitonen O, Uusitupa M. Natural history of peripheral
LB, Longoria R, Rodgers A, et  al. Role of neu-
neuropathy in patients with non-­ insulin-dependent
rologists and diagnostic tests on the management of
diabetes mellitus. N Engl J Med. 1995;333(2):89–94.
distal symmetric polyneuropathy. JAMA Neurol.
14. Nathan DM, Genuth S, Lachin J, Cleary P, Crofford
2014;71(9):1143–9.
O, et  al. The effect of intensive treatment of diabe-
28. Pop-Busui R, Boulton AJM, Feldman EL, Bril V,
tes on the development and progression of long-term
Freeman R, Malik RA, et  al. Diabetic neuropa-
complications in insulin-dependent diabetes mellitus.
thy: a position statement by the American Diabetes
N Engl J Med. 1993;329(14):977–86.
Association. Diabetes Care. 2017;40(1):136–54.
Diagnostic Techniques for Diabetic Peripheral Neuropathy
29. Mulder DW, Lambert EH, Bastron JA, Sprague
RG.  The neuropathies associated with diabetes
mellitus. A clinical and electromyographic study
of 103 unselected diabetic patients. Neurology.
1961;11(4):275–84.
30. Lamontagne A, Buchthal F.  Electrophysiological
studies in diabetic neuropathy. J Neurol Neurosurg
Psychiatry. 1970;33(4):442–52.
31. Dyck PJ, Davies JL, Litchy WJ, O’Brien
PC.  Longitudinal assessment of diabetic polyneu-
ropathy using a composite score in the Rochester
Diabetic Neuropathy Study cohort. Neurology.
1997;49(1):229–39.
32. Dyck PJ, Karnes JL, O’Brien PC, Litchy WJ, Low
PA, Melton LJ.  The Rochester Diabetic Neuropathy
Study: reassessment of tests and criteria for diagnosis
and staged severity. Neurology. 1992;42(6):1164–70.
33. Author. Effect of intensive diabetes treatment on nerve
conduction in the diabetes control and complications
trial. Ann Neurol. 1995;38(6):869–80.
34. Bril V.  Role of electrophysiological studies in dia-
betic neuropathy. Can J Neurol Sci J Can Sci Neurol.
1994;21(4):8–12.
35. Bril V, Ellison R, Ngo M, Bergstrom B, Raynard D,
Gin H. Electrophysiological monitoring in clinical tri-
als. Roche Neuropathy Study Group. Muscle Nerve.
1998;21(11):1368–73.
36. Dyck PJ, Overland CJ, Low PA, Litchy WJ, Davies
JL, Dyck PJB, et  al. Signs and symptoms versus
nerve conduction studies to diagnose diabetic sen-
sorimotor polyneuropathy: Cl vs. Muscle Nerve.
2010;42(2):157–64.
37. Perkins BA, Greene DA, Bril V.  Glycemic con-
trol is related to the morphological severity of dia-
betic sensorimotor polyneuropathy. Diabetes Care.
2001;24(4):748–52.
38. Perkins BA, Olaleye D, Zinman B, Bril V.  Simple
screening tests for peripheral neuropathy in the diabe-
tes clinic. Diabetes Care. 2001;24(2):250–6.
39. Perkins BA, Bril V.  Diabetic neuropathy: a review
emphasizing diagnostic methods. Clin Neurophysiol
Off J Int Fed Clin Neurophysiol. 2003;114(7):1167–75.
40. Dyck PJ, Litchy WJ, Lehman KA, Hokanson JL, Low
PA, O’Brien PC.  Variables influencing neuropathic
endpoints: the Rochester Diabetic Neuropathy Study
of Healthy Subjects. Neurology. 1995;45(6):1115–21.
41. Dyck PJ, Albers JW, Wolfe J, Bolton CF, Walsh N,
Klein CJ, et  al. A trial of proficiency of nerve con-
duction: greater standardization still needed. Muscle
Nerve. 2013;48(3):369–74.
42. Litchy WJ, Albers JW, Wolfe J, Bolton CF, Walsh N,
Klein CJ, et al. Proficiency of nerve conduction using
standard methods and reference values (Cl. NPhys
Trial 4). Muscle Nerve. 2014;50(6):900–8.
43. Dyck PJ, Kratz KM, Karnes JL, Litchy WJ, Klein
R, Pach JM, et al. The prevalence by staged severity
of various types of diabetic neuropathy, retinopathy,
and nephropathy in a population-based cohort: the
Rochester Diabetic Neuropathy Study. Neurology.
1993;43(4):817–24.
109
44. Boulton AJ, Knight G, Drury J, Ward JD.  The
prevalence of symptomatic, diabetic neuropathy
in an insulin-treated population. Diabetes Care.
1985;8(2):125–8.
45. Dyck PJ, Litchy WJ, Daube JR, Harper CM, Dyck
PJB, Davies J, et  al. Individual attributes versus
composite scores of nerve conduction abnormality:
sensitivity, reproducibility, and concordance with
impairment. Muscle Nerve. 2003;27(2):202–10.
46. Feldman EL, Stevens MJ, Thomas PK, Brown MB,
Canal N, Greene DA.  A practical two-step quantita-
tive clinical and electrophysiological assessment for
the diagnosis and staging of diabetic neuropathy.
Diabetes Care. 1994;17(11):1281–9.
47. Dyck PJ, Albers JW, Andersen H, Arezzo JC,
Biessels G-J, Bril V, et al. Diabetic polyneuropathies:
update on research definition, diagnostic criteria
and estimation of severity. Diabetes Metab Res Rev.
2011;27(7):620–8.
48. Malik RA, Williamson S, Abbott C, Carrington
AL, Iqbal J, Schady W, et  al. Effect of angiotensin-­
converting-­enzyme (ACE) inhibitor trandolapril on
human diabetic neuropathy: randomised double-blind
controlled trial. Lancet. 1998;352(9145):1978–81.
49. Wahren J, Foyt H, Daniels M, Arezzo JC.  Long-­
acting c-peptide and neuropathy in type 1 dia-
betes: a 12-month clinical trial. Diabetes Care.
2016;39(4):596–602.
50. Ruggenenti P, Lauria G, Iliev IP, Fassi A, Ilieva AP,
Rota S, et  al. Effects of manidipine and delapril in
hypertensive patients with type 2 diabetes mellitus:
the delapril and manidipine for nephroprotection
in diabetes (DEMAND) randomized clinical trial.
Hypertens Dallas Tex. 2011;58(5):776–83.
51. Dyck PJ, Norell JE, Tritschler H, Schuette K,
Samigullin R, Ziegler D, et  al. Challenges in design
of multicenter trials: end points assessed longitudi-
nally for change and monotonicity. Diabetes Care.
2007;30(10):2619–25.
52. Polydefkis M, Hauer P, Sheth S, Sirdofsky M, Griffin
JW, McArthur JC. The time course of epidermal nerve
fibre regeneration: studies in normal controls and in
people with diabetes, with and without neuropathy.
Brain J Neurol. 2004;127(7):1606–15.
53. Lee JA, Halpern EM, Lovblom LE, Yeung E, Bril V,
Perkins BA. Reliability and validity of a point-of-care
sural nerve conduction device for identification of dia-
betic neuropathy. PLoS One. 2014;9(1):e86515.
54. Langerhans P.  Ueber die Nerven der menschlichen
Haut. Arch Für Pathol Anat Physiol Für Klin Med.
1868;44(2):325–37.
55. Dyck PJ, Winkelmann RK, Bolton CF. Quantitation
of Meissner’s corpuscles in hereditary neurologic
disorders. Charcot-Marie-Tooth disease, Roussy-­
Levy syndrome, Dejerine-Sottas disease, hereditary
sensory neuropathy, spinocerebellar degenerations,
and hereditary spastic paraplegia. Neurology.
1966;16(1):10–7.
56. Dalsgaard CJ, Rydh M, Haegerstrand A.  Cutaneous
innervation in man visualized with protein gene
110
product 9.5 (PGP 9.5) antibodies. Histochemistry.
1989;92(5):385–90.
57. Kanitakis J.  Anatomy, histology and immunohisto-
chemistry of normal human skin. Eur J Dermatol.
2002;12(4):390–9; quiz 400–1
58. Wendelschafer-Crabb G, Kennedy WR, Walk
D. Morphological features of nerves in skin biopsies.
J Neurol Sci. 2006;242(1–2):15–21.
59. McArthur JC, Stocks EA, Hauer P, Cornblath DR,
Griffin JW.  Epidermal nerve fiber density: norma-
tive reference range and diagnostic efficiency. Arch
Neurol. 1998;55(12):1513–20.
60. Liu Y, Fan X, Wei Y, Piao Z, Jiang X. Intraepidermal
nerve fiber density of healthy human. Neurol Res.
2014;36(10):911–4.
61. Lauria G, Cornblath DR, Johansson O, McArthur JC,
Mellgren SI, Nolano M, et al. EFNS guidelines on the
use of skin biopsy in the diagnosis of peripheral neu-
ropathy. Eur J Neurol. 2005;12(10):747–58.
62. Lauria G, Bakkers M, Schmitz C, Lombardi R, Penza
P, Devigili G, et  al. Intraepidermal nerve fiber den-
sity at the distal leg: a worldwide normative reference
study. J Peripher Nerv Syst. 2010;15(3):202–7.
63. Smith AG, Howard JR, Kroll R, Ramachandran P,
Hauer P, Singleton JR, et  al. The reliability of skin
biopsy with measurement of intraepidermal nerve
fiber density. J Neurol Sci. 2005;228(1):65–9.
64. Stocks EA, McArthur JC, Griffen JW, Mouton PR. An
unbiased method for estimation of total epidermal
nerve fibre length. J Neurocytol. 1996;25(11):637–44.
65. Hilliges M, Johansson O.  Comparative analysis of
numerical estimation methods of epithelial nerve
fibers using tissue sections. J Peripher Nerv Syst.
1999;4(1):53–7.
66. Koskinen M, Hietaharju A, Kyläniemi M, Peltola J,
Rantala I, Udd B, et al. A quantitative method for the
assessment of intraepidermal nerve fibers in small-­
fiber neuropathy. J Neurol. 2005;252(7):789–94.
67. Joint Task Force of the EFNS and the PNS. European
Federation of Neurological Societies/Peripheral Nerve
Society Guideline on the use of skin biopsy in the
diagnosis of small fiber neuropathy. Report of a joint
task force of the European Federation of Neurological
Societies and the Peripheral Nerve Society. J Peripher
Nerv Syst. 2010;15(2):79–92.
68. Khoshnoodi MA, Truelove S, Burakgazi A, Hoke A,
Mammen AL, Polydefkis M.  Longitudinal assess-
ment of small fiber neuropathy: evidence of a non-­
length-­dependent distal axonopathy. JAMA Neurol.
2016;73(6):684–90.
69. Quattrini C, Tavakoli M, Jeziorska M, Kallinikos P,
Tesfaye S, Finnigan J, et  al. Surrogate markers of
small fiber damage in human diabetic neuropathy.
Diabetes. 2007;56(8):2148–54.
70. Løseth S, Mellgren SI, Jorde R, Lindal S, Stålberg
E.  Polyneuropathy in type 1 and type 2 diabetes:
comparison of nerve conduction studies, thermal per-
ception thresholds and intraepidermal nerve fibre den-
sities. Diabetes Metab Res Rev. 2010;26(2):100–6.
L. Davalos et al.
71. Sorensen L, Molyneaux L, Yue DK. The relationship
among pain, sensory loss, and small nerve fibers in
diabetes. Diabetes Care. 2006;29(4):883–7.
72. Nebuchennykh M, Løseth S, Lindal S, Mellgren
SI. The value of skin biopsy with recording of intraepi-
dermal nerve fiber density and quantitative sensory
testing in the assessment of small fiber involvement
in patients with different causes of polyneuropathy. J
Neurol. 2009;256(7):1067–75.
73. Vlcková-Moravcová E, Bednarík J, Dusek L, Toyka
KV, Sommer C.  Diagnostic validity of epidermal
nerve fiber densities in painful sensory neuropathies.
Muscle Nerve. 2008;37(1):50–60.
74. Alam U, Jeziorska M, Petropoulos IN, Asghar O,
Fadavi H, Ponirakis G, et  al. Diagnostic utility of
corneal confocal microscopy and intra-epidermal
nerve fibre density in diabetic neuropathy. PLoS One.
2017;12(7):e0180175.
75. Griffin JW, Thompson WJ.  Biology and pathol-
ogy of nonmyelinating Schwann cells. Glia.
2008;56(14):1518–31.
76. Smith AG, Russell J, Feldman EL, Goldstein J, Peltier
A, Smith S, et  al. Lifestyle intervention for pre-dia-
betic neuropathy. Diabetes Care. 2006;29(6):1294–9.
77. Weintraub MI, Herrmann DN, Smith AG, Backonja
MM, Cole SP. Pulsed electromagnetic fields to reduce
diabetic neuropathic pain and stimulate neuronal
repair: a randomized controlled trial. Arch Phys Med
Rehabil. 2009;90(7):1102–9.
78. Rajan B, Polydefkis M, Hauer P, Griffin JW, McArthur
JC.  Epidermal reinnervation after intracutaneous
axotomy in man. J Comp Neurol. 2003;457(1):24–36.
79. Simone DA, Nolano M, Johnson T, Wendelschafer-­
Crabb G, Kennedy WR.  Intradermal injection of
capsaicin in humans produces degeneration and
subsequent reinnervation of epidermal nerve fibers:
correlation with sensory function. J Neurosci.
1998;18(21):8947–59.
80. Nolano M, Simone DA, Wendelschafer-Crabb G,
Johnson T, Hazen E, Kennedy WR. Topical capsaicin
in humans: parallel loss of epidermal nerve fibers and
pain sensation. Pain. 1999;81(1–2):135–45.
81. Siao P, Cros DP.  Quantitative sensory testing. Phys
Med Rehabil Clin N Am. 2003;14(2):261–86.
82. Dyck PJ, Argyros B, Russell JW, Gahnstrom LE,
Nalepa S, Albers JW, et al. Multicenter trial of the pro-
ficiency of smart quantitative sensation tests. Muscle
Nerve. 2014;49(5):645–53.
83. Krumova EK, Geber C, Westermann A, Maier
C.  Neuropathic pain: is quantitative sensory testing
helpful? Curr Diab Rep. 2012;12(4):393–402.
84. Brown MJ, Bird SJ, Watling S, Kaleta H, Hayes L,
Eckert S, et al. Natural progression of diabetic periph-
eral neuropathy in the Zenarestat study population.
Diabetes Care. 2004;27(5):1153–9.
85. Periquet MI, Novak V, Collins MP, Nagaraja HN,
Erdem S, Nash SM, et al. Painful sensory neuropathy:
prospective evaluation using skin biopsy. Neurology.
1999;53(8):1641–7.
Diagnostic Techniques for Diabetic Peripheral Neuropathy
86. Fagius J, Wahren LK.  Variability of sensory thresh-
old determination in clinical use. J Neurol Sci.
1981;51(1):11–27.
87. Shy ME, Frohman EM, So YT, Arezzo JC, Cornblath
DR, Giuliani MJ, et  al. Quantitative sensory test-
ing: report of the Therapeutics and Technology
Assessment Subcommittee of the American Academy
of Neurology. Neurology. 2003;60(6):898–904.
88. Bowling FL, Abbott CA, Harris WE, Atanasov S,
Malik RA, Boulton AJM. A pocket-sized disposable
device for testing the integrity of sensation in the
outpatient setting. Diabet Med J Br Diabet Assoc.
2012;29(12):1550–2.
89. Bracewell N, Game F, Jeffcoate W, Scammell
BE. Clinical evaluation of a new device in the assess-
ment of peripheral sensory neuropathy in diabetes.
Diabet Med J Br Diabet Assoc. 2012;29(12):1553–5.
90. Ziegler D, Siekierka-Kleiser E, Meyer B, Schweers
M.  Validation of a novel screening device
(NeuroQuick) for quantitative assessment of small
nerve fiber dysfunction as an early feature of diabetic
polyneuropathy. Diabetes Care. 2005;28(5):1169–74.
91. Haussleiter IS, Richter H, Scherens A, Schwenkreis P,
Tegenthoff M, Maier C. NeuroQuick–a novel bedside
test for small fiber neuropathy? Eur J Pain Lond Engl.
2008;12(8):1000–7.
92. Ponirakis G, Odriozola MN, Odriozola S, Petropoulos
IN, Azmi S, Fadavi H, et al. NerveCheck: an inexpen-
sive quantitative sensory testing device for patients
with diabetic neuropathy. Diabetes Res Clin Pract.
2016;113:101–7.
93. Inceu GV, Veresiu IA.  Measurement of current
perception thresholds using the Neurometer®  -
applicability in diabetic neuropathy. Clujul Med.
2015;88(4):449–52.
94. Lv S, Fang C, Hu J, Huang Y, Yang B, Zou R, et al.
Assessment of peripheral neuropathy using measure-
ment of the current perception threshold with the
Neurometer® in patients with type 1 diabetes melli-
tus. Diabetes Res Clin Pract. 2015;109(1):130–4.
95. Matsutomo R, Takebayashi K, Aso Y.  Assessment
of peripheral neuropathy using measurement of the
current perception threshold with the neurometer in
patients with type 2 diabetes mellitus. J Int Med Res.
2005;33(4):442–53.
96. Pitei DL, Watkins PJ, Stevens MJ, Edmonds ME. The
value of the Neurometer in assessing diabetic neurop-
athy by measurement of the current perception thresh-
old. Diabet Med J Br Diabet Assoc. 1994;11(9):872–6.
97. Koo BK, Ohn JH, Kwak S-H, Moon MK. Assessment
of diabetic polyneuropathy and autonomic neuropa-
thy using current perception threshold in Korean
patients with diabetes mellitus. Diabetes Metab J.
2014;38(4):285–93.
98. Gierthmühlen J, Binder A, Förster M, Baron R.  Do
we measure what patients feel? An analysis of cor-
respondence between somatosensory modalities upon
quantitative sensory testing and self-reported pain
experience. Clin J Pain. 2018;34(7):610–7.
111
99. Maier C, Baron R, Tölle TR, Binder A, Birbaumer
N, Birklein F, et  al. Quantitative sensory testing
in the German Research Network on Neuropathic
Pain (DFNS): somatosensory abnormalities in 1236
patients with different neuropathic pain syndromes.
Pain. 2010;150(3):439–50.
100. Themistocleous AC, Ramirez JD, Shillo PR, Lees
JG, Selvarajah D, Orengo C, et  al. The Pain in
Neuropathy Study (PiNS): a cross-sectional obser-
vational study determining the somatosensory phe-
notype of painful and painless diabetic neuropathy.
Pain. 2016;157(5):1132–45.
101. Bouhassira D, Wilhelm S, Schacht A, Perrot S,
Kosek E, Cruccu G, et  al. Neuropathic pain phe-
notyping as a predictor of treatment response in
painful diabetic neuropathy: data from the ran-
domized, double-blind, COMBO-DN study. Pain.
2014;155(10):2171–9.
102. Low VA, Sandroni P, Fealey RD, Low PA. Detection
of small-fiber neuropathy by sudomotor testing.
Muscle Nerve. 2006;34(1):57–61.
103. Thaisetthawatkul P, Fernandes Filho JAM,
Herrmann DN.  Contribution of QSART to the
diagnosis of small fiber neuropathy. Muscle Nerve.
2013;48(6):883–8.
104. Abuzinadah AR, Kluding P, Wright D, D’Silva L,
Ryals J, Hendry B, et  al. Less is more in diabetic
neuropathy diagnosis: comparison of quantita-
tive sudomotor axon reflex and skin biopsy. J Clin
Neuromuscul Dis. 2017;19(1):5–11.
105. Berger MJ, Kimpinski K.  Test-retest reliability of
quantitative sudomotor axon reflex testing. J Clin
Neurophysiol Off Publ Am Electroencephalogr Soc.
2013;30(3):308–12.
106. Peltier A, Smith AG, Russell JW, Sheikh K, Bixby B,
Howard J, et al. Reliability of quantitative sudomo-
tor axon reflex testing and quantitative sensory test-
ing in neuropathy of impaired glucose regulation.
Muscle Nerve. 2009;39(4):529–35.
107. Fealey RD, Low PA, Thomas JE. Thermoregulatory
sweating abnormalities in diabetes mellitus. Mayo
Clin Proc. 1989;64(6):617–28.
108. Papanas N, Papatheodorou K, Christakidis D,
Papazoglou D, Giassakis G, Piperidou H, et  al.
Evaluation of a new indicator test for sudomotor
function (Neuropad) in the diagnosis of peripheral
neuropathy in type 2 diabetic patients. Exp Clin
Endocrinol Diabetes Off J Ger Soc Endocrinol Ger
Diabetes Assoc. 2005;113(4):195–8.
109. Papanas N, Boulton AJM, Malik RA, Manes
C, Schnell O, Spallone V, et  al. A simple new
­non-­invasive sweat indicator test for the diagnosis of
diabetic neuropathy. Diabet Med J Br Diabet Assoc.
2013;30(5):525–34.
110. Tsapas A, Liakos A, Paschos P, Karagiannis T,
Bekiari E, Tentolouris N, et  al. A simple plaster
for screening for diabetic neuropathy: a diagnostic
test accuracy systematic review and meta-analysis.
Metabolism. 2014;63(4):584–92.
112
111. Rodríguez-Sánchez B, Peña-Longobardo
LM, Sinclair AJ.  Cost-effectiveness analy-
sis of the Neuropad device as a screening tool
for early diabetic peripheral neuropathy. Eur J
Health Econ HEPAC Health Econ Prev Care.
2020;21(3):335–49.
112. Calvet JH, Dupin J, Winiecki H, Schwarz
PEH. Assessment of small fiber neuropathy through
a quick, simple and non invasive method in a German
diabetes outpatient clinic. Exp Clin Endocrinol
Diabetes Off J Ger Soc Endocrinol Ger Diabetes
Assoc. 2013;121(2):80–3.
113. Casellini CM, Parson HK, Richardson MS, Nevoret
ML, Vinik AI.  Sudoscan, a noninvasive tool for
detecting diabetic small fiber neuropathy and
autonomic dysfunction. Diabetes Technol Ther.
2013;15(11):948–53.
114. Krieger S-M, Reimann M, Haase R, Henkel E,
Hanefeld M, Ziemssen T. Sudomotor testing of dia-
betes polyneuropathy. Front Neurol. 2018;9:803.
115. Rajan S, Campagnolo M, Callaghan B, Gibbons
CH. Sudomotor function testing by electrochemical
skin conductance: does it really measure sudomotor
function? Clin Auton Res Off J Clin Auton Res Soc.
2019;29(1):31–9.
116. Oliveira-Soto L, Efron N.  Morphology of cor-
neal nerves using confocal microscopy. Cornea.
2001;20(4):374–84.
117. Petropoulos IN, Alam U, Fadavi H, Marshall A,
Asghar O, Dabbah MA, et al. Rapid automated diag-
nosis of diabetic peripheral neuropathy with in vivo
corneal confocal microscopy. Invest Ophthalmol Vis
Sci. 2014;55(4):2071–8.
118. Tavakoli M, Ferdousi M, Petropoulos IN, Morris J,
Pritchard N, Zhivov A, et  al. Normative values for
corneal nerve morphology assessed using corneal
confocal microscopy: a multinational normative data
set. Diabetes Care. 2015;38(5):838–43.
119. Dehghani C, Pritchard N, Edwards K, Vagenas D,
Russell AW, Malik RA, et al. Morphometric stability
of the corneal subbasal nerve plexus in healthy indi-
viduals: a 3-year longitudinal study using corneal
confocal microscopy. Invest Ophthalmol Vis Sci.
2014;55(5):3195–9.
120. Petropoulos IN, Manzoor T, Morgan P, Fadavi H,
Asghar O, Alam U, et  al. Repeatability of in  vivo
corneal confocal microscopy to quantify corneal
nerve morphology. Cornea. 2013;32(5):83–9.
121. Efron N, Edwards K, Roper N, Pritchard N,
Sampson GP, Shahidi AM, et  al. Repeatability of
measuring corneal subbasal nerve fiber length in
individuals with type 2 diabetes. Eye Contact Lens.
2010;36(5):245–8.
122. Hertz P, Bril V, Orszag A, Ahmed A, Ng E, Nwe P,
et  al. Reproducibility of in  vivo corneal confocal
microscopy as a novel screening test for early dia-
betic sensorimotor polyneuropathy. Diabet Med J Br
Diabet Assoc. 2011;28(10):1253–60.
L. Davalos et al.
123. Scarr D, Lovblom LE, Ostrovski I, Kelly D, Wu T,
Farooqi MA, et  al. Agreement between automated
and manual quantification of corneal nerve fiber
length: implications for diabetic neuropathy research.
J Diabetes Complicat. 2017;31(6):1066–73.
124. Perkins BA, Lovblom LE, Bril V, Scarr D, Ostrovski
I, Orszag A, et al. Corneal confocal microscopy for
identification of diabetic sensorimotor polyneu-
ropathy: a pooled multinational consortium study.
Diabetologia. 2018;61(8):1856–61.
125. Ahmed A, Bril V, Orszag A, Paulson J, Yeung E,
Ngo M, et  al. Detection of diabetic sensorimotor
polyneuropathy by corneal confocal microscopy in
type 1 diabetes: a concurrent validity study. Diabetes
Care. 2012;35(4):821–8.
126. Tavakoli M, Quattrini C, Abbott C, Kallinikos P,
Marshall A, Finnigan J, et  al. Corneal confocal
microscopy: a novel noninvasive test to diagnose and
stratify the severity of human diabetic neuropathy.
Diabetes Care. 2010;33(8):1792–7.
127. Keller JM, Noben-Trauth K. Genome-wide linkage
analyses identify Hfhl1 and Hfhl3 with frequency-­
specific effects on the hearing spectrum of NIH
Swiss mice. BMC Genet. 2012;13:32.
128. Jiang M-S, Yuan Y, Gu Z-X, Zhuang S-L.  Corneal
confocal microscopy for assessment of diabetic
peripheral neuropathy: a meta-analysis. Br J
Ophthalmol. 2016 Jan;100(1):9–14.
129. Chen X, Graham J, Dabbah MA, Petropoulos IN,
Ponirakis G, Asghar O, et al. Small nerve fiber quan-
tification in the diagnosis of diabetic sensorimo-
tor polyneuropathy: comparing corneal confocal
microscopy with intraepidermal nerve fiber density.
Diabetes Care. 2015;38(6):1138–44.
130. Stem MS, Hussain M, Lentz SI, Raval N, Gardner
TW, Pop-Busui R, et  al. Differential reduction in
corneal nerve fiber length in patients with type 1
or type 2 diabetes mellitus. J Diabetes Complicat.
2014;28(5):658–61.
131. Andersen ST, Grosen K, Tankisi H, Charles M,
Andersen NT, Andersen H, et  al. Corneal confocal
microscopy as a tool for detecting diabetic polyneu-
ropathy in a cohort with screen-detected type 2 dia-
betes: ADDITION-Denmark. J Diabetes Complicat.
2018;32(12):1153–9.
132. Ziegler D, Papanas N, Zhivov A, Allgeier S, Winter
K, Ziegler I, et al. Early detection of nerve fiber loss
by corneal confocal microscopy and skin biopsy
in recently diagnosed type 2 diabetes. Diabetes.
2014;63(7):2454–63.
133. Smith AG, Kowalsky G, Hauer P, Aperghis A,
Singleton J. The diagnostic performance and clinical
relevance of corneal confocal microscopy (CCM)
in patients with diabetic peripheral neuropathy
(S44.001). Neurology. 2016;86:16. Available from
https://n.neurology.org/content/86/16_Supplement/
S44.001
 Sensory Profiles and Diabetic
Neuropathy
Juliane Sachau, Manon Sendel, and Ralf Baron
1 Introduction
Up to 50% of patients with diabetic neuropathy
suffer from neuropathic pain [1], i.e., pain that
arises after a lesion or disease of the somatosen-
sory nervous system. Neuropathic pain can pro-
voke sleep disturbances, limit physical
functioning, lead to psychosocial distress and
thus has a major impact on the quality of life of
affected patients. Pain management is therefore
an important part in the treatment of diabetic neu-
ropathy. Several guidelines have been published
for the treatment of neuropathic pain in general
[2, 3] and painful diabetic neuropathy in particu-
lar [4]. However, even with first-line therapeu-
tics, many affected patients do not receive
sufficient pain relief of their pain, which is
reflected by the high number needed to treat
(NNT) ranging from 3.6 for tricyclic antidepres-
sants to 7.7 for pregabalin [3]. In painful diabetic
neuropathy, it is estimated that approximately
39% of the patients have never even received a
treatment for their pain [5]. In addition, many
clinical trials on new analgesic drugs have failed
in the past. Therefore, an important question that
has become the focus of pain researchers in
recent years is: which patients benefit most from
J. Sachau · M. Sendel · R. Baron (*)
Division of Neurological Pain Research and Therapy,
Department of Neurology, University Hospital
Schleswig-Holstein, Campus Kiel, Kiel, Germany
e-mail: r.baron@neurologie.uni-kiel.de
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_7
a particular pain therapy and how can these
responders be identified a priori?
The correct classification of neuropathic pain
is a prerequisite to answer this question. The fact
that different pathomechanisms can be present
within one pain etiology, while the very same
pathomechanisms can occur across different eti-
ologies, has led to a redefinition of classification
[6]. Contrary to the original etiology-based clas-
sification, neuropathic pain should rather be clas-
sified primarily on the basis of the underlying
pathophysiologic mechanisms. The awareness of
these mechanisms that might underlie a patient's
pain allows the use of drugs that act specifically
on the target structures. This is the concept of an
individualized mechanism-based pain therapy.
Since it is not possible yet to study these
pathomechanisms directly in humans, other
methods that can provide indirect information
must be used. The phenotype of a patient, i.e., the
clinical symptoms and signs, are thought to
reflect certain underlying mechanisms. Heat
hyperalgesia, for example, is related to peripheral
sensitization [7], whereas pinprick hyperalgesia
and dynamic mechanical allodynia can be attrib-
uted to central sensitization [8]. Based on this
idea, efforts have been made to precisely exam-
ine a patient’s phenotype and develop stratifica-
tion approaches in order to predict treatment
responses.
113
114 J. Sachau et al.

2 The DFNS Protocol
of Sensory Profiling
2.1 The DFNS Protocol
of Quantitative Sensory
Testing
One advance in the effort to adequately define
somatosensory phenotypes was the standardiza-
tion of Quantitative sensory testing (QST). The
German Research Network on Neuropathic Pain
(DFNS) developed a standardized test battery
assessing small and large nerve fiber function as
well as parameters of central processing (Table 1).
By using specific techniques to selectively block
nerve fiber classes, adequate tools have been
identified which asses the entire variety of differ-
ent sensory nerve fiber classes (C-, A-delta,
A-beta-fibers) [9–11].
All assessments are performed in a fixed
sequence with standardized patient instructions.
When performed according to protocol, the
DFNS QST protocol shows adequate test–retest
and inter-observer reliability [12] and high homo-
geneity across different centers [13].
The DFNS protocol consists of seven tests
including 13 thermal and mechanical parameters
[14, 15]. The protocol should be performed in
two areas for each patient, i.e., in the most
affected area and in a normal/unaffected one, in
order to detect deviations from control data and
side-to-side differences [16]. The protocol is dis-
played in Fig. 1.

Table 1  Quantitative sensory testing (QST) parameters according to the DFNS protocol
Suspected location of
underlying pathophysiological
mechanism QST finding
QST parameter Device Peripheral Central Loss of function Gain of function
CDT Cold detection Thermotest Aδ (cold) Spinothalamic Thermal Thermal
WDT threshold C (warm) hypoesthesia hyperesthesia
Warm detection
threshold
CPT Cold pain threshold Thermotest Aδ, C Spinothalamic Thermal Thermal
HPT Heat pain thresholda hypoalgesia hyperalgesia
PHS Paradoxial heat Thermotest Central Heat sensation
sensation disinhibition to a cold
stimulus
MDT Mechanical Calibrated Aβ Lemniscal Mechanical Mechanical
detection threshold von Frey hypoesthesia hyperesthesia
hairs
VDT Vibration detection Tuning fork Aβ Lemniscal Pallhypoesthesia
threshold
MPT Mechanical pain Pinprick Aδ, C Spinothalamic Pinprick Pinprick
thresholdb stimuli hypoalgesia hyperalgesia
MPS (Suprathreshold) Pinprick Aδ, C Spinothalamic Pinprick Pinprick
mechanical pain stimuli hypoalgesia hyperalgesia
sensitivityb
DMA Dynamic Cotton wisp, Aβ Lemniscal Allodynia
mechanical cotton wool,
allodyniab brush
WUR Wind-up ratio Pinprick Spinothalamic Temporal
stimuli summation
PPT Pressure pain Pressure Aδ, C Spinothalamic Pressure pain Pressure pain
threshold gauge device hypoalgesia hyperalgesia
 Peripheral sensitization
a

 Central sensitization
b
 Sensory Profiles and Diabetic Neuropathy 115

a b c
CDT WDT TSL & PHS CPT HPT MDT MPT

intensity steps by a factor of 2

y
ramp for all y y
thermal tests y y y y y y
(1 °C/s) y y
n n
n n n n n n
hot? hot? hot? n n
n
paradoxical
warming heat sensations sensation of touch slightly pricking
cooling during cold? no sensation of touch no pricking

~8 minutes ~4 minutes ~4 minutes

d e f g
MPS ALL WUR VDT PPT
100 100 100
Numerical rating 4
Numerical rating

a b Tuning fork Algometer

10 10 10 3 = b (series) (64 Hz) contact area

Ratio
a (single) 1cm2
1 1 1 2

threshold (kPa)
800 ramp (50kPa/s)

Pressure pain
0.1 0.1 0.1 1
600

Vibratory units (/8)

8
8 16 32 64 128 256 512 CW QT BR trains 1 to 5 6 400
1 2 3 4 5
4 200 1st painful
vibratory pressure
single series of 10 stimuli 2
hyperalgesia allodynia to light touch threshold,
normal sensation to pinprick normal sensation 0 tuning fork
hypoalgesia

~8 minutes ~2 minutes ~2 minutes ~2 minutes

Fig. 1  QST-a battery of sensory tests: figure of methods Stimulus-response functions: mechanical pain sensitivity
(reused from Rolke et  al. [15]). The standardized QST (MPS) for pinprick stimuli, and dynamic mechanical allo-
protocol assesses 13 parameters in seven test procedures dynia (ALL) assess A-delta mediated sensitivity to sharp
(a–g). All procedures are presented including a time frame stimuli (pinprick), and also A-beta fiber mediated pain
for testing over one area. (a) Thermal testing comprises sensitivity to stroking light touch (CW cotton wisp, QT
detection and pain thresholds for cold, warm, or hot stim- cotton wool tip, BR brush). (e) Wind-up ratio (WUR)
uli (C- and A-delta fiber mediated): cold detection thresh- compares the numerical ratings within five trains of a
old (CDT); warm detection threshold (WDT); number of single pinprick stimulus (a) with a series (b) of 10 repeti-
paradoxical heat sensations (PHS) during the thermal sen- tive pinprick stimuli to calculate WUR as the ratio: b/a. (f)
sory limen procedure (TSL) for alternating warm and cold Vibration detection threshold (VDT) tests for A-beta fiber
stimuli; cold pain threshold (CPT); heat pain threshold function using a Rydel-Seiffer 64  Hz tuning fork. (g)
(HPT). (b) Mechanical detection threshold (MDT) tests Pressure pain threshold (PPT) is the only test for deep
for A-beta fiber function using von Frey-filaments. (c) pain sensitivity, most probably mediated by muscle C-
Mechanical pain threshold (MPT) tests for A-delta fiber and A-delta fibers. For details regarding the testing proce-
mediated hyper- or hypoalgesia to pinprick stimuli. (d) dures also see Sect. 2

2.1.1 Small Fiber Function s­ensations (PHS) is captured. Thermal thresh-

(C-/A-­delta): Thermal Thresholds olds allow assessing small fiber function or their
Thermal thresholds are assessed using a ther- corresponding central pathways. PHS, however,
mode of a defined size with a baseline tempera- can also occur in case of disturbed central
ture of 32 °C.  Temperature ramps are used to processing.
determine thermal detection (cold detection Heat hyperalgesia is predominantly caused by
threshold, CDT; warm detection threshold, peripheral sensitization [7]. The heat sensitive
WDT) as well as pain thresholds (cold pain ion channel TRPV1 can be sensitized through
threshold, CPT; heat pain threshold, HPT) different mechanisms, e.g., inflammatory media-
using the method of levels. Furthermore, the tors, which causes the heat pain threshold to drop
ability to discriminate temperatures (thermal [17]. The same is true for TRPA1, whose sensiti-
sensory limen, TSL) including paradoxical heat zation causes cold hyperalgesia [18].
116
2.1.2 A-Beta-Fiber-Function:
Mechanical Detection
Thresholds
The mechanical detection threshold (MDT) for
touch is determined using modified von Frey
hairs (0.25 to 512 mN) [19]. The patient’s detec-
tion threshold is identified using the method of
limits, applying the stimuli in ascending and
descending order.
The mechanical detection threshold for vibra-
tion (vibration detection threshold, VDT) is a dis-
appearance threshold. A standard Rydel-Seiffer
tuning fork (64 Hz, 8/8 scale) is administered at a
bone protrusion three times.
2.1.3 A-Delta Fiber Function
and Central Sensitization:
Pinpricks and Pressure Pain
Threshold, Dynamic Mechanic
Allodynia
Weighted pinpricks (8–512 mN) are used to
determine the mechanical pain threshold (MPT),
mechanical pain sensitivity (MPS), and the wind-
­up ratio (WUR).
Pinprick hyperalgesia is evaluated by both the
MPT and MPS. The method of limits is used for
the assessment of MPT.  The patient is asked to
rate each pinprick as “dull” or “sharp.” Pinpricks
are administered in ascending and descending
order of sharpness.
MPS is based on a stimulus-response func-
tion. Subjects have to rate the pain intensity of
each stimulus on a numeric rating scale (NRS, 0
meaning no pain, 100 meaning the worst imagin-
Z − score = ( value of the single patient − mean of controls ) / standard deviation of controls.
Therefore, all QST parameters can be depicted
as standard normal distributions with a mean of
zero. PHS and DMA, that are absent under nor-
mal conditions, are given with original values.
For the majority parameters, which are not
normally distributed (CDT, WDT, TSL, MDT,
MPT, MPS, DMA, WUR, PPT), a logarithmic
transformation is performed beforehand. Z-scores
J. Sachau et al.
able pain). In total, 35 pinprick stimuli are applied
to each testing site.
Pinprick stimuli are alternated with normally
innocuous tactile stimuli (cotton wisp, Q-tip, soft
brush) to test for dynamic mechanic allodynia
(DMA).
Additionally, subjects are asked to rate a sin-
gle pinprick stimulus (256 mN or 128 mN when
testing in the face) as well as ten consecutive
stimuli applied as one per second in a 1 cm2 area.
To determine the WUR the mean rating of five
series of stimuli is divided by the mean rating of
five single stimuli. The WUR correlates to an
increase in excitability of spinal neurons, and is
therefore an indication of the status of spinal pro-
cessing [20].
For determining the pressure pain threshold
(PPT), a pressure algometer is pressed to a mus-
cle with a force up to 20 kg/cm2 and a ramp of 0.5
kg/cm2 per second. PPT is the only QST param-
eter used to assess deep somatic sensitivity.
Abnormal MPT, WUR, MPS, and PPT can
indicate A-delta fiber damage (or their corre-
sponding central pathways). In addition, mechan-
ical hyperalgesia and mechanical allodynia are
surrogates of central sensitization [17].
2.1.4 Z-transformation
and Reference Values
A single patient’s QST values can be compared
with reference values regardless of the different
kinds of measurement. This is possible due to
Z-transformation of the values (except for PHS
and DMA, see below) [21]:
above zero indicate a gain of function, whereas
Z-scores below zero indicate a loss. For this
­purpose, the algebraic sign is changed accord-
ingly in most of the parameters so all values indi-
cating loss of function show a Z-score below zero
and vice versa [15]. This makes it possible to eas-
ily grasp loss and gain of sensory function at a
glance.
Sensory Profiles and Diabetic Neuropathy
As QST values differ significantly across body
regions, between men and women and between
younger and older people, each body area
requires different gender and age specific refer-
ence values [14, 22]. Therefore, reference values
were established for different age groups, sepa-
rately for men and women as well as for specific
body areas including the dorsum of the foot and
hand [14], the face and the trunk [23]. Reference
data are given as means and 95% confidence
intervals (mean ± 1.96 SD), thus values above
+1.96 indicate an abnormal gain and values
below –1.96 an abnormal loss of function.
2.1.5 Advantages and Disadvantages
of QST
There are several disadvantages that should be
kept in mind when performing and evaluating
QST according to the DFNS protocol. First of
all, and perhaps most importantly, QST is a psy-
chophysical measure and therefore requires
patient compliance. Any aspect that hinders
patients’ understanding of the instructions like
cognitive deficits or a language barrier therefore
must be excluded prior to testing. In addition, it
is important to perform QST according to the
protocol using the pre-defined instructions to
guarantee reliable results. As mentioned above,
reference values are required for each area tested
to classify results as normal or abnormal. When
there are no available reference data, comparison
to the contralateral site should be used to judge
sensory function of the tested area [14]. All data
in the DFNS data base have been obtained from
Caucasian subjects. Thus a higher variability of
values can be expected in a cohort of healthy
subjects of different ethnicity [24]. To avoid tis-
sue damage, the range of available values is lim-
ited. For some of the parameters the range of
normal values overlaps with these cut-off values
so that the definition of abnormal values is not
possible. This, to some extent, limits the assess-
ment of sensory loss, especially in older patients
[15]. In addition, QST cannot determine the
exact location of the lesion or disease, thus a dif-
ferentiation between a damage of peripheral
nerve fibers or their correlating central pathways
is not possible.
117
Despite these limitations, QST has some
important advantages. QST enables to determine
a comprehensive somatosensory profile in a sin-
gle test session. This includes the function of
both large and small fibers, which cannot be
assessed using conventional methods of electro-
physiology. Unlike other methods used for the
assessment of small fiber function, QST cannot
only detect loss of function, but is also highly
sensitive to positive signs, i.e. gain of function
[15]. Thus, QST can be used as a potential bio-
marker for the diagnosis of (small fiber) neuropa-
thies, disease progression and prediction of
treatment response [25].
2.2 Sensory Profiling According
to the DFNS QST Protocol
2.2.1 The DFNS QST Protocol
Theoretically, the detection of 26 different abnor-
malities is possible considering the entirety of 13
parameters of the DFNS QST protocol. As DMA
and PHS are absent in healthy subjects, an abnor-
mal loss of function is not possible. Furthermore,
it has to be noted that the upper confidence limits
of the thermal pain thresholds (CPT, HPT) as
well as the lower confidence thresholds of MDT
and VDT are close to the limits of possible attain-
able values. This limits the detection of thermal
hypoalgesia and mechanical hyperesthesia. As
QST values significantly correlate across body
sites in healthy individuals, comparing absolute
QST values to the contralateral site may provide
insight into abnormalities that otherwise could be
missed [14].
All things considered, 20 different phenomena
of hypo- or hyperesthesia can be detected.
The simplest approach of profiling patients is
the description of existence or absence of one
distinct symptom, e.g., allodynia. Considering all
parameters of the DFNS QST protocol, there are
many different profiles that can theoretically
occur, although it is not always appropriate or
necessary to consider all of them [26].
The two most commonly described sensory
profiles are the so-called irritable nociceptor (IN)
phenotype, characterized by hypersensitivity, and
118
the “deafferentation“ or “non-irritable nocicep-
tor” (NIN) phenotype characterized by loss of
function. Those phenotypes will be described
further below.
2.3 Sensory Profiling in Diabetic
Neuropathy
Single parameters of QST, especially thermal
parameters and vibration, have been used in stud-
ies regarding diabetic neuropathy for quite some
time.
The DFNS protocol, however, provides an
opportunity to compile a complete sensory pro-
file in a structured way.
In 2012 Blankenburg et al. used QST to screen
for diabetic neuropathy in children with type 1
diabetes and found subclinical large and small
fiber neuropathies in almost 50% of them [27].
Many studies using QST to stratify patients
especially put an emphasis on finding differences
between patients with painful and painless dia-
betic neuropathy [28]. In two studies patients
were divided into three groups: patients without
neuropathic pain, with mild neuropathic pain,
and those with moderate to severe neuropathic
pain.
Abnormal QST-values, most commonly rep-
resenting loss of function, can be found in most
patients with diabetic neuropathy, although they
are more common when patients report pain [29,
30]. Both large and small fiber functions are
affected [27, 30].
When looking at thermal thresholds, generally
lower Z-values can be found in patients with dia-
betic neuropathy than in healthy controls [30],
regardless of whether patients suffer from pain or
not [29].
Significant gain was generally rare. An excep-
tion was heat hyperalgesia in the study assessing
children with type 1 diabetes [27]. PHS is found
in all groups of patients with diabetic neuropathy.
The lowest Z-scores are present in patients with
moderate to severe pain [30, 31]. Regarding
MDT there is also a high frequency of loss, which
is again most pronounced in patients with painful
neuropathy. In all studies considered VDT, vibra-
J. Sachau et al.
tion sensation was diminished regardless of pain.
Although children with type 1 diabetes did show
a higher MPS, this parameter generally seems to
play a smaller role in diabetic neuropathy [22].
Only a small percentage of patients present a loss
of function. This, again, is more pronounced in
patients with neuropathic pain [30, 31]. DMA is
rare in patients with diabetic neuropathy, it was
only found in a small percentage of patients with
neuropathic pain (17% of patients with mild neu-
ropathic pain, 14% of patients with moderate to
severe neuropathic pain [30]; 7.6% of all patients
with neuropathic pain [31]).
Accordingly, patients with diabetic neuropa-
thy seem to be more likely to fit the deafferenta-
tion profile (e.g., 53.8% of patients with
neuropathic pain [31]). The severity of loss cor-
relates with the presence and severity of neuro-
pathic pain, neuropathy in general and disability
[31]. When differentiating between polyneu-
ropathy patients with and without pain, those
with pain tend to show more significant loss of
function, while the significance correlates with
the severity of neuropathy in general [31]. Gain
of function seems to be much rarer. Only a
small percentage (4.9–7.1% of patients with
mild neuropathic pain, 7.1–7.5% of patients
with moderate to severe neuropathic pain [30];
14.6% in total [31]) of patients display the
characteristics of the irritable nociceptor (IN)
profile. Patients with the IN profile show lower
neuropathy disability scores and a shorter dura-
tion of pain.
DMA and PHS as symptoms of aberrant cen-
tral processing seem to occur more often in
patients with more severe pain [31].
3 Patient Stratification
for Mechanism-Based
Treatment
3.1 Clustering Approach
In 2016, the DFNS has developed a clustering
approach to stratify neuropathic pain patients
into groups of similar clinical signs based on
their QST sensory profile [32]. In total, 902
 Sensory Profiles and Diabetic Neuropathy 119

patients with peripheral neuropathic pain of dif-
ferent etiology were included in this study, more
than half of them (57%) suffered from painful
polyneuropathy. Three clusters, each of them
characterized by similar QST profiles, i.e. pre-
sumably similar pathophysiological mechanisms,
were identified: cluster 1 (sensory loss), cluster 2
(thermal hyperalgesia), and cluster 3 (mechanical
hyperalgesia) (Fig.  2, Table  2). Interestingly,
these three clusters could also be identified in
human surrogate models of neuropathic pain,
supporting the mechanistical relevance of these
sensory phenotypes [33].
Cluster 1 was characterized by a loss of small
and large nerve fiber function, i.e., hypoesthesia
to thermal and mechanical stimuli, as well as par-
adoxical heat sensation. At the mechanism level,
this cluster resembles the deafferentation caused
by a compression nerve block. Cluster 2 showed
intact nerve fiber function combined with thermal
hyperalgesia to heat and cold. This cluster was
found to reflect the human surrogate model of
peripheral sensitization induced by enhanced acti-
vation of TRPV1 receptors. Finally, cluster 3 was
characterized by a loss of small nerve fiber func-
tion in combination with hyperalgesia to blunt

Sensory loss
Thermal hyperalgesia 3
3 10
Mechanical hyperalgesia
n=902
gain of function

Pain rating (NRS)

PHS reports
2 2

1 1
.3
Z-score

0 0 0
DMA PHS

–1
loss of function

–2

–3

CDT WDT TSL CPT HPT PPT MPT MPS WUR MDT VDT

QST parameter

Fig. 2  QST profiles of the 3-cluster solution (adapted
from Baron et al. [32]). Sensory profiles of the three clus-
ters presented as mean Z-scores ± 95% confidence inter-
val (n = 902). Note that Z-transformation eliminates
differences due to test site, sex, and age. Positive Z-scores
indicate positive sensory signs (hyperalgesia), whereas
negative Z-values indicate negative sensory signs (hypo-
esthesia and hypoalgesia). Dashed lines: 95% confidence
interval for healthy subjects (−1.96 < z < +1.96). Note that
if the mean of a cluster is within the shaded area, this does
not imply that it does not differ from a healthy cohort.
Values are significantly different from those of healthy
subjects, if their 95% confidence interval does not cross
the zero line. Insets show numeric pain ratings for
dynamic mechanical allodynia (DMA) on a logarithmic
scale (0–100) and frequency of paradoxical heat sensation
(PHS) (0–3). Blue symbols: cluster 1 “sensory loss”
(42%). Red symbols: cluster 2 “thermal hyperalgesia”
(33%). Yellow symbols: cluster 3 “mechanical hyperalge-
sia” (24%). CDT cold detection threshold, CPT cold pain
threshold, HPT heat pain threshold, MDT mechanical
detection threshold, MPS mechanical pain sensitivity,
MPT mechanical pain threshold, NRS numerical rating
scale, PPT pressure pain threshold, QST quantitative sen-
sory testing, TSL thermal sensory limen, VDT vibration
detection threshold, WDT warm detection threshold,
WUR wind-up ratio
120
Table 2  Characteristics of the three QST clusters
Cluster QST findings
Sensory loss Touch, vibration,
temperature, pain
Paradoxical heat
sensation
Thermal None
hyperalgesia
Heat, cold hyperalgesia
Mechanical In particular warm and
hyperalgesia cold hypoesthesia
Pressure, pinprick
hyperalgesia dynamic
mechanical allodynia
loss of function, gain of function, NSAID non-steroidal anti-inflammatory drug, NMDA N-methyl-D-aspartate
pressure and pinprick as well as dynamic mechan-
ical allodynia, patterns that were also found in
surrogate models of central sensitization.
These three clusters were detected in each of
the included disease etiologies, i.e., polyneuropa-
thy, radiculopathy, peripheral nerve injury, and
postherpetic neuralgia, though with different fre-
quencies. In a second step, an algorithm with
both a deterministic and a probabilistic version
was developed to allocate individual patients to
the three pre-defined clusters [34]. Using the
deterministic approach, each patient is assigned
to exactly one cluster, whereas the probabilistic
approach allows the allocation of one patient to
more than one cluster. By doing so, patients with
painful diabetic polyneuropathy were mostly
assigned to the sensory loss cluster (64% for the
deterministic approach; 82% for the probabilistic
approach), followed by the mechanical hyperal-
gesia cluster (19%; 75%) and least frequently the
thermal hyperalgesia cluster (13%; 33%). This
observation is in agreement with the results of
Themistocleous et al. [30]. Finally, Vollert et al.
proposed recommendations on the sample size,
i.e., the number of patients wo need to be screened
to find a subgroup that is large enough to conduct
a clinical trial (cross-over or parallel design) with
a power of 80%, an alpha-level of 0.05 and an
effect size of 0.3, 0.5, or 0.7 [34]. A detailed table
J. Sachau et al.
Postulated
pathomechanism Therapeutical implications
Loss of small and ++ Antidepressants, opioids
large fiber function
+ Gabapentinoids, sodium channel
blocker
Peripheral ++ Sodium channel blocker
sensitization
+ Antidepressants, gabapentinoids,
capsaicin patches, opioids, botulinum
toxin (NSAID)
Loss of small fiber ++ Gabapentinoids, sodium channel
function blocker
Central sensitization + Antidepressants, opioids, NMDA
antagonists
with the required number of patients for each
condition can be found in the paper. For example,
in order to detect an adequate subgroup of
patients allocated to the sensory loss cluster in a
cross-over study on painful diabetic neuropathy
with an estimated effect size of 0.5, about 41
patients need to be screened. In contrast, the
detection of a mechanical hyperalgesia subgroup
requires more than twice as many patients (n =
103) in the same setting.
The identification of specific patient clusters
is a first step towards a mechanism-based ther-
apy. The Committee for Medicinal Products for
Human Use (CHMP) of the European Medicines
Agency (EMA) has therefore acknowledged the
QST-based sensory subgrouping an adequate
stratification tool for identification of specific
sensory phenotypes of patients in clinical trials
on neuropathic pain. To overcome the well-­
known limitations of QST, i.e., the time-­
consuming protocol and expensive equipment,
several research groups have concentrated on the
development of easy-to-use bedside testing pro-
cedure that can be easily applied in both large
clinical trials and clinical practice [35, 36].
Recently, it was shown that a combination of
only five simple bedside parameters is necessary
for the correct allocation to the three pre-defined
QST clusters [37].
Sensory Profiles and Diabetic Neuropathy 121

3.2 Sensory Testing as Predictor another study, this proportion was even higher
of Treatment Response suggesting that about 50% of all patients with
painful diabetic neuropathy are characterized by
In the last years, several studies have been pub- intact sensory function and would therefore ben-
lished that support the phenotype-based stratifi- efit from a therapy with oxcarbazepine [44].
cation as a possible approach to identify However, other results imply a much smaller
responders in clinical trials on neuropathic pain proportion of this subgroup [30], underscoring
in general, and diabetic neuropathy in particular. that more stratification studies are needed to
Demant et al. performed a placebo-controlled, define the exact phenotype distribution in
cross-over trial to investigate the efficacy of patients with diabetic neuropathy. Nevertheless,
oxcarbazepine, an unselective sodium channel this study nicely shows that the idea of a QST-
blocker, in patients with peripheral neuropathic based subgrouping can improve the treatment of
pain of different etiology [38]. As one of the first peripheral neuropathic pain in general. But there
studies with an a priori stratification, patients are also promising studies for diabetic neuropa-
were prospectively subgrouped according to their thy supporting the phenotype-based stratifica-
QST profile into a NIN (n = 52) and a IN type (n tion as predictive biomarker for treatment
= 31) based on a previously published LoGA response.
classification [16]. Even though this classifica- TRPA1 has been shown to induce cold and
tion does not correspond completely to the three-­ also mechanical hyperalgesia in animal models
cluster sorting algorithm of Baron et  al. [32], of (diabetic) neuropathic pain [45]. In a double-
both approaches share some similarities. The IN blind, placebo-controlled Phase 2a study on pain-
type is characterized by an intact sensory func- ful diabetic neuropathy, the selective TRPA1
tion for cold and warm combined with thermal antagonist GRC 17536 was shown to be effective
and/or mechanical hyperalgesia and hence corre- in a subgroup of patients with moderate to severe
sponds to the thermal hyperalgesia cluster 2, pain and preserved sensory function (similar to
while the NIN is characterized by profound the irritable nociceptor type) as detected via ret-
­sensory loss, although the area can be painful rospective QST-based stratification [44].
nevertheless [39]. Another option to stratify patients based on
The upregulation or increased expression and their phenotype is the assessment of conditioned
activation of TRPV1 [40], voltage-gated sodium pain modulation (CPM). This “pain inhibits pain
channels [41, 42] and voltage-gated calcium phenomenon” describes the reduction in pain
channels [43] are proposed mechanisms respon- intensity of a first painful stimulus (test stimulus)
sible for irritable nociceptors [41, 42]. by a second painful stimulus (conditioning stim-
Consequently, drugs like oxcarbazepine acting as ulus). Based on the hypothesis that certain drugs
sodium channel blockers ought to be more effec- that enhance descending inhibitory pathways
tive in patients with an increased expression and would be more effective in patients with
activation of sodium channels. insufficient CPM effect, Yarnitsky et  al. con-
­
This theoretical assumption was confirmed in ducted an exploratory study in patients with pain-
the study by Demant et al. [38]. While the NNT ful diabetic neuropathy [46]. In total, 30 patients
for a 50% pain reduction was as high as 13 in the received placebo followed by duloxetine, a com-
NIN type, it decreased to 3.9 in the IN subgroup. bined serotonin/norepinephrine reuptake inhibi-
Therefore, oxcarbazepine may be considered as tor. Interestingly, a less efficient CPM effect
therapeutic option in a certain subgroup of before the treatment period was associated with
patients. With respect to pain in diabetic neurop- higher drug efficacy. Therefore, the authors con-
athy, this seems to be the case in about one third cluded that phenotype-based stratification using a
of all patients when applying the probabilistic simple CPM protocol may be a predictor for
approach of the DFNS sorting algorithm. In treatment response to duloxetine.
122
3.3 Sensory Profiling Using
Patient Reported Outcome
Measures
As described above, QST does have its limits
[47]. While its results depict evoked pain very
well, spontaneous pain cannot be assessed. Apart
from that, QST results do not necessarily repre-
sent the symptoms patients actually feel.
Several studies have used questionnaires,
originally designed to screen for neuropathic
pain components [48], to stratify patients consid-
ering the pattern of sensory abnormalities
reported by themselves. The painDETECT ques-
tionnaire as well as the Neuropathic Pain
Symptom Inventory (NPSI) have been able to
successfully identify different subgroups [49].
The Standardized Evaluation of Pain (StEP), a
combination of interview questions and physical
examination, was used to identify eight different
patient clusters. One cluster characterized by
symptoms of loss was specific to diabetic neu-
ropathy [50].
In a large study using the painDETECT ques-
tionnaire, patients with diabetic neuropathy most
commonly exhibited a sensory profile character-
ized by prominent numbness. A considerable
percentage of patients also described suffering
from burning pain and paresthesia. Diabetic neu-
ropathy patients less commonly suffered from
pain attacks and burning pain in combination
with paresthesia and allodynia. Pressure hyper-
algesia in combination with allodynia was least
common [51].
A study using the NPSI [52] found quite simi-
lar results as paresthesia and burning were the pain
descriptors used most often by patients with dia-
betic neuropathy. Paroxysmal pain, evoked pain,
and deep pain were reported less commonly [53].
The pain PREDICT questionnaire [54] was
especially developed to predict treatment
response as well as assess different neuropathic
and nociceptive symptoms and categorize
patients into different subgroups. The commonly
used IN and NIN profiles as well as a third one
“pain attacks with nociceptive component” were
used within this approach as well. Contrary to the
other above-­ mentioned results, most patients
J. Sachau et al.
with diabetic neuropathy exhibited the IN profile
in this study. The profile indicating deafferenta-
tion pain was only second most common. There
are several possible explanations for this result: It
is possible that questionnaires are less suited to
assess negative symptoms like numbness, mak-
ing positive symptoms overrepresented.
Likewise, evoked pain seems to be reported by
more patients in comparison to the number of
patients showing DMA in QST [30, 31]. In addi-
tion, the patients included in this study overall
were in an earlier stage of disease and therefore
displayed the IR profile more often, similar to
what is the case in QST profiles in animal models
[55]. While symptoms and signs of gain seem to
be more common in early disease stages and in
case of small fiber damage, a sensory profile of
loss rather occurs with longer disease duration
[55, 56]. In animal models, however, there are
other causes as well that may explain the lack of
agreement in the QST profile between animal
and patient. Streptozotocin-induced diabetes,
which is the most commonly used model, does
not match the characteristics of diabetes mellitus
in humans exactly [57]. Furthermore, it can be
argued that the assessment of sensory loss in ani-
mals is not as practicable as it is in humans,
meaning that symptoms of loss may be underrep-
resented in animal studies [58].
Recently, a questionnaire-based stratification
approach has been used by Bouhassira and col-
leagues who used the NPSI to identify clusters in
a large cohort of 628 patients with probable or
definite neuropathic pain [59]. First, they identi-
fied three subgroups that were characterized by
different symptom patterns: the deep pain cluster
(higher scores for squeezing and pressure pain),
the evoked pain cluster (higher evoked pain
scores and low deep pain scores), and the pin-
pointed pain cluster (higher scores for tingling
and pins and needles plus moderate burning and
paroxysmal scores). Afterwards they developed
an algorithm to assign patients to one of these
clusters. Finally, they used this approach to per-
form a post-hoc analysis of two placebo-­
controlled studies in order to predict treatment
response to subcutaneous botulinum toxin A
injections. Interestingly, the evoked pain cluster
Sensory Profiles and Diabetic Neuropathy
as well as the deep pain cluster showed a signifi-
cant treatment response, whereas the pinpointed
pain cluster failed to show a significant effect.
While patients with diabetic neuropathy were a
minority in this study, a similar algorithm could
be applied in different cohorts in the future.
Overall, phenotypic-based stratification to
predict treatment response has become more and
more popular in recent years. The results of first
clinical trials in which a prospective subgrouping
was performed have shown promising results,
indicating that this is an appropriate approach to
realize the concept of an individualized
mechanism-­based therapy. In future, the use of
simple bedside tools combined with validated
patient reported outcome measures may help to
apply a priori sensory stratification also in large
clinical trials. This will hopefully result in
improved treatment strategies for neuropathic
pain in general and painful diabetic neuropathy in
particular.
Acknowledgement We would like to thank Dilara
Kersebaum for her excellent help in language editing.
References
1. Sloan G, et  al. A new look at painful diabetic neu-
ropathy. Diabetes Res Clin Pract. 2018;144:177–91.
https://doi.org/10.1016/j.diabres.2018.08.020.
2. Attal N, et  al. EFNS guidelines on the pharmaco-
logical treatment of neuropathic pain: 2010 revi-
sion. Eur J Neurol. 2010;17(9):1113. https://doi.
org/10.1111/j.1468-­1331.2010.02999.x.
3. Finnerup NB, et al. Pharmacotherapy for neuropathic
pain in adults: a systematic review and meta-­analysis.
Lancet Neurol. 2015;14(2):162–73. https://doi.
org/10.1016/S1474-­4422(14)70251-­0.
4. Bril V, et al. Evidence-based guideline: treatment of
painful diabetic neuropathy–report of the American
Association of Neuromuscular and Electrodiagnostic
Medicine, the American Academy of Neurology,
and the American Academy of Physical Medicine
& Rehabilitation. Muscle Nerve. 2011;43(6):910–7.
https://doi.org/10.1002/mus.22092.
5. Daousi C, MacFarlane IA, Woodward A, Nurmikko
TJ, Bundred PE, Benbow SJ.  Chronic painful
peripheral neuropathy in an urban community: a
controlled comparison of people with and without
diabetes. Diabet Med. 2004;21(9):976–82. https://doi.
org/10.1111/j.1464-­5491.2004.01271.x.
123
6. von Hehn CA, Baron R, Woolf CJ.  Deconstructing
the neuropathic pain phenotype to reveal neural
mechanisms. Neuron. 2012;73(4):638–52. https://doi.
org/10.1016/j.neuron.2012.02.008.
7. LaMotte RH, Thalhammer JG, Torebjörk HE,
Robinson CJ. Peripheral neural mechanisms of cuta-
neous hyperalgesia following mild injury by heat. J
Neurosci. 1982;2(6):765–81.
8. LaMotte RH, Shain CN, Simone DA, Tsai
EF.  Neurogenic hyperalgesia: psychophysical stud-
ies of underlying mechanisms. J Neurophysiol.
1991;66(1):190–211. https://doi.org/10.1152/
jn.1991.66.1.190.
9. Fruhstorfer H.  Thermal sensibility changes dur-
ing ischemic nerve block. Pain. 1984;20(4):355–61.
https://doi.org/10.1016/0304-­3959(84)90112-­X.
10. Yarnitsky D, Ochoa JL.  Differential effect of
compression-­ ischaemia block on warm sensation
and heat-induced pain. Brain. 1991;114(2):907–13.
https://doi.org/10.1093/brain/114.2.907.
11. Ziegler EA, Magerl W, Meyer RA, Treede
RD.  Secondary hyperalgesia to punctate mechani-
cal stimuli. Central sensitization to A-fibre nocicep-
tor input. Brain. 1999;122(12):2245–57. https://doi.
org/10.1093/brain/122.12.2245.
12. Geber C, et  al. Test-retest and interobserver reli-
ability of quantitative sensory testing according to
the protocol of the German Research Network on
Neuropathic Pain (DFNS): a multi-centre study.
Pain. 2011;152(3):548–56. https://doi.org/10.1016/j.
pain.2010.11.013.
13. Vollert J, et  al. Quantitative sensory testing using
DFNS protocol in Europe: an evaluation of het-
erogeneity across multiple centers in patients with
peripheral neuropathic pain and healthy subjects.
Pain. 2016;157(3):750–8. https://doi.org/10.1097/j.
pain.0000000000000433.
14. Rolke R, et  al. Quantitative sensory testing in the
German Research Network on Neuropathic Pain
(DFNS): standardized protocol and reference values.
Pain. 2006;123(3):231–43. https://doi.org/10.1016/j.
pain.2006.01.041.
15. Rolke R, et  al. Quantitative sensory testing: a
comprehensive protocol for clinical trials. Eur J
Pain. 2006;10(1):77–88. https://doi.org/10.1016/j.
ejpain.2005.02.003.
16. Maier C, et  al. Quantitative sensory testing in the
German Research Network on Neuropathic Pain
(DFNS): somatosensory abnormalities in 1236
patients with different neuropathic pain syndromes.
Pain. 2010;150(3):439–50. https://doi.org/10.1016/j.
pain.2010.05.002.
17. Treede R-D. Chapter 1 pain and hyperalgesia: defini-
tions and theories. Handb Clin Neurol. 2006;81:3–10.
https://doi.org/10.1016/S0072-­9752(06)80005-­9.
18. Obata K, et  al. TRPA1 induced in sensory neurons
contributes to cold hyperalgesia after inflammation
and nerve injury. J Clin Invest. 2005;115(9):2393–
401. https://doi.org/10.1172/JCI25437.
124
19. Fruhstorfer H, Gross W, Selbmann O. von Frey
hairs: new materials for a new design. Eur J
Pain. 2001;5(3):341–2. https://doi.org/10.1053/
eujp.2001.0250.
20. Herrero JF, Laird JM, López-García JA. Wind-up of
spinal cord neurones and pain sensation: much ado
about something? Prog Neurobiol. 2000;61(2):169–
203. https://doi.org/10.1016/s0301-­0082(99)00051-­9.
21. Glass GV, Stanley JX. Statistical methods in education
and psychology. Englewood Cliffs: Prentice-­Hall; 1970.
22. Blankenburg M, et  al. Reference values for quanti-
tative sensory testing in children and adolescents:
developmental and gender differences of somatosen-
sory perception. Pain. 2010;149(1):76–88. https://doi.
org/10.1016/j.pain.2010.01.011.
23. Pfau DB, et  al. Quantitative sensory testing in the
German Research Network on Neuropathic Pain
(DFNS): reference data for the trunk and applica-
tion in patients with chronic postherpetic neuralgia.
Pain. 2014;155(5):1002–15. https://doi.org/10.1016/j.
pain.2014.02.004.
24. Gierthmühlen J, et  al. Who is healthy? Aspects
to consider when including healthy volunteers in
QST--based studies-a consensus statement by the
EUROPAIN and NEUROPAIN consortia. Pain.
2015;156(11):2203–11. https://doi.org/10.1097/j.
pain.0000000000000227.
25. Smith SM, et al. The potential role of sensory testing,
skin biopsy, and functional brain imaging as biomark-
ers in chronic pain clinical trials: IMMPACT con-
siderations. J Pain. 2017;18(7):757–77. https://doi.
org/10.1016/j.jpain.2017.02.429.
26. Baron R, Förster M, Binder A. Subgrouping of patients
with neuropathic pain according to pain-related sen-
sory abnormalities: a first step to a stratified treatment
approach. Lancet Neurol. 2012;11(11):999–1005.
https://doi.org/10.1016/S1474-­4422(12)70189-­8.
27. Blankenburg M, et  al. Childhood diabetic neuropa-
thy: functional impairment and non-invasive screen-
ing assessment. Diabet Med. 2012;29(11):1425–32.
https://doi.org/10.1111/j.1464-­5491.2012.03685.x.
28. Shillo P, et  al. Painful and painless diabetic neu-
ropathies: what is the difference? Curr Diab Rep.
2019;19(6):32. https://doi.org/10.1007/s11892-­019-­
1150-­5.
29. Krämer HH, Rolke R, Bickel A, Birklein F. Thermal
thresholds predict painfulness of diabetic neuropa-
thies. Diabetes Care. 2004;27(10):2386–91. https://
doi.org/10.2337/diacare.27.10.2386.
30. Themistocleous AC, et  al. The pain in neuropa-
thy study (PiNS): a cross-sectional observational
study determining the somatosensory phenotype
of painful and painless diabetic neuropathy. Pain.
2016;157(5):1132–45. https://doi.org/10.1097/j.
pain.0000000000000491.
31. Raputova J, et  al. Sensory phenotype and risk fac-
tors for painful diabetic neuropathy: a cross-sectional
observational study. Pain. 2017;158(12):2340–53.
https://doi.org/10.1097/j.pain.0000000000001034.
32. Baron R, et  al. Peripheral neuropathic pain: a
mechanism-­ related organizing principle based on
J. Sachau et al.
sensory profiles. Pain. 2017;158(2):261–72. https://
doi.org/10.1097/j.pain.0000000000000753.
33. Vollert J, et  al. Pathophysiological mechanisms of
neuropathic pain: comparison of sensory phenotypes
in patients and human surrogate pain models. Pain.
2018;159(6):1090–102. https://doi.org/10.1097/j.
pain.0000000000001190.
34. Vollert J, et  al. Stratifying patients with peripheral
neuropathic pain based on sensory profiles: algo-
rithm and sample size recommendations. Pain.
2017;158(8):1446–55. https://doi.org/10.1097/j.
pain.0000000000000935.
35. Buliteanu A, et  al. Validation of a bedside quantita-
tive sensory testing (QST) protocol in chronic neu-
ropathic pain. J Pain. 2018;19(3):S52. https://doi.
org/10.1016/j.jpain.2017.12.123.
36. Zhu GC, et al. Concurrent validity of a low-cost and time-
efficient clinical sensory test battery to evaluate somato-
sensory dysfunction. Eur J Pain. 2019;23(10):1826–38.
https://doi.org/10.1002/ejp.1456.
37. Reimer M, et  al. Sensory bedside testing: a simple
stratification approach for sensory phenotyping.
Pain Rep. 2020;5(3):e820. https://doi.org/10.1097/
PR9.0000000000000820.
38. Demant DT, et  al. The effect of oxcarbazepine in
peripheral neuropathic pain depends on pain pheno-
type: a randomised, double-blind, placebo-controlled
phenotype-stratified study. Pain. 2014;155(11):2263–
73. https://doi.org/10.1016/j.pain.2014.08.014.
39. Fields HL, Rowbotham M, Baron R.  Postherpetic
neuralgia: irritable nociceptors and deafferenta-
tion. Neurobiol Dis. 1998;5(4):209–27. https://doi.
org/10.1006/nbdi.1998.0204.
40. Suh Y-G, Oh U.  Activation and activators of
TRPV1 and their pharmaceutical implication. Curr
Pharm Des. 2005;11(21):2687–98. https://doi.
org/10.2174/1381612054546789.
41. Devor M. Sodium channels and mechanisms of neu-
ropathic pain. J Pain. 2006;7(1):3–12. https://doi.
org/10.1016/j.jpain.2005.09.006.
42. Dib-Hajj SD, Cummins TR, Black JA, Waxman
SG.  Sodium channels in normal and pathological
pain. Annu Rev Neurosci. 2010;33:325–47. https://
doi.org/10.1146/annurev-­neuro-­060909-­153234.
43. Bourinet E, Altier C, Hildebrand ME, Trang T, Salter
MW, Zamponi GW. Calcium-permeable ion channels
in pain signaling. Physiol Rev. 2014;94(1):81–140.
https://doi.org/10.1152/physrev.00023.2013.
44. Tandon M et  al. A simple algorithm to identify
likely responders to GRC 17356 in patients of pain-
ful diabetic peripheral neuropathy using sensory
mapping
45. de Araujo DSM, Nassini R, Geppetti P, De Logu
F. TRPA1 as a therapeutic target for nociceptive pain.
Expert Opin Ther Targets. 2020;24(10):997–1008.
https://doi.org/10.1080/14728222.2020.1815191.
46. Yarnitsky D, Granot M, Nahman-Averbuch H,
Khamaisi M, Granovsky Y. Conditioned pain modu-
lation predicts duloxetine efficacy in painful diabetic
neuropathy. Pain. 2012;153(6):1193–8. https://doi.
org/10.1016/j.pain.2012.02.021.
Sensory Profiles and Diabetic Neuropathy
47. Sachau J, Baron R. Neuropathic pain therapy: a puzzle
of different approaches to stratify patients. Pain. 2020.
https://doi.org/10.1097/j.pain.0000000000002120
48. Freynhagen R, Baron R, Gockel U, Tölle TR.  Pain
detect: a new screening questionnaire to identify neu-
ropathic components in patients with back pain. Curr
Med Res Opin. 2006;22(10):1911–20. https://doi.org/
10.1185/030079906X132488.
49. Attal N, Fermanian C, Fermanian J, Lanteri-Minet
M, Alchaar H, Bouhassira D.  Neuropathic pain: are
there distinct subtypes depending on the aetiology or
anatomical lesion? Pain. 2008;138(2):343–53. https://
doi.org/10.1016/j.pain.2008.01.006.
50. Scholz J, et  al. A novel tool for the assessment
of pain: validation in low back pain. PLoS Med.
2009;6(4):e1000047. https://doi.org/10.1371/journal.
pmed.1000047.
51. Baron R, Tölle TR, Gockel U, Brosz M, Freynhagen
R.  A cross-sectional cohort survey in 2100 patients
with painful diabetic neuropathy and postherpetic
neuralgia: differences in demographic data and sen-
sory symptoms. Pain. 2009;146(1):34–40. https://doi.
org/10.1016/j.pain.2009.06.001.
52. Bouhassira D, et  al. Development and validation
of the neuropathic pain symptom inventory. Pain.
2004;108(3):248–57. https://doi.org/10.1016/j.pain.
2003.12.024.
53. Spallone V, Greco C. Painful and painless diabetic neu-
ropathy: one disease or two? Curr Diab Rep. 2013;13:7.
https://doi.org/10.1007/s11892-­013-­0387-­7.
125
54. Tölle TR, et  al. Pain predict: first interim data from
the development of a new patient-reported pain
questionnaire to predict treatment response using
sensory symptom profiles. Curr Med Res Opin.
2019;35(7):1177–85. https://doi.org/10.1080/030079
95.2018.1562687.
55. Rice ASC, Finnerup NB, Kemp HI, Currie GL,
Baron R.  Sensory profiling in animal models of
neuropathic pain: a call for back-translation. Pain.
2018;159(5):819–24. https://doi.org/10.1097/j.
pain.0000000000001138.
56. Sveen KA, et  al. Small- and large-fiber neuropathy
after 40 years of type 1 diabetes: associations with
glycemic control and advanced protein glycation:
the Oslo Study. Diabetes Care. 2013;36(11):3712–7.
https://doi.org/10.2337/dc13-­0788.
57. Goyal SN, et  al. Challenges and issues with
streptozotocin-­induced diabetes  - a clinically rel-
evant animal model to understand the diabetes
pathogenesis and evaluate therapeutics. Chem Biol
Interact. 2016;244:49–63. https://doi.org/10.1016/j.
cbi.2015.11.032.
58. Klinck MP, et  al. Translational pain assessment:
could natural animal models be the missing link?
Pain. 2017;158(9):1633–46. https://doi.org/10.1097/j.
pain.0000000000000978.
59. Bouhassira D, et  al. Stratification of patients based
on the neuropathic pain symptom inventory: develop-
ment and validation of a new algorithm. Pain. 2020.
https://doi.org/10.1097/j.pain.0000000000002130
 Neurotrophic Factors
in the Pathogenesis and Treatment
of Diabetic Neuropathy
Nigel A. Calcutt
1 Introduction
Neurotrophic factors may be broadly described
as polypeptides that signal through receptor tyro-
sine kinases to support survival, growth, and dif-
ferentiation of developing neurons and/or
maintain the phenotype of adult neurons. Nerve
growth factor (NGF), the prototypical neuro-
trophic factor, was initially studied as a critical
molecule for the survival, growth, and differenti-
ation of small sensory neurons in the peripheral
nervous system (PNS). While not required for
survival of adult sensory neurons, NGF regulates
both the phenotype and regenerative capacity of
adult peripheral sensory neurons. The initial
model developed from the study of NGF, ideal-
ized a neurotrophic factor polypeptide produced
and secreted by end organs such as muscle or
skin that was bound and internalized by receptors
at the axonal terminals and retrogradely trans-
ported to the cell body as a signaling endosome
to influence gene expression and neuronal pheno-
type [1, 2]. However, this model has been repeat-
edly decorated by the ever increasing appreciation
of the complexity and subtlety of cell:cell inter-
actions [3, 4]. Within the PNS, neurotrophic fac-
tors may act as autocrine regulators as well as
classical paracrine messengers, while Schwann
N. A. Calcutt (*)
Department of Pathology, University of California
San Diego, La Jolla, CA, USA
e-mail: ncalcutt@ucsd.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_8
cells, satellite cells, macrophages, and neurons
themselves secrete multiple neurotrophic and
other factors and express a veritable cornucopia
of receptors (Fig. 1).
Diabetic neuropathy is commonly described
as a slowly developing dying back axonopathy
that presents as a distal symmetrical polyneurop-
athy [5]. The impact of neurotrophic deficiency
on maintenance of neuronal phenotype, nerve
terminals, and axonal structure in adult rodents is
consistent with this pathology and invites consid-
eration of loss of neurotrophic support as a patho-
genic mechanism contributing to diabetic
neuropathy. There is also a school of thought that
a primary Schwannopathy co-exists with axo-
nopathy in diabetic neuropathy and thus may also
contribute to neurotrophic deficiency given the
growing appreciation that Schwann cells provide
continuous metabolic and trophic support to their
axons [6]. The neurotrophic axis, representing
neurotrophic factor synthesis, secretion, signal-
ing, axonal transport, and gene regulatory mech-
anisms, is complex and offers multiple sites for
lesions that induce a loss of neurotrophic support.
Delivery of exogenous factors to replace defi-
cient endogenous trophic support is an obvious
therapeutic approach, provided that any impair-
ment of receptor mediated uptake or downstream
components of the neurotrophic axis in target
neurons can be overcome. Providing excess neu-
rotrophic support may also allow neurons to
127
128
Fig. 1  Neurotrophic support in the PNS.  The original
model of end-organ derived factors (red arrows), such as
muscle-derived BDNF (upper) and skin-derived NGF
(lower), binding p75NTR/trk receptor complexes before
internalization and retrograde transport to the nucleus has
been augmented by appreciation that factors (orange
resist the multiple other metabolic insults directed
at them by diabetes or promote recovery from the
impact of such insults, irrespective of whether
impaired neurotrophic support is the primary
pathogenic lesion.
This review will cover the impact of diabetes
upon both the classic neurotrophic factor fami-
lies and also other factors that have been exten-
sively studied in the context of their disorder
potentially contributing to the pathogenesis of
diabetic neuropathy and/or their potential as
therapeutics to treat diabetic neuropathy. The
complexities of the impact of diabetes on the
contribution of neurotrophic factors to injury
responses such as nerve regeneration following
physical lesion, inflammation, and wound heal-
ing in skin have been recently reviewed in detail
elsewhere [7–9] and are mentioned only en
passant.
N. A. Calcutt
arrows) may also be supplied by cells of the spinal cord
(neurons, microglia, astrocytes, and oligodendrocytes)
and peripheral nerve (Schwann cells, satellite cells, and
resident macrophages), while neurons also export trophic
factors to their peripheral and central terminals
2 The Neurotrophin Family
All polypeptides in this family share about 50%
sequence homology and signal through receptors
that combine a common low affinity sub-unit
(p75NTR) and a high affinity ligand-specific tyro-
sine kinase sub-unit, the tropomyosin receptor
kinases A-C (Trk’s A-C). Neurotrophin precur-
sors (pro-neurotrophins) may also be secreted
and signal through p75NTR that associates with
other sub-units [10] while the p75NTR sub-unit
acting as a monomer transduces signals that pro-
mote neuronal apoptosis and growth cone col-
lapse [11]. The emergence of the roles of NGF
and other neurotrophins in neuronal development
and maintenance of the adult phenotype and
regeneration has been paralleled by interest in
whether disrupted neurotrophic support
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy 129

c­ ontributes to diabetic neuropathy and the viabil- ings encouraged clinical trials of NGF as a ther-
ity of neurotrophins as therapeutics. apy for diabetic neuropathy. Unfortunately, the
efficacy of injected recombinant human NGF
NGF  In the adult PNS, the TrkA receptor sub-­ suggested by phase II clinical studies was accom-
unit that confers NGF sensitivity is expressed by panied by dose limiting side effects, including
sympathetic and predominantly small diameter pain, which prompted changes in formulation
nociceptive sensory neurons [12] and cells adja- and dose that may have contributed to the failure
cent to nerve ending such as keratinocytes and of subsequent phase III study [36]. While this
muscle secrete NGF.  In contrast to embryonic failure largely discouraged pursuit of NGF injec-
neurons, NGF is not a required factor for survival tion therapy, it is currently allowed for use against
of adult peripheral sensory neurons in vitro [13] diabetic neuropathy and other neurological dis-
and disruption of NGF in adult rodents does not eases in China [37] and in a recent exploratory
induce marked neuronal loss. However, NGF trial recombinant human NGF was reported to
depletion does impact the phenotype of NGF-­ alleviate indices of erectile dysfunction [38].
sensitive adult neurons, for example, by reducing Alternative experimental approaches designed to
expression of the neuropeptide neurotransmitters avoid NGF-induced injection site pain include
substance P and CGRP [14] and by reducing inducing site-specific expression of NGF via a
nociceptor sensitivity [15], thereby impeding viral vector [39] and inducers of local NGF pro-
heat pain perception [16]. Conversely, NGF also duction that concurrently improve indices of neu-
activates inflammatory cells and mediators, sen- ropathy in diabetic rodents. Examples range from
sitizes nociceptors, and induces both myalgia and botanical extracts [40] and vitamin derivatives
chronic injection site pain when given to healthy [41] to small molecules [42, 43], a CB-1 receptor
volunteers [17–19]. antagonist [44] and a PDE5 inhibitor [45].
However, these molecules also induce a complex
Expression of NGF and its receptor sub-units array of alternative potentially neuroprotective
is reduced in skin, and muscle of diabetic rodents responses and ascribing efficacy specifically to
and the retrograde axonal transport of locally induction of NGF has not been possible. As dis-
produced or injected NGF is also reduced [20– cussed below, the ability to modify NGF to
23]. This diminution of NGF support is consis- exclude its algogenic capacity and allow pain
tent with the parallel diabetes-induced reduction free delivery has recently emerged as a novel
in protein levels and axonal transport of NGF-­ therapeutic approach [46].
regulated gene products such as substance P and The suggestion that NGF is involved in gener-
CGRP [20, 24–28] that, in turn, leads to sensory ating some forms of chronic pain has led to anti-­
dysfunction via reduced peripheral and spinal NGF therapies being explored as an approach to
release of these neuropeptides during stimulation suppressing pain states [47, 48]. For example, the
of sensory neurons [29–33]. The capacity of NGF recombinant humanized monoclonal antibody
to both promote sensory nerve growth and regu- Tanezumab, which binds NGF and prevents
late nociceptor phenotype has prompted investi- interaction with its receptor, was developed for
gation of its potential as a therapy for diabetic use against osteoarthritis [49] and chronic low
neuropathy. back pain [50, 51] despite initial concerns regard-
Multiple indices of small fiber neuropathy in ing potential accelerated damage to joints that
diabetic rodents are prevented or reversed by sys- disrupted development programs [52].
temic or intrathecal injection of NGF [22, 26, Tanezumab also produced clinically significant
34]. There are also reports of an impact on indi- improvement of pain scores above placebo lev-
ces of large fiber neuropathy [35], although this is els, without worsening neuropathy, in a small
presumably an indirect effect given the absence cohort of patients with painful diabetic neuropa-
of TrkA receptor expression by most large thy [53] and a more expansive phase 2 clinical
myelinated fibers. These positive preclinical find- trial has recently reported data (NCT01087203).
130
BDNF  BDNF is widely studied because of its
roles in the brain in relation to development and
modulation of synaptic activity and plasticity. In
the context of diabetes, reduced BDNF expres-
sion is implicated in encephalopathy [54],
dementia [55], and retinopathy [56]. A variety of
cells associated with the PNS, including muscle,
keratinocytes, and Schwann cells express BDNF
and induction of BDNF following nerve injury
plays a role in neuronal survival and/or regenera-
tion [57]. Responsiveness to BDNF is conferred
by expression of the TrkB receptor sub-unit
operating in conjunction with the p75NTR. TrkB
is localized to 10–25% of sensory neurons in the
dorsal root ganglia, mainly large myelinated
fibers, and to motor neurons, Schwann cells and
post-synaptic muscle at the neuromuscular junc-
tion [58]. Small, NGF-sensitive sensory neurons
also constitutively express BDNF and transport
it to their central terminals in large dense core
(peptidergic) vesicles where, upon release, it
modulates nociceptive input by enhancing
NMDA receptor-mediated glutamatergic excit-
atory signaling [59–62]. Thus, broadly, in the
PNS, motor and some large sensory fibers are
responsive to BDNF while a proportion of small
sensory neurons produce BDNF and secrete it
centrally to modulate nociception (Fig. 1).
Hyperglycemic rodents show increased
BDNF mRNA and protein in muscle, which has
been proposed to reflect a response to metabolic
injury [21, 63–65]. In contrast, whole nerve
BDNF mRNA and protein levels are reduced
[66–68], possibly a consequence of
hyperglycemia-­induced impaired expression by
Schwann cells [69]. Diabetes also impedes
peripherally-derived support of large fibers by
BDNF as there is a decrease in TrkB receptor
mRNA in sciatic nerve [70] and reduced retro-
grade axonal transport of endogenous BDNF
from the periphery [71]. These observations led
to studies investigating the efficacy of BDNF
replacement therapy against indices of neuropa-
thy in diabetic rodents. BDNF injections did not
prevent indices of small fiber dysfunction, such
as reduced neuropeptide expression [26] and
thermal hypoalgesia [72] in diabetic rats, which
N. A. Calcutt
is consistent with the absence of TrkB on most
small sensory fibers. It also had no effect on
conduction deficits and myelin damage in large
sensory fibers of galactose-fed rats (a model of
polyol pathway hyperactivity) but did protect
against these disorders in large motor fibers
[64]. However, a subsequent clinical trial of
recombinant human BDNF in diabetic subjects
with neuropathy failed to demonstrate improve-
ment of large fiber conduction slowing after 3
months of treatment in a small group of sub-
jects, although improvement in cool thermal
detection thresholds was noted [73]. As with
NGF (see above), BDNF is induced by a variety
of potential therapeutic agents and has been
linked to their efficacy against indices of neu-
ropathy in diabetic rodents, although multiple
other plausible therapeutic pathways are also
modulated and causality has not be established
[66, 67, 74, 75]. BDNF mimetics have been
developed and promote nerve growth [76, 77]
but efficacy against diabetic neuropathy has yet
to be reported.
BDNF and Pain  A role for BDNF in spinally-­
mediated pain during diabetes is suggested by
studies showing that indices of painful neuropa-
thy in diabetic rats can be suppressed by intrathe-
cal delivery of a TrkB/Fc chimera acting as a
BDNF decoy [78, 79]. The source of spinal
BDNF in diabetes has been attributed to activated
microglia [79], glial cells that are increasingly
described as modulators of nociceptive input via
release of a variety of inflammatory and pro-­
nociceptive factors, including BDNF, that impact
both pre- and post-synaptic neurons [80–82].
Primary afferents represent another potential
source of increased BDNF release in the spinal
cord [83]. BDNF expression increases dramati-
cally after inflammatory or physical injury,
including induction in larger sensory neurons
that mediate allodynia to light touch [84, 85].
While there is reduced expression of BDNF by
Schwann cells in the sciatic nerve and spinal
roots in diabetic rodents, diabetes increases
BDNF mRNA and protein in DRG [21, 74, 86,
87] suggesting the potential for increased central
transport and release.
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
Mechanisms by which BDNF promotes
spinally-­mediated pain in diabetes may include
increased NMDA receptor signaling in neurons
and microglia [88] and disruption of GABAergic
inhibitory tone. The latter is a consequence of the
capacity of BDNF to regulate expression of the
membrane ion pump, potassium/chloride co-­
transporter KCC2 [89]. During neuronal develop-
ment, BDNF levels are high, KCC2 pump
expression and activity is low and the post-­
synaptic membrane chloride gradient is such that
activation of the GABAA receptor (a chloride
channel) is excitatory. Upon maturation of the
nervous system, decreased expression of BDNF
results in increased expression of KCC2 which
increases the membrane chloride equilibrium
potential and switches GABAA receptor function
to inhibitory [90]. Elevated spinal BDNF follow-
ing nerve injury in adult animals has been linked
to decreased KCC2, loss of GABAergic inhibi-
Fig. 2  BDNF-mediated spinal disinhibition. Diabetes-­
induced increased BDNF released from primary afferents
and/or activated spinal microglia reduces spinal KCC2
expression, which results in the GABAA receptor losing
inhibitory function and possibly assuming an excitatory
function. Loss of spinal GABAergic inhibition has two
consequences: disinhibition of sensory throughput lead-
131
tion, and a state of spinal disinhibition that drives
or amplifies pain-associated behaviors [91].
KCC2 expression is reduced in diabetic rats with
an attendant shift of spinal GABAergic function
from inhibitory to excitatory, so that spinal deliv-
ery of GABAA antagonists blocks indices of pain
rather than promoting it [92, 93]. Conversely,
incapacitating spinal BDNF in diabetic rats
restores KCC2 expression, GABAergic inhibitory
function and normalizes pain behavior [78, 79].
Whether there is disruption of the spinal
BDNF/KCC2/GABAA receptor axis in humans
with painful diabetic neuropathy is not known.
However, the phenomenon of rate-dependent
depression (RDD) of the electromyogram H
wave following repeated stimulation of periph-
eral nerve also reflects spinal GABAergic inhibi-
tory function in normal animals and there is loss
of RDD in diabetic rodents [78, 92, 94–96]. Loss
of RDD (Fig. 2) has been proposed as an electro-
ing to increased pain perception (purple arrows) and loss
of rate-dependent depression of the H wave (RDD) in the
EMG following repetitive peripheral stimuli (blue
arrows). Loss of RDD may therefore serve as a clinical
biomarker for patients with painful diabetic neuropathy
driven by reduced spinal disinhibition
132 N. A. Calcutt

physiological biomarker of spinally-mediated port of NT-3 to the neuronal perikaryon and

pain arising from an abnormal BDNF/KCC2/ sub-optimal neurotrophic support [70, 108, 109].
GABA axis [78, 86]. Consistent with this hypoth- Similar to BDNF (see above), impaired target-­
esis, cohorts of subjects with type 1 or type 2 dia-derived support is accompanied by increased
betes with painful diabetic neuropathy showed expression of NT-3 mRNA in dorsal root ganglia,
impaired RDD compared to non-diabetic sub- which may reflect a compensatory mechanism
jects or patients with painless diabetic neuropa- [110]. In contrast, both TrkC mRNA and NT-3
thy [97, 98]. As only a sub-group of the painful protein are increased in the skin of diabetic sub-
diabetic neuropathy group showed markedly jects with neuropathy and it has been suggested
abnormal RDD, this disorder has the potential to that this reflects a response to ongoing denerva-
serve as a biomarker for identifying patients in tion [111, 112].
whom pain is mediated primarily by altered spi- The capacity of NT-3 to support a broader
nal disinhibition [86, 97]. Interestingly, a sub-­ array of peripheral neuron types than other neu-
group of patients with painless neuropathy rotrophins makes it an attractive potential ther-
showed exaggerated RDD, raising the possibility apeutic to treat diabetic neuropathy, albeit with
of enhanced spinal GABAergic inhibition con- the potential caveat of having to negotiate
tributing to an absence of pain [98]. reduced TrkC expression by target neurons
[108]. Systemic delivery of NT-3, protected
NT-3  Within the PNS, the TrkC receptor sub-­ large fiber conduction velocity in rat models of
unit that confers NT-3 selectivity is found in hyperglycemia [113, 114] and diminished axo-
motor neurons, large diameter sensory neurons nal transport of diabetes-induced stress signals
innervating muscle, and sympathetic neurons [115] whereas it was without effect on indices
[99, 100]. Low affinity binding to combinations of autonomic neuropathy [116]. Exercise-
of p75NTR and Trk’s A or B also occurs [101] induced induction of NT-3 expression in the
making NT-3 the most promiscuous of the neuro- muscle of diabetic rats is also accompanied by
trophins with potential to provide the broadest prevention of nerve conduction slowing,
therapeutic coverage. NT-3 itself is expressed by although causality was not confirmed [117].
skeletal muscle and retrogradely transported in Delivery of NT-3 to central terminals of pri-
the manner of a classical target-derived trophic mary afferents via intrathecal injection pre-
factor [102]. However, there is also evidence that vented loss of dermal myelinated fibers and
NT-3 undergoes anterograde axonal transport in increased Merkel cell number in diabetic mice;
certain neuronal populations and, as discussed but did not impact mechanical hypoalgesia
above for BDNF, may serve as a modulator of [118]. Interestingly, when NT-3 therapy was
synaptic transmission [103, 104] and synaptic combined with NGF it diminished the efficacy
organization [105] in the spinal cord. TrkC is of NGF alone, possibly by competing with
expressed in the corticospinal tract, fasciculus NGF for the p75NTR [118]. The mechanisms by
cuneatis and fasciculus gracilis of the spinal cord which NT-3 protects neurons from diabetes-­
and NT-3 has been explored to treat spinal cord induced damage include normalization of cal-
injury [106]. cium homeostasis [119], prevention of
mitochondrial dysfunction [120], and mainte-
Reduced NT-3 has been reported in the muscle nance of appropriate neurofilament organiza-
of diabetic patients, especially those with neu- tion [121].
ropathy and correlated with muscle weakness NT-4, like BDNF, signals through the TrkB
[107]. Expression of NT-3 mRNA and protein is receptor but studies of this neurotrophin are lim-
also reduced in the skeletal muscle of type 1 dia- ited to a report of decreased NT-4 in the sciatic
betic rodents and, in combination with reduced nerve of diabetic rats [70], while in human ­muscle
neuronal expression of the TrkC receptor sub-­ NT-4 expression parallels the decrease seen in
unit, results in a reduced retrograde axonal trans- NT-3 [107].
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
3 GDNF Family Ligands
Members of the GDNF family ligands: glial cell
line-derived neurotrophic factor (GDNF), arte-
min, neurturin, and persephin, exhibit relatively
low amino acid sequence homology but are
placed in the same family based on their activa-
tion of the tyrosine kinase receptor “rearranged
during transfection” (RET). RET is activated fol-
lowing binding of a factor to the co-receptor,
GDNF receptor alpha (GRFα), which itself has
four variants (GRFα 1-4) that confer ligand selec-
tivity or specificity [122]. Activation of RET trig-
gers signaling through one or more multiple
cascades, including the MAPK, PI3K, Erk, and
Akt pathways, depending on the specific ligand/
receptor configuration and cell. In Schwann cells
and select neuronal populations, GDNF can also
signal through GRFα in the absence of RET
[123] while interaction of activated RET with the
neurotrophin p75NTR sub-unit has also been
described as part of what is increasingly viewed
as a complex mechanism for cross-talk between a
variety of neurotrophic factors [124].
GDNF can be viewed as a classic target
derived neurotrophic factor for motor neurons
(Fig. 1) as it is expressed by adult skeletal mus-
cle, binds GRFα1/RET located on motor neuron
synaptic terminals, is internalized and then retro-
gradely transported and promotes axonal growth,
synapse formation, and modulation of pre- and
post-synaptic function [125, 126]. GFR/RET sig-
naling also occurs in sub-populations of primary
sensory neurons and postganglionic autonomic
neurons during development [127–129]. In adult
DRG around 35% of neurons, predominantly
small non-peptidergic neurons that label with
IB4 and lack the TrkA receptor sub-unit, are
GDNF responsive [130] while neurons and
Schwann cells increase expression of GDNF fol-
lowing injury [131, 132].
Diabetic rodents [133] and humans [134] have
been reported to exhibit increased mRNA for
GDNF in skin, whereas GDNF mRNA and pro-
tein are decreased in nerve from diabetic rats
[135], suggesting a complex progression of
diabetes-­induced lesions and responses to injury.
The therapeutic potential of GDNF was estab-
133
lished by delivery of the factor direct to the spinal
cord of diabetic mice that demonstrated efficacy
against loss of non-peptidergic neuronal termi-
nals in the spinal cord [136] and skin [133] and
also loss of responses to chemical stimuli [137].
More recently, small molecule RET agonists
[138] or modulators of RET signaling [139] have
shown efficacy against behavioral indices of sen-
sory dysfunction in diabetic rodents. Of other
members of the family, neurturin also protected
against loss of cutaneous nerves [133].
4 IL-6 Cytokine Family
This superfamily of cytokines is defined by their
signaling through a receptor complex that shares
a common gp130 sub-unit that interacts with
cytokine-specific sub-units. It includes
Interleukins 6, 11, 27, and 30, along with ciliary
neurotrophic factor (CNTF), leukemia inhibitory
factor (LIF), oncostatin M (OSM), neuropoietin,
cardiotrophin-like cytokine factor (CLCF1), and
cardiotrophin-1 (CT-1). A sub-group, comprising
CNTF, LIF, OSM, CLCF1, and CT-1 share both
gp130 and the LIF receptor (LIFR) sub-units and
signal via the JAK/STAT, MAPK, and PI3K/Akt
pathways so that there is some overlap in their
phenotypic consequences, such as capacity to
regulate metabolism and impact body weight
[140, 141]. While the ability to reduce food con-
sumption, blood glucose and body weight in
obese models of type 2 diabetes [142] may have
indirect impact on the nervous system and dia-
betic neuropathy, there is also evidence for direct
actions of some members of this group of cyto-
kines on the PNS.
CNTF is a 22-kDa polypeptide neurotrophic
factor that is expressed in a wide range of tissues
including bone marrow, testis, and heart, but
which has been widely studied to date in the con-
text of actions in the nervous system. Cells
responsive to CNTF express the glycosyl-­
phosphatidylinositol (GPI) linked CNTF
­receptor sub-unit [143] which confers high affin-
ity CNTF binding and signaling activity when
combined with both the LIFR and the gp130
sub-­units. CNTF also shows low affinity binding
134
to the IL-6R [144]. Early studies showed CNTF
expression localized to Schwann cells of periph-
eral nerve [145, 146] with levels increasing dur-
ing post-natal development to steady state in
adult nerve. Nerve lesion results in extracellular
accumulation of CNTF at the lesion site fol-
lowed by downregulation of the factor as
Schwann cells enter their non-myelinating pro-
liferative stage and then a restoration of expres-
sion that accompanies re-myelination [147–149].
The ability of CNTF to prevent neuronal death
and promote axonal growth after lesion [150–
152], along with the absence of a known secre-
tion mechanism, prompted the suggestion that
CNTF release from damaged Schwann cells pro-
vides an immediate neuroprotective response to
physical nerve injury [152, 153] that precedes
the temporally-regulated increase in production
of other growth factors by Schwann cells and
other cells at the lesion site. This has led to many
studies that have looked to exploit local delivery
of CNTF and other growth factors to promote
peripheral nerve regeneration after traumatic
injury [154, 155]. Aside from a role following
traumatic injury, CNTF also appears to contrib-
ute to normal maintenance of peripheral motor
neurons. The PNS of CNTF-­null mice develops
normally but there is loss of facial motor neurons
during adult life [156] and morphological abnor-
malities of the paranode and myelin in motor
neurons of L5-L6 spinal ventral roots between 3
and 9 months of age accompanied by impaired
axonal diameter maturation [157]. Depletion of
Schwann cell derived CNTF therefore has the
capacity to contribute to the pathogenesis of
both axonal and glial manifestations of periph-
eral neuropathy.
Rat models of diabetic neuropathy develop
reduced CNTF bioactivity and protein levels in
peripheral nerve over a time period that encom-
passes onset of early physiological and struc-
tural indices of neuropathy, such as NCV
slowing and reduced axonal caliber [158–160].
CNTF depletion and the accompanying reduc-
tion of axonal caliber were prevented by aldose
reductase inhibition [159] a therapeutic
approach that targets metabolism of excess
glucose in Schwann cells [161]. Thus, hyper-
N. A. Calcutt
glycemia driven excess polyol pathway flux
may contribute to the pathogenesis of diabetic
neuropathy by disrupting expansion of axonal
diameter (and hence conduction velocity) sec-
ondary to depletion of Schwann cell derived
CNTF.  How CNTF interacts with axons
remains unclear given the absence of an estab-
lished secretory mechanism, although direct
transfer between Schwann cell and axonal
cytoplasm is plausible given that other cyto-
plasmic constituents undergo such transfer
[162–169].
The therapeutic potential of CNTF was
determined by treatment with recombinant
CNTF, which prevented impaired nerve regen-
eration after sciatic nerve crush in diabetic rats
[170]. Impaired corneal nerve regeneration fol-
lowing physical lesion in diabetic mice and the
degeneration of amacrine and retinal ganglion
cells in diabetic rats were also prevented by
local CNTF delivery [171, 172]. These findings
are consistent with the capacity of CNTF to
promote axonal regeneration, including
enhancement of neurite outgrowth from sen-
sory neurons derived from diabetic rodents
[173]. Exogenous CNTF also ameliorated paw
thermal hypoalgesia, reduced axonal caliber
and the accompanying motor and sensory nerve
conduction slowing of uninjured nerve in dia-
betic rodents [72, 170, 173]. These effects were
seen in the absence of impact of CNTF on
hyperglycemia and suggest neuroprotective
actions independent of nerve injury and repair
programs. Putative mechanisms that may
impact both nerve repair after physical injury
and resistance to metabolic insult involve resto-
ration of normal mitochondrial function in the
nerve of diabetic rodents via activation of JAK/
STAT signaling and NF-KB [173, 174].
Unfortunately, early clinical trials of systemic
CNTF in subjects with motor neuron disease
identified a variety of dose-related and dose
limiting side effects ranging from cough to
anorexia and weight loss [175] that largely dis-
couraged further development of the factor as a
therapy, although potential use as a weight loss
agent was subsequently explored [176]. As with
many neurotrophic factors, systemic delivery of
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
CNTF is unlikely to be viable due the diverse
array of cells and tissues that express the CNTF
receptor and the many biological pathways it
activates. Localized delivery to treat degenera-
tive eye diseases has shown promise [177, 178]
and, as discussed below, use of engineered cells
that distribute to specific parts of the body and
secrete CNTF as part of a cocktail of supportive
factors [179] may offer the most practical
approach for any future use against diabetic
neuropathy.
Of other members of this cytokine family,
the IL-6 receptor sub-unit (IL-6R) is expressed
by myelinating Schwann cells and localizes to
the nodes of Ranvier [180]. There is also a sol-
uble form of the receptor sub-unit (sIL-6R)
that binds secreted IL-6 and can interact with
membrane bound gp130 to confer IL-6 sensi-
tivity on cells that do not themselves express
IL-6R, in a process termed trans-signaling
[181]. Following physical injury, IL-6 is
secreted by denervated Schwann cells and
IL-6R expression increases suggesting roles
for IL-6 in axon:Schwann cell interactions dur-
ing maintenance of steady state and for myelin
clearance via monocyte recruitment and re-
myelination during nerve regeneration [182–
184]. There are reports that IL-6 treatment
ameliorated multiple indices of neuropathy in a
rat model of type 1 diabetes [185–187] and
suggestions that it may be a viable therapy
under controlled release conditions [188].
However, it is becoming increasingly appreci-
ated that IL-6 has differing properties depend-
ing on whether it signals via membrane bound
(classical signaling, broadly anti-­inflammatory)
or soluble (trans-signaling, broadly pro-inflam-
matory) IL-6R [181]. Recent interest in the
role of IL-6 in diabetic neuropathy has largely
shifted to the pro-inflammatory actions of IL-6
and its potential role in diabetic neuropathic
pain that have arisen from reports of increased
expression of IL-6 in peripheral nerve and spi-
nal cord [189, 190].
There have been few studies of other mem-
bers of the IL-6 cytokine family in the context
of diabetic neuropathy. A transient increase
expression of the gp130 and LIFR receptor
135
sub-units in sciatic nerve after onset of diabe-
tes has been reported [191] while attenuated
induction of LIF and IL-6 in sensory and auto-
nomic ganglia with reduced STAT signaling
and reduced expression of regeneration associ-
ated genes is seen following nerve crush injury
to diabetic mice and accompanies impaired
nerve regeneration [192]. However, as yet
these observations have not been exploited to
investigate new therapeutic approaches for dia-
betic neuropathy.
5 Insulin, C-Peptide,
and Insulin-Like Growth
Factors
Unlike muscle and fat, where glucose uptake
occurs via the insulin-regulated GLUT4 trans-
porter, perineurial and endothelial cells of the
blood:nerve barrier utilize the insulin-­
independent GLUT1 glucose transporter [193]
while both GLUT1 the insulin-independent
GLUT3 high affinity glucose transporter are
expressed by axons and Schwann cells [194].
Consequently, nerve glucose levels directly
reflect plasma glucose levels during hyperglyce-
mia. Agents that block metabolism of excess glu-
cose in nerve without impacting systemic insulin
or glucose levels, such as aldose reductase inhibi-
tors, prevent and reverse multiple indices of neu-
ropathy in diabetic rodents [195], although to
date they have failed to translate to clinical use
[196]. This glucotoxicity, along with the occur-
rence of peripheral neuropathy in both type 1 and
type 2 diabetes, was historically cited as a ratio-
nale for hyperglycemia being the primary patho-
genic mechanism that initiated diabetic
neuropathy [197]. However, both secreted prod-
ucts of pro-insulin cleavage, insulin and
C-peptide, have neurotrophic properties and it is
increasingly recognized that insulinopenia may
produce a deficit in trophic support to neurons
that contributes to the pathogenesis of diabetic
neuropathy. Impaired insulin signaling also
occurs in the brain during diabetes [198, 199] and
contributes to cognitive impairments associated
with both diabetes and also other neurodegenera-
136
tive diseases such as Alzheimer’s Disease. The
role of insulin deficiency in CNS pathology, in
terms of both its metabolic and neurotrophic
effects has been extensively reviewed
[200–202].
Insulin  High affinity insulin receptors are pres-
ent on endothelial cells, pericytes, Schwann cell,
and neurons in the PNS [203–205]. In both
Schwann cells and axons, insulin signaling
drives growth and maturation via the phos-
phoinositide 3-kinase (PI 3-kinase)/Akt signal-
ing pathway [206, 207]. The direct effect of
insulin in promoting axonal growth was first
demonstrated using sensory neurons derived
from adult rodents [208]. While embryonic
knockout of neuronal insulin receptors did not
generate a peripheral neuropathy phenotype
[209], dual knockout of receptors for insulin,
and the related factor IGF-1  in Schwann cells
resulted in the development of peripheral nerves
exhibiting both reduced axonal diameter and
thin myelin that persisted into adulthood. Adult
mice also displayed multiple indices of both
large and small fiber neuropathy that replicated
those seen in diabetic mice [206]. Insulin may
therefore be viewed as trophic factor for periph-
eral nerve and insulin deficiency or receptor dys-
function has the potential to contribute to
neuropathy independent of glycemic status [210,
211].
Studies directly comparing rat models of
type-1 and type-2 diabetes with equivalent
hyperglycemia demonstrated that indices of
neuropathy develop earlier, and are more
extreme, in the insulin-deficient animals, sup-
porting the argument that insulin deficiency
contributes to neuropathy in type-1 diabetes
[212, 213]. There is also a growing suspicion
that neuropathy phenotype, and indeed patho-
genesis, may vary between type 1 and type 2
diabetes [214, 215], although both forms of the
disease incorporate impaired insulin signaling
in rodent nerve [216, 217]. Further evidence for
a role of impaired insulin-­mediated trophic sup-
port in neuropathy comes from animal studies in
N. A. Calcutt
type 1 diabetic rodents where insulin was pro-
vided at levels that did not impact hyperglyce-
mia. Low doses of systemically-­ delivered
insulin correct impaired calcium homeostasis in
the perikarya of sensory neurons from diabetic
rats [119] and protect neuronal mitochondrial
phenotype and function [119, 207, 218–220].
This is accompanied by prevention or attenua-
tion of multiple indices of large and small fiber
neuropathy. Similar efficacy against functional
and structural markers of neuropathy is seen
when low dose insulin is delivered by intrana-
sal, intrathecal or intraneuronal injection [204,
221–225].
C peptide  C peptide is a 31 amino acid poly-
peptide that connects the A and B chains of
insulin within the pro-insulin molecule, sepa-
rated from the insulin chains prior to secretion
and released from pancreatic Beta cells in equal
amounts to insulin. While initially considered
inert, it has been argued that C-peptide has a
physiologically relevant role in modulating
insulin signaling and that C-peptide deficiency
contributes to the disease phenotype of type 1
diabetes [226, 227], including the more severe
neuropathy reported in type 1 versus type 2 dia-
betic rats [212]. C-peptide promotes intracellu-
lar signals in multiple cell types, including
peripheral nerves [228] and the orphan G pro-
tein coupled receptor GRP146 has been pro-
posed as one of potentially many C peptide
receptors [229], although concerns of these data
remain [230].
Treating diabetic rodents with C-peptide
prevents or reverses multiple indices of neu-
ropathy with a mechanism of action that
includes potentiation of insulin signal transduc-
tion, induction of other neurotrophic factors,
and stimulation of nitric oxide-mediated vaso-
dilation [231, 232]. Similar efficacy was dem-
onstrated for a long acting analog of C-peptide
[233] and this molecule was advanced to phase
III clinical trial in patients with type 1 diabetes
following promising outcomes when using
native C-peptide in exploratory studies [234–
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
236]. Long acting C-peptide significantly
improved sensory nerve conduction from base-
line but a similar effect was seen in the placebo
group so that the primary efficacy measure was
not supported [237]. C-peptide also signifi-
cantly improved vibration perception threshold
against both baseline and placebo but this posi-
tive effect on a patient reported outcome was
not considered sufficient to advance the thera-
peutic approach further.
Insulin-like Growth Factors  IGF-1 and IGF-2
are single chain polypeptides that share around
45% sequence homology to pro-insulin. They
have a variety of survival and growth related
effects throughout the body, with IGF-2 being
particularly important during development,
while the effects of IGF-1 extend throughout
life [238]. Both IGFs bind to the IGF-1 receptor
(IGF-1R) which signals through tyrosine
kinases and multiple downstream pathways
including the PI 3-kinase/Akt and Ras/Erk cas-
cades. IGF-1 also binds to the insulin receptor
and insulin can activate IGF-1R so, while each
factor has higher affinity for its own cognate
receptor, there is a degree of functional overlap
[239]. A second receptor, IGF-2R selectively
binds IGF-2 but does not transduce a signal,
serving instead to reduce circulating IGF-2 by
sequestering, internalizing, and degrading it.
There is further regulation of circulating IGF’s
by six IGF binding proteins (IGFBP1-6) and
assorted proteases that regulate the IGFBP’s.
The IGF-1R is expressed throughout the periph-
eral nervous system and IGF-1 has effects on
Schwann cells and sensory, motor, and auto-
nomic neurons. Mice deficient in IGF-1, or
IGF-1R or that overexpress IGFBP5 show func-
tional and structural pathology [240] indicating
a role in maturation and maintenance of the
PNS, while IGF-1 also plays a prominent role in
orchestrating many aspects of nerve regenera-
tion after injury [241].
Reduced plasma levels of IGF-1 [242] and
reduced expression of mRNA for IGF-1, IGF-2,
and IGF-1R by peripheral nerve are an early and
137
persistent feature in insulin-deficient diabetic rats
[243–245]. Nerve IGF-2 is also reduced in rat
models of type-2 diabetes and, while plasma
IGF-1 and nerve IGF-1 mRNA remain normal
[246, 247], nerve of type 2 diabetic mice shows
IGF-1 resistance that parallels insulin resistance
[216]. The complex interactions of IGF, IGFBP,
and IGFR have also been investigated [248, 249].
It has been suggested that downregulation of
IGF-1 in Schwann cells is secondary to increased
glucose metabolism through the polyol pathway
[250]. Given the broad impact of IGFs on periph-
eral nerve, such findings prompted multiple pre-
clinical studies to assess efficacy of systemic or
intrathecal delivery of IGF-1 in diabetic rodents
and there are many reports of improved func-
tional and structural indices of neuropathy [79,
204, 225, 247, 251–257]. A recent study found
that IGF-1 causes acute injection site hyperalge-
sia in diabetic mice [258] so reports that local
IGF-1 and/or IGF 1R expression may be induced
by a variety of agents that concurrently impact
neuropathy in diabetic rodents [259–261] may
provide an alternative to direct injection of the
factor.
The pertinence of IGF replacement therapy in
humans is supported by reports that circulating
IGF-1 is reduced in both type 1 and type 2 dia-
betic patients [262, 263] and that the deficit is
more extreme in those with neuropathy [264].
However, the subtleties of the IGF/IGFBP/IGFR
axis remain largely unexplored and the pleiotro-
pic systemic effects of IGFs have caused their
therapeutic development to proceed with caution
in recognition of the potential for multiple side
effects. Another caveat to the development of
IGF’s as potential therapeutics is the need to con-
sider the impact of the IGFBP, which modulate
the bioavailability of circulating IGFs. Circulating
levels of IGFBP1 are increased in type-1 diabetic
patients with neuropathy [265], while nerve from
a small group of diabetic subjects with neuropa-
thy showed reduced mRNA for IGF-1 but dra-
matically increased expression of IGFBP5
protein with the latter proposed to sequester
IGF-1 and thereby reduce it effective concentra-
tions in neurons [266].
138 N. A. Calcutt

There are multiple potential mechanisms by The hedgehog family of morphogens, com-
which IGFs may impact diabetic neuropathy prising sonic (SHH), desert (DHH), and Indian
given their pleiotropic metabolic and neuro- (IHH) regulate many aspects of embryonic
trophic capacity. For example, efficacy of sys- development including growth and patterning of
temic IGF-1 treatment may be secondary to the PNS [277]. They signal via a complex pro-
indirect effects, such as the induction of the cess that begins with binding to the transmem-
angiogenic and neurotrophic polypeptide vas- brane receptor patched that releases tonic
cular endothelial growth factor (VEGF) in inhibition of an associated transmembrane sig-
muscle and nerve of diabetic rodents [267, naling protein, smoothened (SMO). Disinhibition
268]. Direct effects of IGF-1 include protec- of SMO triggers signaling that activates multiple
tion of mitochondrial function [269] and aug- transcription factors to regulate cell phenotype.
mentation of ­ glycolytic capacity in neurons Consistent with their regulation of neuronal
[270], and regulation of Schwann cell differen- development, traumatic injury to the adult PNS
tiation status [250]. induces expression of both DHH and SHH, a
response that has been implicated in multiple
aspects of the nerve survival, repair, and regen-
6 Morphogens and Axonal eration programs [278]. A role in the develop-
Guidance Factors ment of endoneurial blood vessels has also been
suggested by studies in DHH knockout mice
Morphogens are key regulators of developmental [279]. DHH is expressed by Schwann cells in
biology that exert positive or negative influences adult peripheral nerve and mRNA levels are
on the patterns of tissue development by virtue of reduced by diabetes [279, 280], perhaps reflect-
their non-uniform distribution and consequent ing the metabolic damage to these cells that also
establishment of a concentration gradient. In the results in parallel CNTF depletion (see above).
context of development of the nervous system Treating diabetic rats with a SHH-­IgG fusion
there are many molecules that provide guidance protein ameliorated multiple indices of small
cues for axonal pathfinding including the ephrin, fiber neuropathy including the depletion of NGF
semaphorin, slit, and netrin families. Of these, protein and its retrograde axonal transport,
netrin has been reported to correct NCV slowing reduced neuropeptide proteins and reduced large
in adult diabetic mice via a presumed pro-­ fiber conduction velocity and axonal caliber
angiogenic mechanism [271]. Non-conventional [280]. Various small molecule agonists and
axonal guidance molecules include morphogens antagonists of the hedgehog/patched/Smo sig-
of the TGF-β, Wnt ligand, and hedgehog families naling pathways have been produced [281] and
[272]. In the adult PNS, TGF-β is expressed by an agonist of Smo, used to replicate the down-
neurons, Schwann cells, and pericytes and both stream consequences of hedgehog binding to
disrupts the basement membrane of the patched, prevented heat hypoalgesia and tactile
blood:nerve barrier and reduces neurite out- allodynia in diabetic mice [279]. My own expe-
growth in vitro [273, 274]. Expression of TGF-β rience is that efficacy of potent hedgehog path-
mRNA and protein is induced in the DRG and way agonists against indices of neuropathy in
nerve by diabetes, suggesting potential indirect diabetic rodents is offset by unwanted effects
and direct neurotoxic roles [273]. Similarly, it elsewhere (Fig. 3), illustrating the difficulties of
was recently reported that markers of Wnt signal- targeting growth factor therapies to nerve when
ing are increased in sciatic nerve, DRG, and spi- they have more catholic functions. The physical
nal cord of diabetic rats and that Wnt signaling effects illustrated in Fig.  3 are consistent with
inhibitors improved multiple behavioral, func- the role of exaggerated hedgehog signaling in
tional, and structural indices of neuropathy [275, basal cell carcinomas and the use of smo antago-
276]. nists to treat this condition [282].
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy 139

MNCV

a 70 p<0.001 p<0.001 c

60
m/s

50

40
ic

st
l
ro

et

ni
nt

go
ia
co

+A
D

t ic
be
ia
D

Fig. 3 Hedgehog pathway agonism in diabetes.
Streptozotocin-diabetic rats were treated with vehicle or a
potent hedgehog signaling pathway agonist for 8 weeks.
Agonist treatment prevented MNCV slowing (panel a).
However, agonist treatment also caused the dorsal fur to
“fluff” (compare treated diabetic rat at left of panel b vs.
adjacent untreated diabetic rat) while the ventral skin
hypertrophied and depilated (panel c). After 7–8 weeks of
treatment rats began self-mutilating their ears and paws
and were euthanized. Calcutt, unpublished

7 VEGF thelial growth factor (VEGF) also falls into this

For many years, loss of trophic support and
nerve hypoxia secondary to microvascular dis-
ease have served as two parallel, often compet-
ing, hypotheses to explain the pathogenesis of
diabetic neuropathy. More enlightened views
have considered them as two of many ongoing
stressors that cumulatively weaken the capacity
of neurons to sustain their most vulnerable and
energy demanding terminal structures and
make nerves less able to survive and recover
from acute lesions. Angiogenic factors such as
basic fibroblast growth factor (bFGF) show
some efficacy against neuropathy and impaired
regeneration in diabetic rats [283, 284] while
hedgehog proteins may have both direct effects
on neurons and also indirect neuroprotective
effects via regulation of microvascular struc-
ture and function (see above). Vascular endo-
category, as while largely studied as a vasculo-
genic and angiogenic factor that has potential to
repair microvascular disease in diabetes, it also
has growth promoting properties when applied
to neurons derived from adult rodents [285,
286].
VEGFs form a sub-family of the platelet
derived growth factor (PDGF) family that con-
sists of PDGF’s A-D, VEGF’s A-D, and placenta
growth factor (PGF). VEGF-A was the first
member of the VEGF sub-family to be described
and exists in multiple isoforms due to alterna-
tive splicing of the VEGFA gene. All VEGFs
bind VEGF receptors, of which there are at least
three sub-types (VEGFR-1, 2, and 3) and there
is a degree of cross-reactivity between each
VEGF family member and the VEGFR sub-
types such that, for example, VEGF-A binds
both VEGFR-1 and VEGFR-2 while VEGF-D
140
binds VEGFR-2 and VEGFR-3. The vast major-
ity of VEGF-­mediated signaling events studied
to date represent VEGF-A binding to VEGFR-2
with subsequent activation of tyrosine kinases.
In the adult, VEGF-A expression by many cell
types is induced by hypoxia via the transcription
factor hypoxia-inducible factor (HIF) and
secreted VEGF-A binds VEGFR-2 located on
endothelial cells to promote angiogenesis,
increase vascular permeability, and regulate
multiple physiological functions throughout the
body [287].
A role for VEGF in diabetic neuropathy, be it
pathogenic or reparative, is suggested by reports
that certain gene polymorphisms associate with
susceptibility to neuropathy in diabetic patients
[288–290]. In preclinical studies, diabetes
results in transient expression of mRNA for HIF
and its target genes, including VEGF [291],
whereas expression of PDGF is reduced [67].
Increased expression of VEGF-A occurs in both
axons and Schwann cells of the PNS and was
prevented by treatment with insulin or NGF
[292, 293]. This diabetes-induced increase in
VEGF expression could represent an angiogenic
response to injury/hypoxia but may also con-
tribute to neuropathy by promoting vascular
leakage and there is evidence to support both
positive and negative effects of VEGF on periph-
eral nerve. The VEGF-­ sequestering antibody
bevacizumab, which is FDA approved as an
anti-angiogenic agent to treat assorted cancers,
prevented and reversed indices of neuropathy
and neuropathic pain in diabetic rats while
in  vitro studies suggested that Schwann cell
derived VEGF contributed to axonal neuropathy
[294]. Conversely, increasing local VEGF-A
production by either gene transfer [295–297] or
gene transfer of a VEGF transcription enhancer
[298, 299] ameliorated multiple measures of
neuropathy and neuropathic pain in diabetic
rats. Splice variants of VEGF-A have been
reported to either cause (VEGF-A165a) or allevi-
ate (VEGF-A165b) indices of neuropathic pain in
models of traumatic nerve injury [300], while
recombinant VEGF-A165b ameliorated indices of
neuropathy and neuropathic pain in diabetic
rodents [301].
N. A. Calcutt
8 Erythropoietin
Erythropoietin (Epo) is a glycoprotein cytokine
secreted by renal fibroblasts in response to
hypoxia in order to stimulate bone marrow eryth-
ropoiesis. Recombinant Epo is widely used to
treat anemia in chronic kidney disease and during
cancer therapy. Epo is also induced in multiple
cell types during periods of stress and has a vari-
ety of other actions including promoting cell sur-
vival and angiogenesis. Efficacy of Epo extends
to direct effects on neurons, as demonstrated by
in  vitro studies of cholinergic CNS neurons
[302], neuroblastomas [303], and sensory neu-
rons of the DRG [304]. Epo triggers erythropoi-
esis via the high affinity Epo receptor (EpoR)
homodimer but can also bind a lower affinity het-
erodimer comprising EpoR and β common recep-
tor (βcR) sub-units (the innate repair receptor:
IRR) that is proposed to activate tissue protection
and repair [305]. EPO signals in the CNS via
JAK2/STAT and PI3K/Akt, ERK/MAPK and
other pathways that are associated with neuro-
protection [306]. In the PNS, Epo is present in
around 30% of DRG neuronal cell bodies of all
diameters, axons, and Schwann cell cytoplasm
and is upregulated in Schwann cells after physi-
cal nerve injury, while EpoR is expressed by neu-
ronal cell bodies in the DRG, Schwann cells, and
endothelial cells [307]. Proposed functions of
this EPO/EpoR network in the PNS include neu-
ronal survival and regulation of Schwann cell
proliferation following nerve injury [308, 309]
and development of the neuromuscular junction
[303].
Initial studies to determine therapeutic effi-
cacy in diabetic rodents used recombinant EPO
[310–312] or HSV-mediated gene transfer [313,
314] both of which produced broad efficacy
against indices of peripheral neuropathy. As the
neurotrophic properties of Epo are localized to a
particular region of the peptide sequence, modi-
fied EPO can be produced that retains neuro-
trophic properties but lacks erythropoietic
capacity [303, 315, 316]. This allows targeted
neurotrophic activity while minimizing concerns
regarding excess erythropoiesis. EPO derivatives
lacking erythropoietic activity such as carba-
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
mylated EPO and ARA290 (Cibinetide), a modi-
fied 11-mer sequence of the neurotrophic region
of EPO showed efficacy against thermal hypoal-
gesia [316] impaired wound healing [317] and
autonomic neuropathy [318, 319] in diabetic
rodents. A clinical trial of AR290 given to dia-
betic subjects with neuropathy for 4 weeks, with
a further 4 weeks washout, demonstrated allevia-
tion of pain symptoms [320]. This analgesic
property may derive from inhibition of the
TRPV1 receptor [321] or suppression of spinal
microglial activation [322]. Diabetic subjects
with the most severe loss of sensory nerves in the
cornea also showed recovery of innervation
[320], but more extensive clinical data have yet to
be reported.
9 Will Neurotrophic Factors
Ever Be a Viable Therapy
for Diabetic Neuropathy?
The discovery that NGF was not the sole factor
that provided neurotrophic support to the PNS,
and technological advances that allowed synthe-
sis of large quantities of recombinant peptides
with appropriate human sequences, generated
much initial enthusiasm for using neurotrophic
factors as biologically rational and targeted treat-
ments for diabetic neuropathy, and indeed many
other neurodegenerative diseases. A summary of
neurotrophic factors used in clinical trials against
diabetic neuropathy is shown in Table 1. However,
the growing appreciation of the diversity of neu-
rotrophic and growth factors that act on the PNS
was accompanied by recognition that individual
factors were less selective for specific popula-
tions of neurons than anticipated. Consequently,
initial clinical trials of factors that relied on sys-
temic injections of recombinant peptide faced not
only challenges in delivery and dosing [36], but
also concerns regarding potential side effect pro-
files. The clinical trial of recombinant CNTF in
subjects with motor neuron disease [175] and
preclinical studies with a hedgehog agonist
(Fig. 3) illustrate the problems arising from unan-
ticipated systemic effects of neurotrophic factor
treatment. Neurotrophic factors that target small
141
fiber neuropathy due to the distribution of their
receptors, such as NGF [12], or show clinical
efficacy primarily against small fiber neuropathy,
like Epo [320], face the additional obstacle that
prevention or reversal of large fiber conduction
velocity slowing is currently the de facto primary
indicator of efficacy against diabetic neuropathy
used by the US FDA. A number of development
programs have been discontinued following lack
of efficacy against large fiber conduction slow-
ing, despite some, such as BDNF [73] and
C-peptide [237] having significant effects on
other indices of diabetic neuropathy.
Gene transfer  Use of viral vectors to induce
expression of neurotrophic factors in specific
cells has been employed as a means of removing
the need for repeated systemic injections of
recombinant polypeptides [323]. Preclinical
studies have reported efficacy against multiple
indices of neuropathy in diabetic rodents after
inducing expression of NGF [39], NT-3 [324,
325], IGF-1 [326, 327], erythropoietin [313, 314]
or VEGF [295]. A clinical trial using gene ther-
apy to increase local VEGF in diabetic patients
with neuropathy reported efficacy against symp-
tom scores, pain, and sensory loss that impact
quality of life after 6 months [328]. However,
electrophysiological measures of large fiber neu-
ropathy did not improve and a number of serious
adverse events were reported [328], perhaps
reflecting problems inherent to using a factor that
has diverse effects in multiple tissues. A separate
trial involving gene transfer of a VEGF transcrip-
tion enhancer also failed to convince in subjects
with diabetic neuropathy [329]. VEGF treatment
in subjects with diabetes is controversial as anti-
bodies that bind and neutralize VEGF are used to
treat diabetic retinopathy and macular edema
[330, 331]. Moreover, reliance on a single factor
to correct diverse aspects of neuropathy may be
optimistic. A novel variant of the gene therapy
approach was recently reported in which skin
fibroblasts of adult diabetic mice were repro-
grammed into cells expressing neuron-like mark-
ers via nano transfection of genes that code for a
range of transcription factors, with the intent of
modifying the neurotrophic environment of the
142 N. A. Calcutt

Table 1  Clinical trials of neurotrophic and other factors against diabetic neuropathy
Factor Formulation Efficacy vs. Comments Status References
NGF Recombinant Injection site Discontinued Apfel [36]
protein hyperalgesia
Anti-NGF Pain scores Recent data In development Bramson [53]
(tanezumab) (NCT01087203)
BDNF Recombinant Cold detection Injection site Discontinued Wellmer [73]
protein lesions,
increased gut
motility
C-peptide Modified peptide Vibration Placebo effect Discontinued Wahren [237]
perception on NCV
threshold obscured
efficacy
VEGF Gene therapy Pain, sensory Multiple major Discontinued Ropper [328]
loss AE
Transcription Not reported Not reported Discontinued Eisenstein [329]
enhancer
Anti-VEGF Diabetic FDA Approved Glassman [330],
(ranibizumab) retinopathy/ (2015) Striglia [331]
macula edema
Epo Neurotrophic Pain, corneal Small In development Brines [320]
peptide fragment nerve loss exploratory
(Cibinetide) study

epidermis [332]. Nano transfection increased
skin NGF and NT-3 mRNA plus NGF protein
level and was accompanied by an increase in der-
mal/epidermal PGP9.5 +ve staining which iden-
tifies neurons and Langerhans cells. Although
gene therapy remains an advance on systemic
delivery of single recombinant factors, particu-
larly where multiple factors can be induced and
localized, it also comes with its own challenges,
including how to regulate dosing and turn off fac-
tor production as needed, that will need to be
addressed for effective translation to clinical use.
Manipulating the native peptide sequence  One
approach to avoiding unwanted biological effects
of therapeutic peptides is emerging from the recog-
nition that biological proteins and peptides may
contain otherwise unanticipated neurotrophic
sequences that can be isolated from the biological
effects associated with the parent molecule. An
early example was the discovery of a neurotrophic
region in prosaposin, which is a precursor of the
lysosomal saposins that facilitate catabolism of
glycosphingolipds and also shows neurotrophic
properties when secreted in its uncleaved form
[333]. The neurotrophic peptide sequence derived
from prosaposin was effective against multiple
indices of neuropathy in diabetic rodents [334–
339] but a failed clinical trial against neuropathic
pain in patients with HIV halted clinical develop-
ment before it could be tested against diabetic neu-
ropathy [340]. Epo also contains a neurotrophic
sequence [303] that has been manipulated to
develop a neuroprotective therapeutic that lacks
erythropoietic activity and is in clinical develop-
ment [316, 320]. Neurotrophic sequences located
in multiple other biologically active peptides and
proteins were identified using the same algorithm
that was applied to prosaposin and Epo but have
not yet been pursued. Another intriguing finding is
the discovery of trophic factor mutants or splice
variants with distinct biological properties. For
example, studies of the NGFR100W mutation associ-
ated with loss of pain sensation in Hereditary
Sensory and Autonomic Neuropathy type V, have
helped separate the neurotrophic and algogenic
properties of the NGF molecule [341–344] and
may allow development of a “pain free” NGF ther-
apy. Other non-algogenic NGF mutants have also
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
been identified [46] and accelerate wound healing
and reinnervation in diabetic mice [345]. VEGF-A
also exists as splice variants that have been reported
to cause (VEGF-A165a) or alleviate (VEGF-A165b)
indices of neuropathic pain in models of traumatic
nerve injury [300]. Recombinant VEGF-A165b ame-
liorated indices of neuropathy and neuropathic pain
in diabetic rodents [301], suggesting that this vari-
ant may have therapeutic potential. Perhaps most
intriguingly, the balance between endogenous
VEGF-­A165a and VEGF-A165b can be manipulated
by modulators of alternative splicing [346, 347]
and could lead to a small molecule enhancer of
therapeutic VEGF-A165b isoform.
Manipulating neurotrophic signaling  Small
molecule inducers of endogenous neurotrophic
factor production may have a better safety profile
than exogenously delivered factors and dosing
can be better regulated than gene therapy. The
molecules studied to date tend to have a multi-
tude of biological effects outside their induction
of neurotrophic support and it is not yet clear
which are responsible for the beneficial effects on
neuropathy in diabetic rodents—although
impacting a variety of potential pathogenic
mechanisms is not in itself undesirable and may
ultimately prove a more pragmatic approach to
treating a condition that may have multiple con-
tributing lesions. The growing understanding of
the molecular biology of the receptors and sig-
naling pathways associated with neurotrophic
factors has also allowed design of small molecule
mimetics, agonists, and modulators of down-
stream signaling pathways that it is hoped will
obviate the need for peptide injections and allow
easily regulated delivery. Examples include
BDNF mimetics [76, 77], both RET agonists
[138] and modulators of RET signaling [139] to
mimic GDNF bioactivity, and a Smo agonist that
replicates the consequences of hedgehog binding
to patched [279] although, as illustrated in Fig. 3,
they remain susceptible to adverse events and
side effects depending on the profile of receptor
distribution and biological function. Most
recently, there is an interest in assuming the bio-
logical roles of neurotrophic factors by manipu-
143
lating intrinsic growth and repair mechanisms
that are distant from specific neurotrophic factors
and their immediate receptor signaling cascades
and may be common to all neuronal sub-­
populations. A strong case has been made that
nerve growth and repair is tonically restrained by
tumor suppressors such as PTEN, Rb1, and APC
and that lifting this restraint may allow nerve
regeneration and perhaps neuroprotection [8].
Evidence of efficacy against diabetic neuropathy
is currently limited to a study in which PTEN
knockdown improved regeneration after crush
injury in the nerve of diabetic mice [348]. Sensory
neuron regenerative growth is also under tonic
constraint mediated by cholinergic signaling via
muscarinic M1 receptors that depresses mito-
chondrial function, and cellular energy produc-
tion [349–351]. While it is not clear that excess
cholinergic constraint is a contributory mecha-
nism to diabetic neuropathy, removal of the
intrinsic cholinergic constraint on energy produc-
tion using muscarinic antagonists prevents and
reverses multiple indices of neuropathy in dia-
betic rodents [352, 353]. Further preclinical and
clinical studies of the therapeutic potential of
modulating intrinsic neuronal growth restraint
mechanisms are in progress.
Stem cells and secretomes  The impact of dia-
betes on all components of the PNS presents
problems for the therapeutic development of neu-
rotrophic factors that target specific sub-­
populations of neurons. One approach has been
to deliver cocktails of factors that together cover
a broad spectrum of neurons, such as the combi-
nation of NGF and bFGF to enhance nerve regen-
eration in diabetic rats [283]. However, this is not
always beneficial, for example, the addition of
NT-3 to NGF diminished efficacy of NGF possi-
bly by competing with NGF for the p75NTR [118].
Reprogramming keratinocytes to create a local
environment in the skin that is rich in multiple
neurotrophic factors has been discussed above
[332]. Another promising development has been
use of stem cells, not to rebuild peripheral nerve
by engraftment and differentiation but rather to
act as paracrine delivery systems for a cocktail of
144
factors, their secretome, which may not only pro-
vide neurotrophic support but could also impact
other potential pathogenic mechanisms associ-
ated with diabetic neuropathy, such as rebalanc-
ing of a pro-inflammatory environment and repair
of the microvasculature and/or blood:nerve bar-
rier [354]. Studies using stem cells from diverse
sources that were injected into diabetic rodents
have shown that efficacy to alleviate neuropathy
was accompanied by secretion of an array of fac-
tors including NGF, NT3, bFGF, and VEGF that
together can potentially impact a broad cross sec-
tion of cells associated with the PNS, while the
stem cells themselves may also differentiate into
location appropriate cell types [355–357]. In one
study, the improvement in nerve blood flow and
conduction velocity seen in diabetic rats treated
with bone marrow or blood derived mononuclear
cells was blocked by anti-VEGF antibodies
[358]. Adipose-derived stem cells (ASC’s) appear
to show particular promise as their secretome
contains diverse neurotrophic and immunomodu-
latory factors [359] that act on adjacent cells in a
paracrine manner. For example, ASC’s delivered
to the ends of transected nerves secrete multiple
neurotrophic factors, differentiate into Schwann
cells and enhance peripheral nerve repair [179].
In preclinical studies, both human ASC’s and
fluid containing their secretions reversed indices
of neuropathic pain and prevented loss of IENF
in diabetic mice [190] while ASC’s may also
impact both the vascular and neurological con-
tributors to erectile dysfunction in diabetic
rodents [360]. Given the relative ease of harvest-
ing ASC’s and their secretome, the potential for
returning ASC’s as an allograft to treat diabetic
patients is an intriguing extension of the use of
neurotrophic factors as therapies for diabetic
neuropathy that deserves further exploration.
References
1. Howe CL, Mobley WC.  Long-distance retrograde
neurotrophic signaling. Curr Opin Neurobiol.
2005;15(1):40–8.
2. Levi-Montalcini R. The nerve growth factor 35 years
later. Science. 1987;237(4819):1154–62.
N. A. Calcutt
3. LeviMontalcini R, et  al. Nerve growth factor:
from neurotrophin to neurokine. Trends Neurosci.
1996;19(11):514–20.
4. Scott-Solomon E, Kuruvilla R. Mechanisms of neu-
rotrophin trafficking via Trk receptors. Mol Cell
Neurosci. 2018;91:25–33.
5. Thomas PK.  Classification, differential diagno-
sis, and staging of diabetic peripheral neuropathy.
Diabetes. 1997;46:S54–7.
6. Goncalves NP, et al. Schwann cell interactions with
axons and microvessels in diabetic neuropathy. Nat
Rev Neurol. 2017;13(3):135–47.
7. Dong J, et  al. Mast cells in diabetes and diabetic
wound healing. Adv Ther. 2020;37(11):4519–37.
8. Duraikannu A, et al. Beyond trophic factors: exploit-
ing the intrinsic regenerative properties of adult neu-
rons. Front Cell Neurosci. 2019;13:128.
9. Nowak NC, et  al. Cutaneous innervation in
impaired diabetic wound healing. Transl Res.
2021;236:87–108.
10. Hempstead BL.  Deciphering proneurotrophin
actions. Handb Exp Pharmacol. 2014;220:17–32.
11. Marchetti L, et  al. Fast-diffusing p75(NTR) mono-
mers support apoptosis and growth cone collapse
by neurotrophin ligands. Proc Natl Acad Sci U S A.
2019;116(43):21563–72.
12. Fang X, et al. trkA is expressed in nociceptive neu-
rons and influences electrophysiological properties
via Nav1.8 expression in rapidly conducting noci-
ceptors. J Neurosci. 2005;25(19):4868–78.
13. Lindsay RM.  Nerve growth-factors (Ngf, Bdnf)
enhance axonal regeneration but are not required
for survival of adult sensory neurons. J Neurosci.
1988;8(7):2394–405.
14. Lindsay RM, Harmar AJ. Nerve growth-factor regu-
lates expression of neuropeptide genes in adult sen-
sory neurons. Nature. 1989;337(6205):362–4.
15. Bennett DLH, et al. Endogenous nerve growth factor
regulates the sensitivity of nociceptors in the adult
rat. Eur J Neurosci. 1998;10(4):1282–91.
16. Chudler EH, Anderson LC, Byers MR.  Nerve
growth factor depletion by autoimmunization pro-
duces thermal hypoalgesia in adult rats. Brain Res.
1997;765(2):327–30.
17. Barker PA, et al. Nerve growth factor signaling and
its contribution to pain. J Pain Res. 2020;13:1223–41.
18. Lewin GR, Ritter AM, Mendell LM. Nerve growth-­
factor induced hyperalgesia in the neonatal and
adult-rat. J Neurosci. 1993;13(5):2136–48.
19. Petty BG, et  al. The effect of systemically
administered recombinant human nerve growth-­
factor in healthy-human subjects. Ann Neurol.
1994;36(2):244–6.
20. Fernyhough P, et  al. Deficits in sciatic nerve neu-
ropeptide content coincide with a reduction in tar-
get tissue nerve growth factor messenger RNA in
streptozotocin-­diabetic rats: effects of insulin treat-
ment. Neuroscience. 1994;62(2):337–44.
21. Fernyhough P, et  al. Altered neurotrophin mRNA
levels in peripheral nerve and skeletal muscle
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
of experimentally diabetic rats. J Neurochem.
1995;64(3):1231–7.
22. Fernyhough P, et  al. Human recombinant nerve
growth factor replaces deficient neurotrophic
support in the diabetic rat. Eur J Neurosci.
1995;7(5):1107–10.
23. Hellweg R, Hartung HD.  Endogenous levels of
nerve growth-factor (Ngf) are altered in experimen-
tal diabetes-mellitus—a possible role for Ngf in the
pathogenesis of diabetic neuropathy. J Neurosci Res.
1990;26(2):258–67.
24. Brewster WJ, et  al. Reduced sciatic nerve sub-
stance P and calcitonin gene-related peptide in
rats with short-term diabetes or central hypoxae-
mia co-exist with normal messenger RNA levels
in the lumbar dorsal root ganglia. Neuroscience.
1994;58(2):323–30.
25. Calcutt NA, et  al. Axonal transport of substance
P-like immunoreactivity in ganglioside-treated dia-
betic rats. J Neurol Sci. 1990;96(2–3):283–91.
26. Diemel LT, et  al. Expression of neuropeptides in
experimental diabetes; effects of treatment with nerve
growth factor or brain-derived neurotrophic factor.
Brain Res Mol Brain Res. 1994;21(1–2):171–5.
27. Tomlinson DR, Fernyhough P, Diemel LT.  Role of
neurotrophins in diabetic neuropathy and treatment
with nerve growth factors. Diabetes. 1997;46(Suppl
2):S43–9.
28. Robinson JP, et  al. Axonal transport and tis-
sue contents of substance P in rats with long-
term streptozotocin-­ diabetes. Effects of the
aldose reductase inhibitor ‘statil’. Brain Res.
1987;426(2):339–48.
29. Bennett GS, et al. Neurogenic cutaneous vasodilata-
tion and plasma extravasation in diabetic rats: effect
of insulin and nerve growth factor. Br J Pharmacol.
1998;124(7):1573–9.
30. Calcutt NA. Experimental models of painful diabetic
neuropathy. J Neurol Sci. 2004;220(1–2):137–9.
31. Calcutt NA, et al. Elevated substance-P-like immu-
noreactivity levels in spinal dialysates during the
formalin test in normal and diabetic rats. Brain Res.
2000;856(1–2):20–7.
32. Garrett NE, et al. alpha-Lipoic acid corrects neuro-
peptide deficits in diabetic rats via induction of tro-
phic support. Neurosci Lett. 1997;222(3):191–4.
33. Calcutt NA, Chen P, Hua XY.  Effects of diabetes
on tissue content and evoked release of calcitonin
gene-related peptide-like immunoreactivity from rat
sensory nerves. Neurosci Lett. 1998;254(3):129–32.
34. Sango K, et al. Nerve growth factor (NGF) restores
depletions of calcitonin gene-related peptide and
substance P in sensory neurons from diabetic mice
in vitro. J Neurol Sci. 1994;126(1):1–5.
35. Unger JW, et al. Nerve growth factor (NGF) and dia-
betic neuropathy in the rat: morphological investi-
gations of the sural nerve, dorsal root ganglion, and
spinal cord. Exp Neurol. 1998;153(1):23–34.
36. Apfel SC.  Nerve growth factor for the treatment
of diabetic neuropathy: what went wrong, what
145
went right, and what does the future hold? Int Rev
Neurobiol. 2002;50:393–413.
37. Zhao M, et al. Efficacy and safety of nerve growth
factor for the treatment of neurological diseases:
a meta-analysis of 64 randomized controlled tri-
als involving 6,297 patients. Neural Regen Res.
2015;10(5):819–28.
38. Wu YX, et  al. Nerve growth factor improves the
outcome of type 2 diabetes-induced hypotestos-
teronemia and erectile dysfunction. Reprod Sci.
2019;26(3):386–93.
39. Walwyn WM, et al. HSV-1-mediated NGF delivery
delays nociceptive deficits in a genetic model of dia-
betic neuropathy. Exp Neurol. 2006;198(1):260–70.
40. Tsafack EG, et al. Antihypernociceptive and neuro-
protective effects of the aqueous and methanol stem-­
bark extracts of Nauclea pobeguinii (Rubiaceae) on
STZ-induced diabetic neuropathic pain. Evid Based
Complement Alternat Med. 2021;2021:6637584.
41. Riaz S, et  al. A vitamin D-3 derivative (CB1093)
induces nerve growth factor and prevents neu-
rotrophic deficits in streptozotocin-diabetic rats.
Diabetologia. 1999;42(11):1308–13.
42. Calcutt NA, et  al. Protection of sensory function
in diabetic rats by Neotrofin. Eur J Pharmacol.
2006;534(1–3):187–93.
43. Hanaoka Y, et  al. The therapeutic effects of
4-­methylcatechol, a stimulator of endogenous nerve
growth-factor synthesis, on experimental diabetic
neuropathy in rats. J Neurol Sci. 1994;122(1):28–32.
44. Comelli F, et  al. Rimonabant, a cannabinoid CB1
receptor antagonist, attenuates mechanical allodynia
and counteracts oxidative stress and nerve growth
factor deficit in diabetic mice. Eur J Pharmacol.
2010;637(1–3):62–9.
45. Wang L, et  al. Tadalafil promotes the recovery of
peripheral neuropathy in type II diabetic mice. PLoS
One. 2016;11(7):e0159665.
46. Malerba F, et  al. Functional characterization of
human ProNGF and NGF mutants: identification
of NGF P61SR100E as a “painless” lead investiga-
tional candidate for therapeutic applications. PLoS
One. 2015;10(9):e0136425.
47. Ro LS, et  al. Effect of NGF and anti-NGF
on neuropathic pain in rats following chronic
constriction injury of the sciatic nerve. Pain.
1999;79(2–3):265–74.
48. Sevcik MA, et  al. Anti-NGF therapy profoundly
reduces bone cancer pain and the accompanying
increase in markers of peripheral and central sensiti-
zation. Pain. 2005;115(1–2):128–41.
49. Dietz BW, et  al. Targeting nerve growth fac-
tor for pain management in osteoarthritis-clinical
efficacy and safety. Rheum Dis Clin North Am.
2021;47:181–95.
50. Doshi TL, Tesarz J, Cohen SP. Nerve growth factor
inhibitors for low back pain: balancing the risks and
benefits. Pain. 2020;161(9):1941–2.
51. Markman JD, et  al. Tanezumab for chronic low
back pain: a randomized, double-blind, placebo-
146
and active-controlled, phase 3 study of efficacy and
safety. Pain. 2020;161(9):2068–78.
52. Patel MK, Kaye AD, Urman RD. Tanezumab: therapy
targeting nerve growth factor in pain pathogenesis. J
Anaesthesiol Clin Pharmacol. 2018;34(1):111–6.
53. Bramson C, et al. Exploring the role of tanezumab as
a novel treatment for the relief of neuropathic pain.
Pain Med. 2015;16(6):1163–76.
54. Nitta A, et  al. Diabetic neuropathies in brain are
induced by deficiency of BDNF.  Neurotoxicol
Teratol. 2002;24(5):695–701.
55. Kim OY, Song J.  The importance of BDNF and
RAGE in diabetes-induced dementia. Pharmacol
Res. 2020;160:105083.
56. Afarid M, Namvar E, Sanie-Jahromi F.  Diabetic
retinopathy and BDNF: a review on its molecu-
lar basis and clinical applications. J Ophthalmol.
2020;2020:1602739.
57. Richner M, et al. Peripheral nerve injury modulates
neurotrophin signaling in the peripheral and central
nervous system. Mol Neurobiol. 2014;50(3):945–70.
58. Garcia N, et  al. Localization of brain-derived neu-
rotrophic factor, neurotrophin-4, tropomyosin-
related kinase b receptor, and p75NTR receptor by
high-­resolution immunohistochemistry on the adult
mouse neuromuscular junction. J Peripher Nerv
Syst. 2010;15(1):40–9.
59. Garraway SM, Petruska JC, Mendell LM.  BDNF
sensitizes the response of lamina II neurons to high
threshold primary afferent inputs. Eur J Neurosci.
2003;18(9):2467–76.
60. Kerr BJ, et  al. Brain-derived neurotrophic factor
modulates nociceptive sensory inputs and NMDA-­
evoked responses in the rat spinal cord. J Neurosci.
1999;19(12):5138–48.
61. Lever IJ, et  al. Brain-derived neurotrophic fac-
tor is released in the dorsal horn by distinctive
patterns of afferent fiber stimulation. J Neurosci.
2001;21(12):4469–77.
62. Slack SE, et  al. Brain-derived neurotrophic factor
induces NMDA receptor subunit one phosphoryla-
tion via ERK and PKC in the rat spinal cord. Eur J
Neurosci. 2004;20(7):1769–78.
63. Copray S, et  al. Contraction-induced muscle
fiber damage is increased in soleus muscle of
streptozotocin-­diabetic rats and is associated with
elevated expression of brain-derived neurotrophic
factor mRNA in muscle fibers and activated satellite
cells. Exp Neurol. 2000;161(2):597–608.
64. Mizisin AP, et  al. BDNF attenuates functional and
structural disorders in nerves of galactose-fed rats.
J Neuropathol Exp Neurol. 1997;56(12):1290–301.
65. Fernyhough P, Maeda K, Tomlinson DR.  Brain-­
derived neurotrophic factor mRNA levels are
up-regulated in hindlimb skeletal muscle of dia-
betic rats: effect of denervation. Exp Neurol.
1996;141(2):297–303.
66. Li RL, et al. Angelica injection promotes peripheral
nerve structure and function recovery with increased
expressions of nerve growth factor and brain derived
N. A. Calcutt
neurotrophic factor in diabetic rats. Curr Neurovasc
Res. 2010;7(3):213–22.
67. Wang Q, et al. Danhong injection alleviates mechan-
ical allodynia via inhibiting ERK1/2 activation
and elevates BDNF level in sciatic nerve in dia-
betic rat. Evid Based Complement Alternat Med.
2018;2018:5798453.
68. Zhou GJ, et  al. Ameliorative effect of berberine
on neonatally induced type 2 diabetic neuropathy
via modulation of BDNF, IGF-1, PPAR-gamma,
and AMPK expressions. Dose Response.
2019;17(3):1559325819862449.
69. Zhang CH, et  al. The Akt/mTOR cascade medi-
ates high glucose-induced reductions in BDNF via
DNMT1 in Schwann cells in diabetic peripheral neu-
ropathy. Exp Cell Res. 2019;383(1):111502.
70. Rodriguez-Pena A, et  al. Expression of neu-
rotrophins and their receptors in sciatic nerve
of experimentally diabetic rats. Neurosci Lett.
1995;200(1):37–40.
71. Mizisin AP, et al. Decreased accumulation of endog-
enous brain-derived neurotrophic factor against
constricting sciatic nerve ligatures in streptozotocin-­
diabetic and galactose-fed rats. Neurosci Lett.
1999;263(2–3):149–52.
72. Calcutt NA, Freshwater JD, Mizisin AP. Prevention
of sensory disorders in diabetic Sprague-Dawley
rats by aldose reductase inhibition or treatment
with ciliary neurotrophic factor. Diabetologia.
2004;47(4):718–24.
73. Wellmer A, et  al. A double-blind placebo-
controlled clinical trial of recombinant human
brain-derived neurotrophic factor (rhBDNF) in
diabetic polyneuropathy. J Peripher Nerv Syst.
2001;6(4):204–10.
74. Ge HX, et al. Dihydromyricetin affects BDNF lev-
els in the nervous system in rats with comorbid
diabetic neuropathic pain and depression. Sci Rep.
2019;9:14619.
75. Maekawa T, et al. Synthesis and biological activity
of novel 5-(omega-aryloxyalkyl)oxazole derivatives
as brain-derived neurotrophic factor inducers. Chem
Pharm Bull. 2003;51(5):565–73.
76. Fobian K, et  al. Peptides derived from the
solvent-­
exposed loops 3 and 4 of BDNF bind
TrkB and p75(NTR) receptors and stimulate
neurite outgrowth and survival. J Neurosci Res.
2010;88(6):1170–81.
77. Yang SH, et  al. Self-assembling peptide hydrogels
functionalized with LN- and BDNF- mimicking
epitopes synergistically enhance peripheral nerve
regeneration. Theranostics. 2020;10(18):8227–49.
78. Lee-Kubli CAG, Calcutt NA. Altered rate-­dependent
depression of the spinal H-reflex as an indicator of
spinal disinhibition in models of neuropathic pain.
Pain. 2014;155(2):250–60.
79. Morgado C, et  al. Minocycline completely
reverses mechanical hyperalgesia in diabetic rats
through microglia-induced changes in the expres-
sion of the potassium chloride co-transporter 2
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
(KCC2) at the spinal cord. Diabetes Obes Metab.
2011;13(2):150–9.
80. Ferrini F, De Koninck Y. Microglia control neuronal
network excitability via BDNF signalling. Neural
Plast. 2013;2013:429815.
81. Tsuda M.  Microglia-mediated regulation of neu-
ropathic pain: molecular and cellular mechanisms.
Biol Pharm Bull. 2019;42(12):1959–68.
82. Watkins LR, Maier SF. When good pain turns bad.
Curr Dir Psychol Sci. 2003;12(6):232–6.
83. Malcangio M, Montague K. Neuroimmune commu-
nication in chemotherapy-induced neuropathic pain.
Glia. 2017;65:E96–7.
84. Cho HJ, et al. Changes in brain-derived neurotrophic
factor immunoreactivity in rat dorsal root ganglia,
spinal cord, and gracile nuclei following cut or crush
injuries. Exp Neurol. 1998;154(1):224–30.
85. Mannion RJ, et  al. Neurotrophins: peripherally
and centrally acting modulators of tactile stimulus-­
induced inflammatory pain hypersensitivity. Proc
Natl Acad Sci U S A. 1999;96(16):9385–90.
86. Lee-Kubli C, et al. The H-reflex as a biomarker for
spinal disinhibition in painful diabetic neuropathy.
Curr Diab Rep. 2018;18(1):1.
87. Miao B, et al. The implication of transient receptor
potential canonical 6 in BDNF-induced mechanical
allodynia in rat model of diabetic neuropathic pain.
Life Sci. 2021;273:119308.
88. Daulhac L, et al. Phosphorylation of spinal N-methyl-­
D-aspartate receptor NR1 subunits by extracellular
signal-regulated kinase in dorsal horn neurons and
microglia contributes to diabetes-induced painful
neuropathy. Eur J Pain. 2011;15(2):169.e1–169.e12.
89. Rivera C, et  al. BDNF-induced TrkB activation
down-regulates the K+-Cl- cotransporter KCC2
and impairs neuronal Cl- extrusion. J Cell Biol.
2002;159(5):747–52.
90. Rivera C, Voipio J, Kaila K.  Two developmental
switches in GABAergic signalling: the K+-Cl-
cotransporter KCC2 and carbonic anhydrase
CAVII. J Physiol. 2005;562(1):27–36.
91. Miletic G, Miletic V.  Loose ligation of the sciatic
nerve is associated with TrkB receptor-dependent
decreases in KCC2 protein levels in the ipsilateral
spinal dorsal horn. Pain. 2008;137(3):532–9.
92. Jolivalt CG, et  al. Allodynia and hyperalgesia in
diabetic rats are mediated by GABA and depletion
of spinal potassium-chloride co-transporters. Pain.
2008;140(1):48–57.
93. Morgado C, Pinto-Ribeiro F, Tavares I.  Diabetes
affects the expression of GABA and potassium
chloride cotransporter in the spinal cord: a study
in streptozotocin diabetic rats. Neurosci Lett.
2008;438(1):102–6.
94. Boulenguez P, et  al. Down-regulation of the
potassium-­ chloride cotransporter KCC2 contrib-
utes to spasticity after spinal cord injury. Nat Med.
2010;16(3):302–7.
95. Hernandez-Reyes JE, et  al. alpha(5)GABA(A)
receptors play a pronociceptive role and avoid the
147
rate-dependent depression of the Hoffmann reflex in
diabetic neuropathic pain and reduce primary affer-
ent excitability. Pain. 2019;160(6):1448–58.
96. Lee-Kubli CA, et  al. Pharmacological modulation
of rate-dependent depression of the spinal H-reflex
predicts therapeutic efficacy against painful diabetic
neuropathy. Diagnostics. 2021;11(2):283.
97. Marshall AG, et  al. Spinal disinhibition in experi-
mental and clinical painful diabetic neuropathy.
Diabetes. 2017;66(5):1380–90.
98. Worthington A, et al. Spinal inhibitory dysfunction
in patients with painful or painless diabetic neuropa-
thy. Diabetes Care. 2021;44(8):1835–41.
99. Barbacid M. The Trk family of neurotrophin recep-
tors. J Neurobiol. 1994;25(11):1386–403.
100. Lindsay RM.  Role of neurotrophins and trk recep-
tors in the development and maintenance of sensory
neurons: an overview. Philos Trans R Soc Lond B
Biol Sci. 1996;351(1338):365–73.
101. Lewis MA, et  al. Identification and character-
ization of compounds that potentiate NT-3-­
mediated Trk receptor activity. Mol Pharmacol.
2006;69(4):1396–404.
102. Distefano PS, et  al. The neurotrophins Bdnf, Nt-3,
and Ngf display distinct patterns of retrograde
axonal-transport in peripheral and central neurons.
Neuron. 1992;8(5):983–93.
103. Altar CA, DiStefano PS.  Neurotrophin traffick-
ing by anterograde transport. Trends Neurosci.
1998;21(10):433–7.
104. von Bartheld CS, Wang XX, Butowt R. Anterograde
axonal transport, transcytosis, and recycling of
neurotrophic factors—the concept of trophic
currencies in neural networks. Mol Neurobiol.
2001;24(1–3):1–28.
105. Naito Y, Lee AK, Takahashi H. Emerging roles of the
neurotrophin receptor TrkC in synapse organization.
Neurosci Res. 2017;116:10–7.
106. Boyce VS, Mendell LM.  Neurotrophins and spinal
circuit function. Front Neural Circuits. 2014;8:59.
107. Andreassen CS, et  al. Expression of neurotrophic
factors in diabetic muscle-relation to neuropathy and
muscle strength. Brain. 2009;132:2724–33.
108. Fernyhough P, Diemel LT, Tomlinson DR.  Target
tissue production and axonal transport of neuro-
trophin-­3 are reduced in streptozotocin-diabetic rats.
Diabetologia. 1998;41(3):300–6.
109. Ihara C, et al. Decreased neurotrophin-3 expression
in skeletal muscles of streptozotocin-induced dia-
betic rats. Neuropeptides. 1996;30(4):309–12.
110. Cai F, Tomlinson DR, Fernyhough P.  Elevated
expression of neurotrophin-3 mRNA in sensory
nerve of streptozotocin-diabetic rats. Neurosci Lett.
1999;263(2–3):81–4.
111. Kennedy AJ, et  al. Neurotrophin-3 is increased
in skin in human diabetic neuropathy. J Neurol
Neurosurg Psychiatry. 1998;65(3):393–5.
112. Terenghi G, et  al. trkA and trkC expression is
increased in human diabetic skin. Neurosci Lett.
1997;228(1):33–6.
148
113. Mizisin AP, et al. Neurotrophin-3 reverses nerve con-
duction velocity deficits in streptozotocin-diabetic
rats. J Peripher Nerv Syst. 1999;4(3–4):211–21.
114. Mizisin AP, et  al. NT-3 attenuates functional and
structural disorders in sensory nerves of galactose-fed
rats. J Neuropathol Exp Neurol. 1998;57(9):803–13.
115. Middlemas A, et  al. Enhanced activation of axo-
nally transported stress-activated protein kinases in
peripheral nerve in diabetic neuropathy is prevented
by neurotrophin-3. Brain. 2003;126:1671–82.
116. Schmidt RE, et  al. Effect of NGF and neuro-
trophin-­3 treatment on experimental diabetic auto-
nomic neuropathy. J Neuropathol Exp Neurol.
2001;60(3):263–73.
117. Li HW, et al. Muscle NT-3 levels increased by exer-
cise training contribute to the improvement in caudal
nerve conduction velocity in diabetic rats. Mol Med
Rep. 2012;6(1):69–74.
118. Christianson JA, et  al. Neurotrophic modulation
of myelinated cutaneous innervation and mechani-
cal sensory loss in diabetic mice. Neuroscience.
2007;145(1):303–13.
119. Huang TJ, et  al. Diabetes-induced alterations
in calcium homeostasis in sensory neurones of
streptozotocin-­ diabetic rats are restricted to lum-
bar ganglia and are prevented by neurotrophin-3.
Diabetologia. 2002;45(4):560–70.
120. Huang TJ, et  al. Neurotrophin-3 prevents mito-
chondrial dysfunction in sensory neurons
of streptozotocin-­ diabetic rats. Exp Neurol.
2005;194(1):279–83.
121. Sayers NM, et al. Neurotrophin-3 prevents the proxi-
mal accumulation of neurofilament proteins in sen-
sory neurons of streptozocin-induced diabetic rats.
Diabetes. 2003;52(9):2372–80.
122. Sariola H, Saarma M.  Novel functions and
signalling pathways for GDNF.  J Cell Sci.
2003;116(19):3855–62.
123. Paratcha G, Ledda F, Ibanez CF.  The neural cell
adhesion molecule NCAM is an alternative sig-
naling receptor for GDNF family ligands. Cell.
2003;113(7):867–79.
124. Donnelly CR, Pierchala BA.  Plasma membrane
localization of the GFL receptor components:
a nexus for receptor crosstalk. Cell Tissue Res.
2020;382(1):57–64.
125. Stanga S, Boido M, Kienlen-Campard P.  How to
build and to protect the neuromuscular junction: the
role of the glial cell line-derived neurotrophic factor.
Int J Mol Sci. 2021;22(1):136.
126. Zahavi EE, et  al. A compartmentalized micro-
fluidic neuromuscular co-culture system reveals
spatial aspects of GDNF functions. J Cell Sci.
2015;128(6):1241–52.
127. Enomoto H, et  al. RET signaling is essential for
migration, axonal growth and axon guidance of
developing sympathetic neurons. Development.
2001;128(20):3963–74.
128. Luo WQ, et  al. Molecular identification of rap-
idly adapting mechanoreceptors and their devel-
N. A. Calcutt
opmental dependence on ret signaling. Neuron.
2009;64(6):841–56.
129. Luo WQ, et  al. A hierarchical NGF signaling cas-
cade controls Ret-dependent and Ret-independent
events during development of nonpeptidergic DRG
neurons. Neuron. 2007;54(5):739–54.
130. Molliver DC, et al. IB4-binding DRG neurons switch
from NGF to GDNF dependence in early postnatal
life. Neuron. 1997;19(4):849–61.
131. Bar KJ, et al. GDNF and its receptor component Ret
in injured human nerves and dorsal root ganglia.
Neuroreport. 1998;9(1):43–7.
132. Hoke A, et  al. A decline in glial cell-line-derived
neurotrophic factor expression is associated with
impaired regeneration after long-term Schwann cell
denervation. Exp Neurol. 2002;173(1):77–85.
133. Christianson JA, Riekhof JT, Wright DE. Restorative
effects of neurotrophin treatment on diabetes-­
induced cutaneous axon loss in mice. Exp Neurol.
2003;179(2):188–99.
134. Saudek F, et  al. Preserved expression of skin neu-
rotrophic factors in advanced diabetic neuropathy
does not lead to neural regeneration despite pan-
creas and kidney transplantation. J Diabetes Res.
2018;2018:2309108.
135. Liu GS, et  al. Peripheral gene transfer of glial
cell-derived neurotrophic factor ameliorates neu-
ropathic deficits in diabetic rats. Hum Gene Ther.
2009;20(7):715–27.
136. Akkina SK, Patterson CL, Wright DE.  GDNF res-
cues nonpeptidergic unmyelinated primary afferents
in streptozotocin-treated diabetic mice. Exp Neurol.
2001;167(1):173–82.
137. Christianson JA, et  al. Beneficial actions of neuro-
trophin treatment on diabetes-induced hypoalgesia
in mice. J Pain. 2003;4(9):493–504.
138. Viisanen H, et al. Novel RET agonist for the treat-
ment of experimental neuropathies. Mol Pain.
2020;16:1744806920950866.
139. Hedstrom KL, et  al. Treating small fiber neuropa-
thy by topical application of a small molecule
modulator of ligand-induced GFR alpha/RET
receptor signaling. Proc Natl Acad Sci U S A.
2014;111(6):2325–30.
140. Boulton TG, Stahl N, Yancopoulos GD.  Ciliary
neurotrophic factor leukemia inhibitory fac-
tor interleukin-6 oncostatin-M family of cyto-
kines induces tyrosine phosphorylation of a
common set of proteins overlapping those induced
by other cytokines and growth-factors. J Biol Chem.
1994;269(15):11648–55.
141. Pasquin S, Sharma M, Gauchat JF.  Cytokines of
the LIF/CNTF family and metabolism. Cytokine.
2016;82:122–4.
142. Gloaguen I, et  al. Ciliary neurotrophic factor cor-
rects obesity and diabetes associated with leptin
deficiency and resistance. Proc Natl Acad Sci U S A.
1997;94(12):6456–61.
143. Davis S, et al. The receptor for ciliary neurotrophic
factor. Science. 1991;253(5015):59–63.
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
144. Schuster B, et  al. Signaling of human ciliary neu-
rotrophic factor (CNTF) revisited—the interleukin-6
receptor can serve as an alpha-receptor for CNTF. J
Biol Chem. 2003;278(11):9528–35.
145. Dobrea GM, Unnerstall JR, Rao MS. The expression
of Cntf message and immunoreactivity in the central
and peripheral nervous-system of the rat. Dev Brain
Res. 1992;66(2):209–19.
146. Rende M, et al. Immunolocalization of ciliary neu-
ronotrophic factor in adult-rat sciatic-nerve. Glia.
1992;5(1):25–32.
147. Lee DA, Zurawel RH, Windebank AJ. Ciliary neuro-
trophic factor expression in Schwann-cells is induced
by axonal contact. J Neurochem. 1995;65(2):564–8.
148. Sendtner M, et al. Brain-derived neurotrophic factor
prevents the death of motoneurons in newborn rats
after nerve-section. Nature. 1992;360(6406):757–9.
149. Smith GM, et  al. Temporal and spatial expression
of ciliary neurotrophic factor after peripheral-nerve
injury. Exp Neurol. 1993;121(2):239–47.
150. Gravel C, et  al. Adenoviral gene transfer of ciliary
neurotrophic factor and brain-derived neurotrophic
factor leads to long-term survival of axotomized
motor neurons. Nat Med. 1997;3(7):765–70.
151. Sendtner M, et  al. Effect of ciliary neurotrophic
factor (Cntf) on motoneuron survival. J Cell Sci.
1991;15:103–9.
152. Sendtner M, et al. Endogenous ciliary neurotrophic
factor is a lesion factor for axotomized motoneurons
in adult mice. J Neurosci. 1997;17(18):6999–7006.
153. Friedman B, et al. Regulation of ciliary neurotrophic
factor expression in myelin-related Schwann-cells
invivo. Neuron. 1992;9(2):295–305.
154. Godinho MJ, et al. Regeneration of adult rat sensory
and motor neuron axons through chimeric peroneal
nerve grafts containing donor Schwann cells engi-
neered to express different neurotrophic factors. Exp
Neurol. 2020;330:113355.
155. Porzionato A, et al. Development of oxidized poly-
vinyl alcohol-based nerve conduits coupled with
the ciliary neurotrophic factor. Materials (Basel).
2019;12(12):1996.
156. Masu Y, et  al. Disruption of the Cntf gene
results in motor-neuron degeneration. Nature.
1993;365(6441):27–32.
157. Gatzinsky KP, et  al. Early onset of degenerative
changes at nodes of Ranvier in alpha-motor axons of
Cntf null (-/-) mutant mice. Glia. 2003;42(4):340–9.
158. Calcutt NA, et  al. Reduced ciliary neuronotro-
phic factor-like activity in nerves from diabetic or
galactose-­fed rats. Brain Res. 1992;575(2):320–4.
159. Mizisin AP, et  al. Aldose reductase inhibition
increases CNTF-like bioactivity and protein in sci-
atic nerves from galactose-fed and normal rats.
Diabetes. 1997;46(4):647–52.
160. Nakamura H, et  al. Olmesartan medoxomil ame-
liorates sciatic nerve regeneration in diabetic rats.
Neuroreport. 2009;20(16):1481–5.
161. Jiang Y, et al. Novel sites of aldose reductase immu-
nolocalization in normal and streptozotocin-diabetic
rats. J Peripher Nerv Syst. 2006;11(4):274–85.
149
162. Canclini L, et  al. Association of microtubules
and axonal RNA transferred from myelinat-
ing Schwann cells in rat sciatic nerve. PLoS One.
2020;15(5):e0233651.
163. Court FA, et  al. Schwann cell to axon transfer of
ribosomes: toward a novel understanding of the
role of glia in the nervous system. J Neurosci.
2008;28(43):11024–9.
164. Court FA, et  al. Axonal autonomous and glial-­
dependent processes participate in axonal degenera-
tion. Glia. 2011;59:S73.
165. Duncan A, et al. Transfer of horseradish peroxidase
from oligodendrocyte to axon in the myelinating
neonatal rat optic nerve: artefact or transcellular
exchange? Glia. 1996;17(4):349–55.
166. Lasek RJ, Gainer H, Barker J. Transfer of newly syn-
thesized proteins from Schwann-cells to squid giant-­
axon. Anat Rec. 1974;178(2):398.
167. Lopez-Leal R, Diaz P, Court FA.  In vitro analysis
of the role of Schwann cells on axonal degeneration
and regeneration using sensory neurons from dorsal
root ganglia. Methods Mol Biol. 2018;1739:255–67.
168. Shakhbazau A, et  al. Demyelination induces trans-
port of ribosome-containing vesicles from glia to
axons: evidence from animal models and MS patient
brains. Mol Biol Rep. 2016;43(6):495–507.
169. Sotelo JR, et  al. Glia to axon RNA transfer. Dev
Neurobiol. 2014;74(3):292–302.
170. Mizisin AP, et  al. Ciliary neurotrophic fac-
tor improves nerve conduction and ameliorates
regeneration deficits in diabetic rats. Diabetes.
2004;53(7):1807–12.
171. Gao N, et al. Dendritic cell dysfunction and diabetic
sensory neuropathy in the cornea. J Clin Investig.
2016;126(5):1998–2011.
172. Ma MM, et  al. Involvement of ciliary neuro-
trophic factor in early diabetic retinal neuropa-
thy in streptozotocin-induced diabetic rats. Eye.
2018;32(9):1463–71.
173. Saleh A, et  al. Ciliary neurotrophic factor acti-
vates NF-kappa B to enhance mitochondrial bio-
energetics and prevent neuropathy in sensory
neurons of streptozotocin-induced diabetic rodents.
Neuropharmacology. 2013;65:65–73.
174. Chowdhury SR, et  al. Ciliary neurotrophic fac-
tor reverses aberrant mitochondrial bioenergetics
through the JAK/STAT pathway in cultured sensory
neurons derived from streptozotocin-induced dia-
betic rodents. Cell Mol Neurobiol. 2014;34(5):643–9.
175. Lotz B, et al. A double-blind placebo-controlled clin-
ical trial of subcutaneous recombinant human ciliary
neurotrophic factor (rHCNTF) in amyotrophic lat-
eral sclerosis. Neurology. 1996;46(5):1244–9.
176. Ettinger MP, et  al. Recombinant variant of cili-
ary neurotrophic factor for weight loss in obese
adults—a randomized, dose-ranging study. JAMA.
2003;289(14):1826–32.
177. Chew EY, et al. Effect of ciliary neurotrophic factor
on retinal neurodegeneration in patients with macu-
lar telangiectasia type 2 a randomized clinical trial.
Ophthalmology. 2019;126(4):540–9.
150
178. Kauper K, et  al. Two-year intraocular delivery of
ciliary neurotrophic factor by encapsulated cell
technology implants in patients with chronic retinal
degenerative diseases. Invest Ophthalmol Vis Sci.
2012;53(12):7484–91.
179. Rhode SC, Beier JP, Ruhl T.  Adipose tissue stem
cells in peripheral nerve regeneration–in vitro and
in vivo. J Neurosci Res. 2021;99(2):545–60.
180. Lara-Ramirez R, et  al. Expression of interleukin-6
receptor alpha in normal and injured rat sciatic
nerve. Neuroscience. 2008;152(3):601–8.
181. Rothaug M, Becker-Pauly C, Rose-John S. The role
of interleukin-6 signaling in nervous tissue. BBA-­
Mol Cell Res. 2016;1863(6):1218–27.
182. Bolin LM, et  al. Interleukin-6 production by
Schwann-cells and induction in sciatic-nerve injury.
J Neurochem. 1995;64(2):850–8.
183. Gaudet AD, Popovich PG, Ramer MS.  Wallerian
degeneration: gaining perspective on inflam-
matory events after peripheral nerve injury. J
Neuroinflammation. 2011;8:110.
184. Hirota H, et  al. Accelerated nerve regeneration
in mice by upregulated expression of interleukin
(IL) 6 and IL-6 receptor after trauma. J Exp Med.
1996;183(6):2627–34.
185. Andriambeloson E, et al. Interleukin-6 attenuates the
development of experimental diabetes-related neu-
ropathy. Neuropathology. 2006;26(1):32–42.
186. Cameron NE, Cotter MA. The neurocytokine, inter-
leukin-­6, corrects nerve dysfunction in experimental
diabetes. Exp Neurol. 2007;207(1):23–9.
187. Cotter MA, et al. Effects of interleukin-6 treatment
on neurovascular function, nerve perfusion and vas-
cular endothelium in diabetic rats. Diabetes Obes
Metab. 2010;12(8):689–99.
188. Cox AA, et al. Low-dose pulsatile interleukin-6 as a
treatment option for diabetic peripheral neuropathy.
Front Endocrinol. 2017;8:89.
189. Bishnoi M, et al. Streptozotocin-induced early ther-
mal hyperalgesia is independent of glycemic state
of rats: role of transient receptor potential vanilloid
1 (TRPV1) and inflammatory mediators. Mol Pain.
2011;7:52.
190. Brini AT, et al. Therapeutic effect of human adipose-­
derived stem cells and their secretome in experimen-
tal diabetic pain. Sci Rep. 2017;7(1):9904.
191. Toledo-Corral CM, Banner LR.  Early changes of
LIFR and gp130 in sciatic nerve and muscle of dia-
betic mice. Acta Histochem. 2012;114(2):159–65.
192. Niemi JP, et al. Injury-induced gp130 cytokine sig-
naling in peripheral ganglia is reduced in diabetes
mellitus. Exp Neurol. 2017;296:1–15.
193. Muona P, et  al. Glucose transporters of rat
peripheral-­nerve—differential expression of Glut1
gene by Schwann-cells and perineurial cells in vivo
and in vitro. Diabetes. 1992;41(12):1587–96.
194. Magnani P, et al. Glucose transporters in rat periph-
eral nerve: paranodal expression of GLUT1 and
GLUT3. Metabolism. 1996;45(12):1466–73.
N. A. Calcutt
195. Yagihashi S.  Glucotoxic mechanisms and related
therapeutic approaches. Int Rev Neurobiol.
2016;127:121–49.
196. Niimi N, et al. Aldose reductase and the polyol path-
way in Schwann cells: old and new problems. Int J
Mol Sci. 2021;22(3):1031.
197. Tomlinson DR, Gardiner NJ.  Diabetic neuropa-
thies: components of etiology. J Peripher Nerv Syst.
2008;13(2):112–21.
198. Jolivalt CG, et al. Defective insulin signaling path-
way and increased glycogen synthase kinase-3
activity in the brain of diabetic mice: parallels with
Alzheimer’s disease and correction by insulin. J
Neurosci Res. 2008;86(15):3265–74.
199. King MR, et  al. Activation of the insulin-signaling
pathway in sciatic nerve and hippocampus of type 1
diabetic rats. Neuroscience. 2015;303:220–8.
200. Arvanitakis Z, et  al. Brain insulin signaling,
Alzheimer disease pathology, and cognitive func-
tion. Ann Neurol. 2020;88(3):513–25.
201. Ferreira ST.  Brain insulin, insulin-like growth
factor 1 and glucagon-like peptide 1 signal-
ling in Alzheimer’s disease. J Neuroendocrinol.
2021;33(4):e12959.
202. Rebelos E, et al. Brain glucose metabolism in health,
obesity, and cognitive decline-does insulin have any-
thing to do with it? A narrative review. J Clin Med.
2021;10(7):1532.
203. Sugimoto K, et al. Insulin receptor in rat peripheral
nerve: its localization and alternatively spliced iso-
forms. Diabetes Metab Res Rev. 2000;16(5):354–63.
204. Brussee V, Cunningham FA, Zochodne DW. Direct
insulin signaling of neurons reverses diabetic neu-
ropathy. Diabetes. 2004;53(7):1824–30.
205. Sugimoto K, Murakawa Y, Sima AA.  Expression
and localization of insulin receptor in rat dorsal
root ganglion and spinal cord. J Peripher Nerv Syst.
2002;7(1):44–53.
206. Hackett AR, Strickland A, Milbrandt J.  Disrupting
insulin signaling in Schwann cells impairs myelin-
ation and induces a sensory neuropathy. Glia.
2020;68(5):963–78.
207. Huang TJ, Verkhratsky A, Fernyhough P.  Insulin
enhances mitochondrial inner membrane potential
and increases ATP levels through phosphoinosit-
ide 3-kinase in adult sensory neurons. Mol Cell
Neurosci. 2005;28(1):42–54.
208. Fernyhough P, et al. Insulin and insulin-like growth
factor-I enhance regeneration in cultured adult-rat
sensory neurons. Brain Res. 1993;607(1–2):117–24.
209. Grote CW, et al. Deletion of the insulin receptor in
sensory neurons increases pancreatic insulin levels.
Exp Neurol. 2018;305:97–107.
210. Grote CW, Wright DE. A role for insulin in diabetic
neuropathy. Front Neurosci. 2016;10:581.
211. Kobayashi M, Zochodne DW.  Diabetic neuropathy
and the sensory neuron: new aspects of pathogenesis
and their treatment implications. J Diabetes Investig.
2018;9(6):1239–54.
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
212. Kamiya H, et al. Unmyelinated fiber sensory neurop-
athy differs in type 1 and type 2 diabetes. Diabetes
Metab Res Rev. 2005;21(5):448–58.
213. Pierson CR, et  al. Insulin deficiency rather than
hyperglycemia accounts for impaired neurotrophic
responses and nerve fiber regeneration in type 1
diabetic neuropathy. J Neuropathol Exp Neurol.
2003;62(3):260–71.
214. Sima AAF.  The heterogeneity of diabetic neuropa-
thy. Front Biosci. 2008;13:4808–16.
215. Stino AM, et  al. Evolving concepts on the role
of dyslipidemia, bioenergetics, and inflamma-
tion in the pathogenesis and treatment of dia-
betic peripheral neuropathy. J Peripher Nerv Syst.
2020;25(2):76–84.
216. Grote CW, et  al. Peripheral nervous system insu-
lin resistance in ob/ob mice. Acta Neuropathol
Commun. 2013;1:15.
217. Grote CW, et  al. The role of insulin receptor sub-
strate 2  in neuronal insulin resistance in diabetic
neuropathy. J Peripher Nerv Syst. 2011;16:S49.
218. Aghanoori MR, et al. Insulin prevents aberrant mito-
chondrial phenotype in sensory neurons of type 1
diabetic rats. Exp Neurol. 2017;297:148–57.
219. Chowdhury SKR, et  al. Mitochondrial respi-
ratory chain dysfunction in dorsal root gan-
glia of streptozotocin-­ induced diabetic rats and
its correction by insulin treatment. Diabetes.
2010;59(4):1082–91.
220. Huang TJ, et al. Insulin prevents depolarization of the
mitochondrial inner membrane in sensory neurons
of type 1 diabetic rats in the presence of sustained
hyperglycemia. Diabetes. 2003;52(8):2129–36.
221. de la Hoz CL, et  al. A model of chronic diabetic
polyneuropathy: benefits from intranasal insulin are
modified by sex and RAGE deletion. Am J Physiol
Endocrinol Metab. 2017;312(5):E407–19.
222. Francis GJ, et al. Motor end plate innervation loss in
diabetes and the role of insulin. J Neuropathol Exp
Neurol. 2011;70(5):323–39.
223. Guo GF, et al. Local insulin and the rapid regrowth
of diabetic epidermal axons. Neurobiol Dis.
2011;43(2):414–21.
224. Singhal A, et  al. Near nerve local insulin prevents
conduction slowing in experimental diabetes. Brain
Res. 1997;763(2):209–14.
225. Toth C, Brussee V, Zochodne DW.  Remote neuro-
trophic support of epidermal nerve fibres in experi-
mental diabetes. Diabetologia. 2006;49(5):1081–8.
226. Wahren J.  C-peptide and the pathophysiology of
microvascular complications of diabetes. J Intern
Med. 2017;281(1):3–6.
227. Yosten GLC, Kolar GR. The physiology of proinsu-
lin C-peptide: unanswered questions and a proposed
model. Physiology. 2015;30(4):327–32.
228. Hills CE, Brunskill NJ.  Cellular and physi-
ological effects of C-peptide. Clin Sci.
2009;116(7–8):565–74.
229. Yosten GLC, et  al. Evidence for an interaction
between proinsulin C-peptide and GPR146. J
Endocrinol. 2013;218(2):B1–8.
151
230. Lindfors L, et  al. Is GPR146 really the receptor
for proinsulin C-peptide? Bioorg Med Chem Lett.
2020;30(13):127208.
231. Sima AAF, Kamiya H. Is C-peptide replacement the
missing link for successful treatment of neurological
complications in type 1 diabetes? Curr Drug Targets.
2008;9(1):37–46.
232. Wahren J, Kallas A, Sima AAF. The clinical poten-
tial of C-peptide replacement in type 1 diabetes.
Diabetes. 2012;61(4):761–72.
233. Jolivalt CG, et  al. Efficacy of a long-acting
C-peptide analogue against peripheral neuropathy in
streptozotocin-diabetic mice. Diabetes Obes Metab.
2015;17(8):781–8.
234. Ekberg K, et al. Amelioration of sensory nerve dys-
function by C-peptide in patients with type 1 diabe-
tes. Diabetes. 2003;52(2):536–41.
235. Ekberg K, et al. C-peptide replacement therapy and
sensory nerve function in type 1 diabetic neuropathy.
Diabetes Care. 2007;30(1):71–6.
236. Johansson BL, et al. Beneficial effects of C-peptide
on incipient nephropathy and neuropathy in
patients with Type 1 diabetes mellitus. Diabet Med.
2000;17(3):181–9.
237. Wahren J, et  al. Long-acting C-peptide and neu-
ropathy in Type 1 diabetes: a 12-month clinical trial.
Diabetes Care. 2016;39(4):596–602.
238. LeRoith D, Holly JMP, Forbes BE.  Insulin-like
growth factors: ligands, binding proteins, and recep-
tors. Mol Metab. 2021;52:101245.
239. Siddle K.  Molecular basis of signaling specificity
of insulin and IGF receptors: neglected corners and
recent advances. Front Endocrinol. 2012;3:34.
240. Gao WQ, et al. IGF-I deficient mice show reduced
peripheral nerve conduction velocities and decreased
axonal diameters and respond to exogenous IGF-I
treatment. J Neurobiol. 1999;39(1):142–52.
241. Rabinovsky ED.  The multifunctional role of
IGF-1 in peripheral nerve regeneration. Neurol Res.
2004;26(2):204–10.
242. Ekstrom AR, Kanje M, Skottner A.  Nerve regen-
eration and serum levels of insulin-like growth
factor-I in rats with streptozotocin-induced insulin
deficiency. Brain Res. 1989;496(1–2):141–7.
243. Craner MJ, et al. Preferential expression of IGF-I in
small DRG neurons and down-regulation following
injury. Neuroreport. 2002;13(13):1649–52.
244. Kamiya H, et  al. C-Peptide reverses nocicep-
tive neuropathy in type 1 diabetes. Diabetes.
2006;55(12):3581–7.
245. Wuarin L, Guertin DM, Ishii DN.  Early reduction
in insulin-like growth factor gene expression in dia-
betic nerve. Exp Neurol. 1994;130(1):106–14.
246. Schmidt RE, et  al. Analysis of the Zucker
Diabetic Fatty (ZDF) type 2 diabetic rat model
suggests a neurotrophic role for insulin/IGF-I
in diabetic autonomic neuropathy. Am J Pathol.
2003;163(1):21–8.
247. Zhuang HX, et  al. Insulin-like growth factor
(IGF) gene expression is reduced in neural tis-
sues and liver from rats with non-insulin-depen-
152
dent diabetes mellitus, and IGF treatment
ameliorates diabetic neuropathy. J Pharmacol Exp
Ther. 1997;283(1):366–74.
248. Busiguina S, et al. Neurodegeneration is associated
to changes in serum insulin-like growth factors.
Neurobiol Dis. 2000;7(6 Pt B):657–65.
249. Han HJ, Kang CW, Park SH.  Tissue-specific regu-
lation of insulin-like growth factors and insulin-like
growth factor binding proteins in male diabetic rats
in  vivo and in  vitro. Clin Exp Pharmacol Physiol.
2006;33(12):1172–9.
250. Hao W, et al. Hyperglycemia promotes Schwann cell
de-differentiation and de-myelination via sorbitol
accumulation and Igf1 protein down-regulation. J
Biol Chem. 2015;290(28):17106–15.
251. Ishii DN, Lupien SB.  Insulin-like growth factors
protect against diabetic neuropathy: effects on sen-
sory nerve regeneration in rats. J Neurosci Res.
1995;40(1):138–44.
252. Lupien SB, Bluhm EJ, Ishii DN. Systemic insulin-­
like growth factor-I administration prevents cog-
nitive impairment in diabetic rats, and brain IGF
regulates learning/memory in normal adult rats. J
Neurosci Res. 2003;74(4):512–23.
253. Schmidt RE, et  al. Insulin-like growth factor I
reverses experimental diabetic autonomic neuropa-
thy. Am J Pathol. 1999;155(5):1651–60.
254. Schmidt RE, et al. Effect of IGF-I and neurotrophin-3
on gracile neuroaxonal dystrophy in diabetic and
aging rats. Brain Res. 2000;876(1–2):88–94.
255. Zhuang HX, et al. Insulin-like growth factors reverse
or arrest diabetic neuropathy: effects on hyperalgesia
and impaired nerve regeneration in rats. Exp Neurol.
1996;140(2):198–205.
256. Piriz J, Torres-Aleman I, Nunez A. Independent alter-
ations in the central and peripheral somatosensory
pathways in rat diabetic neuropathy. Neuroscience.
2009;160(2):402–11.
257. Wang H, et al. Protection of insulin-like growth fac-
tor 1 on experimental peripheral neuropathy in dia-
betic mice. Mol Med Rep. 2018;18(5):4577–86.
258. Tang ZS, et  al. Peripheral pain is enhanced by
insulin-like growth factor 1 and its receptors in a
mouse model of type 2 diabetes mellitus. J Diabetes.
2019;11(4):309–15.
259. Li JB, et  al. The preventive efficacy of methylco-
balamin on rat peripheral neuropathy influenced
by diabetes via neural IGF-1 levels. Nutr Neurosci.
2010;13(2):79–86.
260. Song W, et al. Jinmaitong, a traditional Chinese com-
pound prescription, ameliorates the streptozocin-­
induced diabetic peripheral neuropathy rats by
increasing sciatic nerve IGF-1 and IGF-1R expres-
sion. Front Pharmacol. 2019;10:255.
261. Wang R, et  al. L-carnitine ameliorates peripheral
neuropathy in diabetic mice with a corresponding
increase in insulin-like growth factor-1 level. Mol
Med Rep. 2019;19(1):743–51.
262. Ekman B, Nystrom F, Arnqvist HJ.  Circulating
IGF-I concentrations are low and not correlated to
N. A. Calcutt
glycaemic control in adults with type 1 diabetes. Eur
J Endocrinol. 2000;143(4):505–10.
263. Tan K, Baxter RC. Serum insulin-like growth factor-
­I levels in adult diabetic-patients—the effect of age.
J Clin Endocrinol Metab. 1986;63(3):651–5.
264. Migdalis IN, et  al. Insulin-like growth-factor-I and
Igf-I receptors in diabetic-patients with neuropathy.
Diabet Med. 1995;12(9):823–7.
265. Crosby SR, et  al. Elevated plasma insulin-like
growth factor binding protein-1 levels in type 1
(insulin-dependent) diabetic patients with peripheral
neuropathy. Diabetologia. 1992;35(9):868–72.
266. Simon CM, et  al. Dysregulated IGFBP5 expres-
sion causes axon degeneration and motoneuron
loss in diabetic neuropathy. Acta Neuropathol.
2015;130(3):373–87.
267. Rabinovsky ED, Draghia-Akli R.  Insulin-like
growth factor I plasmid therapy promotes in  vivo
angiogenesis. Mol Ther. 2004;9(1):46–55.
268. Saboory E, et  al. Exercise and insulin-like growth
factor 1 supplementation improve angiogen-
esis and angiogenic cytokines in a rat model
of diabetes-induced neuropathy. Exp Physiol.
2020;105(5):783–92.
269. Aghanoori MR, et  al. Insulin-like growth fac-
tor-1 activates AMPK to augment mitochondrial
function and correct neuronal metabolism in
sensory neurons in type 1 diabetes. Mol Metab.
2019;20:149–65.
270. Aghanoori MR, et al. Sensory neurons derived from
diabetic rats exhibit deficits in functional glycolysis
and ATP that are ameliorated by IGF-1. Mol Metab.
2021;49:101191.
271. Wilson BD, et  al. Netrins promote develop-
mental and therapeutic angiogenesis. Science.
2006;313(5787):640–4.
272. Yam PT, Charron F.  Signaling mechanisms of
non-conventional axon guidance cues: the Shh,
BMP and Wnt morphogens. Curr Opin Neurobiol.
2013;23(6):965–73.
273. Anjaneyulu M, et  al. Transforming growth factor-­
beta induces cellular injury in experimental diabetic
neuropathy. Exp Neurol. 2008;211(2):469–79.
274. Shimizu F, et  al. Advanced glycation end-products
induce basement membrane hypertrophy in endo-
neurial microvessels and disrupt the blood-nerve
barrier by stimulating the release of TGF-beta and
vascular endothelial growth factor (VEGF) by peri-
cytes. Diabetologia. 2011;54(6):1517–26.
275. Resham K, et  al. Neuroprotective effects of iso-
quercitrin in diabetic neuropathy via Wnt/beta-­
catenin signaling pathway inhibition. Biofactors.
2020;46(3):411–20.
276. Resham K, Sharma SS. Pharmacologic inhibition of
porcupine, disheveled, and beta-catenin in Wnt sig-
naling pathway ameliorates diabetic peripheral neu-
ropathy in rats. J Pain. 2019;20(11):1338–52.
277. Parmantier E, et  al. Schwann cell-derived desert
hedgehog controls the development of peripheral
nerve sheaths. Neuron. 1999;23(4):713–24.
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
278. Moreau N, Boucher Y. Hedging against neuropathic
pain: role of hedgehog signaling in pathological
nerve healing. Int J Mol Sci. 2020;21(23):9115.
279. Chapouly C, et  al. Impaired Hedgehog signalling-­
induced endothelial dysfunction is sufficient
to induce neuropathy: implication in diabetes.
Cardiovasc Res. 2016;109(2):217–27.
280. Calcutt NA, et  al. Therapeutic efficacy of sonic
hedgehog protein in experimental diabetic neuropa-
thy. J Clin Investig. 2003;111(4):507–14.
281. Stanton BZ, Peng LF.  Small-molecule modulators
of the Sonic Hedgehog signaling pathway. Mol
Biosyst. 2010;6(1):44–54.
282. Leavitt E, Lask G, Martin S.  Sonic Hedgehog
pathway inhibition in the treatment of advanced
basal cell carcinoma. Curr Treat Options Oncol.
2019;20(11):84.
283. Li R, et  al. Heparin-poloxamer thermosensitive
hydrogel loaded with bFGF and NGF enhances
peripheral nerve regeneration in diabetic rats.
Biomaterials. 2018;168:24–37.
284. Nakae M, et al. Effects of bFGF on chronic neuropa-
thy in STZ-diabetic rats. Diabetes. 2006;55:A188.
285. Foehring D, Brand-Saberi B, Theiss C.  VEGF-­
induced growth cone enhancement is diminished by
inhibiting tyrosine-residue 1214 of VEGFR-2. Cells
Tissues Organs. 2012;196(3):195–205.
286. Sondell M, Sundler F, Kanje M. Vascular endothelial
growth factor is a neurotrophic factor which stimu-
lates axonal outgrowth through the flk-1 receptor.
Eur J Neurosci. 2000;12(12):4243–54.
287. Bates DO, et al. Physiological role of vascular endo-
thelial growth factors as homeostatic regulators.
Compr Physiol. 2018;8(3):955–79.
288. Arredondo-Garcia VK, et  al. association of the
vascular endothelial growth factor gene polymor-
phism+936 c/t with diabetic neuropathy in patients
with type 2 diabetes mellitus. Arch Med Res.
2019;50(4):181–6.
289. Tavakkoly-Bazzaz J, et  al. VEGF gene polymor-
phism association with diabetic neuropathy. Mol
Biol Rep. 2010;37(7):3625–30.
290. Zhang XH, et  al. Relationship between single
nucleotide polymorphisms in the 3′-untranslated
region of the vascular endothelial growth fac-
tor gene and susceptibility to diabetic periph-
eral neuropathy in China. Arch Med Sci.
2014;10(5):1028–34.
291. Chavez JC, Almhanna K, Berti-Mattera
LN. Transient expression of hypoxia-inducible fac-
tor-­1 alpha and target genes in peripheral nerves from
diabetic rats. Neurosci Lett. 2005;374(3):179–82.
292. Mo FF, et al. Prostaglandin E1 protects the periph-
eral nerve in diabetics through preventing vascu-
lar permeability changes. Exp Clin Endocrinol
Diabetes. 2018;126(2):113–22.
293. Samii A, Unger J, Lange W.  Vascular endothelial
growth factor expression in peripheral nerves and
dorsal root ganglia in diabetic neuropathy in rats.
Neurosci Lett. 1999;262(3):159–62.
153
294. Taiana MM, et  al. Neutralization of Schwann cell-­
secreted VEGF is protective to in vitro and in vivo
experimental diabetic neuropathy. PLoS One.
2014;9(9):e108403.
295. Chattopadhyay M, et al. HSV-mediated gene trans-
fer of vascular endothelial growth factor to dorsal
root ganglia prevents diabetic neuropathy. Gene
Ther. 2005;12(18):1377–84.
296. Murakami T, et al. VEGF 164 gene transfer by elec-
troporation improves diabetic sensory neuropathy in
mice. J Gene Med. 2006;8(6):773–81.
297. Schratzberger P, et  al. Reversal of experimental
diabetic neuropathy by VEGF gene transfer. J Clin
Investig. 2001;107(9):1083–92.
298. Pawson EJ, et  al. Engineered zinc finger protein-­
mediated VEGF-A activation restores deficient
VEGF-A in sensory neurons in experimental diabe-
tes. Diabetes. 2010;59(2):509–18.
299. Price SA, et al. Gene transfer of an engineered tran-
scription factor promoting expression of VEGF-A
protects against experimental diabetic neuropathy.
Diabetes. 2006;55(6):1847–54.
300. Hulse RP, et  al. Regulation of alternative VEGF-A
mRNA splicing is a therapeutic target for analgesia.
Neurobiol Dis. 2014;71:245–59.
301. Hulse RP, et  al. Vascular endothelial growth fac-
tor-­
A(165)b prevents diabetic neuropathic pain
and sensory neuronal degeneration. Clin Sci.
2015;129(8):741–56.
302. Konishi Y, et  al. Trophic effect of erythropoietin
and other hematopoietic factors on central cho-
linergic neurons in  vitro and in  vivo. Brain Res.
1993;609(1–2):29–35.
303. Campana WM, Misasi R, O’Brien JS. Identification
of a neurotrophic sequence in erythropoietin. Int J
Mol Med. 1998;1(1):235–41.
304. Keswani SC, Leitz GJ, Hoke A.  Erythropoietin is
neuroprotective in models of HIV sensory neuropa-
thy. Neurosci Lett. 2004;371(2–3):102–5.
305. Collino M, et  al. Flipping the molecular switch
for innate protection and repair of tissues: long-­
lasting effects of a non-erythropoietic small peptide
engineered from erythropoietin. Pharmacol Ther.
2015;151:32–40.
306. Rey F, et  al. Erythropoietin as a neuroprotective
molecule: an overview of its therapeutic poten-
tial in neurodegenerative diseases. ASN Neuro.
2019;11:1759091419871420.
307. Campana WM, Myers RR.  Erythropoietin and
erythropoietin receptors in the peripheral ner-
vous system: changes after nerve injury. FASEB J.
2001;15(8):1804–6.
308. Campana WM, Myers RR.  Exogenous erythropoi-
etin protects against dorsal root ganglion apoptosis
and pain following peripheral nerve injury. Eur J
Neurosci. 2003;18(6):1497–506.
309. Li XQ, Gonias SL, Campana WM.  Schwann cells
express erythropoietin receptor and represent a
major target for Epo in peripheral nerve injury. Glia.
2005;51(4):254–65.
154 N. A. Calcutt

310. Bianchi R, et al. Erythropoietin both protects from 325. Pradat PF, et al. Continuous delivery of neurotrophin
and reverses experimental diabetic neuropathy. Proc 3 by gene therapy has a neuroprotective effect in
Natl Acad Sci U S A. 2004;101(3):823–8. experimental models of diabetic and acrylamide neu-
311. Demiot C, et al. Erythropoietin restores C-fiber func- ropathies. Hum Gene Ther. 2001;12(18):2237–49.
tion and prevents pressure ulcer formation in diabetic 326. Chu QM, et al. Systemic insulin-like growth factor-1
mice. J Invest Dermatol. 2011;131(11):2316–22. reverses hypoalgesia and improves mobility in a
312. Loesch A, et  al. Sciatic nerve of diabetic rat mouse model of diabetic peripheral neuropathy. Mol
treated with epoetin delta: effects on C-fibers and Ther. 2008;16(8):1400–8.
blood vessels including pericytes. Angiology. 327. Homs J, et  al. Intrathecal administration of IGF-I
2010;61(7):651–68. by AAVrh10 improves sensory and motor deficits
313. Chattopadhyay M, et  al. Neuroprotective effect in a mouse model of diabetic neuropathy. Mol Ther
of herpes simplex virus-mediated gene transfer of Methods Clin Dev. 2014;1:7.
erythropoietin in hyperglycemic dorsal root gan- 328. Ropper AH, et al. Vascular endothelial growth fac-
glion neurons. Brain. 2009;132:879–88. tor gene transfer for diabetic polyneuropathy: a
314. Wu Z, Mata M, Fink DJ.  Prolonged regulat- randomized, double-blinded trial. Ann Neurol.
able expression of EPO from an HSV vector 2009;65(4):386–93.
using the LAP2 promoter element. Gene Ther. 329. Eisenstein M.  Sangamo’s lead zinc-finger ther-
2012;19(11):1107–13. apy flops in diabetic neuropathy. Nat Biotechnol.
315. Brines M, et  al. Nonerythropoietic, tissue-­ 2012;30(2):121–3.
protective peptides derived from the tertiary struc- 330. Glassman AR, et al. Five-year outcomes after initial
ture of erythropoietin. Proc Natl Acad Sci U S A. aflibercept, bevacizumab, or ranibizumab treatment
2008;105(31):10925–30. for diabetic macular edema (protocol T extension
316. Leist M, et al. Derivatives of erythropoietin that are study). Ophthalmology. 2020;127(9):1201–10.
tissue protective but not erythropoietic. Science. 331. Striglia E, et  al. Emerging drugs for the treatment
2004;305(5681):239–42. of diabetic retinopathy. Expert Opin Emerg Drugs.
317. Bitto A, et al. Activation of the EPOR-beta common 2020;25(3):261–71.
receptor complex by cibinetide ameliorates impaired 332. Roy S, et al. Neurogenic tissue nanotransfection in
wound healing in mice with genetic diabetes. BBA-­ the management of cutaneous diabetic polyneuropa-
Mol Basis Dis. 2018;1864(2):632–9. thy. Nanomedicine. 2020;28:102220.
318. Schmidt RE, et  al. Effect of insulin and an 333. Obrien JS, et  al. Identification of prosaposin as
erythropoietin-­derived peptide (ARA290) on estab- a neurotrophic factor. Proc Natl Acad Sci U S A.
lished neuritic dystrophy and neuronopathy in Akita 1994;91(20):9593–6.
(Ins2(Akita)) diabetic mouse sympathetic ganglia. 334. Calcutt NA, et  al. Prosaposin gene expression and
Exp Neurol. 2011;232(2):126–35. the efficacy of a prosaposin-derived peptide in
319. Schmidt RE, et  al. Erythropoietin and its carba- preventing structural and functional disorders of
mylated derivative prevent the development of peripheral nerve in diabetic rats. J Neuropathol Exp
experimental diabetic autonomic neuropathy in Neurol. 1999;58(6):628–36.
STZ-induced diabetic NOD-SCID mice. Exp 335. Calcutt NA, Freshwater JD, O’Brien JS. Protection
Neurol. 2008;209(1):161–70. of sensory function and antihyperalgesic proper-
320. Brines M, et  al. ARA 290, a nonerythropoi- ties of a prosaposin-derived peptide in diabetic rats.
etic peptide engineered from erythropoietin, Anesthesiology. 2000;93(5):1271–8.
improves metabolic control and neuropathic symp- 336. Jolivalt CG, et al. Therapeutic efficacy of prosaposin-­
toms in patients with type 2 diabetes. Mol Med. derived peptide on different models of allodynia.
2014;20:658–66. Pain. 2006;121(1–2):14–21.
321. Zhang WJ, Yu GL, Zhang MY.  ARA 290 relieves 337. Jolivalt CG, et al. Impaired prosaposin secretion dur-
pathophysiological pain by targeting TRPV1 chan- ing nerve regeneration in diabetic rats and protection
nel: integration between immune system and noci- of nerve regeneration by a prosaposin-derived pep-
ception. Peptides. 2016;76:73–9. tide. J Neuropathol Exp Neurol. 2008;67(7):702–10.
322. Swartjes M, et al. ARA 290, a peptide derived from 338. King MR, et  al. Glycogen synthase kinase-3 inhi-
the tertiary structure of erythropoietin, produces bition prevents learning deficits in diabetic mice. J
long-term relief of neuropathic pain coupled with Neurosci Res. 2013;91(4):506–14.
suppression of the spinal microglia response. Mol 339. Mizisin AP, et  al. TX14(A), a prosaposin-derived
Pain. 2014;10:13. peptide, reverses established nerve disorders in
323. Mata M, Chattopadhyay M, Fink DJ. Gene therapy streptozotocin-diabetic rats and prevents them
for the treatment of sensory neuropathy. Expert Opin in galactose-fed rats. J Neuropathol Exp Neurol.
Biol Ther. 2006;6(5):499–507. 2001;60(10):953–60.
324. Chattopadhyay M, et  al. Prolonged preservation 340. Evans SR, et al. A randomized trial evaluating pro-
of nerve function in diabetic neuropathy in mice saptide (TM) for HIV-associated sensory neuropa-
by herpes simplex virus-mediated gene transfer. thies: use of an electronic diary to record neuropathic
Diabetologia. 2007;50(7):1550–8. pain. PLoS One. 2007;2(7):e551.
Neurotrophic Factors in the Pathogenesis and Treatment of Diabetic Neuropathy
341. Sung KJ, et  al. Swedish nerve growth fac-
tor mutation (NGF(R100W)) defines a role for
TrkA and p75(NTR) in nociception. J Neurosci.
2018;38(14):3394–413.
342. Sung KJ, Yang WL, Wu CB.  Uncoupling neuro-
trophic function from nociception of nerve growth
factor: what can be learned from a rare human dis-
ease? Neural Regen Res. 2019;14(4):570–3.
343. Yang WL, et al. Targeted mutation (R100W) of the
gene encoding NGF leads to deficits in the periph-
eral sensory nervous system. Front Aging Neurosci.
2018;10:373.
344. Yang WL, et al. A missense point mutation in nerve
growth factor (NGF(R100W)) results in selective
peripheral sensory neuropathy. Prog Neurobiol.
2020;194:101886.
345. Giuliani A, et  al. Effects of topical application of
CHF6467, a mutated form of human nerve growth
factor, on skin wound healing in diabetic mice. J
Pharmacol Exp Ther. 2020;375(2):317–31.
346. Hulse RP, et  al. The control of alternative splicing
by SRSF1 in myelinated afferents contributes to the
development of neuropathic pain. Neurobiol Dis.
2016;96:186–200.
347. Oltean S, et al. SRPK1 inhibition in vivo: modulation
of VEGF splicing and potential treatment for mul-
tiple diseases. Biochem Soc Trans. 2012;40:831–5.
348. Singh B, et  al. Regeneration of diabetic axons is
enhanced by selective knockdown of the PTEN
gene. Brain. 2014;137:1051–67.
349. Fernyhough P, Calcutt N, Sabbir M.  Muscarinic
receptor antagonism rescues cholinergic-mediated
cytoskeletal defects and enhances neurite out-
growth in sensory neurons. J Peripher Nerv Syst.
2018;23(4):397–8.
350. Sabbir MG, Fernyhough P.  Muscarinic receptor
antagonists activate ERK-CREB signaling to aug-
ment neurite outgrowth of adult sensory neurons.
Neuropharmacology. 2018;143:268–81.
155
351. Saleh A, et  al. Muscarinic toxin 7 signals via
Ca2+/calmodulin-dependent protein kinase
kinase beta to augment mitochondrial function
and prevent neurodegeneration. Mol Neurobiol.
2020;57(6):2521–38.
352. Calcutt NA, et  al. Selective antagonism of musca-
rinic receptors is neuroprotective in peripheral neu-
ropathy. J Clin Investig. 2017;127(2):608–22.
353. Jolivalt CG, et  al. Topical delivery of muscarinic
receptor antagonists prevents and reverses peripheral
neuropathy in female diabetic mice. J Pharmacol
Exp Ther. 2020;374(1):44–51.
354. Conese M, et  al. Harnessing stem cells and neuro-
trophic factors with novel technologies in the treat-
ment of Parkinson’s disease. Curr Stem Cell Res
Ther. 2019;14(7):549–69.
355. Okawa T, et al. Transplantation of neural crest-like
cells derived from induced pluripotent stem cells
improves diabetic polyneuropathy in mice. Cell
Transplant. 2013;22(10):1767–83.
356. Shibata T, et  al. Transplantation of bone marrow-­
derived mesenchymal stem cells improves
diabetic polyneuropathy in rats. Diabetes.
2008;57(11):3099–107.
357. Xie J, et  al. Therapeutic effects of stem cells from
human exfoliated deciduous teeth on diabetic periph-
eral neuropathy. Diabetol Metab Syndr. 2019;11:38.
358. Hasegawa T, et al. Amelioration of diabetic periph-
eral neuropathy by implantation of hematopoietic
mononuclear cells in streptozotocin-induced dia-
betic rats. Exp Neurol. 2006;199(2):274–80.
359. Salgado AJ, et  al. Adipose tissue derived stem
cells secretome: soluble factors and their roles in
regenerative medicine. Curr Stem Cell Res Ther.
2010;5(2):103–10.
360. Yan H, Ding YF, Lu MJ. Current status and prospects
in the treatment of erectile dysfunction by adipose-­
derived stem cells in the diabetic animal model. Sex
Med Rev. 2020;8(3):486–91.
 Treatment Induced Neuropathy
of Diabetes
Nadia McMillan and Christopher H. Gibbons
1 Historical Background
Treatment induced neuropathy of diabetes or
TIND was coined in 2010 to describe the painful
somatosensory and autonomic changes following
aggressive treatment of either type 1 or type 2
diabetes [1]. However, it was first observed by
Caravati in 1933 [2]. He described a 48-year-old
female with diabetes started on insulin and
aggressive lifestyle management who developed
numbness, tingling, and shooting pains in her
lower extremities about 1 month after initiation
of treatment. Within 3 days of discontinuing
insulin her symptoms improved and again
returned with re-initiation of insulin. Initial dis-
cussion attributed these changes to an allergic
skin reaction resulting in the term, “insulin neuri-
tis.” It was later described and termed diabetic
neuropathic cachexia [3] followed by acute pain-
ful diabetic neuropathy (APDN) by Archer et al.
[4] It is important to point out that TIND is dis-
tinct from diabetic neuropathic cachexia (a diag-
N. McMillan
Department of Neurology, Beth Israel Deaconess
Medical Center, Harvard Medical School,
Boston, MA, USA
C. H. Gibbons (*)
Beth Israel Deaconess Medical Center,
Boston, MA, USA
e-mail: cgibbons@bidmc.harvard.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_9
nosis also associated with diabetes mellitus) in
which patients develop profound weight loss, a
symmetric peripheral neuropathy, and painful
dysesthesias over the limbs and trunk without
associated weakness [5]. The term acute painful
diabetic neuropathy encompasses these two
subtypes.
Since then, there have been countless case
reports describing the same changes in a handful
of patients. In 2010, Gibbons and Freeman used
the term “Treatment induced neuropathy of dia-
betes” to describe a case series which described
neuropathy in 16 patients with both type 1 and
type 2 diabetes [1]. In addition to the multiple
case reports, they showed that autonomic symp-
toms correlated with the acute on set of neuropa-
thy associated with glycemic control in all
patients and were able to provide longitudinal
follow up not previously performed.
As it stands, TIND is now defined as “the
acute onset of neuropathic pain and/or autonomic
dysfunction within 8 weeks of a large improve-
ment in glycemic control-specified as a decrease
in glycosylated hemoglobin A1C (HbA1c) of
≥2% points over 3 months” [6]. Diagnostically,
the onset of symptoms can be useful when deter-
mining the etiology of neuropathy. In a patient
with diabetes the acute onset of neuropathy is
likely to be related to TIND or a diabetic lumbo-
sacral radiculoplexus neuropathy (DLRPN).
157
158
Diabetic sixth nerve palsy would also fall into
this category. Diabetic lumbosacral radiculopa-
thy is an acute to subacute neuropathy in which
patients present with proximal or asymmetric
weakness, hypo or areflexia, and deep pain. There
is usually associated weight loss as well as wide-
spread autonomic failure. Nerve biopsy in these
patients shows axonal loss and demyelination
[7]. It is important to point out that both DLRPN
and TIND are distinctly separate from the nerve
damage seen in diabetic polyneuropathy (DPN)
in those with long standing diabetes [8]. Given
the well established presence of DPN, the topic
of TIND itself has been controversial given
­difficulty in teasing out pre-existing nerve dam-
age typically seen in long standing diabetes from
that of an acute neuropathy. See below for clini-
cal features of TIND.
2 Clinical Features
Clinical features of TIND include pain that is
usually described as burning or shooting in a
length dependent, distal or diffuse, or proximal
pattern that is frequently accompanied by allo-
dynia and hyperalgesia. This usually occurs
within 2–8 weeks of initiation of aggressive
treatment of lifestyle changes and correlates
with higher glucose and HbA1C scores. The
larger and more rapid the correction, the greater
the neuropathy distribution and the more severe
the pain. Correction in HbA1c is often >2% in 3
months. To further breakdown symptoms, using
a cohort of 16 patients with TINDs, Gibbons
et  al. found that patients universally suffered
from intractable pain that was typically described
as symmetric and length dependent. However,
3/16 of the patients with type 1 diabetes also
described proximal/generalized pain.
Additionally, patients suffered from pain sensi-
tivity with both hyperalgesia and allodynia
reported in 80% of type 1 subjects and 40% of
type 2 patients. On neurologic examination, sen-
sory changes typically include mild distal loss of
pain and vibratory senses in lower limbs, with
only rare absent ankle and knee reflexes, and
normal reflexes in the upper extremities. Muscle
N. McMillan and C. H. Gibbons
strength is typically preserved [9]. Motor symp-
toms are not considered a feature of TIND. Two
other prominent features of TIND are autonomic
symptoms and microvascular complications that
appear with or shortly after the onset of neuro-
pathic pain [10]. Autonomic symptoms associ-
ated with TIND include impotence, orthostatic
hypotension, syncope, hyperhidrosis, and anhi-
drosis [1, 10, 11]. Microvascular complications
of TIND include both nephropathy and retinopa-
thy [1, 10, 11].
The neuropathy associated with TIND affects
small fibers including autonomic and somatosen-
sory with relative sparing of the large myelinated
nerve fibers as determined by nerve conduction
studies and exam findings. Some patients present
with prominent manifestations of autonomic dys-
function including orthostatic hypotension and
syncope. Gibbons et al. described 16 cases of adult
onset TIND in which all suffered autonomic dys-
function on testing and exhibited symptoms of
autonomic impairment. Autonomic testing showed
mild-to-moderate sympathetic and parasympa-
thetic dysfunction with strong correlations
between the magnitude of decrease in HbA1c over
3 months [10]. A greater change in HbA1c resulted
in worsening cardiovascular parasympathetic
function as determined by the heart rate response
to deep respiration and the Valsalva maneuver and
tilt-table testing to 60° for 45  min. Sympathetic
adrenergic dysfunction was also observed and cor-
related with a greater change in HbA1c as deter-
mined by the fall in systolic blood pressure during
Valsalva and tilt-table testing [10].
In patients with TIND there is a strong cor-
relation with the simultaneous development of
microvascular complications such as retinopa-
thy and nephropathy. The retinopathy develops
rapidly in conjunction with the onset of pain [1,
10]. The Diabetes Control and Complications
Trial (DCCT) defined retinopathy as the devel-
opment of soft exudates and/or intraretinal
microvascular abnormalities sand was consid-
ered “early” if it occurred between baseline and
12-month follow-­up visits [12]. In their 2014
study of 104 patients with TIND, Gibbons and
Freeman determined that at baseline (prior to
TIND development) 65/104 did not have
Treatment Induced Neuropathy of Diabetes
r­etinopathy, 35/104 had non-­proliferative reti-
nopathy, and 4/104 had proliferative retinopa-
thy. One year after developing TIND, 10/104
individuals had no retinopathy, 54/104 had non-
proliferative retinopathy, and 40/104 had prolif-
erative retinopathy (P  <  0.001 vs. baseline
values). Similar to the risk of developing TIND,
the risk of developing retinopathy increases
with every percentage point decrease in the
HbA1c over a 3-month period of time. The
DCCT studies suggests that for every percent-
age point drop in HbA1C, there was a 1.6-fold
increase in the risk for early worsening retinop-
athy and correlates with the initial HbA1c level
[12, 13]. Notably, unstable glycemic control can
be associated with severe proliferative retinopa-
thy resulting in vision loss [13] (Fig. 1).
Nephropathy, as detected by microalbumin-
uria on urine testing, is another prominent feature
Fig. 1  Treatment induced neuropathy of diabetes
159
of patients with TIND.  Gibbons and Freeman
found that 87/104 patients with TIND had evi-
dence of microalbuminuria 12 months after diag-
nosis compared to 18/104 prior to development
of TIND (P < 0.001, χ2). Those with stable glu-
cose did not have significant improvement at
8-year follow-up and those with unstable glyce-
mic control saw worsening in their renal function
with some requiring initiation of hemodialysis
and renal transplantation [1, 6].
3 Pathophysiology
Several theories exist regarding the underlying
pathophysiology that leads to the development
of TIND.  One hypothesis suggests that a ‘rela-
tive’ hypoglycemia in those with chronically
elevated blood glucose levels following ­treatment
160
or aggressive lifestyles changes leads to an
energy dependent failure of axonal transport,
apoptosis (cell death), and neuropathic pain as a
consequence of axon regeneration and ectopic
firing of regenerating nerves [6, 14–16].
Controlled studies of modest hypoglycemia in
humans using an insulin clamp have resulted in
the development of tactile hyperalgesia and tran-
sient autonomic dysfunction associated with the
release of pro-­inflammatory cytokines [17]. In
this theory, hypoglycemia acts as a stress
response resulting in the activation of the hypo-
thalamic-pituitary adrenal axis and sympatho-
adrenal system and release of catecholamines,
glucocorticoids, and cytokines in this disorder.
In particular, cytokines such as interleukin-1β,
interleukin-6, and tumor necrosis factor-α have
been associated with painful neuropathy and are
elevated in experimental hypoglycemia [10, 18,
19]. Using skin biopsies from eight patients
diagnosed with TIND, Gibbons and Freeman
showed that there is diffuse damage to the unmy-
elinated and lightly myelinated nerve fibers that
is temporally related to the rapid improvement in
glucose control [1]. Further exploration of these
models may provider greater understanding of
the potential pathophysiological mechanisms of
the disease.
In another theorized mechanism, hyperglyce-
mic induced microvascular changes including
arteriovenous shunts results in endoneurial isch-
emia and firing of regenerating nerve fibers with
drops in blood glucose levels [15, 18, 19]. Here,
the microvascular changes are unable to remodel
at the same rate of decline in serum glucose
resulting in a sort of “steal syndrome” in which
hyperproliferative vessels shunt blood away from
the endoneurium causing an acute neuropathy. A
role for microvascular factors has been supported
by the finding of abnormal epineurial vessel mor-
phology on sural nerve photographs [15, 18, 19].
In a cohort of five patient with TIND, Tesfaye
found that all had arterial attenuation, arteriove-
nous shunts, and proliferating new vessels when
evaluated by sural nerve photography [15]. These
proposed mechanisms suggest that hypoglyce-
mia or relative hypoglycemia from aggressive
management of diabetes results in neuro-vascular
N. McMillan and C. H. Gibbons
changes. Interestingly, neuropathy often occurs
in parallel with retinopathy and nephropathy sug-
gesting a common pathophysiological mecha-
nism [1]. The possibility of a nutritional
deficiency has been raised but is less likely given
the absence of weight loss in patients who go on
to develop TIND. Further clinical and molecular
level studies need to be performed to evaluate the
underlying mechanism.
4 Epidemiology
The true number of individuals at risk for TIND and
the general prevalence in the population is unknown
likely because the majority of information has been
generated from case reports rather than prospective
or longitudinal studies. Additionally, TIND remains
a relatively unrecognized complication of diabetes
with under reporting of symptoms [6, 20].
Prevalence estimates were initially assumed to
be around 1%. However, more recent studies sug-
gest this number may be much higher. In a single
center study, Gibbons and Freeman found that
10.9% of patients referred to a tertiary care
­diabetic neuropathy clinic over 5 years met crite-
ria for diagnosis of TIND [6]. Potential epidemi-
ology can be extrapolated based on the DCCT
trial that looked at early worsening retinopathy
following initiation of treatment. In this trial,
early worsening of retinopathy was seen in 22%
of patient’s receiving intensive therapy vs. 13%
in those assigned to conventional treatment [12,
21–23]. Recovery was seen in 51% and 55% of
patients at their 18 month, respectively [12].
These findings suggest that TIND may be more
common that previously suggested.
TIND has typically been reported in patient’s
newly diagnosed with type 2 diabetes and after
initiation of insulin as well as those with poorly
controlled type 1 diabetes who do not take their
medication appropriately. Typical age ranges are
between 30 and 60 years at the time of diagnosis
[1]. While typically considered a diagnosis of
adults, Alexandrou et  al. recently published a
case series that included 7 children and young
adults (age 9–22) diagnosed with TINDs [8].
Prior to this only two case reports of TNDs in the
Treatment Induced Neuropathy of Diabetes
pediatric population had been reported [24, 25].
In this case series, children with new onset or
poorly controlled type 1 diabetes and elevated
HbA1c (typically >14%) developed neuropathic
symptoms following initiation of treatment.
More specifically, 57% had new-onset diabetes
while the remainder had lived with it for 3–5
years. All presentations were complicated by
both autonomic symptoms (gastrointestinal
100%; cardiovascular 14%; genitourinary 14%)
and microvascular complications (43% had
microalbuminuria and 2/3 had non-proliferative
retinopathy) [8]. This study was underpowered to
comment on epidemiology but, again, suggests
that number of cases may be larger than expected
due to underreporting.
5 Risk Factors
TIND can develop in individuals with type 2 dia-
betes but is much more common among those
with T1DM. The main risk factors for TIND are
rapid correction of serum glucose with either
medication (insulin or oral hypoglycemic agent)
or diet/lifestyle changes. Those with a higher
baseline HbA1c (>10% or 86 mmol/mol) are at
increased risk of TIND.  The absolute risk of
developing TIND was ~10% if the hemoglobin
A1C (HbA1C) declined by more than 2% points
in 3 months. However, if the HbA1C declined by
4% points or more in 3 months, the absolute risk
of developing TIND exceeded 50% [11, 26]. A
large improvement in glycemic control >2%
points over 3 months is a risk factor and has been
incorporated in the operational definition of
TIDN [6, 11]. No specific treatments for TIND
have been studied [1, 6, 11].
Individuals with uncontrolled diabetes for
prolonged periods, particularly those with a his-
tory of eating disorders involving insulin restric-
tion for calorie purging, termed diabetic anorexia,
are at high risk. In a study by Gibbons and
Goeble-Fabri, 52/60 female and 10/16 males
with T1DM had a history of an eating disorder
[26]. None of the 28 patients with type 2 diabetes
and TIND had a history of eating disorders but 6
individuals with type 2 diabetes with substantial
161
weight loss (>10 lb) developed TIND [26]. It is
possible that diabetic anorexia creates a predis-
position to later nerve injury.
6 Complications
There is a presumption that TIND tends to resolve
with time with few long-term complications [14].
However, most patients experience months to
years of severe neuropathic pain, symptoms of
autonomic dysfunction, and also develop compli-
cations of retinopathy and nephropathy. Long-
term complications include a length dependent
small fiber neuropathy, renal failure, blindness,
gastroparesis, orthostatic hypotension, and recur-
rent syncope [1, 6, 27].
7 Long-Term Outcomes
The long-term outcome of neuropathic pain in
TIND is generally good, but make take anywhere
from 3 to 36 months before pain is under reason-
able control without significant polypharmacy
[11, 26, 28]. In one study with the follow-up of 6
diabetic patients with TIND for 2 years all
reported only mild neuropathic pain in the feet
[28]. In contrast, in a larger longitudinal cohort
study of 26 patients with TIND followed for 8
years, 19 of the individuals studied maintained
stable glycemic control for the 8-year period of
time. The other seven individuals had recurrent
episodes of prolonged durations of hyperglyce-
mia followed by periods of rapid glycemic con-
trol and recurrence of pain, consistent with a
diagnosis of recurrent TIND [11, 26]. Of those
with stable glycemic control, there was a gradual
improvement in neuropathic pain over 12–30
months associated with either discontinuation of
decreased dosing of pain medication [27]. The
improvements were pathologically monitored
with longitudinal skin biopsies that confirmed an
increase in intra-epidermal nerve fiber and
improvement in autonomic function studies.
There was also improvement in autonomic
function testing and other microvascular parame-
ters such as a reduction in retinopathy and trend
162
towards improving GFR over 8 years [1, 26]. In
contrast, those seven individuals with unstable
glycemic control in follow-up and severe progres-
sion of neuropathy with motor involvement, renal
failure, and severe proliferative retinopathy with
blindness [27]. Despite the general improvement
seen in the majority of patients with TIND over
time, more than 25% of patients with TIND expe-
rienced significant morbidity including amputa-
tions, dialysis, and blindness [26]. Identification of
at-risk individuals is an unmet research priority.
8 Disease Management
The current management of TIND is primarily
symptomatic relief while simultaneously main-
taining glucose stability. Symptomatic therapy is
aimed at managing neuropathic pain which is
often refractory to standard treatments.
Neuropathic pain medications include the use of
gabapentinoids, tricyclic antidepressants, or
serotonin-­norepinephrine reuptake inhibitors but
there is no consensus on the management of
TIND. Tricyclic antidepressants are often used in
treatment of neuropathic pain, however, their use
in TIND can be limited due to their anticholinergic
side effects that include orthostatic hypotension
and syncope, which could exacerbate symptoms
of autonomic dysfunction that is often seen in
TIND. Thus, neuropathic pain treatment is based
on the individual case and the associated symp-
toms, but often requires polypharmacy for some
period of time for pain control to be optimized.
It is not advised to allow for permissive hyper-
glycemia to reduce neuropathic pain in TIND due
to the risk for longer term complications of recur-
rent TIND [26]. In general, a multi-disciplinary
approach that includes endocrine, neurology, and
pain is beneficial. Finally, consideration should
also be given regarding prevention. By avoiding
rapid and sizable changes to the HbA1c it may be
possible to prevent TIND development, although
there is not data at this time to support this theory.
While no clear consensus about rates of glycemic
change exist, recommendations include a target
change in HbA1c of <2 percentage points over 3
months if feasible [6, 12, 13, 26].
N. McMillan and C. H. Gibbons
A physical representation of the risk of neu-
ropathy development is graphed by the change in
HbA1c. As the change in HbA1c increases
(X-axis) the absolute risk of TIND development
increases (Y-axis) and is seen with the blue solid
line. The physical distribution of neuropathy is
shown on the figures on the graph. The regions in
red represent typical pain with neuropathy seen
in all patients, with regions in gray representing
pain in some patients. As the change in HbA1c
becomes larger, the risk of TIND increases, the
regional distribution of neuropathy increases and
the total pain scores increase as well.
References
1. Gibbons CH, Freeman R. Treatment-induced diabetic
neuropathy: a reversible painful autonomic neuropa-
thy. Ann Neurol. 2010;67:534–41.
2. Pomeranz B, Macaulay RJ, Caudill MA, et  al.
Assessment of autonomic function in humans
by heart rate spectral analysis. Am J Physiol.
1985;248:H151–3.
3. Ellenberg M.  Diabetic neuropathic cachexia.
Diabetes. 1974;23:418–23.
4. Archer AG, Watkins PJ, Thomas PK, Sharma AK,
Payan J.  The natural history of acute painful neu-
ropathy in diabetes mellitus. J Neurol Neurosurg
Psychiatry. 1983;46:491–9.
5. Neal JM.  Diabetic neuropathic cachexia: a rare
manifestation of diabetic neuropathy. South Med J.
2009;102:327–9.
6. Gibbons CH, Freeman R. Treatment-induced neurop-
athy of diabetes: an acute, iatrogenic complication of
diabetes. Brain. 2015;138:43–52.
7. Dyck PJ, Norell JE, Dyck PJ. Microvasculitis and
ischemia in diabetic lumbosacral radiculoplexus neu-
ropathy. Neurology. 1999;53:2113–21.
8. Alexandrou EG, Corathers SD, Lahoti A, et  al.
Treatment-induced neuropathy of diabetes in youth:
case series of a heterogeneous and challenging com-
plication. J Endocr Soc. 2020;4(12):bvaa154.
9. Castellanos F, Mascias J, Zabala JA, Ricart C, Cabello
A, Garcia-Merino A. Acute painful diabetic neuropa-
thy following severe weight loss. Muscle Nerve.
1996;19:463–7.
10. Gibbons CH, Freeman R. Treatment-induced neurop-
athy of diabetes: an acute, iatrogenic complication of
diabetes. Brain. 2014;138:43–52.
11. Gibbons CH. Treatment induced neuropathy of diabe-
tes. Auton Neurosci. 2020;226:102668.
12. Group DR. Early worsening of diabetic retinopathy in
the Diabetes Control and Complications Trial. Arch
Ophthalmol. 1998;116:874–86.
Treatment Induced Neuropathy of Diabetes
13. Lim SW, van Wijngaarden P, Harper CA, Al-Qureshi
SH.  Early worsening of diabetic retinopathy due
to intensive glycaemic control. Clin Experiment
Ophthalmol. 2019;47:265–73.
14. Knopp M, Srikantha M, Rajabally YA. Insulin neuri-
tis and diabetic cachectic neuropathy: a review. Curr
Diabetes Rev. 2013;9:267–74.
15. Tesfaye S, Malik R, Harris N, et  al. Arterio-venous
shunting and proliferating new vessels in acute pain-
ful neuropathy of rapid glycaemic control (insulin
neuritis). Diabetologia. 1996;39:329–35.
16. Kihara M, Zollman PJ, Smithson IL, Lagerlund TD,
Low PA.  Hypoxic effect of exogenous insulin on
normal and diabetic peripheral nerve. Am J Physiol.
1994;266:E980–5.
17. Gibbons CH, Adler GK, Bonyhay I, Freeman
R. Experimental hypoglycemia is a human model of
stress-induced hyperalgesia. Pain. 2012;153:2204–9.
18. Llewelyn JG, Thomas PK, Fonseca V, King RHM,
Dandona P. Acute painful diabetic neuropathy precip-
itated by strict glycaemic control. Acta Neuropathol.
1986;72:157–63.
19. Llewelyn JG, Thomas P, Fonseca V, Dandona
P. Acute painful diabetic neuropathy. Diabetes Care.
1988;11:748–9.
20. Nicodemus JM, Enriquez C, Marquez A, Anaya
CJ, Jolivalt CG.  Murine model and mechanisms
of treatment-induced painful diabetic neuropathy.
Neuroscience. 2017;354:136–45.
21. Lim SW, van Wijngaarden P, Harper CA, Al-Qureshi
SH.  Early worsening of diabetic retinopathy due to
163
intensive glycaemic control. Clin Exp Ophthalmol.
2019;47:265–73.
22. Chantelau E, Meyer-Schwickerath R, Klabe
K.  Downregulation of serum IGF-1 for treatment
of early worsening of diabetic retinopathy: a long-­
term follow-up of two cases. Ophthalmologica.
2010;224:243–6.
23. Davis MD, Beck RW, Home PD, Sandow J, Ferris
FL. Early retinopathy progression in four randomized
trials comparing insulin glargine and NPH insulin.
Exp Clin Endocrinol Diabetes. 2007;115:240–3.
24. Chandler E, Brown M, Wintergerst K, Doll
E. Treatment-induced neuropathy of diabetes (TIND)
in pediatrics: a case report and review of the literature.
J Clin Endocrinol Metabol. 2019;105:395–8.
25. Varadharaju N, Jeevarathnam D, Rajan M,
Ponnurangam Nagarajan V, James S.  A case of
treatment-induced neuropathy in an adolescent with
type 1 diabetes. Ann Pediatr Endocrinol Metab.
2019;24:203–6.
26. Gibbons CH, Goebel-Fabbri A.  Microvascular
complications associated with rapid improvements
in glycemic control in diabetes. Curr Diab Rep.
2017;17:48.
27. Gibbons CH.  Treatment induced neuropathy of
diabetes-­long term implications in type 1 diabetes. J
Diabetes Complications. 2017;31(4):715–20.
28. Dabby R, Sadeh M, Lampl Y, Gilad R, Watemberg
N.  Acute painful neuropathy induced by rapid cor-
rection of serum glucose levels in diabetic patients.
Biomed Pharmacother. 2009;63:707–9.
 Asymmetric Diabetic Neuropathy:
Radiculoplexus Neuropathies,
Mononeuropathies, and Cranial
Neuropathies

Pariwat Thaisetthawatkul and P. James B. Dyck

1 Introduction neuropathies [2] (Table  1). Asymmetric dia-

Diabetic neuropathy is the most common
acquired neuropathy in the world. It can be
seen in up to 50% in diabetic patients [1]. The
clinical manifestations of diabetic neuropathy
vary, ranging from predominantly sensory
symptoms sometimes with associated neuro-
pathic pain in diabetic sensorimotor polyneu-
ropathy to asymmetrical or multifocal
weakness, pain, and sensory loss in diabetic
radiculoplexus neuropathy. There are different
ways of classifying diabetic neuropathy. One
way is dividing diabetic neuropathy into two
major categories: symmetric and asymmetric
betic neuropathies consist of three major types:
diabetic radiculoplexus neuropathies, diabetic
mononeuropathies, and diabetic cranial neu-
ropathies [2]. Even though asymmetric dia-
betic neuropathies are less common than
symmetric diabetic neuropathies, they are still
important to correctly diagnose as they have
different underlying pathophysiology, may be
misdiagnosed as the other forms of neuropa-
thies with similar clinical manifestations and
perhaps most importantly may respond to
immunotherapy or other therapeutic interven-
tions. This chapter describes the asymmetric
diabetic neuropathies in details.

P. Thaisetthawatkul
Department of Neurological Sciences, University of
Nebraska Medical Center, 988435 Nebraska Medical
Center, Omaha, NE, USA
e-mail: pthaiset@unmc.edu
P. J. B. Dyck (*)
Department of Neurology, Mayo Clinic College of
Medicine, Rochester, MN, USA
e-mail: Dyck.PJames@mayo.edu

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 165
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_10
166
Table 1  Classification of diabetic neuropathies
Symmetrical neuropathies
1. Categories
Diabetic distal sensorimotor neuropathy (DPN)
Diabetic autonomic neuropathy (DAN)
Diabetic small fiber neuropathy
Treatment-induced neuropathy (TIN)
2. Pathophysiology
Metabolic derangement
Altered microvessel functions
Increased oxidative stress
3. Clinical presentation
Symmetrical distal
Predominantly sensory or autonomic symptoms
2 Diabetic Radiculoplexus
Neuropathies (DRPN)
DRPN are clinical syndromes usually described
as acute to subacute onset of pain followed by
weakness and sensory symptoms starting asym-
metrically in one limb and usually spread to the
other corresponding limb in weeks or a few
months in diabetic patients [3]. Anatomically,
DRPN clinically and pathologically involve the
roots, plexus, and peripheral nerves (hence the
terminology of radiculoplexus neuropathy).
DRPN comes in three types that occur in isolation
or present together. If DRPN affects predomi-
nantly the lower limbs, it is called diabetic lumbo-
sacral radiculoplexus neuropathy (DLRPN) while
predominant upper limb involvement gives a clin-
ical syndrome of diabetic cervical radiculoplexus
neuropathy (DCRPN) [3]. In a case of truncal
involvement, the term diabetic thoracic radiculo-
neuropathy (DTRN) is used to describe the clini-
cal syndrome [4]. Among all these subtypes,
DLRPN is probably the most prevalent [4]. DRPN
may occur in isolation affecting the lower limbs,
P. Thaisetthawatkul and P. J. B. Dyck
Asymmetrical neuropathies
1. Categories
Diabetic radiculoplexus neuropathy (DRPN)
 –  Diabetic lumbosacral radiculoplexus neuropathy
(DLRPN)
 –  Diabetic cervical radiculoplexus neuropathy (DCRPN)
 –  Diabetic thoracic radiculoneuropathy (DTRN)
Diabetic mononeuropathies
 –  Median mononeuropathy across the wrist (MNW, carpal
tunnel syndrome)
 –  Ulnar mononeuropathy across the elbow (UNE)
 –  Fibular mononeuropathy (FN)
 –  Lateral femoral cutaneous mononeuropathy (meralgia
paresthetica (MP))
Diabetic cranial neuropathies
 –  Diabetic oculomotor mononeuropathies (OMM)
 –  Bell’s palsy
2. Pathophysiology
Inflammatory microvasculitis
Ischemic nerve injury
Compression
3. Clinical presentation
Asymmetrical distal and/or proximal
Motor greater than sensory or autonomic involvement
upper limbs or truncal area separately or in com-
bination and having a combination of DLRPN,
DCRPN, and DTRN occurring in one setting has
been described in about one-third of cases [5, 6].
3 Diabetic Lumbosacral
Radiculoplexus Neuropathy
(DLRPN)
3.1 Historical Perspective
DLRPN is a long recognized asymmetric painful
paralytic lower limb syndrome of acute to subacute
onset in diabetic patients that has been called by
many different names including diabetic myelopa-
thy [7], diabetic amyotrophy [8], diabetic femoral-
sciatic neuropathy [9], diabetic femoral neuropathy
[10], diabetic mononeuropathy multiplex [11],
proximal diabetic neuropathy [12, 13], Bruns-
Garland syndrome [14, 15], diabetic polyradicu-
lopathy [16], painful lumbosacral plexopathy [17],
diabetic chronic inflammatory demyelinating poly-
radiculoneuropathy [18], multifocal diabetic neu-
ropathy [19], and diabetic lumbosacral
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
radiculoplexus neuropathy [5, 20]. These different
names were given to DLRPN based on different
suspected regions of anatomical involvement (spi-
nal cord, muscles or peripheral nerves) or based on
different purported pathophysiological mechanisms
(metabolic derangement, ischemia of nerve from
non-inflammatory causes or from inflammatory
causes) documented among many studies [7–19].
These studies emphasized different types of clinical
presentations, different anatomical involvements,
and different biopsy or autopsy findings [7–19]. A
large prospective clinical and pathological study by
Dyck et al. in 33 DLRPN patients compared to dia-
betic polyneuropathy patients showed strong evi-
dence to support that ischemic nerve injury from
microvasculitis is the main underlying pathophysi-
ological mechanism of DLRPN [5].
3.2 Epidemiology
As mentioned above, the descriptions of DLRPN
were based on case reports or care series and the
incidence of DLRPN has been unknown due to
lack of population-based epidemiological stud-
ies. Furthermore, experts of DLRPN have
assumed that it is associated with diabetes melli-
tus; however, a non-diabetic form of LRPN has
been identified; it had never been shown that the
rate of diabetes mellitus is increased in LRPN
[21]. Recently, the incidence of DLRPN was
evaluated in a population-based epidemiological
study conducted in a northern United States com-
munity by Ng et al. [22] The study was done in
the well-defined population of Olmsted County
MN and the study assessed the incidence of
LRPN (both diabetic and non-diabetic) and the
relationship of LRPN with diabetes mellitus [22].
The overall incidence of LRPN was found to be
4.13/100,000/year (95% CI 3.12–5.15) with the
incidence of DLRPN was 2.57/100,000/year
(95% CI 1.76–3.38) and non-DLRPN was
1.6/100,000/year (95% CI 0.97–2.24) [22]. The
odds of having LRPN among diabetes patients
was 7.91 (95% CI 4.11–15.21) but the odds of
having LRPN among pre-diabetes patients was
1.006 (95% CI 1.00–1.01) [22]. An important
conclusion of this study is that LRPN is a very
167
common form of inflammatory neuropathy. It is
clear from this study that the incidence of DLRPN
is higher than that of other commonly found
inflammatory polyneuropathies such as Guillain-­
Barre syndrome or chronic inflammatory demy-
elinating polyradiculoneuropathy (CIDP).
Another important conclusion from this study is
that diabetes mellitus is clearly a risk factor for
the development of LRPN as it occurred eight
times more frequently in diabetic persons [22].
3.3 Clinical Presentation
Typically, DLRPN usually starts with the heralding
symptom of focal pain, often severe and sharp,
shooting or burning in quality and associated with
contact allodynia (normal touch producing a pain-
ful experience), followed by weakness that usually
starts proximally more commonly than distally
usually in one lower limb. Over weeks to months,
the pain and weakness spreads to the initially unaf-
fected segments and then often to the other lower
limb [23]. The onset often occurs in the setting of a
large amount of weight loss and often after altera-
tions in the diabetic medications so that by the time
of evaluation the blood sugar is frequently under
good control [5]. When a DLRPN patient is seen
by a neurologist, the patient usually has problem-
atic motor deficits, often sustaining falls and many
are not able to walk and require use of a wheelchair
[23]. The pain of DLRPN usually subsides before
the weakness improves. The clinical course of
DLRPN is that of a monophasic illness with maxi-
mum motor deficits and disease nadir occurring on
average at around 6 months, followed by gradual
improvement occurring over many months (and
sometimes years) [5]. Many patients do not have
complete recovery and proximal segments reinner-
vated more completely than distal segments result-
ing in many patients being left with a footdrop [5].
Autonomic symptoms such as orthostatic hypoten-
sion, urinary bladder dysfunction, gastrointestinal
symptoms, and tachycardia can be seen in about
half of the DLRPN patients [5]. These clinical fea-
tures of DLRPN distinguish it from a more com-
mon diabetic sensorimotor polyneuropathy as
diabetic polyneuropathy has a more insidious
168
onset, more sensory predominant symptoms infre-
quent distal motor weakness without proximal
weakness, more symmetrical presentation and usu-
ally occurring in patients with a prolonged history
of diabetes mellitus or with poor blood glucose
control [23]. The spectrum of DLRPN also extends
to a more uncommon clinical presentation of a
painless motor predominant, diabetic neuropathy
(PMPDN) [24]. PMPDN has slower onset, more
symmetrical findings and more severe motor weak-
ness, more associated upper limb involvement, and
no pain symptoms compared to typical DLRPN
cases which has more patchy distribution and is
multifocal [24]. In an epidemiological study, Pinto
and colleagues noted that a sensory predominant
form of DLRPN occasionally exists in which pain
and sensory loss of the lower limb are prominent
a b
c d
Fig. 1  Serial paraffin sections (a–c) and epoxy semithin
section (d) from a sural nerve biopsy in a patient with dia-
betic lumbosacral radiculoplexus neuropathy (DLRPN)
shows epineurial perivascular inflammatory cell infiltrates
that involve the microvessel wall (a: H&E stain; b: CD 45
P. Thaisetthawatkul and P. J. B. Dyck
that has little associated weakness but with electro-
physiological studies showing motor involvement
that met the DLRPN criteria (involvement in at
least two peripheral nerves from at least two nerve
root distributions) [25]. It is likely that most of
these sensory dominant cases are not correctly
diagnosed as being DLRPN.
3.4 Pathophysiology
DLRPN is caused by ischemic nerve injury asso-
ciated with increased nerve inflammation and
microvasculitis [5, 17, 19, 25]. Nerve biopsies in
DLRPN usually show findings of ischemic nerve
injury including multifocal fiber loss (Fig.  1d),
focal perineurial degeneration and thickening,
stain) causing disruption of muscle layer (c: smooth mus-
cle actin [SMACTIN] stain) and multifocal nerve fiber
loss within a fascicle (d: methylene blue stain). These sec-
tions show finding of microvasculitis and ischemic nerve
injury as are typically seen in DLRPN
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
injury neuroma with abortive regeneration of
axons, neovascularization, and increased rate of
axonal degeneration and secondary segmental
demyelination (grouped areas of demyelination
clustered along one teased nerve fiber due to axo-
nal dystrophy) [5]. Perivascular inflammatory
infiltrates of varying sizes surrounding and
involving epineurial (Fig. 1a, b) more than endo-
neurial blood microvessels are usually present
[5]. In some cases, there is infiltration and
destruction of the vessel wall by the inflamma-
tory infiltrates suggesting microvasculitis; this
can be confirmed with smooth muscle actin
(SMACTIN) staining (Fig.  1c) of these blood
vessels showing distortion of the architecture of
these blood vessels consistent with microvasculi-
tis [5]. Evidence of prior bleeding into the nerve
tissue is seen in two thirds of cases as evidenced
by the presence of hemosiderin-laden macro-
phages in the perineurium and the epineurium
often adjacent to damaged microvessels [5].
These biopsy findings are similar between
DLRPN and non-diabetic LRPN with both sug-
gestive of an inflammatory ischemic pathology
which is evidence that there is a common patho-
physiology between the two clinical entities [21,
26]. PMPDN also has similar nerve biopsy find-
ings to DLRPN providing evidence that PMPDN
is in fact a variant of DLRPN and not diabetic
CIDP [24]. Upregulation of inflammatory media-
tors such as intercellular adhesion molecule-1,
tumor necrosis factor-α, interleukin-6, and
nuclear factor ƙB have been demonstrated in the
nerve tissue of patients with DLRPN and non-­
DLRPN supporting the role of inflammation in
the pathophysiology of DLRPN [27].
3.5 Diagnosis
The diagnosis of DLRPN requires a typical clini-
cal presentation (as described above), compatible
abnormalities in electrophysiologic studies, and
the other causes excluded. The nerve conduction
study and electromyography (NCS/EMG) crite-
ria for DLRPN are neurogenic involvement in
distribution of at least two peripheral nerves from
at least two lumbosacral roots (upper lumbar
169
plexopathy counts as two roots levels) with lum-
bosacral paraspinal denervation being allowed.
These electrophysiological findings are required
to make a definite diagnosis and at the same time
exclude lumbosacral radiculopathy or mononeu-
ropathy which can have similar presentations to
DLRPN [23]. A sensory nerve biopsy from the
lower limb is not required for the diagnosis pro-
vided the clinical presentation is typical.
However, when the clinical presentation is atypi-
cal or the disease course is prolonged and it is
unclear if the inflammatory neuropathy is still
active, a sensory nerve biopsy may provide histo-
logic findings to support the diagnosis. The dif-
ferential diagnosis of DLRPN include neoplastic
lumbosacral plexopathy from adjacent pelvic
organs such as colon, cervix, ovaries, and urinary
bladder or distant metastasis from lungs, breast
and lymphoma [3], infection such as perirectal
abscess [28], tuberculosis [29] or diffuse infiltra-
tive lymphomatosis syndrome in HIV patients
[30], sarcoidosis, traumatic lumbosacral plexopa-
thy [3], radiation-induced lumbosacral plexopa-
thy [3], hematoma and vascular lesions in the
pelvis [3], and postsurgical inflammatory neu-
ropathy [31]. It is therefore important to obtain a
detailed and meticulous history and appropriate
laboratory tests such as blood tests, a spinal tap
and appropriate neuroimaging studies to exclude
the other conditions.
3.6 Management
As the pathophysiology of DLRPN is ischemic
nerve injury related to inflammatory microvascu-
litis, the mainstay of treatment for DLRPN is
immunotherapy. As there are no controlled clini-
cal studies that unequivocally show the efficacy
of immunotherapy in DLRPN, the use of immu-
notherapy is considered a good practice based on
the pathophysiology and clinical experience [23].
In a blinded placebo-controlled study of intrave-
nous methylprednisolone (IVMP) given intermit-
tently over 3  months (pulsed), no significant
differences in time to improvement by 4 points of
the Neuropathy Impairment Score occurred
between the treated and the placebo DLRPN
170
groups were found [32]. However, patient’s per-
ception of pain and positive sensory symptoms
were significantly improved in the methylpred-
nisolone group compared to placebo [32]. In
another study of open trial of pulsed IVMP in
non-diabetic LRPN, there was significant
improvement in neurologic examination in all
patients (but the authors note that LRPN is a
monophasic illness so spontaneous improvement
is expected) [33]. In an open trial of pulsed oral
and intravenous methylprednisolone in DLRPN
patients, improvement of the neuropathy was
noted [34]. The doses, routes, and types of corti-
costeroid used in DLRPN vary in clinical
­practices [35]. In our practice, we generally use a
dose of 1 g IVMP given three times in the first
week (usually every other day) followed by 1 g
IVMP weekly over 12 weeks. Intravenous immu-
noglobulin (IVIG) or plasma pheresis may be
considered but have fewer studies to support their
efficacy in DLRPN [36]. It is important to moni-
tor for and manage possible side effects from cor-
ticosteroid including hyperglycemia (particularly
the use in diabetic patients), weight gain, edema,
bone necrosis, osteoporosis, glaucoma, cataract,
infections, steroid-induced psychosis, sleep dis-
turbance, and mood swing. As severe neuropathic
pain and significant motor weakness may develop
in this condition, pain management, physical
therapy, and rehabilitation must be considered
important management strategies for this group
of patients.
3.7 Outcome and Prognosis
As DLRPN is usually monophasic illness with
self-improvement, most patients’ disease course
will plateau after months of illness and then they
will start to recover. However, many patients do
not experience a complete recovery and some
degree of residual deficits are common [5]. In a
large prospective study on DLRPN patients, at
2-year follow-up, 36% of the patients were able
to walk independently but 9% continued to
require a wheelchair and 48% continued to use
walking aids [5]. Neuropathic pain and long-term
morbidity from weakness continue to be a prob-
P. Thaisetthawatkul and P. J. B. Dyck
lem is many patients. Data from the Olmsted
County epidemiological DLRPN study suggests
that community-based DLRPN is less severe than
referral-based DLRPN; fewer community
DLRPN patients are in wheelchairs or have bilat-
eral disease (37% community compared to 92%
referral) but that the overall features of the
community-­based illness are similar to the pub-
lished series (the illness begins focally with pain,
followed by weakness and begins in proximal or
distal segments but usually eventually involves
both) [25]. The long-term survival is reduced in
DLRPN compared to community control patients
but that this reduction in survival is secondary to
diabetes mellitus and its co-morbidities rather
than the neuropathy itself [25].
4 Diabetic Cervical
Radiculoplexus Neuropathy
(DCRPN)
In 1890, Auche reported 11 cases with different
types of diabetic peripheral neuropathies and
among these described three cases of focal lower
(probable DLRPN) and two cases of focal upper
limb (probable DCRPN) pain and weakness [37].
Other cases of CRPN were described in 1948 as
the shoulder-girdle syndrome or neuralgic amy-
otrophy [38]. The term brachial plexus neuropa-
thy was later adopted for this condition of rapid
onset of pain followed by weakness and sensory
loss in the upper limbs [39]. Most series of CRPN
did not record the occurrence of diabetes mellitus
among those affected [38, 40] but in a large study
of 99 CRPN patients, there were five patients
who also had diabetes mellitus [39]. The inci-
dence of DCRPN in the general population is
unknown. In a large cohort of longitudinal
population-­based study of diabetic neuropathy,
the frequency of DCRPN is estimated to be less
than 1% [1]. DCRPN is a distinct clinical syn-
drome of acute or subacute progressive pain,
weakness, and sensory symptoms that usually
start in one upper extremity and spreads to the
other segment of the same upper limb or to the
other upper limb in an asymmetrical manner
among diabetic patients [3]. In a large cohort of
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
DCRPN seen at one institution, DCRPN usually
starts acutely and unilaterally in one upper limb
and spreads to the other upper limb (similar to
how DLRPN becomes bilateral) [6]. Involvement
of the other body regions producing co-occurring
DLRPN, DTRN or phrenic neuropathy was also
common [6]. Motor weakness is the main cause
of disability in DCRPN and it usually reaches its
maximum severity more quickly than does the
deficits in DLRPN (often within 1 week after the
onset) [6]. Similar to DLRPN, weight loss may
be seen at the onset of the symptoms but less
common than DLRPN [6]. Accompanying auto-
nomic symptoms such as orthostatic lightheaded-
ness and changes in sweating can be seen in
about 30% [6]. DCRPN is diagnosed based on
compatible symptoms and signs and abnormali-
ties in electrophysiologic studies indicating bra-
chial plexus involvement [6]. The
electrophysiologic abnormalities usually are
more widespread than the clinical severity [6].
The other causes of CRPN needs to be excluded
by appropriate blood and CSF tests as well as
neuroimaging studies [3]. Serological markers of
a more widespread inflammatory response such
as sedimentation rate and markers of autoimmu-
nity such as anti-nuclear antibodies are usually
elevated [6]. The pathophysiology of DCRPN
has been reported in a large cohort of DCRPN
patients to show ischemic nerve injury and
immune-mediated microvasculitis similar to
what was reported in DLRPN [6]. Nerve biopsies
in DCRPN shows multifocal fiber loss, focal
perineurial degeneration and thickening and
injury neuroma suggesting ischemic nerve injury
similar to the pathology of DLRPN [6]. Interstitial
abnormalities also include perivascular inflam-
matory infiltrates involving vessel walls, vessel
wall disruption, hemosiderin in macrophages,
bleeding into the nerve interstitial tissue, and
neovascularization that were in some cases diag-
nostic of microvasculitis similar to DLRPN [6].
Similar pathologic changes suggest that DCRPN
and DLRPN are both subtypes of the broader
diagnosis of DRPN. A nerve biopsy is not always
indicated to make a diagnosis but may be needed
in a DCRPN case with atypical clinical presenta-
tion. Because of similar pathophysiology, the
171
management of DCRPN is also similar to the
management of DLRPN [6]. Immunotherapy
such as corticosteroid, IVIg or plasmapheresis
have been used in DCRPN and variable improve-
ment has been reported in this condition even
though treated patients had more severe disability
than untreated patients [6]. There has been no
clinical trial of immunotherapy in DCRPN.
5 Diabetic Thoracic
Radiculoneuropathy (DTRN)
DTRN is a clinical syndrome of acute–subacute
onset of pain, often described as burning, sharp
or shooting, jabbing and throbbing characters,
affecting the thoracic cage, upper abdomen or
scapula areas in a radicular pattern [41]. An early
description of DTRN in diabetic patients was dia-
betic truncal mononeuropathy [42]. The pain
usually started in an acute, usually rapid, or sub-
acute manner and could last for many months or
more than a year before a patient seeks medical
attention [41]. It can be unilateral or bilateral but
asymmetrical and can involve several dermato-
mal segments [41]. It can affect mainly the
abdominal area giving a false impression of acute
abdomen leading patients to wrongly undergo an
unnecessarily exploratory laparotomy [43].
Unilateral abdominal muscle weakness causing
hernia or protrusion of the abdominal wall has
been described [44]. In many cases, the patients
give a history of paresthesia, sensory loss or con-
tact allodynia at the thoracic cage or abdominal
wall suggestive of the diagnosis. In general, the
duration of diabetes mellitus in DTRN cases is of
long duration, with a mean duration of 12 years
in one study [41]. Similar to DLRPN and
DCRPN, accompanying weight loss at the time
of onset can be seen up to 60% of patients [41].
Cachexia at the onset is another common com-
plaint in DTRN [45]. The diagnosis of DTRN
requires the recognition of the entity as the symp-
toms may mimic the other causes of thoracic
radiculopathy, most commonly degenerative tho-
racic disc disease [45] and herpes zoster [46].
The other causes of thoracic radicular pain
include Lyme neuroborreliosis [47], neurosar-
172
coidosis [48], benign [49], and malignant [50,
51] mass lesions and myodil cyst after having a
myelogram [52]. It is therefore important to get a
detailed history and physical examination and
appropriate tests with neuroimaging study to
exclude the other causes. Confirmation of DTRN
requires the findings of fibrillation potentials or
positive waves at the rectus abdominis muscles
and in the thoracic paraspinal muscles [45, 53].
DTRN may coexist with DCRPN or DLRPN in
many cases [5, 6]. The course and prognosis of
DTRN is that of self-limited symptoms in several
months [41]. The main treatment of DTRN is
pain management. In most cases, there is not an
indication for immunotherapy but the authors
have used weekly pulse intravenous methylpred-
nisolone for 6–12 weeks in some cases with sat-
isfactory results. So far, there has been no
reported pathologic study or nerve biopsy study
on DTRN.  However, we have biopsied thoracic
rootlets from individual cases and have seen peri-
vascular inflammatory infiltrates (personal expe-
rience). Furthermore, the similar clinical features
of abrupt onset of pain and weakness, associated
weight loss, and monophasic course make it
likely that DTRN share a similar pathogenesis
with DCRPN and DLRPN and thereby should be
classified as a spectrum of DRPN similar to
DCRPN and DLRPN.
6 Entrapment Neuropathy
Entrapment neuropathies involving the median,
ulnar and fibular nerves, and the lateral cutaneous
nerve of the thigh are commonly seen in diabetic
patients [54]. Diabetes mellitus has been recently
shown to be a risk factor of entrapment neuropa-
thies in the upper extremity (median mononeu-
ropathy across the wrist (MNW) or carpal tunnel
syndrome (CTS) and ulnar mononeuropathy at
the elbow (UNE)) in a large longitudinal,
population-­based Swedish study [55]. The risk of
developing entrapment neuropathy is higher in
diabetic patients who have diabetic sensorimotor
neuropathy (DPN), particularly in females and in
type 2 diabetes mellitus [56]. An increased risk of
entrapment neuropathy in diabetes mellitus can
P. Thaisetthawatkul and P. J. B. Dyck
be explained by several reasons. One morpho-
metric study of posterior interosseous nerve in
diabetic patients who have CTS showed signifi-
cantly increased nerve vessel area, basement
membrane area, and density of unmyelinated
axons and expression of vascular endothelial
growth factor indicating nerve ischemia and
endoneurial hypoxia enhancing increased vascu-
lar permeability and neovascularization in dia-
betic CTS compared to non-diabetic CTS patients
while density of myelinated fibers, myelinated
fiber area, and axon area are significantly
decreased [57]. In addition, ultrasonographic
studies in peripheral nerves in diabetic patients
show diffuse enlargement of nerve cross-­sectional
areas at non-entrapment sites in diabetic patients
with or without DPN compared to healthy con-
trols, significantly higher in patients who have
DPN [58, 59] and even in the nerves that have
normal results of electrophysiological studies
[59]. These evidences suggest structural abnor-
malities, even subclinical, in diabetic nerves
increasing susceptibility to entrapment neuropa-
thies. The causes of nerve enlargement are likely
multifactorial and related to underlying meta-
bolic derangement, oxidative stress and accumu-
lation of glycation products of sorbitol pathway
and inflammatory process described as an under-
lying mechanism in DPN [60]. The following are
the descriptions of each entrapment neuropathy
in association with diabetes mellitus. However, a
detailed account of each entrapment neuropathy
is beyond the scope of this chapter and the read-
ers are recommended to consult the other appro-
priate sources of references.
7 Median Mononeuropathy
Across the Wrist (MNW)
MNW (also known as carpal tunnel syndrome) is
the most common entrapment neuropathy in dia-
betes mellitus [54]. Asymptomatic MNW with
only abnormal electrophysiological findings has
been reported in 22% in type 1 diabetes mellitus
and 29% in type 2 diabetes mellitus while symp-
tomatic MNW in 11% in type 1 diabetes mellitus
and 6% in type 2 diabetes mellitus in a
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
community-­based longitudinal cohort of diabetic
patients (the Rochester Diabetic Neuropathy
Study) [1]. The prevalence of MNW in general
population has been estimated at 2–4% [61] but
in diabetes mellitus, the prevalence has been
reported to be 14% in patients without DPN and
30% in those with DPN [62]. MNW occurs as a
result of median nerve compression at the carpal
ligament [54]. MNW typically causes paresthesia
in the palmar surface of the thumb, second, third,
and the radial half of the fourth digits and, if
severe, weakness in the thumb abductor [63]. Not
uncommonly, paresthesia of the whole hand and
pain in the wrist are also described. These symp-
toms are precipitated by prolonged flexion or
extension of the wrist such as driving or during
sleep [63]. The symptoms of MNW in diabetic
patients are similar to non-diabetic subjects but
coexisting symptoms of DPN (i.e., numbness/
pain in the toes or feet) are usually present [64].
The diagnosis of MNW requires compatible elec-
trodiagnostic abnormalities showing prolonged
distal latencies in the median sensory or motor
studies or slowing of the median conduction
velocity across the wrist [63]. Recently the use of
diagnostic high-resolution ultrasonographic stud-
ies for MNW has become more widespread [65].
An ultrasound finding of an enlarged median
nerve in proximal carpal tunnel, edema around
flexor tendons in cross-sectional images, and
thickening of the flexor retinaculum could sug-
gest CTS [65]. Management of symptomatic
MNW includes conservative treatment such as
the use of wrist splint starting initially at night
and escalating to the daytime and, if unsuccess-
ful, proceeding to local corticosteroid injection
and for more severe cases with weakness or atro-
phy of the thumb abductor, decompressive sur-
gery of the carpel tunnel [64]. A recent 5-year
follow-up study in diabetic patients who under-
went carpal tunnel surgery showed significant
long-term neurophysiological and clinical
improvement in diabetic and non-diabetic
patients alike and this improvement is indepen-
dent of the status of DPN in diabetic CTS patients
[66]. Diabetes mellitus was, however, found to be
an independent risk factor of infection following
carpal tunnel surgery in a recent large cohort of
173
454,987 patients undergoing carpal tunnel sur-
gery [67]. Because underlying metabolic
derangement could make diabetic nerve suscep-
tible to entrapment at the pressure sites, attention
to metabolic control is equally important in addi-
tion to conventional management of MNW in
diabetic patients [54].
8 Ulnar Mononeuropathy
Across the Elbow (UNE)
UNE ranks second to MNW in prevalence in dia-
betic patients with prevalence about 2.1% [54].
Similar to MNW, diabetes mellitus has been
shown to be a risk factor for UNE [55]. The prev-
alence of diabetes mellitus in UNE is about 6%
[68]. Most diabetic UNE cases are asymptomatic
or subclinical and are detected by electrophysio-
logical studies [69, 70]. Similar to MNW, UNE is
found more commonly in diabetic patients with
DPN [70]. The symptoms of UNE in diabetic
patients are similar to non-diabetic UNE.  Most
common symptoms are paresthesia or numbness
at the fifth digit, the medial half of the fourth digit
and the medial half of the volar surface of the
ipsilateral hand [71]. In a more severe case,
intrinsic hand weakness and atrophy of the inter-
ossei resulting in clawing of the fourth and the
fifth digits and the Benediction posture or weak-
ness of third palmar interosseus muscle resulting
in Wartenburg’s sign can be seen [71]. In diabetic
patients with severe UNE, many cases present
with motor weakness and atrophy without sen-
sory symptoms [72]. Inching studies showed that
the most common location for diabetic UNE is at
the retrocondylar groove [70]. The diagnosis of
UNE requires electrodiagnostic abnormalities
showing conduction block or slowing across the
elbow [71]. However, in light of coexisting DPN
in the majority of diabetic UNE patients, electro-
physiologic abnormalities may not be clear cut
due to slowing of the conduction velocity of the
ulnar nerve at the forearm segment. In this situa-
tion electrophysiologic abnormalities may reveal
a non-localizing result and the diagnosis of UNE
may require either inching study across the elbow
[70] or a nerve ultrasound study [73] to make a
174
correct localization. A high-resolution ultrasound
study of the ulnar nerve, similar to its utility in
MNW, has become an important diagnostic tool
in UNE [74]. Similar to an ultrasound study at the
median nerve, a high-resolution ultrasound study
at the ulnar nerve showed significant enlargement
in the cross-sectional area of the nerve in diabetic
patients, particularly in patients with DPN [75].
This could also explain why UNE is also seen
commonly in diabetic patients, particularly in
DPN patients. Management of symptomatic
UNE in diabetic patients is similar to non-­diabetic
UNE. In a mild case, non-operative management
in the form of wearing a soft elbow pad or avoid-
ing prolonged elbow flexion may be considered
and for more severe cases and in the cases that
fail conservative treatment, surgical intervention
may be considered [76]. Another form of ulnar
mononeuropathy seen in diabetic patients is ulnar
neuropathy in the forearm [77]. Ulnar neuropathy
in the forearm has similar clinical symptoms to
UNE but electrodiagnostic studies localize the
lesion in the forearm, i.e. conduction block or
abnormal temporal dispersion are seen between
the wrist and the elbow [77]. It is unclear if this
type of ulnar neuropathy is part of DPN or it has
a structural cause for it.
9 Fibular Mononeuropathy
Across the Fibular Head (FN)
FN is caused by compression of the fibular (pero-
neal) nerve at the bony prominence of the fibular
neck and is the most common entrapment neuropa-
thy of the lower limb [54]. FN commonly occurs
after habitual leg crossing, prolonged hospitaliza-
tion, prolonged bedrest or during a surgical proce-
dure requiring general anesthesia [78]. FN often
occurs when there is a large amount of weight loss
due to loss of subcutaneous fat padding around the
common fibular nerve, such as severe malnutrition
or after having cancers [78]. Diabetes mellitus has
been described as a precipitating factor in a large
case series of FN, seen in 13 out of 103 FN cases
[79]. The prevalence of FN in diabetes mellitus is
unknown but FN was found in 13 out of 29 mono-
neuropathies in diabetic patients [80]. In diabetes
P. Thaisetthawatkul and P. J. B. Dyck
mellitus, FN can be seen as a result of diabetic
muscle infarct involving peroneal muscles [81].
The diagnosis requires elevated creatine kinase and
MRI of the lower extremity to demonstrate muscle
infarct [81]. FN causes weakness of ipsilateral foot
dorsiflexion and eversion but plantar dorsiflexion
and inversion are spared (as these functions are
controlled by the tibial nerve). Ipsilateral footdrop
is a very common clinical presentation. The diag-
nosis requires electrodiagnostic abnormalities
showing slowing or conduction block of the fibular
motor study with or without axonal loss [78]. The
major differential diagnosis for FN is L5 radicu-
lopathy, lumbosacral plexopathy (DLRPN) or sci-
atic mononeuropathy [82]. As some cases of FN
can occur in patients with DPN, electrodiagnostic
abnormalities for FN may not be definite as DPN is
a length-dependent axonal sensorimotor neuropa-
thy and loss of either fibular motor response record-
ing at extensor digitorum brevis muscle or
superficial fibular sensory response or both indicat-
ing axon loss is common in DPN. In these situa-
tions, fibular motor study recording at tibialis
anterior muscle, which is a more proximal muscle,
if recordable, or a detailed needle electromyo-
graphic study may provide helpful diagnostic find-
ings [79]. A neuromuscular ultrasound is also
helpful in diagnosing FN demonstrating enlarge-
ment of nerve cross section area at the fibular head
[83]. Management of FN depends on the etiology
but conservative treatment with the use of an ankle
brace and avoiding compression is the mainstay
strategy. Surgical management is controversial and
should be limited to FN that has a clear structural
cause or severe FN that is prolonged and does not
recover with conservative approach [82].
10 Lateral Femoral Cutaneous
Mononeuropathy [Meralgia
Paresthetica (MP)]
MP is an entrapment syndrome that causes
chronic neuropathic pain in the anterior and lat-
eral thigh. It is the second most common entrap-
ment neuropathy in the lower extremity [84]. It is
caused by entrapment of the lateral femoral cuta-
neous nerve of the thigh near the anterior supe-
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
rior iliac spine while passing beneath the inguinal
ligament [78]. The symptoms of MP consist of
persistent or intermittent positive (tingling,
numbness, burning, stabbing or aching) or nega-
tive (sensory loss or hypoesthesia) symptoms
without muscle weakness in unilateral or ­bilateral
anterolateral thighs [84]. The association between
MP and diabetes mellitus was shown in a popula-
tion-based study in Olmsted County Minnesota
[81]. In this study, the incidence of MP was esti-
mated to be seven times increased in diabetic
patients compared to general population and that
MP patients are two times more likely to go on to
develop diabetes mellitus in the future [85]. In
addition, multivariate analysis showed that dia-
betes mellitus is an independent risk factor for
MP [85]. Moreover, MP has been shown to be
associated with obesity [85] and pregnancy [86].
A femoral neuropathy or upper lumbar radicu-
lopathy or upper lumbar plexopathy (DLRPN)
may present with similar symptoms [78]. The
pathophysiology is likely to be compressive in
most cases but in some retracted cases where
neurectomy was performed and hence pathology
was available, inflammatory infiltrates, perineu-
rial thickening, hemosiderin-laden macrophages,
and multifocal fiber loss were found raising the
possibility of inflammatory mechanisms in some
[87]. The presence of muscle weakness or loss of
knee reflex makes MP unlikely but to exclude
those possibilities, detailed NCS/EMG in the
lower extremities are required [78].
Electrodiagnostic abnormalities in MP may be
aided by performing the sensory conduction
study of the lateral femoral cutaneous nerve,
which is technically difficult to do especially in
an obese subject because of anatomical variation.
There are several methods of sensory studies of
the lateral femoral cutaneous nerve using either
needle or surface stimulators or recorders [88].
The abnormalities consistent with MP consist of
unilateral loss or decrease in the amplitude of the
lateral femoral cutaneous sensory study on the
ipsilateral side of the symptoms [88].
Management of MP consists of mainly conserva-
tive approach with losing weight and the use of
medications to relive neuropathic pain such as
antidepressants, anticonvulsants, and local pain
175
treatment [78]. The symptoms in pregnant
patients usually resolve after delivery. Surgical
decompression of the lateral femoral cutaneous
nerve of the thigh is reserved for persistent pain
after conservative management [78]. In very
severe retractable cases neurectomy of lateral
femoral cutaneous nerve has been performed and
resulted in relief of pain [87].
11 Cranial Neuropathies
Cranial mononeuropathies are uncommon mono-
neuropathies that care occasionally seen in dia-
betic patients, the frequency of which ranging
from none in a population-based cohort of
Rochester Diabetic Neuropathy Study [1] to
0.75% in a more recent hospital-based cohort
[89]. The most common diabetic cranial mono-
neuropathy is oculomotor (CNIII) mononeuropa-
thy (0.35%) while facial (CN VII)
mononeuropathy ranks second (0.21%) followed
by abducens (CN VI) mononeuropathy (0.15%)
and multiple cranial nerve palsies can be seen in
about 0.04% [89]. Trochlear (CN IV) mononeu-
ropathy has been only very rarely described in
diabetic patients [90]. Diabetic patients who
develop cranial mononeuropathies usually have
older ages, prolonged duration of diabetes melli-
tus, type 2, poor control of blood glucose and co-­
morbidities [89].
12 Ocular Motor
Mononeuropathies (OMM)
Despite its rarity, diabetic ophthalmoplegia
causes significant symptoms and anxiety to
affected patients as it usually occurs abruptly
causing diplopia and impairs daily activities
significantly. Most commonly it involves CN
III or CN VI in isolation or less commonly a
combination of these [91] but very rarely CN
IV is involved. In a cohort of unselected cases
of OMM, the frequency of diabetes mellitus
was 7% [92] whereas in a large cohort of iso-
lated CN III palsy, diabetes mellitus was the
cause in 11% [93]. The onset of diabetic OMM
176
usually occurs in hours accompanied by ipsilat-
eral pain or headache, often refractory to anal-
gesics and diplopia [94]. The headache is either
periorbital, retro-­orbital, frontal, temporal or
involves the entire side of the hemi-facial area
[94]. Pain or headache can be seen in about
60% of cases and is most common in CN III
palsy [95]. It can occur about 1 week before the
onset of diplopia in about 35% of cases whereas
65% have pain at the onset of diplopia [95]. The
duration of pain can vary from a few days in a
mild case to more than a month in a severe case
[95]. The other symptom includes ptosis (in CN
III palsy) but severe pupillary dysfunction is
uncommon. In diabetic CN III palsy, impaired
pupillary reaction can be seen in up to 50% and
in about 60% of these, the impairment can be
bilateral suggesting an accompanying diabetic
autonomic neuropathy [93]. Even though
impaired pupillary reaction can be seen in dia-
betic CN III palsy, anisocoria greater than
2 mm is very rare [93]. Similarly, a fixed pupil
ipsilateral to the symptom with opposite normal
pupil is very rare in diabetic CN III palsy while
this phenomenon can be seen in 61% of CN III
palsy from a posterior communicating artery
aneurysm [93]. Diabetes mellitus is among the
most common cause of pupil sparing CN III
palsy [93]. The implication of this matter lies in
the necessity to differentiate diabetic CN III
palsy from the other compressive structural
causes such as a posterior communicating
artery aneurysm or a tumor such as a pituitary
tumor. In a large cohort of patients with iso-
lated CN III palsy, diabetes mellitus was found
to be the cause of CN III palsy in only 11%—
another 6% of patients were diabetic and had
another cause of CN III palsy including isch-
emic stroke such as midbrain infarct, infection
or a posterior communicating artery aneurysm
[93]. In another large prospective cohort of
patients with presumed benign CN III, IV, and
VI palsies, a structural cause was identified in
17% [96]. These findings underscore the need
of neuroimaging studies to identify other pos-
sible structural lesions causing OMM in dia-
betic patients, particularly when there are signs
suggesting a structural lesion such as signifi-
P. Thaisetthawatkul and P. J. B. Dyck
cantly impaired pupillary reaction or aberrant
regeneration when there is CN III involvement.
The underlying pathology causing OMM were
elucidated by three clinic-­pathological studies
[97–99], each of which were performed on a
diabetic patient who developed acute CN III
palsy before death, showing either preservation
of axons [98] or axonal degeneration [97] lim-
ited to the central portion of the nerve trunk
while the peripheral part of the nerve was
spared. In addition, there were extensive patho-
logical changes seen in the intraneural arteri-
oles or small arteries consisting of hyalinization
of the vessel walls and obliteration or marked
narrowing of the lumen without inflammation
[97–99]. The large arteries supplying CN III,
on the contrary, were found to be relatively
unremarkable [97–99]. The pathological
changes were seen only in a very short segment
of CN III, two cases in the intracavernous por-
tion, presumptively a watershed area of blood
supply [97, 98]. The underlying mechanism of
diabetic CN III palsy was postulated to be isch-
emia from microvascular origin [97–99].
Pupillary sparing is explained by the peripheral
localization of the autonomic fibers supplying
the pupils thereby escaping the ischemic injury
[100]. Microvascular ischemia as the cause of
diabetic OMM is also supported by findings of
significant association between ischemic OMM
and the risk factors for cardiovascular diseases
in two studies [96, 101]. Moreover, diabetic
OMM has been found to be an independent risk
factor of ischemic stroke in type 2 diabetes
mellitus in one population-based study [102].
However, the occurrence of diabetic retinopa-
thy was found to be significantly less common
in patients who had diabetic OMM than who
had not suggesting a different pathophysiologic
mechanism for these two microvascular com-
plications [90]. The course and prognosis of
diabetic OMM is self-recovery, either complete
or partial in most cases [94]. In a large,
population-­based study of acquired CN III
palsy, 95% of microvascular CN III palsy had
complete while the rest had incomplete, yet
substantial, recovery [103]. On the contrary,
CN III palsy from the other causes such as post-
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial… 177

surgery, trauma, aneurysm or malignancy has tizing external otitis [112], and mucormycosis in
only about 20–35% chance of complete recov- the parotid gland [113, 114]. These patients usu-
ery [103]. The time to recovery is usually within ally have additional symptoms and signs sugges-
4–6 weeks [94]. Management of diabetic OMM tive of local infections such as local facial
is conservative management and control of car- swelling, otalgia, otorrhea, local tenderness,
diovascular risk factors such as pain manage- fever, and crepitation over the skin [110–114].
ment, blood glucose and lipid control, Facial palsy has been described as part of multi-
management of blood pressure, smoking cessa- ple cranial neuropathies (polyneuritis cranialis)
tion, weight reduction, and the use of eye pad to in diabetic patients [115]. The treatment for
reduce diplopia or adhesive tape for ptosis [94]. Bell’s palsy is oral corticosteroid, usually
60–80  mg/day given in a short course over a
week, preferably started within 72  h after the
13 Facial Paralysis onset [104, 109]. A combination of corticosteroid
and anti-viral agent may be considered in a severe
Facial weakness is another common cranial neu- case but the benefit has been found to be marginal
ropathy seen in diabetic patients [89]. Idiopathic in clinical trials but the risk is minimal [104,
CN VII palsy or Bell’s palsy is the most common 109]. The use of anti-viral agent alone is not rec-
cause of facial weakness, accounting for up to ommended due to lack of benefit over the pla-
75% of cases [104]. Diabetes mellitus is more cebo [104, 109]. The course and prognosis of
prevalent in Bell’s palsy than in general popula- Bells’ palsy is that of self-improvement with
tion, ranging from 10% to 30% among many about 70% having complete recovery and about
studies, and is thought to be an important co-­ 15% having moderate to severe residual deficit
morbidity in the type of facial paralysis [105– [104]. The use of corticosteroid may help the
108] even though routine screening for diabetes most severely affected 30% group (who do not
mellitus is not yet recommended when a patient have a full recovery) to achieve a full recovery in
with Bells’ palsy is encountered at an initial visit about one-third to one-half [104]. The presence
[109]. Bells’ palsy is associated with swelling of diabetes mellitus does not seem to affect the
and compression of the facial nerve at the level of recovery of Bell’s palsy [116]. The severity of
the geniculate ganglion within the meatal seg- Bell’s palsy at the time of maximum deficit is an
ment of the nerve [104]. The cause of Bell’s palsy adverse prognostic factor [94]. Electrophysiologic
is likely from non-specific inflammation, with assessment of the facial motor study can predict
possible contribution in some cases from herpes poor prognosis in that if the facial compound
viral infection [104]. The symptoms consist of muscle action potential (CMAP) is less than 10%
unilateral facial paralysis, facial and retro-­ of the lower limit of normal, the chance of having
auricular pain, altered taste, intolerance to loud meaningful recovery is very unlikely and if the
noise (hyperacusis), and sensory symptoms in facial CMAP is more than 30% of lower normal
some cases [104]. Dry eyes, dry mouth, and croc- limit, the chance of having good recovery is
odile tear syndrome (lacrimation from salivary excellent [94]. In case of severe facial weakness
stimuli) can also be seen [104]. The other causes with guarded chance of recovery, the role of sur-
of facial paralysis that needs to be differentiated gical decompression is still controversial [104,
from typical Bell’s palsy include trauma, 109].
varicella-­
zoster infection (Ramsay Hunt syn-
drome), neoplasm, Lyme disease, and sarcoidosis Disclosure Statement  Pariwat Thaisetthawatkul MD has
[94]. In diabetic patients in particular, certain nothing to disclose.
malignant infections may present predominantly P. James B. Dyck MD has nothing to disclose.
with facial paralysis and needs to be excluded
such as candidal abscess of the parotid gland
[110], cervical necrotizing fasciitis [111], necro-
178
References
1. Dyck PJ, Kratz KM, Karnes JL, et  al. The preva-
lence by staged severity of various types of dia-
betic neuropathy, retinopathy, and nephropathy in
a population-based cohort: the Rochester Diabetic
Neuropathy Study. Neurology. 1993;43:817–24.
2. Tracy JA, Dyck PJB.  The spectrum of diabetic
neuropathies. Phys Med Rehabil Clin N Am.
2008;19:1–26.
3. Thaisetthawatkul P, Dyck PJB.  Cervical and lum-
bosacral radiculoplexus neuropathies. In: Rajabally
Y, editor. Dysimmune neuropathies. San Diego:
Elsevier/Academic Press; 2020. p. 199–224.
4. Sinnreich M, Taylor BV, Dyck PJB.  Diabetic neu-
ropathies. Classification, clinical features, and patho-
physiological basis. Neurologist. 2005;11:63–79.
5. Dyck PJB, Norell JE, Dyck PJ. Microvasculitis and
ischemia in diabetic lumbosacral radiculoplexus
neuropathy. Neurology. 1999;53:2113–21.
6. Massie R, Mauermann ML, Staff NP, et al. Diabetic
cervical radiculoplexus neuropathy: a distinct syn-
drome expanding the spectrum of diabetic radiculo-
plexus neuropathies. Brain. 2012;135:3074–88.
7. Garland H, Taverner D.  Diabetic myelopathy. Br
Med J. 1953;1:1405–8.
8. Garland H.  Diabetic amyotrophy. Br Med J.
1955;2:1287–90.
9. Skanse B, Gydell K. A rare type of femoral-sciatic
neuropathy in diabetes mellitus. Acta Med Scand.
1956;155:463–8.
10. Calverly JR, Mulder DW.  Femoral neuropathy.
Neurology. 1960;10:963–7.
11. Raff MC, Sangalang V, Asbury AK. Ischemic mono-
neuropathy multiplex in association with diabetes
mellitus. Neurology. 1968;18:284.
12. Williams IR, Mayer RF. Subacute proximal diabetic
neuropathy. Neurology. 1976;26:108–16.
13. Asbury AK.  Proximal diabetic neuropathy. Ann
Neurol. 1977;2:179–80.
14. Chokroverty S, Reyes MG, Rubino FA.  Bruns-­
Garland syndrome of diabetic amyotrophy. Trans
Am Neurol Assoc. 1977;102:173–7.
15. Barohn RJ, Sahenk Z, Warmolts JR, et al. The Bruns-­
Garland syndrome (diabetic amyotrophy). Revisited
100 years later. Arch Neurol. 1991;48:1130–5.
16. Bastron JA, Thomas JE.  Diabetic polyradiculopa-
thy: clinical and electromyographic findings in 105
patients. Mayo Clin Proc. 1981;56:725–32.
17. Bradley WG, Chad D, Verghese JP, et  al. Painful
lumbosacral plexopathy with elevated erythrocyte
sedimentation rate: a treatable inflammatory syn-
drome. Ann Neurol. 1984;15:457–64.
18. Krendel DA, Costigan DA, Hopkins LC. Successful
treatment of neuropathies in patients with diabetes
mellitus. Arch Neurol. 1995;52:1053–61.
19. Said G, Lacroix C, Lozeron P, et  al. Inflammatory
vasculopathy in multifocal diabetic neuropathy.
Brain. 2003;126:376–85.
P. Thaisetthawatkul and P. J. B. Dyck
20. Dyck PJB, Windebank AJ.  Diabetic and nondia-
betic lumbosacral radiculoplexus neuropathies: new
insights into pathophysiology and treatment. Muscle
Nerve. 2002;25:477–91.
21. Dyck PJB, Norell JE, Dyck PJ.  Non-diabetic lum-
bosacral radiculoplexus neuropathy: natural history,
outcome and comparison with the diabetic variety.
Brain. 2001;124:1197–207.
22. Ng PS, Dyck PJB, Laughlin RS, et al. Lumbosacral
radiculoplexus neuropathy: Incidence and the
association with diabetes mellitus. Neurology.
2019;92:e1188–94. https://doi.org/10.1212/
WNL.0000000000007020. Epub 2019 Feb 13.
23. Dyck PJB, Thaisetthawatkul P. Lumbosacral plexop-
athy. Continuum (Minneap Minn). 2014;20:1343–58.
24. Garces-Sanchez M, Laughlin RS, Dyck PJB, et  al.
Painless diabetic motor neuropathy: a variant of dia-
betic lumbosacral radiculoplexus Neuropathy? Ann
Neurol. 2011;69:1043–54.
25. Pinto MV, Ng PS, Howe BM, Laughlin RS, Thapa
P, Dyck PJ, Dyck PJB. Lumbosacral radiculo-
plexus neuropathy: neurologic outcomes and sur-
vival in a population-based study. Neurology.
2021;96(16):e2098–e2108. https://doi.org/10.1212/
WNL.0000000000011799.
26. Llewelyn JG, Thomas PK, King RH.  Epineurial
microvasculitis in proximal diabetic neuropathy. J
Neurol. 1998;245:159–65.
27. Dyck PJB, Engelstad J, Norell J, Dyck
PJ.  Microvasculitis in non-diabetic lumbosacral
radiculoplexus neuropathy (LSRPN): similarity to
the diabetic variety (DLSRPN). J Neuropathol Exp
Neurol. 2000;59:525–38.
28. Kawamura N, Dyck PJB, Schmeichel AM, et  al.
Inflammatory mediators in diabetic and non-­diabetic
lumbosacral radiculoplexus neuropathy. Acta
Neuropathol. 2008;115:231–9.
29. Holtzman DM, Davis RE, Greco CM. Lumbosacral
plexopathy secondary to perirectal abscess in a patient
with HIV infection. Neurology. 1989;39:1400–1.
30. Schaller MA, Wicke F, Foerch C, et  al. Central
nervous system tuberculosis: etiology, clinical
manifestations and neuroradiological features. Clin
Neuroradiol. 2019;29:3–18.
31. Chahin N, Temesgen Z, Kurtin PJ, et al. HIV lum-
bosacral radiculoplexus neuropathy mimicking
lymphoma: diffuse infiltrative lymphocytosis syn-
drome (DILS) restricted to nerve? Muscle Nerve.
2010;41:276–82.
32. Staff NP, Engelstad J, Klein CJ, et al. Post-surgical
inflammatory neuropathy. Brain. 2010;133:2866–80.
33. Dyck PJ, O’Brien P, Bosch EP, et  al. The multi-­
centre double-blind controlled trial of IV
methylprednisolone in diabetic lumbosacral radicu-
loplexus neuropathy. Neurology. 2006;66(5 Suppl
2):A191.
34. Dyck PJB, Norell JE, Dyck PJ. Methylprednisolone
may improve lumbosacral radiculoplexus neuropa-
thy. Can J Neurol Sci. 2001;28:224–7.
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
35. Kilfoyle D, Kelkar P, Parry GJ.  Pulsed methyl-
prednisolone is a safe and effective treatment for
diabetic amyotrophy. J Clin Neuromuscul Dis.
2003;4:168–70.
36. Thaisetthawatkul P, Dyck PJB. Treatment of diabetic
and nondiabetic lumbosacral radiculoplexus neurop-
athy. Curr Treat Options Neurol. 2010;12:95–9.
37. Chan YC, Lo YL, Chan ES.  Immunotherapy for
diabetic amyotrophy. Cochrane Database Syst Rev.
2012;6:CD006521.
38. Auche MB. Des alterations des nerfs peripheriques.
Arch Med Exp Anat pathol. 1890;2:635.
39. Parsonage MJ, Turner JW.  Neuralgic amyotrophy;
the shoulder-girdle syndrome. Lancet. 1948;1:973–8.
40. Tsairis P, Dyck PJ, Mulder DW. Natural history of
brachial plexus neuropathy. Report on 99 patients.
Arch Neurol. 1972;27:109–17.
41. van Alfen N, van Engelen BG.  The clinical spec-
trum of neuralgic amyotrophy in 246 cases. Brain.
2006;129:438–50.
42. Kikta DG, Breuer AC, Wilbourn AJ.  Thoracic root
pain in diabetes: the spectrum of clinical and elec-
tromyographic findings. Ann Neurol. 1982;11:80–5.
43. Ellenberg M.  Diabetic truncal mononeuropathy—a
new clinical syndrome. Diabetes Care. 1978;1:10–3.
44. Longstreth GF, Newcomer AD.  Abdominal pain
caused by diabetic radiculopathy. Ann Intern Med.
1977;86:166–8.
45. Chaudhuri KR, Wren DR, Werring D, et  al.
Unilateral abdominal muscle herniation with pain: a
distinctive variant of diabetic radiculopathy. Diabet
Med. 1997;14:803–7.
46. O’Connor RC, Andary MT, Russo RB, et  al.
Thoracic radiculopathy. Phys Med Rehabil Clin N
Am. 2002;13:623–44.
47. Alshekhlee A, Tay E, Buczek M, Shakir ZA, et al.
Herpes zoster with motor involvement: discordance
between the distribution of skin rash and localiza-
tion of peripheral nervous system dysfunction. J Clin
Neuromuscul Dis. 2011;12:153–7.
48. Mormont E, Esselinckx W, De Ronde T, et  al.
Abdominal wall weakness and lumboabdominal
pain revealing neuroborreliosis: a report of three
cases. Clin Rheumatol. 2001;20:447–50.
49. Hamodat H, Tran A.  Neurosarcoidosis resulting in
thoracic radiculopathy: a case report. J Med Case
Rep. 2019;13:130.
50. Perrini P, Benedetto N, Buccoliero AM, et al. Thoracic
radiculopathy from a paravertebral mesothelial cyst.
Acta Neurochir (Wien). 2006;148:989–91.
51. Kim Y, Kim S, Oh K, et al. Thoracic radiculopathy
as initial symptoms of hepatocellular carcinoma: a
case report. Medicine (Baltimore). 2018;97:e11635.
52. Choi HE, Shin MH, Jo GY, et  al. Thoracic radicu-
lopathy due to rare causes. Ann Rehabil Med.
2016;40:534–9.
53. Fitzpatrick MO, Goyal K, Johnston RA.  Thoracic
radiculopathy caused by a myodil cyst. Br J
Neurosurg. 2000;14:351–3.
179
54. Massey EW.  Diabetic truncal mononeuropathy:
electromyographic evaluation. Acta Diabetol Lat.
1980;17:269–72.
55. Vinik A, Mehrabyan A, Colen L, et al. Focal entrap-
ment neuropathies in diabetes. Diabetes Care.
2004;27:1783–8.
56. Rydberg M, Zimmerman M, Gottsäter A, et  al.
Diabetes mellitus as a risk factor for compres-
sion neuropathy: a longitudinal cohort study from
southern Sweden. BMJ Open Diabetes Res Care.
2020;8:e001298.
57. Moon HI, Shin J, Kim YW, et al. Diabetic polyneu-
ropathy and the risk of developing carpal tunnel
syndrome: a nationwide, population-based study.
Muscle Nerve. 2020;62:208–13.
58. Mojaddidi MA, Ahmed MS, Ali R, et al. Molecular
and pathological studies in the posterior interosseous
nerve of diabetic and non-diabetic patients with car-
pal tunnel syndrome. Diabetologia. 2014;57:1711–9.
59. Breiner A, Qrimli M, Ebadi H, et  al. Peripheral
nerve high-resolution ultrasound in diabetes. Muscle
Nerve. 2017;55:171–8.
60. Pitarokoili K, Kerasnoudis A, Behrendt V, et  al.
Facing the diagnostic challenge: nerve ultrasound
in diabetic patients with neuropathic symptoms.
Muscle Nerve. 2016;54:18–24.
61. Figueroa-Romero C, Sadidi M, Feldman
EL.  Mechanisms of disease: the oxidative stress
theory of diabetic neuropathy. Rev Endocr Metab
Disord. 2008;9:301–14.
62. Mediouni Z, Bodin J, Dale AM, et  al. Carpal tun-
nel syndrome and computer exposure at work in
two large complementary cohorts. BMJ Open.
2015;5:e008156.
63. Perkins BA, Olaleye D, Bril V.  Carpal tunnel syn-
drome in patients with diabetic polyneuropathy.
Diabetes Care. 2002;25:565–9.
64. Preston DC, Shapiro BE. Median neuropathy at the
wrist. In: Electromyography and neuromuscular
disorders: clinical-electrophysiologic-ultrasound
correlations. 4th ed. Philadelphia: Elsevier; 2021.
p. 323–57.
65. Smith BE.  Focal and entrapment neuropathies.
Handb Clin Neurol. 2014;126:31–43.
66. Yoshii Y, Zhao C, Amadio PC. Recent advances in
ultrasound diagnosis of carpal tunnel syndrome.
Diagnostics (Basel). 2020;10:596.
67. Thomsen NOB, Andersson GS, Björk J, et  al.
Neurophysiological recovery 5 years after carpal
tunnel release in patients with diabetes. Muscle
Nerve. 2017;56:E59–64.
68. Werner BC, Teran VA, Deal DN.  Patient-related
risk factors for infection following open carpal tun-
nel release: an analysis of over 450,000 Medicare
patients. J Hand Surg Am. 2018;43:214–9.
69. Mondelli M, Aretini A, Rossi S.  Ulnar neuropathy
at the elbow in diabetes. Am J Phys Med Rehabil.
2009;88:278–85.
70. Rota E, Zavaroni D, Parietti L, et al. Ulnar entrap-
ment neuropathy in patients with type 2 diabetes
180
mellitus: an electrodiagnostic study. Diabetes Res
Clin Pract. 2014;104:73–8.
71. Jang JE, Kim YT, Park BK, et al. Subclinical ulnar
neuropathy at the elbow in diabetic patients. Ann
Rehabil Med. 2014;38:64–71.
72. Preston DC, Shapiro BE.  Ulnar neuropathy at the
elbow. In: Electromyography and neuromuscular
disorders: clinical-electrophysiologic-ultrasound
correlations. 4th ed. Philadelphia: Elsevier; 2021.
p. 372–401.
73. Schady W, Abuaisha B, Boulton AJ.  Observations
on severe ulnar neuropathy in diabetes. J Diabetes
Complications. 1998;12:128–32.
74. Pelosi L, Yong V, Blumhardt LD.  Utility of nerve
ultrasound in ulnar neuropathy with abnormal non-­
localizing electrophysiology in diabetes mellitus.
Neurophysiol Clin. 2020;50:387–90.
75. Schertz M, Mutschler C, Masmejean E, et  al.
High-resolution ultrasound in etiological evalua-
tion of ulnar neuropathy at the elbow. Eur J Radiol.
2017;95:111–7.
76. Chen J, Wang CL, Wu S, et  al. The feasibility of
using high-resolution ultrasonography to assess
ulnar nerve in patients with diabetes mellitus. J
Ultrason. 2017;17:160–6.
77. Nakashian MN, Ireland D, Kane PM. Cubital tunnel
syndrome: current concepts. Curr Rev Musculoskelet
Med. 2020;13:520–4.
78. Acosta JA, Hoffman SN, Raynor EM, et  al. Ulnar
neuropathy in the forearm: a possible complication
of diabetes mellitus. Muscle Nerve. 2003;28:40–5.
79. Katirji MB.  Compressive and entrapment neu-
ropathies of the lower extremity. In: Katirji MB,
Kaminski HJ, Ruff RL, editors. Neuromuscular
disorders in clinical practice. 2nd ed. New  York:
Springer; 2014. p. 903–52.
80. Katirji MB, Wilbourn AJ. Common peroneal mono-
neuropathy: a clinical and electrophysiologic study
of 116 lesions. Neurology. 1988;38:1723–8.
81. Mulder DW, Lambert EH, Bastron JA, et al. The neu-
ropathies associated with diabetes mellitus. A clini-
cal and electromyographic study of 103 unselected
diabetic patients. Neurology. 1961;11:275–84.
82. Desmottes MC, Brehier Q, Bertolini E, et  al.
Compression of the common peroneal nerve due to
peroneal muscle infarction in a patient with diabetes.
Joint Bone Spine. 2019;86:265–6.
83. Bowley MP, Doughty CT.  Entrapment neuropa-
thies of the lower extremity. Med Clin North Am.
2019;103:371–82.
84. Hobson-Webb LD, Juel VC.  Common entrap-
ment neuropathies. Continuum (Minneap Minn).
2017; 23(2, Selected Topics in Outpatient
Neurology):487–511.
85. Seror P, Seror R. Meralgia paresthetica: clinical and
electrophysiological diagnosis in 120 cases. Muscle
Nerve. 2006;33:650–4.
86. Parisi TJ, Mandrekar J, Dyck PJB, et  al. Meralgia
paresthetica: relation to obesity, advanced age, and
diabetes mellitus. Neurology. 2011;77:1538–42.
P. Thaisetthawatkul and P. J. B. Dyck
87. van Slobbe AM, Bohnen AM, Bernsen RM, et  al.
Incidence rates and determinants in meralgia pares-
thetica in general practice. J Neurol. 2004;251:294–7.
88. Berini SE, Spinner RJ, Jentoft ME, et  al. Chronic
meralgia paresthetica and neurectomy: a clinical
pathologic study. Neurology. 2014;82:1551–5.
89. Oh S.  Clinical electromyography. 3rd ed.
Philadelphia: Lippincott Williams and Wilkins; 2003.
90. Greco D, Gambina F, Pisciotta M, et  al. Clinical
characteristics and associated comorbidities in dia-
betic patients with cranial nerve palsies. J Endocrinol
Invest. 2012;35:146–9.
91. Trigler L, Siatkowski RM, Oster AS, et  al.
Retinopathy in patients with diabetic ophthalmople-
gia. Ophthalmology. 2003;110:1545–50.
92. Greco D, Gambina F, Maggio F.  Ophthalmoplegia
in diabetes mellitus: a retrospective study. Acta
Diabetol. 2009;46:23–6.
93. Rush JA, Younge BR. Paralysis of cranial nerves III,
IV, and VI. Cause and prognosis in 1,000 cases. Arch
Ophthalmol. 1981;99:76–9.
94. Keane JR.  Third nerve palsy: analysis of 1400
personally-examined inpatients. Can J Neurol Sci.
2010;37:662–70.
95. Smith BE. Cranial neuropathy in diabetes mellitus.
In: Dyck PJ, Thomas PK, editors. Diabetic neu-
ropathy. 2nd ed. Philadelphia: WB Saunders; 1999.
p. 457–67.
96. Wilker SC, Rucker JC, Newman NJ, et  al. Pain in
ischaemic ocular motor cranial nerve palsies. Br J
Ophthalmol. 2009;93:1657–9.
97. Tamhankar MA, Biousse V, Ying GS, et al. Isolated
third, fourth, and sixth cranial nerve palsies from
presumed microvascular versus other causes: a pro-
spective study. Ophthalmology. 2013;120:2264–9.
98. Dreyfus PM, Hakim S, Adams RD.  Diabetic oph-
thalmoplegia; report of case, with postmortem
study and comments on vascular supply of human
oculomotor nerve. AMA Arch Neurol Psychiatry.
1957;77:337–49.
99. Asbury AK, Aldredge H, Hershberg R, et  al.
Oculomotor palsy in diabetes mellitus: a clinico-­
pathological study. Brain. 1970;93:555–66.
100. Weber RB, Daroff RB, Mackey EA.  Pathology of
oculomotor nerve palsy in diabetics. Neurology.
1970;20:835–8.
101. Sunderland S, Hughes ES.  The pupillo-constrictor
pathway and the nerves to the ocular muscles in
man. Brain. 1946;69:301–9.
102. Jung JS, Kim DH. Risk factors and prognosis of iso-
lated ischemic third, fourth, or sixth cranial nerve
palsies in the Korean population. J Neuroophthalmol.
2015;35:37–40.
103. Park SJ, Yang HK, Byun SJ, et al. Risk of ischemic
stroke after third, fourth, and sixth cranial nerve pal-
sies in type 2 diabetes. J Diabetes. 2019;11:379–85.
104. Fang C, Leavitt JA, Hodge DO, et  al. Incidence
and etiologies of acquired third nerve palsy using
a population-based method. JAMA Ophthalmol.
2017;135:23–8.
Asymmetric Diabetic Neuropathy: Radiculoplexus Neuropathies, Mononeuropathies, and Cranial…
105. Glass GE, Tzafetta K.  Bell’s palsy: a summary of
current evidence and referral algorithm. Fam Pract.
2014;31:631–42.
106. Adour KK, Bell DN. Incidence of diabetes mellitus
in Bell’s palsy. Lancet. 1971;2:168.
107. Adour K, Wingerd J, Doty HE. Prevalence of con-
current diabetes mellitus and idiopathic facial paral-
ysis (Bell’s palsy). Diabetes. 1975;24:449–51.
108. Brandenburg NA, Annegers JF.  Incidence and risk
factors for Bell’s palsy in Laredo, Texas: 1974-1982.
Neuroepidemiology. 1993;12:313–25.
109. Zhao H, Zhang X, Tang YD, Zhu J, Wang XH, Li
ST.  Bell’s palsy: clinical analysis of 372 cases
and review of related literature. Eur Neurol.
2017;77:168–72.
110. Albers JR, Tamang S. Common questions about Bell
palsy. Am Fam Physician. 2014;89:209–12.
111. Marioni G, Rinaldi R, de Filippis C, et  al.
Candidal abscess of the parotid gland associated
with facial nerve paralysis. Acta Otolaryngol.
2003;123:661–3.
181
112. Al-Ammar A, Maqbool MS.  Cervical necrotizing
fasciitis with facial nerve paralysis. J Laryngol Otol.
2004;118:573–5.
113. Franco-Vidal V, Blanchet H, Bebear C, et  al.
Necrotizing external otitis: a report of 46 cases. Otol
Neurotol. 2007;28:771–3.
114. Chandu A, MacIsaac RJ, MacGregor DP, Campbell
MC, Wilson MJ, Bach LA.  A case of mucormy-
cosis limited to the parotid gland. Head Neck.
2005;27:1108–11.
115. Gochhait D, Alexender A, Dehuri P, et al. Mucormycotic
parotitis and otitis causing facial nerve palsy in a dia-
betic patient. Cytopathology. 2018;29:582–4.
116. Eshbaugh CG, Siatkowski RM, Smith JL, et  al.
Simultaneous, multiple cranial neuropathies in dia-
betes mellitus. J Neuroophthalmol. 1995;15:219–24.
117. Şevik Eliçora S, Erdem D. Does type 2 diabetes mel-
litus affect the healing of Bell’s palsy in adults? Can
J Diabetes. 2018;42:433–6.
 Motor Neuropathy in Diabetes
Karolina Snopek Khan and Henning Andersen
1 Introduction
Distal symmetric polyneuropathy (DSPN) is the
most common clinical presentation of diabetic
neuropathy [1]. Motor and sensory nerves are
concomitantly affected early in the course of
DSPN when examined by nerve conduction stud-
ies (NCS) [2]. However, clinically motor neurop-
athy first becomes apparent later in the course of
DSPN [3]. This more advanced stage of DSPN
has been denoted type 2b by Dyck et al., defined
by the inability to stand on heels [4]. In
population-­based studies of individuals with type
1 and type 2 diabetes, this level of severe DSPN
was reported in only 6 and 1% of patients, respec-
tively [4]. In recent years, there has been growing
interest in motor dysfunction and muscle weak-
ness in diabetes and in relation to DSPN. Motor
neuropathy leads to muscle weakness and atro-
phy, which ultimately results in motor impair-
ment. Motor neuropathy can be diagnosed based
on a combination of clinical tests and
NCS.  Several methods have been applied to
assess muscle size and quality, including imaging
techniques such as ultrasonography (USG), com-
puted tomography (CT) and magnetic resonance
K. S. Khan · H. Andersen (*)
Department of Neurology, Aarhus University
Hospital, University of Aarhus, Aarhus, Denmark
e-mail: hena@clin.au.dk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_11
imaging (MRI), and histological evaluation based
on muscle biopsies. However, in a clinical set-
ting, these methods are only very rarely imple-
mented, and are mainly used for research
purposes. Identifying motor neuropathy in diabe-
tes is of importance as it may lead to motor dys-
function [5] and has serious clinical consequences
with an increased risk of foot ulcers [4], unstable
gait [6], postural instability [7], frequent falls [8,
9], and impaired joint mobility [10, 11] leading to
increased morbidity. Currently, there are no spe-
cific treatments for motor neuropathy. However,
recent evidence has emerged showing benefits
from training interventions for motor dysfunction
in individuals with DSPN [12–14]. In recent
years new evidence suggests the concomitant
existence and development of a myopathy in dia-
betes [15]. It has been speculated that these
changes may further explain the observed motor
dysfunction and muscle weakness seen in indi-
viduals with diabetes.
In this chapter, we deal with all aspects of
motor neuropathy, including clinical presenta-
tion, pathophysiology, muscle strength, diagnos-
tic tools, functional consequences of motor
impairment, and effects of exercise.
183
184 K. S. Khan and H. Andersen

2 Clinical Presentation viduals with diabetes, including mononeuropa-

In DSPN, sensory signs usually precede clinical
motor dysfunction. Individuals with DSPN and
motor involvement usually do not complain of
muscle weakness until more advanced stages of
the disease, and therefore it may go unnoticed for
a long time.
When motor dysfunction becomes apparent,
signs of sensory abnormalities are usually evi-
dent at the clinical examination. As with sensory
disturbances, motor symptoms and signs ini-
tially present distally in the lower extremities
and progress more proximally. Other less fre-
quent presentations of diabetic neuropathy
should be kept in mind when examining indi-
thies, radiculopathies, and immune-mediated
neuropathies (Table 1). DSPN is a diagnosis of
exclusion, and therefore screening for other
causes of polyneuropathy and a thorough medi-
cal history should be obtained. Motor neuropa-
thies in diabetes with an asymmetrical pattern,
proximal distribution, motor predominant
involvement or an acute onset, should be con-
sidered atypical, requiring thorough examina-
tions for other causes such as inherited
polyneuropathies, motor neuron disease and
acquired immune-mediated polyneuropathies
(Table  1). Furthermore, age-related muscle
weakness due to atrophy and sarcopenia are
other etiologies that need consideration.

Table 1  Atypical diabetic neuropathies and other motor neuropathies

Distribution and
characteristics Clinical features Progression Electrodiagnostic studies
Diabetic
radiculoplexus
neuropathies:
 Lumbosacral Often unilateral Pain followed by Subacute Predominantly axonal
or asymmetric, proximal weakness of degeneration with segmental
may present the lower limb demyelination
with muscles often
symmetrical combined with weight
distribution loss
 Cervicobrachial Asymmetric May involve proximal Predominantly axonal
upper extremity degeneration with segmental
muscles demyelination
 Thoracic May be Pain and dysesthesias EMG of paraspinal muscles
radiculoneuropathy symmetric, involving the chest/ shows signs of denervation
involving abdominal wall can
multiple lead to respiratory
dermatomes weakness
Mononeuropathies Asymmetric Cranial nerve palsies Slow Demyelinating at typical
(oculomotor nerve sites of compression
and facial nerve), in
limbs most
commonly: median,
ulnar and peroneal
nerve
Multifocal motor Asymmetric, Distal muscle Slow Conduction block followed
neuropathy pure motor weakness by axonal degeneration
Motor neuron disease Initially Asymmetric limb Progresses at Initially normal ENG. As
asymmetric, weakness in 80%, and different atrophy becomes severe,
pure motor bulbar weakness in rates, usually axonal loss is seen. EMG
20%, and 1–3% with fast discloses signs of acute and
respiratory weakness chronic muscle denervation
in several muscles and body
regions
Motor Neuropathy in Diabetes 185

Table 1 (continued)
Distribution and
characteristics Clinical features Progression Electrodiagnostic studies
Inflammatory/ Symmetric, AIDP: Hypo- or Acute AIDP: demyelinating with
immune-mediated muscle areflexia sural sparring
Guillain Barre weakness AMAN: Distal and AMAN: motor axonal
Syndrome proximal muscle AMSAN: sensory and motor
weakness axonal
AMSAN: Distal and
proximal muscle
weakness
Chronic inflammatory Symmetric Distal and proximal Subacute and Demyelinating
demyelinating muscle weakness reoccurring
polyradiculoneuropathy
Radiculopathy Asymmetric Depending on Acute ENG: usually normal
location, characterized EMG: axonal denervation
by pain, muscle
weakness, numbness,
and tingling sensation
Hereditary:
  CMT type 1 Symmetric Distal to proximal Slow CMT1: Demyelinating
  CMT type 2 muscle weakness, CMT2: Axonal
severe distal muscle
atrophy, and high foot
arch

Complaints and clinical signs gradually 3 Pathophysiology

become more evident as the disease slowly pro-
gresses. When motor neuropathy becomes clini-
cally apparent, complaints of symptoms from the
lower extremities are typical, such as weakness
of the feet and legs, unstable gait, and difficulty
walking. In the functional and clinical examina-
tion, individuals usually present with an inability
to stand on heels, decreased strength at the toes
and ankles, decreased or absent distal reflexes,
pes cavus, muscle twitching, and muscle cramp-
ing. In the advanced stages of DSPN, proximal
muscles in the lower extremities and the thighs
and even in the upper extremities may become
involved.
Individuals may develop limited mobility
due to changes in connective and subcutaneous
tissue leading to contractures and hammertoes.
This may result in walking difficulties and
impairment of performing daily activities. As
the disease progresses proximally, weakness
may include the upper extremities, much like
the glove-and-­stocking distribution of sensory
neuropathy.
The pathophysiology of motor neuropathy
remains unknown. The muscle weakness in
DSPN can be attributed to both neurogenic and
myopathic abnormalities in the distal muscles of
the lower limbs, which may contribute to
decreased muscle quality and abnormal muscle
structure [15].
In DSPN, there is a combination of axonal
degeneration and demyelination affecting both
motor and sensory nerves. Loss of motor axons
leads to denervation of muscle fibers, causing
muscle weakness, muscular atrophy [16], and
increased fatigability [17], all of which is corre-
lated to the extent and severity of diabetic neu-
ropathy [18, 19]. Chronic hyperglycemia,
hypercholesterolemia, and hypertriglyceridemia
are believed to be the main causes of these patho-
logical changes [1]. Severe demyelination with-
out axonal loss is very rarely seen in DSPN and
suggests concomitant neuropathy, due to an etiol-
ogy other than diabetes. Many factors such as a
high BMI, inactivity, and myopathic changes in
186
the muscles can contribute to the development of
motor impairment.
4 Neurogenic Muscle
Weakness
In DSPN, muscle weakness is related to the
severity of the neuropathy and develops due to
degeneration of motor axons or entire motor
units leading to subsequent muscle fiber loss
with changes in muscle morphology and atro-
phy [20–22]. In contrast to the clinical presenta-
tion, in NCS, motor and sensory nerves are
impaired to a similar degree in the earlier phases
of DPN [23].
This indicates that more subtle clinical signs
of motor dysfunction are not primarily due to
less severe pathology in motor nerves but rather
is the consequence of the compensatory mecha-
nism named collateral re-innervation. Thereby,
denervated muscle fibers are re-innervated by
neighboring motor nerves. However, as this re-
innervation process becomes increasingly inad-
equate, accelerated loss of muscle strength
occurs [21]. If denervated muscle fibers fail to
acquire a new source of innervation, they atro-
phy and undergo apoptosis and ultimately necro-
sis with replacement by intramuscular fat
deposits and other non-contractile tissue.
Furthermore, reduced levels of neurotrophic fac-
tors in striated muscles of individuals with DPN
is related to muscle weakness and neuropathy.
Thus, denervation of muscles in diabetes may be
further ­propagated by reduced levels of muscular
trophic factors and nerve growth factors [24].
5 Diabetic Myopathy
More recently, there has been growing evidence
suggesting the existence of a diabetic myopathy
[15]. Diabetic myopathy is characterized by an
impairment in skeletal muscle structure, func-
tion, and metabolic capacity, which has been
described as a distinct process in diabetic indi-
viduals unrelated to DSPN.
K. S. Khan and H. Andersen
Myopathy is characterized by fibrosis and
atrophy of the muscle. This process causes loss
of energy reserves, decreased muscle quality,
and abnormal muscle structure, ultimately lead-
ing to muscle weakness and functional impair-
ments. Loss of muscle strength in myopathy is a
multifactorial process, which cannot be
explained by muscle atrophy or muscle disuse
alone. Other factors such as non-contractile tis-
sue infiltration and reduced muscle quality of
myocellular proteins may also contribute to
muscle weakness. Individuals with diabetic
myopathy may appear clinically with normal
muscle mass, and nonetheless have lower muscle
strength due to lower quality of the muscles.
Muscular atrophy and hypoxia of the muscle
may contribute through impairment of mito-
chondrial function. Altered mitochondrial func-
tion contributes to insulin resistance of skeletal
muscles due to lipid accumulation within myo-
cytes, leading to a dysfunction of oxidative
enzymes. Abnormal mitochondrial morphology
in diabetes is also evident in tissues other than
muscle, including neurons. However, currently,
there are no studies available in humans, and
most research was performed on diabetic mice.
Another explanation for the development of dia-
betic myopathy is recurrent administration of
subcutaneous insulin [15]. This is thought to
produce a chronic peripheral glucose overload
causing a state of constant intracellular hyper-
glycemia contributing to the development of dia-
betic myopathy.
6 Diagnostic Approach
For the diagnosis of motor neuropathy, the cor-
nerstone is a clinical examination, including
manual muscle testing combined with electro-
physiological examinations. Most validated neu-
ropathy scales assessing DSPN do not focus on
motor involvement. In the Toronto criteria for
confirmed DSPN, early motor symptoms are not
included in the standardized assessment for
DSPN; therefore, motor involvement may be
overlooked. Currently, the Toronto clinical neu-
Motor Neuropathy in Diabetes
ropathy score is the only clinical diabetic neu-
ropathy scale which includes symptoms of
muscle weakness [25].
7 Motor Nerve Dysfunction
Motor nerves can be evaluated using neurophysi-
ological tests such as electroneurography (ENG)
and electromyography (EMG).
In individuals with typical bilateral, distal,
symmetric polyneuropathy that present with both
signs and symptoms of DSPN, there is no need
for further ancillary tests to make the diagnoses
at the level of possible and probable [26].
Ancillary investigations with NCS are needed to
confirm the diagnosis of typical DPSN, to diag-
nose subclinical DSPN, and to diagnose individ-
uals with atypical features and presentations, or
when the diagnosis is uncertain [1, 26, 27].
Confirmation of typical DSPN is usually only
needed for research purposes [26, 28]. NCS can
also be used to quantify the severity of neuropa-
thy [2] and to measure disease progression [29].
Standard ENG examination includes an exami-
nation of both motor and sensory nerves. ENG
techniques, methods, materials, and equipment
used may differ from examiner to examiner.
However, there is some consensus on how an ENG
examination should be performed and interpreted
in DSPN [30]. Generally, a NCS examination of
the lower limb nerves, tibial, peroneal, and sural, is
sufficient. In screening for typical DSPN, a unilat-
eral examination is considered sufficient, provided
that the nerves are either abnormal or normal. The
examination should be modified according to the
individual's signs and symptoms in order to accu-
rately diagnose immune-mediated neuropathies,
mononeuropathies, and radiculopathies (Table 1).
With a standard ENG, the following NCS
parameters are evaluated: Distal motor latency,
motor/sensory conduction velocity, compound
muscle action potential/sensory nerve action
potential amplitude, and minimum F-wave
latency [30].
187
In DSPN, a classical feature seen on ENG is
decreased amplitudes reflecting a loss of axons
combined with slowing in conduction velocity
reflecting abnormalities of the myelin sheath
[31]. Since typical diabetic neuropathy presents
with features of both demyelination and axonal
loss, this is categorized as a mixed neuropathy.
Abnormalities in sural nerve conduction is classi-
cally thought to be a sensitive measure to confirm
the diagnosis [32]. An unrecordable sural nerve
action potential is a frequent finding in DSPN
and can be interpreted as abnormal [33]. F-wave
conduction velocity is an antidromic remnant of
the M-wave and is therefore only present in
motor nerves. It is considered a valuable mea-
surement as it reflects pathology along the entire
nerve and the anterior horn of the spinal cord.
Absent F-waves or prolonged F-wave latency are
thought to be sensitive and reproducible mea-
sures of early DSPN [20, 34, 35]. This further
underscores the notion that motor nerves are
impaired to the same degree as the sensory nerves
in DSPN.
Abnormalities in NCS do not differ between
type 1 and type 2 diabetes [36]; although the lit-
erature in this area is sparse. Balducci et  al.
showed that while muscle strength was more
decreased in patients with type 2 diabetes com-
pared to type 1 diabetes, there was no difference
in NCS measurements [37]. Whether there are
differences in the pathogenesis of type 1 and type
2 DPN is unclear; therefore, nuances and differ-
ences may exist in some studies. A histopatho-
logical study in rats showed that small-fiber
involvement might happen earlier in type 1 dia-
betes [38]. However, this finding is not evident in
conventional NCS, as standard investigations
have limitations and only detect large-fiber
abnormalities. Individuals with isolated small-­
fiber or autonomic neuropathy may have a nor-
mal ENG and a close to normal clinical
examination, requiring ancillary diagnostics with
intraepidermal nerve fiber density (IENFD),
quantitative sensory testing, and corneal confocal
microscopy [39–42].
188
8 Electromyography 9 Muscle Strength in Diabetic
Due to the length-dependent process in DSPN, atro-
phy and weakness of foot muscles are usually the
first clinical signs of motor involvement, followed
by weakness of ankle dorsiflexion resulting in foot
drop. With further progression of the disease, hand
and forearm muscles may be involved. In the clini-
cal setting, EMG can be used to distinguish between
neurogenic weakness, due to axonal loss versus
myogenic weakness from myopathy, and may also
aid as a supplementary investigation in suspicion of
a nerve root lesion. Preferably, EMG is performed
in distal muscles using a concentric needle, as sur-
face electrode EMG cannot quantify muscle qual-
ity. EMG should be performed in both a resting and
an active muscle to evaluate motor unit action
potential, recruitment, and interference patterns. As
described above, loss of muscle strength in diabetes
can be the consequence of both a primary muscular
pathology (diabetic myopathy) [15], or secondary
pathology due to axonal loss leading to neurogenic
weakness. Since the nerve damage found on ENG
is accompanied by neurogenic weakness found on
EMG, performing an EMG is most often not neces-
sary in a clinical setting. Typical EMG findings in
diabetic neurogenic weakness include; resting
fibrillation potentials, reduced recruitment patterns,
and broader, higher, polyphasic motor unit poten-
tials [29, 43]. These findings also correspond with
terminal axonal sprouting resulting in compensa-
tory re-innervation.
Skin biopsies provide an estimate on the num-
ber of distal small nerve fiber segments; however,
for obvious reasons, conventional EMG does not
provide a direct count of the number of motor
axons. Novel methods such as MUNE can
­accurately estimate the entire number of motor
units within a muscle by using surface electrodes
[44]. In a recent study, M-Scan MUNE evalua-
tions were performed in 21 individuals with
DPN, 31 individuals with diabetes without DPN,
and 38 heathy controls. The authors were able to
detect axonal motor nerve damage when collat-
eral re-­innervation limited NCS changes, and the
sensitivity of detecting motor involvement in
DPN, was similar to that of sensory NCS [45].
K. S. Khan and H. Andersen
Distal Symmetric
Polyneuropathy
Individuals with diabetes have reduced muscle
strength when compared to healthy matched
controls [19, 37, 46]. This includes muscle
groups in both upper and lower extremities. The
lack of quantitative methods has disadvantaged
earlier studies when assessing muscle strength.
However, multiple recent studies have confirmed
muscle weakness in diabetes combined with
findings documenting more pronounced muscle
weakness in individuals with DSPN.
Furthermore, in DSPN, muscle weakness is pro-
gressive and related to the presence and espe-
cially the severity of DSPN with a distal to
proximal distribution [18, 47]. Orlando et  al.
confirmed a correlation between motor and sen-
sory nerve conduction velocity and muscle
strength, and fatigability [48]. Several newer
studies have confirmed previous findings, indi-
cating that the severity of DSPN is associated
with foot and ankle muscle weakness and a
reduced range of motion of the lower extremities
[49, 50]. In a recent publication, Orlando et al.
presented a review of the available literature sup-
porting the evidence for reduced muscle strength,
altered muscular endurance, and muscle fiber
composition in DSPN [14].
Increased muscle endurance has been reported
in long-term type 1 diabetes individuals [17, 51].
However, other studies have shown opposite
findings [14]. It has been speculated that changes
in muscle fiber type distributions could explain
the increased muscular endurance. Allen et  al.
2014 described a loss of fast-twitch motor units
and a decreased rate of torque development with
prolonged contractions in individuals with type 2
diabetes and DSPN [52].
Age-related sarcopenia may contribute to
decreased muscle size with increased intramus-
cular adipose tissue leading to alterations in glu-
cose disposal, propagating the development of
type 2 diabetes [53]. Some studies have shown
that individuals with obesity and diabetes develop
muscle weakness and increased intramuscular
Motor Neuropathy in Diabetes 189

fat-accumulations, with hyperglycemia being the 10 Assessment of Muscle

main factor driving this mechanism [54, 55]. Strength
Park et  al., described that decreased muscle
strength was related to longer diabetes duration Muscle strength is defined as the maximum force
and poor glycemic control [19]. a muscle group or an individual muscle can gen-
In a recent meta-analysis, Gundmi et  al. erate [62]. Strength and force output is correlated
found that individuals with type 2 diabetes have to the cross-sectional area of muscle fibers
impaired hand function measured by grip and recruited to generate force and relies on motor
pinch strength when compared to healthy con- unit recruitment [63]. There is a large inter-­
trols [56]. Gutefeldt et  al. evaluated upper individual variation in muscle strength which can
extremity impairment in individuals with long- be explained primarily by differences in muscle
standing type 1 diabetes with and without DPN fiber types, training experience, age, sex, and
[57], and found that two-thirds of the individu- genetic capability.
als had one or more uni- or bilateral clinical Muscle strength can be evaluated by [64]:
signs of upper extremity impairment which cor-
related to a reduced grip force. However, in this 1. Functional tests
study, it remained unknown if the individuals 2. Manual muscle testing
had DPN and it would be speculative to attri- 3. Quantitative techniques.
bute upper extremity impairment to DPN. Lima
et  al. reported that individuals with DPN did Functional tests are easy to perform as they do
not have decreased hand force generation but not require any equipment, and therefore are suit-
impaired hand dexterity and handgrip force able for large-scale population-based studies, e.g.
control [58]. 6-min walk test (Fig.  1). Testing the ability to
Associations between diabetes and decreased
lung function have been reported [53]. Kabitz
et  al. provided one of the first descriptions of
impaired respiratory neuromuscular function
related to DSPN assessed by phrenic nerve stim-
ulation [59]. In a more recent study, Van Eetvelde
et al. described an association between the pres-
ence of DSPN and respiratory muscle strength in
individuals with type 2 diabetes compared to
healthy controls [60]. However, in their study,
results were not stratified according to DSPN;
therefore, no firm conclusion could be made
whether DSPN had a negative impact on respira-
tory muscle strength and function.
In a recent study by Venkataraman et al., indi-
viduals with DSPN had decreased activation of
the motor coordination and visual processing
regions, possibly caused by a reduced sensory
input from the foot [61]. They also found signs of
compensatory activity within the CNS, by greater
activation of motor coordination and sensory
facilitation regions, in individuals with DSPN
having lower muscle strength and decreased joint
mobility. Fig. 1  Six-minute walk test
190
Fig. 2  Dynamometry enables testing of both isometric
and isokinetic movements
walk on heels and toes reflects the strength of the
ankle dorsal and plantar flexors, while rising
from a kneeling position reflects more proximal
muscle strength, including the hip and knee
extensors, and flexors. Although easy to perform,
functional tests have limitations. The exact rela-
tion to muscle strength is unclear, and the inabil-
ity to perform these tests may be due to pain,
obesity, postural instability, and other comorbidi-
ties unrelated to motor dysfunction.
Manual muscle testing (MMT) is part of a stan-
dard neurological examination and provides semi-
quantitative data expressed on an ordinal scale
reflecting muscle strength. MMT requires an expe-
rienced examiner, and even then, manual testing is
insensitive compared to quantitative techniques.
MMT has limited sensitivity in detecting mild to
moderate symmetrical weakness and has adequate
sensitivity in cases with severe weakness.
K. S. Khan and H. Andersen
Dynamometry is a technique enabling testing
with both isometric and isokinetic movements
(Fig. 2). It provides quantitative information on a
linear scale reflecting static and dynamic strength
of all major muscle groups of the upper and lower
extremities. The main limitation of the test is that
the individual being tested has to be fully coop-
erative and factors such as lack of motivation,
pain, depression, and fatigue, may lead to sub-
maximal performance. Furthermore, dynamom-
etry is time-consuming, and equipment is
expensive, and usually only exists at exercise-­
science or neuromuscular specialist research
facilities.
11 Imaging of Muscular
Atrophy and Motor
Neuropathy
In DSPN with motor impairment, initial pre-
sentation is with atrophy of striated muscles
occurring predominantly involving the smaller
muscles of the distal lower extremities. In later,
more advanced stages, proximal involvement
occurs. Muscle size can be quantitatively
determined with CT, USG, and MRI. USG is a
fast and inexpensive method, but has lower
spatial resolution. CT is also a quick examina-
tion, but has low tissue contrast and involves
radiation exposure. A recent study by
Henderson et  al. evaluated muscle size and
muscle strength of the lower limbs by the use
of USG in 15 individuals with diabetes and
early DPN compared to healthy controls [65].
They found that intrinsic muscles of the lower
limb were significantly smaller in the DPN
groups, which correlated to decreased muscle
strength. Henderson suggested that USG is a
cheap and easily re-assessable method that is
useful for the detection of progression of mus-
cle wasting in DPN.  In a previous study,
Severinsen et  al. showed that USG evaluation
of small foot muscles identified considerable
Motor Neuropathy in Diabetes
atrophy which was related to the severity of
DPN based on NCS [66].
MRI has the highest resolution and can be
combined with quantitative techniques that
enable a reliable estimation of muscle volume
and the degree of fatty infiltration, extent of
extracellular edema, and changes in muscle
microstructure. Furthermore, muscle energy
properties have been studied using 31P-MRI
spectroscopy and show lower energy proper-
ties in individuals with DSPN. MRI has several
disadvantages as part of a standardized diag-
nostic tool, including high costs, lower avail-
ability of equipment, and time-consuming
scanning protocols. Individuals with pacemak-
ers, severe obesity, and claustrophobia are dif-
ficult to evaluate by MRI.  In neuromuscular
diseases, MRI studies of striated muscles have
focused on myopathies, whereas in DSPN,
MRI has been applied for research purposes
only.
MRI enables quantification of lower limb
muscular atrophy in individuals with DSPN [16].
Andersen et al. found a reduced volume of foot
muscles on MRI in neuropathic and non-­
neuropathic individuals, suggesting that foot
muscle atrophy can be an early phenomenon,
occurring even before clinical signs of DSPN are
evident [16, 67]. In studies of individuals with
more severe DSPN, there was also a considerable
loss of striated muscle volume in the lower
extremities with a clear distal to proximal gradi-
ent [18, 47]. In long-term follow-up studies, there
was evidence of progressive muscular atrophy
and accelerated loss of muscle tissue and muscle
strength [68].
In another study, individuals with DSPN had
impaired cellular metabolism with decreased
energy reserves in foot muscles, lower energy
reserves of inorganic phosphates, and a lower
phosphocreatine ratio when compared to healthy
191
individuals. It was suggested that prior to the
development of muscular atrophy, there are sub-
stantial changes in the biochemical properties of
the muscles [69].
Almurdhi et  al. showed unexpectedly that
individuals with type 2 diabetes had proximal
rather than distal muscle weakness, related to the
severity of DSPN.  In their study, MRI revealed
reduced proximal muscle volume and weakness,
whereas distal muscle weakness was unrelated to
lower distal muscle volume [70]. The authors
suggested that the finding was due to an increase
in distal intramuscular fat and non-contractile tis-
sue, which was most pronounced in the plantar
flexors. In a more recent MRI study of 40 indi-
viduals with type 2 diabetes and 20 healthy con-
trols, our research group found that individuals
with DPN had decreased muscle strength and
reduced muscle size compared to individuals
without DSPN, also being most pronounced at
the plantar flexors [71]. The muscle weakness
observed was related mainly to muscle atrophy
and to a lesser extent, to decreased tissue quality,
indicating a distal to proximal distribution as
expected in DSPN (Fig. 3).
To visualize peripheral nerves, USG has been
used to evaluate the more superficial sural nerve
[72], but has limitations when trying to visualize
more deeply situated mixed nerves in the lower
extremities. MRI can be used to evaluate struc-
tural findings in larger mixed nerves and
T2-weighted hyperintensities were more frequent
in individuals with DSPN [73]. Diffusion-tensor-­
imaging MR-Neurography enables the detection
of multifocal lesions of mixed peripheral nerves
in DSPN and enables differentiation between
diabetic muscles with or without DSPN (Fig. 4)
[74]. In the future, higher resolution MR images
combined with automated post-processing may
lead to more widespread use of MRI for monitor-
ing DSPN.
192 K. S. Khan and H. Andersen

a HC –DPN +DPN
Calf Thigh Calf Thigh Calf Thigh

W
T1-weighted 4-point Dixon

b HC –DPN +DPN
Calf Thigh Calf Thigh Calf Thigh

TE = 7.6 ms
T2-mapping

TE = 129.2 ms

c HC –DPN +DPN
Calf Thigh Calf Thigh Calf Thigh

b = 0 s/mm3
DTI

b = 600 s/mm3

Fig. 3  Unprocessed MRI scans at calf and thigh levels in
a healthy control (HC) participant, a participant without
diabetic polyneuropathy (-DPN), and a participant with
DPN (+DPN). (a) Water (W) and fat (F) images obtained
with four-point Dixon sequences were used to calculate
fat-fractions. (b) T2 images of two different echo times
(TEs) (in milliseconds) illustrate the muscle signal loss
from the higher echo times. (c) Diffusion images obtained
with two different b-values, illustrate the muscle signal
loss with higher b-values. (Image slices: brightness,
+40%; contrast +20%). (Adapted with permission from
Radiology [71])
Motor Neuropathy in Diabetes
Fig. 4 (a1) Axial MRIs of the thigh, in healthy controls,
(a2) Axial MRIs of the shin, in healthy controls, (a3) DTI
trace images of the thigh, in healthy controls, (a4) DTI
trace images of the shin, in healthy controls. (b1) Axial
MRIs of the thigh, in a patient with diabetes without DPN,
(b2) Axial MRIs of the shin, in a patient with diabetes
without DPN, (b3) DTI trace images of the thigh, in a
patient with diabetes without DPN, (b4) DTI trace images
193
of the shin, in a patient with diabetes without DPN. (c1)
Axial MRIs of the thigh, in a patient with diabetes and
DPN, (c2) Axial MRIs of the shin, in a patient with diabe-
tes and DPN, (c3) DTI trace images of the thigh, in a
patient with diabetes and DPN, (c4) DTI trace images of
the shin, in a patient with diabetes and DPN. (Adapted
with permission from Diabetes [74])
194
12 Motor Function, Postural
Stability, and Falls
Motor function and postural balance are highly
dependent on muscle strength, and adequate
strength is required to prevent a fall [74]. Postural
stability and proper ambulation rely on numerous
factors, including intact function of the visual,
vestibular, somatosensory, and musculoskeletal
systems [75]. DSPN contributes to decreased
stride length, impaired balance, poor coordina-
tion, and unstable gait [76]. To avoid falls and to
maintain postural stability during unexpected
perturbations of movements or positions, quick
compensatory muscle contractions are required.
These compensatory contractions may be
impaired in individuals with motor neuropathy
due to muscle weakness and muscle atrophy of
the lower limbs. Several studies have reported a
higher frequency of falls in individuals with
DSPN [8, 77, 78]. In a recent review of the cur-
rently available literature on DPN in relation to
falls, postural balance, and activities of daily liv-
ing, we found a robust body of evidence demon-
strating lower postural stability and an increased
frequency of falls in individuals with DSPN [79].
However, only two studies have evaluated the
impact of DSPN on activities of daily living.
Several studies have examined muscle strength
as a possible contributor to postural instability in
diabetes. Simoneau et al. reported [80] an associa-
tion between postural instability and lower limb
muscle weakness in individuals with diabetes, but
found no relationship between muscle strength
and severity of DSPN. Nardone et al. found that
postural instability was related to the impairment
of sensory rather than motor nerve fibers [81] and
this has been supported by Menz et al. [82] and
Almurdhi et al. [83], who reported that peripheral
loss of sensation was the main cause of postural
instability. Handsaker et al. found that individuals
with DSPN had impaired strength generation at
the ankle and knee while walking up and down
stairs [84]. This finding was confirmed by
Richardson [7] and Allet [8] et  al., who argued
that both reduced proximal strength and distal
ankle proprioception predict postural instability,
falls, and fall-related injuries.
K. S. Khan and H. Andersen
In a recent large cross-sectional study, we
found an increased number of reported falls
within the previous 12 months among individuals
with type 2 diabetes compared to healthy con-
trols, irrespective of the presence of DPN [85].
Individuals with type 2 diabetes that reported a
fall had impaired postural stability, slower transi-
tional movements and decreased walking speed.
However, to our surprise, muscle strength and
scores of DPN did not differ between fallers and
non-fallers, although muscle strength correlated
to the degree of postural instability and the degree
of DPN in both fallers and non-fallers. Therefore,
physical tests and posturography, which are
highly dependent on muscle strength, should be
incorporated in the design of future fall preven-
tion screening programs.
13 Interventions
Currently, there are no disease-modifying treat-
ments for diabetic motor neuropathy. Resistance-
and aerobic training has been shown to improve
glycemic and metabolic control, with a lower
HbA1c, higher tissue insulin sensitivity, higher
HDL levels, lower plasma triglycerides, and
reduced BMI, all in parallel with lower fat mass
and waist circumference [86–89]. Based on these
findings, exercise is recommended for the pre-
vention of macro- and microvascular complica-
tions of diabetes.
In individuals without DSPN, it is well estab-
lished that exercise leads to gains in muscle
strength of the lower extremities by increasing not
only the quality of the muscles, but also muscle
mass. Until recently, individuals with DSPN were
advised to refrain from resistance and weight
bearing training due to concerns of foot ulcer-
ation. However, recent studies have disproved this
notion, deeming resistance and weight bearing
training to be safe in individuals with DSPN [90].
Resistance training can prevent loss of stri-
ated muscles and improve muscle strength,
insulin sensitivity, neural control in individuals
with diabetes [91]. Further, resistance training
can induce hypertrophy of particularly type IIa
muscle fibers which enhances glucose utiliza-
Motor Neuropathy in Diabetes
tion and ATP generation. Castaneda et  al.
showed that exercise improves glycemic control
and neuromuscular function in individuals with-
out DPN [87].
Exercise training may be a preventive option to
delay the development of neuropathy in diabetes.
In a prospective 4-year study by Balducci et al.,
individuals with diabetes in the non-­intervention
group developed significantly more motor and
sensory neuropathy compared to individuals per-
forming long-term aerobic exercise training [92].
Physical inactivity is associated with higher intra-
muscular adipose tissue in individuals with DPN
and type 2 diabetes [54]. Exercise combined with
dietary interventions improves proximal cutane-
ous re-innervation and regenerative capacities in
individuals with pre-­ diabetes and neuropathy
[93]. Kluding et al. recently reported that individ-
uals with DPN showed improved pain, neuropa-
thy symptoms, and intraepidermal nerve fiber
branching at proximal sites following aerobic and
resistance exercise [94]. Singleton et  al. found
improvements in IENFD in individuals with dia-
betes without neuropathy following combined
aerobic and resistance exercise [95]. In a random-
ized study examining the effect of progressive
resistance training we found improved muscle
strength of the knee extensors and flexors and
motor function in individuals with type 2 diabetic
polyneuropathy at levels comparable with those
seen in individuals with diabetes without DPN
and healthy control individuals. No effects were
observed in small nerve fiber function as reflected
in IENFD [96].
Several studies have reported positive effects
of training in individuals with DSPN; however,
only a very limited number of studies have
included quantitative evaluations of motor func-
tion and/or muscle strength. In a small study of
eleven individuals with type 2 diabetes and DPN,
Praet et al. found an increase in muscle strength
following 10  weeks of progressive resistance
training combined with high-intensity interval
training [12].
In a randomized trial, Allet et  al. found that
individuals with and without DPN improved gait
speed, balance, muscle strength and joint mobil-
ity, following gait and balance exercises twice a
195
week for 12  weeks [97]; however, DSPN was
diagnosed based only on vibration perception
and was not confirmed by NCS. Handsaker et al.
studied nine individuals with DSPN diagnosed
by two clinical tests [98] and found an improve-
ment in speed of ankle and knee strength genera-
tion during stair ascent and descent following
1-h/week high-load dynamic resistance and iso-
metric exercises for 16 weeks.
In the studies by Kruse and Muller, exercise
interventions did not lead to gains in muscle
strength; however, both studies included inter-
ventions of low intensity bodyweight training. In
a randomized trial, Kruse and colleagues [98]
examined the effects of bodyweight leg strength-
ening exercises and balance exercises over
12  months in 79 individuals with DSPN and
found no difference in falls frequency, muscle
strength or balance between the groups following
the intervention. Mueller et al. compared weight-­
bearing versus non-weight-bearing exercises in
individuals with DSPN and found no change in
muscle strength [99]. However, individuals exer-
cised three times a week for 12  weeks, without
resistance, which may explain the lack of an
effect on muscle strength.
Progressive resistance training is the most
efficient way to gain muscle strength. Therefore,
it is surprising that previous studies have not
included resistance throughout the exercise
intervention. Resistance training can prevent
muscle atrophy, improve neural control and pre-
vent age-­ related sarcopenia with an improve-
ment in gait speed, balance, functional mobility
with a reduced risk of falls in healthy individuals
[81, 100, 101].
Several studies have reported improved motor
function measured by walking speed following
balance and gait training, bodyweight exercises,
aerobic exercise, and isokinetic dynamometry
combined with balance and treadmill exercises
[97, 99, 102, 103]. Furthermore, it has been
shown that gait and balance improve following
bodyweight exercises, balance training, or aero-
bic exercise [104–107]. To our knowledge, no
study has examined the effect of exercise in indi-
viduals with severe motor neuropathy. Further
studies should focus on the severity of motor neu-
196
ropathy by using quantitative techniques such as References
imaging and muscle biopsies to examine training
effects at the histopathological and biochemical
levels.
In individuals with DSPN and substantial mus-
cle weakness, it remains unknown whether they
respond to long-term resistance training consider-
ing their severe denervation and incomplete re-
innervation of striated muscles. Individuals with
inherited neuropathies and severe weakness have
been shown to improve their muscle strength fol-
lowing training [108, 109].
Currently, there are no available evidence-­
based treatment recommendations for individuals
with motor involvement and diabetes. Several
multimodal approaches, including lifestyle
changes with exercise, improved glycemic con-
trol, and pharmacological treatments of comor-
bidities that accompany diabetes, have beneficial
effects and may alleviate symptoms of diabetes.
Despite the severity and highly incapacitating
nature of motor neuropathy there is a lack of
large detailed quantitative studies on motor affec-
tion and the assessment of treatment options.
14 Conclusions
• Motor neuropathy presents with distal sym-
metrical impairment always in combination
with sensory disturbances.
• Motor neuropathy presents in a length-­
dependent pattern with a clear proximal-distal
gradient.
• The diagnosis of motor neuropathy is made
based on a combined assessment of motor
nerve function and muscle strength.
• Motor dysfunction is clinically evident in only
a small proportion of individuals with
diabetes.
• Motor dysfunction may develop not only due
to motor neuropathy but also as a consequence
of lower muscle quality due to diabetic
myopathy.
• Resistance training is a hitherto underutilized
non-pharmacological treatment modality that
may be useful for individuals with diabetic
polyneuropathy and motor dysfunction.
K. S. Khan and H. Andersen
1. Pop-Busui R, Boulton AJM, Feldman EL, Bril V,
Freeman R, Malik RA, et  al. Diabetic neuropathy:
a position statement by the American Diabetes
Association. Diabetes Care. 2017;40(1):136–54.
[cited 2019 May 23]. http://www.ncbi.nlm.nih.gov/
pubmed/27999003.
2. Dyck PJ, Carter RE, Litchy WJ.  Modeling nerve
conduction criteria for diagnosis of diabetic poly-
neuropathy. Muscle Nerve. 2011;44(3):340–5.
[cited 2019 May 23]. http://www.ncbi.nlm.nih.gov/
pubmed/21996793.
3. Dyck PJB, Overland CJ, Low PA, Litchy WJ, Davies
JL, Dyck PJB, et  al. Signs and symptoms versus
nerve conduction studies to diagnose diabetic senso-
rimotor polyneuropathy: Cl vs. NPhys trial. Muscle
Nerve. 2010;42(2):157–64. [cited 2019 Dec 10].
http://doi.wiley.com/10.1002/mus.21661.
4. Dyck PJ, Kratz KM, Karnes JL, Litchy WJ, Klein
R, Pach JM, et al. The prevalence by staged severity
of various types of diabetic neuropathy, retinopathy,
and nephropathy in a population-based cohort: the
Rochester Diabetic Neuropathy Study. Neurology.
1993;43(4):817–24. http://www.ncbi.nlm.nih.gov/
pubmed/8469345.
5. Suda EY, Matias AB, Bus SA, Sacco ICN, et  al.
Impact of diabetic neuropathy severity on foot clear-
ance complexity and variability during walking. Gait
Posture. 2019;74:194–9. https://doi.org/10.1016/j.
gaitpost.2019.09.014.
6. Martinelli AR, Mantovani AM, Nozabieli AJL,
Ferreira DMA, Barela JA, de Camargo MR, et  al.
Muscle strength and ankle mobility for the gait
parameters in diabetic neuropathies. Foot (Edinb).
2013;23(1):17–21.
7. Richardson JK, Demott T, Allet L, Kim H, Ashton-­
Miller JA.  Hip strength: ankle proprioceptive
threshold ratio predicts falls and injury in diabetic
neuropathy. Muscle Nerve. 2014;50(3):437–42.
8. Allet L, Kim H, Ashton-Miller J, De Mott T,
Richardson JK.  Step length after discrete pertur-
bation predicts accidental falls and fall-related
injury in elderly people with a range of peripheral
neuropathy. J Diabetes Complications. 2014;28(1
CC-Injuries):79–84. https://www.cochranelibrary.
com/central/doi/10.1002/central/CN-­01038023/full.
9. Cavanagh PR, Derr JA, Ulbrecht JS, Maser RE,
Orchard TJ. Problems with gait and posture in neu-
ropathic patients with insulin-dependent diabetes
mellitus. Diabet Med. 1992;9(5):469–74.
10. Onodera AN, Gomes AA, Pripas D, Mezzarane
RA, Sacco ICN.  Lower limb electromygraphy and
kinematics of neuropathic diabetic patients during
real-life activities: stair negotiation. Muscle Nerve.
2011;44(2):269–77. http://www.embase.com/
search/results?subaction=viewrecord&from=export
&id=L362173554.
Motor Neuropathy in Diabetes
11. Andersen H, Mogensen PM.  Disordered mobil-
ity of large joints in association with neuropathy in
patients with long-standing insulin-dependent diabe-
tes mellitus. Diabet Med. 1997;14(3):221–7.
12. Praet SFE, Jonkers RAM, Schep G, Stehouwer
CDA, Kuipers H, Keizer HA, et al. Long-standing,
insulin-treated type 2 diabetes patients with com-
plications respond well to short-term resistance
and interval exercise training. Eur J Endocrinol.
2008;158(2):163–72. [cited 2019 Jul 9]. http://www.
ncbi.nlm.nih.gov/pubmed/18230822.
13. Balducci S, Zanuso S, Massarini M, Corigliano G,
Nicolucci A, Missori S, et  al. The Italian Diabetes
and Exercise Study (IDES): design and methods for
a prospective Italian multicentre trial of intensive
lifestyle intervention in people with type 2 diabetes
and the metabolic syndrome. Nutr Metab Cardiovasc
Dis. 2008;18(9):585–95. [cited 2019 Jul 17]. http://
www.ncbi.nlm.nih.gov/pubmed/18061415.
14. Orlando G, Balducci S, Bazzucchi I, Pugliese G,
Sacchetti M.  Neuromuscular dysfunction in type
2 diabetes: underlying mechanisms and effect
of resistance training. Diabetes Metab Res Rev.
2016;32(1):40–50. [cited 2019 Jul 9] http://doi.
wiley.com/10.1002/dmrr.2658.
15. Monaco CMF, Perry CGR, Hawke TJ.  Diabetic
myopathy. Curr Opin Neurol. 2017;30(5):545–52.
[cited 2019 Aug 14]. http://insights.ovid.com/
crossref?an=00019052-­201710000-­00016.
16. Andersen H, Gadeberg PC, Brock B, Jakobsen
J.  Muscular atrophy in diabetic neuropathy: a ste-
reological magnetic resonance imaging study.
Diabetologia. 1997;40(9):1062–9.
17. Andersen H.  Muscular endurance in long-term
IDDM patients. Diabetes Care. 1998;21(4):604–9.
http://www.embase.com/search/results?subaction=v
iewrecord&from=export&id=L28172942.
18. Andersen H, Jacobsen J, Andreassen CS.  A pro-
gressive late complication in diabetic distal
symmetric polyneuropathy. Muscle Weakness.
2006;21(2):262–8.
19. Park SW, Goodpaster BH, Strotmeyer ES, de
Rekeneire N, Harris TB, Schwartz AV, et  al.
Decreased muscle strength and quality in older adults
with type 2 diabetes: the health, aging, and body
composition study. Diabetes. 2006;55(6):1813–8.
20. Andersen H, Stålberg E, Falck B.  F-wave latency,
the most sensitive nerve conduction parameter
in patients with diabetes mellitus. Muscle Nerve.
1997;20(10):1296–302.
21. Andersen H, Stålberg E, Gjerstad MD, Jakobsen
J. Association of muscle strength and electrophysi-
ological measures of reinnervation in diabetic neu-
ropathy. Muscle Nerve. 1998;21(12):1647–54.
[cited 2019 Dec 7]. http://www.ncbi.nlm.nih.gov/
pubmed/9843064.
22. Andreassen CS, Jensen JM, Jakobsen J, Ulhøj BP,
Andersen H.  Striated muscle fiber size, composi-
tion, and capillary density in diabetes in relation
197
to neuropathy and muscle strength. J Diabetes.
2014;6(5):462–71.
23. Meijer JWK, Bosma E, Lefrandt JD, Links TP,
Smith AJ, Steward R.  Symptom scoring systems
to diagnose distal polyneuropathy in diabetes: the
diabetic neuropathy symptom score. Diabet Med.
2002;19:962–5.
24. Andreassen CS, Jakobsen J, Flyvbjerg A, Andersen
H.  Expression of neurotrophic factors in diabetic
muscle-relation to neuropathy and muscle strength.
Brain. 2009;132(10):2724–33.
25. Bril V, Perkins BA.  Validation of the Toronto
Clinical Scoring System for diabetic polyneuropa-
thy. Diabetes Care. 2002;25(11):2048–52. [cited
2019 May 24]. http://www.ncbi.nlm.nih.gov/
pubmed/12401755.
26. Tesfaye S, Boulton AJM, Dyck PJ, Freeman R,
Horowitz M, Kempler P, et  al. Diabetic neuropa-
thies: update on definitions, diagnostic criteria, esti-
mation of severity, and treatments. Diabetes Care.
2010;33(10):2285–93. [cited 2019 May 23]. http://
www.ncbi.nlm.nih.gov/pubmed/20876709.
27. Callaghan BC, Xia R, Reynolds E, Banerjee
M, Rothberg AE, Burant CF, et  al. Association
between metabolic syndrome components and
polyneuropathy in an obese population. JAMA
Neurol. 2016;73(12):1468. [cited 2020 May
10]. https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC5217829/#R30.
28. Dyck PJ, Albers JW, Andersen H, Arezzo JC,
Biessels G-J, Bril V, et  al. Diabetic polyneuropa-
thies: update on research definition, diagnostic cri-
teria and estimation of severity. Diabetes Metab Res
Rev. 2011;27(7):620–8.
29. Gregersen G. Latency time, maximal amplitude and
electromyography in diabetic patients. Acta Med
Scand. 1968;183(1–2):55–60.
30. Tankisi H, Pugdahl KBS, Andersen HF-FA, Tankisi
H, Pugdahl K, Beniczky S, Andersen H, et  al.
Evidence-based recommendations for examination
and diagnostic strategies of polyneuropathy electro-
diagnosis. Clin Neurophysiol Pract. 2019;4:214–22.
[cited 2019 Dec 14]. https://www.sciencedirect.com/
science/article/pii/S2467981X19300381.
31. Weisman A, Bril V, Ngo M, Lovblom LE, Halpern
EM, Orszag A, et  al. Identification and prediction
of diabetic sensorimotor polyneuropathy using indi-
vidual and simple combinations of nerve conduction
study parameters. PLoS One. 2013;8(3):e58783.
32. Albers JW, Brown MB, Sima AA, Greene DA. Nerve
conduction measures in mild diabetic neuropathy
in the Early Diabetes Intervention Trial: the effects
of age, sex, type of diabetes, disease duration, and
anthropometric factors. Tolrestat Study Group for
the Early Diabetes Intervention Trial. Neurology.
1996;46(1):85–91.
33. Dyck PJ, Karnes JL, Daube J, O’Brien P, Service
FJ.  Clinical and neuropathological criteria for the
diagnosis and staging of diabetic polyneuropathy.
198
Brain. 1985;108(4):861–80. https://doi.org/10.1093/
brain/108.4.861.
34. Kohara N, Kimura J, Kaji R, Goto Y, Ishii J,
Takiguchi M, et al. F-wave latency serves as the most
reproducible measure in nerve conduction studies
of diabetic polyneuropathy: multicentre analysis in
healthy subjects and patients with diabetic polyneu-
ropathy. Diabetologia. 2000;43(7):915–21.
35. Pan H, Lin J, Chen N, Jian F, Zhang Z, Ding Z, et al.
Normative data of F-wave measures in China. Clin
Neurophysiol. 2013;124(1):183–9.
36. Hendriksen PH, Oey PL, Wieneke GH, Bravenboer
B, Banga JD.  Subclinical diabetic neuropathy:
similarities between electrophysiological results
of patients with Type 1 (insulin-dependent) and
Type 2 (non-insulin-dependent) diabetes melli-
tus. Diabetologia. 1992;35(7):690–5. https://doi.
org/10.1007/BF00400264.
37. Balducci S, Sacchetti M, Orlando G, Salvi L,
Pugliese L, Salerno G, et  al. Correlates of muscle
strength in diabetes: the study on the assessment of
determinants of muscle and bone strength abnormal-
ities in diabetes (SAMBA). Nutr Metab Cardiovasc
Dis. 2014;24(1):18–26.
38. Kamiya H, Murakawa Y, Zhang W, Sima
AAF. Unmyelinated fiber sensory neuropathy differs
in type 1 and type 2 diabetes. Diabetes Metab Res
Rev. 2005;21(5):448–58.
39. Singleton JR, Bixby B, Russell JW, Feldman
EL, Peltier A, Goldstein J, et  al. The Utah Early
Neuropathy Scale: a sensitive clinical scale for early
sensory predominant neuropathy. J Peripher Nerv
Syst. 2008;13(3):218–27. [cited 2019 May 24].
http://www.ncbi.nlm.nih.gov/pubmed/18844788.
40. Devigili G, Tugnoli V, Penza P, Camozzi F, Lombardi
R, Melli G, et  al. The diagnostic criteria for small
fibre neuropathy: from symptoms to neuropathol-
ogy. Brain. 2008;131(7):1912. [cited 2020 May
14]. https://www.ncbi.nlm.nih.gov/pmc/articles/
PMC2442424/.
41. Sivaskandarajah GA, Halpern EM, Lovblom LE,
Weisman A, Orlov S, Bril V, et al. Structure-function
relationship between corneal nerves and conven-
tional small-fiber tests in type 1 diabetes. Diabetes
Care. 2013;36(9):2748–55.
42. Andersson C, Guttorp P, Särkkä A.  Discovering
early diabetic neuropathy from epidermal nerve fiber
patterns. Stat Med. 2016;35(24):4427–42.
43. Braune HJ.  Early detection of diabetic neu-
ropathy: a neurophysiological study on 100
patients. Electromyogr Clin Neurophysiol.
1997;37(7):399–407.
44. Jacobsen AB, Bostock H, Tankisi H.  CMAP scan
MUNE (MScan)—a novel motor unit number esti-
mation (MUNE) method. J Vis Exp. 2018;(136).
45. Kristensen AG, Khan KS, Bostock H, Khan BS,
Gylfadottir S, Andersen H, et  al. MScanFit motor
unit number estimation and muscle velocity recov-
ery cycle recordings in diabetic polyneuropathy.
Clin Neurophysiol Pract. 2020;131(11):2591–9.
K. S. Khan and H. Andersen
46. Lord SR, Caplan GA, Colagiuri R, Colagiuri S,
Ward JA.  Sensori-motor function in older persons
with diabetes. Diabet Med. 1993;10(7):614–8.
47. Andersen H, Poulsen PL, Mogensen CE, Jakobsen
J.  Isokinetic muscle strength in long-term IDDM
patients in relation to diabetic complications.
Diabetes. 1996;45(4):440–5. [cited 2019 Dec 13].
http://www.ncbi.nlm.nih.gov/pubmed/8603765.
48. Orlando G, Balducci S, Bazzucchi I, Pugliese G,
Sacchetti M. The impact of type 1 diabetes and dia-
betic polyneuropathy on muscle strength and fatiga-
bility. Acta Diabetol. 2017;54(6):543–50.
49. Nomura T, Ishiguro T, Ohira M, Ikeda Y.  Diabetic
polyneuropathy is a risk factor for decline of lower
extremity strength in patients with type 2 diabetes. J
Diabetes Investig. 2018;9(1):186–92.
50. Bursac SN, Jandric S, Talic G. Influence of diabetic
distal symmetric polyneuropathy on the perfor-
mance of the musculoskeletal system of lower leg
and foot. Med Arch. 2019;73(4):262–7.
51. IJzerman TH, Schaper NC, Melai T, Meijer K,
Willems PJB, Savelberg HHCM.  Lower extremity
muscle strength is reduced in people with type 2 dia-
betes, with and without polyneuropathy, and is asso-
ciated with impaired mobility and reduced quality
of life. Diabetes Res Clin Pract. 2012;95(3):345–51.
52. Allen MD, Major B, Kimpinski K, Doherty TJ, Rice
CL.  Skeletal muscle morphology and contractile
function in relation to muscle denervation in diabetic
neuropathy. J Appl Physiol. 2014;116(5):545–52.
53. Mesinovic J, Zengin A, De Courten B, Ebeling PR,
Scott D.  Sarcopenia and type 2 diabetes mellitus:
a bidirectional relationship. Diabetes Metab Syndr
Obes. 2019;12:1057–72.
54. Tuttle LJ, Sinacore DR, Cade WT, Mueller
MJ.  Lower physical activity is associated with
higher intermuscular adipose tissue in people with
type 2 diabetes and peripheral neuropathy. Phys
Ther. 2011;91(6):923–30. [cited 2019 Jul 9]. http://
www.ncbi.nlm.nih.gov/pubmed/21474636.
55. Bittel AJ, Bittel DC, Tuttle LJ, Strube MJ, Mueller
MJ, Cade WT, et  al. Explanators of Sarcopenia in
individuals with diabesity: a cross-sectional analy-
sis. J Geriatr Phys Ther. 2017;40(2):86–94.
56. Gundmi S, Maiya AG, Bhat AK, Ravishankar
N, Hande MH, Rajagopal KV.  Hand dysfunc-
tion in type 2 diabetes mellitus: systematic review
with meta-analysis. Ann Phys Rehabil Med.
2018;61(2):99–104.
57. Gutefeldt K, Lundstedt S, Thyberg ISMM,
Bachrach-Lindström M, Arnqvist HJ, Spångeus
A.  Clinical examination and self-reported upper
extremity impairments in patients with long-­
standing type 1 diabetes mellitus. J Diabetes Res.
2020;2020:4172635.
58. Lima KCA, Borges LS, Hatanaka E, Rolim LC, de
Freitas PB. Grip force control and hand dexterity are
impaired in individuals with diabetic peripheral neu-
ropathy. Neurosci Lett. 2017;659:54–9.
Motor Neuropathy in Diabetes
59. Kabitz H-J, Sonntag F, Walker D, Schwoerer A,
Walterspacher S, Kaufmann S, et al. Diabetic poly-
neuropathy is associated with respiratory muscle
impairment in type 2 diabetes. Diabetologia.
2008;51(1):191–7.
60. Van Eetvelde BLM, Cambier D, Vanden Wyngaert
K, Celie B, Calders P.  The influence of clini-
cally diagnosed neuropathy on respiratory muscle
strength in type 2 diabetes mellitus. J Diabetes Res.
2018;2018:8065938.
61. Venkataraman K, Pun V, Mohamed AZ, Luo M,
Wong C, Zong F, et al. Altered motor and motor per-
ceptual cognitive imagery task-related activation in
diabetic peripheral neuropathy: insights from func-
tional MRI. Diabetes Care. 2019;42(10):2004–7.
62. Kell RT, Bell G, Quinney A.  Musculoskeletal fit-
ness, health outcomes and quality of life. Sport Med.
2001;31(12):863–73. [cited 2019 Dec 19]. http://
www.ncbi.nlm.nih.gov/pubmed/11665913.
63. Wakeling JM, Lee SSM, Arnold AS, de Boef Miara
M, Biewener AA. A muscle’s force depends on the
recruitment patterns of its fibers. Ann Biomed Eng.
2012;40(8):1708–20. [cited 2019 Dec 13]. http://
www.ncbi.nlm.nih.gov/pubmed/22350666.
64. Andersen H, Nielsen S, Mogensen CE, Jakobsen
J.  Muscle strength in type 2 diabetes. Diabetes.
2004;53(6):1543–8. [cited 2019 Jul 23]. http://diabe-
tes.diabetesjournals.org/content/53/6/1543.abstract.
65. Henderson AD, Johnson AW, Rasmussen LG, Peine
WP, Symons SH, Scoresby KA, et  al. Early-stage
diabetic neuropathy reduces foot strength and intrin-
sic but not extrinsic foot muscle size. J Diabetes Res.
2020;2020:9536362.
66. Severinsen K. Atrophy of foot muscles in diabetic.
Diabetes Care. 2007;30(12):3053–7.
67. Severinsen K, Andersen H. Evaluation of atrophy of
foot muscles in diabetic neuropathy—a comparative
study of nerve conduction studies and ultrasonogra-
phy. Clin Neurophysiol. 2007;118(10):2172–5.
68. Andreassen CS, Jakobsen J, Ringgaard S, Ejskjaer
N, Andersen H. Accelerated atrophy of lower leg and
foot muscles-a follow-up study of long-term diabetic
polyneuropathy using magnetic resonance imaging
(MRI). Diabetologia. 2009;52(6):1182–91.
69. Greenman RL, Panasyuk S, Wang X, Lyons TE,
Dinh T, Longoria L, et al. Early changes in the skin
microcirculation and muscle metabolism of the dia-
betic foot. Lancet. 2005;366(9498):1711–7.
70. Almurdhi MM, Reeves ND, Bowling FL, Boulton
AJM, Jeziorska M, Malik RA.  Reduced lower-­
limb muscle strength and volume in patients with
type 2 diabetes in relation to neuropathy, intra-
muscular fat, and vitamin D levels. Diabetes Care.
2016;39(3):441–7.
71. Stouge A, Khan KS, Kristensen AG, Tankisi H,
Schlaffke L, Froeling M, et  al. MRI of skeletal
muscles in participants with type 2 diabetes with
or without diabetic polyneuropathy. Radiology.
2020;297(3):608–19.
199
72. Kim K-H, Yoo J-Y, You B-C.  Ultrasonographic
evaluation of sural nerve for nerve conduction study.
Ann Rehabil Med. 2014;38(1):46–51.
73. Pham M, Oikonomou D, Bäumer P, Bierhaus A,
Heiland S, Humpert PM, et al. Proximal neuropathic
lesions in distal symmetric diabetic polyneuropathy:
findings of high-resolution magnetic resonance neu-
rography. Diabetes Care. 2011;34(3):721–3.
74. Vaeggemose M, Pham M, Ringgaard S, Tankisi H,
Ejskjaer N, Heiland S, et  al. Magnetic resonance
neurography visualizes abnormalities in sciatic and
tibial nerves in patients with type 1 diabetes and neu-
ropathy. Diabetes. 2017;66(7):1779–88.
75. Hewston P, Deshpande N. Falls and balance impair-
ments in older adults with type 2 diabetes: think-
ing beyond diabetic peripheral neuropathy. Can J
Diabetes. 2016;40(1):6–9. [cited 2017 Jan 9]. http://
www.embase.com/search/results?subaction=viewre
cord&from=export&id=L607755496.
76. Brown SJ, Handsaker JC, Bowling FL, Boulton AJM,
Reeves ND. Diabetic peripheral neuropathy compro-
mises balance during daily activities. Diabetes Care.
2015;38(6):1116–22. [cited 2019 May 23]. http://
www.embase.com/search/results?subaction=viewre
cord&from=export&id=L609629141.
77. Vaz MM, Costa GC, Reis JG, Junior WM,
Albuquerque de Paula FJ, Abreu DC. Postural control
and functional strength in patients with type 2 diabe-
tes mellitus with and without peripheral neuropathy.
Arch Phys Med Rehabil. 2013;94(12):2465–70.
78. Richardson JK, Eckner JT, Allet L, Kim H, Ashton-­
Miller JA.  Complex and simple clinical reaction
times are associated with gait, balance, and major
fall injury in older subjects with diabetic peripheral
neuropathy. Am J Phys Med Rehabil. 2017;96(1):8–
16. http://www.embase.com/search/results?subactio
n=viewrecord&from=export&id=L617662636.
79. Khan KS, Andersen H. The impact of diabetic neu-
ropathy on activities of daily living, postural balance
and risk of falls—a systematic review. J Diabetes Sci
Technol. 2022;16(2):289–94.
80. Simoneau GG, Ulbrecht JS, Derr JA, Becker MB,
Cavanagh PR.  Postural instability in patients
with diabetic sensory neuropathy. Diabetes Care.
1994;17(12):1411–21. [cited 2019 Jul 23]. https://
doi.org/10.2337/diacare.17.12.1411.
81. Abe T, De Hoyos DV, Pollock ML, Garzarella
L.  Time course for strength and muscle thick-
ness changes following upper and lower body
resistance training in men and women. Eur J Appl
Physiol Occup Physiol. 2000;81(3):0174. [cited
2019 Dec 23]. http://www.ncbi.nlm.nih.gov/
pubmed/10638374.
82. Menz HB, Lord SR, St George R, Fitzpatrick RC,
George RS, Fitzpatrick RC, et  al. Walking stabil-
ity and sensorimotor function in older people with
diabetic peripheral neuropathy. Arch Phys Med
Rehabil. 2004;85(2):245–52. [cited 2019 May 23].
http://www.ncbi.nlm.nih.gov/pubmed/14966709.
200
83. Almurdhi MM, Brown SJ, Bowling FL, Boulton
AJM, Jeziorska M, Malik RA, et al. Altered walking
strategy and increased unsteadiness in participants
with impaired glucose tolerance and Type 2 diabetes
relates to small-fibre neuropathy but not vitamin D
deficiency. Diabet Med. 2017;34(6):839–45.
84. Handsaker JC, Brown SJ, Bowling FL, Cooper G,
Maganaris CN, Boulton AJM, et  al. Contributory
factors to unsteadiness during walking up and down
stairs in patients with diabetic peripheral neuropathy.
Diabetes Care. 2014;37(11):3047–53. [cited 2019
Aug 4]. https://doi.org/10.2337/dc14-­0955.
85. Khan KS, Pop-Busui R, Devantier L, Kristensen
AG, Tankisi H, Dalgas U, et  al. Falls in individu-
als with type 2 diabetes; a cross-sectional study on
the impact of motor dysfunction, postural insta-
bility and diabetic polyneuropathy. Diabet Med.
2020;38(9):e14470.
86. Church TS, Blair SN, Cocreham S, Johannsen
N, Johnson W, Kramer K, et  al. Effects of aero-
bic and resistance training on hemoglobin A1c
levels in patients with type 2 diabetes. JAMA.
2010;304(20):2253. [cited 2019 Dec 20]. http://
www.ncbi.nlm.nih.gov/pubmed/21098771.
87. Castaneda C, Layne JE, Munoz-Orians L, Gordon
PL, Walsmith J, Foldvari M, et  al. A randomized
controlled trial of resistance exercise training to
improve glycemic control in older adults with type
2 diabetes. Diabetes Care. 2002;25(12):2335–41.
[cited 2019 Aug 17]. http://www.ncbi.nlm.nih.gov/
pubmed/12453982.
88. Colberg SR, Sigal RJ, Fernhall B, Regensteiner JG,
Blissmer BJ, Rubin RR, et  al. Exercise and type 2
diabetes: the American College of Sports Medicine
and the American Diabetes Association: joint posi-
tion statement. Diabetes Care. 2010;33(12):e147.
[cited 2019 Aug 23]. https://www.ncbi.nlm.nih.gov/
pmc/articles/PMC2992225/.
89. Dagogo-Jack S, Egbuonu N, Edeoga C.  Principles
and practice of nonpharmacological interventions
to reduce cardiometabolic risk. Med Princ Pract.
2010;19(3):167–75. [cited 2019 Dec 10]. http://
www.ncbi.nlm.nih.gov/pubmed/20357497.
90. LeMaster JW, Mueller MJ, Reiber GE, Mehr DR,
Madsen RW, Conn VS.  Effect of weight-bearing 100. Mcleod JC, Stokes T, Phillips SM. Resistance exer-
activity on foot ulcer incidence in people with dia-
betic peripheral neuropathy: feet first randomized
controlled trial. Phys Ther. 2008;88(11):1385–98. 101. Cadore EL, Rodríguez-Mañas L, Sinclair A,
[cited 2019 Aug 18]. http://www.ncbi.nlm.nih.gov/
pubmed/18801859.
91. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-­
Sceppa C.  Physical activity/exercise and type 2
diabetes. Diabetes Spectr. 2005;18(1):2518–39. 102. Morrison S, Colberg SR, Parson HK, Vinik
[cited 2019 Aug 31]. http://www.ncbi.nlm.nih.gov/
pubmed/15451933.
92. Balducci S, Iacobellis G, Parisi L, Di Biase N,
Calandriello E, Leonetti F, et  al. Exercise train-
ing can modify the natural history of diabetic
peripheral neuropathy. J Diabetes Complications.
K. S. Khan and H. Andersen
2006;20(4):216–23. [cited 2019 Jul 17]. https://
www.sciencedirect.com/science/article/abs/pii/S105
6872705000905?via%3Dihub.
93. Smith AG, Russell J, Feldman EL, Goldstein
J, Peltier A, Smith S, et  al. Lifestyle interven-
tion for pre-diabetic neuropathy. Diabetes Care.
2006;29(6):1294–9.
94. Kluding PM, Pasnoor M, Singh R, Jernigan S,
Farmer K, Rucker J, et  al. The effect of exercise
on neuropathic symptoms, nerve function, and
cutaneous innervation in people with diabetic
peripheral neuropathy. J Diabetes Complications.
2012;26(5):424–9. [cited 2019 Jul 9]. http://www.
ncbi.nlm.nih.gov/pubmed/22717465.
95. Singleton JR, Marcus RL, Jackson JE, Lessard M,
Graham TE, Smith AG.  Exercise increases cutane-
ous nerve density in diabetic patients without neu-
ropathy. Ann Clin Transl Neurol. 2014;1(10):844–9.
[cited 2019 Jul 17]. http://www.ncbi.nlm.nih.gov/
pubmed/25493275.
96. Khan KS, Overgaard K, Tankisi H, Karlsson P,
Devantier L, Gregersen S, et al. Effects of progressive
resistance training in individuals with type 2 diabetic
polyneuropathy: a randomised assessor-blinded con-
trolled trial. Diabetologia. 2022;65(4):620–31.
97. Allet L, Armand S, de Bie RA, Golay A, Monnin D,
Aminian K, et  al. The gait and balance of patients
with diabetes can be improved: a randomised con-
trolled trial. Diabetologia. 2010;53(3):458–66.
[cited 2019 Jul 6]. https://www.ncbi.nlm.nih.gov/
pmc/articles/PMC2815802/.
98. Robin LK, Joseph WL, Richard WM. Fall and
balance outcomes after an intervention to pro-
mote leg strength, balance, and walking in peo-
ple with diabetic peripheral neuropathy: “Feet
First” Randomized Controlled Trial. https://doi.
org/10.2522/ptj.20090362.
99. Mueller MJ, Tuttle LJ, Lemaster JW, Strube MJ,
McGill JB, Hastings MK, et  al. Weight-bearing
versus nonweight-bearing exercise for persons
with diabetes and peripheral neuropathy: a ran-
domized controlled trial. Arch Phys Med Rehabil.
2013;94(5):829–38. [cited 2019 Jul 9]. https://doi.
org/10.1016/j.apmr.2012.12.015.
cise training as a primary countermeasure to age-­
related chronic disease. Front Physiol. 2019;10:645.
Izquierdo M. Effects of different exercise interven-
tions on risk of falls, gait ability, and balance in
physically frail older adults: a systematic review.
Rejuvenation Res. 2013;16(2):105–14.
AI.  Exercise improves gait, reaction time and
postural stability in older adults with type 2 dia-
betes and neuropathy. J Diabetes Complications.
2014;28(5):715–22. [cited 2019 Jul 9]. http://www.
embase.com/search/results?subaction=viewrecord&
from=export&id=L53187427.
Motor Neuropathy in Diabetes
103. Taveggia G, Villafañe JH, Vavassori F, Lecchi C,
Borboni A, Negrini S, et  al. Multimodal treat-
ment of distal sensorimotor polyneuropathy in
diabetic patients: a randomized clinical trial. J
Manipulative Physiol Ther. 2014;37(4):242–52.
[cited 2016 Oct 21]. http://www.ncbi.nlm.nih.gov/
pubmed/24656867.
104. Richardson JK, Sandman D, Vela S. A focused exer-
cise regimen improves clinical measures of balance
in patients with peripheral neuropathy. Arch Phys
Med Rehabil. 2001;82(2):205–9. http://www.ncbi.
nlm.nih.gov/pubmed/11239311.
105. Akbari M, Jafari H, Moshashaee A, Forugh B.  Do
diabetic neuropathy patients benefit from balance
training? J Rehabil Res Dev. 2012;49(2):333–8.
[cited 2019 Jul 9]. http://www.ncbi.nlm.nih.gov/
pubmed/22773533.
106. Dixit S, Maiya A, Shastry BA, Guddattu V. Analysis
of postural control during quiet standing in a
population with diabetic peripheral neuropathy
undergoing moderate intensity aerobic exercise
201
training: a single blind, randomized controlled
trial. Am J Phys Med Rehabil. 2016;95(7):516–24.
[cited 2019 Jul 9]. http://www.ncbi.nlm.nih.gov/
pubmed/26745226.
107. Song CH, Petrofsky JS, Lee SW, Lee KJ, Yim
JE.  Effects of an exercise program on balance
and trunk proprioception in older adults with
diabetic neuropathies. Diabetes Technol Ther.
2011;13(8):803–11. http://www.ncbi.nlm.nih.gov/
pubmed/21561371.
108. Chetlin RD, Gutmann L, Tarnopolsky MA, Ullrich
IH, Yeater RA. Resistance training exercise and cre-
atine in patients with Charcot-Marie-Tooth disease.
Muscle Nerve. 2004;30(1):69–76.
109. Burns J, Sman AD, Cornett KMD, Wojciechowski E,
Walker T, Menezes MP, et al. Safety and efficacy of
progressive resistance exercise for Charcot-Marie-­
Tooth disease in children: a randomised, double-­
blind, sham-controlled trial. Lancet Child Adolesc
Heal. 2017;1(2):106–13.
 Cardiovascular Autonomic
Neuropathy
Lynn Ang and Rodica Pop-Busui
1 Cardiovascular Autonomic
Neuropathy
Autonomic neuropathies affect the autonomic
neurons (parasympathetic, sympathetic, or both)
and manifest as a variety of condition-specific
symptoms and signs involving the cardiovascu-
lar, gastrointestinal, genitourinary, and sudomo-
tor systems in individuals with diabetes [1].
Cardiovascular autonomic neuropathy (CAN) is
the most studied form of the diabetic autonomic
neuropathies, and it is a common and deadly
complication of diabetes mellitus, which is fre-
quently overlooked in clinical practice due to its
characteristic subtle presentation earlier in dis-
ease. Yet, timely detection of CAN may help
implementation of tailored interventions to pre-
vent its progression and mitigate the risk of asso-
ciated complications, including cardiovascular
disease (CVD), cardiac arrhythmias, myocardial
dysfunction leading to congestive heart failure,
and increased mortality [2].
L. Ang · R. Pop-Busui (*)
Division of Metabolism, Endocrinology and
Diabetes, Department of Internal Medicine,
University of Michigan, Ann Arbor, MI, USA
e-mail: rpbusui@umich.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_12
2 Epidemiology
The reported prevalence of CAN varies depend-
ing on the population(s) studied (e.g., type 1 dia-
betes versus type 2 diabetes, intervention trials
versus epidemiological observations, earlier ver-
sus more contemporary cohorts), the available
outcome measures reported, and/or the diagnos-
tic criteria used.
In earlier cohorts of patients, particularly
those with long-standing type 1 diabetes (T1D),
potential candidates for pancreas transplantation,
abnormalities in cardiovascular reflex tests
(CARTs) suggestive of CAN, were detected in up
to 90% of patients [2]. In contrast, prevalence
rates are very low in patients with newly diag-
nosed T1D, as documented by the primary pre-
vention arm of the Diabetes Control and
Complications Trial (DCCT) cohort, arguably
the best phenotyped cohort for CAN with CARTs
as well as with electrocardiograms recordings
obtained repeatedly over more than 20  years of
follow-up [1–4]. During the observational epide-
miological follow-up of the same cohort, known
as the Epidemiology of Diabetes Interventions
and Complications (EDIC) study, the prevalence
of CAN increased steadily over time to as high as
44% of the cohort over 23 years of mean follow-
­up [3]. These data are comparable with data
reported in other observational cohorts with T1D
such as the European Insulin-Dependent Diabetes
Mellitus Prospective Complications Study
203
204
(EURODIAB IDDM), that reported baseline
prevalence rates of autonomic dysfunction,
defined as an abnormality in one or both of
CARTs used, of 36% [5], the STENO T1D study
that reported similar prevalence of CAN (36%)
over mean duration of diabetes of 40 years using
≥ two abnormal CARTs [6], or the Pittsburgh
Epidemiology of Diabetes Complications Study
(EDC) study, that reported a cumulative inci-
dence of symptomatic autonomic neuropathy
over 25  years of approximately 25% [7]. Most
recently, the large contemporary cohort of T1D
enrolled in the T1D Exchange Clinic Network
Registry (T1D Exchange) from years 2010–2017,
reported that autonomic neuropathy that included
CAN defined with the Survey of Autonomic
Symptoms (SAS), was present in ~17% of the
cohort [8].
Similar to T1D, prevalence rates as high as
60% have been reported in earlier cohorts of
patients with long-standing T2D [1, 2]. In more
recent T2D cohorts, however, lower prevalence
rates of 9–25% have been observed. For instance,
in Europe, the Diabetic Cardiovascular
Autonomic Neuropathy Multicenter Study Group
reported CAN prevalence rates of ~22% in T2D
using strict criterion of abnormal findings in
CARTs and indices of heart rate variability
(HRV), while the observational prospective fol-
low-­up of the Danish arm of the Anglo-Danish-­
Dutch Study of Intensive Treatment in People
With Screen-Detected Diabetes in Primary Care
(ADDITION-Denmark), reported prevalence
rates of 9% and 15% for manifest CAN, defined
based on CARTs, at the 6-year and 13-year fol-
low-­up examinations [9]. In the USA, in more
than 8000 participants with T2D at high CVD
risk, with a mean age of ~62  years and mean
duration of 10  years enrolled in the Action to
Control Cardiovascular Risk in Diabetes
(ACCORD) trial; the prevalence of CAN at study
entry defined by indices of HRV derived from
ECG recordings was ~20% [10]. Most recently,
the Glycemia Reduction Approaches in
Diabetes—A Comparative Effectiveness
(GRADE) trial reported that ~10% of this large
contemporary cohort of ~5000 individuals with
T2D of which 36% were female, with a mean
L. Ang and R. Pop-Busui
known T2D duration of 4  years, a mean age of
~57  years, mean HbA1c 8.0% (~64  mmol/mol)
and diverse ethnic representation presented with
CAN at baseline [11] as diagnosed with the same
HRV indices derived from standard 12-lead ECG
recordings using the recently published cut-­
points based on population norms from the Multi-­
Ethnic Study of Atherosclerosis (MESA) [12].
CAN may also present in the pre-diabetic
state, in individuals with impaired glucose toler-
ance, insulin resistance, and/or the metabolic
syndrome as reported from several cohorts dem-
onstrating the presence of subclinical CAN with
either HRV studies or smaller cardiac sympa-
thetic imaging studies [1, 2, 13–16]. Observations
in the population-based cross-sectional KORA
S4 study that included more than 1300 partici-
pants aged 55–74  years phenotyped for CAN
with indices of HRV and QT variability, demon-
strated prevalence rates of CAN as high as 8–11%
among those with prediabetes as defined by
impaired glucose tolerance and or impaired fast-
ing glucose, compared with 11.7% in newly diag-
nosed T2D and 17.5% in those with known T2D
[17]. Increased heart rate (HR) and decreased
HRV were also reported in patients with meta-
bolic syndrome, regardless of fasting glucose lev-
els [18]. CAN is also observed in youth as
documented by the SEARCH for Diabetes in
Youth Cohort study that included ~2000 youth
with T1D and T2D and reported CAN prevalence
rates of ~12% in youth with T1D (mean age
18 years, duration of diabetes 8 years) and 17%
in youth with T2D (mean age 22 years, duration
of diabetes 8 years) [19].
There are several risk factors that were shown
to be strongly associated with CAN in both T1D
and T2D, with some differences also present
among these individuals. Notable and consistent
risk factors for CAN across earlier and more con-
temporary cohorts include older age, markers of
higher glycemic exposure (including longer dia-
betes duration and poor glucose control as docu-
mented by HbA1c), diabetic kidney disease,
hypertension, elevated triglycerides, smoking,
and use of some medications [1–3, 20]. Similar
risk factors for CAN were also described in youth
[20]. In the highly phenotyped DCCT/EDIC
Cardiovascular Autonomic Neuropathy
cohort, in multivariable models, higher mean
HbA1c over time was one of the strongest risk
factors for CAN development [3]. Beyond glyce-
mia, in the same cohort, after adjusting for age
and HbA1c, hypertension and kidney disease
(defined as higher albumin excretion ratio and
lower eGFR) as well as higher triglycerides were
independently associated with CAN [3].
Hypertension, dyslipidemia and high triglycer-
ides were identified as risk factors for CAN in
other cohorts such as the EDC and EURODIAB
IDDM studies for T1D, and as well as in the
ADDITION study and in American Indians with
T2D [21–25].
Furthermore, existing evidence has clearly
established insulin resistance as a positive predic-
tor of abnormal sympathovagal balance in non-
diabetic offspring of patients with T2D, which
suggests that insulin resistance may be an early
contributor to autonomic dysfunction prior to the
onset of diabetes [26, 27]. Recent evidence from
several large cohorts reported an overall increased
CVD risk in women with diabetes, and some
smaller studies suggest that female sex may also
be a risk factor for CAN that confers a higher ini-
tial risk of myocardial injury and dysfunction
[28].
More recently, socio-economic factors includ-
ing lower income, and education, as well as
depression are emerging as risk factors for CAN
[8]. The observation of lower prevalence of CAN
in non-White races reported in the GRADE
cohort adds to an inconsistent body of literature
on race/ethnicity differences, with population-­
based studies reporting lower, matching, or
higher rates of CAN and other complications in
non-White race groups [11, 29, 30].
Key Points
• Cardiovascular autonomic neuropathy is a
prevalent complication in T1D and T2D
including in newly diagnosed T2D.
• High prevalence rates were described in indi-
viduals with prediabetes/metabolic syndrome
as well as in youth.
• Notable risk factors for CAN include older
age, higher glycemic exposure, diabetic kid-
205
ney disease, hypertension, elevated triglycer-
ides, and smoking.
• Socio-economic factors, female gender, as
well as depression are emerging as risk factors
for CAN.
3 Pathophysiology
The potential pathogenic mechanisms leading to
CAN development are broad and may include
complex interactions between indices of glyce-
mic exposure such as chronic hyperglycemia,
impaired insulin signaling, glucose variability
(GV), and disease duration, in conjunction with
several other risk factors, including obesity,
hypertension, dyslipidemia and age-related neu-
ronal deterioration [2]. In concert, these factors
activate critical pathways resulting in oxidative
stress, low-grade inflammation, and advanced
glycation with resultant consequences on nerve
cellular metabolism, bioenergetics, and perfusion
[2, 31–34] (Fig. 1).
Generally, uncontrolled blood glucose, as
indicated by HbA1c levels, is thought of as the
impetus behind complications development in
diabetes [35]. Since the results of the DCCT were
published [35], intensive glucose control has
been the standard clinical practice in T1D [36].
However, evidence from the DCCT/EDIC study
and other diabetes trials cautions that while main-
taining near-normal glucose control is important,
it is not sufficient to prevent all complications nor
to reverse disease once established [2, 4, 35, 37].
Furthermore, the observed increase in mortality
in the intensive glycemic intervention group in
the ACCORD trial raises the possibility that acute
hypoglycemia and/or wide glucose fluctuations
were potential mediators of this unexpected out-
come [2, 38]. Historically, observational and
experimental evidence has implicated hypogly-
cemia in the “dead in bed” syndrome, possibly by
lengthening the QT interval and/or by altering
cardiac vagal protection, with relative sympa-
thetic overactivity in T1D and subsequent fatal
ventricular arrhythmias [2, 39–41]. In addition,
individuals with T1D present with disrupted
compensatory mechanisms in response to hypo-
206 L. Ang and R. Pop-Busui

Hyperglycemia

Hypoglycemia Glucose
Variability

Changes in Oxidative
Metabolic flux Stress

Reactive Inflammation
Nitrogen
Species

Mitochondrial Myeloperoxidase
Pathway; Pathway; Tyrosyl
Superoxide Radical
Hydrogen Hypochlorus
Peroxide Acid

glucoxidation; Endothelial Cell

Polysaturated Fatty Acid; Dysfunction;
Advanced Glycation End Cardivascular Autonomic Superoxide
Products Neuropathy

Cardiac Myocardial
CVD Arrhythmia Heart Failure
Dysfunction

Morbidity Mortality

Fig. 1  The potential complex pathogenic mechanisms of (CVD), cardiac arrhythmia, myocardial dysfunction, and
cardiovascular autonomic neuropathy (CAN) in diabetes. heart failure with subsequent increased morbidity and
CAN may increase the risk of cardiovascular disease mortality

glycemia, such as lack of insulin secretion sup-
pression, impaired counterregulatory hormones
and lack of subjective awareness of hypoglyce-
mia, all increasing the risk of cardiac arrhythmia
and sudden cardiac death [2, 42–44].
Emerging evidence from animal and human
studies reported that glucose variability may have
more pronounced effects on stimulating several
critical pathways involved in the risk of chronic
complications including CAN such as: genera-
tion of reactive oxygen species, endothelial
­dysfunction, and vascular injury [2, 45–48] than
does chronic hyperglycemia.
Moreover, recent human data have suggested
that acute changes in blood glucose levels may be
tied to development of chronic diabetes compli-
cations such as CAN, independent of HbA1c [2,
49, 50]. Our group has reported that increased
glucose variability, particularly with predomi-
nance of hypoglycemic stress, correlates with
reduced HRV, regardless of HbA1c, in T1D
patients [51], which lends support to a potential
role of high glucose variability and hypoglyce-
mia in development of CAN. In concert with our
findings, Rao et al. reported a significant decrease
in baroreflex sensitivity, maximum R-R interval
and maximal range of the R-R interval response
during hypoglycemia induced by a hypoglycemic-­
hyperinsulinemic clamp (2.8 mmol/L) compared
with baseline euglycemia, illustrating decreased
parasympathetic influence and disruption in car-
diovascular homeostasis in the setting of hypo-
glycemia [2, 52]. Increased glucose variability
has been associated with disruptions in sympa-
Cardiovascular Autonomic Neuropathy
thovagal balance in T2D, including in well con-
trolled patients with T2D without overt autonomic
neuropathy; therefore, it may be an early marker
of and contributor to progression of autonomic
dysfunction [2, 53]. In contrast, in the DCCT/
EDIC cohort no effect of glucose variability was
observed on risk of CAN after adjusting for mean
blood glucose and other risk factors [2, 54].
However, the limited number of glucose readings
available as part of the seven-point glucose pro-
files could not accurately characterize the full
spectrum of glucose variability, which is now
possible with the use of continuous glucose mon-
itoring systems (CGMs) [2, 54].
Several other pathways and mechanisms have
been documented in the risk for CAN. Oxidative
stress is an important player in the diabetic state,
and contributes to progressive deterioration and
subsequent complications, including CAN,
through several pathways: (1) the glycoxidation
and polysaturated fatty acid pathways with for-
mation of advanced glycosylation end-products
(AEGs) and reactive intermediates; (2) the reac-
tive nitrogen pathway generating reactive nitro-
gen species; (3) the myeloperoxidase pathway
with production of tyrosyl radical hypochlorous
acid; (4) the mitochondrial pathway with forma-
tion of superoxide hydrogen peroxide; and (5)
endothelial cell dysfunction generating excess
superoxide [2, 55–57]. Systemic oxidative stress
is increased in T1D participants with CAN, and
cardiac sympathetic tone, as measured by 11C-­
HED PET, is altered early in the course of T1D
[2, 33].
In addition, both experimental and clinical
data have implicated low-grade inflammation as
a key mediator in the development of CAN [2,
31, 33, 58–60]. In an experimental model of
streptozotocin-induced diabetes, Wang et  al.
reported that ischemic reperfusion injury (IR)
resulted in overexpression of the redox-sensitive
transcription factor, nuclear factor kappa B (NF-­
κB), in the sciatic endothelial cell and Schwann
cell [2, 60]. This was associated with increased
expression of intercellular adhesion molecule-1
(ICAM-1) and monocyte macrophage infiltration
in diabetic versus control nerve, lending support
to NF-κB activation as a key component in the
207
diabetic nerve inflammatory response. NF-κB is
activated by hyperglycemia, oxidative stress, and
inflammatory cytokines, thereby playing an
important role in regulating other inflammatory
genes and their associated actions, all potentially
leading to subsequent deficits in peripheral and
autonomic nerve fibers [2, 56, 58–60]. In humans,
our group and others reported significantly higher
markers of inflammation, including TNFα, in
patients with T1D and CAN compared with T1D
patients without CAN [2, 31, 33].
There is emerging evidence that alterations in
levels of intermediate metabolites of the glyco-
lytic pathway and tricarboxylic acid (TCA) cycle
are associated with diabetes complications. Our
group has recently reported the specific patterns
of TCA metabolites and amino acids associated
with both baseline and changes in measures of
CAN as well as its progression over 3  years of
follow-up in a cohort with T1D carefully pheno-
typed for CAN using CARTs, indices of HRV
and cardiac sympathetic imaging with 11C-HED
PET [2, 61]. There was a significant association
between higher fumarate and lower asparagine
and glutamine levels at baseline and measures of
CAN. In addition, glutamine and ornithine levels
at baseline correlated with CAN progression
after adjusting for diabetes duration and meta-
bolic parameters [2, 61]. Similarly, Ziegler et al.
analyzed 127 plasma lipid metabolites using MS
in participants from the German Diabetes Study
baseline cohort recently diagnosed with either
T1D or T2D and found that higher plasma levels
of specific lipid metabolites were closely linked
to cardiac autonomic dysfunction in recent-onset
T2D suggesting a role for perturbed lipid metab-
olism in the early development of CAN in T2D
[62]. These most recent reports open the door for
potential therapeutic interventions targeting these
changes to help prevent the development and pro-
gression of CAN in individuals with diabetes.
In summary, the specific mechanisms contrib-
uting to diabetic CAN in any given patient remain
elusive, and results from several clinical trials
evaluating the benefit of potential therapeutic
agents targeting proposed pathogenic pathways
in established CAN were disappointing. However,
with the emerging progress in diabetes technolo-
208 L. Ang and R. Pop-Busui

gies, as well as big data analyses and integration is an independent predictor of both cardiovascu-
of the genotype/phenotype interactions with mul- lar and all-cause mortality [2, 66, 67]. In a large
tiomics approaches, there is promise for identifi- cardiovascular outcome trial involving ~31,000
cation of specific degrees of susceptibility in an participants with stable cardiovascular disease
individual patient’s risk response [2]. with or without diabetes followed for an average
of 5  years, there was a positive correlation
between HR and both cardiovascular and all-­
4 Clinical Impact cause mortality [2, 68].
and Implications Moreover, the presence of CAN, as docu-
mented by deficits in sympathetic innervation
CAN has a broad spectrum of serious clinical with either iodine-123-labeled metaiodobenzyl-
implications. CAN has emerged as an indepen- guanidine (I-123 MIBG) scintigraphy or
dent predictor for both all-cause and CVD mor- 11C-meta-hydroxyephedrine positron emission
tality [2, 32, 63, 64]. In T1D, earlier studies from tomography (11C-HED PET), was associated
the EURODIAB IDDM cohort showed that with significantly higher mortality rates in
CAN, defined by a lower R–R ratio and postural patients with heart failure (HF) [2, 69, 70]. The
hypotension, was a strong risk marker for mortal- AdreView Myocardial Imaging for Risk
ity over 7  years of follow-up [2, 64]. Similar Evaluation in Heart Failure (ADMIRE-HF)
trends are described for prediabetes and T2D. For study described a positive association between
instance, a sub-study of a glucose tolerance-­ abnormal heart-to-mediastinum ratio (H/M) of
stratified sample of 605 participants from the less than 1.6 on I-123 MIBG scan, used as a
larger Hoorn Study population, which consisted measure of relative myocardial sympathetic
of 2484 individuals aged 50–75 years randomly nerve activity at baseline, and HF progression,
selected from the population register of the Dutch cardiac death, arrhythmic event, and all-cause
town of Hoorn, reported that CAN defined on one mortality in 960 patients with New  York Heart
CART and several HRV indices was associated Association functional class II/III HF over
with a twofold increase in the 9-year mortality 2  years of follow-up [2, 70]. Similarly, the
risk in T2D [2]. In the longitudinal, population-­ Prediction of Arrhythmic Events with Positron
based Monitoring of Trends and Determinants in Emission Tomography (PAREPET) study dem-
Cardiovascular Disease (MONICA)/Cooperative onstrated that baseline deficits in sympathetic
Health Research in the Region of Augsburg innervation on 11C-HED PET were a much stron-
(KORA) cohort study, reduced HRV and pro- ger predictor of sudden cardiac arrest in a cohort
longed corrected QT interval (QTc), both reflec- of patients with ischemic cardiomyopathy (50%
tive of CAN, were associated with excess of whom had diabetes) than either deficits in
mortality in a sample of the diabetic elderly gen- myocardial blood flow or infarct size during
eral population after 9 years of follow-up [2, 65]. 4  years of follow-up [2, 69]. Recent evidence
The conclusive evidence was obtained in greater demonstrated associations between cardiac auto-
than 10,000 participants with T2D in the Action nomic dysfunction, assessed by HRV indices,
to Control Cardiovascular Risk in Diabetes CARTs and particularly cardiac I-123 MIBG
(ACCORD) trial, which reported that the pres- imaging as a marker of cardiac sympathetic
ence of CAN at baseline, defined based on indi- integrity loss, and derangements in myocardial
ces of HRV extracted from 12 leads ECG blood flow regulation in T1D [2, 71]. Therefore,
recordings was an independent predictor of in addition to cardiovagal impairment, cardiac
increased mortality during follow-up after adjust- adrenergic denervation has emerged as a poten-
ing for all other traditional CVD risk factors, and tial risk factor in the development of CAN,
multiple medications [2, 63]. In addition, multi- although the current evidence on its actual role is
ple prospective cohort studies found that an ele- notably less robust and further confirmatory
vated resting HR, a surrogate indicator of CAN, studies are needed [2].
Cardiovascular Autonomic Neuropathy
The presence of CAN was also shown to pre-
dict the development of CVD. The DCCT/EDIC
study recently reported that T1D participants
diagnosed with CAN at DCCT closeout were at
significantly higher long-term risk for CVD
events during follow-up, independent of risk fac-
tors, such as age, duration, and BP; although, this
was not independent of glycemic control over
time as documented by the time-weighted HbA1c
[1, 2]. In the analysis of ~1100 patients with T2D
with no symptoms of CVD enrolled in the pro-
spective randomized controlled trial Detection of
Ischemia in Asymptomatic Diabetics (DIAD),
the presence of cardiovascular autonomic dys-
function, defined by being in the lowest quartile
of HR changes during the lying-to-standing test,
was strongly associated with silent myocardial
ischemia and subsequent cardiovascular events
[2, 72].
Another important cause of morbidity in CAN
is its potential role in the development of myocar-
dial dysfunction [32, 73, 74], diabetic cardiomy-
opathy [75], and heart failure, which has emerged
as a hidden and grossly underestimated compli-
cation of diabetes [2]. This has been observed in
both animal and human studies [2, 33, 73, 76,
77]. In humans, a pilot study in T1D reported that
participants with subclinical microangiopathy
demonstrated extensive deficits in left ventricular
sympathetic tone, as measured by 11C-HED PET,
leading to impaired myocardial blood flow regu-
lation and left ventricular diastolic function [33].
Similarly, in a cross-sectional study performed in
110 T2D participants with and without CAN, as
assessed by CARTs, HRV, and I-123 MIBG scan,
participants with CAN demonstrated signifi-
cantly reduced diastolic function on resting and
exercise echocardiography [2, 78]. Early stages
of diabetic cardiomyopathy in both T1D and T2D
are frequently described as deficits in left ventri-
cle relaxation and passive filling, which has been
supported by cohort and population-based stud-
ies [2, 75, 79, 80]. The role of CAN in the devel-
opment of heart failure is further supported by
data obtained from more than 900 participants
with T1D in the DCCT/EDIC study among
whom the presence of CAN was associated with
higher left ventricular mass and mass-to-volume
209
ratio, indices of left ventricular remodeling [2,
73]. The prevailing sympathetic activation and
imbalance associated with CAN may lead to crit-
ical changes in myocardial substrate usage result-
ing in regional ventricular wall motion
abnormalities and dysfunction [2].
Several studies also demonstrated that CAN,
defined based on CARTs and QT interval, is a
risk marker for ischemic stroke [81, 82], with
some studies reporting that CAN was associated
with up to a twofold increase in stroke risk [2, 34,
74].
Cardiovascular autonomic neuropathy may also
contribute to the development of diabetic nephrop-
athy in patients with diabetes, with one possible
mechanism including sympathetic overactivity
leading to constriction of renal pre-­glomerular ves-
sels, proteinuria, and further deterioration in renal
function [2, 83, 84]. More recently, Orlov et  al.
reported a strong association between CAN and
early progressive decline in renal function in
patients with T1D [85]. Similar findings have been
demonstrated in T2D.  In a cohort study of 204
patients with T2D, Tahrani et al. found that patients
with CAN, identified using HRV testing, had a
lower mean estimated glomerular filtration rate
(eGFR) and a higher prevalence rate of albumin-
uria and chronic kidney disease (CKD) at baseline
compared to those without CAN [2, 86].
Furthermore, they established CAN as an indepen-
dent predictor of eGFR decline over a subsequent
follow-up period of 2.5  years. More recent evi-
dence from our group also indicated that CAN is
associated with lesions of diabetic nephropathy as
documented by kidney biopsy in PIMA Indians
with T2D [2, 87]. In the Atherosclerosis Risk in
Communities (ARIC) study, a prospective observa-
tional cohort of 13,241 patients followed for a
median of 16 years, the investigators reported that
increasing resting HR and decreasing HRV corre-
lated with risk of end-stage renal disease (ESRD)
onset and CKD-­related hospitalization, indepen-
dent of diabetes, although notably diabetes was
present in 62% of those who developed ESRD [2,
88]. The authors also observed a pattern of abnor-
mal HRV preceding onset of clinical renal failure
by many years, speculating the potential role auto-
nomic imbalance may play in mediating renal
210
damage [2]. Another indicator of autonomic imbal-
ance is impaired diurnal BP variation, which has
been associated with chronic kidney disease devel-
opment and progression [2, 89–91]. Abnormalities
in the physiologic decrease in nocturnal BP
­compared to daytime BP, termed non-dipping and
reverse dipping, have displayed predictive potential
for progression of nephropathy in T2D, particu-
larly ESRD and dialysis [2, 92–95]. Finally, in
hemodialysis patients with diabetes who presented
with both CAN and left ventricular hypertrophy
(LVH), the associated mortality was increased fur-
ther compared to when CAN was present alone
without LVH [2, 96].
There is considerable evidence demonstrating
that diabetes has early, detrimental effects on reti-
nal nerve structure and function [2, 97]. Stem et al.
evaluated early neuroretinal changes in patients
with T1D and reported that T1D patients with no
to moderate diabetic retinopathy displayed
changes in inner retina anatomy and physiology,
as studied through extensive ophthalmic assess-
ment and optical coherence tomography imaging,
demonstrating the role of retinal autonomic neu-
ropathy in patients with T1D [2, 98, 99]. CAN is
also associated with increased risk of early prolif-
erative diabetic retinopathy (PDR) in individuals
with diabetes [2, 100]. Krolewski et al. evaluated
the association between autonomic dysfunction, as
assessed by R-R variation during deep breathing
and BP variation during position changes, and
presence of PDR in insulin-­ dependent diabetes
mellitus [100]. They reported that patients with
one abnormal test for CAN (mild CAN) and two
or more abnormal tests (significant CAN) had a
35-fold and 77-fold higher risk of PDR, respec-
tively, compared to those without CAN [2].
Moreover, there is emerging evidence support-
ing a positive association between CAN and devel-
opment of erectile dysfunction (ED) and female
sexual dysfunction in individuals with diabetes [2,
101, 102]. The DCCT/EDIC research group studied
the association between measures of CAN, as
assessed by CARTs, and female sexual dysfunction
and urinary incontinence, as evaluated by the
Female Sexual Function Index (FSFI-R) and
Sandvik Severity Index, respectively, in 580 women
with T1D reporting that greater odds of developing
L. Ang and R. Pop-Busui
female sexual dysfunction and urinary incontinence
among those with worse measure of CAN [101].
These findings suggest that the measures of CAN
may be useful for predicting increased risk of
female sexual dysfunction and urinary incontinence
development in women withT1D [2, 101]. The
DCCT/EDIC study also assessed erectile dysfunc-
tion via the International Index of Erectile Function
(IIEF) and lower urinary tract symptoms via the
American Urological Association Symptom Index
(AUASI) in men with T1D [102]. Those with erec-
tile dysfunction and/or lower urinary tract symp-
toms had lower R-R variation and Valsalva ratios
than those without supporting that CAN may be a
risk marker for the onset of urologic complications
in men with long-standing T1D [2, 102].
Key Points
• CAN is an independent risk factor for cardio-
vascular mortality, arrhythmia, silent isch-
emia, and major cardiovascular events.
• Emerging evidence demonstrates that CAN is
an important risk factor for the development
of myocardial dysfunction and heart failure in
both T1D and T2D that is emerging as an
underdiagnosed and important health problem
in diabetes.
• CAN is a predictor of the development of dia-
betic kidney disease and may be a risk marker
for the onset of urologic complications in both
men and women with T1D.
5 Diagnostic
Given the above, a timely diagnosis of CAN has
important clinical consequences.
5.1 Screening and Diagnosis
in Clinical Care
It could be quite challenging to detect CAN in its
early stages in routine clinical care, as patients
may be completely asymptomatic (subclinical
CAN) and may only present with decreased HRV
[1, 2].
Cardiovascular Autonomic Neuropathy 211

5.1.1 Symptoms of CAN symptoms for CAN is quite low requiring an

The most common symptoms of CAN occur upon
standing and include lightheadedness, weakness,
palpitations, faintness, and syncope [1, 2] (Table 1).
Thus, targeted questions to unveil the above symp-
toms should be included while a medical history is
taken in the office, although the specificity of these
appropriate differential [1, 2] (Table 1). In addition,
these symptoms occur late in the disease course [1,
2]. Providers may also consider screening patients
at risk for developing CAN for hypoglycemia
unawareness and vice-versa, as this may be associ-
ated with CAN [1, 103].

Table 1  Diagnostic algorithm for CAN (Adapted from Pop-Busui et  al, Diabetes Care 2017;40:136–154 with the
permission)?
Symptoms Signs/diagnostic tests Differential work-up
Resting Palpitations could Clinical exam: Resting HR  •  Anemia
Tachycardia be asymptomatic >100 bpm  •  Hyperthyroidism
 •  Fever
 •  CVD (atrial fibrillation, flutter,
other)
 •  Dehydration
 •  Adrenal insufficiency
 •  Medications
  – Sympathomimetic agents
(asthma)
  – Over-the-counter cold
agents containing ephedrine or
pseudoephedrine
  – Dietary supplements (e.g.,
ephedra alkaloids)
 •  Smoking, alcohol, caffeine
 •  Recreational drugs (cocaine,
amphetamines, methamphetamine,
mephedrone)
Orthostatic Light headedness Clinical Exam: A reduction of  •  Adrenal insufficiency
Hypotension Weakness >20 mmHg in the systolic BP or  •  Intravascular volume depletion
Faintness >10 mmHg in diastolic BP within   – Blood loss/Acute anemia
Visual impairment 3 min of standing or upright tilt   – Dehydration
Syncope table testing  •  Pregnancy/postpartum
 •  CVD
   –  Arrhythmias
  – Heart failure
  – Myocarditis
  – Pericarditis
   –  Valvular heart disease
 •  Alcohol
 •  Medications
   –  Antiadrenergics
   –  Antianginals
   –  Antiarrhythmics
   –  Anticholinergics
  – Diuretics
   –  ACEi/ARBs
  – Narcotics
  – Neuroleptics
  – Sedatives
Abnormal BP Non-dipping (lack 24-h BP monitoring
regulation of nocturnal BP
drop)
212
5.1.2 Clinical Signs of CAN
The physiologic HRV is a product of both sympa-
thetic and parasympathetic activity [1, 2]. Patients
with subclinical CAN present with a decrease in
HRV, usually with deep breathing or change in
posture, considered the earliest clinical indicator
of CAN [1, 2]. Although HRV testing is largely
confined to the research setting, it may be done in
the office by either (1) taking an ECG recording
as a patient begins to rise from a seated position or
(2) taking an ECG recording during 1–2 min of
deep breathing with calculation of HRV [1].
However, even these relatively simple methods
could be challenging in some clinical settings,
highlighting the need for easily accessible diag-
nostic tools and/or biomarkers for screening and
diagnosis of CAN for general clinical care.
As CAN progresses, individuals may present
with resting tachycardia with fixed heart rate
(>100 bpm), exercise intolerance, non-dipping and
reverse dipping, and in more advanced cases with
orthostatic hypotension (a fall in systolic or dia-
stolic blood pressure by >20 mmHg or >10 mmHg,
respectively, upon standing without an appropriate
increase in heart rate) [1, 104]. Orthostatic hypo-
tension is usually easy to document in the office by
measuring the blood pressure (BP) supine and
after standing, keeping also in mind that in most
cases of CAN, there is no compensatory increase
in the heart rate, despite hypotension [1]. To docu-
ment non-dipping or reverse dipping, the ambula-
tory BP monitoring (ABPM), a standard tool in the
current hypertension management, allows the
assessment of the diurnal BP patterns that is
mainly regulated by sleep-awake changes in the
autonomic cardiovascular function [1, 104].
However, it is important to note that none of these
signs are specific for diabetes related CAN.
CAN diagnosis includes documentation of
symptoms and signs of CAN (Table  1). In a
symptomatic patient with a history of poor glu-
cose control presenting with resting tachycardia
or postural hypotension, clinicians may not need
to perform additional CAN tests given costs and
burden after excluding other potential causes [2].
Importantly, the diagnosis of CAN requires an
extensive differential as many other comorbidi-
ties, such as cardiovascular disease, adrenal insuf-
ficiency, thyroid dysfunction, acute blood loss and
anemia, dehydration, and drug effects/interac-
L. Ang and R. Pop-Busui
tions may present with symptoms and/or signs
mimicking CAN, as illustrated in Table 1 [1, 2].
Key Points
• All patients should be assessed for CAN start-
ing 5 years after diagnosis or in the presence
of other forms of diabetic neuropathy and/or
other diabetic complications with symptoms
and signs.
• Tests excluding other comorbidities or drug
effects/interactions that could mimic CAN
should be performed.
• Screening for CAN should also be considered
in patients with hypoglycemia unawareness
and high glucose variability, prior to insulin
dose adjustments and/or perioperatively.
5.2 Diagnosis for Clinical
Research
Several tests are available to diagnose CAN
including: CARTs, ideally under paced breath-
ing; HRV studies that derive indices in time and
frequency domain using various algorithms; 24-h
BP profiles; baroreflex sensitivity testing; and
cardiac sympathetic imaging [1, 2].
CARTs remain the gold standard for autonomic
testing in both the clinical and research settings [1,
2, 104] and evaluate changes in the heart rate and
BP during several physiological maneuvers such as:
(1) deep breathing; (2) standing; and (3) a Valsalva
maneuver [1, 2]. Several factors may influence the
results of CARTs, such as intake of caffeine, tobacco
products, food, and some medications like beta
blockers. Therefore, these tests should be performed
by uniformly trained personalized personnel and
using standardized protocols to minimize variabil-
ity and bias. In addition, given the very large num-
ber of recording devices and particularly software
­operating with a “black-box” approach (output of
data without reporting the algorithm that calculated
the result) have emerged in recent years, it is neces-
sary to carefully select the most transparent and
verifiable technologies to use for reliable and repro-
ducible data. In general, it is currently accepted that
one abnormal test may indicate early or subclinical
CAN, while two or more abnormal tests are recom-
mended for a definite diagnosis of CAN [104].
Cardiovascular Autonomic Neuropathy
Given that age is one of the most important factors
affecting HR, age-­related normative are needed for
correct classification settings [1, 2, 104].
Indices of HRV are alternative CAN outcome
measures that have been used extensively. The
HRV can be evaluated either as time domain
measures such as: the difference between the lon-
gest and shortest RR interval, the standard devia-
tion of 5-min average of normal RR intervals
(SDNN), and the root-mean square of the differ-
ence of successive R–R intervals (rMSSD), or as
frequency domain measures obtained by spectral
analysis or other mathematical algorithms such
as: low, very low and high frequency power [1, 2,
104]. Although traditionally these indices were
initially derived from longer ECG recordings,
HRV-derived indices from standard 10-s 12-lead
ECG recordings have emerged as more feasible
alternative measures of CAN in population stud-
ies, and have been shown to independently pre-
dict CVD morbidity and mortality in several
large cohorts of people with and without diabetes
[11, 12, 63, 65] and to define prevalence rates.
Until recently, validation studies evaluating the
sensitivity and specificity of these ECG-based
HRV indices compared to the gold standard
CARTs assessment have not been available.
However, rigorous assessment of the sensitivity,
specificity, and probability of correct classifica-
tion of CAN using indices of HRV derived from
short ECG recordings in predicting CAN com-
pared to the more laborious method using the
gold standard CARTs were performed in the
DCCT/EDIC cohort that was phenotyped con-
comitantly with both standardized CARTs and
digitized ECGs. The indices of HRV had modest
predictive performance comparable with other
instruments that have been used successfully to
phenotype diabetic peripheral neuropathy in
large and diverse cohorts of T1D and T2D [105].
Cardiac sympathetic imaging using I-123
MIBG scintigraphy or 11C-HED PET, 24-h ambu-
latory blood pressure profiles, baroreflex sensi-
tivity testing to assess cardiac vagal and
sympathetic baroreflex function, and microneu-
rography, while valid and useful in some specific
research scenarios, require sophisticated infra-
structure, highly trained personnel, and are quite
expensive and time-consuming widely used in
research protocols [1, 2].
213
6 Treatment
As it is the case with diabetic peripheral neuropa-
thy, to date no disease modifying agents are
available for CAN treatment, likely due to the
very complex pathogenic pathways contributing
to development of CAN [1, 2]. Thus, the manage-
ment of CAN includes consideration of glucose
control and modification of other risk factors.
6.1 Prevention
In individuals with T1D, the conclusive data
obtained by the DCCT and later during EDIC study,
as well as evidence from several smaller trials
strongly support that tight control of blood glucose
implemented as early as possible in the disease
course is very effective for preventing CAN and
slowing its progression in patients with T1D [1–4,
106, 107]. For instance, during over an average of
6.5 years DCCT follow-up, intensive glucose con-
trol, with lower hemoglobin A1c levels, reduced the
risk of CAN by 45%, a benefit that persisted during
EDIC with a 30% reduction in incident CAN over
an additional 14  years of follow-up [3, 4, 107].
Similarly, the EURODIAB IDDM Complications
Study, found a significant correlation between
development of CAN and glycemic control [5].
The evidence for patients with T2D is less
conclusive, as in several large cohorts intensive
control of blood glucose had no effect on the risk
of CAN [108–110] although one could argue that
these trials may have not included the most reli-
able CAN outcomes, and most patients with T2D
enrolled had multiple associated comorbidities
and other risk factors, such as hypertension,
hyperlipidemia, obesity, and require
polypharmacy.
In contrast, glucose control as part of a multi-
factorial intervention that targeted besides hyper-
glycemia, hypertension, dyslipidemia, and
lifestyle, demonstrated a 63% reduction in the
rate of progression to CAN in a small T2D cohort
participating in the STENO-2 trial [111]. Most
recently, analyses from the ACCORD trial,
reported that after adjusting for multiple other
risk factors, intensive glucose treatment signifi-
cantly reduced CAN risk by 16% compared to
standard intervention in a large cohort of more
214 L. Ang and R. Pop-Busui

than 8000 T2D participants in the ACCORD trial
over a mean 5 years follow-up [10]. These data
confirmed a beneficial effect of intensive glyce-
mic therapy particularly in individuals without
cardiovascular disease at baseline. A similar ben-
efit on CAN was found for intensive blood pres-
sure control [10], while intensive treatment of
hyperlipidemia with statins and fenofibrate did
not. The population-based Coronary Artery Risk
Development in Young Adults (CARDIA) study
reported benefit of lifestyle interventions in pre-
vention of CAN progression [112]. In the
Diabetes Prevention Program (DPP), lifestyle
interventions improved measures of HRV over
time in participants with prediabetes [113].
6.2 Symptomatic Treatment
Management of orthostatic hypotension involves
both behavioral and pharmacological interventions
(Table 2). Behavioral supportive measures include:
(a) avoiding abrupt changes in body position; (b)
avoiding actions that elevate intra-­abdominal and
intra-thoracic pressures; (c) avoiding medications
that would exacerbate hypotension such as tricyclic
antidepressants, phenothiazines and diuretics; (d)
raise the head of the bed during sleep; (e) follow a
schedule of small and frequent meals to minimize
postprandial hypotension; (f) consider physical
counter-­pressure maneuvers such as leg crossing
and squatting; (g) hydrate with fluids and salt, if not
contraindicated [1, 2]. Pharmacological therapy
includes midodrine and droxidopa, both of which
are FDA approved for the management of ortho-
static hypotension and may be considered in
patients who fail non-pharmacological interven-
tions. However, it is recommended to proceed with
a very slow titration and use the minimally effec-
tive dose to avoid undesirable side effects. In
selected severe cases low dose fludrocortisone may
be also an option [1, 2] (Table 2).

Table 2  Treatment for orthostatic hypotension

Simple measures Physical maneuvers
 •  Avoiding sudden changes in body posture to the  •  Leg crossing
head-up position  •  Squatting
 •  Avoiding medications that aggravate hypotension  •  Muscle pumping
 •  Eating small, frequent meals
 •  Avoiding activities that involve straining
 •  Raising bed head
Pharmacological Dose Major adverse effects Notes
treatment Initial Effective
Midodrine 2.5 mg 2.5–10 mg. p.o, 1–4×/  • Supine –  FDA approved
p.o. day hypertension – Activates α-1 receptors
* First dose to be  •  Piloerection on arterioles and veins,
taken before arising  •  Pruritus increases total peripheral
 •  Paresthesias resistance
 •  Urinary retention
Droxidopa 100 mg 100–600 mg p.o 3×/  • Supine – FDA approved for the
p.o 3×/ day hypertension treatment of neurogenic
day To reduce the potential  •  Headache orthostatic hypotension
for supine  •  Dizziness but not specifically for
hypertension, elevate  •  Nausea individuals with
the head of the bed  • Hyperpyrexia and orthostatic hypotension
and give the last dose Confusion due to diabetes
at least 6 h prior to – α/β adrenergic agonist
bedtime – Effectiveness beyond
2 weeks of treatment has
not been established
Fludrocortisone 0.05 mg 0.1–0.2 mg/day, p.o.  • Supine – Not FDA approved for
po at hypertension orthostatic hypotension
bedtime  •  Hypokalemia – Synthetic
 •  Hypomagnesemia mineralocorticoid
 • Congestive heart – May require up to 1- to
failure 2-week to observe effects
 •  Peripheral edema
Cardiovascular Autonomic Neuropathy
7 Conclusions
CAN is prevalent chronic complication of diabe-
tes with a diverse clinical spectrum of symptoms
and signs. The prevalence rates remain high even
in contemporary cohorts and under the current
standards of care. Prompt recognition of CAN in
the clinical setting is crucial, as CAN is an inde-
pendent predictor of cardiovascular mortality,
cardiac arrhythmia, silent myocardial ischemia,
major cardiovascular events, and myocardial dys-
function leading to increased morbidity, substan-
tial economic burden, and poor quality of life in
patients with diabetes. Despite this knowledge,
screening and diagnosis of the disease in its early
stages remains elusive, and effective manage-
ment strategies and interventions to reverse
development and progression are missing. The
progress in diabetes technologies, combined with
big data analyses and integration of the genotype/
phenotype interactions with multiomics-derived
biomarkers, are likely to unveil specific increased
susceptibilities to or protections against
CAN.  This would also lead to identification of
modifiable factors and targeted techniques pro-
viding the framework for future clinical trials to
test novel interventions to either prevent progres-
sion or reverse disease in a personalized care
approach, ultimately reducing the health care
burden and optimizing patients’ quality of life.
References
1. Pop-Busui R, Boulton AJM, Feldman EL, Bril
V, Freeman R, Malik RA, Sosenko JM, Ziegler
D.  Diabetic neuropathy: a position statement by
the American Diabetes Association. Diabetes Care.
2017;40(1):136–54.
2. Ang L, Dillon B, Mizokami-Stout K, Pop-Busui
R.  Cardiovascular autonomic neuropathy: a
silent killer with long reach. Auton Neurosci.
2020;225:102646.
3. Braffett BH, Gubitosi-Klug RA, Albers JW,
Feldman EL, Martin CL, White NH, Orchard TJ,
Lopes-Virella M, Lachin JM, Pop-Busui R.  Risk
factors for diabetic peripheral neuropathy and car-
diovascular autonomic neuropathy in the Diabetes
Control and Complications Trial/Epidemiology of
Diabetes Interventions and Complications (DCCT/
EDIC) study. Diabetes. 2020;69(5):1000–10.
215
4. Pop-Busui R, Low PA, Waberski BH, Martin CL,
Albers JW, Feldman EL, Sommer C, Cleary PA,
Lachin JM, Herman WH, Group DER.  Effects of
prior intensive insulin therapy on cardiac auto-
nomic nervous system function in type 1 diabetes
mellitus: the Diabetes Control and Complications
Trial/Epidemiology of Diabetes Interventions and
Complications study (DCCT/EDIC). Circulation.
2009;119(22):2886–93.
5. Kempler P, Tesfaye S, Chaturvedi N, Stevens LK,
Webb DJ, Eaton S, Kerényi Z, Tamás G, Ward
J, Fuller J.  Autonomic neuropathy is associated
with increased cardiovascular risk factors: the
EURODIAB IDDM Complications Study. Diabet
Med. 2002;19(11):900–9.
6. Hansen C, Theilade S, Lajer M, Hansen T, Rossing
P.  Cardiovascular autonomic neuropathy and
bone metabolism in type 1 diabetes. Diabet Med.
2018;35(11):1596–604.
7. Pambianco G, Costacou T, Ellis D, Becker DJ,
Klein R, Orchard TJ.  The 30-year natural history
of type 1 diabetes complications: the Pittsburgh
Epidemiology of Diabetes Complications Study
experience. Diabetes. 2006;55(5):1463–9.
8. Mizokami-Stout KR, Bailey R, Foster NC, Ang L,
Aleppo G, Levy CJ, Rickels MR, Shah VN, Polsky
S, Katz M. 563-P: autonomic neuropathy in type 1
diabetes (T1D): findings from the T1D exchange.
Am Diabetes Assoc. 2019;68(Suppl 1):563.
9. Andersen ST, Witte DR, Fleischer J, Andersen H,
Lauritzen T, Jorgensen ME, Jensen TS, Pop-Busui
R, Charles M. Risk factors for the presence and pro-
gression of cardiovascular autonomic neuropathy
in type 2 diabetes: ADDITION-Denmark. Diabetes
Care. 2018;41(12):2586–94.
10. Tang Y, Shah H, Bueno Junior CR, Sun X, Mitri J,
Sambataro M, Sambado L, Gerstein HC, Fonseca V,
Doria A, Pop-Busui R. Intensive risk factor manage-
ment and cardiovascular autonomic neuropathy in
type 2 diabetes: the ACCORD trial. Diabetes Care.
2021;44(1):164–73.
11. Mather KJ, Bebu I, Baker C, Cohen RM, Crandall
JP, DeSouza C, Green JB, Kirkman MS, Krause-­
Steinrauf H, Larkin M.  Prevalence of microvas-
cular and macrovascular disease in the Glycemia
Reduction Approaches in Diabetes-A Comparative
Effectiveness (GRADE) Study cohort. Diabetes Res
Clin Pract. 2020;165:108235.
12. O’Neal WT, Chen LY, Nazarian S, Soliman
EZ. Reference ranges for short-term heart rate vari-
ability measures in individuals free of cardiovascular
disease: the Multi-Ethnic Study of Atherosclerosis
(MESA). J Electrocardiol. 2016;49(5):686–90.
13. Diakakis GF, Parthenakis FI, Patrianakos AP,
Koukouraki SI, Stathaki MI, Karkavitsas NS,
Vardas PE.  Myocardial sympathetic innervation in
patients with impaired glucose tolerance: relation-
ship to subclinical inflammation. Cardiovasc Pathol.
2008;17(3):172–7.
216
14. Papanas N, Vinik AI, Ziegler D. Neuropathy in pre-
diabetes: does the clock start ticking early? Nat Rev
Endocrinol. 2011;7(11):682.
15. Putz Z, Tabák ÁG, Tóth N, Istenes I, Németh N,
Gandhi RA, Hermányi Z, Keresztes K, Jermendy G,
Tesfaye S. Noninvasive evaluation of neural impair-
ment in subjects with impaired glucose tolerance.
Diabetes Care. 2009;32(1):181–3.
16. Wu J-S, Yang Y-C, Lin T-S, Huang Y-H, Chen J-J, Lu
F-H, Wu C-H, Chang C-J. Epidemiological evidence
of altered cardiac autonomic function in subjects
with impaired glucose tolerance but not isolated
impaired fasting glucose. J Clin Endocrinol Metab.
2007;92(10):3885–9.
17. Ziegler D, Voss A, Rathmann W, Strom A, Perz S,
Roden M, Peters A, Meisinger C.  Increased preva-
lence of cardiac autonomic dysfunction at differ-
ent degrees of glucose intolerance in the general
population: the KORA S4 survey. Diabetologia.
2015;58(5):1118–28.
18. Stein P, Barzilay J, Domitrovich P, Chaves P,
Gottdiener J, Heckbert S, Kronmal R. The relation-
ship of heart rate and heart rate variability to non-­
diabetic fasting glucose levels and the metabolic
syndrome: the Cardiovascular Health Study. Diabet
Med. 2007;24(8):855–63.
19. Jaiswal M, Divers J, Urbina EM, Dabelea D, Bell
RA, Pettitt DJ, Imperatore G, Pihoker C, Dolan LM,
Liese AD, Marcovina S, Linder B, Feldman EL,
Pop-Busui R, Group SfDiYS. Cardiovascular auto-
nomic neuropathy in adolescents and young adults
with type 1 and type 2 diabetes: the SEARCH for
Diabetes in Youth Cohort Study. Pediatr Diabetes.
2018;19(4):680–9.
20. Dabelea D, Stafford JM, Mayer-Davis EJ,
D’Agostino R, Dolan L, Imperatore G, Linder
B, Lawrence JM, Marcovina SM, Mottl
AK.  Association of type 1 diabetes vs type 2 dia-
betes diagnosed during childhood and adolescence
with complications during teenage years and young
adulthood. JAMA. 2017;317(8):825–35.
21. Maser RE, Pfeifer MA, Dorman JS, Kuller LH,
Becker DJ, Orchard TJ.  Diabetic autonomic
neuropathy and cardiovascular risk: Pittsburgh
Epidemiology of Diabetes Complications Study
III. JAMA Intern Med. 1990;150(6):1218–22.
22. Stella P, Ellis D, Maser RE, Orchard
TJ.  Cardiovascular autonomic neuropathy (expi-
ration and inspiration ratio) in type 1 diabetes:
incidence and predictors. J Diabetes Complicat.
2000;14(1):1–6.
23. Tesfaye S, Chaturvedi N, Eaton SE, Ward JD,
Manes C, Ionescu-Tirgoviste C, Witte DR, Fuller
JH. Vascular risk factors and diabetic neuropathy. N
Engl J Med. 2005;352(4):341–50.
24. Andersen ST, Witte DR, Andersen H, Bjerg L,
Bruun NH, Jørgensen ME, Finnerup NB, Lauritzen
T, Jensen TS, Tankisi H.  Risk-factor trajectories
preceding diabetic polyneuropathy: ADDITION-­
Denmark. Diabetes Care. 2018;41(9):1955–62.
L. Ang and R. Pop-Busui
25. Reynolds EL, Akinci G, Banerjee M, Looker HC,
Patterson A, Nelson RG, Feldman EL, Callaghan
BC. The determinants of complication trajectories in
American Indians with type 2 diabetes. JCI Insight.
2021;6(10):e146849.
26. Frontoni S, Bracaglia D, Baroni A, Pellegrini F,
Perna M, Cicconetti E, Ciampittiello G, Menzinger
G, Gambardella S.  Early autonomic dysfunc-
tion in glucose-tolerant but insulin-resistant off-
spring of type 2 diabetic patients. Hypertension.
2003;41(6):1223–7.
27. Laitinen T, Vauhkonen I, Niskanen LK, Hartikainen
J, Länsimies EA, Uusitupa M, Laakso M.  Power
spectral analysis of heart rate variability during
hyperinsulinemia in nondiabetic offspring of type 2
diabetic patients: evidence for possible early auto-
nomic dysfunction in insulin-resistant subjects.
Diabetes. 1999;48(6):1295–9.
28. Duvernoy CS, Raffel DM, Swanson SD, Jaiswal
M, Mueller G, Ibrahim E-S, Pennathur S, Plunkett
C, Stojanovska J, Brown MB, Pop-Busui R.  Left
ventricular metabolism, function, and sympathetic
innervation in men and women with type 1 diabetes.
J Nucl Cardiol. 2016;23(5):960–9.
29. Winkley K, Thomas S, Sivaprasad S, Chamley
M, Stahl D, Ismail K, Amiel S.  The clinical
characteristics at diagnosis of type 2 diabetes
­
in a multi-­ ethnic population: the South London
Diabetes cohort (SOUL-D). Diabetologia.
2013;56(6):1272–81.
30. Tahrani AA, Altaf QA, Piya MK, Barnett
AH. Peripheral and autonomic neuropathy in South
Asians and White Caucasians with type 2 diabetes
mellitus: possible explanations for epidemiological
differences. J Diabetes Res. 2017;2017:1273789.
31. Lieb DC, Parson HK, Mamikunian G, Vinik
AI.  Cardiac autonomic imbalance in newly diag-
nosed and established diabetes is associated with
markers of adipose tissue inflammation. Exp
Diabetes Res. 2012;2012:8.
32. Pop-Busui R. What do we know and we do not know
about cardiovascular autonomic neuropathy in dia-
betes. J Cardiovasc Transl Res. 2012;5(4):463–78.
33. Pop-Busui R, Kirkwood I, Schmid H, Marinescu
V, Schroeder J, Larkin D, Yamada E, Raffel DM,
Stevens MJ. Sympathetic dysfunction in type 1 dia-
betes: association with impaired myocardial blood
flow reserve and diastolic dysfunction. J Am Coll
Cardiol. 2004;44(12):2368–74.
34. Vinik AI, Ziegler D.  Diabetic cardiovascular auto-
nomic neuropathy. Circulation. 2007;115(3):387–97.
35. Nathan DM, Genuth S, Lachin J, Cleary P, Crofford
O, Davis M, Rand L, Siebert C. The effect of inten-
sive treatment of diabetes on the development
and progression of long-term complications in
insulin-­dependent diabetes mellitus. N Engl J Med.
1993;329(14):977–86.
36. Glycemic targets: standards of medical care in
diabetes—2021. Diabetes Care. 2021;44(Suppl.
1):S73–84.
Cardiovascular Autonomic Neuropathy
37. Ang L, Jaiswal M, Martin C, Pop-Busui R. Glucose
control and diabetic neuropathy: lessons from recent
large clinical trials. Curr Diab Rep. 2014;14(9):1–15.
38. Gerstein HC, Miller ME, Byington RP, Goff DC Jr,
Bigger JT, Buse JB, Cushman WC, Genuth S, Ismail-­
Beigi F, Grimm RH Jr, Probstfield JL, Simons-­
Morton DG, Friedewald WT.  Effects of intensive
glucose lowering in type 2 diabetes. N Engl J Med.
2008;358(24):2545–59.
39. Gill GV, Woodward A, Casson IF, Weston
PJ.  Cardiac arrhythmia and nocturnal hypoglycae-
mia in type 1 diabetes—the ‘dead in bed’ syndrome
revisited. Diabetologia. 2009;52(1):42–5.
40. Tu E, Twigg SM, Semsarian C. Sudden death in type
1 diabetes: the mystery of the ‘dead in bed’ syn-
drome. Int J Cardiol. 2010;138(1):91–3.
41. Weston PJ, Gill GV.  Is undetected autonomic dys-
function responsible for sudden death in type 1
diabetes mellitus? The ‘dead in bed’ syndrome revis-
ited. Diabet Med. 1999;16(8):626–31.
42. Cryer PE, Davis SN, Shamoon H. Hypoglycemia in
diabetes. Diabetes Care. 2003;26(6):1902–12.
43. Desouza CV, Bolli GB, Fonseca V.  Hypoglycemia,
diabetes, and cardiovascular events. Diabetes Care.
2010;33(6):1389–94.
44. Limberg J, Farni K, Taylor J, Dube S, Basu A, Basu
R, Wehrwein E, Joyner M. Autonomic control dur-
ing acute hypoglycemia in type 1 diabetes mellitus.
Clin Auton Res. 2014;24(6):275–83.
45. Ceriello A, Esposito K, Piconi L, Ihnat MA, Thorpe
JE, Testa R, Boemi M, Giugliano D.  Oscillating
glucose is more deleterious to endothelial func-
tion and oxidative stress than mean glucose in
normal and type 2 diabetic patients. Diabetes.
2008;57(5):1349–54.
46. Horváth E, Benkő R, Kiss L, Muranyi M, Pek T,
Fekete K, Barany T, Somlai A, Csordas A, Szabo
C.  Rapid ‘glycaemic swings’ induce nitrosative
stress, activate poly (ADP-ribose) polymerase and
impair endothelial function in a rat model of diabetes
mellitus. Diabetologia. 2009;52(5):952–61.
47. Monnier L, Mas E, Ginet C, et al. Activation of oxi-
dative stress by acute glucose fluctuations compared
with sustained chronic hyperglycemia in patients
with type 2 diabetes. JAMA. 2006;295(14):1681–7.
48. Saisho Y. Glycemic variability and oxidative stress:
a link between diabetes and cardiovascular disease?
Int J Mol Sci. 2014;15(10):18381–406.
49. Fleischer J.  Diabetic autonomic imbalance and
glycemic variability. J Diabetes Sci Technol.
2012;6(5):1207–15.
50. Fleischer J, Cichosz SL, Hoeyem P, Laugesen E,
Poulsen PL, Christiansen JS, Tarnow L, Hansen
TK. Glycemic variability is associated with reduced
cardiac autonomic modulation in women with type 2
diabetes. Diabetes Care. 2015;38(4):682–8.
51. Jaiswal M, McKeon K, Comment N, Henderson
J, Swanson S, Plunkett C, Nelson P, Pop-Busui
R.  Association between impaired cardiovascu-
lar autonomic function and hypoglycemia in
217
patients with type 1 diabetes. Diabetes Care.
2014;37(9):2616–21.
52. Rao AD, Bonyhay I, Dankwa J, Baimas-George
M, Kneen L, Ballatori S, Freeman R, Adler
GK. Baroreflex sensitivity impairment during hypo-
glycemia: implications for cardiovascular control.
Diabetes. 2016;65(1):209–15.
53. Di Flaviani A, Picconi F, Di Stefano P, Giordani I,
Malandrucco I, Maggio P, Palazzo P, Sgreccia F,
Peraldo C, Farina F. Impact of glycemic and blood
pressure variability on surrogate measures of car-
diovascular outcomes in type 2 diabetic patients.
Diabetes Care. 2011;34(7):1605–9.
54. Lachin JM, Bebu I, Bergenstal RM, Pop-Busui
R, Service FJ, Zinman B, Nathan DM, Group
DER.  Association of glycemic variability in type
1 diabetes with progression of microvascular out-
comes in the diabetes control and complications
trial. Diabetes Care. 2017;40(6):777–83.
55. Hoeldtke RD, Bryner KD, VanDyke K.  Oxidative
stress and autonomic nerve function in early type 1
diabetes. Clin Auton Res. 2011;21(1):19–28.
56. Kellogg AP, Wiggin TD, Larkin DD, Hayes JM,
Stevens MJ, Pop-Busui R.  Protective effects of
cyclooxygenase-2 gene inactivation against periph-
eral nerve dysfunction and intraepidermal nerve
fiber loss in experimental diabetes. Diabetes.
2007;56(12):2997–3005.
57. Pennathur S, Heinecke JW.  Mechanisms for oxi-
dative stress in diabetic cardiovascular disease.
Antioxid Redox Signal. 2007;9(7):955–69.
58. Cameron NE, Cotter MA.  Pro-inflammatory
mechanisms in diabetic neuropathy: focus on the
nuclear factor kappa B pathway. Curr Drug Targets.
2008;9(1):60–7.
59. Pop-Busui R, Ang L, Holmes C, Gallagher K,
Feldman EL.  Inflammation as a therapeutic tar-
get for diabetic neuropathies. Curr Diab Rep.
2016;16(3):29.
60. Wang Y, Schmeichel AM, Iida H, Schmelzer JD,
Low PA. Enhanced inflammatory response via acti-
vation of NF-κB in acute experimental diabetic neu-
ropathy subjected to ischemia–reperfusion injury. J
Neurol Sci. 2006;247(1):47–52.
61. Mathew AV, Jaiswal M, Ang L, Michailidis G,
Pennathur S, Pop-Busui R.  Impaired amino acid
and TCA metabolism and cardiovascular autonomic
neuropathy progression in type 1 diabetes. Diabetes.
2019;68(10):2035–44.
62. Ziegler D, Strom A, Straßburger K, Knebel B, Bönhof
GJ, Kotzka J, Szendroedi J, Roden M. Association
of cardiac autonomic dysfunction with higher levels
of plasma lipid metabolites in recent-onset type 2
diabetes. Diabetologia. 2021;64(2):458–68.
63. Pop-Busui R, Evans GW, Gerstein HC, Fonseca V,
Fleg JL, Hoogwerf BJ, Genuth S, Grimm RH, Corson
MA, Prineas R, Group tAS. Effects of cardiac auto-
nomic dysfunction on mortality risk in the Action to
Control Cardiovascular Risk in Diabetes (ACCORD)
trial. Diabetes Care. 2010;33(7):1578–84.
218
64. Soedamah-Muthu SS, Chaturvedi N, Witte DR,
Stevens LK, Porta M, Fuller JH.  Relationship
between risk factors and mortality in type 1 diabetic
patients in Europe: the EURODIAB Prospective
Complications Study (PCS). Diabetes Care.
2008;31(7):1360–6.
65. Ziegler D, Zentai CP, Perz S, Rathmann W, Haastert
B, Döring A, Meisinger C.  Prediction of mortal-
ity using measures of cardiac autonomic dysfunc-
tion in the diabetic and nondiabetic population: the
MONICA/KORA Augsburg Cohort Study. Diabetes
Care. 2008;31(3):556–61.
66. Ho JE, Bittner V, DeMicco DA, Breazna A,
Deedwania PC, Waters DD. Usefulness of heart rate
at rest as a predictor of mortality, hospitalization
for heart failure, myocardial infarction, and stroke
in patients with stable coronary heart disease (Data
from the Treating to New Targets [TNT] trial). Am J
Cardiol. 2010;105(7):905–11.
67. Nauman J, Janszky I, Vatten LJ, Wisløff U. Temporal
changes in resting heart rate and deaths from isch-
emic heart disease. JAMA. 2011;306(23):2579–87.
68. Lonn EM, Rambihar S, Gao P, Custodis FF, Sliwa K,
Teo KK, Yusuf S, Böhm M. Heart rate is associated
with increased risk of major cardiovascular events,
cardiovascular and all-cause death in patients with
stable chronic cardiovascular disease: an analysis
of ONTARGET/TRANSCEND.  Clin Res Cardiol.
2014;103(2):149–59.
69. Fallavollita JA, Heavey BM, Luisi AJ Jr, Michalek
SM, Baldwa S, Mashtare TL Jr, Hutson AD, Dekemp
RA, Haka MS, Sajjad M, Cimato TR, Curtis AB,
Cain ME, Canty JM Jr. Regional myocardial sym-
pathetic denervation predicts the risk of sudden car-
diac arrest in ischemic cardiomyopathy. J Am Coll
Cardiol. 2014;63(2):141–9.
70. Jacobson AF, Senior R, Cerqueira MD, Wong
ND, Thomas GS, Lopez VA, Agostini D, Weiland
F, Chandna H, Narula J.  Myocardial iodine-123
meta-iodobenzylguanidine imaging and cardiac
events in heart failure: results of the prospective
ADMIRE-HF (AdreView Myocardial Imaging for
Risk Evaluation in Heart Failure) study. J Am Coll
Cardiol. 2010;55(20):2212–21.
71. Zobel EH, Hasbak P, Winther SA, Hansen CS,
Fleischer J, von Scholten BJ, Holmvang L, Kjaer A,
Rossing P, Hansen TW. Cardiac autonomic function
is associated with myocardial flow reserve in type 1
diabetes. Diabetes. 2019;68(6):1277–86.
72. Young LH, Frans JT, Chyun DA, Davey JA, Barrett
EJ, Taillefer R, Heller GV, Iskandrian AE, Wittlin
SD, Filipchuk N. Cardiac outcomes after screening
for asymptomatic coronary artery disease in patients
with type 2 diabetes: the DIAD study: a randomized
controlled trial. JAMA. 2009;301(15):1547–55.
73. Pop-Busui R, Cleary PA, Braffett BH, Martin
CL, Herman WH, Low PA, Lima JAC, Bluemke
DA.  Association between cardiovascular auto-
nomic neuropathy and left ventricular dysfunc-
tion: DCCT/EDIC study (Diabetes Control and
L. Ang and R. Pop-Busui
Complications Trial/Epidemiology of Diabetes
Interventions and Complications). J Am Coll
Cardiol. 2013;61(4):447–54.
74. Spallone V, Ziegler D, Freeman R, Bernardi L,
Frontoni S, Pop-Busui R, Stevens M, Kempler P,
Hilsted J, Tesfaye S, Low P, Valensi P, On behalf
of The Toronto Consensus Panel on Diabetic
N.  Cardiovascular autonomic neuropathy in
diabetes: clinical impact, assessment, diagno-
sis, and management. Diabetes Metab Res Rev.
2011;27(7):639–53.
75. Fang ZY, Prins JB, Marwick TH.  Diabetic cardio-
myopathy: evidence, mechanisms, and therapeutic
implications. Endocr Rev. 2004;25(4):543–67.
76. Mizushige K, Yao L, Noma T, Kiyomoto H, Yu Y,
Hosomi N, Ohmori K, Matsuo H. Alteration in left
ventricular diastolic filling and accumulation of
myocardial collagen at insulin-resistant prediabetic
stage of a type II diabetic rat model. Circulation.
2000;101(8):899–907.
77. Semeniuk LM, Kryski AJ, Severson
DL.  Echocardiographic assessment of cardiac
function in diabetic db/db and transgenic db/
db-hGLUT4 mice. Am J Phys Heart Circ Phys.
2002;283(3):H976–82.
78. Sacre JW, Franjic B, Jellis CL, Jenkins C, Coombes
JS, Marwick TH. Association of cardiac autonomic
neuropathy with subclinical myocardial dysfunc-
tion in type 2 diabetes. JACC Cardiovasc Imaging.
2010;3(12):1207–15.
79. Astorri E, Fiorina P, Astorri A, Contini GA, Albertini
D, Magnati G, Lanfredini M. Isolated and preclini-
cal impairment of left ventricular filling in insulin-­
dependent and non-insulin-dependent diabetic
patients. Clin Cardiol. 1997;20(6):536–40.
80. Schannwell CM, Schneppenheim M, Perings S,
Plehn G, Strauer B.  Left ventricular diastolic dys-
function as an early manifestation of diabetic cardio-
myopathy. Cardiology. 2002;98(1-2):33–9.
81. Cohen JA, Estacio RO, Lundgren RA, Esler AL,
Schrier RW. Diabetic autonomic neuropathy is asso-
ciated with an increased incidence of strokes. Auton
Neurosci. 2003;108(1):73–8.
82. Ko S, Song K, Park S, Kim S, Cha B, Son H, Moon
K, Yoo K, Park Y, Cho J.  Cardiovascular auto-
nomic dysfunction predicts acute ischaemic stroke
in patients with type 2 diabetes mellitus: a 7-year
follow-up study. Diabet Med. 2008;25(10):1171–7.
83. Axelrod S, Lishner M, Oz O, Bernheim J, Ravid
M. Spectral analysis of fluctuations in heart rate: an
objective evaluation of autonomic nervous control in
chronic renal failure. Nephron. 1987;45(3):202–6.
84. Converse RL Jr, Jacobsen TN, Toto RD, Jost CM,
Cosentino F, Fouad-Tarazi F, Victor RG. Sympathetic
overactivity in patients with chronic renal failure. N
Engl J Med. 1992;327(27):1912–8.
85. Orlov S, Cherney DZ, Pop-Busui R, Lovblom LE,
Ficociello LH, Smiles AM, Warram JH, Krolewski
AS, Perkins BA. Cardiac autonomic neuropathy and
early progressive renal decline in patients with non-
Cardiovascular Autonomic Neuropathy
macroalbuminuric type 1 diabetes. Clin J Am Soc
Nephrol. 2015;10(7):1136–44.
86. Tahrani AA, Dubb K, Raymond NT, Begum S, Altaf
QA, Sadiqi H, Piya MK, Stevens MJ. Cardiac auto-
nomic neuropathy predicts renal function decline
in patients with type 2 diabetes: a cohort study.
Diabetologia. 2014;57(6):1249–56.
87. Wheelock KM, Jaiswal M, Martin CL, Fufaa GD,
Weil EJ, Lemley KV, Yee B, Feldman E, Brosius
FC III, Knowler WC.  Cardiovascular autonomic
neuropathy associates with nephropathy lesions in
American Indians with type 2 diabetes. J Diabetes
Complicat. 2016;30(5):873–9.
88. Brotman DJ, Bash LD, Qayyum R, Crews D,
Whitsel EA, Astor BC, Coresh J. Heart rate variabil-
ity predicts ESRD and CKD-related hospitalization.
J Am Soc Nephrol. 2010;21(9):1560–70.
89. Davidson MB, Hix JK, Vidt DG, Brotman
DJ.  Association of impaired diurnal blood pres-
sure variation with a subsequent decline in
glomerular filtration rate. Arch Intern Med.
2006;166(8):846–52.
90. Farmer C, Goldsmith D, Quin JD, Dallyn P, Cox J,
Kingswood JC, Sharpstone P. Progression of diabetic
nephropathy—is diurnal blood pressure rhythm as
important as absolute blood pressure level? Nephrol
Dial Transplant. 1998;13(3):635–9.
91. Lurbe E, Redon J, Kesani A, Pascual JM, Tacons J,
Alvarez V, Batlle D. Increase in nocturnal blood pres-
sure and progression to microalbuminuria in type 1
diabetes. N Engl J Med. 2002;347(11):797–805.
92. Knudsen ST, Laugesen E, Hansen K, Bek T,
Mogensen C, Poulsen P.  Ambulatory pulse pres-
sure, decreased nocturnal blood pressure reduction
and progression of nephropathy in type 2 diabetic
patients. Diabetologia. 2009;52(4):698–704.
93. Nakano S, Ogihara M, Tamura C, Kitazawa M,
Nishizawa M, Kigoshi T, Uchida K. Reversed circa-
dian blood pressure rhythm independently predicts
endstage renal failure in non–insulin-dependent
diabetes mellitus subjects. J Diabetes Complicat.
1999;13(4):224–31.
94. Palmas W, Pickering T, Teresi J, Schwartz JE, Eguchi
K, Field L, Weinstock RS, Shea S. Nocturnal blood
pressure elevation predicts progression of albumin-
uria in elderly people with type 2 diabetes. J Clin
Hypertens. 2008;10(1):12–20.
95. Spallone V. Blood pressure variability and autonomic
dysfunction. Curr Diab Rep. 2018;18(12):1–14.
96. Nishimura M, Hashimoto T, Kobayashi H, Fukuda
T, Okino K, Yamamoto N, Nakamura N, Yoshikawa
T, Takahashi H, Ono T. Association between cardio-
vascular autonomic neuropathy and left ventricu-
lar hypertrophy in diabetic haemodialysis patients.
Nephrol Dial Transplant. 2004;19(10):2532–8.
97. Simó R, Hernández C. Neurodegeneration is an early
event in diabetic retinopathy: therapeutic implica-
tions. Br J Ophthalmol. 2012;96(10):1285–90.
98. Stem MS, Dunbar GE, Jackson GR, Farsiu S, Pop-­
Busui R, Gardner TW. Glucose variability and inner
219
retinal sensory neuropathy in persons with type 1
diabetes mellitus. Eye (Lond). 2016;30(6):825–32.
99. Antonetti DA, Klein R, Gardner TW. Diabetic reti-
nopathy. N Engl J Med. 2012;366(13):1227–39.
100. Krolewski AS, Barzilay J, Warram JH, Martin BC,
Pfeifer M, Rand LI.  Risk of early-onset prolif-
erative retinopathy in IDDM is closely related to
cardiovascular autonomic neuropathy. Diabetes.
1992;41(4):430–7.
101. Hotaling JM, Sarma AV, Patel DP, Braffett BH, Cleary
PA, Feldman E, Herman WH, Martin CL, Jacobson
AM, Wessells H, Pop-Busui R.  Cardiovascular
autonomic neuropathy, sexual dysfunction, and uri-
nary incontinence in women with type 1 diabetes.
Diabetes Care. 2016;39(9):1587–93.
102. Pop-Busui R, Hotaling J, Braffett BH, Cleary PA,
Dunn RL, Martin CL, Jacobson AM, Wessells H,
Sarma AV, Group DER.  Cardiovascular autonomic
neuropathy, erectile dysfunction and lower urinary
tract symptoms in men with type 1 diabetes: findings
from the DCCT/EDIC. J Urol. 2015;193(6):2045–51.
103. Lin YK, Fisher SJ, Pop-Busui R.  Hypoglycemia
unawareness and autonomic dysfunction in diabetes:
lessons learned and roles of diabetes technologies. J
Diabetes Investig. 2020;11(6):1388–402.
104. Spallone V, Bellavere F, Scionti L, Maule S,
Quadri R, Bax G, Melga P, Viviani GL, Esposito K,
Morganti R, Cortelli P, Diabetic Neuropathy Study
Group of the Italian Society of D. Recommendations
for the use of cardiovascular tests in diagnosing dia-
betic autonomic neuropathy. Nutr Metab Cardiovasc
Dis. 2011;21(1):69–78.
105. Pop-Busui R, Backlund JYC, Bebu I, Braffett BH,
Lorenzi G, White NH, Lachin JM, Soliman EZ,
Group DER. The utility of using ECG measures of
heart rate variability as a measure of cardiovascular
autonomic neuropathy in type 1 diabetes. J Diabetes
Investig. 2022;13(1):125–33.
106. Margolis KL, O’Connor PJ, Morgan TM, Buse JB,
Cohen RM, Cushman WC, Cutler JA, Evans GW,
Gerstein HC, Grimm RH Jr, Lipkin EW, Venkat
Narayan KM, Riddle MC Jr, Sood A, Goff DC Jr.
Outcomes of combined cardiovascular risk fac-
tor management strategies in type 2 diabetes:
the ACCORD randomized trial. Diabetes Care.
2014;37(6):1721–8.
107. Martin CL, Albers JW, Pop-Busui R.  Neuropathy
and related findings in the diabetes control and
complications trial/epidemiology of diabetes inter-
ventions and complications study. Diabetes Care.
2014;37(1):31–8.
108. Azad N, Emanuele NV, Abraira C, Henderson WG,
Colwell J, Levin SR, Nuttall FQ, Comstock JP, Sawin
CT, Silbert C, Rubino FA.  The effects of intensive
glycemic control on neuropathy in the VA coopera-
tive study on type II diabetes mellitus (VA CSDM). J
Diabetes Complicat. 1999;13(5-6):307–13.
109. Charles M, Fleischer J, Witte DR, Ejskjaer N, Borch-­
Johnsen K, Lauritzen T, Sandbaek A. Impact of early
detection and treatment of diabetes on the 6-year
220 L. Ang and R. Pop-Busui

prevalence of cardiac autonomic neuropathy in vascular disease in patients with type 2 diabetes. N
people with screen-detected diabetes: ADDITION-­ Engl J Med. 2003;348(5):383–93.
Denmark, a cluster-randomised study. Diabetologia. 112. Carnethon MR, Jacobs DR Jr, Sidney S, Sternfeld
2013;56(1):101–8. B, Gidding SS, Shoushtari C, Liu K.  A longitudi-
110. Duckworth W, Abraira C, Moritz T, Reda D, nal study of physical activity and heart rate recov-
Emanuele N, Reaven PD, Zieve FJ, Marks J, Davis ery: CARDIA, 1987-1993. Med Sci Sports Exerc.
SN, Hayward R, Warren SR, Goldman S, McCarren 2005;37(4):606–12.
M, Vitek ME, Henderson WG, Huang GD. Glucose 113. Carnethon MR, Prineas RJ, Temprosa M, Zhang
control and vascular complications in veterans with ZM, Uwaifo G, Molitch ME. The association among
type 2 diabetes. N Engl J Med. 2009;360(2):129–39. autonomic nervous system function, incident diabe-
111. Gaede P, Vedel P, Larsen N, Jensen GV, Parving HH, tes, and intervention arm in the diabetes prevention
Pedersen O. Multifactorial intervention and cardio- program. Diabetes Care. 2006;29(4):914–9.
 Psychosocial Aspects of Diabetic
Neuropathy: From Description
to Interventions
Loretta Vileikyte and Frans Pouwer
1 Introduction
The diabetic neuropathies are a collection of
nerve disorders caused by diabetes that affect dif-
ferent parts of the nervous system and present
with diverse clinical manifestations of sensorim-
otor and autonomic dysfunction [1]. Distal sym-
metric polyneuropathy (DPN) is one of the
commonest varieties and is the focus of this
chapter. Among the various clinical presentations
of DPN, severe unremitting pain is by far the
most studied by behavioral scientists. Our chap-
ter, while reflecting on this common tendency,
also highlights the importance of other clinical
presentations of DPN.  Postural instability, for
example, is one of the strongest predictors of
poor self-rated health and depression in people
with established DPN [2–4].
Numerous neurobiological similarities tie dia-
betic neuropathies, pain, depression, and anxiety
together—these threads include the neuroendo-
crine system, autonomic system, and the inflam-
L. Vileikyte (*)
Diabetes, Endocrinology and Gastroenterology,
University of Manchester, Manchester, UK
Department of Dermatology, University of Miami,
Miami, FL, USA
e-mail: lvileikyte@med.miami.edu
F. Pouwer
Department of Psychology, Steno Diabetes Center
Odense, University of Southern Denmark,
Odense, Denmark
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_13
matory cytokine system, and a host of other,
interrelated physiological pathways in brain and
body, reviewed elsewhere [5, 6]. This chapter
focuses exclusively on psychosocial factors asso-
ciated with DPN.
Depression has received the lion’s share of
attention from clinical researchers. In people
with diabetes, those with co-morbid complica-
tions are at increased risk to be depressed or
develop depression [7–9]. In two longitudinal
studies, people with both diabetes and depression
had a higher risk to develop DPN [10, 11].
Longitudinal data from nationwide general prac-
tices in Germany showed that individuals with
diabetes and DPN had higher levels of incident
depression [7].
Besides depression, diabetes distress (the neg-
ative affective or emotional experience resulting
from living with the demands of diabetes) and
anxiety are also common in diabetes [12]. In fact,
elevated anxiety symptoms were found to be
more prevalent than depressive symptoms, with
32% of people with diabetes reporting elevated
symptoms of anxiety and 22% for elevated symp-
toms of depression, using the Hospital Anxiety
and Depression Scale [13]. The observed diabe-
tes and anxiety relationship is consistent with
accumulating research linking chronic pain and
anxiety; this pattern mirrors but exceeds the pat-
tern examining depression [14]. These observa-
tions might be of potential importance to
managing painful DPN, as commonly used medi-
221
222
cations are potent mood modulators with distinct
class effects on anxiety and depression.
Accordingly, we discuss studies that examined
the differing contributors to anxiety and depres-
sive symptoms as well as different predictors of
analgesic response to duloxetine and pregabalin
in persons with painful DPN [15, 16]. Again, we
emphasize the importance of distinguishing gen-
eralized distress (depression/anxiety) from dia-
betes/DPN-specific distress, a psychological
concept although related to, but not interchange-
able with, depression or anxiety [12, 17]. A grow-
ing number of researchers studying DPN favor
disease-specific over psychiatric definitions of
emotional distress. Using a diabetic neuropathy
and depression model we illustrate how DPN-­
specific distress mediates the association between
DPN severity and depression [2, 3]. Another
emerging line of inquiry examines DPN-specific
fears such as fear of pain increase and/or fear of
falling and their associations with greater disabil-
ity and reduced quality of life (QoL) in persons
with DPN [18, 19]. Pertaining to this, by compar-
ing and contrasting the performance of generic
and DPN-specific QoL instruments, we make a
case for the need to consider DPN-specific mea-
sures when identifying and managing the psy-
chosocial problems surrounding this devastating
complication of diabetes, both in clinical care
and in clinical research projects. Finally, we pro-
vide a critical overview of the paucity of psycho-
social interventions conducted in this area and
emphasize the need for an integrated, bio-­
psychosocial approach to the management of
DPN. This chapter concludes by discussing psy-
chological treatments and interventions that
could potentially assist in the management of
psychosocial hardships experienced by people
with DPN.
2 The Impact of DPN
on Physical and Mental
Functioning, and QoL
Although DPN is a multifaceted complication of
diabetes, studies examining the impact of DPN
on individuals’ physical and mental functioning
L. Vileikyte and F. Pouwer
have traditionally focused on painful DPN. One
of the earliest, yet probably most comprehensive
assessments to-date of neuropathic pain burden,
was conducted by Gore et al. [20]. This commu-
nity based, cross-sectional self-report survey
described pain severity, as measured by the
Modified Brief Pain Inventory-DPN [21], its
interference with function, as well as prescription
medication use and treatment satisfaction among
patients with painful DPN.  The burden of pain
was reported to be considerable, resulting in a
persistent discomfort despite polypharmacy and
high resource use, and leading to limitations in
daily activities and poor satisfaction with treat-
ments that were often deemed to be inappropri-
ate. Individuals with painful DPN reported
significant impairment in physical and mental
functioning as compared to the general US popu-
lation, although the differences in mental func-
tioning were less pronounced than those observed
in physical functioning. A qualitative study of
persons affected by painful DPN [22] identified
four major areas of impact: (1) physical function,
i.e. walking, exercise, energy, standing, balance,
bending, and mobility; (2) daily life, i.e. produc-
tivity, recreational activities, work, enjoyment,
focus, and chores; (3) social/psychological, i.e.
anxiety, friends/family, irritability, depression,
and fear; and (4) sleep, i.e. falling asleep, waking
in the night and not feeling rested upon
awakening.
In a study by Zelman et  al. [23], individuals
with painful DPN reported impaired sleep rela-
tive to the US population norms and the Medical
Outcomes Study (MOS) chronic patient sample
on both, the sleep problem index and all six sleep
attribute scales. Hierarchical regression indicated
that older age, higher average daily pain, and
higher levels of anxiety and depression were each
significantly associated with, and collectively
accounted for, 47% of variance in the MOS sleep
problem index. More recently, Naranjo et al. [24]
in a cross-sectional study compared sleep charac-
teristics in type 2 diabetes patients with and with-
out painful DPN and investigated the relationship
of sensory phenotypes, anxiety, and depression
with sleep quality. Persons with painful DPN
reported significantly more sleep disturbances
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
than the control subjects in all dimensions of the
MOS sleep scale. Higher scores in anxiety or
depression, greater intensity of pain or a higher
score in the paroxysmal pain phenotype were
associated with lower sleep quality in individuals
with painful DPN. As shown by Tolle et al. [25],
inadequate pain control and severe sleep depriva-
tion resulted in over 60% of persons with painful
DPN who worked either part or full-time report-
ing either missing work or decreased work pro-
ductivity. Worse yet, in this report, pain intensity
disrupted substantially employment status, with
nearly 20% of persons with painful DPN report-
ing job loss or early retirement. A subsequent
prospective study [26] confirming and expanding
upon the literature, showed a significantly worse
trajectory of outcomes over time and long-term
increased total costs for individuals with painful
DPN relative to those without painful DPN and
control subjects. The painful DPN group not only
reported significantly lower levels of physical
functioning, but their physical functioning scores
decreased at a significantly faster rate over a
3-year period relative to other groups. In addi-
tion, those with painful DPN reported signifi-
cantly lower work productivity and activity,
greater resource use, and higher total 3-year per-­
person costs. Resource use and lost productivity
due to health were monetized separately for each
year and then totaled for the 3-year period.
Although the costs were generally comparable
between the diabetes without painful DPN and
control groups, those with painful DPN were
estimated to have significantly higher direct and
total costs than the other groups in each year of
the study. The main contributors to increased cost
in the painful DPN group were the increased
number of physician visits and hospitalizations.
The impact of DPN on individuals’ physical
and mental function originally reported by US
researchers has recently received much attention
in other countries. A study of hospital outpatient
clinics across Belgium [27] showed that both
painless and painful DPN are common, espe-
cially in people with type 2 diabetes. Despite its
profound negative impact on QoL, painful DPN
remains undertreated with only half of individu-
als receiving analgesic treatment, and only 28%
223
of them using anticonvulsants or antidepressants.
Similarly, a French study confirmed the notion
that persons with painful DPN have poorer QoL
and more sleep disturbances, anxiety, and depres-
sion than those without pain, with only 39% of
the persons receiving appropriate treatment for
their painful DPN [28]. Another study conducted
in Wales indicated that the severity of painful
DPN symptoms predicts poor health-related util-
ity and decreased QoL [29]. Happich et al. [30]
described the health-related QoL, using generic
and disease-specific questionnaires, the resource
utilization, and annual costs associated with DPN
in Germany. In this retrospective, observational
study of persons with type 1 and type 2 diabetes
and DPN, the majority of them were severely
impaired with regard to general physical health.
Disease-specific QoL decreased continuously
with increasing DPN severity. In accordance,
costs associated with DPN increased with the
progression of DPN, with costs from the societal
perspective increasing about 50-fold from the
lowest severity stage (those with sensory-motor
neuropathy without symptoms) to persons with
lower extremity amputation. Canadian research-
ers [31] have demonstrated that in persons with
longstanding type 1 diabetes, compared to other
complications, DPN has the greatest association
with both diabetes distress and depression inde-
pendent of clinical and lifestyle confounding
variables. A large Danish study estimated the
prevalence of painless and painful DPN, impor-
tant risk factors, and the association with mental
health in persons with recently diagnosed type 2
diabetes [32]. The group demonstrated that pos-
sible painless and painful DPN were indepen-
dently and additively associated with lower QoL,
poorer sleep, and elevated symptoms of depres-
sion and anxiety.
These findings are important given that health-
care costs are significantly higher for DPN
patients with depression or anxiety relative to
those without such comorbid disorders [33].
Several recent systematic reviews [34, 35] of psy-
chosocial factors related to painful DPN have
also reaffirmed the results of individual studies
indicating that depression, anxiety, sleep quality,
and quality of life are the most studied variables
224
in relation to pain outcomes and are consistently
associated with pain intensity.
3 The Effects of Pain
Medication on Psychosocial
Outcomes: Sleep and Mental
Health
Painful DPN is difficult to treat, with treatment
regimens often inadequate for pain control and
limited by side effects and drug intolerance.
Secondary parameters, such as quality of sleep
and mood, may also be important for successful
painful DPN management. A randomized,
placebo-­ controlled trial study by Boyle et  al.
compared the analgesic efficacy of pregabalin,
amitriptyline, and duloxetine, and their effect on
sleep, daytime functioning, and quality of life in
patients with painful DPN [36]. All medications
reduced pain when compared with placebo, with
pain ratings showing 50% improvement, in line
with previous studies. However, no treatment was
superior to any other. Daytime performance mea-
sures showed no evidence of cognitive impair-
ment during treatment. For sleep, pregabalin
improved sleep continuity, whereas duloxetine
increased wake and reduced total sleep time. The
beneficial effects of pregabalin on pain-related
sleep interference were documented in a review
of clinical trials [37] suggesting that, in addition
to its analgesic properties, pregabalin may affect
pain perception indirectly, that is, through
improving sleep quality. Indeed, a post-hoc anal-
ysis of data pooled from placebo-controlled trials
of pregabalin in persons with painful DPN using
path modeling showed a high degree of associa-
tion between improvements in sleep and pain
relief. Overall, these data suggest that the pres-
ence of comorbid sleep disturbance in persons
with painful DPN might, in part, predict substan-
tial pain relief in response to pregabalin treatment
[38]. The sleep fragmentation with duloxetine, as
observed in the study by Boyle et  al. [36], is
somewhat concerning. It is widely believed that
poor sleep quality may worsen pain, and although
duloxetine has good analgesic efficacy, its effec-
tiveness may be limited by this physiological
L. Vileikyte and F. Pouwer
effect. Therefore, while there was no significant
difference in analgesic efficacy between amitrip-
tyline, duloxetine, and pregabalin, there were sig-
nificant differences in the secondary parameters,
which are of relevance when deciding the opti-
mal treatment for painful DPN.  Somewhat sur-
prisingly, this study [36] found no significant
improvements in mental health as assessed by the
Short Form (SF)-36 which is unexpected given
that all three treatments (pregabalin for general-
ized anxiety disorder and duloxetine and amitrip-
tyline for depression) are indicated for affective
disorders. Moreover, as evident from the studies
reviewed in this chapter, painful DPN is associ-
ated both with depression and anxiety. Boyle and
colleagues therefore proposed that the (SF)-36
measure is not sensitive enough to assess changes
in mood and that more specific measures might
be more appropriate to detect subtle changes in
mood state over time.
4 The Role of Psychosocial
Factors in Pain Experience
and Response to Treatment
While the impact of painful DPN on individuals’
psychosocial functioning is well researched, little
is known about the role of psychosocial factors in
shaping pain experience and persons’ response to
treatment. A study by Otto et al. [39] is one of the
earliest reports that examined the impact of indi-
viduals’ mental and physical functioning, as
measured by the SF-36, on their response to
treatment for painful DPN. The researchers dem-
onstrated that individuals’ physical, social and
emotional status predicts their response to treat-
ment after controlling for demographic variables,
pain duration, and intensity. The authors there-
fore suggested that psychosocial factors might
provide a useful guide in selecting the popula-
tions most responsive to treatment for painful
DPN. Marchettini et al. [16] using data from the
COMBO-DPN study, a multinational clinical
trial in painful DPN, investigated whether there
are different predictors of analgesic response
between duloxetine and pregabalin in painful
DPN.  There were no significant interactions
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
between treatment for subgroups by demo-
graphic, baseline pain characteristics, glycemic
control or presence of comorbidities, and con-
comitant medication use. The only significant
interaction with treatment was observed in the
mood symptom subgroups with a larger pain
reduction in duloxetine-treated persons having
no mood symptoms. Selvarajah et al. [15] exam-
ined the contribution of demographic, social,
clinical, and psychological factors (Pain
Acceptance and Pain Catastrophizing) to anxiety
and depressive symptoms in persons with painful
DPN. This study found a high prevalence of emo-
tional distress with over 50% of individuals expe-
riencing symptoms of anxiety/depression.
Multiple regression analysis showed that cata-
strophic thinking is an independent contributor to
greater symptoms of anxiety and depression.
Being young, single, and unemployed signifi-
cantly contributed to greater anxiety symptoms
while pain-related restrictions of quality of life
were associated with greater depression symp-
tom scores. The report therefore highlighted the
differing independent contributors to anxiety and
depressive symptoms which are largely based on
an individual’s circumstances and experience.
Another study investigated the relationship of
pain catastrophizing with disability and QoL in
persons with painful DPN [40]. Additionally, the
mediating roles of physical activity and/or
decline in physical activity were explored. The
findings emphasize the role of catastrophic think-
ing about pain that was associated with an
increased disability and decreased QoL as well as
with a perceived decline in physical activity,
which in turn, mediated the association between
catastrophizing and disability and QoL.  Kioskli
et  al. [41] investigated whether psychological
flexibility is potentially beneficial for people with
painful DPN. Psychological flexibility is a model
of well-being and performance that includes sev-
eral related processes: acceptance, committed
action, cognitive defusion or ability to separate
one’s thoughts from events as they are directly
experienced and self-as-context that is not based
upon self-evaluations and is separate from one’s
thoughts and feelings. In regression analyses, the
four variables representing psychological flexi-
225
bility accounted for significant variance in pain-
ful DPN experience. However, the variance
accounted for by psychological variables was
less than that accounted for by pain intensity.
These results suggest that psychological flexibil-
ity may play a smaller role, relative to pain inten-
sity, in the context of painful DPN as compared
to the larger populations of chronic, mostly mus-
culoskeletal, pain. A study by Lewko et al. [42]
showed that poorer QoL in people with DPN is
related to their lower levels of illness acceptance.
Among psychosocial factors negatively affecting
illness acceptance were feelings of being a bur-
den to their family and friends and the belief that
people in their company are made anxious by the
person’s illness.
Vileikyte and Gonzalez [43] examined the
role of psychological variables (generalized dis-
tress, personality traits: extroversion, agreeable-
ness, openness, conscientiousness, and
neuroticism, and DPN-specific cognitions) and
sleep deprivation in generating and sustaining
painful DPN. Both cross-sectional and longitudi-
nal models were largely consistent, with depres-
sive and, in particular, anxiety symptoms
independently associated with more severe pain
over time. Sleep impairment and pain unpredict-
ability cognitions were also predictive of
increases in pain. The personality trait “neuroti-
cism” was the only personality characteristic
with significant cross-sectional and longitudinal
relationships to painful DPN, although its effects
were confounded with anxious and depressive
symptoms. The results also indicated the substan-
tial chronicity of pain, with baseline pain severity
explaining 62% of the variance in painful symp-
toms at 18 months. Change-based analyses, con-
trolling for baseline pain severity, showed that
younger age and higher levels of anxiety were the
only significant predictors of increments in pain
over time. These data suggest that although DPN
pain experience is commonly accompanied by
depression, sleep impairment, and negative cog-
nitive appraisals, these factors are unlikely to be
causal contributors or symptom amplifiers to
pain. The most important predictor of pain over
time is baseline pain severity, although anxiety
may also play a causal role. These findings once
226
again emphasize the intimate linkage between
painful DPN and anxiety. The observed painful
DPN and anxiety relationship is consistent with
accumulating research linking chronic pain and
anxiety; this pattern mirrors but exceeds the pat-
tern for research examining depression and
chronic pain [14]. More research is still needed,
particularly that which uses prospective methods,
to delineate the temporal primacy and the mecha-
nisms accounting for the enduring association
between painful DPN and anxiety.
5 Pathways Linking DPN
and Its Symptoms
to Depression and Anxiety
Several earlier reports by Vileikyte et  al. exam-
ined both cross-sectionally [2] and prospectively
[3] the associations between the objective tests of
DPN severity and depression symptoms and
explored the potential physical and psychosocial
mediators of this association. The selection of
psychosocial factors was based on existing
hypotheses linking physical illness to depression.
For example, Williamson [44] postulated that
illness-­related functional disability and restric-
tions in activities of daily living (ADL) can
directly contribute to the development of depres-
sion. Heidrich et  al. [45] argue that the role of
ADL restrictions in generating depression may
depend on the extent to which being unable to
perform daily activities has a negative impact on
sense of self. Furthermore, it has been proposed
that shared negative cognitive schemata underly-
ing physical illness and depression may create
the possibility of a psychological path linking
physical illness to depressive symptoms: both are
perceived as chronic, uncontrollable and poten-
tially having serious consequences [46]. The
cross-sectional results of the study by Vileikyte
et al. [2] demonstrated that DPN severity is sig-
nificantly associated with higher levels of depres-
sion symptoms, with DPN symptoms and
DPN-specific psychosocial factors accounting
for nearly half of the variance in depression
scores. Neuropathic pain, DPN-postural instabil-
ity and reduced feeling in the feet were each
L. Vileikyte and F. Pouwer
independently associated with higher depression
symptoms and together accounted for the rela-
tionship between the clinical measures of DPN
severity and depression scores. The association
between DPN symptoms and depression was par-
tially mediated by two sets of DPN-specific psy-
chosocial factors: (1) restrictions in ADL and
diminished self-worth (self-perception as a fam-
ily burden) and (2) illness cognition, or the per-
ceptions of neuropathic pain unpredictability and
the lack of treatment control. The longitudinal
findings of Vileikyte et al. [3] were largely con-
sistent with their cross-sectional observations
and demonstrated that more severe DPN at base-
line is associated with worsening depression
symptoms over 18-months of follow-up. While
neuropathic pain contributed to depression, it
was particularly DPN-postural instability and its
psychosocial consequences that dominated this
relationship over time. DPN-postural instability
continued to be a significant predictor of depres-
sion symptoms with the hypothesized psychoso-
cial mediators in the equation, thereby suggesting
that it is related to depression through mecha-
nisms that are not fully accounted for by these
studies. It has been proposed, for example, that
perceived postural instability represents not only
balance dysfunction but also other domains of
vulnerability, such as decreased physiological
reserve and psychosocial factors including poor
self-efficacy with walking and fear of falling
[47]. These are established risk factors for func-
tional decline [48] and might have provided addi-
tional linkages of DPN-postural instability to
depression symptoms.
The emergence of DPN-postural instability as
the symptom that is most strongly associated
with depression merits attention from clinicians,
especially as the preliminary evidence presented
in a systematic review [49] suggests that people
with type 2 diabetes and DPN can improve their
balance and walking following exercise training
interventions, such as lower limb strengthening,
balance practice, aerobic exercise, walking pro-
grams, and Tai Chi. We demonstrated that DPN-­
postural instability is common, with nearly a
quarter of DPN patients reporting balance prob-
lems as present either most or all the time [50].
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
Unfortunately, balance deficits may be over-
looked by clinicians, as patients often do not
report balance concerns during medical consulta-
tions owing to the perception that these are an
indicator of diminishing self-resources-a sign of
premature aging rather than illness-related
­disability [50]. Importantly, these patients have a
~20-fold greater risk of falling compared to aged-­
matched controls [51]. The consequences of fall-
ing, however, are not only physical but also
psychological, such as poor self-efficacy with
walking and fear of falling [47]. This highlights
the negative spiral between DPN, postural insta-
bility, falls, and psychological distress. Although
it is acknowledged that psychological factors
such as fear of falling can powerfully impact
upon individuals’ confidence with walking,
resulting in decreased participation in activities
of daily living, elevated depression symptoms
and diminished overall QoL, no psychological
interventions to date have been performed in peo-
ple with DPN-postural instability. Nonetheless,
some positive effects on reducing falls by means
of psychological interventions have been
achieved in high-risk older adults. One multifac-
torial intervention that did significantly reduce
the risk of falling in at-risk elderly patients uti-
lized a cognitive behavioral skills learning
approach [52]. The intervention program aimed
to improve self-efficacy for preventing falls,
encourage behavior change, and provide educa-
tion about risk management. Over 14 months of
follow-up, intervention group participants expe-
rienced a clinically meaningful 31% reduction in
falls. Furthermore, those in the intervention
group were more confident in their ability to
avoid a fall during a variety of functional daily
living tasks and used more protective behavioral
practices than control participants. Similar meth-
odologies adapted for patients with DPN could
be effective and deserve empirical evaluation.
Considering this discussion, clinicians should
therefore address postural instability as a key
symptom when assessing persons with DPN,
especially as his/her perception of their postural
instability appears to be an adequate indicator of
the actual balance impairment [53]. Pending a
definitive intervention study, clinicians should
227
consider a multifaceted approach to the manage-
ment of individuals with DPN-postural instabil-
ity, including a biomechanical component
(physical adaptation to postural instability to
improve safety) and a psychosocial component
(strategies aimed at increasing participation in
daily activities and enhanced perceptions of com-
petence in important social roles). The impor-
tance of postural instability as a symptom of
DPN should receive attention in the education of
both persons with DPN and their healthcare
providers.
6 Generalized Anxiety Vs.
Specific Fears
Persons with chronic diseases commonly report
fears of illness or symptoms recurring or worsen-
ing that have been addressed from several view-
points: an illness-specific perspective (e.g., fear
of hypoglycemia), a generic illness perspective
(e.g., fear of progression), and a psychiatric per-
spective (illness anxiety disorder and somatic
symptom disorder) [54]. A growing number of
researchers, including those studying DPN, favor
disease-specific over psychiatric definitions of
illness anxiety. Fear of amputation, for example,
is emerging as a predominant emotion in people
with diabetic foot ulceration and is a powerful
determinant of preventive foot self-care actions
[55–57]. Similarly, people with DPN-postural
instability experience psychological conse-
quences such as poor self-efficacy with walking
and fear of falling [58]. A group of Dutch investi-
gators conducted a qualitative study that identi-
fied a series of diabetes-specific fears experienced
by people with DPN suffering from pain: fear of
hypoglycemia, fear of pain increase, fear of total
exhaustion, fear of physical injury, fear of falling,
fear of loss of identity, and fear of negative evalu-
ation by others [18]. Subsequently, the same
research group subjected their qualitative obser-
vations to quantitative examination [19]. In a
cross-sectional study of persons with painful
DPN, all fears were independently associated
both with reduced QoL and with greater disabil-
ity, after adjusting for appropriate control vari-
228 L. Vileikyte and F. Pouwer

ables. Linear regression models including all whether DPN is measured by clinical tests of
fears and confounders showed that pain intensity, neurological dysfunction [60] or by patient
pain duration, and fear of falling were signifi- somatic experience with DPN [36, 61]. For
cantly associated with reduced QoL while pain example, Padua et  al. [60] showed that DPN
intensity, male gender and fear of falling were severity, as measured by clinical and neuro-­
significantly associated with greater disability. physiological tests, correlated with physical but
The predominance of fear of falling in persons not mental aspects of SF-36. The authors of this
suffering from painful DPN is an interesting find- paper therefore concluded that DPN is “strictly
ing that requires further investigation. It is plau-
related to the physical aspects of the patients’
sible that people with painful DPN are also quality of life, and not the mental aspects”.
experiencing postural instability related to their Similarly, a prospective investigation reported a
diabetic neuropathy, as it is well-established thattemporal link between the development of pain-
DPN-postural instability is associated with fear ful DPN, foot ulceration and amputations and a
of falling and reduced physical activity [58]. It subsequent decline in SF-36 physical but not
could also be possible that medications used to mental function [61]. More recently, in a large
alleviate pain are negatively impacting the peo- cross-sectional study examining the associations
ples’ balance through their recognized effects on between microvascular and macrovascular com-
the central nervous system [59]. Nonetheless, it isplications of diabetes and health status, DPN was
an interesting report confirming that persons with associated with the greatest reduction in physical
painful DPN suffer from a variety of fears that but not mental component score as measured by
could be tackled by a carefully designed psycho-­ SF-36 [62]. However, the conclusion that DPN
educational intervention, thereby improving and its clinical manifestations do not impact
physical and psychosocial well-being of these mental function is not necessarily valid, espe-
individuals. cially as a number of studies reviewed in this
chapter reported on a significant association
between DPN and elevated levels of both gener-
7 Measuring QoL in DPN: alized (anxiety/depression) and DPN-specific
Generic, DPN-Specific or emotional distress. The alternative explanation
Combined Approach? could be that the generic measures of health sta-
tus are less sensitive and thus do not capture ade-
7.1 Limitations of the Generic quately neither generalized nor DPN-specific
Health status/QoL emotional disturbance. Vileikyte et al. [50] com-
Instruments pared the performance of the SF-12 and a neu-
ropathy and foot ulcer-specific quality of life
The main reason for using a generic measure of questionnaire, the NeuroQoL, providing support
health status or QoL is that the potential impact for the latter. This study demonstrated that while
of different diseases can be compared. For more the mental functioning scale from the SF-12 was
refined assessment of the impact of a certain con- not associated with DPN severity, a DPN-­
dition, validated disease-specific assessment emotional distress scale from the NeuroQoL
tools are usually the preferred option. Cross-­ showed a strong association with DPN severity
sectional and longitudinal studies in DPN that and was the most important link between DPN
employed the generic health status/QoL instru- and diminished QoL.  The results of a compre-
ments, such as the most commonly used, SF-36, hensive review that assessed the impact of neuro-
revealed a common pattern, that is, that DPN pathic pain on QoL supported these observations.
affects health predominantly in the domain of It provided evidence that the impact of neuro-
physical functioning: the impact of DPN on men- pathic pain varies in part as a function of the QoL
tal functioning was non-significant in most of domain being considered, with the condition-­
these reports. This pattern holds regardless specific QoL measures being more sensitive to
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions 229

the effects of neuropathic pain than the generic with various peripheral neuropathies including
instruments of physical and especially mental DPN (PN-QOL-97). This instrument consists of
functioning [63]. Taken together, it would seem 97 items and encompasses several scales with
appropriate to supplement the generic tools of varying specificity of assessment, such as an
health status with validated anxiety and depres- item on overall health from the generic
sion scales when assessing the levels of EuroQoL, the entire SF-36 along with items
­generalized distress while considering condition-
measuring symptoms specific to peripheral neu-
specific questionnaires to study DPN-specific ropathies and dysfunction. The 47-item Norfolk
emotional distress. QoL-DPN, in addition to the generic health sta-
Another inherent limitation of the generic tus and general information items includes a
instruments is that their content was imposed bycomprehensive measure of DPN symptoms and
the investigators and did not emerge from inter-the specific ways these symptoms impact on
views with people affected by DPN.  Although ADL [66]. The important limitation of this
the generic measures allow the comparisons instrument is that it does not incorporate DPN-
between different disease states at the most specific emotional distress which is central to
abstract level, they lack specificity, that is, they
DPN experience. The 35-item NeuroQoL [50],
do not describe the specific clinical features of
derived from the discussions with focus groups
DPN and the ways these features impair of persons with DPN and developed in collabo-
QoL. Therefore, the findings from these studies ration neuropathy and psychology experts, is a
leave a gap between DPN as abstractly defined hierarchically organized scale that assesses an
and the person’s experience of DPN that is essen-
individual’s subjective reports of functioning
tial for framing effective interventions. and QoL in six specific domains. Following the
It is also important to note that although hierarchical model [67], the base of each domain
appraisals of mental, physical and social func- is assessed with items that measure DPN-
tioning are important in QoL decision-making, specific somatic experiences (pain, unsteadi-
they are not direct measures of QoL. Individualsness, and symptoms of reduced feeling in the
can report similar levels of dysfunction and differ
feet), DPN-specific social and personal dys-
in their subjective judgments as to the impact of
function, and emotional states. The two final
these functional impairments on their items in the scale complete the hierarchical
QoL. Therefore, QoL is what it appears to be: “aapproach by requesting that the patients appraise
subjective judgment of quality of one’s life andthe impact of DPN on their QoL and provide an
overall QoL evaluation. The psychometric anal-
not a measure of physical or cognitive function, a
report of emotional state, or a measure of a yses that compared the performance of the
patient’s integration into a social network” [64].
NeuroQoL to the SF-12 demonstrated that the
Such a definition of QoL necessitates therefore NeuroQoL was more strongly associated with
the inclusion of items directly asking respondents
DPN severity (criterion validity), more fully
to subjectively evaluate their QoL. mediated the relationship of DPN with QoL
(construct validity) and significantly increased
explained variance in QoL judgment over the
7.2 The Shift from Generic SF-12 (incremental validity) [50].
to DPN-Specific QoL Following the development of DPN-specific
Assessments QoL instruments, a number of systematic
reviews and DPN expert guidelines subjected
In an attempt to overcome the limitations of the these measures to scientific scrutiny. Two sys-
generic instruments, three DPN-specific QoL tematic reviews of patient reported outcomes
measures have been developed. The health-­ (PROs) concluded that although the DPN-
related QoL measure by Vickrey et al. [65] was specific tools were better than generic in quanti-
developed for the assessment of QoL in persons fying temporal changes in QoL and showed
230 L. Vileikyte and F. Pouwer

greater sensitivity to neuropathy severity, no one position statement regarding psychosocial care
measure was identified as a “gold standard” for for people with diabetes by the American
assessing QoL in DPN.  The authors therefore Diabetes Association (ADA) therefore recom-
recommended that the instrument selection mended that in general: “…psychosocial care
should depend on the clinical and social context should be integrated with collaborative, patient-­
of the assessment [68, 69]. An international centered medical care and provided to all people
workshop of neurology experts on the selection with diabetes, with the goals of optimizing health
of outcome measures for clinical trials in periph- outcomes and health-related quality of life” [72,
eral neuropathy recommended the use of 73]. A key element of collaborative, patient-­
PN-QOL-97 and the NeuroQoL [70], while a centered diabetes care is that levels of diabetes
recent Position Statement by the American distress, depression, anxiety, and disordered eat-
Diabetes Association covering diabetic neuropa- ing are monitored and discussed. This should be
thies proposed the Norfolk QoL-DPN and the done using patient-appropriate validated tools “at
NeuroQoL for QoL assessment in DPN [1]. the initial visit, at periodic intervals, and when
Examination by neurology experts of the content there is a change in disease, treatment, or life cir-
validity of symptom-based measures for dia- cumstance” [72]. In collaborative or stepped
betic, chemotherapy, and HIV peripheral neu- care, treatment for depression is intensified if the
ropathies concluded that given significant symptoms still persist [74].
overlap in symptoms between neuropathy etiol- In the ADA position statement [72], a specific
ogy, a measure with content validity for multiple paragraph focuses on diabetes complications and
neuropathies with supplemental disease-specific functional limitations. People with diabetes com-
modules could be of great value in the develop- plications, including neuropathy, foot ulcers or
ment of disease-modifying treatments for limb amputation not only experience depression
peripheral neuropathies [71]. Therefore, it would and anxiety, but also have lower overall physical
appear that in order to advance this complex function and impaired QoL.  This can result in
area, it would seem appropriate that the develop- reduced autonomy and serious role impairments
ment and/or refinement of QoL instruments and [72, 75]. It is therefore recommended that in dia-
their critical appraisals were conducted by a betes care settings the levels of chronic pain asso-
team consisting of persons with DPN, experts ciated with diabetes complications and the impact
both in DPN and in developing PRO measures. of painful DPN or discomfort on QoL are moni-
Additional research on the psychometric charac- tored. After a discussion of questionnaire scores,
teristics of existing measures is warranted, these appropriate pain management therapies should
future studies should follow rigorous guidelines be offered, including a consultation with a mental
for reporting on PROs [69]. health provider for pain self-management strate-
gies [72].
In their comment on the above ADA position
8 How to Manage statement, Snoek et al. propose that it is key to
Psychological Problems emphasize that the absence of a serious mental
of Persons with DPN health problem is not the same as emotional
in Clinical Care? well-­being or “a good quality of life” [76].
Another important remark by these authors is
As described in this chapter, psychological prob- that the use of screening tools should never
lems are common in people with diabetes, espe- become a routine procedure where health care
cially in those with diabetes complications, such providers are simply “ticking the box.” The
as neuropathy. Diabetes distress, depression, and score(s) on a questionnaire should always be
disordered eating can not only seriously impair regarded as a tool that can help to facilitate and
quality of life, but also negatively impact self-­ improve the clinical conversation. This is impor-
care behaviors and long-term outcomes [8]. A tant, as a questionnaire score that is indicative of
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
a mental health problem does not by definition
equal a need for psychiatric or psychological
care. Diabetes health care teams should be
trained to carefully explore this with an individ-
ual with diabetes. If needed, culturally accept-
able, affordable evidence-based treatment
options should be offered in response to the
identified psychosocial needs [76].
Many questionnaires can be used to support
clinical diabetes care. In the ADA position state-
ment, several questionnaires are recommended,
including the Problem Areas in Diabetes Survey
[77] or Diabetes Distress Scale [78] for diabetes
distress, Patient Health Questionnaire (PHQ)-9
or Beck Depression Inventory (BDI-II) for
depression [79]. The conversation after the
assessment can lead to further evaluation or a
structured psychiatric diagnostic interview. For
the assessment of chronic pain, it is recom-
mended to use the Short-form McGill Pain
Questionnaire (SF-MPQ-2) [80]. This tool can
also be used to determine in clinical care whether
the treatment for painful DPN is effective. It is
also important to emphasize that pain is by defi-
nition a subjective, personal experience and that
emotional difficulties such as loneliness, stress or
depression can definitely negatively impact the
way people experience and act as symptom
amplifiers of pain [81].
9 Psychological Treatment
Options for People
with Painful DPN
Different psychological interventions have
been designed to aid people to improve the way
they cope with chronic pain by changing: (1)
the way they perceive or appraise pain and (2)
the way they behave, in order to diminish both
distress and disability. A systematic review of
75 studies focused on the effectiveness of psy-
chological treatments in a number of chronic
pain conditions, such as chronic low back pain,
rheumatoid arthritis, fibromyalgia, and a mix-
ture of persistent pain conditions [82]. The
most commonly investigated psychological
intervention was cognitive behavioral therapy
231
(CBT, 59 studies). The systematic review fur-
thermore has included eight studies that tested
behavioral therapy (BT), and five studies that
investigated Acceptance and Commitment
Therapy (ACT). The results of the systematic
review suggest that there is moderate-­quality
evidence showing that individuals treated with
CBT experience less pain and distress by the
end of the treatment and 6–12  months later,
compared to those who had received no treat-
ment for their pain, though effect sizes were
relatively small. Those in the intervention
groups also reported less disability on average
(low-quality evidence). Furthermore, moder-
ate-quality evidence showed that CBT reduced
pain, disability, and distress in comparison with
individuals receiving a non-psychological
intervention for their pain, such as an exercise
program, or education about how to manage
pain, but here too, the effect sizes were rela-
tively small. At 6–12  months follow-up, pain
and distress were still significantly lower (mod-
erate-quality evidence), while disability levels
did not differ between CBT and non-psycho-
logical treatment (low-quality evidence) [82].
Another systematic review with meta-analy-
sis specifically focused on the effectiveness of
different psychological interventions on pain
and related outcomes in adults with diabetic
peripheral neuropathy [83]. In this review, nine
trials were included, showing that in the short-
term, psychological therapies had a large posi-
tive effect on pain severity, and a small effect
on pain interference. In adults with DPN, psy-
chological interventions were concluded to
have a moderate effect on depressive symp-
toms. When the effects on medium-term fol-
low-up assessments were summarized, a large
effect on pain severity and on pain interference
was observed, while the effect on depressive
symptoms was moderate. Furthermore, for
long-term follow-up assessments, improve-
ments in pain interference, mood, and self-care
behaviors were reported [83].
Three recent randomized controlled trials
(RCTs) were not included in the above system-
atic review. The first small RCT from the USA
(n  =  47) by Higgins et  al. compared CBT with
232
diabetes education in people with diabetic neuro-
pathic pain [84]. The findings of the study
rejected the primary hypothesis that those receiv-
ing CBT for painful DPN would experience a
larger reduction in pain intensity than those
receiving education. The study reported some
benefits of CBT relative to education on second-
ary outcomes, such as, for example, reduction in
neuropathic pain intensity in the CBT arm.
Therefore, CBT is currently the leading
­psychological treatment for chronic pain, with
small to medium effect sizes for CBT relative to
controls.
One potential new candidate that could help
to improve clinical care for people with painful
DPN is Acceptance and Commitment Therapy
(ACT). ACT is to be considered as a third wave
of cognitive behavioral therapies. When ACT is
applied to people with chronic pain, the goal is
to increase valued action in the presence of pain
and also accomplish improvements to function-
ing. The goal is to increase “psychological flex-
ibility” through six key processes: (1)
acceptance, (2) values-based action, (3) contact
with the present moment, (4) cognitive defu-
sion, (5) develop an observer self that can
change dependent on the context, and (6) com-
mitted action in line with important values [85].
ACT was tested in an RCT, where 50 individu-
als with painful DPN were randomized to (1)
medication alone to manage neuropathic pain
(2) medication and eight sessions of ACT [85].
The authors reported that ACT increased the
level of pain acceptance and reduced pain per-
ception. The third study was a single arm pilot
study (n = 30) from the UK [86]. Kioskli et al.
[87] were the first to investigate an online ACT
for painful DPN (https://www.act4painonline.
co.uk), in order to explore whether a larger RCT
is warranted. The initial results of this study are
promising. Participants who completed the ther-
apy reported clinically meaningful effects at
post-treatment for 100% of participants for pain
intensity and pain distress, 67% for depressive
symptoms, 58% for functional impairment, 42%
for cognitive fusion, 67% for committed action,
58% for self-as-context, and 42% for pain
acceptance.
L. Vileikyte and F. Pouwer
10 Concluding Remarks
and Directions for Future
Research
Distal symmetric polyneuropathy, especially
when accompanied by pain, has a profound and
wide-reaching impact on the lives of people, with
decrements in physical and mental functioning,
sleep quality, participation in activities of daily
living, employment, and overall quality of life.
Unlike painful DPN, postural instability is largely
underappreciated clinical manifestation of dia-
betic neuropathy. Nonetheless, DPN-postural
instability is the strongest determinant of depres-
sive symptoms in this population leading to
impaired QoL not only through biomechanical
difficulties but also through associated psycho-
logical distress. There is therefore an unmet need
for the development of multifaceted interventions
that address both psychological distress and bio-
mechanical challenges experienced by patients
with this incapacitating complication of
diabetes.
The reviewed studies have important implica-
tions for the care of people with DPN.  They
define several possible psychosocial routes link-
ing DPN to depression/anxiety and impaired
QoL thereby providing clinicians with the spe-
cific points for interventions to alleviate emo-
tional distress and improve QoL. Notably, these
reports identify several aspects of DPN-specific
emotional distress, such as diminished self-worth
and the perceptions of self as a family burden that
are experienced by persons with painful DPN and
with DPN-postural instability. Furthermore, a set
of specific fears related to DPN experience is also
prominent and includes fear of falling and fear of
worsening of pain, just to mention a few. It is
therefore important to consider not only general-
ized distress (depression/anxiety) or diabetes dis-
tress that is associated with the management of
glycemia. Importantly, persons’ emotional
responses that are specific to diabetic neuropathy
should also be assessed and monitored. Although
several DPN-specific instruments capture some
of the aspects of neuropathy-related distress [50,
56], more work is needed to further develop and
refine these scales. The area that warrants strong
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
support from researchers, funding organizations,
and companies is further standardization of PROs
that can be used in clinical care and research of
people with DPN. The psychometric characteris-
tics of existing tools should be compared and fur-
ther developed in line with rigorous guidelines
[85, 86, 88, 89]. New, more effective drugs with
fewer side effects should be developed and tested,
and in these trials, the most optimal PROs should
be used. Future research should also rigorously
test the effectiveness of newer interventions such
as ACT and online ACT or Mindfulness-Based
Cognitive Therapy [87, 90]. Offering online
treatment has several advantages, such as easy
access, avoidance of waiting lists, people with
painful DPN can participate in their own pace,
from their own home and during a time of the day
that they prefer. However, it is also known that
drop-out rates also tend to be relatively high.
Blended care, where online support is offered in
combination with a clinical intervention should
also be studied. Future research should also
investigate how we can become better at prevent-
ing the development of painful DPN, and how we
can become better at choosing the right interven-
tion for each individual. Precision medicine stud-
ies are needed to learn more about how clinicians
can quickly find the most successful or most
promising intervention for each individual. This
applies to both pharmacological and psychologi-
cal interventions.
References
1. Pop-Busui R, Boulton AJ, Feldman EL, et  al.
Diabetic neuropathy: a position statement by the
American Diabetes Association. Diabetes Care.
2017;40:136–54.
2. Vileikyte L, Leventhal H, Gonzalez JS, Peyrot M,
Rubin RR, Ulbrecht JS, Garrow A, Waterman C,
Cavanagh PR, Boulton AJ.  Diabetic peripheral neu-
ropathy and depressive symptoms: the association
revisited. Diabetes Care. 2005;28:2378–83.
3. Vileikyte L, Peyrot M, Gonzalez JS, et al. Predictors of
depressive symptoms in persons with diabetic periph-
eral neuropathy: a longitudinal study. Diabetologia.
2009;52:1265–73.
4. Hoogendoorn CJ, Gonzalez JS, Schechter CB, et al.
Correlates of self-rated health in people with diabetic
peripheral neuropathy: a longitudinal study. Diabet
233
Med. 2020;13:e14383. https://doi.org/10.1111/
dme.14383.
5. Jain R, Jain S, Raison CL, Maletic V. Painful diabetic
neuropathy is more than pain alone: examining the
role of anxiety and depression as mediators and com-
plicators. Curr Diab Rep. 2011;11:275–84.
6. Lees JG, Fivelman B, Duffy SS, Makker PG,
Perera CJ, Moalem-Taylor G.  Cytokines in neuro-
pathic pain and associated depression. Mod Trends
Pharmacopsychiatry. 2015;30:51–66.
7. Jacob L, Kostev K. Prevalence of depression in type 2
diabetes patients in German primary care practices. J
Diabetes Complicat. 2016;30:432–7.
8. Pouwer F, Schram MT, Iversen MM, Nouwen A,
Holt RIG.  How 25 years of psychosocial research
has contributed to a better understanding of the
links between depression and diabetes. Diabet Med.
2020;37(3):383–92. Epub 2020 Jan 24. PMID:
31909844. https://doi.org/10.1111/dme.14227.
9. Nouwen A, Adriaanse MC, van Dam K, et  al.
European depression in diabetes (EDID) research
consortium. Longitudinal associations between
depression and diabetes complications: a sys-
tematic review and meta-analysis. Diabet Med.
2019;36:1562–72.
10. Black SA, Markides KS, Ray LA.  Depression pre-
dicts increased incidence of adverse health outcomes
in older Mexican Americans with type 2 diabetes.
Diabetes Care. 2003;26:2822–8.
11. Lin EH, Rutter CM, Katon W, et al. Depression and
advanced complications of diabetes: a prospective
cohort study. Diabetes Care. 2010;33:264–9.
12. Skinner TC, Joensen L, Parkin T.  Twenty-five
years of diabetes distress research. Diabet Med.
2020;37:393–400.
13. Collins MM, Corcoran P, Perry IJ.  Anxiety and
depression symptoms in patients with diabetes.
Diabet Med. 2009;26:153–61.
14. Asmundson GJ, Katz J.  Understanding the co-­
occurrence of anxiety disorders and chronic pain:
state-of-the-art. Depress Anxiety. 2009;26:888–901.
15. Selvarajah D, Cash T, Sankar A, et al. The contribu-
tors of emotional distress in painful diabetic neuropa-
thy. Diab Vasc Dis Res. 2014;11:218–25.
16. Marchettini P, Wilhelm S, Petto H, et al. Are there dif-
ferent predictors of analgesic response between anti-
depressants and anticonvulsants in painful diabetic
neuropathy? Eur J Pain. 2016;20:472–82.
17. Snoek FJ, Bremmer MA, Hermanns N.  Constructs
of depression and distress in diabetes: time
for an appraisal. Lancet Diabetes Endocrinol.
2015;3:450–60.
18. Kanera IM, van Laake-Geelen CCM, Ruijgrok JM,
et al. Living with painful diabetic neuropathy: insights
from focus groups into fears and coping strategies.
Psychol Health. 2019;34:84–105.
19. Geelen CC, Smeets RJEM, Schmitz S, van den Bergh
JP, Goossens MEJB, Verbunt JA. Anxiety affects dis-
ability and quality of life in patients with painful dia-
betic neuropathy. Eur J Pain. 2017;21:1632–41.
234
20. Gore M, Brandenburg NA, Dukes E, Hoffman DL, Tai
KS, Stacey B. Pain severity in diabetic peripheral neu-
ropathy is associated with patient functioning, symp-
tom levels of anxiety and depression, and sleep. J Pain
Symptom Manag. 2005;30:374–85.
21. Zelman DC, Gore M, Dukes E, Tai KS, Brandenburg
N.  Validation of a modified version of the brief
pain inventory for painful diabetic peripheral neu-
ropathy. J Pain Symptom Manag. 2005;29(4):401–
10. PMID: 15857744. https://doi.org/10.1016/j.
jpainsymman.2004.06.018.
22. Brod M, Pohlman B, Blum SI, Ramasamy A, Carson
R. Burden of illness of diabetic peripheral neuropathic
pain: a qualitative study. Patient. 2015;8:339–48.
23. Zelman DC, Brandenburg NA, Gore M. Sleep impair-
ment in patients with painful diabetic peripheral neu-
ropathy. Clin J Pain. 2006;22:681–5.
24. Naranjo C, Dueñas M, Barrera C, Moratalla G, Failde
I. Sleep characteristics in diabetic patients depending
on the occurrence of neuropathic pain and related fac-
tors. Int J Environ Res Public Health. 2020;17:8125.
25. Tölle T, Xu X, Sadosky AB. Painful diabetic neuropa-
thy: a cross-sectional survey of health state impair-
ment and treatment patterns. J Diabetes Complicat.
2006;20:26–33.
26. daCosta DBM, Cappelleri JC, Joshi AV.  A longitu-
dinal assessment of painful diabetic peripheral neu-
ropathy on health status, productivity, and health care
utilization and cost. Pain Med. 2011;12:118–26.
27. Van Acker K, Bouhassira D, De Bacquer D,
Weiss S, Matthys K, Raemen H, Mathieu C, Colin
IM. Prevalence and impact on quality of life of periph-
eral neuropathy with or without neuropathic pain in
type 1 and type 2 diabetic patients attending hospital
outpatients clinics. Diabetes Metab. 2009;35:206–13.
28. Bouhassira D, Letanoux M, Hartemann A.  Chronic
pain with neuropathic characteristics in diabetic
patients: a French cross-sectional study. PLoS One.
2013;8:e74195.
29. Currie CJ, Poole CD, Woehl A, et al. The health-related
utility and health-related quality of life of hospital-­
treated subjects with type 1 or type 2 diabetes with
particular reference to differing severity of peripheral
neuropathy. Diabetologia. 2006;49:2272–80.
30. Happich M, John J, Stamenitis S, Clouth J, Polnau
D.  The quality of life and economic burden of
neuropathy in diabetic patients in Germany in
2002—results from the diabetic microvascular com-
plications (DIMICO) study. Diabetes Res Clin Pract.
2008;81:223–30.
31. Bai JW, Lovblom LE, Cardinez M, et al. Neuropathy
and presence of emotional distress and depression
in longstanding diabetes: results from the Canadian
study of longevity in type 1 diabetes. J Diabetes
Complicat. 2017;31:1318–24.
32. Gylfadottir SS, Christensen DH, Nicolaisen SK, et al.
Diabetic polyneuropathy and pain, prevalence, and
patient characteristics: a cross-sectional questionnaire
study of 5,514 patients with recently diagnosed type 2
diabetes. Pain. 2020;161:574–83.
L. Vileikyte and F. Pouwer
33. Boulanger L, Zhao Y, Bao Y, et  al. A retrospec-
tive study on the impact of comorbid depression or
anxiety on healthcare resource use and costs among
diabetic neuropathy patients. BMC Health Serv Res.
2009;9:111.
34. Kioskli K, Scott W, Winkley K, Kylakos S, McCracken
LM. Psychosocial factors in painful diabetic neuropa-
thy: a systematic review of treatment trials and survey
studies. Pain Med. 2019;20:1756–73.
35. Naranjo C, Del Reguero L, Moratalla G, Hercberg
M, Valenzuela M, Failde I.  Anxiety, depression and
sleep disorders in patients with diabetic neuropathic
pain: a systematic review. Expert Rev Neurother.
2019;19:1201–9.
36. Boyle J, Eriksson ME, Gribble L, et al. Randomized,
placebo-controlled comparison of amitriptyline,
duloxetine, and pregabalin in patients with chronic
diabetic peripheral neuropathic pain: impact on pain,
polysomnographic sleep, daytime functioning, and
quality of life. Diabetes Care. 2012;35:2451–8.
37. Roth T, van Seventer R, Murphy TK.  The effect of
pregabalin on pain-related sleep interference in dia-
betic peripheral neuropathy or postherpetic neuralgia:
a review of nine clinical trials. Curr Med Res Opin.
2010;26:2411–9.
38. Vinik A, Emir B, Parsons B, Cheung R.  Prediction
of pregabalin-mediated pain response by severity of
sleep disturbance in patients with painful diabetic
neuropathy and post-herpetic neuralgia. Pain Med.
2014;15:661–70.
39. Otto M, Bach FW, Jensen TS, Sindrup SH.  Health-­
related quality of life and its predictive role for anal-
gesic effect in patients with painful polyneuropathy.
Eur J Pain. 2007;11:572–8.
40. Geelen CC, Kindermans HP, van den Bergh JP,
Verbunt JA. Perceived physical activity decline as a
mediator in the relationship between pain catastro-
phizing, disability, and quality of life in patients
with painful diabetic neuropathy. Pain Pract.
2017;17:320–8.
41. Kioskli K, Winkley K, McCracken LM.  Might psy-
chological flexibility processes and acceptance and
commitment therapy (ACT) apply in adults with pain-
ful diabetic neuropathy? A cross-sectional survey. J
Contextual Behav Sci. 2019;13:66–73.
42. Lewko J, Polityńska B, Kochanowicz J, et al. Quality
of life and its relationship to the degree of illness
acceptance in patients with diabetes and peripheral
diabetic neuropathy. Adv Med Sci. 2007;52(Suppl
1):144–6.
43. Vileikyte L, Gonzalez JS.  Recognition and manage-
ment of psychosocial issues in diabetic neuropathy.
Handb Clin Neurol. 2014;126:195–209.
44. Williamson GM.  The central role of restricted nor-
mal activities in adjustment to illness and disabil-
ity: a model of depressed affect. Rehabil Psychol.
1998;43:327–47.
45. Heidrich SM, Forsthoff CA, Ward SE. Psychological
adjustment in adults with cancer: the self as mediator.
Health Psychol. 1994;13:346–53.
Psychosocial Aspects of Diabetic Neuropathy: From Description to Interventions
46. Brownlee S, Leventhal EA, Leventhal H. Regulation,
self-regulation and regulation of the self in maintain-
ing physical health. In: Boekartz M, Pintrich PR,
ZeiDPNer M, editors. Handbook of self-regulation.
San Diego, CA: Academic; 2000. p. 369–416.
47. McKee KJ, Orbell S, Austin CA, et  al. Fear of fall-
ing, falls efficacy, and health outcomes in older
people following hip fracture. Disabil Rehabil.
2002;15(24):327–33.
48. Lindenberger EC, Landefeld CS, Sands LP, et  al.
Unsteadiness reported by older hospitalized patients
predicts functional decline. J Am Geriatr Soc.
2003;51:621–6.
49. Gu Y, Dennis SM.  Are falls prevention programs
effective at reducing the risk factors for falls in people
with type-2 diabetes mellitus and peripheral neuropa-
thy: a systematic review with narrative synthesis. J
Diabetes Complicat. 2017;31:504–16.
50. Vileikyte L, Peyrot M, Bundy C, et al. The develop-
ment and validation of a neuropathy–and foot ulcer-­
specific quality of life instrument. Diabetes Care.
2003;26:2549–55.
51. Schwartz AV, Vittinghoff E, Sellmeyer DE, et  al.
Health, aging, and body composition study. Diabetes-­
related complications, glycemic control, and falls in
older adults. Diabetes Care. 2008;31:391–6.
52. Clemson L, Cumming RG, Kendig H, Swann M,
Heard R, Taylor K. The effectiveness of a community-­
based program for reducing the incidence of falls in
the elderly: a randomized trial. J Am Geriatr Soc.
2004;52:1487–94.
53. Reeves ND, Brown SJ, Petrovic M, Boulton AJM,
Vileikyte L.  How does self-perceived unsteadiness
influence balance and gait in people with diabetes?
Preliminary observations. Diabetes Care. 2017;40:e51–2.
54. Lebel S, Mutsaers B, Tomei C, et al. Health anxiety
and illness-related fears across diverse chronic ill-
nesses: a systematic review on conceptualization,
measurement, prevalence, course, and correlates.
PLoS One. 2020;15:e0234124.
55. Vileikyte L, Pouwer F, Gonzalez JS.  Psychosocial
research in the diabetic foot: are we making progress?
Diabetes Metab Res Rev. 2020;36(Suppl 1):e3257.
56. Vileikyte L, Gonzalez JS, Leventhal H, et al. Patient
interpretation of neuropathy (PIN) questionnaire: an
instrument for assessment of cognitive and emotional
factors associated with foot self-care. Diabetes Care.
2006;29:2617–24.
57. Wukich DK, Raspovic KM, Suder NC.  Patients with
diabetic foot disease fear major lower-­extremity amputa-
tion more than death. Foot Ankle Spec. 2018;11:17–21.
58. Vileikyte L, Crews RT, Reeves ND.  Psychological
and biomechanical aspects of patient adaptation to
diabetic neuropathy and foot ulceration. Curr Diab
Rep. 2017;23(17):109.
59. Tesfaye S, Vileikyte L, Rayman G, et al. Painful dia-
betic peripheral neuropathy: consensus recommen-
dations on diagnosis, assessment and management.
Diabetes Metab Res Rev. 2011;21:629–38.
235
60. Padua L, Saponara C, Ghirlanda G, et  al. Health-­
related quality of life in type 1 diabetic patients and
influence of peripheral nerve involvement. Neurol
Sci. 2001;22:239–45.
61. Ahroni JH, Boyko EJ.  Responsiveness of the SF-36
among veterans with diabetes mellitus. J Diabetes
Complicat. 2000;14:31–9.
62. Venkataraman K, Wee HL, Leow MK, et  al.
Associations between complications and health-­
related quality of life in individuals with diabetes.
Clin Endocrinol. 2013;78:865–73.
63. Jensen MP, Chodroff MJ, Dworkin RH.  The
impact of neuropathic pain on health-related qual-
ity of life: review and implications. Neurology.
2007;68:1178–82.
64. Leventhal H, Colman S.  Quality of life: a process
view. Psychol Health. 1997;12:753–67.
65. Vickrey BG, Hays RD, Beckstrand M. Development
of a health-related quality of life measure for
peripheral neuropathy. Neurorehabil Neural Repair.
2000;14:93–104.
66. Vinik EJ, Hayes RP, Oglesby A, et  al. The devel-
opment and validation of the Norfolk QOL-DPN,
a new measure of patients’ perception of diabetes
and diabetic neuropathy. Diabetes Technol Ther.
2005;7:497–508.
67. Fries JF, Sigh G.  The hierarchy of patient out-
comes. In: Spilker B, editor. Quality of life and
Pharmacoeconomics in clinical trials. 2nd ed.
Philadelphia: Lippincott-Raven; 1996. p. 11–24.
68. Hogg FR, Peach G, Price P, Thompson MM,
Hinchliffe RJ.  Measures of health-related quality
of life in diabetes-related foot disease: a systematic
review. Diabetologia. 2012;55:552–65.
69. Bredfeldt C, Altschuler A, Adams AS, Portz JD,
Bayliss EA.  Patient reported outcomes for dia-
betic peripheral neuropathy. J Diabetes Complicat.
2015;29:1112–8.
70. Merkies IS, Lauria G. 131st ENMC international
workshop: selection of outcome measures for periph-
eral neuropathy clinical trials 10-12 December 2004,
Naarden, the Netherlands. Neuromuscul Disord.
2006;16:149–56.
71. Gewandter JS, Burke L, Cavaletti G, et  al. Content
validity of symptom-based measures for diabetic, che-
motherapy, and HIV peripheral neuropathy. Muscle
Nerve. 2017;55:366–72.
72. Young-Hyman D, de Groot M, Hill-Briggs F,
Gonzalez JS, Hood K, Peyrot M.  Psychosocial care
for people with diabetes: a position statement of
the American Diabetes Association. Diabetes Care.
2016;39:2126–40.
73. American Diabetes Association. 5. Facilitating behav-
ior change and well-being to improve health out-
comes: standards of medical care in diabetes—2021.
Diabetes Care. 2021;44:S53–72.
74. Katon WJ, Lin EH, Von Korff M, et al. Collaborative
care for patients with depression and chronic ill-
nesses. N Engl J Med. 2010;363:2611–20.
236
75. Argoff CE, Cole BE, Fishbain DA, Irving GA. Diabetic
peripheral neuropathic pain: clinical and quality-of-­
life issues. Mayo Clin Proc. 2006;81(Suppl):S3–S11.
76. Snoek FJ, Hermanns N, de Wit M, et al. Psychosocial
aspects of Diabetes Study Group of the European
Association for the Study of Diabetes. Comment on
Young-Hyman et al. Diabetes Care. 2018;41:e31–2.
77. Kenny E, O'Malley R, Roche K, et al. Diabetes dis-
tress instruments in adults with type 1 diabetes: a sys-
tematic review using the COSMIN ­(COnsensus-­based
standards for the selection of health status mea-
surement INstruments) checklist. Diabet Med.
2020;24:e14468.
78. Polonsky WH, Fisher L, Earles J, et  al. Assessing
psychosocial distress in diabetes: development
of the diabetes distress scale. Diabetes Care.
2005;28(3):626–31. https://doi.org/10.2337/
diacare.28.3.626.
79. van Dijk SEM, Adriaanse MC, van der Zwaan L, et al.
Measurement properties of depression questionnaires
in patients with diabetes: a systematic review. Qual
Life Res. 2018;27:1415–30.
80. Dworkin RH, Turk DC, Revicki DA, et  al.
Development and initial validation of an expanded
and revised version of the short-form McGill pain
questionnaire (SF-MPQ-2). Pain. 2009;144:35–42.
81. Speight J, Holmes-Truscott E, Hendrieckx C,
Skovlund S, Cooke D.  Assessing the impact of dia-
betes on quality of life: what have the past 25 years
taught us? Diabet Med. 2020;37:483–92.
82. Williams ACC, Fisher E, Hearn L, Eccleston
C.  Psychological therapies for the management of
chronic pain (excluding headache) in adults. Cochrane
Database Syst Rev. 2020;12(8):CD007407.
83. Racaru S, Sturt J, Celik A. The effects of psychologi-
cal interventions on diabetic peripheral neuropathy:
a systematic review and meta-analysis. Pain Manag
Nurs. 2020;S1524-9042(20):30210–1.
L. Vileikyte and F. Pouwer
84. Higgins DM, Heapy AA, Buta E, et  al. A random-
ized controlled trial of cognitive behavioral therapy
compared with diabetes education for diabetic
peripheral neuropathic pain. J Health Psychol.
2020;19:1359105320962262.
85. Hughes LS, Clark J, Colclough JA, Dale E, McMillan
D.  Acceptance and commitment therapy (ACT) for
chronic pain: a systematic review and meta-analyses.
Clin J Pain. 2017;33:552–68.
86. Taheri AA, Foroughi AA, Mohammadian Y, et al. The
effectiveness of acceptance and commitment therapy
on pain acceptance and pain perception in patients
with painful diabetic neuropathy: a randomized con-
trolled trial. Diabetes Ther. 2020;11:1695–708.
87. Kioskli K, Scott W, Winkley K, Godfrey E,
McCracken LM. Online acceptance and commitment
therapy for people with painful diabetic neuropathy
in the United Kingdom: a single-arm feasibility trial.
Pain Med. 2020;21:2777–88.
88. U.S Department of Health and Human Services Food
and Drug Administration (FDA). Guidance for indus-
try: patient reported outcome measures–use in medic-
inal product development to support labelling claims.
2009. https://www.fda.gov/media/77832/download.
Accessed 1 Dec 2018.
89. Frost MH, Reeve BB, Liepa AM, Stauffer JW, Hays
RD. Mayo/FDA patient-reported outcomes Consensus
meeting group; what is sufficient evidence for the reli-
ability and validity of patient-reported outcome mea-
sures? Value Health. 2007;10(Suppl 2):S94–S105.
90. van Laake-Geelen CCM, Smeets RJEM, Quadflieg
SPAB, Kleijnen J, Verbunt JA. The effect of exercise
therapy combined with psychological therapy on
physical activity and quality of life in patients with
painful diabetic neuropathy: a systematic review.
Scand J Pain. 2019;19:433–9.
 Part II
Pathophysiology
 The Genomics of Diabetic
Neuropathy
Abirami Veluchamy, Blair H. Smith,
and David L. Bennett
1 Introduction
For the vast majority of human diseases individ-
ual susceptibility is to some degree influenced by
genetic variation. Understanding the genetics of
common disorders such as diabetes mellitus,
heart disease and auto-immune disorders has pro-
vided novel insights into disease pathogenesis,
identified new drug targets and, through combin-
ing multiple genetic risk variants into polygenic
risk scores, enabling risk stratification [1]. In this
chapter we will discuss how genomics is inform-
ing our knowledge of diabetic neuropathy and
also key consequences of diabetic neuropathy
such as neuropathic pain. Compared to our
understanding of the genomics of diabetes per se
this field (and indeed the genomics of other dia-
betes complications such as retinopathy and
nephropathy) is still very much in its infancy [2].
The most important reason for this is the fact that
large cohorts with detailed harmonised pheno-
typing of neuropathy and neuropathic pain have
not yet been fully developed. As will be apparent
from other chapters in this book, the phenotype
A. Veluchamy · B. H. Smith
Division of Population Health and Genomics, School
of Medicine, University of Dundee,
Dundee, Scotland, UK
D. L. Bennett (*)
The Nuffield Department of Clinical Neuroscience,
University of Oxford, Oxford, UK
e-mail: david.bennett@ndcn.ox.ac.uk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_14
of neuropathy is ideally captured by a combina-
tion of relevant patient reported outcomes, exam-
ination findings and investigations such as nerve
conduction studies with harmonised definitions.
Given the subjective nature of pain, neuropathic
pain is arguably an even harder phenotype to cap-
ture, and important information such as the dura-
tion, quality and localisation of pain and whether
this is likely attributable to diabetic neuropathy
(e.g., bilateral pain in the feet) is required [3, 4].
These issues are now beginning to be appreciated
in large population cohorts such as UK-Biobank,
and such cohorts, allied to advances in sequenc-
ing technology/analytics and the development of
human cellular models, should significantly
advance our understanding of the genomics of
diabetic neuropathy over the next decade. We
will briefly highlight the broad approaches used
in human genomics and how they have been
applied to diabetes before focusing on the current
state of knowledge in diabetic neuropathy.
2 A Brief Introduction
to Diabetes Genetics
Partly for historical reasons there has been a
dichotomy of approach in human genetics into,
firstly, the study of rare genetic disorders inher-
ited in a monogenic (Mendelian) fashion, and
secondly, common disorders which are polygenic
with multiple genes each having a small impact
239
240
on disease risk [1]. In Mendelian disorders the
disease allele is absent or extremely rare in the
general population and has a large impact on
gene function, for instance, through changes in
the amino acid sequence or markedly reduced
gene expression. A relevant example is neonatal
diabetes due to gain of function mutations in sub-
units of the ATP-sensitive potassium channel, a
critical determinant of excitability in pancreatic
islet cells [5]. These mutations lead to reduced
ATP sensitivity, reduced cellular excitability and
impaired insulin secretion. This finding provided
a molecular rationale for the use of Sulfonylurea
drugs in this disorder, and is an excellent example
of how gene identification can enlighten under-
standing of disease pathophysiology and thera-
peutics [6].
The second broad approach to human genetics
is based on the fact that many common disorders
are related to risk determined by multiple genes
(also interacting with environmental factors), and
this can be revealed by investigating the associa-
tion of the allele frequency of single nucleotide
polymorphisms (SNPs) in individuals with dis-
ease phenotype. Such associations can be studied
using a candidate gene approach or preferably
across the whole genome using a Genome-Wide
Association Study (GWAS) approach. Careful
attention to bias, appropriate significance thresh-
olds and large sample sizes have generated
robust, reproducible findings across a range of
common diseases. In this context the SNPs iden-
tified each have a small effect size and in most
cases are non-coding, and their effect is thought
to be mediated through subtle effects on gene
expression through altering the function of gene
promoters and enhancers. Determining whether
and how a variant can impact on gene expression
is now greatly facilitated by large datasets being
generated by consortia such as ENCODE, which
define the elements in DNA and histones that
regulate gene expression in multiple cell types
[7]. The GWAS approach requires large sample
sizes to generate sufficient statistical power.
Because diabetes is common and has a large
impact on health, diabetes has been at the fore-
front of advances in genomics. More than 50
genetic loci have been found to be significantly
A. Veluchamy et al.
associated with type 1 diabetes and more than
200 with type 2 diabetes [8]. These are emphasis-
ing important biological pathways such as pan-
creatic islet cell dysfunction in the case of type 2
diabetes [9]. These advances are likely to con-
tinue with increased application of genomics at
population level (e.g., UK Biobank [10]) and
within national healthcare systems [11]. As we
develop larger population cohorts, we are realis-
ing that, rather than a dichotomy of monogenic/
Mendelian genetics versus polygenic/common
disease genetics, there is a whole spectrum of
effect sizes and allele frequencies within the pop-
ulation; gene variants may also have pleiotropic
effects influencing multiple phenotypes/disor-
ders. We are also rapidly seeing a transitioning of
sequencing technologies (as costs drop) from
array sequencing of SNPs and whole exome
sequencing to whole genome sequencing. Just as
important as these changes in technologies are
analysis techniques and availability of very large
datasets of genetic variation in adult populations
(gnomAD [12]) and in those with clinical disor-
ders (ClinVar [13]).
3 The Genomics of Diabetic
Neuropathy in Humans
Genomic studies may help us to identify the
genetic risk factors which result in 30–50% of
patients with diabetes developing neuropathy
[14]. This will help in our understanding of the
pathogenesis of neuropathy, and any shared
genetic architecture with other diabetes compli-
cations, implying common pathophysiological
pathways.
For the past two decades, many genetic asso-
ciation studies have identified potential risk vari-
ants for diabetic neuropathy. These indicate that
genetics has a significant contribution to its
development, though there are currently few
studies with heritability estimates. Table 1 pres-
ents the known genetic variants associated with
neuropathy in people with type 2 diabetes, from
25 human genetic association studies. The dia-
betic neuropathy phenotype was defined in most
of the studies using at least one of the following:
The Genomics of Diabetic Neuropathy 241

Table 1  Known genetic variants associated with neuropathy in type 2 diabetes patients
Effect of
Study (year) Ethnicity N (cases/controls) Phenotyping Variants Gene name association
Arredondo- Mexican 218(90/128) Monofilament rs3025039 Vascular Protective
García et al. and tuning endothelial
(2019) [15] fork test growth factor
(VEGF)
Basol et al. Turkish 468(227/241) DNS and NDS VNTR Interleukin 4 Risk
(2013) [16] polymorphism (IL-4)
Buraczynska Polish 926(406/520) CNE and Pro198Leu Glutathione Risk
et al. (2017) EMG Peroxidase 1
[17] (GPx-1)
Chen et al. Chinese 160(80/80) WHO rs3821799 Adiponectin Risk
(2015) [18] diagnostic rs3774261 (ADPN)
criteria
Ciccacci et al. Italy 130(62/68) VPT and rs2910164 MIR146a Protective
(2014) [19] Neuropathic rs11888095 MIR128a Risk
questionnaire
Ciccacci et al. Italy 149(69/80) MNSI, VPT rs3746444 MIR499a Risk
(2018) [20]
Gupta and North 650(356/294) VPT and C106T Aldo-keto Risk
Singh (2017) Indian Monofilament rs759853 reductase
[21] family 1
member B
(AKR1B1)
Inanir et al. Turkish 516(235/281) NSS and NDS I/D Angiotensin- Risk
(2013) [22]a polymorphism converting
enzyme (ACE)
Ji et al. (2015) Chinese 180(90/90) Clinical T45G Adiponectin Risk
[23] assessment G276T (ADPN)
Kakavand Iranian 248(141/107) UKST C667T Methylene- Risk
Hamidi et al. score > 2, tetrahydrofolate
(2018) [24] microfilament reductase
(MTHFR)
Kolla et al. Indian 400(198/202) VPT +874A/T Interleukin 10 Risk
(2009) [25]a 1083G/A (IL-10)
Interferon-
gamma
(IFN-G)
Mansoor et al. Pakistan 772(276/496) Clinical I/D Angiotensin- Risk
(2012) [26] assessment converting
enzyme (ACE)
II
Marzban et al. Iranian 106(49/57) NSS and NDS HLA- Human Risk
(2016) [27] DQB1*02 and Leukocyte
HLA- Antigen (HLA)
DRB1*07
Monastiriotis Greece 234(54/180) NDS Epsilon 4 Apolipoprotein Risk
et al. (2013) E (APOE)
[28]
Papanas et al. Greek 130(70/60) DN index I/D Alpha2B Risk
(2007) [29] adrenergic
receptor
(ADRA2B)
Ren et al. Han 787(402/385) Monofilament rs5498 Intercellular Risk
(2015) [30] Chinese test, VPT, adhesion
Clinical molecule-1
assessment (ICAM-1)
(continued)
242 A. Veluchamy et al.

Table 1 (continued)
Effect of
Study (year) Ethnicity N (cases/controls) Phenotyping Variants Gene name association
Shah et al. Asian 495(139/356) VPT T-786C Endothelial- Risk
(2013) [31] Indian 475(133/342) (rs2070744), derived nitric
288(81/207) 27 VNTR oxide synthase
Intron 4, (eNOS)
G894T
(rs1799983)
Stephens et al. British 572(173/399) Clinical I/D Angiotensin- Risk
(2006) [32] assessment converting
enzyme (ACE)
Stoian et al. Spanish 174(84/90) Clinical I/D Vascular Risk
(2014) [33]a assessment Endothelial
Growth Factor
(VEGF)
Sun et al. Chinese 281(143/138) WHO rs2248069 Calcium Risk
(2018) [34] diagnostic rs16030 Voltage-Gated
Criteria rs216008 Channel
rs2239050 Subunit Alpha1
rs3794619 A (CACNA1A)
rs7191246 Calcium
Voltage-Gated
Channel
Subunit Alpha1
C (CACNA1C)
Calcium
Voltage-Gated
Channel
Subunit Alpha1
H (CACNA1H)
Tang et al. British 773(211/558) CNE rs1050450 Glutathione Risk
(2012) [35] peroxidase-1
(GPx-1)
Tang et al. American/ 5168(4384/784) MNSI >2.0 rs13417783 Sodium Risk
(2019) [36]b Canadian Voltage-Gated
Channel Alpha
Subunit 2
(SCN2A)
Ukinc et al. Turkish 52(37/15) Neurological C677T Methylene- Risk
(2009) [37] assessment and tetrahydrofolate
symptoms reductase
(MTHFR)
Wang et al. Chinese 251(101/150) Toronto C667T Methylene- Risk
(2012) [38] Diabetic tetrahydrofolate
Neuropathy reductase
Consensus (MTHFR)
Yigit et al. Turkish 512(230/282) NSS and NDS C667T Methylene- Risk
(2013) [22] tetrahydrofolate
reductase
(MTHFR)
ACCORD action to control cardiovascular risk in diabetes, BARI 2D bypass angioplasty revascularization investigation
in type 2 diabetes, CNE clinical neurological examination, DN diabetic neuropathy, CNE clinical neurological examina-
tion, EMG electromyography, MNSI Michigan neuropathy screening instrument, MPQ McGill pain questionnaire, NDS
neuropathy disability score, NSS neuropathy symptom score, UKST the United Kingdom screening test, VPT vibration
perception threshold, WHO World Health Organisation
a
DN vs healthy controls
b
Genome-wide association studies
The Genomics of Diabetic Neuropathy 243

monofilament test, vibration perception thresh- ined in four cohorts [22, 24, 37, 38]. The MTHFR
old, Toronto Diabetic Neuropathy Expert C677T polymorphism was found to be signifi-
Consensus, the United Kingdom screening test, cantly associated with diabetic neuropathy in a
Michigan neuropathy screening instrument meta-analysis of these cohorts (OR, 1.43;
(MNSI), neuropathy symptom score (NSS) and/ P = 0.014) [42]. Although the exact role of this
or neuropathy disability score (NDS). Most of gene in diabetic neuropathy is not yet known,
the candidate gene association studies investi- in vitro studies showed that hyper homocystein-
gated the association of genetic variants with dia- aemia affects nerve function.
betic neuropathy using cohorts of type 2 diabetes A recent meta-analysis [41], the GPx-1 vari-
patients with and without neuropathy. Very few ant (rs1050450) showed a significant association
compared diabetic neuropathy cases only with with diabetic neuropathy (OR = 1.43, P < 0.001).
healthy controls. The studies were conducted GPx-1 encodes an antioxidant enzyme that pro-
predominantly in cohorts of European and Asian tects cells against oxidative damage. Two other
ancestry. studies reported the association of two polymor-
Single nucleotide polymorphisms (SNPs) in phisms (I/D and rs3025039) in VEGF with dia-
angiotensin-converting enzyme (ACE), betic neuropathy [15, 33]. VEGF modulates
methylene-­tetrahydrofolate reductase (MTHFR), vascular permeability and contributes to neuro-
vascular endothelial growth factor (VEGF) and genesis. Further investigation is needed to vali-
glutathione peroxidase-1 (GPx-1) genes were date these genetic variants. Other reported genes
found to be associated with diabetic neuropathy associated with diabetic neuropathy in just a sin-
susceptibility in multiple studies of individuals gle study are presented in Table  1. ADRA2B,
from different ethnic backgrounds [39–41] (see CACNA1C, CACNA1H and CACNA1A are
Table  1). A recent systematic review and meta-­ mainly involved in neurotransmission. HLA-­
analysis reported that ACE I  >  D and MTHFR DQB1, HLA-DRB1-DQB1, IL-4, IFN-G, ICAM-­
1298A/C variants were significantly associated 1 and eNOS represent the immune response
with increased risk of diabetic neuropathy [42]. pathway. The other genes (APOE, ADPN and
The DD genotype of ACE was shown to increase AKR1B1) are relevant to cell signalling and
the risk of developing diabetic peripheral neu- metabolism. Notably, two studies implicated spe-
ropathy in Caucasian, South Asian, Turkish and cific microRNAs (rs2910164, rs11888095 and
Egyptian cohorts, with an odds ratio of 1.43  in rs3746444) involved in the regulation of inflam-
the meta-analysis (P  =  0.004) [42]. The ACE mation [19, 20]. None of these findings have yet
gene has been the most frequently studied in dia- been replicated in large cohorts.
betic neuropathy. It encodes angiotensin-­ Some candidate genes have been found to
converting enzyme which induces oxidative be associated with neuropathy in type 1 diabe-
stress, inflammation and vascular changes. ACE tes cohorts in Russia. These include the
inhibitors were shown to be effective in treating −262 T > C polymorphism of the catalase gene
diabetic neuropathy in experimental studies [43]. (CAT) [45, 46], the Ala(−9)Val SNP of super-
A recent study examined the efficacy of ACE oxide dismutase 2 (SOD) [47] and the
inhibitors for 2  years on 63 patients with auto- Arg213Gly polymorphism of superoxide dis-
nomic and/or peripheral diabetic neuropathy and mutase 3 (SOD3) [48, 49]. All three genes
found that they improved cardiovascular auto- (CAT, SOD2 and SOD3) protect the body
nomic diabetic neuropathy but did not improve against oxidative stress. A genetic variant
peripheral neuropathy [44]. More studies are (rs1001179) in CAT was shown to decrease the
therefore needed to elucidate the role of ACE risk of developing neuropathy in type 1 diabe-
inhibitors in diabetic neuropathy. The MTHFR tes (OR, 0.68; CI, 0.53–0.86; P  <  0.05) in a
gene, which encodes an enzyme that catalyses recent meta-analysis [41]. Two studies found
the conversion of homocysteine to methionine that the Arg allele of rs1799895 (Arg213Gly)
through the re-methylation pathway, was exam- in SOD3 increased the risk of developing neu-
244
ropathy in type 1 diabetes (OR, 5.91; P = 0.01
[48] and OR, 1.64; P  =  0.02 [49]). Sample
sizes in these candidate gene association stud-
ies ranged from 100 to 1000 and the findings
are yet to be replicated.
Recently, the first GWAS of diabetic neu-
ropathy found a novel genome-wide significant
locus at chromosome 2q24 (P < 5 × 10−8) in the
Action to Control Cardiovascular Risk in
Diabetes (ACCORD) discovery cohort of 4384
white cases and 784 controls without neuropa-
thy. The T allele of the lead SNP, r13417783,
showed a protective effect with an odds ratio of
0.64 (P = 1.9 × 10−9). The closest genes to the
intergenic variant are Xin Actin Binding Repeat
Containing 2 (XIRP2) and sodium voltage-
gated channel alpha subunit 2 (SCN2A) [36].
The protective allele of this SNP influences the
expression of SCN2A, which can increase neu-
ronal excitability. It was replicated in an inde-
pendent cohort of people with type 2 diabetes
and neuropathy (791 cases and 158 controls) in
the Bypass Angioplasty Revascularization
Investigation in Type 2 Diabetes (BARI2D)
trial (OR, 0.57; P = 0.0009). It retained genome-
wide significance (P = 7.9 × 10−12) in the meta-
analysis (OR, 0.63). An MNSI score >2 was
used to define neuropathy cases in both cohorts.
More GWAS are needed to validate this finding
and identify more associated genetic variants.
3.1 Insights into Diabetic
Neuropathy Provided by Gene
Variants Associated
with Charcot-Marie-Tooth
Charcot-Marie-Tooth (CMT) is an umbrella
terms for inherited neuropathies and can poten-
tially inform our understanding of diabetic neu-
ropathy. The most common form of CMT is a
slowly progressive, length-dependent sensorim-
otor neuropathy with onset in the first two
decades of life. There are sub-types with pre-
dominantly motor, sensory or autonomic pre-
sentations. More than 100 genes have now been
linked to CMT. An interesting example of how
understanding CMT can inform our understand-
A. Veluchamy et al.
ing of diabetic neuropathy is hereditary sensory
neuropathy type-1 (HSN1) a sub-type of CMT
with predominant sensory involvement. This is
caused by heterozygous mutations in the genes
SPTLC1 [50, 51] or SPTLC2 [52] which encode
two of the three subunits which constitute the
enzyme Serine Palmitoyl Transferase (SPT).
SPT has the highest affinity for l-serine and
palmitoyl-CoA forming formation of sphingoid
bases (SBs) which are a class of lipid that form
the backbone of all sphingolipids (SLs).
Interestingly HSN1 mutations alter the substrate
specificity of SPT from the amino acid l-serine
to l-alanine or l-glycine, resulting in the pro-
duction of the toxic deoxysphingoid bases
(DSBs) [53]. The toxic DSBs can neither be
converted to complex sphingolipids nor be
degraded, and hence they accumulate. HSN1
patients with SPTLC1 or 2 mutations have ele-
vated plasma DSBs levels. There is a potential
link between this metabolic pathway and dia-
betic neuropathy because elevated plasma DSB
levels are also detected in the plasma of patients
with metabolic syndrome and type 2 diabetes
[54, 55]. The DSB levels in patients with diabe-
tes are not as high as those observed in HSN1
and as yet there is no direct causative link to dia-
betic neuropathy. However these could be part
of the metabolic milieu which is injurious to
neurons in diabetes and understanding toxicity
mechanisms in HSN1 could be relevant to dia-
betic neuropathy.
A further example of how understanding of
CMT can inform our knowledge of diabetic neu-
ropathy is the recent finding that bi-allelic loss-­
of-­
function mutations in the sorbitol
dehydrogenase (SORD) gene is the most com-
mon cause of autosomal recessive inherited neu-
ropathy [56]. SORD is an enzyme that converts
sorbitol into fructose in the two-step polyol path-
way which has previously been implicated in dia-
betic neuropathy. These mutations lead to loss of
enzymatic activity, and participants with these
mutations were found to have higher fasting
serum sorbitol levels. A Drosophila model could
recapitulate the phenotype which was rescued by
treatment with aldose reductase inhibitors (this
drug class was originally developed for targeting
The Genomics of Diabetic Neuropathy
the polyol pathway in diabetic neuropathy). In rat
models of diabetes, inhibition of SORD worsens
neuropathy [57] and the toxicity of raised sorbitol
is thought to be mediated by increased cellular
osmolarity and oxidative stress. Better under-
standing of the mechanisms by which reduced
SORD levels result in CMT could therefore
inform our understanding of diabetic
neuropathy.
3.2 Insights Provided into
Diabetic Neuropathy
from Transcriptional Profiling
in Humans
A further functional genomics approach to
understand diabetic neuropathy is to assess
genome-­wide changes in gene expression evoked
by diabetic neuropathy. Changes in gene expres-
sion reflect a complex mixture of genomic varia-
tion, epigenetic changes and the reaction to
metabolic changes/neural injury, but given the
strong link between gene expression and biolog-
ical function helps provide important insight into
disease pathogenesis. This approach is, however,
limited by access to human nerve tissue from
subjects with diabetic neuropathy and relevant
controls. One study which met this challenge of
tissue availability used tissue from participants
who had taken part in a negative clinical trial
which included two sural nerve biopsies,
52  weeks apart. The authors compared gene
expression (assessed using micro-arrays) in the
sural nerves in those that had developed progres-
sive neuropathy versus those who had not, as
defined by the change in myelinated fibre den-
sity [58]. Over 500 genes were found to be sig-
nificantly differentially expressed in the patient
samples when comparing the progressive dia-
betic neuropathy to non-progressive groups.
These were enriched for pathways involved in
the inflammatory response and lipid metabolism
and likely predominantly reflect gene expression
changes in Schwann cells, but it should be
remembered that these nerve biopsies will also
have contained fibroblasts, vascular and immune
cells; furthermore, some mRNAs are actively
245
transported down axons, potentially distorting
the results. This study also identified gene
expression patterns within nerves which were
predictive of neuropathy progression. Further
studies are needed to see if such an approach can
be replicated in more accessible bio-samples
(such as skin biopsy or even gene expression
profiling in blood).
The same sural nerve tissue was then used for
investigation of epigenetic factors by assessment
of DNA methylation [59]. DNA methylation
refers to the covalent attachment of a methyl
group to a cytosine adjacent to a guanine (CpG)
by DNA methyltransferases. Methylation of
CpG islands (DNA regions with a high density
of CpG motifs) in the promoter of a gene
represses gene expression. A large number of
differentially methylated CpGs were identified,
confirming epigenetic changes in the context of
progressive versus non-progressive diabetic neu-
ropathy. Functional analysis indicated that epi-
genes were enriched in biological pathways
related to nervous system development, axon
guidance, glycerophospholipid metabolism and
MAPK signalling; several of which have previ-
ously been shown to be involved in diabetic
peripheral neuropathy (DPN). A subsequent
data-driven clustering approach was used and
found clusters defined by patterns of gene
expression which were related to high and low
levels of glycosylated haemoglobin (HbA1c)
[60] which was then integrated with assessment
of DNA methylation. There were 998 differen-
tially expressed and 929 differentially methyl-
ated genes when comparing the high and low
HbA1c clusters. The immune system was identi-
fied as an enriched pathway and the extracellular
matrix and axon guidance were also identified.
In summary, tissue availability has limited exam-
ination of gene expression changes within the
nerve due to diabetic neuropathy, however, this
one cohort has been extensively studied and
demonstrates extensive gene expression and epi-
genetic changes associated with neuropathy pro-
gression. These changes particularly highlight
alterations in the immune system/inflammation,
axon growth, extracellular matrix and lipid
metabolism.
246
4 The Genomics
of Neuropathic Pain
in Diabetic Neuropathy
in Humans
Neuropathic pain was shown to have a heritabil-
ity of 37% in a twins study [61]. In recent years,
genetic studies have identified a number of
genetic polymorphisms associated with various
common neuropathic pain conditions but few of
these studies have examined neuropathic pain in
diabetic populations [62]. These genes are asso-
ciated with ion channels, and/or are involved
with neurotransmission, metabolic and immune
response pathways. There is a general lack of
replication of these findings, again through lack
of harmonised phenotyping and sample sizes
with sufficient statistical power.
Meng et al. reported the first GWAS of neu-
ropathic pain in patients with diabetic neuropa-
thy (572 cases and 2491 controls) phenotyped
using medication and monofilament test records.
They found a suggestive variant at chromosome
8p21.3 (rs17428041; OR, 0.67; P  =  1  ×  10−7)
near the glial cell-derived neurotrophic factor
family receptor alpha 2 gene (GFRA2), which is
involved in receptor signalling and the immune
response [63]. In the same diabetes cohort, a
gender-­ specific GWAS reported a SNP-based
heritability of 30% and 14.7% in males and
females respectively. This study identified a
suggestive SNP at chromosome 8q23.1
(rs6986153; OR, 1.67; P = 8.02 × 10−7) near the
high-mobility group box 1 pseudogene 46
(HMGB1P46) in males (470 cases and 2021
controls) and at chromosome 1q35.1
(rs71647933; OR, 2.31; P  =  2.75  ×  10−7) near
the zinc-finger and SCAN domain-encoding-
toll-like receptor 12 pseudogene (ZSCAN20-
TLR12P) in females (491 cases and 1239
controls) [63]. These genes are involved in the
immune response pathway. Both studies lack
genome-wide signals for neuropathic pain sus-
ceptibility and failed to replicate in other neuro-
pathic pain conditions [64]. Moreover, these
suggestive signals were not shown to be associ-
ated with diabetic neuropathy in a more recent
GWAS of diabetic peripheral neuropathy [36].
A. Veluchamy et al.
These signals may be specific to neuropathic
pain and warrant further studies.
A candidate gene association study by Ursu
et al. found an association between the presence
of the gain of function SNP, rs1718119
(Ala348Thr), in the purinergic receptor 7 gene
(P2RX7) and higher pain intensity score in
females with diabetic neuropathy (P  =  0.039)
[65]. In this study, 156 patients with diabetic neu-
ropathic pain were identified by their MNSI score
and 24-h average pain severity rating. One of the
most studied variants in other chronic pain condi-
tions, rs1799971, A118G in the Mu-opioid recep-
tor gene, was shown to be associated with
neuropathic pain in patients with a diabetic foot
ulcer (OR, 0.24; P = 0.038) [66]. However, each
of these variants has only been identified by a
single study, each with a small sample size. More
studies are required to find reliable genetic varia-
tions for diabetic neuropathic pain. Harnessing
modern genomic technologies and aggregating
large cohorts with a validated phenotype can pro-
vide better understanding of the genetic architec-
ture of neuropathic pain in diabetic neuropathy.
4.1 Sodium Channel Variants
and Painful Diabetic
Neuropathy
Voltage-gated sodium channels are key determi-
nants of sensory neuron excitability and are
required for the initial amplification of the gen-
erator potential within nociceptor terminals, the
production of regenerative action potentials to
transmit information along axons to the CNS,
and ultimately the ingress of action potentials
into the central terminals and neurotransmitter
release [67]. The voltage-gated sodium channel
consists of a pore-forming α-subunit complexed
to two non-pore-forming β-subunits (which
modulate trafficking and function of the
α-subunit). There are nine genes encoding the
α-subunits (Nav1.1–1.9) and a number of these
(Nav1.7, 1.8 and 1.9) are highly and relatively
selectively expressed within nociceptors. Human
genetics has demonstrated a strong association
between variants in these genes and pain percep-
The Genomics of Diabetic Neuropathy
tion. Bi-allelic loss-of-function mutations in
Nav1.7 results in congenital insensitivity to pain
[68] whilst gain of function variants in Nav1.7
have been linked with the pain disorders erythro-
melalgia (painful, red extremities exacerbated
by warmth [69]), paroxysmal extreme pain dis-
order (paroxysmal, pain and erythema of the
mandibular, ocular and perineal regions often
triggered by mechanical stimulation [70]) and
small fibre neuropathy (pain associated with
degeneration of small fibre terminals in the skin
[71]). Gain of function variants in Nav1.8 and
Nav1.9 have also been linked to painful neuropa-
thies [72, 73].
Given this human genetic data linking sodium
channel variants to pain, it is a natural question as
to whether variants in sodium channels alter risk
or severity of neuropathic pain in the context of
diabetic neuropathy. A number of studies have
investigated this issue. Li et  al., compared a
cohort of almost 1000 participants with painful
diabetic neuropathy that had been recruited in tri-
als to over 3000 population controls [74] and
found a higher frequency of the Nav1.7 variants
Val991Leu/Met932Leu in cases versus controls;
these variants were also associated with more
severe neuropathic pain. These variants are in
complete linkage disequilibrium (i.e., always
inherited together). Interestingly Val991Leu/
Met932Leu had previously been identified in
patients with idiopathic small fibre neuropathy
and shown to produce gain of function in terms of
enhancing resurgent currents and causing hyper-­
excitability when expressed in rat dorsal root
ganglion (DRG) neurons [71]. A further study
used targeted next generation sequencing to
examine rare variants (present in less than 0.5%
of the general population) in a cohort of partici-
pants with diabetic neuropathy in which detailed
phenotyping for neuropathic pain had been per-
formed [75]. Rare variants in Nav1.7 were more
common in the painful versus painless diabetic
neuropathy group. A number of the variants iden-
tified in the painful diabetic neuropathy group
had previously been identified in painful idio-
pathic small fibre neuropathy and found to pro-
duce gain of function. A number of novel variants
were also identified, which had not previously
247
been linked to painful neuropathy (M1852T and
T1596I), which demonstrated markedly impaired
channel fast inactivation on patch clamp analysis
resulting in gain of function. This is likely to be a
gene-environment interaction in that participants
did not report pain until they developed diabetic
neuropathy. Thus, although the Nav1.7 variants
resulted in hyper-excitability this was tolerated in
early life, but the added ‘stress’ of diabetes
appeared to trigger a chronic pain state. Not all
studies have found an association between rare
variants in voltage-gated sodium channels
(VGSCs) and painful neuropathy. Wadgawan
et al., reported no difference in the frequency of
rare variants in Nav1.7, 1.8 and 1.9  in painful
versus painless neuropathy of mixed aetiology
[76]. This total cohort of 457 included 179 with
diabetic neuropathy, but the diabetic neuropathy
group was not analysed independently. Further
studies on larger well-phenotyped cohorts are
needed.
Genetic variation in VGSCs relevant to neuro-
pathic pain may not be restricted to α-subunits.
Recently a variant (D109N) in the β2-subunit
was identified in a participant with painful dia-
betic neuropathy [77]. The β-subunits are not
pore-forming but complex with and sculpt the
functional properties of pore-forming α-subunits.
This variant was found to result in hyper-­
excitability when expressed in DRG neurons due
to a gain of function effect on Nav1.7.
Identification of variants in VGSCs may have
implications for treatment of neuropathic pain.
Idiopathic small fibre neuropathy has already
provided an example of a stratified pain medi-
cine approach in which lacosamide, which acts
as a blocker of VGSCs but has limited efficacy
in unselected neuropathic pain patients, was
shown to be effective versus placebo in a
selected group of patients with painful small
fibre neuropathy associated with rare variants in
Nav1.7 [78]. In certain rare cases personalised
treatment approaches may be taken: for exam-
ple, an S242T variant in Nav1.8 was identified
in a participant with painful diabetic neuropathy
[79]. Interestingly the equivalent amino acid
change (S241T) in Nav1.7 had previously been
linked to inherited erythromelalgia and found to
248
cause gain of function which was responsive to
carbamazepine (which has activity against
VGSCs but is not routinely used in the treatment
of painful diabetic neuropathy). Structural mod-
elling comparing S242T Nav1.8 to S241T
Nav1.7 showed that these were indeed equiva-
lent amino acids in these two ion channels, in
the same orientation and electrophysiological
analysis demonstrated that S242T Nav1.8 also
resulted in gain of function which could be
reversed by carbamazepine.
5 Conclusion
We are in the early phase of applying genomic
technologies to diabetic neuropathy and the asso-
ciated complication of neuropathic pain. Large
cohorts with detailed harmonised phenotyping of
neuropathy and neuropathic pain have only
recently been developed and so we are now likely
to see an accelerating pace of discovery. There
have been some early encouraging findings, for
instance, the GWAS of diabetic neuropathy iden-
tifying a SNP which is in proximity to and alters
the expression of SCN2A (encoding Nav1.2).
Rare variants in voltage-gated sodium channels
(especially Nav1.7) have also been linked to
painful (versus painless) diabetic neuropathy in
candidate gene studies. These lead to gain of
channel function so providing a plausible func-
tional link to sensory neuron hyper-excitability
and pain. We have also learned lessons as to how
metabolism can have a major impact on nerve
function through understanding the genetic basis
of neuropathies inherited in a Mendelian fashion
(such as CMT) and interestingly parallels are
arising with some of these metabolic changes to
those observed in diabetes. As this field matures
advances in genomics could have significant
impact in the clinical management of diabetic
neuropathy and pain: including improved under-
standing of the pathogenesis of these disorders
and discovery of novel treatment targets, the
identification of those most at risk of neuropathy
and pain in order to intervene at an early stage
and a more personalised approach to treatment
selection.
A. Veluchamy et al.
Acknowledgements This work was supported by the
European Union’s Horizon 2020 research and innovation
programme under grant agreement No 633491
(DOLORisk).
References
1. Claussnitzer M, Cho JH, Collins R, Cox NJ,
Dermitzakis ET, Hurles ME, Kathiresan S, Kenny EE,
Lindgren CM, MacArthur DG, et al. A brief history
of human disease genetics. Nature. 2020;577:179–89.
2. Ahlqvist E, van Zuydam NR, Groop LC, McCarthy
MI. The genetics of diabetic complications. Nat Rev
Nephrol. 2015;11:277–87.
3. Finnerup NB, Haroutounian S, Kamerman P, Baron
R, Bennett DL, Bouhassira D, Cruccu G, Freeman R,
Hansson P, Nurmikko T, et  al. Neuropathic pain: an
updated grading system for research and clinical prac-
tice. Pain. 2016;157:1599–606.
4. van Hecke O, Kamerman PR, Attal N, Baron R,
Bjornsdottir G, Bennett DL, Bennett MI, Bouhassira
D, Diatchenko L, Freeman R, et al. Neuropathic pain
phenotyping by international consensus (NeuroPPIC)
for genetic studies: a NeuPSIG systematic review,
Delphi survey, and expert panel recommendations.
Pain. 2015;156:2337–53.
5. Gloyn AL, Pearson ER, Antcliff JF, Proks P, Bruining
GJ, Slingerland AS, Howard N, Srinivasan S, Silva
JM, Molnes J, et al. Activating mutations in the gene
encoding the ATP-sensitive potassium-channel sub-
unit Kir6.2 and permanent neonatal diabetes. N Engl
J Med. 2004;350:1838–49.
6. Pipatpolkai T, Usher S, Stansfeld PJ, Ashcroft
FM. New insights into KATP channel gene mutations
and neonatal diabetes mellitus. Nat Rev Endocrinol.
2020;16:378–93.
7. Consortium EP.  An integrated encyclopedia of
DNA elements in the human genome. Nature.
2012;489:57–74.
8. Mahajan A, Taliun D, Thurner M, Robertson NR,
Torres JM, Rayner NW, Payne AJ, Steinthorsdottir V,
Scott RA, Grarup N, et al. Fine-mapping type 2 diabe-
tes loci to single-variant resolution using high-density
imputation and islet-specific epigenome maps. Nat
Genet. 2018;50:1505–13.
9. Krentz NAJ, Gloyn AL. Insights into pancreatic islet
cell dysfunction from type 2 diabetes mellitus genet-
ics. Nat Rev Endocrinol. 2020;16:202–12.
10. Sudlow C, Gallacher J, Allen N, Beral V, Burton P,
Danesh J, Downey P, Elliott P, Green J, Landray M,
et al. UK biobank: an open access resource for identi-
fying the causes of a wide range of complex diseases
of middle and old age. PLoS Med. 2015;12:e1001779.
11. Turro E, Astle WJ, Megy K, Graf S, Greene D,
Shamardina O, Allen HL, Sanchis-Juan A, Frontini
M, Thys C, et  al. Whole-genome sequencing of
patients with rare diseases in a national health system.
Nature. 2020;583:96–102.
The Genomics of Diabetic Neuropathy
12. Karczewski KJ, Francioli LC, Tiao G, Cummings
BB, Alfoldi J, Wang Q, Collins RL, Laricchia KM,
Ganna A, Birnbaum DP, et  al. The mutational con-
straint spectrum quantified from variation in 141,456
humans. Nature. 2020;581:434–43.
13. Landrum MJ, Chitipiralla S, Brown GR, Chen C, Gu
B, Hart J, Hoffman D, Jang W, Kaur K, Liu C, et al.
ClinVar: improvements to accessing data. Nucleic
Acids Res. 2020;48:D835–44.
14. Feldman EL, Callaghan BC, Pop-Busui R, Zochodne
DW, Wright DE, Bennett DL, Bril V, Russell JW,
Viswanathan V.  Diabetic neuropathy. Nat Rev Dis
Primers. 2019;5:41.
15. Arredondo-García VK, Cepeda-Nieto AC, Batallar-­
Gómez T, Salinas-Santander M, Zugasti-Cruz A,
Ramírez-Calvillo L, Maldonado-Sánchez K, Morlett-­
Chávez J, Barajas-Martínez H.  Association of the
vascular endothelial growth factor gene polymor-
phism +936 C/T with diabetic neuropathy in Mexican
patients with Type 2 diabetes mellitus. Arch Med Res.
2019;50:181–6.
16. Basol N, Inanir A, Yigit S, Karakus N, Kaya SU. High
association of IL-4 gene intron 3 VNTR polymorphism
with diabetic peripheral neuropathy. J Mol Neurosci.
2013;51(2):437–41. https://doi.org/10.1007/s12031-
013-0048-y. Epub 2013 Jun 16. PMID: 23771787.
17. Buraczynska M, Buraczynska K, Dragan M, Ksiazek
A. Pro198Leu Polymorphism in the glutathione per-
oxidase 1 gene contributes to diabetic peripheral neu-
ropathy in type 2 diabetes patients. Neuromolecular
Med. 2017;19(1):147–53. https://doi.org/10.1007/
s12017-016-8438-2. Epub 2016 Sep 3. PMID:
27592002.
18. Chen Y, Wang J, Wang L, Huang P, Tan ZX, Liu HJ.
Adiponectin gene polymorphisms are associated with
increased susceptibility to diabetic peripheral neu-
ropathy. Biomarkers. 2015;20(6–7):474–80. https://
doi.org/10.3109/1354750X.2015.1101784. PMID:
26616148.
19. Ciccacci C, Morganti R, Di Fusco D, D'Amato C,
Cacciotti L, Greco C, Rufini S, Novelli G, Sangiuolo
F, Marfia GA, et  al. Common polymorphisms in
MIR146a, MIR128a and MIR27a genes contribute
to neuropathy susceptibility in type 2 diabetes. Acta
Diabetol. 2014;51:663–71.
20. Ciccacci C, Latini A, Greco C, Politi C, D’Amato
C, Lauro D, Novelli G, Borgiani P, Spallone
V.  Association between a MIR499A polymorphism
and diabetic neuropathy in type 2 diabetes. J Diabetes
Complicat. 2018;32(1):11–7.
21. Gupta B, Singh SK. Association of aldose reductase
gene polymorphism (C-106T) in susceptibility of dia-
betic peripheral neuropathy among north Indian popu-
lation. J Diabetes Complications. 2017;31(7):1085–9.
https://doi.org/10.1016/j.jdiacomp.2017.04.011.
Epub 2017 Apr 28. PMID: 28495421.
22. Yigit S, Karakus N, Inanir A.  Association of
MTHFR gene C677T mutation with diabetic periph-
eral neuropathy and diabetic retinopathy. Mol Vis.
2013;19:1626–30.
249
23. Ji ZY, Li HF, Lei Y, Rao YW, Tan ZX, Liu HJ, Yao
GD, Hou B, Sun ML. Association of adiponectin
gene polymorphisms with an elevated risk of diabetic
peripheral neuropathy in type 2 diabetes patients. J
Diabetes Complications. 2015;29(7):887–92. https://
doi.org/10.1016/j.jdiacomp.2015.06.008. Epub 2015
Jun 19. PMID:26144281.
24. Kakavand Hamidi A, Radfar M, Amoli
MM.  Association between MTHFR variant and dia-
betic neuropathy. Pharmacol Rep. 2018;70:1–5.
25. Kolla VK, Madhavi G, Pulla Reddy B, Srikanth Babu
BM, Yashovanthi J, Valluri VL, Ramesh J, Akka J.
Association of tumor necrosis factor alpha, interferon
gamma and interleukin 10 gene polymorphisms with
peripheralneuropathy in South Indian patients with
type 2 diabetes. Cytokine. 2009;47(3):173–7. https://
doi.org/10.1016/j.cyto.2009.06.007. Epub 2009 Jul
15. PMID: 19608431.
26. Mansoor Q, Javaid A, Bilal N, Ismail M. Angiotensin-
converting enzyme (ACE) gene II genotype pro-
tects against the development of diabetic peripheral
neuropathy in type 2 diabetes mellitus. J Diabetes.
2012;4(3):257–61. https://doi.org/10.1111/j.1753-
0407.2012.00205.x. PMID: 22607040.
27. Marzban A, Kiani J, Hajilooi M, Rezaei H,
Kahramfar Z, Solgi G. HLA class II alleles and
risk for peripheral neuropathy in type 2 diabetes
patients. Neural Regen Res. 2016;11(11):1839–44.
https://doi.org/10.4103/1673-5374.194756. PMID:
28123430.
28. Monastiriotis C, Papanas N, Trypsianis G,
Karanikola K, Veletza S, Maltezos E. The ε4
allele of the APOE gene is associated with more
severe peripheral neuropathy in type 2 diabetic
patients. Angiology. 2013;64(6):451–5. https://doi.
org/10.1177/0003319712453645. Epub 2012 Jul 23.
PMID: 22826377.
29. Papanas N, Papatheodorou K, Papazoglou D,
Kotsiou S, Christakidis D, Maltezos E. An insertion/
deletion polymorphism in the alpha2B adrenocep-
tor gene is associated with peripheral neuropathy
in patients with type 2 diabetes mellitus. Exp Clin
Endocrinol Diabetes. 2007;115(5):327–30. https://
doi.org/10.1055/s-2007-967084. PMID:17516297.
30. Ren Z, Ji N, Jia K, et al. Association of the intercel-
lular adhesion molecule-1 gene polymorphisms with
type 2 diabetes and diabetic peripheral neuropa-
thy in a Chinese Han population. Genes Genomics.
2015;37:69–75.
31. Shah VN, Cheema BS, Kohli HS, Sharma R, Khullar
M, Bhansali A. Endothelial nitric oxide synthase gene
polymorphism and the risk of diabetic neuropathy in
Asian Indian patients with type 2 diabetes. J Diabetes
Metab. 2013;4:243.
32. Stephens JW, Dhamrait SS, Acharya J, Humphries
SE, Hurel SJ. A common variant in the ACE gene is
associated with peripheral neuropathy in women with
type 2 diabetes mellitus. J Diabetes Complications.
2006;20(5):317–21. https://doi.org/10.1016/j.jdia-
comp.2005.07.010. PMID: 16949519.
250
33. Stoian A, Bacârea A, Moţǎţǎianu A, Stoian M, Gliga
F, Bacârea V, Carmen D, Bǎnescu C. Vascular endo-
thelial growth factor insertion/deletion gene polymor-
phism in patients with type 2 diabetes and diabetic
peripheral polyneuropathy. Rev Romana Med Lab.
2014;22:165–72.
34. Sun L, Ma J, Mao Q, Yang YL, Ma LL, Niu L, Liu LF.
Association of single nucleotide polymorphisms in
CACNA 1A/CACNA 1C/CACNA 1H calcium channel
genes with diabetic peripheral neuropathy in Chinese
population. Biosci Rep. 2018;38(3):BSR20171670.
https://doi.org/10.1042/BSR20171670. Print 2018
Jun 29. PMID: 29581247.
35. Tang TS, Prior SL, Li KW, Ireland HA, Bain SC,
Hurel SJ, Cooper JA, Humphries SE, Stephens JW.
Association between the rs1050450 glutathione per-
oxidase-1 (C > T) gene variant and peripheral neu-
ropathy in two independent samples of subjects
with diabetes mellitus. Nutr Metab Cardiovasc Dis.
2012;22(5):417–25. https://doi.org/10.1016/j.
numecd.2010.08.001. Epub 2010 Dec 24. PMID:
21185702.
36. Tang Y, Lenzini PA, Pop-Busui R, Ray PR, Campbell
H, Perkins BA, Callaghan B, Wagner MJ, Motsinger-­
Reif AA, Buse JB, et al. A genetic locus on chromo-
some 2q24 predicting peripheral neuropathy risk in
type 2 diabetes: results from the ACCORD and BARI
2D studies. Diabetes. 2019;68:1649–62.
37. Ukinc K, Ersoz HO, Karahan C, Erem C,
Eminagaoglu S, Hacihasanoglu AB, Yilmaz M,
Kocak M.  Methyltetrahydrofolate reductase C677T
gene mutation and hyperhomocysteinemia as a novel
risk factor for diabetic nephropathy. Endocrine.
2009;36:255–61.
38. Wang H, Fan D, Hong T.  Is the C677T polymor-
phism in methylenetetrahydrofolate reductase gene or
plasma homocysteine a risk factor for diabetic periph-
eral neuropathy in Chinese individuals? Neural Regen
Res. 2012;7:2384–91.
39. Politi C, Ciccacci C, D'Amato C, Novelli G, Borgiani
P, Spallone V.  Recent advances in exploring the
genetic susceptibility to diabetic neuropathy. Diabetes
Res Clin Pract. 2016;120:198–208.
40. Witzel II, Jelinek HF, Khalaf K, Lee S, Khandoker
AH, Alsafar H.  Identifying common genetic risk
factors of diabetic neuropathies. Front Endocrinol.
2015;6:1–18.
41. Zhao Y, Zhu R, Wang D, Liu X. Genetics of diabetic
neuropathy: systematic review, meta-analysis and
trial sequential analysis. Ann Clin Transl Neurol.
2019;6:1996–2013.
42. Wu S, Han Y, Hu Q, Zhang X, Cui G, Li Z, Guan
Y. Effects of common polymorphisms in the MTHFR
and ACE genes on diabetic peripheral neuropa-
thy progression: a meta-analysis. Mol Neurobiol.
2017;54:2435–44.
43. James E, Vincent A, Cheng T, Feldman EL. Diabetic
neuropathy: mechanisms to management. Pharmacol
Ther. 2008;120:1–34.
44. Didangelos T, Tziomalos K, Margaritidis C,
Kontoninas Z, Stergiou I, Tsotoulidis S, Karlafti E,
A. Veluchamy et al.
Mourouglakis A, Hatzitolios AI. Efficacy of adminis-
tration of an angiotensin converting enzyme inhibitor
for two years on autonomic and peripheral neuropa-
thy in patients with diabetes mellitus. J Diabetes Res.
2017;2017:6719239.
45. Babizhayev MA, Strokov IA, Nosikov VV, Savel’yeva
EL, Sitnikov VF, Yegorov YE, Lankin VZ. The role
of oxidative stress in diabetic neuropathy: generation
of free radical species in the glycation reaction and
gene polymorphisms encoding antioxidant enzymes
to genetic susceptibility to diabetic neuropathy in
population of type I diabetic patients. Cell Biochem
Biophys. 2015;71(3):1425–43.
46. Kasznicki J, Sliwinska A, Kosmalski M, Merecz A,
Majsterek I, Drzewoski J. Genetic polymorphisms
(Pro197Leu of Gpx1, +35A/C of SOD1, -262C/T of
CAT), the level of antioxidant proteins (GPx1, SOD1,
CAT) and the risk of distal symmetric polyneuropathy
in Polish patients with type 2 diabetes mellitus. Adv
Med Sci. 2016;61(1):123–9.
47. Chistyakov DA, Savost’anov KV, Zotova EV, Nosikov
VV.  Polymorphisms in the Mn-SOD and EC-SOD
genes and their relationship to diabetic neuropathy in
type 1 diabetes mellitus. BMC Med Genet. 2001;2:4.
48. Strokov IA, Bursa TR, Drepa OI, Zotova VE, Nosikov
VV, Ametov AS. Predisposing genetic factors for dia-
betic polyneuropathy in patients with type 1 diabetes:
a population-based case-control study. Acta Diabetol.
2003;40:375–9.
49. Zotova VE, Chistyakov DA, Savost'yanov VK, Bursa
TR, Galeev VI, Strokov IA, Nosikov VV. Association
of the SOD2 Ala(−9)Val and SOD3 Arg213Gly poly-
morphisms with diabetic polyneuropathy in diabetes
mellitus type 1. Mol Biol. 2003;37:404–8.
50. Bejaoui K, Wu C, Scheffler MD, Haan G, Ashby P,
Wu L, de Jong P, Brown RH Jr. SPTLC1 is mutated
in hereditary sensory neuropathy, type 1. Nat Genet.
2001;27:261–2.
51. Dawkins JL, Hulme DJ, Brahmbhatt SB, Auer-­
Grumbach M, Nicholson GA. Mutations in SPTLC1,
encoding serine palmitoyltransferase, long chain base
subunit-1, cause hereditary sensory neuropathy type
I. Nat Genet. 2001;27:309–12.
52. Rotthier A, Auer-Grumbach M, Janssens K, Baets J,
Penno A, Almeida-Souza L, Van Hoof K, Jacobs A,
De Vriendt E, Schlotter-Weigel B, et al. Mutations in
the SPTLC2 subunit of serine palmitoyltransferase
cause hereditary sensory and autonomic neuropathy
type I. Am J Hum Genet. 2010;87:513–22.
53. Penno A, Reilly MM, Houlden H, Laura M, Rentsch
K, Niederkofler V, Stoeckli ET, Nicholson G, Eichler
F, Brown RH Jr, et al. Hereditary sensory neuropathy
type 1 is caused by the accumulation of two neurotoxic
sphingolipids. J Biol Chem. 2010;285:11178–87.
54. Bertea M, Rutti MF, Othman A, Marti-Jaun J,
Hersberger M, von Eckardstein A, Hornemann
T. Deoxysphingoid bases as plasma markers in diabe-
tes mellitus. Lipids Health Dis. 2010;9:84.
55. Dohrn MF, Othman A, Hirshman SK, Bode H,
Alecu I, Fahndrich E, Karges W, Weis J, Schulz JB,
Hornemann T, et  al. Elevation of plasma 1-deoxy-­
The Genomics of Diabetic Neuropathy
sphingolipids in type 2 diabetes mellitus: a suscepti-
bility to neuropathy? Eur J Neurol. 2015;22:806–14,
e855
56. Cortese A, Zhu Y, Rebelo AP, Negri S, Courel S, Abreu
L, Bacon CJ, Bai Y, Bis-Brewer DM, Bugiardini E,
et al. Biallelic mutations in SORD cause a common
and potentially treatable hereditary neuropathy with
implications for diabetes. Nat Genet. 2020;52:473–81.
57. Schmidt RE, Dorsey DA, Beaudet LN, Plurad SB,
Parvin CA, Yarasheski KE, Smith SR, Lang HJ,
Williamson JR, Ido Y. Inhibition of sorbitol dehydro-
genase exacerbates autonomic neuropathy in rats with
streptozotocin-induced diabetes. J Neuropathol Exp
Neurol. 2001;60:1153–69.
58. Hur J, Sullivan KA, Pande M, Hong Y, Sima AA,
Jagadish HV, Kretzler M, Feldman EL.  The identi-
fication of gene expression profiles associated with
progression of human diabetic neuropathy. Brain.
2011;134:3222–35.
59. Guo K, Elzinga S, Eid S, Figueroa-Romero C, Hinder
LM, Pacut C, Feldman EL, Hur J.  Genome-wide
DNA methylation profiling of human diabetic periph-
eral neuropathy in subjects with type 2 diabetes mel-
litus. Epigenetics. 2019;14:766–79.
60. Guo K, Eid SA, Elzinga SE, Pacut C, Feldman EL,
Hur J.  Genome-wide profiling of DNA methyla-
tion and gene expression identifies candidate genes
for human diabetic neuropathy. Clin Epigenetics.
2020;12:123.
61. Momi SK, Fabiane SM, Lachance G, Livshits G,
Williams FM.  Neuropathic pain as part of chronic
widespread pain: environmental and genetic influ-
ences. Pain. 2015;156:2100–6.
62. Veluchamy A, Hebert HL, Meng W, Palmer CNA,
Smith BH.  Systematic review and meta-analysis
of genetic risk factors for neuropathic pain. Pain.
2018;159:825–48.
63. Meng W, Deshmukh H, Donnelly LA, Torrance N,
Colhoun HM, Palmer CNA, Smith BH.  A genome-­
wide association study provides evidence of
sex-­specific involvement of Chr1p35.1 (ZSCAN20-­
TLR12P) and Chr8p23.1 (HMGB1P46) with diabetic
neuropathic pain. EBioMedicine. 2015;2:1386–93.
64. Warner SC, van Meurs JB, Schiphof D, Bierma-­
Zeinstra SM, Hofman A, Uitterlinden AG, Richardson
H, Jenkins W, Doherty M, Valdes AM. Genome-wide
association scan of neuropathic pain symptoms post
total joint replacement highlights a variant in the
protein-­kinase C gene. Eur J Hum Genet. 2017;44:1–6.
65. Ursu D, Ebert P, Langron E, Ruble C, Munsie L, Zou
W, Fijal B, Qian Y-W, McNearney TA, Mogg A, et al.
Gain and loss of function of P2X7 receptors: mecha-
nisms, pharmacology and relevance to diabetic neuro-
pathic pain. Mol Pain. 2014;10:37.
66. Cheng KI, Lin SR, Chang LL, Wang JY, Lai
CS.  Association of the functional A118G polymor-
phism of OPRM1 in diabetic patients with foot ulcer
pain. J Diabetes Complicat. 2010;24:102–8.
67. Bennett DL, Clark AJ, Huang J, Waxman SG, Dib-­
Hajj SD. The role of voltage-gated sodium channels
in pain signaling. Physiol Rev. 2019;99:1079–151.
251
68. Cox JJ, Reimann F, Nicholas AK, Thornton G, Roberts
E, Springell K, Karbani G, Jafri H, Mannan J, Raashid
Y, et al. An SCN9A channelopathy causes congenital
inability to experience pain. Nature. 2006;444:894–8.
69. Yang Y, Wang Y, Li S, Xu Z, Li H, Ma L, Fan J, Bu
D, Liu B, Fan Z, et al. Mutations in SCN9A, encod-
ing a sodium channel alpha subunit, in patients with
primary erythermalgia. J Med Genet. 2004;41:171–4.
70. Fertleman CR, Baker MD, Parker KA, Moffatt S,
Elmslie FV, Abrahamsen B, Ostman J, Klugbauer N,
Wood JN, Gardiner RM, et al. SCN9A mutations in
paroxysmal extreme pain disorder: allelic variants
underlie distinct channel defects and phenotypes.
Neuron. 2006;52:767–74.
71. Faber CG, Hoeijmakers JG, Ahn HS, Cheng X,
Han C, Choi JS, Estacion M, Lauria G, Vanhoutte
EK, Gerrits MM, et  al. Gain of function Nanu1.7
mutations in idiopathic small fiber neuropathy. Ann
Neurol. 2012a;71:26–39.
72. Faber CG, Lauria G, Merkies IS, Cheng X, Han C,
Ahn HS, Persson AK, Hoeijmakers JG, Gerrits MM,
Pierro T, et  al. Gain-of-function Nav1.8 mutations
in painful neuropathy. Proc Natl Acad Sci U S A.
2012b;109:19444–9.
73. Huang J, Han C, Estacion M, Vasylyev D, Hoeijmakers
JG, Gerrits MM, Tyrrell L, Lauria G, Faber CG, Dib-­
Hajj SD, et al. Gain-of-function mutations in sodium
channel Na(v)1.9  in painful neuropathy. Brain.
2014;137:1627–42.
74. Li QS, Cheng P, Favis R, Wickenden A, Romano
G, Wang H.  SCN9A variants may be implicated in
neuropathic pain associated with diabetic periph-
eral neuropathy and pain severity. Clin J Pain.
2015;31:976–82.
75. Blesneac I, Themistocleous AC, Fratter C, Conrad
LJ, Ramirez JD, Cox JJ, Tesfaye S, Shillo PR, Rice
ASC, Tucker SJ, et al. Rare NaV1.7 variants associ-
ated with painful diabetic peripheral neuropathy. Pain.
2018;159:469–80.
76. Wadhawan S, Pant S, Golhar R, Kirov S, Thompson
J, Jacobsen L, Qureshi I, Ajroud-Driss S, Freeman R,
Simpson DM, et al. NaV channel variants in patients
with painful and nonpainful peripheral neuropathy.
Neurol Genet. 2017;3:e207.
77. Alsaloum M, Estacion M, Almomani R, Gerrits MM,
Bonhof GJ, Ziegler D, Malik R, Ferdousi M, Lauria
G, Merkies IS, et al. A gain-of-function sodium chan-
nel beta2-subunit mutation in painful diabetic neurop-
athy. Mol Pain. 2019;15:1744806919849802.
78. de Greef BTA, Hoeijmakers JGJ, Geerts M, Oakes M,
Church TJE, Waxman SG, Dib-Hajj SD, Faber CG,
Merkies ISJ.  Lacosamide in patients with Nav1.7
mutations-related small fibre neuropathy: a random-
ized controlled trial. Brain. 2019;142:263–75.
79. Han C, Themistocleous AC, Estacion M, Dib-Hajj
FB, Blesneac I, Macala L, Fratter C, Bennett DL,
Waxman SG, Dib-Hajj SD.  The novel activity of
carbamazepine as an activation modulator extends
from NaV1.7 mutations to the NaV1.8-S242T mutant
channel from a patient with painful diabetic neuropa-
thy. Mol Pharmacol. 2018;94:1256–69.
 Metabolic Mechanisms in Diabetic
Neuropathy

Mark Yorek

1 Introduction dependent neuropathy [4, 5]. This decrease in

sensory perception is the most common and ear-
Diabetic peripheral neuropathy is a heterogenous liest form of diabetic peripheral neuropathy and
condition that can manifest as many different is gradual with symptoms of tingling, pain, and
symptoms and are the most common complica- loss of sensation in the toes [6, 7]. Clinical evi-
tions of diabetes mellitus with an estimated prev- dence of motor dysfunction is less prevalent with
alence ranging up to 50% and possibly higher only 1–6% of diabetic patients displaying clinical
depending on the diagnostic criteria, whether the symptoms and generally occurs in patients with
subjects have type 1 or type 2 diabetes and dura- established diabetic peripheral neuropathy [7]. A
tion of diabetes [1–3]. Diabetic peripheral neu- decrease in motor nerve conduction velocity
ropathy affects both sensorimotor and autonomic early in diabetic animal models is a common
parts of the peripheral nervous system [1]. The finding but there is also evidence in animal mod-
most common clinically recognized form is dia- els and humans of a decrease in compound mus-
betic distal symmetric sensorimotor polyneurop- cle action potential amplitudes and reduced
athy which is characterized by the progressive muscle strength but this has been much less stud-
loss of nerve fibers, both large and small [1, 4]. In ied in the pre-clinical or clinical setting [7, 8].
this progressive disorder the most distal nerve The Diabetes Control and Complications Trial
segments of the feet and hands are affected first (DCCT) and United Kingdom Prospective
and involve retraction of terminal sensory axons Diabetes Study (UKPDS) that focused on type 1
in the periphery with relative preservation of the and type 2 diabetes, respectively, demonstrated
perikarya. This phenomenon is often referred to that hyperglycemia is an important contributing
as “dying back syndrome” or the “stocking and factor to the onset and progression of nerve dam-
glove” pattern reflects damage to the longest sen- age especially in those subjects with type 1 dia-
sory axons first and thus is considered a length-­ betes [9, 10]. However, other reports/studies have
shown that good glycemic control provides little
M. Yorek (*) benefit in towards peripheral neuropathy in those
Department of Internal Medicine, University of Iowa, with type 2 diabetes and recent evidence also
Iowa City, IA, USA suggests that small nerve fiber damage occurs in
Department of Veterans Affairs, Iowa City VA Health individuals with impaired glucose tolerance,
Care System, Iowa City, IA, USA independent of chronic hyperglycemia, and the
Fraternal Order of Eagles Diabetes Research Center, diagnosis of diabetes [11–14]. Thus, other condi-
University of Iowa, Iowa City, IA, USA tions, in addition to hyperglycemia, must contrib-
e-mail: myorek@icva.gov

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 253
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_15
254
ute to the onset and progression of diabetic
peripheral neuropathy in subjects with type 2 dia-
betes. Because of the multiple clinical manifesta-
tions associated with diabetic peripheral
neuropathy that can include pain in 15–30% of
diabetic patients with neuropathy determining
the pathophysiology and therapy for diabetic
neuropathies is challenging [15].
Studies using animal models of both type 1
and type 2 diabetes have resulted in the
identification of wide array of pathological
­ second step is the oxidation of sorbitol to form
mechanisms as contributing to diabetes periph-
eral neuropathy. Diabetic peripheral neuropathy
has been described by some investigators to be a
disease of the vasculature leading to nerve isch-
emia and altered nerve function [16–20]. Other
investigators have proposed that diabetic periph-
eral neuropathy is caused by a combination of
metabolic defects associated with an increased
flux of glucose through the aldose reductase
pathway leading to a defect in Na+/K+-ATPase
and protein kinase C activities and an alteration
of signal transduction pathways in the nerve
[21, 22]. Additional pathologic contributors to
diabetic peripheral neuropathy have been
reported to include increased formation of
advanced glycation endproducts, hexosamine
pathway dysregulation, reduced neurotrophic
support, increased inflammatory, and oxidative
stress and dyslipidemia [23, 24]. Overall, these
mechanisms and likely others cause damage to
neurons, Schwann cells, and the vasculature.
Ultimately, relentless damage to the nerve com-
plex and surrounding vasculature leads to dia-
betic peripheral neuropathy. Given the complex
etiology of diabetic peripheral neuropathy a
successful treatment will likely require a combi-
nation of early detection, lifestyle changes, and
pharmaceutical interventions targeting the
mechanisms deemed most responsible for the
pathogenesis. Before this can occur additional
studies are needed to determine the most rele-
vant and targetable causes of diabetic peripheral
neuropathy.
The present review will focus on a number of
the mechanisms introduced above with an
emphasis on those impacted by metabolic
dysregulation.
M. Yorek
2 Role of Aldose Reductase
and Polyol Pathway
The polyol pathway is catalyzed by two enzymes.
Aldose reductase, the first enzyme of this path-
way, is believed to have a primary role in early
metabolic damage to peripheral nerves [25]. In
this initial step glucose is reduced to sorbitol by a
reaction that requires NADPH, which indirectly
contributes to an increase in oxidative stress. The
fructose a reaction catalyzed by NAD-dependent
sorbitol dehydrogenase (see Fig. 1). During peri-
ods of excess glucose such as diabetes tissues
independent of insulin for glucose transport
accumulate glucose intracellularly leading to an
increased metabolism of glucose by this pathway
and the accumulation of sorbitol causing osmotic
stress and a decrease the intracellular levels of
myo-inositol and taurine. The decrease in myo-­
inositol and taurine occurs in response to the
excess intracellular accumulation of the osmolyte
sorbitol and is referred to the compatible osmo-
lyte hypothesis (to be discussed below). Other
negative consequences of this pathway when
activated are the excess generation of fructose,
which is a more potent glycation agent than glu-
cose leading to an increase in glycative stress.
The polyol pathway remains an interesting
therapeutic target for the treatment of diabetic
peripheral neuropathy even though early clinical
trials were unsuccessful in achieving meaningful
improvement in human subjects in spite of wide
spread success of pre-clinical studies [25–27].
Pre-clinical studies with a variety of diabetic ani-
mal models using multiple approaches including
inhibitors of both aldose reductase and sorbitol
dehydrogenase and gene and dietary manipula-
tion have provided overwhelming evidence of an
important role for this pathway in diabetes com-
plications including peripheral neuropathy. In
rodent studies a wide variety of aldose reductase
inhibitors have been shown to improve multiple
endpoints associated with diabetic peripheral
neuropathy. In this article I will focus on results
from pre-clinical and clinical studies performed
with three different aldose reductase inhibitors;
sorbinil, epalrestat and ranirestat.
Metabolic Mechanisms in Diabetic Neuropathy
Fig. 1  Illustration of polyol pathway with blue ovals representing conditions that contributes to diabetic peripheral
neuropathy
Sorbinil was one of the earliest aldose reduc-
tase inhibitors to undergo extensive pre-clinical
and clinical studies for diabetic peripheral neu-
ropathy. In studies with type 1 diabetic rats
treated with streptozotocin treatment with
sorbinil prevented as well as reversed defective
axonal transport and slowing of motor nerve con-
duction velocity [28, 29]. Sorbinil treatment also
normalized myo-inositol levels in the sciatic
nerve in these studies. Another group of investi-
gators demonstrated that treating streptozotocin-­
induced diabetic rats with sorbinil corrected both
motor and sensory nerve conduction velocity as
well as endoneurial nutritive blood flow of the
sciatic nerve, metabolic abnormalities that
included the NAD+/NADH redox imbalance and
energy deficiency and improved oxidative stress
[30]. Cameron et  al. [31] in studies also using
streptozotocin-diabetic rats demonstrated that
sorbinil treatment protected axon growth
retardation.
In early studies of the sciatic nerve untreated
diabetic rats were found to have a decrease in
255
Na+/K+ ATPase activity and myo-inositol con-
tent that was corrected following sorbinil treat-
ment [32, 33]. All these studies reported that
sorbinil treatment reduced tissue sorbitol levels.
These studies led to the compatible osmolyte
hypothesis that stated myo-inositol depletion and
abnormal signaling by inositol phospholipids
contribute to a decrease in Na+/K+ ATPase activ-
ity in the sciatic nerve, which generates the trans-
membrane sodium and potassium potentials
necessary for nerve impulse conduction and the
sodium gradient needed for sodium-dependent
uptake of substrates [32]. The compatible osmo-
lyte hypothesis was derived from studies demon-
strating that sorbitol accumulation caused by
hyperglycemia/diabetes and activation of the
polyol pathway leads to a corresponding decrease
in myo-inositol and taurine levels in peripheral
nerves [34]. The diabetes-induced decrease in
myo-inositol and taurine in peripheral nerves was
preventable by treatment with an aldose reduc-
tase inhibitor such as sorbinil that prevented the
increase in sorbitol [35, 36]. This hypothesis as
256
being a contributing factor to peripheral neuropa-
thy is supported by additional studies demon-
strating the replenishing myo-inositol or taurine
through the diet partially reversed slowing of
motor nerve conduction velocity in diabetic rats
[34, 37, 38]. Using another approach my labora-
tory has shown that treating rats with a diet con-
taining a high concentration of l-fucose caused a
slowing of nerve conduction velocity and
decrease sciatic nerve Na+/K+ ATPase activity
and myo-inositol content [39]. l-Fucose is a
potent competitive inhibitor of myo-inositol
transport by neural and endothelial cells [40–42].
Restoring myo-inositol levels through the diet of
rats fed the l-fucose diet restored nerve function
and Na+/K+ ATPase activity [39]. My laboratory
has also demonstrated that treating streptozotocin-­
induced diabetic rats with sorbinil or myo-­inositol
improved endoneurial blood flow, motor nerve
conduction velocity, and vascular function of epi-
neurial arterioles of the sciatic nerve [43]. The
ladder finding is important because we had previ-
ously demonstrated that decrease of vascular
relaxation to acetylcholine by epineurial arteri-
oles precedes the slowing of nerve conduction
velocity indicating that vascular dysregulation is
a contributing factor to diabetes-induced nerve
dysfunction [44]. Adding to the theory that myo-­
inositol depletion contributes to diabetic periph-
eral neuropathy a recent study has demonstrated
that mRNA and protein expression of myo-­
inositol cotransporters in the sciatic nerve are sig-
nificantly decreased in experimental diabetes
[45].
Other animal studies that lend support of the
accumulation of polyols causing slowing of
motor nerve conduction velocity are galactose
fed rats. Feeding rats a diet enriched with galac-
tose leads to a large accumulation of the polyol,
galactitol and slowing of motor nerve conduction
velocity [46]. Interestingly in this study the
authors were not able to demonstrate a decrease
in sensory nerve conduction velocity with galac-
tose intoxication. In contrast, in streptozotocin-­
induced diabetic rats both motor and sensory
nerve conduction velocity was decreased in this
study [46]. The effects of galactose intoxication
on polyol accumulation and myo-inositol deple-
M. Yorek
tion are prevented by and aldose reductase inhibi-
tor [47, 48]. Studies have also been done using an
inhibitor of sorbitol dehydrogenase the second
enzyme in the polyol pathway that converts sor-
bitol to fructose (see Fig.  1). In those studies it
has been found that blocking sorbitol’s conver-
sion to fructose exacerbated sympathetic auto-
nomic neuropathy in streptozotocin-induced
diabetic rats and Zucker diabetic rats [49, 50].
The effects of the sorbitol dehydrogenase inhibi-
tor were prevented by the addition of sorbinil.
Lastly, in mice overexpressing human aldose
reductase compared to wild type mice the induc-
tion of diabetes using streptozotocin caused a
significantly greater increase in sorbitol and fruc-
tose in peripheral nerves even though both sets
mice had comparable levels of hyperglycemia
[51]. Both the diabetic wild type and aldose
reductase transgenic mice had defective nerve
conduction velocities but this was significantly
more severe in the diabetic aldose reductase
transgenic mice. Treating these mice with an
aldose reductase inhibitor significantly prevented
the accumulation of sorbitol and slowing of nerve
conduction velocity [51]. In contrast, studies of
diabetes in mice deficient in aldose reductase
revealed significantly lower levels of sorbitol
compared to diabetic wild type mice and protec-
tion from slowing of nerve conduction velocities
[52]. Furthermore, aldose reductase deficiency in
diabetic mice significantly reduced several mark-
ers of oxidative stress that were significantly
increased in diabetic wild type mice [52].
The effect of sorbinil has also been exten-
sively studied in human subjects with diabetes.
The design of these studies varied as did the num-
ber of subjects enrolled, treatment period, dose,
duration, and endpoints examined. Many of the
clinical studies conducted with sorbinil were lim-
ited to a small number of subjects and duration of
treatment of a year or less. The results from these
studies were mixed with some studies reporting a
significant improvement in nerve conduction
velocity and axonal atrophy [53–55] while other
studies reported limited to no benefit [56–58].
The results from these trials was summarized
nicely by Pfeifer et  al. [59] stating that “future
trials should be designed with adequate statistical
Metabolic Mechanisms in Diabetic Neuropathy
power, with consideration of the variability of the
endpoint measurements for long enough dura-
tion, and with rigorous quality control to defini-
tively confirm the utility of aldose reductase
inhibitors in the treatment of diabetic distal sym-
metrical polyneuropathy and autonomic neurop-
athy.” This review was written in 1997 and the
same problems are still a challenge for adequate
clinical trials for diabetic peripheral neuropathy
in 2021.
Epalrestat is another aldose reductase inhibi-
tor that has an extensive history but unlike
sorbinil it is being used clinically for treatment of
diabetic peripheral neuropathy primarily in Japan
and recent pre-clinical and clinical studies pro-
vide evidence that it may also be beneficial for
diabetic nephropathy [60–62]. In pre-clinical
studies conducted in rats using a combination of
a high fat and high carbohydrate diet followed by
a low dose of streptozotocin, a model for type 2
diabetes, it was found that epalrestat treatment
protected the diabetic rats from peripheral neu-
ropathy through inhibition of the polyol pathway
and by alleviating oxidative stress [63]. Results
from several clinical studies report that epalrestat
treatment of 150 mg/day may improve motor and
sensory nerve conduction velocity and subjective
neuropathy symptoms with minimal side effects
[62, 64].
In collaboration with the Obrosova laboratory
we have also demonstrated that inhibition of
aldose reductase with fidarestat of type 1 diabetic
rats attenuates oxidative-nitrosative stress and
activation of poly(ADP-ribose) polymerase while
improving multiple endpoints associated with
diabetic peripheral neuropathy [65]. This ener-
gized studies in my laboratory to examine the
effect of the combination of α-lipoic acid, an
antioxidant, and fidarestat on vascular and neural
complications in a type 1 diabetic rat model [66].
The results from this study demonstrated that the
combination therapy of α-lipoic acid and fidares-
tat was more efficacious in preventing diabetes-­
induced vascular and neural dysfunction than
monotherapy that required higher doses to be
equally effective. Our studies attributed this to
the combination therapy allowing for the
increased conversion of α-lipoic acid to the more
257
effective antioxidant dihydrolipoic acid, which
provides a greater protection from oxidative
damage than does glutathione (GSH) (Fig.  2
[67]). Fidarestat blocked aldose reductase activ-
ity thereby conserving NADPH levels, which
were then available for the production of dihy-
drolipoic acid whose availability is important for
the production of GSH [67]. Interestingly, several
groups have reported that the combination of
α-lipoic acid and epalrestat is better than mono-
therapy clinically for diabetic peripheral neurop-
athy and improvement of motor and sensory
nerve conduction velocity [68, 69].
Ranirestat is one of the more recent aldose
reductase inhibitors to be tested pre-clinically and
clinically for diabetic peripheral neuropathy.
Ranirestat is an uncompetitive/reversible inhibitor
of aldose reductase [70]. In long-term studies
with streptozotocin-diabetic rats ranirestat
reduced sorbitol accumulation in the sciatic nerve
and improved the decrease in motor nerve con-
duction velocity [71]. Treatment with ranirestat
also prevented the deformity of myelinated fibers
and the decrease in their axonal and myelin areas
(atrophy) in sural nerves as well as the changes in
the size frequency histogram of myelinated fibers
[70]. In another independent study with a study
design more relevant to clinical practice ranirestat
treatment was started 12 weeks after the onset of
hyperglycemia using streptozotocin-treated rats.
At the time of treatment both motor and sensory
nerve conductions were decreased and the
untreated diabetic rats were hypoalgesic in
response to a thermal stimulus [72]. Following
only 6 weeks of treatment both motor and sensory
nerve conduction was improved as was the foot
withdrawal latency. Ranirestat treatment also
improved the intraepidermal nerve fiber density.
The authors of this study concluded that ranirestat
has the potential for regeneration in the peripheral
nervous system [72]. Ota et  al. [73] conducted
studies evaluating the effect of ranirestat com-
pared to epalrestat treatment on peripheral neu-
ropathy and cataract formation in spontaneously
diabetic torii rats. In this study they found that
ranirestat and epalrestat prevented diabetic neu-
ropathy but only ranirestat prevented cataract for-
mation. There have been several clinical studies
258
Fig. 2  Lipoic acid (LA) and dihydrolipoic acid (DHLA) metabolism (modified from [67])
performed using ranirestat. Sekiguchi et  al. [74]
found that 52 weeks of treatment with ranirestat
(40  mg/day) was well tolerated and improved
nerve conduction velocity but was not able to
detect any improvement in symptoms and signs.
In two separate studies led by Dr. Vera Bril with
ranirestat treatment for up to 52–60  weeks with
dosing at 20–40  mg/day it was found treatment
improved motor nerve conduction velocity and
sensory nerve conduction velocity in one study
but not the other [74, 75]. In both studies ranires-
tat was reported to be well tolerated with no dif-
ferences compared to placebo in adverse events.
Ranirestat is the furthest advanced aldose reduc-
tase inhibitor for clinical trials except for epalres-
tat and reproducibly exhibits some degree of
measurable objective beneficial outcomes [76].
Its favorable safety profile makes it an attractive
choice for further exploration.
3 Non-enzymatic Glycation
Another common pathway that was recognized
early to contribute to diabetic peripheral neurop-
athy was non-enzymatic glycation and the forma-
tion of advanced glycation endproducts.
Advanced glycation endproducts are created
from non-enzymatic reactions of reducing sugars
M. Yorek
such as glucose or fructose with free amino
groups of proteins, lipids, or nucleic acids to ini-
tially form Schiff bases or Amadori adducts that
in the early stages are reversible (Fig. 3) [77–79].
Overtime these products continue to undergo
reactions that include dehydration, fragmenta-
tion, and cross-linking to form irreversible
advanced glycation endproducts [77]. The accu-
mulation of these compounds is associated with
many disease states including diabetes. Their
accumulation causes structural damage to tissues
and organs including components of peripheral
nerves such as Schwann cells and cytoskeletal
proteins; tubulin, neurofilaments, and actin [80,
81]. The accumulating damage to these tissues
culminates in abnormal nerve function including
slowed axonal transport, atrophy and degenera-
tion, and slowing of nerve conduction velocity
[77, 82]. Advanced glycation endproducts can
also indirectly affect peripheral nerves by alter-
ing vascular structure and function and ultimately
affecting blood flow causing localized ischemia.
Advanced glycation endproducts can also elicit
their effects through binding of the receptor for
advanced glycation endproducts (RAGE) [83].
RAGE is expressed in endothelial cells and
Schwann cells [84]. Activation of RAGE has
been shown to stimulate NF-κB and increase oxi-
dative stress [85–87]. Advanced glycation end-
Metabolic Mechanisms in Diabetic Neuropathy
Fig. 3  Pathogenic mechanisms associated with non-enzymatic glycation (modified from [77])
products have also been shown to induce
basement membrane hypertrophy in endoneurial
microvessels and disrupt the blood-nerve barrier
[88]. In epineurial arterioles, resistance size
blood vessels that provide circulation to the sci-
atic nerve, endothelium-dependent vascular
relaxation is mediated in part by acetylcholine
via nitric oxide generating mechanism important
for regulating endoneurial blood flow [89]. We
and others have demonstrated that endoneurial
blood flow is improved with inhibitors of
advanced glycation endproduct formation as well
as oxidative stress; thereby linking these two
potential mechanisms of diabetes vascular and
neural complications [43, 90]. Studies with mice
deficient in RAGE and use of a competitive decoy
for advanced glycation endproducts, soluble
RAGE (sRAGE), have contributed to our under-
standing of the role of advanced glycation end-
products may have in the development of diabetic
peripheral neuropathy [91–93]. In diabetic
RAGE-null mice nerve conduction velocity was
improved as was nerve generation compared to
diabetic wild type mice [94, 95]. Also, treating
259
diabetic wild type mice with sRAGE with the
intent to sequester RAGE ligands also improved
diabetic peripheral neuropathy [95]. Interestingly,
it has been shown that RAGE is expressed in
about 30% of all nerve fibers in normal, healthy
human subjects, and expression is higher in those
subjects with peripheral neuropathy [96].
The studies above imply that molecules that
may inhibit the action of advanced glycation end-
products may be effective approach for treatment
of diabetic peripheral neuropathy. One of the ear-
lier compounds used to inhibit the effects of
advanced glycation endproducts was aminogua-
nidine. Several laboratories including my own
have shown that treating diabetic rats with ami-
noguanidine improve peripheral neuropathy with
mechanisms contributing to inhibition of free
radical formation and improvement in vascular
function [22, 43, 97–101]. Pyridoxamine and
analogues have also been shown to be effective
inhibitors of advanced glycation endproduct for-
mation and treatment for diabetes complications
including diabetic peripheral neuropathy [90,
102, 103].
260 M. Yorek

Much like the study of the polyol pathway in Protein kinase C (PKC) is another potential tar-
animal models of diabetes there is considerable get for diabetic peripheral neuropathy treatment.
support from pre-clinical studies with diabetic Protein kinase C comprises a superfamily of iso-
rodents for a pathologic role of advanced glyca- enzymes, many of which are activated by 1,2-dia-
tion endproducts in the development and progres- cylglycerol (DAG) [111]. Increased PKC activity
sion of diabetic peripheral neuropathy. is another pathogenic mechanism that has been
Unfortunately, there have been few clinical trials linked to diabetic vascular disease via animal stud-
testing the effects of inhibitors of advanced gly- ies using PKC inhibitors. In the vasculature
cation endproducts on diabetic peripheral increased production of diacylglycerol (DAG)
neuropathy. leads to activation of PKC isoforms (α, β1 and 2,
δ) [111]. Increased PKC activity has been associ-
ated with changes in blood flow, basement mem-
4 Hexosamine and Protein brane thickening, extracellular matrix expansion,
Kinase C increased vascular permeability, and changes in
enzyme activities including Na+-K+ ATPase, phos-
The hexosamine pathway contributes to an phatidylinositol 3-kinase, and mitogen activated
increase in oxidative stress and has been impli- protein kinase [111]. In the nerve DAG levels are
cated in the pathogenesis of diabetes complica- decreased with diabetes and no consistent change
tions including diabetic peripheral neuropathy in activity of PKC isoforms has been observed
[104]. Under conditions of elevated intracellular [111, 112]. Nonetheless, inhibition of PKC β of
glucose and flux into glycolysis, excess, fructose-­ diabetic rats has been reported to correct reduced
6-­phosphate is diverted from glycolysis to pro- nerve blood flow and slowing of nerve conduction
duce glucosamine-6-phosphate, which in turn velocity suggesting a major role of the neurovas-
converts into uridine diphosphate-N-acetyl glu- culature in diabetic peripheral neuropathy [111,
cosamine, a substrate for the formation of proteo- 113–115]. This is further supported by studies
glycans and other glycoproteins [105]. The showing that treatment of diabetic rats with a
diphosphate-N-acetyl glucosamine can also PKCβ inhibitor also improved vascular dysfunc-
increase glycosylation of proteins such as endo- tion of mesenteric arteries to acetylcholine [115].
thelial nitric oxide synthase causing impaired Furthermore, the benefits of PKC inhibition on
enzyme activity, reduced formation of nitric oxide nerve blood flow and nerve conduction velocity
and vascular dysfunction [106]. It is not entirely were attenuated by a nitric oxide synthase inhibi-
clear what kinds of peripheral nerve proteins may tor [116]. Studies of diabetic peripheral neuropa-
be modified by the hexosamine pathway during thy in human subjects treated with a PKC β
diabetes, although there is emerging evidence that inhibitor, ruboxistaurin, for 6  months to 1  year
imbalances in the glycolated state of nervous sys- reported mixed effects with improvement in skin
tem proteins is associated with neurodegenerative microvascular blood flow in diabetes subjects with
disease [107]. One of the more promising treat- neuropathy but limited improvement in sensory
ments based on animal studies of the hexosamine symptoms, vibration detection threshold and
pathway is benfotiamine, a fat-soluble derivative Neuropathy Total Symptom Score-6 [117–119].
of thiamine [108]. In diabetic rats benfotiamine
was found to relieve inflammatory and neuro-
pathic pain [109]. In a phase III trial clinical trial 5 Oxidative and Nitrosative
of 165 patients with symmetrical, distal diabetic Stress
polyneuropathy benfotiamine treatment for
6 weeks reported a significant improvement in the Oxidative stress is a condition resulting from an
Neuropathy Symptom Score with best results imbalance between the generation of reactive
obtained with the high dose according to pain oxygen species (ROS) and the ability of antioxi-
patient-reported symptoms [110]. dant mechanisms to neutralize these compounds
Metabolic Mechanisms in Diabetic Neuropathy
[120]. The most common forms of ROS are
superoxide (O2−), hydrogen peroxide (H2O2),
hydroxyl radical (OH−), and peroxynitrite
(ONOO−) [121]. Enzymes and pathways located
throughout the cell, including the plasma
membrane, cytosol, mitochondria, and peroxi-
­ guanidine and transition metal chelators, or
somes have the ability to generate these com-
pounds under both normal and pathological
conditions [121]. Superoxide is the most biologi-
cally important ROS. It can be produced by the
electron transport chain of the mitochondria, and
by NADH oxidase, NAD(P)H oxidase, xanthine
oxidase, cyclooxygenase, lipoxygenase, cyto-
chrome P-450, and, during periods of tetrahydro-
biopterin deficiency, by nitric oxide synthase
[121]. Superoxide can spontaneously acquire an
electron to form H2O2. The formation of H2O2
from O2− can also occur via a reaction catalyzed
by superoxide dismutase (SOD) of which there is
three isoforms: Mn-SOD, which is located in the
mitochondria and two isoforms of Cu,Zn-SOD,
which are located either in the cytosol or extra-
cellularly [121]. Hydrogen peroxide can be con-
verted to water by the action of catalase or by
glutathione peroxidase in the presence of reduced
glutathione [121]. However, in the presence of
trace metals such as Fe, H2O2 can form OH− via a
process known as the Fenton reaction [121]. The
formation of ONOO−, which is the result of a
reaction between O2− and NO, is pathologically
important in vascular disease and has been dem-
onstrated to be enhanced in diabetes [121, 122].
It is ONOO− that is responsible for much of the
cytotoxicity of nitric oxide (NO) that contributes
to nitrosative stress [123]. Peroxynitrite has a
short half-life but is able to diffuse across cell
membranes and depending on the cell environ-
ment can cause a nitrosylation of proteins, which
generally reduces enzyme activity, oxidation of
glutathione, an important antioxidant, and
increased peroxidation of lipids [123].
As discussed above, endothelial dysfunction
contributes significantly to diabetic vascular dis-
ease, which is an important factor in the develop-
ment of diabetic peripheral neuropathy. This is
supported by studies such as Cameron and Cotter
who demonstrated that reduced nerve perfusion
is a contributing factor in the etiology of diabetic
261
peripheral neuropathy [124]. Free radicals such
as O2− and OH− cause vascular endothelial dam-
age and reduced NO-mediated vasodilation.
Inhibition of advanced glycosylation and autoxi-
dation, major sources of free radicals, by amino-
antioxidants and free radical scavengers have
been demonstrated to improve the diabetes-­
induced decrease in endoneurial blood flow and
improve neural dysfunction such as slowed nerve
conduction velocity [3, 124–132]. The endothe-
lium, via the release of vasodilators and vasocon-
strictors, controls the vascular tone. The three
major factors produced by the endothelium that
contribute to the regulation of vascular relaxation
are NO, prostacyclin and the as yet an unidenti-
fied factor referred to as endothelium-derived
hyperpolarizing factor (EDHF). Impaired
endothelium-­ dependent vasodilation has been
demonstrated in various vascular beds of animal
models of diabetes and humans with type 1 and
type 2 diabetes [133, 134]. The mechanisms
induced by hyperglycemia/diabetes considered to
contribute to endothelial dysfunction as dis-
cussed above are the activation of PKC, increased
activity of the polyol pathway, increased forma-
tion of advanced glycation endproducts, and
increased oxidative stress. Interestingly, studies
by Brownlee and colleagues have suggested
hyperglycemia-induced production of O2− by
mitochondria of endothelial cells as the common
link for mechanisms of diabetes-induced vascu-
lar dysfunction [135, 136]. Fernyhough has
described mitochondrial dysfunction in diabetic
neuropathy as a “series of unfortunate metabolic
events” [137]. Other chapters in this publication
will review the impact of mitochondrial dysfunc-
tion, nutrient stress and loss of insulin-dependent
growth factor support. Our studies conducted
with intact vascular tissue consisting of epineu-
rial arterioles of the sciatic nerve lend support to
the studies by Brownlee and colleagues con-
ducted with cultured endothelial cells [138].
Studies by my laboratory have provided evidence
that the generation of oxidative stress through the
production of O2− and ONOO− impairs vascular
function and endothelium-dependent vascular
relaxation of epineurial arterioles of the sciatic
262
nerve from diabetic rats, which precedes the
slowing of motor nerve conduction velocity [43,
44, 125, 139]. Studies designed to investigate the
source of O2− formation provided results suggest-
ing that complex I of the mitochondrial electron
transport chain and possibly NAD(P)H oxidase
are responsible for the increase in O2− formation
observed with epineurial arterioles from the sci-
atic nerve [138]. It was shown that pretreating
epineurial arterioles from diabetic rats with the
PKC inhibitor bisindolylmaleimide (GF
109203X) improved acetylcholine-mediated vas-
cular relaxation but did not prevent the increase
in O2− formation suggesting that activation of
PKC by oxidative stress is downstream of O2−
formation [138]. We have also demonstrated that
treating diabetic rats with three different types of
antioxidants prevented the diabetes-induced
increase in O2− and ONOO− formation in aorta
and epineurial arterioles of the sciatic nerve and
diabetes-induced vascular and neural dysfunc-
tion, thereby providing additional evidence that
increased oxidative stress contributes to diabetes-­
induced vascular and neural disease [125, 126].
Studies from other laboratories have provided
further evidence that antioxidants may prevent
vascular complications in diabetes. Treating dia-
betic rats with tempol, a stable superoxide dis-
mutase mimic compound, abolished the
diabetes-induced increase in vascular O2−, malo-
ndialdehyde and 8-epi-prostaglandin F(2α), and
also the impairment in relaxation of aortic rings
to acetylcholine [125]. Cameron and colleagues
have demonstrated that treating diabetic rats with
α-lipoic acid or the metal chelators hydroxyethyl
starch deferoxamine or trientine prevented the
diabetes-induced impairment in vascular relax-
ation associated with hyperalgesia and neurovas-
cular deficits [128, 129, 140–142]. In addition,
Keegan et al. demonstrated that treating diabetic
rats with α-lipoic acid improved endothelium-­
dependent vascular relaxation of corpus caverno-
sum smooth muscle [131]. In another study we
demonstrated that diabetic peripheral neuropathy
in a rat model of type 2 diabetes was improved by
treatment with mitoquinone (Mito-Q) but inter-
estingly in this study treatment did not improve
vascular reactivity by epineurial arterioles to ace-
M. Yorek
tylcholine [143]. Because metals chelators and
OH− scavengers have also been demonstrated to
be effective in preventing diabetes-induced vas-
cular and neural dysfunction it is likely that the
formation of OH− may also contribute to impair-
ment of vascular reactivity and nerve function in
diabetes [124, 128, 129, 132, 140, 141, 144].
As discussed above in regard to improving
diabetes impaired vascular function prevention of
oxidative stress is a promising approach for inter-
vention and halting of diabetic peripheral neu-
ropathy. A variety of antioxidants including
vitamin E have been demonstrated to have bene-
ficial effects in treating neuropathy in diabetes
patients and diabetic animal models [145–147].
More recently α-lipoic acid has shown promise
as a potential antioxidant treatment for diabetic
neuropathy [148, 149]. Our studies have demon-
strated that α-lipoic acid provides good protec-
tion against oxidative stress in diabetic rats of
4–6  week duration [125]. The treatment of dia-
betic rats with α-lipoic acid significantly
improved diabetes-induced decrease in endoneu-
rial blood flow, endothelium-dependent vascular
relaxation in arterioles that provide circulation to
the sciatic nerve, and motor nerve conduction
velocity. α-Lipoic acid treatment also reduced the
production of O2− by the aorta and O2− and
ONOO− by epineurial arterioles. Treating dia-
betic rats with α-lipoic acid prevented the
diabetes-­induced increase in thiobarbituric acid
reactive substances in serum and significantly
improved lens glutathione levels. α-Lipoic acid is
a good metal chelator and is capable of scaveng-
ing hydroxyl radicals, hypochlorous acid and sin-
glet oxygen, but not O2− or peroxyl radicals
[149–152]. However, in its reduced form, as
dihydrolipoic acid, it is a good scavenger of O2−
and prevents initiation of lipid peroxidation
[149–152]. In vivo, α-lipoic acid can be con-
verted into dihydrolipoic acid [149, 150] (Fig. 2).
This reaction requires NADPH, which is reduced
in diabetes due to the increased flux of glucose
through the aldose reductase pathway [120, 153].
Therefore, one potential form of combination
therapy for the treatment of diabetic neuropathy
may be combining an aldose reductase inhibitor
with α-lipoic acid [66]. This combination should
Metabolic Mechanisms in Diabetic Neuropathy 263

promote the formation of dihydrolipoic acid, smaller molecule (molecular weight 483 vs.
thereby enhancing the antioxidant potential of 30,000 for M40403 and the native enzyme,
α-lipoic acid and possibly providing a synergistic respectively) [126, 158, 159].
effect. In a study by Nakamura et al. of diabetic Another pathway for targeting as a potential
neuropathy in streptozotocin-induced diabetic treatment for diabetic peripheral neuropathy is
rats, they found that treating diabetic rats with the nitrosative stress and the activation of poly(ADP-­
aldose reductase inhibitor NZ-314 improved ribose) polymerase (PARP) [160, 161]. Studies
nerve function and reduced oxidative stress using peroxynitrite decomposition catalysts as
[154]. They concluded that the efficacious effect treatments of diabetic rodents have shown
of aldose reductase inhibition on diabetic neu- improvement in nerve conduction velocity as
ropathy may be mediated by decreasing oxygen well as other sensory related neuropathic deficits
free radicals. This would agree with our studies and improvement of vascular relaxation by epi-
described above demonstrating that treating dia- neurial, coronary, and mesenteric arterioles
betic rats with fidarestat and α-lipoic acid was a [162–165]. These treatments were also shown to
better treatment for diabetic peripheral neuropa- reduce nitrotyrosine and PARP immunofluores-
thy than either monotherapy [66]. Efficacy of cence in the sciatic nerve and dorsal root gan-
α-lipoic acid for diabetic peripheral neuropathy glion neurons [165]. Others studies have
has also been examined in human subjects. In a demonstrated that the PARP inhibitor
review article by Papanas and Ziegler [155] that 3-­aminobenzamide reduced nitrotyrosine immu-
examined placebo-controlled clinical trials and noreactivity in vascular tissue and improved vas-
meta-analyses they concluded that there is evi- cular reactivity of epineurial arterioles and neural
dence supporting α-lipoic acid as an efficacious function [166–168]. Studies from my laboratory
and safe treatment for diabetic neuropathy. that lend support to PARP activation and meta-
The development of superoxide dismutase bolic stress contributing to diabetic peripheral
mimetics are another class of antioxidants with neuropathy have shown that treating dietary
potential for treatment of diabetes complications obese or type 2 diabetic mice with nicotinamide
including diabetic peripheral neuropathy [126, riboside improved multiple endpoints associated
156, 157]. Because of limitations associated with with peripheral neuropathy [169].
enzyme therapies these non-peptidyl compounds
may offer advantages resulting in better clinical
therapies and outcomes for diseases mediated by 6 Additional Mechanisms
O2− radicals such as diabetes [158]. In our studies and Treatment Options
we demonstrated that treating diabetic rats with
M40403 inhibited the generation of O2− by aorta Other pathways and downstream mediators of
and epineurial vessels of the sciatic nerve, the pathology associated with diabetic peripheral
formation of ONOO− by epineurial vessels of the neuropathy include mitogen activated protein
sciatic nerve, the reduction in endoneurial blood kinase (MAPK), nuclear factor κB (NF-κB),
flow, the slowing of motor nerve conduction cyclooxygenase 2 (COX-2), 12/15-lipoxygenase
velocity and impairment of endothelium-­ (12/15-LOX), and Na+/H+-exchanger-1 (NHE-1).
dependent vasodilation of epineurial arterioles. It These downstream mediators of neural impair-
also improved the diabetes induced increase in ment have in common their activation via oxida-
serum TBARS and sciatic nerve conjugated diene tive/inflammatory stress. Activation of the MAPK
level, two additional markers of oxidative stress pathway has been documented in neuron and
[126]. M40403 is a prototypic example of a sta- Schwann cells by hyperglycemia as well as in
ble, low molecular weight, manganese-­sciatic nerve of diabetic rodents and human sub-
containing, non-peptidic molecule possessing the jects with type 1 and type 2 diabetes [170, 171].
function and catalytic rate of native SOD Different treatment modalities have been
enzymes, but with the advantage of being a much employed in pre-clinical studies for treatment of
264
diabetic peripheral neuropathy showing improve-
ment in neural endpoints as well as reduction in
activation of MAPK [172–174]. These studies
indicate a potential therapeutic approach for
treatment of diabetic peripheral neuropathy
through inhibition of MAPK pathway.
Activation of PARP and MAPK can contrib-
ute to regulation of gene expression through acti-
vation of NF-κB [175]. Nuclear factor-2 erythroid
related factor-2 (Nrf2) and NF-κB pathways are
potential therapeutic targets for diabetic neuropa-
thy [176]. Pre-clinical studies therapeutically tar-
geting Nrf2 and NF-κB have demonstrated by
inhibiting NF-κB and/or activating Nrf2 can
reduce inflammation and oxidative stress and
improve endpoints related to diabetic peripheral
neuropathy [177–182].
Other potential targets for treatment of dia-
betic peripheral neuropathy include inhibition of
Na+/K+ exchangers [183]. Treating streptozotocin-­
induced diabetic rats with cariporide reversed
multiple endpoints associated with diabetic
peripheral neuropathy and interestingly also
reduced markers linked to increased oxidative
and nitrosative stress and advanced glycation
endproducts suggesting that the pathology of
activation of Na+/K+ exchangers in diabetes
occurs upstream of these common mediators of
diabetic peripheral neuropathy.
Impairment in lipid metabolism is another
potential therapeutic target for treatment of dia-
betic peripheral neuropathy [184].
Cyclooxygenase-2 pathway is upregulated in the
peripheral nerves and dorsal root ganglia neurons
in experimental diabetes and chemical inhibition
or gene deletion improves peripheral nerve func-
tion including nerve conduction velocity and loss
of epidermal nerve fibers while reducing oxida-
tive stress and inflammation [185, 186]. Treating
diabetic mice with the cyclooxygenase-2 inhibi-
tor meloxicam has been shown to reduce neuro-
pathic pain [187]. Other anti-inflammatory or
anti-oxidative stress reagents thymoquinone and
sulforaphane have been shown to improve in pre-­
clinical studies diabetic peripheral neuropathy
while reducing cyclooxygenase-2 expression in
Schwann cells or sciatic nerves [188, 189].
12/15-Lipoxygenase overexpression is evident in
M. Yorek
the sciatic nerve and spinal cord of diabetic
rodents [190]. Pharmacological inhibition of
12/15-lipoxygenase or gene deficiency has been
shown to alleviate some but not all pathological
changes associated with diabetic peripheral neu-
ropathy of large and small fibers while reducing
MAPK activation and relieving oxidative and
inflammatory stress [191–193].
In recent years my laboratory has focused on
two additional pathways of intervention as a ther-
apeutic target for diabetic peripheral neuropathy.
In the next two sections I will briefly review our
studies of the effect of inhibition of neutral endo-
peptidase or neprilysin and nutritional supple-
mentation of omega-3 polyunsaturated fatty acids
(PUFA) primarily found in fish oil eicosapentae-
noic acid (EPA) and docosahexaenoic acid
(DHA) and their metabolites. I have previously
provided reviews for both of these topics [194,
195].
Angiotensin converting enzyme (ACE),
dipeptidyl peptidase IV (DPP-IV), and neprilysin
(neutral endopeptidase) are all proteases. As a
group these proteases can contribute to disease
conditions through the breakdown of important
vasoactive and neuroprotective peptides as well
as other peptides such as glucagon-like peptide 1.
DPP-IV inhibitors (gliptins) are a class of drugs
approved by the FDA for hyperglycemia-related
treatment of type 2 diabetes [196]. Pre-clinical
studies have demonstrated that dipeptidyl pepti-
dase IV inhibitors are also an effective treatment
for diabetic peripheral neuropathy [197–200].
The renin-angiotensin system (RAS) plays a
major role in the pathophysiology of cardiovas-
cular and renal disease [201–203]. Most known
effects of angiotensin II are mediated via activa-
tion of the AT1-receptor [202]. Activation of the
AT1-receptor is involved in vasoconstriction,
inactivation of bradykinin, water and salt homeo-
stasis, control of neuro-humoral systems, reac-
tive oxygen species production, cellular
hypertrophy, and hyperplasia and apoptosis
[203]. ACE inhibitors and angiotensin receptor
blockers are well-known treatments for hyperten-
sion, atherosclerosis, congestive heart failure,
stroke, and myocardial infarction including those
patients with diabetes [201, 204]. In diabetic ani-
Metabolic Mechanisms in Diabetic Neuropathy
mal models we and others have demonstrated
that ACE inhibitors and to a lesser extent angio-
tensin receptor blockers can alleviate many defi-
cits associated with diabetic peripheral
neuropathy as well as vascular dysfunction of
epineurial arterioles of the sciatic nerve [205–
210]. In addition to pre-clinical studies in animal
models of diabetes there are also results from
three small clinical trials demonstrating that dia-
betic neuropathy can be improved by treatment
of patients with quinapril, trandolapril or lisino-
pril [211–213].
The other lesser known protease that is
included in this discussion is neprilysin.
Neprilysin is found in many tissues including
vascular and renal tissue and its activity is
increased by fatty acids and glucose in human
microvascular cells [214–217]. Interestingly,
neprilysin activity has been shown to be activated
by protein kinase C, which is increased in vascu-
lar tissues by diabetes including endothelial cells
[218, 219]. Importantly for this discussion nepri-
lysin degrades vasoactive neuropeptides includ-
ing; natriuretic peptides, adrenomedullin,
bradykinin, endothelin, glucagon-like peptide 1,
substance P, and calcitonin gene-related peptide
[220, 221]. Thus, inhibition of this protease pro-
vides great potential for the treatment of diabetes
complications. To test this question several phar-
maceutical companies created what was termed
vasopeptidase inhibitors. At the time these inhibi-
tors were a new class of drug designed to simul-
taneously inhibit neprilysin and ACE activity
[222]. They were first developed for treating
hypertension and heart failure [223, 224]. In
regard to diabetes complications it has been
shown that vascular conductance in the femoral
artery of streptozotocin-induced diabetic rats was
improved by a vasopeptidase inhibitor [225].
Furthermore, it has been demonstrated that vaso-
peptidase inhibitors are neuroprotective and pre-
vent nephropathy in ZDF rats [226, 227].
Vasopeptidase inhibitors have also been reported
to decrease matrix metalloproteinases and AGE
accumulation/formation in type 2 diabetes and
improve wound healing [228, 229]. However,
there was no information available about the
potential benefits of vasopeptidase inhibitors in
265
diabetic peripheral neuropathy until we per-
formed our studies. Using the vasopeptidase
inhibitor ilepatril (AVE7688) we demonstrated in
both pre-diabetic and type 1 or type 2 diabetic
mice and rats that the combined inhibition of
ACE and neprilysin was more effective than
monotherapy in improving neurovascular func-
tion and slowing/reversing motor and sensory
peripheral nerve deficits including diabetes-­
induced decrease in innervation and sensitivity of
corneal nerves [207, 230–234]. Furthermore, we
demonstrated that deletion of the neprilysin gene
protected mice from the development of neuro-
pathologic deficits and loss of corneal nerves
upon induction of diabetes [235]. However,
despite greater efficacy for hypertension than
monotherapy, development of vasopeptidase
inhibitors was halted due to significant off-target
effects in some cohorts, most notably increased
frequency of angioedema in hypertensive sub-
jects [236]. This led to the development of
LCZ696 by Novartis a combination of sacubitril
(neprilysin inhibitor) and valsartan (angiotensin
receptor blocker) (Entresto). In 2015 the FDA
approved the use of sacubitril/valsartan in
patients with heart failure with reduced ejection
fraction [237]. We obtained this drug to study its
effect on vascular and neural complications in a
rat model of type 2 diabetes [238]. Using an early
and late intervention study design the results
demonstrated efficacy of sacubitril/valsartan in
improving vascular and neural function that was
superior to valsartan alone. In the early interven-
tion protocol, sacubitril/valsartan treatment was
found to slow progression of these deficits and,
with late intervention treatment, was found to
stimulate restoration of vascular reactivity, motor
and sensory nerve conduction velocities, and sen-
sitivity/regeneration of sensory nerves of the skin
and cornea. We concluded that sacubitril/valsar-
tan may be an effective treatment for diabetic
peripheral neuropathy and due to its safety profile
and FDA approval could advance quickly to clin-
ical trials.
Recently we have also been interested in the
potential for omega-3 PUFA as a treatment for
diabetic peripheral neuropathy. Since inflamma-
tory stress is considered to be a primary mecha-
266 M. Yorek

nism for diabetic peripheral neuropathy and
omega-3 PUFA and their metabolites are known
to have anti-inflammatory properties it was rea-
sonable to expect that fish oil, a common source
for omega-3 PUFA, may be an effective treat-
ment [239, 240]. Although trials using treatment
with omega-3 PUFA have primarily focused on
cardiovascular disease risk reduction [241–246],
there is significant evidence of their benefits
beyond cardiovascular disease. For instance, sev-
eral recent studies have reported potential bene-
fits of omega-3 PUFA consumption on several
chronic inflammatory and autoimmune diseases,
stroke, muscle atrophy, and neurodegenerative
disease [247–254]. Nonetheless, the consump-
tion of omega-3 PUFA remains low in the
Western diet due to historically increased con-
sumption of omega-6 PUFA.
In our own most recent pre-clinical studies, we
found that long-chain omega-3 PUFA, primarily
EPA and DHA derived from fish (menhaden) oil,
were effective in reversing impaired nerve conduc-
tion velocities and nerve fiber density and sensitiv-
ity in the skin and cornea [255–257]. Furthermore,
our studies have demonstrated that E and D series
resolvins (resolution-phase interaction products),
metabolites of EPA and DHA, respectively, (see
Fig.  4) which are known to have anti-inflamma-
tory and neuroprotective properties, may be the
active mediator of the EPA and DHA effects on
diabetic peripheral neuropathy and nerve regener-
ation [255, 258, 259]. Thus, optimizing production
of these metabolites may be an effective approach
to increasing the efficacy of EPA and DHA sup-
plements in the diet, and thus ensuring a better out-
come for diabetic peripheral neuropathy. In that
regard we have demonstrated that the combination
of menhaden oil and salsalate, a highly effective
agent in blocking proinflammatory chemokines
and cytokines, with a large margin of safety and
low costs [260, 261] increase production of resol-
vin D1 in  vivo vs. menhaden oil alone and this

Elcosapentaenoic acid Docosahexaenoic acid

(EPA, 20:5n-3) (DHA, 22:6n-3)
COOH
COOH

3 3
COX-2/Aspirin CX-2/Aspirin 12-LOX
CYP450 15-lipoxgenase

Maresins
18-HEPE
Maresins 1 OH

5-LOX 15-LOX
COOH
Resolvin
E series
OH
Resolvin E1 Resolvin E3
Protectins
OH OH Protectins D1
COOH
COOH
5-LOX COOH

OH OH OH HO OH
Resolvin D series
Resolvin E2
Resolvin D1
OH
HO OH
COOH
COOH

OH
OH
Metabolism of EPA and DHA

Fig. 4  Metabolism of EPA and DHA

Metabolic Mechanisms in Diabetic Neuropathy
combination is more efficacious toward improving
peripheral neuropathy in diabetic rodents [259,
262]. Another important evidence, supporting
future clinical trials of omega-3 PUFA is that our
pre-­clinical studies have found that the circulating
levels of EPA and DHA in rats receiving menha-
den oil was comparable to the EPA and DHA
serum levels of human subjects taking 4  g of
EPA + DHA per day (fish oil capsules) for 4 weeks
[257, 263]. For instance, the omega-3 PUFA levels
in serum increased from 5.6% to 14.4% and the
EPA to AA ratio increased from 0.12 to 0.88  in
human participants treated with 4 g of fish oil for
4 weeks [263]. This is similar to the 3.9% to 15.4%
increase in the serum omega-3 PUFA levels and
0.04 to 0.54 increase in the EPA to AA ratio we
found in our study of diabetic rats treated with
menhaden oil [256]. Thus, increasing the circula-
tory levels of EPA and DHA, and favorably alter-
ing the omega-3 index to a healthy range of 8–12%
could be an effective disease modifying therapy
for treating diabetic peripheral neuropathy [264].
In humans, a proof-of-concept study that
tested the effects of 12  months of ~2  g daily
omega-3 PUFA supplementation on the progres-
sion of diabetic peripheral neuropathy in patients
with type 1 diabetes, reported a 29% significant
increase in corneal nerve fiber length (primary
outcome), but no effect on sensory nerve conduc-
tion velocity [265]. Unfortunately, it is unclear if
omega-3 PUFA therapeutic levels were achieved
in these participants given that serum omega-3
index or omega-3 PUFA metabolites were not
assessed. However, the above change in corneal
fiber length was comparable to the increase in
corneal fiber length we observed in type 1 diabe-
tes rats treated with menhaden oil and provides
encouraging information to support future clini-
cal trials using omega-3 PUFA and salsalate as an
intervention for diabetic peripheral neuropathy
[256].
7 Conclusion
Over the years our knowledge and understanding
of the different mechanisms that contribute to
the etiology of diabetic peripheral neuropathy
267
has expanded considerably. However, this infor-
mation has not provided us with a successful
treatment. There are numerous reasons why this
may be the case with some of this being covered
in other chapters. From my point of view some
of the reasons that may factor into these past fail-
ures are that animals are not suitable models for
diabetic neuropathy as it occurs in humans,
many of the human studies that have been con-
ducted to date are flawed with patients being to
advance in disease, treatment periods to short
and endpoints not adequate to detect early status
of disease and reversibility. However, a primary
reason I believe for the failure of finding an
effective treatment for diabetic peripheral neu-
ropathy is the monotherapy approach that has
been taken for many of the studies. It has been
repeatedly demonstrated and reported that the
etiology of diabetic peripheral neuropathy is
complex with many mechanisms, some dis-
cussed in this chapter as well as in other chap-
ters, contributing to the development and
progression of the disease. Our cardiology col-
leagues routinely treat their patients with hyper-
tension by advising them to consider lifestyle
and dietary changes as well as prescribing sev-
eral different drugs that target the different
potential mechanisms responsible for high blood
pressure. Thus, it seems irrational to think that
anything other than a combination of lifestyle
changes that include increased exercise and bet-
ter dietary choices and a combination of inter-
ventional drugs will be needed to successfully
treat diabetic peripheral neuropathy. I have pro-
vided you evidence of this from different animal
studies in this chapter. What investigators and
clinicians need to be asking are what the primary
targets are for successfully treating diabetic
peripheral neuropathy and what the best drugs to
use for these targets are?
References
1. Zenker J, Ziegler D, Chrast R.  Novel pathogenic
pathways in diabetic neuropathy. Trends Neurosci.
2013;36:439–49.
2. Vinik A.  Clinical practice. Diabetic sensory and
motor neuropathy. N Engl J Med. 2016;374:1455–64.
268
3. Pop-Busui R, Boulton A, Feldman E, et  al.
Diabetic neuropathy: a position statement by the
American Diabetes Association. Diabetes Care.
2017;40:136–54.
4. Perkins B, Bril V.  Electrophysiology test-
ing in diabetic neuropathy. Handb Clin Neurol.
2014;126:235–48.
5. Feldman E, Callaghan B, Pop-Busui R, et al. Diabetic
neuropathy. Nat Rev Dis Primers. 2019;5:41.
6. Kobayashi M, Zochodne DW.  Diabetic neuropathy
and the sensory neuron: new aspects of pathogenesis
and their treatment implications. J Diabetes Investig.
2018;9:1239–54.
7. Wilson NM, Wright DE.  Experimental motor
neuropathy in diabetes. Handb Clin Neurol.
2014;126:461–7.
8. Zochodne DW, Ramji N, Toth C. Neuronal targeting
in diabetes mellitus: a story of sensory neurons and
motor neurons. Neuroscientist. 2008;14:311–8.
9. Writing team for the Diabetes Control and
Complications Trial/Epidemiology of Diabetes
Interventions and Complications Research Group.
Effect of intensive therapy on the microvascular
complications of type 1 diabetes mellitus. JAMA.
2002;287:2563–9.
10. Stratton I, Adler AI, Neil H, et  al. Association
of glycaemia with macrovascular and microvas-
cular complications of type 2 diabetes (UKPDS
35): prospective observational study. BMJ.
2000;321:405–12.
11. Bönhof GJ, Herder C, Strom A, Papanas N, Roden
M, Ziegler D. Emerging biomarkers, tools, and treat-
ments for diabetic polyneuropathy. Endocr Rev.
2019;40:153–92.
12. Zochodne D.  Diabetes mellitus and the peripheral
nervous system: manifestations and mechanisms.
Muscle Nerve. 2007;36:144–66.
13. Tavakoli M, Mojaddidi M, Fadavi H, Malik
R. Pathophysiology and treatment of painful diabetic
neuropathy. Curr Pain Headache Rep. 2008;12:192–7.
14. Smith A, Singleton J.  Impaired glucose tolerance
and neuropathy. Neurologist. 2008;14:23–9.
15. Malik R. Wherefore art thou, O treatment for diabetic
neuropathy? Int Rev Neurobiol. 2016;127:287–317.
16. Cameron NE, Cotter MA, Low PA.  Nerve blood
flow in early experimental diabetes in rats: relation
to conduction deficits. Am J Phys. 1991;261:E1–8.
17. Nukada H, Dyck PJ.  Microsphere embolization
of nerve capillaries and fiber degeneration. Am J
Pathol. 1984;115:275–87.
18. Cameron NE, Cotter MA, Dines KC, Maxfield
EK, Carey F, Mirrlees DJ. Aldose reductase inhibi-
tion, nerve perfusion, oxygenation and function in
streptozotocin-diabetic rats: dose-response consider-
ations and independence from a myo-inositol mech-
anism. Diabetologia. 1994;37:651–63.
19. Cameron NE, Cotter MA, Archibald V, Dines
KC, Maxfield EK.  Anti-oxidant and pro-oxidant
effects on nerve conduction velocity, endoneu-
rial blood flow and oxygen tension in non-diabetic
M. Yorek
and streptozotocin-­ diabetic rats. Diabetologia.
1994;37:449–59.
20. Kles KA, Vinik AI. Pathophysiology and treatment
of diabetic peripheral neuropathy: the case for dia-
betic neurovascular function as an essential compo-
nent. Curr Diabetes Rev. 2006;2:131–45.
21. Cotter MA, Dines KC, Cameron NE.  Prevention
and reversal of motor and sensory peripheral nerve
conduction abnormalities in streptozotocin-diabetic
rats by the prostacyclin analogue iloprost. Naunyn
Schmiedebergs Arch Pharm. 1993;347:534–40.
22. Cameron NE, Cotter MA, Dines KC, Love A. Effects
of aminoguanidine on peripheral nerve function
and polyol pathway metabolites in streptozotocin-­
diabetic rats. Diabetologia. 1992;35:946–50.
23. Sima AA.  Pathological mechanisms involved in
diabetic neuropathy: can we slow the process? Curr
Opin Investig Drugs. 2006;7:324–37.
24. Pop-Busui R, Sima A, Stevens M. Diabetic neuropa-
thy and oxidative stress. Diabetes Metab Res Rev.
2006;22:257–73.
25. Gabbay KH. Aldose reductase inhibition in the treat-
ment of diabetic neuropathy: where are we in 2004?
Curr Diab Rep. 2004;4:405–8.
26. Oates PJ.  Aldose reductase, still a compelling tar-
get for diabetic neuropathy. Curr Drug Targets.
2008;9:14–36.
27. Chalk C, Benstead TJ, Moore F.  Aldose reductase
inhibitors for the treatment of diabetic polyneuropa-
thy. Cochrane Database Syst Rev. 2007;CD004572.
28. Tomlinson DR, Moriarty RJ, Mayer JH. Prevention
and reversal of defective axonal transport and motor
nerve conduction velocity in rats with experimen-
tal diabetes by treatment with the aldose reductase
inhibitor Sorbinil. Diabetes. 1984;33:470–6.
29. Willars GB, Tomlinson DR, Robinson JP.  Studies
of sorbinil on axonal transport in streptozotocin-­
diabetic rats. Metabolism. 1986;35(4 Suppl
1):66–70.
30. Obrosova IG, Huysen CV, Fathallah L, Cao XC,
Greene DA, Stevens MJ.  As aldose reductase
inhibitor reverses early diabetes-induced changes in
peripheral nerve function, metabolism, and antioxi-
dative defense. FASEB J. 2002;16:123–5.
31. Cameron NE, Leonard MB, Ross IS, Whiting
PH. The effects of sorbinil on peripheral nerve con-
duction velocity, polyol concentrations and morphol-
ogy in the streptozotocin-diabetic rat. Diabetologia.
1986;29:168–74.
32. Greene DA, Lattimer SA.  Recent advances in
the therapy of diabetic peripheral neuropathy by
means of a aldose reductase inhibitor. Am J Med.
1985;79:13–7.
33. Greene DA.  A sodium-pump defect in diabetic
peripheral nerve corrected by sorbinil administra-
tion: relationship to myo-inositol metabolism and
nerve conduction slowing. Metabolism. 1986;35(4
Suppl 1):60–5.
34. Stevens MJ, Lattimer SA, Kamijo M, Van Huysen
C, Sima AA, Greene DA.  Osmotically-induced
Metabolic Mechanisms in Diabetic Neuropathy
nerve taurine depletion and the compatible osmolyte
hypothesis in experimental diabetic neuropathy in
the rat. Diabetologia. 1993;36:608–14.
35. Greene DA, Mackway AM. Decreased myo-­inositol
content and Na+-K+-ATPase activity in superior
cervical ganglion of STZ-diabetic rat and pre-
vention by aldose reductase inhibition. Diabetes.
1986;35:1106–8.
36. Finegold D, Lattimer SA, Nolle S, Bernstein M,
Greene DA.  Polyol pathway activity and myo-­
inositol metabolism. A suggested relationship in
the pathogenesis of diabetic neuropathy. Diabetes.
1983;32:988–92.
37. Gillon KR, Hawthorne JN, Tomlinson DR. Myo-­
inositol and sorbitol metabolism in relation to
peripheral nerve function in experimental diabe-
tes in the rat: effect of aldose reductase inhibition.
Diabetologia. 1983;25:365–71.
38. Pop-Busui R, Sullivan KA, Van Huysen CV, et  al.
Depletion of taurine in experimental diabetic neurop-
athy: implications for nerve metabolic, vascular, and
functional deficits. Exp Neurol. 2001;168:259–72.
39. Yorek MA, Wiese TJ, Davidson EP, et al. Reduced
motor nerve conduction velocity and Na(+)-K(+)-
ATPase activity in rats maintained on L-fucose diet.
Reversal by myo-inositol supplementation. Diabetes.
1993;42:1401–6.
40. Yorek MA, Stefani MR, Dunlap JA, Ro KS,
Davidson EP.  Trans-hydroxyl group configuration
on carbons 2 and 3 of glucose. Responsible for
acute inhibition of myo-inositol transport? Diabetes.
1991;40:1016–23.
41. Yorek MA, Dunlap JA, Stefani MR, Davidson
EP.  L-fucose is a potent inhibitor of myo-inositol
transport and metabolism in cultured neuroblastoma
cells. J Neurochem. 1992;58:1626–36.
42. Stefani MR, Dunlap JA, Yorek MA.  Effect of
L-fucose on proliferation and myo-inositol metabo-
lism in cultured cerebral microvessel and aortic
endothelial cells. J Cell Physiol. 1992;153:321–31.
43. Coppey LJ, Gellett JS, Davidson EP, Dunlap JA,
Yorek MA. Effect of treating streptozotocin-induced
diabetic rats with sorbinil, myo-inositol or amino-
guanidine on endoneurial blood flow, motor nerve
conduction velocity and vascular function of epineu-
rial arterioles of the sciatic nerve. Int J Diabetes Res.
2002;3:21–36.
44. Coppey LJ, Davidson EP, Dunlap JA, Lund DD,
Yorek MA.  Slowing of motor nerve conduction
velocity in streptozotocin-induced diabetic rats is
preceded by impaired vasodilation in arterioles that
overlie the sciatic nerve. Int J Exp Diabetes Res.
2000;1:131–43.
45. Farias VX, Uchoa PN, Aquino CP, et al. Expression
of myo-inositol cotransporters in the sciatic nerve
and dorsal root ganglia in experimental diabetes.
Braz J Med Biol Res. 2019;52:e8589.
46. Kalichman MW, Dines KC, Bobik M, Mizisin
AP.  Nerve conduction velocity, laser Doppler flow,
and axonal caliber in galactose and streptozotocin
diabetes. Brain Res. 1998;810:130–7.
269
47. Dvornik E, Simard-Duquesne N, Krami M, et  al.
Polyol accumulation in galactosemic and dia-
betic rats: control by an aldose reductase inhibitor.
Science. 1973;182:1146–8.
48. Kamijo M, Basso M, Cherian PV, Hohman TC, Sima
AA. Galactosemia produces ARI-preventable nodal
changes similar to those of diabetic neuropathy.
Diabetes Res Clin Pract. 1194;25:117–29.
49. Schmidt RE, Dorsey DA, Beaudet LN, et  al. A
potent sorbitol dehydrogenase inhibitor exacer-
bates sympathetic autonomic neuropathy in rats
with streptozotocin-induced diabetes. Exp Neurol.
2005;192:407–19.
50. Schmidt RE, Dorsey DA, Beaudet LN, et  al.
Inhibition of sorbitol dehydrogenase exacerbates
autonomic neuropathy in rats with streptozotocin-­
induced diabetes. J Neuropathol Exp Neurol.
2001;60:1153–69.
51. Yagihashi S, Yamagishi SI, Wada R, et al. Neuropathy
in diabetic mice overexpressing human aldose reduc-
tase and effects of aldose reductase inhibitor. Brain.
2001;124:2448–58.
52. Ho ECM, Lam KSL, Chen YS, et  al. Aldose
reductase-­deficient mice are protected from delayed
motor nerve conduction velocity, increased c-Jun
NH2-terminal kinase activation, depletion of reduced
glutathione, increased superoxide accumulation, and
DNA damage. Diabetes. 2006;55:1946–53.
53. Sima AA, Bril V, Nathaniel V, et  al. Regeneration
and repair of myelinated fibers in sural-nerve
biopsy specimens from patients with diabetic
neuropathy treated with sorbinil. N Engl J Med.
1988;319:548–55.
54. Sima AA, Prashar A, Nathaniel V, Bril V, Werb MR,
Greene DA.  Overt diabetic neuropathy: repair of
axo-glial dysjunction and axonal atrophy by aldose
reductase inhibition and its correlation to improve-
ment in nerve conduction velocity. Diabet Med.
1993;10:115–21.
55. Gieron MA, Malone JI, Lowitt S, Korthals
JK. Improvement in peripheral nerve function after
one year of Sorbinil. Neuroreport. 1991;2:348–50.
56. Lewin IG, O’Brien IA, Morgan MH, Corrall
RJ. Clinical and neurophysiological studies with the
aldose reductase inhibitor, sorbinil, in symptomatic
diabetic neuropathy. Diabetologia. 1984;26:445–8.
57. O’Hare JP, Morgan MH, Alden P, Chissel S, O’Brien
IA, Corrall RJ.  Aldose reductase inhibition in dia-
betic neuropathy: clinical and neurophysiological
studies of one year’s treatment with sorbinil. Diabet
Med. 1988;5:537–42.
58. Fagius J, Brattberg A, Jameson S, Berne C. Limited
benefit of treatment of diabetic polyneuropathy with
an aldose reductase inhibitor: a 24-week controlled
trial. Diabetologia. 1985;28:323–9.
59. Pfeifer MA, Schumer MP, Gelber DA.  Aldose
reductase inhibitors: the end of an era or the need
for different trial designs? Diabetes. 1997;46(Suppl
2):S82–9.
60. He J, Gao H-X, Yang N, et al. The aldose reductase
inhibitor epalrestat exerts nephritic protection on dia-
270
betic nephropathy in db/db mice through metabolic
modulation. Acta Pharmacol Sin. 2019;40:86–97.
61. Hotta N, Kawamori R, Fukuda M, Shigeta Y. Long-­
term clinical effects of epalrestat, an aldose reduc-
tase inhibitor, on progression of diabetic neuropathy
and other microvascular complications: multivariate
epidemiological analysis based on patient back-
ground factors and severity of diabetic neuropathy.
Diabet Med. 2012;29:1529–33.
62. Ramirez MA, Borja NL. Epalrestat: an aldose reduc-
tase inhibitor for the treatment of diabetic neuropa-
thy. Pharmacotherapy. 2008;28:646–55.
63. Li Q-R, Wang Z, Zhou W, et al. Epalrestat protects
against diabetic peripheral neuropathy by alleviat-
ing oxidative stress and inhibiting polyol pathway.
Neural Regen Res. 2016;11:345–51.
64. Hotta N, Akanuma Y, Kawamori R, et al. Long-term
clinical effects of epalrestat, an aldose reductase
inhibitor, on diabetic peripheral neuropathy: the
3-year, multicenter, comparative Aldose Reductase
Inhibitor-Diabetes Complications Trial. Diabetic
Care. 2006;29:1538–44.
65. Obrosova IG, Pacher P, Szabo C, et al. Aldose reduc-
tase inhibition counteracts oxidative-nitrosative
stress and poly(ADP-ribose) polymerase activation
in tissue sites for diabetes complications. Diabetes.
2005;54:234–42.
66. Yorek MA, Coppey LJ, Gellett JS, Davidson EP,
Lund DD. Effect of fidarestat and alpha-lipoic acid
on diabetes-induced epineurial arteriole vascular
dysfunction. Exp Diabesity Res. 2004;5:123–35.
67. Rochette L, Ghibu S, Muresan A, Vergely C. Alpha-­
lipoic acid: molecular mechanisms and therapeu-
tic potential in diabetes. Can J Physiol Pharmacol.
2015;93:1021–7.
68. Wang X-T, Lin H-X, Xu S-A, Lu Y-K. Lipoic acid
combined with epalrestat versus lipoic acid in
treating diabetic peripheral neuropathy: a meta-­
analysis. Zhongguo Yi Xue Ke Xue Yuan Xue Bao.
2017;39:656–64.
69. Zhao M, Chen J-Y, Chu Y-D, Zhu Y-B, Luo L, Bu
S-Z. Efficacy of epalrestat plus α-lipoic acid combi-
nation therapy versus monotherapy in patients with
diabetic peripheral neuropathy: a meta-analysis of
20 randomized controlled trials. Neural Regen Res.
2018;13:1087–95.
70. Matsumoto T, Ono Y, Kurono M, Kuromiya A,
Nakamura K, Bril V. Ranirestat (AS-3201), a potent
aldose reductase inhibitor, reduces sorbitol lev-
els and improves motor nerve conduction velocity
in streptozotocin-diabetic rats. J Pharmacol Sci.
2008;107:231–7.
71. Matsumoto T, Ono Y, Kurono M, Kuromiya A,
Toyosawa K, Ueda Y, Bril V.  Long-term treat-
ment with ranirestat (AS-3201), a potent aldose
reductase inhibitor, suppresses diabetic neuropathy
and cataract formation in rats. J Pharmacol Sci.
2008;107:340–8.
72. Asano S, Himeno T, Hayami T, et  al. Ranirestat
improved nerve conduction velocities, sensory per-
M. Yorek
ception, and intraepidermal nerve fiber density in
rats with overt diabetic polyneuropathy. J Diabetes
Res. 2019;2019:2756020.
73. Ota A, Kakehashi A, Toyoda F, et al. Effects of long-­
term treatment with ranirestat, a potent aldose reduc-
tase inhibitor, on diabetic cataract and neuropathy in
spontaneously diabetic torii rats. J Diabetes Res.
2013;2013:175901.
74. Sekiguchi K, Kohara N, Baba M, et al. Aldose reduc-
tase inhibitor ranirestat significantly improves nerve
conduction velocity in diabetic polyneuropathy: a
randomized double-blind placebo-controlled study
in Japan. J Diabetes Investig. 2019;10:466–74.
75. Bril V, Hirose T, Tomioka S, Buchanan R. Ranirestat
for the management of diabetic sensorimotor poly-
neuropathy. Diabetes Care. 2009;32:1256–60.
76. Giannoukakis N.  Evaluation of ranirestat for the
treatment of diabetic neuropathy. Expert Opin Drug
Metab Toxicol. 2014;10:1051–9.
77. Sima AAF, Sugimoto K.  Experimental dia-
betic neuropathy: an update. Diabetologia.
1999;42:773–88.
78. Brownlee M, Cerami A, Vlassara H. Advanced gly-
cosylation end products in tissue and the biochemi-
cal basis of diabetic complications. N Engl J Med.
1988;318:1315–21.
79. Zochodne D. Mechanisms of diabetic neuron dam-
age: molecular pathways. Handb Clin Neurol.
2014;126:379–99.
80. Ryle C, Leow CK, Donaghy M.  Nonenzymatic
glycation of peripheral and central nervous system
proteins in experimental diabetes mellitus. Muscle
Nerve. 1997;20:577–84.
81. Cullum NA, Mahon J, Stringer K, McLean
WG.  Glycation of rat sciatic nerve tubulin in
experimental diabetic mellitus. Diabetologia.
1991;34:387–9.
82. Ozturk G, Sekeroglu MR, Erdogan E, Ozturk M. The
effect of non-enzymatic glycation of extracellular
matrix proteins on axonal regeneration in vitro. Acta
Neuropathol. 2006;112:627–32.
83. Forbes J, Cooper M. Mechanisms of diabetic com-
plications. Physiol Rev. 2013;93:137–88.
84. Wada R, Yagihashi S.  Role of advanced glyca-
tion end products and their receptors in develop-
ment of diabetic neuropathy. Ann N Y Acad Sci.
2005;1043:598–604.
85. Bekircan-Kurt CB, Tan E, Ozdamar SE.  The acti-
vation of RAGE and NF-κB in nerve biopsies of
patients with axonal and vasculitic neuropathy. Noro
Psikiyatr Ars. 2015;52:279–82.
86. Yamagishi S, Nakamura K, Matsui T, Noda Y,
Imaizumi T.  Receptor for advanced glycation end
products (RAGE): a novel therapeutic target for
diabetic vascular complication. Curr Pharm Des.
2008;14:487–95.
87. Yamagishi S, Nakamura K, Matsui T, Ueda S,
Fukami K, Okuda S.  Agents that block advanced
glycation end product (AGE)-RAGE (receptor for
AGEs)-oxidative stress system: a novel therapeutic
Metabolic Mechanisms in Diabetic Neuropathy
strategy for diabetic vascular complications. Expert
Opin Investig Drugs. 2008;17:983–96.
88. Shimizu F, Sano Y, Haruki H, Kanda T.  Advanced
glycation end-products induce basement mem-
brane hypertrophy in endoneurial microvessels
and disrupt the blood-nerve barrier by stimulat-
ing the release of TGF-β and vascular endothelial
growth factor (VEGF) by pericytes. Diabetologia.
2011;54:1517–26.
89. Terata K, Coppey LJ, Davidson EP, Dunlap JA,
Gutterman DD, Yorek MA.  Acetylcholine-induced
arteriolar dilation is reduced in streptozotocin-­
induced diabetic rats with motor nerve dysfunction.
Br J Pharmacol. 1999;128:837–43.
90. Cameron NE, Gibson TM, Nangle MR, Cotter
MA.  Inhibitors of advanced glycation end prod-
uct formation and neurovascular dysfunction
in experimental diabetes. Ann N Y Acad Sci.
2005;1043:784–92.
91. Sugimoto K, Yasujima M, Yagihashi S.  Role of
advanced glycation end products in diabetic neu-
ropathy. Curr Pharm Des. 2008;14:953–61.
92. Lukic I, Humpert P, Nawroth P, Bierhaus A.  The
RAGE pathway: activation and perpetuation in the
pathogenesis of diabetic neuropathy. Ann N Y Acad
Sci. 2008;1126:76–80.
93. Toth C, Martinez J, Zochodne D.  RAGE, dia-
betes, and the nervous system. Curr Mol Med.
2007;7:766–76.
94. Juranek JK, Geddis MS, Song F, et al. RAGE defi-
ciency improves postinjury sciatic nerve regeneration
in type 1 diabetic mice. Diabetes. 2013;62:931–43.
95. Bierhaus A, Humpert PM, Morcos M, et  al.
Understanding RAGE, the receptor for advanced
glycation end products. J Mol Med. 2005;83:876–86.
96. Juranek JK, Kothary P, Mehra A, Hays A, Brannagan
TH, Schmidt AM. Increased expression of the recep-
tor for advanced glycation end-products in human
peripheral neuropathies. Brain Behav. 2013;3:701–9.
97. Yagihashi S, Kamijo M, Baba M, Yagihashi N, Nagai
K.  Effect of aminoguanidine on functional and
structural abnormalities in peripheral nerve of STZ-­
induced diabetic rats. Diabetes. 1992;41:47–52.
98. Miyauchi Y, Shikama H, Takasu T, et al. Slowing of
peripheral motor nerve conduction was ameliorated
by aminoguanidine in streptozotocin-induced dia-
betic rats. Eur J Endocrinol. 1996;134:467–73.
99. Sugimoto K, Yagihashi S. Effects of aminoguanidine
on structural alterations of microvessels in periph-
eral nerve of streptozotocin diabetic rats. Microvasc
Res. 1997;53:105–12.
100. Kihara M, Schmelzer JD, Poduslo JF, Curran GL,
Nickander KK, Low PA.  Aminoguanidine effects
on nerve blood flow, vascular permeability, elec-
trophysiology, and oxygen free radicals. Proc Natl
Acad Sci U S A. 1991;88:6107–11.
101. Cameron NE, Cotter MA. Rapid reversal by amino-
guanidine of the neurovascular effects of diabetes in
rats: modulation by nitric oxide synthase inhibition.
Metabolism. 1996;45:1147–52.
271
102. Metz TO, Alderson NL, Thorpe SR, Baynes
JW. Pyridoxamine, an inhibitor of advanced glyca-
tion and lipoxidation reactions: a novel therapy for
treatment of diabetic complications. Arch Biochem
Biophys. 2003;419:41–9.
103. Nagai R, Schirakawa JI, Ohno RI, Moroishi N,
Nagai M.  Inhibition of AGEs formation by natural
products. Amino Acids. 2014;46:261–6.
104. Du XL, Edelstein D, Rossetti L, et al. Hyperglycemia-­
induced mitochondrial superoxide overproduction
activates the hexosamine pathway and induces plas-
minogen activator inhibitor-1 expression by increas-
ing Sp1 glycosylation. Proc Natl Acad Sci U S A.
2000;97:12222–6.
105. Román-Pintos LM, Villegas-Rivera G, Rodríguez-­
Carrizalez AD, Miranda-Díaz AG, Cardona-­
Muñoz EG.  Mellitus: inflammation, oxidative
stress, and mitochondrial function. J Diabetes Res.
2016;3425617
106. Yagihashi S.  Glucotoxic mechanisms and related
therapeutic approaches. Int Rev Neurobiol.
2016;127:121–49.
107. Heath JM, Sun Y, Yuan K, et al. Activation of AKT
by O-linked N-acetylglucosamine induces vas-
cular calcification in diabetes mellitus. Circ Res.
2014;114:1094–102.
108. Hammes H, Du X, Edelstein D, et al. Benfotiamine
blocks three major pathways of hyperglycemic dam-
age and prevents experimental diabetic retinopathy.
Nat Med. 2003;9:294–9.
109. Sanchez-Ramirez GM, Caram-Salas NL, Rocha-­
Ganzalez HI, et al. Benfotiamine relieves inflamma-
tory and neuropathic pain in rats. Eur J Pharmacol.
2006;530:48–53.
110. Stracke H, Gaus W, Achenbach U, Federlin K,
Bretzel R.  Benfotiamine in diabetic polyneuropa-
thy (BEN-DIP): results of a randomized, double-­
blind, placebo-controlled clinical study. Exp Clin
Endocrinol Diabetes. 2008;116:600–5.
111. Eichberg J.  Protein kinase C changes in diabe-
tes: is the concept relevant to neuropathy? Int Rev
Neurobiol. 2002;50:61–82.
112. Evcimen ND, King GL. The role of protein kinase C
activation and the vascular complications of diabe-
tes. Pharmacol Res. 2007;55:498–510.
113. Sasase T, Morinaga H, Abe T, et al. Protein kinase C
beta inhibitor prevents diabetic peripheral neuropa-
thy, but not histopathological abnormalities of retina
in spontaneously diabetic Torii rat. Diabetes Obes
Metab. 2009;11:1084–7.
114. Nakamura J, Kato K, Hamada Y, et  al. A protein
kinase C-beta-selective inhibitor ameliorates neural
dysfunction in streptozotocin-induced diabetic rats.
Diabetes. 1999;48:2090–5.
115. Cotter MA, Jack AM, Cameron NE.  Effects of the
protein kinase C beta inhibitor LY333531 on neural
and vascular function in rats with streptozotocin-­
induced diabetes. Clin Sci. 2002;103:311–21.
116. Cameron NE, Cotter MA, Jack AM, Basso MD,
Hohman TC. Protein kinase C effects on nerve func-
272
tion, perfusion, Na(+),K(+)-ATPase activity and
glutathione content in diabetic rats. Diabetologia.
1999;42:1120–30.
117. Casellini CM, Barlow PM, Rice AL, et  al. A
6-month randomized, double-blind-masked,
placebo-­controlled study evaluating the effects of
the protein kinase C-beta inhibitor ruboxistaurin on
skin microvascular blood flow and other measures
of diabetic peripheral neuropathy. Diabetes Care.
2007;30:896–902.
118. Brooks B, Delaney-Robinson C, Molyneaux L,
Yue DK.  Endothelial and neural regulation of skin
microvascular blood flow in patients with diabetic
peripheral neuropathy: effect of treatment with the
isoform-specific protein kinase C beta inhibitor,
ruboxistaurin. J Diabetes Complicat. 2008;22:88–95.
119. Vinik AI, Bril V, Kempler P, et  al. Treatment of
symptomatic diabetic peripheral neuropathy with
the protein kinase C beta-inhibitor ruboxistaurin
mesylate during a 1-year randomized, placebo-­
controlled, double-blind clinical trial. Clin Ther.
2005;27:1164–80.
120. van Dam PS.  Oxidative stress and diabetic neu-
ropathy: pathophysiological mechanisms and
treatment perspectives. Diabetes Metab Res Rev.
2002;18:176–84.
121. Schnackenberg CG.  Physiological and pathophysi-
ological roles of oxygen radicals in the renal micro-
vasculature. Am J Phys. 2002;282:R335–42.
122. Radi R, Peluffo G, Alvarez MN, Naviliat M, Cayota
A. Unraveling peroxynitrite formation in biological
systems. Free Radic Biol Med. 2001;30:463–88.
123. Fredstrom S.  Nitric oxide, oxidative stress, and
dietary antioxidants. Nutrition. 2002;18:537–9.
124. Cameron NE, Cotter MA.  Metabolic and vascular
factors in the pathogenesis of diabetic neuropathy.
Diabetes. 1997;46:S31–7.
125. Coppey LJ, Gellett JS, Davidson EP, Dunlap JA,
Lund DD, Yorek MA.  Effect of antioxidant treat-
ment of streptozotocin-induced diabetic rats on
endoneurial blood flow, motor nerve conduc-
tion velocity and vascular reactivity of epineu-
rial arterioles of the sciatic nerve. Diabetes.
2001;50:1927–37.
126. Coppey LJ, Gellett JS, Davidson EP, et  al. Effect
of M40403 treatment of diabetic rats on endoneu-
rial blood flow, motor nerve conduction velocity and
vascular function of epineurial arterioles of the sci-
atic nerve. Br J Pharmacol. 2001;134:21–9.
127. Yorek MA, Coppey LJ, Gellett JS, et al. Effect of treat-
ment of diabetic rats with Dehydroepiandrosterone
(DHEA) on vascular and neural function. Am J
Phys. 2002;283:E1067–75.
128. Cameron NE, Cotter MA, Maxfield EK. Antioxidant
treatment prevents the development of peripheral
nerve dysfunction in streptozotocin-diabetic rats.
Diabetologia. 1993;36:299–304.
129. Cameron NE, Cotter MA.  Neurovascular dys-
function in diabetic rats: potential contribution
of autoxidation and free radicals examined using
M. Yorek
transition metal chelating agents. J Clin Invest.
1995;96:1159–63.
130. Cameron NE, Cotter MA.  Effects of antioxidants
on nerve and vascular dysfunction in experimental
diabetes. Diabetes Res Clin Pract. 1999;45:137–46.
131. Keegan A, Cotter MA, Cameron NE. Effects of dia-
betes and treatment with the antioxidant α-lipoic acid
on endothelial and neurogenic responses of corpus
cavernosum in rats. Diabetologia. 1999;42:343–50.
132. Karasu C, Dewhurst M, Stevens EJ, Tomlinson
DR.  Effects of anti-oxidant treatment on sciatic
nerve dysfunction in streptozotocin-diabetic rats;
comparison with essential fatty acids. Diabetologia.
1995;38:129–34.
133. Pieper GM, Siebeneich W. Diabetes-induced endo-
thelial dysfunction is prevented by long-term treat-
ment with the modified iron chelator, hydroxyethyl
starch conjugated-deferoxamine. J Cardiovasc
Pharmacol. 1997;30:734–8.
134. De Vriese AS, Verbeuren TJ, Van de Voorde J,
Lameire NH. Vanhoutte PM endothelial dysfunction
in diabetes. Br J Pharmacol. 2000;130:963–74.
135. Brownlee M.  Biochemistry and molecular cell
biology of diabetic complications. Nature.
2001;414:813–20.
136. Nishikawa T, Edelstein D, Du XL, et al. Normalizing
mitochondrial superoxide production blocks three
pathways of hyperglycaemic damage. Nature.
2000;404:787–90.
137. Fernyhough P.  Mitochondrial dysfunction in dia-
betic neuropathy: a series of unfortunate metabolic
events. Curr Diab Rep. 2015;15:89.
138. Coppey LJ, Gellett JS, Davidson EP, Yorek
MA.  Preventing superoxide formation in epineu-
rial arterioles of the sciatic nerve from diabetic rats
restores endothelium-dependent vasodilation. Free
Radic Res. 2003;37:33–40.
139. Nassar T, Kadery B, Lotan C, Da’as N, Kleinman Y,
Haj-Yehia A.  Effects of the superoxide dismutase-­
mimetic compound tempol on endothelial dysfunc-
tion in streptozotocin-induced diabetic rats. Eur J
Pharmacol. 2002;436:111–8.
140. Cameron NE, Jack AM, Cotter MA.  Effect of
alpha-­lipoic acid on vascular responses and noci-
ception in diabetic rats. Free Radic Biol Med.
2001;31:125–35.
141. Cameron NE, Cotter MA. Effects of an extracellular
metal chelator on neurovascular function in diabetic
rats. Diabetologia. 2001;44:621–8.
142. Inkster ME, Cotter MA, Cameron NE. Effects of tri-
entine, a metal chelator, on defective endothelium-­
dependent relaxation in the mesenteric vasculature
of diabetic rats. Free Radic Res. 2002;36:1091–9.
143. Fink B, Coppey L, Davidson E, et al. Effect of mito-
quinone (Mito-Q) on neuropathic endpoints in an
obese and type 2 diabetic rat model. Free Radic Res.
2020;54:311–8.
144. Cameron NE, Tuck Z, McCabe L, Cotter MA. Effect
of the hydroxyl radical scavenger, dimethylthiourea,
on peripheral nerve tissue perfusion, conduction
Metabolic Mechanisms in Diabetic Neuropathy
velocity and nociception in experimental diabetes.
Diabetologia. 2001;44:1161–9.
145. Manzella D, Barbieri M, Ragno E, Paolisso
G.  Chronic administration of pharmacologic doses
of vitamin E improves the cardiac autonomic ner-
vous system in patients with type 2 diabetes. Am J
Clin Nutr. 2001;73:1052–7.
146. van Dam PS, Bravenboer B, van Asbeck BS, Marx
JJ, Gispen WH.  High rat food vitamin E content
improves nerve function in streptozotocin-diabetic
rats. Eur J Pharmcol. 1999;376:217–22.
147. Nicklander KK, Schmelzer JD, Rohwer DA, Low
PA. Effect of alpha-tocopherol deficiency on indices
of oxidative stress in normal and diabetic peripheral
nerve. J Neurol Sci. 1994;126:6–14.
148. Kocak G, Aktan F, Canbolat O, et  al. Alpha-lipoic
acid treatment ameliorates metabolic parameters,
blood pressure, vascular reactivity and morphol-
ogy of vessels already damaged by streptozotocin-­
diabetes. Diabetes Nutr Metab. 2000;13:308–18.
149. Coleman MD, Eason RC, Bailey CJ. The therapeutic
use of lipoic acid in diabetes: a current perspective.
Environ Toxicol Pharmacol. 2001;10:167–72.
150. Packer L, Kraemer K, Rimbach G.  Molecular
aspects of lipoic acid in the prevention of diabetes
complications. Nutrition. 2001;17:888–95.
151. Dincer Y, Telci A, Kayah R, Yilmaz IA, Cakatay U,
Akcay T. Effect of α-lipoic acid on lipid peroxida-
tion and anti-oxidant enzyme activities in diabetic
rats. Clin Exp Pharmacol Physiol. 2002;29:281–4.
152. Jones W, Li X, Zhi-Chao Q, Perriott L, Whitesell
RR, May JM.  Uptake, recycling , and antioxidant
actions of α-lipoic acid in endothelial cells. Free Rad
Biol Med. 2002;33:83–93.
153. Bravenboer B, Kappelle AC, Hamers FPR, van
Buren T, Erkelens DW, Gispen WH.  Potential use
of glutathione for the prevention and treatment of
diabetic neuropathy in the streptozotocin-induced
diabetic rat. Diabetologia. 1992;35:813–7.
154. Nakamura J, Hamada Y, Chaya S, et  al. Transition
metals and polyol pathway in the development of
diabetic neuropathy in rats. Diabetes Metab Res Rev.
2002;18:395–402.
155. Papanas N, Ziegler D.  Efficacy of α-lipoic acid in
diabetic neuropathy. Expert Opin Pharmacother.
2014;15:2721–31.
156. Cuzzocrea S, Riley DP, Caputi AP, Salvemini
D.  Antioxidant therapy: a new pharmacological
approach in shock, inflammation, and ischemia/
reperfusion injury. Pharmacol Rev. 2001;53:135–59.
157. Doggrell SA.  Therapeutic potential of selective
superoxide dismutase mimetics. Drugs Future.
2002;27:385–90.
158. Salvemini D, Wang ZQ, Zweier JL, et al. A nonpep-
tidyl mimic of superoxide dismutase with therapeu-
tic activity in rats. Science. 1999;286:304–6.
159. Salvemini D, Riley DP.  M40403. Drugs Fut.
2002;25:1027–33.
160. Pacher P, Obrosova IG, Mabley JG, Szabo C. Role
of nitrosative stress and peroxynitrite in the patho-
273
genesis of diabetic complications. Emerging
new therapeutical strategies. Curr Med Chem.
2005;12:267–75.
161. Cowell RM, Russell JW. Nitrsative injury and anti-
oxidant therapy in the management of diabetic neu-
ropathy. J Investig Med. 2004;52:33–44.
162. Vareniuk I, Pavlov IA, Drel VR, et  al. Nitrosative
stress and peripheral diabetic neuropathy in
leptin-deficient (ob/ob) mice. Exp Neurol.
2007;205:425–36.
163. Obrosova IG, Drel VR, Oltman CL, et  al. Role of
nitrosative stress in early neuropathy and vascular
dysfunction in streptozotocin-diabetic rats. Am J
Physiol Endocrinol Metab. 2007;293:E1645–55.
164. Obrosova IG, Mabley JG, Zsengeller Z, et al. Role
for nitrosative stress in diabetic neuropathy: evi-
dence from studies with a peroxynitrite decomposi-
tion catalyst. FASEB J. 2005;19:401–3.
165. Drel VR, Pacher P, Vareniuk I, et  al. A peroxyni-
trite decomposition catalyst counteracts sensory
neuropathy in streptozotocin-diabetic mice. Eur J
Pharmacol. 2007;569:48–58.
166. Obrosova IG, Drel VR, Pacher P, et  al. Oxidative-­
nitrosative stress and poly(ADP-ribose) poly-
merase (PARP) activation in experimental diabetic
neuropathy: the relation is revisited. Diabetes.
2005;54:3435–41.
167. Obrosova IG, Li F, Abatan OI, et  al. Role of
poly(ADP-ribose) polymerase activation in diabetic
neuropathy. Diabetes. 2004;53:711–20.
168. Ilnytska O, Lyzogubov VV, Stevens MJ, et  al.
Poly(ADP-ribose) polymerase inhibition alleviates
experimental diabetic sensory neuropathy. Diabetes.
2006;55:1686–94.
169. Trammell SAJ, Weidemann BJ, Chadda A, et  al.
Nicotinamide riboside opposes type 2 diabetes and
neuropathy in mice. Sci Rep. 2016;6:26933.
170. Purves T, Middlemas A, Agthong S, et al. A role for
mitogen-activated protein kinases in the etiology of
diabetic neuropathy. FASEB J. 2001;15:2508–14.
171. Zhu H, Yu W-J, Le Y, et  al. High glucose levels
increase the expression of neurotrophic factors
associated with p-p42/p44 MAPK in Schwann cells
in vitro. Mol Med Rep. 2012;6:179–84.
172. Stavniichuk R, Drel VR, Shevalye H, et  al.
Baicalein alleviates diabetic peripheral neuropa-
thy through inhibition of oxidative-nitrosative
stress and p38 MAPK activation. Exp Neurol.
2011;230:106–13.
173. Zhou J, Du X, Long M, et al. Neuroprotective effect
of berberine is mediated by MAPK signaling path-
way in experimental diabetic neuropathy in rats. Eur
J Pharmacol. 2016;774:87–94.
174. Sweitzer SM, Medicherla S, Almirez R, et  al.
Antinociceptive action of a p38alpha MAPK inhibi-
tor, SD-282, in a diabetic neuropathy model. Pain.
2004;109:409–19.
175. Kovacs K, Vaczy A, Fekete K, et  al. PARP inhibi-
tor protects against chronic hypoxia/reoxygenation-­
induced retinal injury by regulation of MAPKs,
274
HIF1α, Nrf2, and NFκB. Invest Ophthalmol Vis Sci.
2019;60:1478–90.
176. Yerra VG, Negi G, Sharma SS, Kumar A. Potential
therapeutic effects of the simultaneous targeting of
the Nrf2 and NF-κB pathways in diabetic neuropa-
thy. Redox Biol. 2013;1:394–7.
177. Li J, Hu X, Liang F, et  al. Therapeutic effects of
moxibustion simultaneously targeting Nrf2 and
NF-κB in diabetic peripheral neuropathy. Appl
Biochem Biotechnol. 2019;189:1167–82.
178. Shay KP, Moreau RF, Smith EJ, Smith AR, Hagen
TM.  Alpha-lipoic acid as a dietary supplement:
molecular mechanisms and therapeutic potential.
Biochim Biophys Acta. 2009;1790:1149–60.
179. Agca CA, Tuzcu M, Hayirli A, Sahin K.  Taurine
ameliorates neuropathy via regulating NF-κB and
Nrf2/HO-1 signaling cascades in diabetic rats. Food
Chem Toxicol. 2014;71:116–21.
180. Ma J, Shi M, Zhang X, et al. GLP-1R agonists ame-
liorate peripheral nerve dysfunction and inflamma-
tion via p38 MAPK/NF-κB signaling pathways in
streptozotocin-induced diabetic rats. Int J Mol Med.
2018;41:2977–85.
181. Kumar A, Negi G, Sharma SS.  Suppression of
NF-κB and NI-κB regulated oxidative stress and
neuroinflammation by BAY11-7082 (IκB phosphor-
ylation inhibitor) in experimental diabetic neuropa-
thy. Biochimie. 2012;94:1158–65.
182. Kumar A, Sharma SS.  NF-kappaB inhibitory
action of resveratrol: a probable mechanism
of neuroprotection in experimental diabetic
neuropathy. Biochem Biophys Res Commun.
2010;394:360–5.
183. Lupachyk S, Watcho P, Shevalye H, et  al. Na+/
K+ exchanger 1 inhibition reverses manifesta-
tion of peripheral diabetic neuropathy in type 1
diabetic rats. Am J Physiol Endocrinol Metab.
2013;305:E396–404.
184. Kellogg AP, Cheng HT, Pop-Busui
R. Cyclooxygenase-2 pathway as a potential thera-
peutic target in diabetic peripheral neuropathy. Curr
Drug Targets. 2008;9:68–76.
185. Kellogg AP, Pop-Busui R. Peripheral nerve dysfunc-
tion in experimental diabetes is mediated by cyclo-
oxygenase-­2 and oxidative stress. Antioxid Redox
Signal. 2005;7:1521–9.
186. Kellogg AP, Wiggin TD, Larkin DD, Hayes JM,
Stevens MJ, Pop-Busui R.  Protective effects of
cyclooxygenase-2 gene inactivation against periph-
eral nerve dysfunction and intraepidermal nerve
fiber loss in experimental diabetes. Diabetes.
2007;56:2997–3005.
187. Kimura S, Kontani H.  Demonstration of antial-
lodynic effects of the cyclooxygenase-2 inhibitor
meloxicam on established diabetic neuropathic pain
in mice. J Pharmacol Sci. 2009;110:213–7.
188. Chen L, Li B, Chen B, et  al. Thymoquinone alle-
viates the experimental diabetic peripheral neu-
ropathy by modulation of inflammation. Sci Rep.
2016;6:31656.
M. Yorek
189. Moustafa P, Abdelkader NF, El-Awdan SA,
El-Shabrawy OA, Zaki HF.  Extracellular matrix
remodeling and modulation of inflammation and
oxidative stress by sulforaphane in experimen-
tal diabetic peripheral neuropathy. Inflammation.
2018;41:1460–76.
190. Stavniichuk R, Drel VR, Shevalye H, et al. Role of
12/15-lipoxygenase in nitrosative stress and periph-
eral prediabetic and diabetic neuropathies. Free
Radic Biol Med. 2010;49:1036–45.
191. Stavniichuk R, Obrosov AA, Drel VR, Nadler JL,
Obrosova IG, Yorek MA. 12/15-Lipoxygenase inhi-
bition counteracts MAPK phosphorylation in mouse
and cell culture models of diabetic peripheral neu-
ropathy. J Diabetes Mellitus. 2013;3:33015. https://
doi.org/10.4236/jdm.2013.33015.
192. Obrosova IG, Stavniichuk R, Drel VR, et  al.
Different roles of 12/15-lipoxygenase in diabetic
large and small fiber peripheral and autonomic neu-
ropathies. Am J Pathol. 2010;177:1436–47.
193. Stavniichuk R, Shevalye H, Hirooka H, Nadler JL,
Obrosova IG. Interplay of sorbitol pathway or glu-
cose metabolism, 12/15-lipoxygenase, and mitogen-­
activated protein kinase in the pathogenesis of
diabetic peripheral neuropathy. Biochem Pharmacol.
2012;83:932–40.
194. Yorek MA.  The potential role of angiotensin con-
verting enzyme and vasopeptidase inhibitors in the
treatment of diabetic neuropathy. Curr Drug Targets.
2008;9:77–84.
195. Yorek MA. Is fish oil a potential treatment for dia-
betic peripheral neuropathy? Curr Diabetes Rev.
2018;14:339–49.
196. Deacon CF.  Dipeptidyl peptidase 4 inhibitors in
the treatment of type 2 diabetes mellitus. Nat Rev
Endocrinol. 2020;16:642–53.
197. Jin HY, Liu WJ, Park JH, Baek HS, Park
TS.  Effect of dipeptidyl peptidase-IV (DPP-IV)
inhibitor (Vildagliptin) on peripheral nerves in
streptozotocin-­induced diabetic rats. Arch Med Res.
2009;40:536–44.
198. Tsubi K, Mizukami H, Inaba W, Baba M, Yagihashi
S. The dipeptidyl peptidase IV inhibitor vildagliptin
suppresses development of neuropathy in diabetic
rodents: effects on peripheral sensory nerve func-
tion, structure and molecular changes. J Neurochem.
2016;136:859–70.
199. Bianchi R, Cervellini C, Porretta-Serapiglia C, et al.
Beneficial effects of PKF275-055, a novel, selec-
tive, orally bioavailable, long-lasting dipeptidyl
peptidase IV inhibitor in streptozotocin-induced dia-
betic peripheral neuropathy. J Pharmacol Exp Ther.
2012;340:64–72.
200. Davidson EP, Coppey LJ, Dake B, Yorek
MA.  Treatment of streptozotocin-induced dia-
betic rats with alogliptin: effect on vascular
and neural complications. Exp Diabetes Res.
2011;2011:810469.
201. Volpe M, Savoia C, De Paolis P, Ostrowska B, Tarasi
D, Rubattu S. The renin-angiotensin system as a risk
Metabolic Mechanisms in Diabetic Neuropathy
factor and therapeutic target for cardiovascular and
renal disease. J Am Soc Nephrol. 2002;13:S173–8.
202. Nickenig G, Ostergren J, Struijker-Boudier
H. Clinical evidence for the cardiovascular benefits of
angiotensin receptor blockers. J Renin-Angiotensin-­
Aldosterone Syst. 2006;7(Suppl 1):S1–7.
203. Mogensen CE. Microalbuminuria and hypertension
with focus on type 1 and type 2 diabetes. J Intern
Med. 2003;254:45–66.
204. Li EC, Heran BS, Wright JM. Angiotensin converting
enzyme (ACE) inhibitors versus angiotensin recep-
tor blockers for primary hypertension. Cochrane
Database Syst Rev. 2014;2014:CD009096.
205. Cameron NE, Cotter MA, Robertson S. Angiotensin
converting enzyme inhibition prevents develop-
ment of muscle and nerve dysfunction and stimu-
lates angiogenesis in streptozotocin-diabetic rats.
Diabetologia. 1992;35:12–8.
206. Coppey LJ, Davidson EP, Rinehart TW, et al. ACE
inhibitor or angiotensin II receptor antagonist atten-
uates diabetic neuropathy in streptozotocin-induced
diabetic rats. Diabetes. 2006;55:341–8.
207. Davidson EP, Coppey LJ, Holmes A, Yorek
MA.  Effect of inhibition of angiotensin convert-
ing enzyme and/or neutral endopeptidase on vas-
cular and neural complications in high fat fed/low
dose streptozotocin-diabetic rats. Eur J Pharmacol.
2012;677:180–7.
208. Manschot SM, Gispen WH, Kappelle LJ, Biessels
GJ.  Nerve conduction velocity and evoked poten-
tial latencies in streptozotocin-diabetic rats: effects
of treatment with an angiotensin converting enzyme
inhibitor. Diabetes Metab Res Rev. 2003;19:469–77.
209. Aggarwal M, Singh J, Sood S, Arora B.  Effects of
lisinopril on streptozotocin-induced diabetic neu-
ropathy in rats. Methods Find Exp Clin Pharmacol.
2001;23:131–4.
210. Oltman CL, Davidson EP, Coppey LJ, Kleinschmidt
TL, Lund DD, Yorek MA. Attenuation of vascular/
neural dysfunction in Zucker rats treated with enala-
pril or rosuvastatin. Obesity. 2008;16:82–9.
211. Didangelos T, Tziomalos K, Margaritidis C, et  al.
Efficacy of administration of an angiotensin convert-
ing enzyme inhibitor for two years on autonomic and
peripheral neuropathy in patients with diabetes mel-
litus. J Diabetes Res. 2017;2017:6719239.
212. Malik RA, Williamson S, Abbott C, et  al. Effect
of angiotensin-converting-enzyme (ACE) inhibi-
tor trandolapril on human diabetic neuropathy:
randomized double-blind controlled trial. Lancet.
1998;352:1978–81.
213. Reja A, Tesfaye S, Harris ND, Ward JD. Is ACE inhi-
bition with lisinopril helpful in diabetic neuropathy?
Diabet Med. 1995;12:307–9.
214. Ebihara F, Di Marco GS, Juliano MA, Casarini
DE.  Neutral endopeptidase expression in mesan-
gial cells. J Renin-Angiotensin-Aldosterone Syst.
2003;4:228–33.
215. Muangman P, Spenny ML, Tamura RN, Gibran
NS. Fatty acids and glucose increase neutral endo-
275
peptidase activity in human microvascular endothe-
lial cells. Shock. 2003;19:508–12.
216. Edwards RM, Pullen M, Nambi P.  Distribution of
neutral endopeptidase activity along the rat and rab-
bit nephron. Pharmacology. 1999;59:45–50.
217. Gonzalez W, Soleilhac JM, Fournie-Zaluski MC,
Roques BP, Michel JB.  Characterization of neu-
tral endopeptidase in vascular cells, modulation
of vasoactive peptide levels. Eur J Pharmacol.
1998;345:323–31.
218. Kikkawa F, Shibata K, Suzuki T, et  al. Signal
pathway involved in increased expression of neu-
tral endopeptidase 24.11 by gonadotropin releas-
ing hormone in choriocarcinoma cells. Placenta.
2004;25:176–83.
219. Suzki T, Ino K, Kikkawa F, et al. Neutral endopepti-
dase/CD10 expression during phorbol ester-induced
differentiation of choriocarcinoma cells through the
protein kinase C- and extracellular signal-regulated
kinase-dependent signalling pathway. Placenta.
2002;23:475–82.
220. Pu Q, Schiffrin EL.  Effect of ACE/NEP inhibition
on cardiac and vascular collagen in stroke-prone
spontaneously hypertensive rats. Am J Hypertens.
2001;14:1067–72.
221. Plamboeck A, Holst JJ, Carr RD, Deacon CF. Neutral
endopeptidase 24.11 and dipeptidyl peptidase IV are
both mediators of the degradation of glucagon-like
peptide 1  in the anaesthetised pig. Diabetologia.
2005;48:1882–90.
222. Weber M.  Emerging treatments for hypertension:
potential role for vasopeptidase inhibition. Am J
Hypertens. 1999;12:139S–47S.
223. Trindade PT, Rouleau JL. Vasopeptidase inhibitors:
potential role in the treatment of heart failure. Heart
Fail Monit. 2001;2:2–7.
224. Nathisuwan S, Talbert RL. A review of vasopeptidase
inhibitors: a new modality in the treatment of hyper-
tension and chronic heart failure. Pharmacotherapy.
2002;22:27–42.
225. Arbin V, Claperon N, Fournie-Zaluski MC, Roques
BP, Peyroux J.  Effects of combined neutral endo-
peptidase 24-11 and angiotensin-converting enzyme
inhibition on femoral vascular conductance in
streptozotocin-­induced diabetic rats. Br J Pharmacol.
2000;130:1297–304.
226. Schafer S, Schmidts H-L, Bleich M, Busch AE,
Linz W.  Nephroprotection in Zucker diabetic fatty
rats by vasopeptidase inhibition is partly brady-
kinin B2 receptor dependent. Br J Pharmacol.
2004;143:27–32.
227. Schäfer S, Linz W, Vollert H, et al. The vasopepti-
dase inhibitor AVE7688 ameliorates type 2 diabetic
nephropathy. Diabetologia. 2004;47:98–103.
228. Spenny ML, Muangman P, Sullivan SR, et  al.
Neutral endopeptidase inhibition in diabetic wound
repair. Wound Repair Regen. 2002;10:295–301.
229. Wihler C, Schafer S, Schmid K, et  al. Renal accu-
mulation and clearance of advanced glycation end-­
products in type 2 diabetic nephropathy: effect of
276
angiotensin-converting enzyme and vasopeptidase
inhibition. Diabetologia. 2005;48:1645–53.
230. Oltman CL, Davidson EP, Coppey LJ, Kleinschmidt
TL, Dake B, Yorek MA. Role of the effect of inhibi-
tion of neutral endopeptidase on vascular and neural
complications in streptozotocin-induced diabetic
rats. Eur J Pharmacol. 2001;650:556–62.
231. Coppey L, Davidson E, Lu B, Gerard C, Yorek
M.  Vasopeptidase inhibitor ilepatril (AVE7688)
prevents obesity- and diabetes-induced neu-
ropathy in C57Bl/6J mice. Neuropharmacology.
2011;60:259–66.
232. Davidson EP, Kleinschmidt TL, Oltman CL, Lund
DD, Yorek MA.  Treatment of streptozotocin-
induced diabetic rats with AVE7688, a vasopepti-
dase inhibitor: effect on vascular and neural disease.
Diabetes. 2007;56:355–62.
233. Oltman CL, Davidson EP, Coppey LJ, Kleinschmidt
TL, Yorek MA.  Treatment of Zucker diabetic fatty
rats with AVE7688 improves vascular and neural
dysfunction. Diabetes Obes Metab. 2009;11:223–33.
234. Davidson EP, Coppey LJ, Yorek MA. Early loss of
innervation of cornea epithelium in streptozotocin-­
induced type 1 diabetic rats: improvement with
ilepatril treatment. Invest Ophthalmol Vis Sci.
2012;53:8067–74.
235. Yorek MS, Obrosov A, Lu B, Gerard C, Kardon
RH, Yorek MA.  Effect of inhibition or deletion of
neutral endopeptidase on neuropathic endpoints in
high fat fed/low dose streptozotocin-treated mice. J
Neuropathol Exp Neurol. 2016;75:1072–80.
236. von Lueder TG, Atar D, Krum H.  Current role of
neprilysin inhibitors in hypertension and heart fail-
ure. Pharmacol Ther. 2014;144:41–9.
237. Sangaralingham LR, Sangaralingham SJ, Shah ND,
Yao X, Dunlay SM. Adoption of sacubitril/valsartan
for the management of patients with heart failure.
Circ Heart Fail. 2018;11:e004302.
238. Davidson EP, Coppey LJ, Shevalye H, Obrosov
A, Yorek MA.  Vascular and neural complications
in type 2 diabetic rats: improvement by sacubitril/
valsartan greater than valsartan alone. Diabetes.
2018;67:1616–26.
239. Serhan CN, Chiang N, Dalli J.  New pro-­resolving
n-3 mediators bridge resolution of infectious
inflammation to tissue regeneration. Mol Asp Med.
2018;64:1–17.
240. Serhan CN, Yacoubian S, Yang R. Anti-inflammatory
and proresolving lipid mediators. Annu Rev Pathol.
2008;3:279–312.
241. Grisold A, Callaghan BC, Feldman EL.  Mediators
of diabetic neuropathy: is hyperglycemia the only
culprit? Curr Opin Endocrinol Diabetes Obes.
2017;24:103–11.
242. Bittner V.  Implications for REDUCE IT in clinical
practice. Prog Cardiovasc Dis. 2019;62:395–400.
243. Kris-Etherton PM, Richter CK, Bowen KJ, et  al.
Recent clinical trials shed new light on the cardio-
vascular benefits of omega-3 fatty acids. Methodist
Debakey Cardiovasc J. 2019;15:171–8.
M. Yorek
244. Shaikh SR, Kinnun JJ, Leng X, Williams JA, Wassall
SR. How polyunsaturated fatty acids modify molec-
ular organization in membranes: insight from NMR
studies of model systems. Biochim Biophys Acta.
2015;1848:211–9.
245. Simopoulos AP.  The importance of the omega-6/
omega-3 fatty acid ratio in cardiovascular dis-
ease and other chronic diseases. Exp Biol Med.
2008;233:674–88.
246. Gil A, Serra-Majem L, Calder PC, Uauy
R. Systematic reviews of the role of omega-3 fatty
acids in the prevention and treatment of disease. Br J
Nutr. 2012;107:S1–2.
247. Kaliannan K, Li X-Y, Wang B, et  al. Multi-omic
analysis in transgenic mice implicates omega-6/
omega-3 fatty acid imbalance as a risk factor for
chronic disease. Commun Biol. 2019;2:276.
248. Lanchais K, Capel F, Tournadre A. Could omega-3
fatty acids preserve muscle health in rheumatoid
arthritis? Nutrients. 2020;12:E223.
249. Ueno Y, Miyamoto N, Yamashiro K, Tanaka R,
Hattori N. Omega-3 polyunsaturated fatty acids and
stroke burden. Int J Mol Sci. 2019;20:E5549.
250. Li X, Bi X, Wang S, Zhang Z, Li F, Zhao
AZ.  Therapeutic potential of ω-3 polyunsaturated
fatty acids in human autoimmune diseases. Front
Immunol. 2019;10:2241.
251. Marton LT, Goulart RA, Carvalho ACA,
Barbalho SM.  Omega fatty acids and inflamma-
tory bowel diseases: an overview. Int J Mol Sci.
2019;20:E4851.
252. McGlory C, Calder PC, Nunes EA.  The influence
of omega-3 fatty acids on skeletal muscle protein
turnover in health, disuse, and disease. Front Nutr.
2019;6:144.
253. Avallone R, Vitale G, Bertolotti M.  Omega-3 fatty
acids and neurodegenerative diseases: new evidence
in clinical trials. Int J Mol Sci. 2019;20:E4256.
254. Abdolmaleki F, Kovanen PT, Mardani R, Gheibi-­
Hayat SM, Bo S, Sahebkar A. Resolvins: emerging
players in autoimmune and inflammatory diseases.
Clin Rev Allergy Immunol. 2020;58:82–91.
255. Shevalye H, Yorek MS, Coppey LJ, et  al. Effect
of enriching the diet with menhaden oil or daily
treatment with resolvin D1 on neuropathy in a
mouse model of type 2 diabetes. J Neurophysiol.
2015;114:199–208.
256. Coppey LJ, Davidson EP, Obrosov A, Yorek
MA. Enriching the diet with menhaden oil improves
peripheral neuropathy in streptozotocin-induced type
1 diabetic rats. J Neurophysiol. 2015;113:701–8.
257. Coppey LJ, Holmes A, Davidson EP, Yorek
MA.  Partial replacement with menhaden oil
improves peripheral neuropathy in high fat fed low
dose streptozotocin type 2 diabetic rat. J Nutr Metab.
2012;2012:950517.
258. Obrosov A, Coppey LJ, Shevalye H, Yorek
MA.  Effect of fish oil vs. resolvin D1, E1, methyl
esters of resolvins D1 or D2 on diabetic peripheral
neuropathy. J Neurol Neurophysiol. 2017;8(453)
Metabolic Mechanisms in Diabetic Neuropathy
259. Yorek MS, Coppey LJ, Shevalye H, Obrosov A,
Kardon RH, Yorek MA. Effect of treatment with sal-
salate, menhaden oil, combination of salsalate and
menhaden oil, or resolvin D1 of C57Bl/6J type 1 dia-
betic mouse on neuropathic endpoints. J Nutr Metab.
2016;2016:5905891.
260. Goldfine AB, Fonseca V, Jablonski KA, et  al.
Salicylate (salsalate) in patients with type 2 diabetes:
a randomized trial. Ann Intern Med. 2013;159:1–12.
261. Goldfine AB, Fonseca V, Jablonski KA, et  al. The
effects of salsalate on glycemic control in patients
with type 2 diabetes: a randomized trial. Ann Intern
Med. 2010;152:346–57.
262. Davidson EP, Coppey LJ, Shevalye H, Obrosov
A, Yorek MA.  Effect of dietary content of menha-
den oil with or without salsalate on neuropathic
277
endpoints in high-fat-fed/low-dose streptozotocin-­
treated Sprague Dawley rats. J Diabetes Res.
2018;2018:2967127.
263. Superko HR, Superko SM, Nasir K, Agatston
A, Garrett BC.  Omega-3 fatty acid blood levels.
Circulation. 2013;128:2154–61.
264. Laidlaw M, Holub BJ.  Effects of supplemen-
tation with fish oil-derived n-3 fatty acids and
alpha-­linolenic acid on circulating plasma lipids
and fatty acid profiles in women. Am J Clin Nutr.
2003;77:37–42.
265. Lewis EJH, Perkins BA, Lovblom LE, Bazinet RP,
Wolever TMS, Bril V. Effect of omega-3 supplemen-
tation on neuropathy in type 1 diabetes: a 12-month
pilot trial. Neurology. 2017;88:2294–301.
 Mechanisms of Nerve Injury
in Diabetes: Dyslipidemia,
Bioenergetics, and Oxidative
Damage
Stephanie A. Eid, Mohamed Noureldein,
Masha G. Savelieff, and Eva L. Feldman
1 Introduction
Peripheral neuropathy (PN) is a common compli-
cation in patients with prediabetes and diabetes.
The most common PN is a distal symmetric poly-
neuropathy, which occurs in a distal-to-proximal
pattern of nerve damage, starting in the feet
(Fig.  1a). Patients experience either positive
symptoms, e.g., prickling, tingling, or negative
symptoms, e.g., numbness. When PN has reached
mid-calf, patients begin to experience symptoms
in the fingers (Fig.  1b). While there are other
types of PN in this patient population, particu-
larly autonomic neuropathy, this chapter will
focus on distal symmetric polyneuropathy,
referred to as PN.
Stephanie A.  Eid, Mohamed Noureldein and Masha
G.  Savelieff contributed equally with all other
contributors.
S. A. Eid · M. Noureldein
Department of Neurology, University of Michigan
Medical School, Ann Arbor, MI, USA
M. G. Savelieff
NeuroNetwork for Emerging Therapies, University of
Michigan, Ann Arbor, MI, USA
E. L. Feldman (*)
Department of Neurology, University of Michigan
Medical School, Ann Arbor, MI, USA
NeuroNetwork for Emerging Therapies, University of
Michigan, Ann Arbor, MI, USA
e-mail: efeldman@umich.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_16
PN lacks effective treatments; currently, stan-
dard management consists of glucose control.
Systematic analysis of randomized, controlled
clinical studies indicate, however, that well-­
controlled glucose more effectively prevents PN
development in type 1 (T1D) versus type 2 (T2D)
diabetes patients [1]. Several clinical studies have
since evaluated additional PN risk factors, which
have centered on the metabolic syndrome, and its
components, obesity and dyslipidemia. Several
studies in diverse population have found the meta-
bolic syndrome to be a PN risk factor, independent
of glycemic status (Fig. 1a, Table 1), in the United
States [2–8], China [9–11], Denmark [12, 13],
Germany [14, 15], and the Netherlands [16].
Furthermore, PN risk increases with the number of
metabolic syndrome components [9, 11, 17, 18].
Although hyperglycemia remains a major risk
factor for persons with T2D, the emergence of
the metabolic syndrome, especially dyslipidemia
and obesity as additional PN risk factors, suggest
a distinct or additional pathophysiology underly-
ing neuropathy in T2D versus T1D. This is fur-
ther supported by multiple studies showing that
good or intensive control does not necessarily
prevent PN development in T2D [1, 19, 20]. In
addition to impaired glucose handling, persons
with T2D frequently have comorbid obesity and
dyslipidemia [3, 21], suggesting that, even if dia-
betes is a major PN driver, the effects of lipids on
nerve health also warrant investigation to fully
appreciate the mechanisms of neurodegeneration
279
280
a b
Fig. 1  The peripheral nervous system, PN, and sensory
neurons. (a) Outline of lean (silhouette on the right) and
obese (silhouette on the left) body. PN starts in the feet
(deep red) and progresses to the calves (fading red). At
this point, PN commences in the finger tips (deep red) and
progresses to the hands (fading red). Schematic lumbar
spinal cord section is shown with the dorsal root ganglion
(DRG), which houses sensory neuron cell bodies.
Simplified representation of peripheral sciatic nerves
shown, which runs dorsally. The nerve actually branches
in the lower leg. (b) Anatomy of a sensory peripheral
nerve. Nerves are structured into bundles of neurons
embedded in concentric layers of connective tissue (epi-
neurium, perineurium, endoneurium). The epineurium
covers the entire nerve structure and holds within it fasci-
cles, blood vessels (arterial, red; venous, blue), and addi-
tional loose connective tissue. A perineurium covers each
in PN. However, prior research focused solely on
the effects of hyperglycemia on peripheral ner-
vous system function, particularly of constituent
neurons and Schwann cells as isolated systems,
which only reveals a fraction of the complete pic-
ture. Newer studies addressing the role of dyslip-
idemia and other metabolic syndrome
components and whole nerve bioenergetics are
identifying novel aspects of PN pathogenesis in
T2D.
These advancements have rested on the devel-
opment of animal models, which better mirror
diet-induced metabolic dysfunction in humans
S. A. Eid et al.
c
fascicle, which houses unmyelinated neurons (C fibers in
Remak bundles) and myelinated neurons (A fibers). An
endoneurium covers neuron fibers. Close-up examination
of a myelinated sensory neuron (magnifying glass) reveals
the myelinating Schwann cells, supportive glia, which
wrap around the neuron axon. The neuronal cell body
(otherwise called the soma) resides with the DRG and the
dendrites forge connections with spinal cord neurons. Not
shown are the neuron’s axon termini, which reside within
the target organ, e.g., skin sensory receptors. Red inset:
Neuron cross-section showing central axon and wrapped
Schwann cell. (c) Schematic of a peripheral sensory nerve
to illustrate the aspect ratio. For the longest nerves, e.g.,
sciatic, the axon is around 20,000 times longer than the
cell body diameter. Schwann cells only shown in the start
of the figure, but occur along the entire neuron axon, till
the axon terminus. Created with BioRender.com
[22]. Feeding a high-fat diet (HFD) rich in
­saturated fatty acids to wild-type mice induces
­obesity and impaired glucose tolerance resem-
bling prediabetes. Additional administration of
low-­dose streptozotocin (STZ) abrogates insulin
production and generates a T2D mouse model.
These diet-based models are devoid of genetic
mutations, like the leptin signaling deficient
db/db and ob/ob mice, and instead maintain
leptin production [23, 24], which is more akin to
hyperleptinemia in obesity [25]. The obese HFD
prediabetes and HFD-STZ T2D mice develop
small fiber PN, through intraepidermal nerve
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage 281

Table 1  Summary of clinical studies, which identified components of the metabolic syndrome, including obesity and
dyslipidemia, as PN risk factors
Study Country Population Main MetS and diabetes findings related to PN
Callaghan United States The Diabetes Study of Northern TG level was an independent, stepwise risk
et al. (2011) California (DISTANCE) of factor for non-traumatic lower-extremity
28,701 participants with T1D amputations; TG 150–199 mg/dl, HR 1.10
(~3.5%) or T2D (~96.5%) with (95%CI 0.92–1.32); 200–499 mg/dl, 1.27
baseline TG (1.10–1.47); >500 mg/dl, 1.65 (1.30–2.10);
reference <150 mg/dl
Smith et al. United States The Utah Diabetic Neuropathy Multivariate analysis found obesity and TG
(2013) Study (UDNS) of 218 T2D were related to IENFD loss while elevated
participants HbA1c was related to motor NCV defects
Callaghan United States University of Michigan Weight Diabetes OR 4.90 (95%CI 1.06–22.63) and
et al. (2016) Management Program of 102 WC OR 1.24 (95%CI 1.00–1.55) but not
obese participants, 44.1% prediabetes OR 3.82 (95%CI 0.95–15.41)
normoglycemic, 30.4% significantly associated with PN in
prediabetic, 25.5% T2D multivariable models. PN prevalence was high
in this obese population, even in
normoglycemic individuals
Callaghan United States Health, Aging, and Body Diabetes OR 1.65 (95%CI 1.18–2.31) and
et al. (2016) Composition (Health ABC) HbA1c OR 1.42 (95%CI 1.15–1.75) were the
study of 2382 participants, only Metabolic syndrome measures
21.0% diabetic, 29.9% significantly associated with PN. WC and
prediabetic, 52.8% metabolic HDL significantly associated with multiple
syndrome secondary PN outcomes
Jaiswal et al. United States 79 T2D Pima Indian Participants with versus without PN were more
(2016) participants obese (BMI 36 ± 9 vs 32 ± 5 kg/m2, p = 0.027)
Jaiswal et al. United States SEARCH for Diabetes in Youth Of Metabolic syndrome components, higher
(2017) study of 1734 T1D youths and LDL-c, TG, obesity, diastolic BP and lower
258 T2D youths HDL-c were PN risk factors in T1D youths;
lower HDL-c was a PN risk factor in T2D
youths. Poor glycemic control correlated with
PN in T1D OR 1.53 (95%CI 1.24–1.88) but
not T2D youths OR 1.05 (95%CI 0.7–1.56)
Callaghan United States University of Michigan bariatric BMI was comparable in obese participants
et al. (2020) surgery clinic of 138 obese with and without PN (p = 0.86). WC OR 1.39
participants (95%CI 1.10–1.75), TGs OR 1.31 (95%CI
1.00–1.70), systolic BP OR 2.89 (95%CI
1.49–5.61) were significantly associated with
PN
Lu et al. China Shanghai Diabetic Neuropathy Participants with known T2D had the highest
(2013) Epidemiology and Molecular PN frequency (13.1%). Among participants
Genetics Study (SH-DREAMS) without T2D, higher WC, 2-h postprandial
of 2035 participants, 22.5% plasma glucose, and hypertension were more
normoglycemic, 51.3% likely in those with PN
prediabetic, 26.2% T2D
Han et al. China Shanghai Diabetic Neuropathy Increase Metabolic syndrome in components
(2015) Epidemiology and Molecular linearly proportional to PN (p = 0.001 for
Genetics Study (SH-DREAMS) trend). Metabolic syndrome OR 2.0 (95%CI
of 2035 participants 1.2–3.2) for PN; however, Metabolic syndrome
no longer associated with PN when HOMA-IR
was added to the binary logistic regression,
HOMA-IR OR 1.2 (95%CI 1.1–1.4) associated
with PN
Callaghan China Pinggu, China study of 4002 Diabetes OR 2.60 (95%CI 1.77–3.80), weight
et al. (2018) participants, 37.2% OR 1.09 (95%CI 1.02–1.18), and the number
normoglycemic, 18.9% of metabolic syndrome components OR 1.17
diabetic, 44.0% prediabetic (95%CI 1.03–1.32) were significantly
associated with PN
(continued)
282 S. A. Eid et al.

Table 1 (continued)
Study Country Population Main MetS and diabetes findings related to PN
Andersen Denmark Danish arm of the Anglo- Baseline weight HR 1.09 (95%CI 1.03–1.16),
et al. (2018) Danish-Dutch study of WC HR 1.14 (95%CI 1.05–1.24), BMI HR
Intensive Treatment of Diabetes 1.14 (95%CI 1.06–1.23), log2(methylglyoxal)
in Primary Care (ADDITION) HR 1.45 (95%CI 1.12–1.89), HDL-c HR 0.82
of 1256 T2D participants (95%CI 0.69–0.99), and LDL-c HR 0.92
without PN at baseline (95%CI 0.86–0.98) significantly associated
with incident PN
Christensen Denmark Danish Centre for Strategic In regression analyses, central obesity (WC,
et al. (2020) Research in Type 2 Diabetes WHR, waist-to-­height ratio) associated with
(DD2) Cohort of 5249 T2D PN. Of Metabolic syndrome components, TG
participants ≥1.7 mmol/l aPR 1.36 (95%CI 1.17–1.59),
HDL-c <1.0/1.2 mmol/l (male/female) aPR
1.35 (95%CI 1.12–1.62), hs-CRP ≥3.0 mg/l
aPR 1.66 (95%CI 1.42–1.94), and HbA1c ≥78
mmol/mol (9.3%) aPR 1.42 (95%CI 1.06–
1.88) associated with PN
Ziegler et al. Germany MONICA (Monitoring Trends WC and diabetes were independent factors
(2008) and Determinants on significantly associated with PN in the entire
Cardiovascular Diseases)/ population. Peripheral arterial disease was
KORA (Cooperative Research additionally associated with PN in the diabetic
in the Region of Augsburg) of population
195 diabetes, 46 IGT, 71 IFG,
81 NGT participants
Schlesinger Germany Cooperative Health Research in Overweight OR 3.06 (95%CI 1.57–5.97),
et al. (2019) the Region of Augsburg obese OR 3.47 (95%CI 1.72–7.00), WC OR
(KORA) F4/FF4 cohort of 513 1.22 (95%CI 1.07–1.38). Interaction analyses
participants did not reveal any differences by diabetes
status
Hanewinckel Netherlands Rotterdam Study of 908 BMI independently associated with lower sural
et al. (2017) participants SNAP (OR 1.53 95%CI 1.13–2.09) and
peroneal CMAP (OR 1.49 95%CI 1.11–1.99)
amplitudes
aPR adjusted prevalence ratio, BMI body mass index, BP blood pressure, CI confidence interval, CMAP compound
motor action potential, HbA1c hemoglobin A1c, HDL-c high-­density lipoprotein cholesterol, HOMA-IR homeostatic
model assessment of insulin resistance, HR hazard ratio, hs-CRP high-sensitivity C-reactive protein, IENFD intraepi-
dermal nerve fiber density, IFG impaired fasting glucose, IGT impaired glucose tolerance, LDL-c low-density lipopro-
tein cholesterol, Metabolic syndrome metabolic syndrome, NCV nerve conduction velocity, NGT normal glucose
tolerance, OR odds ratio, SNAP sural sensory nerve action potential, T2D type 2 diabetes, TG triglycerides, WC waist
circumference, WHR waist-to-hip ratio

fiber density loss, and large fiber PN, with
decreased nerve conduction velocities in both
sensory sural and motor sciatic nerves [26–29].
Furthermore, omics and integrated multi-omics
platforms are providing insight through unbiased
interrogation of dysregulated pathways in neuro-
pathic nerve from animal models as well as
human biopsy samples. These studies are uncov-
ering participation of multiple pathways includ-
ing inflammation and oxidative stress in PN
pathophysiology.
Lastly, the association of metabolic syndrome
and its components with PN development high-
lights new therapeutic avenues. Although evi-
dence from large, randomized, and controlled
clinical trials remains scarce, the emerging body
of work which suggests lifestyle changes to cor-
rect the metabolic syndrome and promote weight
loss could prevent PN onset or slow progression.
The American Diabetes Association now recom-
mends diet and exercise as the first-line PN treat-
ment to reverse metabolic syndrome components
and improve glucose handling and lipid profiles
[30–34].
This chapter will examine these three emer-
gent facets of PN pathophysiology. First, we will
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
examine bioenergetics in the diabetic nerve and
the effects of nutrient overload, followed by a
discussion of bioenergetics with a lipid-centric
focus to lay out the effects of dyslipidemia and
altered nerve lipid metabolism on PN develop-
ment. Finally, this chapter will round off with the
most promising therapeutic avenues for PN based
on the lipid-centric perspective, particularly for
T2D patients that also have the metabolic
syndrome.
2 Bioenergetics and Nutrient
Overload in the Nerve
and PN
2.1 Introduction to Nerve
Bioenergetics
Peripheral neuron morphology and its high
energy demands for signal transmission present a
unique challenge to the peripheral nervous sys-
tem. Structurally, peripheral nerve axons can be
as long as 3 feet and are 20,000 times the size of
their cell body (Fig. 1b); thus, energy utilization
and production must be coordinated over propor-
tionately large distances. Functionally, peripheral
neurons require massive amounts of energy to
maintain their membrane potential across a large
surface area [35] and high energy density at spe-
cialized structures, e.g., at terminal synapses, to
perform their roles [36, 37]. Thus, neurons need
to orchestrate energy production according to
their distinct morphology and at areas of highest
energy consumption [38]. Even under resting
conditions, a single neuron can consume 4.7 bil-
lion adenosine triphosphate (ATP) molecules per
second [39]. To meet this immense energy
demand, neurons rely on efficient bioenergetics
through multiple pathways, which is essential to
their health and survival and for peripheral ner-
vous system functioning.
To generate this energy, sensory neurons and
axons can utilize glucose and fatty acids,
(Fig. 2a). Glucose uptake is facilitated by glucose
transporters (GLUTs), which exist in different
isoforms (Fig. 2b). Peripheral neurons primarily
express insulin-independent GLUT3 [42]. Once
283
transported into neurons, glucose catabolism
commences anaerobically within the cytoplasm
by glycolysis, which releases two ATP molecules
and pyruvate [38]. The pyruvate is shuttled into
the mitochondrial matrix by mitochondrial pyru-
vate carrier [43], where it is further metabolized
through the tricarboxylic acid cycle, generating
guanosine triphosphate (GTP), reduced nicotin-
amide adenine dinucleotide (NADH), and flavin
adenine dinucleotide (FADH2; Fig. 2c) [44, 45].
Oxidative phosphorylation next utilizes NADH
and FADH2 as cofactors to complexes I and II,
respectively, which are electron transport chain
proteins that reside within the inner mitochon-
drial membrane [46, 47]. Electrons continue to
move along the electron transport chain, on to
complexes III and IV [48, 49], while protons are
transported from the matrix to the intermembrane
space [45]. This transport generates a proton gra-
dient and the mitochondrial membrane potential
across the inner mitochondrial membrane. The
mitochondrial membrane potential is then dissi-
pated as protons move though complex V (ATP
synthase), generating ATP [50]. In sum, a single
glucose molecule optimally releases 36 ATP after
tricarboxylic acid cycle and oxidative
phosphorylation.
Mitochondrial aerobic oxidation releases
more energy per glucose molecule than anaero-
bic glycolysis; however, It remains uncertain
what pathway neurons prefer under basal con-
ditions, although periods of high-intensity
activity may require both [38, 51, 52]. Glucose
is believed to be the preferred energy source;
however, nerves can develop insulin resistance
[53–55], hindering their capacity for glucose
uptake.
Fatty acids are another fuel substrate; they are
catabolized in mitochondria by β-oxidation
(Fig.  2c) [56]. The fatty acid is first ligated to
CoA by acyl-CoA synthetase long-chain family
member 1 (ACSL1) to generate fatty acyl-CoA,
which is shuttled across the outer mitochondrial
membrane by carnitine palmitoyltransferase I
(CPT1) that in turn substitutes the CoA for carni-
tine. The acylcarnitine is then translocated across
the inner mitochondrial membrane by CPT2,
which substitutes the carnitine back to CoA,
284
b
Fig. 2  Homeostatic peripheral nerve bioenergetics. (a)
Anatomy of a myelinated sensory neuron. Sensory neu-
rons under basal and active conditions require constant
ATP to meet their energy demands. (b) Neurons mainly
express insulin-independent glucose transporter (GLUT)
3 as well as GLUT1 to transport glucose (GLU) into
axons as fuel substrate under homeostatic conditions.
GLU is catabolized to pyruvate (PYR) by glycolysis in the
axon cytoplasm, releasing two ATP molecules.
Alternatively, neurons do express fatty acid (FA) trans-
porters (FAT) [40] and may metabolize fatty acids. Under
homeostatic conditions, Schwann cells utilize GLU,
whereas FAs are used mainly to maintain myelin, although
they can also be metabolized [41]. Under conditions of
especially high neuronal energy demands, Schwann cells
may supplement neuronal energy reserves via axo-glial
crosstalk. PYR is converted to lactate (LAC), which is
shuttled to neurons through monocarboxylate transporters
(MCT). Within neurons, LAC is reverted to PYR, which
joins the pool of substrates entering the tricarboxylic acid
(TCA) cycle. Vital mitochondria (colored red) are traf-
ficked along axons to meet local energy demands, e.g., at
releasing the fatty acyl-CoA into the mitochon-
drial matrix where it undergoes β-oxidation.
Even-chained acyl-CoAs undergo a repeated
four-step process, which cleaves off two carbon
atoms at a time in the form of acetyl-CoA, as well
as NADH and FADH2 [56, 57]. Odd-chained
S. A. Eid et al.
c
axon terminals. (c) PYR is transported by mitochondrial
pyruvate carrier (MPC) across the inner mitochondrial
membrane (IMM) into the mitochondrial matrix. The
pyruvate dehydrogenase complex converts PYR into
acetyl-­CoA, which enters the TCA cycle, generating gua-
nosine triphosphate (GTP), reduced nicotinamide adenine
dinucleotide (NADH), and flavin adenine dinucleotide
(FADH2). NADH generated from the TCA cycle drives
oxidative phosphorylation (OXPHOS) within the IMM
and intermembrane space, releasing additional ATP mol-
ecules. Alternatively, pyruvate carboxylase converts PYR
into oxaloacetate, an anaplerotic reaction, which boosts
TCA capacity. FAs are ligated to CoA by ACSL1 (acyl-­
CoA synthetase long chain family member 1) to generate
fatty acyl-CoA, which is shuttled across the outer mito-
chondrial membrane (OMM) by carnitine palmitoyltrans-
ferase I (CPT1), which substitutes the CoA for carnitine.
The acylcarnitine is then translocated across the IMM by
CPT2, which reverts the carnitine back to CoA, releasing
the fatty acyl-CoA into the mitochondrial matrix to enter
β-oxidation. Created with BioRender.com
acyl-CoAs go through the same process, except
that the final cleavage yields propionyl-CoA in
addition to acetyl-CoA. Subsequently, the gener-
ated acetyl-CoA feeds into the tricarboxylic acid
cycle followed by oxidative phosphorylation to
generate ATP.
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
2.2 Axo-glial Metabolic
Interactions in the Nerve
Schwann cells are the primary glia in the periph-
eral nerves, which support and nurture neurons.
They wrap around axons forming the myelin
sheath (Fig. 2a) that aids in neuronal signal trans-
duction. Schwann cells are thought to primarily
utilize glucose as an energy source to conserve
fatty acids for myelin synthesis [41, 58], although
this remains incompletely known. Indeed, under
normal physiologic conditions, Schwann cells
express GLUT1, which may be insulin responsive
[59], and GLUT3 (Fig.  2b) [60, 61]. Thus, they
undergo insulin-sensitive, facilitated glucose
transport via GLUT1 [62] and insulin-­independent
glucose transport via GLUT3 [63]. In addition to
their myelinating role, Schwann cells can also
sense axonal activity and supplement axons with
energy substrates, such as lactate, during periods
of intense neuronal activity [64, 65] via monocar-
boxylate transporters (MCTs).
Numerous model systems have demonstrated
this communication channel of Schwann cell-­
axon crosstalk through MCTs [60, 66, 67], which
are currently believed to be involved in axo-glial
metabolic coupling in the peripheral nervous sys-
tem [68]. In an axotomy model of peripheral
nerve injury, Schwann cells stabilize neurons and
slow axon degeneration post injury by increasing
glycolysis and transferring lactate to axons via
MCTs [69]. Schwann cell-restricted knockout of
glycolytic enzymes or GLUT1 renders Schwann
cells incapable of lactate production and blocks
their protective effect in axotomized nerves. This
occurs without affecting nerve myelination,
indicative specifically of Schwann cell-axon met-
abolic crosstalk. MCT1 is widely expressed by
Schwann cells, but not by sciatic nerve axons
[70]. Sciatic nerve axon crush in mice with het-
erozygous MCT1 knockout slows nerve regener-
ation and recovery of neuronal compound motor
action potentials, indicating MCT1 is important
to neuronal function. In neurodevelopmental
models, Schwann cell-restricted MCT1 ablation
leads to defects in motor end-plate innervation,
although it does not impact motor and sensory
nerve conduction velocities nor myelination [71].
285
The impact of MCT1 knockout may be modest
under baseline conditions, but aggravated in neu-
ropathic conditions. For instance, Schwann cell-­
specific heterozygous MCT1 knockout in mice
with STZ-induced T1D worsens PN, assessed by
tail nerve conduction velocity [72]. Complete
Schwann cell-specific MCT1 deletion in mice
preferentially impairs sensory over motor nerve
function, but also generates a hypomyelinated
phenotype [73]. Furthermore, MCT1 deletion
disrupts lipid synthesis and glycolysis in Schwann
cells, which revert to a less mature state.
Overall, evidence indicates an intimate bioen-
ergetics cooperation between axon function and
Schwann cells, suggesting prediabetic and dia-
betic PN progression may be noncell autono-
mous. Thus, metabolic insults that are injurious
to Schwann cells may also contribute to PN.
2.3 Mitochondrial Dynamics
in the Nerve and PN
In order for axons to function properly, they
require an extensive amount of energy, which
they obtain from mitochondria, the “power-
house” of the cell. As a result, well-coordinated
mitochondrial dynamics are essential to neurons,
and comprise biogenesis (fission/fusion bal-
ance), trafficking, and mitophagy. Mitochondrial
biogenesis occurs primarily in the cell body, fol-
lowed by trafficking of newly generated mito-
chondria to locations of high energy demand
(Fig. 2b) [74]. This transport can occur over very
long distances, as is the case for the sciatic nerve
in humans, which extends over 3 feet. Once
mitochondria arrive at their destination, e.g.,
nodes of Ranvier, growth cones, axonal branch-
ing points, or synaptic terminals, they start to
participate in  local energy production through
oxidative phosphorylation [75–78]. Additionally,
local mitochondrial biogenesis can occur, e.g., in
neurites, although the extent this occurs remains
a point of continued research [79–81]. Terminally
damaged mitochondria, generally from exces-
sive oxidative stress, are eliminated by mitoph-
agy to clear the way for new, functioning
mitochondria [82].
286
Since mitochondria are crucial to maintaining
proper nerve function, mitochondrial dysfunction
and bioenergetics failure are central to and con-
sistently associated with PN pathogenesis
(Fig.  3a) [88–90]. Diabetes and the metabolic
syndrome affect all aspects of mitochondrial
dynamics in neurons. Mitochondrial length and/
or size are regulated by the balance between fis-
b
Fig. 3 Diabetic peripheral nerve bioenergetics. (a)
Sensory neurons undergo distal-to-proximal axon degen-
eration due to bioenergetics and mitochondrial trafficking
breakdown and buildup of reactive oxygen species (ROS)
and lipotoxic species. (b) Under hyperglycemic condi-
tions, glucose (GLU) accumulates in axons through
insulin-­independent glucose transporters (GLUT).
Excessive glucose boosts flux through the polyol and hex-
osamine pathways, increases protein crosslinking through
advanced glycation end products (AGEs) and receptor of
AGEs (RAGE) activation, and raises diacylglycerol levels
and protein kinase C (PKC) activation. Neurons can also
utilize fatty acids as a fuel substrate; however, large influx
and catabolism of fatty acids under dyslipidemic condi-
tions increases ROS generation and lipotoxic acylcarni-
tine production from mitochondria (inset C). Excessive
fatty acids (FA) also impair mitochondrial trafficking,
depolarize mitochondria, and alter their bioenergetics [83,
84], leading to immobile, damaged mitochondria (colored
brown). Under hyperglycemic conditions, GLU uptake
S. A. Eid et al.
sion versus fusion. The former is generally
favored in the presence of severe stress or nutri-
ent excess, the latter usually during nutrient star-
vation [82]. For example, hyperglycemia induces
mitochondrial fission in dorsal root ganglia in
vitro and the dorsal and ventral root axons of
db/db T2D mice in vivo [91]. On the other hand,
dyslipidemia alters the fission–fusion balance
c
may be impaired in Schwann cells since they express
insulin-dependent GLUTs. Under dyslipidemic condi-
tions, Schwann cells have increased expression of ACSL1
(acyl-CoA synthetase long-chain family member 1; inset
(c)) and carnitine palmitoyltransferase (CPT1), concomi-
tant with a boost in production of acylcarnitines, oxidative
stress, and apoptosis [85]. Mitochondrial dysfunction in
Schwann cells shifts lipid metabolism from lipogenesis to
lipolysis with a consequent increase acylcarnitines, which
can diffuse into axons [86]. Acylcarnitines can also trigger
a decrease in endoplasmic reticulum calcium stores [87]
and an increase in intracellular calcium [86]. (c) Impaired
bioenergetics and altered metabolic flux through the tri-
carboxylic acid (TCA) cycle and β-oxidation increases
ROS production and lipotoxic intermediates. IMM inner
mitochondrial membrane, LAC lactate, MCT monocar-
boxylate transporter, MPC mitochondrial pyruvate carrier,
OMM outer mitochondrial membrane. Created with
BioRender.com
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
[83, 84]. Nutrient overload also influences mito-
chondrial trafficking (Fig.  3b). For example,
excess saturated fatty acids, a hallmark of dyslip-
idemia, impairs mitochondrial axonal transport
in dorsal root ganglia neurons in vitro [83, 84]. In
vivo, HFD induces obesity and prediabetes in
mice and alters mitochondrial dynamics, includ-
ing energy production and trafficking in saphe-
nous nerves [29].
Interestingly, Charcot-Marie-Tooth disease,
the most prevalent inherited neuropathy [74], is
also a length-dependent polyneuropathy and is
caused, in a large subset of patients, by mutations
to genes regulating mitochondrial dynamics,
including bioenergetics, trafficking, fission, and
fusion. Mitochondrial dysfunction is also present
in chemotherapy-induced peripheral neuropathy
[92, 93], which manifests in a glove and stocking
pattern. Thus mitochondrial dysfunction is a cen-
tral theme in not only prediabetic and diabetic PN
in both mouse models and humans but also in
other well-established neuropathic disorders.
2.4 Mitochondrial Oxidative
Stress in the Nerve and PN
Mitochondria are a major source of reactive oxy-
gen species, such as superoxide anions, which
can be produced unintentionally during oxidative
phosphorylation [94], even under normal basal
conditions. Cellular antioxidant defense mecha-
nisms scavenge these reactive oxygen species to
maintain cellular redox homeostasis and prevent
oxidative damage [95, 96]. However, oxidative
stress occurs when reactive oxygen species pro-
duction outstrips the ability of cells to mount an
antioxidant defense. Reactive oxygen species
production from mitochondrial respiration is
influenced by the mitochondrial membrane
potential. Low mitochondrial membrane ­potential
slows respiration and reduces reactive oxygen
species production [97], while high mitochon-
drial membrane potential boosts respiration and
increases reactive oxygen species generation
[98]. The mitochondrial membrane potential is
related to substrate flux and level of oxidative
phosphorylation through mitochondria. Under
287
diabetic or the metabolic syndrome conditions,
hyperglycemia stimulates mitochondrial oxida-
tive phosphorylation, leading to reactive oxygen
species overproduction, mitochondrial oxidative
damage, and compromised intracellular pro-
cesses (Fig. 3c) [98]. In dorsal root ganglia neu-
rons, hyperglycemia induces reactive oxygen
species production and oxidative stress, as well
as caspase-dependent apoptosis [99].
Reactive oxygen species overproduction in
diabetes and the metabolic syndrome addition-
ally inhibit key mitochondrial enzymes and con-
tribute further to mitochondrial dysfunction and
damage [100]. Reactive oxygen species, such as
superoxide and hydrogen peroxide, inhibit a key
tricarboxylic acid cycle enzyme, aconitase [100,
101], which disrupts the tricarboxylic acid cycle.
It also results in α-ketoglutarate dehydrogenase
(α-KGDH) overactivation that further truncates
the tricarboxylic acid cycle [100]. α-KGDH itself
is a source of reactive oxygen species, which fur-
ther exacerbates mitochondrial oxidative damage
[100].
Fortunately, peripheral nervous system
Schwann cells possess a high antioxidant capac-
ity and can detoxify reactive oxygen species to a
certain limit, even under conditions of
hyperglycemia-­induced metabolic dysregulation
[102]. Schwann cells are extremely resilient to
hyperglycemia and only become susceptible to
mitochondrial injury at very high glucose con-
centrations (150 mM) [103]. Furthermore, the
basal activity of antioxidant enzymes in Schwann
cells rivals the maximum stimulated activities of
the same enzymes in dorsal root ganglia neurons
[102].
In summary, neurons and their axons are
highly reliant on normal mitochondrial dynamics
and are susceptible to injury if mitochondrial
function is disrupted by nutrient overload.
Hyperglycemia and/or dyslipidemia shifts the
dynamics from a homeostatic to a pathological
state, disrupting the balance in mitochondrial fis-
sion/fusion, trafficking, and reactive oxygen spe-
cies production. These aspects are present in both
diabetic, type 1 and 2, and prediabetic obesity
models, highlighting mitochondrial dysfunction
generally as a PN hallmark.
288
2.5 Glucose Excess
and Inflammation
in the Nerve and PN
As mentioned above, neurons mostly express
insulin-independent GLUT3 [42]; therefore,
insulin-resistant, hyperglycemic, or impaired
glucose tolerance conditions do not affect glu-
cose transport into neurons, resulting in neuronal
glucose accumulation [104–106]. This glucose
buildup has numerous downstream effects
(Fig. 4), which ultimately lead to neuronal dam-
age [107, 108].
One well-known pathway leading to injury is
the polyol pathway. Glycolysis commences in the
cytoplasm by hexokinase, which phosphorylates
glucose, committing it to glycolysis [109–111].
However, glucose excess outstrips hexokinase
activity and glycolytic capacity, leading to surplus
glucose, which is instead converted to sorbitol by
aldose reductase through the polyol pathway [106].
The accumulating sorbitol disrupts the osmotic
balance inside axons, inducing osmotic stress and
an efflux of myo-inositol and taurine in an effort to
restore osmotic balance [112]. Myo-inositol is
essential for Na+/K+ ATPase channel functioning;
thus, myo-inositol loss disrupts Na+/K+ ATPase
function and consequently membrane resting
potential and axon electrophysiology [36].
Additionally, aldose reductase utilizes NADPH as
a cofactor; hence, excessive aldose reductase activ-
ity depletes NADPH energy stores, interfering with
the synthesis of nitric oxide, an important signaling
molecule, and glutathione, an antioxidant defense
molecule [36]. Antioxidant glutathione deficiency
exacerbates the damaging effects of reactive oxy-
gen species, mediating further cellular damage.
This “metabolic flux” due to polyol pathway
activation is considered a central pathophysio-
logic mechanism of metabolic PN [112]. In addi-
tion, aldose reductase activation decreases ATP
production, precipitates neuronal injury [113]
and interferes with energy influx from Schwann
cells [109, 110, 114]. Further literature supports
the deleterious effects of aldose reductase hyper-
activation and the polyol pathway [115]. In mice
with STZ-induced T1D PN, lack of aldose reduc-
tase preserves nerve function [116], while
S. A. Eid et al.
o­ verexpression worsens outcomes [117].
Disappointingly, pharmacological inhibition of
aldose reductase in 32 clinical trials conducted
during the 1990s failed to improve diabetic PN
outcomes [118], which may have arisen from the
inability of these inhibitors to reach the periph-
eral nervous system [119].
Another prominent pathway stemming from
axonal glucose accumulation, which leads to
nerve injury, is the advanced glycation end prod-
ucts (AGEs) pathway. Glucose buildup in the
nerve under hyperglycemic conditions can go
through the Maillard reaction by reacting with
protein amino groups to give rise to Schiff bases
or Amadori products. These rearrangement, gen-
erating reactive groups, which stably crosslink
proteins to generate AGEs in the nerve [36, 120,
121]. AGEs cause cellular damage by crosslink-
ing critical proteins, inhibiting their function.
They also bind to a specific receptor, RAGE
(receptor for AGEs), which triggers a signaling
cascade leading to nuclear factor-kappa B (NF-­
κB) activation and inflammation, vasoconstric-
tion, and loss of neurotrophic support in the
peripheral nervous system [122].
Excess glucose can also increase flux through
the hexosamine pathway. During the first few gly-
colytic steps, glucose is converted to fructose-­6-­
phosphate, which can either remain within
glycolysis or be directed into the hexosamine path-
way. Although the shunt into the hexosamine path-
way normally occurs at a low rate, it dramatically
increases under hyperglycemic conditions [107].
The hexosamine pathway utilizes fructose-­ 6-­
phosphate and glutamine in the first step, ultimately
producing uridine 5'-diphosphate-N-acetylglucos-
amine (UDP-­GlcNac). UDP-GlcNac binds to ser-
ine/threonine residues of multiple proteins, a
process termed O-GlcNAcylation, altering their
function and causing downstream injury. For
instance, O-GlcNAcylation of specificity protein 1
(Sp1), a key transcription factor, disrupts cellular
homeostasis by activating transforming growth
factor-β (TGF-β) and plasminogen activator inhibi-
tor-1 (PAI-1), resulting in peripheral nerve inflam-
mation [123].
Finally, another glycolytic intermediate
involved in neuronal dysfunction in hyperglyce-
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage 289

a b

Fig. 4  Hyperglycemia and inflammation in the nerve and pathway: Excess F-6-P can exit glycolysis and react with
PN. (a) Under homeostatic conditions, insulin-­ glutamine, ultimately producing uridine 5′-diphosphate-
independent glucose transporter 1 (GLUT1, green trans- N-­acetylglucosamine (UDP-GlcNac). UDP-GlcNac binds
porter) imports glucose (GLU) into axons. GLU is to serine/threonines altering protein function and causing
phosphorylated by hexokinase (HK) into fructose-6-­ downstream injury, e.g., of key transcription factors (TF),
phosphate (F-6-P), committing it to glycolysis and even- disrupting cellular homeostasis by activating transforming
tual pyruvate (PYR) production. The PYR is transported growth factor-β (TGF-β) and plasminogen activator inhib-
into mitochondria, where it enters the tricarboxylic acid itor-­1 (PAI-1), resulting in peripheral nerve inflammation.
(TCA) cycle followed by oxidative phosphorylation (3) Advanced glycation end products (AGEs) pathway:
(OXPHOS) to generate energy. Mitochondria remain vital Excess GLU can go through the Maillard reaction eventu-
and functioning (red). (b) Under hyperglycemic condi- ally giving rise to AGEs, which stably crosslink proteins,
tions, excess GLU floods neuron axons; the subsequent inhibiting their function. AGEs also bind to a specific
increased catabolic flux through the TCA cycle increases receptor, RAGE (receptor for AGEs), triggering signaling
reactive oxygen species (ROS) production, damaging leading to nuclear factor-kappa B (NF-κB) activation,
mitochondria (brown). GLU accumulation also boosts resulting in inflammation, vasoconstriction, and loss of
flux through multiple pathways. (1) Polyol pathway: neurotrophic support in the peripheral nervous system. (4)
When GLU excess outstrips HK activity and glycolytic Protein kinase C (PKC) activation pathway: The accumu-
capacity, surplus GLU is instead converted to sorbitol lating glycolytic intermediate, dihydroxyacetone phos-
(SOR) by aldose reductase (AR). Accumulating SOR dis- phate (DHAP), is converted to diacylglycerol (DAG),
rupts axon osmotic balance, leading to myo-inositol and which in turn activates protein kinase C (PKC), leading to
taurine efflux. Myo-­inositol is essential for Na+/K+ ATPase nerve dysfunction, through insulin resistance, impaired
channel functioning, disrupting membrane resting poten- Na+/K+ ATPase, hypoxia, TGF-β overexpression, and dis-
tial and axon electrophysiology. Additionally, excessive rupted vascular endothelial growth factor. IMM, inner
AR activity depletes NADPH, interfering with antioxidant mitochondrial membrane; MPC, mitochondrial pyruvate
defense, e.g., glutathione (GSH) synthesis, further exacer- carrier; OMM, outer mitochondrial membrane. Created
bating ROS levels and oxidative stress. (2) Hexosamine with BioRender.com
290
mia is dihydroxyacetone phosphate (DHAP).
Upon accumulation, DHAP is converted to diac-
ylglycerol, which in turn activates protein kinase
C (PKC) [124]. Activated PKC is responsible for
multiple metabolic and inflammatory pathways
that can result in nerve dysfunction, such as insu-
lin resistance, impaired Na+/K+ ATPase, hypoxia,
TGF-β overexpression, and vascular endothelial
growth factor disruption [125]. Once again, the
idea of using a PKC inhibitor to treat diabetic PN
failed in clinical trials [125]. Similar to the clini-
cal narrative for aldose reductase inhibitors, the
failure of PKC inhibitors was attributed to the
inability of the drugs to penetrate into the periph-
eral nervous system [126].
3 Dyslipidemia Is Associated
with PN
Although glycemic control is generally effective
for preventing or slowing PN progression in T1D,
the risk of PN persists in prediabetic and T2D
patients, even with well-controlled glycemia
[127]. Dyslipidemia, commonly seen clinically in
both prediabetes and T2D patients, has recently
emerged as an independent PN risk factor [8, 128–
131]. Dyslipidemia refers to a range of plasma
lipid abnormalities, characterized by the triad of
increased triglycerides and low-density lipopro-
tein cholesterol (LDL-c) and decreased high-den-
sity lipoprotein cholesterol (HDL-c) [132].
Several clinical studies have identified dyslip-
idemia, either in sum or through its individual
components (elevated triglycerides and LDL-c or
lowered HDL-c), to contribute to PN risk (Table 1).
This is perhaps best exemplified by the Danish
arm of the Anglo-Danish-Dutch study of Intensive
Treatment of Diabetes in Primary Care
(ADDITION), a longitudinal study of 1256 T2D
participants without PN at baseline, which were
assessed for incident PN at a mean follow-up of 13
years [12]. Low baseline HDL-c was protective for
future PN, whereas higher waist circumference
and body mass index were predictive of future
PN.  Another study conducted in Denmark, the
Danish Centre for Strategic Research in Type 2
Diabetes (DD2) Cohort of 5249 T2D participants
S. A. Eid et al.
additionally identified elevated triglycerides and
anthropometric measures waist-to-hip and waist-
to-height ratios as PN risks [13]. We found that
HDL associated with multiple secondary PN out-
comes in the Health, Aging, and Body Composition
(Health ABC) study [3] and triglycerides were a
stepwise risk factor for nontraumatic lower-
extremity amputations [7]. In a population of
youths, lower HDL-c was a PN risk factor in T2D
[5]; notably, however, higher LDL-c and triglycer-
ides and lower HDL-c were PN risk factors in T1D
youths, although this cohort was significantly less
obese than the T2D.  Therefore, dyslipidemia in
patients with T1D may also have implications for
PN onset, although this has not been extensively
studied. Patients with T1D historically suffer from
poor weight, but obesity, as in T2D, is similarly
becoming more prevalent in T1D [133, 134].
It is also important to note that obesity is intri-
cately tied to dyslipidemia through fat depots and
energy balance and flux in the body [135, 136]. It
is therefore unsurprising that multiple measures
of obesity, such as body mass index and waist cir-
cumference, have also correlated with PN risk
(Table 1). Specifically, central obesity is the most
significant PN driver [2, 4] compared to other
anthropometric measures. This is anticipated,
since the visceral, intra-abdominal fat depot is
more metabolically impaired, hypertrophied, and
pro-inflammatory, compared to other depots,
such as gluteal, femoral, and subcutaneous [135,
137–140]. Thus, an understanding of systemic
energy balance and lipid trafficking between the
visceral fat depot and circulation is relevant to
dyslipidemia and downstream events.
The association of dyslipidemia, as well as
obesity, with PN is strengthened by preclinical
studies in HFD-induced obese mice, which
develop dyslipidemia and PN independent of
hyperglycemia [26, 28, 141]. Overall, these find-
ings suggest that the underlying PN pathophysi-
ology may differ in T1D versus T2D [127], and
this increasingly may be the more accurate per-
spective that dyslipidemia may add an additional
layer of pathological processes over those
induced by hyperglycemia alone. In fact, glucose
and lipid metabolism acts in concert systemically
in the body overall; thus, an understanding of
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
systemic metabolic dynamics is needed to impart
an understanding in target organs.
As a more recent avenue of investigation, the
mechanisms underlying the effects of dyslipid-
emia and altered lipid metabolism on PN remain
incompletely understood. This section will
review several aspects of lipid metabolism, start-
ing from global systemic effects of dyslipidemia
and circulating lipid levels, to the localized
impact of dyslipidemia in the peripheral nervous
a b
Fig. 5  Lipid metabolism in obesity and dyslipidemia. (a)
Under homeostatic conditions, a balance exists between
energy production and energy expenditure. Surplus
energy is stored in adipose tissue depots, which act as
short- and long-term storage units and as endocrine
sources. Under excessive caloric intake, visceral fat depot
hypertrophies and inflames (red inset, immune cell
recruitment and activation), increasing circulating free
fatty acids (FA) and pro-inflammatory adipokines. This
causes insulin resistance, triglyceride release from the
liver, and ectopic fatty deposition (shown in the fatty
liver). The changes in circulating lipids leads to dyslipid-
emia and the metabolic syndrome. (b) Systemic dyslipid-
emic conditions have localized effects in the nerve (yellow
inset), e.g., lipid uptake (inset (c)). (c) Possible dysregu-
lated aspects of lipid metabolism in the nerve in PN (blue
inset). (1) De novo lipogenesis: The LXR-SREBP1-FAS
axis (green) is dysregulated in PN, impairing lipid synthe-
sis. (2) Lipid uptake: Increased expression of fatty acid
transport proteins like CD36 (gray transporter) occurs in
291
system, beginning with lipid uptake, lipogenesis,
and lipotoxic species accumulation. Next, within
the nerve, we will discuss the cellular and molec-
ular influence of lipids on mitochondrial bioener-
getics and trafficking, oxidative stress, and
inflammation, and their implications on nerve
health and PN (Fig.  5). Whenever possible, we
will also highlight cell-specific behavior, e.g.,
neuronal, Schwann cells, and potential lipid-­
centric axo-glial metabolic crosstalk.
PN. (3) Lipid accumulation and lipotoxic species: Lipid
build-up and upregulated lipid metabolism occurs in neu-
ropathic nerve. High metabolic flux promotes lipid inter-
mediate formation, e.g., ceramides and diacylglycerols.
Schwann cells (SC) overexpress ACSL1 (acyl-CoA syn-
thetase long chain family member 1), which boosts acyl-
carnitine production. (4) Dyslipidemia induced
inflammation: Saturated fatty acids (SFA) activate toll-­
like receptors (TLR); dyslipidemia increases oxidized
low-density lipoproteins (oxLDLs), which activate lectin-­
like oxLDL receptor (LOX-1), the receptor for oxLDLs;
these interactions increase inflammation downstream. (5)
Dyslipidemia impairs mitochondrial bioenergetics and
dynamics: Dyslipidemia impairs bioenergetics, biogene-
sis, trafficking (stop sign over microtubule motor protein),
and reactive oxygen species production, causing mito-
chondrial damage (brown), ultimately leading to PN. FAS,
fatty acid synthase; LXR, liver X receptor; SREBP1, ste-
rol regulatory element-binding proteins 1. Created with
BioRender.com
292
3.1 Global Energy Imbalance
and Metabolic Disruption
During Dyslipidemia
Body weight is maintained by a balance between
energy production and energy expenditure.
Surplus energy substrate from extra food intake
is stored in adipose tissue depots, which act both
as short- and long-term storage units and as endo-
crine sources by releasing adipokines [142, 143].
Adipokines constitute a type of cytokines and
hormones regulating appetite, insulin sensitivity,
inflammation, and immune cell recruitment [142,
144]. During periods of fasting or starvation, adi-
pose tissue releases energy in the form of free
fatty acids, which are utilized by other organs for
energy production [143]. This normal physiolog-
ical process occurs as long as the adipose tissue is
healthy and caloric intake is balanced against
energy expenditure.
In case of excessive caloric intake, adipocytes
enlarge (hypertrophy) and proliferate ­(hyperplasia)
(Fig.  5a). Particularly, visceral fat depot adipo-
cytes hypertrophy, impairing their ability to store
fat [135, 137–140] and activating adipose triglyc-
eride lipase, which increases free fatty acid release
from adipocytes [145]. The rise in circulating free
fatty acids contributes to insulin resistance and
dyslipidemia along with intracellular fatty acid
accumulation in target tissues or organs [146,
147]. Additionally, the hypertrophied visceral adi-
pose tissue inflames along with resident macro-
phage activation [139]. This alters adipokine
secretion from the visceral depot, leading to ele-
vated systemic inflammation, another mechanism
causing insulin resistance [148]. Since insulin
promotes hepatic fatty acid esterification into tri-
glycerides and their release into the circulation
[149], insulin resistance also leads to changes in
plasma triglycerides, which correlate with T2D
risk [150]. Thus, overall, enlarged, hypertrophied
visceral fat depots impair insulin sensitivity and
induce dyslipidemia and the metabolic syndrome
by altering the circulating lipid profile [135].
We can infer a few critical points from these
observations. One, the global systemic effects
underscore the close relationship between glucose,
e.g., insulin resistance, and lipid metabolism, e.g.,
S. A. Eid et al.
circulating free fatty acids. Thus, dyslipidemia and
hyperglycemia work in tandem to exacerbate meta-
bolic dysfunction, with consequent dual effects on
the nerve. Two, obesity, e.g., through a hypertro-
phied visceral depot, elevates free fatty acids and
pro-inflammatory cytokines, which can ultimately
affect the nerves and promote PN [135]. Three,
elevated free fatty acids can be converted by reac-
tive oxygen species to toxic metabolites, such as
lipid peroxides, leading to systemic inflammation
and localized neuronal apoptosis [99]. Four, fatty
acids released from visceral fat depots can accumu-
late ectopically in various tissues leading to dam-
age, e.g., muscle, liver as fatty liver disease, as well
as in the nerve (Fig. 5b) [28]. Cumulatively, sys-
temic global energy imbalance can result in a
vicious feed forward cycle of mounting inflamma-
tion, insulin resistance, and dyslipidemia, culmi-
nating in peripheral nerve injury.
3.2 De Novo Lipogenesis
in the Nerve and PN
Systemic energy consumption is critical to our
understanding of dyslipidemia development; but,
equally important is a thorough understanding of
energy utilization at the level of the nerve itself.
Lipids are required in the nerve for manifold pur-
poses. Perhaps foremost, Schwann cells utilize
lipids to generate myelin, a structural and insulat-
ing material, which facilitates axon signal trans-
duction [41]. Lipids are also essential components
of organelles, e.g., mitochondria, and influence
membrane composition and fluidity [151, 152],
which affect nerve conductance [153].
Additionally, neurons express fatty acid trans-
porters [40] and can utilize fatty acids as fuel
substrates in β-oxidation to generate energy for
nerve function.
Under homeostatic conditions, neurons and
their supporting glia acquire lipids through two
mechanisms: de novo lipogenesis and/or uptake
of plasma lipoproteins or fatty acids. There is
currently an increased understanding of the
capacity for Schwann cells to synthesize fatty
acids de novo, which are required for myelina-
tion, a dynamic process with high-lipid demands
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
[154, 155]. Schwann cells express fatty acid syn-
thase (FAS), the key rate-limiting step in de novo
lipogenesis, which is essential for the proper
onset and efficiency of myelination [154]. FAS
activity is regulated by sterol regulatory element-­
binding proteins 1 (SREBP1) [156], which works
hand-in-hand with the liver X receptors (LXR),
an established regulatory pathway of fatty acid
lipogenesis [157]. LXR has well-characterized
roles in maintaining fatty acid homeostasis in
metabolically active tissues [158].
Pathologic conditions dysregulates the process
of de novo lipogenesis and the LXR-­SREBP1-­
FAS axis in nerves, which is evident from PN
models (Fig. 5c). A recent study found that inhibit-
ing FAS adversely affected lipid synthesis and
axonal growth in neuronal cultures [159]. SREBP1
and FAS expression are disrupted in sciatic nerve
from rodents with T1D-induced PN and in cul-
tured Schwann cells exposed to metabolic insults
[160, 161]. LXR activation is associated with anti-
inflammatory, anti-­ apoptotic, and antioxidative
properties in diabetic complications [162, 163].
Gavini et al. showed that LXR is transcriptionally
active in dorsal root ganglia neurons, and its acti-
vation reduces saturated fatty acid-induced endo-
plasmic reticulum stress and allodynia in a HFD
prediabetic mouse model of PN [164]. These find-
ings are complemented by comparative transcrip-
tomics profiling in peripheral nerves of mice and
humans with PN, which demonstrate that the LXR
pathway is dysregulated in PN and conserved
across species in T1D and T2D [165]. Importantly,
emerging data have demonstrated a role for LXR
activation in reducing reactive oxygen species
generation and preserving Schwann cell integrity
in T1D PN [166]. Together, these data suggest that
a dysregulated LXR-SREBP1-FAS axis, by
impairing crucial lipogenesis, may be involved in
PN, an area requiring further investigation.
3.3 Lipid Uptake in the Nerve
and PN
In addition to de novo lipid synthesis, neurons
and Schwann cells acquire lipids from the circu-
lation. Under homeostatic conditions, triglycer-
293
ides are initially hydrolyzed by lipoprotein lipase
(LPL) in endothelial cells, which are widely
found in adipose tissue and skeletal muscle. This
releases non-esterified free fatty acids, which are
taken up either passively by diffusion or facili-
tated by transporters, like CD36 and fatty acid
transport proteins (FATP) expressed in multiple
tissue types, including peripheral nerves [41].
Subsequently, acquired fatty acids are either
incorporated into myelin in Schwann cells,
directly oxidized for energy production in mito-
chondria, or stored in lipid droplets [66, 167].
While lipid uptake for myelin has been exten-
sively studied and recently reviewed [41], less is
known about its pathological implications in
PN. The peripheral nervous system is exposed to
excess fatty acids in prediabetes and T2D, since
most patients also have dyslipidemia. Additionally,
early studies suggested that dysregulated LPL
activity may contribute to lipotoxicity in PN [168,
169]. Altered CD36 expression is another poten-
tial mechanism for increased lipid delivery into
the nerve, which we have consistently shown is
elevated in murine and human PN (Fig. 5c) [28,
170, 171]. Concomitant increases in mitogen acti-
vated protein kinase (MAPK) signaling in neuro-
pathic peripheral nerves from db/db support a role
for CD36  in mediating a MAPK-dependent
inflammatory response [170]. Peroxisome
proliferator-­activated receptor-gamma coactivator
(PGC) 1α, a master regulator of mitochondrial
biogenesis, potentially controls β-oxidation
through its actions on the fatty acid transporter
CD36 [172]. We have shown increased PGC1α
[29] and CD36 expression in murine and human
PN [28, 170, 171], suggesting dysregulation of
lipid uptake and β-oxidation occurs in PN.
Overall, lipid accumulation may occur in
nerves resulting from the combination of excess
circulating fatty acid levels and overexperssion of
uptake transporters, which could, in part, under-
lie PN pathogenesis, a view that requires direct
confirmation in human PN. Furthermore, manip-
ulating key lipogenic targets in sensory neurons
of PN-relevant mouse models is critical to under-
stand the effect of de novo lipogenesis and lipid
uptake on axo-glial interactions, independent of
myelination.
294 S. A. Eid et al.

3.4 Dyslipidemia, Lipid found that the overall lipid content increased in
Metabolism, and Lipotoxic peripheral nerves of T2D mice, but with few sig-
Species in the Nerve nificant correlations in specific lipid species with
During PN plasma. However, there was an intra-class correla-
tion between plasma and nerve triglycerides [175],
Excess lipid accumulation and metabolic flux which are also associated with PN progression in
increase the likelihood that lipotoxic species the experimental and clinical settings [28, 129].
form, especially when fatty acid uptake outpaces These findings elucidate a complex interaction
β-oxidation [173, 174]. Therefore, lipid accumu- between plasma and nerve lipid levels and reveal
lation in peripheral nerves, as may occur during patterns of altered lipid metabolism, including tri-
obesity and dyslipidemia [28], could lead to lipo- glyceride classes, which could be used as PN bio-
toxic species generation. Preclinical studies in markers in the context of prediabetes and T2D.
PN-relevant culture and mouse models are begin- Lipid build-up directly promotes increased lipid
ning to provide insights into these mechanisms in intermediate levels, such as diacylglycerols [176]
PN, ranging from detecting lipid accumulation, and ceramides [177, 178], which could exert lipo-
overrepresentation of fatty acid metabolism, and toxic downstream effects. Indeed, this is the case in
the generation of lipotoxic species. neuropathic nerve from prediabetic and T2D mice,
Omics technologies are aiding these investiga- which accumulates diacylglycerols [28], and has
tions by interrogating dysregulated pathways in several downstream effects relevant to PN, includ-
neuropathic nerve. For instance, transcriptomics ing PKC activation [125]. Clinical studies of lipo-
profiling of peripheral nerves from db/db T2D toxic species in nerve are lacking in humans, given
mice identified fatty acid metabolism as one of the constraints of tissue accessibility, although cor-
the highly enriched pathways in PN [170]. This relation of lipotoxic plasma species with PN have
result has been consistently replicated in human been performed. Given the relevance of ceramides
sural nerve biopsies and robust HFD-induced pre- to obesity and drawing parallels to hereditary neu-
diabetes and T2D mouse models [28, 165]. ropathies [179], we analyzed plasma
Lipidomics analysis of sciatic nerve from HFD 1-­deoxydihydroceramide levels, which increased
prediabetes and HFD-STZ T2D mice revealed step-wise from lean controls, to obese participants,
abnormal accumulation of saturated fatty acid-­ to obese T2D participants, and were highest in
containing triglycerides, demonstrating that per- obese T2D participants with DN [180]. We also
sistent dyslipidemia increases lipid delivery to the conducted an untargeted metabolomics study of
nerve [28]. Integrating lipidomics with transcrip- plasma from T2D participants to identify species
tomics in neuropathic nerves identified elevated correlating with PN status [181]. Although few
expression of diacylglycerol acyltransferase 2 species differed significantly, an overview analysis
(DGAT2), the enzyme required for the last and of all plasma lipids identified a shift in abundance,
committed step in triglyceride synthesis, a finding chain length, and saturation level of fatty acids and
mirrored in sural nerve biopsies from hyperlipid- numerous complex and lipotoxic species, includ-
emic T2D patients with PN [28]. These adverse ing ceramides and acylcarnitines, by T2D and PN
effects were improved by dietary reversal from status. The influence of complex lipid chain length
saturated fatty acid-rich HFD back to a standard and saturation has also been implicated in diabetic
low-fat diet, underscoring the therapeutic benefit nephropathy [182]. Therefore, moving forward,
of restoring nerve lipid homeostasis in PN. identifying these trends and potential specific spe-
To better understand how systemic dyslipid- cies may unlock therapeutic avenues, e.g., inhibi-
emia affects localized nerve lipid metabolism, we tors of complex lipid biosynthesis or signaling,
conducted an integrated transcriptomic and lipido- such as ceramides, obesity treatments [183, 184],
mic analysis of plasma and peripheral nerves from and potential targets for PN treatment.
db/db T2D mice to identify alterations in shared Growing interest in the non-cell autonomous
and unique lipid species by tissue types [175]. We nature of PN and emergent studies are starting to
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
point toward a potential axo-glial communication
in a lipid-mediated manner [66, 185], including
lipotoxic species transfer between Schwann cells
and axons. This concept is reinforced by our results,
which demonstrate that Schwann cells increase
ACSL1 expression in vivo in db/db sciatic nerve
and in vitro under dyslipidemic conditions [85].
This was accompanied by elevated expression of
mitochondrial fatty acid transporters, CPT1a and
CPT1b, concommitant with a boost in production
of acylcarnitines, which may be lipotoxic [186,
187]. We also found that dyslipidemia impaired
Schwann cells mitochondrial bioenergetics and
induced oxidative stress and apoptosis in vitro [85].
In mice, Schwann cell-restricted knockout of
TFAM (mitochondrial transcription factor A gene),
a transcription factor that regulates mitochondrial
DNA replication and downstream oxidative phos-
phorylation protein production, does not affect
Schwann cell survival, but does induce progressive
PN [86]. Loss of TFAM disrupts Schwann cell
lipid metabolism, which shifts from lipogenesis to
lipolysis, with a consequent accumulation of
medium- and long-chain acylcarnitines in sciatic
nerves. In turn, explants of these nerves release
multiple long-chain acylcarnitine species into cul-
ture media, suggesting these acylcarnitnes could
diffuse from Schwann cells into axons in vivo.
Furthermore, treating dorsal root ganglia cultures
with the acylcarnitine palmitoyl-carnitine signifi-
cantly elevates intracellular calcium and induces
axonal blebbing and degeneration upon long-term
incubation [86]. Overall, abnormal lipid metabo-
lism and consequent injury of Schwann cells, dis-
rupting normal axo-glial communication, could be
damaging to axons and lead to PN.
3.5 Dyslipidemia-Induced
Inflammation in the Nerve
During PN
Hypertrophied fat depots increase systemic
inflammation by releasing pro-inflammatory adi-
pokines, such as interleukin-6 (IL-6) and tumor
necrosis factor-α (TNFα) [140, 142, 144].
Locally, in neuropathic nerve, our transcriptomic
studies have consistently found enrichment of
295
genes and pathways related to inflammation and
immune system dysregulation in db/db T2D mice
[188]. A combined epigenomic-transcriptomic
analysis of human sural nerves from T2D patients
with progressive versus non-progressive PN
identified enrichment of immune system response
[189]. Comparing gene expression in neuropathic
peripheral nerves from db/db T2D mice versus
human also highlights inflammation as a highly
conserved pathway [165]. In our HFD prediabe-
tes mouse model of PN, transcriptomic analysis
of sciatic nerve identified involvement of toll-like
receptor (TLR) pathways [190]. TLR2 and TLR4
knockout delayed PN onset in obese prediabetic
mice, implicating inflammation in the early
stages of PN.  TLRs are receptors on innate
immune cells, which sense danger signals, e.g.,
pathogens, emitting pro-inflammatory signals.
Saturated fatty acids activate TLRs [191–193]
and constitute an immunometabolic link between
dyslipidemia and inflammation.
Additionally, dyslipidemia may indirectly
affect nerve function by mediating oxidative stress,
and altering downstream signaling pathways
[194]. For example, we previously showed that
dyslipidemia increases oxidized low-density lipo-
proteins (oxLDLs) in dorsal root ganglia, which in
turn interact with the lectin-like oxLDL receptor,
LOX-1 to induce reactive oxygen species genera-
tion and damage [141]. This mechanism may be
shared in other complication-prone tissues in dia-
betes, such as the retina and kidney [195, 196].
oxLDLs can also exert their apoptotic effects in
dorsal root ganglia neurons via TLR4 [197].
Therefore, altered energy metabolism, genera-
tion of toxic lipid intermediates, inflammation,
mitochondrial dysfunction, and oxidative stress
can act in concert in the dyslipidemic milieu to
promote PN.
3.6 Dyslipidemia Impairs
Mitochondrial Bioenergetics
and Dynamics in the Nerve
During PN
We have outlined how systemic dyslipidemia can
affect localized nerve lipid uptake, lipogenesis,
296
and accumulation, with consequent lipotoxic
intermediate generation and inflammation. In a
previous section, we also discussed how hyper-
glycemia in diabetes and the metabolic syndrome
elevate flux through the polyol and hexosamine
pathways, alter RAGE and PKC activation,
increase reactive oxygen species production, and
produce mitochondrial dysfunction, all key
mechanisms proposed for PN pathogenesis [107,
108]. With respect to comparing peripheral ner-
vous system mitochondrial bioenergetics in the
setting of excess glucose and lipids, the db/db
T2D mouse provides an optimal experimental
model, because these animals become hypergly-
cemic and dyslipidemic and develop PN [198,
199]. Treating db/db mice with pioglitazone, a
peroxisome proliferator-activated receptor
gamma (PPARγ) agonist, preserves nerve func-
tion, lowers lipid levels, and alters expression of
genes related to oxidative stress, fatty acid
β-oxidation, and the tricarboxylic acid cycle
[188, 200, 201]. However, fluxomics experiments
in sciatic nerves from db/db T2D mice offer the
best comparison of different energy substrates
[175, 202]. Administering db/db mice radiola-
beled [U-13C6] glucose revealed no change in gly-
colysis, but a significant decrease in tricarboxylic
acid cycle intermediates in neuropathic sciatic
nerve, underscoring mitochondrial dysfunction.
Using [13C16] palmitate, neuropathic sciatic
nerves had increased β-oxidation and acylcarni-
tine accumulation, followed by a drop in tricar-
boxylic acid cycle substrates. Thus, excess
glucose and lipids alter mitochondrial bioener-
getics in T2D nerves in distinct and complimen-
tary ways. Future fluxomic studies in the
diet-induced HFD prediabetic model would shed
more insight on nutrient flux and mitochondrial
bioenergetics specifically in the dyslipidemia set-
ting, an avenue warranting investigation.
Cell culture models have also proven useful
for increasing our understanding of mitochon-
drial bioenergetics and dynamics. Simulating
dyslipidemic conditions in vitro through long-­
chain saturated fatty acid palmitate treatment
depolarizes axonal mitochondria, reduces intra-
cellular ATP levels, induces oxidative stress, and
raises apoptosis of dorsal root ganglia neurons
S. A. Eid et al.
[83]. Dyslipidemia also affects mitochondrial
trafficking in sensory dorsal root ganglion axons,
decreasing the number of motile mitochondria in
the presence of palmitate [83, 84]. The effect is
chain-length-dependent; long-chain saturated
fatty acids palmitate or stearate impair mitochon-
drial transport, whereas medium-chain saturated
fatty acids laurate and myristate exert no effect
[83]. In contrast to saturated fatty acids, unsatu-
rated fatty acids are neuroprotective. For exam-
ple, the monounsaturated fatty acid oleate
prevents palmitate-induced mitochondrial defects
and apoptosis in vitro in dorsal root ganglia and
preserves nerve function in prediabetic HFD
mice [167]. Oleate is thought to mediate these
neuroprotective effects through lipid droplet for-
mation [167]. Polyunsaturated fatty acid diet
supplementation exerts similar protective effects
on PN in prediabetes and T2D rat models [203,
204].
We have also investigated the influence of
HFD in saphenous nerve in vivo in prediabetic
mice [29]. HFD affected the number, speed, and
size of mitochondria trafficking in the antero-
grade and retrograde directions. It also depolar-
ized axonal mitochondria and impaired their
ability to dissipate mitochondrial membrane
potential in response to physiological demand.
Moreover, HFD lowered sensory axonal calcium
levels, lengthened mitochondria, and increased
PGC1α expression, overall suggesting that mito-
chondrial dysfunction and an imbalance of axo-
nal energy underlies PN in prediabetes.
Schwann cell metabolism is also susceptible to
hyperglycemic and dyslipidemic conditions.
Hyperglycemia induces Schwann cell oxidative
stress, mitochondrial dysfunction, and apoptosis
in vitro [205–208] and in vivo in diabetes animal
models of PN [209]. Dyslipidemia also elicits
Schwann cell injury; treatment with the saturated
fatty acid palmitate in vitro lowers Schwann cell
endoplasmic reticulum calcium stores, followed
by mitochondrial membrane depolarization, and
an increase in oxidative stress [87]. Pretreatment
with elevated glucose levels potentiates the detri-
mental effect of palmitate; thus, similar to the sce-
nario in neurons, hyperglycemia and dyslipidemia
may synergize to induce Schwann cell damage.
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
In summary, given the high energy demands
of the PNS, neurons rely heavily on mitochon-
dria. Therefore, the susceptibility of mitochon-
dria to hyperglycemia and dyslipidemia on
bioenergetics, substrate utilization, biogenesis,
trafficking, and reactive oxygen species produc-
tion ultimately lead to PN.
4 Disease-Modifying
Therapies for PN Associated
with Dyslipidemia
Our increased understanding of the lipid-centric
molecular mechanisms underlying PN and the
emerging role of obesity as a causal factor advo-
cate lifestyle and dietary interventions as PN
treatments [131, 210]. Although large, random-
ized controlled clinical trials are lacking, the
emergent evidence is that lifestyle interventions
may exert beneficial effects, and exercise is now
recommended for persons with T2D and PN by
the American Diabetes Association [30]. Here,
we will review the current level of evidence for
lifestyle interventions as well as lipid-centric
pharmacological approaches for PN.
4.1 Exercise
The potential of exercise as a therapeutic strategy
for persons with PN has been evaluated in three
non-randomized and one small randomized trials
[211–214]. Intraepidermal nerve fiber density
increased in patients with impaired glucose toler-
ance and PN after an individualized 12-month
exercise intervention program along with diet
[214]. More recent trials showed similar improve-
ments in cutaneous reinnervation rates and nerve
fiber branching in PN following aerobic exercise
intervention in prediabetes and diabetes partici-
pants [212, 213]. Interestingly, these interven-
tions were not associated with significant weight
loss and used intraepidermal nerve fiber density
as a primary outcome measure. Balducci et  al.
showed improved nerve conduction velocity and
vibration perception after a 4-year aerobic exer-
cise intervention in a small randomized trial of
297
T1D and T2D participants [211]; however, one
limitation was that the study did not stratify par-
ticipants based on their diabetes status, i.e., T1D
from T2D, and did not assess intraepidermal
nerve fiber density. The results from these four
trials are promising, but call for confirmatory
studies on larger cohorts to validate the role of
exercise as a disease-modifying PN therapy.
Several additional studies have also investigated
the benefit of exercise on steadiness and gait, and
quality of life [reviewed in [210]]. Moving for-
ward, research comparing the efficacy of aerobic
versus high-intensity interval training will be
critical to fully elucidate the optimal exercise
regimen for PN improvement.
4.2 Dietary Interventions
Besides exercise, preclinical studies strongly
support a role for dietary interventions for revers-
ing PN [26, 28, 167, 203, 204, 215]. Switching
prediabetic or T2D mice from a saturated to a
monounsaturated fatty acid-rich diet reverses
metrics of both small and large fiber PN, aligning
with the differences these fatty acids have on
mitochondrial bioenergetics [28, 167]. Although
studies on PN in humans are lacking, diets rich in
monounsaturated or polyunsaturated fatty acid
are beneficial to metabolic health and T2D and
cardiovascular disease prevention [34, 216–218].
One small open-label study of omega-3 supple-
mentation in T1D participants with PN improved
corneal nerve fiber length by corneal confocal
microscopy [219]. However, corneal nerve fiber
length is not a validated PN metric; indeed, nei-
ther nerve conduction nor cooling or vibration
threshold improved upon omega-3
supplementation.
Preclinical studies further demonstrate a link
between fatty acid dietary source and nerve func-
tion [28, 167], a finding that was confirmed in the
Netherlands Epidemiology of Obesity (NEO)
study, which found a positive correlation between
animal-based but not dairy- or plant-based satu-
rated and unsaturated fatty acids with markers of
insulin resistance [220]. Consistent with these
findings, patients with T2D and painful PN
298
placed on a low-fat, plant-based diet experienced
lower pain symptoms after 20 weeks of interven-
tion [221]. However, firm conclusions could not
be drawn because these participants also lost
weight, which likely contributed to improvement
in nerve function. Although data on the effects of
dietary-only interventions in PN are sparse, we
posit that a plant-based diet enriched in unsatu-
rated fatty acids with or without exercise is essen-
tial for peripheral nerve health and function, in
line with the recommendations of the American
Diabetes Association [30].
4.3 Weight Loss
Interventions designed to evaluate the effect of
medical or surgical weight loss on PN are lim-
ited. Recently, the Look AHEAD (Action for
Health in Diabetes) trial, which aimed to achieve
and maintain long-term weight loss through
intensive lifestyle intervention program, showed
PN improvement in overweight and obese
patients with T2D according to the Michigan
Neuropathy Screening Instrument questionnaire
and light touch sensation [222]. The benefits,
however, mostly did not persist 1–2.3 years after
study termination. Another long-term r­ andomized
Diabetes Prevention Program Outcome Study
(DPPOS) compared lifestyle intervention to met-
formin or placebo [223]. At a mean 15-year fol-
low-up, there were no differences among groups
for composite PN, nephropathy, and retinopathy
outcomes, although there was a trending decrease
in PN in the lifestyle group. When stratified by
sex, lifestyle intervention significantly lowered
composite microvascular outcomes in females.
Additionally, an ongoing clinical trial by our
group will shed light on the efficacy of medical
weight loss on PN progression in obese and pre-
diabetic participants (NCT02689661) [4].
Regarding surgical weight loss, a small pro-
spective study of T2D participants found that
Roux-en-Y gastric bypass surgery improved PN
at 6-month follow-up, according to the
Neuropathy Symptom Score and the Neuropathy
Deficit Score [224]. We are also examining the
effects of bariatric surgery as well as high inten-
S. A. Eid et al.
sity interval training, alone or combined, in a ran-
domized trial on PN in obese participants
(NCT03617185) [2]. This study will utilize the
gold standard intraepidermal nerve fiber density
as the primary outcome and constitutes a critical
step to determine if either intervention, bariatric
surgery or high-intensity exercise, will generate
superior PN outcomes.
4.4 Lipid-Lowering Agents
In terms of pharmacological interventions, it
remains unclear whether single- or multi-­factorial
approaches are needed to achieve sufficient thera-
peutic efficacy. Incorporating lipid-lowering
agents may be effective for treating PN. Fenofibrate,
a PPARα agonist, which reduces plasma triglycer-
ides and modestly increases HDL-c [128], has
shown promise for treating diabetic microvascular
complications [225, 226], including PN [227]. The
Fenofibrate Intervention and Event Lowering in
Diabetes (FIELD) study showed that fenofibrate
treatment significantly reduced the risk of lower-
limb amputations in T2D patients [227]. Additional
mechanistic studies in db/db mice demonstrated
that these neuroprotective properties may be medi-
ated by the AMP activated protein kinase (AMPK)-
PGC1α pathway [228], an effect that needs to be
validated in other T2D mouse models and in
human T2D PN.
On the other hand, the efficacy of statin ther-
apy, aimed at reducing plasma LDL-c, remains
controversial. While Rajamani et al. showed that
statin use reduces the risk of developing PN and
foot ulcers in T2D patients [229], increasingly,
data reveal a positive association between pro-
longed statin treatment duration and elevated PN
risk [230, 231]. Additionally, a recent review sug-
gested a potential link between statin therapy and
the development of neuropathic pain [232]. Our
large population-based study of 259,625 T2D
participants found that statin therapy neither
worsened nor improved PN, although it could not
discount a small acute risk of harm [233]. Thus,
more mechanistic studies are needed to deter-
mine if statin use is a meaningful therapeutic PN
strategy.
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
Lastly, it is important to emphasize that these
drugs are not tissue-specific and may have dif-
ferential effects on complication-prone tissues, in
addition to pleiotropic actions independent of
lipid-lowering [234, 235]. Thus, besides drug
refinement, identifying cellular and subcellular
targets of lipotoxicity may be the key to attain
full efficacy against nerve damage. Of interest,
we recently applied network analysis of per-
turbed and interacting pathways to identify
PN-induced rewiring between pathogenic path-
ways, which offers alternative pharmacological
strategies for PN treatment [236]. Cross-species
comparison of human versus mouse transcrip-
tomes (detailed above) is another critical
approach for offering mechanism-based targets
conserved across mouse and human and improv-
ing therapeutic interventions [165].
5 Conclusions
The past decade has seen PN research move past
the purely glucose-centric view to also encom-
pass lipid-centric concepts. This parallels fre-
quent comorbid obesity and dyslipidemia present
in prediabetes and T2D patients, and the complex
intersection of glucose and lipid metabolism at
the molecular level. Large clinical studies have
outlined that these additional risk factors for PN,
obesity, dyslipidemia, and the metabolic syn-
drome, can also occur independent of hypergly-
cemia. The clinical findings also suggest that
glucose control alone may not suffice to prevent
PN onset or progression in T2D patients, and that
the metabolic syndrome is a critical driver of PN
pathogenesis. Preclinical studies are uncovering
the pathomechanisms of dyslipidemia on nerves
through lipogenesis, mitochondrial bioenergetics
and transport, as well as oxidative stress and
inflammation. A clarification of the mechanisms
in the appropriate models are needed, since many
previous studies utilized T1D or T2D animals
possessing genetic mutations, e.g., leptin signal-
ing deficient db/db [22], which do not fully model
the human condition [25].
Additionally, emergent research avenues
require deeper understanding, including the
299
intersection between glucose- and lipid-centric
mechanisms in PN and axo-glial metabolic com-
munication and the non-cell autonomous nature
of PN. Adoption of omics and multi-omics tech-
niques followed by target validation will advance
our understanding of pathophysiology and iden-
tify actionable targets for developing disease-­
modifying therapy for PN secondary to metabolic
dysfunction. Additionally, the contribution of the
metabolic syndrome, obesity, and dyslipidemia
to PN in the T1D setting will need to be addressed
as obesity rises in this population as well.
Presently, diet, weight loss, and exercise are
recommended as the first-line PN treatment to
reverse metabolic syndrome components and
improve glucose handling and lipid profiles,
anticipating this will prevent, slow, or reverse PN
[30]. Preliminary clinical evidence indicates
these may be promising avenues. Indeed, life-
style interventions have long been regarded as
beneficial for T2D patients, as one review stated
even as long as 20 years ago, “The question is no
longer whether diet and exercise can benefit the
individual with type 2 diabetes” but rather to
what extent and on the “…appropriate feasibility
and acceptability of an intervention program.”
We await the findings from randomized, con-
trolled clinical trials, which will evaluate the effi-
cacy of diet, exercise, and surgical weight loss on
PN to identify the optimal regimen.
Acknowledgments The authors thank Bhumsoo Kim,
PhD, for revising content and for critical input.
References
1. Callaghan BC, et  al. Enhanced glucose control
for preventing and treating diabetic neuropathy.
Cochrane Database Syst Rev. 2012;6:CD007543.
2. Callaghan BC, et  al. Central obesity is associated
with neuropathy in the severely obese. Mayo Clin
Proc. 2020;95(7):1342–53.
3. Callaghan BC, et  al. Metabolic syndrome compo-
nents are associated with symptomatic polyneuropa-
thy independent of glycemic status. Diabetes Care.
2016;39(5):801–7.
4. Callaghan BC, et  al. Association between meta-
bolic syndrome components and polyneuropa-
thy in an obese population. JAMA Neurol.
2016;73(12):1468–76.
300
5. Jaiswal M, et al. Prevalence of and risk factors for
diabetic peripheral neuropathy in youth with type 1
and type 2 diabetes: SEARCH for Diabetes in Youth
Study. Diabetes Care. 2017;40(9):1226–32.
6. Jaiswal M, et  al. Burden of diabetic peripheral
neuropathy in pima indians with type 2 diabetes.
Diabetes Care. 2016;39(4):e63–4.
7. Callaghan BC, et  al. Triglycerides and amputa-
tion risk in patients with diabetes: ten-year fol-
low-­up in the DISTANCE study. Diabetes Care.
2011;34(3):635–40.
8. Smith AG, Singleton JR.  Obesity and hyperlipid-
emia are risk factors for early diabetic neuropathy. J
Diabetes Complicat. 2013;27(5):436–42.
9. Callaghan BC, et  al. Diabetes and obesity are the
main metabolic drivers of peripheral neuropathy.
Ann Clin Transl Neurol. 2018;5(4):397–405.
10. Lu B, et  al. Determination of peripheral neuropa-
thy prevalence and associated factors in Chinese
subjects with diabetes and pre-diabetes - ShangHai
Diabetic neuRopathy Epidemiology and Molecular
Genetics Study (SH-DREAMS). PLoS One.
2013;8(4):e61053.
11. Han L, et  al. Peripheral neuropathy is associated
with insulin resistance independent of metabolic
syndrome. Diabetol Metab Syndr. 2015;7:14.
12. Andersen ST, et al. Risk factors for incident diabetic
polyneuropathy in a cohort with screen-detected
type 2 diabetes followed for 13 years: ADDITION-­
Denmark. Diabetes Care. 2018;41(5):1068–75.
13. Christensen DH, et  al. Metabolic factors, life-
style habits, and possible polyneuropathy in early
type 2 diabetes: a nationwide study of 5,249
patients in the danish centre for strategic research
in type 2 diabetes (DD2) cohort. Diabetes Care.
2020;43(6):1266–75.
14. Schlesinger S, et al. General and abdominal obesity
and incident distal sensorimotor polyneuropathy:
insights into inflammatory biomarkers as potential
mediators in the KORA F4/FF4 cohort. Diabetes
Care. 2019;42(2):240–7.
15. Ziegler D, et  al. Prevalence of polyneuropathy in
pre-diabetes and diabetes is associated with abdomi-
nal obesity and macroangiopathy: the MONICA/
KORA Augsburg Surveys S2 and S3. Diabetes Care.
2008;31(3):464–9.
16. Hanewinckel R, et  al. High body mass and kidney
dysfunction relate to worse nerve function, even in
adults without neuropathy. J Peripher Nerv Syst.
2017;22(2):112–20.
17. Grisold A, Callaghan BC, Feldman EL.  Mediators
of diabetic neuropathy: is hyperglycemia the only
culprit? Curr Opin Endocrinol Diabetes Obes.
2017;24(2):103–11.
18. Stino AM, Smith AG. Peripheral neuropathy in pre-
diabetes and the metabolic syndrome. J Diabetes
Investig. 2017;8(5):646–55.
19. Duckworth W, et  al. Glucose control and vascular
complications in veterans with type 2 diabetes. N
Engl J Med. 2009;360(2):129–39.
S. A. Eid et al.
20. Ismail-Beigi F, et al. Effect of intensive treatment of
hyperglycaemia on microvascular outcomes in type
2 diabetes: an analysis of the ACCORD randomised
trial. Lancet. 2010;376(9739):419–30.
21. Callaghan BC, et  al. The prevalence and determi-
nants of cognitive deficits and traditional diabetic
complications in the severely obese. Diabetes Care.
2020;43(3):683–90.
22. O'Brien PD, Sakowski SA, Feldman EL.  Mouse
models of diabetic neuropathy. ILAR J.
2014;54(3):259–72.
23. Elzinga SE, et  al. Sex differences in insulin resis-
tance but not peripheral neuropathy in a diet-­
induced prediabetes mouse model. Dis Model Mech.
2021;14(4):48909.
24. Medrikova D, et  al. Sex differences during the
course of diet-induced obesity in mice: adipose tis-
sue expandability and glycemic control. Int J Obes.
2012;36(2):262–72.
25. Farr OM, Gavrieli A, Mantzoros CS.  Leptin appli-
cations in 2015: what have we learned about leptin
and obesity? Curr Opin Endocrinol Diabetes Obes.
2015;22(5):353–9.
26. Hinder LM, et al. Dietary reversal of neuropathy in
a murine model of prediabetes and metabolic syn-
drome. Dis Model Mech. 2017;10(6):717–25.
27. O’Brien PD, et  al. Juvenile murine models of pre-
diabetes and type 2 diabetes develop neuropathy. Dis
Model Mech. 2018;11(12):37374.
28. O'Brien PD, et al. Integrated lipidomic and transcrip-
tomic analyses identify altered nerve triglycerides in
mouse models of prediabetes and type 2 diabetes.
Dis Model Mech. 2020;13(2):42101.
29. Sajic M, et al. High dietary fat consumption impairs
axonal mitochondrial function in  vivo. J Neurosci.
2021;41(19):4321–34.
30. Pop-Busui R, et al. Diabetic neuropathy: a position
statement by the American Diabetes Association.
Diabetes Care. 2017;40(1):136–54.
31. Tuomilehto J, et  al. Prevention of type 2 diabetes
mellitus by changes in lifestyle among subjects
with impaired glucose tolerance. N Engl J Med.
2001;344(18):1343–50.
32. Aucott L, et al. Effects of lifestyle interventions and
long-term weight loss on lipid outcomes - a system-
atic review. Obes Rev. 2011;12(5):e412–25.
33. Tjønna AE, et  al. Aerobic interval training versus
continuous moderate exercise as a treatment for
the metabolic syndrome: a pilot study. Circulation.
2008;118(4):346–54.
34. Salas-Salvado J, et al. Reduction in the incidence of
type 2 diabetes with the Mediterranean diet: results
of the PREDIMED-Reus nutrition intervention ran-
domized trial. Diabetes Care. 2011;34(1):14–9.
35. Sasaki Y.  Metabolic aspects of neuronal degenera-
tion: from a NAD+ point of view. Neurosci Res.
2019;139:9–20.
36. Feldman EL, et al. New horizons in diabetic neurop-
athy: mechanisms, bioenergetics, and pain. Neuron.
2017;93(6):1296–313.
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
37. Wang JT, Medress ZA, Barres BA.  Axon degen-
eration: molecular mechanisms of a self-destruction
pathway. J Cell Biol. 2012;196(1):7–18.
38. Rossi MJ, Pekkurnaz G.  Powerhouse of the mind:
mitochondrial plasticity at the synapse. Curr Opin
Neurobiol. 2019;57:149–55.
39. Zhu XH, et  al. Quantitative imaging of energy
expenditure in human brain. NeuroImage.
2012;60(4):2107–17.
40. Mitchell RW, et  al. Fatty acid transport protein
expression in human brain and potential role in fatty
acid transport across human brain microvessel endo-
thelial cells. J Neurochem. 2011;117(4):735–46.
41. Poitelon Y, Kopec AM, Belin S. Myelin fat facts: an
overview of lipids and fatty acid metabolism. Cell.
2020;9(4):812.
42. Simpson IA, et  al. The facilitative glucose trans-
porter GLUT3: 20 years of distinction. Am J Physiol
Endocrinol Metab. 2008;295(2):E242–53.
43. Bender T, Martinou JC. The mitochondrial pyruvate
carrier in health and disease: to carry or not to carry?
Biochim Biophys Acta. 2016;1863(10):2436–42.
44. Krebs HA, Johnson WA. Metabolism of ketonic acids
in animal tissues. Biochem J. 1937;31(4):645–60.
45. Chaban Y, Boekema EJ, Dudkina NV.  Structures
of mitochondrial oxidative phosphorylation super-
complexes and mechanisms for their stabilisation.
Biochim Biophys Acta. 2014;1837(4):418–26.
46. Cecchini G.  Function and structure of complex
II of the respiratory chain. Annu Rev Biochem.
2003;72:77–109.
47. Hirst J. Energy transduction by respiratory complex
I–an evaluation of current knowledge. Biochem Soc
Trans. 2005;33(Pt 3):525–9.
48. Crofts AR.  The cytochrome bc1 complex: func-
tion in the context of structure. Annu Rev Physiol.
2004;66:689–733.
49. Tsukihara T, et  al. The whole structure of the
13-subunit oxidized cytochrome c oxidase at 2.8
A. Science. 1996;272(5265):1136–44.
50. Boyer PD. The ATP synthase–a splendid molecular
machine. Annu Rev Biochem. 1997;66:717–49.
51. Rangaraju V, Calloway N, Ryan TA. Activity-driven
local ATP synthesis is required for synaptic function.
Cell. 2014;156(4):825–35.
52. Sobieski C, Fitzpatrick MJ, Mennerick
SJ.  Differential presynaptic ATP supply for basal
and high-demand transmission. J Neurosci.
2017;37(7):1888–99.
53. Grote CW, Ryals JM, Wright DE. In vivo peripheral
nervous system insulin signaling. J Peripher Nerv
Syst. 2013;18(3):209–19.
54. Grote CW, et  al. Peripheral nervous system insu-
lin resistance in ob/ob mice. Acta Neuropathol
Commun. 2013;1:15.
55. Grote CW, Wright DE. A role for insulin in diabetic
neuropathy. Front Neurosci. 2016;10:581.
56. Houten SM, Wanders RJ. A general introduction to
the biochemistry of mitochondrial fatty acid beta-­
oxidation. J Inherit Metab Dis. 2010;33(5):469–77.
301
57. Kastaniotis AJ, et  al. Mitochondrial fatty acid
synthesis, fatty acids and mitochondrial physiol-
ogy. Biochim Biophys Acta Mol Cell Biol Lipids.
2017;1862(1):39–48.
58. Kidd GJ, Ohno N, Trapp BD. Biology of Schwann
cells. Handb Clin Neurol. 2013;115:55–79.
59. Schulingkamp RJ, et al. Insulin receptors and insulin
action in the brain: review and clinical implications.
Neurosci Biobehav Rev. 2000;24(8):855–72.
60. Jha MK, Morrison BM. Lactate transporters mediate
glia-neuron metabolic crosstalk in homeostasis and
disease. Front Cell Neurosci. 2020;14:589582.
61. Magnani P, et al. Glucose transporters in rat periph-
eral nerve: paranodal expression of GLUT1 and
GLUT3. Metabolism. 1996;45(12):1466–73.
62. Magnani P, et  al. Regulation of glucose transport
in cultured Schwann cells. J Peripher Nerv Syst.
1998;3(1):28–36.
63. Jha MK, Morrison BM. Glia-neuron energy metabo-
lism in health and diseases: New insights into the
role of nervous system metabolic transporters. Exp
Neurol. 2018;309:23–31.
64. Brown AM, et al. Schwann cell glycogen selectively
supports myelinated axon function. Ann Neurol.
2012;72(3):406–18.
65. Véga C, et  al. Uptake of locally applied deoxyglu-
cose, glucose and lactate by axons and Schwann
cells of rat vagus nerve. J Physiol. 2003;546(Pt
2):551–64.
66. Bouçanova F, Chrast R.  Metabolic interaction
between schwann cells and axons under physiologi-
cal and disease conditions. Front Cell Neurosci.
2020;14:148.
67. Gonçalves NP, et al. Schwann cell interactions with
axons and microvessels in diabetic neuropathy. Nat
Rev Neurol. 2017;13(3):135–47.
68. Domenech-Estevez E, et  al. Distribution of mono-
carboxylate transporters in the peripheral nervous
system suggests putative roles in lactate shuttling
and myelination. J Neurosci. 2015;35(10):4151–6.
69. Babetto E, Wong KM, Beirowski B.  A glycolytic
shift in Schwann cells supports injured axons. Nat
Neurosci. 2020;23(10):1215–28.
70. Morrison BM, et al. Deficiency in monocarboxylate
transporter 1 (MCT1) in mice delays regeneration of
peripheral nerves following sciatic nerve crush. Exp
Neurol. 2015;263:325–38.
71. Bouçanova F, et  al. Disrupted function of lactate
transporter MCT1, but not MCT4, in Schwann cells
affects the maintenance of motor end-plate innerva-
tion. Glia. 2021;69(1):124–36.
72. Jha MK, et  al. Reducing monocarboxylate trans-
porter MCT1 worsens experimental diabetic periph-
eral neuropathy. Exp Neurol. 2020;333:113415.
73. Jha MK, et  al. Monocarboxylate transporter 1  in
Schwann cells contributes to maintenance of
sensory nerve myelination during aging. Glia.
2020;68(1):161–77.
74. Rumora AE, et  al. Disorders of mitochondrial
dynamics in peripheral neuropathy: Clues from
302
hereditary neuropathy and diabetes. Int Rev
Neurobiol. 2019;145:127–76.
75. Courchet J, et  al. Terminal axon branching is
regulated by the LKB1-NUAK1 kinase path-
way via presynaptic mitochondrial capture. Cell.
2013;153(7):1510–25.
76. Spillane M, et  al. Mitochondria coordinate sites of
axon branching through localized intra-axonal pro-
tein synthesis. Cell Rep. 2013;5(6):1564–75.
77. Sun T, et al. Motile axonal mitochondria contribute
to the variability of presynaptic strength. Cell Rep.
2013;4(3):413–9.
78. Sheng Z-H. The interplay of axonal energy homeo-
stasis and mitochondrial trafficking and anchoring.
Trends Cell Biol. 2017;27(6):403–16.
79. Amiri M, Hollenbeck PJ. Mitochondrial biogenesis
in the axons of vertebrate peripheral neurons. Dev
Neurobiol. 2008;68(11):1348–61.
80. Schwarz TL.  Mitochondrial trafficking in neurons.
Cold Spring Harb Perspect Biol. 2013;5(6):11304.
81. Lentz SI, et  al. Mitochondrial DNA (mtDNA) bio-
genesis: visualization and duel incorporation of
BrdU and EdU into newly synthesized mtDNA
in vitro. J Histochem Cytochem. 2010;58(2):207–18.
82. Yu SB, Pekkurnaz G.  Mechanisms orchestrating
mitochondrial dynamics for energy homeostasis. J
Mol Biol. 2018;430(21):3922–41.
83. Rumora AE, et  al. Chain length of saturated fatty
acids regulates mitochondrial trafficking and function
in sensory neurons. J Lipid Res. 2019;60(1):58–70.
84. Rumora AE, et al. Dyslipidemia impairs mitochon-
drial trafficking and function in sensory neurons.
FASEB J. 2018;32(1):195–207.
85. Hinder LM, et  al. Long-chain acyl coenzyme A
synthetase 1 overexpression in primary cultured
Schwann cells prevents long chain fatty acid-­
induced oxidative stress and mitochondrial dysfunc-
tion. Antioxid Redox Signal. 2014;21(4):588–600.
86. Viader A, et  al. Aberrant Schwann cell lipid
metabolism linked to mitochondrial deficits leads
to axon degeneration and neuropathy. Neuron.
2013;77(5):886–98.
87. Padilla A, et  al. Hyperglycemia magnifies
Schwann cell dysfunction and cell death trig-
gered by PA-induced lipotoxicity. Brain Res.
2011;1370:64–79.
88. Nicholls DG, Budd SL. Mitochondria and neuronal
survival. Physiol Rev. 2000;80(1):315–60.
89. Fernyhough P.  Mitochondrial dysfunction in dia-
betic neuropathy: a series of unfortunate metabolic
events. Curr Diab Rep. 2015;15(11):89.
90. Chowdhury SK, Smith DR, Fernyhough P. The role
of aberrant mitochondrial bioenergetics in diabetic
neuropathy. Neurobiol Dis. 2013;51:56–65.
91. Vincent AM, et  al. Mitochondrial biogenesis and
fission in axons in cell culture and animal mod-
els of diabetic neuropathy. Acta Neuropathol.
2010;120(4):477–89.
92. Colvin LA. Chemotherapy-induced peripheral neu-
ropathy: where are we now? Pain. 2019;160(Suppl
1):1–10.
S. A. Eid et al.
93. Trecarichi A, Flatters SJL.  Mitochondrial dys-
function in the pathogenesis of chemotherapy-­
induced peripheral neuropathy. Int Rev Neurobiol.
2019;145:83–126.
94. Angelova PR, Abramov AY.  Role of mitochondrial
ROS in the brain: from physiology to neurodegen-
eration. FEBS Lett. 2018;592(5):692–702.
95. Oyewole AO, Birch-Machin MA.  Mitochondria-­
targeted antioxidants. FASEB J.
2015;29(12):4766–71.
96. Snezhkina AV, et  al. ROS generation and antioxi-
dant defense systems in normal and malignant cells.
Oxidative Med Cell Longev. 2019;2019:6175804.
97. Akude E, et  al. Diminished superoxide genera-
tion is associated with respiratory chain dysfunc-
tion and changes in the mitochondrial proteome
of sensory neurons from diabetic rats. Diabetes.
2011;60(1):288–97.
98. Nishikawa T, et al. Normalizing mitochondrial super-
oxide production blocks three pathways of hypergly-
caemic damage. Nature. 2000;404(6779):787–90.
99. Vincent AM, et  al. Short-term hyperglycemia pro-
duces oxidative damage and apoptosis in neurons.
FASEB J. 2005;19(6):638–40.
100. Tretter L, Adam-Vizi V.  Alpha-ketoglutarate dehy-
drogenase: a target and generator of oxidative stress.
Philos Trans R Soc B. 2005;360(1464):2335–45.
101. Gardner P, Fridovich I.  Inactivation-reactivation
of aconitase in Escherichia coli. A sensitive
measure of superoxide radical. J Biol Chem.
1992;267(13):8757–63.
102. Vincent AM, et  al. Sensory neurons and Schwann
cells respond to oxidative stress by increasing
antioxidant defense mechanisms. Antioxid Redox
Signal. 2009;11(3):425–38.
103. Delaney CL, et  al. Insulin-like growth factor-I and
over-expression of Bcl-xL prevent glucose-mediated
apoptosis in Schwann cells. J Neuropathol Exp
Neurol. 2001;60(2):147–60.
104. Thurston JH, et  al. Effects of acute, subacute, and
chronic diabetes on carbohydrate and energy metab-
olism in rat sciatic nerve: relation to mechanisms of
peripheral neuropathy. Diabetes. 1995;44(2):190–5.
105. Kishi Y, et  al. Alpha-lipoic acid: effect on glucose
uptake, sorbitol pathway, and energy metabolism
in experimental diabetic neuropathy. Diabetes.
1999;48(10):2045–51.
106. Obrosova IG, et  al. Aldose reductase inhibi-
tion counteracts oxidative-nitrosative stress and
poly (ADP-ribose) polymerase activation in tis-
sue sites for diabetes complications. Diabetes.
2005;54(1):234–42.
107. Yagihashi S.  Glucotoxic mechanisms and related
therapeutic approaches. Int Rev Neurobiol.
2016;127:121–49.
108. Dewanjee S, et al. Molecular mechanism of diabetic
neuropathy and its pharmacotherapeutic targets. Eur
J Pharmacol. 2018;833:472–523.
109. Greene DA, Lattimer S.  Impaired rat sciatic nerve
sodium-potassium adenosine triphosphatase in
acute streptozocin diabetes and its correction by
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
dietary myo-inositol supplementation. J Clin Invest.
1983;72(3):1058–63.
110. Greene DA, Lattimer SA, Sima AA. Sorbitol, phos-
phoinositides, and sodium-potassium-ATPase in the
pathogenesis of diabetic complications. N Engl J
Med. 1987;316(10):599–606.
111. Greene DA, et  al. Nerve Na+-K+-ATPase,
conduction, and myo-inositol in the insulin-­
deficient BB rat. Am J Physiol Endocrinol Metab.
1984;247(4):E534–9.
112. Oates PJ.  Aldose reductase, still a compelling tar-
get for diabetic neuropathy. Curr Drug Targets.
2008;9(1):14–36.
113. Stevens MJ, et  al. The linked roles of nitric oxide,
aldose reductase and (Na+, K+)-ATPase in the slow-
ing of nerve conduction in the streptozotocin dia-
betic rat. J Clin Invest. 1994;94(2):853–9.
114. Kern TS, Engerman RL.  Immunohistochemical
distribution of aldose reductase. Histochem J.
1982;14(3):507–15.
115. Cameron N, Cotter M.  Potential therapeutic
approaches to the treatment or prevention of diabetic
neuropathy: evidence from experimental studies.
Diabet Med. 1993;10(7):593–605.
116. Ho EC, et  al. Aldose reductase–deficient mice are
protected from delayed motor nerve conduction
velocity, increased c-Jun NH2-terminal kinase acti-
vation, depletion of reduced glutathione, increased
superoxide accumulation, and DNA damage.
Diabetes. 2006;55(7):1946–53.
117. Yagihashi S, et  al. Neuropathy in diabetic mice
overexpressing human aldose reductase and
effects of aldose reductase inhibitor. Brain.
2001;124(12):2448–58.
118. Tsai SC, Burnakis TG. Aldose reductase inhibitors:
an update. Ann Pharmacother. 1993;27(6):751–4.
119. Chalk C, Benstead TJ, Moore F.  Aldose reductase
inhibitors for the treatment of diabetic polyneuropa-
thy. Cochrane Database Syst Rev. 2007;4:CD004572.
120. Misur I, et  al. Advanced glycation endproducts in
peripheral nerve in type 2 diabetes with neuropathy.
Acta Diabetol. 2004;41(4):158–66.
121. Popova EA, Mironova RS, Odjakova MK.  Non-­
enzymatic glycosylation and deglycating enzymes.
Biotechnol Biotechnol Equip. 2010;24(3):1928–35.
122. Lukic IK, et al. The RAGE pathway: activation and
perpetuation in the pathogenesis of diabetic neurop-
athy. Ann N Y Acad Sci. 2008;1126(1):76–80.
123. Du X-L, et al. Hyperglycemia-induced mitochondrial
superoxide overproduction activates the hexosamine
pathway and induces plasminogen activator inhibi-
tor-­1 expression by increasing Sp1 glycosylation.
Proc Natl Acad Sci U S A. 2000;97(22):12222–6.
124. Eichberg J.  Protein kinase C changes in diabe-
tes: is the concept relevant to neuropathy? Int Rev
Neurobiol. 2002;50:61–82.
125. Geraldes P, King GL. Activation of protein kinase C
isoforms and its impact on diabetic complications.
Circ Res. 2010;106(8):1319–31.
126. Bansal D, et  al. Ruboxistaurin for the treatment of
diabetic peripheral neuropathy: a systematic review
303
of randomized clinical trials. Diabetes Metab J.
2013;37(5):375.
127. Callaghan BC, Hur J, Feldman EL.  Diabetic neu-
ropathy: one disease or two? Curr Opin Neurol.
2012;25(5):536–41.
128. Vincent AM, et al. Hyperlipidemia: a new therapeu-
tic target for diabetic neuropathy. J Peripher Nerv
Syst. 2009;14(4):257–67.
129. Wiggin TD, et  al. Elevated triglycerides correlate
with progression of diabetic neuropathy. Diabetes.
2009;58(7):1634–40.
130. Eid S, et  al. New insights into the mechanisms of
diabetic complications: role of lipids and lipid
metabolism. Diabetologia. 2019;62(9):1539–49.
131. Savelieff MG, Callaghan BC, Feldman EL.  The
emerging role of dyslipidemia in diabetic microvas-
cular complications. Curr Opin Endocrinol Diabetes
Obes. 2020;27(2):115–23.
132. Franssen R, et  al. Obesity and dyslipidemia. Med
Clin North Am. 2011;95(5):893–902.
133. Corbin KD, et al. Obesity in type 1 diabetes: patho-
physiology, clinical impact, and mechanisms.
Endocr Rev. 2018;39(5):629–63.
134. Mottalib A, et  al. Weight management in patients
with type 1 diabetes and obesity. Curr Diab Rep.
2017;17(10):92.
135. O'Brien PD, et  al. Neurological consequences of
obesity. Lancet Neurol. 2017;16(6):465–77.
136. Ebbert JO, Jensen MD. Fat depots, free fatty acids,
and dyslipidemia. Nutrients. 2013;5(2):498–508.
137. Tchoukalova YD, et  al. Regional differences
in cellular mechanisms of adipose tissue gain
with overfeeding. Proc Natl Acad Sci U S A.
2010;107(42):18226–31.
138. Acosta JR, et al. Increased fat cell size: a major phe-
notype of subcutaneous white adipose tissue in non-­
obese individuals with type 2 diabetes. Diabetologia.
2016;59(3):560–70.
139. Wensveen FM, et  al. The “Big Bang” in obese
fat: events initiating obesity-induced adi-
pose tissue inflammation. Eur J Immunol.
2015;45(9):2446–56.
140. Fontana L, et  al. Visceral fat adipokine secretion
is associated with systemic inflammation in obese
humans. Diabetes. 2007;56(4):1010–3.
141. Vincent AM, et  al. Dyslipidemia-induced neuropa-
thy in mice: the role of oxLDL/LOX-1. Diabetes.
2009;58(10):2376–85.
142. Ouchi N, et  al. Adipokines in inflamma-
tion and metabolic disease. Nat Rev Immunol.
2011;11(2):85–97.
143. Bartness TJ, et al. Sensory and sympathetic nervous
system control of white adipose tissue lipolysis. Mol
Cell Endocrinol. 2010;318(1-2):34–43.
144. Unamuno X, et  al. Adipokine dysregulation and
adipose tissue inflammation in human obesity. Eur J
Clin Investig. 2018;48(9):e12997.
145. Schweiger M, et al. Adipose triglyceride lipase and
hormone-sensitive lipase are the major enzymes
in adipose tissue triacylglycerol catabolism. J Biol
Chem. 2006;281(52):40236–41.
304
146. Chen SC, Tseng CH. Dyslipidemia, kidney disease,
and cardiovascular disease in diabetic patients. Rev
Diabet Stud. 2013;10(2-3):88–100.
147. Bardini G, Rotella CM, Giannini S.  Dyslipidemia
and diabetes: reciprocal impact of impaired lipid
metabolism and Beta-cell dysfunction on micro-
and macrovascular complications. Rev Diabet Stud.
2012;9(2-3):82–93.
148. Hotamisligil GS, Shargill NS, Spiegelman
BM.  Adipose expression of tumor necrosis factor-­
alpha: direct role in obesity-linked insulin resistance.
Science. 1993;259(5091):87–91.
149. Czech MP, et  al. Insulin signalling mecha-
nisms for triacylglycerol storage. Diabetologia.
2013;56(5):949–64.
150. Rhee EP, et al. Lipid profiling identifies a triacylg-
lycerol signature of insulin resistance and improves
diabetes prediction in humans. J Clin Invest.
2011;121(4):1402–11.
151. Basu Ball W, Neff JK, Gohil VM. The role of non-
bilayer phospholipids in mitochondrial structure and
function. FEBS Lett. 2018;592(8):1273–90.
152. Spector AA, Yorek MA.  Membrane lipid com-
position and cellular function. J Lipid Res.
1985;26(9):1015–35.
153. Djemli-Shipkolye A, et al. Na, K-atpase alterations
in diabetic rats: relationship with lipid metabolism
and nerve physiological parameters. Cell Mol Biol.
2001;47(2):297–304.
154. Montani L, et  al. De novo fatty acid synthe-
sis by Schwann cells is essential for periph-
eral nervous system myelination. J Cell Biol.
2018;217(4):1353–68.
155. Liang G, Cline GW, Macica CM. IGF-1 stimulates
de novo fatty acid biosynthesis by Schwann cells
during myelination. Glia. 2007;55(6):632–41.
156. Camargo N, Smit AB, Verheijen MH.  SREBPs:
SREBP function in glia-neuron interactions. FEBS
J. 2009;276(3):628–36.
157. Shimano H, Sato R. SREBP-regulated lipid metabo-
lism: convergent physiology - divergent pathophysi-
ology. Nat Rev Endocrinol. 2017;13(12):710–30.
158. Wang B, Tontonoz P.  Liver X receptors in lipid
signalling and membrane homeostasis. Nat Rev
Endocrinol. 2018;14(8):452–63.
159. Ewan EE, et  al. Ascending dorsal column sensory
neurons respond to spinal cord injury and down-
regulate genes related to lipid metabolism. Sci Rep.
2021;11(1):374.
160. de Preux AS, et al. SREBP-1c expression in Schwann
cells is affected by diabetes and nutritional status.
Mol Cell Neurosci. 2007;35(4):525–34.
161. Zhu L, et  al. Prolonged high-glucose exposure
decreased SREBP-1/FASN/ACC in Schwann cells
of diabetic mice via blocking PI3K/Akt pathway. J
Cell Biochem. 2019;120(4):5777–89.
162. Tachibana H, et  al. Activation of liver X
receptor inhibits osteopontin and amelio-
rates diabetic nephropathy. J Am Soc Nephrol.
2012;23(11):1835–46.
S. A. Eid et al.
163. Hazra S, et  al. Liver X receptor modulates dia-
betic retinopathy outcome in a mouse model
of streptozotocin-induced diabetes. Diabetes.
2012;61(12):3270–9.
164. Gavini CK, et  al. Liver X receptors protect dorsal
root ganglia from obesity-induced endoplasmic
reticulum stress and mechanical allodynia. Cell Rep.
2018;25(2):271–277 e4.
165. McGregor BA, et al. Conserved transcriptional sig-
natures in human and murine diabetic peripheral
neuropathy. Sci Rep. 2018;8(1):17678.
166. Eid SA, et  al. Targeting the NADPH oxidase-
4 and liver X receptor pathway preserves
­
schwann cell integrity in diabetic mice. Diabetes.
2020;69(3):448–64.
167. Rumora AE, et  al. The divergent roles of dietary
saturated and monounsaturated fatty acids on nerve
function in murine models of obesity. J Neurosci.
2019;39(19):3770–81.
168. Wang H, Eckel RH.  Lipoprotein lipase: from
gene to obesity. Am J Physiol Endocrinol Metab.
2009;297(2):E271–88.
169. Verheijen MH, et  al. Local regulation of fat
metabolism in peripheral nerves. Genes Dev.
2003;17(19):2450–64.
170. Pande M, et al. Transcriptional profiling of diabetic
neuropathy in the BKS db/db mouse: a model of
type 2 diabetes. Diabetes. 2011;60(7):1981–9.
171. Hur J, et  al. The identification of gene expres-
sion profiles associated with progression of
human diabetic neuropathy. Brain. 2011;134(Pt
11):3222–35.
172. Supruniuk E, Miklosz A, Chabowski A. The impli-
cation of PGC-1alpha on fatty acid transport across
plasma and mitochondrial membranes in the insulin
sensitive tissues. Front Physiol. 2017;8:923.
173. DeFronzo R.  Insulin resistance, lipotoxicity, type
2 diabetes and atherosclerosis: the missing links.
The Claude Bernard Lecture 2009. Diabetologia.
2010;53(7):1270–87.
174. D'Souza K, Nzirorera C, Kienesberger PC.  Lipid
metabolism and signaling in cardiac lipotoxicity.
Biochim Biophys Acta. 2016;1861(10):1513–24.
175. Sas KM, et al. Shared and distinct lipid-lipid inter-
actions in plasma and affected tissues in a diabetic
mouse model. J Lipid Res. 2018;59(2):173–83.
176. Szendroedi J, et  al. Role of diacylglycerol activa-
tion of PKCtheta in lipid-induced muscle insulin
resistance in humans. Proc Natl Acad Sci U S A.
2014;111(26):9597–602.
177. Chaurasia B, Summers SA.  Ceramides  - lipotoxic
inducers of metabolic disorders. Trends Endocrinol
Metab. 2015;26(10):538–50.
178. Hu W, et  al. Differential regulation of dihydrocer-
amide desaturase by palmitate versus monounsatu-
rated fatty acids: implications for insulin resistance.
J Biol Chem. 2011;286(19):16596–605.
179. Fridman V, et  al. Randomized trial of l-serine in
patients with hereditary sensory and autonomic neu-
ropathy type 1. Neurology. 2019;92(4):e359–70.
Mechanisms of Nerve Injury in Diabetes: Dyslipidemia, Bioenergetics, and Oxidative Damage
180. Fridman V, et  al. Altered plasma serine and
1-­deoxydihydroceramide profiles are associated
with diabetic neuropathy in type 2 diabetes and obe-
sity. J Diabetes Complicat. 2021;35(4):107852.
181. Rumora AE, et al. Plasma lipid metabolites associate
with diabetic polyneuropathy in a cohort with type 2
diabetes. Ann Clin Transl Neurol. 2021. https://doi.
org/10.1002/acn3.51367.
182. Afshinnia F, et  al. Increased lipogenesis and
impaired beta-oxidation predict type 2 diabetic kid-
ney disease progression in American Indians. JCI
Insight. 2019;4(21):e130317.
183. Stith JL, Velazquez FN, Obeid LM.  Advances in
determining signaling mechanisms of ceramide and
role in disease. J Lipid Res. 2019;60(5):913–8.
184. Raichur S, et al. The role of C16:0 ceramide in the
development of obesity and type 2 diabetes: CerS6
inhibition as a novel therapeutic approach. Mol
Metab. 2019;21:36–50.
185. Barber CN, Raben DM. Lipid metabolism crosstalk
in the brain: Glia and neurons. Front Cell Neurosci.
2019;13:212.
186. Zlobine I, Gopal K, Ussher JR.  Lipotoxicity in
obesity and diabetes-related cardiac dysfunction.
Biochim Biophys Acta. 2016;1861(10):1555–68.
187. McCoin CS, Knotts TA, Adams SH. Acylcarnitines—
old actors auditioning for new roles in metabolic
physiology. Nat Rev Endocrinol. 2015;11(10):617.
188. Hinder LM, et al. Comparative RNA-Seq transcrip-
tome analyses reveal distinct metabolic pathways in
diabetic nerve and kidney disease. J Cell Mol Med.
2017;21(9):2140–52.
189. Guo K, et  al. Genome-wide profiling of DNA
methylation and gene expression identifies can-
didate genes for human diabetic neuropathy. Clin
Epigenetics. 2020;12(1):123.
190. Elzinga S, et al. Toll-like receptors and inflammation
in metabolic neuropathy; a role in early versus late
disease? Exp Neurol. 2019;320:112967.
191. Rocha DM, et al. Saturated fatty acids trigger TLR4-­
mediated inflammatory response. Atherosclerosis.
2016;244:211–5.
192. Lee JY, et al. Saturated fatty acid activates but poly-
unsaturated fatty acid inhibits Toll-like receptor
2 dimerized with Toll-like receptor 6 or 1. J Biol
Chem. 2004;279(17):16971–9.
193. Hwang DH, Kim JA, Lee JY.  Mechanisms for the
activation of Toll-like receptor 2/4 by saturated fatty
acids and inhibition by docosahexaenoic acid. Eur J
Pharmacol. 2016;785:24–35.
194. Stino AM, et  al. Evolving concepts on the role of
dyslipidemia, bioenergetics, and inflammation in the
pathogenesis and treatment of diabetic peripheral
neuropathy. J Peripher Nerv Syst. 2020;25(2):76–84.
195. Yokoi H, Yanagita M.  Targeting the fatty acid
transport protein CD36, a class B scavenger recep-
tor, in the treatment of renal disease. Kidney Int.
2016;89(4):740–2.
196. Fu D, et  al. Immune complex formation in human
diabetic retina enhances toxicity of oxidized LDL
305
towards retinal capillary pericytes. J Lipid Res.
2014;55(5):860–9.
197. Nowicki M, et al. Oxidized low-density lipoprotein
(oxLDL)-induced cell death in dorsal root ganglion
cell cultures depends not on the lectin-like oxLDL
receptor-1 but on the toll-like receptor-4. J Neurosci
Res. 2010;88(2):403–12.
198. Hinder LM, et al. Mitochondrial uncoupling has no
effect on microvascular complications in type 2 dia-
betes. Sci Rep. 2019;9(1):881.
199. Eid SA, et al. Differential effects of empagliflozin
on microvascular complications in murine mod-
els of type 1 and type 2 diabetes. Biology.
2020;9(11):347.
200. Hinder LM, et  al. Transcriptional networks of pro-
gressive diabetic peripheral neuropathy in the db/db
mouse model of type 2 diabetes: an inflammatory
story. Exp Neurol. 2018;305:33–43.
201. Eid SA, et  al. Gene expression profiles of diabetic
kidney disease and neuropathy in eNOS knockout
mice: predictors of pathology and RAS blockade
effects. FASEB J. 2021;35(5):e21467.
202. Sas KM, et al. Tissue-specific metabolic reprogram-
ming drives nutrient flux in diabetic complications.
JCI Insight. 2016;1(15):e86976.
203. Coppey L, et al. Effect of dietary oils on peripheral
neuropathy-related endpoints in dietary obese rats.
Diabetes Metab Syndr Obes. 2018;11:117–27.
204. Coppey L, et  al. Progressive loss of corneal nerve
fibers and sensitivity in rats modeling obesity and
type 2 diabetes is reversible with omega-3 fatty acid
intervention: supporting cornea analyses as a marker
for peripheral neuropathy and treatment. Diabetes
Metab Syndr Obes. 2020;13:1367–84.
205. Cinci L, et  al. Oxidative, metabolic, and apoptotic
responses of Schwann cells to high glucose levels. J
Biochem Mol Toxicol. 2015;29(6):274–9.
206. Kato A, et  al. Recurrent short-term hypoglycemia
and hyperglycemia induce apoptosis and oxidative
stress via the ER stress response in immortalized
adult mouse Schwann (IMS32) cells. Neurosci Res.
2019;147:26–32.
207. Liu YP, Shao SJ, Guo HD.  Schwann cells apopto-
sis is induced by high glucose in diabetic peripheral
neuropathy. Life Sci. 2020;248:117459.
208. Zhang L, et  al. Hyperglycemia alters the Schwann
cell mitochondrial proteome and decreases coupled
respiration in the absence of superoxide production.
J Proteome Res. 2010;9(1):458–71.
209. Russell JW, et al. Neurons undergo apoptosis in ani-
mal and cell culture models of diabetes. Neurobiol
Dis. 1999;6(5):347–63.
210. Zilliox LA, Russell JW. Physical activity and dietary
interventions in diabetic neuropathy: a systematic
review. Clin Auton Res. 2019;29(4):443–55.
211. Balducci S, et  al. Exercise training can modify the
natural history of diabetic peripheral neuropathy. J
Diabetes Complicat. 2006;20(4):216–23.
212. Kluding PM, et  al. The effect of exercise on neu-
ropathic symptoms, nerve function, and cutaneous
306
innervation in people with diabetic peripheral neu-
ropathy. J Diabetes Complicat. 2012;26(5):424–9.
213. Singleton JR, et  al. Supervised exercise improves
cutaneous reinnervation capacity in metabolic syn-
drome patients. Ann Neurol. 2015;77(1):146–53.
214. Smith AG, et  al. Lifestyle intervention for
pre-­diabetic neuropathy. Diabetes Care.
2006;29(6):1294–9.
215. Cooper MA, et  al. A ketogenic diet reduces meta-
bolic syndrome-induced allodynia and promotes
peripheral nerve growth in mice. Exp Neurol.
2018;306:149–57.
216. Estruch R, et al. Primary prevention of cardiovascu-
lar disease with a Mediterranean diet supplemented
with extra-virgin olive oil or nuts. N Engl J Med.
2018;378(25):e34.
217. Guasch-Ferré M, et al. Total and subtypes of dietary
fat intake and risk of type 2 diabetes mellitus in the
Prevención con Dieta Mediterránea (PREDIMED)
study. Am J Clin Nutr. 2017;105(3):723–35.
218. Qian F, et al. Metabolic effects of monounsaturated
fatty acid-enriched diets compared with carbohy-
drate or polyunsaturated fatty acid-enriched diets in
patients with type 2 diabetes: a systematic review
and meta-analysis of randomized controlled trials.
Diabetes Care. 2016;39(8):1448–57.
219. Lewis EJH, et al. Effect of omega-3 supplementation
on neuropathy in type 1 diabetes: a 12-month pilot
trial. Neurology. 2017;88(24):2294–301.
220. Wanders AJ, et  al. Fatty acid intake and its
dietary sources in relation with markers of type
2 diabetes risk: The NEO study. Eur J Clin Nutr.
2017;71(2):245–51.
221. Bunner AE, et al. A dietary intervention for chronic
diabetic neuropathy pain: a randomized controlled
pilot study. Nutr Diabetes. 2015;5:e158.
222. The Look AHEAD Research Group. Effects of a
long-term lifestyle modification programme on
peripheral neuropathy in overweight or obese adults
with type 2 diabetes: the Look AHEAD study.
Diabetologia. 2017;60(6):980–8.
223. Diabetes Prevention Program Research Group.
Long-term effects of lifestyle intervention or met-
formin on diabetes development and microvascular
complications over 15-year follow-up: the Diabetes
Prevention Program Outcomes Study. Lancet
Diabetes Endocrinol. 2015;3(11):866–75.
S. A. Eid et al.
224. Müller-Stich BP, et  al. Gastric bypass leads to
improvement of diabetic neuropathy independent
of glucose normalization–results of a prospec-
tive cohort study (DiaSurg 1 study). Ann Surg.
2013;258(5):760–5; discussion 765–6
225. Davis TM, et  al. Effects of fenofibrate on renal
function in patients with type 2 diabetes mel-
litus: the Fenofibrate Intervention and Event
Lowering in Diabetes (FIELD) Study. Diabetologia.
2011;54(2):280–90.
226. Keech AC, et  al. Effect of fenofibrate on the need
for laser treatment for diabetic retinopathy (FIELD
study): a randomised controlled trial. Lancet.
2007;370(9600):1687–97.
227. Rajamani K, et al. Effect of fenofibrate on amputation
events in people with type 2 diabetes mellitus (FIELD
study): a prespecified analysis of a randomised con-
trolled trial. Lancet. 2009;373(9677):1780–8.
228. Cho YR, et al. Therapeutic effects of fenofibrate on
diabetic peripheral neuropathy by improving endo-
thelial and neural survival in db/db mice. PLoS One.
2014;9(1):e83204.
229. Kang EY, et  al. Association of statin therapy with
prevention of vision-threatening diabetic retinopa-
thy. JAMA Ophthalmol. 2019;137(4):363–71.
230. de Langen JJ, van Puijenbroek EP.  HMG-CoA-­
reductase inhibitors and neuropathy: reports to the
Netherlands Pharmacovigilance Centre. Neth J Med.
2006;64(9):334–8.
231. Tierney EF, et  al. Association of statin use with
peripheral neuropathy in the U.S. population 40
years of age or older. J Diabetes. 2013;5(2):207–15.
232. Pergolizzi JV, et al. Statins and neuropathic pain: a
narrative review. Pain Ther. 2020;9(1):97–111.
233. Kristensen FP, et al. Statin therapy and risk of poly-
neuropathy in type 2 diabetes: a Danish Cohort
Study. Diabetes Care. 2020;43(12):2945–52.
234. Oesterle A, Laufs U, Liao JK.  Pleiotropic effects
of statins on the cardiovascular system. Circ Res.
2017;120(1):229–43.
235. Tsimihodimos V, Liberopoulos E, Elisaf
M.  Pleiotropic effects of fenofibrate. Curr Pharm
Des. 2009;15(5):517–28.
236. de Anda-Jauregui G, et al. Pathway crosstalk pertur-
bation network modeling for identification of con-
nectivity changes induced by diabetic neuropathy
and pioglitazone. BMC Syst Biol. 2019;13(1):1.
 Targeting the Mitochondrion
in Diabetic Neuropathy

Ahmad Hedayat, Krish Chandrasekaran,
Lindsay A. Zilliox, and James W. Russell

1 Introduction mitochondrial function are encoded by nuclear

Mitochondria are critical in regulating energy
homeostasis in both neurons and Schwann cells
in diabetic neuropathy. They synthesize ATP
through oxidative phosphorylation but also have
an important role in integrating cellular metabo-
lism and cell signaling pathways. In addition to
the bioenergetic function of mitochondria in dia-
betes, research over the past 20 years shows that
mitochondria have ever more complex roles
(Fig.  1) in that mitochondrial bioenergetics is
integrated with other metabolic pathways, pro-
tein biogenesis, cellular signaling, quality con-
trol, immune and stress responses, and apoptosis.
Importantly, 99% of the proteins involved in
A. Hedayat · K. Chandrasekaran
Department of Neurology, University of Maryland
School of Medicine, Baltimore, MD, USA
L. A. Zilliox
Department of Neurology, University of Maryland
School of Medicine, Baltimore, MD, USA
Maryland VA Healthcare System,
Baltimore, MD, USA
J. W. Russell (*)
Department of Neurology, University of Maryland
School of Medicine, Baltimore, MD, USA
Maryland VA Healthcare System,
Baltimore, MD, USA
CAMC Institute for Academic Medicine,
Charleston, WV, USA
e-mail: James.Russell@CAMC.org
genes and are then manufactured in the cytosol,
where specific targeting signals direct the pro-
teins to mitochondrial surface receptors and then
into specific mitochondrial compartments. Thus,
the proteins are very specifically directed to form
a highly complex molecular machinery in which
mitochondrial bioenergetics is only a small part.
Mitochondria have been shown to be highly
dynamic both in the cytosol and within the axon,
and the mitochondria constantly undergo changes
in morphology due to fission and fusion as part of
the process of mitophagy related to mitochon-
drial quality control. These changes have been
shown to be important in diabetic neuropathy.
Furthermore, mitochondria are able to communi-
cate with other organ systems, for example, the
endoplasmic reticulum. Importantly, several
studies have shown over the past 20 years that in
diabetic neuropathy there is loss of absolute
mitochondria, a decrease in mitochondrial DNA,
and dysfunction of mitochondria that occurs at
multiple levels [1–6].

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 307
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_17
308
Fig. 1  Mitochondrial respiratory complexes, biogenesis,
and quality control. Mitochondrial respiratory chain
supercomplexes are integrated into functional networks
with the translocase of the inner membrane (TIM23) and
the ATP-dependent AAA proteases of the inner mem-
brane. The ATP-dependent AAA proteases (not shown)
act as a quality control system in mitochondria to degrade
select inner membrane (IM), intermembrane space (IMS),
matrix proteins, and outer membrane (OM) proteins. The
coenzyme Q (CoQ) biosynthetic complex is on the IM
2 Mitochondrial Mechanisms
and Potential Drug Targets
in DN
2.1 Role of Mitochondrial
Dysfunction in DN
Diabetes-induced-oxidative damage in neurons,
axons, and Schwann cells has been proposed as a
unifying DN mechanism. An increased metabolic
influx into mitochondria increases respiration and
A. Hedayat et al.
matrix side and provides CoQ for the respiratory chain
and enzymes. CoQ complex precursors are imported by
the translocase of the outer membrane (TOM) and TIM
machineries. The mitochondrial respiratory chain is the
main generator for reactive oxygen species (ROS) that
increases oxidation of proteins, reduces protein synthesis,
but can also has a role in signaling. CI complex I, CII
complex II, CIII complex III, CIV complex IV, MPP mito-
chondrial processing peptidase, Oct1 octapeptidyl pepti-
dase, PAM presequence translocase-associated motor
results in a high proton gradient, resulting in
increased reactive oxygen species (ROS) produc-
tion. The increase in generation of ROS results
from mitochondrial dysfunction that is induced by
an increase in mitochondrial inner membrane
depolarization and degeneration of mitochondria in
peripheral nerves [4, 6–10]. The regeneration of
new mitochondria may rescue these impaired mito-
chondria. Thus, activation of mitochondrial regen-
eration may be protective under conditions where
significant mitochondrial degeneration is present.
Targeting the Mitochondrion in Diabetic Neuropathy 309

Molecular methods allow manipulations of chondrial respiration proteins and protects and
the mouse genome to overexpress or knockout replicates the mitochondrial genome [3, 12, 15,
expression of individual genes. The ease of 16]. PGC-1α has been mapped to chromosome
genetic manipulation has provided important 4P15.1, a region associated with basal insulin
tools for study of the pathogenesis, prevention, levels in Pima Indians that are identified as hav-
and treatment of DN. There are three key genes, ing a high risk of developing diabetes and
which code for proteins that are transcriptional diabetic-­ related complications [17]. Common
factors or activators of transcriptional factors and polymorphisms of PGC-1α are associated with
promote function and regeneration of mitochon- conversion from impaired glucose tolerance to
dria (Fig.  2). These are: the NAD+-dependent diabetes [18]. Thus, it is likely that PGC-1α and
deacetylase SIRT1 (Silent mating type informa- its downstream signaling intermediates are
tion regulation 2 homolog1), Peroxisome important in normal mitochondrial regulation in
proliferator-­
activated receptor-gamma co-­diabetic subjects [19]. We used PGC-1α knock-
activator 1α (PGC-1α), and mitochondrial out mice [PGC-1α (−/−)] to study the effect in
Transcription Factor A (mtTFA or TFAM). DN. Loss of PGC-1α, a protein that is essential
Peroxisome proliferator-activated receptor-­ for mitochondrial function and energy homeosta-
gamma co-activator 1α (PGC-1α) is a transcrip- sis, causes neuropathy that is worsened by diabe-
tional co-activator and a master regulator for tes [4]. This observation was supported by
mitochondrial biogenesis in many tissues includ- evidence from other groups showing that phos-
ing the nervous system [11–13]. PGC-1α is a phorylation and expression of adenosine
promising target for therapy for neurological dis- monophosphate-­ activated protein kinase
ease. For example, the pan-PPAR agonist, bezafi- (AMPK)/PGC-1α and mitochondrial respiratory
brate, upregulates PGC-1α and exerts beneficial chain complex proteins are downregulated in
effects in a transgenic mouse model of dorsal root ganglia of both diabetic rats and mice
Huntington’s disease [14]. PGC-1α activates [6, 20, 21]. Control of bioenergetics in the dia-
transcriptional factors such as nuclear respiratory betic peripheral nerve is in part controlled by the
factor 1 (NRF1) and mitochondrial transcription SIRT1-AMPK and PGC-1α signaling axis in
factor A (TFAM), which in turn induces mito- which the upstream members (SIRT1-AMPK)

Fig. 2  The majority of mitochondrial proteins are synthe-
sized in the nucleus and shuttled to the mitochondria. Key
transcription factors such as SIRT1, PGC-1α, and TFAM
are important in regulating mitochondrial function and
preserving mitochondrial DNA in the peripheral nervous
system in response to increased metabolic loads provided
by hyperglycemia or hyperlipidemia in diabetes. SIRT1
regulates PGC-1α activity via deacetylation, whereas
PGC-1α regulates expression of TFAM via nuclear respi-
ratory factor, NRF1
310
sense the metabolic demands of cells, whereas
the downstream member, PGC-1α, regulates
mitochondrial biogenesis and function [6, 20–
23]. There was no difference in plasma glucose,
hemoglobin A1C, or insulin levels between
PGC-1α wild-type diabetic and PGC-1α knock-
out diabetic mice, indicating that exacerbation of
peripheral nerve injury associated with reduced
PGC-1α levels is independent of glycemic con-
trol. However, there was a significant increase in
total cholesterol and triglyceride levels in nondia-
betic PGC-1α knockout compared to PGC-1α
wild-type mice and PGC-1α knockout diabetic
compared to PGC-1α wild-type diabetic mice.
These findings are consistent with current obser-
vations that increased lipid levels are associated
with neuropathy [24] possibly by ox-LDL inter-
action with the receptor LOX-1 [25–27]. Impaired
fatty acid oxidation is strongly correlated to
impaired activation of PGC-1α [28] and adipose-­
specific ablation of PGC-1α results in mitochon-
drial dysfunction [29]. Thus, mitochondrial
PGC-1α has a critical role in lipid metabolism.
Loss of PGC-1α may also worsen neuropathy
due to impaired antioxidant defense related to
impaired mitochondrial function. Several lines of
evidence support this idea: PGC-1α was mark-
edly reduced in diabetic DRG neurons; the mito-
chondria in neurons from PGC-1α nondiabetic
mice contained abnormal vacuoles consistent
with mitochondrial degeneration; there was a
decrease in mitochondrial respiratory function;
gene expression of glutathione peroxidase and
SOD2 were reduced; protein levels of transcrip-
tion factors that regulate mitochondrial transcrip-
tion and are important for normal mitochondrial
homeostasis (for example, TFAM and NRF1)
were significantly decreased in dorsal root gan-
glion neurons of both PGC-1α knockout nondia-
betic and diabetic mice; levels of oxidized
proteins consistent with increased oxidative
stress were significantly increased; and overex-
pression of PGC-1α, using an adenoviral con-
struct, prevented the generation of ROS in adult
mouse DRG neuron cultures [4]. The role of
PGC-1α in mitochondrial regulation are sup-
ported by: (1) data showing an inverse correlation
between muscle PGC-1α levels and mitochon-
A. Hedayat et al.
drial activity in humans with insulin resistance
and diabetes [30, 31] and (2) previous studies
showing PGC-1α knockout mice have degenerat-
ing mitochondria in the CNS [32].
TFAM lies downstream of PGC-1α and has an
important role in protecting the integrity of mito-
chondria. In humanized TFAM overexpressing
transgenic mouse model, levels of TFAM are
increased only twofold in DRG neurons [3].
Overexpression of TFAM reversed experimental
DN.  TFAM prevents slowing of nerve conduc-
tion velocity, decreases intra epidermal nerve
fibers’ loss, and reduces mechanical allodynia.
The transgenic mice expressing human TFAM
protein elevated mitochondrial DNA (mtDNA)
copy number and total TFAM levels to the same
extent. This suggests that the mtDNA copy num-
ber itself does not have influence upon TFAM
gene expression. The suggested explanation is
that the interaction between mtDNA and TFAM
is interrelated and that the existence of one aug-
ments the stability of the other. This interaction is
possibly valuable from a regulatory point of view
since minor changes in mtDNA or TFAM protein
levels result in swift modification to preserve a
constant ideal ratio between TFAM and mtDNA
[33]. This is supported by the studies that mtDNA
interacts with TFAM to wrap mtDNA to consti-
tute a nucleoid formation similar to histones in
the nucleosome [34–36].
Meanwhile, mitochondria can generate ROS
due to their respiratory activity, contingently
TFAM functions in multiple roles to promote
mtDNA transcription, replication, and wraps
mtDNA to protect it from ROS attack [34–36].
The blood studies demonstrated that there was no
difference in glucose or insulin levels between
WT diabetic and TFAM transgenic diabetic mice.
Although human TFAM is likely to be expressed
in all tissues because the promoter is not tissue-
or cell-specific, TFAM overexpression does not
reduce the severity of diabetes. Thus, protection
against peripheral nerve injury is localized to
peripheral nervous system mitochondria and is
independent of glycemic control. The results
showed that there is a net loss of both mtDNA
and TFAM in chronic experimental diabetes.
After acute diabetic exposure, there is an attempt
Targeting the Mitochondrion in Diabetic Neuropathy
to upregulate both TFAM and mtDNA, whereas
the increase is not significant. It could be argued
that exposure to acute hyperglycemia rises TFAM
levels, mtDNA, and mitochondrial biogenesis to
meet the high-energy demand in neurons [37].
However, in chronic hyperglycemia with a con-
current generation of ROS, the regulation might
change from physiological to pathological and
could eventually lead to a decline in mtDNA,
TFAM, and mitochondrial function. MtDNA is
particularly susceptible to oxidative injury due in
part to the following factors: first, its location
within mitochondria where the respiratory com-
plexes I and III are potential sites for the genera-
tion of superoxide radicals and second, the
limited repair activity against DNA damage
within mitochondria [38]. Under normal condi-
tions, the toxic effects of ROS are prevented by
such scavenging enzymes as superoxide dis-
mutase, glutathione peroxidase, catalase, and
other non-enzymatic antioxidants. However,
when ROS production becomes excessive, or if
the levels of antioxidant enzymes decrease, then
oxidative stress might harm the functional and
structural integrity of biological tissue. DRG
neurons from TFAM transgenic mice are able to
scavenge high glucose-induced ROS much more
efficiently than those from WT mice. However,
this scavenging ability did not appear to be due
an increase in the expression of the antioxidant
enzymes SOD2 or glutathione peroxidase. Our
results with mitochondrial respiration showed no
significant increase in ADP-stimulated State 3
and resting State 4 respiration. All these findings
suggest that the protective effect of TFAM over
expression is apparent only under chronic condi-
tions of oxidative stress.
SIRT1 acts upstream to coordinate the SIRT1/
PGC-1α/TFAM pathway to regulate mitochon-
drial oxidative energy metabolism and neuronal
protection. Using an inducible neuron-specific
SIRT1 overexpressing (SIRT1OE) C57BL6
mouse, SIRT1 is able to reverse DN in high fat
fed diabetic mice [39]. Diabetes reduces NAD+,
SIRT1, and PGC-1α levels in DRG neurons.
SIRT1 overexpression normalizes these levels.
These findings are consistent with the concept
that the SIRT1/PGC-1α/TFAM pathway can
311
reverse neuropathy in a model of type 2 diabetes
mellitus (T2DM). SIRT1 requires NAD+ as a
cofactor to deacetylate PGC-1α and other tran-
scriptional factors. Treatment with a precursor of
NAD+, Nicotinamide Riboside (NR), can reverse
neuropathy in the HFD mouse that is a model of
T2DM [40–42]. Adult C57BL6 mice were fed a
high-fat diet (HFD, 60% calories from fat) for 2
months until they developed neuropathy. Then,
150 mg/kg or 300 mg/kg NR was mixed with
HFD and fed every day for 2 months [40, 41].
Neuropathy improved with treatment in both
groups but more effectively with 300 mg/kg
NR. There was no change in the control diet ani-
mals. NR treatment decreased the HFD-induced
increase in triglycerides and non-esterified fatty
acids and normalized the impaired glucose toler-
ance test. In HFD mice, there was a decrease in
the NAD+ level, in SIRT1 activity, and in PGC-1α
levels in DRG neurons. NR normalized these
measurements.
2.2 Changes in Redox Potential
in the Mitochondrion in DN
Oxidative stress is one of the most important
potential mechanisms of injury to the peripheral
nervous system and most is generated from the
mitochondrion [7, 43]. Mitochondrial stress is
particularly important in T2DM but to a lesser
extent is important in type 1 diabetes mellitus
(T1DM) [3, 4, 6, 23]. Stress states and dysfunc-
tion of the respiratory chain can lead to increased
reactive oxygen species (ROS) production that
causes oxidative damage in proteins, DNA, and
membranes. As ROS also have a signaling role,
mitochondrially produced ROS can signal the
functional state of mitochondria [44–47]. In
experimental diabetes, oxidative stress levels
and reduced antioxidant defense are associated
with neuropathy. In comparison, blocking oxi-
dative stress in diabetic animals re-establishes
normal blood flow and sciatic and saphenous
nerve conduction velocities. Augmented meta-
bolic mitochondrial flux due to high glucose
results in increased reactive oxygen species
(ROS) formation such as superoxide, peroxyni-
312
trite, hydroxyl radicals, and impaired mitochon-
drial respiratory function [21]. ROS generation
enhances mitochondrial permeability transition,
opening the adenine nucleotide translocase/
voltage-­dependent anion channel resulting in
mitochondrial swelling that distorts the outer
membrane’s integrity [1, 4]. The resulting mem-
brane lipid peroxidation and DNA degradation
are linked to neuronal or axonal injury [48].
Mitochondria become swollen and develop dis-
rupted cristae [48], and there is loss of mito-
chondria [4].
In contrast, upregulation of pathways that
block oxidative stress precludes axonal injury in
diabetic neuropathy. For example, overexpres-
sion of manganese superoxide dismutase (SOD2),
the enzyme responsible for mitochondrial detoxi-
fication of oxygen radicals, reduces superoxide in
cultured dorsal root ganglion (DRG) neurons and
inhibits cellular injury. Diabetic neuropathy is
observed in the C57BL/6Jdb/db diabetic mouse,
and decreased expression of SOD2 in these ani-
mals increased diabetic neuropathy [49]. The
generation of ROS can be decreased by stabiliz-
ing the inner mitochondrial membrane potential
and reducing generation of superoxide and other
radicals or by increasing antioxidant defense.
Rising levels of uncoupling proteins such as
UCP1 or UCP3 that prevent hyper-polarization
of the inner mitochondrial membrane potential
will reduce ROS formation and ameliorate neuro-
nal injury [1].
An important marker of oxidative stress is a
reduction in the glutathione system that protects
diabetic neurons and Schwann cells from injury.
Diabetic patients with distal symmetric polyneu-
ropathy exhibited significantly lower values of
reduced glutathione (GSH) and a reduced gluta-
thione/oxidized glutathione (GSH/GSSG) ratio
[50]. In contrast, malondialdehyde and oxidized-­
LDL levels were unchanged in diabetic neuropa-
thy. Another recent approach has been to
manipulate Nrf2, which mediates various anti-
oxidant proteins’ expression via antioxidant
response element (ARE) binding sites. Nrf2
appears to bridge the link between neuroinflam-
mation and apoptotic pathways impacting the
progression of diabetic neuropathy [51]. Under
A. Hedayat et al.
normal conditions, Nrf2 is inactive and remains
in the cytosol. Acute hyperglycemia increases
the expression of Nrf2, but persistent hypergly-
cemia decreases Nrf2 expression.
Downregulation of Nrf2 causes various micro-
vascular changes that result in diabetic neuropa-
thy. Targeting Nrf2 activators as potential
therapies may provide insight into new therapies
for diabetic neuropathy. For example, natural
compounds such as curcumin, sulforaphane, res-
veratrol, and vitamin D can activate Nrf2 and
promote antioxidant pathways to mitigate oxida-
tive stress [52]. Another factor that can trigger
Nrf2 is taurine, found naturally in the body.
Supplementing diabetic rats with taurine did not
alter body weight and blood glucose concentra-
tion, but reduced the serum malondialdehyde
concentration in diabetic rats and alleviated neu-
roinflammation by suppressing NF-kappaB
expression and enhancing Nrf2 [53].
Fisetin, a phytoflavonoid that simultaneously
targets NF-kappaB and Nrf2, shows a beneficial
effect in experimental DN [54]. Fisetin was
administered at 5 and 10 mg/kg for 2 weeks in
streptozotocin (STZ) diabetic rats. As deter-
mined by nerve conduction studies and sciatic
nerve blood flow deficits, neuropathy was ame-
liorated in treated rats. Importantly, fisetin
reduced levels of interleukin-6 and tumor necro-
sis factor-alpha in sciatic nerves of diabetic rats
(p < 0.001), suggesting that a combined approach
to modifying oxidative stress and neuroinflam-
mation may be a more effective approach. The
mechanistic target of rapamycin (mTOR) is
another promising agent for the development of
novel regenerative strategies for the treatment of
diabetes mellitus. The mTOR pathway impacts
numerous cellular processes including cellular
metabolic homeostasis, insulin resistance, insu-
lin secretion, apoptosis, and autophagy [55].
mTOR is critical for many signaling pathways
that involve phosphoinositide 3-kinase (PI 3-K),
protein kinase B (Akt), AMP-­activated protein
kinase (AMPK), sirtuin 1 (SIRT1), Wnt1 induc-
ible signaling pathway protein 1 (WISP1), and
insulin/insulin growth factor 1 [56]. mTOR, sir-
tuins, and insulin/insulin growth factor 1 signal-
ing are longevity pathways involved in an array
Targeting the Mitochondrion in Diabetic Neuropathy 313

of different processes, including metabolism, peripheral nervous tissue [65]. Tissue glucose is
and neuronal plasticity, that are critical in pre- metabolized to 3-deoxyglucosone (3-DG),
venting neuropathy and in axonal repair. As a methylglyoxal, and Nε-(carboxymethyl) lysine.
result, mTOR represents an exciting target for Binding of these AGE ligands to the receptor for
the treatment of DN. advanced glycation end-products (RAGE) can
Nitrosative stress can increase nerve injury in result in diabetic complications. AGE and
diabetes [57]. Peroxynitrite resulting from the RAGE accumulation in nerve perineurial, epi-
reaction of nitric oxide with superoxide causes neurial, and endoneurial microvessels of
oxidative stress and leads to poly ADP-­ patients with impaired glucose tolerance (IGT)
ribosylation, mitochondrial dysfunction, results in increased expression of the transcrip-
impaired stress signaling, and protein nitration. tion factor nuclear factor κB (NFκB) [66, 67].
In diabetic mice treated with the peroxynitrite AGE:RAGE interactions further lead to the gen-
decomposition catalyst Fe(III) tetramesitylpor- eration of ROS (ROS) and the activation of pro-
phyrin octasulfonate (FeTMPS, 10 mg/kg/day) tein kinase C (PKC), transforming growth factor
or protein nitration inhibitor(-)-epicatechin gal- β (TGF-β) [68], mitogen-activated protein
late (20 mg/kg/day) for 4 weeks, there was partial kinase (MAPK), and transcription factors like
correction of neuropathy [58]. This supports the nuclear factor κB (NFκB) [69].
concept that peroxynitrite decomposition cata- RAGE is thought to be one of the potential
lysts may be used in the treatment of diabetic contributors to neurotoxicity [70]. It has been
peripheral neuropathy. In humans, gastric bypass shown that RAGE activation leads to an increase
and in particular Roux-Y gastric bypass (RYGB) in proinflammatory molecules, oxidative stress-
can improve neuropathic symptoms independent ors, and cytokines. The proinflammatory effect of
from glycemic control. The procedure is associ- RAGE is mediated by its binding to ligands such
ated with improved nitrosative injury [59], and as AGE, S100/calgranulin, and amphoterin.
the decrease in nitrotyrosine is correlated with Activation of these ligands leads to subsequent
improvement in the neuropathy disability scale activation of downstream pathways such as
up to 12 months. NF-kappaB, STAT, and JKN. An important target
for the treatment of AGE-induced diabetic nerve
injury is the glyoxalase system. The glyoxalase
2.3 Advanced Glycation End system is a highly specific enzyme system exist-
Products ing in all mammalian cells that is responsible for
and the Mitochondrion in DN the detoxification of dicarbonyl species, primar-
ily methylglyoxal. It has been implicated to play
Dicarbonyl metabolites, for example methylgly- an essential role in preventing the increased for-
oxal, glyoxal and 3-deoxyglucosone, accumula- mation of AGEs. When glyoxalase 1 is knocked
tion has been associated with metabolic and out in cells, the formation of methylglyoxal and
age-related diseases in which there is a pro-­ associated protein modifications were blocked
inflammatory and pro-oxidant state [60]. [71], suggesting that glyoxalase 1 could be an
Methylglyoxal is the most reactive dicarbonyl important target for treatment in DN [72]. Other
and is the precursor of the major quantitative studies have shown that pyridoxamine prevents
advanced glycation end products (AGEs). the formation of AGEs by trapping carbonyl
Improved mitochondrial function has been intermediates and can reduce development of
linked to a decrease in generation of AGEs in a vascular disease and nephropathy in the nondia-
variety of tissues [61], and specifically in rela- betic, Zucker obese rat [73], but little is known
tion to diabetic neuropathy and other diabetic about its effect in reducing human diabetic neu-
complications [62–64]. In contrast, increased ropathy in human studies. Other natural inhibi-
glycemia results in reduced mitochondrial func- tors of AGEs have not been tested in diabetic
tion and induces the generation of AGEs in neuropathy.
314
2.4 Mitochondrial Regulation
of Molecular Chaperones
in DN
Heat shock proteins (HSPs) or molecular chaper-
ones also have roles in cell signaling and regula-
tion of metabolism. Overexpression and
pharmacological induction of HSP72 prevent
high-fat diet-induced glucose intolerance and
skeletal muscle insulin resistance [74].
Furthermore, overexpression of skeletal muscle
HSP72 in mice increases endurance nearly two-
fold and increases mitochondrial content by 50%.
HSP72 mRNA expression correlates well with
mitochondrial enzyme activity in human skeletal
muscle. In contrast, HSP72 expression is mark-
edly decreased in skeletal muscle in humans with
type 2 diabetes.
Individual molecular chaperones and chaper-
one complexes are expressed under normal and
disease states. Molecular chaperones like heat
shock protein 90 (Hsp90) and Hsp70 assist in
folding nascent polypeptides into their final bio-
logically active conformations and in the refold-
ing of aggregated and denatured proteins. As a
result, the chaperones help direct proteins toward
degradation via the proteasome or by autophagy
[75, 76]. Pharmacologic modulation of the
molecular chaperones can improve insulin resis-
tance [77] and peripheral neuropathy [78]. Hsp90
modulators such as DMAG, KU-32, and KU-596
show promise in treating diabetic complications
in animal models and may afford a novel and
effective disease-modifying approach for humans
[79].
2.5 TGF-β and DN
Transforming growth factor-β1 (TGF-β1) is an
important mediator of fibrosis in diabetes. Data
from human biopsy material and experimental
models indicate increased expression of IHG-1 is
a critical component of fibrogenesis as it ampli-
fies TGF-β1 signaling and biogenesis [80]. IHG-1
is expressed in mitochondria, stabilizes PGC-1α
protein, and increases mitochondrial biogenesis.
TGF-β1 has been associated with diabetic
A. Hedayat et al.
nephropathy and retinopathy because of its abil-
ity to increase extracellular matrix.
There is evidence that TGF-β is upregulated in
STZ diabetic rats with neuropathy. In DRG and
sciatic nerve from diabetic animals, TGF-beta1
and TGF-beta2 are increased and are associated
with increased neuronal death [68]. In diabetic
DRG neurons using quantitative real-time PCR
(QRT-PCR), TGF-β1 and TGF-β2 mRNA, but
not TGF-β3, were increased at the 4 and 12 week
time point [68]. In the sciatic nerve, TGF-β3
mRNA was primarily increased. In DRG neu-
rons, pretreatment with TGF-β neutralizing anti-
body prevented the increase in total TGF-β
protein observed with high glucose. In DRG neu-
rons exposed to high glucose, TGF-β2 > β1
increased the percent of cleaved caspase-3 com-
pared to high glucose alone, and TGF-β neutral-
izing antibody inhibited the increase. TGF-β
isoforms applied directly to DRG neurons
reduced neurite outgrowth, and this effect was
partially reversed by TGF-β neutralizing anti-
body. These findings implicated upregulation of
TGF-beta in experimental diabetic peripheral
neuropathy and suggested a potential new target
for the treatment of diabetic peripheral
neuropathy.
Patients with T2DM and clinically detectable
serum TGF-β1 showed a positive correlation with
nerve conduction velocities, suggesting that this
cytokine might be used as a biomarker for dia-
betic peripheral neuropathy [81]. TGF-β increases
Smad3 signaling, thereby affecting metabolism
and energy homeostasis and reducing inflamma-
tion and ROS production [82].
2.6 Metabotropic Glutamate
Receptors, Oxidative Injury,
and DN
Metabotropic glutamate receptors (mGluRs) are
a subfamily of glutamate receptors that are
G-protein-coupled and linked to second messen-
ger systems [83]. Glutamate is the principal neu-
rotransmitter in the CNS that is transported to the
axon terminals [84]. The mechanism for produc-
tion, paracrine release, and recycling of gluta-
Targeting the Mitochondrion in Diabetic Neuropathy
mate happens in sensory ganglia and contains the
enzymes amidohydrolase, glutaminase [85, 86],
glutamate aspartate transporter (GLAST), gluta-
mate transporter 1 (GLT1) [87], as well as the
recycling enzyme glutamine synthetase [85, 88].
The Glutamate carboxypeptidase II (GCP II)
inhibitor 2-(phosphonomethyl)pentanedioic acid
(2-PMPA), which activates mGluR2/3, protects
against glucose-induced programmed cell death
(PCD) and neurite degeneration in DRG neurons
in a cell culture model of DN [83]. Preclinical
data indicate that GCPII inhibitors ameliorate
DN in animal models. Direct or indirect activa-
tion of mGluR2/3 protects against development
Fig. 3  Glutamate transporters GLT-1 and GLAST are
present in Satellite Glial Cells (SGC) cells to transport
extracellular glutamate into the SGC where glutamine
synthetase (GS) converts it to glutamine, which is eventu-
ally recycled to the neuron for conversion into glutamate.
In diabetes, hyperglycemia-induced oxidative stress
affects glutamate transport proteins, increasing extracel-
315
of DN in animal models [89]. mGluR2/3 ago-
nists prevent glucose-induced neuronal injury in
DRG neuronal cultures by increasing glutathi-
one and maintaining mitochondrial function
only in the presence of Schwann/SGC cells [83,
90, 91]. Treatment with a mGluR2/3 agonist nor-
malizes levels of glutathione and oxidized pro-
teins, while increasing levels of superoxide
dismutase 2 (SOD2), SIRT1, PGC-1alpha,
TFAM, glutamate transporter proteins, and glu-
tamine synthetase in DRG neurons. Thus,
mGluR3 agonists have the ability to protect
against cellular injury by regulating oxidative
stress in models of DN (Fig. 3).
lular glutamate. LY379268 treatment promotes glutamate
uptake, and it likely decreases extracellular glutamate. In
addition, mGluR 2/3 receptors present in DRG neurons
decrease extracellular glutamate. Thus, activation of
mGluR2/3 receptor by an agonist can likely constrain sen-
sory transmission and importantly reduce nociceptive
transmission
316
2.7 Mitophagy,
the Inflammasome,
and Innate Immune Pathways
in DN
Recent results suggest that mitochondrial impair-
ment contributes to inflammasome formation and
activation of the innate immune system.
Maintaining functional mitochondria is essential
to generate energy (ATP) for maintaining resting
potentials, action potentials, and neurotransmis-
sion at both the pre- and post-synaptic sites in the
neuron [92, 93]. Mitochondria, therefore, need to
be repaired and replaced [94]. Mitochondrial bio-
genesis, axonal transport, and mitochondrial fis-
sion and fusion contribute to this rejuvenation
[95]. Mitochondrial stress has been proposed as a
major mediator of diabetic complications such as
neuropathy in humans and animal models [7, 43,
96]. Therefore, in order to maintain a healthy
mitochondrial population, damaged proteins and
organelles must be cleared. Mitochondria
undergo a process of regeneration and autophagy
(clearance), termed mitophagy. Impaired mito-
chondrial quality control can cause disease, for
example, mitofusin 2 mutations in Charcot-Marie
Tooth disease type 2A [97, 98] and Opa1 muta-
tions in autosomal-dominant optic atrophy [99–
101]. In addition, impairment of mitochondrial
quality control has been demonstrated to activate
innate immune pathways, including inflamma-
some (NLRP3)-mediated signaling and the anti-
viral cyclic GMPAMP synthase (cGAS)/
stimulator of interferon genes (STING)–regu-
lated interferon response [reviewed in [95]].
Innate immune-signaling pathways have evolved
in complex multicellular eukaryotes to recognize
invasive pathogens, including bacteria, viruses,
fungi, and protists. Damage-associated mito-
chondrial protein (DAMP) is recognized by the
cell’s immune-recognition pathways as foreign
due to the resemblance to ancient prokaryotes
[102]. Sensing of these DAMPs by NOD-, LRR-,
and pyrin domain–containing proteins (NLRPs)
has been shown to activate inflammatory immune
responses [103]. Like the problems posed by
mitochondria’s resemblance to ancient prokary-
otes, the presence of a second nonnuclear com-
A. Hedayat et al.
plement of DNA in the mitochondria presents a
range of challenges to the host immune system
[104]. Release of mtDNA has been shown to trig-
ger DNA-sensing antiviral pathways [104].
Activation of the innate immune system occurs in
peripheral neuropathy and is of importance. For
example, vincristine-induced peripheral neuropa-
thy is driven by activation of the NLRP3 inflam-
masome and subsequent release of interleukin-1β
from macrophages. However, if inflammation is
reduced by treatment with the NLRP3 inhibitor
MCC950, or if there is knockout of NLRP3  in
mice, then there is a significant reduction in
mechanical allodynia and gait impairment [105,
106]. Similarly, paclitaxel can induce high-level
expression of NLRP3 inflammasomes in the
infiltrated macrophages within DRG and sciatic
nerve. This in turn promotes IL-1β production
and mechanical allodynia in a chemotherapy-­
induced neuropathic pain model [107].
The NF-kappaB pathway is part of the central
machinery initiating and propagating inflamma-
tory responses, and this is true also for DN. The
NF-kappaB inflammatory cascade may be inhib-
ited by BAY 11-7082, an IkappaB phosphoryla-
tion inhibitor [108]. BAY 11-7082 (1 and 3 mg/
kg) was administered to STZ diabetic rats for 14
days starting from the end of the 6th weeks post
diabetic induction. Diabetic rats developed
altered nociceptive parameters and deficits in
nerve functions and also demonstrated elevated
expression of NF-kappaB (p65), IkappaB, and
p-IkappaB along with elevated levels of IL-6 and
TNF-alpha and inducible enzymes (COX-2 and
iNOS). In addition, DN was associated with a
surge in oxidative stress, a reduction in Nrf2/
HO-1 expression, and an amelioration in GSH
levels. BAY 11-7082 improved abnormal sensory
responses and deficits. BAY 11-7082 also facili-
tated the surge in the expression of NF-kappaB,
IkappaB, and p-IkappaB. BAY 11-7082 restrained
the levels of IL-6, TNF-alpha, COX-2, and iNOS
in the sciatic nerve. Thus, it can be concluded that
NF-kappaB expression and downstream expres-
sion of proinflammatory mediators are significant
features of nerve damage resulting in inflamma-
tion and oxidative stress. Furthermore, BAY
11-7082 is able to improve experimental DN by
Targeting the Mitochondrion in Diabetic Neuropathy
modifying neuroinflammation and improving
antioxidant protection.
3 Current Treatment
Approaches That Target
Mitochondrial Function
in DN
3.1 Lifestyle Modification
and Improvement
in Mitochondrial Function
in DN
There are currently no disease-modifying treat-
ments that have been definitively shown, in
­randomized clinical trials, to reduce or reverse
diabetic sensory polyneuropathy. However, indi-
viduals with impaired glucose regulation and
neuropathy may benefit from aggressive diabetic
control and lifestyle interventions. These inter-
ventions can postpone the onset of diabetes and
may reverse small fiber neuropathy associated
with early diabetes mellitus. Exercise not only
increases insulin sensitivity and glucose control,
but also increases end organ perfusion, reduces
lipid and protein oxidation, inhibits adipocyte
production of free fatty acids and deleterious adi-
pokines, and reduces humoral inflammation [1, 9,
109–111]. The mechanisms by which lifestyle
changes can affect mitochondrial function have
previously been described [109]. Although
dietary and lifestyle interventions are likely to
affect multiple pathways, there is evidence that
improved mitochondrial function in muscle
might exert a remote effect. AMP-activated pro-
tein kinase (AMPK) and the protein deacetylase,
sirtuin 1 (SIRT1), are fuel sensing molecules.
During energy deprivation, AMPK activation
restores energy balance by increasing levels of
ATP, for example by increasing fatty acid oxida-
tion, and reducing processes that consume
ATP.  AMP-activated protein kinase has been
shown to circulate in the blood after release from
muscle [112]. This could explain why exercise,
which improves muscle function, can also have
remote effects on somatic and autonomic nerve
fibers. Furthermore, in exercised muscle, nutrient
317
deprivation leads to an increase in NAD+ and a
decrease in nicotinamide (NAM), which in turn
leads to activation of SIRT1 and 35 kDa peroxi-
some proliferator-activated receptor-gamma
co-activator-1α (PGC1-α) in the mitochondrion,
resulting in an increase in oxidative phosphoryla-
tion and fatty acid metabolism [2, 39, 109, 113].
NAM and NADH act as inhibitors of SIRT1.
NAM phosphoribosyltransferase (Nampt) is
important in the conversion of NAM to NAD+.
Exercise increases Nampt activity in muscle but
also circulates outside the muscle to increase
insulin sensitivity. Nampt then decreases NAM
and increases NAD+ resulting in further activa-
tion of the SIRT1-PGC1-α pathway and increas-
ing mitochondrial function. AMP-activated
protein kinase (AMPK) also likely regulates
SIRT1 by acting as an energy sensor and acti-
vates Nampt. Thus, the net effect of nutritional
restriction and exercise is to increase NAD+ and
drive the SIRT1-PGC1-α pathway toward
improved mitochondrial function.
In addition, changes in fatty acid metabolism
likely affect mitochondrial function in the mus-
cle. Translational research indicates that high-fat
diets can result in neuropathy in diabetes and that
withdrawing or otherwise manipulating a high fat
diet can reduce neuropathy [114–116]. In
humans, a nutritional intervention study that was
targeted to improve essential fatty acid dys-
metabolism in type 1 diabetes mellitus examined
supplementation with seal oil omega-3 polyun-
saturated fatty acids in individuals with type 1
diabetes mellitus and neuropathy. After 12
months of supplementation, there was an increase
in corneal nerve fiber length [117]. There was no
progression of clinical disease symptoms, and
there were no declines in small and large fiber
sensory and functional measures. However, there
was no improvement detected in nerve conduc-
tion studies or sensory function.
3.2 Improved Glycemic Control
and DN
As discussed above, improved glycemic control
will reduce mitochondrial dysfunction and oxida-
318
tive or nitrosative stress, consequently reducing
protein and lipid oxidation, neuroinflammation,
and neuronal and Schwann cell injury signaling.
Early hyperglycemic insult can lead to perma-
nent, cumulative damage. Mitochondrial injury
takes place after just 1 h of hyperglycemic expo-
sure due to an increase in mitochondrial mem-
brane potential likely caused by a decrease in the
cells’ NAD+/NADH ratio [118]. The Diabetes
Control and Complications Trial (DCCT) deter-
mined a clear link between impaired glycemic
control, neuropathy, and retinopathy. This DCCT
prospectively followed 1441 participants with
T1DM for a mean of 6.5 years to study the effect
of intensive insulin on the development of dia-
betic complications [reviewed in [111]]. The
study divided participants into a primary (preven-
tion) and a secondary (intervention) group and
treated them with intensive or conventional insu-
lin therapy. In the secondary-intervention cohort,
intensive insulin therapy reduced clinical neu-
ropathy’s appearance by 60% at the 5-year fol-
low-­up. If patients had neither retinopathy nor
significant albuminuria at the start of the study
(primary-prevention cohort), then the results
were even more impressive: intensive therapy
reduced the appearance of neuropathy by 69%
compared to only 10% with conventional therapy
and showed that early optimal glycemic control
could prevent the development of neuropathy
before patients developed microvascular injury
and retinopathy.
Although studies show clear improvement in
diabetic and somatic neuropathy outcomes for
T1DM, the data for T2DM is less clear [119].
Despite these reservations, improvement in
T2DM was observed in the UK Prospective
Diabetes Study (UKPDS) study. In this study,
3867 patients with newly diagnosed T2DM were
randomly assigned to intensive therapy with a
sulfonylurea (chlorpropamide, glibenclamide, or
glipizide), with insulin, or conventional diet ther-
apy. After 10 years, risk of microvascular end-
points was reduced by 25% (p = 0.0099) [120].
Thus, intensive blood glucose control by either
sulfonylureas or insulin substantially decreased
the risk of microvascular complications in
patients with T2DM; however, there was a higher
A. Hedayat et al.
rate of hypoglycemic complications, particularly
in the insulin treatment group [120].
Other studies have shown a less definitive
change in neuropathy in T2DM with improved
glycemic control. For example, the VA Diabetes
Trial randomized 1791 military veterans with
T2DM to either intensive or standard glucose
control. The development of neuropathy was
determined based upon self-report. While the
study did not find any difference in the rate of
new microvascular complications, such as neu-
ropathy, there was a non-significant 5% reduction
in the incidence of neuropathy [121]. Not all
studies have shown that aggressive glucose con-
trol reduces the development or progression of
neuropathy in T2DM. The VA cooperative study
on type II diabetes mellitus (VA-CSDM) com-
pared standard insulin treatment to an intensive
therapy group in men with T2DM who required
insulin. The enrolled men had poorly controlled
T2DM with an average duration 7.8 ± 4 years and
about half had neuropathy at baseline. Intensive
insulin therapy did result in an improvement in
HgBA1c after 6 months, but no difference was
found in the prevalence of DN between the inten-
sive vs. standard arms (64% intensive vs 69%
standard) after 2 years [122].
Importantly, improved glycemic control has
been shown to have a sustained benefit on diabe-
tes and its complications. The NeuroEDIC study
found a reduced prevalence of neuropathy in a
group of type 1 diabetics that received intensive
treatment compared to the standard treatment
group during the DCCT study. However, although
intensive glycemic control reduced the severity
of neuropathy, 34% of subjects in the former
intensive treatment group and 41% of those in the
former conventional treatment group still devel-
oped clinical neuropathy [123].
3.3 Diet and Lifestyle
Interventions in DN
Glucose-modifying drug therapy is typically not
appropriate in patients with impaired glucose
regulation due to cost and potential for serious
side effects such as hypoglycemia. A more suit-
Targeting the Mitochondrion in Diabetic Neuropathy
able approach for these patients would be a life-
style intervention that could arrest the underlying
process that leads to neuropathy and its associ-
ated functional disability. At present there is no
evidence from a randomized study that a lifestyle
intervention would reverse somatic neuropathy.
However, there is evidence that a lifestyle inter-
vention can be more effective than a drug inter-
vention in preventing conversion from IGT to
diabetes. The Diabetes Prevention Program
(DPP) study was designed to determine if a more
intensive dietary and exercise intervention or
metformin treatment effectively prevented or
delayed the onset of T2DM in people with IGT or
impaired fasting glucose (IFG). The DPP demon-
strated that lifestyle changes reduced the risk of
developing T2DM by 58% in adults with IFG or
IGT who were at high risk of developing diabetes
[124]. Metformin drug therapy did also reduce
the risk of developing T2DM, but was less effec-
tive than weight loss and increased physical
activity [124]. To prevent one case of diabetes in
3 years, only 6.9 persons would have to partici-
pate in the lifestyle intervention program,
whereas 13.9 would have to receive metformin
[124], which clearly showed that the lifestyle
intervention was almost twice as effective as drug
therapy. Examining the 10-year cost-­effectiveness
of the DPP interventions, lifestyle was found to
be cost-effective, and metformin was marginally
cost-saving compared to placebo [125].
Optimally, lifestyle changes should be intro-
duced in patients with IGT, who are early in the
course of developing neuropathy. The Impaired
Glucose Tolerance Causes Neuropathy Study
(IGTN) was a natural history study that gave gen-
eral dietary and physical activity advice (similar
to those in the DPP “lifestyle intervention”
group) to participants with IGT or IFG and mild
neuropathy. In the IGTN study, weight loss and/
or an increase in physical activity was associated
with slower progression of neuropathy based on
the intraepidermal nerve fiber density and with
the ability to re-grow epidermal nerve fibers
[110, 126]. More recent studies have underscored
the importance of intervention during the earliest
stages of impaired glucose regulation, and a life-
style intervention can be more effective than a
319
drug intervention in preventing conversion from
IGT to diabetes [124]. Furthermore, studies have
shown that an exercise intervention can prevent
the development of neuropathy in subjects who
have T2DM but do not have neuropathy [127–
129], may affect pain and symptom outcomes in
subjects with neuropathy, and is safe [130, 131].
3.4 Alpha-Lipoic Acid and DN
Alpha-lipoic acid (ALA) is biosynthesized in
small amounts from octanoic acid in the mito-
chondrion, where it is used as a cofactor for pyru-
vate dehydrogenase and α-ketoglutarate
dehydrogenase complexes [132, 133]. ALA is a
strong antioxidant that functions by removing
already formed free radicals [134]. ALA has a
disease modifying effect, and in several studies, it
has been shown to improve DN symptoms.
However, the evidence to support its use is not
strong because clinical trials with ALA have been
completed using a variety of study designs, routes
of administration, and sample sizes [135–138]. In
the multicenter, randomized, double-blind,
placebo-­controlled ALADIN III trial, there was a
small but significant improvement in the
Neuropathy Impairment Score (NIS) in ALA-­
treated patients, but no significant improvement
in the Total Symptom Score (TSS) [136]. In the
Deutsche Kardiale Autonome Neuropathie
(DEKAN) Study, there were small improvements
in the cardiac autonomic spectral analysis in
ALA treated patients [139]. In the SYDNEY2
trial, 181 diabetic patients were treated for 5
weeks after a 1-week placebo run-in period with
the following daily oral doses of ALA: 600 mg (n
= 45) (ALA600), 1200 mg (n = 47) (ALA1200),
and 1800 mg (n = 46) (ALA1800) or placebo (n
= 43) [137]. The change from the baseline TSS
was the primary outcome measure. The
Neuropathy Symptoms and Change (NSC) score
and the NIS were secondary end points. The
mean TSS decreased by 51% in ALA600, 48% in
ALA1200, and 52% in ALA1800 compared with
32% in the placebo group (P < 0.05 vs. placebo).
The NSC significantly improved in all three ALA
groups, and the NIS was numerically reduced.
320
Thus, oral treatment with ALA for 5 weeks
improved neuropathy measures in patients with
DN. The optimum risk-to-benefit ratio occurred
with 600 mg alpha-lipoic acid once daily [137].
Further support for ALA therapy is provided by a
meta-analysis of four trials (ALADIN I, ALADIN
III, SYDNEY, NATHAN II) comprising 1258
patients (ALA n = 716; placebo n = 542) [140].
In the Nathan I study [138], 460 diabetic patients
with mild-to-moderate diabetic sensory polyneu-
ropathy were assigned randomly to oral treat-
ment with 600 mg ALA once daily (n = 233) or
placebo (n = 227) for 4 years. A composite score
of the NIS-LL and seven neurophysiologic tests
was the primary end point. The NIS, NIS-LL,
nerve conduction, and quantitative sensory tests
(QSTs) were included as secondary outcome
measures. There was no significant difference
between treatment groups for the change in pri-
mary end point from baseline to 4 years (P =
0.105). When sub-scores of the composite score
were measured independently, the change from
baseline was significantly better for ALA com-
pared to placebo for: NIS (P = 0.028), NIS-LL (P
= 0.05), and the NIS-LL muscular weakness sub-­
score (P = 0.045). With ALA compared to pla-
cebo, more patients showed a clinically
meaningful improvement, and fewer showed pro-
gression in NIS (P = 0.013) and NIS-LL (P =
0.025). Part of the reason that the primary study
endpoint failed to detect a difference was that
nerve conduction and QST results did not signifi-
cantly worsen in the placebo group. Global
assessment of treatment tolerability did not differ
between the groups.
The ALA studies’ overall results indicate that
chronic treatment with ALA at an optimal dose of
at least 600 mg/day is safe and improves some
neuropathic deficits in patients with DN. However,
limitations in end point measures have precluded
a definitive conclusion about whether ALA
reverses or prevents DPN. In the analysis of the
NATHAN 1 trial, improvement and prevention of
progression of NIS-LL with ALA vs. placebo
after 4 years was predicted by lower BMI, higher
age, male sex, history of cardiovascular disease
(CVD), normal blood pressure, insulin treatment,
longer duration of diabetes and neuropathy, and
A. Hedayat et al.
higher neuropathy stage. A more recent study
used a combination therapy with superoxide dis-
mutase, ALA, acetyl L-carnitine, and vitamin
B12  in a prospective, double-blind, placebo-­
controlled study, randomized study in 85 patients
with T2DM over 12 months [141]. In this study,
the Michigan Neuropathy Screening Instrument
Questionnaire, vibration perception threshold,
sural nerve conduction velocity and amplitude,
pain, and quality of life questionnaires were mea-
sured endpoints and showed statistically signifi-
cant improvement after 12 months compared to
the placebo group. The cardiovascular autonomic
reflex tests and Michigan Neuropathy Screening
Instrument Examination did not significantly
change. However, similar results were obtained
with normalization of just B12 levels [142] sug-
gesting that the main active component in the
combination tablet was B12.
3.5 Benfotiamine and DN
Mitochondria take up and use thiamine pyro-
phosphate, the active form of thiamine. In dia-
betic tissues, only about 10% of a cell’s thiamine
is in the cytosol, available to transketolase, while
the remainder is inside the mitochondrion [118].
Thiamine plays a crucial role in the metabolism
of energy. Benfotiamine is a synthetic derivative
of thiamine. It elevates the levels of thiamine
diphosphate in the cell, a cofactor necessary for
the activation of transketolase, resulting in the
reduction of tissue levels of AGEs. The anti-AGE
effect of benfotiamine makes it effective for DN
treatment [143].
Benfotiamine is able to reduce the accumula-
tion of triosephosphates in diabetes. Excess trios-
ephosphates can be removed via the reductive
pentosephosphate pathway. However, this path-
way is impaired in diabetes by mild thiamine
deficiency. The activity and expression of the
thiamine-dependent enzyme, transketolase, are
consequently decreased in the pentosephosphate
pathway. Benfotiamine therapy in experimental
diabetes restores disposal of triosephosphates by
the reductive pentosephosphate pathway in
hyperglycemia. This prevents activation of mul-
Targeting the Mitochondrion in Diabetic Neuropathy
tiple biochemical pathways that cause injury in
diabetes: PKC, hexosamine, glycation and oxida-
tive stress pathways [144]. High-dose thiamine
also corrects dyslipidemia in experimental diabe-
tes, in part by normalizing cholesterol and tri-
glycerides. IGT is observed with thiamine
deficiency, and thus dietary thiamine may help
prevent T2DM.
Benfotiamine has been assessed as a clinical
therapy in a double-blind, placebo-controlled,
phase-III study [145]. A total of 165 patients with
DPN were randomized to one of three treatment
groups: benfotiamine 600 mg/day, benfotiamine
300 mg/day, or placebo. After 6 weeks of treat-
ment, the primary outcome parameter, the
Neuropathy Symptom Score, differed
­significantly between the treatment groups in the
per-­protocol but not in the intention to treat popu-
lation (p = 0.055). The Total Symptom Score
showed no significant differences after 6 weeks
of treatment. The improvement was more notice-
able at the higher benfotiamine dose and
increased with treatment duration. In the TSS,
“pain” showed the best response to treatment.
Treatment was well tolerated in all groups.
4 Conclusion
Many lines of evidence indicate that aberrant
mitochondrial structure and function contrib-
ute to cellular dysfunction, cell death, and dis-
ease pathology including diabetic neuropathy.
In general, disease may occur within several
levels of mitochondrial dysfunction. These
include impaired oxidative phosphorylation,
lipid metabolism, imbalanced mitochondrial
dynamics, fragmented mitochondrial networks,
impaired mitochondrial signaling, neuroin-
flammation, and mutations in the mitochon-
drial or nuclear genome. Exploration of some
of the mitochondrial directed mechanisms for
diabetic neuropathy reveals targets for treat-
ment of disease. If these targets are promising
in preclinical studies, they may be utilized in
human clinical trials. Dietary and exercise
interventions are not only effective but have
significant effects on mitochondrial function
321
and signaling, although it remains to be seen
whether these effects can be sustained. As with
all interventions, therapeutics targeting mito-
chondria will face problems with delivery, and
specificity challenges to ameliorate pathologi-
cal symptoms without inducing negative side
effects.
Acknowledgments Supported in part by the National
Institute of Diabetes and Digestive and Kidney Diseases,
National Institutes of Health 1R01DK107007-01A1,
Office of Research Development, Department of Veterans
Affairs (Biomedical and Laboratory Research Service and
Rehabilitation Research and Development,
101RX001030), Diabetes Action Research and Education
Foundation, University of Maryland Institute for Clinical
& Translational Research (ICTR), and the Baltimore
GRECC (JWR), 1K2RX001651 (LAZ).
References
1. Vincent AM, Olzmann JA, Brownlee M, Sivitz WI,
Russell JW.  Uncoupling proteins prevent glucose-­
induced neuronal oxidative stress and programmed
cell death. Diabetes. 2004;53:726–34.
2. Chandrasekaran K, Anjaneyulu M, Choi J, Kumar
P, Salimian M, Ho CY, et al. Role of mitochondria
in diabetic peripheral neuropathy: Influencing the
NAD(+)-dependent SIRT1-PGC-1alpha-TFAM
pathway. Int Rev Neurobiol. 2019;145:177–209.
3. Chandrasekaran K, Anjaneyulu M, Inoue T, Choi J,
Sagi AR, Chen C, et al. Mitochondrial transcription
factor A regulation of mitochondrial degeneration
in experimental diabetic neuropathy. Am J Physiol
Endocrinol Metab. 2015;309(2):E132–E41.
4. Choi J, Chandrasekaran K, Inoue T, Muragundla A,
Russell JW.  PGC-1α regulation of mitochondrial
degeneration in experimental diabetic neuropathy.
Neurobiol Dis. 2014;64:118–30.
5. Rumora AE, Savelieff MG, Sakowski SA, Feldman
EL. Disorders of mitochondrial dynamics in periph-
eral neuropathy: clues from hereditary neuropathy
and diabetes. Int Rev Neurobiol. 2019;145:127–76.
6. Fernyhough P.  Mitochondrial dysfunction in dia-
betic neuropathy: a series of unfortunate metabolic
events. Curr Diab Rep. 2015;15(11):89–0671.
7. Russell JW, Berent-Spillson A, Vincent AM,
Freimann CL, Sullivan KA, Feldman EL. Oxidative
injury and neuropathy in diabetes and impaired glu-
cose tolerance. Neurobiol Dis. 2008;30(3):420–9.
8. Cowell RM, Kennedy E, Berent-Spillson A,
Sundkvist G, Russell JW.  Regulation of mitochon-
drial biogenesis in Schwann cells: implications for
diabetic neuropathy. Ann Neurol. 2005;58:S29.
9. Russell JW, Golovoy D, Vincent AM, Mahendru P,
Olzmann JA, Mentzer A, et al. High glucose-induced
322
oxidative stress and mitochondrial dysfunction in
neurons. FASEB. 2002;16(13):1738–48.
10. Vincent AM, Calabek B, Roberts L, Feldman
EL.  Biology of diabetic neuropathy. Handb Clin
Neurol. 2013;115:591–606.
11. Puigserver P. Tissue-specific regulation of metabolic
pathways through the transcriptional coactivator
PGC1-alpha. Int J Obes. 2005;29(1):5–9.
12. Puigserver P, Spiegelman BM.  Peroxisome
proliferator-­activated receptor-gamma coactivator 1
alpha (PGC-1 alpha): transcriptional coactivator and
metabolic regulator. Endocr Rev. 2003;24(1):78–90.
13. St-Pierre J, Drori S, Uldry M, Silvaggi JM, Rhee J,
Jager S, et al. Suppression of reactive oxygen species
and neurodegeneration by the PGC-1 transcriptional
coactivators. Cell. 2006;127(2):397–408.
14. Johri A, Calingasan NY, Hennessey TM, Sharma
A, Yang L, Wille E, et  al. Pharmacologic activa-
tion of mitochondrial biogenesis exerts wide-
spread beneficial effects in a transgenic mouse
model of Huntington’s disease. Hum Mol Genet.
2012;21(5):1124–37.
15. Scarpulla RC, Vega RB, Kelly DP.  Transcriptional
integration of mitochondrial biogenesis. Trends
Endocrinol Metab. 2012;23(9):459–66.
16. Wu Z, Puigserver P, Andersson U, Zhang C,
Adelmant G, Mootha V, et  al. Mechanisms con-
trolling mitochondrial biogenesis and respiration
through the thermogenic coactivator PGC-1. Cell.
1999;98(1):115–24.
17. Esterbauer H, Oberkofler H, Krempler F, Patsch
W. Human peroxisome proliferator activated recep-
tor gamma coactivator 1 (PPARGC1) gene: cDNA
sequence, genomic organization, chromosomal
localization, and tissue expression. Genomics.
1999;62(1):98–102.
18. Andrulionyte L, Kuulasmaa T, Chiasson JL, Laakso
M.  Single nucleotide polymorphisms of the per-
oxisome proliferator-activated receptor-alpha gene
(PPARA) influence the conversion from impaired
glucose tolerance to type 2 diabetes: the STOP-­
NIDDM trial. Diabetes. 2007;56(4):1181–6.
19. Mootha VK, Lindgren CM, Eriksson KF,
Subramanian A, Sihag S, Lehar J, et al. PGC-1alpha-­
responsive genes involved in oxidative phosphory-
lation are coordinately downregulated in human
diabetes. Nat Genet. 2003;34(3):267–73.
20. Roy Chowdhury SK, Smith DR, Saleh A,
Schapansky J, Marquez A, Gomes S, et al. Impaired
adenosine monophosphate-activated protein kinase
signalling in dorsal root ganglia neurons is linked to
mitochondrial dysfunction and peripheral neuropa-
thy in diabetes. Brain. 2012;135(6):1751–66.
21. Chowdhury SK, Smith DR, Fernyhough P. The role
of aberrant mitochondrial bioenergetics in diabetic
neuropathy. Neurobiol Dis. 2013;51:56–65.
22. Canto C, Auwerx J.  Calorie restriction: is
AMPK a key sensor and effector? Physiology.
2011;26(4):214–24.
23. Fernyhough P, Jonathan M. Mechanisms of disease:
mitochondrial dysfunction in sensory neuropathy
A. Hedayat et al.
and other complications in diabetes. Handb Clin
Neurol. 2014;126:353–77.
24. Callaghan BC, Cheng HT, Stables CL, Smith AL,
Feldman EL.  Diabetic neuropathy: clinical mani-
festations and current treatments. Lancet Neurol.
2012;11(6):521–34.
25. Vincent AM, Callaghan BC, Smith AL, Feldman
EL.  Diabetic neuropathy: cellular mecha-
nisms as therapeutic targets. Nat Rev Neurol.
2011;7(10):573–83.
26. Vincent AM, Hinder LM, Pop-Busui R, Feldman
EL.  Hyperlipidemia: a new therapeutic target
for diabetic neuropathy. J Peripher Nerv Syst.
2009;14(4):257–67.
27. Vincent AM, Hayes JM, McLean LL, Vivekanandan-­
Giri A, Pennathur S, Feldman EL.  Dyslipidemia-­
induced neuropathy in mice: the role of oxLDL/
LOX-1. Diabetes. 2009;58(10):2376–85.
28. Heilbronn LK, Gregersen S, Shirkhedkar D, Hu D,
Campbell LV.  Impaired fat oxidation after a single
high-fat meal in insulin-sensitive nondiabetic indi-
viduals with a family history of type 2 diabetes.
Diabetes. 2007;56(8):2046–53.
29. Kleiner S, Mepani RJ, Laznik D, Ye L, Jurczak MJ,
Jornayvaz FR, et  al. Development of insulin resis-
tance in mice lacking PGC-1alpha in adipose tissues.
Proc Natl Acad Sci U S A. 2012;109(24):9635–40.
30. Mootha VK, Handschin C, Arlow D, Xie X, St Pierre
J, Sihag S, et al. Erralpha and Gabpa/b specify PGC-­
1alpha-­dependent oxidative phosphorylation gene
expression that is altered in diabetic muscle. Proc
Natl Acad Sci U S A. 2004;101(17):6570–5.
31. Patti ME, Butte AJ, Crunkhorn S, Cusi K, Berria R,
Kashyap S, et al. Coordinated reduction of genes of
oxidative metabolism in humans with insulin resis-
tance and diabetes: Potential role of PGC1 and NRF1.
Proc Natl Acad Sci U S A. 2003;100(14):8466–71.
32. Leone TC, Lehman JJ, Finck BN, Schaeffer PJ,
Wende AR, Boudina S, et al. PGC-1a deficient mice
exhibit multi-system energy metabolic derange-
ments: muscle dysfunction, abnormal weight con-
trol, and hepatic steatosis. PLoS Biol. 2005;3:e101.
33. Farge G, Mehmedovic M, Baclayon M, van den
Wildenberg SM, Roos WH, Gustafsson CM, et  al.
In vitro-reconstituted nucleoids can block mitochon-
drial DNA replication and transcription. Cell Rep.
2014;8(1):66–74.
34. Alam TI, Kanki T, Muta T, Ukaji K, Abe Y, Nakayama
H, et  al. Human mitochondrial DNA is packaged
with TFAM. Nucleic Acids Res. 2003;31(6):1640–5.
35. Ohgaki K, Kanki T, Fukuoh A, Kurisaki H, Aoki
Y, Ikeuchi M, et  al. The C-terminal tail of mito-
chondrial transcription factor a markedly strength-
ens its general binding to DNA.  J Biochem.
2007;141(2):201–11.
36. Wang YE, Marinov GK, Wold BJ, Chan
DC.  Genome-wide analysis reveals coating of
the mitochondrial genome by TFAM.  PLoS One.
2013;8(8):e74513.
37. Morimoto N, Miyazaki K, Kurata T, Ikeda Y,
Matsuura T, Kang D, et al. Effect of mitochondrial
Targeting the Mitochondrion in Diabetic Neuropathy
transcription factor a overexpression on motor neu-
rons in amyotrophic lateral sclerosis model mice. J
Neurosci Res. 2012;90(6):1200–8.
38. Murphy MP.  How mitochondria produce reactive
oxygen species. Biochem J. 2009;417(1):1–13.
39. Chandrasekaran K, Salimian M, Konduru SR,
Choi J, Kumar P, Long A, et  al. Overexpression
of Sirtuin 1 protein in neurons prevents and
reverses experimental diabetic neuropathy. Brain.
2019;142(12):3737–52.
40. Chandrasekaran K, Chen C, Sagi AR, Russell JW. A
nicotinamide adenine nucleotide (NAD+) precur-
sor is a potential therapy for diabetic neuropathy. J
Neuromusc Dis. 2016;3(1):S86.
41. Chandrasekaran K, Najimi N, Kumar P, Russell
JW.  Nicotinamide riboside is a potential therapy
for diabetic neuropathy. Neurodiab; Bucharest,
Romania. 2016.
42. Trammell SA, Weidemann BJ, Chadda A, Yorek MS,
Holmes A, Coppey LJ, et al. Nicotinamide riboside
opposes type 2 diabetes and neuropathy in mice.
Sci Rep. 2016;6:26933. https://doi.org/10.1038/
srep26933.:26933.
43. Feldman EL, Callaghan BC, Pop-Busui R, Zochodne
DW, Wright DE, Bennett DL, et al. Diabetic neurop-
athy. Nat Rev Dis Primers. 2019;5(1):41.
44. Sena LA, Chandel NS.  Physiological roles of
mitochondrial reactive oxygen species. Mol Cell.
2012;48(2):158–67.
45. Milenkovic D, Blaza JN, Larsson NG, Hirst J. The
enigma of the respiratory chain supercomplex. Cell
Metab. 2017;25(4):765–76.
46. Topf U, Suppanz I, Samluk L, Wrobel L, Boser A,
Sakowska P, et al. Quantitative proteomics identifies
redox switches for global translation modulation by
mitochondrially produced reactive oxygen species.
Nat Commun. 2018;9(1):324.
47. Pfanner N, Warscheid B, Wiedemann
N.  Mitochondrial proteins: from biogenesis to
functional networks. Nat Rev Mol Cell Biol.
2019;20(5):267–84.
48. Cowell RM, Russell JW.  Peripheral neuropathy
and the Schwann cell. In: Kettenmann H, Ransom
BR, editors. Neuroglia. 2nd ed. Oxford: Oxford
University Press; 2004. p. 573–85.
49. Vincent AM, Russell JW, Sullivan KA, Backus C,
Hayes JM, McLean LL, et al. SOD2 protects neurons
from injury in cell culture and animal models of dia-
betic neuropathy. Exp Neurol. 2007;208(2):216–27.
50. Mendez MM, Folgado J, Tormo C, Artero A, Ascaso
M, Martinez-Hervas S, et al. Altered glutathione sys-
tem is associated with the presence of distal sym-
metric peripheral polyneuropathy in type 2 diabetic
subjects. J Diabetes Complicat. 2015;29(7):923–7.
51. Kumar A, Mittal R. Nrf2: a potential therapeutic tar-
get for diabetic neuropathy. Inflammopharmacology.
2017;25(4):393–402.
52. Jimenez-Osorio AS, Gonzalez-Reyes S, Pedraza-­
Chaverri J. Natural Nrf2 activators in diabetes. Clin
Chim Acta. 2015;448:182–92.
323
53. Agca CA, Tuzcu M, Hayirli A, Sahin K.  Taurine
ameliorates neuropathy via regulating NF-kappaB
and Nrf2/HO-1 signaling cascades in diabetic rats.
Food Chem Toxicol. 2014;71:116–21.
54. Sandireddy R, Yerra VG, Komirishetti P, Areti A,
Kumar A. Fisetin imparts neuroprotection in experi-
mental diabetic neuropathy by modulating Nrf2
and NF-kappaB pathways. Cell Mol Neurobiol.
2016;36(6):883–92.
55. Maiese K.  Novel nervous and multi-system
regenerative therapeutic strategies for diabe-
tes mellitus with mTOR.  Neural Regen Res.
2016;11(3):372–85.
56. Mazucanti CH, Cabral-Costa JV, Vasconcelos AR,
Andreotti DZ, Scavone C, Kawamoto EM. Longevity
pathways (mTOR, SIRT, Insulin/IGF-1) as key mod-
ulatory targets on aging and neurodegeneration. Curr
Top Med Chem. 2015;15(21):2116–38.
57. Cowell RM, Russell JW. Nitrosative injury and anti-
oxidant therapy in the management of diabetic neu-
ropathy. J Investig Med. 2004;52(1):33–44.
58. Stavniichuk R, Shevalye H, Lupachyk S, Obrosov
A, Groves JT, Obrosova IG, et  al. Peroxynitrite
and protein nitration in the pathogenesis of diabetic
peripheral neuropathy. Diabetes Metab Res Rev.
2014;30(8):669–78.
59. Muller-Stich BP, Billeter AT, Fleming T, Fischer
L, Buchler MW, Nawroth PP. Nitrosative stress but
not glycemic parameters correlate with improved
neuropathy in nonseverely obese diabetic patients
after Roux-Y gastric bypass. Surg Obes Relat Dis.
2015;11(4):847–54.
60. Nigro C, Leone A, Fiory F, Prevenzano I, Nicolo A,
Mirra P, et al. Dicarbonyl stress at the crossroads of
healthy and unhealthy aging. Cell. 2019;8(7):749.
61. Jiating L, Buyun J, Yinchang Z.  Role of metfor-
min on osteoblast differentiation in type 2 diabetes.
Biomed Res Int. 2019;2019:9203934.
62. Park S, Kang HJ, Jeon JH, Kim MJ, Lee IK. Recent
advances in the pathogenesis of microvascu-
lar complications in diabetes. Arch Pharm Res.
2019;42(3):252–62.
63. Toth C, Rong LL, Yang C, Martinez J, Song F, Ramji
N, et al. Receptor for advanced glycation end prod-
ucts (RAGEs) and experimental diabetic neuropathy.
Diabetes. 2008;57(4):1002–17.
64. Goldin A, Beckman JA, Schmidt AM, Creager
MA. Advanced glycation end products: sparking the
development of diabetic vascular injury. Circulation.
2006;114(6):597–605.
65. Vincent AM, Perrone L, Sullivan KA, Backus C,
Sastry AM, Lastoskie C, et al. Receptor for advanced
glycation end products activation injures primary
sensory neurons via oxidative stress. Endocrinology.
2007;148(2):548–58.
66. Kanwar YS, Sun L, Xie P, Liu FY, Chen S.  A
glimpse of various pathogenetic mechanisms
of diabetic nephropathy. Annu Rev Pathol.
2011;6:395–423. https://doi.org/10.1146/annurev.
pathol.4.110807.092150.:395-­423.
324
67. Haslbeck KM, Schleicher E, Bierhaus A, Nawroth
P, Haslbeck M, Neundorfer B, et  al. The AGE/
RAGE/NF-(kappa)B pathway may contribute to the
pathogenesis of polyneuropathy in impaired glucose
tolerance (IGT). Exp Clin Endocrinol Diabetes.
2005;113(5):288–91.
68. Anjaneyulu M, Berent-Spillson A, Inoue T, Choi J,
Cherian K, Russell JW. Transforming growth factor-­
beta induces cellular injury in experimental diabetic
neuropathy. Exp Neurol. 2008;211(2):469–79.
69. Toth C, Rong LL, Yang C, Martinez J, Song F, Ramji
N, et al. Receptor for advanced glycation end prod-
ucts (RAGEs) and experimental diabetic neuropathy.
Diabetes. 2007;57:1002–17.
70. Ray R, Juranek JK, Rai V.  RAGE axis in neuroin-
flammation, neurodegeneration and its emerging
role in the pathogenesis of amyotrophic lateral scle-
rosis. Neurosci Biobehav Rev. 2016;62:48–55.
71. Morgenstern J, Fleming T, Schumacher D, Eckstein
V, Freichel M, Herzig S, et al. Loss of glyoxalase 1
induces compensatory mechanism to achieve dicar-
bonyl detoxification in mammalian schwann cells. J
Biol Chem. 2017;292(8):3224–38.
72. Jack MM, Ryals JM, Wright DE. Characterisation of
glyoxalase I in a streptozocin-induced mouse model
of diabetes with painful and insensate neuropathy.
Diabetologia. 2011;54(8):2174–82.
73. Alderson NL, Chachich ME, Youssef NN, Beattie
RJ, Nachtigal M, Thorpe SR, et al. The AGE inhibi-
tor pyridoxamine inhibits lipemia and development
of renal and vascular disease in Zucker obese rats.
Kidney Int. 2003;63(6):2123–33.
74. Archer AE, Von Schulze AT, Geiger PC.  Exercise,
heat shock proteins and insulin resistance. Philos
Trans R Soc Lond Ser B Biol Sci. 2018;373:1738.
75. Mayer MP, Bukau B.  Hsp70 chaperones: cellular
functions and molecular mechanism. Cell Mol Life
Sci. 2005;62(6):670–84.
76. Peterson LB, Blagg BS. To fold or not to fold: modu-
lation and consequences of Hsp90 inhibition. Future
Med Chem. 2009;1(2):267–83.
77. Henstridge DC, Whitham M, Febbraio
MA.  Chaperoning to the metabolic party: the
emerging therapeutic role of heat-shock pro-
teins in obesity and type 2 diabetes. Mol Metab.
2014;3(8):781–93.
78. Farmer KL, Li C, Dobrowsky RT. Diabetic periph-
eral neuropathy: should a chaperone accom-
pany our therapeutic approach? Pharmacol Rev.
2012;64(4):880–900.
79. Dobrowsky RT.  Targeting the diabetic chaperome
to improve peripheral neuropathy. Curr Diab Rep.
2016;16(8):71.
80. Murphy M, Hickey F, Godson C.  IHG-1 amplifies
TGF-beta1 signalling and mitochondrial biogenesis
and is increased in diabetic kidney disease. Curr
Opin Nephrol Hypertens. 2013;22(1):77–84.
81. Hussain G, Rizvi SA, Singhal S, Zubair M, Ahmad
J. Serum levels of TGF-beta1 in patients of diabetic
peripheral neuropathy and its correlation with nerve
A. Hedayat et al.
conduction velocity in type 2 diabetes mellitus.
Diabetes Metab Syndr. 2016;10(1):135–9.
82. Tan CK, Chong HC, Tan EH, Tan NS.  Getting
‘Smad’ about obesity and diabetes. Nutr Diabetes.
2012;2:e29.
83. Berent-Spillson A, Robinson A, Golovoy D, Slusher
B, Rojas C, Russell JW. Protection against glucose-­
induced neuronal death by NAAG and GCP II
inhibition is regulated by mGluR3. J Neurochem.
2004;89:90–9.
84. Choi DW.  Excitotoxic cell-death. J Neurobiol.
1992;23(9):1261–76.
85. Miller KE, Richards BA, Kriebel RM. Glutamine-,
glutamine synthetase-, glutamate dehydrogenase-
and pyruvate carboxylase-immunoreactivities in the
rat dorsal root ganglion and peripheral nerve. Brain
Res. 2002;945(2):202–11.
86. Cangro CB, Sweetnam PM, Wrathall JR, Haser WB,
Curthoys NP, Neale JH. Localization of elevated glu-
taminase immunoreactivity in small DRG neurons.
Brain Res. 1985;336(1):158–61.
87. Carozzi VA, Canta A, Oggioni N, Ceresa C,
Marmiroli P, Konvalinka J, et  al. Expression and
distribution of ‘high affinity’ glutamate trans-
porters GLT1, GLAST, EAAC1 and of GCPII
in the rat peripheral nervous system. J Anat.
2008;213(5):539–46.
88. Ohara PT, Vit JP, Bhargava A, Romero M,
Sundberg C, Charles AC, et  al. Gliopathic pain:
when satellite glial cells go bad. Neuroscientist.
2009;15(5):450–63.
89. Anjaneyulu M, Berent-Spillson A, Russell
JW.  Metabotropic glutamate receptors (mGluRs)
and diabetic neuropathy. Curr Drug Targets.
2008;9(1):85–93.
90. Berent Spillson A, Russell JW. Metabotropic gluta-
mate receptor regulation of neuronal cell death. Exp
Neurol. 2003;184:S97–S105.
91. Berent-Spillson A, Russell JW.  Metabotropic
glutamate receptor 3 protects neurons from
glucose-­induced oxidative injury by increasing intra-
cellular glutathione concentration. J Neurochem.
2007;101(2):342–54.
92. Howarth C, Gleeson P, Attwell D.  Updated energy
budgets for neural computation in the neocor-
tex and cerebellum. J Cereb Blood Flow Metab.
2012;32(7):1222–32.
93. Rangaraju V, Calloway N, Ryan TA. Activity-driven
local ATP synthesis is required for synaptic function.
Cell. 2014;156(4):825–35.
94. Misgeld T, Schwarz TL.  Mitostasis in neurons:
maintaining mitochondria in an extended cellular
architecture. Neuron. 2017;96(3):651–66.
95. Moehlman AT, Youle RJ. Mitochondrial quality con-
trol and restraining innate immunity. Annu Rev Cell
Dev Biol. 2020;36:265–89.
96. Chowdhury SK, Zherebitskaya E, Smith DR,
Akude E, Chattopadhyay S, Jolivalt CG, et  al.
Mitochondrial respiratory chain dysfunction in dor-
sal root ganglia of streptozotocin-induced diabetic
Targeting the Mitochondrion in Diabetic Neuropathy
rats and its correction by insulin treatment. Diabetes.
2010;59(4):1082–91.
97. Zuchner S, Mersiyanova IV, Muglia M, Bissar-­
Tadmouri N, Rochelle J, Dadali EL, et al. Mutations
in the mitochondrial GTPase mitofusin 2 cause
charcot-marie-tooth neuropathy type 2A. Nat Genet.
2004;36(5):449–51.
98. Kijima K, Numakura C, Izumino H, Umetsu K,
Nezu A, Shiiki T, et al. Mitochondrial GTPase mito-
fusin 2 mutation in Charcot-Marie-Tooth neuropathy
type 2A. Hum Genet. 2005;116(1-2):23–7.
99. Alexander C, Votruba M, Pesch UE, Thiselton
DL, Mayer S, Moore A, et  al. OPA1, encoding a
dynamin-­related GTPase, is mutated in autosomal
dominant optic atrophy linked to chromosome 3q28.
Nat Genet. 2000;26(2):211–5.
100. Delettre C, Lenaers G, Griffoin JM, Gigarel N,
Lorenzo C, Belenguer P, et al. Nuclear gene OPA1,
encoding a mitochondrial dynamin-related protein,
is mutated in dominant optic atrophy. Nat Genet.
2000;26(2):207–10.
101. Yu-Wai-Man P, Griffiths PG, Gorman GS, Lourenco
CM, Wright AF, Auer-Grumbach M, et  al. Multi-­
system neurological disease is common in patients
with OPA1 mutations. Brain. 2010;133(3):771–86.
102. Dyall SD, Brown MT, Johnson PJ.  Ancient inva-
sions: from endosymbionts to organelles. Science.
2004;304(5668):253–7.
103. Iyer SS, He Q, Janczy JR, Elliott EI, Zhong Z, Olivier
AK, et  al. Mitochondrial cardiolipin is required
for Nlrp3 inflammasome activation. Immunity.
2013;39(2):311–23.
104. West AP, Shadel GS. Mitochondrial DNA in innate
immune responses and inflammatory pathology. Nat
Rev Immunol. 2017;17(6):363–75.
105. Starobova H, Monteleone M, Adolphe C, Batoon
L, Sandrock CJ, Tay B, et  al. Vincristine-induced
peripheral neuropathy is driven by canonical
NLRP3 activation and IL-1beta release. J Exp Med.
2021;218(5):e20201452.
106. Savelieff MG, Feldman EL.  Immune-mediated
vincristine-­induced neuropathy: unlocking thera-
pies. J Exp Med. 2021;218(5):e20210286.
107. Jia M, Wu C, Gao F, Xiang H, Sun N, Peng P, et al.
Activation of NLRP3 inflammasome in peripheral
nerve contributes to paclitaxel-induced neuropathic
pain. Mol Pain. 2017;13:1744806917719804.
108. Kumar A, Negi G, Sharma SS.  Suppression
of NF-kappaB and NF-kappaB regulated oxi-
dative stress and neuroinflammation by BAY
11-7082 (IkappaB phosphorylation inhibitor) in
experimental diabetic neuropathy. Biochimie.
2012;94(5):1158–65.
109. Zilliox LA, Russell JW. Physical activity and dietary
interventions in diabetic neuropathy: a systematic
review. Clin Auton Res. 2019;4:443–55.
110. Smith AG, Russell JW, Feldman EL, Goldstein
J, Peltier A, Smith S, et  al. Lifestyle interven-
tion for prediabetic neuropathy. Diabetes Care.
2006;29:1294–9.
325
111. Russell JW, Kaminsky AJ. Oxidative injury in dia-
betic neuropathy. In: Opara E, editor. Nutrition and
diabetes: pathophysiology and management. Boca
Raton: Taylor & Francis; 2005. p. 381–97.
112. Kobilo T, Guerrieri D, Zhang Y, Collica SC, Becker
KG, van Praag H. AMPK agonist AICAR improves
cognition and motor coordination in young and aged
mice. Learn Mem. 2014;21(2):119–26.
113. Chandrasekaran K, Choi J, Arvas MI, Salimian
M, Singh S, Xu S, et  al. Nicotinamide mono-
nucleotide administration prevents experimen-
tal diabetes-induced cognitive impairment and
loss of hippocampal neurons. Int J Mol Sci.
2020;21(11):3756.
114. Hinder LM, O’Brien PD, Hayes JM, Backus C,
Solway AP, Sims-Robinson C, et al. Dietary rever-
sal of neuropathy in a murine model of prediabe-
tes and metabolic syndrome. Dis Model Mech.
2017;10(6):717–25.
115. Cooper MA, Menta BW, Perez-Sanchez C, Jack
MM, Khan ZW, Ryals JM, et  al. A ketogenic diet
reduces metabolic syndrome-induced allodynia
and promotes peripheral nerve growth in mice. Exp
Neurol. 2018;306:149–57.
116. Coppey L, Davidson E, Shevalye H, Torres ME,
Yorek MA.  Effect of dietary oils on peripheral
neuropathy-related endpoints in dietary obese rats.
Diabetes Metab Syndr Obes. 2018;11:117–27.
117. Lewis EJH, Perkins BA, Lovblom LE, Bazinet RP,
Wolever TMS, Bril V. Effect of omega-3 supplemen-
tation on neuropathy in type 1 diabetes: a 12-month
pilot trial. Neurology. 2017;88(24):2294–301.
118. Teodoro JS, Gomes AP, Varela AT, Duarte FV, Rolo
AP, Palmeira CM. Uncovering the beginning of dia-
betes: the cellular redox status and oxidative stress as
starting players in hyperglycemic damage. Mol Cell
Biochem. 2013;376(1-2):103–10.
119. Ang L, Jaiswal M, Martin C, Pop-Busui R. Glucose
control and diabetic neuropathy: lessons from
recent large clinical trials. Curr Diab Rep.
2014;14(9):528.
120. UK Prospective Diabetes Study (UKPDS). Intensive
blood-glucose control with sulphonylureas or insu-
lin compared with conventional treatment and risk
of complications in patients with type 2 diabetes
(UKPDS 33). Lancet. 1998;352(9131):837–53.
121. Duckworth W, Abraira C, Moritz T, Reda D,
Emanuele N, Reaven PD, et al. Glucose control and
vascular complications in veterans with type 2 dia-
betes. N Engl J Med. 2009;360(2):129–39.
122. Azad N, Emanuele NV, Abraira C, Henderson WG,
Colwell J, Levin SR, et al. The effects of intensive
glycemic control on neuropathy in the VA coopera-
tive study on type II diabetes mellitus (VA CSDM). J
Diabetes Complicat. 1999;13:307–13.
123. Albers JW, Herman WH, Pop-Busui R, Feldman EL,
Martin CL, Cleary PA, et al. Effect of prior intensive
insulin treatment during the Diabetes Control and
Complications Trial (DCCT) on peripheral neuropa-
thy in type 1 diabetes during the Epidemiology of
326
Diabetes Interventions and Complications (EDIC)
Study. Diabetes Care. 2010;33(5):1090–6.
124. Knowler WC, Barrett-Connor E, Fowler SE,
Hamman RF, Lachin JM, Walker EA, et  al.
Reduction in the incidence of type 2 diabetes with
lifestyle intervention or metformin. N Engl J Med.
2002;346:393–403.
125. Diabetes Prevention Program Research Group. The
10-year cost-effectiveness of lifestyle intervention
or metformin for diabetes prevention: an intent-to-­
treat analysis of the DPP/DPPOS.  Diabetes Care.
2012;35(4):723–30.
126. Singleton JR, Smith AG, Russell JW, Feldman
EL.  Microvascular complications of impaired glu-
cose tolerance. Diabetes. 2003;52:2867–76.
127. Balducci S, Iacobellis G, Parisi L, Di Biase N,
Calandriello E, Leonetti F, et al. Exercise training can
modify the natural history of diabetic peripheral neu-
ropathy. J Diabetes Complicat. 2006;20(4):216–23.
128. Singleton JR, Marcus RL, Jackson JE, Lessard K,
Graham TE, Smith AG.  Exercise increases cutane-
ous nerve density in diabetic patients without neu-
ropathy. Ann Clin Transl Neurol. 2014;1(10):844–9.
129. Singleton JR, Marcus RL, Lessard MK, Jackson JE,
Smith AG.  Supervised exercise improves cutane-
ous reinnervation capacity in metabolic syndrome
patients. Ann Neurol. 2015;77(1):146–53.
130. Kluding PM, Pasnoor M, Singh R, D'Silva LJ, Yoo
M, Billinger SA, et al. Safety of aerobic exercise in
people with diabetic peripheral neuropathy: single-­
group clinical trial. Phys Ther. 2015;95(2):223–34.
131. Kluding PM, Pasnoor M, Singh R, Jernigan S, Farmer
K, Rucker J, et al. The effect of exercise on neuro-
pathic symptoms, nerve function, and cutaneous
innervation in people with diabetic peripheral neu-
ropathy. J Diabetes Complicat. 2012;26(5):424–9.
132. Ikuta N, Chikamoto K, Asano Y, Yasui Y, Yokokawa
H, Terao K, et al. Time course effect of r-alpha-lipoic
acid on cellular metabolomics in cultured hepatoma
cells. J Med Food. 2017;20(3):211–22.
133. Karalis DT, Karalis T, Karalis S, Kleisiari AS,
Malakoudi F, Maimari KEV.  The effect of alpha-­
lipoic acid on diabetic peripheral neuropathy and the
upcoming depressive disorders of type II diabetics.
Cureus. 2021;13(1):e12773.
134. Papanas N, Ziegler D. Efficacy of alpha-lipoic acid
in diabetic neuropathy. Expert Opin Pharmacother.
2014;15(18):2721–31.
135. Reljanovic M, Reichel G, Rett K, Lobisch M,
Schuette K, Moller W, et  al. Treatment of diabetic
polyneuropathy with the antioxidant thioctic acid
(alpha-lipoic acid): a two year multicenter random-
ized double-blind placebo-controlled trial (ALADIN
A. Hedayat et al.
II). Alpha Lipoic Acid in Diabetic Neuropathy. Free
Radic Res. 1999;31(3):171–9.
136. Ziegler D, Hanefeld M, Ruhnau KJ, Hasche H,
Lobisch M, Schutte K, et al. Treatment of symptom-
atic diabetic polyneuropathy with the antioxidant
alpha-lipoic acid: a 7-month multicenter random-
ized controlled trial (ALADIN III Study). ALADIN
III Study Group. Alpha-Lipoic Acid in Diabetic
Neuropathy. Diabetes Care. 1999;22:1296–301.
137. Ziegler D, Ametov A, Barinov A, Dyck PJ, Gurieva
I, Low PA, et  al. Oral treatment with alpha-lipoic
acid improves symptomatic diabetic polyneu-
ropathy: the SYDNEY 2 trial. Diabetes Care.
2006;29(11):2365–70.
138. Ziegler D, Low PA, Litchy WJ, Boulton AJ, Vinik
AI, Freeman R, et al. Efficacy and safety of antioxi-
dant treatment with alpha-lipoic acid over 4 years
in diabetic polyneuropathy: the NATHAN 1 trial.
Diabetes Care. 2011;34(9):2054–60.
139. Ziegler D, Schatz H, Conrad F, Gries FA, Ulrich H,
Reichel G. Effects of treatment with the antioxidant
alpha-lipoic acid on cardiac autonomic neuropathy
in NIDDM patients. A 4-month randomized con-
trolled multicenter trial (DEKAN Study). Deutsche
Kardiale Autonome Neuropathie. Diabetes Care.
1997;20(3):369–73.
140. Ziegler D, Nowak H, Kempler P, Vargha P, Low
PA. Treatment of symptomatic diabetic polyneurop-
athy with the antioxidant alpha-lipoic acid: a meta-­
analysis. Diabet Med. 2004;21(2):114–21.
141. Didangelos T, Karlafti E, Kotzakioulafi E,
Kontoninas Z, Margaritidis C, Giannoulaki P, et al.
Efficacy and safety of the combination of superoxide
dismutase, alpha lipoic acid, vitamin B12, and carni-
tine for 12 months in patients with diabetic neuropa-
thy. Nutrients. 2020;12(11):3254.
142. Didangelos T, Karlafti E, Kotzakioulafi E, Margariti
E, Giannoulaki P, Batanis G, et  al. Vitamin B12
supplementation in diabetic neuropathy: a 1-year,
randomized, double-blind, placebo-controlled trial.
Nutrients. 2021;13(2):395.
143. Balakumar P, Rohilla A, Krishan P, Solairaj P,
Thangathirupathi A.  The multifaceted therapeu-
tic potential of benfotiamine. Pharmacol Res.
2010;61(6):482–8.
144. Thornalley PJ. The potential role of thiamine (vita-
min B1) in diabetic complications. Curr Diabetes
Rev. 2005;1(3):287–98.
145. Stracke H, Gaus W, Achenbach U, Federlin K,
Bretzel RG. Benfotiamine in diabetic polyneuropa-
thy (BENDIP): results of a randomised, double
blind, placebo-controlled clinical study. Exp Clin
Endocrinol Diabetes. 2008;116(10):600–5.
 Diabetic Sensory Neurons, Dorsal
Root Ganglia, and Neuropathy

Aparna Areti and Douglas W. Zochodne

1 Introduction to injury, risk of falls, intractable neuropathic

Worldwide, there are few more compelling clini-
cal problems than burgeoning diabetes mellitus
(DM) and its irreversible complications, particu-
larly neurological. Diabetic polyneuropathy
(DPN) is not only the most common DM compli-
cation, but it is also the most common form of
damage to the peripheral nervous system and its
prevalence dwarfs that of many other neurologi-
cal disorders. Over 6% of Canadians have DM,
and the WHO estimates that 366 million people
worldwide will develop it within 10 years, with
diabetic polyneuropathy (DPN) in over 50% [1,
2]. It is a progressive irreversible neurological
problem that begins with foot numbness and
pain. Investigators of this unfortunate disorder
recognize the urgency of finding new solutions,
none successful to date, to prevent DPN’s major
clinical burdens: loss of sensation, insensitivity
pain, and foot ulceration leading to amputation
[3–10]. From a neurosciences perspective, there
is also exquisite neurobiology, not well under-
stood, that dictates how this condition, a unique
form of sensory predominant axon neurodegen-
eration, develops. DPN speaks to the singular
behavior of sensory axon terminals in the epider-
mis that possess a dynamic plasticity required to
maintain their investment of skin and its turnover
of epidermal keratinocytes [11]. In DPN, this
plasticity is abnormal and sensory axons change
their phenotype, retract, and disappear. Human
DPN is described as “stocking and glove”
because the distal terminals of sensory axons that
innervate the skin of the toes and fingers are
retracted first (see our reviews [12–16]). Essential
features of DPN are slowing of conduction veloc-
ity, altered sensory behavior, and prominent loss
of dermal and epidermal skin axons [17–20].

A. Areti · D. W. Zochodne (*)

Peripheral Nerve Research Laboratory, Division of
Neurology, Department of Medicine and the
Neurosciences and Mental Health Institute,
University of Alberta, Edmonton, AB, Canada
e-mail: zochodne@ualberta.ca

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 327
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_18
328
The material in this chapter overlaps with a
number of papers and reviews from our labora-
tory alone or collaboratively including some
published recently, which the reader is referred
to for greater depth [7, 14, 16, 21–27]. These
related reviews cover similar material but are
updated in this chapter with several nuances in
their interpretation. Some of the materials on
molecular alterations in diabetic neurons are not
replicated in this review. In discussing how DM
targets the nervous system, we begin with dorsal
root ganglia, critical but understudied sites that
house the cell bodies of peripheral sensory
neurons.
2 Structure and Physiology
of Dorsal Root Ganglia
Dorsal root ganglia (DRGs) are paired paraspinal
collections of peripheral neurons, cell bodies
(perikarya) of sensory axons that supply sensa-
tion to the head, trunk, and limbs. Identified
within or adjacent to spinal foramina, they are
susceptible to mechanical compression by disc
material or osteophytes, an outcome that may be
important in common symptoms such as back or
neck discomfort. DRGs have a vascular supply
subserved by segmental radicular branches that
supply nerve roots and by anastomoses from the
spinal artery [28]. They are unique structures, not
often studied in humans or animals because of
their relative inaccessibility to dissection.
Covered by a capsule, within is a subjacent layer
of neurons of varying sizes and molecular
makeup with their processes directed inward
toward the center of the ganglion. The capsule is
a derivative of the perineurium of the nerve trunk
and has a slightly different structure and proper-
ties when it surrounds DRG versus exiting nerve
root axons [29]. Neuron DRG are also placed at
varying depths within the ganglia interspersed
among the exiting axons. While less studied, it is
inferred that peripheral trigeminal ganglia share
characteristics of paraspinal DRGs. While we are
aware of few studies on quantitative measures of
blood flow in DRGs, the findings have suggested
interesting characteristics [30]. Blood flow of rat
A. Areti and D. W. Zochodne
DRGs is higher than that of the endoneurium of
peripheral nerve trunks, almost double in a range
of 35–40  mL/100g/min compared to
15–17 mL/100g/min in the endoneurium respec-
tively. Moreover, there is evidence of partial
“autoregulation” of flow, with relative stability
down to mean arterial pressures of 60  mmHg.
This differs from the nerve trunk endoneurium
where a gradual curvilinear relationship between
mean arterial pressure and blood flow is observed.
It is also different than CNS, where flows are
more tightly preserved at low mean arterial blood
pressures. Mean oxygen tensions directly mea-
sured in DRGs were comparable to the endoneu-
rium but lower than brain. While speculative,
these unique characteristics of DRG physiology
may render them susceptible to microangiopathy
in DM.  A second important characteristic of
DRGs is that their barrier characteristics to
ingress of blood-borne molecules are less robust
than both the blood nerve and blood brain barri-
ers [31]. This is important translational informa-
tion given that yet undeveloped therapies of DPN
might be capable of accessing DRG neurons
through a “leaky” barrier following systemic
administration.
Dorsal root ganglia house populations of
large, medium, and smaller sensory neurons, the
perikaryal size of which determines the caliber
of their axons. These neurons are classified as
“pseudo-unipolar” because one perikaryal
branch forms into a central branch that enters
the spinal cord and a peripheral branch joins
with motor and autonomic axons to form mixed
spinal nerve roots that blend into peripheral
nerve trunks. The central branches of large, neu-
rofilament rich expressing neurons join the dor-
sal columns of the spinal cord where they travel
upward to the brainstem. In DM, this anatomy is
important because detailed MR images of the
spinal cord have identified dorsal column atro-
phy, suggesting degeneration of the central
branches of large afferent sensory fibers [32].
Further resolution in these studies will be
required to determine whether column atrophy
relates to increased fiber packing, myelin thin-
ning, axon loss, or axonal atrophy. Early images
of DM DRGs have also been captured by MR
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
but have not had the resolution yet to clarify
their changes beyond gross volume alterations
[33]. Large caliber neurons give rise to large
myelinated “Aα” axons that subserve discrimi-
native touch, vibration, proprioception, and hair
movement. Their peripheral branches, also
myelinated, innervate sensory organs such as
Pacinian corpuscles, Meissner’s corpuscles, and
hair follicles transmitting rapidly adapting
mechanical sensation. The central branches of
alternative, small neurofilament poor neurons
synapse within the dorsal horn of the spinal
cord. All sensory neurons express neurofila-
ments, but the content of this scaffold protein
varies with neuronal size, in turn influencing its
signal in studies using immunohistochemical
labeling, hence the inaccurate terminology of
“neurofilament negative” small neurons. Small
neurons send unmyelinated axons to the skin
and other target organs. In the skin, the axons
invade and innervate the epidermis, sensing
pain, mechanical damage, and probably chemo-
sensitivity. The detailed physiology and inner-
vation of sensory neurons in the skin are
reviewed elsewhere [34, 35]. Unmyelinated
afferent sensory axons, so-­called C fibers that
innervate the epidermis as free nerve endings,
are divided into two categories, peptidergic and
nonpeptidergic. Peptidergic axons and neurons
express substance P (SP) and CGRP (calcitonin
gene-related peptide) and ­terminate in deeper
layers of the skin known as the stratum basale.
Nonpeptidergic axons are also called Mrgprs
(Mas-related g protein-coupled receptors) or
IB4 (named for lectin binding that labels these
neurons) [36]. This is of interest because in DM,
peptidergic and nonpeptidergic central axons
entering the dorsal horn of the spinal cord may
be targeted differentially [37]. Medium caliber
neurons contain CGRP and through intermedi-
ate caliber myelinated axons (Aβ and γ) trans-
mit mechanical sensitivity and pain. While the
normal differential protein content of DRG sen-
sory neurons has expanded, not reviewed here,
other common classifications include their
expression of neurotrophin family receptors
[38–40]. TrkA (Tropomyosin receptor kinase
A), the high affinity NGF receptor, is expressed
329
on small peptidergic axons. TrkB is the receptor
for BDNF and NT4/5, and it is expressed on
medium sized neurons, whereas TrkC, the
receptor for NT3, is found on larger caliber sen-
sory neurons. IB4 neurons are responsive to
GDNF and express its receptor, GFRα1.
3 DRG Responses to Injury
DRG neurons sense injury to their distal axon
branches by developing a series of structural and
molecular changes known as the “axotomy”
response or axon reaction [41]. Structural
changes include a change in the histological
appearance called central “chromatolysis,” attrib-
uted to displacement of ribosomes to the periph-
ery of the perikarya. Nuclei similarly are shifted
to the periphery of the cell. More important are a
series of molecular changes initially described by
Richardson, Hokfelt, Verge, and others [40, 42,
43] known as “RAGs” or regeneration-associated
genes. Prominent and early characterized RAGs
were downregulation of neurofilament proteins,
Trk receptors and upregulation of tubulin,
GAP43,c-Jun, and others. More recent RNA
sequencing data has identified an extensive list of
upregulated and downregulated sensory neuron
transcripts [44]. In turn, the trigger for RAG
development is likely a retrograde signal from
the injured axon, initially an electrical and cal-
cium wave, then retrogradely transported signals
including importins α and β and pErk [45]. The
major conceptual idea behind altered gene regu-
lation after axotomy is that it represents a major
shift in sensory axons from a stable maintenance
role to a growth and regenerative state. Both
upregulation and downregulation of neuron
mRNAs may be important in facilitating a growth
response. It is also a characteristic of the CNS
that similar retrograde RAG induction is attenu-
ated, in concert with a dramatically diminished
regenerative potential.
An important reason for addressing RAGs in
this review is the remarkable absence of RAG
development in experimental diabetes despite an
“axotomy-like” loss of their distal terminals.
Missing in particular are the upregulation of
330
growth-related proteins such as tubulin and
GAP43 [19]. The general downturn in gene
expression has suggested a degenerative pheno-
type and the possibility that mRNA generation
and protein translation are impaired.
There are other remarkable features of DRGs,
but of unknown significance in DM, given the
paucity of studies of both human and chronic
animal models. DRG neurons are intimately
surrounded by a special class of glial cell known
as the perineuronal satellite cells. These are
placed in close circumferential relationship to
DRG sensory neurons and have the curious
property of enlarging and proliferating in
response to distal axotomy. This is despite the
lack of a direct cellular connection to the injured
neuron. These glial cells thereby “sense” distal
neuronal injury albeit by unclear mechanisms.
Despite the lack of an overt physical connec-
tion, perineuronal satellite cells have direct
molecular communication with their neuron
partners. For example, administration of a fluo-
rochrome dye to distal axon branches not only is
retrogradely transported to their parent neuron
perikarya, but over time, the perineuronal glial
cells also express the fluorochrome, having been
transferred from neurons. Perineuronal satellite
cells are more dynamic than their neuron part-
ners and exhibit ongoing turnover-cell death and
proliferation, even in the absence of injury. They
also have close relationships with intrinsic resi-
dent macrophages within DRGs. The latter are a
newly recognized constituent of DRGs, broadly
classified as among DRCCs (dorsal root ganglia
resident cycling cells) [46]. DRCCs are popula-
tions of self-­renewing cells that include resident
macrophages as well as satellite glial cells that
colabel with Sox2, indicating cells with stem-
like characteristics. Moreover, the close physi-
cal relationships of these cells with adult DRG
neurons suggest an ongoing role in the spatial
and temporal arrangements of the
DRG.  Interestingly, despite prior assumptions
that DRG macrophages are largely blood borne,
the populations that proliferate within DRGs are
local. For example, their ongoing replication
stimulated by local CSF1 (colony-­stimulating
factor 1) signaling can be identified in explant
A. Areti and D. W. Zochodne
cultures without a connection to the systemic
circulation. Macrophages are important sources
of growth factors, of relevance if their endoge-
nous function within DRGs were impaired by
DM.  Schwann cell (SC) dysfunction, not dis-
cussed in this review, is a well-established fea-
ture of DPN, making it highly likely that related
perineuronal satellite cells are similarly targeted
[47–49].
The Sox2 expressing cell populations in
DRG likely overlap with progenitor cells in
spheroids isolated from DRGs [50] and are
interesting in that they may offer avenues to
replenish DRG sensory neurons when lost.
However at present, there is insufficient evi-
dence that this would be required in
DPN.  Overall, the extent and potential role of
altered DRG cellular dynamics and signaling
have not been explored in DM.
One important outcome of DM may be DNA
damage, either nuclear or mitochondrial. BRCA1
(Breast Cancer 1) is a DNA repair molecule that
repairs double-strand DNA breaks but also has
transcriptional activity, acts as a ubiquitin ligase,
and has heterochromatin-related gene silencing
activities [51, 52]. Surprising new findings indi-
cate that axotomy alone is sufficient to trigger a
DNA damage response in DRG sensory neurons
and that its repair is important for viable regrowth.
Along these lines, axotomized DRG sensory neu-
rons had a substantial rise in γH2A.X (H2A his-
tone family member X; >60% of neurons)
expression, a marker of DNA damage, a change
that indicates a dramatic rise in the extent of
DNA damage after injury. Whether DNA damage
can account for alterations in gene expression in
DPN is speculative and has not been examined.
Nuclear BRCA1 colocalizes in neurons with
γHSA.X expression where it may offer a repair
and stabilizing action that allows neurons to
recover and regenerate. For example, elimination
of BRCA1 in knockdown experiments impaired
neurite growth of adult sensory neurons in vitro
and peripheral nerve regeneration in vivo. Further
work is required to discover whether DNA dam-
age may be important in DM, whether it is prop-
erly repaired by BRCA, and whether it could also
account for diabetic mitochondriopathy [53, 54].
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
4 Dorsal Root Ganglia
and Diabetes Mellitus
Beyond DM-related declines in overall DRG
volume by MR analysis [33], neither gross
structural alterations of ganglia nor intra-gan-
glion axonal damage have been described in
diabetic DRGs. In a chronic model of mouse
DM type 1, perikaryal area and nuclear area
were reduced [55]. Careful nonbiased counting
of DRG neurons in a long-term models of diabe-
tes in rats has not identified evidence of sensory
neuron loss [19]: absence of loss in rats of
12 months diabetes duration and a maximum of
10% loss in mice of 9 months diabetes duration
[18]. This finding has argued against reports of
short-term DRG neuron apoptosis in cell culture
models, a finding not substantiated and poten-
tially linked to high and nonphysiological acute
glucose exposure [56]. Relative preservation of
neuron numbers and axon counts during these
long-term chronic models indicate that neuron
drop out is unlikely to contribute to the
phenotype.
Despite the absence of significant neuron
dropout, an important and positive finding,
DRGs and sensory neurons are nonetheless tar-
geted by chronic DM.  While our group failed
to identify declines in nerve blood flow in an
extensive series of DM models, we did identify
significant declines in DRG blood flow in
chronic models of DM in rats. This was noted
in 17- to 23-week diabetic BBW rats, a model
of type 2 DM [57] and 16-week diabetic type 1
STZ diabetic rats [58]. These changes, interest-
ingly, developed in the absence of declines of
DRG oxygen tensions. While reductions in
DRG blood flow might contribute to neuronal
stress, they may also arise from declines in
oxygen demand, with lesser oxygen extraction.
In an anatomical study of vascular architecture
in rats (examined in the absence of fixation and
harvested under anesthesia prior to euthana-
sia), numbers of perfused microvessels were
elevated in the endoneurium but not DRG of
diabetics [59]. The findings suggest that gan-
331
glia and nerves may adapt to microvascular
abnormalities in DM, maintaining essential
perfusion through angiogenesis perhaps trig-
gered by hypoxia. Recent work has also sug-
gested the possibility of altered capillary
dynamics in chronic DM, but it is uncertain
whether these are sufficient to cause neurode-
generation [60] (Table 1).
Endothelin-1 (ET-1) is a highly potent vaso-
constrictor peptide that when topically applied
over the epineurial vascular plexus induced tem-
porary ischemic conduction block in nondiabetic
nerves. In this model of ischemia, intense vaso-
constriction of the epineurial circulation impaired
downstream blood flow within the endoneurial
compartment. In models of DM however, the
impact had more significant consequences [62–
64]. ET-1 generated endoneurial infarction and
axonal degeneration in diabetics but not control
nondiabetic rats. DM microvascular dysfunction
in arterial and arteriolar preparations has been
linked to potentiated microvascular vasoconstric-
tion, in turn attributed to attenuated vasodilata-
tion [65, 66]. These physiological alterations
develop in the absence of overt structural micro-
angiopathy but nonetheless may render ischemic
damage to diabetic neurons.
Selective DRG ischemia was studied using
topical ET-1-induced vasoconstriction of capsu-
lar and radicular ganglia feeding vessels. In
DM, ET-1 generated intense vasoconstriction of
the feeding vessels of ganglia with large and
prolonged declines in local ganglion blood flow,
exceeding the changes observed in nondiabetics
[67]. Sensory neuron damage was manifest as
loss of neurofilaments, dissolution of neurons,
nuclear TUNEL labeling (indicating apoptosis),
“nests of Nageotte,” replacing lost neurons, and
chromatolytic changes such as displacement of
nuclei to the periphery of neuron cell bodies.
Intraganglionic axons and downstream sural
sensory axons developed degeneration. These
findings provided evidence that DRGs are
indeed sensitive to ischemia secondary to
diabetic-­related changes in microvascular
physiology.
Table 1  Phenotype and interventions in chronic diabetic models, Zochodne Laboratory
332

DM duration MCV SCV Thermal Mechanical Footpad

Title Model (pretreatment) Treatment slowing slowing sensation sensation axons Citation
GLP-1 SW mice 8 weeks 4 weeks, insulin Y Y Loss Loss ± [148]
Type 1 STZ pellets, exendin
Exendin-4 + + + + −
Low dose insulin − − + + −
High dose insulin − − − − −
dbdb 8 weeks Exendin-4 + − + + −
Low dose insulin − − − − −
High dose insulin − − − − −
Intranasal C57B6/L 8 weeks Intranasal Y Y F gain M loss [134]
Insulin, RAGE−/− mice Insulin for
8 weeks
Male IN insulin + + − −
Female IN insulin + + + −
RAGE−/− + + − +
male
RAGE−/− IN insulin ± + − −
DM, PTEN KD CD-1 mice 10 weeks None Y Y Loss Loss [163]
regeneration
Epigenetic studies CD-1 mice 20 weeks None Y Y Loss Loss Loss [73]
STZ
12–16 weeks IN miRNA + + + + +
3 doses/2 weeks
Let-7i
MiR-341 + + + −
Spliceosome C57B6/L 12 weeks Y Y Loss Loss [55]
mice STZ
Intraplantar − + + −
CWC siRNAx4; 28d
endpoint
ES studies, CD-1 male 4 or 8 weeks None ± Y Loss Loss [158]
regeneration mice STZ
HSP27 overexpression C57BL/6 24 weeks Y Y Loss Elevated Loss [61]
mice STZ
A. Areti and D. W. Zochodne
 DM duration MCV SCV Thermal Mechanical Footpad
Title Model (pretreatment) Treatment slowing slowing sensation sensation axons Citation
HSP overexpression − + + + −
Plantar insulin 8–12 weeks Paw insulin No change Loss (CD-1 Loss [132]
only)
C57BL/6J +
STZ
dbdb +
CD-1 STZ − + +
ZDF model ZDF rats 16 weeks None Y Y No loss Loss Loss [208]
Type 2
Intrathecal SD male rats 8 weeks IT insulin or Y Y Loss [116,
STZ IGF-1- 4 weeks 133]
0.1 IU insulin + − +
0.2 IU insulin + +
IGF-1 + + +
Spontaneous SW mice 36 weeks None + ±
regression STZ
Neurofilament−/− B6C3 mice 8 weeks None Y (incl Y (incl
DM Nf−/−) Nf−/−)
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy

Chronic DM SD rat STZ 52 weeks None Y Y [19, 209,

210]
Chronic DM SD male rats 26 weeks None Y Y [17]
STZ
Near nerve insulin SD male rats 4 weeks Near nerve insulin Y Y [131]
STZ
0.1 IU 3×/week for 4 + Different
weeks nerve
BBW model Female 20 weeks Y Y [57]
BBW rats
Indomethacin SD male 16 weeks Indomethacin± ± Y [211]
rats STZ Guanethidine
8 weeks
(continued)
333
Table 1 (continued)
334

DM duration MCV SCV Thermal Mechanical Footpad

Title Model (pretreatment) Treatment slowing slowing sensation sensation axons Citation
Indomethacin + +
Guanethidine − − worsens
Indo+Guaneth − −
Sulindac SD male rats 12 weeks Sulindac onset Y Y [58, 212]
STZ 16 weeks Sulindac later
Insulin 1.0 IU
Sulindac onset − +
16 weeks
Sulindac later ± ±
1 month
Insulin − +
Bolded entries are findings in the models, with specific interventionsm; (unbolded) from each study listed below; Y = yes; ‘Loss’ refers to an impairment in the listed modality.
For interventions, +indicates that the intervention did improve this parameter, – had no improvement. Nf -/- are neurofilament lacking animals
A. Areti and D. W. Zochodne
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy 335

5 Sensory Neurons in Diabetes contribute to the disease and might be amenable

to intervention. Peptide loss may explain the
As discussed above, there is very little direct evi- alterations in vascular reflexes that accompany
dence for significant sensory neuron loss or drop- DPN [72], and shifts in sodium channel expres-
out in chronic DM despite their susceptibility to sion may contribute to neuropathic pain, trigger-
acute ischemia and their vulnerability to short-­ ing anion overload and degeneration in diabetic
term exposure to high glucose levels. Neither neurons. Finally, heightened expression of PARP,
scenario is informative about their fate during NFκB, and activated caspase-3 demonstrate neu-
long-term diabetes. Human studies of diabetic rons under duress, despite their capacity to sur-
DRGs are limited. An older report by Greenbaum, vive. PARP is an energy-consuming DNA repair
lacking immunohistochemical or newer histo- molecule. However, in light of all of these
logical stains, studied six patients at postmortem changes, the finding that most sensory neurons in
[68]. Only three patients with clinical diabetic chronic DM survive remains important news.
polyneuropathy had evidence of mild neuron loss Restoring function and growth to surviving and
with nests of Nageotte, axon retraction bulbs, and existing neurons is an easier task than replenish-
vacuolation of sensory neurons. Vacuolation ing them.
however may have been a processing artifact More recent RNA microarray analyses of
[69]. In a later and separate report using EM, mRNAs in experimental diabetes have added
Schmidt identified degenerative changes in new targets and information [73, 74]. While rela-
DRGs from both chronic DM and aging includ- tively few of these assays have been carried out,
ing dystrophic proximal axons that accumulated it is likely the lists will vary with the choice of the
neurofilament [70]. DM model and its sex and the duration of hyper-
A previous review from our laboratory catego- glycemia. Moreover, unexplored factors such as
rized known molecular changes in experimental the gut microbiome of housed study rodents may
DM [71] including loss of structural proteins alter findings. Cheng et al. [73] identified a series
such as neurofilament subunits, tubulin, of up- and downregulated mRNAs in chronic
CGRP,SP,PACAP, and other neuropeptides, type 1 DM mice (streptozotocin [STZ] induced)
sodium channel Nav1.8, neurotrophin receptors, of duration 5 months. Prolonged models are pre-
GAP43 and NFκB among others. Upregulated ferred given that chronic human disease requires
molecules included activated caspase-3, Poly a significant exposure to DM, not addressed by
(ADP-ribose) polymerase (PARP), sodium chan- short duration studies of 1–2 months. In the long
nels Nav1.3,1.6,1.9 and subunit β3, Heat shock duration STZ model, theoretical concerns raised
protein HSP27, MAPK second messengers over STZ toxicity are obviated. The list of altered
(ERKs, JNKs, P38), and the insulin receptor. mRNAs in the Cheng model was accompanied
This original list, now expanded, remains incom- by two additional features: (1) upregulation of
plete but has identified a range of molecular alter- GW bodies, cytoplasmic sites of RNA processing
ations associated with chronic DM. The loss of within DRGs; and (2) a concurrent list of differ-
neurofilament structural protein mRNA in DRGs entially altered miRNAs, the short RNA supra-
paralleled loss of neurofilament investment in regulatory system of epigenetic RNA control
distal axons and their associated atrophy [17]. (discussed further below). Both the findings offer
Despite this loss, mice lacking neurofilaments additional and alternative explanations for the
nonetheless develop conduction slowing and altered molecular architecture of DM sensory
accelerated changes of DPN, indicating that this neuron-altered mRNA handling by GW bodies
protein is probably not part of the pathogenesis of and shifted supraregulatory control by miRNAs.
the disease [20]. The loss of growth-related These mechanisms contributing to altered protein
GAP43 and changes in MAPK signaling mole- synthesis may be additive to other disruptions
cules are important in providing impetus to sub- such as altered DNA repair considered above.
sequent ideas that attenuated growth programs However, the findings have added yet another
336
neurodegenerative mechanism, that of altered
spliceosomes, considered below.
Among the differentially regulated mRNA
“hits” in chronic diabetic DRGs, only a few
members have been examined in depth to date.
Price and Tomlinson identified evidence for a
role of Txnip (thioredoxin interacting protein) in
experimental DM [75], upregulation also con-
firmed in our work [73]. The significance of this
finding is that the protein may be a mediator of
oxidative stress in sensory neurons but also act as
a tumor suppressor, potentially limiting growth.
From our array list, we have also focused on two
lesser understood upregulated proteins, DUSPs
(dual specificity phosphatases) and CWC22, a
spliceosome protein.
CWC22, an upregulated protein involved in
the spliceosome, has proved interesting and rele-
vant to DM and DPN in different ways [55].
CWC22 is largely expressed in sensory neuron
nuclei, and its knockdown enhanced neurite out-
growth of adult sensory neurons in dissociated
cultures. This finding suggested that the protein
was a “brake” on neuron growth and plasticity. In
a chronic model of DPN in mice, unilateral hind-
paw dosing of CWC siRNA designed to knock-
down CWC improved ipsilateral sensory
conduction and thermal sensation. The contralat-
eral paw treated with scrambled control siRNA
retained neuropathic abnormalities, indicating a
pathogenic role for CWC22 upregulation in
DPN. Aberrant CWC action may disrupt the spli-
ceosome with which it is associated, a nuclear
complex that determines the fate of transcribed
mRNAs. Additional spliceosome abnormalities
in chronic DPN mice included upregulated num-
bers of Cajal bodies, sites of RNA splicing in
nuclei, altered localization of survival motor neu-
ron (SMN) protein, and abnormal nuclear foci of
small nuclear ribonucleoprotein particles
(snRNPs), also parts of the spliceosome.
Dysregulated motor neuron SMN is linked to the
development of spinal muscular atrophy.
DUSPs (dual specificity phosphatases; MAP
kinase phosphatases, MKPs), proteins that inter-
act with the MAPK signaling pathway, are a sep-
arate group of proteins we discovered in our
chronic DM mRNA array to be upregulated in
A. Areti and D. W. Zochodne
DRGs. Similarly understudied, DUSPs have
proven of interest in unexpected ways, albeit not
yet in the context of DPN [76]. Sensory neurons,
both injured and noninjured, express isoforms
DUSP1 and 4 in both perikarya and axons. Their
knockdown (KD) in sensory neurons, unlike
CWC22, suppresses neurite outgrowth, suggest-
ing a role by DUSPs in supporting normal axonal
“plasticity” of adult neurons. Along these lines,
DUSP KD or inhibition intensified axonal dam-
age in the setting of capsaicin neurotoxicity, indi-
cating a protective role by DUSPs for axons.
Moreover, in vivo, DUSP knockdown or inhibi-
tion accelerated axonal degeneration following
axon transection-loss of axon excitability, neuro-
filament dissolution, and morphological features
of axon breakdown. DUSPs are part of the
recently and partly untangled process of active
axonal degeneration, for which several member
proteins have been identified. A protein known as
SARM1 (sterile alpha and TIR motif containing
1) appears central to axonal degeneration. Along
these lines, accelerated axonal degeneration sec-
ondary to DUSP inhibition or KD was rescued by
concurrent SARM1 KD. In DM, DUSP upregu-
lation within sensory neurons might provide a
compensatory response to incipient axon dam-
age. There is also early evidence that SARM1
KD may protect against DPN [77]. The cascade
of axonal degeneration, not reviewed here, likely
involves mechanisms shared by many neuropa-
thies; its unraveling has offered several new tar-
gets for axonal protection, including SARM1
[78]. Downstream of SARM1, further proteins
also participate in axon breakdown including cal-
pains and caspase-6, the latter identified in early
degenerating skin axon profiles [79]. Overall,
there are close connections between the reper-
toire of proteins also found mainly in perikarya
that nonetheless have major impacts on down-
stream axon survival. A full understanding of
how DM might impact the full complex of axonal
degeneration related proteins requires further
work, but may offer opportunities to intervene.
Understanding the roles of supraregulatory
miRNAs in rendering diabetic neuron dysfunc-
tion has only just begun. Their role might relate
to the problem of fitting together diverse pieces
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
of the DPN phenotype into a single molecular
trigger. These pieces include, for example, elec-
trophysiological slowing of motor and sensory
conduction, distal loss of sensory axons in the
skin and other targets, axon, perikaryal atrophy,
and the generation of neuropathic pain despite
fiber loss. Moreover, the range of mRNA changes
discussed above is difficult to link together
beyond a generalized down run of cellular
machinery. miRNAs characteristically have
many mRNA targets and could be important in
wide “down tuning” of neurons or alternatively in
supporting inappropriate upregulation, as in the
case for CWC22. Cheng et  al. [73] identified a
series of differentially regulated miRNAs in the
same 5-month-old type 1 cohort of diabetic mice
that underwent mRNA analysis described above.
These mice had well-characterized phenotypic
features of DPN. One of the most prominent dif-
ferentially expressed miRNAs was mmu-Let7i,
reduced by 39% in diabetic DRGs. mmu-Let 7i
has a large range of mRNA targets (>900) that
include IGF-1 and its receptor among 84 other
growth pathway mRNAs, contrasted to 21 metab-
olism and diabetes pathway mRNAs. Exogenous
application of a mmu-Let 7i mimic had a trophic
impact on adult sensory neurons, encouraging
both neurite outgrowth and branching. Moreover,
intranasal administration of a mmu-Let 7i mimic
improved thermal and mechanical sensation,
motor and sensory conduction velocity and epi-
dermal innervation in a model of chronic type 1
DM.  In the same work, intranasal treatment of
anti-miR to instead tune down an alternative
upregulated miRNA, mmu-miR-341, improved
thermal sensation, and conduction velocities. The
overall findings lent support to the idea that
miRNA manipulation may influence the DPN
phenotype and to the idea that intranasal nucleo-
tides, may have promise in DPN or other disor-
ders, not unlike that offered by intranasal
insulin.
Several findings described above emphasize
the role of growth and plasticity molecules that
may have impaired actions in DPN. Although the
term “plasticity” has wide use, here we apply it to
the capacity of neurons to extend processes and
grow or regenerate. This property may provide
337
critical capacity for sensory axons to maintain
their investment in endorgans like skin. In keep-
ing with this theme, we next discuss several new
examples of strategies to enhance neuron growth
in the setting of damage or disease.
6 Regrowth Strategies
for Sensory Neurons
Diabetic neurons targeted by diverse metabolic
impacts from DM on sensory neurons may with-
stand damage through strategies that support
their overall growth properties. In several
instances, these approaches, largely not yet tested
in humans, have remarkable impacts on experi-
mental DPN. Proposed mechanisms of DPN are
not reviewed here in depth. A key tenet however
is “neurotoxic stress,” arising from glucotoxicity,
increased polyol flux, free radical toxicity, nitrer-
gic damage, mitochondrial abnormalities, lipid
toxicity, and deficits in growth factors [80–102].
In many putative therapies for DPN, alleviation
of the impact of one or more of these mechanisms
has been highlighted. An important example has
been the long history of aldose reductase inhibi-
tors (ARIs) to eliminate polyol flux. While ARIs
have had some impact on human disease, interest
in their use has waned and none have been fully
accepted therapeutically [103, 104].
Growth Factors and Insulin  The limited impact
of traditional growth factors in experimental and
human DPN has been reviewed elsewhere [15,
23]. In the cases of neurotrophins, lack of enthu-
siasm has stemmed from negative clinical trials,
uncertain dosing, downregulation of their recep-
tor expression in models, and their capability to
support only subsets of neurons [105]. Hepatic
growth factor has recently been trialed, albeit
with limited published pretrial preclinical data as
to its mechanism, with results suggesting an
impact on human pain [106].
We have previously made the case for exploit-
ing the neurotrophic properties of insulin. Almost
50  years ago, Frazier et  al. in “Nerve Growth
Factor and Insulin,” published by Science,
338
emphasized the evolutionary structural similari-
ties between NGF and insulin [107]. Unlike the
neurotrophins, of which NGF is a member, insu-
lin receptors (IRs) are widely expressed on
peripheral neurons and are upregulated in dia-
betic models. IRs share downstream transduction
cascades with other growth factors [108–117]
and on ligation they undergo autophosphoryla-
tion, exhibit tyrosine kinase activity, and utilize
downstream activators known as IRS-1 and 2
(IRS=insulin receptor substrate) [118, 119].
There is eventual activation of the p85 subunit of
PI3K, a central growth signal for neurons [119–
127]. Insulin is associated with increased neurite
outgrowth from adult neurons studied in  vitro
[128]. It also improves nerve regeneration in non-
diabetic injury models, a property that may be
appropriate for addressing DPN [129, 117].
Either loss of insulin in type 1 DM or resistance
to its actions in type 2 DM may have a role in the
development of peripheral neuron dysfunction.
Finally, the neuronal actions of insulin as a
growth factor are separate from its impact on gly-
cemia. The preclinical data suggesting a role for
insulin signaling in DPN are several: low insulin
doses (that do not alter blood glucose levels)
applied near nerve, within the skin, or intrathe-
cally to access DRGs all improve DPN [130–
133]. Intranasal insulin, a route that accesses
DRG through the CSF, improved experimental
DPN [134, 135]. Local corneal insulin restored
innervation as assessed using corneal confocal
microscopy [136]. Insulin-like Growth Factors
(IGFs), family members, use similar downstream
signaling pathways [137–139].
While the range of impacts of insulin in pre-
clinical work is impressive, there are potential
barriers to its translational success. Insulin resis-
tance is a central deficit in most type 2 DM
patients characterized by normal or high systemic
insulin levels in relationship to failed insulin-­
mediated glucose uptake by muscle and adipose
tissue [140–143]. We described evidence for
insulin “resistance” in neurons, referring to a
deficit in its ability to display trophic actions fol-
lowing high dose or prolonged insulin exposure
[144, 145]. These findings were confirmed by
others [146, 147]. Whereas several mechanisms
A. Areti and D. W. Zochodne
may account for neuronal insulin resistance, it is
possible that combined use with insulin sensitiz-
ers may overcome this limitation.
Glucagon-like peptide 1 (GLP-1) operates by
enhancing systemic insulin sensitivity in type 2
DM, but its receptors are also expressed on adult
sensory neurons, and upregulated in a model of
DM type 2 [148]. Moreover, a GLP-1 agonist,
exendin-4 has trophic actions on adult sensory
neurons in vitro, associated with increased neu-
rite outgrowth. In the setting of only a mild
impact on glycemia, subcutaneous exendin-4
given over 8 weeks in mice with DM of 2 months
duration was examined for its impact on DPN. In
contrast to partial impacts of either low- or high-­
dose insulin, exendin-4 improved motor and sen-
sory conduction velocity in type 1 DM mice and
reversed thermal and mechanical sensory loss.
Similar benefits, excepting sensory conduction
velocity, were noted in type 2 DM mice. There
were no significant impacts on epidermal inner-
vation but the shorter model durations limited
this endpoint. The overall findings paralleled
similar results in two other laboratories [149,
150] and provide preclinical support for human
studies of GLP-1 agonism in DPN. Human trials
focused on GLP-1 or its analogs in DPN have not
yet been reported.
Supporting Intrinsic Neuronal Growth
Properties  That sensory neurons require an
ongoing level of “plasticity” or growth in the
absence of injury is supported by newer findings.
For example, changes in epidermal and hair fol-
licle innervation may readjust at short notice
from noninvasive stimuli, such as hair shaving
[11]. In DM models, short-term local hindpaw
injections of low-dose insulin, signaling dermal
axon insulin receptors, can rapidly restore axon
numbers to nondiabetic values [132]. Finally,
human skin biopsies of normal subjects display
markers of growth and plasticity on epidermal
axons, indicating evidence of ongoing growth.
These included GAP43, Shh, SCG10, and others
[151]. Overall, it may be that sensory axons
require ongoing growth and plasticity simply to
deal with the wear and tear of skin and routine
shedding of superficial skin keratinocytes.
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
Removing intrinsic “brakes” on neuron
growth as a novel strategy to enhance plasticity
allows consideration of several targets. The idea
has been to unleash the growth potential of stable
and regeneration reluctant adult sensory neurons
temporarily and locally. Several “brakes” are
“tumor suppressor” molecules and pathways
identified as being gatekeepers to uncontrolled
tumor cell growth. Moreover, some “brakes” are
not among injury-related “RAGs” described
above, indicating that their expression is ongo-
ing, and stable irrespective of injury. Others how-
ever are tuned down after injury to foster
regrowth. “Preconditioning” refers to a height-
ened regenerative response of peripheral neurons
that have been subjected to an axon injury (axot-
omy) within the previous 5–7 days; it ramps up
early RAG expression, allowing a more robust
response to a subsequent injury [152]. While the
mechanism of preconditioning is unconfirmed,
rises in BDNF neurotrophic signaling have been
considered. In vitro preparations of adult neurons
(see below) demonstrate a dramatic rise in out-
growth if their axons are subjected to an axotomy
prior to their harvesting and growth in culture.
How dissociated cultured adult neurons respond
to interventions in vitro is an important investiga-
tive screening step as a robust predictor of in vivo
regeneration. “Neurites” are axon-like out-
growths from neurons in  vitro and their growth
along with overall neuron survival, neurite initia-
tion, branching, and directional conduct are all
important deliverables from in vitro analysis.
Replicating the molecular reprogramming
generated by preconditioning might encourage
better regrowth of peripheral sensory neurons.
While the triggers for preconditioning are
debated, axon injuries generate a retrograde
directed injury discharge and a calcium wave. On
the supposition that exogenous electrical stimula-
tion (ES) might mimic or augment this signal,
Gordon, Brushart, and colleagues identified that
a specific ES paradigm, delivered immediately
after axotomy to the intact proximal stump was
associated with improved motor axon regenera-
tion and preferential motor branch reinnervation
of muscles [153, 154]. Follow on work confirmed
a similar benefit from a 1 h epoch of 20 Hz stimu-
339
lation on sensory axon regrowth [155–157].
Additional findings were that ES indeed evoked a
calcium transient, as occurs during precondition-
ing, and that it increased induction of specific
RAG mRNAs including GAP43 and tubulin
while reversing the changes in ATF3, an injury
marker [158]. In vitro ES enhanced the transcrip-
tion of BDNF mRNA, indicating a close connec-
tion between ES and preconditioning [156, 159].
It also lowered PTEN levels (see below) and was
sensitive to PI3K inhibition.
Superimposed axon injury in the setting of
DPN offers insights into the regenerative poten-
tial of neurons, not unlike those gained by pre-
conditioning. In the peripheral nervous system,
DM is associated with a “double hit” comprised
of degenerative neuropathy as well as a specific
failure in regeneration. This topic has been
reviewed in detail separately [160]. Nonetheless,
ES, as a surrogate for preconditioning, was asso-
ciated with improved regeneration from a super-
imposed sciatic nerve injury in chronic diabetic
mice [158]. Yet to be tested is whether ES reverses
overall features of DPN, but since this is a focal
strategy, thought will be required as how best to
exploit it for generalized nerve dysfunction in
DM.
PTEN (phosphatase and tensin homolog
deleted on chromosome 10) was among the first
tumor suppressor molecules examined in periph-
eral neurons [161]. Mutations in PTEN are asso-
ciated with Cowden’s syndrome, a condition that
predisposes persons to oncogenesis with solid
tumors. PTEN is expressed in adult DRG neu-
rons, but more prominently in small caliber non-
peptidergic IB4 or Mpgprs neurons that have
slower regrowth properties [162]. We noted that
PTEN pharmacological inhibition or knockdown
by siRNA was associated with dramatic rises in
neurite outgrowth from dissociated adult neurons
in vitro. PTEN knockdown or inhibition was an
important example of the impact of removing a
key “brake” on neuron plasticity, demonstrating
robust enhanced outgrowth of their neurites.
Moreover, this impact was striking irrespective of
whether the neurons were previously uninjured
(intact) or already preconditioned.
Preconditioning has been assumed to generate
340
the peak regenerative response that neurons are
capable of. However, both the ES work, superim-
posing stimulation on an injury, and PTEN inhi-
bition have indicated that it is possible to drive
neuron regenerative plasticity significantly higher
than achieved with preconditioning alone.
Indeed, in the case of PTEN inhibition or knock-
down, heightened neurite outgrowth was yet
more pronounced if superimposed on an injury
state. Thus, this was the first example we were
aware of in which neuron outgrowth could be
ramped up substantially by a single molecular
trigger beyond the preconditioning level. In vivo
PTEN knockdown or inhibition increased axon
outgrowth from transected sciatic nerves. In both
type 1 and type 2 DM mice, PTEN was upregu-
lated in DRG sensory neurons and its knockdown
enhanced their neurite outgrowth in vitro. Finally,
we confirmed that PTEN knockdown in  vivo
could repair the regenerative deficit of mice with
type 1 DM and others confirmed similar findings
in a type 2 DM model in vivo [163, 164]. Taken
together these findings indicate that diabetic neu-
rons are responsive to strategies that improve
their plasticity, an important property to exploit
in treatment.
The next “tumor suppressor” analyzed was
chosen for its central role in transcriptional regu-
lation, Rb1(retinoblastoma 1), known to be
mutated in childhood retinoblastoma tumors.
Rb1 is expressed in the cytoplasm and nuclei of
adult DRG primary sensory neurons [165], but
unlike PTEN, its expression appears to involve
most sensory neuron subtypes, including those
from mice with chronic DM. In its nonphosphor-
ylated state Rb1 binds to a divergent transcription
factor E2F1 and prevents its activation [166].
While E2F1 signaling is best established in facili-
tating S phase mitotic DNA replication stage
entry, its actions are less well understood in
growth [167, 168]. Following phosphorylation by
cyclin-CDKs Rb1 releases E2F1 to permit its sig-
naling. Like PTEN, Rb1 is not among the known
RAGs indicating a constitutive “braking” role
independent of injury or axotomy [44, 165, 169].
Rb1 knockdown also does not activate PI3k-pAkt
[161, 170, 171], but its knockdown enhanced
adult neuron outgrowth, branching and initiation
A. Areti and D. W. Zochodne
in  vitro, in part through increased levels of
PPARϒ [Peroxisome proliferator-activated
receptor gamma]. PPARϒ, in turn, was expressed
in sensory neuron nuclei and cytoplasm, and it
enhanced in vitro neurite growth. In addition to
being a mediator of Rb1 knockdown’s impact on
neurons, PPARϒ is a known insulin sensitizing
agent [172, 173], recognized to promote insulin
signaling at several levels: IRS-1 and IRS-2 [174,
175], the p85 subunit of PI3K [176] and else-
where [177, 178]. Given this series of findings,
Rb1 knockdown, E2F1 activation, and PPARγ
may allow neurons to overcome insulin resis-
tance and respond to insulin especially in more
common type 2 DM. This role has yet to be con-
firmed. However, preliminary evidence indicates
that Rb1 knockdown has synergistic impacts on
neuronal growth with insulin and that it improves
chronic experimental DPN (unpublished data).
There are additional, less well-explored mol-
ecules and pathways that may offer roles in
restoring neuron plasticity. Adenomatous polyp-
osis coli (APC) is mutated in colorectal carcino-
mas [179] and suppresses tumorigenesis by
forming a destruction protein complex with
β-catenin, also a divergent transcription factor.
APC thereby blocks β-catenin signaling to sup-
press growth. APC expression rises in DRG neu-
rons after axotomy injury, and it is expressed in
the sensory neuron cytoplasm, axons, and SCs.
Like PTEN, APC is prominent within nonpepti-
dergic IB4 neurons that have attenuated growth
behavior. APC knockdown facilitated β-catenin
translocation to the neuron nucleus and enhanced
neurite outgrowth in vitro and nerve regeneration
in  vivo. The analysis of its impact on nerve
regrowth involved local application of siRNA at a
nerve injury site and ipsilateral knockdown of its
mRNA in the DRG and sciatic nerve. As in sev-
eral siRNA studies to date, the findings speak to
the mobility of small nucleotides exploiting ret-
rograde transport to knockdown and obviating
the need for viral vector directed therapies [180].
Sonic hedgehog protein (Shh) is a morphogen
that directs developmental motor pathway forma-
tion. It is expressed in adult DRG sensory neu-
rons, axons, and Schwann cells. Unlike the tumor
suppressor molecules, constitutive Shh may act
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
to support regeneration, as its knockdown attenu-
ates growth. Shh has been identified within epi-
dermal axons from normal human subjects,
supporting ideas that these axons are normally in
a state of ongoing growth [151]. Finally, but not
discussed here are a range of SC mitogens and
signals that may support axon regrowth indirectly
through the close partnership of axons and SCs
during regrowth [181, 182].
Topical Manipulation of Sensory Axons  While
this review has focused on overall strategies to
improve the growth and plasticity of sensory neu-
rons, manipulating their distal terminals may be
of interest. For example, a substantial number of
mRNAs are transported from perikarya (cell bod-
ies) distally to axon terminals and their growth
cones where they undergo local translation [183,
184]. CGRP is an interesting example of a local
axon synthesized peptide that signals to SCs and
microvessels [185, 186]. Several growth factor
receptors, including the insulin receptor, are also
expressed on skin axons and discussed above.
Local intracutaneous insulin restored epidermal
innervation in type 1 and 2 DM models of mice
[132]. Beyond these findings, growth cones also
express a number of intrinsic proteins, not
reviewed here, that determine whether they
advance and grow or collapse and retract. An
important example is RhoA, a GTPase that acti-
vates ROCK (Rho kinase) to collapse growth
cones exposed to inhibitor cues such as myelin or
chondroitin sulfate proteoglycans. RhoA or
ROCK inhibition increases growth of adult sen-
sory neurons in  vitro and outgrowth of axons
from a nerve injury [187].
A novel and exciting approach to directly sup-
port regrowth of skin axons has emerged from
work by Fernyhough and Calcutt exploiting axo-
nal muscarinic acetylcholine type 1 receptors
(M1Rs) [188–190]. Mice lacking the receptors
had enhanced neurite outgrowth in vitro and were
protected from DPN.  Moreover, M1R-specific
antagonists reversed mitochondrial dysfunction
in diabetes and features of neuropathy by mecha-
nisms operating in part through activation of
CAMKKβ, AMPK, and PGC-1α. In female dia-
341
betic mice, the M1R antagonist, pirenzepine
applied at 2% to the paw improved tactile allo-
dynia, thermal sensory loss, and loss of epider-
mal axons. Interestingly pirenzepine does not
cross the blood brain barrier and has already been
used extensively as a therapy in the clinic for
reflux and other clinical conditions. A clinical
trial of pirenzepine topical in patients with type 2
DM is underway.
7 Sensory Neurons and Pain
This topic is reviewed in detail elsewhere [15,
191, 192]. The pain phenotype of DPN is of com-
plex origin involving alterations within DRGs,
axons, and their ion channels together with
altered pain pathway signaling. Within DRG neu-
rons, the generation of pacemaker ectopic dis-
charges, reflecting inappropriate excitability has
been recognized for several decades as contribut-
ing to pain [193]. Both peripherally generated
and perikaryal originating discharges may under-
lie the positive sensory symptoms that contribute
to painful DPN.  Current ideas have particularly
focused on the roles of sodium channels and
T-type calcium channels. Mutated SCN9A
Nav1.7 sodium channel α subunits that are associ-
ated with a gain of function and hyperexcitability
have been identified by Faber, Merkies, and
Waxman in painful idiopathic sensory polyneu-
ropathies [194]. Nav1.7 channels are localized to
DRG neurons and axon terminals [195].
Alternative sodium channel α subunits desig-
nated Nav1.8 are expressed on DRG cell bodies
and terminals and have activation properties
linked to Nav1.7. Both have been described in
diabetic models linked to methylglyoxal, a meta-
bolic byproduct of DM [195, 196]. β-Subunits of
the sodium channel that modulate channel open-
ing have also been implicated [197] and a spe-
cific mutation, associated with neuron
hyperexcitability, has recently been described in
a diabetic patient [198].
Similarly, hyperexcitability of DRG neurons
may develop from T-type voltage-gated calcium
channels (Cav3.2) that are found in small- and
medium-sized neurons. These channels may also
342
contribute toward ectopic painful discharges in
DM [199, 200]. Their upregulated currents have
been described in both types 1 and 2 DM models
[201, 202].
Abnormalities in other channels have also
been described in DPN including HCN
(hyperpolarization-­ activated cyclin nucleotide-­
gated) channels [203] and potassium ion chan-
nels, specifically Kv1.2 [204]. Related changes in
the function of TRP channels including TRPV1
and TRPA1 have also been described in relation
to pain from DPN [205, 206].
Taken together, the evidence for several forms
of ion channelopathy in DM that contribute to
pain is substantial. It is unclear whether axon
ectopic discharges, DRG neuron pacemaking, or
remodeling of upstream CNS pain pathways are
primarily responsible for DM pain syndromes. It
is recognized that peripherally originating dis-
charges are capable of widespread secondary
remodeling of ascending nociceptive pathways.
Additional questions one might ask are how spe-
cific ion channel abnormalities develop in overall
DPN degeneration and what mechanisms gener-
ate these changes: transcriptional, translational,
or epigenetic? Since not all DPN phenotypes
have pain, it seems difficult to attribute neurode-
generation to pain-associated cation overload
within axons or perikarya. In DPN models, some
mouse strains routinely develop thermal and
mechanical sensory loss, even early in their
course, whereas others are characterized by
hyperalgesia. In patients, there may be an impor-
tant interaction between genotype and the likeli-
hood to develop the painful subtype of DPN.
8 Conclusions
The interaction between sensory neurons and
chronic DM remains complex and not neatly cat-
egorized. Sensory neurons develop a range of mor-
phological (axon retraction), gene expression and
protein changes that are difficult to link into a uni-
fying mechanism. Only briefly mentioned here are
primary targeting of mitochondria, reviewed else-
where [207] and dysfunction of Schwann cells, the
critical neuron glial partner. Important is the rec-
A. Areti and D. W. Zochodne
ognition that DPN is a unique neurodegenerative
disorder with sensory, but also autonomic and later
motor neuron targeting. Despite an onslaught of
multiple mechanisms that generate neurotoxic
“stress” with efforts to mitigate them, this review
instead emphasizes the promise of enhancing the
overall plasticity of the sensory neuron, enabling it
to withstand diabetes and thrive.
Acknowledgments  The effort devoted to this chapter was
supported by current operating grants from the Canadian
Institutes of Health Research (FRN148675 and 168929).
The authors acknowledge the experimental work cited
here and contributions by colleagues and trainees in the
Zochodne laboratory. Work described has been supported
since 1989 by the Canadian Institutes of Health Research,
Canadian Diabetes Association, the Alberta Heritage
Foundation for Medical Research, Muscular Dystrophy
Association of Canada, University of Alberta Hospital
Foundation, Department of Medicine and Division of
Neurology, University of Alberta, NIDDK Complications
Consortium, and the Juvenile Diabetes Foundation.
Disclosures  None.
References
1. Wild S, Roglic G, Green A, et al. Global prevalence
of diabetes: estimates for the year 2000 and projec-
tions for 2030. Diabetes Care. 2004;27:1047–53.
2. Capes S, Anand S.  What is type 2 diabetes? In:
Gerstein HC, Haynes RB, editors. Evidence-­
based diabetes care. Hamilton: Bc Decker; 2001.
p. 151–63.
3. Zochodne DW, Said G.  Recombinant human
nerve growth factor and diabetic polyneuropathy.
Neurology. 1998;51:662–3.
4. Apfel SC, Kessler JA, Adornato BT, et  al.
Recombinant human nerve growth factor in the
treatment of diabetic polyneuropathy. NGF Study
Group. Neurology. 1998;51:695–702.
5. Bril V. Invited review: status of current clinical tri-
als in diabetic polyneuropathy. Can J Neurol Sci.
2001;28:191–8.
6. Brown MJ, Asbury AK.  Diabetic neuropathy. Ann
Neurol. 1984;15:2–12.
7. Zochodne DW. Diabetic neuropathies: features and
mechanisms. Brain Pathol. 1999;9:369–91.
8. Dyck PJ, Kratz KM, Karnes JL, et  al. The preva-
lence by staged severity of various types of dia-
betic neuropathy, retinopathy, and nephropathy in
a population-based cohort: the Rochester Diabetic
Neuropathy Study. Neurology. 1993;43:817–24.
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
9. Mulder DW, Lambert EH, Bastron JA, et  al. The
neuropathies associated with diabetes mellitus.
Neurology. 1961;11:275–84.
10. Rundles RW.  Diabetic neuropathy—general review
with report of 125 cases. Medicine. 1945;24:111–60.
11. Cheng C, Guo GF, Martinez JA, et al. Dynamic plas-
ticity of axons within a cutaneous milieu. J Neurosci.
2010;30:14735–44.
12. Zochodne DW. Diabetes mellitus and the peripheral
nervous system: manifestations and mechanisms.
Muscle Nerve. 2007;36:144–66.
13. Anderson D, Zochodne DW.  Current ideas on the
treatment of diabetic neuropathies. Expert Rev
Endocrinol Metab. 2016;11(2):187–05.
14. Zochodne DW.  Sensory neurodegeneration in dia-
betes: beyond glucotoxicity. Int Rev Neurobiol.
2016;127:151–80.
15. Zochodne DW.  Mechanisms of diabetic neuron
damage: molecular pathways. Handb Clin Neurol.
2014;126:379–99.
16. Zochodne DW.  Diabetes and the plasticity of sen-
sory neurons. Neurosci Lett. 2015;596:60–5.
17. Scott JN, Clark AW, Zochodne DW. Neurofilament
and tubulin gene expression in progressive experi-
mental diabetes: failure of synthesis and export by
sensory neurons. Brain. 1999;122:2109–18.
18. Kennedy JM, Zochodne DW. Experimental diabetic
neuropathy and spontaneous recovery: is there irrep-
arable damage? Diabetes. 2005;54:830–7.
19. Zochodne DW, Verge VMK, Cheng C, et  al. Does
diabetes target ganglion neurons? Progressive sen-
sory neuron involvement in long term experimental
diabetes. Brain. 2001;124:2319–34.
20. Zochodne DW, Sun H-S, Cheng C, et al. Accelerated
diabetic neuropathy in axons without neurofila-
ments. Brain. 2004;127:2193–200.
21. Feldman EL, Callaghan BC, Pop-Busui R, et  al.
Diabetic neuropathy. Nat Rev Dis Primers.
2019;5:41.
22. Kobayashi M, Zochodne DW. Diabetic polyneurop-
athy: bridging the translational gap. J Peripher Nerv
Syst. 2020;25:66–75.
23. Kobayashi M, Zochodne DW.  Diabetic neuropathy
and the sensory neuron: new aspects of pathogenesis
and their treatment implications. J Diabetes Investig.
2018;9:1239–54.
24. Zochodne DW.  The challenges of diabetic poly-
neuropathy: a brief update. Curr Opin Neurol.
2019;32:666–75.
25. Zochodne DW.  Is early diabetic neuropathy a
disorder of the dorsal root ganglion? A hypoth-
esis and critique of some current ideas on the eti-
ology of diabetic neuropathy. J Peripher Nerv Syst.
1996;1:119–30.
26. Zochodne DW.  Nerve and ganglion blood flow in
diabetes: an appraisal. In: Tomlinson D, editor.
Neurobiology of diabetic neuropathy, vol. 50. San
Diego: Academic Press; 2002. p. 161–202.
27. Kobayashi, M, Cheng, C, De La Hoz, C, et  al.
Diabetic neuropathy and the sensory neuron: new
343
molecular targets. American Academy of Neurology
Abstracts P2.010. 2015.
28. Adams WE. The blood supply of nerves. I. Historical
review. J Anat. 1942;76:323–41.
29. Reinhold AK, Rittner H.  Characteristics of the
nerve barrier and the blood dorsal root gan-
glion barrier in health and disease. Exp Neurol.
2020;327:113244.
30. Zochodne DW, Ho LT. Unique microvascular char-
acteristics of the dorsal root ganglion in the rat.
Brain Res. 1991;559:89–93.
31. Arvidson B. Distribution of protein tracers in periph-
eral ganglia. A light and electron microscopic study
in rodents after various modes of tracer administra-
tion. Acta Univ Ups. 1979;344:1–72.
32. Selvarajah D, Wilkinson ID, Emery CJ, et al. Early
involvement of the spinal cord in diabetic peripheral
neuropathy. Diabetes Care. 2006;29:2664–9.
33. Jende JME, Kender Z, Rother C, et  al. Diabetic
polyneuropathy is associated with pathomorpho-
logical changes in human dorsal root ganglia: a
study using 3T MR neurography. Front Neurosci.
2020;14:570744.
34. Willis WD, Coggeshall RE.  Sensory mechanisms
of the spinal cord. 3rd ed. New  York: Kluwer
Academic/Plenum; 2004.
35. Zochodne DW. Our wired nerves. The human nerve
connectome. San Diego: Academic Press; 2020.
36. Zylka MJ, Rice FL, Anderson DJ. Topographically
distinct epidermal nociceptive circuits revealed
by axonal tracers targeted to Mrgprd. Neuron.
2005;45:17–25.
37. Akkina SK, Patterson CL, Wright DE. Gdnf rescues
nonpeptidergic unmyelinated primary afferents in
streptozotocin-treated diabetic mice. Exp Neurol.
2001;167:173–82.
38. Richardson PM, Verge VMK, Riopelle
RJ.  Quantitative radioautography for NGF recep-
tors. In: Rush RA, editor. Nerve growth factors.
Chichester: Wiley; 1989. p. 315–26.
39. Verge VMK, Riopelle RJ, Richardson PM.  Nerve
growth factor receptors on normal and injured sen-
sory neurons. J Neurosci. 1989;9:914–22.
40. Funakoshi H, Frisen J, Barbany G, et al. Differential
expression of mRNAs for neurotrophins and their
receptors after axotomy of the sciatic nerve. J Cell
Biol. 1993;123:455–65.
41. Lieberman AR.  The axon reaction: a review of the
principal features of perikaryal responses to axon
injury. Int Rev Neurobiol. 1971;14:49–124.
42. Richardson PM, Verge VM.  The induction of a
regenerative propensity in sensory neurons fol-
lowing peripheral axonal injury. J Neurocytol.
1986;15:585–94.
43. Verge VMK, Gratto KA, Karchewski LA, et  al.
Neurotrophins and nerve injury in the adult. Philos
Trans R Soc Lond B Biol Sci. 1996;351:423–30.
44. Chandran V, Coppola G, Nawabi H, et al. A systems-­
level analysis of the peripheral nerve intrinsic axonal
growth program. Neuron. 2016;89:956–70.
344
45. Perlson E, Medzihradszky KF, Darula Z, et  al.
Differential proteomics reveals multiple components
in retrogradely transported axoplasm after nerve
injury. Mol Cell Proteomics. 2004;3:510–20.
46. Krishnan A, Bhavanam S, Zochodne D.  An inti-
mate role for adult dorsal root ganglia resident
cycling cells in the generation of local macrophages
and satellite glial cells. J Neuropathol Exp Neurol.
2018;77:929–41.
47. Kalichman MW, Powell HC, Mizisin AP. Reactive,
degenerative, and proliferative Schwann cell
responses in experimental galactose and human dia-
betic neuropathy. Acta Neuropathol. 1998;95:47–56.
48. Goncalves NP, Vaegter CB, Pallesen LT. Peripheral
glial cells in the development of diabetic neuropathy.
Front Neurol. 2018;9:268.
49. Gumy LF, Bampton ET, Tolkovsky
AM.  Hyperglycaemia inhibits Schwann cell pro-
liferation and migration and restricts regeneration
of axons and Schwann cells from adult murine
DRG. Mol Cell Neurosci. 2008;37:298–311.
50. Geuna S, Borrione P, Fornaro M, et al. Neurogenesis
and stem cells in adult mammalian dorsal root gan-
glia. Anat Rec. 2000;261:139–40.
51. Starita LM, Parvin JD.  The multiple nuclear func-
tions of BRCA1: transcription, ubiquitination and
DNA repair. Curr Opin Cell Biol. 2003;15:345–50.
52. Krishnan A, Purdy K, Chandrasekhar A, et  al. A
BRCA1-dependent DNA damage response in the
regenerating adult peripheral nerve milieu. Mol
Neurobiol. 2018;55(5):4051–67.
53. Huang TJ, Price SA, Chilton L, et  al. Insulin pre-
vents depolarization of the mitochondrial inner
membrane in sensory neurons of type 1 diabetic
rats in the presence of sustained hyperglycemia.
Diabetes. 2003;52:2129–36.
54. Roy CS, Djordjevic J, Thomson E, et al. Depressed
mitochondrial function and electron transport
complex II-mediated H2O2 production in the cor-
tex of type 1 diabetic rodents. Mol Cell Neurosci.
2018;90:49–59.
55. Kobayashi M, Chandrasekhar A, Cheng C, et  al.
Diabetic polyneuropathy, sensory neurons, nuclear
structure and spliceosome alterations: a role for
CWC22. Dis Model Mech. 2017;10:215–24.
56. Russell JW, Sullivan KA, Windebank AJ, et  al.
Neurons undergo apoptosis in animal and cell
culture models of diabetes. Neurobiol Dis.
1999;6:347–63.
57. Zochodne DW, Ho LT, Allison JA.  Dorsal root
ganglia microenvironment of female BB Wistar
diabetic rats with mild neuropathy. J Neurol Sci.
1994;127:36–42.
58. Zochodne DW, Ho LT. The influence of sulindac on
experimental streptozotocin-induced diabetic neu-
ropathy. Can J Neurol Sci. 1994;21:194–202.
59. Zochodne DW, Nguyen C.  Increased peripheral
nerve microvessels in early experimental diabetic
neuropathy: quantitative studies of nerve and dorsal
root ganglia. J Neurol Sci. 1999;166:40–6.
A. Areti and D. W. Zochodne
60. Ostergaard L, Finnerup NB, Terkelsen A, et al. The
effects of capillary dysfunction on oxygen and glu-
cose extraction in diabetic neuropathy. Diabetologia.
2015;58:666–77.
61. Korngut L, Ma CH, Martinez JA, et al.Overexpression
of human HSP27 protects sensory neurons from dia-
betes. Neurobiol Dis. 2012;47(3):436–43.
62. Zochodne DW, Ho LT, Gross PM. Acute endoneu-
rial ischemia induced by epineurial endothelin in the
rat sciatic nerve. Am J Physiol. 1992;263:H1806–10.
63. Zochodne DW, Cheng C, Sun H. Diabetes increases
sciatic nerve susceptibility to endothelin-induced
ischemia. Diabetes. 1996;45:627–32.
64. Zochodne DW, Cheng C. Diabetic peripheral nerves
are susceptible to multifocal ischemic damage from
endothelin. Brain Res. 1999;838:11–7.
65. Zochodne DW, Ho LT.  Diabetes mellitus prevents
capsaicin from inducing hyperaemia in the rat sciatic
nerve. Diabetologia. 1993;36:493–6.
66. Thomsen K, Rubin I, Lauritzen M.  NO- and non-
NO-, non-prostanoid-dependent vasodilatation in
­
rat sciatic nerve during maturation and develop-
ing experimental diabetic neuropathy. J Physiol.
2002;543:977–93.
67. Xu Q-G, Cheng C, Sun H, et al. Local sensory gan-
glion ischemia induced by endothelin vasoconstric-
tion. Neuroscience. 2003;122:897–905.
68. Greenbaum D, Richardson PC, Salmon MV, et  al.
Pathological observations on six cases of diabetic
neuropathy. Brain. 1964;87:201–14.
69. Li X-G, Zochodne DW.  Microvacuolar neuronopa-
thy is a post-mortem artifact of sensory neurons. J
Neurocytol. 2003;32:393–8.
70. Schmidt RE, Beaudet LN, Plurad SB, et al. Axonal
cytoskeletal pathology in aged and diabetic
human sympathetic autonomic ganglia. Brain Res.
1997;769:375–83.
71. Toth C, Brussee V, Cheng C, et al. Diabetes mellitus
and the sensory neuron. J Neuropathol Exp Neurol.
2004;63:561–73.
72. Caselli A, Rich J, Hanane T, et  al. Role of
C-nociceptive fibers in the nerve axon reflex-­
related vasodilation in diabetes. Neurology.
2003;60:297–300.
73. Cheng C, Kobayashi M, Martinez JA, et al. Evidence
for epigenetic regulation of gene expression and
function in chronic experimental diabetic neuropa-
thy. J Neuropathol Exp Neurol. 2015;74:804–17.
74. Price SA, Zeef LA, Wardleworth L, et  al.
Identification of changes in gene expression in dor-
sal root ganglia in diabetic neuropathy: correlation
with functional deficits. J Neuropathol Exp Neurol.
2006;65:722–32.
75. Price SA, Gardiner NJ, Duran-Jimenez B, et  al.
Thioredoxin interacting protein is increased in sen-
sory neurons in experimental diabetes. Brain Res.
2006;1116(1):206–14.
76. Chandrasekhar A, Komirishetty P, Areti A, et  al.
Dual specificity phosphatases support axon plastic-
ity and viability. Mol Neurobiol. 2021;58:391–407.
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
77. Cheng Y, Liu J, Luan Y, et al. Sarm1 gene deficiency
attenuates diabetic peripheral neuropathy in mice.
Diabetes. 2019;68:2120–30.
78. Gerdts J, Summers DW, Milbrandt J, et al. Axon self-­
destruction: new links among SARM1, MAPKs, and
NAD+ metabolism. Neuron. 2016;89:449–60.
79. Woo V, Cheng C, Duraikannu A, et al. Caspase-6 is
a dispensable enabler of adult mammalian axonal
degeneration. Neuroscience. 2018;371:242–53.
80. Mizisin AP, Powell HC.  Schwann cell changes
induced as early as one week after galactose intoxi-
cation. Acta Neuropathol. 1997;93:611–8.
81. Borghini I, Ania-Lahuerta A, Regazzi R, et  al.
Alpha, beta I, beta II, delta, and epsilon protein
kinase C isoforms and compound activity in the sci-
atic nerve of normal and diabetic rats. J Neurochem.
1994;62:686–96.
82. Brett FM, Kalichman MW, Calcutt NA, et al. Effects
of seven days of galactose feeding and aldose reduc-
tase inhibition on mast cells and vessel morphometry
in rat sciatic nerve. J Neurol Sci. 1996;141:6–12.
83. Greene DA, Lattimer SA, Sima AA. Sorbitol, phos-
phoinositides, and sodium-potassium-ATPase in the
pathogenesis of diabetic complications. N Engl J
Med. 1987;316:599–606.
84. Gupta S, Sussman I, Mcarthur CS, et  al.
Endothelium-dependent inhibition of Na(+)-K+
AtPase activity in rabbit aorta by hyperglycemia.
possible role of endothelium-derived nitric oxide. J
Clin Invest. 1992;90:727–32.
85. Mizisin AP, Li L, Calcutt NA.  Sorbitol accumu-
lation and transmembrane efflux in osmotically
stressed JS1 Schwannoma cells. Neurosci Lett.
1997;229:53–6.
86. Roberts RE, Mclean WG.  Protein kinase C iso-
zyme expression in sciatic nerves and spinal
cords of experimentally diabetic rats. Brain Res.
1997;754:147–56.
87. Sima AAF, Brismar T, Yagihashi S.  Neuropathies
encountered in the spontaneously diabetic BB Wistar
rat. In: Dyck PJ, Thomas PK, Asbury AK, et al., edi-
tors. Diabetic neuropathy. Toronto: W.B. Saunders;
1987.
88. Cherian PV, Kamijo M, Angelides KJ, et al. Nodal
Na(+)-channel displacement is associated with
nerve-conduction slowing in the chronically diabetic
BB/W rat: prevention by aldose reductase inhibition.
J Diabetes Complications. 1996;10:192–200.
89. Ahlgren SC, Levine JD. Protein kinase C inhibitors
decrease hyperalgesia and C-fiber hyperexcitability
in the streptozotocin-diabetic rat. J Neurophysiol.
1994;72:684–92.
90. Ishii H, Jirousek MR, Koya D, et al. Amelioration of
vascular dysfunctions in diabetic rats by an oral PKC
beta inhibitor. Science. 1996;272:728–31.
91. Judzewitsch RG, Jaspan JB, Polonsky KS, et  al.
Aldose reductase inhibition improves nerve con-
duction velocity in diabetic patients. N Engl J Med.
1983;308:119–25.
345
92. Mizisin AP, Powell HC, Myers RR.  Edema and
increased endoneurial sodium in galactose neurop-
athy. reversal with an aldose reductase inhibitor. J
Neurol Sci. 1986;74:35–43.
93. Schmidt RE, Plurad SB, Sherman WR, et al. Effects
of aldose reductase inhibitor sorbinil on neuroaxonal
dystrophy and levels of myo-inositol and sorbital in
symapthetic autonomic ganglia of streptozotocin-­
induced diabetic rats. Diabetes. 1989;38:569–79.
94. Sima AA, Bril V, Nathaniel V, et  al. Regeneration
and repair of myelinated fibers in sural-nerve
biopsy specimens from patients with diabetic
neuropathy treated with sorbinil. N Engl J Med.
1988;319:548–55.
95. Zenon GJ, Abobo CV, Carter BL, et al. Potential use
of aldose reductase inhibitors to prevent diabetic
complications. Clin Pharm. 1990;9:446–57.
96. Zherebitskaya E, Akude E, Smith DR, et  al.
Development of selective axonopathy in adult
sensory neurons isolated from diabetic rats: role
of glucose-induced oxidative stress. Diabetes.
2009;58:1356–64.
97. Fernyhough P, Gallagher A, Averill SA, et  al.
Duplicate use 14199 aberrant neurofilament phos-
phorylation in sensory neurons of rats with diabetic
neuropathy. Diabetes. 1999;48:881–9.
98. Huang TJ, Sayers NM, Verkhratsky A, et  al.
Neurotrophin-3 prevents mitochondrial dysfunction
in sensory neurons of streptozotocin-diabetic rats.
Exp Neurol. 2005;194:279–83.
99. Low PA, Nickander KK, Tritschler HJ.  The roles
of oxidative stress and antioxidant treatment
in experimental diabetic neuropathy. Diabetes.
1997;46:S38–42.
100. Ceriello A, Giugliano D.  Oxidative stress and dia-
betic complications. In: Alberti KGGM, Zimmet P,
Defronzo RA, et al., editors. International textbook
of diabetes mellitus. Chichester, New  York: Wiley;
1997. p. 1453–61.
101. Lyons TJ, Jenkins AJ.  Glycation, oxidation, and
lipoxidation in the development of the complications
of diabetes: a carbonyl stress hypothesis. Diabetes
Rev. 1997;5:365–91.
102. Zalba G, Beaumont J, San Jose G, et  al. Vascular
oxidant stress: molecular mechanisms and patho-
physiological implications. J Physiol Biochem.
2000;56:57–64.
103. Oates PJ.  Polyol pathway and diabetic peripheral
neuropathy. Int Rev Neurobiol. 2002;50:325–92.
104. Chalk C, Benstead TJ, Moore F.  Aldose reduc-
tase inhibitors for the treatment of diabetic
polyneuropathy. Cochrane Database Syst Rev.
2007;2007:Cd004572.
105. Apfel SC, Schwartz S, Adornato BT, et al. Efficacy
and safety of recombinant human nerve growth factor
in patients with diabetic polyneuropathy: a random-
ized controlled trial. rhNGF Clinical Investigator
Group. JAMA. 2000;284:2215–21.
106. Kessler JA, Smith AG, Cha BS, et al. Double-blind,
placebo-controlled study of HGF gene therapy
346
in diabetic neuropathy. Ann Clin Transl Neurol.
2015;2:465–78.
107. Frazier WA, Angeletti RH, Bradshaw RA.  Nerve
growth factor and insulin. Science. 1972;176:482–8.
108. Ishii DN. Implication of insulin-like growth factors
in the pathogenesis of diabetic neuropathy. Brain
Res Brain Res Rev. 1995;20:47–67.
109. Ishii DN. Insulin and related neurotrophic factors in
diabetic neuropathy. Diabet Med. 1993;10:14s–5s.
110. Sugimoto K, Murakawa Y, Zhang W, et  al. Insulin
receptor in rat peripheral nerve: its localization and
alternatively spliced isoforms. Diabetes Metab Res
Rev. 2000;16:354–63.
111. Sugimoto K, Murakawa Y, Sima AA.  Expression
and localization of insulin receptor in rat dorsal
root ganglion and spinal cord. J Peripher Nerv Syst.
2002;7:44–53.
112. Fernyhough P, Mill JF, Roberts JL, et al. Stabilization
of tubulin mRNAs by insulin and insulin-like growth
factor I during neurite formation. Brain Res Mol
Brain Res. 1989;6:109–20.
113. Chiu SL, Chen CM, Cline HT.  Insulin recep-
tor signaling regulates synapse number, dendritic
plasticity, and circuit function in  vivo. Neuron.
2008;58:708–19.
114. Havrankova J, Roth J, Brownstein M. Insulin recep-
tors are widely distributed in the central nervous sys-
tem of the rat. Nature. 1978;272:827–9.
115. Karagiannis SN, King RH, Thomas
PK. Colocalisation of insulin and IGF-1 receptors in
cultured rat sensory and sympathetic ganglion cells.
J Anat. 1997;191:431–40.
116. Brussee V, Cunningham A, Zochodne DW.  Direct
insulin signalling of neurons reverses diabetic neu-
ropathy. Diabetes. 2004;53:1824–30.
117. Xu Q-G, Li X-Q, Kotecha SA, et al. Insulin as an
in vivo growth factor. Exp Neurol. 2004;188:43–51.
118. White MF. The insulin signalling system and the IRS
proteins. Diabetologia. 1997;40:S2–17.
119. Thirone AC, Huang C, Klip A. Tissue-specific roles
of IRS proteins in insulin signaling and glucose
transport. Trends Endocrinol Metab. 2006;17:72–8.
120. Heidenreich KA.  Insulin and IGF-I receptor sig-
naling in cultured neurons. Ann N Y Acad Sci.
1993;692:72–88.
121. Fadool DA, Tucker K, Phillips JJ, et  al. Brain
insulin receptor causes activity-dependent current
suppression in the olfactory bulb through mul-
­ binding proteins in transected nerves and cultured
tiple phosphorylation of Kv1.3. J Neurophysiol.
2000;83:2332–48.
122. Folli F, Bonfanti L, Renard E, et al. Insulin receptor
substrate-1 (IRS-1) distribution in the rat central ner-
vous system. J Neurosci. 1994;14:6412–22.
123. Edstrom A, Ekstrom PA.  Role of phosphatidylino-
sitol 3-kinase in neuronal survival and axonal out-
growth of adult mouse dorsal root ganglia explants.
J Neurosci Res. 2003;74:726–35.
124. Brunet A, Datta SR, Greenberg ME. Transcription-­
dependent and -independent control of neuronal sur-
A. Areti and D. W. Zochodne
vival by the PI3K-Akt signaling pathway. Curr Opin
Neurobiol. 2001;11:297–305.
125. Namikawa K, Honma M, Abe K, et al. Akt/protein
kinase B prevents injury-induced motoneuron death
and accelerates axonal regeneration. J Neurosci.
2000;20:2875–86.
126. Franke TF, Hornik CP, Segev L, et  al. PI3K/
Akt and apoptosis: size matters. Oncogene.
2003;22:8983–98.
127. Hanada M, Feng J, Hemmings BA. Structure, regu-
lation and function of PKB/Akt—a major therapeu-
tic target. Biochim Biophys Acta. 2004;1697:3–16.
128. Fernyhough P, Willars GB, Lindsay RM, et  al.
Insulin and insulin-like growth factor I enhance
regeneration in cultured adult rat sensory neurones.
Brain Res. 1993;607:117–24.
129. Toth C, Brussee V, Martinez JA, et  al. Rescue and
regeneration of injured peripheral nerve axons by
intrathecal insulin. Neuroscience. 2006;139:429–49.
130. Brussee V, Cunningham FA, Zochodne DW. Direct
insulin signaling of neurons reverses diabetic neu-
ropathy. Diabetes. 2004;53:1824–30.
131. Singhal A, Cheng C, Sun H, et al. Near nerve local
insulin prevents conduction slowing in experimental
diabetes. Brain Res. 1997;763:209–14.
132. Guo G, Kan M, Martinez JA, et  al. Local insulin
and the rapid regrowth of diabetic epidermal axons.
Neurobiol Dis. 2011;43:414–21.
133. Toth C, Brussee V, Zochodne DW.  Remote neuro-
trophic support of epidermal nerve fibres in experi-
mental diabetes. Diabetologia. 2006;49:1081–8.
134. De La Hoz CL, Cheng C, Fernyhough P, et  al. A
model of chronic diabetic polyneuropathy: benefits
from intranasal insulin are modified by sex and
RAGE deletion. Am J Physiol Endocrinol Metab.
2017;312(5):E407–19.
135. Francis GJ, Martinez JA, Liu WQ, et  al. Motor
end plate innervation loss in diabetes and the
role of insulin. J Neuropathol Exp Neurol.
2011;70:323–39.
136. Chen DK, Frizzi KE, Guernsey LS, et al. Repeated
monitoring of corneal nerves by confocal micros-
copy as an index of peripheral neuropathy in type-1
diabetic rodents and the effects of topical insulin. J
Peripher Nerv Syst. 2013;18:306–15.
137. Cheng HL, Randolph A, Yee D, et al. Characterization
of insulin-like growth factor-I and its receptor and
Schwann cells. J Neurochem. 1996;66:525–36.
138. Zochodne DW, Cheng C.  Neurotrophins and other
growth factors in the regenerative milieu of proximal
nerve stump tips. J Anat. 2000;196:279–83.
139. Cheng HL, Russell JW, Feldman EL.  IGF-I pro-
motes peripheral nervous system myelination. Ann
N Y Acad Sci. 1999;883:124–30.
140. Morino K, Petersen KF, Shulman GI.  Molecular
mechanisms of insulin resistance in humans and
their potential links with mitochondrial dysfunction.
Diabetes. 2006;55(Suppl 2):S9–S15.
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
141. Zeggini E, Parkinson J, Halford S, et al. Association
studies of insulin receptor substrate 1 gene (IRS1)
variants in type 2 diabetes samples enriched for
family history and early age of onset. Diabetes.
2004;53:3319–22.
142. Esposito DL, Li Y, Vanni C, et  al. A novel T608R
missense mutation in insulin receptor substrate-1
identified in a subject with type 2 diabetes impairs
metabolic insulin signaling. J Clin Endocrinol
Metab. 2003;88:1468–75.
143. Imoto K, Kukidome D, Nishikawa T, et  al. Impact
of mitochondrial reactive oxygen species and apop-
tosis signal-regulating kinase 1 on insulin signaling.
Diabetes. 2006;55:1197–204.
144. Singh B, Xu Y, Guo GF, et al. Insulin signaling and
insulin resistance in sensory neurons. J Peripher
Nerv Syst. 2009;13:254.
145. Singh B, Xu Y, McLaughlin T, et al. Resistance to tro-
phic neurite outgrowth of sensory neurons exposed
to insulin. J Neurochem. 2012;121(2):263–76.
146. Kim B, Mclean LL, Philip SS, et al. Hyperinsulinemia
induces insulin resistance in dorsal root ganglion
neurons. Endocrinology. 2011;152:3638–47.
147. Grote CW, Morris JK, Ryals JM, et al. Insulin recep-
tor substrate 2 expression and involvement in neu-
ronal insulin resistance in diabetic neuropathy. Exp
Diabetes Res. 2011;2011:212571.
148. Kan M, Guo G, Singh B, et al. Glucagon-like peptide
1, insulin, sensory neurons, and diabetic neuropathy.
J Neuropathol Exp Neurol. 2012;71:494–510.
149. Himeno T, Kamiya H, Naruse K, et  al. Beneficial
effects of exendin-4 on experimental polyneuropa-
thy in diabetic mice. Diabetes. 2011;60:2397–406.
150. Jolivalt CG, Fineman M, Deacon CF, et  al. GLP-1
signals via ERK in peripheral nerve and prevents
nerve dysfunction in diabetic mice. Diabetes Obes
Metab. 2011;13:990–1000.
151. Bautista J, Chandrasekhar A, Komirishetty PK, et al.
Regenerative plasticity of intact human skin axons. J
Neurol Sci. 2020;417:117058.
152. Senger JB, Verge VMK, Chan KM, et al. The nerve
conditioning lesion: a strategy to enhance nerve
regeneration. Ann Neurol. 2018;83:691–702.
153. Brushart TM, Hoffman PN, Royall RM, et  al.
Electrical stimulation promotes motoneuron regen-
eration without increasing its speed or conditioning
the neuron. J Neurosci. 2002;22:6631–8.
154. Al-Majed AA, Neumann CM, Brushart TM, et  al.
Brief electrical stimulation promotes the speed and
accuracy of motor axonal regeneration. J Neurosci.
2000;20:2602–8.
155. Brushart TM, Jari R, Verge V, et al. Electrical stimu-
lation restores the specificity of sensory axon regen-
eration. Exp Neurol. 2005;194:221–9.
156. Geremia NM, Gordon T, Brushart TM, et  al.
Electrical stimulation promotes sensory neuron
regeneration and growth-associated gene expression.
Exp Neurol. 2007;205:347–59.
157. Singh B, Xu QG, Franz CK, et  al. Accelerated
axon outgrowth, guidance, and target reinnerva-
347
tion across nerve transection gaps following a
brief electrical stimulation paradigm. J Neurosurg.
2012;116:512.
158. Singh B, Krishnan A, Micu I, et al. Peripheral neuron
plasticity is enhanced by brief electrical stimulation
and overrides attenuated regrowth in experimental
diabetes. Neurobiol Dis. 2015;83:134–51.
159. Geremia NM, Pettersson LM, Hasmatali JC, et  al.
Endogenous BDNF regulates induction of intrinsic
neuronal growth programs in injured sensory neu-
rons. Exp Neurol. 2010;223:128–42.
160. Kennedy JM, Zochodne DW. The regenerative defi-
cit of peripheral nerves in experimental diabetes:
its extent, timing and possible mechanisms. Brain.
2000;123:2118–29.
161. Christie KJ, Webber CA, Martinez JA, et al. PTEN
inhibition to facilitate intrinsic regenerative out-
growth of adult peripheral axons. J Neurosci.
2010;30:9306–15.
162. Tucker BA, Rahimtula M, Mearow KM.  Laminin
and growth factor receptor activation stimulates dif-
ferential growth responses in subpopulations of adult
DRG neurons. Eur J Neurosci. 2006;24:676–90.
163. Singh B, Singh V, Krishnan A, et al. Regeneration of
diabetic axons is enhanced by selective knockdown
of the PTEN gene. Brain. 2014;137:1051–67.
164. Pham VM, Tu NH, Katano T, et al. Impaired periph-
eral nerve regeneration in type-2 diabetic mouse
model. Eur J Neurosci. 2018;47:126–39.
165. Christie KJ, Krishnan A, Martinez JA, et  al.
Enhancing adult nerve regeneration through the
knockdown of retinoblastoma protein. Nat Commun.
2014;5:3670.
166. Futatsugi A, Utreras E, Rudrabhatla P, et al. Cyclin-­
dependent kinase 5 regulates E2F transcription
factor through phosphorylation of Rb protein in neu-
rons. Cell Cycle. 2012;11:1603–10.
167. Dimova DK, Dyson NJ. The E2F transcriptional net-
work: old acquaintances with new faces. Oncogene.
2005;24:2810–26.
168. Dick FA, Rubin SM. Molecular mechanisms under-
lying Rb protein function. Nat Rev Mol Cell Biol.
2013;14:297–306.
169. Abe N, And Cavalli V. Nerve injury signaling. Curr
Opin Neurobiol. 2008;18:276–83.
170. Jones DM, Tucker BA, Rahimtula M, et  al. The
synergistic effects of NGF and IGF-1 on neurite
growth in adult sensory neurons: convergence on
the PI 3-kinase signaling pathway. J Neurochem.
2003;86:1116–28.
171. Park KK, Liu K, Hu Y, et al. Promoting axon regen-
eration in the adult CNS by modulation of the PTEN/
mTOR pathway. Science. 2008;322:963–6.
172. Culman J, Zhao Y, Gohlke P, et  al. PPAR-gamma:
therapeutic target for ischemic stroke. Trends
Pharmacol Sci. 2007;28:244–9.
173. Bordet R, Ouk T, Petrault O, et  al. PPAR: a new
pharmacological target for neuroprotection in stroke
and neurodegenerative diseases. Biochem Soc Trans.
2006;34:1341–6.
348
174. Iwata M, Haruta T, Usui I, et al. Pioglitazone ame-
liorates tumor necrosis factor-alpha-induced insulin
resistance by a mechanism independent of adipo-
genic activity of peroxisome proliferator—activated
receptor-gamma. Diabetes. 2001;50:1083–92.
175. Smith U.  Pioglitazone: mechanism of action. Int J
Clin Pract Suppl. 2001;13–18.
176. Rieusset J, Auwerx J, Vidal H.  Regulation of
gene expression by activation of the peroxisome
proliferator-­activated receptor gamma with rosigli-
tazone (BRL 49653) in human adipocytes. Biochem
Biophys Res Commun. 1999;265:265–71.
177. Ribon V, Johnson JH, Camp HS, et  al.
Thiazolidinediones and insulin resistance: peroxi-
some proliferatoractivated receptor gamma activa-
tion stimulates expression of the CAP gene. Proc
Natl Acad Sci U S A. 1998;95:14751–6.
178. Baumann CA, Chokshi N, Saltiel AR, et al. cloning
and characterization of a functional peroxisome pro-
liferator activator receptor-gamma-responsive ele-
ment in the promoter of the CAP gene. J Biol Chem.
2000;275:9131–5.
179. Duraikannu A, Martinez JA, Chandrasekhar A,
et al. expression and manipulation of the APC-beta-­
catenin pathway during peripheral neuron regenera-
tion. Sci Rep. 2018;8:13197.
180. Chitwood DH, Timmermans MC.  Small RNAs are
on the move. Nature. 2010;467:415–9.
181. Chen YY, Mcdonald D, Cheng C, et  al. Axon and
Schwann cell partnership during nerve regrowth. J
Neuropathol Exp Neurol. 2005;64:613–22.
182. Webber CA, Christie KJ, Cheng C, et al. Schwann
cells direct peripheral nerve regeneration through
the Netrin-1 receptors, DCC and Unc5H2. Glia.
2011;59:1503–17.
183. Verma P, Chierzi S, Codd AM, et  al. Axonal pro-
tein synthesis and degradation are necessary for
efficient growth cone regeneration. J Neurosci.
2005;25:331–42.
184. Willis DE, Twiss JL. The evolving roles of axonally
synthesized proteins in regeneration. Curr Opin
Neurobiol. 2006;16:111–8.
185. Zochodne DW, Allison JA, Ho W, et  al. Evidence
for CGRP accumulation and activity in experimental
neuromas. Am J Physiol. 1995;268:H584–90.
186. Toth CC, Willis D, Twiss JL, et al. Locally synthe-
sized calcitonin gene-related peptide has a critical
role in peripheral nerve regeneration. J Neuropathol
Exp Neurol. 2009;68:326–37.
187. Cheng C, Webber CA, Wang J, et  al. Activated
RHOA and peripheral axon regeneration. Exp
Neurol. 2008;212:358–69.
188. Calcutt NA, Smith DR, Frizzi K, et  al. Selective
antagonism of muscarinic receptors is neuropro-
tective in peripheral neuropathy. J Clin Invest.
2017;127:608–22.
189. Saleh A, Sabbir MG, Aghanoori MR, et  al.
Muscarinic toxin 7 signals via Ca(2+)/calmodulin-­
dependent protein kinase kinase beta to augment
A. Areti and D. W. Zochodne
mitochondrial function and prevent neurodegenera-
tion. Mol Neurobiol. 2020;57:2521–38.
190. Jolivalt CG, Frizzi KE, Han MM, et al. Topical deliv-
ery of muscarinic receptor antagonists prevents and
reverses peripheral neuropathy in female diabetic
mice. J Pharmacol Exp Ther. 2020;374:44–51.
191. Lee-Kubli CA, Calcutt NA.  Painful neuropathy:
mechanisms. Handb Clin Neurol. 2014;126:533–57.
192. Greig M, Tesfaye S, Selvarajah D, et al. Insights into
the pathogenesis and treatment of painful diabetic
neuropathy. Handb Clin Neurol. 2014;126:559–78.
193. Wall PD, Devor M. Sensory afferent impulses origi-
nate from dorsal root ganglia as well as from the
periphery in normal and nerve injured rats. Pain.
1983;17:321–39.
194. Faber CG, Hoeijmakers JG, Ahn HS, et al. Gain of
function Na(V) 1.7 mutations in idiopathic small
fiber neuropathy. Ann Neurol. 2012;71:26–39.
195. Lauria G, Ziegler D, Malik R, et  al. The role of
sodium channels in painful diabetic and idiopathic
neuropathy. Curr Diab Rep. 2014;14:538.
196. Bierhaus A, Fleming T, Stoyanov S, et  al.
Methylglyoxal modification of Nav1.8 facilitates
nociceptive neuron firing and causes hyperalgesia in
diabetic neuropathy. Nat Med. 2012;18:926–33.
197. Shah BS, Gonzalez MI, Bramwell S, et al. Beta3, a
novel auxiliary subunit for the voltage gated sodium
channel is upregulated in sensory neurones follow-
ing streptozocin induced diabetic neuropathy in rat.
Neurosci Lett. 2001;309:1–4.
198. Alsaloum M, Estacion M, Almomani R, et  al. A
gain-of-function sodium channel beta2-subunit
mutation in painful diabetic neuropathy. Mol Pain.
2019;15:1744806919849802.
199. Todorovic SM, Lingle CJ. Pharmacological proper-
ties of T-type Ca2+ current in adult rat sensory neu-
rons: effects of anticonvulsant and anesthetic agents.
J Neurophysiol. 1998;79:240–52.
200. Latham JR, Pathirathna S, Jagodic MM, et  al.
Selective T-type calcium channel blockade alle-
viates hyperalgesia in ob/ob mice. Diabetes.
2009;58:2656–65.
201. Todorovic SM, Jevtovic-Todorovic V.  Targeting of
Cav3.2 T-type calcium channels in peripheral sen-
sory neurons for the treatment of painful diabetic
neuropathy. Pflugers Arch. 2014;466:701–6.
202. Orestes P, Osuru HP, Mcintire WE, et  al. Reversal
of neuropathic pain in diabetes by targeting gly-
cosylation of Ca(V)3.2 T-type calcium channels.
Diabetes. 2013;62:3828–38.
203. Tsantoulas C, Lainez S, Wong S, et  al.
Hyperpolarization-activated cyclic nucleotide-­
gated 2 (HCN2) ion channels drive pain in mouse
models of diabetic neuropathy. Sci Transl Med.
2017;9:Eaam6072.
204. Zenker J, Poirot O, De Preux Charles AS, et  al.
Altered distribution of juxtaparanodal Kv1.2 subunits
mediates peripheral nerve hyperexcitability in type 2
diabetes mellitus. J Neurosci. 2012;32:7493–8.
Diabetic Sensory Neurons, Dorsal Root Ganglia, and Neuropathy
205. Andersson DA, Gentry C, Light E, et  al.
Methylglyoxal evokes pain by stimulating TRPA1.
PLos One. 2013;8:E77986.
206. Dull MM, Riegel K, Tappenbeck J, et  al.
Methylglyoxal causes pain and hyperalge-
sia in human through C-fiber activation. Pain.
2019;160:2497–507.
207. Chowdhury SK, Smith DR, Fernyhough P. The role
of aberrant mitochondrial bioenergetics in diabetic
neuropathy. Neurobiol Dis. 2013;51:56–65.
208. Brussee V, Guo GF, Dong YY, et al. Distal degenera-
tive sensory neuropathy in a long term type 2 diabe-
tes rat model. Diabetes. 2008;57:1664–73.
349
209. Zochodne DW, Verge VM, Cheng C, et  al. Nitric
oxide synthase activity and expression in experimen-
tal diabetic neuropathy. J Neuropathol Exp Neurol.
2000;59:798–807.
210. Cheng C, Zochodne DW. Sensory neurons with acti-
vated caspase-3 survive long-term experimental dia-
betes. Diabetes. 2003;52:2363–71.
211. Zochodne DW, Ho LT.  The influence of indo-
methacin and guanethidine on experimental strep-
tozotocin diabetic neuropathy. Can J Neurol Sci.
1992;19:433–41.
212. Van Der Sloot P, Mizisin A, Zochodne DW. Sulindac
in established experimental diabetes: a follow-up
study. Can J Neurol Sci. 1995;22(3):198–201.
 Micro- and Macrovascular Disease
in Diabetic Neuropathy

Lihong Chen and Aristidis Veves

1 Introduction Chronic diabetic complications are the result

Diabetes is often defined a “vascular disease”
because of the early and extensive involvement
of the vascular tree observed in diabetic
patients and, even, in those at risk of develop-
ing diabetes. Both the micro- and macrocircu-
lation are affected, though the pathophysiology,
histology, clinical history, and clinical sequelae
at the two vascular levels appear to be quite
different.
Historically, there have been two competing
hypotheses regarding the origins of diabetic neu-
ropathy. According to the first one, diabetic neu-
ropathy is secondary to the impairment of nerve
and Schwann cells, while the second one stated
that diabetic neuropathy is mostly ascribed to
microvascular disease. However, it is currently
realized that nerve and microvascular injury are
both among the factors that contribute to nerve
dysfunction.
of small vessel disease. Diabetic microangiopa-
thy has been considered the main anatomic alter-
ation leading to the development of retinopathy,
nephropathy, and neuropathy. Nevertheless, mac-
roangiopathy, i.e., atherosclerosis of peripheral
arteries, is also a prominent feature of long-­
lasting diabetes and is characterized as involving
predominantly distal arteries. The possible links
between diabetic micro- and macrovascular alter-
ations and nerve damage will be the focus of this
chapter.
2 Microvascular Disease:
Overview and Anatomic
Changes
Lesions specific for diabetes have been observed
in the arterioles and capillaries of the foot and
other organs that are the typical targets of dia-

L. Chen
Division of Endocrinology, West China Hospital,
Sichuan University, Chengdu, China
A. Veves (*)
Department of Surgery, Beth Israel Deaconess
Medical Center, Boston, MA, USA
e-mail: aveves@bidmc.harvard.edu

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 351
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_19
352 L. Chen and A. Veves

betic chronic complications. A nowadays-­ ity of the vessels further limiting their ability to
historical retrospective histological study dilate in response to different stimuli [12].
demonstrated the presence of PAS-positive mate- In the kidney, nonenzymatic glycosylation
rial in the arterioles of amputated limb specimens reduces the charge on the basement membrane,
from diabetic patients [1]. Though it was believed which may account for transudation of albumin,
for several years that the anatomic changes an expanded mesangium, and albuminuria [13].
described were occlusive in nature, in 1984 Similar increases in vascular permeability occur
LoGerfo and Coffmann recognized that, in dia- in the eye and probably contribute to macular
betic patients, there is no evidence of an occlu- exudate formation and retinopathy [14]. In sim-
sive microvascular disease [2]. Subsequent plest terms, microvascular structural alterations
prospective anatomic staining and arterial casting in diabetes result in an increased vascular perme-
studies have demonstrated the absence of an arte- ability and impaired autoregulation of blood flow
riolar occlusive lesion, thus dispelling the hope- and vascular tone.
less notion of diabetic “occlusive small vessel Previous studies have shown a correlation
disease” [3, 4]. between the development of diabetic chronic
While there is no occlusive lesion in the dia- complication and metabolic control with perhaps
betic microcirculation, other structural changes the strongest evidence coming from the Diabetes
do exist. Arteriolar hyalinization, corresponding Control and Complications Trial (DCCT), which
to the thickening of walls of arterioles, was enrolled type 1 diabetic patients, and the United
observed in patients with diabetes, regardless of Kingdom Prospective Diabetes Study (UKPDS),
hypertension [5]. These abnormalities are char- which enrolled type 2 diabetic patients [15, 16].
acterized by hypertrophic remodeling and are The results from both clinical trials clearly
associated with impaired endothelium-depen- showed a delay in the development and progres-
dent vasodilation in  vitro. Furthermore, endo- sion of retinopathy, nephropathy, and neuropathy
neurial capillary density was also found to be with intensive glycemic control, thus supporting
increased in patients with diabetes, suggesting the direct causal relationship between hypergly-
that ­capillary density may respond to diabetes- cemia and microcirculation impairment. In con-
induced nerve ischemia [6]. The thickening of trast, there was less evident for macrovascular
the capillary basement membrane, which is disease, which was assessed only in the UKPDS.
accompanied by pericyte degeneration and It should be emphasized hyperglycemia alone
endothelial cell hyperplasia, is the dominant is not sufficient to trigger generalized diabetic
structural change in diabetic neuropathy. The microvascular pathologies. The Joslin Diabetes
composition of basement membrane in diabetes Center 50-year Medalist Study has shown that
suggested that accumulation of glycosylated 30–35% patients with type 1 diabetes did not have
and cross-linked proteins contributed to the significant microvascular complications [17].
expanded membrane structure and impaired This suggests that there may be unidentified
function. This can affect endothelial cell and genetic or other endogenous protective factors
extracellular matrix interactions and represents that diminish the adverse microvascular effects of
a response to the metabolic changes related to hyperglycemia. It has been shown that multiple
diabetes and hyperglycemia [7–9]. Furthermore, mechanisms are involved in the adverse effects of
the thickness of basement membrane of skin hyperglycemia with vascular complications,
and nerve correlates with the extent of neuropa- including nonenzymatic glycation and the forma-
thy in diabetes [10]. However, this alteration tion of advanced glycation end products (AGEs);
does not lead to occlusion of the capillary enhanced reactive oxygen production and actions;
lumen, and arteriolar blood flow may be normal endoplasmic reticulum (ER) stress; and the acti-
or even increased despite these changes [11]. vation of the polyol pathway, the diacylglycerol
On the contrary, it might act as a barrier to the (DAG)–protein kinase C (PKC) pathway [5], Src
exchange of nutrients and/or increase the rigid- homology-2 domain-containing phosphatase-1
Micro- and Macrovascular Disease in Diabetic Neuropathy 353

Hyperglycemia

Genetic and
LocaI tissue
epigenetic
responses
modulators

Increase intracellular glucose metabolism

Cell signaling dysfunction Toxic metabolites Altered osmols and

(Activation of PKC, MAP kinase, (AGE, oxidants, redox potentiaI
SHP-1, and phosphatases) methylglyoxal (polyol pathway)

Oxidative stress, ER stress, Inflammatory cytokines

Loss of insulin actions Kallikrein-bradykinin activation

Vascular and specific tissue cells dysfunction and

abnormal turnover

Nephropathy Retinopathy Neuropathy

Fig. 1  Schema of hyperglycemia’s induced pathways to microvascular complications. MAP, mitogen-activated

protein

(SHP-1), and the reninangiotensin system (RAS)
and kallikrein-bradykinin (BK) systems (Fig. 1).
Although the structural alterations observed in
diabetic capillaries do not affect the basal blood
flow, some functional abnormalities of the micro-
vascular circulation, which may eventually result
in a relative ischemia, have been extensively doc-
umented. This aspect will be deeply discussed in
the next section.
3 Pathophysiology
of Microvascular Disease
and Endothelial Dysfunction
in Diabetes
3.1 Functional Changes
Diabetes mellitus, even in the absence of com-
plications, impairs the vascular reactivity, that
is the endothelium dependent and independent
vasodilation, in the skin microcirculation [18].
Many glucose-related metabolic pathways can
determine endothelium dysfunction: increased
aldose reductase activity leading to the imbal-
ance in nicotinamide adenine dinucleotide
phosphate (NADP)/nicotinamide adenine dinu-
cleotide phosphate, reduced form (NADPH);
auto-­oxidation of glucose leading to the forma-
tion of reactive oxygen species (ROS);
“advanced glycation end-products” (AGEs)
produced by nonenzymatic glycation of pro-
teins; abnormal n6-fatty acid metabolism and
inappropriate activation of protein kinase C. All
these different pathways lead to an increase of
oxidative stress which is responsible for a
reduced availability of nitric oxide and, in turn,
for a functional tissue hypoxia and the develop-
ment of diabetic chronic complications [19]
(Fig. 2).
354 L. Chen and A. Veves

DIABETES

Hyperglycemia

DAG Impaired
n-6 fatty acid Polyol Sugar autoxidation Advanced
metabolism pathway glycation

Triglycerides PKC
LDL

OXIDATIVE STRESS

ENDOTHELIAL DYSFUNCTION

Capillary blood flow endoneurial hypoxia

NERVE DYSFUNCTION

NCV, Regeneration, Structural

Fig. 2  New concepts in the pathogenesis of diabetic neuropathy

3.2 Microvascular Dysfunction
and Diabetic Neuropathy
Microvascular reactivity is further reduced at the
foot level in the presence of peripheral diabetic
neuropathy. Endothelial nitric oxide synthase
(eNOS) is a key regulator of vascular nitric oxide
production. Immunostaining of foot skin biopsies
in our unit, with antiserum to human eNOS, glu-
cose transporter I (GLUT I), which is a functional
marker of the endothelium and von Willebrand
factor, an anatomical marker, showed no differ-
ence among diabetic patients with or without
peripheral neuropathy in the staining of GLUT I
and von Willebrand factor, while the staining for
the eNOS was reduced in neuropathic patients
(Fig.  3) [20]. Another study documented
increased levels of iNOS and reduced eNOS lev-
els in skin from the foot of diabetic patients with
severe neuropathy and foot ulceration [21].
Differences in the microcirculation between
the foot and forearm levels have also been inves-
tigated, the main hypothesis being that increased
hydrostatic pressure in distal microcirculatory
beds, related to the orthostatic posture, affects the
foot microcirculation more than at the forearm
level. The endothelium-dependent and -indepen-
dent vasodilation is, in fact, lower at the foot level
when compared to the forearm in healthy sub-
jects and both non-neuropathic and neuropathic
diabetic patients [22]. This forearm–foot gradient
exists despite a similar baseline blood flow at the
foot and forearm level. Therefore, it is reasonable
to believe that erect posture may be a contribut-
ing factor for the early development of the nerve
damage at the foot, compared to the forearm.
Micro- and Macrovascular Disease in Diabetic Neuropathy 355

Fig. 3  Expression of 80
eNOS in patients with
diabetic neuropathy
(black columns),
patients with both
diabetic neuropathy and 60
peripheral vascular
disease (hatched
columns) and healthy
subjects (white 40
columns). The
expression of eNOS was
reduced in both the
diabetic groups
compared to the healthy 20
subjects

0
absent reduced normal

Endoneurial microangiopathy was related to
the severity of neuropathy, suggesting that there
might be a causal relationship between impaired
microvasculature and diabetic neuropathy [6,
23]. In a large observational study, it was found
that endothelial function was a strong indepen-
dent predictor of DPN [24]. Experimental data
also showed that endothelium impairment could
cause neuropathy through involvement of the
Desert Hedgehog pathway [25].
3.2.1 Endothelium Dysfunction
Previous studies have suggested that endothelium-­
dependent vasodilation is impaired in diabetes,
regardless of the presence of long-term compli-
cations [26–28]. It has also been shown that
endothelium dysfunction precedes the develop-
ment of macro- and microvascular diabetic com-
plications, indicating that endothelial impairment
could be a critical factor toward the development
of diabetic vascular complications [29]. However,
these changes mainly involve the ability of
microcirculation under stress, such as foot injury
while they do not affect the baseline blood flow
[20]. This feature, combined with the arteriove-
nous shunt that is present at the foot level due to
autonomic neuropathy, can account for observa-
tions in clinical practice, in which, diabetic neu-
ropathy with or without ulceration presents with
a warm limb with distended veins and palpable
pulses. The final result is that despite no reduc-
tion in the total lower extremity blood flow, there
is functional ischemia that promoted injury and
impairs wound healing.
3.3 Autonomic Denervation
Autonomic nerve fibers take part in the modula-
tion of blood circulation. At rest, noradrenergic
sympathetic nerves are tonically active in normo-
thermic environments and increase their activity
during cold exposure, releasing both norepineph-
rine and co-transmitters to decrease skin blood
flow [30]. Autonomic neuropathy can compro-
mise the diabetic microcirculation because of the
development of arterio-venous shunting due to
sympathetic denervation. The opening of these
shunts leads to a maldistribution of blood between
the nutritional capillaries and subpapillary ves-
sels, and consequent aggravation of microvascu-
lar ischemia. Studies employing sural nerve
photography and fluorescein angiography as well
as other elegant techniques seem to support this
concept [31, 32]. In addition to the above, these
changes also result in impaired thermoregulation
[33].
A loss of sympathetic tone is also responsible
for an increased capillary permeability in diabetic
patients with neuropathy [34, 35]. This might
356 L. Chen and A. Veves

Fig. 4  The nerve axon

reflex-related NERVE AXON REFLEX
vasodilation or
neurovascular response:
stimulation of the Acetylcholine
c-nociceptive nerve
fibers by acetylcholine
or other noxious stimuli
leads to antidromic
stimulation of the
adjacent c-fibers, which Nerve
secrete Calcitonin cell
Gene-Related Peptide
(CGRP) that causes
vasodilation and
increased local blood
flow
Vasodilation
CGRP

cause endoneurial edema, as demonstrated by % 300 286

employing magnetic resonance spectroscopy,
which can in turn represent another mechanism
leading to a reduction of endoneurial perfusion
and a worsening of the nerve damage [35]. The
increased lower extremity capillary pressure upon
assuming the erect posture, due to early loss of
postural vasoconstriction (mediated by the sympa-
thetic fibers), might amplify this edematous effect.
3.4 Nerve Axon Reflex
Diabetic somatic neuropathy can further affect
the skin microcirculation by the impairment of
the axon reflex-related vasodilatation (Lewis’
flare). Under normal conditions, the stimulus of
the c-nociceptive nerve fibers not only travels in
the normal direction, centrally toward the spinal
cord, but also peripherally (“antidromic conduc-
tion”) to local cutaneous blood vessels, causing a
vasodilatation by the release of vasoactive sub-
stances, such as calcitonin gene-related peptide
(CGRP), Neuropeptide Y, substance P, and bra-
dykinin by the c fibers initiate neurogenic inflam-
mation (Fig. 4). This short circuit, or nerve axon
reflex, is responsible for the Lewis’ triple flare
response to injury and plays an important role in
increasing local blood flow when it is mostly
needed, i.e., in condition of stress.
268
200
100
41*
29* 25*
0
DN DA DV DC C
Fig. 5  The neurovascular response (expressed as percent-
age of blood flow increase over the baseline blood flow) is
significantly reduced at the foot level of diabetic patients
with peripheral somatic neuropathy (DN), autonomic neu-
ropathy (DA), and peripheral artery disease (DV) com-
pared to diabetic patients without complications (DC) and
healthy controls (C) *p < 0.001
This neurovascular (N-V) response is signifi-
cantly reduced at the foot level in diabetic patients
with peripheral somatic neuropathy, autonomic
neuropathy, and peripheral artery disease com-
pared to diabetic patients without complications
and healthy control subjects (Fig.  5) [22, 36].
Moreover, local anesthesia significantly reduces
the nerve axon reflex-related vasodilation at the
foot of patients without peripheral neuropathy,
while it has no effect on the amount of the pre-­
anesthesia N-V vasodilation—which is already
Micro- and Macrovascular Disease in Diabetic Neuropathy
very low—at the foot of neuropathic patients
[37]. This suggests that the main determinant of
the presence of the neurovascular vasodilation is
c-fiber function and that its measurement could
be used as a surrogate measure of the function of
these fibers.
Neurovascular response is associated with
measures of peripheral nerve function [37, 38].
Studies in our units have shown that an N-V
response lower than 50% is highly sensitive
(90%) and adequately specific (74%), in identify-
ing diabetic patients with peripheral neuropathy
[38]. Besides, the finding that this response is sig-
nificantly reduced even in the early stages of
peripheral neuropathy supports the hypothesis
that small fiber damage is a precocious event in
the clinical history of diabetic neuropathy—even
preceding large fibers’ impairment (Fig. 6). This
leads to impaired vasodilation under conditions
of stress, such as injury or inflammation.
Therefore, it is possible to speculate that small
fiber neuropathy might further contribute to nerve
hypoxic damage by the impairment of this hyper-
emic response, determining a vicious cycle of
injury.
180
% blood flow increase over baseline
*
160
*
140
120
100
80
60
40
20
0
C D
Fig. 6  The nerve axon reflex-related vasodilation at the
foot level in a diabetic population stratified on the basis of
the degree of peripheral somatic neuropathy in patients
without neuropathy (D), with mild neuropathy (DN mild),
with moderate neuropathy (DN moderate), and with
severe neuropathy (DN severe) compared to healthy con-
357
The previous conclusions are supported by
recent studies in experimental diabetes which
have demonstrated that epineurial arterioles of
the sciatic nerve are innervated by sensory nerves
that contain CGRP and mediate a hyperemic
response at this level [39]. Furthermore, it has
been shown that in long-term diabetic rats the
amount of CGRP present in epineurial arterioles
is diminished, which could be due to a denerva-
tion process [40]. Exogenous CGRP-mediated
vasodilation of these arterioles is also impaired in
experimental diabetes, indicating a reduced
CGRP bioactivity [40]. All these findings further
support a role of small sensory nerve fibers’
impairment in the development and progression
of diabetic neuropathy.
The impairment of the nerve axon reflex-­
related vasodilation is not affected by successful
bypass surgery in patients with peripheral arterial
disease. In addition, the endothelium-dependent
and -independent vasodilation, which is not
related to the nerve axon reflex, remains impaired
after successful revascularization. Therefore,
despite correction in obstructive lesions and res-
toration of normal blood flow in the large vessels,
DN mild DN moderate DN severe
trols (C). Median (25–75 percentile). The nerve axon
reflex-related vasodilation is already significantly reduced
in the early stages of neuropathy (subclinical neuropathy),
supporting the belief that small fiber dysfunction might
precede large fiber impairment in the natural history of
diabetic nerve damage
358
the changes in microcirculation continue to be
present and cause tissue hypoxia under condi-
tions of stress [41].
In summary, both the functional and structural
changes observed in diabetic microcirculation
contribute to the shift of blood flow away from
the nutritive capillaries to low-resistance arterio-
venous shunts, leading to functional ischemia of
tissues including peripheral nerves and, conse-
quently, to the development of diabetic periph-
eral neuropathy and other diabetic chronic
complications.
4 Lower Extremity Arterial
Disease and Diabetes
Smoking, diabetes mellitus, and dyslipidemia are
the most associated risk factors of lower extrem-
ity artery disease [42]. Diabetes mellitus is esti-
mated to be associated with two- to fourfold
increase in the prevalence of peripheral arterial
disease (PAD) [43]. The concomitant occurrence
of atherosclerotic peripheral vascular disease and
peripheral neuropathy in patients with diabetes is
the main factor in the development of diabetic
foot pathology. Furthermore, peripheral arterial
disease (PAD) of the lower extremities is consid-
ered the major risk factor for lower-extremity
amputation, and it is also accompanied by a high
likelihood for cardiovascular and cerebrovascular
diseases [44]. In a large, multinational
­meta-­analysis, the presence of diabetes was asso-
ciated with nearly a twofold increased odds for
PAD [45]. The risk of amputation in patients with
diabetes is 0.2–2.0/1000, which is 10–15 times
greater in diabetes than in the nondiabetic popu-
lation [46].
Although the underlying pathogenesis of ath-
erosclerotic disease in diabetics is similar to that
noted in nondiabetics, there are still some sig-
nificant differences. Diabetic patients present
with atherosclerotic disease at a significantly
younger age than their nondiabetic counterparts,
tend to present with more advanced disease and
have a worse prognosis. The prevalence of con-
comitant of PAD and diabetes is high in patients
with critical lower limb ischemia, with more
L. Chen and A. Veves
than 50% patients with critical limb ischemia
having diabetes [47]. In addition, diabetes is also
associated with more severe below-the-knee
PAD.  The anterior tibial, posterior tibial, and
peroneal arteries are most commonly affected
and tend to have more diffuse lesions [48–50].
Equally important is the observation that the
arteries of the foot, specifically the dorsalis
pedis, are often spared of occlusive disease. This
provides an excellent option for a distal revascu-
larization target [51].
The clinical presentation and prognosis of
PAD with diabetes is also implicated by the coex-
istence of peripheral neuropathy. Microvascular
disease (i.e., retinopathy, nephropathy, neuropa-
thy) independently associated with limb amputa-
tion after controlling for traditional atherogenic
risk factors. Concomitant presence of microvas-
cular disease and PAD synergistically amplifies
the risk of amputation. Thus, it is imperative to
take into consideration the importance of micro-
vascular disease in consideration of limb salvage
strategies [52]. In addition, diabetic patients are
less likely to present with typical claudication
symptoms than patients without diabetes. As a
consequence, the development of tissue loss (foot
ulceration or gangrene) may represent the first
sign of lower limb ischemia and because of its
limb-threatening potential, it is termed as critical
limb ischemia. Therefore, diabetic patients with a
foot ulcer should always be evaluated for isch-
emia, irrespective of their symptoms, particularly
for the increased risk of limb-threatening infec-
tion and faulty healing related to PVD [53].
The observations that distal vessels (i.e., distal
peroneal, dorsal pedal, and plantar arteries) are
often spared from arterial occlusive disease had a
crucial impact on the manner in which peripheral
vascular disease is approached in the diabetic
population. In the past, based upon the false pre-
sumption of small vessel disease, diabetics were
not treated as aggressively with revascularization
as is now standard. A more aggressive attempt to
correct the vascular deficit in diabetic ischemic
limbs in addition to more aggressive measures to
control local infection has radically altered the
prognosis of peripheral vascular disease in the
diabetic extremity.
Micro- and Macrovascular Disease in Diabetic Neuropathy
5 Conclusions
Changes in the micro- and macrocirculation, both
anatomical and functional, contribute to the
development of diabetic neuropathy. On the other
hand, the development of diabetic neuropathy
also affects the vasodilatory capacity of the
microcirculation and can interfere with the clini-
cal presentation of peripheral obstructive arterial
disease. Thus, the interaction between changes in
the vasculature and peripheral nerves is bidirec-
tional and results in changes in both blood flow
and neuronal function.
References
1. Goldenberg S, Alex M, Joshi RA, Blumenthal
HT. Nonatheromatous peripheral vascular disease of
the lower extremity in diabetes mellitus. Diabetes.
1959;8:261–73.
2. LoGerfo FW, Coffman JD.  Current concepts.
Vascular and microvascular disease of the foot in
diabetes. Implications for foot care. N Engl J Med.
1984;311:1615–9.
3. Conrad MC. Large and small artery occlusion in dia-
betics and nondiabetics with severe vascular disease.
Circulation. 1967;36:83–91.
4. Strandness DE Jr, Priest RE, Gibbons GE. Combined
clinical and pathologic study of diabetic and non-
diabetic peripheral arterial disease. Diabetes.
1964;13:366–72.
5. Rizzoni D, Porteri E, Guelfi D, Muiesan ML,
Valentini U, Cimino A, Girelli A, Rodella L, Bianchi
R, Sleiman I, Rosei EA. Structural alterations in sub-
cutaneous small arteries of normotensive and hyper-
tensive patients with non-insulin-dependent diabetes
mellitus. Circulation. 2001;103:1238–44.
6. Thrainsdottir S, Malik RA, Dahlin LB, Wiksell
P, Eriksson KF, Rosen I, Petersson J, Greene DA,
Sundkvist G.  Endoneurial capillary abnormali-
ties presage deterioration of glucose tolerance and
accompany peripheral neuropathy in man. Diabetes.
2003;52:2615–22.
7. LeBleu VS, Macdonald B, Kalluri R.  Structure and
function of basement membranes. Exp Biol Med.
2007;232:1121–9.
8. Haitoglou CS, Tsilibary EC, Brownlee M, Charonis
AS. Altered cellular interactions between endothelial
cells and nonenzymatically glucosylated laminin/type
IV collagen. J Biol Chem. 1992;267:12404–7.
9. Williamson JR, Tilton RG, Chang K, Kilo
C.  Basement membrane abnormalities in diabetes
mellitus: relationship to clinical microangiopathy.
Diabetes Metab Rev. 1988;4:339–70.
359
10. Malik RA, Newrick PG, Sharma AK, Jennings A,
Ah-See AK, Mayhew TM, Jakubowski J, Boulton
AJ, Ward JD.  Microangiopathy in human diabetic
neuropathy: relationship between capillary abnor-
malities and the severity of neuropathy. Diabetologia.
1989;32:92–102.
11. Parving HH, Viberti GC, Keen H, Christiansen
JS, Lassen NA.  Hemodynamic factors in the gen-
esis of diabetic microangiopathy. Metabolism.
1983;32:943–9.
12. Flynn MD, Tooke JE.  Aetiology of diabetic foot
ulceration: a role for the microcirculation? Diabet
Med. 1992;9:320–9.
13. Mogensen CE, Schmitz A, Christensen
CK.  Comparative renal pathophysiology relevant to
IDDM and NIDDM patients. Diabetes Metab Rev.
1988;4:453–83.
14. Cunha-Vaz JG.  Studies on the pathophysiology of
diabetic retinopathy. The blood-retinal barrier in dia-
betes. Diabetes. 1983;32(Suppl 2):20–7.
15. Diabetes C, Complications Trial Research G, Nathan
DM, Genuth S, Lachin J, Cleary P, Crofford O, Davis
M, Rand L, Siebert C.  The effect of intensive treat-
ment of diabetes on the development and progression
of long-term complications in insulin-dependent dia-
betes mellitus. N Engl J Med. 1993;329:977–86.
16. Group UPDSU.  Intensive blood-glucose control
with sulphonylureas or insulin compared with con-
ventional treatment and risk of complications in
patients with type 2 diabetes (UKPDS 33). Lancet.
1998;352:837–53.
17. Keenan HA, Costacou T, Sun JK, Doria A, Cavellerano
J, Coney J, Orchard TJ, Aiello LP, King GL. Clinical
factors associated with resistance to microvascular
complications in diabetic patients of extreme disease
duration: the 50-year medalist study. Diabetes Care.
2007;30:1995–7.
18. Morris SJ, Shore AC, Tooke JE.  Responses of the
skin microcirculation to acetylcholine and sodium
nitroprusside in patients with NIDDM. Diabetologia.
1995;38:1337–44.
19. Cameron NE, Eaton SE, Cotter MA, Tesfaye
S. Vascular factors and metabolic interactions in the
pathogenesis of diabetic neuropathy. Diabetologia.
2001;44:1973–88.
20. Veves A, Akbari CM, Primavera J, Donaghue VM,
Zacharoulis D, Chrzan JS, DeGirolami U, LoGerfo
FW, Freeman R.  Endothelial dysfunction and the
expression of endothelial nitric oxide synthetase in
diabetic neuropathy, vascular disease, and foot ulcer-
ation. Diabetes. 1998;47:457–63.
21. Jude EB, Boulton AJ, Ferguson MW, Appleton I. The
role of nitric oxide synthase isoforms and arginase
in the pathogenesis of diabetic foot ulcers: possible
modulatory effects by transforming growth factor
beta 1. Diabetologia. 1999;42:748–57.
22. Arora S, Smakowski P, Frykberg RG, Simeone LR,
Freeman R, LoGerfo FW, Veves A.  Differences in
foot and forearm skin microcirculation in diabetic
360
patients with and without neuropathy. Diabetes Care.
1998;21:1339–44.
23. Malik RA, Tesfaye S, Thompson SD, Veves A,
Sharma AK, Boulton AJ, Ward JD. Endoneurial local-
isation of microvascular damage in human diabetic
neuropathy. Diabetologia. 1993;36:454–9.
24. Roustit M, Loader J, Deusenbery C, Baltzis D, Veves
A. Endothelial dysfunction as a link between cardio-
vascular risk factors and peripheral neuropathy in
diabetes. J Clin Endocrinol Metab. 2016;101:3401–8.
25. Chapouly C, Yao Q, Vandierdonck S, Larrieu-Lahargue
F, Mariani JN, Gadeau AP, Renault MA.  Impaired
Hedgehog signalling-induced endothelial dysfunction
is sufficient to induce neuropathy: implication in dia-
betes. Cardiovasc Res. 2016;109:217–27.
26. Williams SB, Cusco JA, Roddy MA, Johnstone MT,
Creager MA. Impaired nitric oxide-mediated vasodi-
lation in patients with non-insulin-dependent diabetes
mellitus. J Am Coll Cardiol. 1996;27:567–74.
27. Palmer RM, Ashton DS, Moncada S. Vascular endo-
thelial cells synthesize nitric oxide from L-arginine.
Nature. 1988;333:664–6.
28. Johnstone MT, Creager SJ, Scales KM, Cusco JA, Lee
BK, Creager MA.  Impaired endothelium-dependent
vasodilation in patients with insulin-dependent diabe-
tes mellitus. Circulation. 1993;88:2510–6.
29. Elhadd TA, Kennedy G, Hill A, McLaren M, Newton
RW, Greene SA, Belch JJ.  Abnormal markers of
endothelial cell activation and oxidative stress in chil-
dren, adolescents and young adults with type 1 diabe-
tes with no clinical vascular disease. Diabetes Metab
Res Rev. 1999;15:405–11.
30. Charkoudian N. Mechanisms and modifiers of reflex
induced cutaneous vasodilation and vasoconstriction
in humans. J Appl Physiol. 2010;109:1221–8.
31. Tesfaye S, Harris N, Jakubowski JJ, Mody C, Wilson
RM, Rennie IG, Ward JD.  Impaired blood flow and
arterio-venous shunting in human diabetic ­neuropathy:
a novel technique of nerve photography and fluores-
cein angiography. Diabetologia. 1993;36:1266–74.
32. Flynn MD, Tooke JE.  Diabetic neuropathy and the
microcirculation. Diabet Med. 1995;12:298–301.
33. Rutkove SB, Veves A, Mitsa T, Nie R, Fogerson PM,
Garmirian LP, Nardin RA.  Impaired distal thermo-
regulation in diabetes and diabetic polyneuropathy.
Diabetes Care. 2009;32:671–6.
34. Lefrandt JD, Bosma E, Oomen PH, Hoeven JH, Roon
AM, Smit AJ, Hoogenberg K. Sympathetic mediated
vasomotion and skin capillary permeability in diabetic
patients with peripheral neuropathy. Diabetologia.
2003;46:40–7.
35. Eaton RP, Qualls C, Bicknell J, Sibbitt WL Jr, King
MK, Griffey RH.  Structure-function relationships
within peripheral nerves in diabetic neuropathy: the
hydration hypothesis. Diabetologia. 1996;39:439–46.
36. Hamdy O, Abou-Elenin K, LoGerfo FW, Horton ES,
Veves A.  Contribution of nerve-axon reflex-related
vasodilation to the total skin vasodilation in diabetic
patients with and without neuropathy. Diabetes Care.
2001;24:344–9.
L. Chen and A. Veves
37. Caselli A, Rich J, Hanane T, Uccioli L, Veves
A.  Role of C-nociceptive fibers in the nerve axon
reflex-related vasodilation in diabetes. Neurology.
2003;60:297–300.
38. Caselli A, Spallone V, Marfia GA, Battista C, Pachatz
C, Veves A, Uccioli L.  Validation of the nerve axon
reflex for the assessment of small nerve fibre dysfunc-
tion. J Neurol Neurosurg Psychiatry. 2006;77:927–32.
39. Calcutt NA, Chen P, Hua XY. Effects of diabetes on
tissue content and evoked release of calcitonin gene-­
related peptide-like immunoreactivity from rat sen-
sory nerves. Neurosci Lett. 1998;254:129–32.
40. Yorek MA, Coppey LJ, Gellett JS, Davidson
EP. Sensory nerve innervation of epineurial arterioles
of the sciatic nerve containing calcitonin gene-related
peptide: effect of streptozotocin-induced diabetes.
Exp Diabetes Res. 2004;5:187–93.
41. Arora S, Pomposelli F, LoGerfo FW, Veves
A. Cutaneous microcirculation in the neuropathic dia-
betic foot improves significantly but not completely
after successful lower extremity revascularization. J
Vasc Surg. 2002;35:501–5.
42. Criqui MH, Aboyans V.  Epidemiology of peripheral
artery disease. Circ Res. 2015;116:1509–26.
43. Beckman JA, Creager MA, Libby P.  Diabetes and
atherosclerosis: epidemiology, pathophysiology, and
management. JAMA. 2002;287:2570–81.
44. Reiber GE, Pecoraro RE, Koepsell TD. Risk factors for
amputation in patients with diabetes mellitus. A case-­
control study. Ann Intern Med. 1992;117:97–105.
45. Fowkes FG, Rudan D, Rudan I, Aboyans V,
Denenberg JO, McDermott MM, Norman PE,
Sampson UK, Williams LJ, Mensah GA, Criqui
MH. Comparison of global estimates of prevalence
and risk factors for peripheral artery disease in 2000
and 2010: a systematic review and analysis. Lancet.
2013;382:1329–40.
46. Vadiveloo T, Jeffcoate W, Donnan PT, Colhoun
HC, McGurnaghan S, Wild S, McCrimmon R,
Leese GP.  Scottish Diabetes Research Network
Epidemiology G. Amputation-free survival in 17,353
people at high risk for foot ulceration in diabe-
tes: a national observational study. Diabetologia.
2018;61:2590–7.
47. Dick F, Diehm N, Galimanis A, Husmann M,
Schmidli J, Baumgartner I. Surgical or endovascular
revascularization in patients with critical limb isch-
emia: influence of diabetes mellitus on clinical out-
come. J Vasc Surg. 2007;45:751–61.
48. Jude EB, Oyibo SO, Chalmers N, Boulton
AJ. Peripheral arterial disease in diabetic and nondia-
betic patients: a comparison of severity and outcome.
Diabetes Care. 2001;24:1433–7.
49. American Diabetes A.  Peripheral arterial dis-
ease in people with diabetes. Diabetes Care.
2003;26:3333–41.
50. Armstrong EJ, Waltenberger J, Rogers
JH.  Percutaneous coronary intervention in patients
with diabetes: current concepts and future directions.
J Diabetes Sci Technol. 2014;8:581–9.
Micro- and Macrovascular Disease in Diabetic Neuropathy
51. Van Damme H. Crural or pedal artery revascularisa-
tion for limb salvage: is it justified? Acta Chir Belg.
2004;104:148–57.
52. Behroozian A, Beckman JA.  Microvascular dis-
ease increases amputation in patients with periph-
eral artery disease. Arterioscler Thromb Vasc Biol.
2020;40:534–40.
361
53. Weitz JI, Byrne J, Clagett GP, Farkouh ME, Porter JM,
Sackett DL, Strandness DE Jr, Taylor LM. Diagnosis
and treatment of chronic arterial insufficiency of
the lower extremities: a critical review. Circulation.
1996;94:3026–49.
 The Spinal Cord in Diabetic
Neuropathy

Andrew G. Marshall, Anne Worthington,
and Corinne G. Jolivalt

1 Introduction flow of afferent information from the periphery

to the higher central nervous system, it is increas-
The spinal cord role in diabetic neuropathy has ingly evident that the spinal cord is a complex
been described since early 1900 and reviewed in sensorimotor processing interface. The spinal
detail in the previous edition [1]. In contrast to cord is the first site of integration of sensory input
the vulnerable peripheral nerves, the spinal cord from the periphery and the last site of descending
is exposed to less glycemic stress as a conse- control of sensory and motor systems; therefore,
quence of lower glucose levels in the cerebral disruption of its function may impede appropri-
spinal fluid than in the plasma [2]. However, the ate CNS control systems and contribute to appar-
spinal cord is not protected from diabetes-­ ent peripheral neuropathy [5].
induced injury. Indeed, structural and functional The number of studies on the function and
damage are discernable in the spinal cord of dia- contribution of the spinal cord to diabetes neu-
betic patients with histological evidence of neu- ropathy has more than trippled in the last 15 years
ropathy, radiculopathy, and myelopathy [1] and (from 15  articles/year in 2006 to 45–50/year in
clear atrophy detectable by magnetic resonance the last 5 years for spinal cord and diabetic neu-
imaging (MRI) scans [3, 4]. Structural damages ropathy in PubMed search). The main progress
are also detected in the spinal cord of animal areas in the last 15 years will be developed in this
models of diabetes, although not to the extent of chapter.
structural degeneration seen in human [1]. Rather
than merely acting as a passive conduit for the • Inflammation
• Molecular mechanisms
A. G. Marshall • Spinal disinhibition
Musculoskeletal Biology, Institute of Life Course and • Spinal cord stimulation
Medical Sciences, University of Liverpool,
Liverpool, UK
A. Worthington 2 Inflammation
Division of Diabetes, Endocrinology and
Gastroenterology, School of Medical Sciences,
Faculty of Biology, Medicine and Health, University 2.1 Microglia Activation in Type 1
of Manchester, Manchester, UK Diabetes
C. G. Jolivalt (*)
University of California San Diego, Several mechanisms taking place at the spinal cord
La Jolla, CA, USA level may underline the development and mainte-
e-mail: cjolivalt@ucsd.edu

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 363
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_20
364
nance of diabetic peripheral neuropathy. Although
diabetic neuropathy is not an inflammatory neu-
ropathy per se, among these mechanisms, activa-
tion of microglia and subsequent release of
proinflammatory cytokines and reactive oxygen
intermediates have been suggested as playing an
important role in spinal sensitization. Systemic
changes such as hyperglycemia may be involved
in microglial activation. Under culture conditions,
hyperglycemia can cause morphological changes
indicative of activation of microglia as well as
increased release of inflammatory cytokines [6, 7].
However, hyperglycemia was shown to be neces-
sary but not sufficient to induce hyperalgesia,
which was associated with increase of extracellu-
lar signal-regulated kinase (ERK), cJun N-terminal
kinase (JNK) and p38 phosphorylation in the spi-
nal cord and dorsal root ganglia (DRG), more par-
ticularly in neurons and microglia of streptozotocin
(STZ) diabetic rats [8]. Microglial density
increased along with phosphorylated p38 MAPK
(a marker of microglial activation) in diabetic
rodent spinal cord, correlating with mechanical
hyperalgesia [9, 10], and a cannabinoid receptor
B2 (CB2) agonist attenuated the neuropathic state,
with reduced elevation of microglial density and
phosphorylated p38  in diabetic mice [10].
Similarly, exercise, gabapentin, and lidocaine
attenuated microglia activation, as measured by
p38 phosphorylation in diabetic rodents [11–13].
However, in our group, we have shown that differ-
ences exist between microglial activation observed
in diabetic rats and that observed in nerve-ligated
rats (Dr. Lee-Kubli, unpublished). Unlike nerve
injury that caused a large increase in the total num-
ber of microglia, diabetes was not associated with
a change in microglia number. However, the per-
centage of microglia with retracted processes was
similar in both diabetic and nerve-ligated rats and
was consistent with descriptions of morphological
activation depicted in the literature [8, 14, 15].
Additionally, the mitogen-activated protein
kinase (MAPK) phosphorylation was not detected
in diabetic non-hyperalgesic rats, and N-methyl-
D-aspartate receptor (NMDAR) activation was
required for the MAPK phosphorylation in neu-
rons and microglia and consequent hyperalgesia
in the diabetic hyperalgesic rats [8]. The NMDAR
contributes to central sensitization in the spinal
A. G. Marshall et al.
cord, a phenomenon which comprises various
pathophysiological mechanisms responsible for
neuropathic pain-like signs in animal models.
NMDAR function is modulated by post-transla-
tional modifications including phosphorylation,
and this is proposed to underlie its involvement in
the production of pain hypersensitivity. In particu-
lar, increased phosphorylation of NMDAR1 was
found in neurons and microglial cells in the spinal
cord of STZ rats developing hyperalgesia.
Blockers of MEK (an upstream kinase of ERK)
and of NMDAR suppressed hyperalgesia and
decreased NMDAR phosphorylation, demonstrat-
ing involvement of neuronal and microglial
NMDA and ERK in central sensitization and
mechanical hyperalgesia in diabetes [16].
Microglia involvement in allodynia in type 1
diabetic rats was further supported by adminis-
tration of gabapentin or minocycline, which
attenuated microglial activation (reduction of
expression of the early gene cFos) that correlated
with reduced allodynia in STZ diabetic rats [13,
14, 17–19]. Inhibition of microglial activation by
minocycline restored normal spinal expression of
OX-42, BDNF, and DREAM (a calcium regu-
lated transcriptional repressor) proteins [20].
Increased microglial activation and protein
expression of BDNF and DREAM were not pres-
ent in diabetic rats without painful neuropathy
(hypoalgesia after formalin injection) [21].
Peripheral infiltrated macrophage may also
contribute to the spinal cord inflammation as their
number increased throughout the development of
diabetic neuropathy in STZ mice. Administration
of clodronate liposomes to deplete monocyte
reduced infiltrated macrophages along with alle-
viation of tactile allodynia [22].
Similarly, alteration of astrocytes activation
may be involved in diabetic neuropathic pain.
However, the role of astrogliosis in the hyperal-
gesia detected in type 1 diabetic rodents is not
clear, with report showing decreased GFAP-­
positive astrocytes [23, 24] while other reported
increased activation of astrocytes in addition to
microglial activation in the spinal cord of STZ
diabetic rats [25]. Reduced activation of microg-
lia and astrogliosis were also observed in the spi-
nal cord of the HFD/STZ diabetes model only at
the late stage (after 30 days of diabetes) [26].
The Spinal Cord in Diabetic Neuropathy 365

2.2 Proinflammatory Peptides to reduce microglial activation and cytokines

and Cytokines
Various studies have shown increased proinflam-
matory cytokines levels in diabetic rodent spinal
cord. Here we report only recent studies for
kinins and TNFα, while studies showing efficacy
release are reported in Table 1.
Kinins are defined as pro-inflammatory and
vasoactive peptides, which act through the activa-
tion of two G-protein-coupled receptors denoted
as B1 and B2 [42, 43]. The pro-nociceptive kinin
B1 receptor (B1R) is upregulated on sensory

Table 1  Studies showing efficacy on microglial activation and proinflammatory cytokines

Efficacy on neuropathic
Compound, extract, drug (main Animal pain indices [not assessed
function) Reduction of levels model (NA)] References
Curcumin (antioxidant) TNFα, TNFα receptor 1 STZ Mechanical allodynia [27]
diabetic and thermal hyperalgesia
rats
Puerarin (Chinese remedy) Microglia and astrocytes STZ Mechanical and thermal [28]
expression, NFκB, Il1β, diabetic nociceptive responses
IL6, TNFα rats
Gabapentin (anticonvulsant) Neutrophils, l-selectin STZ Allodynia and thermal [29]
(only one among 29 diabetic hyperalgesia
cytokines) rats
Hydralazine (scavenger of Microglia activation, Il1β, STZ Mechanical allodynia [30]
Acrolein, a lipid peroxidation TNFα diabetic and thermal hyperalgesia
product) rats
Danggui Sini decoction Microglia and astrocytes STZ Allodynia and thermal [31]
activation, Il1β, IL6, diabetic hyperalgesia
TNFα, NFκΒ rats
Ammoxetine (SNRI) Microglia activation, Il1β, STZ Mechanical allodynia [32]
IL6, TNFα diabetic
rats
Dexmedetomidine (sedative, Il1β, TNFα STZ Mechanical and thermal [33]
analgesic) diabetic nociceptive responses
rats
GS-KG9-Ginseng Microglia activation STZ Mechanical and thermal [34]
diabetic nociceptive responses
rats
Gabapentin, Levetiracetam Microglia and astrocytes STZ Allodynia and thermal [35]
(anticonvulsant) expression diabetic hyperalgesia
mice
Bone marrow-derived Microglia and astrocytes STZ Mechanical allodynia [36]
mesenchymal stem cells activation, Il1β, TNFα diabetic and thermal hypoalgesia
mice
Tramadol (opiate) and Il1β STZ Thermal hyperalgesia [37]
gabapentin diabetic
mice
Duloxetine (SSRI) Microglia and astrocytes Alloxan Allodynia and thermal [38]
expression diabetic hyperalgesia
mice
Valproate sodium (antiepileptic) Il1β, TNFα Alloxan Allodynia and thermal [39]
diabetic hyperalgesia
mice
Salidroside (anti-inflammatory Il1β, TNFα ZDF Mechanical allodynia [40]
phenylpropanoid glucoside) diabetic and thermal hyperalgesia
rats
MK801 (NMDAR antagonist) IL6, TNFα Db/db mice Allodynia and thermal [41]
hyperalgesia
366
C-fibers, astrocytes, and microglia in the spinal
cord of STZ diabetic rat. This upregulation was
associated with increased Iba1 staining and
reversed by microglia inhibitors (fluorocitrate
and minocycline), while improving tactile allo-
dynia [44].
Tumor necrosis factor alpha (TNFα) is a pro-
inflammatory cytokine that has been implicated
as a key pain mediator in the development and
maintenance of neuropathic pain conditions.
TNFα was significantly increased in STZ dia-
betic rats that had painful diabetic neuropathy
and to a lesser extent in diabetic rats with non-­
painful neuropathy as determined by their
response to von Frey filaments [45]. The produc-
tion of TNFα in the spinal cord has been shown to
be increased in diabetic mice, and attenuation of
this overproduction by a selective cannabinoid
CB1 receptor antagonist prevented mechanical
allodynia [46]. Systemic or intrathecal adminis-
tration of etanercept, a TNFα inhibitor, alleviated
tactile allodynia in diabetic mice [47]. The
increased TNFα levels in diabetic rat spinal cord
was also reduced by treatment with a Poly(ADP-­
ribose)Polymerase-1 (PARP) inhibitor [48].
TNFα has been shown to be elevated in diabetic
spinal cord, and its modulation improved painful
diabetic neuropathy.
Numerous studies have shown activation of
microglia and release of proinflammatory cyto-
kines (Il1β, TNFα) in STZ rodents that were
reduced by administration of various compounds,
extracts, and drugs (Table 1).
Modulation of spinal cord inflammation in
rodent models of diabetic neuropathy supports
the contribution of inflammation to painful dia-
betic neuropathy and shows it could be a benefi-
cial therapeutical avenue to explore for treatment
of diabetic neuropathy.
2.3 Inflammation in Type 2
Diabetes
In contrast to type 1 diabetic rodent models, spi-
nal astrocytes, but not microglia, were shown to
be activated in type 2 diabetic mice. The astro-
cyte activation correlated with increased expres-
A. G. Marshall et al.
sion of IL1β and increased phosphorylation of
NMDAR in spinal dorsal horn neurons of type 2
diabetic mice [49]. In type 2 diabetic mice, astro-
cytic inhibition attenuated allodynia while
microglia inhibition had no effect [49].
Phosphorylated ERK was increased in the spi-
nal cord and DRG of type 2 diabetic mice (db/db)
[50] and inhibition of MEK (ERK Kinase) atten-
uated mechanical allodynia, thermal hyperalge-
sia, and formalin response in these mice [51].
The enhanced ERK phosphorylation was detected
in projection sensory neurons and was associated
with astrocyte activation, increased NMDAR1
phosphorylation in projection neurons and upreg-
ulation of nitric oxide synthase (neuronal and
inducible) in interneurons and astrocytes peaking
up at 10 weeks of diabetes [50]. Inhibition of
NMDAR, MAPK, or NOS decreased ERK phos-
phorylation, NOS upregulation, and reduced
astrocytosis and GFAP upregulation in db/db
mice spinal cord while inhibiting mechanical
allodynia [50]. Increased ERK activation was
demonstrated in ZDF rats following pressure of
the hindpaw. This was reduced after 7 weeks of
treatment with pioglitazone (a peroxisome
proliferator-­activated receptor γ (PPARγ) ligand)
[52]. However, the treatment reversed hypogly-
cemia, therefore the effect on spinal sensitization
cannot be attribute to pioglitazone alone but par-
tially to a reversal of diabetes.
3 Molecular Mechanisms
3.1 Mitochondrial Function
Mitochondrial respiratory chain activity is
reduced in heart, kidney, skeletal muscle, and
DRG of patients with type 2 diabetes and/or dia-
betic rodents (reviewed in [53]) and contributes
to impaired bioenergetics due to alteration of the
AMP-activated protein kinase (AMPK)/sirtuin
(SIRT)/peroxisome proliferator-activated recep-
tor γ coactivator α (PGC1α) pathway. In contrast
to the peripheral nerves and the brain [53–55],
little is known about mitochondria dynamics, dis-
tribution, and function in the diabetic spinal cord.
Mitochondrial fission was increased, resulting in
The Spinal Cord in Diabetic Neuropathy
the reduction of mitochondria in neuronal axons,
and was accompanied by an abnormal distribu-
tion of mitochondria away from the axons toward
the cell bodies in the spinal cord neurons of type
1 diabetic rats and in ventral spinal cord neurons
in culture in high glucose condition [56]. The
mitochondria also appeared fragmented. Another
recent study showed reduction in mitochondrial
mass in the STZ diabetic rat spinal cord [57, 58].
More studies on spinal cord mitochondria are
required to fully characterize the role of the spi-
nal cord in diabetic neuropathy.
3.2 Synaptic Plasticity
In contrast to the spinal mitochondria, synaptic
plasticity has been extensively studied in the last
15 years.
Imaging of the spinal cord of type 1 diabetic
rats by BOLD functional MRI demonstrated a
decreased blood-oxygen-level-dependent activ-
ity, indicative of changes in synapses between
primary afferents and second order neurons [59].
Synaptic and dendritic alterations were detected
after established diabetes in type 1 diabetic rats
with reduced axonal stabilization and dendritic
structures associated with decreased expression
of ILK-PINCH proteins levels in the spinal cord
[60]. In STZ diabetic rats, alterations in dendritic
spine morphology have been shown in the dorsal
spinal cord and were associated with indices of
neuropathic pain [61]. Morphologic changes
were not observed after 1 week of diabetes when
hyperglycemia was present without evidence of
pain [61]. Additionally, evoked responses to
brush and low force von Frey filaments, but not
higher force static von Frey stimulation, were
recorded from wide dynamic range (WDR) neu-
rons in STZ diabetic rat spinal cord while, in the
genetic BB/Wor rat model, responses to brush
and von Frey stimulation were not affected, cor-
responding to their lack of tactile hypersensitivity
[62].
In neuronal cells, RhoA, a member of the
small molecular G proteins family, Ras super-
family (see review [63]), is involved in the guid-
ance and extension of axons as well as the
367
development and structural plasticity of dendrites
and dendritic spines [64, 65]. Several studies
have suggested that RhoA regulates the stability
of dendritic branches and spines in neurons [64,
66]. High glucose levels can induce the activation
of RhoA through multiple mechanisms mediated
by PKC, cyclic guanosine monophosphatase
(cGMP)-dependent protein kinase G (PKG), and
reactive oxygen species [67]. It was also shown
that RhoA and its downstream kinase, Rho kinase
(ROCK), play important roles in the development
and/or maintenance of chronic pain [68, 69], and
the direct activation of ROCK is involved in dia-
betic painful neuropathy. The expression of
cleaved product of ROCK was increased in rats
fed a high-fructose diet [70]. Furthermore, ROCK
inhibitors have shown efficacy in STZ diabetic
rat with attenuation of mechanical allodynia and
thermal responses, and reduction of WDR neuron
hyperexcitability [61, 71]. Other intracellular
pathways and proteins play a role in the altered
synaptic plasticity in diabetes. APPL1, a neuro-
nal adaptor protein, affects synaptic plasticity,
but its defined role in the pathogenesis of painful
diabetic neuropathy is under investigation.
APPL1 expression was decreased in neurons and
microglia of diabetic rats that presented mechani-
cal and thermal hyperalgesia. Reduced APPL1
was associated with activation of mTor, reduction
of AMPK phosphorylation and regulation of den-
dritic spines. APPL1 deficiency exacerbated the
hyperalgesia in STZ-rats, and restoration of
APPL1 levels ameliorated painful diabetic neu-
ropathy [72].
Sirtuins also play a role in synaptic plasticity
and pain. Sirtuin levels were shown to be
decreased along with enhanced expression of
structural synaptic plasticity protein (PSD95,
GAP 43 and synaptophysin) in spinal dorsal horn
of diabetic neuropathic rats and db/db mice [73,
74]. Activator of Sirtuin 1 relieved pain behavior
and inhibited the enhanced structural synaptic
plasticity in diabetic rats and db/db mice while
­silencing Sirtuin 1-induced pain behavior in nor-
mal rats [73].
Preclinical work in animal models of diabetes
demonstrated alteration of synaptic plasticity at
the spinal cord levels contributing to painful dia-
368
betic neuropathy, which corroborated with human
data, where MRI imaging showed spinal cord
atrophy as an early process only in patients with
clinically detectable diabetic painful neuropathy
[4].
3.3 Receptors and Ion Channels
Pain, and more particularly here, diabetic painful
neuropathy, is linked to aberrant, exaggerated, or
reduced receptors and ion channels activity.
Preclinical models of painful diabetic neuropathy
have allowed studies of the pathological modula-
tion of ion channels at the spinal cord level and
the development of therapies.
3.3.1 Glutamatergic Receptors
mRNA for subunits of glutamatergic NMDA and
AMPA receptors are increased in the spinal cord
of diabetic rodents [75]. More specifically,
NMDA R1 subunit levels were increased in the
spinal dorsal horn of STZ diabetic rats which
were hyperalgesic/allodynic, and magnesium, a
voltage-dependent blocker of NMDA, prevented
the increased of phosphorylated NR1 (activation)
and thermal and tactile allodynia [16, 76]. Similar
results were obtained with the NMDA antagonist
MK-801 [77]. Additionally, higher expression of
phosphorylated spinal NR2B subunit was accom-
panied by tactile allodynia and increased noci-
ceptive response in the formalin test in diabetic
rats and was reduced by intrathecal treatment
with ifenprodil, an atypical non-competitive
NMDA R2B antagonist [78]. Expression of spi-
nal NMDA R2B was reduced along with TRPV1
expression in diabetic mice after treatment with
Ginger extracts and was associated with partial
alleviation of allodynia and thermal hyperalgesia
[79]. Activation of NMDA R was also demon-
strated in prediabetes and type 2 diabetes. Spinal
NMDA R2 expression and activation were
increased in high-fat-diet-fed mice at 24 weeks
when tactile allodynia and thermal hyperalgesia
occurred [80]. Furthermore, inhibition of NMDA
R2B alleviated tactile allodynia but not thermal
hypoalgesia in the prediabetic mice [80]. In type
2 diabetic mice (db/db), NMDA inhibition by
A. G. Marshall et al.
intrathecal MK801 reduced mechanical allodynia
along with reduction of pERK, pAKT, TNFα,
and IL6 levels in the spinal cord [41].
Synaptic calcium-permeable AMPA receptors
are increased in the spinal dorsal neurons of dia-
betic rats and blocking them reduced nociceptive
hypersensitivity [81]. Increased metabotropic
glutamate receptors (mGluR5) in spinal cord of
STZ diabetic rats may also contribute to increased
glutamatergic input and nociceptive transmission
in diabetic neuropathic pain [82].
3.3.2 Adrenergic Receptors
Adrenergic receptors are G-protein-coupled
receptors, with two main groups, α and β, in
which activation can generate analgesic effect or
contribute to chronic neuropathic pain [83]. The
increased amplitude of glutamatergic excitatory
postsynaptic current in spinal cord slides from
STZ diabetic rats was inhibited by a specific
α2-adrenoceptor agonist, which concurrently
alleviated hyperalgesia. The electrophysiological
study suggested an upregulation of
α2-adrenoceptors activity in the spinal cord horn
of STZ diabetic rats and a somewhat contrastive
antinociceptive effect of the α2 adrenoceptor
agonist that resulted in attenuation of glutamater-
gic transmission in the spinal dorsal horn [84]. In
allodynic db/db mice, decreased levels of neuro-
nal α2 adrenergic receptor protein were detected
during the period of mechanical allodynia, which
was alleviated by systemic administration of dex-
medetomidine, a selective α2 adrenergic receptor
agonist [85], possibly via the Wnt10a/β catenin
signaling pathway [33].
3.3.3 Ligand-Gated Channels
Transient receptor potential (TRP) channels
transduce extracellular stimuli into neuronal
responses through influx of calcium [86] and
when dysfunctional may contribute to painful
diabetic neuropathy. Protein expression of TRP
vanilloid 1 (TRPV1) was increased at an early
stage in the spinal cord, but not the DRG, of STZ
rats developing allodynia. The increased TRPV1
levels were detected in small size CGRP neurons
and intrathecal administration of TRPV1 antago-
nists alleviated allodynia [87]. Similarly,
The Spinal Cord in Diabetic Neuropathy
increased expression of TRPV1 receptor in the
spinal cord of STZ diabetic rats was reduced by
treatment with the antidepressant mirtazapine
[88]. Recently, in contrast, TRPV1 and CGRP
expressions were shown to be reduced in the spi-
nal cord of diabetic rats and further decreased
with administration of Ropivacaine, a commonly
used clinical anesthetic, dramatically affecting
NCV of the sciatic nerve [89], suggesting greater
precaution when planning surgery for patients
with diabetes.
Spinal TRP ankyrin 1 (A1) channels, on cen-
tral terminals of primary afferent nerve fibers,
were shown to play an important role in mainte-
nance of mechanical hypersensitivity and con-
tribute to cutaneous neurogenic inflammation
while cutaneous TRPA1 channels contributed to
mechanical hypersensitivity in STZ rats [90, 91].
Methylglyoxal, a reactive glucose metabolite,
was reported to facilitate painful diabetic neu-
ropathy via sensitization of TRPA1-adenylyl
cyclase type 1 (AC1) signaling cascade in type 2
diabetic db/db mice [92].
3.3.4 Voltage-Gated Sodium
Channels
Voltage-gated sodium channels (Nav) play a criti-
cal role in controlling cellular excitability and
have gain interest for their role in painful diabetic
neuropathy particularly Nav 1.3, 1.7, and 1.8
among the 10 sodium channel α subunits
(Nav1.1-­1.9 and NavX).
Intrathecal administration of Nav1.7 and
Nav1.8 antagonists alleviated thermal hyperalge-
sia in diabetic mice but not in control mice,
although changes in the spinal sodium channels
proteins were not different between control and
diabetic mice [93].
3.3.5 Voltage-Gated Calcium
Channels
Voltage-gated calcium channels of L, N, and T
type are widely studied for their role in neuro-
pathic pain, with the T-type channel Cav3.2 par-
ticularly studied in painful diabetic neuropathy,
mainly in DRG sensory neurons [94]. Limited
numbers of studies have indirectly shown a role
of dorsal horn T-channels in neuropathic pain
369
responses with intrathecal administrations of
T-channel blockers rapidly reducing neural excit-
ability and pain responses [95, 96]. Only one
study has demonstrated that specific inhibition of
Cav3.2 T channels in superficial dorsal horn neu-
rons suppressed spontaneous excitatory synaptic
transmission in diabetic rats to a greater extent
than in healthy age-matched animals [97].
Aberrant regulation of nociceptive receptors
and ion channels in the dorsal horn of diabetic
rats contribute to increased pronociception via
ion channels in sensory neurons. Modulation of
spinal receptors and channels may lead to novel
therapies for painful diabetic neuropathy.
3.4 Oxidative Stress
Reactive oxygen species (ROS) are produced by
physiological functions and scavenged by enzy-
matic and non-enzymatic antioxidant system. In
diabetes, chronic hyperglycemia leads to an
imbalance of the oxidative status [98] and
increased oxidative stress occurs in peripheral
and central tissues [99]. Several studies have
demonstrated the role that oxidative stress plays
at the level of the spinal cord to contribute to dia-
betic neuropathic pain, mainly by demonstrating
the beneficial effect of antioxidants.
Poly(ADP-ribose)Polymerase-1 (PARP) is an
nuclear enzyme that cleaves nicotinamide ade-
nine dinucleotide (NAD+) and contributes to
NAD+ deletion and energy failure, impaired sig-
nal transduction, and apoptosis [100]. Enhanced
activation of PARP was demonstrated in diabetic
rat spinal cord, with increased nitrotyrosine lev-
els in neurons, oligodendrocytes, and astrocytes.
A PARP inhibitor prevented sciatic nerve and
spinal PARP activation, nitrotyrosine accumula-
tion, and ameliorated the neuropathy in the STZ
rats [48].
Curcumin is a natural polyphenol with multi-
ple properties, the most recognized being
antioxidant and anti-inflammatory [101].
­
Curcumin was shown to ameliorate diabetic neu-
ropathy by suppression of oxidative stress in
brain and sciatic nerves of diabetic rats [102] and
recently was shown to reduce oxidative stress
370
markers in the spinal cord of STZ diabetic rats
via inhibition of the spinal NADPH oxidases
[103]. Consistently with its anti-inflammatory
properties, curcumin was shown to reduce
expression of TNFα and TNFα receptor 1 in the
dorsal horn of STZ diabetic rats [27].
Increased oxidative stress damage and neuro-
nal activation (Fos) in the spinal cord neurons
were normalized by 10 weeks of preventative
treatment of STZ rats with an antioxidant
(epigallocatechin-­gallate) [104], by mexiletine, a
sodium channel blocker with antioxidative prop-
erties [105] and by treatment with ozone and/or
insulin [106].
Increased oxidative stress has been demon-
strated in diabetic neuropathy, and the ability to
scavenge ROS ameliorates diabetic neuropathy
in rodents.
3.5 Serotonin/Noradrenaline
Reuptake Inhibitors
Among the four currently FDA-approved drugs
to improve painful diabetic polyneuropathy
[107], two are serotonin/noradrenaline reuptake
inhibitors (SSRI/SNRI), duloxetine and fluox-
etine, although fluoxetine is supported by only a
small study. Duloxetine was the first SSRI pre-
scription drug approved by FDA for the manage-
ment of pain associated with diabetic neuropathy
[108] and is the most recommended and pre-
scribed treatment for painful diabetic neuropathy
while studies continued to define its mechanism(s)
of action. Duloxetine exerted its anti-allodynic
effect in diabetic rats predominantly via indirect
activation of 5HT2A receptors in the spinal cord
[109]. Duloxetine is a balanced serotonin/nor-
adrenaline reuptake inhibitor, and its efficacy
may also depend on modulation of noradrenaline,
as it reduced noradrenergic signals in the spinal
cord by inhibiting the norepinephrine transport-
ers [110]. Other mechanisms may play a role in
the analgesic effect of duloxetine. In diabetic
mice, duloxetine treatment for 4 weeks alleviated
allodynia and thermal hyperalgesia, and this was
associated with reduction of microglia and astro-
cytes expression [38] and reduced overexpres-
A. G. Marshall et al.
sion of TLR4-MYd88-dependent pathway [111].
Other antidepressants, milnacipan (SNRI), par-
oxetine and fluvoxamine (SSRI) [112], and
ammoxetine (SNRI) [32] alleviated allodynia in
STZ diabetic rats after intrathecal administration.
Despite the influence of serotonin in pain modu-
lation, SSRIs are less effective than tricyclic anti-
depressants. 5HT2A receptor density was not
affected in the spinal cord of diabetic rats [109,
113], but the PDZ domains of PSD-95 that inter-
acts with 5HT2A receptors was upregulated. This
interaction PDZ/5HT2AR might contribute to the
resistance of SSRI-induced analgesia in painful
diabetic neuropathy, as disrupting this interaction
enhanced fluoxetine antihyperalgesic effect
[113].
Aberrant regulation of nociceptive receptors
and ion channels in the dorsal horn of diabetic
rats contributes to increased pronociception via
ion channels in sensory neurons. Modulation of
spinal receptors and channels may lead to novel
therapies for painful diabetic neuropathy.
4 Spinal Disinhibition
Depending on the physiological and pathophysi-
ological state, processing in the spinal cord dor-
sal horn may dampen down or enhance output
from nociceptive projection neurons. Several
possible mechanisms of spinally mediated
enhancement of ascending nociceptive drive
including wind-up, long-term potentiation, glial
inflammation, altered descending pain modula-
tion, and spinal disinhibition have been impli-
cated in centrally mediated pain. The following
section will focus on the accumulating evidence
suggesting that spinal disinhibition plays an
important role in pain generation in diabetic
neuropathy.
The major inhibitor transmitters in the spinal
cord are γ-aminobutyric acid (GABA) and gly-
cine. GABA acts via GABA-A and GABA-B to
produce inhibition either at pre-synaptic termi-
nals, thereby reducing transmitter release (i.e.,
pre-synaptic inhibition) or on the post-synaptic
dendrites (i.e., post-synaptic inhibition). GABA-­
A-­mediated inhibition is ionotropic resulting
The Spinal Cord in Diabetic Neuropathy 371

from the passage of ions through voltage-gated ioral indices of pain in rodent pain models
channels [114]. The main ion that permeates the [128–131]. A reduction in this tonic GABAergic/
pore of the GABA-A receptor is chloride (Cl−), glycinergic inhibition of ascending nociceptive
the concentration of which is much greater in the transmission, a phenomenon termed disinhibi-
extracellular space. The polarity of the GABA-A tion, is a putative mechanism of central sensitiza-
response is determined largely by the intracellu- tion and could potentially result from loss of
lar Cl− concentration. In the post-developmental inhibitory interneurons or attenuation of inhibi-
nervous system, the low intracellular Cl− concen- tory neurotransmitter storage/release [132, 133].
tration is controlled by the potassium (K+)/chlo- Indeed, a reduction in GABAergic tone in the
ride (Cl−) co-transporter KCC2 which extrudes dorsal horn of the spinal cord has been demon-
intracellular Cl− [115]. Following GABA-A acti- strated in models of peripheral nerve injury [134,
vation, Cl− enters the cell according to its electro- 135]. However, this is not the case in the spinal
chemical concentration gradient, causing cord of diabetic rats exhibiting behavioral indices
inhibitory post-synaptic potentials (IPSP), lead- of pain. In the STZ rat model of type 1 diabetes,
ing to phasic hyperpolarization [114]. It is concentrations of GABA in the spinal cord are
increasingly recognized that subtypes of increased in both basal state and evoked responses
GABA-A also occur at sites other than the syn- in the formalin-induced pain model. In contrast,
apse. These extrasynaptic GABA-A receptors, in the same study, levels of the excitatory neu-
which are activated by GABA derived from glial rotransmitter glutamate are diminished [136].
cells and “circulating” GABA [116], produce Given that spinal cord GABA-A receptor protein
tonic hyperpolarization and consequently tonic levels are not significantly altered in STZ rats
inhibition of spinal cord cells [117, 118] and pri- [137], the paradoxical co-existence of increased
mary afferents [119]. GABA and increased nociceptive drive suggest
GABA-B-mediated inhibition is, in contrast, that there is impaired/altered post-synaptic
metabotropic. The GABA-B receptor is G-protein responsiveness to GABA.
coupled, and the binding of GABA triggers a sec- Alterations in the post-synaptic responsive-
ond messenger cascade [120, 121]. In the post-­ ness to GABA were first described in a peripheral
synaptic cell, this results in activation of nerve injury model of pain in nondiabetic rats
G-protein-activated inwardly rectifying K+ [138]. Increasing evidence indicates that the dys-
(GIRK) channels which are highly co-localized functional spinal inhibitory processing seen in
with GABA-B receptors [120–122]. The result- STZ rats results from a similar mechanism.
ing increase in K+ conductance promotes K+ Allodynia and other behavioral indices of pain in
efflux from the cell and hyperpolarization. In STZ rats are associated with downregulation of
contrast to the fast inhibitory signaling mediated post-synaptic KCC2  in the dorsal, although not
by GABA-A receptors, GABA-B receptor-­ ventral, spinal cord [137, 139]. This causes
mediated hyperpolarization is slow and pro- GABA, acting via spinal GABA-A receptors, to
longed [121]. become pro-nociceptive rather than inhibitory.
Inhibitory interneurons in the spinal cord are Accordingly, spinal GABA-A blockade, which
intimately associated with nociceptive circuits under normal circumstances is pro-nociceptive,
[123]. The effect of this inhibition is to reduce the reverses allodynia in STZ rats [137]. The mecha-
output of projection neurons in response to nox- nism, like that seen in the peripheral nerve injury
ious peripheral stimulation. Accordingly, in nor- model, also involves BDNF.  Spinal administra-
mal rodents, spinal application of GABA or tion of BDNF in normal rats phenocopies the
glycine antagonists result in behavioral indices of findings in STZ rats with the emergence of allo-
hypersensitivity and pain to both noxious and dynia and reduced spinal KCC2 expression
innocuous tactile stimulation [124–127]. [139]. Furthermore, sequestration of spinal
Conversely, enhancement of spinal GABA ele- BDNF in STZ rats reverses allodynia [139]. The
vates nociceptive thresholds and reverses behav- source of the BDNF in STZ rats is currently not
372
clear, although, interestingly, BDNF expression
is elevated in the dorsal root ganglia of STZ rats,
raising the possibility that it may derive from the
primary afferent fibers [140].
Inhibitory function in the spinal cord can be
assessed using rate-dependent depression (RDD)
of the Hoffmann-reflex (H-reflex) [141]. The
H-reflex is a trans-spinal reflex that is elicited by
electrical stimulation of a mixed motor and sen-
sory nerve trunk. The H-reflex can be measured
in mammalian species including rodents and
humans using a simple modification of traditional
nerve conduction studies [142–145]. When a sec-
ond H-wave is elicited with an inter-stimulus
interval less than 2 s following the initial stimula-
tion, the H-wave amplitude decreases relative to
the first response. The measure of the change in
amplitude is referred to as H-reflex RDD—also
known as paired-pulse depression or frequency-­
dependent depression [146–151].
The plausibility of RDD as a mechanistic
biomarker of loss of GABAergic inhibition due
to GABA reversal, spinal disinhibition, and
resultant pain-related phenomena in painful
diabetic neuropathy has been demonstrated in a
number of studies in rodents. STZ rats exhibit-
ing mechanical hypersensitivity/allodynia and
reduced KCC2 in the dorsal horn show attenua-
tion of RDD [137]. Furthermore, spinal deliv-
ery of either the KCC2 blocker DIOA or BDNF
to control rats, interventions that mechanisti-
cally recapitulate the spinal disinhibition phe-
notype in STZ rats, are also associated with
impaired RDD [137, 139]. In keeping with the
underlying mechanism of GABA reversal, both
the behavioral indices of pain and loss of RDD
in all these models are reversed by spinal
administration of GABA-A antagonists
[137]—a pharmacological intervention that
both induces hypersensitivity and impairs RDD
in control rats. RDD and allodynia in STZ rats
are also reversed by the carbonic anhydrase
inhibitor acetazolamide, which mitigates the
effects of reduced KCC2 and Cl− dysregulation
by decreasing bicarbonate efflux through
GABA-A receptors [152].
Spinal disinhibition and loss of RDD in pain-
ful diabetic neuropathy are not just dependent
A. G. Marshall et al.
on synaptic GABA-A.  For example, altered
responsiveness of the α5 subunit-containing
GABA-A receptor is also implicated in allo-
dynia and impaired RDD in STZ rats [153].
Normal levels of RDD can also occur in the
presence of GABA-A blockade. As we have
already discussed, spinal disinhibition and
impaired RDD are reversed by GABA-A block-
ing drugs in normal rats treated with KCC2
antagonist DOIA or BDNF as well as in diabetic
rats [137]. This implies that other inhibitory
mechanisms can also contribute to RDD in the
spinal cord of STZ rats. Further evidence for
this is revealed in studies investigating the time
course of RDD and spinal disinhibition in STZ
rats which indicate that the development of spi-
nal disinhibition in diabetes is a dynamic time-
dependent process involving GABA-A- and
GABA-B-mediated responses [152]. Impaired
RDD due to GABA-A reversal is initially com-
pensated by GABA-B mediated inhibition.
However, over time, due to increasing GABA-A
reversal or attenuation of GABA-B inhibitory
function, impaired RDD and the full spinal dis-
inhibition phenotype become evident. The role
of glycinergic inhibition in spinal disinhibition
in diabetic rats has not been studied extensively
although glycine release in the spinal cord of
STZ rats are mildly reduced [136].
The disclosure of spinal disinhibition as a
putative pain mechanism in STZ rats raises the
question as to whether similar mechanisms occur
in clinical populations with painful diabetic neu-
ropathy. This is of potential clinical importance
as it may lead to mechanistically targeted drug
discovery or individualized therapy in patient
populations. However, this would require identi-
fication of populations or sub-populations of
patients with painful diabetic neuropathy in
whom spinal disinhibition is a major pain mecha-
nism. Currently it is not possible to measure lev-
els of dorsal horn GABA or KCC2  in humans.
However, RDD is conserved across mammalian
species including humans raising the possibility
that results obtained in rodents may be translated
to diabetic patients.
To assess the translational potential of the
preclinical findings, we measured the magni-
The Spinal Cord in Diabetic Neuropathy
tude of H-reflex RDD in patients with type 1
diabetes and painful or painless neuropathy
[154]. In a subset of patients with painful dia-
betic neuropathy, there was loss of RDD com-
pared to both healthy controls and patients with
painless diabetic neuropathy. Importantly, the
impairment of RDD was independent of mea-
sures of both large and small fiber neuropathy,
indicating that is not merely a reflection of
severity of neuropathy. Loss of RDD was also
independent of glycemic control. These findings
support the hypothesis that impaired RDD may
serve as clinical biomarker in a subset of patients
where pain arises primarily from spinal disinhi-
bition. We have also recently demonstrated that
impaired RDD is also seen in patients with type
2 diabetes and painful neuropathy (unpublished
observations). Like the findings in type 1 diabe-
tes, not all subjects with type 2 diabetes and
neuropathic pain demonstrated impairment of
RDD. Approximately 60% of patients with dia-
betes will develop neuropathy, 30% of those
with neuropathy will develop neuropathic pain
and, from our exploratory study, 40% of those
will show RDD deficits. In contrast, diabetic
rodents exhibit much more homogeneous neu-
ropathy, neuropathic pain, and impaired RDD
phenotypes [141]. While this may reflect a more
complex etiopathogenesis of painful diabetic
neuropathy in humans, it is also plausible that
this heterogeneity can be used to enable defini-
tion of abnormal values and predict therapeutic
response to medications that target spinal inhi-
bition, one of these drugs being duloxetine. The
selective serotonin-norepinephrine re-uptake
inhibitor duloxetine, used to treat painful dia-
betic neuropathy [155], alleviates the tactile
allodynia and also restores RDD in STZ rats in
a spinal 5HT2A receptor-dependent manner
[109, 154], and diabetic patients with impaired
spinal disinhibition seemed to find duloxetine to
be an efficacious treatment for their neuropathic
pain [154].
To summarize, there is accumulating preclini-
cal evidence that loss of RDD and indices of neu-
ropathic pain share a common pathogenic
mechanism involving spinal KCC2 depletion and
disinhibition caused by inversion of GABA-A
373
receptor function. Although the exact neural cir-
cuitry has yet to be defined, both RDD and
behavioral indices of neuropathic pain also
exhibit common responses to spinally acting
analgesics. RDD status, which is easily measur-
able in clinical populations, may be a viable bio-
marker for identifying the dominant generator
site in individual patients with painful diabetic
neuropathy and for predicting efficacy of thera-
peutic strategies that alleviate spinal
disinhibition.
5 Spinal Cord Stimulation
Spinal cord stimulation (SCS) is proposed to
relieve chronic intractable pain by stimulating
nerve fibers in the spinal cord. The resulting
impulses in the fibers may inhibit the conduction
of pain signals to the brain, according to the pain
gate theory proposed by Melzack and Wall [156]
and block the sensation of pain. In general, SCS
is part of an overall treatment strategy and is used
only after the more conservative treatments have
failed (review in [157]). SCS is an effective ther-
apy for different chronic painful conditions that
has gained interest for painful peripheral diabetic
neuropathy within the last 15 years. In addition to
the pain gate theory, several other mechanisms
have been proposed for the beneficial effect of
SCS in diabetes (review in [158]). SCS may pro-
duce peripheral vasodilatation via stimulation of
the sympathetic system and release of neuro-
trophic factors [159, 160] and may involve small
diameter nociceptive afferents expressing TRPV1
receptor which stimulation induces release of
calcitonin gene-related peptide (CGRP), nitric
oxide [161] and activation of ERK [162].
Vasodilatation was observed after SCS in STZ
diabetic rats; however, hyperglycemic conditions
attenuated the blood flow responses to SCS [163],
possibly due to the decreased CGRP levels [164,
165], or reduced TRPV1 containing C fibers
[161]. Another mechanism may involve altera-
tion of spinal neurochemistry, such as increased
GABA and acetylcholine after SCS resulting in
the activation of inhibitory interneurons with the
reduction of allodynia in neuropathic pain mod-
374
els [166, 167]. This plausible mechanism has not
been studied in diabetic models and may not
apply to efficacy of SCS in diabetes. GABA has
been shown to be excitatory in diabetic rats due
to impairment in spinal KCC2, resulting in spinal
inhibitory dysfunction (see Sect. 4 and [136, 137,
154]).
Mechanisms by which SCS improves neuro-
pathic pain, particularly in diabetes, are not
clearly understood. However, studies in diabetic
patients have shown efficacy. Neuropathic pain
was significantly reduced after 6 months of SCS
in diabetic patients with chronic pain in their
lower limbs that was not responsive to conven-
tional treatment, and this was not associated with
changes in microcirculatory perfusion [168].
SCS showed pain relief of more than 50% in
50–77% of patients [169, 170] and up to 5 years
[171].
Optimization of SCS parameters has shown
improvement over conventional SCS (30–
120  Hz). One of these optimization is high-­
frequency SCS (150–500  Hz). In a preclinical
rodent model of type 1 diabetes, alleviation of
mechanical hypersensitivity was independent of
the frequency of stimulation applied [172].
However, it was shown that high-frequency SCS
demonstrated delayed effect after cessation of
SCS on mechanical allodynia in STZ diabetic
rats [173].
Recently, another paradigm of stimulation has
been developed, burst SCS (5 pulses at 500 Hz,
delivered 40  times/s). Burst SCS reduced pain
further than tonic stimulation (30–120 Hz) with
reduced local paresthesia [174]. About 58% of
the patients experienced significant additional
pain reduction [175].
SCS has improved in the last 15 years with
high frequency and burst protocols applied to
diabetes-induced neuropathic pain not respond-
ing to current treatments. However, little is
known in diabetic conditions and the mecha-
nisms by which SCS is effective in diabetic
patients remain unknown. Further studies are
needed to understand how SCS works in diabetes
and to predict which patients may benefit from
SCS, possibly patients with evidence of spinal
disinhibition detected by RDD.
A. G. Marshall et al.
6 Summary/Conclusions
It is clear, particularly from the preclinical
rodent studies, that diabetes is a multifactorial
disease. Many mechanisms take place in the
spinal cord, contributing to neuropathy and
painful neuropathy that can be alleviated in ani-
mal models by a multitude of drugs. However,
little has been translated to human with many
failed clinical trials. Nevertheless, progress has
been made with spinal cord stimulation to ben-
efit intractable pain, four FDA-approved drugs
are now available, and a new method of detec-
tion of pain generation site may facilitate physi-
cians’ prescription ability of the current drugs.
The detection of pain generation site by RDD
may also enhance clinical trial design and
patient selection for central versus peripheral
effectors and therefore contribute to more suc-
cessful clinical trials.
References
1. Mizisin AP, Jolivalt CG, Calcutt NA (2007) Spinal
cord: structure and function in diabetes. In: Veves A,
Malik R (eds) Contemporary diabetes: diabetic neu-
ropathy: clinical management, 2nd edn
2. Calcutt NA, Freshwater JD, O'Brien JS.  Protection
of sensory function and antihyperalgesic proper-
ties of a prosaposin-derived peptide in diabetic rats.
Anesthesiology. 2000;93:1271–8.
3. Eaton SE, Harris ND, Rajbhandari SM, et al. Spinal-­
cord involvement in diabetic peripheral neuropathy.
Lancet. 2001;358:35–6.
4. Selvarajah D, Wilkinson ID, Emery CJ, et al. Early
involvement of the spinal cord in diabetic peripheral
neuropathy. Diabetes Care. 2006;29:2664–9.
5. Tesfaye S, Selvarajah D, Gandhi R, et  al. Diabetic
peripheral neuropathy may not be as its name sug-
gests: evidence from magnetic resonance imaging.
Pain. 2016;157(Suppl 1):S72–80.
6. Quan Y, Du J, Wang X.  High glucose stimu-
lates GRO secretion from rat microglia via ROS,
PKC, and NF-kappaB pathways. J Neurosci Res.
2007;85:3150–9.
7. Quan Y, Jiang CT, Xue B, Zhu SG, Wang X. High
glucose stimulates TNFalpha and MCP-1 expression
in rat microglia via ROS and NF-kappaB pathways.
Acta Pharmacol Sin. 2011;32:188–93.
8. Daulhac L, Mallet C, Courteix C, et  al. Diabetes-­
induced mechanical hyperalgesia involves
spinal mitogen-activated protein kinase activa-
tion in neurons and microglia via N-methyl-D-­
The Spinal Cord in Diabetic Neuropathy
aspartate-dependent mechanisms. Mol Pharmacol.
2006;70:1246–54.
9. Cheng KI, Wang HC, Chuang YT, et  al. Persistent
mechanical allodynia positively correlates with
an increase in activated microglia and increased
P-p38 mitogen-activated protein kinase activation
in streptozotocin-induced diabetic rats. Eur J Pain.
2014;18:162–73.
10. Toth CC, Jedrzejewski NM, Ellis CL, Frey WH 2nd.
Cannabinoid-mediated modulation of neuropathic
pain and microglial accumulation in a model of
murine type I diabetic peripheral neuropathic pain.
Mol Pain. 2010;6:16.
11. Suzuki N, Hasegawa-Moriyama M, Takahashi Y,
Kamikubo Y, Sakurai T, Inada E. Lidocaine attenu-
ates the development of diabetic-induced tactile
allodynia by inhibiting microglial activation. Anesth
Analg. 2011;113:941–6.
12. Thakur V, Gonzalez M, Pennington K, Nargis S,
Chattopadhyay M. Effect of exercise on neurogenic
inflammation in spinal cord of type 1 diabetic rats.
Brain Res. 2016;1642:87–94.
13. Wodarski R, Clark AK, Grist J, Marchand F,
Malcangio M. Gabapentin reverses microglial acti-
vation in the spinal cord of streptozotocin-induced
diabetic rats. Eur J Pain. 2009;13:807–11.
14. Morgado C, Pereira-Terra P, Cruz CD, Tavares
I.  Minocycline completely reverses mechanical
hyperalgesia in diabetic rats through microglia-­
induced changes in the expression of the potassium
chloride co-transporter 2 (KCC2) at the spinal cord.
Diabetes Obes Metab. 2011;13:150–9.
15. Tsuda M, Ueno H, Kataoka A, Tozaki-Saitoh H,
Inoue K.  Activation of dorsal horn microglia con-
tributes to diabetes-induced tactile allodynia via
extracellular signal-regulated protein kinase signal-
ing. Glia. 2008;56:378–86.
16. Daulhac L, Maffre V, Mallet C, et al. Phosphorylation
of spinal N-methyl-d-aspartate receptor NR1 sub-
units by extracellular signal-regulated kinase in
dorsal horn neurons and microglia contributes to
diabetes-induced painful neuropathy. Eur J Pain.
2011;15:169.e161–12.
17. Johnson MS, Ryals JM, Wright DE.  Diabetes-­
induced chemogenic hypoalgesia is paralleled by
attenuated stimulus-induced fos expression in the
spinal cord of diabetic mice. J Pain. 2007;8:637–49.
18. Morgado C, Terra PP, Tavares I.  Neuronal hyper-
activity at the spinal cord and periaqueductal grey
during painful diabetic neuropathy: effects of gaba-
pentin. Eur J Pain. 2010;14:693–9.
19. Pabreja K, Dua K, Sharma S, Padi SS, Kulkarni
SK. Minocycline attenuates the development of dia-
betic neuropathic pain: possible anti-inflammatory
and anti-oxidant mechanisms. Eur J Pharmacol.
2011;661:15–21.
20. Ismail CAN, Suppian R, Aziz CBA, Long
I.  Minocycline attenuates the development of dia-
betic neuropathy by modulating DREAM and BDNF
375
protein expression in rat spinal cord. J Diabetes
Metab Disord. 2019;18:181–90.
21. Ismail CAN, Suppian R, Ab Aziz CB, Long
I. Expressions of spinal microglia activation, BDNF,
and DREAM proteins correlated with formalin-­
induced nociceptive responses in painful and
painless diabetic neuropathy rats. Neuropeptides.
2020;79:102003.
22. Sun JJ, Tang L, Zhao XP, Xu JM, Xiao Y, Li
H.  Infiltration of blood-derived macrophages con-
tributes to the development of diabetic neuropathy.
J Immunol Res. 2019;2019:7597382.
23. Afsari ZH, Renno WM, Abd-El-Basset E. Alteration
of glial fibrillary acidic proteins immunoreactivity in
astrocytes of the spinal cord diabetic rats. Anat Rec
(Hoboken). 2008;291:390–9.
24. Renno WM, Alkhalaf M, Afsari Z, Abd-El-Basset
E, Mousa A.  Consumption of green tea alters glial
fibriliary acidic protein immunoreactivity in the
spinal cord astrocytes of STZ-diabetic rats. Nutr
Neurosci. 2008;11:32–40.
25. Kim SH, Kwon JK, Kwon YB.  Pain modality and
spinal glia expression by streptozotocin induced dia-
betic peripheral neuropathy in rats. Lab Anim Res.
2012;28:131–6.
26. Byrne FM, Cheetham S, Vickers S, Chapman
V. Characterisation of pain responses in the high fat
diet/streptozotocin model of diabetes and the analge-
sic effects of antidiabetic treatments. J Diabetes Res.
2015;2015:752481.
27. Li Y, Zhang Y, Liu DB, Liu HY, Hou WG, Dong
YS.  Curcumin attenuates diabetic neuropathic pain
by downregulating TNF-alpha in a rat model. Int J
Med Sci. 2013;10:377–81.
28. Liu M, Liao K, Yu C, Li X, Liu S, Yang S. Puerarin
alleviates neuropathic pain by inhibiting neuro-
inflammation in spinal cord. Mediators Inflamm.
2014;2014:485927.
29. Newton VL, Guck JD, Cotter MA, Cameron NE,
Gardiner NJ.  Neutrophils infiltrate the spinal cord
parenchyma of rats with experimental diabetic neu-
ropathy. J Diabetes Res. 2017;2017:4729284.
30. Yao L, Wu YT, Tian GX, Xia CQ, Zhang F, Zhang
W.  Acrolein scavenger hydralazine prevents
streptozotocin-­ induced painful diabetic neuropa-
thy and spinal neuroinflammation in rats. Anat Rec
(Hoboken). 2017;300:1858–64.
31. Liu M, Qiang QH, Ling Q, et al. Effects of Danggui
Sini decoction on neuropathic pain: experimental
studies and clinical pharmacological significance of
inhibiting glial activation and proinflammatory cyto-
kines in the spinal cord. Int J Clin Pharmacol Ther.
2017;55:453–64.
32. Zhang TT, Xue R, Fan SY, et al. Ammoxetine atten-
uates diabetic neuropathic pain through inhibiting
microglial activation and neuroinflammation in the
spinal cord. J Neuroinflammation. 2018;15:176.
33. Zhong JM, Lu YC, Zhang J.  Dexmedetomidine
reduces diabetic neuropathy pain in rats through the
376
Wnt 10a/beta-catenin signaling pathway. Biomed
Res Int. 2018;2018:9043628.
34. Lee JY, Choi HY, Park CS, et al. GS-KG9 amelio-
rates diabetic neuropathic pain induced by strepto-
zotocin in rats. J Ginseng Res. 2019;43:58–67.
35. Reda HM, Zaitone SA, Moustafa YM.  Effect of
levetiracetam versus gabapentin on peripheral neu-
ropathy and sciatic degeneration in streptozotocin-­
diabetic mice: influence on spinal microglia and
astrocytes. Eur J Pharmacol. 2016;771:162–72.
36. Evangelista AF, Vannier-Santos MA, de Assis Silva
GS, et al. Bone marrow-derived mesenchymal stem/
stromal cells reverse the sensorial diabetic neuropa-
thy via modulation of spinal neuroinflammatory cas-
cades. J Neuroinflammation. 2018;15:189.
37. Miranda HF, Poblete P, Sierralta F, Noriega V,
Prieto JC, Zepeda RJ.  Interleukin-1beta in syner-
gism gabapentin with tramadol in murine model
of diabetic neuropathy. Inflammopharmacology.
2019;27:151–5.
38. Tawfik MK, Helmy SA, Badran DI, Zaitone
SA. Neuroprotective effect of duloxetine in a mouse
model of diabetic neuropathy: role of glia suppress-
ing mechanisms. Life Sci. 2018;205:113–24.
39. Elsherbiny NM, Ahmed E, Kader GA, et  al.
Inhibitory effect of valproate sodium on pain behav-
ior in diabetic mice involves suppression of spinal
histone deacetylase 1 and inflammatory mediators.
Int Immunopharmacol. 2019;70:16–27.
40. Ni GL, Cui R, Shao AM, Wu ZM. Salidroside ame-
liorates diabetic neuropathic pain in rats by inhibiting
neuroinflammation. J Mol Neurosci. 2017;63:9–16.
41. Liu S, Liu X, Xiong H, et  al. CXCL13/CXCR5
signaling contributes to diabetes-induced tactile
allodynia via activating pERK, pSTAT3, pAKT
pathways and pro-inflammatory cytokines produc-
tion in the spinal cord of male mice. Brain Behav
Immun. 2019;80:711–24.
42. Leeb-Lundberg LM, Marceau F, Muller-Esterl W,
Pettibone DJ, Zuraw BL.  International union of
pharmacology. XLV.  Classification of the kinin
receptor family: from molecular mechanisms to
pathophysiological consequences. Pharmacol Rev.
2005;57:27–77.
43. Regoli D, Nsa Allogho S, Rizzi A, Gobeil
FJ. Bradykinin receptors and their antagonists. Eur
J Pharmacol. 1998;348:1–10.
44. Talbot S, Chahmi E, Dias JP, Couture R.  Key role
for spinal dorsal horn microglial kinin B1 receptor in
early diabetic pain neuropathy. J Neuroinflammation.
2010;7:36.
45. Ismail CAN, Aziz CBA, Suppian R, Long
I. Imbalanced oxidative stress and pro-inflammatory
markers differentiate the development of diabetic
neuropathy variants in streptozotocin-induced dia-
betic rats. J Diabetes Metab Disord. 2018;17:129–36.
46. Comelli F, Bettoni I, Colombo A, Fumagalli P,
Giagnoni G, Costa B.  Rimonabant, a cannabinoid
CB1 receptor antagonist, attenuates mechani-
cal allodynia and counteracts oxidative stress and
A. G. Marshall et al.
nerve growth factor deficit in diabetic mice. Eur J
Pharmacol. 2010;637:62–9.
47. Dogrul A, Gul H, Yesilyurt O, Ulas UH, Yildiz
O.  Systemic and spinal administration of etaner-
cept, a tumor necrosis factor alpha inhibitor, blocks
tactile allodynia in diabetic mice. Acta Diabetol.
2011;48:135–42.
48. Drel VR, Lupachyk S, Shevalye H, et al. New thera-
peutic and biomarker discovery for peripheral dia-
betic neuropathy: PARP inhibitor, nitrotyrosine,
and tumor necrosis factor-{alpha}. Endocrinology.
2010;151:2547–55.
49. Liao YH, Zhang GH, Jia D, et al. Spinal astrocytic
activation contributes to mechanical allodynia
in a mouse model of type 2 diabetes. Brain Res.
2011;1368:324–35.
50. Dauch JR, Yanik BM, Hsieh W, Oh SS, Cheng
HT.  Neuron-astrocyte signaling network in spinal
cord dorsal horn mediates painful neuropathy of type
2 diabetes. Glia. 2012;60:1301–15.
51. Xu X, Chen H, Ling BY, Xu L, Cao H, Zhang
YQ.  Extracellular signal-regulated protein kinase
activation in spinal cord contributes to pain hyper-
sensitivity in a mouse model of type 2 diabetes.
Neurosci Bull. 2014;30:53–66.
52. Griggs RB, Donahue RR, Adkins BG, Anderson
KL, Thibault O, Taylor BK.  Pioglitazone inhibits
the development of hyperalgesia and sensitization of
spinal nociresponsive neurons in type 2 diabetes. J
Pain. 2016;17:359–73.
53. Fernyhough P.  Mitochondrial dysfunction in dia-
betic neuropathy: a series of unfortunate metabolic
events. Curr Diab Rep. 2015;15:89.
54. Hamid HS, Mervak CM, Munch AE, et  al.
Hyperglycemia- and neuropathy-induced changes
in mitochondria within sensory nerves. Ann Clin
Transl Neurol. 2014;1:799–812.
55. Vincent AM, Edwards JL, McLean LL, et  al.
Mitochondrial biogenesis and fission in axons in cell
culture and animal models of diabetic neuropathy.
Acta Neuropathol. 2010;120:477–89.
56. Chen L, Huang J, Li XC, et al. High-glucose induced
mitochondrial dynamics disorder of spinal cord
neurons in diabetic rats and its effect on mitochon-
drial spatial distribution. Spine (Phila Pa 1976).
2019;44:E715–22.
57. Rashedinia M, Alimohammadi M, Shalfroushan
N, et  al. Neuroprotective effect of syringic acid
by modulation of oxidative stress and mito-
chondrial mass in diabetic rats. Biomed Res Int.
2020;2020:8297984.
58. Rashedinia M, Khoshnoud MJ, Fahlyan BK,
Hashemi SS, Alimohammadi M, Sabahi Z. Syringic
acid: a potential natural compound for the manage-
ment of renal oxidative stress and mitochondrial bio-
genesis in diabetic rats. Curr Drug Discov Technol.
2021;18(3):405–13.
59. Malisza KL, Jones C, Gruwel ML, Foreman D,
Fernyhough P, Calcutt NA.  Functional magnetic
resonance imaging of the spinal cord during sensory
The Spinal Cord in Diabetic Neuropathy
stimulation in diabetic rats. J Magn Reson Imaging.
2009;30:271–6.
60. Jiang Y, Mizisin AP, Rearden A, Jolivalt CG.
Diabetes induces changes in ILK, PINCH and com-
ponents of related pathways in the spinal cord of
rats. Brain Res. 2010;1332:100–9.
61. Tan AM, Samad OA, Fischer TZ, Zhao P, Persson
AK, Waxman SG.  Maladaptive dendritic spine
remodeling contributes to diabetic neuropathic pain.
J Neurosci. 2012;32:6795–807.
62. Schuelert N, Gorodetskaya N, Just S, Doods H,
Corradini L.  Electrophysiological characteriza-
tion of spinal neurons in different models of diabe-
tes type 1- and type 2-induced neuropathy in rats.
Neuroscience. 2015;291:146–54.
63. Takai Y, Sasaki T, Matozaki T. Small GTP-binding
proteins. Physiol Rev. 2001;81:153–208.
64. Nakayama AY, Harms MB, Luo L. Small GTPases
Rac and Rho in the maintenance of dendritic spines
and branches in hippocampal pyramidal neurons. J
Neurosci. 2000;20:5329–38.
65. Threadgill R, Bobb K, Ghosh A.  Regulation of
dendritic growth and remodeling by Rho, Rac, and
Cdc42. Neuron. 1997;19:625–34.
66. Billuart P, Winter CG, Maresh A, Zhao X, Luo
L. Regulating axon branch stability: the role of p190
RhoGAP in repressing a retraction signaling path-
way. Cell. 2001;107:195–207.
67. Ohsawa M, Kamei J. RhoA/Rho kinase signaling in
the spinal cord and diabetic painful neuropathy. Eur
J Pharmacol. 2010;644:1–4.
68. Inoue M, Rashid MH, Fujita R, Contos JJ, Chun
J, Ueda H.  Initiation of neuropathic pain requires
lysophosphatidic acid receptor signaling. Nat Med.
2004;10:712–8.
69. Tatsumi S, Mabuchi T, Katano T, et al. Involvement
of Rho-kinase in inflammatory and neuropathic pain
through phosphorylation of myristoylated alanine-­
rich C-kinase substrate (MARCKS). Neuroscience.
2005;131:491–8.
70. Mustafa S, Vasudevan H, Yuen VG, McNeill
JH. Renal expression of arachidonic acid metaboliz-
ing enzymes and RhoA/Rho kinases in fructose insu-
lin resistant hypertensive rats. Mol Cell Biochem.
2010;333:203–9.
71. Ohsawa M, Aasato M, Hayashi SS, Kamei J. RhoA/
Rho kinase pathway contributes to the pathogen-
esis of thermal hyperalgesia in diabetic mice. Pain.
2011;152:114–22.
72. He WY, Zhang B, Zhao WC, et  al. mTOR activa-
tion due to APPL1 deficiency exacerbates hyperal-
gesia via Rab5/Akt and AMPK signaling pathway
in streptozocin-induced diabetic rats. Mol Pain.
2019;15:1744806919880643.
73. Zhang Z, Ding X, Zhou Z, et  al. Sirtuin 1 allevi-
ates diabetic neuropathic pain by regulating syn-
aptic plasticity of spinal dorsal horn neurons. Pain.
2019;160:1082–92.
74. Zhou CH, Zhang MX, Zhou SS, et al. SIRT1 attenu-
ates neuropathic pain by epigenetic regulation of
377
mGluR1/5 expressions in type 2 diabetic rats. Pain.
2017;158:130–9.
75. Tomiyama M, Furusawa K, Kamijo M, Kimura T,
Matsunaga M, Baba M.  Upregulation of mRNAs
coding for AMPA and NMDA receptor subunits and
metabotropic glutamate receptors in the dorsal horn
of the spinal cord in a rat model of diabetes mellitus.
Brain Res Mol Brain Res. 2005;136:275–81.
76. Rondon LJ, Privat AM, Daulhac L, et al. Magnesium
attenuates chronic hypersensitivity and spinal
cord NMDA receptor phosphorylation in a rat
model of diabetic neuropathic pain. J Physiol.
2010;588:4205–15.
77. Wang J, Sun Z, Wang Y, Wang H, Guo Y. The role
and mechanism of glutamic NMDA receptor in the
mechanical hyperalgesia in diabetic rats. Neurol
Res. 2017;39:1006–13.
78. Ismail CAN, Suppian R, Abd Aziz CB, Haris K, Long
I. Increased nociceptive responses in streptozotocin-­
induced diabetic rats and the related expression
of spinal NR2B subunit of N-methyl-D-aspartate
receptors. Diabetes Metab J. 2019;43:222–35.
79. Fajrin FA, Nugroho AE, Nurrochmad A, Susilowati
R.  Ginger extract and its compound, 6-shogaol,
attenuates painful diabetic neuropathy in mice via
reducing TRPV1 and NMDAR2B expressions in the
spinal cord. J Ethnopharmacol. 2020;249:112396.
80. Suo M, Wang P, Zhang M.  Role of Fyn-mediated
NMDA receptor function in prediabetic neuropathy
in mice. J Neurophysiol. 2016;116:448–55.
81. Chen SR, Zhang J, Chen H, Pan HL. Streptozotocin-­
induced diabetic neuropathic pain is associated with
potentiated calcium-permeable AMPA receptor
activity in the spinal cord. J Pharmacol Exp Ther.
2019;371:242–9.
82. Li JQ, Chen SR, Chen H, Cai YQ, Pan HL. Regulation
of increased glutamatergic input to spinal dorsal
horn neurons by mGluR5  in diabetic neuropathic
pain. J Neurochem. 2010;112:162–72.
83. Carroll I, Gaeta R, Mackey S. Multivariate analysis
of chronic pain patients undergoing lidocaine infu-
sions: increasing pain severity and advancing age
predict likelihood of clinically meaningful analge-
sia. Clin J Pain. 2007;23:702–6.
84. Chen SR, Chen H, Yuan WX, Pan HL.  Increased
presynaptic and postsynaptic alpha2-­ adrenoceptor
activity in the spinal dorsal horn in pain-
ful diabetic neuropathy. J Pharmacol Exp Ther.
2011;337:285–92.
85. Chen H, Xu X, Yang XY, et  al. Systemic dexme-
detomidine attenuates mechanical allodynia through
extracellular sign db type 2 diabetic mice. Neurosci
Lett. 2017;657:126–33.
86. Julius D.  TRP channels and pain. Annu Rev Cell
Dev Biol. 2013;29:355–84.
87. Cui YY, Xu H, Wu HH, Qi J, Shi J, Li YQ. Spatio-­
temporal expression and functional involvement
of transient receptor potential vanilloid 1  in dia-
betic mechanical allodynia in rats. PLoS One.
2014;9:e102052.
378
88. Bektur E, Sahin E, Ceyhan E, et al. Beneficial effect
of mirtazapine on diabetes-induced hyperalgesia:
involvement of TRPV1 and ASIC1 channels in the
spinal cord and dorsal root ganglion. Neurol Res.
2019;41:544–53.
89. Zhang N, Wei H, Wu W, et al. Effect of ropivacaine
on peripheral neuropathy in streptozocin diabetes-­
induced rats through TRPV1-CGRP pathway. Biosci
Rep. 2019;39:BSR20190817.
90. Wei H, Chapman H, Saarnilehto M, Kuokkanen K,
Koivisto A, Pertovaara A. Roles of cutaneous versus
spinal TRPA1 channels in mechanical hypersensitiv-
ity in the diabetic or mustard oil-treated non-diabetic
rat. Neuropharmacology. 2010;58:578–84.
91. Wei H, Koivisto A, Pertovaara A. Spinal TRPA1 ion
channels contribute to cutaneous neurogenic inflam-
mation in the rat. Neurosci Lett. 2010;479:253–6.
92. Griggs RB, Santos DF, Laird DE, et al. Methylglyoxal
and a spinal TRPA1-AC1-Epac cascade facilitate
pain in the db/db mouse model of type 2 diabetes.
Neurobiol Dis. 2019;127:76–86.
93. Tanaka K, Sekino S, Ikegami M, Ikeda H, Kamei
J.  Antihyperalgesic effects of ProTx-II, a Nav1.7
antagonist, and A803467, a Nav1.8 antagonist, in
diabetic mice. J Exp Pharmacol. 2015;7:11–6.
94. Calcutt NA.  Diabetic neuropathy and neuropathic
pain: a (con)fusion of pathogenic mechanisms? Pain.
2020;161:S65–86.
95. Maeda Y, Aoki Y, Sekiguchi F, et  al. Hyperalgesia
induced by spinal and peripheral hydrogen sulfide:
evidence for involvement of Cav3.2 T-type calcium
channels. Pain. 2009;142:127–32.
96. Matthews EA, Dickenson AH.  Effects of ethosuxi-
mide, a T-type Ca(2+) channel blocker, on dorsal
horn neuronal responses in rats. Eur J Pharmacol.
2001;415:141–9.
97. Jacus MO, Uebele VN, Renger JJ, Todorovic
SM.  Presynaptic Cav3.2 channels regulate excit-
atory neurotransmission in nociceptive dorsal horn
neurons. J Neurosci. 2012;32:9374–82.
98. Obrosova IG.  Diabetic painful and insensate neu-
ropathy: pathogenesis and potential treatments.
Neurotherapeutics. 2009;6:638–47.
99. Sandireddy R, Yerra VG, Areti A, Komirishetty P,
Kumar A.  Neuroinflammation and oxidative stress
in diabetic neuropathy: futuristic strategies based on
these targets. Int J Endocrinol. 2014;2014:674987.
100. Andrabi SA, Kim NS, Yu SW, et  al. Poly(ADP-­
ribose) (PAR) polymer is a death signal. Proc Natl
Acad Sci U S A. 2006;103:18308–13.
101. Hewlings SJ, Kalman DS. Curcumin: a review of its
effects on human health. Foods. 2017;6:92.
102. Acar A, Akil E, Alp H, et  al. Oxidative dam-
age is ameliorated by curcumin treatment in brain
and sciatic nerve of diabetic rats. Int J Neurosci.
2012;122:367–72.
103. Zhao WC, Zhang B, Liao MJ, et al. Curcumin ame-
liorated diabetic neuropathy partially by inhibition
of NADPH oxidase mediating oxidative stress in the
spinal cord. Neurosci Lett. 2014;560:81–5.
A. G. Marshall et al.
104. Raposo D, Morgado C, Pereira-Terra P, Tavares
I.  Nociceptive spinal cord neurons of laminae I-III
exhibit oxidative stress damage during diabetic
neuropathy which is prevented by early antioxidant
treatment with epigallocatechin-gallate (EGCG).
Brain Res Bull. 2015;110:68–75.
105. Ates O, Cayli SR, Altinoz E, et al. Neuroprotective
effect of mexiletine in the central nervous system of
diabetic rats. Mol Cell Biochem. 2006;286:125–31.
106. Karademir M, Sonmez MA, Akcilar R, Kocak E,
Yay A, Eser O. Evaluation of therapeutic potential of
intraperitoneal ozone gas in combination with insulin
above cranial and spinal neuropathy in rats with dia-
betes mellitus. Bratisl Lek Listy. 2018;119:636–41.
107. Alam U, Sloan G, Tesfaye S.  Treating pain in dia-
betic neuropathy: current and developmental drugs.
Drugs. 2020;80:363–84.
108. Lunn MP, Hughes RA, Wiffen PJ (2009) Duloxetine
for treating painful neuropathy or chronic pain.
Cochrane Database Syst Rev:CD007115.
109. Mixcoatl-Zecuatl T, Jolivalt CG. A spinal mechanism
of action for duloxetine in a rat model of painful dia-
betic neuropathy. Br J Pharmacol. 2011;164:159–69.
110. Kinoshita J, Takahashi Y, Watabe AM, Utsunomiya
K, Kato F.  Impaired noradrenaline homeostasis in
rats with painful diabetic neuropathy as a target of
duloxetine analgesia. Mol Pain. 2013;9:59.
111. Zhou DM, Zhuang Y, Chen WJ, Li W, Miao
B.  Effects of duloxetine on the toll-like recep-
tor 4 signaling pathway in spinal dorsal horn in a
rat model of diabetic neuropathic pain. Pain Med.
2018;19:580–8.
112. Ikeda T, Ishida Y, Naono R, et  al. Effects of intra-
thecal administration of newer antidepressants on
mechanical allodynia in rat models of neuropathic
pain. Neurosci Res. 2009;63:42–6.
113. Pichon X, Wattiez AS, Becamel C, et al. Disrupting
5-HT(2A) receptor/PDZ protein interactions reduces
hyperalgesia and enhances SSRI efficacy in neuro-
pathic pain. Mol Ther. 2010;18:1462–70.
114. Vithlani M, Terunuma M, Moss SJ.  The dynamic
modulation of GABA(A) receptor trafficking and
its role in regulating the plasticity of inhibitory syn-
apses. Physiol Rev. 2011;91:1009–22.
115. Kaila K, Price TJ, Payne JA, Puskarjov M, Voipio
J. Cation-chloride cotransporters in neuronal devel-
opment, plasticity and disease. Nat Rev Neurosci.
2014;15:637–54.
116. Bright DP, Renzi M, Bartram J, et  al. Profound
desensitization by ambient GABA limits activation
of δ-containing GABAA receptors during spillover.
J Neurosci. 2011;31:753–63.
117. Delgado-Lezama R, Loeza-Alcocer E, Andrés
C, Aguilar J, Guertin PA, Felix R.  Extrasynaptic
GABA(A) receptors in the brainstem and spi-
nal cord: structure and function. Curr Pharm Des.
2013;19:4485–97.
118. Perez-Sanchez J, Lorenzo LE, Lecker I, et  al.
α5GABA. J Neurosci Res. 2017;95:1307–18.
The Spinal Cord in Diabetic Neuropathy
119. Loeza-Alcocer E, Canto-Bustos M, Aguilar J,
González-Ramírez R, Felix R, Delgado-Lezama
R. α(5)GABA(A) receptors mediate primary affer-
ent fiber tonic excitability in the turtle spinal cord. J
Neurophysiol. 2013;110:2175–84.
120. Bettler B, Kaupmann K, Mosbacher J, Gassmann
M.  Molecular structure and physiological func-
tions of GABA(B) receptors. Physiol Rev.
2004;84:835–67.
121. Terunuma M.  Diversity of structure and function
of GABA.  Proc Jpn Acad Ser B Phys Biol Sci.
2018;94:390–411.
122. Kaupmann K, Schuler V, Mosbacher J, et al. Human
gamma-aminobutyric acid type B receptors are dif-
ferentially expressed and regulate inwardly rec-
tifying K+ channels. Proc Natl Acad Sci U S A.
1998;95:14991–6.
123. Bohlhalter S, Weinmann O, Mohler H, Fritschy
JM.  Laminar compartmentalization of GABAA-­
receptor subtypes in the spinal cord: an immunohis-
tochemical study. J Neurosci. 1996;16:283–97.
124. Roberts LA, Beyer C, Komisaruk BR.  Nociceptive
responses to altered GABAergic activity at the spinal
cord. Life Sci. 1986;39:1667–74.
125. Yaksh TL.  Behavioral and autonomic correlates of
the tactile evoked allodynia produced by spinal gly-
cine inhibition: effects of modulatory receptor sys-
tems and excitatory amino acid antagonists. Pain.
1989;37:111–23.
126. Sivilotti L, Woolf CJ. The contribution of GABAA
and glycine receptors to central sensitization: dis-
inhibition and touch-evoked allodynia in the spinal
cord. J Neurophysiol. 1994;72:169–79.
127. Sorkin LS, Puig S, Jones DL.  Spinal bicuculline
produces hypersensitivity of dorsal horn neurons:
effects of excitatory amino acid antagonists. Pain.
1998;77:181–90.
128. Wilson PR, Yaksh TL.  Baclofen is antinociceptive
in the spinal intrathecal space of animals. Eur J
Pharmacol. 1978;51:323–30.
129. Hammond DL, Drower EJ.  Effects of intrathecally
administered THIP, baclofen and muscimol on noci-
ceptive threshold. Eur J Pharmacol. 1984;103:121–5.
130. Miletic G, Draganic P, Pankratz MT, Miletic
V.  Muscimol prevents long-lasting potentiation of
dorsal horn field potentials in rats with chronic con-
striction injury exhibiting decreased levels of the
GABA transporter GAT-1. Pain. 2003;105:347–53.
131. Gwak YS, Tan HY, Nam TS, Paik KS, Hulsebosch
CE, Leem JW. Activation of spinal GABA receptors
attenuates chronic central neuropathic pain after spi-
nal cord injury. J Neurotrauma. 2006;23:1111–24.
132. Basbaum AI, Bautista DM, Scherrer G, Julius
D. Cellular and molecular mechanisms of pain. Cell.
2009;139:267–84.
133. Costigan M, Scholz J, Woolf CJ. Neuropathic pain: a
maladaptive response of the nervous system to dam-
age. Annu Rev Neurosci. 2009;32:1–32.
134. Castro-Lopes JM, Tavares I, Coimbra A.  GABA
decreases in the spinal cord dorsal horn after periph-
eral neurectomy. Brain Res. 1993;620:287–91.
379
135. Moore KA, Kohno T, Karchewski LA, Scholz J,
Baba H, Woolf CJ.  Partial peripheral nerve injury
promotes a selective loss of GABAergic inhibition
in the superficial dorsal horn of the spinal cord. J
Neurosci. 2002;22:6724–31.
136. Malmberg AB, O’Connor WT, Glennon JC, Cesena
R, Calcutt NA.  Impaired formalin-evoked changes
of spinal amino acid levels in diabetic rats. Brain
Res. 2006;1115:48–53.
137. Jolivalt CG, Lee CA, Ramos KM, Calcutt
NA.  Allodynia and hyperalgesia in diabetic rats
are mediated by GABA and depletion of spi-
nal potassium-chloride co-transporters. Pain.
2008;140:48–57.
138. Coull JA, Beggs S, Boudreau D, et  al. BDNF
from microglia causes the shift in neuronal anion
gradient underlying neuropathic pain. Nature.
2005;438:1017–21.
139. Lee-Kubli CA, Calcutt NA. Altered rate-dependent
depression of the spinal H-reflex as an indicator of
spinal disinhibition in models of neuropathic pain.
Pain. 2014;155:250–60.
140. Fernyhough P, Diemel LT, Brewster WJ, Tomlinson
DR.  Altered neurotrophin mRNA levels in periph-
eral nerve and skeletal muscle of experimentally dia-
betic rats. J Neurochem. 1995;64:1231–7.
141. Lee-Kubli C, Marshall AG, Malik RA, Calcutt
NA. The H-reflex as a biomarker for spinal disinhi-
bition in painful diabetic neuropathy. Curr Diab Rep.
2018;18:1.
142. Wager EW, Buerger AA.  A linear relationship
between H-reflex latency and sensory conduc-
tion velocity in diabetic neuropathy. Neurology.
1974;24:711–4.
143. Stanley EF.  Sensory and motor nerve conduction
velocities and the latency of the H reflex during
growth of the rat. Exp Neurol. 1981;71:497–506.
144. Cliffer KD, Tonra JR, Carson SR, et al. Consistent
repeated M- and H-Wave recording in the hind limb
of rats. Muscle Nerve. 1998;21:1405–13.
145. Palmieri RM, Ingersoll CD, Hoffman MA.  The
hoffmann reflex: methodologic considerations and
applications for use in sports medicine and athletic
training research. J Athl Train. 2004;39:268–77.
146. Lloyd DP, Wilson VJ. Reflex depression in rhythmi-
cally active monosynaptic reflex pathways. J Gen
Physiol. 1957;40:409–26.
147. Ishikawa K, Ott K, Porter RW, Stuart D.  Low fre-
quency depression of the H wave in normal and spi-
nal man. Exp Neurol. 1966;15:140–56.
148. Meinck HM.  Occurrence of the H reflex and
the F wave in the rat. Electroencephalogr Clin
Neurophysiol. 1976;41:530–3.
149. Schindler-Ivens S, Shields RK.  Low frequency
depression of H-reflexes in humans with acute
and chronic spinal-cord injury. Exp Brain Res.
2000;133:233–41.
150. Ho SM, Waite PM.  Effects of different anesthetics
on the paired-pulse depression of the h reflex in adult
rat. Exp Neurol. 2002;177:494–502.
380
151. Kakinohana O, Hefferan MP, Nakamura S, et  al.
Development of GABA-sensitive spasticity and
rigidity in rats after transient spinal cord isch-
emia: a qualitative and quantitative electrophysi-
ological and histopathological study. Neuroscience.
2006;141:1569–83.
152. Lee-Kubli CA, Zhou X, Jolivalt CG, Calcutt
NA. Pharmacological modulation of rate-dependent
depression of the spinal H-reflex predicts thera-
peutic efficacy against painful diabetic neuropathy.
Diagnostics (Basel). 2021;11:283.
153. Hernandez-Reyes JE, Salinas-Abarca AB, Vidal-­
Cantu GC, et  al. alpha5GABAA receptors play a
pronociceptive role and avoid the rate-dependent
depression of the Hoffmann reflex in diabetic neuro-
pathic pain and reduce primary afferent excitability.
Pain. 2019;160:1448–58.
154. Marshall AG, Lee-Kubli C, Azmi S, et al. Spinal dis-
inhibition in experimental and clinical painful dia-
betic neuropathy. Diabetes. 2017;66:1380–90.
155. Sultan A, Gaskell H, Derry S, Moore RA. Duloxetine
for painful diabetic neuropathy and fibromyalgia
pain: systematic review of randomised trials. BMC
Neurol. 2008;8:29.
156. Melzack R, Wall PD. Pain mechanisms: a new the-
ory. Science. 1965;150:971–9.
157. Vannemreddy P, Slavin KV. Spinal cord stimulation:
current applications for treatment of chronic pain.
Anesth Essays Res. 2011;5:20–7.
158. McGreevy K, Williams KA. Contemporary insights
into painful diabetic neuropathy and treatment with
spinal cord stimulation. Curr Pain Headache Rep.
2012;16:43–9.
159. McCarthy KF, McCrory C. Cerebrospinal fluid lev-
els of glial cell-derived neurotrophic factor correlate
with spinal cord stimulation frequency in patients
with neuropathic pain: a preliminary report. Spinal
Cord. 2014;52(Suppl 2):S8–10.
160. Wu M, Linderoth B, Foreman RD. Putative mecha-
nisms behind effects of spinal cord stimulation on
vascular diseases: a review of experimental studies.
Auton Neurosci. 2008;138:9–23.
161. Wu M, Komori N, Qin C, Farber JP, Linderoth B,
Foreman RD.  Sensory fibers containing vanilloid
receptor-1 (VR-1) mediate spinal cord stimulation-­
induced vasodilation. Brain Res. 2006;1107:177–84.
162. Wu M, Komori N, Qin C, Farber JP, Linderoth B,
Foreman RD.  Extracellular signal-regulated kinase
(ERK) and protein kinase B (AKT) pathways
involved in spinal cord stimulation (SCS)-induced
vasodilation. Brain Res. 2008;1207:73–83.
163. Wu M, Thorkilsen MM, Qin C, Farber JP, Linderoth
B, Foreman RD.  Effects of spinal cord stimula-
tion on peripheral blood circulation in rats with
streptozotocin-induced diabetes. Neuromodulation.
2007;10:216–23.
164. Davidson EP, Coppey LJ, Yorek MA.  Activity and
expression of the vanilloid receptor 1 (TRPV1) is
A. G. Marshall et al.
altered by long-term diabetes in epineurial arterioles
of the rat sciatic nerve. Diabetes Metab Res Rev.
2006;22:211–9.
165. Yorek MA, Coppey LJ, Gellett JS, Davidson
EP.  Sensory nerve innervation of epineurial arteri-
oles of the sciatic nerve containing calcitonin gene-­
related peptide: effect of streptozotocin-induced
diabetes. Exp Diabesity Res. 2004;5:187–93.
166. Schechtmann G, Song Z, Ultenius C, Meyerson BA,
Linderoth B.  Cholinergic mechanisms involved in
the pain relieving effect of spinal cord stimulation
in a model of neuropathy. Pain. 2008;139:136–45.
167. Stiller CO, Cui JG, O'Connor WT, Brodin E,
Meyerson BA, Linderoth B.  Release of gamma-­
aminobutyric acid in the dorsal horn and suppres-
sion of tactile allodynia by spinal cord stimulation in
mononeuropathic rats. Neurosurgery. 1996;39:367–
74; discussion 374-365.
168. de Vos CC, Rajan V, Steenbergen W, van der Aa
HE, Buschman HP. Effect and safety of spinal cord
stimulation for treatment of chronic pain caused
by diabetic neuropathy. J Diabetes Complications.
2009;23:40–5.
169. Pluijms WA, Slangen R, Joosten EA, et al. Electrical
spinal cord stimulation in painful diabetic polyneu-
ropathy, a systematic review on treatment efficacy
and safety. Eur J Pain. 2011;15:783–8.
170. Slangen R, Schaper NC, Faber CG, et  al. Spinal
cord stimulation and pain relief in painful dia-
betic peripheral neuropathy: a prospective two-­
center randomized controlled trial. Diabetes Care.
2014;37:3016–24.
171. van Beek M, Geurts JW, Slangen R, et al. Severity
of neuropathy is associated with long-term spi-
nal cord stimulation outcome in painful diabetic
peripheral neuropathy: five-year follow-up of a
prospective two-center clinical trial. Diabetes Care.
2018;41:32–8.
172. Pluijms WA, van Kleef M, Honig WM, Janssen SP,
Joosten EA.  The effect of spinal cord stimulation
frequency in experimental painful diabetic polyneu-
ropathy. Eur J Pain. 2013;17:1338–46.
173. van Beek M, van Kleef M, Linderoth B, van Kuijk
SM, Honig WM, Joosten EA.  Spinal cord stimu-
lation in experimental chronic painful diabetic
polyneuropathy: delayed effect of high-frequency
stimulation. Eur J Pain. 2017;21:795–803.
174. de Vos CC, Bom MJ, Vanneste S, Lenders MW,
de Ridder D.  Burst spinal cord stimulation evalu-
ated in patients with failed back surgery syndrome
and painful diabetic neuropathy. Neuromodulation.
2014;17:152–9.
175. Tjepkema-Cloostermans MC, de Vos CC, Wolters
R, Dijkstra-Scholten C, Lenders MW.  Effect of
burst stimulation evaluated in patients familiar
with spinal cord stimulation. Neuromodulation.
2016;19:492–7.
 Brain Changes in Diabetes and
Cognitive Dysfunction
Geert Jan Biessels
The first reports on cognitive dysfunction in rela-
tion to diabetes date back a century ago [1]. Yet,
only over the past two decades, the spectrum of
cognitive changes occurring in people with dia-
betes in terms of clinical features and prognosis
has become more clear. While diabetes-­associated
changes have been observed in all age groups,
from young children to the oldest old, this chap-
ter will primarily discuss findings in adults. For
cognitive changes that are observed in children
with diabetes, the reader is referred to dedicated
reviews [2, 3].
1 Stages of Cognitive
Dysfunction: Definitions
When considering different stages of cognitive
dysfunction, it is essential to discern cognitive
impairment from more subtle cognitive changes.
The term cognitive impairment is best reserved for
acquired cognitive dysfunction where the actual
cognitive performance on neuropsychological
testing falls outside the normal range. While there
is no universally accepted definition of “abnormal
test results” among neuropsychologist, usually
scores below the 5th to the 10th percentile on
G. J. Biessels (*)
Department of Neurology G03.232, UMC Utrecht
Brain Center, University Medical Center Utrecht,
Utrecht, the Netherlands
e-mail: g.j.biessels@umcutrecht.nl
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_21
norm referenced tests are considered as abnormal.
In clinical research and care, acquired cognitive
impairment is generally classified with the diag-
nostic labels “Mild Cognitive Impairment (MCI)”
or dementia. MCI refers to impairment affecting
one or more cognitive domains with—by defini-
tion—largely preserved activities of daily life [4,
5]. This construct was devised over two decades
ago to capture a stage between normal cognition
and dementia [5]. Individuals with MCI are at risk
of transition to dementia, but not all people with
MCI show cognitive decline and in some cogni-
tion may even revert back to normal. Dementia is
defined as acquired objective cognitive impair-
ment affecting multiple cognitive domains, severe
enough to affect activities of daily life [6]. Most
people with dementia will show cognitive decline,
although this clearly depends on the aetiology. It
is important to note that neither the definition of
MCI nor that of dementia refer to any particular
underlying aetiology. Yet, for lay people, as well
as for many clinicians, a diagnosis of dementia is
often implicitly equated with “Alzheimer’s dis-
ease”. Although Alzheimer’s disease is indeed the
most common form of dementia, it is important to
realize that MCI or dementia is principally syn-
dromal diagnosis, defined by symptoms. Such
symptoms may be suggestive of particular demen-
tia aetiologies, but are known to lack specificity.
Therefore, the current standard in research—and
increasingly also in clinical care—is to diagnose
underlying aetiologies with dedicated biomarkers
for the actual disease process [7]. Such biomarker
381
382
approaches have also shown that the pathologies
ultimately contributing to the development of
dementia may start to develop many years before
cognitive symptoms develop. Alzheimer pathol-
ogy, for example, may occur in individuals who
are cognitively normal already two decades before
dementia becomes clinically manifest [8, 9]. Yet,
as long as treatment for most types of dementia is
primarily symptomatic and supportive, it can be
argued that for clinical practice, biomarker-sup-
ported diagnosis is not essential. Clearly, this is
different for aetiological research. For example, it
is important to be aware that approximately one in
six people who have a clinical diagnosis of “prob-
able Alzheimer’s disease”, based on a syndromal
diagnosis without biomarker support [10], do not
have biomarker or pathological evidence of under-
lying Alzheimer pathology [11, 12]. This is clearly
relevant for studies of dementia occurring in asso-
ciation with diabetes, as will be described below,
as most epidemiological studies on that topic
based their diagnoses on dementia types on syn-
dromal, rather than biomarker supported,
diagnoses.
While diagnostic constructs for cognitive
impairment are by nature dichotomous classifiers,
which are present or absent in an individual
patient, the more subtle cognitive changes that
occur in association with diabetes rather present
themselves as a continuum. It has been noted that
as a group, people with diabetes perform on aver-
age slightly worse on cognitive testing than people
without diabetes. This has been noted across all
age groups, for people with both type 1 and type 2
diabetes [13, 14]. The magnitude of the group dif-
ference is typically around 0.3–0.5 standard devia-
tion units [13, 14]. In other words, mean
performance on the domains that are most typi-
cally affected by diabetes is at the 40th to the 30th
percentile of that of reference groups of people
without diabetes. This is well above the threshold
of the 5th to 10th percentile that denote cognitive
impairment. These subtle cognitive changes have
been coined “diabetes-associated cognitive decre-
ments” [15]. The subtle changes in cognitive func-
tion to which this term refers might result in
cognitive complaints (usually expressed only by
the patient), but activities of daily life should by
G. J. Biessels
definition be preserved [15]. Longitudinal studies
show that diabetes-­ associated cognitive decre-
ments show only slow progression over years [15,
16]. Because the construct “diabetes-associated
cognitive decrements” refers to cognitive changes
that can be detected at a group level, it cannot be
readily used as a diagnostic label in individual
patients. Yet, it may serve to understand cognitive
complaints expressed by an individual patient if
the severity and trajectories of the deficits fit with
the construct’s definition.
In my view, diabetes-associated cognitive dec-
rements, on the one hand, and MCI and demen-
tia, on the other, should not be regarded as a
continuum [16]. These different stages of cogni-
tive dysfunction affect different age groups, have
different trajectories, and appear to have different
underlying mechanisms [17]. Diabetes-associated
decrements can occur in all age groups, from
young adults and even adolescents with T2DM
[18, 19] to the oldest old [20]. Cognitive decline
over time is typically slow over the course of
many years [16]. These decrements affect people
with diabetes as a group: they are characterized
by a modest shift in average cognitive perfor-
mance across individuals, rather than abnormal
performance in a subset of individuals. Dementia
and MCI, on the other hand, are defined by abnor-
mal cognitive performance in individual patients.
Occurrence of dementia and MCI is strongly age-­
dependent [21], and particularly dementia mostly
involves relentless further cognitive decline.
Clearly, most people who develop MCI or
dementia will generally pass through a stage of
cognitive performance deficits that equate to cog-
nitive decrements. Yet, there are really no argu-
ments to generically consider all people with
diabetes-associated cognitive decrements to be
on a path towards dementia.
2 Stages of Cognitive
Dysfunction in Diabetes
2.1 Type 1 Diabetes
At a group level, adults with T1DM present with
subtle cognitive decrements relative to age-­
Brain Changes in Diabetes and Cognitive Dysfunction
matched controls [13, 22]. These decrements par-
ticularly affect the domain intelligence (effect
size Cohen’s d  =  0.7), psychomotor efficiency
(d = 0.6), and cognitive flexibility (d = 0.5) [13]
(see also Table  1). Long-term follow-up of the
participants of the Diabetes Control and
Complications Trial (DCCT) shows that cogni-
tive decline over time is generally limited up to a
diabetes duration of ~25  years, although at an
average diabetes duration of 37 years, substantial
deterioration of psychomotor and mental effi-
ciency was observed [23]. There may also be
subgroups of patients, particularly those with
advanced microvascular complications, who may
show more pronounced cognitive decline over
time [24]. Risk factors for accelerated cognitive
decline in people with type 1 diabetes also
include exposure to higher HbA1c levels, more
episodes of severe hypoglycaemia, and elevated
systolic blood pressure [23].
With increasing numbers of people with type
1 diabetes reaching old age, the question if type 1
diabetes increases the risk of dementia becomes
more and more relevant, and there is an emerging
Table 1  Cognitive decrements in type 1 and type 2
diabetes
Cognitive domain Type 1 Type 2
General cognition 0.2– 0.3
0.3
Attention 0–0.3 0.2–
0.4
Psychomotor speed 0.3– 0.3
0.6
Verbal memory 0–0.2 0.2–
0.3
Non-verbal memory 0–0.2 0.2–
0.3
Mental flexibility—executive 0.4– 0.3
function 0.6
Visuospatial 0.4 0.2
Estimates of the difference in average performance in
people with diabetes compared to age-matched reference
groups per domain expressed in standardized effect sizes
(Cohen’s d; i.e. the difference between the means of two
groups divided by the pooled standard deviation of the
two groups), based on meta-analyses by Brands [13],
Palta [14], and Monette [30] et al. In neuropsychological
studies, an effect size of 0.2 is generally considered to cor-
respond to small effects, 0.5 to medium, and 0.8 to large
effects [106]
383
literature on this topic. A retrospective cohort
study of administrative hospital records from
England observed that the relative risk for demen-
tia in people admitted with type 1 diabetes was
1.65 (95% CI 1.61, 1.68), compared to a refer-
ence group [25]. This increased risk included
both a syndromal diagnosis of Alzheimer’s dis-
ease (relative risk 1.40; 95% CI 1.27, 1.53) and
vascular dementia (2.66; 95% CI 2.48, 2.86).
Although the relative risk of dementia was high-
est in people with type 1 diabetes admitted to
hospital when aged 30–39  years (7.10; 95% CI
4.65, 10.6) and decreased as age increased, for
example to 1.61 (95% CI 1.57, 1.66) in those
aged 70–79 years, the absolute risk of dementia
occurrence was clearly much lower in the young-
est age band than in older people (30–39 years:
1.0/1000; 50–59  years: 8.5/1000; 70–79  years:
59.2/1000) [25], showing that also among people
with diabetes occurrence of dementia below the
age of 60  years is relatively rare. Exposure to
higher HbA1c levels [26], episodes of severe
hypo- or hyperglycaemia [27], and presence of
diabetic retinopathy [28] or depression [29] have
been identified as risk factors for dementia among
people with type 1 diabetes, but this topic war-
rants further investigation.
2.2 Type 2 Diabetes
Subtle cognitive decrements in people with type
2 diabetes can be seen across cognitive domains,
with little specificity in terms of domains affected
(Table  1). Relative to people without diabetes,
effect sizes are mostly small to moderate (Cohen’s
d 0.2–0.3) [14, 30]. Numerous studies have noted
these decrements in patients with type 2 diabetes
of all age groups [14, 30]. Onset is likely already
in pre-diabetic stages [16]. Further diabetes-­
related cognitive decline generally evolves only
slowly over time, at a rate that is up to 50% faster
than that of normal cognitive ageing [16, 31, 32].
Population-based studies report an increased
risk of MCI in people with type 2 diabetes, com-
pared to people without diabetes (e.g. Ref. [33]).
A recent meta-analysis estimated the relative risk
at 1.49 (95% CI 1.26–1.77) [34]. This included
384
both so-called amnestic MCI (where the princi-
pally affected domain is memory; relative risk
1.50; 95% CI 1.17–1.92) and non-amnestic MCI
(where another non-memory domain is princi-
pally affected; relative risk 1.34; 95% CI 1.04–
1.73) [34]. The MCI risk appears to be higher in
those with higher HbA1c levels (i.e. HbA1c ≥7%,
compared with HbA1c <7%, hazard ratio 1.56
(95% CI 1.10–2.21) [33]. There is also consistent
evidence that type 2 diabetes increases the rate of
conversion from MCI to dementia. The odds ratio
was estimated at 1.53 (95% CI 1.20–1.97) in a
recent meta-analysis of ten studies [35], whereas
another meta-analysis of nine studies found a rel-
ative risk of 1.91 (95%CI 1.54–2.36) [34].
There have also been many studies on the risk
of dementia in relation to diabetes [34, 36].
Although not all of these studies formally distin-
guished between type 1 and type 2 diabetes, the
vast majority of the patients involved will have
had type 2 diabetes [36]. A recent meta-analysis,
pooling 31 studies, with a total sample size of
4.3  million, showed a relative risk of all-cause
dementia of 1.43 (95% CI 1.33–1.53). Meta-­
regression analyses showed that the degree to
which cardiovascular disease was controlled for
explained heterogeneity between studies: the
pooled relative risk of studies not controlling for
cardiovascular disease was 1.66 (95% CI 1.53–
1.80), compared to 1.28 (95% CI 1.20–1.36) in
studies controlling for cardiovascular disease
[34]. Twenty-four studies included in the meta-­
analysis reported data for diabetes and the risk of
a clinical diagnosis of Alzheimer’s disease, with
a relative risk of 1.43 (95% CI 1.25–1.62) [34].
Seventeen studies reported on the relative risk of
vascular dementia, which was estimated at 1.91
(95% CI 1.61–2.25) [34]. Importantly, diagnoses
of Alzheimer’s disease in the studies included in
the meta-analysis relied on a syndromal diagno-
sis, rather than biomarkers, which is known to
have limited specificity for the actual presence of
Alzheimer’s pathology [7, 11].
Risk factors for dementia for dementia in the
general population include several factors that
are closely related to type 2 diabetes, including
midlife hypertension, midlife obesity, physical
inactivity, depression, smoking, and low educa-
G. J. Biessels
tional attainment [37, 38]. Moreover, among
patients with type 2 diabetes episodes of severe
hypo- or hyperglycaemia, presence of micro- or
macrovascular complications, including stroke,
and a low body mass index are known risk factors
for dementia [39, 40]. The relation between gly-
cemia and dementia risk in people with type 2
diabetes is not completely clear [41]. There are
indications that both high and low HbA1c levels
may negatively impact cognition particularly in
older individuals [42], as has also been noted in
type 1 diabetes [26]. There are also indications
that in people without diabetes fasting plasma
glucose is related to dementia risk [34]. More
recently, however, evidence form Mendelian ran-
domization studies questions if glucose levels
and even type 2 diabetes itself (as defined by
abnormal glucose levels) are causally related to
dementia, in particular Alzheimer’s disease [43,
44]. In the light of these latter observations it
been suggested that type 2 diabetes essentially
represents a risk marker for dementia and that the
excess dementia risk in people with type 2 diabe-
tes primarily reflects lower socioeconomic posi-
tion and childhood and educational attainment
[45]. On the other hand, Mendelian randomiza-
tion studies do suggest that type 2 diabetes is
associated with small vessel disease-related brain
injury [46]. All in all these combined observa-
tions suggest that the interrelations between type
2 diabetes and dementia are complex, likely
depending on multiple factors, in part with non-­
linear relationships, and—as will be discussed
further on in this chapter—likely depending on
multiple underlying disease processes.
3 Mechanisms: Structural
Brain Changes and Dementia
Aetiologies
Studying the aetiology and underlying brain
changes of cognitive dysfunction and dementia
has long relied primarily on post-mortem neuro-
pathological examination. Fortunately, over the
past decades new neuroimaging methods and
other biomarkers have been established that pro-
vide much more options for mechanistic studies,
Brain Changes in Diabetes and Cognitive Dysfunction
importantly, also during life. This has proven to
be of great value for understanding the impact of
diabetes on the brain [47]. The plethora of differ-
ent markers can, however, provide challenges
with interpretation. A key element in this context
is to consider biomarker specificity for the pro-
cess one intends to measure. For example, if one
is interested in detecting Alzheimer-related brain
changes, a conventional approach has long been
to look at atrophy of the hippocampus, because
atrophy in this region is closely linked to
Alzheimer’s disease. The point is, however, that
hippocampal atrophy is not specific to Alzheimer’s
disease [7]. Hence, if one truly wishes to assess
Alzheimer’s pathology in relation to diabetes, it is
preferred to use currently available biomarkers for
ß-amyloid (Aß) deposition or pathologic tau, as
these proteins are the constituents of the plaques
and tangles that define Alzheimer’s disease as a
unique disease among several that can lead to
dementia [7]. To support design of further bio-
marker studies in diabetes and facilitate interpre-
tation, we have recently proposed a framework to
arrange them in three broad classes [47]. Firstly,
biomarkers reflecting pathologies considered to
define major causes of dementia, in particular
Alzheimer’s and vascular disease. Secondly, bio-
markers of brain parenchymal injury that may
occur as a consequence of different disease pro-
cesses. Thirdly, biomarkers of brain blood flow
and metabolism, abnormalities of which may con-
tribute to brain injury. We made the point that
although biomarkers of brain parenchymal injury
may show patterns of injury suggestive of certain
aetiologies, caution is warranted to derive aetiol-
ogy from such markers, because of limited aetio-
logical specificity. Moreover, while abnormal
patterns of brain blood flow and metabolism may
reflect processes that can cause further brain
injury, brain injury by itself can affect cerebral
blood flow and metabolism [47].
3.1 Diabetes and Alzheimer
Pathologies
Brain Aβ and tau load can be imaged using PET
technology [48, 49]. In addition Aβ42, Aβ40,
385
total tau and phosphorylated tau (P-tau) levels
can be measured in cerebrospinal fluid (CSF)
samples, obtained by lumbar puncture. These
protein levels in the CSF provide an indication of
the burden of Aβ and tau pathologies in the brain
[7]. Evaluation of plasma levels of Aβ and tau
proteins as markers of Alzheimer pathologies is
still in development [50].
Despite the extensive epidemiological evi-
dence for an association between diabetes and
the risk of a syndromal diagnosis of Alzheimer’s
disease, post-mortem neuropathological studies
do not observe a higher burden of Aβ plaques and
tau tangles in people with diabetes compared to
those without diabetes [51, 52]. Moreover, PET
imaging studies show no influence of diabetes on
Aβ deposition [47, 53]. For example, in a pro-
spective population-based cohort study, neither
mid-life nor late-life diabetes was associated
with abnormal Aβ tracer uptake on PET scans
acquired in late-life [53]. This suggests that the
excess dementia risk in people with diabetes is
likely attributable to non-Alzheimer pathologies.
3.2 Diabetes and Vascular Brain
Injury
It is important to realize that most methods to
assess possible vascular contributions to cognitive
decline and dementia involve assessment of
parenchymal brain lesions that are the conse-
quence of cerebrovascular disease. This is because
the primary vessel populations involved are so
small that they cannot be examined by conven-
tional neuroimaging techniques. Manifestations
of these so-called cerebral small vessel diseases
(SVDs) in the form of brain parenchymal lesions
can be detected both with CT and MRI, although
sensitivity of MRI is superior [54]. The principal
manifestations on MRI include lacunes, white
matter hyperintensities, microbleeds, microin-
farcts and enlarged perivascular spaces (Figs.  1
and 2) [54]. There are also emerging techniques to
assess SVDs at the level of the small cerebral
blood vessels themselves [55], but at the time of
writing these have not yet been evaluated in rela-
tion to diabetes.
386
SVDs are common, and the main vascular
aetiology in dementia [56, 57] and vascular dis-
ease is a key contributor to other diabetic compli-
cations. Evaluation of SVD-related brain injury
is therefore of clear interest for diabetes-­
associated cognitive dysfunction [58].
For type 1 diabetes, few neuroimaging studies
have thus far have reported on SVD lesions. This
is likely due to the fact that most brain MRI stud-
ies of patients with type 1 diabetes included pop-
ulations under the age of 40–50 years, and sample
sizes were generally modest. Because SVD
lesions typically are relatively rare or low in bur-
den at that age, this limits the power of such stud-
ies. A study with the largest sample size thus far
(191 patients, 30 matched controls, average age
40  years) did observe that microbleeds were
seven times as common in patients (24%) as con-
trols (3.3%), and some form of white matter
hyperintensities was seen in 23% of patients and
Fig. 1  Grading white matter hyperintensities. White mat-
ter hyperintensities (WMH) on Fluid Attenuated Inversion
Recovery (FLAIR) or other T2-weighted sequences can
be rated visually according to the Fazekas scale [60].
Periventricular WMHs are graded as: 0  =  absence,
1  =  caps or pencil-thin lining, 2  =  smooth halo, and
3  =  irregular periventricular WMH extending into the
G. J. Biessels
6.7% of controls [59]. As expected at this age, the
burden of white matter hyperintensities was
mostly low (Fazekas grade 1 [60], see Fig. 1) and
only 2.1% of patients and 0 controls had lacunes,
too few to permit formal statistical testing [59].
Another study, including 97 patients (average
diabetes duration 41 years) and 81 controls, aver-
age age 50 years, did observe a higher burden of
white matter hyperintensities, with 33% of
patients having a Fazekas score of 2 or 3, com-
pared to 7% of controls, with 5% of patients
showing lacunes versus none of the controls [61].
Among patients with type 1 diabetes, micro-
bleeds were found to be more common in those
with than in those without proliferative retinopa-
thy [62].
The literature on vascular brain lesions in
patients with type 2 diabetes is more extensive
[63, 64]. Autopsy studies report a more frequent
occurrence of brain infarcts in people with type
deep white matter. Deep (subcortical) WMH are graded
separately as: 0 = absence, 1 = punctuate foci, 2 = begin-
ning confluence of foci, and 3  =  large confluent areas.
Examples of each grade are provided. The patient with
grade 3 WMH also has multiple lacunes, examples marked
by white arrowheads
Brain Changes in Diabetes and Cognitive Dysfunction 387

a b

c d

Fig. 2  Examples of vascular brain injury on MRI. (a)
Right frontal lacune (white arrowhead). On FLAIR
images lacunes of presumed vascular origin generally
have a central CSF-like cystic core with a surrounding rim
of hyperintensity [54]. This patient also has Fazekas grade
3 WMH [60]. (b) Cerebral microbleeds (white arrow-
heads) on a T2*-weighted image. Microbleeds appear as
areas of signal void on T2*-weighted MRI sequences
[54]. They can be rated according to location as strictly
deep, strictly lobar, and mixed (as is the case in this
patient). It is important to distinguish microbleeds from
other sources of signal void such as blood vessels (open
arrowheads) and mineral deposits in the basal ganglia
(star), based on shape and location. (c/d) An acute cortical
infarct showing up as a high signal on FLAIR (c) with a
corresponding lesion on diffusion MRI (d). This patient
also has Fazekas grade 3 WMH (c)
388
2 diabetes compared to people without diabetes
[64]. These studies also report that vessel
pathology is more common in those with type 2
diabetes, in particular cerebral atherosclerosis
[64]. By contrast, type 2 diabetes does not
appear to be associated with cerebral arteriolo-
sclerosis at autopsy [64, 65]. This requires fur-
ther investigation and seems counterintuitive, in
the light of type 2 diabetes’ known association
with peripheral arteriolosclerosis and with sub-
cortical infarcts in the brain [64]. Indeed, type 2
diabetes is an established risk factor for lacunes
on CT and MRI [63, 66, 67], and Mendelian
randomization analyses also suggest a causal
association between type 2 diabetes and lacunar
stroke [68]. The volume of white matter hyper-
intensities also appears to be modestly increased
in people with type 2 diabetes, compared to
those without [17, 67, 69]. White matter hyper-
intensities appear to be higher in patients with
higher HbA1c levels [70]. There is as yet no
convincing evidence that type 2 diabetes affects
the burden of microbleeds, microinfarcts and
enlarged perivascular spaces [17, 67]. While
occurrence of these different SVD lesions
appeared to be a mediator in type 2 diabetes-
associated cognitive decrements on several cog-
nitive domains [71], others found that the
association between type 2 diabetes and cogni-
tive impairment was not attenuated when SVD
lesions were taken into account [66].
3.3 Diabetes
and Neurodegeneration/
Global Brain Injury
Modest reductions in cerebral volume, generally
considered to reflect cerebral atrophy, are the
most extensively documented neuroimaging fea-
ture of type 1 as well as type 2 diabetes. Of note,
atrophy can reflect underlying neurodegenerative
disease, but it is also a manifestation of SVD
[54]. Moreover, for studies in diabetes, it should
be considered that fluctuations in blood glucose
levels can induce osmotic changes, which can
also affect brain volumes [72], thus confounding
atrophy assessment.
G. J. Biessels
Studies on brain volumes in people with type
1 diabetes generally have modest sample sizes. A
meta-analysis of volumetric data found no sig-
nificant differences in total brain volume, total
grey matter volume, or hippocampal volume
between people with type 1 and controls [73].
Some studies do show regional differences in
cortical grey matter volume or density within the
frontal [74–76], posterior [76, 77], and temporal
cortex [76]. The patterns of affected regions vary
across studies, likely reflecting the subtle nature
of the changes, variation in study selection crite-
ria, and modest sample sizes. There are indica-
tions that lower volumes are associated with
worse glycaemic control [76, 77] and also with
worse cognitive performance [77], but these top-
ics warrant further studies with larger sample
sizes, also including people with more longstand-
ing diabetes.
Modest brain atrophy, cortical as well as sub-
cortical, is a consistent finding of MRI studies
in type 2 diabetes. Cross-sectional studies report
reductions in average total brain volume of 0.5–
2.0% relative to controls, which is the equiva-
lent of 2–5 years of normal ageing [70, 73, 78].
A meta-analysis reported regional reductions in
orbitofrontal cortex grey matter, the hippocam-
pus, and basal ganglia [73]. It is questionable,
however, if hippocampal atrophy is really out of
proportion to global atrophy [79]. Several [78,
80], but not all [81], longitudinal studies show
that atrophy in people with type 2 diabetes
exceeds the rate of brain volume loss of normal
ageing, although effects are modest. In this
respect, the trajectories of brain volume changes
and those of diabetes-associated cognitive dec-
rements are much the same. Lower brain vol-
ume is the neuroimaging marker that most
clearly differentiates patients with type 2 diabe-
tes and MCI from those with type 2 diabetes
without MCI [82].
Diffusion tensor imaging (DTI) is a marker of
brain parenchymal injury, primarily used to eval-
uate white matter changes [83]. The technique
allows quantification of global white matter
injury, but can also be used to evaluate the integ-
rity of specific tract abnormalities and brain net-
work integrity [54, 63]. DTI studies in people
Brain Changes in Diabetes and Cognitive Dysfunction
with type 1 and type 2 diabetes show changes in
white matter microstructure and connectivity
compared to controls, which relate to cognitive
dysfunction [67, 84, 85].
Besides brain imaging, alternative biomarker
approaches to detect neurodegeneration are
emerging that may also be relevant for diabetes.
These include assessment of retinal nerve fibre
layer thickness with optical coherence tomogra-
phy [86] and blood-based markers such as neuro-
filament light [87, 88].
3.4 Markers of Cerebral Blood
Flow and Metabolism
Given the fact that diabetes is defined by meta-
bolic disturbances and is known to affect vascular
function throughout the body, biomarkers of
brain blood flow and metabolism are clearly of
interest to explore the effects of diabetes on the
brain. However, as mentioned earlier, a caveat in
the interpretation of abnormal patterns of cere-
bral blood flow and metabolism is that they may
represent both cause and effect of disease pro-
cesses in the brain.
Early studies, mostly with small sample sizes,
showed variable results on cerebral perfusion in
people with diabetes [63, 89, 90]. More recent
studies, primarily involving using arterial spin
labelling MRI, show that at least in people with
type 2 diabetes, global cerebral blood flow
appears to be largely preserved [91]. Yet, there
are reports of regional flow reductions relative to
controls that also relate to worse cognitive per-
formance [92, 93].
Cerebral glucose metabolism is clearly also of
interest in people with diabetes [94]. It is impor-
tant, however, that inter-subject variation in blood
glucose levels may affect tracer uptake in FDG-­
PET and may thus confound results [47].
Advanced glycation end-products (AGEs)
result from nonenzymatic chemical reactions
between reduced sugars and proteins. Formation
of AGEs is accelerated by hyperglycaemia [95].
Circulating AGEs can be measured in plasma, but
their accumulation in tissues can also be esti-
mated through measurement of so-called skin
389
autofluorescence [95]. In a large population-­
based cohort study, both higher levels of the cir-
culating AGEs skin autofluorescence were
associated with worse cognitive performance in
people with and without type 2 diabetes [95].
Another study observed that higher skin autoflu-
orescence was associated with grey matter atro-
phy in people with type 2 diabetes [96].
4 Implications for Clinical
Practice
4.1 Emerging Clinical Guidelines
Over the past years, several professional societies
have published guidelines on the management of
cognitive impairment in relation to diabetes
(Reviewed in Biessels and Whitmer [97]; see also
Text Box 1 [98–103]). Essentially, all of these
guidelines recommend that cognitive impairment
in people with diabetes:
1. Should be actively sought for (by screening)
2. Dictates an individualized diabetes manage-
ment regimen, considering the patient’s capa-
bilities, generally with more lenient treatment
targets and simplified treatment schemes to
improve compliance and reduce treatment-­
related risks
The rationale for these recommendations is
that among people with diabetes those with
­cognitive impairment represent a risk group for
poor health outcomes, that cognitive impairment
hampers diabetes self-management, and that it is
related with a more frequent occurrence of treat-
ment related complications, in particular severe
hypoglycaemic episodes [97]. By screening for
cognitive impairment and offering individualized
care to those affected, these risks might be
mitigated.
The emergence of this guidance is clearly
important. Yet, implementation in clinical prac-
tice, particularly with regard to routine screening
for cognitive impairment, does currently not
appear to be widespread. Further development of
effective and efficient screening strategies that
390
can be used in a diabetes care setting still required,
as well as further evidence that screening and
personalized management of cognitive impair-
ment in people with diabetes indeed improve
patient outcomes [97].
Text Box 1 Key Points from Clinical
Guidelines on Screening, Diagnosis, and
Clinical Management of Cognitive
Impairment Diabetes (Adapted from [97])
International Diabetes Federation
(IDF): Managing Older People with
Type 2 Diabetes Global Guideline
2013 [98]
• Assessment of older people with diabe-
tes should be multidisciplinary, aiming
to collect information on medical, psy-
chosocial, and functional capabilities
and how these may limit activities.
Annual assessment with cognitive
screening tests can be considered.
Enquiry about functional capacity and
cognitive and mental health should be a
standard part of patient assessment.
• In patients with cognitive impairment
treatment targets should be more lenient,
particularly for glycaemic control,
Diabetes and Dementia in Older
People: a Best Clinical Practice
Statement by a Multidisciplinary
National (United Kingdom) Expert
Working Group 2014 [101]
• Routinely employ a brief cognitive
screening test in annual patient
assessments
• In patients with cognitive impairment
therapy risk should be minimized:
–– Avoid overly intensive management
–– Use therapies that reduce risk of
hypoglycaemia
–– Focus education and support on both
patients and carers
The ADA guideline is an update from earlier recommen-
1 
dations [112].
G. J. Biessels
Glycaemic Targets for Elderly
Patients with Diabetes Japan
Diabetes Society (JDS)/Japan
Geriatrics Society (JGS) 2016 [100]
• Individualized targets, with consideration
of age, severity of cognitive impairment
• Relatively higher glycaemic targets for
use of drugs potentially associated with
severe hypoglycaemia
American Diabetes Association
(ADA): “Standards of Medical Care
in Diabetes” 20191 [99]
• Annual screening for early detection of
mild cognitive impairment or dementia
is indicated for adults 65  years of age
and older
• Older adults who are otherwise healthy
with few coexisting chronic illnesses
and intact cognitive function and func-
tional status should have lower glycae-
mic goals (A1C<7.5% [58 mmol/mol]),
while those with multiple coexisting
chronic illnesses, cognitive impairment,
or functional dependence should have
less stringent glycaemic goals
(A1C < 8.0–8.5% [64–69 mmol/mol]).
Endocrine Society: Treatment of
Diabetes in Older Adults 2019 [102]
• Perform cognitive screening in patients
aged 65  years and older. Repeat every
2–3 years after a normal screening test
result for patients without cognitive
complaints or repeat 1 year after a bor-
derline normal test result.
• In patients with a diagnosis of cognitive
impairment (i.e. MCI or dementia), it is
suggested to simplify medication regi-
mens and tailor glycaemic targets (i.e.
less stringent).
For full details see original publications.
Brain Changes in Diabetes and Cognitive Dysfunction
4.2 Diagnosing Cognitive
Dysfunction in People
with Diabetes
The optimal diagnostic strategy in people with
diabetes and cognitive complaints depends on the
prior likelihood of presence of cognitive impair-
ment. This likelihood is clearly strongly related
to age, but also determined by history taking.
Examples of “what went wrong” due to the cog-
nitive complaints can be helpful to estimate the
likely severity of cognitive dysfunction. One
should collect information on impact on profes-
sional and social activities and on other aspects
of daily life. Trajectories of the complaints in
terms of their onset and progression over time
should be determined.
In people below the age of 60  years, with
long-standing (i.e. several years) cognitive com-
plaints, that showed only slow progression over
time, with preserved social and professional
activities, although possibly at the cost of a per-
ceived higher effort, presence of diabetes-­
associated cognitive decrements may be
considered. Accompanying fatigue is not uncom-
mon. Affective complaints should be evaluated
and presence of depression should also be con-
sidered. In case of such subtle cognitive com-
plaints, ancillary investigations are mostly
indicated to rule out alternative explanations.
Neuropsychological examination can be per-
formed to formally quantify the severity of the
cognitive deficits. Brief cognitive screening tests
are generally less useful in this setting, as they
are not sensitive to subtle cognitive changes.
Neuroimaging should be used on indication.
Routine MRI scanning of people with subtle cog-
nitive changes will likely yield a lot of chance
findings, like low-grade white matter hyperinten-
sities or an occasional lacune. This may generate
worries rather than comfort patients. Most impor-
tantly, it should be acknowledged that diabetes-­
associated cognitive decrements are not a formal
diagnostic label, for lack of specific criteria that
can reliably pinpoint affected individuals. This
label should therefore be offered as a possible
explanation for a patient’s complaints rather than
a formal diagnosis [15]. Re-evaluation of the
391
complaints after 6–12  months is generally
recommended.
Patients suspected of the presence of MCI and
dementia clearly warrant another diagnostic
approach [15, 104]. As there are no
­diabetes-­specific features to MCI and dementia,
the same diagnostic tests are indicated as in
patients without diabetes.
4.3 Treatment
As indicated above and summarized in Text Box
1, current guidelines recommend tailored diabe-
tes treatment of people with diabetes and cogni-
tive impairment.
At present there are no treatments that can halt
cognitive decline due to neurodegenerative dis-
eases. Clearly, adequate cardiovascular risk fac-
tor management is indicated, but this should be
guided by the generic diabetes management
guidelines. In patients with manifestations of
vascular injury on MRI, but no history of stroke
or other cardiovascular disease, primary preven-
tion strategies for prevention of cardiovascular
disease generally apply. White matter hyperin-
tensities as such, for example, are no solid indica-
tion for prescription of antiplatelet drugs. The
reader is referred to published recommendations
for the management of such “covert” cerebrovas-
cular disease [105].
At present there is no evidence that maintain-
ing strict glycaemic control can prevent cognitive
decline in people with diabetes [41, 106]. Avoiding
severe hypoglycaemic episodes is important, but
that is already a cornerstone of regular diabetes
care. There is an expanding literature of class
effects of different glucose-­ lowering drugs on
cognitive impairment, reporting benefit or poten-
tial harm for different drugs [107, 108].
Importantly, most of the available data is from
observational studies and possibly subject by
bias, such as confounding by indication.
Interestingly though, more and more head to head
comparisons of different drugs become available,
largely as an spin-off of the large cardiovascular
outcome trials that have been conducted for these
drugs over the past decades. While many of these
392
trials have not reported benefit of one drug over
another (e.g. Refs [109, 110]), a recent study (the
REWIND trial) that compared the glucagon-like
peptide-1 receptor agonist dulaglutide to placebo
found an indication that dulaglutide was superior
in preventing cognitive impairment [111],
although it should be acknowledged that the two
treatment arms in the trial also differed in terms of
HbA1c levels.
References
1. Miles WR, Root HF.  Psychologic tests applied in
diabetic patients. Arch Intern Med. 1922;30:767–77.
2. Cameron FJ, Northam EA, Ryan CM.  The effect
of type 1 diabetes on the developing brain. Lancet
Child Adolesc Health. 2019;3(6):427–36.
3. Cato A, Hershey T.  Cognition and type 1 diabe-
tes in children and adolescents. Diabetes Spectr.
2016;29(4):197–202.
4. Petersen RC, Lopez O, Armstrong MJ, et al. Practice
guideline update summary: mild cognitive impair-
ment: report of the guideline development, dis-
semination, and implementation subcommittee of
the American Academy of Neurology. Neurology.
2018;90(3):126–35.
5. Petersen RC. Mild cognitive impairment as a diag-
nostic entity. J Intern Med. 2004;256(3):183–94.
6. American Psychiatric Association. Diagnostic and
statistical manual of mental disorders, fourth edition 20. van den Berg E, de Craen AJ, Biessels GJ, Gussekloo
(DSM-IV). Washington DC: American Psychiatric
Association; 1994.
7. Jack CR Jr, Bennett DA, Blennow K, et al. NIA-AA
research framework: toward a biological defini-
tion of Alzheimer’s disease. Alzheimers Dement.
2018;14(4):535–62.
8. Jansen WJ, Ossenkoppele R, Knol DL, et  al.
Prevalence of cerebral amyloid pathology in per-
sons without dementia: a meta-analysis. JAMA.
2015;313(19):1924–38.
9. Bateman RJ, Xiong C, Benzinger TL, et  al.
Clinical and biomarker changes in dominantly
inherited Alzheimer’s disease. N Engl J Med.
2012;367(9):795–804.
10. McKhann G, Drachman D, Folstein M, Katzman
R, Price D, Stadlan EM.  Clinical diagnosis of
Alzheimer’s disease: report of the NINCDS-­ 24. Nunley KA, Rosano C, Ryan CM, et al. Clinically
ADRDA Work Group under the auspices of
Department of Health and Human Services
Task Force on Alzheimer’s Disease. Neurology.
1984;34(7):939–44.
11. Ossenkoppele R, Jansen WJ, Rabinovici GD,
et  al. Prevalence of amyloid PET positivity in
dementia syndromes: a meta-analysis. JAMA.
2015;313(19):1939–49.
G. J. Biessels
12. Kapasi A, DeCarli C, Schneider JA. Impact of mul-
tiple pathologies on the threshold for clinically overt
dementia. Acta Neuropathol. 2017;134(2):171–86.
13. Brands AM, Biessels GJ, de Haan EH, Kappelle
LJ, Kessels RP.  The effects of type 1 diabetes on
cognitive performance: a meta-analysis. Diabetes
­
Care. 2005;28(3):726–35.
14. Palta P, Schneider AL, Biessels GJ, Touradji P, Hill-­
Briggs F.  Magnitude of cognitive dysfunction in
adults with type 2 diabetes: a meta-analysis of six
cognitive domains and the most frequently reported
neuropsychological tests within domains. J Int
Neuropsychol Soc. 2014;20(3):278–91.
15. Koekkoek PS, Kappelle LJ, van den Berg E, Rutten
GE, Biessels GJ. Cognitive function in patients with
diabetes mellitus: guidance for daily care. Lancet
Neurol. 2015;14(3):329–40.
16. Biessels GJ, Strachan MW, Visseren FL, Kappelle
LJ, Whitmer RA.  Dementia and cognitive decline
in type 2 diabetes and prediabetic stages: towards
targeted interventions. Lancet Diabetes Endocrinol.
2014;2(3):246–55.
17. Biessels GJ, Despa F. Cognitive decline and dementia
in diabetes mellitus: mechanisms and clinical impli-
cations. Nat Rev Endocrinol. 2018;14(10):591–604.
18. Brady CC, Vannest JJ, Dolan LM, et al. Obese ado-
lescents with type 2 diabetes perform worse than
controls on cognitive and behavioral assessments.
Pediatr Diabetes. 2017;18(4):297–303.
19. Yates KF, Sweat V, Yau PL, Turchiano MM, Convit
A. Impact of metabolic syndrome on cognition and
brain: a selected review of the literature. Arterioscler
Thromb Vasc Biol. 2012;32(9):2060–7.
J, Westendorp RG.  The impact of diabetes melli-
tus on cognitive decline in the oldest of the old: a
prospective population-based study. Diabetologia.
2006;49(9):2015–23.
21. Prince M, Bryce R, Albanese E, Wimo A, Ribeiro
W, Ferri CP.  The global prevalence of dementia: a
systematic review and metaanalysis. Alzheimers
Dement. 2013;9(1):63–75 e2.
22. Li W, Huang E, Gao S.  Type 1 diabetes mellitus
and cognitive impairments: a systematic review. J
Alzheimers Dis. 2017;57(1):29–36.
23. Jacobson AM, Ryan CM, Braffett BH, et  al.
Cognitive performance declines in older adults with
type 1 diabetes: results from 32 years of follow-
­up in the DCCT and EDIC Study. Lancet Diabetes
Endocrinol. 2021;9(7):436–45.
relevant cognitive impairment in middle-aged adults
with childhood-onset type 1 diabetes. Diabetes Care.
2015;38(9):1768–76.
25. Smolina K, Wotton CJ, Goldacre MJ. Risk of demen-
tia in patients hospitalised with type 1 and type 2
diabetes in England, 1998-2011: a retrospective
national record linkage cohort study. Diabetologia.
2015;58(5):942–50.
Brain Changes in Diabetes and Cognitive Dysfunction
26. Lacy ME, Gilsanz P, Karter AJ, Quesenberry CP,
Pletcher MJ, Whitmer RA. Long-term glycemic con-
trol and dementia risk in type 1 diabetes. Diabetes
Care. 2018;41(11):2339–45.
27. Whitmer RA, Gilsanz P, Quesenberry CP, Karter
AJ, Lacy ME.  Association of type 1 diabetes and
hypoglycemic and hyperglycemic events and risk of
dementia. Neurology. 2021;97(3):e275–83.
28. Rodill LG, Exalto LG, Gilsanz P, Biessels GJ,
Quesenberry CP Jr, Whitmer RA. Diabetic retinopa-
thy and dementia in type 1 diabetes. Alzheimer Dis
Assoc Disord. 2018;32(2):125–30.
29. Gilsanz P, Schnaider Beeri M, Karter AJ, Quesenberry
CP Jr, Adams AS, Whitmer RA. Depression in type
1 diabetes and risk of dementia. Aging Ment Health.
2019;23(7):880–6.
30. Monette MC, Baird A, Jackson DL. A meta-­analysis
of cognitive functioning in nondemented adults
with type 2 diabetes mellitus. Can J Diabetes.
2014;38(6):401–8.
31. Pappas C, Andel R, Infurna FJ, Seetharaman
S.  Glycated haemoglobin (HbA1c), diabetes
and trajectories of change in episodic memory
performance. J Epidemiol Community Health.
2017;71(2):115–20.
32. Yaffe K, Falvey C, Hamilton N, et  al. Diabetes,
glucose control, and 9-year cognitive decline
among older adults without dementia. Arch Neurol.
2012;69(9):1170–5.
33. Rawlings AM, Sharrett AR, Albert MS, et  al. The
association of late-life diabetes status and hyper-
glycemia with incident mild cognitive impairment
and dementia: the ARIC study. Diabetes Care.
2019;42(7):1248–54.
34. Xue M, Xu W, Ou YN, et al. Diabetes mellitus and
risks of cognitive impairment and dementia: a sys-
tematic review and meta-analysis of 144 prospective
studies. Ageing Res Rev. 2019;55:100944.
35. Pal K, Mukadam N, Petersen I, Cooper C.  Mild
cognitive impairment and progression to demen-
tia in people with diabetes, prediabetes and
metabolic syndrome: a systematic review and
meta-analysis. Soc Psychiatry Psychiatr Epidemiol.
2018;53(11):1149–60.
36. Biessels GJ, Staekenborg S, Brunner E, Brayne C,
Scheltens P. Risk of dementia in diabetes mellitus: a
systematic review. Lancet Neurol. 2006;5(1):64–74.
37. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne
C.  Potential for primary prevention of Alzheimer’s
disease: an analysis of population-based data. Lancet
Neurol. 2014;13(8):788–94.
38. Deckers K, van Boxtel MP, Schiepers OJ, et  al.
Target risk factors for dementia prevention: a sys-
tematic review and Delphi consensus study on the
evidence from observational studies. Int J Geriatr
Psychiatry. 2015;30(3):234–46.
39. Exalto LG, Quesenberry CP, Barnes D, Kivipelto
M, Biessels GJ, Whitmer RA.  Midlife risk score
for the prediction of dementia four decades later.
Alzheimers Dement. 2014;10(5):562–70.
393
40. Yu JH, Han K, Park S, et  al. Incidence and risk
factors for dementia in type 2 diabetes mellitus:
a nationwide population-based study in Korea.
Diabetes Metab J. 2020;44(1):113–24.
41. Geijselaers SLC, Sep SJS, Stehouwer CDA, Biessels
GJ. Glucose regulation, cognition, and brain MRI in
type 2 diabetes: a systematic review. Lancet Diabetes
Endocrinol. 2015;3(1):75–89.
42. Janssen J, van den Berg E, Zinman B, et al. HbA1c,
insulin resistance, and beta-cell function in rela-
tion to cognitive function in type 2 diabetes: the
CAROLINA cognition substudy. Diabetes Care.
2019;42(1):e1–3.
43. Larsson SC, Traylor M, Malik R, et al. Modifiable
pathways in Alzheimer’s disease: Mendelian ran-
domisation analysis. BMJ. 2017;359:j5375.
44. Benn M, Nordestgaard BG, Tybjaerg-Hansen
A, Frikke-Schmidt R.  Impact of glucose on
risk of dementia: Mendelian randomisation
studies in 115,875 individuals. Diabetologia.
2020;63(6):1151–61.
45. James SN, Wong A, Tillin T, Hardy R, Chaturvedi
N, Richards M. The effect of mid-life insulin resis-
tance and type 2 diabetes on older-age cognitive
state: the explanatory role of early-life advantage.
Diabetologia. 2019;62(10):1891–900.
46. Liu J, Rutten-Jacobs L, Liu M, Markus HS, Traylor
M.  Causal impact of type 2 diabetes mellitus on
cerebral small vessel disease: a Mendelian random-
ization analysis. Stroke. 2018;49(6):1325–31.
47. Biessels GJ, Nobili F, Teunissen CE, Simo R,
Scheltens P.  Understanding multifactorial brain
changes in type 2 diabetes: a biomarker perspective.
Lancet Neurol. 2020;19(8):699–710.
48. Smith R, Scholl M, Leuzy A, et  al. Head-to-head
comparison of tau positron emission tomography
tracers [(18)F]flortaucipir and [(18)F]RO948. Eur J
Nucl Med Mol Imaging. 2020;47(2):342–54.
49. Herholz K, Ebmeier K.  Clinical amyloid imag-
ing in Alzheimer’s disease. Lancet Neurol.
2011;10(7):667–70.
50. Palmqvist S, Janelidze S, Stomrud E, et  al.
Performance of fully automated plasma assays
as screening tests for Alzheimer disease-related
beta-­amyloid status. JAMA Neurol. 2019;76(9):
1060–9.
51. Abner EL, Nelson PT, Kryscio RJ, et al. Diabetes is
associated with cerebrovascular but not Alzheimer’s
disease neuropathology. Alzheimers Dement.
2016;12(8):882–9.
52. Dos Santos Matioli MNP, Suemoto CK, Rodriguez
RD, et  al. Diabetes is not associated with
Alzheimer’s disease neuropathology. J Alzheimers
Dis. 2017;60(3):1035–43.
53. Gottesman RF, Schneider AL, Zhou Y, et  al.
Association between midlife vascular risk factors
and estimated brain amyloid deposition. JAMA.
2017;317(14):1443–50.
54. Wardlaw JM, Smith EE, Biessels GJ, et  al.
Neuroimaging standards for research into small ves-
394
sel disease and its contribution to ageing and neu-
rodegeneration. Lancet Neurol. 2013;12(8):822–38.
55. Zwanenburg JJM, van Osch MJP.  Targeting cere-
bral small vessel disease with MRI.  Stroke.
2017;48(11):3175–82.
56. O’Brien JT, Thomas A. Vascular dementia. Lancet.
2015;386(10004):1698–706.
57. Wardlaw JM, Smith C, Dichgans M.  Small ves-
sel disease: mechanisms and clinical implications.
Lancet Neurol. 2019;18(7):684–96.
58. van Sloten TT, Sedaghat S, Carnethon MR, Launer
LJ, Stehouwer CDA.  Cerebral microvascular
complications of type 2 diabetes: stroke, cogni-
tive dysfunction, and depression. Lancet Diabetes
Endocrinol. 2020;8(4):325–36.
59. Thorn LM, Shams S, Gordin D, et al. Clinical and
MRI features of cerebral small-vessel disease in type
1 diabetes. Diabetes Care. 2019;42(2):327–30.
60. Fazekas F, Chawluk JB, Alavi A, Hurtig HI,
Zimmerman RA. MR signal abnormalities at 1.5 T
in Alzheimer’s dementia and normal aging. AJR Am
J Roentgenol. 1987;149(2):351–6.
61. Nunley KA, Ryan CM, Orchard TJ, et  al. White
matter hyperintensities in middle-aged adults
with childhood-onset type 1 diabetes. Neurology.
2015;84(20):2062–9.
62. Woerdeman J, van Duinkerken E, Wattjes MP,
et al. Proliferative retinopathy in type 1 diabetes is
associated with cerebral microbleeds, which is part
of generalized microangiopathy. Diabetes Care.
2014;37(4):1165–8.
63. Brundel M, Kappelle LJ, Biessels GJ.  Brain imag-
ing in type 2 diabetes. Eur Neuropsychopharmacol.
2014;24(12):1967–81.
64. Pruzin JJ, Nelson PT, Abner EL, Arvanitakis
Z.  Invited review: relationship of type 2 diabe-
tes to human brain pathology. Neuropathol Appl
Neurobiol. 2018;44(4):347–62.
65. Blevins BL, Vinters HV, Love S, et  al. Brain arte-
riolosclerosis. Acta Neuropathol. 2021;141(1):1–24.
66. Gerstein HC, Smith EE, Ramasundarahettige
C, et  al. Diabetes, brain infarcts, cognition, and
small vessels in the Canadian Alliance for Healthy
Hearts and Minds study. J Clin Endocrinol Metab.
2021;106(2):e891–e8.
67. Moran C, Beare R, Phan T, et al. Neuroimaging and
its relevance to understanding pathways linking dia-
betes and cognitive dysfunction. J Alzheimers Dis.
2017;59(2):405–19.
68. Traylor M, Persyn E, Tomppo L, et al. Genetic basis
of lacunar stroke: a pooled analysis of individual
patient data and genome-wide association studies.
Lancet Neurol. 2021;20(5):351–61.
69. Garfield V, Farmaki AE, Fatemifar G, et  al. The
relationship between glycaemia, cognitive function,
structural brain outcomes and dementia: a Mendelian
Randomisation Study in the UK Biobank. Diabetes.
2021;70(10):2313–21.
70. Schneider ALC, Selvin E, Sharrett AR, et al. Diabetes,
prediabetes, and brain volumes and subclinical cere-
G. J. Biessels
brovascular disease on MRI: the Atherosclerosis Risk
in Communities Neurocognitive Study (ARIC-­NCS).
Diabetes Care. 2017;40(11):1514–21.
71. Qiu C, Sigurdsson S, Zhang Q, et al. Diabetes, mark-
ers of brain pathology and cognitive function: the
Age, Gene/Environment Susceptibility-Reykjavik
Study. Ann Neurol. 2014;75(1):138–46.
72. Puri BK, Lewis HJ, Saeed N, Davey NJ. Volumetric
change of the lateral ventricles in the human
brain following glucose loading. Exp Physiol.
1999;84(1):223–6.
73. Moulton CD, Costafreda SG, Horton P, Ismail K, Fu
CH.  Meta-analyses of structural regional cerebral
effects in type 1 and type 2 diabetes. Brain Imaging
Behav. 2015;9(4):651–62.
74. Frokjaer JB, Brock C, Softeland E, et  al.
Macrostructural brain changes in patients with long-
standing type 1 diabetes mellitus—a cortical thick-
ness analysis study. Exp Clin Endocrinol Diabetes.
2013;121(6):354–60.
75. Hughes TM, Ryan CM, Aizenstein HJ, et al. Frontal
gray matter atrophy in middle aged adults with
type 1 diabetes is independent of cardiovascular
risk factors and diabetes complications. J Diabetes
Complications. 2013;27(6):558–64.
76. Musen G, Lyoo IK, Sparks CR, et  al. Effects of
type 1 diabetes on gray matter density as mea-
sured by voxel-based morphometry. Diabetes.
2006;55(2):326–33.
77. Franc DT, Kodl CT, Mueller BA, Muetzel RL,
Lim KO, Seaquist ER.  High connectivity between
reduced cortical thickness and disrupted white mat-
ter tracts in long-standing type 1 diabetes. Diabetes.
2011;60(1):315–9.
78. de Bresser J, Tiehuis AM, van den Berg E, et  al.
Progression of cerebral atrophy and white matter
hyperintensities in patients with type 2 diabetes.
Diabetes Care. 2010;33(6):1309–14.
79. Wisse LE, de Bresser J, Geerlings MI, et al. Global
brain atrophy but not hippocampal atrophy is related
to type 2 diabetes. J Neurol Sci. 2014;344(1–2):32–6.
80. Espeland MA, Bryan RN, Goveas JS, et al. Influence
of type 2 diabetes on brain volumes and changes in
brain volumes: results from the Women’s Health
Initiative Magnetic Resonance Imaging studies.
Diabetes Care. 2013;36(1):90–7.
81. Callisaya ML, Beare R, Moran C, Phan T, Wang W,
Srikanth VK. Type 2 diabetes mellitus, brain atrophy
and cognitive decline in older people: a longitudinal
study. Diabetologia. 2019;62(3):448–58.
82. Groeneveld O, Reijmer Y, Heinen R, et  al. Brain
imaging correlates of mild cognitive impair-
ment and early dementia in patients with type 2
diabetes mellitus. Nutr Metab Cardiovasc Dis.
2018;28(12):1253–60.
83. Reijmer YD, Brundel M, de Bresser J, et  al.
Microstructural white matter abnormalities and
cognitive functioning in type 2 diabetes: a dif-
fusion tensor imaging study. Diabetes Care.
2013;36(1):137–44.
Brain Changes in Diabetes and Cognitive Dysfunction
84. Sanjari Moghaddam H, Ghazi Sherbaf F, Aarabi
MH.  Brain microstructural abnormalities in type 2
diabetes mellitus: a systematic review of diffusion
tensor imaging studies. Front Neuroendocrinol.
2019;55:100782.
85. Alotaibi A, Tench C, Stevenson R, et al. Investigating
brain microstructural alterations in type 1 and type 2
diabetes using diffusion tensor imaging: a system-
atic review. Brain Sci. 2021;11(2):140.
86. den Haan J, Verbraak FD, Visser PJ, Bouwman
FH.  Retinal thickness in Alzheimer’s disease: a
systematic review and meta-analysis. Alzheimers
Dement (Amst). 2017;6:162–70.
87. Khalil M, Teunissen CE, Otto M, et  al.
Neurofilaments as biomarkers in neurological disor-
ders. Nat Rev Neurol. 2018;14(10):577–89.
88. Bridel C, van Wieringen WN, Zetterberg H, et  al.
Diagnostic value of cerebrospinal fluid neuro-
filament light protein in neurology: a system-
atic review and meta-analysis. JAMA Neurol.
2019;76(9):1035–48.
89. van Golen LW, Huisman MC, Ijzerman RG, et  al.
Cerebral blood flow and glucose metabolism mea-
sured with positron emission tomography are
decreased in human type 1 diabetes. Diabetes.
2013;62(8):2898–904.
90. Mankovsky BN, Metzger BE, Molitch ME,
Biller J.  Cerebrovascular disorders in patients
with diabetes mellitus. J Diabetes Complications.
1997;10:228–42.
91. Jansen JF, van Bussel FC, van de Haar HJ, et  al.
Cerebral blood flow, blood supply, and cognition in
type 2 diabetes mellitus. Sci Rep. 2016;6(1):10.
92. Cui Y, Liang X, Gu H, et al. Cerebral perfusion alter-
ations in type 2 diabetes and its relation to insulin
resistance and cognitive dysfunction. Brain Imaging
Behav. 2017;11(5):1248–57.
93. Bangen KJ, Werhane ML, Weigand AJ, et  al.
Reduced regional cerebral blood flow relates to
poorer cognition in older adults with type 2 diabetes.
Front Aging Neurosci. 2018;10:270.
94. Roberts RO, Knopman DS, Cha RH, et al. Diabetes
and elevated hemoglobin A1c levels are associated
with brain hypometabolism but not amyloid accu-
mulation. J Nucl Med. 2014;55(5):759–64.
95. Spauwen PJ, van Eupen MG, Kohler S, et  al.
Associations of advanced glycation end-products
with cognitive functions in individuals with and
without type 2 diabetes: the maastricht study. J Clin
Endocrinol Metab. 2015;100(3):951–60.
96. Moran C, Munch G, Forbes JM, et  al. Type 2 dia-
betes, skin autofluorescence, and brain atrophy.
Diabetes. 2015;64(1):279–83.
97. Biessels GJ, Whitmer RA.  Cognitive dysfunction
in diabetes: how to implement emerging guidelines.
Diabetologia. 2020;63(1):3–9.
98. Federation ID.  Global guideline for managing
older people with type 2 diabetes. 2013. https://
www.idf.org/e-­l ibrary/guidelines/78-­g lobal-­
guideline-­f or-­m anaging-­o lder-­p eople-­w ith-­t ype-­
2-­diabetes.html.
395
99. American Diabetes Association. 12. Older adults:
standards of medical care in diabetes-2019. Diabetes
Care. 2019;42(Suppl 1):S139–S47.
100. Committee Report: Glycemic targets for elderly
patients with diabetes: Japan Diabetes Society (JDS)/
Japan Geriatrics Society (JGS) Joint Committee on
Improving Care for Elderly Patients with Diabetes. J
Diabetes Investig. 2017;8(1):126–8.
101. Sinclair AJ, Hillson R, Bayer AJ.  Diabetes and
dementia in older people: a Best Clinical Practice
Statement by a multidisciplinary National Expert
Working Group. Diabet Med. 2014;31(9):1024–31.
102. LeRoith D, Biessels GJ, Braithwaite SS, et  al.
Treatment of diabetes in older adults: an Endocrine
Society* clinical practice guideline. J Clin
Endocrinol Metab. 2019;104(5):1520–74.
103. Brown AF, Mangione CM, Saliba D, Sarkisian
CA. Guidelines for improving the care of the older
person with diabetes mellitus. J Am Geriatr Soc.
2003;51(5 Suppl Guidelines):S265–S80.
104. Srikanth V, Sinclair AJ, Hill-Briggs F, Moran
C, Biessels GJ.  Type 2 diabetes and cognitive
dysfunction-towards effective management of
both comorbidities. Lancet Diabetes Endocrinol.
2020;8(6):535–45.
105. Smith EE, Saposnik G, Biessels GJ, et  al.
Prevention of stroke in patients with silent cere-
brovascular disease: a scientific statement for
Healthcare Professionals From the American Heart
Association/American Stroke Association. Stroke.
2017;48(2):e44–71.
106. Areosa Sastre A, Vernooij RW, Gonzalez-Colaco
Harmand M, Martinez G. Effect of the treatment of
type 2 diabetes mellitus on the development of cog-
nitive impairment and dementia. Cochrane Database
Syst Rev. 2017;6:CD003804.
107. Patrone C, Eriksson O, Lindholm D. Diabetes drugs
and neurological disorders: new views and thera-
peutic possibilities. Lancet Diabetes Endocrinol.
2014;2(3):256–62.
108. McMillan JM, Mele BS, Hogan DB, Leung
AA.  Impact of pharmacological treatment of dia-
betes mellitus on dementia risk: systematic review
and meta-analysis. BMJ Open Diabetes Res Care.
2018;6(1):e000563.
109. Biessels GJ, Verhagen C, Janssen J, et al. Effects of
linagliptin vs glimepiride on cognitive performance
in type 2 diabetes: results of the randomised double-­
blind, active-controlled CAROLINA-COGNITION
study. Diabetologia. 2021;64(6):1235–45.
110. Biessels GJ, Verhagen C, Janssen J, et  al. Effect
of linagliptin on cognitive performance in patients
with type 2 diabetes and cardiorenal comorbidities:
the CARMELINA randomized trial. Diabetes Care.
2019;42(10):1930–8.
111. Gerstein HC, Hart R, Colhoun HM, et al. The effect
of dulaglutide on stroke: an exploratory analysis
of the REWIND trial. Lancet Diabetes Endocrinol.
2020;8(2):106–14.
112. Kirkman MS, Briscoe VJ, Clark N, et al. Diabetes in
older adults. Diabetes Care. 2012;35(12):2650–64.
 Lifestyle and Dietary
Modifications: Relevance
in the Management of Diabetic
Neuropathy
Jonathan Enders and Douglas E. Wright
1 Introduction
Diabetic peripheral neuropathy (DPN) is a com-
mon complication of prediabetes, type 1 diabetes
mellitus (T1DM), and type 2 diabetes mellitus
(T2DM). DPN remains largely irreversible, and
there are no known curative treatments for
patients with DPN [1, 2]. The discovery of preci-
sion therapeutics remains an important goal since
a clear understanding of the breadth of pathogen-
esis remains elusive. It is known that strict glyce-
mic control using oral agents or insulin therapy
can slow or prevent the progression of DPN [3,
4]. Worldwide, the incidence of diabetes and obe-
sity is linked and increasing dramatically, helping
to drive an increase in the prevalence of
DPN.  Driven by axon degeneration, demyelin-
ation, and nerve dysfunction, common symptoms
of DPN include loss of sensation, limb pain, gait
disturbance, and autonomic dysfunction [5].
Collectively, these symptoms reduce the quality
of life for patients with DPN.  This chapter will
highlight experimental approaches to explore
dietary interventions and lifestyle changes to
treat DPN.  The chapter will also address their
limitations and discuss how these interventions
J. Enders ∙ D. E. Wright (*)
Department of Anesthesiology, University of Kansas
Medical Center, Kansas City, KS, USA
e-mail: dwright@kumc.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_22
could lead to the identification of new drug
targets.
2 Pharmacologic Treatment
of Diabetic Peripheral
Neuropathy
For many patients with T2DM, the biguanide
metformin is commonly recommended as first
line therapy following lifestyle modifications.
However, other medications—including insulin
injection, alpha-glucosidase inhibitors, dipepti-
dyl peptidase-4 (DPP-4) inhibitors, glucagon-­
like peptide-1 receptor agonists, sodium-glucose
transporter (SGLT) 2 inhibitors—can also be
effective by themselves or in combination ther-
apy [6]. Glycemic control in patients with T1DM
is generally addressed with insulin injections [7].
Achieving tight glycemic control is challenging,
as meaningful reductions in blood glucose is the
desired goal, but this outcome is strongly influ-
enced by patient compliance, activity level, diet,
and overall health.
For pain associated with DPN, there are sev-
eral pharmacologic therapies used with variable
success. A subset (10–26%) of patients with DPN
report pain that is described as stabbing, electri-
cal pain, pins and needles, or burning sensations
[8]. It is thought that tight glycemic control is not
sufficient to completely prevent painful symp-
toms [9, 10]. Pain medications include tricyclic
397
398
antidepressants (TCAs), serotonin and
­norepinephrine reuptake inhibitors (SNRIs), and
anticonvulsants (gabapentin or pregabalin) [11,
12]. As an over-the-counter supplement, α-lipoic
acid has shown moderate efficacy in studies of
DPN-­associated pain [13, 14]. Finally, topical
treatments can be used that includes application
of a lidocaine patch or capsaicin cream [15].
There are also other off-label medications used to
treat painful DPN, including herbal supplements
like nutmeg and Citrullus colocynthis extract [16,
17]. Opioids generally are contraindicated for
DPN-associated pain. Overall, these classes of
medications are not sufficiently effective and
commonly have physiological or psychological
side effects that limit their use. Thus, additional
treatments to clinically treat DPN are needed.
Hopefully, when coupled with lifestyle or dietary
changes, the effectiveness of these existing
approaches may be improved.
3 Obesity and Diabetic
Peripheral Neuropathy
It is important to understand the role of obesity in
DPN, as obesity is increasing rapidly and is rele-
vant to lifestyle and dietary modifications that
impact DPN.  Both preclinical and clinical
research have clearly identified obesity as a key
risk factor for DPN.  Recent work by Callahan
and colleagues has shown in studies from various
countries that obesity strongly associates with
DPN [11]. Other metabolic syndrome compo-
nents, such as hypertriglyceridemia, hyperten-
sion, and low high-density lipoprotein cholesterol,
appear less strongly commonly associated with
DPN compared with obesity and hyperglycemia.
Importantly normoglycemic patients with obe-
sity have a higher prevalence of DPN than lean,
healthy patients, strongly suggesting obesity
itself is an independent risk factor of DSP [11].
This point should strongly influence preclinical
studies (discussed below) that use dietary inter-
ventions to model DPN. Weight loss and exercise
are still believed to be one of the best treatments
for patients that suffer from obesity-related DPN
[18]. The lifestyle and dietary changes described
J. Enders and D. E. Wright
below in this chapter should all consider obesity
as a key factor in DPN improvements.
4 High-Fat Diets as a Tool
to Study Diabetic Peripheral
Neuropathy
A healthy diet provides clear benefit to prevent-
ing diabetic neuropathy and using a high-fat diet
(HFD) in the setting of DPN has revealed impor-
tant insights into mechanisms and outcomes
associated with a poor diet. Poor diet is risk fac-
tor for the development of T2DM but has the
potential to be modified. Using a high-fat diet in
animal models has revealed extensive informa-
tion about how elevated carbohydrates and fats
can lead to symptoms that strongly resemble pre-
diabetes and, if carried out long enough, manifes-
tations of diabetes and neuropathy. This has been
covered extensively in a number of reviews [19].
One advantage of using dietary interventions is
that neuropathy progression is more gradual and
leads to metabolic imbalances resembling the
natural progression from obesity to T2DM.  In
addition, HFDs also provide stronger relevance
to increased dietary fat/dyslipidemia in both met-
abolic syndrome in patients with prediabetes and
T2DM patients [20–23].
In general terms, chronic use of experimental
HFD paradigms can lead to progressive impair-
ments in sensori-behavioral changes, nerve con-
duction, and nerve pathology. These dietary
formulations generally range from 25% fat/40%
carbohydrates up to 60%fat/20% carbohydrates
[19] (24% from fat, 41% from carbohydrates).
Common complications in HFD-fed mice include
weight gain and glucose intolerance, and as well
as impairments in aforementioned neurological
impairments that include, including behavioral
(hindpaw thermal and mechanical threshold
changes), slowed nerve conduction velocity, and
sensory nerve terminal degeneration (intraepider-
mal nerve fiber density: IENFD), leading to
reduced innervation of the cutaneous tissue [19].
Some of the potential mechanisms identified that
contribute to these deficits include weight gain,
insulin resistance of sensory ­neurons [24, 25],
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
dyslipidemia [20], mitochondrial deficits [26],
and growth factor changes [27].
In addition to mice, adult Sprague-Dawley
(SD) rats have been studied extensively when
exposed to a HFD.  These studies have largely
revealed similar findings that include abnormal
changes that include the appearance of mechani-
cal allodynia, reduced thermal nociception as
decreased IENFD and changes in subepithelial
corneal nerves [28]. Another rat model that
experimentally highlights metabolic complica-
tions includes rats selectively bred for high (high-­
capacity runners; HCR) or low endurance
exercise capacity (low-capacity runners; LCR)
[29]. Selective breeding of these rats has resulted
in divergent intrinsic aerobic capacities and sus-
ceptibility for metabolic complications. Analysis
of these rats revealed that LCR female rats had
higher overall fat content, surprisingly higher
IENFD, reduced mechanical sensitivity, and
increased numbers of mast and Langerhans cells
in skin compared to female HCR rats. These
results suggest that poor metabolic status and
reduced aerobic capacity influence peripheral
sensation and inflammation in peripheral tissues
that are involved in DPN [29]. Continued study
of the HCR and LCR rats should prove fruitful to
understand how metabolic complications impact
sensory function and pain. Finally, sex differ-
ences in the quality and sensitivity of male and
female rats have been reported [30], and potential
sex differences likely require increased rigor and
attention. Overall, the use of dietary intervention
and genetic breeding to model metabolism asso-
ciated DPN features has provided unique insight
to the role of obesity, insulin resistance, dyslipid-
emia, inflammation in DPN.
Diet is critical in modifying the redox state of
sensory neurons and, thus, the progression of
DPN. Consumption of a HFD leads to hypergly-
cemia and dyslipidemia. Hyperglycemia results
in the accumulation of toxic metabolites, such as
sorbitol [31, 32] and reactive dicarbonyls that
include methylglyoxal and glyoxal [33]. These
toxic metabolites increase the generation of [34]
and impair the scavenging of ROS [35–37].
HFD-induced dyslipidemia is also a key risk fac-
tor for DPN, resulting in elevated ROS genera-
399
tion [38] and poor mitochondrial trafficking/
health [39–41]. Relevant to DPN, ROS can also
impact several transient receptor potential (TRP)
channels [42–45] and also sensitize voltage-gated
channels [46, 47] to evoke abnormal sensory sig-
naling that is evident in DPN.
5 Dietary Interventions
to Improve Diabetic
Peripheral Neuropathy
5.1 Mediterranean Diets
Although relatively understudied, a dietary inter-
vention gaining preclinical interest in DPN is the
Mediterranean diet. The Mediterranean diet is
less stringent compared to other diets in terms of
macronutrient abundance. Namely, the
Mediterranean diet is rich in whole grains, vege-
tables, and legumes, with olive oils as a primary
source of fats and poultry and fish as the main
source of protein. As a result, adherence to a
Mediterranean diet results in reduced sucrose
intake and elevated consumption of monounsatu-
rated and omega-3 polyunsaturated fatty acids.
One potential benefit of the Mediterranean
diet for treating DPN is relieving hyperglycemia.
Two studies from Esposito and colleagues sug-
gest that patients with T2DM consuming a
Mediterranean diet were less likely to experience
hyperglycemia and exhibit reduced hemoglobin
A1C.  These benefits may in part be due to
improved insulin sensitivity [48, 49]. With whole
grain as the primary carbohydrate source of
Mediterranean diets, sucrose and fructose intake
are reduced, which has been linked preclinically
to spared insulin resistance and rescued neuro-
pathic symptoms such as mechanical allodynia
and intraepidermal fiber loss [50, 51].
Mediterranean diets may also offer benefit in
DPN by reversing dyslipidemia. Feldman and
colleagues have demonstrated many harmful
effects of dyslipidemia in DPN, such as poor
mitochondrial trafficking and health, that contrib-
ute to disease progression [39, 40]. Preclinical
studies by Rumora et  al. reveal that mice fed a
HFD composed primarily of saturated fatty acids
400 J. Enders and D. E. Wright

go on to develop symptoms of peripheral neurop-
athy, while mice fed a diet high in monounsatu-
rated fats, like those found in olive oil, did not
develop neuropathy [52]. Moreover, monounsatu-
rated fat-rich diets reversed the neuropathy caused
by high saturated fat diets in these animals and
were able to rescue mitochondrial motility and
health [52]. Thus, preclinical models that resem-
ble a Mediterranean diet prevent and reverse the
onset of DPN and obesity-related neuropathies.
While there is accumulating preclinical evi-
dence that suggests potential efficacy of a
Mediterranean diet in treating DPN, to the best of
our knowledge no explicit clinical study of
Mediterranean diet in DPN or other neuropathic
conditions has been completed. There is, how-
ever, limited evidence that increased omega-3
fatty acid consumption is negatively correlated
with incidence of DPN [53], again supporting a
putative protective effect for Mediterranean diets.
There remains, however, a significant gap in
research that precludes a definitive statement
about the efficacy of Mediterranean diets as inter-
ventions for DPN.
5.2 Ketogenic Diets
Ketogenic diets are very low carbohydrates that
lead to ketogenesis by the liver. Ketogenic diets
have been used to treat neurodegenerative disor-
ders [54, 55], epilepsy and seizures [56, 57], and
chronic pain [58–60]. The outcome of a keto-
genic diet is to displace the carbohydrate with
oils and fats, and the resulting effect of excess fat
consumption causes increased liver fatty acid
oxidation and production of the two major ketone
bodies β-hydroxybutyrate (β-OHB) and acetoac-
etate. Once synthesized, these two ketone bodies
enter circulation and are taken up by extra-hepatic
tissues to replace glucose as a principal fuel
source.
Because ketogenic diets reduce glucose intake
and circulating levels of glucose, ketogenic diets

Fig. 1 Systemic
changes induced by a
ketogenic diet. A Reduced
ketogenic diet induces Hyperglycemia
widespread benefits to
many organ systems
involved in
DPN. Reduced
consumption of Detoxify Toxic
carbohydrates due to Metabolites
consuming a ketogenic
diet reduces
hyperglycemia and
accumulation of
advanced glycation Nutritional Reduced
endproducts. Moreover,
Ketosis Inflammation
ketogenic diets
contribute to
detoxification of toxic
metabolites, reduced
inflammation and Reduced
generation of reactive Oxidative Stress
oxygen species, and
support peripheral axon
growth

Support
Peripheral
Axon Growth
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
have been used in the treatment of patients with
diabetes (Fig. 1). Prior to the widespread use of
insulin treatment, low-carbohydrate diets were
commonly used to reduce elevated glucose levels
through dietary reduction of carbohydrates in
mild cases of diabetes and more severe elimina-
tion of protein and carbohydrate consumption in
severe cases of diabetes [61]. Patients with diabe-
tes that were fed low-carbohydrate meals did not
exhibit post-prandial hyperglycemia similar to
diabetic patients consuming a standard diet [62].
Importantly, these diets did not originally replace
lost caloric intake from reduced carbohydrate
consumption with elevated fat intake, leading to
severe morbidity. Complications of this starva-
tion approach often included patient weakness
and death. In the early 1920s, it became more
common to use high-fat, low-carbohydrate diets
in weak and starvation-resistant diabetics [63].
These early low-carbohydrate diets resemble
modern ketogenic diets, and their use was suc-
cessful in improving diabetes parameters in these
patients (Fig. 1). However, following the discov-
ery of and practice of using insulin, emphasis on
high-fat, low-carbohydrate dietary interventions
for diabetes became less common.
5.2.1 Addressing Health Concerns
of Using a Ketogenic Diet
in Patients with Diabetes
Consumption of low-carbohydrate/high-fat diets
generates appropriate health concerns, as there is
the potential to cause diabetic ketoacidosis, a
common life-threatening complication of dia-
betes. Diabetic ketoacidosis is associated with
an accumulation of excess blood ketone bodies
and lactate. β-OHB, acetoacetate, and lactate
deprotonate at physiologic pH, causing reduc-
tions in blood pH. With a reduction in carbohy-
drate consumption, however, lactate production
is actually decreased [64], thus physiologically
sparing patients from acidosis. In fact, acidosis
was reversed in patients with diabetes consum-
ing very low-­carbohydrate, high-fat diets [63].
Additional studies in humans reveal [57, 65, 66]
the benefits of ketogenic diets in diabetes and
401
their safety in patients with diabetes. These clini-
cal studies are strongly consistent with findings
in preclinical models of prediabetes [58], T1DM
and T2DM [67–69]. Although more investigation
is needed, there is increasing and powerful evi-
dence that ketogenic diets can be safely used in
patients with DPN.
5.2.2 Ketogenic Diet and Neural
Disease
Many of the benefits of using ketogenic diets to
treat neural disorders were grounded in their use
for the treatment of pharmacologically non-­
responsive epilepsy. Now, ketogenic diets have
been shown to be effective in treating a range of
neurodegenerative diseases and traumatic injury
[22, 35, 70–73]. Despite their use, however,
there remains a limited understanding of the
mechanisms of action provided by a ketogenic
diet on chronic painful conditions and
DPN.  Regarding various pain conditions we
reported that consumption of a ketogenic diet
abruptly improved abnormal mechanical sensa-
tion without substantial changes in weight using
a HFD mouse model [58–60, 67]. In addition,
we have recently demonstrated significant bene-
fits of using a ketogenic diet in reversing
mechanical allodynia and abnormal thermal sen-
sation in a mouse model of T1DM [67]. Both of
these studies also demonstrated, in  vitro, that
administration of β-OHD enhanced neurite out-
growth from cultured primary sensory neurons
[58]. Also, both studies also demonstrated,
in  vivo, that consumption of a ketogenic diet
could prevent reductions in IENFD and, impor-
tantly, stimulate epidermal nerve fiber regenera-
tion (Fig.  1) [67]. These studies hold promise
that using ketogenic diets to treat DPN can pro-
vide benefits, yet mechanisms by which the diet
or ketone bodies directly or indirectly reduce
nociception and normalize epidermal fiber inner-
vation are not clearly understood. However, data
from other studies described below point to plau-
sible mechanisms that should be explored and
will likely provide insight into potential mecha-
nisms and identify new drug targets.
402
5.2.3 Ketone Bodies as Signaling
Molecules
Ketone bodies serve as important bioenergetic
fuel sources; yet they are also very effective
extra-hepatic signaling molecules [74]. It is
becoming clear that these signaling functions are
involved modulating peripheral tissues These
include activation of G-protein coupled receptor
signaling cascades, modulation of cellular
inflammation, neurotrophin signaling, and epi-
genetic modifications. β-OHB serves an extracel-
lular ligand for hydroxycarboxylic acid receptor
2 (HCAR2) [75], a G-protein coupled receptor
widely expressed in a range of tissues, including
DPN-relevant nociceptors and circulating
immune cells (Fig.  2). Dimethyl fumarate is
another HCAR2 ligand and has shown promise in
preclinical pain models [75] and axonal degen-
eration following nerve injury [70]. Axonal
Inflammation
Nociceptive
Priming
HCAR2
αi β/y
IL-1b
NRF2
NLRP3
lnflammasome
Antioxidant
Genes
Fig. 2 Molecular functions of β-hydroxybutyrate
(β-OHB). β-OHB mediates its effects by extracellular sig-
naling through G-protein-coupled receptors like HCAR2
and through intracellular interactions after entering the
cell through the monocarboxylic acid transporter 1
(MCT1). These effects include reduced activation of the
J. Enders and D. E. Wright
regeneration induced by dimethyl fumarate is
linked to a shift in CD4+ T cell activation toward
a TH2 subtype [70]. This is interesting since
β-OHB is known to inhibit the NOD- LRR- and
Pyrin domain-containing protein 3 (NLRP3)
inflammasome [73, 76]. NLRP3 activation
cleaves pro-interleukin-1 to interleukin-1β (IL-­
1β), which promotes T cell polarization toward a
TH1 phenotype [77, 78] and pro-nociceptive
responses [79, 80]. Together, these studies pro-
vide support that ketone bodies may improve
sensory neuron function by modulating inflam-
mation, a key risk factor in DPN (Figs. 1 and 2).
An additional signaling function of β-OHB is
related to its activity as a histone deacetylase
(HDAC) inhibitor. Exercised-induced elevations
of β-OHB are associated with increased skeletal
muscle expression of brain-derived neurotrophic
factor (BDNF) [81]. β-OHB increased activity at
b-HB
K
K-ATP
MCT1 K
Channels K
ATP
b-HB
Glycolysis
Membrane
Associated
ATP
Cytosolic ATP
ROS
HDACs Mitochondrial
Biogenesis &
Mass
Bdnf
NLRP3 inflammasome, activation of NRF2-mediated
transcriptional profiles, inhibition of histone deacetylases,
improved mitochondrial biogenesis and function, and
reduced glycolysis, resulting in activation of KATP
channels
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
the BDNF promoter and decreased HDAC activ-
ity and occupancy of the BDNF promoter by
HDAC2 and HDAC3. β-OHB also rescued
HDAC hyperactivity in settings of neuropathic
pain, spinal cord injury [72, 82], and cisplatin-­
induced neuropathy [83]. Thus, β-OHB modula-
tion of the epigenetic landscape via HDAC
inhibition provides an additional mechanism that
merits exploration (Fig. 2).
In addition to β-OHB, acetoacetate may also
serve important functions associated with
DPN.  In a metabolism-independent fashion,
acetoacetate is known to stimulate extracellular-­
signal related kinase (ERK) activation [23].
Treatment with acetoacetate increased prolifer-
ation of skeletal muscle satellite cells [84] that
was associated with increased entry into S
phase and phosphorylation of the kinases ERK1
and ERK2. These effects occurred in the pres-
ence of knockdown of Oxct1, the enzyme
responsible for catalyzing the generation of
acetyl-CoA from acetoacetate, suggesting the
effects were independent of acetoacetate
metabolism [85].
5.2.4 Ketogenic Diets and TRP
Channels
The modulation of TRP ion channels is a poten-
tial mechanism to explain the actions of keto-
genic diets. TRP ankyrin 1 (TRPA1) and TRP
vanilloid 1 (TRPV1) are members of the transient
receptor potential family of ion channels and are
strongly implicated in painful conditions, includ-
ing painful DPN [6, 31, 86–89]. These channels
are responsive to toxic metabolites known to be
upregulated in diabetes, such as methylglyoxal
[90] and ROS [42–44]. Capsaicin sensitivity of
TRPV1 is heightened in DRG neurons in diabetic
rats [91]. The heightened activity can be rescued
by supplementation of the antioxidant melatonin,
implicating ROS and ROS-mediated TRPVI sen-
sitization in DPN.  Methylglyoxal can activate
TRPA1 to induce abnormal neuronal activity and
nociception [92]. Elevated ROS in DRG neurons
[43] and spinal dorsal horn [44] both increase
spontaneous neuronal activity that can be
reversed with TRPA1 and TRPV1 antagonists
[43, 44] (Fig. 1).
403
Heightened nociception caused by toxic gly-
colysis metabolites and ROS may be potentially
modulated by a ketogenic diet or ketone adminis-
tration. The antioxidant effects of a ketogenic diet
are well established [93]. As an example relevant
to DPN, we reported that a ketogenic diet reduces
H2O2 generation in the sciatic nerve [94] of dia-
betic mice, this effect was seen in the distal nerve
segment only, suggesting a length-­dependent pro-
cess that could be associated with impaired mito-
chondrial trafficking in DPN [94]. Finally,
methylglyoxal synthesis occurs as a normal con-
sequence of glycolysis; thus, methylglyoxal accu-
mulation should be reduced during ketosis.
Recent evidence now suggests that a ketogenic
diet may enhance axon degeneration and regen-
eration in DPN. DPN leads to reductions in epi-
dermal nerve fibers [95–98] and demyelination of
large-diameter axons [99, 100]. Hyperglycemia
contributes to this neurodegenerative change in
peripheral innervation, and Schwann cells cul-
tured in high glucose conditions undergo apopto-
sis [101] and de-­ differentiation [102].
Additionally, the Schwann cells secrete less NGF,
leading to reduced DRG neurite outgrowth [103].
We recently published that consumption of a
ketogenic diet stimulates the regeneration of epi-
dermal nerve fibers in a mouse model of DPN
[67]. The increased axon regeneration is likely
associated with both improved glycemic status
and direct actions of ketone bodies on DRG neu-
rons. This latter point is based on in vitro studies,
where β-OHB stimulated neurite outgrowth in
cultured DRG neurons in both the presence and
absence of glucose [58]. This in vitro work sug-
gests that ketone bodies have direct actions to
stimulate axon growth, perhaps through serving
as a potent fuel source (Fig. 1) [58].
5.2.5 Ketones and Modulation
of ATP-Gated Ion Channels
Ketogenic diets influence neuronal excitability
via regulation of ATP-gated potassium (KATP)
channels, and this action is believed to underlie
benefits of ketogenic diets in modulation of neu-
ronal firing rates in models of seizure and epi-
lepsy (Fig. 2). In the presence of ATP, these ion
channels remain closed. There are several pro-
404
posed mechanisms thought to be involved in
mechanisms through which a ketogenic diet
modulates KATP channel activity. Both β-OHB
and acetoacetate reduce the firing rates of
GABAergic neurons in the substantia nigra pars
reticula [88]. Importantly, this effect occurs in the
presence of glucose, and the inhibitory effect of
the ketones could be reversed following ketone
washout, administration of the KATP antagonist
tolbutamide, or genetic knockout of the Kir6.2
channel subunit. Additional support is provided
from studies of hippocampal slices, where reduc-
tions in available glucose lead to hyperpolariza-
tion of CA3 pyramidal neurons and decreased
frequencies of post-synaptic currents [104], The
KATP antagonist tolbutamide similarly prevented
and reversed outward potassium currents in these
neurons in reduced glucose.
Through the reduction of carbohydrate intake,
evidence suggests that a ketogenic diet reduces
glycolysis that then leads to reductions in plasma
membrane-proximal ATP. The result of this effect
would then increase the opening of KATP chan-
nels. Because glycolysis is partially localized to
the plasma membrane, it can provide a localized
source of ATP adjacent to cellular pumps that
includes the Na+/K+-ATPase pump [105, 106].
Ketone bodies are known to reduce cellular glu-
cose utilization and glycolytic flux [107, 108],
which reduces ATP gradients and increases KATP
channel activity. Similarly, reducing total glu-
cose, in essence, decreases substrate availability,
leading to decreased ATP and decreased opening
of KATP channels.
Importantly, KATP channels have been
reported to play a role in modulation of pain and
nociception. KATP channels are expressed by
sensory neurons [109], and opening of KATP
channels reduces neuronal current amplitude
following electrical stimulus. Inhibiting KATP
channels with the antagonist glibenclamide
exerts an opposite effect to increase current
amplitude. Ligation of the sciatic nerve closes
KATP channels in sensory neurons [110] and
leads to a decrease in KATP channel expression in
the spinal cord [109, 111]. Conversely, intrathe-
cal administration of a KATP channel activator
reverses thermal algesia and mechanical allo-
J. Enders and D. E. Wright
dynia following nerve injury [111].
Glibenclamide use also increases hindpaw
mechanical sensitivity in a model of bone
cancer-­associated pain in rats [112].
Interestingly, glibenclamide, also known as gly-
buride, is an available but infrequently used
medication used to treat T2DM [113].
Specifically related to DPN, a role for KATP chan-
nels has not yet been identified. However, the
expression of KATP channels in sensory neurons
and the consequences of hyperglycemia and dys-
lipidemia on channel activity suggest KATP chan-
nels may be an important cellular target. It is
known that KATP channel opening is affected by
glucose concentration [114, 115], and
hyperglycemia-­ evoked neuronal depolarization
of DRG neurons can be reversed by diazoxide, a
KATP channel activator [116]. As a surrogate for
dyslipidemia, palmitate supplementation lowers
the membrane-localization of the KATP channel
subunit SUR1, resulting in cellular lipotoxicity
[117]. Collectively, the effect of ketogenic diets
and ketone bodies to reduce glucose utilization,
glycolytic flux, and hyperglycemia [69, 107,
108] could increase opening of KATP channels
and ultimately reduce neuronal firing in neurons
[88, 104]. Also, ketogenic diets also reduce dys-
lipidemia [85, 118]. Based on this growing lit-
erature, it is plausible that a ketogenic diet could
function in an antinociceptive action in DPN and
KATP channels may be an important target to
explore.
5.2.6 Exogenous Ketone
Supplementation as a Tool
to Treat Diabetic Peripheral
Neuropathy
Evidence is growing that ketogenic diets may
serve as a promising intervention for treating
DPN and other chronic pain conditions. However,
long term adherence to a ketogenic diet remains
difficult and may ultimately be difficult for
patients to follow. This limitation may be over-
come by direct supplement with exogenous
ketone bodies. There is a growing market that
promotes commercially available ketone salts,
ketone salts with medium-chain triglyceride
(MCT) oil, or ketone esters [91, 99]. At this point,
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
there is no clinical evidence regarding ketone
supplementation in DPN.  However, evidence
suggests that exogenous ketone supplementation
mirrors some of the positive actions of a keto-
genic diet. Addition a β-OHB salt with MCT oil
or ketone ester can reduce blood glucose [119],
and supplementation with ketone salts reduces
oxidative stress in select organ systems [120]. It
is likely that certain actions (axon growth,
G-protein receptor activation, epigenetic regula-
tion, ion channel modulation) of a ketogenic diet
on DPN may be independent of hyperglycemia
and could be mimicked through ketone supple-
mentation. This will be an important area for
research as it may be an easier intervention for
patients with diabetes and can be easily coupled
with existing medications.
6 Dietary Influences
on the Microbiome Relevant
to Diabetic Peripheral
Neuropathy
Another mechanism by which lifestyle interven-
tion might affect the severity of DPN is through
regulation of gut microbiota. The microbiome is
emerging as a novel perspective in several pathol-
ogies, and its involvement in pain and neuropathy
is emerging. Mansuy-Aubert recently used a
moderately high-fat Western diet to induce neu-
ropathy in mice. They then assessed differences
in microbiota populations and performed fecal
transplantation [71]. Mice that were fed a Western
diet developed symptoms of neuropathy and,
importantly, displayed reduced diversity in gut
flora and abundance of butyrate-producing bacte-
rial species. Antibiotics and subsequent fecal
transfer rescued mouse thermal and mechanical
sensory symptoms, increased intraepidermal
innervation, and induced transcriptional changes
in the DRG and sciatic nerves. These effects are
consistent with the effects of antibiotic treatment
[121] and fecal transfer [122] in reversing
chemotherapy-­ induced neuropathy, suggesting
that alterations in gut microbiota may contribute
to painful neuropathies generally, not just in dia-
betes (Fig. 3).
405
One mechanism by which alterations in
microbiota may contribute to the onset of neuro-
pathic pain is through dysregulation of pathogen-
or damage-associated molecular pattern
receptors, such as the toll-like receptor (TLR)
family (Fig. 3). These receptors play an essential
role in innate immunity by detecting molecular
motifs normally associated with infection and
initiating an inflammatory response. TLR4, for
instance, is the canonical receptor for lipopoly-
saccharide (LPS) and, when activated, drives
symptoms of neuropathy in healthy mice [123]
and contributes to onset and progression of
obesity-­induced [124, 125] and chemotherapy-­
induced neuropathy [126].
Another potential mechanism includes signal-
ing by metabolites produced by microbiota.
Butyrate, for instance, is commonly produced as a
metabolic byproduct and signaling metabolite in
gut flora [127]. Butyrate is a potent inhibitor of his-
tone deacetylases (HDACs) [128, 129] and may
play a similar role to the closely related
β-hydroxybutyrate in regulating gene expression of
neurotrophins [81]. Butyrate is directly correlated
with improvements in allodynia and intraepidermal
nerve density in obese mice following fecal trans-
fer [71]. Interestingly, butyrate also demonstrates
activity as a partial agonist of the capsaicin recep-
tor, transient receptor channel vanilloid 1 (TRPV1),
and increases desensitization of TRPV1 [71].
TRPV1 contributes to DPN and other neuropathic
conditions [87, 91, 95, 97, 130]; thus, its modula-
tion by butyrate, and presumably butyrate-produc-
ing bacteria, provides an exciting avenue of
research moving forward (Fig. 3).
While sedentary lifestyles and Western diet
contribute to changes in microbiota that are pro-­
pathogenic, lifestyle interventions also alter the
gut microbiota. As previously discussed, keto-
genic diets are gaining popularity culturally and
as therapeutic interventions for a variety of pain
conditions. There is strong evidence that these
diets profoundly alter gut microbiota, though
these changes have yet to be documented in the
context of diabetes and DPN.  In children with
severe epilepsy, a ketogenic diet decreased
Actinobacteria and Eubacterium rectale, a major
butyrate-producing species [131]. There are con-
406
Butyrate
HDAC TRPV1
Inhibition Regulation
Altered Gene
Partial
Expression
Agonism
Increased Increased
Neurotrophins Desensitization
Fig. 3  The effect of the microbiome on DPN. The micro-
biota is known to impact neuropathy and play important
putative roles in DPN through varied mechanisms.
Increased systemic lipopolysaccharide (LPS) is pro-­
nociceptive, activating TLR4 signaling and driving
inflammation. Conversely, butyrate production by gut
flicting reports whether ketogenic diets decrease
[132] or increase [133] butyrate production in the
gut, though these differences may be due to adher-
ence to the ketogenic diet and the clinical popula-
tions being examined. If ketogenic diets indeed
reduce microbiota butyrate production, it may be
that the structurally similar β-hydroxybutyrate
may compensate for reduced butyrate, acting as
an HDAC inhibitor and possibly engaging in
TRPV1 modulation as well. These exciting lines
of research clearly merit more investigation.
7 Exercise and Physical
Activity to Improve Diabetic
Peripheral Neuropathy
Exercise and physical activity interventions are
important and increasingly studied lifestyle inter-
ventions for DPN. Exercise interventions can be
reasonably modeled preclinically and preclinical
studies in rodents have revealed beneficial effects
J. Enders and D. E. Wright
Lipopolysaccharide
(LPS)
TLR4 Signaling
Reduced
Inflammation
Reduced Nociceptive
Priming
flora appears to have antinociceptive actions. Butyrate
acts as a partial agonist for the capsaicin receptor TRPV1
and increases desensitization of this receptor. Additionally,
butyrate acts as a histone deacetylase (HDAC) inhibitor,
where it contributes to cell survival programs and tran-
scription of neurotrophins
related to the onset and progression of DPN. It is
known that voluntary exercise positively affects
hyperglycemia, insulin resistance [134, 135],
dyslipidemia [86], and advanced glycation
­endproducts (AGE) accumulation [135, 136]. In
addition IENFD [27, 136] and nerve conduction
velocity (NCV) [137] are altered by exercise
intervention. Here, we discuss the impact of exer-
cise and physical activity on non-neuronal and
neuronal systems as they relate to DPN.
7.1 Exercise Benefits on Non-­
neuronal Tissues
Exercise plays an important role in opposing obe-
sity and metabolic syndrome, major contributors
to prediabetes and T2DM.  Exercise improves
body composition, a risk factor relevant to DPN
in T2DM [138]. Exercise can improve bone min-
eral density in premenopausal women and in
males [89, 139, 140]. Additionally, the Health,
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
Aging, and Body Composition study identified a
relationship between bone mineral density and
sensory threshold, NCV, and nerve conduction
amplitude [141]. Although interesting, relation-
ship between reduced bone mineral density and
DPN is not clear and requires further study.
Dyslipidemia induces mitochondrial damage
[39, 40], increased ROS, and inflammation in
many tissues. In addition, dyslipidemia and insu-
lin resistance engage in a positive feedback loop
[142, 143]. Exercise reduces circulating triglyc-
eride levels in metabolic syndrome and diabetes
[86]. Finally, aerobic exercise consistently
improves insulin resistance in both preclinical
studies of diabetes [135] and in clinical settings
[134]. Importantly, exercise reduces insulin resis-
tance and hyperglycemia in DPN in T1DM and
T2DM [135]. Importantly, this leads to reduced
AGE accumulation in individuals that exercise
[136].
7.2 Exercise Benefits on Neuronal
Tissues
Exercise has direct and demonstrable actions on
sensory neurons relevant to DPN.  Exercise is
known to powerfully regulate neurotrophin
expression in tissues innervated by sensory
axons. For example, exercise increases the
expression and release of brain-derived neuro-
trophic factor (BDNF) [81] and neurotrophin-3
(NT-3) [144] in skeletal muscle. Both neuro-
trophins attract proprioceptive axons and are vital
to their growth and maintenance. These fibers are
commonly affected in large-fiber DPN [145–
147], resulting in poor proprioception, altered
gait, and balance control. Preclinical studies have
shown that exercise leads to improvements in
these sensory deficits [148] possibly through
improved muscle spindle innervation, likely
associated with skeletal muscle neurotrophin
availability. However, additional studies are nec-
essary to clearly elucidate the connection between
exercise and muscle spindle innervation in DPN.
Exercise also impacts IENFD, which is
reduced in DPN and often used as a measure of
disease progression. Following aerobic and resis-
tance training for 10  weeks, clinical studies in
407
patients with DPN have reported increased
IENFD and sprouting in proximal leg skin biop-
sies [136]. Consistent with these findings, a par-
allel study using 12 weeks of a similar exercise
program reported a similar, significant increase
in distal leg IENFD [149]. Here, clinical research
seems to indicate a positive effect of IENFD and
suggests additional studies are needed.
One important and clear effect of exercise is
improvements in mitochondrial health. The
extreme morphology and axon length of DRG
axons require a significant energetic demand. It
has been proposed that in DPN, mitochondrial
damage disrupts energy production in sensory
axons, leading to axon damage and degeneration
[39, 40, 150]. Exercise is known to increase the
expression of heat shock protein 70 (HSP70) in
the DRG of diabetic rats [151]. HSP70 plays an
essential role in chaperone-mediated autophagy
mitophagy [152]. Moreover, exercise has been
shown to upregulate peroxisome proliferator acti-
vated receptor gamma associated coactivator 1
alpha (PGC1α), a critical mediator of mitochon-
drial biogenesis [153, 154]. Overexpression of
downstream targets of PGC1α in sensory neurons
is sufficient to reverse the painful symptoms and
IENFD reductions in mouse models of DPN
[155, 156]. Overall, the evidence for using exer-
cise as an intervention to improve symptoms of
DPN is exciting. Yet exercise recommendations
for T1DM and T2DM patients are not universal
and improved behavioral interventions and edu-
cation for patients with DPN are needed to help
them successfully engage in safe, appropriate
exercise regimens.
8 Conclusion
With the worldwide epidemic increase in obesity
and diabetes, impaired quality of life for patients
that suffer from DPN demands new and better
non-pharmacological interventions to help slow,
reduce or possibly reverse symptoms of
DPN.  Interventions such as dietary changes or
exercise point to important lifestyle changes that
can reduce the burden of DPN, but also highlight
a rich new area to explore novel mechanisms that
could lead to the development of new drug
408
Obesity
Insulin Resistance
Hyperglycemia
AGE accumulation
Neurotrophins
ROS reduction
Dyslipidemia
Mitochondrial
function
Fig. 4  Points of convergence for dietary- and exercise-­
induced mechanisms. The benefits of dietary intervention
and exercise appear to share many common mechanisms
relevant to DPN, including beneficial effects on obesity,
reductions in insulin resistance, reductions in glycemic
levels, reductions in AGE accumulation, increased neuro-
targets. Exploration of these mechanisms merits
further research, particularly related to the over-
lapping mechanism between exercise and dietary
changes (Fig. 4).
Acknowledgments This work was supported by NIH
grants RO1 NS043314 (D.E.W.), T32 HD057850 (J.E.
and D.E.W.), the Kansas Institutional Development Award
(IDeA) P20 GM103418 (D.E.W.). All figures were cre-
ated using BioRender (biorender.com).
Conflict of Interest  The authors declare no
competing financial interests.
Author Contributions  J.E. and D.E.W. co-
wrote the chapter.
References
1. Coppini DV, Spruce MC, Thomas P, Masding
MG.  Established diabetic neuropathy seems irre-
versible despite improvements in metabolic and
vascular risk markers—a retrospective case-control
study in a hospital patient cohort. Diabet Med.
2006;23:1016–20.
2. Zochodne DW.  The challenges of diabetic poly-
neuropathy: a brief update. Curr Opin Neurol.
2019;32:666–75.
J. Enders and D. E. Wright
trophin production, reduced ROS production, and
improved mitochondrial function. These core mecha-
nisms appear to be central to improving DPN and contin-
ued study of these risk factors should help identify new
pharmacological targets that can be targeted and coupled
with existing medical treatments
3. Ang L, Jaiswal M, Martin C, Pop-Busui
R.  Glucose control and diabetic neuropathy: les-
sons from recent large clinical trials. Curr Diab
Rep. 2014;14:528.
4. Skyler JS.  Diabetic complications. The importance
of glucose control. Endocrinol Metab Clin N Am.
1996;25:243–54.
5. Feldman EL, Nave KA, Jensen TS, Bennett
DLH.  New horizons in diabetic neuropathy:
mechanisms, bioenergetics, and pain. Neuron.
2017;93:1296–313.
6. American Diabetes A. 9. Pharmacologic
approaches to glycemic treatment: standards of
medical care in diabetes-2020. Diabetes Care.
2020;43:S98–S110.
7. Mathieu C, Gillard P, Benhalima K.  Insulin ana-
logues in type 1 diabetes mellitus: getting better all
the time. Nat Rev Endocrinol. 2017;13:385–99.
8. Tesfaye S, Boulton AJ, Dickenson AH. Mechanisms
and management of diabetic painful distal
symmetrical polyneuropathy. Diabetes Care.
2013;36:2456–65.
9. Oyibo SO, Prasad YD, Jackson NJ, Jude EB, Boulton
AJ. The relationship between blood glucose excur-
sions and painful diabetic peripheral neuropathy: a
pilot study. Diabet Med. 2002;19:870–3.
10. Smith AG, Russell J, Feldman EL, Goldstein J, Peltier
A, Smith S, Hamwi J, Pollari D, Bixby B, Howard J,
Singleton JR.  Lifestyle intervention for pre-diabetic
neuropathy. Diabetes Care. 2006;29:1294–9.
11. Callaghan BC, Xia R, Reynolds E, Banerjee M,
Rothberg AE, Burant CF, Villegas-Umana E,
Pop-Busui R, Feldman EL.  Association between
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
metabolic syndrome components and polyneu-
ropathy in an obese population. JAMA Neurol.
2016;73:1468–76.
12. Feldman EL, Callaghan BC, Pop-Busui R, Zochodne
DW, Wright DE, Bennett DL, Bril V, Russell JW,
Viswanathan V.  Diabetic neuropathy. Nat Rev Dis
Primers. 2019;5:41.
13. Agathos E, Tentolouris A, Eleftheriadou I, Katsaouni
P, Nemtzas I, Petrou A, Papanikolaou C, Tentolouris
N. Effect of alpha-lipoic acid on symptoms and qual-
ity of life in patients with painful diabetic neuropa-
thy. J Int Med Res. 2018;46:1779–90.
14. Papanas N, Ziegler D. Efficacy of alpha-lipoic acid
in diabetic neuropathy. Expert Opin Pharmacother.
2014;15:2721–31.
15. Yang XD, Fang PF, Xiang DX, Yang YY.  Topical
treatments for diabetic neuropathic pain. Exp Ther
Med. 2019;17:1963–76.
16. Heydari M, Homayouni K, Hashempur MH, Shams
M.  Topical Citrullus colocynthis (bitter apple)
extract oil in painful diabetic neuropathy: a double-­
blind randomized placebo-controlled clinical trial. J
Diabetes. 2016;8:246–52.
17. Ostovar M, Akbari A, Anbardar MH, Iraji A,
Salmanpour M, Hafez Ghoran S, Heydari M,
Shams M.  Effects of Citrullus colocynthis L. in
a rat model of diabetic neuropathy. J Integr Med.
2020;18:59–67.
18. The Look AHEAD Research Group. Effects of a
long-term lifestyle modification programme on
peripheral neuropathy in overweight or obese adults
with type 2 diabetes: the Look AHEAD study.
Diabetologia. 2017;60:980–8.
19. Preguiça I, Alves A, Nunes S, Gomes P, Fernandes
R, Viana SD, Reis F.  Diet-induced rodent models
of diabetic peripheral neuropathy, retinopathy and
nephropathy. Nutrients. 2020;12:250.
20. Guilford BL, Ryals JM, Wright DE.  Phenotypic
changes in diabetic neuropathy induced by a high-­
fat diet in diabetic C57BL/6 mice. Exp Diabetes Res.
2011;2011:848307.
21. Obrosova IG, Ilnytska O, Lyzogubov VV, Pavlov
IA, Mashtalir N, Nadler JL, Drel VR. High-fat diet
induced neuropathy of pre-diabetes and obesity:
effects of "healthy" diet and aldose reductase inhibi-
tion. Diabetes. 2007;56:2598–608.
22. Stino AM, Smith AG. Peripheral neuropathy in pre-
diabetes and the metabolic syndrome. J Diabetes
Investig. 2017;8:646–55.
23. Vincent AM, Hayes JM, McLean LL, Vivekanandan-­
Giri A, Pennathur S, Feldman EL.  Dyslipidemia-­
induced neuropathy in mice: the role of oxLDL/
LOX-1. Diabetes. 2009;58:2376–85.
24. Grote CW, Groover AL, Ryals JM, Geiger PC,
Feldman EL, Wright DE. Peripheral nervous system
insulin resistance in ob/ob mice. Acta Neuropathol
Commun. 2013;1:15.
25. Grote CW, Morris JK, Ryals JM, Geiger PC,
Wright DE.  Insulin receptor substrate 2 expres-
sion and involvement in neuronal insulin resis-
409
tance in diabetic neuropathy. Exp Diabetes Res.
2011;2011:212571.
26. Guilford BL, Ryals JM, Lezi E, Swerdlow RH,
Wright DE. Dorsal root ganglia mitochondrial bio-
chemical changes in non-diabetic and streptozotocin-­
induced diabetic mice fed with a standard or high-fat
diet. J Neurol Neurosci. 2017;8:180.
27. Groover AL, Ryals JM, Guilford BL, Wilson NM,
Christianson JA, Wright DE.  Exercise-mediated
improvements in painful neuropathy associated with
prediabetes in mice. Pain. 2013;154:2658–67.
28. Coppey LJ, Shevalye H, Obrosov A, Davidson EP,
Yorek MA. Determination of peripheral neuropathy
in high-fat diet fed low-dose streptozotocin-treated
female C57Bl/6J mice and Sprague-Dawley rats. J
Diabetes Investig. 2018;9:1033–40.
29. Cooper MA, Jack MM, Ryals JM, Hayley P, Escher
T, Koch LG, Britton SL, Raupp SM, Winter MK,
McCarson KE, Geiger PC, Thyfault JP, Wright
DE.  Rats bred for low and high running capacity
display alterations in peripheral tissues and nerves
relevant to neuropathy and pain. Brain Behav.
2017;7:e00780.
30. Stavniichuk R, Drel VR, Shevalye H, Vareniuk
I, Stevens MJ, Nadler JL, Obrosova IG.  Role of
12/15-lipoxygenase in nitrosative stress and periph-
eral prediabetic and diabetic neuropathies. Free
Radic Biol Med. 2010;49:1036–45.
31. Hubinont C, Sener A, Malaisse WJ. Sorbitol content
of plasma and erythrocytes during induced short-­
term hyperglycemia. Clin Biochem. 1981;14:19–20.
32. Stavniichuk R, Shevalye H, Hirooka H, Nadler
JL, Obrosova IG.  Interplay of sorbitol pathway
of glucose metabolism, 12/15-lipoxygenase, and
­mitogen-­activated protein kinases in the pathogen-
esis of diabetic peripheral neuropathy. Biochem
Pharmacol. 2012;83:932–40.
33. Bierhaus A, Fleming T, Stoyanov S, Leffler A, Babes
A, Neacsu C, Sauer SK, Eberhardt M, Schnölzer M,
Lasitschka F.  Methylglyoxal modification of Na v
1.8 facilitates nociceptive neuron firing and causes
hyperalgesia in diabetic neuropathy. Nat Med.
2012;18:926–33.
34. Amano S, Yamagishi S-i, Kato N, Inagaki Y,
Okamoto T, Makino M, Taniko K, Hirooka H,
Jomori T, Takeuchi M. Sorbitol dehydrogenase over-
expression potentiates glucose toxicity to cultured
retinal pericytes. Biochem Biophys Res Commun.
2002;299:183–8.
35. Kikuchi S, Shinpo K, Moriwaka F, Makita Z, Miyata
T, Tashiro K.  Neurotoxicity of methylglyoxal and
3-deoxyglucosone on cultured cortical neurons:
synergism between glycation and oxidative stress,
possibly involved in neurodegenerative diseases. J
Neurosci Res. 1999;57:280–9.
36. Kuntz S, Kunz C, Rudloff S. Carbonyl compounds
methylglyoxal and glyoxal affect interleukin-8
secretion in intestinal cells by superoxide anion gen-
eration and activation of MAPK p38. Mol Nutr Food
Res. 2010;54:1458–67.
410
37. Polykretis P, Luchinat E, Boscaro F, Banci
L.  Methylglyoxal interaction with superoxide dis-
mutase 1. Redox Biol. 2020;30:101421.
38. Lupachyk S, Watcho P, Hasanova N, Julius U,
Obrosova IG.  Triglyceride, nonesterified fatty
acids, and prediabetic neuropathy: role for oxi-
dative–nitrosative stress. Free Radic Biol Med.
2012;52:1255–63.
39. Rumora AE, Lentz SI, Hinder LM, Jackson
SW, Valesano A, Levinson GE, Feldman
EL.  Dyslipidemia impairs mitochondrial traffick-
ing and function in sensory neurons. FASEB J.
2018;32:195–207.
40. Rumora AE, LoGrasso G, Haidar JA, Dolkowski
JJ, Lentz SI, Feldman EL.  Chain length of satu-
rated fatty acids regulates mitochondrial traffick-
ing and function in sensory neurons. J Lipid Res.
2019;60:58–70.
41. Viader A, Sasaki Y, Kim S, Strickland A, Workman
Cayce S, Yang K, Gross Richard W, Milbrandt
J.  Aberrant Schwann cell lipid metabolism linked
to mitochondrial deficits leads to axon degeneration
and neuropathy. Neuron. 2013;77:886–98.
42. Miyake T, Nakamura S, Zhao M, So K, Inoue K,
Numata T, Takahashi N, Shirakawa H, Mori Y,
Nakagawa T, Kaneko S. Cold sensitivity of TRPA1
is unveiled by the prolyl hydroxylation blockade-­
induced sensitization to ROS.  Nat Commun.
2016;7:12840.
43. Nazıroğlu M.  Activation of TRPM2 and TRPV1
channels in dorsal root ganglion by NADPH oxidase
and protein kinase C molecular pathways: a patch
clamp study. J Mol Neurosci. 2017;61:425–35.
44. Nishio N, Taniguchi W, Sugimura YK, Takiguchi
N, Yamanaka M, Kiyoyuki Y, Yamada H, Miyazaki
N, Yoshida M, Nakatsuka T.  Reactive oxygen spe-
cies enhance excitatory synaptic transmission in rat
spinal dorsal horn neurons by activating TRPA1 and
TRPV1 channels. Neuroscience. 2013;247:201–12.
45. Sözbir E, Nazıroğlu M. Diabetes enhances oxidative
stress-induced TRPM2 channel activity and its con-
trol by N-acetylcysteine in rat dorsal root ganglion
and brain. Metab Brain Dis. 2016;31:385–93.
46. Schink M, Leipold E, Schirmeyer J, Schönherr R,
Hoshi T, Heinemann SH.  Reactive species modify
Na V 1.8 channels and affect action potentials in
murine dorsal root ganglion neurons. Pflügers Arch.
2016;468:99–110.
47. Schlüter F, Leffler A.  Oxidation differentially
modulates the recombinant voltage-gated Na+
channel α-subunits Nav1.7 and Nav1.8. Brain Res.
2016;1648:127–35.
48. Esposito K, Maiorino M, Di Palo C, Giugliano D,
Group CPHS.  Adherence to a Mediterranean diet
and glycaemic control in type 2 diabetes mellitus.
Diabet Med. 2009;26:900–7.
49. Esposito K, Maiorino MI, Petrizzo M, Bellastella G,
Giugliano D. The effects of a Mediterranean diet on
the need for diabetes drugs and remission of newly
J. Enders and D. E. Wright
diagnosed type 2 diabetes: follow-up of a random-
ized trial. Diabetes Care. 2014;37:1824–30.
50. Barrière DA, Noll C, Roussy G, Lizotte F, Kessai
A, Kirby K, Belleville K, Beaudet N, Longpré J-M,
Carpentier AC, Geraldes P, Sarret P.  Combination
of high-fat/high-fructose diet and low-dose strepto-
zotocin to model long-term type-2 diabetes compli-
cations. Sci Rep. 2018;8:424.
51. García G, Gutiérrez-Lara EJ, Centurión D,
Granados-Soto V, Murbartián J.  Fructose-induced
insulin resistance as a model of neuropathic pain in
rats. Neuroscience. 2019;404:233–45.
52. Rumora AE, LoGrasso G, Hayes JM, Mendelson FE,
Tabbey MA, Haidar JA, Lentz SI, Feldman EL. The
divergent roles of dietary saturated and monounsatu-
rated fatty acids on nerve function in murine models
of obesity. J Neurosci. 2019;39:3770–81.
53. Tao M, McDowell MA, Saydah SH, Eberhardt
MS. Relationship of polyunsaturated fatty acid intake
to peripheral neuropathy among adults with diabetes
in the National Health and Nutrition Examination
Survey (NHANES) 1999–2004. Diabetes Care.
2008;31:93.
54. Brownlow ML, Benner L, D’Agostino D, Gordon
MN, Morgan D.  Ketogenic diet improves motor
performance but not cognition in two mouse
models of Alzheimer’s pathology. PLoS One.
2013;8:e75713.
55. Taylor MK, Sullivan DK, Mahnken JD, Burns JM,
Swerdlow RH. Feasibility and efficacy data from a
ketogenic diet intervention in Alzheimer’s disease.
Alzheimers Dement. 2018;4:28–36.
56. Bough KJ, Schwartzkroin PA, Rho JM.  Calorie
restriction and ketogenic diet diminish neuronal
excitability in rat dentate gyrus in  vivo. Epilepsia.
2003;44:752–60.
57. Dressler A, Trimmel-Schwahofer P, Klbermaxz K,
Prayer D, Kaspiran G, Rami B, Schober E, Feucht
M. Type 1 diabetes and epilepsy: efficacy and safety
of the ketogenic diet. Epilepsia. 2010;51:1806–9.
58. Cooper MA, Menta BW, Perez-Sanchez C, Jack
MM, Khan ZW, Ryals JM, Winter M, Wright DE. A
ketogenic diet reduces metabolic syndrome-induced
allodynia and promotes peripheral nerve growth in
mice. Exp Neurol. 2018;306:149–57.
59. Di Lorenzo C, Coppola G, Bracaglia M, Di Lenola
D, Sirianni G, Rossi P, Di Lorenzo G, Parisi V,
Serrao M, Cervenka MC. A ketogenic diet normal-
izes interictal cortical but not subcortical responsiv-
ity in migraineurs. BMC Neurol. 2019;19:1–9.
60. Ruskin DN, Sturdevant IC, Wyss LS, Masino
SA. Ketogenic diet effects on inflammatory allodynia
and ongoing pain in rodents. Sci Rep. 2021;11:1–8.
61. Holmden HF. Thesis on diabetes mellitus; its treat-
ment by abstinence from food. Annexe thesis digiti-
sation project 2016 Block 8. 1913.
62. Mosenthal HO, Clausen SW, Hiller A. The effect of
diet on blood sugar in diabetes mellitus. Arch Intern
Med. 1918;XXI:93–108.
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
63. Newburgh L, Marsh PL. The use of a high fat diet in
the treatment of diabetes mellitus: first paper. Arch
Intern Med. 1920;26:647–62.
64. Luukkonen PK, Dufour S, Lyu K, Zhang X-M,
Hakkarainen A, Lehtimäki TE, Cline GW, Petersen
KF, Shulman GI, Yki-Järvinen H. Effect of a keto-
genic diet on hepatic steatosis and hepatic mitochon-
drial metabolism in nonalcoholic fatty liver disease.
Proc Natl Acad Sci. 2020;117:7347–54.
65. Chandrasekaran P, Rani PK.  Reversal of diabetic
tractional retinal detachment attributed to keto diet.
BMJ Case Rep. 2020;13:e235873.
66. Tóth CC, Z.  Type 1 diabetes mellitus successfully
managed with the paleolithic ketogenic diet. Int J
Case Rep Images. 2014;5:699–703.
67. Enders J, Swanson MT, Ryals J, Wright D.  A
ketogenic diet reduces mechanical allodynia and
improves epidermal innervation in diabetic mice.
Pain. 2022;163(4):682–9.
68. Hussain TA, Mathew TC, Dashti AA, Asfar S,
Al-Zaid N, Dashti HM. Effect of low-calorie versus
low-carbohydrate ketogenic diet in type 2 diabetes.
Nutrition. 2012;28:1016–21.
69. Poplawski MM, Mastaitis JW, Isoda F, Grosjean
F, Zheng F, Mobbs CV.  Reversal of diabetic
nephropathy by a ketogenic diet. PLoS One.
2011;6:e18604.
70. Bombeiro AL, Pereira BTN, Bonfanti AP, de
Oliveira ALR.  Immunomodulation by dimethyl
fumarate treatment improves mouse sciatic nerve
regeneration. Brain Res Bull. 2020;160:24–32.
71. Bonomo RR, Cook TM, Gavini CK, White CR,
Jones JR, Bovo E, Zima AV, Brown IA, Dugas
LR, Zakharian E, Aubert G, Alonzo F, Calcutt NA,
Mansuy-Aubert V.  Fecal transplantation and butyr-
ate improve neuropathic pain, modify immune cell
profile, and gene expression in the PNS of obese
mice. Proc Natl Acad Sci. 2020;117:26482–93.
72. Kong G, Huang Z, Ji W, Wang X, Liu J, Wu X,
Huang Z, Li R, Zhu Q.  The ketone metabolite
β-hydroxybutyrate attenuates oxidative stress in
spinal cord injury by suppression of class I histone
deacetylases. J Neurotrauma. 2017;34:2645–55.
73. Shang S, Wang L, Zhang Y, Lu H, Lu X. The beta-­
hydroxybutyrate suppresses the migration of glioma
cells by inhibition of NLRP3 inflammasome. Cell
Mol Neurobiol. 2018;38:1479–89.
74. Puchalska P, Crawford PA. Multi-dimensional roles
of ketone bodies in fuel metabolism, signaling, and
therapeutics. Cell Metab. 2017;25:262–84.
75. Boccella S, Guida F, De Logu F, De Gregorio D,
Mazzitelli M, Belardo C, Iannotta M, Serra N,
Nassini R, de Novellis V.  Ketones and pain: unex-
plored role of hydroxyl carboxylic acid receptor
type 2  in the pathophysiology of neuropathic pain.
FASEB J. 2019;33:1062–73.
76. Youm Y-H, Nguyen KY, Grant RW, Goldberg EL,
Bodogai M, Kim D, D'agostino D, Planavsky N,
Lupfer C, Kanneganti TD.  The ketone metabo-
lite β-hydroxybutyrate blocks NLRP3 inflamma-
411
some–mediated inflammatory disease. Nat Med.
2015;21:263–9.
77. Gris D, Ye Z, Iocca HA, Wen H, Craven RR,
Gris P, Huang M, Schneider M, Miller SD, Ting
JP-Y. NLRP3 plays a critical role in the development
of experimental autoimmune encephalomyelitis by
mediating Th1 and Th17 responses. J Immunol.
2010;185:974–81.
78. Hatscher L, Lehmann CHK, Purbojo A, Onderka
C, Liang C, Hartmann A, Cesnjevar R, Bruns H,
Gross O, Nimmerjahn F, Ivanović-Burmazović I,
Kunz M, Heger L, Dudziak D. Select hyperactivat-
ing NLRP3 ligands enhance the T_H1-
and T_H17-inducing potential of human
type 2 conventional dendritic cells. Sci Signal.
2021;14:eabe1757.
79. Jia M, Wu C, Gao F, Xiang H, Sun N, Peng P, Li
J, Yuan X, Li H, Meng X.  Activation of NLRP3
inflammasome in peripheral nerve contributes to
paclitaxel-induced neuropathic pain. Mol Pain.
2017;13:1744806917719804.
80. Sweitzer S, Colburn R, Rutkowski M, DeLeo
J.  Acute peripheral inflammation induces moder-
ate glial activation and spinal IL-1β expression that
correlates with pain behavior in the rat. Brain Res.
1999;829:209–21.
81. Sleiman SF, Henry J, Al-Haddad R, El Hayek L,
Abou Haidar E, Stringer T, Ulja D, Karuppagounder
SS, Holson EB, Ratan RR, Ninan I, Chao
MV.  Exercise promotes the expression of brain
derived neurotrophic factor (BDNF) through the
action of the ketone body β-hydroxybutyrate. eLife.
2016;5:e15092.
82. Wang X, Wu X, Liu Q, Kong G, Zhou J, Jiang J, Wu
X, Huang Z, Su W, Zhu Q.  Ketogenic Metabolism
Inhibits Histone Deacetylase (HDAC) and reduces
oxidative stress after spinal cord injury in rats.
Neuroscience. 2017;366:36–43.
83. Mikami D, Kobayashi M, Uwada J, Yazawa
T, Kamiyama K, Nishimori K, Nishikawa Y,
Nishikawa S, Yokoi S, Taniguchi T, Iwano M.
β-Hydroxybutyrate enhances the cytotoxic effect of
cisplatin via the inhibition of HDAC/survivin axis in
human hepatocellular carcinoma cells. J Pharmacol
Sci. 2020;142:1–8.
84. Zou X, Meng J, Li L, Han W, Li C, Zhong R, Miao
X, Cai J, Zhang Y, Zhu D. Acetoacetate accelerates
muscle regeneration and ameliorates muscular dys-
trophy in mice*. J Biol Chem. 2016;291:2181–95.
85. Volek JS, Fernandez ML, Feinman RD, Phinney
SD.  Dietary carbohydrate restriction induces a
unique metabolic state positively affecting athero-
genic dyslipidemia, fatty acid partitioning, and met-
abolic syndrome. Prog Lipid Res. 2008;47:307–18.
86. Goldhaber-Fiebert JD, Goldhaber-Fiebert SN,
Tristán ML, Nathan DM.  Randomized controlled
community-based nutrition and exercise interven-
tion improves glycemia and cardiovascular risk fac-
tors in type 2 diabetic patients in rural Costa Rica.
Diabetes Care. 2003;26:24.
412
87. Luo J, Bavencoffe A, Yang P, Feng J, Yin S, Qian A,
Yu W, Liu S, Gong X, Cai T. Zinc inhibits TRPV1 to
alleviate chemotherapy-induced neuropathic pain. J
Neurosci. 2018;38:474–83.
88. Ma W, Berg J, Yellen G. Ketogenic diet metabolites
reduce firing in central neurons by opening KATP
channels. J Neurosci. 2007;27:3618–25.
89. Marques EA, Mota J, Carvalho J.  Exercise effects
on bone mineral density in older adults: a meta-­
analysis of randomized controlled trials. Age.
2012;34:1493–515.
90. Andersson DA, Gentry C, Light E, Vastani N,
Vallortigara J, Bierhaus A, Fleming T, Bevan
S.  Methylglyoxal evokes pain by stimulating
TRPA1. PLoS One. 2013;8:e77986.
91. Kahya MC, Nazıroğlu M, Övey İS.  Modulation of
diabetes-induced oxidative stress, apoptosis, and
Ca(2+) entry through TRPM2 and TRPV1 channels
in dorsal root ganglion and hippocampus of diabetic
rats by melatonin and selenium. Mol Neurobiol.
2017;54:2345–60.
92. Griggs RB, Laird DE, Donahue RR, Fu W, Taylor
BK.  Methylglyoxal requires AC1 and TRPA1 to
produce pain and spinal neuron activation. Front
Neurosci. 2017;11:679.
93. Greco T, Glenn TC, Hovda DA, Prins ML. Ketogenic
diet decreases oxidative stress and improves mito-
chondrial respiratory complex activity. J Cereb
Blood Flow Metab. 2016;36:1603–13.
94. Cooper MA, McCoin C, Pei D, Thyfault JP,
Koestler D, Wright DE.  Reduced mitochondrial
reactive oxygen species production in peripheral
nerves of mice fed a ketogenic diet. Exp Physiol.
2018;103:1206–12.
95. Cheng HT, Dauch JR, Porzio MT, Yanik BM, Hsieh
W, Smith AG, Singleton JR, Feldman EL. Increased
axonal regeneration and swellings in intraepidermal
nerve fibers characterize painful phenotypes of dia-
betic neuropathy. J Pain. 2013;14:941–7.
96. Christianson JA, Ryals JM, Johnson MS, Dobrowsky
RT, Wright DE.  Neurotrophic modulation of
myelinated cutaneous innervation and mechani-
cal sensory loss in diabetic mice. Neuroscience.
2007;145:303–13.
97. Hsieh Y-L, Lin C-L, Chiang H, Fu Y-S, Lue J-H,
Hsieh S-T.  Role of peptidergic nerve terminals in
the skin: reversal of thermal sensation by calcitonin
gene-related peptide in TRPV1-depleted neuropa-
thy. PLoS One. 2012;7:e50805.
98. Johnson MS, Ryals JM, Wright DE.  Early loss of
peptidergic intraepidermal nerve fibers in an STZ-­
induced mouse model of insensate diabetic neuropa-
thy. Pain. 2008;140:35–47.
99. Johnson PC, Doll SC, Cromey DW. Pathogenesis of
diabetic neuropathy. Ann Neurol. 1986;19:450–7.
100. Malik RA, Tesfaye S, Newrick PG, Walker D,
Rajbhandari SM, Siddique I, Sharma AK, Boulton
AJM, King RHM, Thomas PK, Ward JD. Sural nerve
pathology in diabetic patients with minimal but pro-
gressive neuropathy. Diabetologia. 2005;48:578–85.
J. Enders and D. E. Wright
101. Zhu L, Hao J, Cheng M, Zhang C, Huo C, Liu Y, Du
W, Zhang X.  Hyperglycemia-induced Bcl-2/Bax-­
mediated apoptosis of Schwann cells via mTORC1/
S6K1 inhibition in diabetic peripheral neuropathy.
Exp Cell Res. 2018;367:186–95.
102. Hao W, Tashiro S, Hasegawa T, Sato Y, Kobayashi
T, Tando T, Katsuyama E, Fujie A, Watanabe R,
Morita M, Miyamoto K, Morioka H, Nakamura M,
Matsumoto M, Amizuka N, Toyama Y, Miyamoto
T.  Hyperglycemia promotes schwann cell de-­
differentiation and de-myelination via sorbitol accu-
mulation and Igf1 protein down-regulation*. J Biol
Chem. 2015;290:17106–15.
103. Tosaki T, Kamiya H, Yasuda Y, Naruse K, Kato K,
Kozakae M, Nakamura N, Shibata T, Hamada Y,
Nakashima E, Oiso Y, Nakamura J.  Reduced NGF
secretion by Schwann cells under the high glucose
condition decreases neurite outgrowth of DRG neu-
rons. Exp Neurol. 2008;213:381–7.
104. Kawamura M, Ruskin DN, Masino SA.  Metabolic
autocrine regulation of neurons involves coop-
eration among pannexin hemichannels, adenos-
ine receptors, and KATP channels. J Neurosci.
2010;30:3886–95.
105. Campanella ME, Chu H, Low PS.  Assembly and
regulation of a glycolytic enzyme complex on the
human erythrocyte membrane. Proc Natl Acad Sci.
2005;102:2402–7.
106. Epstein T, Xu L, Gillies RJ, Gatenby RA. Separation
of metabolic supply and demand: aerobic glycoly-
sis as a normal physiological response to fluctuating
energetic demands in the membrane. Cancer Metab.
2014;2:7.
107. Puchalska P, Martin SE, Huang X, Lengfeld JE,
Daniel B, Graham MJ, Han X, Nagy L, Patti GJ,
Crawford PA. Hepatocyte-macrophage acetoacetate
shuttle protects against tissue fibrosis. Cell Metab.
2019;29:383–398.e7.
108. Valdebenito R, Ruminot I, Garrido-Gerter P,
Fernández-Moncada I, Forero-Quintero L,
Alegría K, Becker HM, Deitmer JW, Barros
LF.  Targeting of astrocytic glucose metabolism by
beta-­hydroxybutyrate. J Cereb Blood Flow Metab.
2015;36:1813–22.
109. Kawano T, Zoga V, McCallum JB, Wu HE, Gemes
G, Liang MY, Abram S, Kwok WM, Hogan QH,
Sarantopoulos CD.  ATP-sensitive potassium cur-
rents in rat primary afferent neurons: biophysical,
pharmacological properties, and alterations by pain-
ful nerve injury. Neuroscience. 2009;162:431–43.
110. Sarantopoulos C, McCallum B, Sapunar D, Kwok
W-M, Hogan Q.  ATP-sensitive potassium chan-
nels in rat primary afferent neurons: the effect of
neuropathic injury and gabapentin. Neurosci Lett.
2003;343:185–9.
111. Wu X-F, Liu W-T, Liu Y-P, Huang Z-J, Zhang Y-K,
Song X-J.  Reopening of ATP-sensitive potassium
channels reduces neuropathic pain and regulates
astroglial gap junctions in the rat spinal cord. Pain.
2011;152:2605–15.
Lifestyle and Dietary Modifications: Relevance in the Management of Diabetic Neuropathy
112. Xia H, Zhang D, Yang S, Wang Y, Xu L, Wu J, Ren
J, Yao W, Fan L, Zhang C, Tian Y, Pan H-L, Wang
X.  Role of ATP-sensitive potassium channels in
modulating nociception in rat model of bone cancer
pain. Brain Res. 2014;1554:29–35.
113. Fernandes VHR, Chaves FRP, Soares AAS, Breder
I, Kimura-Medorima ST, Munhoz DB, Cintra RMR,
Breder JC, Barreto J, Nadruz W, Carvalho LSF,
Quinaglia T, Arieta CEL, Sposito AC. Dapagliflozin
increases retinal thickness in type 2 diabetic patients
as compared with glibenclamide: a randomized con-
trolled trial. Diabetes Metab. 2021;47:101280.
114. Henquin JC.  ATP-sensitive K+ channels may con-
trol glucose-induced electrical activity in pan-
creatic B-cells. Biochem Biophys Res Commun.
1988;156:769–75.
115. Henquin JC. Role of voltage- and Ca2(+)-dependent
K+ channels in the control of glucose-induced elec-
trical activity in pancreatic B-cells. Pflugers Arch.
1990;416:568–72.
116. de Campos Lima T, DdO S, JBP L, Chiovato LM,
CMdC L.  Hyperglycemia induces mechanical
hyperalgesia and depolarization of the resting mem-
brane potential of primary nociceptive neurons: Role
of ATP-sensitive potassium channels. J Neurol Sci.
2019;401:55–61.
117. Ruan J-S, Lin J-K, Kuo Y-Y, Chen Y-W, Chen
P-C.  Chronic palmitic acid-induced lipotoxicity
correlates with defective trafficking of ATP sensi-
tive potassium channels in pancreatic β cells. J Nutr
Biochem. 2018;59:37–48.
118. Feinman RD, Volek JS.  Low carbohydrate diets
improve atherogenic dyslipidemia even in the
absence of weight loss. Nutr Metab. 2006;3:24.
119. Kesl SL, Poff AM, Ward NP, Fiorelli TN, Ari C,
Van Putten AJ, Sherwood JW, Arnold P, D’Agostino
DP.  Effects of exogenous ketone supplementation
on blood ketone, glucose, triglyceride, and lipo-
protein levels in Sprague–Dawley rats. Nutr Metab.
2016;13:9.
120. Kephart WC, Mumford PW, Mao X, Romero MA,
Hyatt HW, Zhang Y, Mobley CB, Quindry JC, Young
KC, Beck DT.  The 1-week and 8-month effects of
a ketogenic diet or ketone salt supplementation on
multi-organ markers of oxidative stress and mito-
chondrial function in rats. Nutrients. 2017;9:1019.
121. Shen S, Lim G, You Z, Ding W, Huang P, Ran
C, Doheny J, Caravan P, Tate S, Hu K, Kim H,
McCabe M, Huang B, Xie Z, Kwon D, Chen L,
Mao J.  Gut microbiota is critical for the induc-
tion of chemotherapy-­induced pain. Nat Neurosci.
2017;20:1213–6.
122. Ramakrishna C, Corleto J, Ruegger PM, Logan GD,
Peacock BB, Mendonca S, Yamaki S, Adamson
T, Ermel R, McKemy D, Borneman J, Cantin
EM.  Dominant role of the gut microbiota in che-
motherapy induced neuropathic pain. Sci Rep.
2019;9:20324.
123. Szabo-Pardi TA, Barron LR, Lenert ME, Burton
MD.  Sensory neuron TLR4 mediates the develop-
413
ment of nerve-injury induced mechanical hyper-
sensitivity in female mice. Brain Behav Immun.
2021;97:42–60.
124. Chen T, Li H, Yin Y, Zhang Y, Liu Z, Liu
H. Interactions of Notch1 and TLR4 signaling path-
ways in DRG neurons of in vivo and in vitro models
of diabetic neuropathy. Sci Rep. 2017;7:1–12.
125. Reichardt F, Chassaing B, Nezami BG, Li G,
Tabatabavakili S, Mwangi S, Uppal K, Liang B,
Vijay-Kumar M, Jones D.  Western diet induces
colonic nitrergic myenteric neuropathy and dys-
motility in mice via saturated fatty acid-and
lipopolysaccharide-­ induced TLR4 signalling. J
Physiol. 2017;595:1831–46.
126. Li Y, Zhang H, Kosturakis AK, Cassidy RM, Zhang
H, Kennamer-Chapman RM, Jawad AB, Colomand
CM, Harrison DS, Dougherty PM.  MAPK signal-
ing downstream to TLR4 contributes to paclitaxel-­
induced peripheral neuropathy. Brain Behav Immun.
2015;49:255–66.
127. Louis P, Flint HJ.  Formation of propionate and
butyrate by the human colonic microbiota. Environ
Microbiol. 2017;19:29–41.
128. Kim HJ, Leeds P, Chuang DM. The HDAC inhibi-
tor, sodium butyrate, stimulates neurogenesis in the
ischemic brain. J Neurochem. 2009;110:1226–40.
129. Tong X, Yin L, Giardina C.  Butyrate suppresses
Cox-2 activation in colon cancer cells through
HDAC inhibition. Biochem Biophys Res Commun.
2004;317:463–71.
130. Liu C, Li C, Deng Z, Du E, Xu C. Long non-coding
RNA BC168687 is involved in TRPV1-mediated
diabetic neuropathic pain in rats. Neuroscience.
2018;374:214–22.
131. Lindefeldt M, Eng A, Darban H, Bjerkner A,
Zetterström CK, Allander T, Andersson B,
Borenstein E, Dahlin M, Prast-Nielsen S. The keto-
genic diet influences taxonomic and functional com-
position of the gut microbiota in children with severe
epilepsy. NPJ Biofilms Microbiomes. 2019;5:5.
132. Ferraris C, Meroni E, Casiraghi MC, Tagliabue A,
De Giorgis V, Erba D. One month of classic thera-
peutic ketogenic diet decreases short chain fatty
acids production in epileptic patients. Front Nutr.
2021;8:613100.
133. Nagpal R, Neth BJ, Wang S, Craft S, Yadav
H.  Modified Mediterranean-ketogenic diet modu-
lates gut microbiome and short-chain fatty acids
in association with Alzheimer's disease mark-
ers in subjects with mild cognitive impairment.
EBioMedicine. 2019;47:529–42.
134. Cuff DJ, Meneilly GS, Martin A, Ignaszewski A,
Tildesley HD, Frohlich JJ. Effective exercise modal-
ity to reduce insulin resistance in women with type 2
diabetes. Diabetes Care. 2003;26:2977.
135. Dall'aglio E, Chang F, Chang H, Wright D, Reaven
GM. Effect of exercise training and sucrose feeding
on insulin-stimulated glucose uptake in rats with
streptozotocin-induced insulin-deficient diabetes.
Diabetes. 1983;32:165–8.
414
136. Kluding PM, Pasnoor M, Singh R, Jernigan S,
Farmer K, Rucker J, Sharma NK, Wright DE.  The
effect of exercise on neuropathic symptoms, nerve
function, and cutaneous innervation in people
with diabetic peripheral neuropathy. J Diabetes
Complicat. 2012;26:424–9.
137. Balducci S, Iacobellis G, Parisi L, Di Biase N,
Calandriello E, Leonetti F, Fallucca F.  Exercise
training can modify the natural history of dia-
betic peripheral neuropathy. J Diabetes Complicat.
2006;20:216–23.
138. Oh TJ, Lee J-E, Choi SH, Jang HC.  Association
between body fat and diabetic peripheral neuropa-
thy in middle-aged adults with type 2 diabetes mel-
litus: a preliminary report. J Obes Metab Syndr.
2019;28:112–7.
139. Bassey EJ, Rothwell MC, Littlewood JJ, Pye
DW.  Pre- and postmenopausal women have dif-
ferent bone mineral density responses to the
same high-impact exercise. J Bone Miner Res.
1998;13:1805–13.
140. Kelley GA, Kelley KS, Tran ZV. Exercise and bone
mineral density in men: a meta-analysis. J Appl
Physiol. 2000;88:1730–6.
141. Strotmeyer ES, Cauley JA, Schwartz AV, de
Rekeneire N, Resnick HE, Zmuda JM, Shorr RI,
Tylavsky FA, Vinik AI, Harris TB. Reduced periph-
eral nerve function is related to lower hip BMD and
calcaneal QUS in older white and black adults: the
Health, Aging, and Body Composition Study. J Bone
Miner Res. 2006;21:1803–10.
142. Lombardo YB, Hein G, Chicco A.  Metabolic syn-
drome: effects of n-3 PUFAs on a model of dys-
lipidemia, insulin resistance and adiposity. Lipids.
2007;42:427–37.
143. Semple RK, Sleigh A, Murgatroyd PR, Adams CA,
Bluck L, Jackson S, Vottero A, Kanabar D, Charlton-­
Menys V, Durrington P, Soos MA, Carpenter TA,
Lomas DJ, Cochran EK, Gorden P, O’Rahilly S,
Savage DB. Postreceptor insulin resistance contrib-
utes to human dyslipidemia and hepatic steatosis. J
Clin Invest. 2009;119:315–22.
144. Gómez-Pinilla F, Ying Z, Opazo P, Roy R, Edgerton
V. Differential regulation by exercise of BDNF and
NT-3  in rat spinal cord and skeletal muscle. Eur J
Neurosci. 2001;13:1078–84.
145. Feczko JD, Klueber KM. Cytoarchitecture of muscle
in a genetic model of murine diabetes. Am J Anat.
1988;182:224–40.
146. Muller KA, Ryals JM, Feldman EL, Wright
DE.  Abnormal muscle spindle innervation and
J. Enders and D. E. Wright
large-fiber neuropathy in diabetic mice. Diabetes.
2008;57:1693.
147. van Deursen RWM, Sanchez MM, Ulbrecht JS,
Cavanagh PR. The role of muscle spindles in ankle
movement perception in human subjects with dia-
betic neuropathy. Exp Brain Res. 1998;120:1–8.
148. Allet L, Armand S, de Bie RA, Golay A, Monnin
D, Aminian K, Staal JB, de Bruin ED. The gait and
balance of patients with diabetes can be improved:
a randomised controlled trial. Diabetologia.
2010;53:458–66.
149. Singleton JR, Marcus RL, Jackson JE, Lessard M,
Graham TE, Smith AG.  Exercise increases cutane-
ous nerve density in diabetic patients without neu-
ropathy. Ann Clin Transl Neurol. 2014;1:844–9.
150. Roy Chowdhury SK, Smith DR, Saleh A,
Schapansky J, Marquez A, Gomes S, Akude E,
Morrow D, Calcutt NA, Fernyhough P.  Impaired
adenosine monophosphate-activated protein kinase
signalling in dorsal root ganglia neurons is linked to
mitochondrial dysfunction and peripheral neuropa-
thy in diabetes. Brain. 2012;135:1751–66.
151. Yoon H, Thakur V, Isham D, Fayad M, Chattopadhyay
M. Moderate exercise training attenuates inflamma-
tory mediators in DRG of Type 1 diabetic rats. Exp
Neurol. 2015;267:107–14.
152. Zheng Q, Huang C, Guo J, Tan J, Wang C, Tang B,
Zhang H.  Hsp70 participates in PINK1-mediated
mitophagy by regulating the stability of PINK1.
Neurosci Lett. 2018;662:264–70.
153. Schnyder S, Svensson K, Cardel B, Handschin
C. Muscle PGC-1α is required for long-term systemic
and local adaptations to a ketogenic diet in mice. Am
J Physiol Endocrinol Metab. 2017;312:E437–46.
154. Wrann CD, White JP, Salogiannnis J, Laznik-­
Bogoslavski D, Wu J, Ma D, Lin JD, Greenberg
ME, Spiegelman BM.  Exercise induces hippocam-
pal BDNF through a PGC-1α/FNDC5 pathway. Cell
Metab. 2013;18:649–59.
155. Chandrasekaran K, Anjaneyulu M, Inoue T, Choi J,
Sagi AR, Chen C, Ide T, Russell JW. Mitochondrial
transcription factor A regulation of mitochondrial
degeneration in experimental diabetic neuropathy.
Am J Physiol Endocrinol Metab. 2015;309:E132–41.
156. Chandrasekaran K, Salimian M, Konduru SR, Choi
J, Kumar P, Long A, Klimova N, Ho C-Y, Kristian T,
Russell JW.  Overexpression of Sirtuin 1 protein in
neurons prevents and reverses experimental diabetic
neuropathy. Brain. 2019;142:3737–52.
 Pathophysiology of Neuropathic
Pain
Andreas C. Themistocleous
and Miroslav Misha Backonja
1 Introduction
Neuropathic pain is defined as “pain arising from
disease or lesion of somatosensory nervous sys-
tem” [1]. It is a heterogenous condition in which
the aetiology of the inciting nerve lesion interacts
with individual contributory factors such as gen-
otype and environmental factors [2]. The indi-
vidual neuropathic pain phenotype is the end
result of combinations of pathophysiological
mechanisms that include sensory nerve damage,
ectopic activity, sensitisation of the peripheral
and central nervous system, and altered brain
connectivity. A single aetiology, such as diabetes
mellitus, has a broad range of pathological pre-
sentations and does not cause neuropathic pain in
a uniform way.
Neuropathic pain is a common complication
caused by diabetic neuropathy. It affects between
25 and 50% of patients, depending on study
inclusion criteria [3]. Painful diabetic neuropathy
is often described as burning. Other descriptors
include stabbing, shooting, stinging, or aching
sensation and are often associated with tingling
A. C. Themistocleous
Nuffield Department of Clinical Neurosciences,
University of Oxford, Oxford, UK
e-mail: andreas.themistocleous@ndcn.ox.ac.uk
M. M. Backonja (*)
NCCIH, NIH, Bethesda, MD, USA
e-mail: misha.backonja@nih.gov
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_23
and pins-and-needles, i.e. paraesthesias or dys-
aesthesias. In diabetic neuropathy, the pattern of
neuropathic pain depends on the nerves affected.
In distal symmetric polyneuropathy, the most
common form of diabetic neuropathy, neuro-
pathic pain is found in the toes, feet, and legs that
can progress to involve the fingers and hands [4],
thus evolving into a “glove-and-stocking” distri-
bution synonymous with diabetic neuropathy.
A number of risk factors from large cross-­
sectional cohort studies are linked to painful dia-
betic neuropathy (Table  1). Poorer diabetic
control, obesity, polyneuropathy severity,
increasing age, female gender, and ethnicity are
possible risk factors for development and persis-
tence of painful diabetic neuropathy [5–9]. The
correlation of these risk factors to neuropathic
pain is not linear, as patients with severe neuropa-
Table 1  Risk factors for development of painful symp-
toms in diabetic polyneuropathy
Demographic factors
 • Advancing age
 • Female sex
 • Ethnicity
Clinical
 • Poorer diabetic control
 • Polyneuropathy severity
 • Obesity
Genetics
 • Heritability of pain traits
 • Voltage gated sodium channels
Psychological factors: Depression, anxiety
415
416 A. C. Themistocleous and M. M. Backonja

thy or very poor diabetic control can have pain- 2 Peripheral Mechanisms
less diabetic neuropathy. Several studies have of Painful Diabetic
shown associations between genotype and the Neuropathy
risk of developing neuropathic pain [10].
Although a number of findings still need to be In human studies, microneurography recordings
replicated, the heritability of pain traits shows in painful diabetic neuropathy have shown spon-
that genetics has an important role [11]. taneous activity in nociceptive fibres, mechano-­
Psychological factors are emerging as critical insensitive C fibre sensitisation, altered
factors in the expression of chronic painful dia- distribution of C afferent nerve fibres, and loss of
betic neuropathy. The processing of peripheral function within polymodal nociceptors (Table 2)
nociceptive signals by brain networks involved in [15–17]. However, in the only study comparing
processing pain related experiences will vary microneurography recordings between patients
between individuals and shape how individuals with painful and painless diabetic distal symmet-
experience, report, and manage chronic pain. rical polyneuropathy, some degree of spontane-
Longitudinal prospective natural history studies, ous activity and sensitisation was found in
combined with reliable measures of relevant patients with painless diabetic neuropathy [16].
pathophysiological mechanisms are required to This observation does not have clear interpreta-
explore and understand the relationship between tion, as it was a small study of only eight partici-
risk factors and pain in diabetic neuropathy. pants, of which five were diagnosed with
However, these studies are yet to be done. neuropathic pain. An important assumption based
A paradox of painful diabetic neuropathy is on other human neuropathic pain microneurogra-
the combination of ongoing, stimulus-­phy studies is that spontaneous activity and sensi-
independent neuropathic pain and sensory loss in tisation are important for the development and
a painful area. A typical clinical presentation is maintenance of neuropathic pain [18, 19]. C fibre
spontaneous ongoing pain and numbness with polymodal receptors with spontaneous activity
hyposensitivity to evoked pinprick, touch, and and altered stimulus-response function properties
temperature stimuli. Detailed studies that com- (lowered response threshold, increased response
bine interviews, clinical assessment, and quanti- magnitude, and a less adaptive response pattern
tative sensory testing show that sensory loss is to mechanical stimulation) are seen in rodent
the most common sensory phenotype in diabetic
polyneuropathy [6, 8, 9]. Mechanical hypersensi- Table 2 Neuronal and non-neuronal contributor to
tivity to evoked stimuli, i.e. mechanical hyperal- peripheral neuropathic pain mechanisms
gesia or allodynia, is an uncommon finding, Neuronal factors
present in only 15% of study participants [8].  • Spontaneous activity due to abnormal ectopic
However, dynamic mechanical allodynia is the activity
only discriminator between painless and painful  • Mechano-insensitive C fibres sensitisation
 • Altered distribution of C afferent nerve fibres
diabetic neuropathy, found only in those with  • Polymodal nociceptor loss function
neuropathic pain [12]. This contrasts to other  • Increased voltage gated sodium expression—
neuropathic pain conditions in which mechanical enhanced currents
hypersensitivity is a common finding and can  • Post-translational modification of ion channels, e.g.
methylglyoxal
guide treatment [13, 14].  • TRPA1 activation
In the following section we will discuss the pos-  • Peripheral neuronal sensitisation of TRPV1 in dorsal
sible pathophysiological mechanisms, within the root ganglion neurons
peripheral and central nervous systems, that cause  • Increased CaV3.2 T-type channel activity
Non-neuronal factors
painful diabetic neuropathy. The majority of mech-  • Higher epineural oxygen saturation
anisms discussed will apply to distal symmetrical  • Faster epineural blood flow—arterio-venous
polyneuropathy because it is the most common shunting
presentation of painful diabetic neuropathy.  • Pro-inflammatory cytokines
Pathophysiology of Neuropathic Pain
models of diabetic neuropathy [18, 20].
Therefore, abnormal ectopic activity and periph-
eral nociceptor sensitisation are pathophysiologi-
cal hallmarks of nerve injury and aberrant
signalling within nociceptive fibres.
Abnormal ectopic activity can occur at multi-
ple sites that include site of nerve injury, periph-
eral axon and dorsal root ganglia of nociceptive
neurones [21]. Sensory neuronal excitability is
controlled by the expression, trafficking, and
function of ligand and voltage gated ion chan-
nels. These channels are responsible for sensory
stimuli transduction, action potential generation
and propagation, and neurotransmitter release
with the dorsal horn. Voltage gated sodium (NaV)
channels are important drivers of excitability
within nociceptor neurones, as they are crucial
for action potential electrogenesis. Clustering of
voltage gated sodium channels at sites of injury
lowers the action potential thresholds and causes
an increase in excitability within primary
afferents.
Nine different NaV channels (NaV1.1–NaV1.9)
are expressed in humans and each has a unique
pattern of expression across different types of
excitable cells. The sodium channels NaV 1.7,
1.8, and 1.9 are the most highly expressed
α-subunits in nociceptors [22]. NaV 1.3 is
expressed in embryonic tissue only. However,
NaV 1.3 expression is significantly upregulated
after nerve injury [23]. Rare extreme inherited
channelopathies show the importance of voltage
gated sodium channel in nociceptive pathways
[24]. For example, loss of function mutations in
NaV 1.7 causes congenital insensitivity to pain,
while gain of function mutations causes erythro-
melalgia and paroxysmal extreme pain disorder.
Patch clamp recordings from small diameter dor-
sal root ganglion neurones show that Na  +  cur-
rent characteristics vary between neurones. These
experiments show that subunit expression and
post-translational modification are important in
regulating neuronal firing. NaV 1.7 is a threshold
channel for action potential generation, amplifies
subthreshold depolarisation, and contributes to
the rising phase of the action potential. NaV 1.8 is
the main contributor to the rising phase of action
potentials, and rapid repriming supports high-­
417
frequency firing of action potentials. NaV 1.9
amplifies subthreshold depolarisation, the large
window current may contribute to resting mem-
brane potential. NaV 1.3, when expressed after
peripheral nerve injury, amplifies subthreshold
depolarisation, and rapid repriming supports
high-frequency firing of action potentials. In
experimental rodent models of diabetic neuropa-
thy, NaV1.8 expression is increased and enhanced
currents mediated by the channel reduce C fibre
nociceptor conduction failure, which leads to
increased neuronal conduction to the central ner-
vous system [25].
Emerging evidence shows that inherited NaV
1.7 and NaV 1.8 mutations that cause dorsal root
ganglion hyperexcitability may increase the risk
for the development of painful diabetic neuropa-
thy [26–29]. Gain of function variants may also
cause axonal degeneration through increased
energetic demand and reactive oxygen species
production [30, 31]. The G856D mutation in
NaV1.7 produces significant changes in channel
biophysical properties, which contribute to the
hyperexcitability exhibited by neurones contain-
ing Nav1.8. The biophysical channel changes
lead to ATP depletion, increased reactive oxygen
species, elevated intracellular Na+ and Ca2+.
These changes cause increased axonal depolari-
sation with subsequent axonal degeneration.
Blockade of the reverse mode of the sodium/cal-
cium exchanger protects against neurite degen-
eration. This variant has not been studied in the
context of diabetes, but it is conceivable that the
energetic demands of diabetes may amplify the
changes caused by G856D mutation in NaV1.7.
The role of NaV 1.9 and NaV 1.3 has not been
studied in painful diabetic neuropathy and their
role is unclear.
Post-translational modification of ion chan-
nels, caused by metabolites that increase in dia-
betes, is an important potential mechanism in
neuropathic pain. Methylglyoxal is a highly reac-
tive metabolite that increases in diabetes and
causes non-enzymatic glycation of proteins [32].
Furthermore, the enzyme that metabolises meth-
ylglyoxal, gyloxylase-1, is lowly expressed in the
peripheral nervous system, and activity is further
reduced in diabetes. In human microneurography
418
studies methylglyoxal causes pain, axon-reflex
erythema, and long lasting hyperalgesia through
activation of mechano-insensitive C fibres [33].
In vitro data shows that methylglyoxal directly
activates cultured dorsal root ganglia nocicep-
tors, evokes the release of calcitonin-gene related
peptide, and causes pain hypersensitivity in
rodents by post-translational modification of
NaV1.8 that reduces inactivation of the channel
with subsequent nociceptor hyperexcitability
[34].
Other channels implicated in painful diabetic
neuropathy include: transient receptor potential
cation channel, subfamily A member 1 (TRPA1);
nonselective transient receptor potential cation
channel subfamily V member 1 (TRPV1); and
T-type Ca2+ channels.
TRPA1 is a nonselective ligand gated cation
channel expressed in a subset of nociceptors. It is
activated by exogenous compounds such as mus-
tard oil and cinnamaldehyde, noxious cold, and
endogenous ligands such as prostaglandins [35].
TRPA1 is implicated in painful diabetic neuropa-
thy. In rodent models of streptozotocin induced
diabetic neuropathy, pain related behaviour was
attenuated after administration of a TRPA1 chan-
nel antagonist [36]. However, a major caveat of
this model is that streptozotocin directly activates
TRPA1 channels [37]. In human studies, pre-­
injection with a TRPA1 antagonist blocks pain
sensation and hyperalgesia after methylglyoxal
administration [33]. Methylglyoxal binds and
modifies intracellular cysteine residues that acti-
vate TRPA1 with resultant hyperexcitability of
nociceptive neurones [38]. The reactive oxygen
species hydrogen peroxide and 4-­hydroxyhexenal,
which are increased in diabetes, also activate
TRPA1 [35]. Although further validation is
required, TRPA1 is a promising therapeutic
target.
TRPV1 is a nonselective ligand gate cation
channel that is activated by endogenous and
exogenous chemicals such as capsaicin and high
noxious temperatures. Its role in peripheral
inflammatory pain is well characterised [39].
After partial nerve injury, TRPV1 is upregulated
in uninjured sensory fibres [40, 41], and inhibi-
tion of TRPV1 activity or reducing levels of
A. C. Themistocleous and M. M. Backonja
TRPV1 attenuate nociceptive behaviours and
thermal hyperalgesia in animal models of neuro-
pathic pain [42–44]. Changes in the expression of
TRPV1, in rodent diabetic neuropathy models,
correlate with mechanical allodynia [45], devel-
opment of thermal hyperalgesia and TRPV1
expression on large myelinated A-fibres, an
injury-induced phenotypic shifts in low-­threshold
sensory neurones [46]. A possible mechanism of
TRPV1 mediated neuropathic pain is peripheral
neuronal sensitisation of TRPV1  in dorsal root
ganglion neurones [47]. High serum glucose con-
centration and the inflammatory cytokine high-­
mobility group box-protein 1, which is
upregulated in the skin of diabetic rats, enhance
TRPV1-evoked Ca2+ responses in dorsal gan-
glion neurones. This is achieved through the
receptor for advanced glycation end products in a
protein kinase C dependent manner and is
blocked by vascular endothelial growth factor
[47]. In human studies, pre-injection with a
TRPV1 antagonist does not block pain sensation
and hyperalgesia after methylglyoxal administra-
tion [33].
T-type Ca2+channels regulate the subthreshold
excitability of nociceptors [48].
In streptozotocin models of diabetic neuropa-
thy, CaV3.2 T-type channel activity is increased.
Glycosylation of extracellular arginine residues
of CaV3.2  T-type channel causes increased cur-
rent density, increased surface expression, and
altered kinetics [49–51]. All of which contribute
to dorsal root ganglion hyperexcitability.
Furthermore, downregulation of CaV3.2  T-type
channel improves pain related behaviours in
streptozotocin model of rodent diabetic neuropa-
thy [50, 51].
2.1 Endothelial Dysfunction
and Inflammation
Nerve blood flow impairment, hypoxia, and car-
diovascular risk factors are important in the
pathogenesis of diabetic neuropathy [52]. Nerve
haemodynamics differ between patients with
painless and painful diabetics neuropathy [53,
54]. Higher epineural oxygen saturation and
Pathophysiology of Neuropathic Pain 419

faster epineural blood flow are present in those Table 3  Neuronal and non-neuronal contributor to cen-
with painful diabetic neuropathy [54]. These tral neuropathic pain mechanisms
findings may be related to arterio-venous shunt- Neuronal factors
 • Exaggerated response to synaptic input
ing, but it is not clear how haemodynamic  • Lower threshold for neuronal activation
changes contribute to neuropathic pain. Skin  • Increased response to suprathreshold stimuli
microvascular vasodilator and vasoconstrictor  • Receptive field expansion
responses are altered in diabetic neuropathy;  • Morphological alterations in postsynaptic structures
within second-order sensory neurons
however, no consistent relationship to painful  • Dysregulation of potassium-chloride cotransporter 2
diabetic neuropathy has been found. Pro-­ activity
inflammatory cytokines predict the incidence and  • Decreased pain thresholds and wide dynamic range
progression of distal sensorimotor polyneuropa- neuron hyperexcitability
thy [55], and serum markers of endothelial dys- Non-neuronal contributors to central neuropathic pain
mechanisms
function and inflammation are elevated in patients  • Segmental microglial activation
with painful diabetic neuropathy [56, 57]. In  • Neuroinflammation
rodent models of painful diabetic neuropathy,
pro-inflammatory cytokines expression is ele-
vated in serum and dorsal root ganglia neurones microglial activation, and neuroinflammation in
[58, 59]. These data suggest a link between local the central nervous system [61–63]. In strepto-
blood flood dysregulation, inflammation, and zotocin induced diabetic neuropathy dendritic
painful diabetic neuropathy. Further studies are spine remodelling was related to decreased pain
required to better understand these relationships thresholds and wide dynamic range neuron
and delineate the underlying mechanisms. hyperexcitability [60]. A Rac1-specific inhibitor
In summary, there is a clear role for ion chan- known to interfere with spine plasticity decreased
nel dysfunction in the pathogenesis of painful the presence of malformed spines, attenuated
diabetic neuropathy. The increase in neuronal neuronal hyperresponsiveness to peripheral stim-
activity from sensitised nociceptors and ectopic uli, reduced spontaneous firing activity from
foci in degenerating nerve fibres will cause wide dynamic range neurones, and improved
increased synaptic transmission within the dorsal nociceptive mechanical pain thresholds [60]. The
horn of the spinal cord. The phenomenon of cen- role of dorsal horn microglia has been studied
tral sensitisation will be discussed in the next through the administration of a minocycline, a
section. selective inhibitor of microglial activation, in
streptozotocin induced neuropathy in rats.
Administration of minocycline attenuated the
3 Central Mechanisms development of diabetic neuropathy and pain
of Painful Diabetic behaviours and reduced levels of interleukin-1β
Neuropathy and tumour necrosis factor-α, lipid peroxidation,
and nitrite within the spinal cord [62]. In a similar
Central sensitisation of dorsal horn neurones is experiment minocycline reduced the expression
characterised by exaggerated response to synap- of the potassium chloride cotransporter 2 within
tic input, lower threshold for neuronal activation, the spinal cord [61]. Inhibition of microglial acti-
increased response to suprathreshold stimuli, and vation through inhibition of extracellular signal-­
receptive field expansion (Table 3) [14]. There is regulated protein kinase signalling ameliorated
emerging evidence that similar mechanisms play allodynia is a streptozotocin induced diabetic
a role in painful diabetic neuropathy. Possible neuropathy [63]. Microglia release several cyto-
mechanisms include morphological alterations in kines and chemokines that increase dorsal horn
postsynaptic structures within second-order sen- neurone excitability after nerve injury. For exam-
sory neurons [60], dysregulation of potassium-­ ple, the ionotropic ATP receptor, P2X4, in
chloride cotransporter 2 activity [61], segmental microglial cells upregulates after nerve injury.
420
After upregulation, ATP-induced activation of
P2X4 on microglia results in the release of brain
derived neurotrophic factor (BDNF) from these
cells. BDNF via its receptor, TrkB present on
dorsal horn neurones, decreases K+/Cl− cotrans-
porter KCC2 activity [64]. The net effect being
dorsal horn neurone excitability. The role of
NMDA receptors in painful diabetic neuropathy
is unclear. In animal models NMDA receptors
show increased expression and activation within
the spinal cord; however, NMDA receptor antag-
onist drug trials in patients have shown limited
efficacy [65, 66]. Human studies of the spinal
cord are limited. Structural neuroimaging of the
spine has shown early involvement of the spinal
cord in diabetic neuropathy, but no differences
were found between those with painful and pain-
less neuropathy [65, 67].
The dorsal horn of the spinal cord is under the
control of a descending pain modulatory system
that can either inhibit or facilitate transmission of
nociceptive information [68]. Once signals are
transmitted along primary afferent neurones to
the superficial dorsal horn, sensory information is
integrated by projection neurones in the superfi-
cial and deep lamina of the spinal cord. This
information is then transmitted via ascending
pathways to the periaqueductal grey, rostroven-
tral medulla, and thalamus of the brain. The peri-
aqueductal grey and rostroventral medulla in turn
send projections back down to the spinal cord
that can additionally facilitate or inhibit spinal
excitability. Neuropathic pain can result from
signal amplification or loss of inhibition within
these central pathways. Conditioned pain modu-
lation, a clinical measure of these descending
controls, is impaired in some patients with pain-
ful diabetic neuropathy [69]. Functional brain
imaging has shown that the ventrolateral periaq-
ueductal grey functional connectivity is altered in
those with painful diabetic polyneuropathy [70].
The magnitude of the changes correlates to spon-
taneous and evoked pain intensity. Therefore,
dysfunction of the ventrolateral periaqueductal
A. C. Themistocleous and M. M. Backonja
grey mediated descending pain modulatory sys-
tem may signal a brain-based pain facilitation
mechanism for painful diabetic polyneuropathy.
In streptozotocin rodent models’ electrophysio-
logical analysis of the rostroventromedial
medulla shows an increase in the spontaneous
activity of pronociceptive ON-like cells and
decreased spontaneous activity of OFF-like cells
[71]. These changes are similar to traumatic neu-
ropathic pain models and suggest that enhanced
descending pain facilitation may contribute to
rodent pain behaviours. Immunohistochemistry
analysis of the spinal cord and ventrolateral peri-
aqueductal grey shows an increase in c-Fos
expression in streptozotocin model of neuropa-
thy. In this model gabapentin reversed mechani-
cal hypersensitivity and c-Fos expression [72].
Thalamic changes are seen in human and ani-
mal studies. Magnetic resonance spectroscopy
shows a reduction of N-acetyl-aspartate, a marker
for neuronal and axonal activity, in the thalamus
of only those with painful diabetic neuropathy
[73]. Magnetic resonance perfusion shows an
increase in thalamic vascularity with associated
sluggish flow in painful diabetic neuropathy [74].
In streptozotocin models of diabetic neuropathy
electrophysiological recordings of the ventropos-
terolateral nucleus of thalamus show increased
spontaneous and evoked activity to mechanical
stimulation of the hind paw [75]. Neurones in the
thalamic ventral posteromedial nucleus from rats
with experimental diabetic neuropathy showed
increased firing to precisely graded, multidirec-
tional whisker deflection, in the absence of overt
primary afferent dysfunction of trigeminal gan-
glion neurones [76]. These findings suggest that
central thalamic neurones are hyperexcitable in
the absence of increased or aberrant primary
afferent input.
Functional neuroimaging shows altered func-
tional connectivity and somatosensory cortical
thickness in patients with painful diabetic neu-
ropathy [77–80]. For example, studies of the
anterior cingulate cortex show increased func-
Pathophysiology of Neuropathic Pain
tional activity [81] and cerebral blood flow [82]
in patients with painful diabetic neuropathy. In
earlier studies, sensorimotor areas showed
reduced connectivity bilaterally, while several
frontal areas, insulae, and thalami showed greater
connectivity [77]. In more recent studies func-
tional activity and somatosensory cortical thick-
ness are related to the sensory phenotype of the
patient [83]. Patients with diabetic distal sym-
metrical neuropathy can be divided into two phe-
notypes based on quantitative sensory assessment
called irritable and non-irritable nociceptor/deaf-
ferentation phenotype. Irritable nociceptor phe-
notype presents with relatively preserved sensory
function associated with thermal and/or mechani-
cal hyperalgesia, and the non-irritable nociceptor
or deafferentation phenotype presents with sen-
sory loss to thermal and mechanical stimuli.
Current cohort studies show that the irritable
nociceptor phenotype is far less common than the
non-irritable nociceptor phenotype [6, 8]. Patients
with irritable nociceptor phenotype show greater
thalamus–insular cortex functional connectivity
and decreased thalamus–somatosensory cortex
functional connectivity compared with those
with the non-irritable nociceptor phenotype.
Thalamus–insular cortex functional connectivity
was positively correlated to pain scores. Those
with a more severe polyneuropathy show a
greater reduction in thalamus–somatosensory
cortex functional connectivity and reduction in
somatosensory grey matter cortical volume [83].
This data suggests that distal axonopathy or
peripheral deafferentation leads to reduced
421
peripheral sensory input and subsequently
reduced cortical volumes and functional connec-
tivity. While in those with less severe neuropathy
this phenomenon does not occur. It is not known
whether the CNS changes observed in painful
diabetic neuropathy are a maladaptive response
of increased peripheral nociceptive afferent input
or an independent site for the initiation and main-
tenance of neuropathic pain in diabetic
neuropathy.
4 Conclusions
Over the last decade there have been significant
advances in our understanding of the pathophysi-
ological mechanisms of painful diabetic neuropa-
thy. The importance of voltage gated sodium
channels and advances in structural and functional
neuroimaging have provided novel insights into
peripheral and central mechanisms of painful dia-
betic neuropathy. It is clear from both clinical and
pre-clinical studies that an array of pathogenic
mechanism underlies neuropathic pain and mul-
tiple mechanisms take place and are likely to con-
tribute to some degree to the neuropathic pain
experience on the level of an individual patient
(Tables 2 and 3, Fig. 1). What is needed are large
longitudinal studies utilising deep phenotyping
that will prospectively investigate the develop-
ment of painful diabetic neuropathy. Although
these studies may be logistically challenging and
expensive, these difficulties pale in contrast to the
cost of painful diabetic neuropathy.
422 A. C. Themistocleous and M. M. Backonja

Cortex

PERCEPTION Thalamus

Brain
Stem
DECENDING
MODULARION

CENTRAL
TRANSMISSION Spinal Cord

K+/CL– cotransporter
Microglia

SYNAPTIC
Nav1.7, 1.8 TRANSMISSION
Cav3.2
TRANSDUCTION
PERIPHERAL
TRANSMISSION Dorsal Root
TRPA1 Ganglion (DRG)
TRPV1
Peripheral Nerves (PNS)

Pro-inflammatory cytokines

Fig. 1  Pain mechanisms represented along the neuroaxis

mation of severity, and treatments. Diabetes Care.

References
1. Finnerup NB, Haroutounian S, Kamerman P, Baron
R, Bennett DLH, Bouhassira D, et  al. Neuropathic
pain: an updated grading system for research and
clinical practice. Pain. 2016;157(8):1599–606.
2. von Hehn CA, Baron R, Woolf CJ. Deconstructing the
neuropathic pain phenotype to reveal neural mecha-
nisms. Neuron. 2012;73(4):638–52.
3. Abbott CA, Malik RA, van Ross ERE, Kulkarni
J, Boulton AJM.  Prevalence and characteristics of
painful diabetic neuropathy in a large community-­
based diabetic population in the UK. Diabetes Care.
2011;34(10):2220–4.
4. Tesfaye S, Boulton AJM, Dyck PJ, Freeman R,
Horowitz M, Kempler P, et  al. Diabetic neuropa-
thies: update on definitions, diagnostic criteria, esti-
2010;33(10):2285–93.
5. Hébert HL, Veluchamy A, Torrance N, Smith
BH. Risk factors for neuropathic pain in diabetes mel-
litus. Pain. 2017;158(4):560–8.
6. Raputova J, Srotova I, Vlckova E, Sommer C, Üçeyler
N, Birklein F, et al. Sensory phenotype and risk fac-
tors for painful diabetic neuropathy: a cross-sectional
observational study. Pain. 2017;158(12):2340–53.
7. Sieberg CB, Taras C, Gomaa A, Nickerson C, Wong
C, Ward C, et  al. Neuropathic pain drives anxiety
behavior in mice, results consistent with anxiety
levels in diabetic neuropathy patients. Pain Rep.
2018;3(3):e651-e.
8. Themistocleous AC, Ramirez JD, Shillo PR, Lees
JG, Selvarajah D, Orengo C, et al. The pain in neu-
ropathy study (PiNS): a cross-sectional observational
study determining the somatosensory phenotype
Pathophysiology of Neuropathic Pain
of painful and painless diabetic neuropathy. Pain.
2016;157(5):1132–45.
9. Truini A, Spallone V, Morganti R, Tamburin S,
Zanette G, Schenone A, et al. A cross-sectional study
investigating frequency and features of definitely
diagnosed diabetic painful polyneuropathy. Pain.
2018;159(12):2658–66.
10. Calvo M, Davies AJ, Hébert HL, Weir GA, Chesler
EJ, Finnerup NB, et  al. The genetics of neuropathic
pain from model organisms to clinical application.
Neuron. 2019;104(4):637–53.
11. Norbury TA, MacGregor AJ, Urwin J, Spector TD,
McMahon SB.  Heritability of responses to pain-
ful stimuli in women: a classical twin study. Brain.
2007;130(11):3041–9.
12. Scholz J, Rathmell JP, David WS, Chad DA, Broderick
AC, Perros SG, et al. A standardized clinical evalua-
tion of phenotypic diversity in diabetic polyneuropa-
thy. Pain. 2016;157(10):2297–308.
13. Demant DT, Lund K, Vollert J, Maier C, Segerdahl
M, Finnerup NB, et  al. The effect of oxcarbaze-
pine in peripheral neuropathic pain depends on
pain phenotype: a randomised, double-blind,
placebo-­ controlled phenotype-stratified study. Pain.
2014;155(11):2263–73.
14. Jensen TS, Finnerup NB.  Allodynia and hyperalge-
sia in neuropathic pain: clinical manifestations and
mechanisms. Lancet Neurol. 2014;13(9):924–35.
15. Orstavik K, Jørum E.  Microneurographic findings
of relevance to pain in patients with erythromelalgia
and patients with diabetic neuropathy. Neurosci Lett.
2010;470(3):180–4.
16. Ørstavik K, Namer B, Schmidt R, Schmelz M,
Hilliges M, Weidner C, et  al. Abnormal function
of C-Fibers in patients with diabetic neuropathy. J
Neurosci. 2006;26(44):11287–94.
17. Serra J, Duan WR, Locke C, Solà R, Liu W, Nothaft
W. Effects of a T-type calcium channel blocker, ABT-­
639, on spontaneous activity in C-nociceptors in
patients with painful diabetic neuropathy: a random-
ized controlled trial. Pain. 2015;156(11):2175–83.
18. Serra J, Bostock H, Solà R, Aleu J, García E, Cokic
B, et al. Microneurographic identification of sponta-
neous activity in C-nociceptors in neuropathic pain
states in humans and rats. Pain. 2012;153(1):42–55.
19. Serra J, Collado A, Solà R, Antonelli F, Torres X,
Salgueiro M, et  al. Hyperexcitable C nociceptors in
fibromyalgia. Ann Neurol. 2014;75(2):196–208.
20. Suzuki Y, Sato J, Kawanishi M, Mizumura K. Tissue
glucose level modulates the mechanical responses of
cutaneous nociceptors in streptozotocin-diabetic rats
but not normal rats in vitro. Pain. 2002;99(3):475–84.
21. Devor M.  Ectopic discharge in Abeta afferents
as a source of neuropathic pain. Exp Brain Res.
2009;196(1):115–28.
22. Bennett DL, Clark AJ, Huang J, Waxman SG, Dib-­
Hajj SD. The role of voltage-gated sodium channels
in pain Signaling. Physiol Rev. 2019;99(2):1079–151.
23. Waxman SG, Kocsis JD, Black JA. Type III sodium
channel mRNA is expressed in embryonic but not
423
adult spinal sensory neurons, and is reexpressed fol-
lowing axotomy. J Neurophysiol. 1994;72(1):466–70.
24. Bennett DL, Woods CG.  Painful and painless chan-
nelopathies. Lancet Neurol. 2014;13(6):587–99.
25. Craner MJ, Klein JP, Renganathan M, Black JA,
Waxman SG.  Changes of sodium channel expres-
sion in experimental painful diabetic neuropathy. Ann
Neurol. 2002;52(6):786–92.
26. Blesneac I, Themistocleous AC, Fratter C, Conrad LJ,
Ramirez JD, Cox JJ, et al. Rare NaV1.7 variants asso-
ciated with painful diabetic peripheral neuropathy.
Pain. 2018;159(3):469–80.
27. Han C, Themistocleous AC, Estacion M, Dib-
Hajj FB, Blesneac I, Macala L, et  al. The novel
activity of carbamazepine as an activation modu-
lator extends from Na(V)1.7 mutations to the
Na(V)1.8-S242T Mutant Channel from a patient
with painful diabetic neuropathy. Mol Pharmacol.
2018;94(5):1256–69.
28. Hoeijmakers JGJ, Faber CG, Merkies ISJ, Waxman
SG. Channelopathies, painful neuropathy, and diabe-
tes: which way does the causal arrow point? Trends
Mol Med. 2014;20(10):544–50.
29. Li QS, Cheng P, Favis R, Wickenden A, Romano
G, Wang H.  SCN9A variants may be implicated in
neuropathic pain associated with diabetic periph-
eral neuropathy and pain severity. Clin J Pain.
2015;31(11):976–82.
30. Estacion M, Vohra BPS, Liu S, Hoeijmakers J, Faber
CG, Merkies ISJ, et al. Ca2+ toxicity due to reverse
Na+/Ca2+ exchange contributes to degeneration of
neurites of DRG neurons induced by a neuropathy-­
associated Nav1.7 mutation. J Neurophysiol.
2015;114(3):1554–64.
31. Rolyan H, Liu S, Hoeijmakers JG, Faber CG, Merkies
IS, Lauria G, et  al. A painful neuropathy-associated
Nav1.7 mutant leads to time-dependent degeneration
of small-diameter axons associated with intracellular
Ca2+ dysregulation and decrease in ATP levels. Mol
Pain. 2016;12:1744806916674472.
32. Feldman EL, Nave K-A, Jensen TS, Bennett
DLH.  New horizons in diabetic neuropathy:
mechanisms, bioenergetics, and pain. Neuron.
2017;93(6):1296–313.
33. Düll MM, Riegel K, Tappenbeck J, Ries V, Strupf
M, Fleming T, et  al. Methylglyoxal causes pain and
hyperalgesia in human through C-fiber activation.
Pain. 2019;160(11):2497–507.
34. Bierhaus A, Fleming T, Stoyanov S, Leffler A, Babes
A, Neacsu C, et  al. Methylglyoxal modification of
Nav1.8 facilitates nociceptive neuron firing and
causes hyperalgesia in diabetic neuropathy. Nat Med.
2012;18(6):926–33.
35. Andersson DA, Gentry C, Moss S, Bevan S. Transient
receptor potential A1 is a sensory receptor for
multiple products of oxidative stress. J Neurosci.
2008;28(10):2485–94.
36. Wei H, Hämäläinen Minna M, Saarnilehto M,
Koivisto A, Pertovaara A.  Attenuation of mechani-
cal hypersensitivity by an antagonist of the TRPA1
424
Ion Channel in diabetic animals. Anesthesiology.
2009;111(1):147–54.
37. Andersson DA, Filipović MR, Gentry C, Eberhardt
M, Vastani N, Leffler A, et  al. Streptozotocin
stimulates the Ion Channel TRPA1 directly:
INVOLVEMENT OF PEROXYNITRITE*. J Biol
Chem. 2015;290(24):15185–96.
38. Andersson DA, Gentry C, Light E, Vastani N,
Vallortigara J, Bierhaus A, et  al. Methylglyoxal
evokes pain by stimulating TRPA1. PLoS One.
2013;8(10):e77986.
39. Huang J, Zhang X, McNaughton PA. Modulation of
temperature-sensitive TRP channels. Semin Cell Dev
Biol. 2006;17(6):638–45.
40. Hudson LJ, Bevan S, Wotherspoon G, Gentry C, Fox
A, Winter J.  VR1 protein expression increases in
undamaged DRG neurons after partial nerve injury.
Eur J Neurosci. 2001;13(11):2105–14.
41. Kim HY, Park C-K, Cho I-H, Jung SJ, Kim JS,
Oh SB.  Differential changes in TRPV1 expres-
sion after trigeminal sensory nerve injury. J Pain.
2008;9(3):280–8.
42. Christoph T, Grünweller A, Mika J, Schäfer MKH,
Wade EJ, Weihe E, et al. Silencing of vanilloid recep-
tor TRPV1 by RNAi reduces neuropathic and vis-
ceral pain in  vivo. Biochem Biophys Res Commun.
2006;350(1):238–43.
43. Ta LE, Bieber AJ, Carlton SM, Loprinzi CL, Low PA,
Windebank AJ. Transient receptor potential Vanilloid
1 is essential for cisplatin-induced heat hyperalgesia
in mice. Mol Pain. 2010;6:15.
44. Watabiki T, Kiso T, Kuramochi T, Yonezawa K, Tsuji
N, Kohara A, et al. Amelioration of neuropathic pain
by novel transient receptor potential Vanilloid 1
antagonist AS1928370 in rats without Hyperthermic
effect. J Pharmacol Exp Ther. 2011;336(3):743–50.
45. Cui Y-Y, Xu H, Wu H-H, Qi J, Shi J, Li Y-Q. Spatio-­
temporal expression and functional involvement
of transient receptor potential Vanilloid 1  in dia-
betic mechanical allodynia in rats. PLoS One.
2014;9(7):e102052.
46. Pabbidi RM, Yu S-Q, Peng S, Khardori R, Pauza ME,
Premkumar LS.  Influence of TRPV1 on diabetes-­
induced alterations in thermal pain sensitivity. Mol
Pain. 2008;4:1744-8069-4-9.
47. Bestall SM, Hulse RP, Blackley Z, Swift M, Ved N,
Paton K, et al. Sensory neuronal sensitisation occurs
through HMGB-1-RAGE and TRPV1 in high-glucose
conditions. J Cell Sci. 2018;131(14):jcs215939.
48. Bourinet E, Francois A, Laffray S.  T-type calcium
channels in neuropathic pain. Pain. 2016;157:S15–22.
49. Joksimovic SL, Evans JG, McIntire WE, Orestes P,
Barrett PQ, Jevtovic-Todorovic V, et al. Glycosylation
of ca(V)3.2 channels contributes to the hyperalgesia
in peripheral neuropathy of type 1 diabetes. Front Cell
Neurosci. 2020;14:605312.
50. Messinger RB, Naik AK, Jagodic MM, Nelson
MT, Lee WY, Choe WJ, et  al. In vivo silencing
of the CaV3.2 T-type calcium channels in sen-
sory neurons alleviates hyperalgesia in rats with
A. C. Themistocleous and M. M. Backonja
streptozocin-induced diabetic neuropathy. Pain.
2009;145(1):184–95.
51. Orestes P, Osuru HP, McIntire WE, Jacus MO,
Salajegheh R, Jagodic MM, et  al. Reversal
of Neuropathic Pain in Diabetes by Targeting
Glycosylation of Ca_v3.2 T-Type
Calcium Channels. Diabetes. 2013;62(11):3828–38.
52. Cameron NE, Eaton SE, Cotter MA, Tesfaye
S. Vascular factors and metabolic interactions in the
pathogenesis of diabetic neuropathy. Diabetologia.
2001;44(11):1973–88.
53. Archer AG, Roberts VC, Watkins PJ. Blood flow pat-
terns in painful diabetic neuropathy. Diabetologia.
1984;27(6):563–7.
54. Eaton SE, Harris ND, Ibrahim S, Patel KA, Selmi
F, Radatz M, et  al. Increased sural nerve epineurial
blood flow in human subjects with painful diabetic
neuropathy. Diabetologia. 2003;46(7):934–9.
55. Herder C, Kannenberg JM, Huth C, Carstensen-Kirberg
M, Rathmann W, Koenig W, et  al. Proinflammatory
cytokines predict the incidence and progression of dis-
tal sensorimotor polyneuropathy: KORA F4/FF4 study.
Diabetes Care. 2017;40(4):569–76.
56. Doupis J, Lyons TE, Wu S, Gnardellis C, Dinh T,
Veves A.  Microvascular reactivity and inflamma-
tory cytokines in painful and painless peripheral
diabetic neuropathy. J Clin Endocrinol Metabol.
2009;94(6):2157–63.
57. Purwata TE. High TNF-alpha plasma levels and mac-
rophages iNOS and TNF-alpha expression as risk
factors for painful diabetic neuropathy. J Pain Res.
2011;4:169–75.
58. Cheng HT, Dauch JR, Oh SS, Hayes JM, Hong Y,
Feldman EL. p38 mediates mechanical allodynia
in a mouse model of type 2 diabetes. Mol Pain.
2010;6:28.
59. Zhang C, Ward J, Dauch JR, Tanzi RE, Cheng
HT. Cytokine-mediated inflammation mediates pain-
ful neuropathy from metabolic syndrome. PLoS One.
2018;13(2):e0192333.
60. Tan AM, Samad OA, Fischer TZ, Zhao P, Persson
A-K, Waxman SG.  Maladaptive dendritic spine
Remodeling contributes to diabetic neuropathic pain.
J Neurosci. 2012;32(20):6795–807.
61. Morgado C, Pereira-Terra P, Cruz CD, Tavares
I. Minocycline completely reverses mechanical hyper-
algesia in diabetic rats through microglia-induced
changes in the expression of the potassium chloride
co-transporter 2 (KCC2) at the spinal cord. Diabetes
Obes Metab. 2011;13(2):150–9.
62. Pabreja K, Dua K, Sharma S, Padi SS, Kulkarni
SK.  Minocycline attenuates the development of dia-
betic neuropathic pain: possible anti-inflammatory
and anti-oxidant mechanisms. Eur J Pharmacol.
2011;661(1–3):15–21.
63. Tsuda M, Ueno H, Kataoka A, Tozaki-Saitoh H, Inoue
K.  Activation of dorsal horn microglia contributes
to diabetes-induced tactile allodynia via extracel-
lular signal-regulated protein kinase signaling. Glia.
2008;56(4):378–86.
Pathophysiology of Neuropathic Pain
64. Gold MS, Backonja MM.  Pain: from neurobiology
to disease. In: Zigmond MJ, Rowland LP, Coyle JT,
editors. Neurobiology of brain disorders. San Diego:
Academic Press; 2015. p. 674–92.
65. Calcutt NA.  Diabetic neuropathy and neuropathic
pain: a (con)fusion of pathogenic mechanisms? Pain.
2020;161(Suppl 1):S65–s86.
66. Zhou H-Y, Chen S-R, Pan H-L.  Targeting
N-methyl-­ D-aspartate receptors for treatment of
neuropathic pain. Expert Rev Clin Pharmacol.
2011;4(3):379–88.
67. Selvarajah D, Wilkinson ID, Emery CJ, Harris ND,
Shaw PJ, Witte DR, et  al. Early involvement of
the spinal cord in diabetic peripheral neuropathy.
Diabetes Care. 2006;29(12):2664–9.
68. Kuner R, Kuner T. Cellular circuits in the brain and
their modulation in acute and chronic pain. Physiol
Rev. 2021;101(1):213–58.
69. Yarnitsky D, Granot M, Nahman-Averbuch H,
Khamaisi M, Granovsky Y. Conditioned pain modu-
lation predicts duloxetine efficacy in painful diabetic
neuropathy. Pain. 2012;153(6):1193–8.
70. Segerdahl AR, Themistocleous AC, Fido D, Bennett
DL, Tracey I. A brain-based pain facilitation mecha-
nism contributes to painful diabetic polyneuropathy.
Brain. 2018;141(2):357–64.
71. Silva M, Amorim D, Almeida A, Tavares I, Pinto-­
Ribeiro F, Morgado C. Pronociceptive changes in the
activity of rostroventromedial medulla (RVM) pain
modulatory cells in the streptozotocin-diabetic rat.
Brain Res Bull. 2013;96:39–44.
72. Morgado C, Terra PP, Tavares I. Neuronal hyperactiv-
ity at the spinal cord and periaqueductal grey during
painful diabetic neuropathy: effects of gabapentin.
Eur J Pain. 2010;14(7):693–9.
73. Sorensen L, Siddall PJ, Trenell MI, Yue
DK.  Differences in metabolites in pain-processing
brain regions in patients with diabetes and painful
neuropathy. Diabetes Care. 2008;31(5):980–1.
74. Selvarajah D, Wilkinson ID, Gandhi R, Griffiths PD,
Tesfaye S.  Microvascular perfusion abnormalities
of the thalamus in painful but not painless diabetic
425
p­ olyneuropathy: a clue to the pathogenesis of pain in
type 1 diabetes. Diabetes Care. 2011;34(3):718–20.
75. Fischer TZ, Tan AM, Waxman SG.  Thalamic neu-
ron hyperexcitability and enlarged receptive fields
in the STZ model of diabetic pain. Brain Res.
2009;1268:154–61.
76. Freeman OJ, Evans MH, Cooper GJ, Petersen RS,
Gardiner NJ.  Thalamic amplification of sensory
input in experimental diabetes. Eur J Neurosci.
2016;44(1):1779–86.
77. Cauda F, Sacco K, Duca S, Cocito D, D'Agata F,
Geminiani GC, et al. Altered resting state in diabetic
neuropathic pain. PLoS One. 2009;4(2):e4542.
78. Selvarajah D, Wilkinson ID, Fang F, Sankar A, Davies
J, Boland E, et  al. Structural and functional abnor-
malities of the primary somatosensory cortex in dia-
betic peripheral neuropathy: a multimodal MRI study.
Diabetes. 2019;68(4):796–806.
79. Teh K, Wilkinson ID, Heiberg-Gibbons F, Awadh
M, Kelsall A, Pallai S, et al. Somatosensory network
functional connectivity differentiates clinical pain
phenotypes in diabetic neuropathy. Diabetologia.
2021;64(6):1412–21.
80. Wilkinson ID, Teh K, Heiberg-Gibbons F, Awadh M,
Kelsall A, Shillo P, et  al. Determinants of treatment
response in painful diabetic peripheral neuropathy: a
combined deep sensory phenotyping and multimodal
brain MRI study. Diabetes. 2020;69(8):1804–14.
81. Tseng MT, Chiang MC, Chao CC, Tseng WY, Hsieh
ST. fMRI evidence of degeneration-induced neuro-
pathic pain in diabetes: enhanced limbic and striatal
activations. Hum Brain Mapp. 2013;34(10):2733–46.
82. Watanabe K, Hirano S, Kojima K, Nagashima K,
Mukai H, Sato T, et  al. Altered cerebral blood flow
in the anterior cingulate cortex is associated with
neuropathic pain. J Neurol Neurosurg Psychiatry.
2018;89(10):1082–7.
83. Selvarajah D, Wilkinson ID, Maxwell M, Davies
J, Sankar A, Boland E, et  al. Magnetic resonance
neuroimaging study of brain structural differences
in diabetic peripheral neuropathy. Diabetes Care.
2014;37(6):1681–8.
 Central Nervous System
Involvement in Painful Diabetic
Neuropathy
Dinesh Selvarajah, Joyce Lim, Kevin Teh, Xin Chen,
Jing Wu, and Solomon Tesfaye
Abbreviations
1
H-MRS Proton magnetic resonance
spectroscopy
ACC Anterior cingulate cortex
BOLD Blood oxygen level dependent
CNS Central nervous system
DPN Diabetic peripheral neuropathy
DSP Diabetic sensorimotor polyneuropathy
fMRI Functional magnetic resonance
imaging
GM Frey matter
HSMN Hereditary sensorimotor neuropathy
IR Irritable
MR Magnetic resonance
NIR Non-irritable
PAG Periaqueductal grey
PET Positron emission tomography
QST Quantitative sensory testing
S1 Primary somatosensory cortex
D. Selvarajah · J. Lim · K. Teh
Department of Oncology and Metabolism, Medical
School, University of Sheffield, Sheffield, UK
X. Chen · J. Wu
Department of Endocrinology, Xiangya Hospital,
Central South University,
Changsha, People’s Republic of China
S. Tesfaye (*)
Royal Hallamshire Hospital, Sheffield Teaching
Hospitals and the University of Sheffield,
Sheffield, UK
e-mail: Solomon.Tesfaye@sth.nhs.uk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_24
S2 Secondary somatosensory cortex
vlPAG Ventrolateral periaqueductal grey
WM White matter
1 Painful Diabetic Neuropathy
Painful distal symmetrical polyneuropathy (pain-
ful DPN) affects about 20% of people with diabe-
tes [1] and is defined as ‘pain as a direct
consequence of peripheral nerve injury’ [2]. It
often results in a significant morbidity and finan-
cial burden on the individual and a wider socio-
economic impact on society [3]. In the UK, the
average annual healthcare cost per patient with
painful diabetic neuropathy is £2511 [3].
Approximately 60% of painful DPN patients
gainfully employed felt being less productive at
work [4]. Pain also interferes with their daily
activities, work, mood and sleep and is associated
with a reduced quality of life [4]. Not surpris-
ingly, there is a high prevalence of depression
and emotional distress [4]. Hence, pain is a com-
plex multidimensional construct which not only
affects sensory but also emotional/cognitive pro-
cessing. Although nociceptive signals may be ini-
tiated/generated in the periphery, the central
nervous system has an integral role in the patho-
genesis and maintenance of the chronic pain
state/experience. Not surprisingly, the central
nervous system is the main site of action of many
427
428
commonly used pharmacotherapy agents in neu-
ropathic pain. This further emphasises the impor-
tant role of the central nervous system in painful
DPN [5–7].
Despite recent advances and considerable
research, treatment for neuropathic pain is largely
ineffective. A meta-analysis which evaluated the
commonly used drugs in painful DPN found only
modest outcomes, with the number needed to
treat to achieve 50% pain relief ranging from 4 to
10 [8]. One reason for this poor treatment
response is that these agents do not target the
pathogenesis of neuropathic pain in diabetes. The
development of the current first-line agents fol-
lowed a therapeutic pathway, rather than a molec-
ular target pathway. Specifically, these agents
were developed for other indications before dem-
onstration of analgesic efficacy in neuropathic
pain (e.g. pregabalin [anti-epileptic], tricyclic
antidepressants). Although the CNS binding site
for these drugs has been identified (e.g. alpha-­
2delta-­1 calcium channel accessory protein for
gabapentinoids) the mechanism whereby this
interaction results in analgesia and the functional
significance of this binding are still not clear. At
present, there is no pathogenetic treatment found
to provide sufficient pain relief or prevent the
development of neuropathic pain in diabetes [9].
The main reason for this is because the pathogen-
esis of painful DPN is not fully understood.
Given the importance of the central nervous sys-
tem, a better understanding of the structural/func-
tional neuroplastic changes that occur will serve
as a catalyst for future drug discovery pro-
grammes in painful DPN. This would also serve
as robust biomarkers capable of assessing the
benefits of future novel therapies [10].
Another reason for the wide variability or
poor treatment response seen in clinical practice
and the failure of numerous Phase III ran-
domised controlled trials of novel compounds
for neuropathic pain is due to an underlying het-
erogeneity in clinical pain phenotypes [11].
Painful DPN is not a homogenous condition but
is comprised of many different clinical pain
phenotypes. This variation is thought to result
from the extent and distribution of nerve fibre
D. Selvarajah et al.
involvement (see Fig.  1 below). One approach
to assess this is by using quantitative sensory
testing (QST) which has shown that painful
DPN patients can be broadly subdivided into
two phenotypes: ‘irritable’ (IR, with relatively
preserved sensory function associated with ther-
mal and/or mechanical hyperalgesia) and ‘non-
irritable’ (NIR, dominated by thermal and
mechanical sensory loss) [12]. Subsequent stud-
ies appear to suggest that some treatments are
seemingly more effective in those with the irri-
table (IR) compared to the non-­irritable (NIR)
nociceptor phenotype (for a review see [13] and
for examples see [14–17]). Consequently, pain
phenotyping may, in the future, become impor-
tant in guiding individual patients’ treatment;
although the exact approach is heavily under
debate. Whilst mechanistic approaches (e.g.
QST) that require carefully evaluating specific
responses to guide therapy have significant
appeal (e.g. cold, heat, von Frey, etc.), in prac-
tice, these are time consuming and may be dif-
ficult to implement in busy clinical practices.
Furthermore, these are psychophysical measures
which rely on patient responses and may be sub-
jective and biased. Sensory profiling methods
also do not capture the complex and multidimen-
sional pain experience, which affects emotional
and cognitive processing in addition to sensory
processing. For example, chronic pain patients
often undergo neuropsychological changes,
which include changes in emotion and motiva-
tion or changes in cognition [18]. Chronic pain
may also arise after the onset of depression, even
in patients without a prior history of pain or
depression. Collectively, these clinical insights
suggest a better strategy is needed for assessing
and treating painful DPN, given it is a chronic
disease of dynamic process (e.g. evolution of co-­
morbid phenotypes such as anxiety or depres-
sion), which is not easily reversed in most
patients. This chapter will begin by examining
the alterations in the central nervous system
(CNS) in painful DPN and describe how neuro-
imaging could be used to phenotype patients with
painful DPN for personalised therapy before
highlighting knowledge gaps and future work.
Central Nervous System Involvement in Painful Diabetic Neuropathy
Fig. 1  Schematic representation of the generation of
pain: (a) Normal: Central terminals of c-afferents project
into the dorsal horn and make contact with secondary
pain-signalling neurons. Mechanoreceptive Aβ afferents
project without synaptic transmission into the dorsal col-
umns (not shown) and also contact secondary afferent
dorsal horn neurons. (b) C-fibre sensitisation: Spontaneous
activity in peripheral nociceptors (peripheral sensitisation,
black stars) induces changes in the central sensory pro-
cessing, leading to spinal cord hyperexcitability (central
2 CNS Changes in Response
to Peripheral Nerve Injury
and How These Changes
Result in Chronic Pain
in Diabetes
Neuropathic pain is the result of maladaptive
changes in the peripheral and central nervous
systems [2, 19, 20]. Alterations in ion channels
and ectopic activity at the periphery lead to
enhanced synaptic transmission within the dorsal
horn of the spinal cord and amplification of noci-
ceptive information, a process known as central
sensitisation [21]. Central sensitisation and the
altered balance between inhibitory interneurons
429
sensitisation, grey star) that causes input from mechanore-
ceptive Aβ (light touch) and Aδ fibres (punctuate stimuli)
to be perceived as pain (allodynia). (c) C-fibre loss:
C-nociceptor degeneration and novel synaptic contacts of
Aβ fibres with ‘free’ central nociceptive neurons, causing
dynamic mechanical allodynia. (d) Central disinhibition:
Selective damage of cold-sensitive Aδ fibres that leads to
central disinhibition, resulting in cold hyperalgesia.
Sympat, sympathetic nerve [23]
and the descending pain modulatory system can
either facilitate or inhibit the transmission of
nociceptive information at spinal levels [22]. The
altered balance within these systems contributes
to the development and persistence of neuro-
pathic pain. A more detailed schematic represen-
tation of the current theory for the generation of
different symptoms and signs of neuropathic pain
can be found in Fig. 1.
2.1 Changes in the Spinal Cord
Spinal cord involvement was first recognised in
the 1960s when post-mortem autopsy studies in
430
patients with advanced diabetes reported atrophy,
demyelination and axonal [24, 25] loss. These
findings were largely dismissed as being second-
ary to poor diabetes control and infection (e.g.
syphilis) rather than DPN.  Indeed, the
pathological abnormalities in the spinal cord
­ painful and painless DPN [30]. There were also
were reported in isolation and not examined in
the context of DPN related peripheral nerve
changes. Subsequent studies performed in the
late 1970s and 1980s utilised advances in somato-
sensory evoked potentials and demonstrated cen-
tral (brain and spinal cord) slowing in humans
with DPN [26] and rodent models [27]. With the
advent and accessibility of magnetic resonance
(MR) imaging in the 1990s and early 2000s,
investigators were able to demonstrate clear spi-
nal cord involvement in the form of cervical cord
atrophy not only in patients with established
DPN but also in those with early subclinical
DPN. Eaton et al. revealed significant reduction
in the spinal cord cross-sectional area in the cer-
vical and upper thoracic regions in patients with
advanced DPN compared to non-diabetic con-
trols [28]. A subsequent, larger study confirmed
these findings but also revealed spinal cord atro-
phy was present in subjects with early, subclini-
cal DPN [29]. Moreover, subjects with hereditary
sensorimotor neuropathy (HSMN), an inherited
neuropathy localised to the peripheral nerves,
had preserved cord cross-sectional area [29].
This argues against the ‘dying-back’ phenome-
non, i.e. peripheral nerve injury leading to subse-
quent cord atrophy in DPN rather it suggests that
diabetes causes a generalised, concomitant insult
to both the peripheral and central nervous sys-
tems [29].
2.2 Changes in the Brain
2.2.1 Structural/Metabolic
Structural and functional cortical plasticity is a
fundamental property of the vertebrate central
nervous system in response to nerve injury.
However, it can have maladaptive consequences,
possibly resulting in chronic pain. Studies using
structural MR neuroimaging have demonstrated
D. Selvarajah et al.
a clear reduction in peripheral (pre- and postcen-
tral gyrus) and deep grey matter (GM) nuclei
(caudate, putamen, medial pallidum, thalamus
and ventral nuclear) in regions involved with
somatosensory/nociception perception in both
alterations in the spinothalamic, corticospinal
and thalamocortical white matter projections to
the cerebral cortex indicative of the involvement
of somatosensory, motor and pain-related path-
ways [31]. Specific alterations in patients with
painful DPN have been reported in the cingulate
cortex, insular cortex, prefrontal lobe, thalamus,
periaqueductal WM and external capsule [31].
The involvement of the anterior cingulate cortex
(ACC) and prefrontal cortex has been implicated
in the unpleasant perception associated with allo-
dynia (pain due to a stimulus that normally does
not provoke pain) [32]. Patients with painful- and
painless DPN were also found to have reduced
GM volume localised to regions involved in
somatosensory perception compared to control
patients with diabetes and no DPN and healthy
volunteers [30]. Another modality of MR imag-
ing is proton (H+) MR spectroscopy. This method
can be used to examine neurochemical altera-
tions in the brain. Using this method, patients
with painless DPN had increased thalamic neuro-
nal dysfunction compared to those with no DPN
and painful DPN [33]. This finding suggests that
the preservation of thalamic neuronal function
may be a pre-requisite for the perception of pain
in diabetic neuropathy. This finding was also sup-
ported by structural MR imaging which demon-
strated reduction in thalamic grey matter volume
in subjects with painless diabetic neuropathy but
not painful diabetic neuropathy [30]. The thala-
mus plays an important role in somatosensory
perception. It not only acts as a relay station but
also modulates/processes nociceptive informa-
tion from the periphery to other areas of the brain
[34]. Animal models of painful DPN have dem-
onstrated increased spontaneous activity from the
thalamic nuclei [35]. Magnetic resonance perfu-
sion imaging of the brain has shown increased
thalamic vascularity in patients with painful DPN
[36]. These findings imply the possible involve-
Central Nervous System Involvement in Painful Diabetic Neuropathy
ment of the thalamus in central amplification of
somatosensory signals. Similar changes have
also been observed in the ACC suggesting a role
for ACC activation in the development of central
sensitisation [37]. In addition, patients with pain-
ful DPN have altered functional connectivity in
the ventrolateral PAG (vlPAG), which correlates
with their pain intensity and cerebral blood flow
in response to tonic heat stimulation [38]. This
supports the idea that abnormalities in the PAG
may result not only in reduced inhibition, but also
facilitation of pain [38]. Collectively, these find-
ings, supported by numerous studies in other
chronic pain conditions, demonstrate dynamic
neuronal changes that have profound effects on
the brain in patients with DPN.
2.2.2 Functional
Functional magnetic resonance imaging (fMRI) is
a modality of MR imaging which examines brain
activity by detecting changes in blood flow [blood
oxygen level dependent (BOLD) signal] that is
coupled with neuronal activation, i.e. when an area
of the brain is in use, blood flow to that region also
increases and this can be examined using
fMRI. Using this method several studies have iden-
tified, the neurological signature for both physical
and neuropathic pain (capsaicin model). The brain
regions activated in response to a nociceptive stim-
ulus can be divided into somatic region responsible
for the localisation of pain and coding intensity
such as the ventrolateral thalamus, somatosensory
cortex, dorsal posterior insula and the emotional/
affective regions responsible for emotional pain
processing such as the anterior insula, dorsal lateral
prefrontal cortex and ACC [39].
A case-controlled, fMRI study of carefully
phenotyped patients with DPN demonstrated a
pathophysiological relationship between anatom-
ical and functional changes in the brain and clini-
cal pain phenotypes. Subjects with painful DPN
were divided into sensate (preserved nocicep-
tion) and insensate (abnormal nociception or
painful ‘painless DPN’) cohorts. Subjects with
insensate painful DPN had the lowest S1 cortical
thickness and greatest S1 cortical functional reor-
ganisation with expansion of the area represent-
431
ing pain in the lower limb region to include face
and lips regions [40] compared to painful DPN
subjects with relatively preserved sensation.
Furthermore, the extent to which S1 cortical
structure and function are altered was related to
the severity of neuropathy and the magnitude of
self-reported pain. These data suggest a dynamic
plasticity of the brain in DPN, driven by the neu-
ropathic process and may ultimately determine
an individual’s clinical pain phenotype [40].
Over the last decade, resting-state fMRI
(RS-fMRI), a relatively quick (6  min) and sim-
ple, non-invasive technique, has become an
increasingly appealing method to examine spon-
taneous brain activity in individuals without rely-
ing on experimental external stimulation tasks.
During a typical RS-fMRI examination, the hae-
modynamic response to spontaneous neuronal
activity (bold oxygen level dependent, BOLD
signal) is acquired, whilst subjects are instructed
to simply rest in the MRI scanner [41]. Data
acquired is used to evaluate the connectivity
between different brain regions, i.e. functional
connectivity when a subject is at rest. RS-fMRI
experiments in painful DPN have reported greater
thalamic-insula functional connectivity and
decreased thalamic-somatosensory cortical func-
tional connectivity in patients with the irritable
nociceptor phenotype compared to the non-­
irritable nociceptor phenotype (Fig.  2) [42].
There was a significant positive correlation
between thalamic-insula functional connectivity
with self-reported pain scores. Conversely, there
was a greater reduction in thalamic-­somatosensory
cortical functional connectivity in those with
more severe neuropathy (Fig.  2). This demon-
strates how RS-fMRI measures of functional
connectivity relate to both the somatic and non-­
somatic assessments of painful DPN.
Multimodal MR imaging combining both
structural and RS-fMRI has also been used to
predict treatment response in painful DPN
(Fig.  2). Responders to neuropathic pain treat-
ment have significantly greater S1 cortical vol-
ume and greater functional connectivity between
the insular cortex and corticolimbic system com-
pared to non-responders [43]. The insula cortex
432 D. Selvarajah et al.

Non Irritable Nociceptor Irritable Nociceptor

R Thal-Post CG FC
R PostCG R PostCG R Thal-lC FC
0.4
0.1

0.3

Effect size

Effect size
0.0
IC IC
0.2
Thalamus Thalamus
–0.1
0.1

–0.2 0
IR NIR IR NIR

NTSS-6 TCNS
F 1 2 3 4
0.6
Partial correlation (r)

0.4 * R Thal – IC FC

0.2 R Thal – PostCG FC

0
–0.2

*
–0.4

Responders Non-Responders

R IC-Forb FC R IC-Amy FC
0.5 0.4
Effect size

AC 0.3 AC

Effect size
0.2

IC IC
0.1
0.0
FOrb FOrb
Amy Amy
–0.1
Res N-Res Res N-Res

High IC to cortical limbic connectivity Low IC to cortical limbic connectivity

Fig. 2 (a) Right view of resting-state functional connec-
tivity in non-irritable nociceptor and irritable nociceptor
painful diabetic peripheral neuropathy phenotype. R right,
IC insula cortex (MNI coordinates: 44, 4, 0), PostCG post-
central gyrus (2–36 62), thal thalamus (10–19 6) [1]. (b,
c) Bar charts demonstrating effect size of differences in
thalamic-insula and thalamic-somatosensory cortical
functional connectivity. IR irritable nociceptor, NIR non-­
irritable nociceptor. (b) The bar graph plots 4 FC and
behaviour partial correlation coefficients (r Pearson)
derived from the functional connectivity of the right
thalamus-­insula cortex (R thal-IC FC, white) and FC of
the right thalamus-somatosensory cortex (R thal-PostCG
FC, black). Bars 1–2 and 3–4 indicate correlations involv-
ing the neuropathic total symptom score 6 (NTSS-6) and
Toronto Clinical Neuropathy Score (TCNS), respectively.
Each partial correlation coefficient, e.g. bar 1, is obtained
by correlating a given behaviour, e.g. self-reported pain
scores (NTSS-6) with the functional connectivity derived
from a given network, e.g. right thal-IC FC. *p < 0.05. FC
functional connectivity [1]. (c) Right view of bilateral
insular cortices resting-state FC in responders and non-­
responders to intravenous lidocaine treatment [2]. AC
anterior cingulate gyrus, Acu nucleus accumbens, Amy
amygdala, FOrb orbital frontal cortex, IC insula cortex, L
left, R right [42, 43]

plays a pivotal role in processing the emotion and 3 Clinical Implications

cognitive dimensions of the chronic pain experi-
ence [44]. The corticolimbic circuits have also There is now increasing evidence that brain
long been implicated in reward, decision making imaging could serve as a reliable biomarker in
and fear learning. Hence, these findings suggest preclinical and clinical trials in pain therapeutics
that this network may have a role in determining [13]. It could be used to increase the probability
treatment response in painful DPN.  Examining of novel compounds advancing to Phase II trials,
neurotransmitter activity and receptor ­distribution reduce the variability of the therapeutic response
within these networks may provide clues to why and reduce the overall expense and time of drug
some patients respond, whilst others are non- development. As highlighted in this chapter, MR
responders to neuropathic pain treatments. brain imaging has demonstrated alterations in
Central Nervous System Involvement in Painful Diabetic Neuropathy 433

brain structure, biochemistry and perfusion
between painful and painless DPN. More recent
studies have demonstrated significant structural
and functional cortical changes linked not only to
different clinical phenotypes but also treatment
response in painful DPN. This has the potential
to fast-track the development of new therapeutic
interventions and alter the management approach
for painful DPN.  For example, MR imaging
could provide specific targets that are relevant to
pain and modulating activation in these targets
would provide objective evidence that a novel
compound engages and attenuates nociceptive
processing.
There are of course limitations and the useful-
ness of biomarkers remains a controversial topic.
Focused research is now needed to standardise
the application of neuroimaging biomarkers to
address the heterogeneity in their application in
terms of methodological approach, outcomes
examined and time points used for assessments.
More robust and better designed neuroimaging
studies are also required to develop and validate
central nervous system biomarkers capable of
assessing the benefits of future novel therapies.
The direct clinical application of brain imaging
in practice will also be dependent on a robust
health economic analysis of screening and inter-
ventions with study designs that allow relevant
and appropriate cost comparisons. Another cru-
cial element is the patient’s perspective on the
usefulness, acceptance and feasibility of these
novel imaging technologies in the provision of
future services.
4 Future Direction
Future research can be divided into 3-key areas:
mechanistic or pathophysiological, performing
clinical trials relevant to an individual with pain-
ful diabetic neuropathy and influencing federal
agencies/commissioners of diabetes and pain ser-
vices. Brain imaging techniques such as func-
tional magnetic resonance imaging (fMRI),
positron emission tomography (PET) and proton
magnetic resonance spectroscopy (1H-MRS)
provide neurochemical, structural or functional
information on the processing and modulation of
nociceptive inputs in the brain which result in the
perception of pain [13]. A summary of the bene-
fits and limitations of these different imaging
modalities can be found in Table  1 [45–51].
Neuroimaging studies have demonstrated differ-

Table 1  Some of the benefits and limitations of various neuroimaging modalities used in the investigation of CNS
lesions [46–52]
Neuroimaging
techniques Benefits Limitations
Functional • Detects local increased cerebral blood flow and • Low temporal resolution
MRI changes in blood oxygen concentration (BOLD • Signals may be affected by scanner’s
contrast) loud noise and physiologic noise
• Two primary types: Task-­based fMRI (utilisation of • Neurovascular uncoupling: False
visual, auditory or other stimuli to induce different negative BOLD responses due to
cognitive states in subjects) and resting-state fMRI interference of coupling between
(emphasis on spontaneous BOLD signal fluctuation) neuronal activity and adjacent
• Relatively high spatial resolution microvasculature
• Non-invasive
PET scan • Neuronal metabolic activity measured using • Involves the use of radioactive tracers
18
F-FDG can act as a marker of nerve injury and • Lower spatial resolution compared to
neuropathic pain fMRI
• Non-invasive
1
H-MRS • Non-invasive • Low sensitivity
• Does not involve ionising radiation • Lack of standardisation and quality
• Able to investigate a wide range of biological assurance of data acquisition and
processes analysis methods
• Combination of spectroscopy and imaging enables
the acquisition of metabolic, physiological and
anatomical data in a single experiment
434
ences in neuronal activation in response to acute
pain and chronic pain as well as differences in
functional connectivity between patients and
healthy controls. This provides an important
segue into a better understanding of the underly-
ing mechanisms involved in the generation,
maintenance and exacerbation of chronic pain
[13]. As a prognostic tool, brain imaging could
also be used to identify individuals who are more
‘vulnerable’ or ‘resilient’ against developing per-
sistent pain after being exposed to factors which
can trigger the onset of chronic pain. For exam-
ple, increased functional connectivity of the
nucleus accumbens with prefrontal cortex pre-
dicts transition from acute pain to chronic pain in
subjects with sub-acute back pain [52]. Similar
studies need to be performed in painful DPN to
determine if the same or other brain regions are
implicated. Brain imaging can also be used to
predict treatment response, hence may enable
stratification of participants in clinical trials [13].
Proof of concept for this has been demonstrated
in fibromyalgia where fMRI data derived from
visual stimulation has shown a greater than 80%
accuracy in distinguishing whether a patient with
fibromyalgia improved when administered pre-
gabalin vs. placebo [53]. In painful DPN, a quick
(6-min) RS-fMRI scan could also be used to
determine response to neuropathic pain treatment
[42]. Hence, brain imaging can be a useful diag-
nostic, prognostic and predictive tool and its use
as a potential biomarker in chronic pain clinical
trials requires further research focussing on stan-
dardisation of outcomes, validation, reproduc-
ibility and evaluation of outcomes.
The next important step in delineating CNS
involvement in diabetic neuropathy is to develop
methods that allow non-invasive assessment of
the functional changes in the spinal cord in-vivo.
This will be an important step in exploring inter-­
relationships between the peripheral and central
nervous system in greater detail. As detailed
above, there have been a number of anatomical
studies which have demonstrated increased cord
atrophy in diabetic neuropathy—however, accu-
rate assessment of (1) neuronal activity within
the spinal cord in relation to evoked/spontaneous
pain and (2) top-down inhibition of afferent
D. Selvarajah et al.
n­ ociceptive inputs will complete the picture of
CNS modulation of neuropathic pain. This poses
an enormous imaging challenge because the
small size of the spinal cord results in low signal
to noise and there is considerable motion artefact
from respiration, cardiac cycle and cerebrospinal
fluid pulsations [53]. These challenges are now
being addressed with high-field strength MR
imaging at 7 and 14-tesla opening up this part of
the neural axis for future mechanistic studies and
clinical applications [54, 55].
Another promising imaging innovation that
will significantly impact clinical and experi-
mental pain research is simultaneous positron
emission tomography and MR imaging (PET/
MR). This is a powerful non-invasive tool with
strong translational value. PET imaging serves
as a bridge from in  vitro pharmacology to
in vivo molecular readouts, whereas fMRI is an
indicator of dynamic brain function with a close
link to behaviour (Fig. 3). The PET radiotracer
2-[18F] fluoro-2-deoxy-d-glucose ([18F]FDG)
is able to quantify not only changes in brain
activity that is reflected by increases and
decreases in cerebral metabolic rate of glucose
but also the brain activity that stems primarily
from synaptic activity [56]. Combining [18F]
FDG-PET with fMRI allows the study of neu-
ronal activity on a slow time scale (minutes)
with metabolic-level PET data, and on a fast
time scale (seconds) with vascular-­level fMRI
data (Fig.  2) [56]. Thus, PET-MR provides a
more robust, direct readout of neuronal activity
in-vivo. It represents a key bridging technology,
whereby functional markers with molecular
and/or pharmacological specificity can be
quantified and tracked with anatomical preci-
sion. Combined with advanced computational
models, PET/MR provides a level of insight
into disease processes and drug effects in vivo
which corresponds much more closely to what
is possible in preclinical research. Another key
advance is the availability of PET radio-­ligands
for opioid [57] and dopamine receptors [58].
This has allowed the study of these receptor
systems in a number of clinical pain states but
presently remains solely a research tool.
Preliminary studies have showed abnormal
Central Nervous System Involvement in Painful Diabetic Neuropathy 435

Neurophysiological Haemodynamic
Neuropharmacological Neurometabolic
Glycogen
HO CH2CHCO2H

HO NH2 Glucose Glucose

NO

Pyruvate K+
Stimulation
CO2
Protein Fat
O2
TCA
Energy cycle Nutrients

Pharmacon- Increase in
Increase energy Increase in
receptor neuronal
consumption blood flow
interaction activity

PET MEG, EEG PET fMRI

Imaging

Receptor FMRI Lorenz Glucose Blood Flow Blood

Occupation effect imaging consumption oxygenation (BOLD)
imaging diffusion

Fig. 3  An illustration of the different specificity of func- lism and haemodynamics. PET positron emission
tional neuroimaging modalities to physiological processes tomography, MEG magnetoencephalography, EEG elec-
(receptor activity, electromagnetic, metabolic and haemo- troencephalography, fMRI functional magnetic resonance
dynamic) involved in neuronal activity, cerebral metabo- imaging [61]

receptor binding in a number of different
­neuropathic pain states [59] and reversal of this
abnormality following the resolution of painful
symptoms [60]. The main limitation with many
of these cross-sectional, observational studies
is whether the changes described are the cause
or consequence of the chronic pain state.
However, it is evident from studies in painful
diabetic neuropathy that the brain is ailing both
in terms of functional reorganisation and exten-
sive neurodegeneration.
Increase assay sensitivity in clinical trial—
exposing fewer patients to risk of new drugs and
reducing costs of Phase 3 studies. Another impor-
tant aspect of neuroimaging research especially
in the context of chronic pain trials is the need to
standardise the use of neuroimaging tools and
post-processing methods. Significant steps have
been made towards addressing this limitation
with the publication of various consensus state-
ments [61, 62]. Emerging approaches to imaging
analyses such as machine learning algorithms
(e.g. support vector machines, convolutional neu-
ral networks, etc.) may also prove useful in
addressing concerns about replication [13, 42].
These methods enable pooling of MR data from
studies utilising different imaging equipment and
analysis procedures and have produced informa-
tive inferences about brain structure and function
in diabetic neuropathy. These are essential steps
towards gaining regulatory approvals from fed-
eral agencies for the use of neuroimaging as a
validated biomarker in diabetic painful DPN
[13]. This would lead to increased assay sensitiv-
ity in later stage clinical trials (e.g. phase 2b or 3)
studies which will minimise patient exposure to
experimental drugs and has the potential to
reduce the overall costs of drug development.
436
Disease modifying treatments that target the
pathogenesis of diabetic neuropathy and effective
symptoms control remain the highest unmet
needs. As of yet, standard recommendations to
optimise cardiovascular risk factors effectively
(in some countries where there is local regulatory
approval for the use of agents such as aldose
reductase inhibitors, alpha-lipoic acid and benfo-
tiamine) and symptom control are the corner-
stones in practice [63]. In the future, these
recommendations could be augmented by per-
sonalised, targeted approaches towards restoring
neurotransmitter imbalance, cortical neuronal
degeneration/functional reorganisation and fur-
ther potential factors addressed in this chapter
that might represent treatment targets. As in other
microvascular complications for diabetes, e.g.
diabetic nephropathy with microalbuminuria or
retinopathy with digital retinal photography, the
identification of a robust biomarker for painful
DPN will lead to early identification and also
accelerate the development of targets of the spe-
cific mechanism that leads to chronic pain.
Moreover, the disease processes that are not
unique to diabetes but also contribute to the per-
sistence and resistance to treatment of the chronic
pain condition will be elucidated further. For
example, the CNS processes described in this
chapter in patients with painful DPN are proba-
bly still important contributors to chronic pain in
people without diabetes.
5 Conclusions
In conclusion, clinicians and patients acknowl-
edge that painful DPN is an important complica-
tion of diabetes that leads to significant
comorbidity and suffering. Hitherto considered a
disease of the peripheral nervous system, research
over the past decades has outlined the clinical
features and brain/spinal cord imaging correlates
of painful DPN.  Insights gained from clinical
research are beginning to be translated into appli-
cations but there are still gaps in our knowledge.
Current challenges include replication to provide
further support for the neuroimaging tools. It is
important to acknowledge that most of the stud-
D. Selvarajah et al.
ies described have been relatively small, cohort
studies that were exploratory that could lead to
false positive results as a consequence of an
inflated risk of type 1 error. Nevertheless, neuro-
imaging led developments for people with diabe-
tes will have wider implications for other chronic
pain aetiologies outside the field of diabetes and
hence are highly relevant.
References
1. Veves A, Backonja M, Malik RA. Painful diabetic neu-
ropathy: epidemiology, natural history, early diagno-
sis, and treatment options. Pain Med. 2008;9:660–74.
2. Sloan G, et al. A new look at painful diabetic neuropa-
thy. Diabetes Res Clin Pract. 2018;144:177–91.
3. Alleman CJM, et  al. Humanistic and economic bur-
den of painful diabetic peripheral neuropathy in
Europe: a review of the literature. Diabetes Res Clin
Pract. 2015;109:215–25.
4. Selvarajah D, et  al. The contributors of emotional
distress in painful diabetic neuropathy. Diab Vasc Dis
Res. 2014;11:218–25.
5. Feldman EL, et al. Diabetic neuropathy. Nat Rev Dis
Primers. 2019;5(1):41.
6. Alam U, Sloan G, Tesfaye S.  Treating pain in dia-
betic neuropathy: current and developmental drugs.
Drugs. 2020;80(4):363–84. https://doi.org/10.1007/
s40265-­020-­01259-­2.
7. Singh R, Rao HK, Singh TG.  Neuropathic pain in
diabetes mellitus: challenges and future trends. Obes
Med. 2020;18:100215.
8. Finnerup NB, et al. Pharmacotherapy for neuropathic
pain in adults: systematic review, meta-analysis and
updated NeuPSig recommendations. Lancet Neurol.
2015;14:162–73.
9. Kioskli K, Scott W, Winkley K, Kylakos S, McCracken
LM. Psychosocial factors in painful diabetic neuropa-
thy: a systematic review of treatment trials and survey
studies. Pain Med. 2019;20:1756–73.
10. Bönhof GJ, et  al. Emerging biomarkers, tools, and
treatments for diabetic polyneuropathy. Endocr Rev.
2018;40:153–92.
11. Finnerup NB, Sindrup SH, Jensen TS. The evidence
for pharmacological treatment of neuropathic pain.
Pain. 2010;150:573–81.
12. Backonja MM, et  al. Value of quantitative sensory
testing in neurological and pain disorders: NeuPSIG
consensus. Pain. 2013;154:1807–19.
13. Smith SM, Dworkin RH, Turk DC, Baron R,
Polydefkis M, Tracey I, Borsook D, Edwards RR,
Harris RE, Wager TD, Arendt-Nielsen L. The poten-
tial role of sensory testing, skin biopsy, and func-
tional brain imaging as biomarkers in chronic pain
clinical trials: IMMPACT considerations. J Pain.
2017;18:757–77.
Central Nervous System Involvement in Painful Diabetic Neuropathy
14. Mainka T, et  al. Presence of hyperalgesia predicts
analgesic efficacy of topically applied capsaicin 8% in
patients with peripheral neuropathic pain. Eur J Pain.
2016;20:116–29.
15. Edwards RR, et al. Patient phenotyping in clinical tri-
als of chronic pain treatments: IMMPACT recommen-
dations. Pain. 2016;157:1851–71.
16. Wasner G, Kleinert A, Binder A, Schattschneider J,
Baron R.  Postherpetic neuralgia: topical lidocaine
is effective in nociceptor-deprived skin. J Neurol.
2005;252:677–86.
17. Demant DT, Lund K, Vollert J, Maier C, Segerdahl M,
Finnerup NB, Jensen TS, Sindrup SH. The effect of
oxcarbazepine in peripheral neuropathic pain depends
on pain phenotype: a randomised, double-blind,
placebo-controlled phenotype-stratified study. Pain.
2014;155(11):2263–73.
18. Tracey I. “Seeing” how our drugs work brings
translational added value. Anesthesiology.
2013;119(6):1247–8.
19. Jensen TS, Finnerup NB.  Neuropathic pain: periph-
eral and central mechanisms. Eur J Pain Suppl.
2009;3:33–6.
20. Meacham K, Shepherd A, Mohapatra DP,
Haroutounian S.  Neuropathic pain: central vs.
peripheral mechanisms. Curr Pain Headache Rep.
2017;21:28.
21. Feldman EL, Nave KA, Jensen TS, Bennett DLH. New
horizons in diabetic neuropathy: mechanisms, bioen-
ergetics, and pain. Neuron. 2017;93:1296–313.
22. Colloca L, et  al. Neuropathic pain. Nat Rev Dis
Primers. 2017;3:17002.
23. Vinik A, Casellini C, Nevoret ML.  Diabetic neu-
ropathies. In: Feingold KR, Anawalt B, Boyce
A, et  al., editors. Endotext [Internet]. South
Dartmouth, MA: MDText.com; 2000. https://
www.ncbi.nlm.nih.gov/books/NBK279175/figure/
diab-­neuropathies_f_diab-­neuropathies_figure8/.
24. Reske-Nielsen E, Lundbaek K.  Pathological
changes in the central and peripheral nervous
system of young long-term diabetics. II.  The spi-
nal cord and peripheral nerves. Diabetologia.
1968;4:34–43.
25. Reske-Nielsen E, Lundbaek K, Gregersen G, Harmsen
A. Pathological changes in the central and peripheral
nervous system of young long-term diabetics: the
terminal neuro-muscular apparatus. Diabetologia.
1970;6:98–103.
26. Ziegler D, Mühlen H, Dannehl K, Gries FA.  Tibial
nerve somatosensory evoked potentials at various
stages of peripheral neuropathy in insulin dependent
diabetic patients. J Neurol Neurosurg Psychiatry.
1993;56(1):58–64.
27. Biessels GJ, et  al. Neurophysiological changes
in the central and peripheral nervous system of
streptozotocin-­diabetic rats. Course of development
and effects of insulin treatment. Brain. 1999;122(Pt
4):757–68.
28. Eaton SEM, et al. Spinal-cord involvement in diabetic
peripheral neuropathy. Lancet. 2001;358:35–6.
437
29. Selvarajah D, et  al. Early involvement of the spinal
cord in diabetic peripheral neuropathy. Diabetes Care.
2006;29:2664–9.
30. Selvarajah D, Wilkinson ID, Maxwell M, Davies
J, Sankar A, Boland E, et  al. Magnetic resonance
neuroimaging study of brain structural differences
in diabetic peripheral neuropathy. Diabetes Care.
2014;37(6):1681–8.
31. Zhang Y, et al. Sensorimotor and pain-related altera-
tions of the gray matter and white matter in type 2 dia-
betic patients with peripheral neuropathy. Hum Brain
Mapp. 2020;41:710–25.
32. Jensen TS, Finnerup NB.  Allodynia and hyperalge-
sia in neuropathic pain: clinical manifestations and
mechanisms. Lancet Neurol. 2014;13:924–35.
33. Selvarajah D, Wilkinson ID, Emery CJ, Shaw PJ,
Griffiths PD, Gandhi R, Tesfaye S. Thalamic neuronal
dysfunction and chronic sensorimotor distal symmet-
rical polyneuropathy in patients with type 1 diabetes
mellitus. Diabetologia. 2008;51(11):2088–92.
34. Shillo P, et  al. Painful and painless diabetic neu-
ropathies: what is the difference? Curr Diab Rep.
2019;19:32.
35. Freeman OJ, Evans MH, Cooper GJS, Petersen
RS, Gardiner NJ.  Thalamic amplification of sen-
sory input in experimental diabetes. Eur J Neurosci.
2016;44:1779–86.
36. Selvarajah D, Wilkinson ID, Gandhi R, Griffiths
PD, Tesfaye S.  Microvascular perfusion abnor-
malities of the thalamus in painful but not painless
diabetic polyneuropathy: a clue to the pathogen-
esis of pain in type 1 diabetes. Diabetes Care.
2011;34(3):718–20.
37. Watanabe K, et  al. Altered cerebral blood flow in
the anterior cingulate cortex is associated with
neuropathic pain. J Neurol Neurosurg Psychiatry.
2018;89:1082–7.
38. Segerdahl AR, Themistocleous AC, Fido D, Bennett
DL, Tracey I. A brain-based pain facilitation mecha-
nism contributes to painful diabetic polyneuropathy.
Brain. 2018;141:357–64.
39. Wager TD, et  al. An fMRI-based neurologic
signature of physical pain. N Engl J Med.
2013;368:1388–97.
40. Selvarajah D, et  al. Structural and functional abnor-
malities of the primary somatosensory cortex in dia-
betic peripheral neuropathy: a multimodal MRI study.
Diabetes. 2019;68:796–806.
41. Lv H, et al. Resting-state functional MRI: everything
that nonexperts have always wanted to know. Am J
Neuroradiol. 2018;39:1390–9.
42. Teh K, Wilkinson ID, Heiberg-Gibbons F, Awadh M,
Kelsall A, Pallai S, Sloan G, Tesfaye S, Selvarajah
D.  Somatosensory network functional connectivity
differentiates clinical pain phenotypes in diabetic neu-
ropathy. Diabetologia. 2021;64(6):1412–21.
43. Wilkinson ID, et  al. Determinants of treatment
response in painful diabetic peripheral neuropathy: a
combined deep sensory phenotyping and multimodal
brain MRI study. Diabetes. 2020;69:1804–14.
438
44. Lu C, et al. Insular cortex is critical for the perception,
modulation, and Chronification of pain. Neurosci
Bull. 2016;32:191–201.
45. Glover GH.  Overview of functional magnetic reso-
nance imaging. Neurosurg Clin N Am. 2011;22:133–9.
46. Chen JE, Glover GH.  Functional magnetic reso-
nance imaging methods. Neuropsychol Rev.
2015;25:289–313.
47. Shruti A, Sair HI, Pillai JJ. Limitations of rsfMRI in
the setting of focal brain lesions. Neuroimaging Clin
N Am. 2017;27:645–61.
48. Slough C, Masters SC, Hurley RA, Taber KH. Clinical
positron emission tomography (PET) neuroimag-
ing: advantages and limitations as a diagnostic tool. J
Neuropsychiatry Clin Neurosci. 2016;28:1–6.
49. Behera D, Jacobs KE, Behera S, Rosenberg J, Biswal
S. 18F-FDG PET/MRI can be used to identify injured
peripheral nerves in a model of neuropathic pain. J
Nucl Med. 2011;52:1308–12.
50. Chatham JC, Blackband SJ.  Nuclear magnetic reso-
nance spectroscopy and imaging in animal research.
ILAR J. 2001;42:189–208.
51. Öz G, et  al. Clinical proton MR spectroscopy
in central nervous system disorders. Radiology.
2014;270:658–79.
52. Baliki MN, et  al. Corticostriatal functional connec-
tivity predicts transition to chronic back pain. Nat
Neurosci. 2012;15:1117–9.
53. López-Solà M, Woo CW, Pujol J, Deus J, Harrison
BJ, Monfort J, Wager TD.  Towards a neuro-
physiological signature for fibromyalgia. Pain.
2017;158(1):34–47.
54. Brooks JCW, et al. Physiological noise modelling for
spinal functional magnetic resonance imaging stud-
ies. Neuroimage. 2008;39:680–92.
D. Selvarajah et al.
55. Maieron M, et al. Functional responses in the human
spinal cord during willed motor actions: evidence
for side- and rate-dependent activity. J Neurosci.
2007;27:4182–90.
56. Wehrl HF, Hossain M, Lankes K, Liu CC, Bezrukov
I, Martirosian P, Schick F, Reischl G, Pichler
BJ.  Simultaneous PET-MRI reveals brain function
in activated and resting state on metabolic, hemo-
dynamic and multiple temporal scales. Nat Med.
2013;19:1184–9.
57. Jones AKP, et  al. Measurement of changes in opi-
oid receptor binding in  vivo during trigeminal neu-
ralgic pain using [11C]diprenorphine and positron
emission tomography. J Cereb Blood Flow Metab.
1999;19:803–8.
58. Hagelberg N, et  al. Striatal dopamine D2 receptors
in modulation of pain in humans: a review. Eur J
Pharmacol. 2004;500:187–92.
59. Harris RE, et  al. Decreased central μ-opioid
receptor availability in fibromyalgia. J Neurosci.
2007;27:10000–6.
60. Jones AKP, et  al. Changes in central opioid recep-
tor binding in relation to inflammation and pain in
patients with rheumatoid arthritis. Rheumatology.
1994;33:909–16.
61. Uludağ K, Roebroeck A.  General overview on the
merits of multimodal neuroimaging data fusion.
Neuroimage. 2014;102:3–10.
62. Huang L, et al. Brain white matter changes associated
with urological chronic pelvic pain syndrome: multi-
site neuroimaging from a MAPP case-control study.
Pain. 2016;157:2782–91.
63. Sloan G, Selvarajah D, Tesfaye S. Pathogenesis, diag-
nosis and clinical management of diabetic peripheral
neuropathy. Nat Rev Endocrinol. 2021;17(7):400–20.
 Part III
Clinical Consequences and Treatments
 Characteristics and Treatment
of Painful Diabetic Neuropathy
Sandra Sif Gylfadottir and Nanna Brix Finnerup
1 Definition, Epidemiology,
Diagnosis,
and Characteristics
Painful diabetic neuropathies include painful dia-
betic radiculoplexopathy, mononeuropathies,
treatment-induced neuropathies induced by fast
improvement of glycemic control after hypergly-
cemia, entrapment neuropathies, and distal sym-
metrical sensorimotor polyneuropathies [1]. The
focus of this chapter will be on painful diabetic
polyneuropathy (P-DPN) as this group represents
the majority of the patients in clinical setting [2]
and the diagnostic and treatment approaches to
the different types of painful diabetic neuropa-
thies are similar, except for treatment-induced
neuropathy (discussed in chapter “Asymmetric
Diabetic Neuropathy: Radiculoplexus
Neuropathies, Mononeuropathies, and Cranial
Neuropathies”).
Pain is defined as “An unpleasant sensory and
emotional experience associated with, or resem-
bling that associated with, actual or potential tis-
sue damage” [3]. P-DPN is a subgroup of
neuropathic pain, which is defined as “pain
S. S. Gylfadottir (*) · N. B. Finnerup
Department of Clinical Medicine, Danish Pain
Research Center, Aarhus University,
Aarhus, Denmark
Department of Neurology, Aarhus University
Hospital, Aarhus, Denmark
e-mail: sifgyl@clin.au.dk; finnerup@clin.au.dk
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_25
caused by a lesion or disease of the somatosen-
sory nervous system” [4]. Of patients with DPN,
25–50% experience pain due to their DPN [5–
11]. Many factors influence the prevalence esti-
mates of P-DPN, including different definitions
and assessment methods of P-DPN as well as dif-
ferences in duration and type of diabetes [8,
12–19].
Patients with P-DPN often complain of both
painful and non-painful sensations. Non-painful
sensations include paresthesia, which are not
unpleasant, e.g. tingling, and dysesthesia, which
are unpleasant sensations, e.g. pricking. The pain
can be spontaneous or evoked and examples of
neuropathic pain symptom descriptions can be
burning, squeezing, shooting, or pricking pain,
although the pain can be described in a similar
way as in other pain conditions [20]. Therefore
the pain description itself is normally not suffi-
cient to diagnose P-DPN and it has to be con-
firmed by a clinical examination and supportive
diagnostic tests. Evoked pain to, e.g. touch or
cold is uncommon in P-DPN compared to some
other neuropathic pain conditions. Pain and other
abnormal sensory symptoms usually start in the
feet and may spread to the legs and hands in the
most severe cases.
The clinical examination consists of a stan-
dard neurological examination with a focus on
sensory examination of feet and hands, or any
other affected area. In Fig. 1 there are examples
of tools that can be used in the bedside examina-
441
442
tion of a patient presenting with P-DPN. In 2016
the International Association for the Study of
pain special interest group on neuropathic pain
(NeuPSIG) proposed an upgraded grading sys-
tem for neuropathic pain, both for research and
Fig. 1  Example of tools that can be used in the bedside
examination of the patient presenting with P-DPN. Left to
right: tuning fork for vibration sense, brush for light brush
stroking, Neuropen® for light touch and pinprick sensa-
tion, reflex hammer, temperature rolls, cotton wool swab,
and ear pin
Painful diabetic polyneuropathy
Possible painful DPN
Diabetes and pain in both feet
Probable painful DPN
Pain is associated with sensory signs in the feet/
legs, neuroanatomically plausible with a
polyneuropathy
Definite painful DPN
Abnormal NCS and/or IENFD
Fig. 2  The diagnosis of P-DPN [21]. At all steps, other causes of pain and neuropathy should be considered
S. S. Gylfadottir and N. B. Finnerup
clinical practice [21, 22]. Applying this for
P-DPN (Fig. 2), possible P-DPN requires a his-
tory of P-DPN, pain in the feet, and a diagnosis of
diabetes. Probable P-DPN additionally requires
sensory signs in the same area, and definite
P-DPN requires abnormal tests that confirm a
structural or functional damage to the sensory
nerves (examples: skin biopsy and nerve conduc-
tion studies). At all stages it is important to con-
sider other causes of polyneuropathy with similar
or the same symptom presentation (i.e. alcohol or
chemotherapy induced, infective or inflammatory
polyneuropathy) or conditions presenting with
painful feet or legs (i.e. arthritis, spinal stenosis,
herniated disc, peripheral arterial diseases) [10].
It is difficult to replace bedside clinical exami-
nation with standardized assessment tools,
although they can be useful in screening, epide-
miological research, and to aid in the diagnosis of
neuropathic pain [23]. An example of screening
and diagnostic tool is the douleur neuropathique
en 4 questions (DN4). The DN4 consists of seven
questions and a clinical examination part and was
designed to discriminate neuropathic pain from
other types of pain [24]. The questions are focused
on pain and paresthesia (pins and needles, numb-
Other causes of pain
excluded
Characteristics and Treatment of Painful Diabetic Neuropathy
ness, itching, burning, painful cold, and electric
shock sensation) and the examination of brush
and pinprick hypoesthesia and dynamic mechani-
cal allodynia (DMA) in the feet/legs. The DN4
questionnaire part has been validated and used for
P-DPN [19, 24, 25]. In a recent validation study in
type 2 diabetes patients, the sensitivity and speci-
ficity of the DN4 questionnaire for definitely
diagnosed P-DPN were 80.0 (44.4; 97.5) and 89.9
(83.6; 94.3), ­respectively, for a cut of point ≥3/7
[10]. An example of a questionnaire that can be
used to quantify and characterize the symptoms of
neuropathic pain is the neuropathic pain symptom
inventory (NPSI) with 12 questions about the
presence, character, and intensity of pain and the
answers can be converted into sum scores [26].
Other neuropathic pain tools are available, such as
the PainDETECT, the Leeds Assessment of
Neuropathic Symptoms and Signs questionnaire,
and the Neuropathic Pain Questionnaire [23]. The
numerical rating scale (NRS) from 0 to 10 or a
Visual Analog Scale (VAS) is used to measure the
intensity of the pain with 0 being no pain and ten
the worst pain imaginable and is simple to admin-
ister [27]. In the assessment of pain it is also
important to assess the impact of pain on daily
activities, sleep, mood, function, and quality of
life.
It is a puzzle why some patients with DPN
develop painful symptoms, while others do not
[28, 29]. Most studies are cross-sectional, and
there is generally a lack of longitudinal studies to
elucidate the natural course of the disease and the
risk factors for developing neuropathic pain in
diabetes. Type 2 diabetes compared to type 1 dia-
betes and higher measures of HbA1c have been
associated with P-DPN. Recent studies have also
found correlation of P-DPN with increased sever-
ity of neuropathy [7, 8, 30–32]. Other risk factors
that have been linked to P-DPN are smoking,
female sex, increasing age, obesity, and the co-­
occurrence of other diabetic complications [16,
33, 34]. The genetic component in the risk profile
of P-DPN is described in chapter “The Genomics
of Diabetic Neuropathy”.
Psychosocial factors such as reduced quality
of life, depression, sleep deprivation, and anxiety
have consistently been shown to be associated
443
with P-DPN and pain intensity [7, 12, 17, 31, 32,
35–40]. Most studies are cross-sectional and
have therefore not examined the causal relation-
ship between having a chronic neuropathic pain
and mental health. A large questionnaire study of
Danish patients with type 2 diabetes and neurop-
athy found that both patients with DPN with and
without pain reported worse mental health and
quality of life, although patients with pain were
more severely affected [38].
2 Treatment of P-DPN
Information about pain and pain mechanisms and
discussion of aggravating and alleviating factors
and pain coping strategies are often the first step
in pain treatment. While the evidence for psycho-
logical and self-care interventions are limited for
chronic pain, most guidelines recommend this in
the treatment of chronic pain [41]. For any treat-
ment, realistic expectations for the outcome
should be discussed with the patient, as often
only partial pain relief can be expected. A multi-
disciplinary approach with physicians, pain
nurses, psychologists, social workers, and phys-
iotherapists may be needed in severe pain that is
not easily treated [42–44]. Furthermore, a close
collaboration between diabetes treatment units
and primary care physicians, which patients
attend regularly, and pain clinics may be helpful
and convenient for the patients.
3 Pharmacological Treatment
Currently there is no generally established dis-
ease modifying therapy for P-DPN and the aim of
the treatment is therefore mainly improvement of
modifiable risk factors, quality of life, and symp-
tomatic pain relief.
3.1 Pathogenesis Orientated
Treatment
Optimal glycemic control and the assessment of
vascular and lifestyle risk factors are important.
444
It is well known that hyperglycemia plays a role
in the pathogeneses of neuropathy [45], and it
may contribute to the pathogeneses of neuro-
pathic pain, although the evidence for improved
glycemic control as a pain relief is limited as
there is a lack of randomized controlled studies
[43, 46, 47]. In a meta-analysis, enhanced glyce-
mic control was significantly effective in the pre-
vention of neuropathy in type 1 diabetes patients,
but not in type 2 diabetes [48]. A clinical study in
type 2 diabetes however showed improvement in
neurophysiological parameters with strict blood
sugar control without hypoglycemia for a longer
period of time in patients with poor glycemic
control [49].
Alpha-lipoic acid (ALA) is suggested to be a
causative treatment of DPN by having beneficial
effects on vascular dysfunction, tissue hypoxia,
and glucose utilization. It is also suggested to
prevent painful symptoms by possibly suppress-
ing the activation of TRPV1 channel via activa-
tion of the NF-κB, a cellular nuclear transcription
factor protein complex, but with limited effect
from clinical trials [50–53]. Another potential
neuroprotective agent in DPN is Actovegin that
consists of filtered extract from calf serum and is
thought to have antihypoxic effect on tissue. In a
multicenter, double blinded, randomized trial,
treatment with Actovegin improved symptoms of
neuropathy including pain, paresthesia, and
numbness and the improvement was considered
to be clinically relevant [54, 55].
Vitamin D has been thought to play a role in
peripheral neuropathy by stimulating keratino-
cytes in the skin to produce nerve growth factor,
with D vitamin as a potential therapeutic agent
[56]. No randomized, double-blinded studies
have assessed the efficacy of vitamin D in
P-DPN [57, 58].
3.2 Symptomatic Treatment
The challenges in the medical treatment of neu-
ropathic pain lie in the relatively high numbers
needed to treat (NNT) which is the number of
patients needed to treat for one to have 50% pain
reduction and often intolerable side effects of the
S. S. Gylfadottir and N. B. Finnerup
medication [59]. Additionally patients with
P-DPN can be elderly and suffer from other
chronic diseases or other macro- or microvascu-
lar diabetes complications, for example, heart
and kidney diseases, potentially reinforcing the
side effects or rendering the recommended treat-
ment contraindicated.
In 2016 the neuropathic pain special interest
group of the International Association for the
Study of Pain (NeuPSIG) presented treatment
guidelines for neuropathic pain based on system-
atic review and meta-analysis [59]. These recom-
mendations are based on a large number of
studies and apply to patients with neuropathic
pain of many etiologies, based on the conclusion
made by the authors that the effect of the treat-
ment does not depend on the etiology [59]. These
recommendations are in line with other recom-
mendations for painful DPN, from the European
Federation of Neurological Societies (EFNS)
[60], the American Academy of Neurology [61],
the Agency for Health Research and Quality
(AHRQ) [62], the Toronto Expert Panel on
Diabetic Neuropathy [46], and a recent meta-­
analysis on P-DPN [63].
3.3 Guideline Recommendations
Pregabalin and gabapentin, tricyclic antidepres-
sants (TCA), and serotonin noradrenaline reup-
take inhibitors (SNRIs) are recommended as a
first-line therapy with a strong recommendation
and moderate to high quality of evidence [60,
61, 63]. In terms of safety, there is no evidence
to suggest one of these drugs is above others
[59, 64]. Duloxetine has been recommended for
painful DPN.  Capsaicin 8% and lidocaine 5%
patches are recommended as a second-line ther-
apy with moderate to high quality of evidence
and a week’s recommendation for use. Finally,
botulin toxin A subcutaneous injection is rec-
ommended as third-­ line treatment option
(Table 1 and Fig. 3) [59].
The results for combination therapy were
inconclusive in both meta-analyses [59, 63],
although the search for sufficient pain relief in
the clinical setting often results in a combination
Characteristics and Treatment of Painful Diabetic Neuropathy 445

Table 1  Guidelines recommendation

Daily doses Common side effects Precautions
First-line recommendations
Tricyclic 25-(100) 150 mg, Sedation, dry mouth, constipation, People with seizures,
antidepressants starting with doses of blurred vision, weight gain, glaucoma, cardiac diseases,
(TCAs) 10–25 mg orthostatic hypotension with risk and dysuria and the elderly
of falls in elderly patients. >65 years.
ECG before starting
treatment for QT interval.
Serotonin-­ 60–120 mg for Nausea, somnolence, dizziness, In people with seizures,
norepinephrine duloxetine and constipation, dry mouth, glaucoma, uncontrolled
inhibitors (SNRIs) 150–225 mg for hypertension. Often mild and hypertension, and in the
venlafaxine in one transient side effects elderly.
dose
Pregabalin 300–600 mg in two Dizziness, somnolence, headache, Reduced doses in people
doses/day peripheral edema, and weight gain. with impaired kidney
Risk of suicidal thoughts function
Gabapentin 900–3600 mg in three Same as for pregabalin Same as for pregabalin
doses/day
Second-line recommendations
Lidocaine 5% One to three patches in Mild redness, swelling, irritation, Avoid contact with eyes,
patches the painful area up to change in skin color, and nose, and mouth
12 h numbness
Capsaicin 8% One to four patches in Local discomfort and skin Should not be used in the
patches the painful area for reaction/burning face and on or near mucous
30–60 min every membranes
3 months
Examples of SNRIs: Venlafaxine, Duloxetine, TCAs: Amitriptyline, Nortriptyline, Imipramine. For detailed informa-
tion on doses, side effects, and precautions, please refer to individual product information

of more than one of the recommended drugs, 3.4 Modes of Action,

preferable using drugs with different modes of
action. The question whether it is better to
increase the drug dose of a monotherapy with one
of the recommended drugs or if it is better to
combine two drugs in patients that do not respond
to a standard dose was addressed in the
COMBO-DN study of patients with P-DPN [65].
The comparison between high recommended
dose of pregabalin or duloxetine in patients that
did not respond to standard dose was compared
to the effect of a combination therapy of the two
drugs at moderate dosages, with no differences
between the groups, although the combination
therapy was considered well tolerated and safe
[65]. It is worth noting that patients with diabetes
often receive multiple medication for glycemic
control, hypertension, and dyslipidemia and in
some cases medication for sleep disturbances,
depression, or anxiety. Polypharmacy may there-
fore raise concerns about compliance, adherence,
and drug–drug interactions [66].
Recommended Dose, and Side
Effects
Tricyclic antidepressants (TCAs) act on the nor-
adrenergic system by inhibiting presynaptic
reuptake of monoamines (noradrenalin and sero-
tonin) and the place of action for the TCAs is
thought to be both central and peripheral [20, 67].
Other mechanisms may also be involved, includ-
ing blockade of voltage-gated sodium channels
and NMDA (N-methyl-d-aspartate) receptors
[68, 69]. Experimental data have suggested that
TCAs work through an effect on β2 adrenergic
receptors. However, a recent randomized con-
trolled trial found that the tricyclic antidepressant
imipramine but not the β-agonist terbutaline had
effect on painful polyneuropathy suggesting that
β2 agonisms are not a clinical important mecha-
nism of action of TCAs in painful polyneuropa-
thy [70]. Examples of TCAs are amitriptyline,
nortriptyline, and imipramine and the overall
446 S. S. Gylfadottir and N. B. Finnerup

Painful diabetic polyneuropathy

Before initiating drug treatment asess

comorbidity, contraindictations and side effects.

First line drug therapy

TCA Non-pharmacological treatment

SNRIs Provide patient education, encourage self-care
and consider psychological help and
Gabapentin/pregabalin physiotherapy.

If there is partiel, but inadequate pain

control with monotherapy with one ore
more of the first line therapy options in a
recommended dose, the combination of
TCA+gabapentin or pregabalin, or
SNRIs+gabapentin or pregabalin can be
considered.

Second line drug therapy

Lidocain 5% patches

Capsacain 8% patches
Topical treatment can be indicated in a
selected group of patients.

Fig. 3  Treatment algorithm for P-DPN

number of patients needed to treat (NNT) to
achieve at least 50% reduction in pain intensity
for one patient for TCAs in neuropathic pain was
3.6 (95%CI 3.0–4.4) where most trials included
amitriptyline [59]. A starting dose of 10–25 mg
increasing with 10  mg every fifth day until
desired pain relief is obtained where maximum
dose of 100–150 mg is recommended. The most
common side effects are dry mouth, constipation,
sedation, blurred vision, weight gain, and ortho-
static hypotension with the risk of falls in elderly
patients. An ECG before treatment is recom-
mended to exclude prolongation of PR and QT
intervals. Secondary amines (nortriptyline and
desipramine) have lower affinity for muscarinic
acetylcholine receptors and imipramine has
lower affinity to block histaminergic H1 recep-
tors than amitriptyline and may therefore cause
less sedative side effects [69, 71]. Caution should
be shown in the treatment of elderly people
>65 years where lower doses are required because
of risk of sedation, falls, and sudden cardiac
death [72, 73].
SNRIs also target the serotonin noradrenergic
system by inhibiting presynaptic reuptake of sero-
tonin and noradrenalin, augmenting descending
inhibitory pathways. Examples are duloxetine and
venlafaxine. Most common adverse effects are
nausea, somnolence, dizziness, constipation, and
dry mouth; these side effects are often mild and
transient. Recommended daily doses are
60–120 mg/day for duloxetine and 150–225 mg/
day for venlafaxine. The NNT for 50% pain reduc-
tion for SNRIs is 6.4 (95% CI 5.2–8.4) [20, 59].
Characteristics and Treatment of Painful Diabetic Neuropathy 447

The group of antidepressants should be tried [76, 78]. Tramadol, which is a centrally acting
for some time for pain relief before switching to synthetic opioid working on both opioid and
another drug because lack of effect [69]. In monoaminergic pathways, may be considered
patients with concomitant depression TCAs and for flares of pain. A range of drugs have incon-
SNRIs may have some effect on depressive clusive evidence based on conflicting results
symptoms [41]. from randomized trials, including lamotrigine,
Gabapentin and pregabalin have similar oxcarbazepine, topiramate, and selective sero-
modes of action and bind to the α2δ subunit of tonin-reuptake inhibitors, and thus there is no
the presynaptic voltage-gated calcium channels evidence to support their use for P-DPN, and
and thereby reduce Ca2+ influx resulting in further studies are needed to support their use,
decreased neurotransmitter release [20, 67]. The e.g. in specific subtypes of patients [59]. One
site of action can be peripheral, spinal, and study found oxcarbamazepine to be effective in
supraspinal. The NNT for pregabalin is 7.7 (6.5; a subgroup of patients with painful polyneu-
9.4) and for gabapentin 7.2 (5.9; 9.1) [59]. ropathy and a sensory phenotype with pre-
Common side effects are dizziness, somnolence, served small-fiber sensation (cold and warm
peripheral edema, headache, and weight gain. sensation and sensory hypersensitivity com-
Daily doses for gabapentin are 900–3600  mg pared to those without), but these results have
daily in three doses starting with 300  mg/day not been confirmed [79]. Recent systematic
and for pregabalin 300–600  mg in two doses reviews find no evidence for the use of cannabi-
starting with 75  mg/day. There are increasing noids, cannabis, and cannabis-based medicines
concerns and evidence for misuse of gabapentin in chronic pain [80, 81]. The NeuPSIG guide-
and pregabalin [74]. lines have a weak recommendation against the
Topical treatment with capsaicin 8% and lido- use of cannabis-based medicines in neuropathic
caine 5% patches can be beneficial in a selected pain because of lack of effect in clinical trials,
group of patients. Capsaicin binds to the transient potential misuse, diversion, and long-term
receptor potential vanilloid 1 (TRPV1) on noci- mental health risks [59].
ceptors desensitizing them for 12  weeks.
Normally one to four patches are applied in the
painful area for 30–60  min every 3  months. 3.6 Emerging Treatment
Lidocaine inhibits ectopic nerve activity by Possibilities
blocking voltage-gated sodium channels [20].
One to three lidocaine 5% patches are applied in Several drug classes are currently in clinical
the painful area for up to 12 h. development for neuropathic pain, although to
date, we have seen a failure in developing new
drug classes for neuropathic pain [82, 83]. A
3.5 Other Pharmacological peripherally acting drug EMA401, which is a
Treatment Options competitive antagonist to angiotensin II type 2
receptor, showed promising effects in peripheral
Tramadol and strong opioids have also shown neuropathic pain [84], but recently phase 2 trials
some effect in randomized controlled trials of have been terminated due to safety reasons.
short duration, but are generally not recom- Antibodies targeting the activity of human nerve
mended for the treatment of chronic non-­ growth factors are also in development for pain.
malignant pain because of long-term Recombinant human nerve growth factor had no
consequences, risk of abuse and dependence, effect on diabetic polyneuropathy in a large trial
and lack of effect on functional status [75–78]. but showed some effect on pain in the legs, which
If opioids are indicated for medium-term use, a was one of many secondary outcomes [85].
low dose in a monitored setting following local Tanezumab, a humanized monoclonal antibody
opioid prescribing guidelines is recommended against nerve growth factor was found to reduce
448
P-DPN despite early study termination due to a
partial clinical hold by the US Food and Drug
Administration (FDA) due to joint-related safety
[86], but currently the drug is only being devel-
oped for osteoarthritis and other pain types.
Subtype-selective sodium channel inhibitors, in
particular Nav1.7 blockers, are also being investi-
gated for neuropathic pain.
4 Nonpharmacological
Treatment
The first treatment of neuropathic pain offered to
patients is most often pharmacological, but it is
also important to consider nonpharmacological
treatment options, either as the only treatment or
as an addition to medical treatment.
With a general agreement of having P-DPN
being associated with worse mental health, there
are paradoxically few studies that have examined
the effect of psychological treatment in P-DPN
[39, 87]. Psychological therapy for chronic pain
may consist of cognitive behavioral therapy
(CBT), a traditional therapy form that helps
patients to identify, change, and provide coping
strategies for negative thoughts or behavior asso-
ciated with pain. Another therapy form derived
from CBT also used in the treatment of pain is
acceptance and commitment therapy (ACT)
which integrates acceptance, cognitive defusion,
and mindfulness to produce more psychological
flexibility [88]. In a recently published single arm
primarily feasibility trial with online ACT treat-
ment in patients with P-DPN the clinical out-
comes showed beneficial effect on psychological
functions and pain intensity, although the study
has limitations due to design [89]. In a newly
updated systematic review of the effect of psy-
chological therapies on the broad spectrum of
chronic pain, compared to standard treatment or
waiting list, there was only very small to small
beneficial effect of CBT in pain reduction, dis-
ability, and stress [90]. In separate studies of neu-
ropathic pain and P-DPN the evidence was too
sparse and insufficient to make any conclusions
on the effect of psychological treatment in pain
relief [39, 87]. Therefore psychological treatment
S. S. Gylfadottir and N. B. Finnerup
may have a place in the treatment of painful DPN
although the evidence is sparse.
Physiotherapy treatment options can be aimed
at pain reduction, preserve function level, and
prevent immobilization or it can consist of thera-
peutic exercise [91]. In a qualitative study,
patients with P-DPN described walking difficul-
ties and other impacts of pain on physical func-
tions [92]. A review of physical therapy in
chemotherapy induced neuropathy showed that
there was possible a symptom effect, although
with limited evidence [93].
In general, physical activity in adjutancy to
other treatments or alone could have positive
effects on P-DPN, although the evidence is lim-
ited with a high risk of bias [94, 95].
Yoga, mindfulness, and meditation are gain-
ing increased popularity as a treatment option in
various conditions, often for the improvement of
mental health and quality of life, but data on the
efficacy for pain relief are few or missing in the
treatment of neuropathic pain, the same is for
acupuncture [95–97].
Neurostimulation has been used as a treatment
option in P-DPN. Examples are spinal cord stim-
ulation (SCS) and transcutaneous electrical stim-
ulation (TENS). According to the newly
published guidelines from the European Academy
of Neurology (EAN), there is weak evidence for
adding spinal cord stimulation (SCS) to the phar-
macological treatment in P-DPN compared with
pharmacological treatment alone [98] and a
recent meta-analysis including randomized con-
trolled trials concluded that SCS was effective for
pain relief compared to standard care [99]. A
Cochrane review of TENS treatment for neuro-
pathic pain could not conclude whether the treat-
ment was effective compared to sham therapy or
not, because of the size and quality of the included
studies, neither were side effects nor adverse
events adequately addressed in the studies [100].
5 Conclusion
Painful neuropathy is a common and often dis-
abling consequence of diabetes. It is a neuro-
pathic pain and is caused by the lesion of the
Characteristics and Treatment of Painful Diabetic Neuropathy
peripheral nerves. The pharmacological treat-
ment is challenging and often provides only par-
tial pain relief at best. Therefore both
nonpharmacological and pharmacological treat-
ments are recommended.
References
1. Izenberg A, Perkins BA, Bril V.  Diabetic neuropa-
thies. Semin Neurol. 2015;35:424–30.
2. Gylfadottir SS, Weeracharoenkul D, Andersen
ST, Niruthisard S, Suwanwalaikorn S, Jensen
TS.  Painful and non-painful diabetic polyneuropa-
thy: clinical characteristics and diagnostic issues. J
Diabetes Investig. 2019;10:1148–57.
3. Raja SN, Carr DB, Cohen M, et  al. The revised
International Association for the Study of Pain defi-
nition of pain: concepts, challenges, and compro-
mises. Pain. 2020;161:1976.
4. Taxonomy I. https://www.iasp-­pain.org/terminology
?navItemNumber=576
5. Tesfaye S, Boulton AJ, Dyck PJ, et al. Diabetic neu-
ropathies: update on definitions, diagnostic criteria,
estimation of severity, and treatments. Diabetes
Care. 2010;33:2285–93.
6. Callaghan BC, Cheng HT, Stables CL, Smith AL,
Feldman EL.  Diabetic neuropathy: clinical mani-
festations and current treatments. Lancet Neurol.
2012;11:521–34.
7. Themistocleous AC, Ramirez JD, Shillo PR, et  al.
The Pain in Neuropathy Study (PiNS): a cross-­
sectional observational study determining the
somatosensory phenotype of painful and painless
diabetic neuropathy. Pain. 2016;157:1132–45.
8. Abbott CA, Malik RA, van Ross ER, Kulkarni J,
Boulton AJ. Prevalence and characteristics of pain-
ful diabetic neuropathy in a large community-based
diabetic population in the U.K.  Diabetes Care.
2011;34:2220–4.
9. Andersen ST, Witte DR, Dalsgaard EM, et al. Risk
factors for incident diabetic polyneuropathy in a
cohort with screen-detected type 2 diabetes followed
for 13 years: addition-Denmark. Diabetes Care.
2018;41:1068–75.
10. Gylfadottir SS, et  al. Diagnosis and prevalence of
diabetic polyneuropathy: a cross-sectional study of
Danish patients with type 2 diabetes. Eur J Neurol.
2020;27:2575.
11. van Hecke O, Austin SK, Khan RA, Smith BH,
Torrance N. Neuropathic pain in the general popula-
tion: a systematic review of epidemiological studies.
Pain. 2014;155:654–62.
12. Feldman EL, Callaghan BC, Pop-Busui R, et  al.
Diabetic neuropathy. Nat Rev Dis Primers.
2019;5:41.
13. Bonhof GJ, Herder C, Strom A, Papanas N, Roden
M, Ziegler D. Emerging biomarkers, tools, and treat-
449
ments for diabetic polyneuropathy. Endocr Rev.
2019;40:153–92.
14. Partanen J, Niskanen L, Lehtinen J, Mervaala
E, Siitonen O, Uusitupa M.  Natural history of
peripheral neuropathy in patients with non-insu-
lin-dependent diabetes mellitus. N Engl J Med.
1995;333:89–94.
15. Ziegler D, Strom A, Lobmann R, Reiners K, Rett
K, Schnell O.  High prevalence of diagnosed and
undiagnosed polyneuropathy in subjects with and
without diabetes participating in a nationwide edu-
cational initiative (PROTECT study). J Diabetes
Complicat. 2015;29:998–1002.
16. Truini A, Spallone V, Morganti R, et al. A cross sec-
tional study investigating frequency and features of
definitely diagnosed diabetic painful polyneuropa-
thy. Pain. 2018;159:2658.
17. Van Acker K, Bouhassira D, De Bacquer D, et  al.
Prevalence and impact on quality of life of periph-
eral neuropathy with or without neuropathic
pain in type 1 and type 2 diabetic patients attend-
ing hospital outpatients clinics. Diabetes Metab.
2009;35:206–13.
18. Aslam A, Singh J, Rajbhandari S.  Prevalence of
painful diabetic neuropathy using the self-completed
Leeds assessment of neuropathic symptoms and
signs questionnaire in a population with diabetes.
Can J Diabetes. 2015;39:285–95.
19. Bouhassira D, Letanoux M, Hartemann A. Chronic
pain with neuropathic characteristics in diabetic
patients: a French cross-sectional study. PLoS One.
2013;8:e74195.
20. Finnerup NB, Kuner R, Jensen TS.  Neuropathic
pain: from mechanisms to treatment. Physiol Rev.
2020;101:259.
21. Finnerup NB, Haroutounian S, Kamerman P, et  al.
Neuropathic pain: an updated grading system for
research and clinical practice. Pain. 2016;157:1599.
22. Treede RD, Jensen TS, Campbell JN, et  al.
Neuropathic pain: redefinition and a grading sys-
tem for clinical and research purposes. Neurology.
2008;70:1630–5.
23. Attal N, Bouhassira D, Baron R.  Diagnosis and
assessment of neuropathic pain through question-
naires. Lancet Neurol. 2018;17:456–66.
24. Bouhassira D, Attal N, Alchaar H, et al. Comparison
of pain syndromes associated with nervous or
somatic lesions and development of a new neuro-
pathic pain diagnostic questionnaire (DN4). Pain.
2005;114:29–36.
25. Spallone V, Morganti R, D’Amato C, Greco C,
Cacciotti L, Marfia GA.  Validation of DN4 as a
screening tool for neuropathic pain in painful dia-
betic polyneuropathy. Diabet Med. 2012;29:578–85.
26. Bouhassira D, Attal N, Fermanian J, et  al.
Development and validation of the neuropathic pain
symptom inventory. Pain. 2004;108:248–57.
27. Jensen MP, Karoly P, Braver S. The measurement of
clinical pain intensity: a comparison of six methods.
Pain. 1986;27:117–26.
450
28. Spallone V, Greco C.  Painful and painless diabetic
neuropathy: one disease or two? Curr Diab Rep.
2013;13:533–49.
29. Shillo P, Sloan G, Greig M, et al. Painful and pain-
less diabetic neuropathies: what is the difference?
Curr Diab Rep. 2019;19:32.
30. Tesfaye S, Boulton AJ, Dickenson AH. Mechanisms
and management of diabetic painful distal
symmetrical polyneuropathy. Diabetes Care.
2013;36:2456–65.
31. Davies M, Brophy S, Williams R, Taylor A.  The
prevalence, severity, and impact of painful diabetic
peripheral neuropathy in type 2 diabetes. Diabetes
Care. 2006;29:1518–22.
32. Raputova J, Srotova I, Vlckova E, et  al. Sensory
phenotype and risk factors for painful diabetic neu-
ropathy: a cross-sectional observational study. Pain.
2017;158:2340–53.
33. Gylfadottir SS, Christensen DH, Nicolaisen SK,
et al. Diabetic polyneuropathy and pain, prevalence,
and patient characteristics: a cross-sectional ques-
tionnaire study of 5,514 patients with recently diag-
nosed type 2 diabetes. Pain. 2019;161:574.
34. Hebert HL, Veluchamy A, Torrance N, Smith
BH.  Risk factors for neuropathic pain in diabetes
mellitus. Pain. 2016;158:560.
35. Naranjo C, Del Reguero L, Moratalla G, Hercberg
M, Valenzuela M, Failde I. Anxiety, depression and
sleep disorders in patients with diabetic neuropathic
pain: a systematic review. Expert Rev Neurother.
2019;19:1201–9.
36. Gore M, Brandenburg NA, Dukes E, Hoffman DL,
Tai KS, Stacey B.  Pain severity in diabetic periph-
eral neuropathy is associated with patient function-
ing, symptom levels of anxiety and depression, and
sleep. J Pain Symptom Manag. 2005;30:374–85.
37. D’Amato C, Morganti R, Greco C, et  al. Diabetic
peripheral neuropathic pain is a stronger predictor
of depression than other diabetic complications and
comorbidities. Diab Vasc Dis Res. 2016;13:418–28.
38. Gylfadottir SS, Christensen DH, Nicolaisen SK,
et al. Diabetic polyneuropathy and pain, prevalence,
and patient characteristics: a cross-sectional ques-
tionnaire study of 5,514 patients with recently diag-
nosed type 2 diabetes. Pain. 2020;161:574–83.
39. Kioskli K, Scott W, Winkley K, Kylakos S,
McCracken LM.  Psychosocial factors in pain-
ful diabetic neuropathy: a systematic review of
treatment trials and survey studies. Pain Med.
2019;20:1756–73.
40. Geelen CC, Smeets R, Schmitz S, van den Bergh JP,
Goossens M, Verbunt JA. Anxiety affects disability
and quality of life in patients with painful diabetic
neuropathy. Eur J Pain. 2017;21:1632–41.
41. Finnerup NB. Nonnarcotic methods of pain manage-
ment. N Engl J Med. 2019;380:2440–8.
42. Scascighini L, Toma V, Dober-Spielmann S, Sprott
H.  Multidisciplinary treatment for chronic pain: a
systematic review of interventions and outcomes.
Rheumatology (Oxford). 2008;47:670–8.
S. S. Gylfadottir and N. B. Finnerup
43. Sloan G, Shillo P, Selvarajah D, et  al. A new look
at painful diabetic neuropathy. Diabetes Res Clin
Pract. 2018;144:177–91.
44. Shaygan M, Böger A, Kröner-Herwig B. Predicting
factors of outcome in multidisciplinary treat-
ment of chronic neuropathic pain. J Pain Res.
2018;11:2433–43.
45. Calcutt NA.  Diabetic neuropathy and neuropathic
pain: a (con)fusion of pathogenic mechanisms? Pain.
2020;161:S65–s86.
46. Tesfaye S, Vileikyte L, Rayman G, et al. Painful dia-
betic peripheral neuropathy: consensus recommen-
dations on diagnosis, assessment and management.
Diabetes Metab Res Rev. 2011;27:629–38.
47. Boulton AJ, Drury J, Clarke B, Ward JD. Continuous
subcutaneous insulin infusion in the management
of painful diabetic neuropathy. Diabetes Care.
1982;5:386–90.
48. Callaghan BC, Little AA, Feldman EL, Hughes
RA.  Enhanced glucose control for preventing and
treating diabetic neuropathy. Cochrane Database
Syst Rev. 2012;6:CD007543.
49. Ishibashi F, Taniguchi M, Kosaka A, Uetake H,
Tavakoli M.  Improvement in neuropathy outcomes
with normalizing HbA(1c) in patients with type 2
diabetes. Diabetes Care. 2019;42:110–8.
50. Papanas N, Ziegler D. Emerging drugs for diabetic
peripheral neuropathy and neuropathic pain. Expert
Opin Emerg Drugs. 2016;21:393–407.
51. Papanas N, Ziegler D. Efficacy of alpha-lipoic acid
in diabetic neuropathy. Expert Opin Pharmacother.
2014;15:2721–31.
52. Zhang BY, Zhang YL, Sun Q, et  al. Alpha-lipoic
acid downregulates TRPV1 receptor via NF-κB and
attenuates neuropathic pain in rats with diabetes.
CNS Neurosci Ther. 2020;26:762–72.
53. Ziegler D, Nowak H, Kempler P, Vargha P, Low
PA. Treatment of symptomatic diabetic polyneurop-
athy with the antioxidant alpha-lipoic acid: a meta-­
analysis. Diabet Med. 2004;21:114–21.
54. Ziegler D, Edmundson S, Gurieva I, Mankovsky B,
Papanas N, Strokov I. Predictors of response to treat-
ment with actovegin for 6 months in patients with
type 2 diabetes and symptomatic polyneuropathy. J
Diabetes Complicat. 2017;31:1181–7.
55. Ziegler D, Movsesyan L, Mankovsky B, Gurieva I,
Abylaiuly Z, Strokov I.  Treatment of symptomatic
polyneuropathy with actovegin in type 2 diabetic
patients. Diabetes Care. 2009;32:1479–84.
56. Fukuoka M, Sakurai K, Ohta T, Kiyoki M, Katayama
I.  Tacalcitol, an active vitamin D3, induces nerve
growth factor production in human epidermal
keratinocytes. Skin Pharmacol Appl Ski Physiol.
2001;14:226–33.
57. Shillo P, Selvarajah D, Greig M, et al. Reduced vita-
min D levels in painful diabetic peripheral neuropa-
thy. Diabet Med. 2019;36:44–51.
58. Yammine K, Wehbe R, Assi C.  A systematic
review on the efficacy of vitamin D supplementa-
Characteristics and Treatment of Painful Diabetic Neuropathy
tion on diabetic peripheral neuropathy. Clin Nutr.
2020;39:2970–4.
59. Finnerup NB, Attal N, Haroutounian S, et  al.
Pharmacotherapy for neuropathic pain in adults: a
systematic review and meta-analysis. Lancet Neurol.
2015;14:162–73.
60. Attal N, Cruccu G, Baron R, et al. EFNS guidelines on
the pharmacological treatment of neuropathic pain:
2010 revision. Eur J Neurol. 2010;17:1113–e1188.
61. Bril V, England J, Franklin GM, et al. Evidence-based
guideline: treatment of painful diabetic neuropa-
thy: report of the American Academy of Neurology,
the American Association of Neuromuscular and
Electrodiagnostic Medicine, and the American
Academy of Physical Medicine and Rehabilitation.
Neurology. 2011;76:1758–65.
62. Agency for Health Research and Quality (AHRQ).
Accessed 15 Jan 2021.
63. Waldfogel JM, Nesbit SA, Dy SM, et  al.
Pharmacotherapy for diabetic peripheral neuropa-
thy pain and quality of life: a systematic review.
Neurology. 2017;88:1958–67.
64. Enomoto H, Yasuda H, Nishiyori A, et al. Duloxetine
in patients with diabetic peripheral neuropathic pain
in Japan: a randomized, doubleblind, noninferior-
ity comparative study with pregabalin. J Pain Res.
2018;11:1857–68.
65. Tesfaye S, Wilhelm S, Lledo A, et al. Duloxetine and
pregabalin: high-dose monotherapy or their combi-
nation? The “COMBO-DN study”--a multinational,
randomized, double-blind, parallel-group study in
patients with diabetic peripheral neuropathic pain.
Pain. 2013;154:2616–25.
66. Finnerup NB. Add-on therapy: when two are not bet-
ter than one. Pain. 2013;154:2579–80.
67. Kremer M, Salvat E, Muller A, Yalcin I, Barrot
M.  Antidepressants and gabapentinoids in neuro-
pathic pain: mechanistic insights. Neuroscience.
2016;338:183–206.
68. Javed S, Alam U, Malik RA.  Burning through the
pain: treatments for diabetic neuropathy. Diabetes
Obes Metab. 2015;17:1115–25.
69. Sindrup SH, Otto M, Finnerup NB, Jensen
TS.  Antidepressants in the treatment of neu-
ropathic pain. Basic Clin Pharmacol Toxicol.
2005;96:399–409.
70. Gillving M, Demant D, Holbech JV, et al. A random-
ized, controlled trial of a β2-agonist in painful poly-
neuropathy. Pain. 2020;162:1364.
71. Gillman PK.  Tricyclic antidepressant pharmacol-
ogy and therapeutic drug interactions updated. Br J
Pharmacol. 2007;151:737–48.
72. American Geriatrics Society. Updated beers criteria
for potentially inappropriate medication use in older
adults. J Am Geriatr Soc. 2015;2015(63):2227–46.
73. Ray WA, Meredith S, Thapa PB, Hall K, Murray
KT. Cyclic antidepressants and the risk of sudden car-
diac death. Clin Pharmacol Ther. 2004;75:234–41.
74. Evoy KE, Sadrameli S, Contreras J, Covvey JR,
Peckham AM, Morrison MD.  Abuse and misuse
451
of pregabalin and gabapentin: a systematic review
update. Drugs. 2020;81:125.
75. Franklin GM.  Opioids for chronic noncancer pain:
a position paper of the American Academy of
Neurology. Neurology. 2014;83:1277–84.
76. Dowell D, Haegerich TM, Chou R. CDC guideline
for prescribing opioids for chronic pain--United
States, 2016. JAMA. 2016;315:1624–45.
77. Hoffman EM, Watson JC, St Sauver J, Staff NP,
Klein CJ. Association of Long-term Opioid Therapy
With Functional Status, adverse outcomes, and mor-
tality among patients with polyneuropathy. JAMA
Neurol. 2017;74:773–9.
78. Ballantyne JCBS, Blyth F, Cordosa M, Finley
A, Furlan A, et  al. https://www.iasp-­pain.org/
Advocacy/OpioidPositionStatement.2018. Accessed
19 Jan 2021.
79. Demant DT, Lund K, Vollert J, et  al. The effect
of oxcarbazepine in peripheral neuropathic pain
depends on pain phenotype: a randomised, double-­
blind, placebo-controlled phenotype-stratified study.
Pain. 2014;155:2263–73.
80. Fisher E, Moore RA, Fogarty AE, et al. Cannabinoids,
cannabis, and cannabis-based medicine for pain
management: a systematic review of randomised
controlled trials. Pain. 2020;162:S45.
81. Moore RA, Fisher E, Finn DP, et al. Cannabinoids,
cannabis, and cannabis-based medicines for pain
management: an overview of systematic reviews.
Pain. 2020;162:S67.
82. Attal N, Bouhassira D.  Translational neuropathic
pain research. Pain. 2019;160(Suppl 1):S23–s28.
83. Yekkirala AS, Roberson DP, Bean BP, Woolf
CJ. Breaking barriers to novel analgesic drug devel-
opment. Nat Rev Drug Discov. 2017;16:545–64.
84. Rice ASC, Dworkin RH, McCarthy TD, et  al.
EMA401, an orally administered highly selective
angiotensin II type 2 receptor antagonist, as a novel
treatment for postherpetic neuralgia: a randomised,
double-blind, placebo-controlled phase 2 clinical
trial. Lancet. 2014;383:1637–47.
85. Apfel SC, Schwartz S, Adornato BT, et al. Efficacy
and safety of recombinant human nerve growth factor
in patients with diabetic polyneuropathy: a random-
ized controlled trial. rhNGF Clinical Investigator
Group. JAMA. 2000;284:2215–21.
86. Bramson C, Herrmann DN, Carey W, et  al.
Exploring the role of tanezumab as a novel treat-
ment for the relief of neuropathic pain. Pain Med.
2015;16:1163–76.
87. Eccleston C, Hearn L, Williams AC. Psychological
therapies for the management of chronic neuro-
pathic pain in adults. Cochrane Database Syst Rev.
2015;2015:CD011259.
88. Hayes SC, Luoma JB, Bond FW, Masuda A, Lillis
J. Acceptance and commitment therapy: model, pro-
cesses and outcomes. Behav Res Ther. 2006;44:1–25.
89. Kioskli K, Scott W, Winkley K, Godfrey E,
McCracken LM. Online acceptance and commitment
therapy for people with painful diabetic neuropathy
452
in the United Kingdom: a single-arm feasibility trial.
Pain Med. 2020;21:2777–88.
90. Williams ACC, Fisher E, Hearn L, Eccleston
C.  Psychological therapies for the management
of chronic pain (excluding headache) in adults.
Cochrane Database Syst Rev. 2020;8:CD007407.
91. Akyuz G, Kenis O.  Physical therapy modalities
and rehabilitation techniques in the management
of neuropathic pain. Am J Phys Med Rehabil.
2014;93:253–9.
92. Brod M, Pohlman B, Blum SI, Ramasamy A,
Carson R.  Burden of illness of diabetic periph-
eral neuropathic pain: a qualitative study. Patient.
2015;8:339–48.
93. Jesson T, Runge N, Schmid AB. Physiotherapy for
people with painful peripheral neuropathies: a nar-
rative review of its efficacy and safety. Pain Rep.
2020;5:1–e834.
94. Singleton JR, Smith AG, Marcus RL.  Exercise as
therapy for diabetic and prediabetic neuropathy.
Curr Diab Rep. 2015;15:120.
95. Davies B, Cramp F, Gauntlett-Gilbert J, Wynick D,
McCabe CS. The role of physical activity and psy-
chological coping strategies in the management of
S. S. Gylfadottir and N. B. Finnerup
painful diabetic neuropathy--a systematic review of
the literature. Physiotherapy. 2015;101:319–26.
96. Baute V, Zelnik D, Curtis J, Sadeghifar
F.  Complementary and alternative medicine for
painful peripheral neuropathy. Curr Treat Options
Neurol. 2019;21:44.
97. van Laake-Geelen CCM, Smeets R, Quadflieg S,
Kleijnen J, Verbunt JA. The effect of exercise ther-
apy combined with psychological therapy on physi-
cal activity and quality of life in patients with painful
diabetic neuropathy: a systematic review. Scand J
Pain. 2019;19:433–9.
98. Cruccu G, Garcia-Larrea L, Hansson P, et al. EAN
guidelines on central neurostimulation therapy in
chronic pain conditions. Eur J Neurol. 2016;23:1489.
99. Amato Nesbit S, Sharma R, Waldfogel JM, et  al.
Non-pharmacologic treatments for symptoms of
diabetic peripheral neuropathy: a systematic review.
Curr Med Res Opin. 2019;35:15–25.
100. Gibson W, Wand BM, O’Connell NE.
Transcutaneous electrical nerve stimulation (TENS)
for neuropathic pain in adults. Cochrane Database
Syst Rev. 2017;9:CD011976.
 Orthostatic Hypotension
and Sudomotor Dysfunction
in Diabetes
Lauren F. Fanty and Christopher H. Gibbons
1 Introduction
Diabetic autonomic neuropathy (DAN) is the
leading cause of autonomic neuropathy in the
USA. There are multiple mechanisms underlying
the pathophysiology of diabetic neuropathy
including neurovascular insufficiency, metabolic
insult to nerve fibers, neurohormonal growth fac-
tor insufficiency, and bioenergy failure [1].
Generalized DAN typically is gradual in onset
and leads to progressive autonomic impairment
to vasomotor (vascular), visceromotor (gastroin-
testinal), parasympathetic, sympathetic adrener-
gic, and sympathetic cholinergic autonomic
fibers [2–4]. Clinical features are broad, as DAN
affects the cardiovascular, gastrointestinal, pupil-
lomotor, thermoregulatory, sudomotor, and uro-
genital systems. Many large epidemiologic
studies have established the increased morbidity,
mortality, and sudden death in patients with dia-
betic autonomic neuropathy [4–9].
The prevalence of diabetic autonomic neurop-
athy varies depending on diagnostic criteria,
patient cohort, and testing modality with esti-
L. F. Fanty
Department of Neurology, Beth Israel Deaconess
Medical Center, Harvard Medical School,
Boston, MA, USA
C. H. Gibbons (*)
Beth Israel Deaconess Medical Center,
Boston, MA, USA
e-mail: cgibbons@bidmc.harvard.edu
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_26
mates ranging from 7% to 90% [10]. Prevalence
of autonomic impairment also varies between the
types of diabetes, with an estimated prevalence of
54% and 73% in type 1 and type 2 diabetes,
respectively [10]. Cardiac autonomic neuropathy
(CAN) is the most life-threatening form of dia-
betic autonomic neuropathy with a 56% 10-year
mortality [5, 10, 11]. In CAN, Vagus nerve dener-
vation leads to reduced heart rate variability and
presents initially as a resting tachycardia [6].
With continued disease progression, sympathetic
cardiac denervation leads to orthostatic hypoten-
sion and chronotropic incompetence [6, 12, 13].
Sudomotor (sweating) dysfunction is often
comorbid with CAN and one of the earliest signs
of DAN. Sudomotor dysfunction typically occurs
in a length-dependent fashion, often in parallel
with sensory neuropathy.
In DAN, orthostatic hypotension is due to
sympathetic adrenergic failure, while sudomotor
dysfunction function is due to sympathetic cho-
linergic failure [6]. This chapter will focus on the
pathophysiology, clinical presentation, diagnosis,
and treatment of orthostatic hypotension and
sudomotor dysfunction in patients with diabetes.
2 Orthostatic Hypotension
Orthostatic hypotension, OH, is defined as a fall
in systolic blood pressure (SBP) of 20 mmHg or
fall in diastolic blood pressure (DBP) of 10 mmHg
453
454
within 3  min of standing or a head-up tilt to at
least 60° on a tilt table [14]. It is one of the most
incapacitating manifestations of autonomic fail-
ure [6]. Orthostatic hypotension in isolation may
be a clinical sign indicating a variety of medical
conditions such as severe anemia, intravascular
volume depletion, physical deconditioning, anti-
hypertensive medication use, or impairment of
peripheral vasoconstriction [15, 16].
Neurogenic orthostatic hypotension, nOH, is a
subset of OH that occurs when the autonomic ner-
vous system fails to provide sufficient autonomic
postural response via sympathetic vasoconstric-
tion and fails to increase heart rate to maintain
blood pressure [15]. In diabetes, there is degener-
ation of sympathetic pre- and postganglionic
fibers leading to failure in both the vagal and
adrenergic limbs of the baroreflex resulting in
nOH [6]. In a normal response to moving to the
upright position, there is an increase in plasma
norepinephrine as well as increased cardiac out-
put and peripheral vasoconstriction. In nOH, there
is impaired release of peripheral norepinephrine
with impaired peripheral vasoconstriction and
impaired cardiac acceleration in response to a fall
in blood pressure [17, 18]. Inadequate blood pres-
sure response is most often attributable to
decreased norepinephrine released from postgan-
glionic sympathetic nerves leading to pooling of
blood in the splanchnic region, pelvis, and depen-
dent areas [15]. Compensatory increase in heart
rate when standing from a seated position is aug-
mented by norepinephrine and sympathetic nerve
endings, which may also be impaired with sympa-
thetic dysfunction [6, 19]. The Rochester Diabetic
Study, which evaluated a cohort of 83 Type I dia-
betics and 148 Type II diabetics, found that of the
approximately 50% of patient with clinical mani-
festations of neuropathy, about 10% additionally
had clinical evidence of autonomic neuropathy
[20, 21]. Orthostatic hypotension was present in
8.4% of type 1 diabetics and 7.4% of type 2 dia-
betics [2, 20, 21]. Other estimates report the prev-
alence of orthostatic hypotension to be as high as
20–25% in patients with long term diabetes [2].
A symptom that is sometimes associated with
nOH is worsening of orthostatic tolerance after
eating secondary to a drop in blood pressure,
L. F. Fanty and C. H. Gibbons
known as postprandial hypotension. Postprandial
hypotension is defined as a fall of at least 10 mmHg
within 2 h of eating [22]. This is caused by degen-
eration of the fibers supplying the splanchnic mes-
enteric bed, a large vascular bed that increases in
volume by 200–300% after a carbohydrate heavy
meal, resulting in orthostatic stress from venous
pooling [22]. Normally, this increase in volume is
compensated for by an increase in sympathetic
response leading to splanchnic vasoconstriction
[19]. However, this compensatory mechanism is
deficient in patients with nOH.
Supine hypertension is another common com-
plication detected in an estimated 50% of patients
with nOH [15]. Supine hypertension is defined as
a supine SBP  ≧  150 or DBP  ≧  90 [15, 16]. In
patients with supine hypertension, a reduction in
SBP of 30 mmHg or DBP of 15 mmHg may be a
more appropriate criterion for nOH [15]. Supine
hypertension is a product of both impaired baro-
reflex function and chronically increased activa-
tion of the renin–angiotensin system from
orthostatic hypotension [15, 19]. nOH and supine
hypertension are associated conditions with the
same underlying pathophysiology but create a
challenge to effectively treat both conditions
simultaneously [15, 19].
While our focus of discussion will be on nOH
secondary to disruption of peripheral sympatheti-
cally mediated reflex vasoconstriction attributed
to diabetic neuropathy, it is important to note that
other neuropathies, neurodegenerative disorders,
and high spinal cord lesions also affect peripheral
sympathetic neurons [19]. Some neurodegenera-
tive disorders, such as multiple systems atrophy,
affect central sympathetic neurons resulting in
nOH from central autonomic dysfunction, which
will not be a focus of this chapter [15].
3 Clinical Presentation
and Initial Evaluation
The prevalence of orthostatic hypotension in dia-
betes is estimated to range from 6% to 30% and
increases with age, disease duration, and disease
control [2, 6, 23]. It is more common in institution-
alized than community dwelling elderly and is
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes
normally a late manifestation of cardiovascular
autonomic neuropathy [10]. Neurogenic OH can
be either symptomatic or asymptomatic, thus
­creating a barrier to both recognition and treatment
[24]. It is the recommendation of the American
Autonomic Society that patients with peripheral
neuropathies known to be associated with auto-
nomic dysfunction, such as diabetes, be screened
for nOH [15]. At minimum, patients should be
asked if they have symptoms, they commonly
experience within 3–5 min of standing that resolve
when sitting or lying down. There are additionally
different variants of OH, and while some occur
within the first 15  s of standing (initial OH),
delayed OH occurs after 3 min of standing [15].
Please see Table 1 for a full list of recommended
screening questions in patients with suspect nOH.
3.1 Symptoms
Orthostatic hypotension is thought to be one of
the most incapacitating manifestations of auto-
nomic failure [6]. Symptoms are limited to stand-
ing and resolve when sitting or lying down.
Inadequate perfusion of the brain may cause diz-
ziness, lightheadedness, cognitive slowing,
blurry or dimmed vision, “coat hanger” pain in
Table 1  Screening questions for patients with suspect
orthostatic hypotension
1. Have you fainted/blacked out recently?
2. Do you feel dizzy or lightheaded upon standing?
3. Do you have vision disturbances when standing?
4. Do you have difficulty breathing when standing?
5. Do you have leg buckling or leg weakness when
standing?
6. Do you ever experience neck pain or aching when
standing?
7. Do the above symptoms improve or disappear when
you sit or lay down?
8. Are the above symptoms worse in the morning or
after meals?
9. Have you experienced a fall recently?
10. Are there any other symptoms you commonly
experience when you stand up or within 3–5 min of
standing and get better when you sit or lay down?
Any of the questions that are answered “yes” should
prompt measurement of orthostatic vital signs
Adapted from Gibbons et al. 2017 [15]
455
the upper body, fatigue, dyspnea, chest pain, and
syncope [16]. Patients may also report worsening
of symptoms during exercise, within 2 h of eat-
ing, or that symptoms are exacerbated by large,
carbohydrate rich meals and alcohol [16].
Symptoms may show diurnal variability (worse
in the morning) and are influenced by hydration
and ambient temperature [14, 15].
3.2 Initial Steps
If any of the screening questions for OH are posi-
tive, orthostatic vitals should be obtained in the
clinic. If possible, have patients track their ortho-
static vital signs at home upon awakening, when
symptomatic, and prior to bedtime for around a
week to aid in accurate clinical assessment of the
condition [15]. In addition to recording changes
in blood pressure, orthostatic changes to heart
rate should be evaluated. A change in heart rate of
<15 bpm within 3 min of moving from lying to
standing in the setting of OH suggests that the
OH is due to neurogenic causes because the com-
pensatory sympathetic heart rate increase is
impaired [25]. In nOH, a ΔHR/ΔSBP ratio
<0.5  beats/min/mmHg is suggestive of neuro-
genic orthostatic hypotension [25]. Conversely,
heart rate change >15 bpm or ΔHR/ΔSBP ratio
>0.5  beats/min/mmHg is suggestive of non-­
neurogenic cause of orthostatic hypotension [25].
The relevance of the heart rate ratio was evalu-
ated in a prospective study of 423 patients and
neurogenic OH was reliably distinguished from
other causes of OH when the ΔHR/ΔSBP ratio at
3  min of tilt was <0.5  beats/min/mmHg [25].
However, this criterion can be confounded by
medications that interfere with heart rate, intrin-
sic cardiac rhythm disturbances that impair car-
diac compensatory increase in heart rate, and by
baseline variability in age and disease status [15].
3.3 Evaluation for Additional
Causes of OH
In diabetic patients with changes in blood pres-
sure and heart rate suggestive of OH, history,
456 L. F. Fanty and C. H. Gibbons

physical exam, and basic diagnostics should be
obtained to evaluate for other non-neurogenic
causes of OH. An electrocardiogram can be used
to evaluate for cardiac dysfunction, complete
blood count to look for infection and anemia,
basic metabolic panel for electrolyte distur-
bances, acid base disorders, volume depletion
and renal impairment, thyroid stimulating hor-
mone, vitamin B12, albumin to identify chronic
illness or poor nutrition, and liver enzyme testing
to evaluate for liver dysfunction [15]. Serum and
urine protein electrophoresis can be considered if
not already completed as part of a neuropathy
workup. Neurologic antibodies can be sent rarely
if clinical suspicion is high [15]. It is important to
review the patient’s medication list as polyphar-
macy is common in patients with diabetes and
increases the risk of OH [15]. See Table 2 for a
list of medication types that may cause or exacer-
bate OH.
3.4 Additional Autonomic Testing
If clinical suspicion is high for OH and vital sign
testing is unrevealing, clinicians should consider
conducting further testing to evaluate for delayed
OH [13, 26]. These tests might include 24-h
ambulatory blood pressure monitoring, supine
and standing norepinephrine levels, and more
specialized autonomic testing such as tilt table
testing, autonomic reflex testing (heart rate vari-
ability to paced breathing, heart rate, and blood
pressure response to a Valsalva maneuver), and
sudomotor function testing [15].
3.4.1 Valsalva Maneuver
In the Valsalva maneuver, the patient exhales
against resistance in order to increase intratho-
racic pressure resulting in reduced cardiac pre-
load. The Valsalva maneuver is done in a
standardized manner. The patient rests in a

Table 2  Medications that may cause or worsen orthostatic hypotension

Medication class Common examples
Alcohol
Anticholinergics Atropine, glycopyrrolate, hyoscyamine
Antidepressants/tricyclics Amitriptyline, nortriptyline, imipramine, desipramine, trazodone,
selective serotonin reuptake inhibitors
Benzodiazepines Alprazolam, chlordiazepoxide, clonazepam diazepam, lorazepam
Central sympatholytic agents Centrally acting α2 agonists: clonidine
False neurotransmitters: alpha-methyldopa
Dopaminergic agents Levodopa, bromocriptine, cabergoline, apomorphine, pramipexole,
ropinirole, rotigotine
Drugs that induce autonomic neuropathy Amiodarone, vincristine, cisplatin
Insulin
Muscle relaxants Tizanidine
Negative inotropic/chronotropic agents Beta-adrenergic blockers: propranolol, metoprolol, atenolol, carvedilol,
labetalol
Non-dihydropyridine calcium channel blockers
Neuroleptics Chlorpromazine, quetiapine
Opioids (rare side effect) Oxycodone, hydrocodone, morphine, hydromorphone, heroin
Preload reducers Diuretics: acetazolamide, furosemide, hydrochlorothiazide,
spironolactone, torsemide
Nitrates: isosorbide dinitrate, nitroglycerine, nitroprusside
Phosphodiesterase 5 inhibitors: sildenafil, tadalafil, vardenafil
Renin-angiotensin system (RAS) Angiotensin converting enzyme inhibitors: captopril, enalapril, lisinopril
antagonists Angiotensin receptor blockers: losartan, telmisartan, candesartan, valsartan
Vasodilators α1 adrenoceptor antagonists: alfuzosin, doxazosin, prazosin, terazosin,
tamsulosin, carvedilol, labetalol
Dihydropyridine calcium channel blockers: amlodipine, nifedipine,
nicardipine
Other vasodilators: hydralazine, minoxidil
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes 457

Fig. 1  The phases of

Phase of Valsalva
the Valsalva maneuver.
In the top panel the heart I II early II late III IV
rate is shown in a solid
black line in beats/min. 120

Heart Rate
In the middle panel the
range of systolic and 90
diastolic blood pressures
(mmHg) during the 60
Valsalva maneuver are Heart Rate
shown with the thick
grey line. In the lower
panel, the expiratory
pressure (mmHg) is
120
shown with the solid
black line
60
Blood Pressure

30
mmHg

0
Expiratory Pressure
In

Ex

En
sp

pir

d
ira

Ex
at
tio

ion

pir
n

at
ion

recumbent position and is asked to blow into a
mouthpiece connected to a manometer and main-
tain a column of mercury at 40 mmHg for 15 s
[27]. Beat-to-beat blood pressure recordings are
obtained and the four main phases of Valsalva
maneuver are analyzed [see Fig.  1]. The four
phases include:
• Phase I: At the beginning of strain/expira-
tion, there is a transient rise in blood pres-
sure and a fall in heart rate due to propulsion
of blood into peripheral circulation and
compression of the aorta [27].
• Phase II: This phase has an early fall and
then recovery of blood pressure. Positive
intrathoracic pressure leads to impaired
venous return, reduced preload and stroke
volume, and overall fall in blood pressure.
The fall in blood pressure activates aortic
and carotid baroreceptors leading to
increased sympathetic discharges, increased
heart rate, increased cardiac output, and
peripheral vasoconstriction [27].
• Phase III: Marks the end of strain/expira-
tion. With release in positive intrathoracic
pressure, blood pressure falls as there is
expansion of the pulmonary vascular bed,
and heart rate increases [27].
• Phase IV: Characterized by an overshoot of
blood pressure above baseline. Normal
venous return of the heart coupled with
sympathetic activity from phase II leads to
overshoot of blood pressure from baseline.
Increased blood pressure stimulates barore-
flex and return to normal blood pressure.
Baroreflex also results in bradycardia
before return to baseline [27].
In sympathetic adrenergic failure, there is a
loss of phase II recovery and phase IV overshoot
and a delay in blood pressure recovery time. The
sensitivity and specificity of this test to detect
458 L. F. Fanty and C. H. Gibbons

adrenergic failure and separate neurogenic from caution in patients with a sensory neuropathy who
non-neurogenic OH are both 80% [27]. are at risk for ulcer formation) [15, 16, 19]. Based
on the clinical response, pharmacologic interven-
3.4.2 Tilt Table tions may be considered. Please note that it is rec-
The tilt table allows for a controlled environment ommended to wait 2  weeks between any
to monitor the autonomic response to orthostatic interventions to fully assess efficacy [17].
stress. During a tilt table test, a patient is pas-
sively transitioned to the upright position to at
least 60°. Gravity redistributes an estimated 500– 5 Medication Review
8000 mL of blood (10% of total blood volume) and Adjustment
from cerebral and thoracic vasculature to the
splanchnic and lower extremities [27]. Upon Medications that reduce intravascular volume,
transitioning to the upright posture, the force of promote vasodilation, reduce preload, or are neg-
gravity causes the blood volume to redistribute ative inotropic/chronotropic agents can exacer-
away from the cerebral and thoracic vasculature bate symptoms of OH. These medications should
toward the splanchnic and lower extremity vascu- be reviewed with patients and prescribing physi-
lature. The tilt test is a stronger orthostatic stress cians and slowly reduced/discontinued as clini-
because it removes the leg contraction that occurs cally safe. Blood pressure and heart rate responses
during standing [28]. In healthy individuals, should be monitored in response to changes to
moving to the upright position unloads the carotid help identify offending agents. Please refer to
and cardiopulmonary baroreceptors leading to a Table 2 for comprehensive list of medication cat-
reflex increase in sympathetic adrenergic output egories that should be removed if possible [15].
[27, 28]. The sympathetic adrenergic response
leads to increased peripheral vasoconstriction,
increased heart rate, and increased cardiac con- 6 Nonpharmacological
tractility. In nOH, there is an insufficient auto- Interventions
nomic response to orthostatic stress leading to a
drop in blood pressure without a compensatory 6.1 Lifestyle
tachycardia. Tilt table testing often detects OH
within 3–5 min, but up to 60 min may be required 6.1.1 Volume Repletion: Increase
to detect delayed OH [29]. Hydration and Salt Intake
Many patients are chronically volume depleted
and benefit from increasing daily water intake. A
4 Treatment minimum of 2 L daily (64 oz) is recommended to
ensure adequate hydration and if cardiac status
In patients with nOH secondary to diabetes, man- permits, up to 3  L daily is recommended [15].
agement of glycemic index should be optimized Patients with nOH can also bolus water intake
[2, 6, 30]. Further treatment of OH should be with rapid results. A 16  oz (500  mL) bolus of
geared toward optimal management of symptoms water can increase SBP by 30  mmHg in about
and not toward a specific target blood pressure. 5–10 min [15, 16, 31]. Effects however are short-­
Initial treatments should first focus on nonphar- lasting, wearing off after 45–60  min [16].
macologic measures such as removal of offending However, this can be a useful tool to counteract
medications, increase in dietary salt and fluid postprandial hypotension or OH in the setting of
intake, education on the use of physical counter-­ specific brief activities.
maneuvers, regular exercise, sleeping with head The 24 h urine sodium for patients not taking
of bed elevated (to reduce supine hypertension diuretics or fludrocortisone can be checked to
and reduce nocturia), and use of specific compres- ensure adequate daily sodium intake is occurring.
sion garments (but these must be considered with If urine sodium is <100 mEq/24 h, then consider
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes
salt repletion. It is recommended 1–2 teaspoons
of salt be added to daily diet; however, this can be
increased up to 10 g for those that required vol-
ume repletion. For those at risk for cardiovascu-
lar complications, salt intake and clinical
response should be monitored closely and down-­
titrated as needed. An alternative is salt tabs
0.5–1 g, with the caveat they can cause abdomi-
nal discomfort [15, 16].
6.1.2 Avoid Warm, Humid
Environments and Increase
in Core Body Temperature
Elevation in body temperature can lead to vasodi-
lation and loss of intravascular volume through
sweating, both contributing to more severe drop
in orthostatic blood pressure. Thus, saunas, hot
tubs, and hot showers should be avoided. Shower
chairs can be utilized for an extra safety measure
[15, 16]. In addition, sudden exposure to hot
humid environments (such as winter vacations to
a warm environment) can be fraught with peril
without adequate preparation and hydration.
Patients should be counseled on the need to avoid
outdoor activities during the hottest part of the
day and should take it slowly until they know
how their blood pressure will respond.
6.1.3 Remain Active
High intensity activity which can raise body tem-
perature should be avoided. However, regular
exercise is encouraged with adequate prehydra-
tion to prevent deconditioning [15]. Exercises
such as swimming, rowing, and recumbent bike
may be better tolerated than standing exercises
because they remove the orthostatic stress of
standing. Exercising in pools is beneficial as
hydrostatic pressure from water counteracts grav-
ity induced pooling of blood [16]. However,
patients must take extra care when emerging
from the pool because the sudden loss of hydro-
static pressure can result in dramatic OH [16].
Prolonged bedrest (such as during a hospitaliza-
tion) can result in significant deconditioning in a
short period of time and dramatically worsen
OH.  Patients will require aggressive physical
therapy in order to gain back the orthostatic toler-
ance that is lost during deconditioning.
459
6.1.4 Purposeful Movement
Valsalva maneuvers or straining with closed glot-
tis can lead to decreased preload and worsening
of orthostatic symptoms and should be performed
cautiously (e.g. fainting on the toilette is a com-
mon situation in the setting of constipation) [15,
16]. Patients should change positions gradually
and sit before standing. Activities like leg cross-
ing, standing on tiptoes, squatting, and buttock
clenching can all help to promote venous blood
return [16].
6.1.5 Eat Small, Low Carbohydrate
Meals
Orthostatic symptoms may worsen after meals
due to pooling of blood in the splanchnic bed.
To minimize this effect, recommend eating
small meals with low glycemic index.
Avoiding sugary beverages will also be bene-
ficial [15, 16].
6.1.6 Avoid Alcohol and Caffeine
Alcohol can act as a vasodilator, thereby worsen-
ing symptoms of nOH.  Although caffeine is a
stimulant, it also has diuretic properties and can
lead to intravascular depletion and worsening
symptoms of nOH if not consumed with appro-
priate water hydration [15, 16]. In general prac-
tice, avoiding both alcohol and caffeine helps to
prevent complications, but can be considered in
special situations with appropriate forethought
(this often helps to preserve the physician–patient
relationship).
6.1.7 Sleep with Head of Bed
Elevated
Vascular depletion occurs overnight secondary
to pressure natriuresis in the setting of supine
hypertension [16]. Sleeping with the head of
bed elevated approximately 6  in. can help to
reduce nocturnal diuresis and symptomatic
orthostatic hypotension in the morning.
Elevating head of bed also helps to maintain
renin–angiotensin–aldosterone system to main-
tain higher blood pressure in the morning [15,
16]. Additionally, sleeping with the head of bed
elevated helps reduce neurogenic supine
hypertension.
460
6.1.8 Compression Garments
Compression garments applying 30–40  mmHg
can be used to help prevent venous pooling. With
most venous pooling occurring in the splanchnic
bed, waist high compression garments or abdom-
inal binders are the most effective. A 2016 study
found that an inflatable automated abdominal
binder which provided 40  mmHg splanchnic
venous compression was non-inferior to mido-
drine in improving orthostatic symptoms. While
promising, it should be noted that sample size
was small [16, 32]. Limitations to compression
garments include poor compliance due to dis-
comfort and difficulty putting on and ulcers in
patients with neuropathy impairing sensation in
areas of compression [15, 16].
7 Pharmacologic Interventions
When nonpharmacological measures fail to
improve symptomatic nOH, pharmacological
intervention should be considered. There are two
complementary strategies used: expanding intra-
vascular volume and increasing peripheral vascu-
lar resistance [16]. Expanding intravascular
volume has been traditionally achieved by using
the mineralocorticoid fludrocortisone and treat-
ment of underlying anemia, if present. Peripheral
vascular resistance can be increased with mido-
drine, droxidopa, pseudoephedrine, or norepi-
nephrine reuptake inhibitors. It is also worth
noting that all patients being treated for nOH
with pharmacologic agents should be wary of
supine hypertension and so should avoid horizon-
tal positions and sleep/rest with the head of bed
elevated 30–45°. See Table  3 for a review of
potential therapies.
7.1 Treatment of Anemia
Decreased blood viscosity and oxygen carrying
capacity caused by anemia can exacerbate nOH
[15]. Iron deficiency anemia leads to increased
nitric oxide, a potent vasodilator and possibly
another contributor to nOH. Therefore, investiga-
L. F. Fanty and C. H. Gibbons
tion and correction of anemia via repletion of
iron, vitamins, or erythropoietin as indicated can
help ameliorate symptomatic nOH [15, 19].
7.2 Fludrocortisone
Although one of the most frequently prescribed
agents for orthostatic hypotension, fludrocorti-
sone is not approved by US Food and Drug
Administration for this indication [15]. As a syn-
thetic mineralocorticoid, fludrocortisone
increases renal sodium and water reabsorption
resulting in intravascular volume expansion and
increased blood pressure [15, 16]. It also works
synergistically with adrenergic agonists to
enhance pressor effect [16]. Starting dose is typi-
cally 0.05–0.2 mg/day with little therapeutic ben-
efit above 0.2–0.3  mg/day [15, 16]. Effects of
fludrocortisone can take up to 3–7 days to appre-
ciate [15, 16]. Common side effects include
supine hypertension, hypokalemia, and ankle
edema [16]. Long term use worsens hypertension
and can result in end-organ damage, especially
left ventricular hypertrophy and renal failure
[16]. This medication should be used cautiously
in patients with congestive heart failure and is
associated with a higher risk of all cause hospital-
ization [15, 16].
7.3 Midodrine
Midodrine has been FDA approved for the treat-
ment of OH since 1996 [16, 33]. Midodrine is a
prodrug that is converted peripherally into its
metabolite desglymidodrine, a selective alpha-1
adrenoceptor agonist [16]. Desglymidodrine
increases vascular resistance and blood pressure
through arteriolar and venous constriction [16].
Midodrine is short acting lasting 1–3 h and typi-
cal dosing is 2.5–15 mg up to three times daily
during waking hours [16, 33]. A major side effect
is supine hypertension, and it is recommended
that midodrine not be taken within 5 h of bedtime
[15, 16]. Other side effects include piloerection,
scalp itching, and urinary retention [16]. Extra
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes 461

Table 3  Treatment of symptomatic orthostatic hypotension and related pathology

Treatment of symptomatic orthostatic hypotension
Nonpharmacologic interventions
• Avoid medications that could worsen or cause orthostatic hypotension
• Avoid increase in core body temperature (warm and humid environments such as saunas, hot tubs, and hot
showers)
• Move with purpose:
 – Dorsiflex the feet or perform hand grip exercise before standing
 – Change positions slowly
 – Perform exercises to promote venous return such as standing on both legs and crossing them, clenching
buttocks
• Remain active
 – Trial water exercises, rowing, recumbent bike
• Elevate head of bed by 10–20°
• Increase hydration
 – 2–3 L of water daily
 – 500 mL boluses can lead to instant improvement in low blood pressure
• Increase daily sodium; 1 teaspoon daily additional to heart healthy diet; can increase Na intake to 6–10 g/daily
• Eat small, carbohydrate low meals
• Avoid alcohol as it can cause vasodilation
• Avoid coffee as diuretic properties can lead to decreased intravascular volume
• Trial abdominal compression garments
Pharmacologic interventions
Medication Mechanism of action Dose
Midodrine α-1 adrenoceptor agonist 2.5–15 mg up to three times a day during day time;
not to take within 5 h of bedtime
Droxidopa Synthetic norepinephrine precursor 100–600 mg up to three times a day during day time;
not to take within 5 h of bedtime
Fludrocortisone Synthetic mineralocorticoid 0.05–0.2 mg daily
Pyridostigmine Acetylcholinesterase inhibitor 30–60 mg up to three times daily
Atomoxetine Norepinephrine reuptake inhibitor 10–18 mg twice daily
Consideration: Treatment of anemia recommended

caution should be utilized when prescribing to
patients with congestive heart failure and chronic
renal failure [15].
7.4 Droxidopa
Droxidopa has been FDA approved for the treat-
ment of OH since 2014 [15, 34]. Droxidopa is a
norepinephrine prodrug that is converted to nor-
epinephrine by l-amino decarboxylase in the
kidneys, the central nervous system, and in
peripheral sympathetic postganglionic nerve end-
ings [15, 16]. The increase in peripheral neural
norepinephrine is believed the primary mecha-
nism of action for increasing blood pressure.
Droxidopa is short acting with a peak response of
3.5–6  h [15, 16]. Starting dose is 100–600  mg
three times daily during waking hours [16, 34].
Common side effects include supine hyperten-
sion, headache, nausea, dizziness, and fatigue.
Extra caution should be utilized when prescrib-
ing to patients with congestive heart failure and
chronic renal failure [15, 16].
It is important to note that patients with a
supine plasma norepinephrine level <220 pg/mL
have a greater blood pressure response to droxi-
dopa treatment [16]. Palma et al. found an inverse
relationship between supine plasma norepineph-
rine and blood pressure increase following droxi-
dopa [2, 16]. This is explained by denervation
sensitivity of postganglionic adrenergic ­receptors,
462
which occurs in around 25% of DAN patients,
resulting in an exaggerated response to adrener-
gic agonists such as droxidopa and in theory,
midodrine [16]. Droxidopa notionally may be
less effective in patients on carbidopa, which
blocks the conversion of droxidopa to norepi-
nephrine [16]. There may not be all that many
patients with both diabetic autonomic neuropathy
and Parkinson’s disease, but in those situations
the use of carbidopa with droxidopa should be
considered with caution.
7.5 Norepinephrine Reuptake
Inhibitors
Norepinephrine reuptake inhibitors, such as ato-
moxetine, increase the availability of norepi-
nephrine at the neurovascular junction. In patients
with intact central autonomic function, the effect
on peripheral sympathetic postganglionic fibers
leading to vasoconstriction is counteracted by the
vasodilatory effect of central α2 receptors [16,
35]. In patients with central autonomic dysfunc-
tion, norepinephrine reuptake inhibitors result in
peripheral vasoconstriction without vasodilation
from central α2 receptors [16]. This makes atom-
oxetine a treatment for nOH from central auto-
nomic failure seen in diseases such as multiple
system atrophy, but atomoxetine is less likely to
be effective for a primarily peripheral process
such as DAN [2, 16].
7.6 Other
Pyridostigmine is used off-label for the treatment
of nOH. Pyridostigmine is a cholinesterase inhib-
itor and enhances cholinergic neurotransmission
at peripheral cholinergic synapses, including
sympathetic ganglia [36]. It has been shown to
increase systolic blood pressure by only 4 mmHg
and is more useful in patients whose symptoms
are less severe and when used as an adjunctive
[36]. Dosing starts at 30–60  mg once to three
times a day [15]. Adverse side effects include
cholinergic symptoms such as abdominal cramp-
ing, diarrhea, sweating, sialorrhea, and urinary
L. F. Fanty and C. H. Gibbons
incontinence [15]. One of the benefits of the use
of pyridostigmine is it does not contribute to
supine hypertension.
8 Special Considerations
8.1 Neurogenic Supine
Hypertension
Most patients with neurogenic OH will experi-
ence supine hypertension which is defined as sys-
tolic BP ≥140  mmHg and/or diastolic BP
≥90 mmHg [37]. All patients with supine hyper-
tension should be advised to avoid horizontal
supine position, including during naps, and ele-
vate head of bed at least 6 in. at night [5, 16]. A
carbohydrate rich snack or alcoholic beverage
right before bedtime can aide in vasodilation of
the splanchnic bed and lower blood pressure [15,
16]. Fludrocortisone should be avoided if possi-
ble, as a main side effect is supine hypertension
[15]. A consensus of expert opinion released in
2017 recommends prioritizing treatment of symp-
tomatic nOH over supine hypertension as symp-
tomatic nOH can be immediately debilitating and
lead to increased falls and potentially death [15].
Short acting antihypertensives (captopril, cloni-
dine, hydralazine, losartan) may be given at night
time if the decision is made to treat supine hyper-
tension. For supine SBP 160–180, the decision to
treat should be individualized. For severe supine
hypertension (SBP > 180, DBP > 110) treatment
should be initiated [17]. Patients started on anti-
hypertensives at night should be warned to take
extra precautions when getting out of bed at night
and use a urinal or bedside commode to decrease
falls [15]. See Table 4 for a review of treatments
for supine hypertension.
8.2 Postprandial Hypotension
The treatment of postprandial hypotension dif-
fers from the treatment of orthostatic hypoten-
sion because of shorter anticipated periods of
time where blood pressure may be low.
Nonpharmacologic measures include eating
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes 463

Table 4  Treatment of supine hypertension Table 5  Treatment of postprandial hypotension

Nonpharmacologic interventions Nonpharmacologic interventions
• Carbohydrate snack or if drinking alcohol, have a • Smaller meals with lower glycemic index to reduce
glass right before bedtime splanchnic pooling
• No medications for orthostatic hypotension within • Drink 12–18 oz of water 15 min before mealtime
5 h of bedtime • Avoid sugary beverages and alcohol
• Short acting antihypertensives prior to bedtime • Monitoring of symptoms and avoidance of strenuous
• Extra precaution when getting out of bed at night if exercise 60 min after a meal
pharmacologic intervention pursued Pharmacologic interventions
Pharmacologic interventions Medication Mechanism of Dose
Medication Mechanism of Dose action
action Acarbose α-glucosidase 50–100 mg
Captopril ACE inhibitor 25 mg at night inhibitor before meals
Clonidine Central α-2 0.2 mg with Midodrine α-1 2.5–15 mg
agonist evening meal adrenoceptor before meals
Hydralazine Relaxation of 10–25 mg at agonist
peripheral night Octreotide Somatostatin 50 μg, 30 min
smooth muscles analogue before each
Losartan Angiotensin II 50 mg at night meal
receptor
antagonist
Nitroglycerin Vasodilator 0.1 mg/h patch
Patch at night to be 9 Sudomotor Dysfunction
removed in the
morning
Sudomotor function refers to the regulation of
sweating through the stimulation of eccrine sweat
small, low carbohydrate meals, throughout the glands. Sudomotor dysfunction can result in an
day and the avoidance of alcohol and sugary increase in sweat production (hyperhidrosis),
drinks [15, 16]. Patients can also drink 12–18 oz decrease in sweat production (hypohydrosis), or
of water after a meal, but before they stand up. lack of sweat production (anhidrosis). Sweating
Blood pressure is usually lowest 30–60 min after abnormalities are common in many types of neu-
a meal, and patients should monitor for symp- ropathy and sudomotor dysfunction is one of the
toms and encouraged to sit or lie down if needed. earliest detectable abnormalities in distal small
If nonpharmacological interventions fail to fiber neuropathy. The sudomotor system serves
improve symptoms, clinicians can consider phar- as the primary mechanism for evaporative heat
macologic intervention [16]. Midodrine, taken loss, and an integral part of the autonomic system
right before or during a meal can help decrease maintaining thermal homeostasis [38, 39].
venous pooling [16]. Insulin is a vasodilator, so
acarbose, a medication which reduces the gastro-
intestinal absorption of glucose, helps decrease 10 The Anatomy
postprandial hypotension by reducing insulin and Pathophysiology
release [16]. Another option is octreotide, a med- of Sudomotor Function
ication that induces vasoconstriction of splanch-
nic vessels and can attenuate postprandial The body’s autonomic system maintains an inter-
hypotension [22]. However Octreotide is admin- nal temperature around 37 °C [38]. In response to
istered subcutaneously, is often cost prohibitive, a rise in body temperature, the autonomic ner-
and can induce nausea and abdominal pain. vous system helps cool body temperature through
Midodrine, taken right before or during a meal sympathetic responses such as generalized sweat-
can help decrease venous pooling [16]. See ing, vasodilation, and hyperpnea [38]. The body
Table 5 for review of treatments for postprandial has cutaneous, visceral, and spinal cord thermo-
hypotension. receptors which initially detect an increase in
464
temperature [38]. Thermal receptors and thermal
nociceptors are typically A-delta and C fibers,
both small, unmyelinated axons [40]. These
fibers enter the spinal cord along Lissauer’s tract
before traveling one to two vertebral levels above
the level of entry to synapse on the gray matter of
the dorsal horn [38]. Axons of second order neu-
rons then decussate and join the spinothalamic
tract to ascend to the reticular formation, thala-
mus, and preoptic areas of the hypothalamus
[38].
The hypothalamus integrates the new thermal
information with references to changes in fluid
volume and electrolyte concentration, resulting
in thermoregulation [39]. If cooling is required,
warm sensing neurons in the medial preoptic area
will activate efferent sympathetic neurons. These
neurons descend ipsilaterally to synapse on the
preganglionic neurons in the intermediolateral
cell column of the spinal cord [38]. The pregan-
glionic neurons exit through the ventral root and
white rami T1-L2 and synapse on the paraverte-
bral sympathetic chain ganglia [38].
Postganglionic sudomotor and vasomotor fibers
then travel to the dermis and synapse on sweat
glands and cutaneous arterioles to regulate evap-
orative and radiant heat loss, respectively [38].
The greatest density of thermoregulatory eccrine
glands is on the forehead, then the upper limbs,
the trunk, and the lower limbs. Sweat gland den-
sity is high in the palms and soles of the feet, but
these glands mediate emotional sweating as well
as thermoregulation [38, 39]. Eccrine sweat
glands are primarily innervated by unmyelinated
C fibers synapsing on cholinergic M3-type recep-
tors. Sympathetic sudomotor neurons can also
co-express markers of noradrenergic neurons, but
parasympathetic response on sudomotor function
is thought to be negligible [38, 39].
11 Clinical Presentation
Patients with diabetic autonomic neuropathy
often present with concerns of increased sweat-
ing on their face and chest attributable to focal
compensatory hyperhidrosis [41]. This form of
hyperhidrosis is a compensatory mechanism and
L. F. Fanty and C. H. Gibbons
occurs adjacent or distal to areas of decreased
sweating. In diabetic autonomic neuropathy, this
leads to increased sweating adjacent to the palms
and soles of the feet as well as more centrally
over the chest and forehead [41]. Focal compen-
satory hyperhidrosis is different from primary
focal hyperhidrosis. Primary focal hyperhidrosis
affects the palms, soles, and axilla and often
occurs in response to an emotional trigger. It is
independent of thermoregulatory needs [38].
There have been a few cases of diabetic neu-
ropathy associated with gustatory sweating,
another focal hyperhidrosis which presents as
unilateral facial sweating in response to spicy
foods [42–44]. The proposed mechanism is aber-
rant rewiring of facial parasympathetic salivary
glands to denervated sympathetic facial sweat
glands [38]. Diabetic neuropathy can also be
associated with nocturnal generalized diaphore-
sis, although this clinical presentation is less
common [38, 41]. It is important to note that
hyperhidrosis typically affects quality of life
rather than necessitating urgent medical atten-
tion. Rarely, severe generalized hyperhidrosis
can lead to hypothermia and hypovolemia, while
generalized anhidrosis can result in
hyperthermia.
As mentioned above, focal compensatory
hyperhidrosis results from underlying hypohi-
drosis or anhidrosis that is typically in a length-­
dependent fashion [41]. In diabetic autonomic
neuropathy, the most frequent pattern of sweat
loss is in the traditional stocking and glove distri-
bution, conforming to the pattern of length
dependency [38]. Thoracic nerve roots may also
be involved resulting in asymmetric sweat loss in
affected areas. Progressive loss of sudomotor
function can result in global anhidrosis with
resultant thermoregulatory impairment and
hyperthermia. At this stage, a patient normally
has profound generalized autonomic neuropathy
[6]. On exam, skin changes often occur due to
denervation of sweat glands, which results in tro-
phic skin changes and hair loss. Peripheral auto-
nomic dysfunction may also lead to observable
physical exam findings such as edema, rough
patches of skin, loss of nails, and poor wound
healing which can be contributing factors in the
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes
development of foot ulcers. One study found the
odds ratio of foot ulcers increased to 15.3 if the
patient lacked sympathetic skin response [45].
12 Diagnostics
Quantitative analysis of sudomotor dysfunction
is an important component of autonomic testing
and can help confirm, monitor progression, and
identify treatment success of autonomic dysfunc-
tion. Assessment of sudomotor dysfunction by
eliciting direct or indirect sweat response remains
one of the most sensitive and specific means to
detect small fiber neuropathy. Direct sweat
response can be obtained by using iontophoresis
of cholinergic agonists to stimulate M3 musca-
rinic receptors [39]. An indirect axon reflex sweat
response can also be observed with direct stimu-
lation by observation in adjacent regions. When
cholinergic agonists bind the nicotine receptors
on sudomotor nerves, an impulse travels anti-
dromically up the postganglionic sudomotor
axon. When it reaches a branch point, the impulse
then travels orthodromically until it reaches a dif-
ferent population of sweat glands [38, 39, 46].
Tests of sudomotor function include quantita-
tive sudomotor axon reflex testing (QSART),
quantitative direct and indirect axon reflex testing
(QDIRT), silicone impressions, acetylcholine
sweat-spot, thermoregulatory sweat testing
(TST), and sympathetic skin response (SSR).
The choice of test to perform often is dictated by
resources and lesion localization [39]. It is impor-
tant to note that all tests of sudomotor function
are susceptible to artifact and confounding vari-
ables such as ambient temperature, hydration,
medication use, age, and gender [39].
12.1 Quantitative Sudomotor Axon
Reflex Test (QSART)
The most widely utilized test of autonomic dys-
function is QSART, which is thought to be a sen-
sitive and reproducible test of sudomotor function
[39, 46]. QSART tests the integrity of the post-
ganglionic sympathetic neuron by measuring the
465
axon reflex mediated sweat response. Typical
testing sites include forearm proximal and distal
leg and dorsum of foot [39]. A multi-­
compartmental capsule is placed on the skin. The
outer ring is typically filled with 10% acetylcho-
line and the inner well has nitrogen gas with a
hygrometer on the outflow tract. Iontophoresis of
acetylcholine is started for 5 min and humidity is
recorded from baseline to 15  min poststimula-
tion. Results are analyzed by assessing change in
relative humidity and sweat latency [39, 46].
Although it is the most utilized test, QSART
requires specialized equipment and is expensive
[39]. In patients with diabetes, there is a length-­
dependent reduction to sweat volume beginning
at the foot and progressing to more proximal sites
with disease progression [20].
12.2 Quantitative Direct
and Indirect Axon Reflex
Testing (QDIRT)
QDIRT measures the postganglionic sudomotor
function by quantifying direct and axon reflex
mediated sweat response. Sweat glands are stim-
ulated by acetylcholine iontophoresis. Sweat
glands are visualized by indicator dye and a digi-
tal image is taken every 15 s for 7 min. The sweat
droplets are then quantified by number, size,
location, and latency. As it is a relatively new
technique, a limitation of this technique is pau-
city of data on normative values [47]. In patients
with diabetes, there is typically a reduction in
sweat droplet number and sweat area in the set-
ting of neuropathy [47].
12.3 Silicone Impressions
Silicone impressions measure the postganglionic
sudomotor function by quantifying direct and
axon reflex mediated sweat response [38]. During
this test, sweat glands are stimulated by choliner-
gic iontophoresis [48]. A thin layer of moldable
silicone material is then placed on the skin. Sweat
droplets displace the silicone material during
polymerization resulting in impressions that can
466
be quantified [39]. Data is reported as droplet
number, size, distribution, and volume (which
can be estimated by assuming droplets form a
hemisphere) [38]. When interpreting results, it is
important to know that this testing is prone to
artifact as abnormal impressions can be caused
by hair, dirt, and air bubbles and the reader may
not be able to differentiate between actual and
artifactual impressions [39]. While it is one of the
easier tests to conduct, changes in moldable
material (with more rapid polymerization) have
led to reduced ability for droplets to leave inden-
tations [39]. Silicone impressions have not been
systematically evaluated in patients with diabetic
neuropathy.
12.4 Acetylcholine Sweat-Spot
This test is used to evaluate postganglionic sudo-
motor function by measuring axon reflex medi-
ated sweat response [39]. To perform this test, the
skin is first painted with an indicator dye. An
intradermal injection of acetylcholine is then
used to stimulate sweat glands. Sweat spots can
be visualized from the indicator dye and a picture
is taken 6 min after injection and quantified on a
grid [39]. Limitations to this method are that it
represents a single point in time and that some
patients report the injection to be painful [39].
The acetylcholine sweat-spot test also has not
been studied in detail in patients with diabetic
neuropathy.
12.5 Thermoregulatory Sweat
Testing (TST)
TST interrogates the sympathetic sudomotor
pathway from the central nervous system to the
cutaneous sweat gland. During this test, the
patient is placed in a temperature and humidity
controlled chamber with ceiling mounted infra-
red heaters controlled by probes measuring the
patient’s temperature on the head, trunk, and
proximal leg. There is also an oral cutaneous
probe [38, 49]. The patient is covered in an indi-
cator dye which changes color in response to the
L. F. Fanty and C. H. Gibbons
lower pH of sweat [38]. Heat and humidity are
increased so the patient’s oral temperature
increases by 1.0 °C above their baseline tempera-
ture. Digital photographs help generate a sweat
density map [38]. Results are expressed as a
TST% (measured area of anterior body anhidro-
sis divided by area of anatomic figure, multiplied
by 100) [39]. The patterns of sweating are easily
noted and TST% can provide a severity index,
but also track severity of disease. It also allows
evaluation of areas often not assessed by other
methods such as the trunk, head, fingers, and toes
[39]. However, TST cannot differentiate between
preganglionic and postganglionic lesions [50].
For this reason TST is often combined with tests
which evaluate for postganglionic lesions such as
QSART, QDIRT, and silicone impressions. TST
is time-consuming, requires specialized equip-
ment, and is often poorly reimbursed. In patients
with diabetes, TST results often provide a graphi-
cal depiction of a distribution of neuropathy, with
reduced sweat production distally [38, 49].
12.6 Sympathetic Skin Response
(SSR)
SSR, also known as galvanic skin response, pro-
vides a surrogate measure of sympathetic cholin-
ergic sudomotor function through the measure of
electrodermal activity. Recording electrodes are
placed on the dorsal and ventral hand and foot
[51, 52]. Changes in skin potential are triggered
by inspiratory gasp, forceful expiration, startle
response, or electrical stimulation and recordings
are taken via EMG [49, 50]. SSRs are reported to
be absent or present; amplitude and latency are
also observed [49, 50]. This method is easy to
perform but is not thought to be a reliable mea-
sure of sudomotor dysfunction. There is high
variability within and between subjects, and
SSRs decline with age and may not be seen in
patients older than 50 years. Further, SSR is only
a surrogate measure of sweat gland function and
patients with congenital absence of sweat glands
will still have an SSR response, thus creating a
clear discrepancy with other measures of true
sudomotor function [49, 50]. The sympathetic
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes
skin response (SSR) evaluates the momentary
change in the electrical potential associated with
induced sweating in the palms and soles. The
response can be evoked by psychological or local
nociceptive stimuli [53]. As the response is emo-
tionally activated rather than thermoregulatory, it
is highly variable and has limited sensitivity and
specificity in the diagnosis of sudomotor nerve
impairment. SSR is a sensitive, but not specific,
measure of sudomotor dysfunction in patients
with diabetes. Abnormalities to both amplitude
and latency have been detected across a range of
diabetic neuropathy [51]. However, SSR ampli-
tude and latency measures do not seem to accu-
rately estimate the severity of a neuropathy and
cannot be used to track longitudinal changes
[54].
13 Treatment
Therapy for sudomotor dysfunction related to
diabetic autonomic neuropathy primarily con-
sists of education and lifestyle modifications.
Although patients’ primary concern may be
proximal compensatory sweating, clinicians
should avoid pharmaceutical treatment as this
can increase the risk of hyperthermia. Patients
should be counseled on lifestyle modifications
such as wearing cooling garments, carrying cool-
ant packs and spray water bottles, and avoidance
of warm environments when possible. Patients
should also partake in careful foot care to prevent
ulceration and infection.
References
1. Eid S, Sas KM, Abcouwer SF, et  al. New insights
into the mechanisms of diabetic complications:
role of lipids and lipid metabolism. Diabetologia.
2019;62:1539–49.
2. Sasaki H, Kawamura N, Dyck PJ, Dyck PJB, Kihara
M, Low PA.  Spectrum of diabetic neuropathies.
Diabetol Int. 2020;11:87–96.
3. Vinik AI, Erbas T, Casellini CM.  Diabetic cardiac
autonomic neuropathy, inflammation and cardiovas-
cular disease. J Diabetes Investig. 2013;4:4–18.
4. Spallone V, Ziegler D, Freeman R, et al. Cardiovascular
autonomic neuropathy in diabetes: clinical impact,
467
assessment, diagnosis, and management. Diabetes
Metab Res Rev. 2011;27:639.
5. Vinik AI, Casellini C, Parson HK, Colberg SR,
Nevoret ML.  Cardiac autonomic neuropathy in dia-
betes: a predictor of cardiometabolic events. Front
Neurosci. 2018;12:591.
6. Freeman R.  Diabetic autonomic neuropathy. Handb
Clin Neurol. 2014;126:63–79.
7. Soedamah-Muthu SS, Chaturvedi N, Witte DR,
Stevens LK, Porta M, Fuller JH.  Relationship
between risk factors and mortality in type 1 diabetic
patients in Europe: the EURODIAB Prospective
Complications Study (PCS). Diabetes Care.
2008;31:1360–6.
8. Ziegler D, Zentai CP, Perz S, et al. Prediction of mor-
tality using measures of cardiac autonomic dysfunc-
tion in the diabetic and nondiabetic population: the
MONICA/KORA Augsburg Cohort Study. Diabetes
Care. 2008;31:556–61.
9. Astrup AS, Tarnow L, Rossing P, Hansen BV, Hilsted
J, Parving HH.  Cardiac autonomic neuropathy pre-
dicts cardiovascular morbidity and mortality in type 1
diabetic patients with diabetic nephropathy. Diabetes
Care. 2006;29:334–9.
10. Kaur D, Tiwana H, Stino A, Sandroni P. Autonomic
neuropathies. Muscle Nerve. 2021;63:10–21.
11. Dimitropoulos G, Tahrani AA, Stevens MJ.  Cardiac
autonomic neuropathy in patients with diabetes mel-
litus. World J Diabetes. 2014;5:17–39.
12. Maser RE, Lenhard MJ.  Cardiovascular autonomic
neuropathy due to diabetes mellitus: clinical mani-
festations, consequences, and treatment. J Clin
Endocrinol Metab. 2005;90:5896–903.
13. Gibbons CH, Freeman R.  Clinical implications of
delayed orthostatic hypotension: a 10-year follow-up
study. Neurology. 2015;85:1362.
14. Freeman R, Wieling W, Axelrod FB, et al. Consensus
statement on the definition of orthostatic hypoten-
sion, neurally mediated syncope and the postural
tachycardia syndrome. Auton Neurosci Basic Clin.
2011;161:46–8.
15. Gibbons CH, Schmidt P, Biaggioni I, et al. The rec-
ommendations of a consensus panel for the screening,
diagnosis, and treatment of neurogenic orthostatic
hypotension and associated supine hypertension. J
Neurol. 2017;264:1567–82.
16. Palma JA, Kaufmann H.  Management of ortho-
static hypotension. Continuum (Minneap Minn).
2020;26:154–77.
17. Freeman R. Clinical practice. Neurogenic orthostatic
hypotension. N Engl J Med. 2008;358:615–24.
18. Low PA, Singer W.  Management of neurogenic
orthostatic hypotension: an update. Lancet Neurol.
2008;7:451–8.
19. Freeman R, Abuzinadah AR, Gibbons C, Jones
P, Miglis MG, Sinn DI.  Orthostatic hypotension:
JACC state-of-the-art review. J Am Coll Cardiol.
2018;72:1294–309.
20. Low PA, Benrud-Larson LM, Sletten DM, et  al.
Autonomic symptoms and diabetic neuropa-
468
thy: a population-­ based study. Diabetes Care.
2004;27:2942–7.
21. Dyck PJ, Kratz KM, Karnes JL, et al. The prevalence
by staged severity of various types of diabetic neurop-
athy, retinopathy, and nephropathy in a population-­
based cohort: the Rochester Diabetic Neuropathy
Study. Neurology. 1993;43:817–24. Published erra-
tum appears in Neurology 1993;43(11):2345.
22. Jansen RW, Lipsitz LA.  Postprandial hypoten-
sion: epidemiology, pathophysiology, and clinical
management. Ann Intern Med. 1995;122:286–95.
Review.
23. Gannon J, Claffey P, Laird E, Newman L, Kenny RA,
Briggs R.  The cross-sectional association between
diabetes and orthostatic hypotension in community-­
dwelling older people. Diabet Med. 2019;37:1299.
24. Freeman R, Illigens BMW, Lapusca R, et al. Symptom
recognition is impaired in patients with orthostatic
hypotension. Hypertension. 2020;75:1325–32.
25. Norcliffe-Kaufmann L, Kaufmann H, Palma JA, et al.
Orthostatic heart rate changes in patients with auto-
nomic failure caused by neurodegenerative synucle-
inopathies. Ann Neurol. 2018;83:522–31.
26. Gibbons CH, Freeman R. Delayed orthostatic hypo-
tension: a frequent cause of orthostatic intolerance.
Neurology. 2006;67:28–32.
27. Cheshire WP, Freeman R, Gibbons CH, et  al.
Electrodiagnostic assessment of the autonomic ner-
vous system: a consensus statement endorsed by the
American Autonomic Society, American Academy
of Neurology, and the International Federation
of Clinical Neurophysiology. Clin Neurophysiol.
2020;132:666.
28. Illigens BMW, Gibbons CH.  Autonomic test-
ing, methods and techniques. Handb Clin Neurol.
2019;160:419–33.
29. Gibbons CH, Freeman R. Delayed orthostatic hypoten-
sion. Auton Neurosci Basic Clin. 2020;229:102724.
30. Andersen ST, Witte DR, Fleischer J, et al. Risk factors
for the presence and progression of cardiovascular
autonomic neuropathy in type 2 diabetes: addition-­
Denmark. Diabetes Care. 2018;41:2586–94.
31. Mathias CJ, Young TM.  Water drinking in the man-
agement of orthostatic intolerance due to orthostatic
hypotension, vasovagal syncope and the postural
tachycardia syndrome. Eur J Neurol. 2004;11:613–9.
32. Okamoto LE, Diedrich A, Baudenbacher FJ, et  al.
Efficacy of servo-controlled splanchnic venous
compression in the treatment of orthostatic hypo-
tension: a randomized comparison with midodrine.
Hypertension. 2016;68:418–26.
33. McClellan KJ, Wiseman LR, Wilde MI.  Midodrine.
A review of its therapeutic use in the management of
orthostatic hypotension. Drugs Aging. 1998;12:76–
86. Review.
34. Kaufmann H, Norcliffe-Kaufmann L, Palma
JA. Droxidopa in neurogenic orthostatic hypotension.
Expert Rev Cardiovasc Ther. 2015;13:875–91.
35. Ramirez CE, Okamoto LE, Arnold AC, et al. Efficacy
of atomoxetine versus midodrine for the treatment
L. F. Fanty and C. H. Gibbons
of orthostatic hypotension in autonomic failure.
Hypertension. 2014;64:1235–40.
36. Singer W, Sandroni P, Opfer-Gehrking TL, et  al.
Pyridostigmine treatment trial in neurogenic ortho-
static hypotension. Arch Neurol. 2006;63:513–8.
37. Fanciulli A, Jordan J, Biaggioni I, et  al. Consensus
statement on the definition of neurogenic supine
hypertension in cardiovascular autonomic failure
by the American Autonomic Society (AAS) and the
European Federation of Autonomic Societies (EFAS):
Endorsed by the European Academy of Neurology
(EAN) and the European Society of Hypertension
(ESH). Clin Auton Res. 2018;28:355–62.
38. Coon EA, Cheshire WP Jr. Sweating disorders.
Continuum (Minneap Minn). 2020;26:116–37.
39. Illigens BM, Gibbons CH. Sweat testing to evaluate
autonomic function. Clin Auton Res. 2009;19:79–87.
40. Mandadi S, Roufogalis BD.  ThermoTRP channels
in nociceptors: taking a lead from capsaicin receptor
TRPV1. Curr Neuropharmacol. 2008;6:21–38.
41. Stewart JD.  Sweating abnormalities and other auto-
nomic disorders in diabetes mellitus. Mayo Clin Proc.
1989;64:712–5.
42. Blair DI, Sagel J, Taylor I. Diabetic gustatory sweat-
ing. South Med J. 2002;95:360–2.
43. Urman JD, Bobrove AM.  Diabetic gustatory sweat-
ing successfully treated with topical glycopyrrolate:
report of a case and review of the literature. Arch
Intern Med. 1999;159:877–8.
44. Shaw JE, Parker R, Hollis S, Gokal R, Boulton
AJ.  Gustatory sweating in diabetes mellitus. Diabet
Med. 1996;13:1033–7.
45. Tentolouris N, Marinou K, Kokotis P, Karanti A,
Diakoumopoulou E, Katsilambros N. Sudomotor dys-
function is associated with foot ulceration in diabetes.
Diabet Med. 2009;26:302–5.
46. Low PA, Caskey PE, Tuck RR, Fealey RD, Dyck
PJ.  Quantitative sudomotor axon reflex test in
normal and neuropathic subjects. Ann Neurol.
1983;14:573–80.
47. Gibbons CH, Illigens BM, Centi J, Freeman
R.  QDIRT: quantitative direct and indirect test of
sudomotor function. Neurology. 2008;70:2299–304.
48. Vilches JJ, Navarro X. New silicones for the evalua-
tion of sudomotor function with the impression mold
technique. Clin Auton Res. 2002;12:20–3.
49. Fealey RD, Low PA, Thomas JE.  Thermoregulatory
sweating abnormalities in diabetes mellitus. Mayo
Clin Proc. 1989;64:617–28.
50. Hilz MJ, Dutsch M. Quantitative studies of autonomic
function. Muscle Nerve. 2006;33:6–20.
51. Braune HJ, Horter C.  Sympathetic skin response in
diabetic neuropathy: a prospective clinical and neu-
rophysiological trial on 100 patients. J Neurol Sci.
1996;138:120–4.
52. Levy DM, Reid G, Rowley DA, Abraham
RR.  Quantitative measures of sympathetic skin
response in diabetes: relation to sudomotor and neu-
rological function. J Neurol Neurosurg Psychiatry.
1992;55:902–8.
Orthostatic Hypotension and Sudomotor Dysfunction in Diabetes 469

53. Shahani BT, Halperin JJ, Boulu P, Cohen
J.  Sympathetic skin response--a method of assess-
ing unmyelinated axon dysfunction in periph-
eral neuropathies. J Neurol Neurosurg Psychiatry.
1984;47:536–42.
54. Vetrugno R, Liguori R, Cortelli P, Montagna
P.  Sympathetic skin response: basic mecha-
nisms and clinical applications. Clin Auton Res.
2003;13:256–70.
 Gastrointestinal Neuropathy
Karen L. Jones, Chinmay S. Marathe, Tongzhi Wu,
Christopher K. Rayner, and Michael Horowitz
1 Introduction
Gastrointestinal neuropathy is now appreciated
as of major importance in patients with type 1 or
type 2 diabetes and to have diverse clinical mani-
festations. Disordered gastric motor function is
the best characterised manifestation, but auto-
nomic neuropathy can affect all regions of the
gastrointestinal tract and may be associated with
heterogeneous symptoms, including heartburn,
nausea and vomiting, postprandial fullness, diar-
rhoea, constipation, and faecal incontinence.
K. L. Jones (*) · C. S. Marathe · T. Wu ·
M. Horowitz
Adelaide Medical School, The University of
Adelaide, Adelaide, Australia
Centre of Research Excellence in Translating
Nutritional Science to Good Health, The University
of Adelaide, Adelaide, Australia
Endocrine and Metabolic Unit, Royal Adelaide
Hospital, Adelaide, Australia
e-mail: karen.jones@adelaide.edu.au; chinmay.
marathe@adelaide.edu.au; tongzhi.wu@adelaide.edu.au;
michael.horowitz@adelaide.edu.au
C. K. Rayner
Adelaide Medical School, The University of
Adelaide, Adelaide, Australia
Centre of Research Excellence in Translating
Nutritional Science to Good Health, The University
of Adelaide, Adelaide, Australia
Department of Gastroenterology and Hepatology,
Royal Adelaide Hospital, Adelaide, Australia
e-mail: chris.rayner@adelaide.edu.au
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_27
Disordered gastric emptying (delayed and accel-
erated) has important implications for the optimi-
sation of glycaemic control in diabetes.
Gastrointestinal autonomic impairment may also
impact the regulation of postprandial blood pres-
sure. The outcomes of current approaches to the
management of gastrointestinal autonomic neu-
ropathy are, unfortunately for the main part, sub-­
optimal, partly because the underlying
pathophysiology is still poorly defined.
This chapter addresses the epidemiology,
pathophysiology, and management of gastroin-
testinal autonomic neuropathy, including the
impact of gastrointestinal function on glycaemic
control in type 1 and type 2 diabetes.
2 Assessment
of Gastrointestinal
Symptoms and Autonomic
Function in Diabetes
While gastrointestinal symptoms in diabetes are
frequently classified in an organ-specific manner,
i.e. ‘oesophageal’ (heartburn, dysphagia). ‘gas-
tric’ (nausea, vomiting, postprandial fullness),
and ‘intestinal’ (constipation, diarrhoea, faecal
incontinence), there is, in many cases, substantial
overlap so that the attribution of a particular
symptom to a specific region of the gastrointesti-
nal tract often lacks validity. Moreover, gastroin-
testinal symptoms are also prevalent in the
471
472
general population (particularly women and indi-
viduals who are obese), may not be volunteered
(particularly when regarded as embarrassing),
there is substantial symptom ‘turnover’ (i.e. with
time, symptoms are both ‘gained’ and ‘lost’) and
symptoms are associated with both psychological
dysfunction and depression [1].
An important issue is that symptoms should, if
possible, not be evaluated using the so-called
patient self-report, as this is known to be
­unreliable [1]; rather, a validated symptom ques-
tionnaire should be used, as is now mandated by
regulatory bodies for clinical trials of functional
gastrointestinal disorders, e.g. irritable bowel
syndrome and functional dyspepsia. A major lim-
itation in the majority of studies, which compro-
mises their interpretation, has been the use of
‘self-report’ to evaluate symptoms, despite the
availability of these validated instruments, some
specific to diabetes [1–3], including the Diabetes
Bowel Symptom Questionnaire [3] and a diary,
the Gastroparesis Cardinal Symptom Index
(GSCI-DD) [4]. This issue applies, in particular,
to glucose-lowering drugs that are clearly associ-
ated with adverse gastrointestinal effects, most
recently, glucagon-like peptide-1 (GLP-1) recep-
tor agonists (RAs) [5]. The use of self-report also
increases the potential for ‘precebo’ (expectation
of benefit) and ‘nocebo’ (expectation of harm)
effects.
Despite this limitation, it is clear that gastroin-
testinal symptoms occur more frequently in both
type 1 and type 2 diabetes, affect quality of life
adversely, ‘turnover’, and are associated with
depression [1, 2], the use of specific glucose-­
lowering medications [1], and poor, chronic gly-
caemic control [1].
It should also be appreciated that the frequent
attribution of gastrointestinal symptoms to auto-
nomic dysfunction in diabetes is essentially one
of exclusion, because of the lack of diagnostic
tests that can be used to specifically evaluate gas-
trointestinal autonomic function in humans.
Furthermore, these tests are all indirect, e.g. the
plasma pancreatic polypeptide response to sham
feeding [1, 2] and the postprandial increase in
superior mesenteric blood flow as measured by
Doppler ultrasound [1, 2]. Non-invasive, vali-
K. L. Jones et al.
dated, cardiovascular reflex tests are used widely
as a surrogate but, by definition, are primarily an
index of cardiovascular, rather than gastrointesti-
nal, function [6].
3 Oesophagus
While the oesophagus has received much less
attention than the stomach, it is clear that both
disordered oesophageal motility and ‘oesopha-
geal symptoms’, e.g. dysphagia and gastroesoph-
ageal reflux, occur frequently, particularly in
patients with evidence of autonomic neuropathy.
Gastroesophageal reflux is also, not infrequently,
asymptomatic, even when associated with ero-
sive disease [7].
Oesophageal transit is delayed in ~50%
patients with longstanding diabetes [8, 9] and
associated weakly with symptoms of reflux and
dysphagia. The prevalence of oesophageal symp-
toms in patients with diabetes is highly dependent
on the population studied. A higher prevalence of
symptoms of reflux has been observed in tertiary
referral centres (~40%), than in the community
(~15%). Symptoms of dysphagia occur in ~25%
of patients with diabetes seen in tertiary referral
centres, while community-­based studies report a
prevalence of 5% [10]. Whether diabetes is asso-
ciated with an increased risk of Barrett’s oesopha-
gus remains contentious [11].
3.1 Diagnosis
Endoscopy is required to identify mucosal dis-
ease. As in non-diabetic patients, oesophageal
motility can be assessed by conventional or high-­
resolution manometry [12], to evaluate lumen-­
occlusive pressure waves, and contrast video
swallow radiology demonstrates both structural
and functional disorders. Scintigraphy is less
invasive and can be used to measure oesophageal
transit of swallowed radiolabelled food [8], but
the technique lacks standardisation and is not
routinely employed clinically. Oesophageal pH
and impedance measurements may be useful in
the assessment of gastroesophageal reflux.
Gastrointestinal Neuropathy
3.2 Pathophysiology
The pathogenesis of oesophageal dysfunction is
likely to be similar to that of gastroparesis (to be
discussed), with evidence of vagal dysfunction
[13], although the relationship of oesophageal
transit with the rate of gastric emptying is rela-
tively weak [6]. Oesophageal motility has also
been shown to be affected by acute elevations in
blood glucose [13, 14]. Reported motor abnor-
malities include reductions in the number and
amplitude of peristaltic waves [15] and increased
stiffness of the oesophageal body [16]. There is
also evidence that changes in central processing
may contribute to symptoms [17].
3.3 Treatment
The management of disordered oesophageal
function in diabetes is for the main part empiri-
cal. Gastroesophageal reflux is usually treated
with proton pump inhibitors [18] and, in some
cases, elevating the head-end of the bed by
30–45° may prove helpful [19]. Patients with
delayed oesophageal transit, particularly the
elderly, are at risk for ‘pill-induced oesophagitis’
and should be advised to always drink water
immediately after taking oral medications. The
use of pharmacological options other than proton
pump inhibitors, including prokinetic drugs,
lacks an evidence base. Patients with diabetes,
particularly when poorly controlled, have an
increased propensity for oesophageal candidia-
sis, which may present as odynophagia, or be
asymptomatic.
4 Stomach
Disordered gastric emptying is the most widely
recognised manifestation of gastrointestinal
autonomic neuropathy in diabetes. Kassander, in
1958 [20], reported markedly increased, and also
asymptomatic, intragastric retention of barium
in six insulin-treated patients (5 with type 1 dia-
betes) and coined the descriptive term ‘gastropa-
resis diabeticorum’ [20]. His report was prescient
473
in refuting the dogma that symptoms were the
outcome of delay in gastric emptying. Earlier
Rundles (in 1945) [21] suggested that gastropa-
resis represented a manifestation of autonomic
neuropathy. Diabetes is now recognised as the
major cause of chronic gastroparesis; abnor-
mally delayed, and more rapid, gastric emptying
both occur frequently in type 1 and type 2 diabe-
tes [6, 22]. While it has been suggested that a
diagnosis of gastroparesis should be dependent
on the concomitant presence of upper gastroin-
testinal symptoms, such as nausea and vomiting,
as well as delayed gastric emptying. We consider
this inappropriate, particularly given that (as will
be discussed) not infrequently, patients with
markedly delayed gastric emptying have few, or
no, symptoms, as Kassander initially observed
[20] and delayed emptying, even when asymp-
tomatic, has substantial clinical implications,
particularly in relation to the management of
glycaemia.
In health, gastric emptying of solid and liquid
nutrients exhibits a wide inter-individual, but
much lower intra-individual, variation, approxi-
mating an overall linear rate (in the case of solids,
following an initial lag phase), ranging between 1
and 4 kcal/min [23]. Initial studies performed in
the 1980s, which capitalised on the development
of scintigraphic techniques to measure gastric
emptying around that time, demonstrated that
gastric emptying is delayed in 30–50% of indi-
viduals with longstanding (and then, usually
poorly controlled) type 1 or type 2 diabetes [6].
Accordingly, gastroparesis became a common,
rather than a, hitherto assumed, relatively rare,
disorder in diabetes [24]. A recent analysis of
DCCT-EDIC data, where gastric emptying was
measured using a stable isotope breath test,
reported a comparable prevalence of gastropare-
sis of 47% [25]. These studies also established
that, in cases of gastroparesis, the magnitude of
the delay in gastric emptying, while significant,
is often modest, the relationship of rates of gas-
tric emptying of solids and liquids is relatively
weak, gastric emptying is occasionally acceler-
ated, gastric emptying is reproducible in the
longer-­term (up to 25  years) and even marked
delay in emptying may not be associated with a
474
poor prognosis [26]. Healthcare costs and hospi-
talisation related to diabetic gastroparesis have
continued to increase in the USA [27], but that
this may, at least in part, reflect increased aware-
ness. In contrast to longstanding, complicated
diabetes, it has been appreciated more recently
that in both well-controlled type 2 diabetes [22]
and adolescents with type 1 diabetes [28], gastric
emptying is usually normal or mildly accelerated.
It has been suggested, probably correctly, that the
latter may predispose to the development of type
2 diabetes [22, 29], a concept supported by the
demonstration that in some ethnic populations
predisposed to diabetes, gastric emptying is faster
than in Caucasians [30, 31].
The relationship of upper gastrointestinal
symptoms, such as nausea, vomiting, and post-
prandial fullness, with the rate of gastric empty-
ing and the intragastric distribution of a meal [32]
are statistically significant but, in general, weak
[6]. It has recently been reported that symptoms
occurring during measurement of gastric empty-
ing, but not the rate of gastric emptying per se,
are predictive of daily symptoms [33]. Of the
symptoms, early satiety and fullness appear to be
most strongly associated with delayed gastric
emptying [34]. There is evidence that vomiting is
more prominent and abdominal pain less in
patients with diabetic gastroparesis than is the
case in patients with idiopathic gastroparesis
[35]. Accordingly, it is clear that delay in gastric
emptying is best regarded as a marker of gastro-
duodenal motor/sensory abnormality, rather than
as a direct cause of symptoms, as was previously
believed.
Despite a high prevalence of obesity, particu-
larly in type 2, but increasingly also in type 1,
diabetes, malnutrition occurs frequently in
patients with gastroparesis. Gastroparesis may
also impact on the absorption of oral medica-
tions, which is of particular relevance to drugs
which have a short half-life, or where a rapid
onset of action is needed, as is the case with some
oral hypoglycaemic agents [36]. In insulin-
treated patients (i.e. type 1 or insulin-treated type
2), delayed gastric emptying has the potential to
affect glycaemic control adversely by predispos-
ing to a mismatch between the delivery of nutri-
K. L. Jones et al.
ents to the small intestine and the action of
exogenous insulin. Kassander observed in 1958
that ‘the retention of stomach contents in a dia-
betic obviously may cause confusion as far as
food intake and utilisation are concerned’ [20].
Gastric emptying, even variations within the nor-
mal range, may affect both the timing and dosage
of insulin. As early as 1994 [37], it was reported
that the insulin requirement to maintain eugly-
caemia in type 1 patients with gastroparesis com-
pared to those with normal emptying was less
between 0 and 120 min and greater between 180
and 240 min after a meal [37]. In insulin-treated
patients, unexplained episodes of hypoglycae-
mia, particularly postprandially, may be the sole
feature of gastroparesis [38], the so-called gastric
hypoglycaemia [39].
4.1 Diagnosis
Because the prevalence of upper gastrointestinal
symptoms is not strongly predictive of gastropa-
resis, and gastroparesis may be asymptomatic,
the diagnosis is dependent on measurement of
gastric emptying [24]. Gastroduodenal pathol-
ogy, particularly gastric outlet obstruction, must
always be excluded by upper gastrointestinal
endoscopy. A number of methods are available to
measure gastric emptying (Table 1). Irrespective
of the choice, attempts should be made to cease
medications that may slow gastric emptying (e.g.
opioids [40] and both short- [41] and long-acting
[42] GLP-1 agonists) and maintain blood glucose
in the range of 4–10 mmol/L during the measure-
ment [43]. Metformin appears to slow gastric
emptying, but only modestly [44]. Scintigraphy,
Table 1  Methods to assess gastric emptying
• Scintigraphy
• Stable isotope breath tests
• Ultrasound
• SPECT
• MRI
• Absorption kinetics of oral drugs, e.g. paracetamol
(acetaminophen)
• Radiology (barium studies/radiopaque markers)
• Smart Pill™
Gastrointestinal Neuropathy
the use of radiolabelled meals and a gamma cam-
era, remains the ‘gold standard’ technique and
allows measurement of emptying of both solid
and liquid meal components, potentially concur-
rently [45] as well as (in research studies) intra-
gastric meal distribution. There have been recent
steps towards standardisation of methodology for
scintigraphic measurement, internationally [43].
The major limitations of scintigraphy are the
necessity for expensive equipment and the asso-
ciated radiation burden. Stable isotope breath
tests are now well validated and used increas-
ingly for measurement of gastric emptying [46,
47] with the advantages of simplicity and avoid-
ance of radiation exposure [46]. The methodol-
ogy used to evaluate breath test data remains
inconsistent and, in general, gastric emptying
times are best regarded as notional, rather than
precise, although correction using the so-called
Wagner-Nelson method may generate data com-
parable to that of scintigraphy [48]. A further
limitation is that the effects of gastric emptying
and small intestinal absorption of the label can-
not be discriminated. The wireless motility cap-
sule, or Smart Pill™, can evaluate gastric
emptying [49], but because it is a non-digestible
solid, it is only emptied from the stomach after
emptying of digestible solids and nutrient liquids
and is also insensitive in the identification of
rapid emptying [50]. The plasma absorption
Fig. 1 Relationship
between solid gastric
emptying (intragastric
Solid retention at 100 min (%)
retention at 100 min),
measured with
scintigraphy, and
autonomic function
assessed by
cardiovascular reflexes
in unselected patients
with longstanding type 1
and type 2 diabetes
(n = 87). The
relationship is
significant but weak [6]
475
kinetics of oral paracetamol absorption have been
used widely, particularly by the pharmaceutical
industry, as a simple surrogate marker of gastric
emptying, but unfortunately, this technique has
major limitations [51, 52] and, from our perspec-
tive, should not be used for clinical (and ideally
not for research) purposes.
4.2 Pathophysiology
While the pathophysiology of diabetic gastropa-
resis is complex, there have been substantial
insights in this area in the last ~10 to 15 years,
particularly reflecting outcomes of the work of
the Gastroparesis Clinical Research Consortium
in the USA [53]. Earlier studies demonstrating an
association of delayed gastric emptying with an
impaired pancreatic polypeptide response to
sham feeding [54] supported the concept that
gastroparesis is a sequela of autonomic neuropa-
thy. However, the relationship of gastric empty-
ing, measured with scintigraphy, with
cardiovascular autonomic function was shown to
be relatively weak [6] (Fig.  1). In both patients
with diabetes and rodent models, reductions in
cells in the vagal motor and sensory sympathetic
ganglia as well as structural abnormalities of
vagal nerve fibres, both intrinsic and extrinsic to
the gut wall, have been reported [55]. In patients
120
r = 0.35, p < 0.01
100
80
60
40
20
0
0 1 2 3 4 5 6
Score for autonomic nerve function
476 K. L. Jones et al.

with gastroparesis, a spectrum of abnormalities
in gastroduodenal motor/sensory function have
been demonstrated, including impaired accom-
modation of the proximal stomach (i.e. the relax-
ation of the stomach in response to meal
ingestion), antral hypomotility, increased con-
tractile activity localised to the pylorus, abnor-
malities of the gastric electrical rhythm, and
‘hypersensitivity’ to both small intestinal nutri-
ents and gastric distension [36, 56, 57]. The full-­
thickness gastric biopsies obtained by the
Gastroparesis Clinical Research Consortium,
often at the time of gastric pacemaker insertion
(to be discussed), have shown abnormalities in
>80% of cases of severe gastroparesis resistant to
conventional therapy, which are, not surprisingly,
heterogeneous. The dominant abnormalities, in
addition to changes in vagal innervation, are loss
or damage to the Interstitial Cells of Cajal (ICC),
which are integral to the function of the gastric
‘pacemaker’ [58], an immune infiltrate in the
myenteric plexus, a reduction in intrinsic nerves
and loss of inhibitory neurons expressing nitric
oxide synthase (nNOS) (Fig. 2). The basal lamina
around smooth muscle cells and nerves has been
noted to be characteristically thickened [58]. The
cause of the loss of ICCs, which is now regarded
as the ‘dominant’ abnormality in diabetic gastro-
paresis, remains uncertain but may be secondary
to changes in immune function with a shift from
protective M2 to, activated, M1 macrophages,
reductions in insulin or IGF-1 signalling and the
production of stem cell factor [54], and impaired
regulation of heme-oxygenase-1, resulting in
increased oxidative stress [60, 61]. The expres-
sion of the Ano-1 gene, expressed in the ICCs
and which encodes a calcium-activated chloride
channel, may also be abnormal [62]. Abnormal
duodenal mucosal mitochondrial gene expres-
sion has recently been reported to be associated
with a delay in gastric emptying [63], suggesting
that mitochondrial function may be important.
While it should be appreciated that the studies of
the Gastroparesis Clinical Research Consortium
relate primarily to patients with severe gastropa-
resis, the heterogeneous nature of the abnormali-
ties demonstrated has substantial implications for
effective management.

a b

Fig. 2  Haematoxylin and eosin staining of the inter-­ myenteric plexus (original magnification 340) in a patient
myenteric plexus from patients with gastroparesis. (a) with severe diabetic gastroparesis [59]. (Figure courtesy
Normal ganglia and nerve fibres (original magnification of Prof H Parkman)
340). (b) Moderate lymphocytic infiltrate in the inter-­
Gastrointestinal Neuropathy 477

4.3 Effects of the Glycaemic but significant, effect to slow gastric emptying of

Environment on Gastric both solids and liquids in longstanding type 1
Emptying diabetes, measured using scintigraphy in 1990
[65] and, subsequently, elevations in blood glu-
Acute changes in the blood glucose concentra- cose within the normal postprandial range were
tion represent a reversible determinant of the rate shown to delay gastric emptying modestly in
of gastric emptying in diabetes [64–66], so that uncomplicated type 1 diabetes [64]. Acute eleva-
gastric emptying is slowed by hyperglycaemia tions in blood glucose were also shown to attenu-
and accelerated by insulin-induced hypoglycae- ate the effects of drugs which accelerate [70],
mia [24]. Accordingly, the relationship of gastric potentiate the effects of drugs that delay, gastric
emptying with glycaemia is bidirectional (Fig. 3). emptying [71], and affect both gastro-­
Guidelines for measurement of gastric emptying pyloroduodenal motility [72] and gastric electri-
state that blood glucose concentrations should be cal activity, in health. Prostaglandins [73] and
<15 mmol/L at the time of the test while avoiding nitric oxide (NO) [74] have been suggested as
hypoglycaemia [43]. potential mediators of these effects. These stud-
Acute hyperglycaemia (16–20  mmol/L) was ies employed glucose insulin ‘clamps’ to induce
first reported to slow gastric emptying in 1962 hyperglycaemia rapidly. Aigner et  al. [75]
[68], although Stunkard and Wolff had observed recently evaluated the effects of spontaneous
in 1956 that intravenous glucose, presumably fluctuations in fasting glucose on gastric empty-
leading to hyperglycaemia, abolished the so-­ ing in type 1 diabetes and failed to observe a rela-
called gastric hunger contractions [69]. tionship, although the elevations in blood glucose
Hyperglycaemia was shown to have a variable, that they observed were, in most cases, modest.

More rapid gastric emptying increases

postprandial glycaemia

Gastric emptying Postprandial blood

glucose

Acute hyperglycaemia slows gastric

emptying, while hypoglycaemia
accelerates it − +

GLP-1 slows
More rapid gastric emptying Beta cell Alpha cell
gastric emptying (insulin) (glucagon)
increases GLP-1
and GIP secretion

GLP-1, GIP increase

insulin
GLP-1, GIP
GLP-1 suppresses
glucagon

Fig. 3  Schema of the inter-dependent relationships between gastric emptying, postprandial glycaemia, and incretin
hormones [67]
478
Accordingly, further studies, to clarify the effect
of hyperglycaemia on gastric emptying, are
required, including evaluation of the effect of
treatment-induced, improved glycemic control
where current information is limited. There is
uncontrolled evidence that successful pancreas-­
kidney transplantation may reverse gastroparesis
[76]; there is no information about the effect of
islet cell transplantation.
In contrast to the effect of hyperglycaemia,
acute hypoglycaemia (blood glucose ~2.0 to
2.6 mmol/L) markedly accelerates gastric empty-
ing of solids and liquids in healthy subjects [77]
and both complicated [77] and longstanding [78],
type 1 diabetes, the latter including patients with
cardiovascular autonomic dysfunction and, in
some cases, gastroparesis (Fig.  4). The magni-
tude of this acceleration has been shown to be
dependent on the degree of hypoglycaemia [79],
as well as vagal function [80] and, unlike the neu-
roendocrine responses to hypoglycaemia (e.g.
stimulation of catecholamines), not attenuated by
a single episode of antecedent hypoglycaemia
[81]. There is no information about the impact of
insulin-induced hypoglycaemia on gastric emp-
tying in a large number of individuals with
insulin-­treated diabetes in whom the awareness
of hypoglycaemia is impaired. The more rapid
intestinal carbohydrate absorption resulting from
Solid
(100g minced beef)
100
80 p <0.01
Retention 60
(%)
40
20
0 0
0 15 30 45 60 75 90 105 120
Time (min)
Fig. 4  Gastric emptying of solid (minced beef) and liquid
(water) components of a meal, measured with scintigra-
phy, during hypoglycaemia (blood glucose ~2.6 mmol/L
K. L. Jones et al.
the robust acceleration of gastric emptying by
insulin-induced hypoglycaemia is likely to repre-
sent a hitherto, poorly appreciated, but important
counter-regulatory mechanism.
4.4 Treatment
The insights summarised in the preceding section
attest to the implicit importance of definition of
the pathophysiology of a disease in order to
develop approaches to effective management.
Given that many of the significant advances are
recent, it is perhaps not surprising that, despite its
substantial health burden, few treatment options
for diabetic gastroparesis are available and their
efficacy is, in general, limited. The initial dogma
that making the stomach pump/empty ‘better’
would result in symptom improvement, albeit
intuitively illogical, has compromised progress
[82]. Unfortunately, much of the evidence base is
also derived from poorly designed studies and
there is a lack of high quality randomised, con-
trolled trials (Table 2).
Management of gastroparesis, whether
directed at symptom relief, improvement in gly-
caemic control and/or nutritional status should,
like all therapies, ideally be individualised.
General dietary recommendations, consumption
Liquid
(150ml water)
100 Hypoglycaemia
Euglycaemia
80
60 p <0.05
40
20
0 15 30 45 60 75 90 105 120
Time (min)
from t  =  15–45  min) and euglycaemia in longstanding
type 1 diabetes (n = 20). Hypoglycaemia accelerates gas-
tric emptying [78]
Gastrointestinal Neuropathy 479

Table 2  Management of symptomatic diabetic prokinetic, was withdrawn from the market in
gastroparesis
2000 because of the potential for cardiac
• Optimise glycemic control arrhythmias.
  – In insulin-treated patients, address the timing of The dopamine (D2) antagonist, metoclo-
delivery of exogenous insulin to that of nutrient
absorption; consider continuous subcutaneous insulin pramide, can be administered subcutaneously,
delivery with concomitant continuous subcutaneous intranasally, or orally (subcutaneous metoclo-
glucose monitoring (with high- and low-glucose pramide has been used to manage attacks of vom-
alarms) iting on an outpatient basis). The intranasal route
• Address nutritional status
may be effective in women, but not men [87].
  – Modify the size, composition, and frequency of
meals The use of metoclopramide is associated with a
  – Consider jejunal feeding risk of QTc prolongation and tardive dyskinesia
• Address delayed gastric emptying with a trial of [88]. Because of the latter, in many countries,
prokinetic therapy including the USA, metoclopramide can cur-
• Address the dominant symptom(s) (e.g. nausea) with rently only be used for short-term treatment (e.g.
‘symptom-specific’ therapy
12  weeks). Domperidone is also a D2 receptor
• Address psychological co-morbidities
antagonist, but does not cross the blood–brain
barrier readily and is, accordingly, less prone to
of frequent, small meals and avoidance of food central nervous system adverse effects than meto-
that may provoke symptoms (particularly those clopramide [89]. Domperidone may also affect
high in fibre or fat), have, for the main part, not the QT interval and is not readily available in the
been evaluated rigorously [83, 84]. Moreover, USA.
there is anecdotal evidence that outcomes are The macrolide antibiotics, erythromycin and
only rarely satisfactory. In malnourished patients azithromycin, are motilin receptor agonists, fre-
in whom symptoms are refractory, a trial of naso-­ quently used ‘off-label’ for their prokinetic
jejunal feeding should certainly be considered. effects. Their long-term efficacy has not been
Modestly accelerated gastric emptying, as often established, tachyphylaxis to the acceleration of
observed in type 2 diabetes, does not appear to be emptying probably occurs, and there is also a risk
associated with symptoms. As discussed, drugs of QTc prolongation, particularly as the result of
which may slow gastric emptying should, ideally, drug interactions [90]. Mitemcinal, a motilin
be ceased, but this may not prove feasible. agonist, but without antibiotic properties, was
Optimising glycaemic control is logical and sup- shown to accelerate gastric emptying over
ported by uncontrolled data of the effects of sub- 3  months in diabetic gastroparesis but, unfortu-
cutaneous administration of insulin via a pump nately, failed to improve symptoms [36].
linked to continuous monitoring of subcutaneous Novel agents have been shown to accelerate
glucose concentrations [85]. Further studies are gastric emptying. One of these, relamorelin,
required. administered subcutaneously slowed efficacy for
As indicated, pharmacological treatment has symptom improvement, as well as accelerating
focussed on prokinetic medications designed to gastric emptying in phase 2 [91, 92]; phase 3 tri-
accelerate gastric emptying and, despite substan- als were, however, subsequently terminated. The
tial limitations, this remains the mainstay of cur- oral, highly selective 5HT-4 agonists, prucalo-
rent management. All of the prokinetic pride and velusetrag, were developed for the
medications have a significant adverse effect pro- management of constipation, but have been
file and with some, there is a reduction in effect shown to also accelerate gastric emptying.
with sustained administration (tachyphylaxis) Prucalopride has also been reported to accelerate
[86]. Not surprisingly, the relationship of symp- emptying in a small study of diabetic gastropare-
tomatic improvement with effects to accelerate sis [93]. Further studies are required. Recently,
gastric emptying is, at best, weak [84]. Cisapride, the selective 5HT-4 agonist, TAK-954, was
a widely used, and possibly the most effective, reported to normalise gastric emptying of a liquid
480
meal in more critically ill patients with enteral
feeding intolerance than receiving metoclo-
pramide [94]. Again, further studies are required.
Pharmacological therapy directed at specific
symptoms may be useful, including antiemetics
such as 5HT-3 antagonists, phenothiazines, and
cannabinoids, none of which has been evaluated
adequately. The tachykinin receptor (neurokinin-
1 receptor) antagonist, tradipitant, has been
­ trial. Acupuncture has been advocated for symp-
reported to improve nausea in patients with gas-
troparesis, but the study was not powered ade-
quately to evaluate effects in the diabetic subgroup
alone [95]. The tricyclic antidepressant, nortripty-
line, has been reported to be ineffective in symp-
tomatic, idiopathic gastroparesis [96]. Abell et al.
[97] recently reported that sepiapterin, which
increases levels of NO, is associated with an
improvement in gastric accommodation in women
with diabetic gastroparesis, as evaluated by a
nutrient satiety test [98].
Drugs used in the management of chronic
abdominal pain, such as gabapentin and pregaba-
lin, have not been evaluated formally in diabetic
gastroparesis.
An improved understanding of the mecha-
nisms underlying the gastric motor and sensory
dysfunctions in diabetes is likely to be central to
the development of novel, and more effective,
pharmacotherapy. The design of future clinical
trials should also focus on the most relevant end-
points, which will influence trial design and sub-
ject selection criteria. The US FDA has recently
recommended that trials evaluating novel treat-
ments for gastroparesis should have patient-­
reported outcomes as the primary endpoint [99].
Invasive therapies for diabetic gastroparesis
include gastric electrical stimulation, intrapyloric
injections of botulinum toxin, and pyloromyot-
omy, none of which has established efficacy.
Gastric electrical stimulation (approved by the
FDA on humanitarian grounds) involves the
delivery of low energy, high frequency pulses to
the antrum and does not accelerate gastric empty-
ing. The outcomes of initial ‘open-label’ studies
were promising, but not supported by subsequent,
more rigorous, evaluation [100]. However, a
recent study of patients with refractory vomiting
(with or without delayed gastric emptying) found
K. L. Jones et al.
that the frequency of vomiting was less when the
device was turned on [101]. Initial enthusiasm for
the potential use of intrapyloric botulinum toxin
injection was dampened by the negative outcome
of sham-controlled trials [102, 103]. Gastric per-
oral endoscopic pyloromyotomy (G-POEM),
which is being evaluated as a treatment for gas-
troparesis, has not been evaluated in a controlled
tom relief in gastroparesis but, in a recent
Cochrane review the major limitations of the evi-
dence are apparent [104].
5 Gallbladder
The prevalence of cholelithiasis is increased
markedly in type 2 diabetes, which is not surpris-
ing given the high prevalence of well recognised
risk factors, including obesity and hypertriglyc-
eridaemia. Autonomic neuropathy (and possibly,
delayed gastric emptying) is also likely to be
important in the pathogenesis of gallstones, sup-
ported by observations that postprandial gallblad-
der emptying is frequently impaired [105] and
that acute hyperglycaemia inhibits gallbladder
emptying in health [106]. Interestingly, the use of
GLP-1RAs, particularly liraglutide, is associated
with an increased risk of gallbladder/biliary dis-
ease for uncertain reasons [107], but possibly, in
part, by slowing gallbladder contraction.
6 Small and Large Intestine
While diabetic enteropathy occurs frequently
and also represents an important cause of mor-
bidity, there is much less information about it
when compared to gastroparesis. This is despite
the fact that ‘diabetic diarrhoea’, attributed to
autonomic neuropathy [108], was first described
in 1936 [109], and the prevalence of symptoms
of diarrhoea, constipation, and faecal inconti-
nence is much higher in people with diabetes in
community studies [10]. In tertiary populations,
the prevalence of symptoms is even greater
about 20% for diarrhoea and ~60% for constipa-
tion [110].
Gastrointestinal Neuropathy
6.1 Diagnosis
In most cases, diabetic enteropathy represents a
diagnosis of exclusion. Medications used fre-
quently in the management of diabetes are par-
ticularly metformin (which may cause diarrhoea
by inhibiting bile acid absorption),
α-glucosidase inhibitors (acarbose, miglitol,
voglibose), GLP-­1RAs and SGLT-2 inhibitors
(Table 3). Small intestinal bacterial overgrowth
can be diagnosed by jejunal aspiration and cul-
ture, and/or breath tests, but both approaches
have substantial limitations [111]. The preva-
lence of coeliac disease is increased substan-
tially (~5%) in type 1 diabetes, so that
serological screening is justified, even in the
absence of symptoms. Diabetes, usually inap-
propriately regarded as type 2, is not infre-
quently associated with pancreatic exocrine
insufficiency (type 3c diabetes), where the
demonstration of a low level of faecal elastase
is diagnostically useful and pancreatic enzyme
replacement has the potential to improve both
symptoms and glycaemic control, as well as
nutritional status. In patients with diarrhoea,
microscopic colitis, albeit rare and associated
with autoimmune disorders, should be consid-
ered [112]. Hyperthyroidism and Addison’s
disease occur more frequently in type 1 diabe-
tes and may cause diarrhoea. Importantly, dis-
orders that frequently affect the general
population, particularly irritable bowel syn-
drome, also occur frequently in people who
have diabetes. Diabetes has been reported to be
a risk factor for Clostridium difficile infection,
where metformin may be protective [113].
Approaches to the diagnosis of constipation
and faecal incontinence in diabetes are similar to
those in individuals without diabetes. In patients
with constipation, medication should be reviewed
and those that frequently cause constipation ide-
ally ceased (e.g. opioids, anticholinergics, cal-
cium channel antagonists) (Table  4). For mild
constipation, lifestyle modifications, including
increases in physical exercise and dietary fibre,
are usually recommended. A therapeutic trial of a
laxative (bulk, osmotic, or stimulatory) is often
indicated. If this proves ineffective, evaluation
481
Table 3  Potential causes of diarrhoea in patients with
diabetes
• Medications
  – Metformin
 – α-glucosidase inhibition—acarbose, voglibose, or
miglitol
  – GLP-1 receptor agonists
  – Orlistat
  – Sorbitol, mannitol, or xylitol
• Small intestinal bacterial overgrowth
• Coeliac disease (especially in type 1 diabetes)
• Pancreatic exocrine insufficiency
• Microscopic colitis
• Endocrine disorders (especially in type 1 diabetes)
  – Hyperthyroidism
  – Adrenal insufficiency
• ‘Diabetic diarrhoea’
• Faecal incontinence
Table 4  Medications associated with slowing of gastric
emptying
• Opiates
• GLP-1 receptor agonists
• Pramlintide
• Anticholinergics
• Levodopa
• Metformin (?)
• Calcium channel antagonists
• Beta-blockers
• Somatostatin analogs (e.g. octreotide)
for slow transit, or a defecatory disorder, should
be considered.
6.2 Pathophysiology
Intestinal contractile activity may be increased,
decreased, or both (as well as uncoordinated), as
a result of autonomic neuropathy in diabetes,
which the inference is that transit may be ‘nor-
mal’, accelerated, or delayed. Gastroparesis has
been reported to be associated with abnormal
small intestinal motility in ~80% of cases [114].
Rodent models of diabetes have demonstrated
that adrenergic innervation may be impaired and
contribute to diarrhoea, in part by modifying
fluid and electrolyte absorption [112]. Diabetic
diarrhoea is associated with alterations in both
482
gastrointestinal transit and intraluminal pH [115].
In health, acute elevations in blood glucose have
been reported to affect small intestinal motility
[116], but the amount of information is limited.
In type 1 diabetes, small intestinal glucose
absorption is abnormally increased during hyper-
glycaemia for uncertain reasons [117]. As is the
case with gastroparesis, the concept that vagal
dysfunction is the dominant abnormality in dia-
betic enteropathy has been broadened with the
recognition of abnormalities in the ICCs and
nNOS [54].
Small intestinal bacterial overgrowth occurs
frequently, albeit with variable estimates, and is
likely to be secondary to altered small intestinal
motility. While there is relatively little informa-
tion about colonic motility, it is clear that
colonic transit is frequently delayed, and consti-
pation, more likely [118], in individuals with
cardiac autonomic neuropathy [110]. In patients
undergoing colectomy, a loss of myenteric, par-
ticularly inhibitory nitrergic neurons, has been
reported [119].
The abnormalities in anorectal function,
investigated most often in patients with faecal
incontinence, are heterogeneous with changes in
internal and external anal sphincter function and
impaired rectal sensation to distension [120,
121]. Acute hyperglycaemia may reduce anal
sphincter pressures [122].
6.3 Treatment
As indicated, the diagnosis of diabetic enteropa-
thy is essentially one of exclusion and there is a
lack of specific therapy. Small intestinal bacterial
overgrowth should be treated with antibiotics
(most commonly rifaximin), but frequently
relapses. When ‘diabetic diarrhoea’ is refractory,
oral clonidine (an α-2 agonist) and subcutaneous
octreotide may prove effective [112]. There is
limited information about the treatment of consti-
pation in diabetes. There is evidence, albeit lim-
ited, that flaxseed may be more effective than
psyllium and also improve glycaemic control in
type 2 diabetes [123]. The outcomes of small
studies support the efficacy of the 5HT-4 agonist,
K. L. Jones et al.
mosapride [124] and the cholinesterase inhibitor,
pyridostigmine [125]. Lubiprostone, which acti-
vates CIC-2 chloride channels, has been reported
to accelerate colonic transit and increase bowel
frequency in diabetes [126]. There is also evi-
dence that linaclotide, which increases guanylate
cyclase-C, may be useful [127], but further stud-
ies are required. In some cases, anorectal dys-
function responds to biofeedback therapy [128].
7 Modulation of Gastric
Emptying and Small
Intestinal Motility to Improve
Glycaemic Control
Gastric emptying is a major determinant of post-
prandial glycaemia in health, as well as diabetes,
accounting for about a third of the variance in the
initial rise in blood glucose [129–131]. This is of
particular relevance to the achievement of good
glycaemic control in diabetes (often classified as
a glycated haemoglobin (HbA1c) level <7.0%),
which is highly desirable to minimise the risk of
the development and progression of the micro-
vascular complications of diabetes, i.e. retinopa-
thy, nephropathy, and neuropathy, markedly.
When HbA1c is >7.0% but <8.0%, the postpran-
dial rise, rather than preprandial (fasting), blood
glucose, is the dominant determinant (even in
patients managed with insulin) [132]. In a 75  g
oral glucose tolerance test, it is not widely appre-
ciated that the 60 min plasma glucose (rather than
the fasting and 120  min glucose levels that are
used diagnostically) is the strongest predictor of
the future development of type 2 diabetes [133]
and this value is closely related to the rate of gas-
tric emptying in subjects with impaired glucose
tolerance or type 2 diabetes [131]. It is, accord-
ingly, surprising that the importance of the rate of
gastric emptying which, by definition, determines
the delivery of carbohydrates to the small intes-
tine, has until recently received little attention, in
contrast to the focus on the relevance of the car-
bohydrate content of a meal. Modulation of
­gastric emptying, by dietary or pharmacological
means, has been shown to have the capacity to
affect postprandial glycaemic excursions pro-
Gastrointestinal Neuropathy
foundly [40, 134, 135]. For example, in type 2
patients, slowing of emptying by a nutrient ‘pre-
load’ [134] or administration of morphine [40]
reduces the glycaemic response to a subsequent
meal, while acceleration of emptying, as induced
by erythromycin [40], increases it significantly.
Such observations have stimulated the develop-
ment, and now widespread, use of drugs targeting
slowing of gastric emptying to improve glycae-
mic control in type 2 diabetes, particularly ‘short-­
acting’ GLP-1RAs (exenatide BID [136] and
lixisenatide [41, 137]) and the amylin agonist,
pramlintide [138]. The magnitude of the slowing
of gastric emptying induced by these drugs is
dependent on the baseline rate of gastric empty-
ing [136] (so that patients with gastroparesis do
not appear to be susceptible to further slowing)
and predictive of the extent of the reduction in
postprandial glucose [41, 136] (Fig. 5). While it
has been suggested that the more recently devel-
oped ‘long-acting’ GLP-1RAs (e.g. liraglutide,
exenatide QW) do not slow gastric emptying
with chronic administration, i.e. their glucose-­
lowering capacity solely reflects insulinotropic
and glucagonostatic properties [139], recent
studies have shown that this is not the case, i.e.
there is a sustained slowing, albeit probably less
than that induced by short-acting GLP-1RAs [42,
140]. These considerations are of particular rele-
vance to the frequent use of a combination of a
‘short’- or ‘long’-acting GLP-1RA with a basal
insulin in the management of type 2 diabetes. In
this situation, it is possible that routine assess-
Fig. 5 Relationship
between the magnitude
Postprandial glucose AUC(0-3h)
of glucose lowering and
slowing of gastric
emptying in type 2
(mmol/L*h/L)
diabetes (n = 17) [136]
483
ment of gastric emptying, perhaps using a stable
isotope breath test, would allow more person-
alised, and hence, effective management. Studies
to explore this issue are indicated.
It should be appreciated that, in addition to the
rate of gastric emptying, small intestinal motility
is a determinant of the glycaemic response to a
meal [141], albeit very poorly characterised. In
the majority of studies, the relative importance of
changes in gastric emptying from those on small
intestinal motility to postprandial glycaemia has
not been discriminated, in part because of the
substantial technical challenges. The relevance of
changes in small intestinal motility/transit has
been established by studies in which carbohy-
drate has been infused directly into the small
intestine [142]. It is also recognised that GLP-­
1RAs may affect small intestinal motility [143].
This is, again, an area where further research is
indicated.
8 Relevance of Gastric
Emptying and Small
Intestinal Motility
to Postprandial Hypotension
Postprandial hypotension, usually defined as a
fall in systolic blood pressure of >20  mmHg
within 2 h of a meal, has been estimated to occur
in 20–40% of individuals with longstanding type
1 or type 2 diabetes; a prevalence which is prob-
ably greater than that of orthostatic hypotension,
60 Placebo
5µg exenatide
50 10µg exenatide
40 r = -0.49, P=0.0004
30
20
10
0
0 1 2 3 4 5 6
T50 (h)
484
a Gastric emptying
100 Placebo
Lixisenatide
80
% Retention
60
40
20
P <0.001
0 105
0 30 60 90 120 150 180
Time (min)
Fig. 6  Acute effects of lixisenatide (10 μg sc) on (a). gas-
tric emptying of a 75  g oral glucose load, measured by
scintigraphy and (b). postprandial blood pressure in
which is much better characterised [144]. In
addition to diabetes, postprandial hypotension
occurs frequently in a number of neurological
disorders associated with autonomic dysfunc-
tion, such as Parkinson’s disease and is also
common in apparently ‘healthy’ older individu-
als [145]. While there is, unfortunately, a lack of
intervention studies to support direct causality,
postprandial hypotension is strongly associated
with an increased risk of a number of adverse
sequelae, particularly syncope and falls [146]
and it also appears to be an independent risk fac-
tor for mortality [147]. In the broadest sense,
postprandial hypotension can be regarded as a
manifestation of autonomic dysfunction in dia-
betes—a substantial increase in sympathetic
activity is required to maintain blood pressure
postprandially [148] consequent to the postpran-
dial diversion of blood to increase splanchnic
blood flow. However, it is now appreciated that
‘gastrointestinal’ factors unrelated to increased
intestinal blood flow per se, specifically gastric
distension, gastric emptying and the release of
gastrointestinal hormones, are important in the
pathogenesis of postprandial hypotension and
have implications for strategies for effective
management, which is currently lacking [145].
In type 2 diabetes the magnitude of the fall in
systolic blood pressure after an oral glucose load
is greater when the rate of gastric emptying is
relatively more rapid [149], while an increase in
gastric distension, such as induced by drinking
K. L. Jones et al.
b Systolic blood pressure
145 Placebo
Lixisenatide
135
mmHg
125
115
P < 0.001
BL 15 30 45 60 75 90 105 120
Time (min)
patients with type 2 diabetes (n = 15). Lixisenatide mark-
edly slows gastric emptying and attenuates the magnitude
of the postprandial fall in blood pressure [41]
water, attenuates the fall [150]. It has recently
been shown that acute administration of both
exogenous (intravenous) GLP-1 [151] and the
short-acting GLP-1RA, lixisenatide [41] mark-
edly attenuates the fall in systolic blood pressure
induced by oral glucose in type 2 diabetes, prob-
ably primarily by slowing gastric emptying.
While additional studies are required prior to
clinical translation, the outcomes are certainly
promising and support the concept that, in many
ways, postprandial hypotension should be
regarded as a ‘gastrointestinal’, rather than ‘car-
diovascular’, disorder and that dietary and/or
pharmacological strategies which slow gastric
emptying are likely to prove useful in manage-
ment (Fig. 6).
9 Conclusions
Gastrointestinal function, even when normal, is
now recognised to be central to the management
of type 1 and type 2 diabetes. The outcomes of
recent studies capitalising on the novel investiga-
tive techniques represent a major paradigm shift
with the recognition that disordered gastrointesti-
nal autonomic function occurs frequently, that
the abnormalities are heterogeneous and often
associated with clinically relevant impacts on
gastrointestinal symptoms, glycaemic control
and/or postprandial blood pressure. Therapeutic
development has been compromised, to some
Gastrointestinal Neuropathy
extent, by the pursuit of concepts which proved
oversimplistic and, in retrospect, were somewhat
naive. An improved characterisation of dysfunc-
tions appears pivotal to the development of novel
and more effective therapies. The latter must also
be evaluated in well-designed controlled trials,
which is, itself, challenging.
References
1. Du YT, Rayner CK, Jones KL, Talley NJ, Horowitz
M.  Gastrointestinal symptoms in diabetes: preva-
lence, assessment, pathogenesis, and management.
Diabetes Care. 2018;41:627–37.
2. Quan C, Talley NJ, Jones MP, Spies J, Horowitz
M.  Gain and loss of gastrointestinal symptoms in
diabetes mellitus: associations with psychiatric
disease, glycemic control, and autonomic neuropa-
thy over 2 years of follow-up. Am J Gastroenterol.
2008;103:2023–30.
3. Quan C, Talley NJ, Cross S, Jones M, Hammer J,
Giles N, et  al. Development and validation of the
diabetes bowel symptom questionnaire. Aliment
Pharmacol Ther. 2003;17:1179–87.
4. Revicki DA, Speck RM, Lavoie S, Puelles J, Kuo
B, Camilleri M, et al. The American neurogastroen-
terology and motility society gastroparesis cardinal
symptom index-daily diary (ANMS GCSI-DD):
psychometric evaluation in patients with idiopathic
or diabetic gastroparesis. Neurogastroenterol Motil.
2019;31:e13553.
5. Rayner CK, Jones KL, Horowitz M. Semaglutide vs
placebo as an adjunct to intensive behavioral therapy
and body weight in adults with overweight or obe-
sity. JAMA. 2021;326:1213–4.
6. Horowitz M, Maddox AF, Wishart JM, Harding PE,
Chatterton BE, Shearman DJ. Relationships between
oesophageal transit and solid and liquid gastric
emptying in diabetes mellitus. Eur J Nucl Med.
1991;18:229–34.
7. Lluch I, Ascaso JF, Mora F, Minguez M, Pena A,
Hernandez A, et al. Gastroesophageal reflux in dia-
betes mellitus. Am J Gastroenterol. 1999;94:919–24.
8. Horowitz M, Smout AJPM.  Disordered esopha-
geal and gastric motility in diabetes mellitus. Adv
Endocrinol Metab. 1993;4:81–114.
9. Gokturk S, Akyuz F, Arici S, Alpaslan B, Ormeci A,
Soyer OM, et al. Gastroesophageal reflux in asymp-
tomatic patients with diabetes: an impedance study
diabetes, obesity and gastroesophageal reflux. Exp
Clin Endocrinol Diabetes. 2020;128:52–8.
10. Bytzer P, Talley NJ, Leemon M, Young LJ, Jones
MP, Horowitz M.  Prevalence of gastrointestinal
symptoms associated with diabetes mellitus: a
population-­based survey of 15,000 adults. Arch
Intern Med. 2001;161:1989–96.
485
11. Petrick JL, Li N, Anderson LA, Bernstein L, Corley
DA, El Serag HB, et  al. Diabetes in relation to
Barrett’s esophagus and adenocarcinomas of the
esophagus: a pooled study from the International
Barrett’s and Esophageal Adenocarcinoma
Consortium. Cancer. 2019;125:4210–23.
12. Roman S, Marjoux S, Thivolet C, Mion
F. Oesophageal function assessed by high-resolution
manometry in patients with diabetes and inadequate
glycaemic control. Diabet Med. 2014;31:1452–9.
13. Kinekawa F, Kubo F, Matsuda K, Fujita Y, Tomita
T, Uchida Y, et al. Relationship between esophageal
dysfunction and neuropathy in diabetic patients. Am
J Gastroenterol. 2001;96:2026–32.
14. Boeckxstaens GE, Horowitz M, Bermingham H,
Holloway RH. Physiological variations in blood glu-
cose concentration affect oesophageal motility and
sensation in normal subjects. Neurogastroenterol
Motil. 1997;9:239–46.
15. Holloway RH, Tippett MD, Horowitz M, Maddox
AF, Moten J, Russo A. Relationship between esoph-
ageal motility and transit in patients with type I dia-
betes mellitus. Am J Gastroenterol. 1999;94:3150–7.
16. Frokjaer JB, Brock C, Brun J, Simren M, Dimcevski
G, Funch-Jensen P, et  al. Esophageal distension
parameters as potential biomarkers of impaired
gastrointestinal function in diabetes patients.
Neurogastroenterol Motil. 2012;24:1016–e544.
17. Brock C, Softeland E, Gunterberg V, Frokjaer JB,
Lelic D, Brock B, et al. Diabetic autonomic neuropa-
thy affects symptom generation and brain-gut axis.
Diabetes Care. 2013;36:3698–705.
18. Dumic I, Nordin T, Jecmenica M, Stojkovic Lalosevic
M, Milosavljevic T, Milovanovic T. Gastrointestinal
tract disorders in older age. Can J Gastroenterol
Hepatol. 2019;2019:6757524.
19. Badillo R, Francis D.  Diagnosis and treatment of
gastroesophageal reflux disease. World J Gastrointest
Pharmacol Ther. 2014;5:105–12.
20. Kassander P. Asymptomatic gastric retention in dia-
betics (gastroparesis diabeticorum). Ann Intern Med.
1958;48:797–812.
21. Rundles RW.  Diabetic neuropathy. General review
with report of 125 cases. Medicine. 1945;24:111–60.
22. Watson LE, Xie C, Wang X, Li Z, Phillips LK, Sun
Z, et  al. Gastric emptying in patients with well-­
controlled type 2 diabetes compared to non-diabetic
young and older controls. J Clin Endocrinol Metab.
2019;104:3311–9.
23. Brener W, Hendrix TR, McHugh PR. Regulation of
the gastric emptying of glucose. Gastroenterology.
1983;85:76–82.
24. Phillips LK, Deane AM, Jones KL, Rayner CK,
Horowitz M.  Gastric emptying and glycaemia in
health and diabetes mellitus. Nat Rev Endocrinol.
2015;11:112–28.
25. Bharucha AE, Batey-Schaefer B, Cleary PA, Murray
JA, Cowie C, Lorenzi G, et al. Delayed gastric empty-
ing is associated with early and long-term hypergly-
486
cemia in type 1 diabetes mellitus. Gastroenterology.
2015;149:330–9.
26. Chang J, Russo A, Bound M, Rayner CK, Jones
KL, Horowitz M.  A 25-year longitudinal evalua-
tion of gastric emptying in diabetes. Diabetes Care.
2012;35:2594–6.
27. Wang YR, Fisher RS, Parkman HP.  Gastroparesis-­
related hospitalizations in the United States: trends,
characteristics, and outcomes, 1995-2004. Am J
Gastroenterol. 2008;103:313–22.
28. Perano SJ, Rayner CK, Kritas S, Horowitz M,
Donaghue K, Mpundu-Kaambwa C, et  al. Gastric
emptying Is more rapid in adolescents with type 1
diabetes and impacts on postprandial glycemia. J
Clin Endocrinol Metab. 2015;100:2248–53.
29. Phillips WT, Schwartz JG, McMahan CA.  Rapid
gastric emptying of an oral glucose solu-
tion in type 2 diabetic patients. J Nucl Med.
1992;33:1496–500.
30. Schwartz JG, McMahan CA, Green GM, Phillips
WT.  Gastric emptying in Mexican Americans
compared to non-Hispanic whites. Dig Dis Sci.
1995;40:624–30.
31. Wang X, Xie C, Marathe CS, Malbert CH, Horowitz
M, Jones KL, et al. Disparities in gastric emptying
and postprandial glycaemia between Han Chinese
and Caucasians with type 2 diabetes. Diabetes Res
Clin Pract. 2020;159:107951.
32. Jones KL, Horowitz M, Wishart JM, Maddox AF,
Harding PE, Chatterton BE. Relationships between
gastric emptying, intragastric meal distribution and
blood glucose concentrations in diabetes mellitus. J
Nucl Med. 1995;36:2220–8.
33. Kuwelker S, Prichard DO, Bailey K, Bharucha
AE. Relationship between symptoms during a gastric
emptying study, daily symptoms and quality of life in
patients with diabetes mellitus. Neurogastroenterol
Motil. 2021;33:e14154.
34. Jones KL, Russo A, Stevens JE, Wishart JM, Berry
MK, Horowitz M. Predictors of delayed gastric emp-
tying in diabetes. Diabetes Care. 2001;24:1264–9.
35. Parkman HP, Yates K, Hasler WL, Nguyen L,
Pasricha PJ, Snape WJ, et al. Similarities and differ-
ences between diabetic and idiopathic gastroparesis.
Clin Gastroenterol Hepatol. 2011;9:1056–64; quiz
e133–4.
36. Stevens JE, Jones KL, Rayner CK, Horowitz
M. Pathophysiology and pharmacotherapy of gastro-
paresis: current and future perspectives. Expert Opin
Pharmacother. 2013;14:1171–86.
37. Ishii M, Nakamura T, Kasai F, Onuma T, Baba T,
Takebe K. Altered postprandial insulin requirement
in IDDM patients with gastroparesis. Diabetes Care.
1994;17:901–3.
38. Lysy J, Israeli E, Strauss-Liviatan N, Goldin
E. Relationships between hypoglycaemia and gastric
emptying abnormalities in insulin-treated diabetic
patients. Neurogastroenterol Motil. 2006;18:433–40.
39. Horowitz M, Jones KL, Rayner CK, Read NW.
‘Gastric’ hypoglycaemia--an important concept in
K. L. Jones et al.
diabetes management. Neurogastroenterol Motil.
2006;18:405–7.
40. Gonlachanvit S, Hsu CW, Boden GH, Knight LC,
Maurer AH, Fisher RS, et al. Effect of altering gas-
tric emptying on postprandial plasma glucose con-
centrations following a physiologic meal in type-II
diabetic patients. Dig Dis Sci. 2003;48:488–97.
41. Jones KL, Rigda RS, Buttfield MDM, Hatzinikolas
S, Pham HT, Marathe CS, et al. Effects of lixisenatide
on postprandial blood pressure, gastric emptying and
glycaemia in healthy people and people with type 2
diabetes. Diabetes Obes Metab. 2019;21:1158–67.
42. Jones KL, Huynh LQ, Hatzinikolas S, Rigda RS,
Phillips LK, Pham HT, et al. Exenatide once weekly
slows gastric emptying of solids and liquids in
healthy, overweight people at steady-state concen-
trations. Diabetes Obes Metab. 2020;22:788–97.
43. Abell TL, Camilleri M, Donohoe K, Hasler WL, Lin
HC, Maurer AH, et al. Consensus recommendations
for gastric emptying scintigraphy: a joint report of
the American Neurogastroenterology and Motility
Society and the Society of Nuclear Medicine. J Nucl
Med Technol. 2008;36:44–54.
44. Borg MJ, Bound M, Grivell J, Sun Z, Jones KL,
Horowitz M, et al. Comparative effects of proximal
and distal small intestinal administration of metfor-
min on plasma glucose and glucagon-like peptide-1,
and gastric emptying after oral glucose, in type 2
diabetes. Diabetes Obes Metab. 2019;21:640–7.
45. Collins PJ, Horowitz M, Cook DJ, Harding PE,
Shearman DJ.  Gastric emptying in normal sub-
jects--a reproducible technique using a single
scintillation camera and computer system. Gut.
1983;24:1117–25.
46. Lee JS, Camilleri M, Zinsmeister AR, Burton DD,
Kost LJ, Klein PD.  A valid, accurate, office based
non-radioactive test for gastric emptying of solids.
Gut. 2000;46:768–73.
47. Viramontes BE, Kim DY, Camilleri M, Lee JS,
Stephens D, Burton DD, et al. Validation of a stable
isotope gastric emptying test for normal, accelerated
or delayed gastric emptying. Neurogastroenterol
Motil. 2001;13:567–74.
48. Sanaka M, Nakada K, Nosaka C, Kuyama Y.  The
Wagner-Nelson method makes the [13C]-breath test
comparable to radioscintigraphy in measuring gastric
emptying of a solid/liquid mixed meal in humans.
Clin Exp Pharmacol Physiol. 2007;34:641–4.
49. Saad RJ, Hasler WL. A technical review and clini-
cal assessment of the wireless motility capsule.
Gastroenterol Hepatol. 2011;7:795–804.
50. Hasler WL, May KP, Wilson LA, Van Natta M,
Parkman HP, Pasricha PJ, et  al. Relating gastric
scintigraphy and symptoms to motility capsule tran-
sit and pressure findings in suspected gastroparesis.
Neurogastroenterol Motil. 2018;30:e13196.
51. Bartholome R, Salden B, Vrolijk MF, Troost FJ,
Masclee A, Bast A, et al. Paracetamol as a post pran-
dial marker for gastric emptying, A food-drug inter-
action on absorption. PLoS One. 2015;10:e0136618.
Gastrointestinal Neuropathy
52. Horowitz M, Rayner CK, Marathe CS, Wu T, Jones
KL.  Glucagon-like peptide-1 receptor agonists and
the appropriate measurement of gastric emptying.
Diabetes Obes Metab. 2020;22:2504–6.
53. Grover M, Bernard CE, Pasricha PJ, Lurken
MS, Faussone-Pellegrini MS, Smyrk TC, et  al.
Clinical-histological associations in gastropa-
resis: results from the Gastroparesis Clinical
Research Consortium. Neurogastroenterol Motil.
2012;24(531-9):e249.
54. Kashyap P, Farrugia G. Diabetic gastroparesis: what
we have learned and had to unlearn in the past 5
years. Gut. 2010;59:1716–26.
55. Yarandi SS, Srinivasan S.  Diabetic gastrointesti-
nal motility disorders and the role of enteric ner-
vous system: current status and future directions.
Neurogastroenterol Motil. 2014;26:611–24.
56. Thazhath SS, Jones KL, Horowitz M, Rayner
CK.  Diabetic gastroparesis: recent insights into
pathophysiology and implications for man-
agement. Expert Rev Gastroenterol Hepatol.
2013;7:127–39.
57. Chakraborty S, Halland M, Burton D, Desai A, Neja
B, Low P, et al. GI dysfunctions in diabetic gastro-
enteropathy, their relationships with symptoms, and
effects of a GLP-1 antagonist. J Clin Endocrinol
Metab. 2019;104:1967–77.
58. Faussone-Pellegrini MS, Grover M, Pasricha
PJ, Bernard CE, Lurken MS, Smyrk TC, et  al.
Ultrastructural differences between diabetic
and idiopathic gastroparesis. J Cell Mol Med.
2012;16:1573–81.
59. Harberson J, Thomas RM, Harbison SP, Parkman
HP.  Gastric neuromuscular pathology in gastropa-
resis: analysis of full-thickness antral biopsies. Dig
Dis Sci. 2010;55:359–70.
60. Camilleri M, Chedid V, Ford AC, Haruma K,
Horowitz M, Jones KL, et al. Gastroparesis. Nat Rev
Dis Primers. 2018;4:41.
61. Kashyap P, Farrugia G.  Oxidative stress: key
player in gastrointestinal complications of diabetes.
Neurogastroenterol Motil. 2011;23:111–4.
62. Mazzone A, Bernard CE, Strege PR, Beyder A,
Galietta LJ, Pasricha PJ, et  al. Altered expression
of Ano1 variants in human diabetic gastroparesis. J
Biol Chem. 2011;286:13393–403.
63. Puthanmadhom Narayanan S, O’Brien D, Sharma
M, Miller K, Adams P, Passos JF, et  al. Duodenal
mucosal mitochondrial gene expression is associated
with delayed gastric emptying in diabetic gastroen-
teropathy. JCI. Insight. 2021;6:e143596.
64. Schvarcz E, Palmer M, Aman J, Horowitz M,
Stridsberg M, Berne C.  Physiological hyperglyce-
mia slows gastric emptying in normal subjects and
patients with insulin-dependent diabetes mellitus.
Gastroenterology. 1997;113:60–6.
65. Fraser RJ, Horowitz M, Maddox AF, Harding PE,
Chatterton BE, Dent J. Hyperglycaemia slows gas-
tric emptying in type 1 (insulin-dependent) diabetes
mellitus. Diabetologia. 1990;33:675–80.
487
66. Schvarcz E, Palmer M, Aman J, Lindkvist B,
Beckman KW. Hypoglycaemia increases the gastric
emptying rate in patients with type 1 diabetes mel-
litus. Diabet Med. 1993;10:660–3.
67. Marathe CS, Rayner CK, Jones KL, Horowitz
M.  Relationships between gastric emptying, post-
prandial glycemia, and incretin hormones. Diabetes
Care. 2013;36:1396–405.
68. Aylett P. Gastric emptying and change of blood glu-
cose level, as affected by glucagon and insulin. Clin
Sci. 1962;22:171–8.
69. Stunkard AJ, Wolff HG. Studies on the physiology
of hunger. I. The effect of intravenous administration
of glucose on gastric hunger contractions in man. J
Clin Invest. 1956;35:954–63.
70. Jones KL, Berry M, Kong MF, Kwiatek MA,
Samsom M, Horowitz M. Hyperglycemia attenuates
the gastrokinetic effect of erythromycin and affects
the perception of postprandial hunger in normal sub-
jects. Diabetes Care. 1999;22:339–44.
71. Plummer MP, Jones KL, Annink CE, Cousins CE,
Meier JJ, Chapman MJ, et  al. Glucagon-like pep-
tide 1 attenuates the acceleration of gastric empty-
ing induced by hypoglycemia in healthy subjects.
Diabetes Care. 2014;37:1509–15.
72. Rayner CK, Su YC, Doran SM, Jones KL, Malbert
CH, Horowitz M.  The stimulation of antral motil-
ity by erythromycin is attenuated by hyperglycemia.
Am J Gastroenterol. 2000;95:2233–41.
73. Hasler WL, Soudah HC, Dulai G, Owyang
C. Mediation of hyperglycemia-evoked gastric slow-­
wave dysrhythmias by endogenous prostaglandins.
Gastroenterology. 1995;108:727–36.
74. Kuo P, Gentilcore D, Nair N, Stevens JE, Wishart
JM, Lange K, et al. The nitric oxide synthase inhibi-
tor, Ng-nitro-L-arginine-methyl-ester, attenuates the
delay in gastric emptying induced by hyperglycae-
mia in healthy humans. Neurogastroenterol Motil.
2009;21:1175–e103.
75. Aigner L, Becker B, Gerken S, Quast DR, Meier
JJ, Nauck MA.  Day-to-day variations in fasting
plasma glucose do not influence gastric emptying
in subjects with type 1 diabetes. Diabetes Care.
2021;44:479–88.
76. Gaber AO, Oxley D, Karas J, Cardoso S, Hathaway
D, Shokouh-Amiri MH, et  al. Changes in gas-
tric emptying in recipients of successful com-
bined pancreas-­ kidney transplants. Dig Dis.
1991;9:437–43.
77. Schvarcz E, Palmer M, Aman J, Berne
C. Hypoglycemia increases the gastric emptying rate
in healthy subjects. Diabetes Care. 1995;18:674–6.
78. Russo A, Stevens JE, Chen R, Gentilcore D, Burnet
R, Horowitz M, et al. Insulin-induced hypoglycemia
accelerates gastric emptying of solids and liquids
in long-standing type 1 diabetes. J Clin Endocrinol
Metab. 2005;90:4489–95.
79. Murthy TA, Grivell J, Hatzinikolas S, Chapple LS,
Chapman MJ, Stevens JE, et  al. Acceleration of
gastric emptying by insulin-induced hypoglycemia
488
is dependent on the degree of hypoglycemia. J Clin
Endocrinol Metab. 2021;106:364–71.
80. Hjelland IE, Oveland NP, Leversen K, Berstad A,
Hausken T.  Insulin-induced hypoglycemia stimu-
lates gastric vagal activity and motor function with-
out increasing cardiac vagal activity. Digestion.
2005;72:43–8.
81. Kar P, Jones KL, Plummer MP, Ali Abdelhamid Y,
Giersch EJ, Summers MJ, et  al. Antecedent hypo-
glycemia does not attenuate the acceleration of gas-
tric emptying by hypoglycemia. J Clin Endocrinol
Metab. 2017;102:3953–60.
82. Rayner CK, Jones KL, Horowitz M. Is making the
stomach pump better the answer to gastroparesis?
Gastroenterology. 2019;156:1555–7.
83. Olausson EA, Storsrud S, Grundin H, Isaksson
M, Attvall S, Simren M.  A small particle size diet
reduces upper gastrointestinal symptoms in patients
with diabetic gastroparesis: a randomized controlled
trial. Am J Gastroenterol. 2014;109:375–85.
84. Jalleh R, Marathe CS, Rayner CK, Jones KL,
Horowitz M.  Diabetic gastroparesis and glycaemic
control. Curr Diab Rep. 2019;19:153.
85. Koch KL, Hasler WL, Van Natta M, Calles-Escandon
J, Grover M, Pasricha PJ, et al. Satiety testing in dia-
betic gastroparesis: effects of insulin pump therapy
with continuous glucose monitoring on upper gas-
trointestinal symptoms and gastric myoelectrical
activity. Neurogastroenterol Motil. 2020;32:e13720.
86. Tesfaye S, Boulton AJ, Dyck PJ, Freeman R,
Horowitz M, Kempler P, et  al. Diabetic neuropa-
thies: update on definitions, diagnostic criteria, esti-
mation of severity, and treatments. Diabetes Care.
2010;33:2285–93.
87. Parkman HP, Carlson MR, Gonyer
D. Metoclopramide nasal spray is effective in symp-
toms of gastroparesis in diabetics compared to
conventional oral tablet. Neurogastroenterol Motil.
2014;26:521–8.
88. Lee A, Kuo B.  Metoclopramide in the treatment
of diabetic gastroparesis. Expert Rev Endocrinol
Metab. 2010;5:653–62.
89. Tornblom H.  Treatment of gastrointestinal auto-
nomic neuropathy. Diabetologia. 2016;59:409–13.
90. Rayner CK, Horowitz M.  New management
approaches for gastroparesis. Nat Clin Pract
Gastroenterol Hepatol. 2005;2:454–62; quiz 93.
91. Lembo A, Camilleri M, McCallum R, Sastre R,
Breton C, Spence S, et al. Relamorelin reduces vom-
iting frequency and severity and accelerates gas-
tric emptying in adults with diabetic gastroparesis.
Gastroenterology. 2016;151(87-96):e6.
92. Camilleri M, McCallum RW, Tack J, Spence SC,
Gottesdiener K, Fiedorek FT.  Efficacy and safety
of relamorelin in diabetics with symptoms of gas-
troparesis: a randomized, placebo-controlled study.
Gastroenterology. 2017;153:1240–50.e2.
93. Andrews CN, Woo M, Buresi M, Curley M, Gupta
M, Tack J, et  al. Prucalopride in diabetic and con-
nective tissue disease-related gastroparesis: ran-
K. L. Jones et al.
domized placebo-controlled crossover pilot trial.
Neurogastroenterol Motil. 2021;33:e13958.
94. Chapman MJ, Jones KL, Almansa C, Barnes CN,
Nguyen D, Deane AM.  Blinded, double-dummy,
parallel-group, phase 2a randomized clinical trial
to evaluate the efficacy and safety of a highly
selective 5-Hydroxytryptamine Type 4 receptor
agonist in critically ill patients with enteral feed-
ing intolerance. JPEN J Parenter Enteral Nutr.
2021;45:115–24.
95. Carlin JL, Lieberman VR, Dahal A, Keefe MS, Xiao
C, Birznieks G, et al. Efficacy and safety of tradipi-
tant in patients with diabetic and idiopathic gastro-
paresis in a randomized, placebo-controlled trial.
Gastroenterology. 2021;160(76-87):e4.
96. Parkman HP, Van Natta ML, Abell TL, McCallum
RW, Sarosiek I, Nguyen L, et  al. Effect of nor-
triptyline on symptoms of idiopathic gastropare-
sis: the NORIG randomized clinical trial. JAMA.
2013;310:2640–9.
97. Abell TL, Garcia LM, Wiener GJ, Wo JM, Bulat
RS, Smith N. Effect of oral CNSA-001 (sepiapterin,
PTC923) on gastric accommodation in women
with diabetic gastroparesis: a randomized, placebo-­
controlled, phase 2 trial. J Diabetes Complicat.
2021;35:107961.
98. Jalleh RJ, Marathe CS, Jones KL, Horowitz
M, Rayner CK.  Digesting the pathogenesis of
diabetic gastroparesis. J Diabetes Complicat.
2021;35:107992.
99. Pasricha PJ, Camilleri M, Hasler WL, Parkman
HP.  White paper AGA: gastroparesis: clinical
and regulatory insights for clinical trials. Clin
Gastroenterol Hepatol. 2017;15:1184–90.
100. McCallum RW, Snape W, Brody F, Wo J, Parkman
HP, Nowak T.  Gastric electrical stimulation with
Enterra therapy improves symptoms from dia-
betic gastroparesis in a prospective study. Clin
Gastroenterol Hepatol. 2010;8:947–54; quiz e116.
101. Ducrotte P, Coffin B, Bonaz B, Fontaine S, Bruley
Des Varannes S, Zerbib F, et  al. Gastric electri-
cal stimulation reduces refractory vomiting in
a randomized crossover trial. Gastroenterology.
2020;158:506–14.e2.
102. Arts J, Holvoet L, Caenepeel P, Bisschops R, Sifrim
D, Verbeke K, et  al. Clinical trial: a randomized-­
controlled crossover study of intrapyloric injec-
tion of botulinum toxin in gastroparesis. Aliment
Pharmacol Ther. 2007;26:1251–8.
103. Friedenberg FK, Palit A, Parkman HP, Hanlon A,
Nelson DB.  Botulinum toxin A for the treatment
of delayed gastric emptying. Am J Gastroenterol.
2008;103:416–23.
104. Kim KH, Lee MS, Choi TY, Kim TH. Acupuncture
for symptomatic gastroparesis. Cochrane Database
Syst Rev. 2018;12:CD009676.
105. Pazzi P, Scagliarini R, Gamberini S, Pezzoli
A.  Review article: gall-bladder motor function
in diabetes mellitus. Aliment Pharmacol Ther.
2000;14(Suppl 2):62–5.
Gastrointestinal Neuropathy
106. Gielkens HAJ, Van Oostayen JA, Frölich M, Biemond
I, Lamers CBHW, Masclee AAM.  Dose-dependent
inhibition of postprandial gallbladder motility and
plasma hormone secretion during acute hyperglyce-
mia. Scand J Gastroenterol. 1998;33:1074–9.
107. Nauck MA, Muus Ghorbani ML, Kreiner E,
Saevereid HA, Buse JB, Investigators LP, Cobot
LT.  Effects of liraglutide compared with placebo
on events of acute gallbladder or biliary disease in
patients with type 2 diabetes at high risk for cardio-
vascular events in the LEADER randomized trial.
Diabetes Care. 2019;42:1912–20.
108. Dooley CP, el Newihi HM, Zeidler A, Valenzuela
JE.  Abnormalities of the migrating motor complex
in diabetics with autonomic neuropathy and diar-
rhea. Scand J Gastroenterol. 1988;23:217–23.
109. Bargen JA, Bollman JL, Kepler EJ. The “diarrhea of
diabetes” and steatorrhea of pancreatic insufficiency.
Mayo Clin Proc. 1936;11:737–42.
110. Phillips LK, Rayner CK, Jones KL, Horowitz M. An
update on autonomic neuropathy affecting the gas-
trointestinal tract. Curr Diab Rep. 2006;6:417–23.
111. Quigley EMM, Murray JA, Pimentel M.  AGA
Clinical practice update on small intestinal bacte-
rial overgrowth: expert Review. Gastroenterology.
2020;159:1526–32.
112. Selby A, Reichenbach ZW, Piech G, Friedenberg
FK.  Pathophysiology, differential diagnosis,
and treatment of diabetic diarrhea. Dig Dis Sci.
2019;64:3385–93.
113. Eliakim-Raz N, Fishman G, Yahav D, Goldberg E,
Stein GY, Zvi HB, et al. Predicting Clostridium diffi-
cile infection in diabetic patients and the effect of met-
formin therapy: a retrospective, case-­control study.
Eur J Clin Microbiol Infect Dis. 2015;34:1201–5.
114. Camilleri M, Malagelada JR.  Abnormal intestinal
motility in diabetics with the gastroparesis syn-
drome. Eur J Clin Investig. 1984;14:420–7.
115. Sangnes DA, Dimcevski G, Frey J, Softeland
E.  Diabetic diarrhoea: a study on gastrointestinal
motility, pH levels and autonomic function. J Intern
Med. 2021;290:1206.
116. Russo A, Fraser R, Horowitz M. The effect of acute
hyperglycaemia on small intestinal motility in nor-
mal subjects. Diabetologia. 1996;39:984–9.
117. Rayner CK, Samsom M, Jones KL, Horowitz
M. Relationships of upper gastrointestinal motor and
sensory function with glycemic control. Diabetes
Care. 2001;24:371–81.
118. Prasad VG, Abraham P. Management of chronic con-
stipation in patients with diabetes mellitus. Indian J
Gastroenterol. 2017;36:11–22.
119. Chandrasekharan B, Anitha M, Blatt R, Shahnavaz
N, Kooby D, Staley C, et  al. Colonic motor dys-
function in human diabetes is associated with
enteric neuronal loss and increased oxidative stress.
Neurogastroenterol Motil. 2011;23(131-8):e26.
120. Schiller LR, Santa Ana CA, Schmulen AC, Hendler
RS, Harford WV, Fordtran JS. Pathogenesis of fecal
incontinence in diabetes mellitus: evidence for
489
internal-anal-sphincter dysfunction. N Engl J Med.
1982;307:1666–71.
121. Sun WM, Katsinelos P, Horowitz M, Read
NW.  Disturbances in anorectal function in patients
with diabetes mellitus and faecal incontinence. Eur J
Gastroenterol Hepatol. 1996;8:1007–12.
122. Russo A, Sun WM, Sattawatthamrong Y, Fraser R,
Horowitz M, Andrews JM, et al. Acute hyperglyce-
mia affects anorectal motor and sensory function in
normal subjects. Gut. 1997;41:494–9.
123. Soltanian N, Janghorbani M.  Effect of flaxseed or
psyllium vs. placebo on management of constipa-
tion, weight, glycemia, and lipids: a randomized
trial in constipated patients with type 2 diabetes.
Clin Nutr ESPEN. 2019;29:41–8.
124. Ueno N, Inui A, Satoh Y.  The effect of mosapride
citrate on constipation in patients with diabetes.
Diabetes Res Clin Pract. 2010;87:27–32.
125. Bharucha AE, Low P, Camilleri M, Veil E, Burton D,
Kudva Y, et al. A randomised controlled study of the
effect of cholinesterase inhibition on colon function
in patients with diabetes mellitus and constipation.
Gut. 2013;62:708–15.
126. Christie J, Shroff S, Shahnavaz N, Carter LA,
Harrison MS, Dietz-Lindo KA, et al. A randomized,
double-blind, placebo-controlled trial to examine the
effectiveness of lubiprostone on constipation symp-
toms and colon transit time in diabetic patients. Am
J Gastroenterol. 2017;112:356–64.
127. Yang J, Lei Y. Comparison of the efficacy and safety
of different doses of linaclotide for patients with
chronic constipation: a meta-analysis and Bayesian
analysis. Evid Based Complement Alternat Med.
2021;2021:9923879.
128. Wald A, Tunuguntla AK.  Anorectal sensorimotor
dysfunction in fecal incontinence and diabetes mel-
litus. Modification with biofeedback therapy. N Engl
J Med. 1984;310:1282–7.
129. Horowitz M, Edelbroek MA, Wishart JM, Straathof
JW.  Relationship between oral glucose tolerance
and gastric emptying in normal healthy subjects.
Diabetologia. 1993;36:857–62.
130. Jones KL, Horowitz M, Carney BI, Wishart JM,
Guha S, Green L.  Gastric emptying in “early”
noninsulin-dependent diabetes mellitus relationship
to oral glucose tolerance and appetite. J Nucl Med.
1996;37:1643–8.
131. Marathe CS, Horowitz M, Trahair LG, Wishart JM,
Bound M, Lange K, et al. Relationships of early and
late glycemic responses with gastric emptying dur-
ing an oral glucose tolerance test. J Clin Endocrinol
Metab. 2015;100:3565–71.
132. Riddle M, Umpierrez G, DiGenio A, Zhou R,
Rosenstock J. Contributions of basal and postpran-
dial hyperglycemia over a wide range of A1C levels
before and after treatment intensification in type 2
diabetes. Diabetes Care. 2011;34:2508–14.
133. Bergman M.  The 1-hour plasma glucose: common
link across the glycemic spectrum. Front Endocrinol
(Lausanne). 2021;12:752329.
490
134. Ma J, Stevens JE, Cukier K, Maddox AF, Wishart
JM, Jones KL, et al. Effects of a protein preload on
gastric emptying, glycemia, and gut hormones after
a carbohydrate meal in diet-controlled type 2 diabe-
tes. Diabetes Care. 2009;32:1600–2.
135. Kamruzzaman M, Horowitz M, Jones KL, Marathe
CS.  Gut-based strategies to reduce postprandial
glycaemia in type 2 diabetes. Front Endocrinol
(Lausanne). 2021;12:661877.
136. Linnebjerg H, Park S, Kothare PA, Trautmann
ME, Mace K, Fineman M, et  al. Effect of exena-
tide on gastric emptying and relationship to post-
prandial glycemia in type 2 diabetes. Regul Pept.
2008;151:123–9.
137. Lorenz M, Pfeiffer C, Steinstrasser A, Becker RH,
Rutten H, Ruus P, et  al. Effects of lixisenatide
once daily on gastric emptying in type 2 diabetes-
-relationship to postprandial glycemia. Regul Pept.
2013;185:1–8.
138. Samsom M, Szarka LA, Camilleri M, Vella A,
Zinsmeister AR, Rizza RA. Pramlintide, an amylin
analog, selectively delays gastric emptying: potential
role of vagal inhibition. Am J Physiol Gastrointest
Liver Physiol. 2000;278:G946–51.
139. Drucker DJ, Buse JB, Taylor K, Kendall DM,
Trautmann M, Zhuang D, et  al. Exenatide once
weekly versus twice daily for the treatment of type
2 diabetes: a randomised, open-label, non-inferiority
study. Lancet. 2008;372:1240–50.
140. Halawi H, Khemani D, Eckert D, O’Neill J, Kadouh
H, Grothe K, et al. Effects of liraglutide on weight,
satiation, and gastric functions in obesity: a ran-
domised, placebo-controlled pilot trial. Lancet
Gastroenterol Hepatol. 2017;2:890–9.
141. Thazhath SS, Marathe CS, Wu T, Chang J, Khoo
J, Kuo P, et al. The glucagon-like peptide 1 recep-
tor agonist exenatide inhibits small intestinal
motility, flow, transit, and absorption of glucose
in healthy subjects and patients with type 2 dia-
betes: a randomized controlled trial. Diabetes.
2016;65:269–75.
142. Kuo P, Bellon M, Wishart J, Smout AJ, Holloway
RH, Fraser RJ, et al. Effects of metoclopramide on
duodenal motility and flow events, glucose absorp-
tion, and incretin hormone release in response
K. L. Jones et al.
to intraduodenal glucose infusion. Am J Physiol
Gastrointest Liver Physiol. 2010;299:G1326–33.
143. Thazhath SS, Marathe C, Wu T, Chang J, Khoo J,
Kuo P, et al. Effects of the glucagon-like peptide-1
(GLP-1) receptor agonist, exenatide, on small intes-
tinal motility, flow and glucose absorption in healthy
subjects and type 2 diabetes. United European
Gastroenterol J. 2014;2:A29.
144. Mathias CJ, Jones KL. Postprandial hypotension in
autonomic disorders. In: Mathias CJ, Bannister R,
editors. Autonomic failure: a textbook of clinical
disorders of the autonomic nervous system. Oxford:
Oxford University Press; 2013. p. 354–70.
145. Trahair LG, Horowitz M, Jones KL.  Postprandial
hypotension: a systematic review. J Am Med Dir
Assoc. 2014;15:394–409.
146. Puisieux F, Bulckaen H, Fauchais AL, Drumez S,
Salomez-Granier F, Dewailly P.  Ambulatory blood
pressure monitoring and postprandial hypotension in
elderly persons with falls or syncopes. J Gerontol A
Biol Sci Med Sci. 2000;55A:M535–M40.
147. Fisher AA, Davis MW, Srikusalanukul W, Budge
MM.  Postprandial hypotension predicts all-cause
mortality in older, low-level care residents. J Am
Geriatr Soc. 2005;53:1313–20.
148. Masuda Y, Kawamura A. Role of the autonomic ner-
vous system in postprandial hypotension in elderly
persons. J Cardiovasc Pharmacol. 2003;42(Suppl
1):S23–6.
149. Jones KL, Tonkin A, Horowitz M, Wishart JM,
Carney BI, Guha S, et  al. Rate of gastric empty-
ing is a determinant of postprandial hypotension in
non-insulin-dependent diabetes mellitus. Clin Sci
(Lond). 1998;94:65–70.
150. Shannon JR, Diedrich A, Biaggioni I, Tank J,
Robertson RM, Robertson D, et  al. Water drinking
as a treatment for orthostatic syndromes. Am J Med.
2002;112:355–60.
151. Trahair LG, Horowitz M, Stevens JE, Feinle-Bisset
C, Standfield S, Piscitelli D, et al. Effects of exog-
enous glucagon-like peptide-1 on blood pressure,
heart rate, gastric emptying, mesenteric blood flow
and glycaemic responses to oral glucose in older
individuals with normal glucose tolerance or type 2
diabetes. Diabetologia. 2015;58:1769–78.
 Diabetic Neuropathy: Clinical
Management—Genitourinary
Dysfunction in Diabetes
Vincenza Spallone, Enrico Finazzi Agrò,
Roberta Centello, Claudio Lecis, Luca Orecchia,
and Andrea M. Isidori
1 Introduction
Bladder and sexual dysfunction in diabetes
include a variety of conditions dealing with two
distinct functions. In addition to sharing changes
in vascularization and innervation, however, they
hold additional similarities. These are common
complications in people with diabetes reaching in
some manifestations a prevalence of over 50%,
the highest prevalence among diabetic complica-
tions. They are burdened with a relevant impact
on quality of life (QoL) and on survival.
Notwithstanding, they are underdiagnosed and
underestimated. For both bladder and sexual dys-
function, a shift has occurred from the attribution
of main pathogenetic responsibility to diabetic
neuropathy to a multiorgan and multifactorial
pathophysiology. Their nature is recognized as
multidimensional: organic, relational, and intra-
psychic. However, both bladder and sexual dys-
function are still characterized by poor knowledge
and clarity in definition, in diagnostic criteria and
V. Spallone (*)
Endocrinology, Department of Systems Medicine,
University of Rome Tor Vergata, Rome, Italy
e-mail: vincenza.spallone@uniroma2.it
E. Finazzi Agrò · L. Orecchia
Urology, Department of Surgical Sciences, University
of Rome Tor Vergata, Rome, Italy
R. Centello · C. Lecis · A. M. Isidori
Department of Experimental Medicine, Sapienza
University of Rome, Rome, Italy
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7_28
testing standardization, with consequent discrep-
ancies in epidemiological data, although some
advancement has been made mainly for erectile
dysfunction (ED). Finally, they coexist in many
patients pointing to common pathogenesis and an
interrelationship at multiple levels including psy-
chological aspects.
2 Bladder Dysfunction
in Diabetes
The traditional concept of diabetic cystopathy as
the gradual development of decreased bladder
sensation, increased capacity, and impaired
detrusor contractility and emptying [1] has given
way to the actual definition of diabetic bladder
dysfunction as a broad spectrum of abnormalities
and lower urinary tract symptoms (LUTS) [2].
The watershed moment that changed the view of
pathophysiology of diabetic cystopathy was the
observation that in patients with diabetes and
voiding symptoms the most frequent urodynamic
finding was detrusor instability (55%) and not
detrusor areflexia (10%) [3].
2.1 Epidemiology of Diabetic
Bladder Dysfunction
In the general population urinary incontinence
has been reported to occur in 5–69% of women
491
492
and in 1–39% of men, while overactive bladder in
8–42% of women and in 10–26% of men [4].
Risk factors of urinary incontinence in women
are pregnancy, vaginal delivery, pelvic organ pro-
lapse, diabetes, and genetic factors, while in men
risk factors include increasing age, other LUTS,
urinary tract infections, cognitive impairment,
neurologic disorders, and prostatectomy [4]. Risk
factors of overactive bladder in men are age,
other LUTS, depression, constipation, neurologic
conditions, and erectile dysfunction [4]. The
presence of LUTS impairs QoL and is associated
with increased risk of depression with a bidirec-
tional relationship between LUTS and both
depression and anxiety [5]. Moreover, in a meta-­
analysis of 15 studies the presence of moderate to
severe LUTS (without excluding benign prostatic
hyperplasia) was associated with major adverse
cardiac events in both cross-sectional (odds ratio
2.38) and longitudinal studies (odds ratio 1.68),
with the risk being lower in older patients and
higher in those with diabetes [6].
Epidemiological data on bladder dysfunction
in diabetes are scarce and with wide variations
reflecting the lack of standardized diagnostic
modalities and possible selection bias of the stud-
ied population. Over 50% of women and men
with diabetes have bladder dysfunction, with
bladder instability or hypersensitivity being the
most frequent finding, ranging from 39% to 61%
of subjects [2].
2.1.1 Bladder Dysfunction in Men
In African American men, a history of diabetes
was associated with an adjusted odds ratio for
moderate to severe LUTS of 1.95 [7]. In patients
with type 2 diabetes from Taiwan, 22.5% reported
overactive bladder [8], with age, male sex, and
waist circumference being independent risk fac-
tors for overactive bladder [8]. In the UroEDIC
study, an ancillary study of DCCT/EDIC study, at
EDIC year 17, LUTS were present in 25% of
men with type 1 diabetes [9]. Abnormal Valsalva
ratio at DCCT closeout independently predicted
the presence of ED and LUTS at EDIC year 17
(odds ratio 1.50) [9]. At EDIC year 17, LUTS
were associated with increasing age, micro-
(odds ratio 1.86) and macroalbuminuria (odds
V. Spallone et al.
ratio 2.56), possible diabetic polyneuropathy
(odds ratio 1.97), and cardiovascular autonomic
neuropathy (CAN) (odds ratio 2.07) [9, 10].
Interestingly, in 71% of cases LUTS and ED
occurred concomitantly [10].
In men with benign prostatic hyperplasia, dia-
betes was associated with greater severity of
LUTS and lower maximum flow rate [11]. The
clinical overlap between benign prostatic hyper-
plasia and LUTS and variable definitions of these
conditions might explain conflicting reports of an
epidemiological association between benign
prostatic hyperplasia and diabetes [2, 6].
2.1.2 Bladder Dysfunction in Women
In two small studies, abnormal voiding patterns
were more common in women with type 2 diabe-
tes than in control women, with a prevalence of
13.9% and 22.2% compared to 1.8% and 5.6%
and an odds ratio of 4.8 [12, 13]. In three large
population studies, women with diabetes had a
higher incidence and prevalence of urinary incon-
tinence compared to those without diabetes (39%
vs. 26%) and an odds ratio between 1.2 and 1.32
with the greater risk for urge incontinence (up to
3.5) and for severe forms of incontinence (odds
ratio 1.54) [14–17]. In the Look AHEAD study in
overweight/obese women with type 2 diabetes,
weekly incontinence was reported by 27%, with
race/ethnicity (non-Hispanic whites), BMI
≥30 kg/m2, prior hysterectomy, and urinary tract
infection in the prior year being the risk factors
[18]. In the Diabetes RRISK study in women with
type 2 diabetes at least one urinary symptom was
present in 59% and post-void residual volume
≥100  mL in 12%, with this latter finding being
associated with worse glycemic control (odds
ratio 1.30 per % increase in HbA1c) [19]. In the
DCCT/EDIC study at EDIC year 17, urinary
incontinence was present in 30% of women with
type 1 diabetes [20]. Abnormal Valsalva ratio at
EDIC year 13/14 was independently associated
with increased risk of developing urinary inconti-
nence at EDIC year 17 (odds ratio 1.68), while at
EDIC year 17 urinary incontinence was associ-
ated with age and BMI ≥30 kg/m2 [10, 20]. In the
same study, HbA1c predicted the development of
urinary incontinence from year 10 to year 17 with
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
an adjusted odds ratio of 1.41 per % increase in
mean HbA1c in EDIC years 1–10 [21].
2.1.3 Urinary Tract Infections
In a large population-based German study, the
incidence of urinary tract infections in patients
with type 2 diabetes was 111.8/55.8 per 1000
patient-years for women/men, with age
>79 years, female gender, history of urinary tract
infections and cystoscopy, number of comorbidi-
ties, HbA1c >9% and GFR <60  mL/min being
associated with increased risk [22]. A meta-anal-
ysis of 22 studies documented an odds ratio for
asymptomatic bacteriuria of 3.0 and 3.2 in type 1
and type 2 diabetes, respectively, and of 2.6 and
3.7  in women and men, respectively [23].
Moreover, in patients with diabetes a greater risk
of developing a symptomatic urinary infection,
of having unusual infecting organisms, and of
progressing to serious or uncommon conse-
quences was described [2, 24]. Postmenopausal
state and diabetes treatment complexity were
associated with a higher risk of urinary tract
infections in type 2 diabetes, whereas in type 1
diabetes sexual activity and not duration of dia-
betes or HbA1c increased the risk [2]. In the
DCCT/EDIC study, urinary tract infection, pres-
ent in 17% of women with type 1 diabetes, was
associated with microalbuminuria (odds ratio of
2.08) and CAN (odds ratio of 1.57) [10, 20].
In summary, bladder dysfunction findings are
present in >50% of people with diabetes, with
bladder instability being the most common. Men
with diabetes have a twofold risk of LUTS and a
prevalence of storage LUTS of 22.5% in type 2
diabetes and of any LUTS of 25% in type 1 diabe-
tes. Risk factors are age in type 2 diabetes and
age, microalbuminuria, and CAN in type 1 diabe-
tes. The most common urinary symptom in women
is incontinence present in 27% to 39% in type 2
and in 30% in type 1 diabetes, with an odds ratio
from 1.2 to 3.5. Risk factors for incontinence in
type 2 diabetes are obesity, ethnicity, hysterec-
tomy, and urinary infection, while in type 1 diabe-
tes age, BMI, glycemic control, and CAN. Urinary
infections are more common in people with diabe-
tes, with diabetes being a promoter of their pro-
gression. In most men with type 1 diabetes LUTS
overlap with sexual dysfunction.
493
2.2 Physiology of Micturition
The lower urinary tract is composed of the blad-
der and the urethra, which are responsible for
urine storage and voiding. Micturition is regulated
by a complex neural circuitry involving the
peripheral and central nervous system. The blad-
der is mainly innervated by the parasympathetic
pelvic nerve, while the urethra by the sympathetic
hypogastric nerve and somatic pudendal nerve.
Bladder storage depends on the sympathetic acti-
vation of β3-adrenoceptors and simultaneous
deactivation of parasympathetic M2-M3 musca-
rinic receptors, resulting in bladder relaxation.
Closure of the bladder neck is due to the sympa-
thetic activation of α1A/D adrenoceptors. Storage
function is maintained by the spinal cord inhibi-
tory reflex, which is facilitated by the pontine
continence center and also by the hypothalamus,
cerebellum, basal ganglia, and frontal cortex.
Bladder voiding depends on a spino-­bulbo-­spinal
excitatory reflex that is triggered by afferents
fibers from tensor receptors in the bladder and
involves the periaqueductal gray in the midbrain
and the pontine micturition center. Fibers project-
ing from pontine micturition center activate the
sacral bladder preganglionic nucleus responsible
for bladder contraction (through M2-M3 musca-
rinic receptors of acetylcholine) and inhibit the
sacral urethral motor nucleus (Onuf’s nucleus),
causing striated sphincter relaxation. The frontal
cortex and hypothalamus initiate the bladder
voiding reflex and are therefore involved in both
storage and voiding [25]. Due to this complexity,
bladder function is sensitive to numerous injuries
or diseases affecting the nervous system.
2.3 Pathophysiology of Diabetic
Bladder Dysfunction
Expectedly, diabetic damage of both somatic
and autonomic fibers was first considered as the
main pathogenetic mechanism of diabetic blad-
der dysfunction underlying its classical form of
diabetic cystopathy, through the involvement of
(1) sensory afferent fibers responsible for
decreased bladder sensation and increased
bladder capacity and (2) parasympathetic effer-
494
ent fibers leading to impaired detrusor contrac-
tility and reduced bladder emptying. However,
urodynamic and animal studies have suggested
a spectrum of different facets to bladder dys-
function and a multifocal influence of diabetes.
Moreover, they have founded a temporal theory
of diabetic bladder with a progression from an
early compensated storage phase characterized
by bladder hypertrophy and overactivity and a
late decompensated voiding phase with detru-
sor underactivity [26]. However, a clear demon-
stration of this phase progression in humans is
lacking, not all the studies fill in this framework
[27] and this view might turn out to be overly
simplistic.
2.3.1 Diabetic Neuropathy
Diabetic autonomic neuropathy and polyneurop-
athy have been associated with LUTS and uri-
nary incontinence in type 1 diabetes [9, 20], as
well as diabetic polyneuropathy being associated
with low flow rate on urodynamic evaluation
and, somewhat controversially, with high post-
void residual in type 2 diabetes [12, 19, 27].
Abnormal electrophysiologic findings were
shown in patients with diabetes and LUTS, i.e.,
peripheral neuropathy in urethral striated sphinc-
ter and bulbo-cavernosus muscles, increased
latency of bulbo-cavernosus reflexes, and
delayed somatosensor evoked responses of tibial
and pudendal nerves [28, 29]. In addition to the
finding of the reduced density of acetylcholines-
terase positive staining nerves in the bladder
wall in patients with severe type 1 diabetes [30],
an impairment in Aδ and C fiber bladder afferent
pathways was found in women with type 2 dia-
betes using urodynamics with intravesical cur-
rent perception threshold testing, which was
associated with poor emptying function and
detrusor hypoactivity [31].
2.3.2 Neurotrophins
Studies in diabetic animals have shown an
increase in bladder nerve growth factor (NGF)
levels in the early compensatory phases and a fall
in the decompensated phases of bladder dysfunc-
tion [32] leading to the hypothesis that increased
NGF is involved in detrusor overactivity and
V. Spallone et al.
bladder pain, while reduced NGF is a conse-
quence of a deficiency in bladder production or
nerve transport to the bladder [33]. Moreover,
NGF gene therapy by herpes-virus-mediated
NGF gene delivery into the bladder prevented
excessive increases in bladder capacity and
reduced post-void residual volume in diabetic
rats [32]. In addition, the bladders of diabetic
mice showed an increase in the proNGF/NGF
ratio and in the coexpression of proNGF and p75
neurotrophin receptor (p75NTR) in the mucosa
in association with high levels of tumor necrosis
factor-α and bladder changes, which were
reverted by the inhibition of proNGF and antago-
nism of p75NTR [34]. These findings suggest an
involvement of NGF and the proNGF/p75NTR
axis in bladder dysfunction pathogenesis.
2.3.3 Detrusor Smooth Muscle
Dysfunction
In animal models of diabetes, early adaptive
modifications of detrusor muscle have been
described: changes in biological pathways and
up- and downregulation of proteins involved in
muscle function including the proapoptopic path-
way, altered membrane lipid composition and
neurotransmitter release, enhanced muscarinic
sensitivity, abnormal purinergic transmission,
and hyperglycemia-driven pathways suggesting
elevated oxidative stress and tissue remodeling
[33, 35]. Chronic administration in mice of meth-
ylglyoxal, a product of glycolysis involved in the
advanced glycation end products (AGEs) forma-
tion, induced bladder inflammation and increased
micturition frequency and number of non-­voiding
contractions at the urodynamic evaluation, sug-
gestive of an overactive bladder [36].
2.3.4 Urothelial Dysfunction
Urothelial cells have sensor molecules and,
when activated by mechanical or chemical
stimuli via these receptors and ion channels,
they can in turn release mediators, such as nitric
oxide (NO), ATP, acetylcholine, prostaglan-
dins, and substance P and thus interact with
nerves endings, interstitial cells, smooth mus-
cle, and blood vessels with an amplifying effect
of stimuli [32, 35]. Chronic hyperglycemia can
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes 495

alter receptor expression and cause a breach of potentially leading to increased outlet resistance
the barrier and may influence afferent nerves, and voiding dysfunction [39]. The urethra is sub-
thus contributing to bladder instability, hyper- ject to time-dependent changes as with bladder:
activity, and altered bladder sensation [35]. A in the early stage increased urethral pressure dur-
diabetes-induced urothelial senescence has ing micturition, subsequent detrusor-sphincter
been also hypothesized, capable of both dis- dyssynergia, and finally reduced thickness of the
rupting the barrier function and altering signal- urothelium and atrophy of striated muscle bundle
ing and sensation of bladder fullness, resulting impairing voiding efficiency [40] (Fig. 1).
in overactive symptoms [37]. Finally, adher- In summary, somatic and autonomic neurop-
ence of Escherichia Coli with type 1 fimbriae to athy play a major role in bladder dysfunction,
the uroepithelial cells from women with diabe- mainly involving voiding function, but other
tes was twofold compared to controls and was diabetes-­related mechanisms directly affect
related to HbA1c, suggesting an effect of gly- bladder and urethra. The traditional model of
cosylation on bacteria receptors [38]. diabetic cystopathy, including neuropathy as
the main culprit, has been replaced by the con-
2.3.5 Urethra Dysfunction cept of a multilevel spectrum of mechanisms,
In diabetic rats, a condition of urethropathy has and a two-­phase model of bladder dysfunction
been described including external urethral (Fig. 1). Pathophysiology of diabetic bladder
sphincter dysfunction, decreased NO-induced remains, however, incompletely explored
relaxation and instead sensitization to α1-­ together with differences between type 1 and
adrenergic agonists of urethral smooth muscle, type 2 diabetes.

Early Late
Storage Phase Voiding Phase

Factors Polyuria Factors Hyperglycaemia (oxidative stress)

Neuropathy
↓ Neuropeptides
Changes ↑ Bladder mass Detrusor changes
↑ Urethra tickness Bladder-sphincter incoordination
Instrumental Detrusor Overactivity Changes ↓ Urothelium thickness
findings ↓ Striated muscle

Symptoms Urgency, incontinence, Instrumental Detrusor Underactivity

frequency, nocturia findings ↑ post-void residual volume
Possible concomitant BPH (in
men)
Symptoms ↓ sensation, difficulty in voiding
(weak stream, dribbling),
sensation of incomplete voiding

Fig. 1  A temporal progression model of diabetic bladder
dysfunction. A two-phase temporal progression model of
bladder dysfunction has been proposed. In an early phase,
polyuria consequent to hyperglycemia would promote an
adaptive remodeling of the bladder and urethra, including
hypertrophy, while other functional changes regard neuro-
nal transmitter release and excitability and NO production
and control with increased protein nitration [33]. Bladder
and urethra overactivity and storage LUTS are the corre-
spondent instrumental and symptomatic findings. In a late
stage, hyperglycemia-driven pathogenetic mechanisms,
like oxidative stress, induce structural changes in smooth
muscle cells of detrusor and the urethra, including apopto-
sis [41], neuropathy and neurodegeneration in the bladder
[42], and possibly changes in urethro-vesical crosstalk
[40]. This stage of decompensated bladder function is
characterized by an underactive bladder and urethra,
decreased micturition pressure, and increased post-void
residual with the emergence of voiding symptoms [35]
496
2.4 Clinical Manifestations
of Diabetic Bladder
Dysfunction
LUTS in patients with diabetes are often indolent
and unrecognized despite the impact on QoL and
the prognostic meaning. In the early stage,
urgency, increased daytime frequency, and noctu-
ria constitute the reason for an early urologic
referral, as well as urinary incontinence, while
voiding symptoms are generally later complaints
(Fig. 1).
2.4.1 Manifestations in Men
and Women
The distribution of LUTS differs between women
and men, with a predominance of an overactive
bladder in men and of incontinence in women. In
men, bladder storage symptoms and voiding
symptoms might be related to both benign pros-
tatic hyperplasia and diabetic bladder dysfunc-
tion, often coexisting in men over 50, making
their differential diagnosis challenging in the pri-
mary care setting.
In women, symptoms are increased daytime
frequency, nocturia, but also weak urinary
streams and incomplete voiding with increased
post-void residual [12, 13, 19]. This latter finding
in the Diabetes RRISK study was associated with
both obstructive voiding symptoms (2.47 odds
ratio) and urge incontinence (2.18 odds) [19].
However, the most common symptom in women
is urinary incontinence [14]. While in type 2 dia-
betes urge incontinence is the most frequent form
[14, 16], in type 1 diabetes mixed or stress incon-
tinence seems to be prevalent [21].
2.4.2 Urinary Tract Infections
Diabetes causes a threefold risk of asymptomatic
bacteriuria independent of glycemic control, with
pooled prevalence of 14.2% in women and 2.3%
in men [23]. Preliminary treatment of asymptom-
atic bacteriuria does not reduce the risk of symp-
tomatic infections and complications and it is not
recommended by scientific societies together
with its screening [43, 44]. Nonetheless, diabe-
tes, mainly in presence of poor glycemic control
and high post-void residual, is a host-related risk
V. Spallone et al.
factor for symptomatic infections, complications,
and antibiotic resistance [44]. Due to their mech-
anism of action, SGLT2i have been associated
with urinary infections with a relative risk of
1.29  in a meta-analysis of seven clinical trials
[45], although a subsequent meta-analysis did not
confirm this finding and also disproved a dose–
response relation between most SGLT2 inhibi-
tors and urinary infections with the exception of
dapagliflozin [46]. However, conclusions cannot
be made regarding serious urinary infections as
pyelonephritis, which a safety announcement of
the Food and Drug Administration (FDA)
reported as associated with SGLT2i [47].
2.5 Diagnosis of Diabetic Bladder
Dysfunction
Diagnostic pathway includes a preliminary
assessment and a specialist neuro-urological
assessment (Fig. 2).
2.5.1 History, Questionnaires,
and Diaries
Patients history is of paramount importance in the
initial diagnostic evaluation of LUTS and should
include: (1) BMI, diabetes history; (2) lifestyle
and habits (smoking, alcohol, addictive drugs),
the number of absorbent pads used per day, QoL,
disabilities; (3) comorbidities, m
­ edication history
(opioids, anticholinergic agents, α-adrenoceptor
agonists); (4) surgical history (abdominopelvic
and urologic surgery), and in women parity,
obstetric and gynecological surgery; (5) a detailed
investigation of LUTS with regard to onset, initia-
tion (hesitancy, straining, Credé maneuver), and
interruption (paradoxical, passive) of micturition,
voiding frequency, nocturia, urgency, stress,
urgency or mixed urinary incontinence; (6) mac-
rohematuria and signs of urinary infections; (7)
genital, sexual, and bowel dysfunction symptoms.
Standardized questionnaires validated in different
languages represent a useful tool to quantify
symptoms severity, including: (1) the International
Prostatic Symptoms Score (IPSS), validated in
men with benign prostate hypertrophy, allows for
easy assessment of LUTS (excluding inconti-
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes 497

Preliminary diagnostic assessment

History, questionnaires and bladder diary
Physical examination
Urinalysis +/- urine culture and eGFR
Uroflowmetry and post-void residual measurement (ultrasound)

Symptomatic urinary tract Overactive bladder ± Urgency Stress urinary incontinence Voiding-phase symptoms
infection incontinence Lifestyle intervention Lifestyle intervention
Treat infection Lifestyle intervention Pelvic floor muscle training α1-blockers (in men)
Repeat preliminary assessment Antimuscarinics or β-3 agonists Duloxetine Intermittent catheterization
(Do not treat asymptomatic
bacteriuria)

Yes Are symptoms No

refractory to 1st
line therapy?

2nd level diagnostic and specialist Confirm 1st line therapy

neurourological assessment Follow-up of symptoms and QoL using
Urodynamics questionnaires
+/- Urethrocystoscopy Repeat preliminary assessment if
+/- Upper and Lower urinary tract ultrasound symptoms change
+/- Neuro-urophysiological tests

Refractory detrusor overactivity +/- Stress urinary BPH-related symptoms

Detrusor underactivity
incontinence incontinence Stepwise therapeutic approach:
Stepwise therapeutic
Stepwise therapeutic approach: Surgery tailored to Associate 5-ARIs when
st
approach:
Titrate dose of 1 line therapy patient-specific factors indicated
Intermittent catheterization
Switch to or combine with 2nd line therapy BPH surgery
Tibial nerve stimulation
Tibial nerve stimulation Sacral neuromodulation
Onabotulinum A toxin Injection
Sacral neuromodulation
Augmentation cystoplasty
Urinary diversion
Periodical specialist symptom and QoL reassessment is always advisable in patients with symptoms refractory to 1st line therapy. Therapeutic approach in patients with mixed presentations
(i.e., urgency and detrusor underactivity, stress and detrusor overactivity incontinence) may vary and require expert neuro-urological assessment.
Consider screening of diabetic polyneuropathy and autonomic neuropathy.

Fig. 2  An algorithm for a diagnostic and therapeutic
approach to diabetic bladder dysfunction. A preliminary
assessment in a patient with diabetes and LUTS is based
on history, questionnaires, bladder diary, urinalysis, uro-
flowmetry, and post-void residual measurement (by supra-
pubic ultrasound). This relatively inexpensive approach
often supplies all of the information needed.
Questionnaires and a bladder diary at scheduled intervals
(i.e., 6 to 12-monthly) allow for the monitoring of treat-
ment outcomes [48, 49]. Uroflowmetry and post-void
residual measurement can be repeated as well to monitor
the urodynamic treatment effects or in the event of clinical
changes. Urinalysis with urine culture should verify the
efficacy of antibiotic treatment or the suspicion of new
infection. Invasive tests like urodynamics are advisable
only when the preliminary approach is inconclusive or the
initial symptom-based treatment is ineffective or in order
to plan surgery. Upper urinary tract ultrasound and ure-
throcystoscopy are reserved for differential diagnosis
[50–52]. The first-line treatment includes lifestyle inter-
ventions for all LUTS and antimuscarinics or β-3 agonists
for overactive bladder, pelvic floor muscle training and
duloxetine for stress incontinence, and intermittent cathe-
terization for voiding symptoms. In men with bladder out-
let obstruction α1-blockers can be used. In the event of
inefficacy, when a definite diagnosis is obtained, an
upgrading of treatment for detrusor overactivity includes
combination of first-line drugs, tibial nerve stimulation,
onabotulinumtoxinA, sacral neuromodulation and, as a
last resort, surgery. A similar stepwise approach is applied
for detrusor underactivity with tibial nerve stimulation
and sacral neuromodulation [53–56]. eGFR estimated
glomerular filtration rate, BPH benign prostatic hyperpla-
sia, 5-ARIs 5-α reductase inhibitors

nence) and LUTS-­related QoL; (2) the Overactive of preliminary assessment and follow-up of
Bladder Questionnaire Short Form (OAB-q SF) LUTS.  It is a semi-objective tool that quantifies
and the International Consultation on Incontinence LUTS and urodynamic variables such as every
Questionnaire-Urinary Incontinence Short Form voided volume, nocturnal, and daytime voided
(ICI-Q UI SF) are focused on urgency symptoms, volume, for a recommended duration of at least
incontinence and related QoL.  Gender specific 3  days [48, 49]. Patient’s motivation can favor
extended versions of the ICI-Q are also available accurate data collection.
(ICIQ-MLUTS and ICIQ-FLUTS) [57, 58]. A Given the insidiousness and possible patient
bladder/voiding diary is an essential component unawareness of abnormal micturition patterns,
498
questions on urinary symptoms should be
included in the yearly assessment of people with
diabetes together with urine analysis.
Questionnaires for autonomic symptoms such as
Compass 31 [59] contain questions on inconti-
nence and difficult and incomplete voiding and
might be useful for a preliminary identification of
these symptoms, but a comparison with specific
questionnaires is lacking.
2.5.2 Physical Examination
Physical examination constitutes an essential
bedside evaluation of patients with LUTS, includ-
ing: (1) in males, abdominal, inguinal, genital
examination to rule out acute urinary retention or
infections, mass or alteration of the external ure-
thral orifice, digital rectal examination as part of
a standard neuro-urological assessment; (2) in
females, alongside abdominal and pelvic exami-
nation, vaginal examination to assess estrogen
status, pelvic organs prolapse, and urethral hyper-
mobility  with or without  stress urinary inconti-
nence during the cough test. When appropriate, a
more focused neuro-urological exam should
investigate sensation in the S2-S5 dermatomes
bilaterally, the bulbocavernosus and perianal
reflex, anal sphincter tone and voluntary contrac-
tion, and pelvic floor muscle maximal and
repeated contractions with the pubococcygeal
test [50].
2.5.3 Laboratory Investigations
Urinalysis is always recommended to detect
signs of infection, together with urine culture and
antibiotic sensitivity tests in the presence of
symptomatic urinary infections. Kidney function
is usually included in the monitoring of diabetic
nephropathy.
2.5.4 Instrumental Diagnostic
Uroflowmetry with measurement of post-void
residual is a non-invasive test that is necessary in
the preliminary evaluation and requested prior to
any further invasive test in patients with LUTS
who are able to void. Uroflowmetry measures
voided volume and Qmax. Post-void residual is
measured by catheterization or preferably esti-
V. Spallone et al.
mated through a suprapubic ultrasound scan.
Results of uroflowmetry and post-void residual
are needed to plan medical or surgical treatment
of LUTS [51]. While suprapubic ultrasound can
be integrated in the uroflowmetry to estimate
post-void residual, upper urinary tract imaging is
not necessary in the preliminary assessment
unless indicated by clinical features, taking into
account that patients with neurogenic LUTS may
develop urinary tract stones [50].
2.5.5 Invasive Urodynamics
and Urethrocystoscopy
Urodynamic studies are minimally invasive tests
in the outpatient setting that complete the diag-
nostic pathway of LUTS, guide treatment
choices, and assist in patient’ counseling. A
combination of multichannel filling cystometry
and pressure/flow study is generally used in
patients with diabetes. Urodynamics assesses
detrusor and bladder outlet function and can lead
to the diagnosis of urodynamic incontinence or
severe detrusor overactivity in the absence of
incontinence, can suggest treatment in patients
resistant to first-line therapy, and can help to dif-
ferentiate the cause of bladder outlet obstruction
symptoms in males (due to either benign pros-
tatic hyperplasia, detrusor underactivity or both)
[51]. Invasive urethrocystoscopy can be suitable
for the diagnosis of bladder outlet obstruction
(i.e., urethral stenosis) and in the event of urethra
or bladder pathologies are suspected (i.e., gross
hematuria).
2.5.6 Uro-Neurophysiological Tests
Electrophysiological testing of the motor or sen-
sory function of the pelvic floor muscles, urethral
and anal sphincter, pudendal nerves, bladder, and
urethra is available and able to identify peripheral
neuropathy, but they are considered elective pro-
cedures [52].
Given the association between lower urinary
tract dysfunction with CAN and polyneuropa-
thy, screening for these complications, through
the assessment of sensorimotor and autonomic
symptoms and signs, is advisable in patients
with diabetes and LUTS and vice versa screen-
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
ing for LUTS in those with diabetic neuropathy
[29, 60].
A stepwise algorithm for diagnostic manage-
ment of patients with diabetes and LUTS is illus-
trated in Fig. 2.
2.6 Treatment of Diabetic Bladder
Dysfunction
Treatment is multimodal and focuses on the pres-
ervation of renal function, relief of LUTS, pro-
motion of continence, prevention of infectious
complications, and amelioration of QoL [53].
Management strategies include behavioral inter-
vention, pharmacological treatment, and surgery.
Patient reported outcomes and diaries are useful
in the follow-up to monitor changes and treat-
ment outcomes.
2.6.1 Lifestyle Interventions
General behavioral modifications are advised in
any patient presenting with LUTS.  A reduction
of fluid intake in certain day periods might help
to control nocturia. Scheduled time voiding
might be employed in the presence of decreased
bladder sensitivity and high voided volumes.
Reduction of alcohol and caffeine intake might
prevent diuretic effects and their irritative conse-
quences. A review of medications and their time
of administration (especially diuretics) is advis-
able to reduce bothersome effects on LUTS.  In
addition to glycemic control, whose association
with incontinence in type 1 diabetes [21] and
voiding function in type 2 diabetes [19] is well
documented, lifestyle interventions play a spe-
cific therapeutic role in patients with urinary
incontinence. Results from the DPPOS study
showed a lower prevalence of urinary inconti-
nence in overweight women with impaired glu-
cose tolerance enrolled in intensive lifestyle
intervention group compared to oral metformin
and placebo groups over the course of a 6-year
follow-up period (46.7% vs. 53.1% vs. 49.9%
urinary incontinence/week) with a mediating
effect of reduction in BMI and waist circumfer-
ence [61]. Intensive lifestyle intervention has
499
shown efficacy on urinary incontinence (but not
nocturia or frequency) also in overweight/obese
men with type 2 diabetes in the Look AHEAD
trial with almost double the odds of urinary
incontinence resolving in the intensive lifestyle
intervention group compared to only diabetes
support and education [62]. Pelvic muscle train-
ing can improve symptoms of stress and other
forms of urinary incontinence and should be con-
sidered in first-line conservative management
program for patients with urinary incontinence
[52, 54].
2.6.2 Treatment of Symptoms
Related to Benign Prostatic
Hyperplasia
Men with diabetes presenting with symptoms of
bladder outlet obstruction might benefit from
α1-adrenoceptor antagonists as first-line therapy,
with a preference for selective α1-blockers (tam-
sulosin, silodosin) and similar efficacy as in men
without diabetes [11]. Drug-induced vasoactive
effects could exacerbate orthostatic hypotension
in those with CAN. Treatment with 5-α reductase
inhibitors (5-ARIs) reduces acute urinary reten-
tion episodes and the need for surgery and is con-
sidered in association with α1-blockers in patients
with prostate volumes >40  mL and prostate-­
specific antigen levels >1.4–1.6 ng/mL. Specialist
evaluation and prostate cancer screening, assess-
ment of sexual function, and proper counseling
on the possible impact on sexual function are
advisable in patients with diabetes prior to pre-
scription [63]. When indicated, surgery can be
performed, taking into account in preoperative
counseling that diabetes is associated with some
persistence of LUTS and voiding dysfunction at
3 months after surgery [64].
2.6.3 Treatment of Urgency, Detrusor
Overactivity, and Urgency
Incontinence
Muscarinic receptor antagonists (antimuscarin-
ics) are the cornerstone of treatment of storage
symptoms in the general population and can be
also offered to patients with diabetes. All avail-
able formulations have similar clinical efficacy
500
[65]. M3 selective inhibitor darifenacin has
shown similar efficacy in patients with overac-
tive bladder and diabetes vs. those without
comorbidities, and a similar class effect for all
antimuscarinics might be suggested [65].
Patients’ counseling on side effects like xerosto-
mia, constipation, dizziness, blurred vision,
nasopharyngitis, and occasionally tachycardia
is advisable, as well as on the possible worsen-
ing effect on post-void residual, which requires
regular voiding symptoms monitoring and
1-month re-­evaluation and represents a contra-
indication if ≥150 mL. Combination therapy of
antimuscarinics with α1-blockers can be effec-
tive in men with moderate-severe LUTS not
responding to a single first-line drug, taking into
account the safety profile of both. Mirabegron is
the only licensed oral β-3 adrenoceptor agonist
with efficacy in treating voiding frequency,
urgency, and urgency incontinence in the gen-
eral population. A pooled analysis of trials
showed a better tolerability profile compared to
antimuscarinics in elderly patients [66], with
less xerostomia and constipation. However,
hypertension, headache, and nasopharyngitis
are drug specific adverse effects, while severe
uncontrolled hypertension is a contraindication.
Evidence on combination therapy with mirabe-
gron is lacking.
Electrical neuromodulation is a therapeutic
option in patients with an overactive bladder.
Tibial nerve stimulation (TNS), via a percuta-
neous or transcutaneous approach, delivers
electrical stimuli to the sacral micturition cen-
ter via the S2-S4 sacral nerve plexus through a
fine gauge needle or a patch positioned above
the medial malleolus. Treatment is performed
in an outpatient setting in cycles of 10–12 once-
weekly stimulations of 30 min. TNS was effec-
tive and comparable to the antimuscarinic drug
tolterodine in females with urgency urinary
incontinence [67, 68] and also effective in those
not benefiting from antimuscarinics. TNS is a
minimally invasive and safe option available as
a second therapeutic line or in combination
with oral treatment. However, data on the long-
V. Spallone et al.
term durability of treatment [69] and on the
efficacy in patients with diabetes are needed.
Implantation of a permanent sacral neuromodu-
lator (SNM) seems to modulate spinal cord
reflexes and brain networks by peripheral affer-
ents. It was similarly effective in patients with
diabetes as in those without in a 29-month fol-
low-up study, with ­success rates of 69.2% for
urgency urinary incontinence and 85.7% for
urgency-frequency [70]. The finding of a higher
number of explants due to infection in patients
with diabetes in this study (16.7% vs. 4.3%)
[70] was not however confirmed in another
study (3.1% vs. 6.8%) [71]. While studies com-
paring TNS and SNM are lacking, lower risk
profile, minimally invasive nature, and lower
(initial) costs might favor TNS, although SNM
was more intensively investigated. Patient pref-
erence and compliance are other aspects to con-
sider when choosing the neuromodulation
procedure.
OnabotulinumtoxinA (onabotA) injection in
the detrusor muscle represents a well-established
therapeutic alternative for refractory detrusor
overactivity and urgency incontinence both in
the idiopathic and neurogenic forms. The proce-
dure is minimally invasive (by cystoscopy under
local anesthesia) and can be performed in the
outpatients setting. In a retrospective study, ona-
botA obtained in 48 patients with diabetes and
refractory detrusor overactivity a similar success
rate as in patients without diabetes (56% vs.
61%), but those with diabetes had a higher
6 month rate of large post-void residual and of
general weakness [72]. Thus, patients should
receive preliminary information on the drug
safety profile and adequate training in clean
intermittent catheterization if motivated to start
treatment.
Augmentation cystoplasty is another option to
treat refractory detrusor overactivity. No data are
available on comparison of this alternative with
SNM or onabotA injection in the general or dia-
betic population. Urinary diversion is the last sur-
gical choice to treat symptoms that are refractory
to any other therapeutic approach.
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
2.6.4 Treatment of Detrusor
Underactivity
Despite preclinical studies with cholinergic drugs
and intravesical cannabinoids, no pharmacologi-
cal treatment has been validated for detrusor
underactivity. Clean intermittent self- or aided-­
catheterization still represents the preferred
option for patients with underactive detrusor and
high post-void residual or no detrusor contractil-
ity. Treatment reduces bacteriuria and infections
compared to indwelling catheterization [73].
Frequency of daily catheterizations should be tai-
lored to symptoms and degree of underactivity
and to maintain voiding volumes less than 400–
500 mL. Patient education on technique is essen-
tial and a third-party catheterization for patients
with poor manual dexterity or cognitive dysfunc-
tion is a preferable option to indwelling catheter-
ization. In a systematic review, TNS showed
limited efficacy in patients with idiopathic non-­
obstructive urinary retention with an objective
success rate (reduction in number of catheteriza-
tions or catheterized volume) up to 41% and a
subjective success rate (patient’s request for con-
tinued chronic treatment) up to 59% [55]. No
studies are available in patients with diabetes.
SNM represents a therapeutic option in patients
with diabetes and detrusor underactivity with a
success rate of 66.7% [70] but with a possible
increased risk of infection. SNM might enable
simultaneous pancreas and kidney transplanta-
tion in candidate patients with detrusor acontrac-
tility requiring clean intermittent catheterization
and at high risk of recurrent urinary infections by
restoring voiding and preventing urinary infec-
tions [74].
2.6.5 Treatment of Stress Urinary
Incontinence
The selective serotonin-norepinephrine reuptake
inhibitor (SNRI) duloxetine increases the stimu-
lation of the spinal cord receptors of the pudendal
motor neurons and improves the resting tone and
contraction strength of the striated urinary
sphincter. Duloxetine was found to improve
female stress urinary incontinence and related
QoL. Side effects are nausea and vomiting, xero-
501
stomia, constipation, dizziness, insomnia, som-
nolence, and fatigue and can lead to premature
discontinuation [75]. Surgical treatment includes,
depending on the underlying cause of inconti-
nence, adjustable continence therapy balloons,
urethral slings, artificial urinary sphincter, and
other open or laparoscopic assisted incontinence
surgeries with progressively increasing invasive-
ness. The surgical procedure should be tailored to
the patient and performed by expert functional
urologist in referral centers. As for other pros-
thetic implants, diabetes and obesity were associ-
ated with mesh erosion for trans-vaginal and
trans-obturator sling surgery [76]. Stress urinary
incontinence rarely occurs in males unless as a
consequence of prostatic surgery. Surgical
options are similar in the general and diabetic
populations and include balloons, slings, and
artificial sphincter implantation.
Figure 2 presents a therapeutic algorithm for
bladder dysfunction in diabetes.
2.7 Barriers to Effective
Management of Diabetic
Bladder Dysfunction
Despite significant advancement in the last two
decades, the natural history and pathophysiology
of diabetic bladder dysfunction are still poorly
known. It remains underrepresented in research,
epidemiological studies do not adopt standard-
ized diagnostic criteria and are most cross-­
sectional, therapeutic trials are very few, thus
limiting the possibility to build treatment evi-
dence, with data on tolerability and treatment
adherence very limited. Cost-effectiveness stud-
ies on diagnostic and treatment pathways are
lacking and awareness of bladder dysfunction is
not widespread in clinical practice. Diagnosis
often occurs at an advanced stage, and manage-
ment is almost exclusively in the hands of urolo-
gists [56]. Nonetheless, the evidence of a
beneficial role of the lifestyle intervention, of
glycemic control and even of the choice of anti-
hyperglycemic drugs [77] as well as the prognos-
tic cardiovascular meaning of the LUTS in males
502
[6], points to the need for a holistic and multidis-
ciplinary approach to the prevention and manage-
ment of diabetic bladder dysfunction (including
the general practitioner, urologist, endocrinolo-
gist, neurologist, gynecologist, and cardiologist).
3 Male Sexual Dysfunction
in Diabetes
Male sexual dysfunction includes ED, abnormal-
ities of orgasmic, ejaculatory function, sexual
desire, enjoyment, and failure of genital response.
Although diabetes may well affect different
aspects of sexual health, most attention has been
focused on ED.
3.1 Epidemiology of Diabetic
Erectile Dysfunction
3.1.1 Prevalence of Diabetic Erectile
Dysfunction
According to the Fourth International
Consultation on Sexual Medicine [78] ED is
defined as the consistent or recurrent inability to
attain and/or maintain penile erection sufficient
for sexual satisfaction. In the population-based
Massachusetts Male Aging Study (MMAS), the
prevalence of moderate and severe ED was 35%
[79]. Many studies documented an association
between ED and diabetes. A meta-analysis of 145
studies showed a prevalence of ED of 59.1% in
diabetes, of 37.5% in type 1 diabetes and of
66.3% in type 2 diabetes [80]. This meta-analysis
highlighted the heterogeneity of the included
studies with regard to the prevalence rate (from
35% to 80%), the countries, the diagnostic
approach, and the information on diabetic com-
plications. In the eight studies comparing men
with diabetes and healthy controls, the preva-
lence was 51.6% and 25.5%, respectively, with
an odds ratio for ED of 3.62 [80]. In a large
Italian cross-sectional study, the mean prevalence
of ED in men with diabetes was 35.8% [81],
compared to 12.8% in the general population
[82]. Available data suggest that ED develops in
people with diabetes 10–15 years earlier than in
V. Spallone et al.
those without [80]. Moreover, although related to
age and diabetes duration, ED starts early in dia-
betes, might be present in 32% of patients with
recently diagnosed type 2 diabetes, as docu-
mented by SUBITO-DE study [83], and may be
associated with undiagnosed diabetes, as in the
NHANES study, with an odds ratio of 2.2 that
reached a value of 8.7  in the 40–59 age group
[84]. The MMAS provided an incidence rate of
ED of 26 cases/1000 man-years [85]. In Italian
men with diabetes the incidence of ED was 68
cases/1000 person-years, more than twice that
seen in the MMAS [86].
In summary, both the prevalence and inci-
dence of ED are more than tripled and doubled,
respectively, in males with diabetes compared to
healthy people. ED develops 10–15 years earlier
in men with diabetes. ED starts early in diabetes
and can be present at diabetes diagnosis or even
precede it.
3.1.2 Clinical Correlates,
Comorbidities, and Risk Factors
of Diabetic Erectile Dysfunction
Population-based studies have shown that in
addition to age, clinical correlates or comorbidi-
ties of ED were pelvic surgery, diabetes (an age-­
adjusted odds ratio of 3.95), LUTS, hypertension,
heart disease, heavy smoking, and depression,
with instead a protective role of education, physi-
cal activity, and alcohol drinking [87, 88].
Clinical correlates of ED in diabetes were found
to be advancing age, diabetes duration, poor gly-
cemic control, hypertension, hyperlipidemia,
sedentary lifestyle, smoking, and diabetic com-
plications [89]. However, in the cited meta-­
analysis by Kouidrat et al. [80] only hypertension
was a moderator of ED presence in men with dia-
betes. When considering longitudinal studies, in
the MMAS, lower levels of education, diabetes,
heart disease, and hypertension were indepen-
dent predictors of ED [85], while in people with
diabetes, age, diabetes duration, renal disease,
and hypertension played a predictive role for ED
[86]. In the UroEDIC II study, ED was present in
45% of men with type 1 diabetes and in more
than half the cases it was associated with other
urological complications, with age, poorer glyce-
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
mic control, and other diabetic complications
including CAN [10]. Moreover, during a 16-year
follow-up the rate of incident ED was 43.4 cases
per 1000 person-years [90]. Risk of ED increased
by 21% with each 10 mmHg of systolic BP eleva-
tion but only in those not under antihypertensive
medications [90]. Similarly, in type 2 diabetes in
the Look AHEAD trial, at a cross-sectional level,
every 10 mmHg increase in systolic/diastolic BP
was associated with a 30%/10% increased preva-
lence of ED [91].
In summary, in men with diabetes, clinical
correlates of ED are age, diabetes duration, gly-
cemic control, hypertension, hyperlipidemia,
sedentary lifestyle, smoking, diabetic complica-
tions, and LUTS, with age, duration, glycemic
control, hypertension, and renal disease being
longitudinal predictors of ED, and with an over-
all prominent role of hypertension.
Depression and QoL
In the general population, depression is common
among patients with ED with a frequency rang-
ing from 8.7% to 43.1% [92], with an interrela-
tionship between depression and ED [92].
Depression and diabetes share reciprocal suscep-
tibility and comorbidities and the prevalence of
symptoms of depression in patients with diabetes
is around 30% [93]. In this context, the effect of
ED on mental health is of extreme relevance. In
the SUBITO-DE study, ED and severe depres-
sion were already associated in patients with
newly diagnosed type 2 diabetes [83]. In Italian
QuED study in type 2 diabetes, ED was associ-
ated with higher diabetes-specific health distress,
worse psychological adaptation to diabetes,
severe depressive symptoms (present in 45.6% of
patients with frequent ED), lower scores in the
mental components of QoL, and a less satisfac-
tory sex life [94]. The impact of ED on QoL was
independent of the presence of depression [94].
After 3 years of follow-up, a worsening of physi-
cal functioning scores was observed only in
patients with ED at baseline, while the newly
developed ED was preceded by a deterioration in
all QoL dimensions and a worsening in depres-
sive symptoms and was in turn associated with an
increase in depressive symptoms and decrease in
503
QoL [95]. In the UroEDIC study, at EDIC year
17, the presence of ED was associated with worse
QoL (an odds ratio of 3.01) and higher level of
psychiatric symptoms independently of depres-
sion or microangiopathic complications [96].
Moreover, men with history of effective treat-
ment of ED had almost identical QoL as those
who had never experienced ED [96]. Possible
mechanisms linking depression and ED in diabe-
tes are both biological and psychological and
might also include hypogonadism, autonomic
dysfunction, and the presence of comorbidities.
In summary, in the diabetic scenario where
depression and diabetes already have a recipro-
cal relationship, the presence of ED increases the
risk of depression and vice versa, the association
might be present at diagnosis of type 2 diabetes.
ED has an adverse impact on QoL, independent
of depression, with a tripled risk of having low
QoL in type 1 diabetes. The relationship, also in
this case, is bidirectional given that ED predicts
low QoL and low QoL (together with depression)
predicts the development of ED.
Cardiovascular Disease
There is enough evidence that ED shares the
same risk factors as cardiovascular diseases and
in addition, ED in itself is actually considered an
early sign of forthcoming cardiovascular disease
with a time window of 2–5 years before its devel-
opment [97]. At least five meta-analyses have
confirmed the association between metabolic
syndrome and its components with up to 2.6-fold
increase in ED, in particular fasting blood glu-
cose [98] and hypertension [99, 100]. Past and
current smoking and cannabis use were associ-
ated with higher risk, while physical activity and
moderate alcohol exerted a protective role in
other meta-analyses focused on these lifestyle
factors [97, 101]. Conversely, two large meta-­
analyses showed that ED was associated with
increased risk of cardiovascular disease, coro-
nary heart disease, stroke, and all-cause mortality
with the highest pooled relative risk of coronary
heart disease (up to 1.62) [102, 103]. Moreover,
the risk associated with ED was higher at younger
ages, and a grading effect of ED severity in pre-
dicting cardiovascular events was suggested
504 V. Spallone et al.

[103]. In the ED population not only conven- gonadism associated with both ED and cardio-
tional cardiovascular risk factors but also other vascular morbidity [97].
factors may increase the cardiovascular risk
including organic factors such as low testoster- Diabetic Neuropathy
one and PRL levels, psychological and relational Considering, among the many available studies,
aspects like depression and reduced sexual inter- those with larger sample sizes and standard
course [97]. assessment of CAN and ED, the overall findings
In summary, cardiovascular disease increases suggest that CAN is associated with ED [9, 81,
the risk of ED in diabetes, and in the general 109]. In a cross-sectional Italian study, peripheral
population there is evidence that (1) traditional vascular disease, cardiac disease, nephropathy,
cardiovascular risk factors increase the risk of autonomic neuropathy, polyneuropathy, diabetic
ED, (2) lifestyle changes are able to improve ED, foot, and retinopathy were all independent cor-
(3) ED is associated with a relative risk of devel- relates of ED in diabetes, with the greatest odds
oping cardiovascular disease ranging from 1.34 for autonomic neuropathy (4.3 vs. 3.6, 3.0 and
to 1.62 (a magnitude similar to that of main car- 2.5 of diabetic foot, polyneuropathy, and arteri-
diovascular risk factors), and (4) ED can be con- opathy) [81]. However, in a subsequent 2.8-year
sidered a low-cost biomarker for the need for follow-up study, the relative risk of ED associ-
more intensive cardiovascular risk factor ated with autonomic neuropathy was 1.16 com-
modification. pared to 3.79 of diabetic foot, 2.02 of coronary
artery disease, 1.97 of renal disease, 1.86 of poly-
Mechanisms Linking Erectile Dysfunction neuropathy, and 1.75 of peripheral vascular dis-
and Cardiovascular Risk ease [86]. In the UroEDIC study, ED development
The common risk factors, their mutual associa- at EDIC year 17 was associated with lower car-
tion, and the role of ED as a harbinger of cardio- diovascular reflex tests (CARTs) at DCCT close-
vascular events suggest a shared out and EDIC year 16/17 [9]. CAN diagnosis at
pathophysiological basis. The first link is vascu- EDIC year 16/17 was associated with 2.65 greater
lar disease: the artery size hypothesis suggests adjusted odds of ED or LUTS (not of ED only)
that an equally sized atherosclerotic plaque [9]. In the same cohort at EDIC year 17, CAN
would be tolerated less and compromise flow had unadjusted odds for ED of 2.82 [10].
more in the smaller penile arteries than in coro- Regarding diabetic polyneuropathy, the
nary arteries, which is the reason why ED will UroEDIC study showed an association between
precede by 2–5  years coronary artery disease confirmed diabetic polyneuropathy and ED
[104]. Another possibility regards an impairment [110]. Moreover, at EDIC year 17, the presence
of endothelium-dependent and endothelium-­ of neuropathic symptoms was associated with
independent vasodilation before the development unadjusted odds for ED of 2.15, while retinopa-
of atherosclerotic plaque and more likely to cause thy and microalbuminuria had odds of 2.11 and
ED than angina. The observed benefit of type 5 2.83, respectively [10]. In two small studies in
phosphodiesterase inhibitors (PDE5Is) on the patients with type 1 and type 2 diabetes, ED was
incidence of cardiovascular disease in men with associated with impairment in measures of both
ED [105] and of major adverse cardiac events in large (neuropathic symptoms, vibration percep-
men with diabetes and silent coronary artery dis- tion threshold, sural and peroneal nerve ampli-
ease [106] might involve endothelial dysfunc- tude, and conduction velocity) and small fiber
tion. In fact, PDE5Is in patients with diabetes or function (thermal thresholds, deep breathing test,
high cardiovascular risk were able to improve intraepidermal nerve fiber density, and corneal
penile and systemic endothelial function, reduce confocal microscopy parameters) [111, 112].
ET-1 and endothelial inflammatory mediators, These and other studies have advocated a greater
and increase circulating endothelial progenitor association of ED with abnormalities in small
cells [107, 108]. Another linking factor is hypo- than in large nerve fiber function [111–113].
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
In summary, CAN diagnosis in cross-sectional
studies is associated with ED with increased
odds up to 4, while in longitudinal studies its pre-
dictive power is lower and equivalent or less than
that of other diabetic complications. Similarly,
the presence of diabetic polyneuropathy is asso-
ciated with at least double the risk of ED in both
cross-sectional and longitudinal studies, with a
possible (but not definitively shown) preferential
link with small fiber function abnormalities.
 echanisms Linking Erectile Dysfunction
M corpora cavernosa, causing a compression of the
and Diabetic Neuropathy
The association of diabetic neuropathy with ED
has biological soundness. Impaired sensation of
the penis takes away local information to the
brain and the afferent arm of reflex erection,
abnormal motor function impacts on skeletal
muscles participating in erection; autonomic neu-
ropathy affects arterial dilation through impaired
NO release and NO-synthase function of nonad-
renergic noncholinergic (NANC) neurons, and
trabecular smooth muscle relaxation of the cor-
pus cavernosum [114]. In vitro examination of
the corpus cavernosum tissue of men with diabe-
tes showed impairment of autonomically medi-
ated relaxation of smooth muscle as well as
reduction in endothelium-dependent relaxation
[115]. Dysfunction and degeneration of the
NANC nerves of the corpus cavernosum contrib-
ute to the NO defect, the main culprit of ED,
which follows also the loss of NO endothelial
production/activity induced by many factors like
the increase in AGEs formation, protein kinase C
(PKC) activation, oxidative stress, arginase II,
and RhoA/Rho-kinase activity (involved in penile
flaccidity), and the presence of androgen defi-
ciency [116].
3.2 Physiology
and Pathophysiology
of Erectile Function
The erectile mechanism is based on the mutual
interaction of vascular, neural, hormonal, and
psychological processes [117]. The penis remains
in its flaccid state as long as the vascular smooth
505
muscle cells are contracted. Sexual stimulation
causes NO release from nervous terminations
(NANC neurons). The NO activates soluble gua-
nylate cyclase raising intracellular cyclic guano-
sine monophosphate (cGMP) and then
intracellular Ca2+, resulting in a relaxation of the
smooth muscle in the arteries and arterioles sup-
plying the erectile tissue (Fig. 3). The process is
also enhanced by concomitant acetylcholine
(Ach)-derived adenylate cyclase activation.
Increased blood influx fills the lacunar spaces of
subtunical compressive venules, resulting in
almost complete occlusion of venous outflow
(venous-occlusion phenomenon). cGMP is
hydrolyzed by cellular phosphodiesterase-5
(PDE5), restoring smooth muscle contraction
and allowing the penis to return to its flaccid
state. Detumescence is also mediated by adrener-
gic receptor reactivation, which leads to rhythmic
contractions of the cavernosal smooth muscle, a
decrease in arterial diameter and thereby induc-
ing venous outflow [118].
Testosterone plays a role in nearly every level
of the erectile function, from central genesis of
sexual desire to pelvic ganglions neurotransmis-
sion and smooth muscle and endothelial cells
function in the corpora cavernosa [119]. At the
central level testosterone promotes the release of
stimulatory neurotransmitters such as dopamine,
oxytocin, and NO, with an influence during sex-
ual development and later in puberty and adult-
hood mating behaviors. At the peripheral level,
testosterone modulates structure, function, and
innervation of trabecular smooth muscle cells,
the endothelial function of penile vessels, and
fibro-elastic properties of the corpus cavernosum
[119, 120].
ED can result from three main abnormalities:
(1) an inability to induce the erectile mechanism,
(2) a deficiency in the arterial blood supply of the
penis, and (3) insufficient maintenance of a
proper lacunar blood concentration (venous-­
occlusive dysfunction). These conditions are not
mutually exclusive, and different degrees of each
can occur in the natural history of ED. While in
the past, in the general population under 40, ED
was considered in most cases a purely psycho-
506
NITRIC OXIDE
PDE5i
Ca2+
Sequestration
5’ -GMP GC
GTP Phospholamban
PDE5 cGMP
Fig. 3  The cyclic guanosine monophosphate (cGMP)-
signaling pathway in smooth muscle cell in penile vascu-
lature. Nitric oxide activates guanylate cyclase (GC)
causing increased synthesis of cGMP. cGMP binds to and
activates cGMP-dependent protein kinase (PKG), thereby
promoting a cascade of phosphorylation events resulting
in the activation of myosin light chain phosphatase
(MLCP) that dephosphorylates the light chain of myosin
genic disorder, up to 70% of these cases can actu-
ally have organic etiologies [121]. Primary
psychogenic ED is dependent on the erectolytic
(or anti-erectile) effect of noradrenaline neuro-
transmission. Stressful or depressive events, gen-
eralized anxiety disorders, or other
psychological-impacting factors are generally
associated with this type of ED.  Organic ED
often involves or determines a psychological
component, affecting general mood, interper-
sonal relationships, and overall individual QoL
[122]. Vasculogenic is largely the most common
nonendocrine organic cause of ED, can involve
arterial inflow disorders and/or venous outflow
changes (venous-occlusion dysfunction) gener-
ally in the setting of generalized atherosclerosis
[97, 118]. Neurogenic causes derive from central
or autonomic neural damage, which converge in
V. Spallone et al.
PKG IP3
Receptor
Cytoplasmatic Ca2+
MYPT reduction
MLKC
(MLKC), thereby inducing smooth muscle relaxation.
PKG promotes the lowering of intracellular calcium
through the inhibition of inositol trisphosphate (IP3)
receptor and phosphorylation of phospholamban in sarco-
plasmic reticulum. cGMP is broken down at the catalytic
site of PDE5, PDE5Is have a high affinity for the PDE5
catalytic site, thereby blocking cGMP access to the site
and fostering cGMP accumulation
reducing NO release from terminal nerves [118].
Radical pelvic surgery (nerve-sparing or non-­
nerve sparing) is still a common form of iatro-
genic ED [118]. Calcium channel blockers
directly act on cavernosal ion channel flux, while
β-blockers and thiazide diuretics exercise an indi-
rect reduction in arterial blood flow to the inter-
nal pudendal artery. Finally, hypogonadism
represents a relevant cause of poor response to
first-line therapy of ED [97].
3.3 Multifactorial Pathogenesis
of Diabetic Erectile
Dysfunction
In diabetes many pathogenetic factors of ED may
coexist and potentiate each other. Diabetes-­related
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes 507

mechanisms and comorbidities—i.e., the compo- 3.3.1 Hypogonadotropic

nents of metabolic syndrome—can affect endo-
thelial function, micro- and macrovascular
function, peripheral nerve function, smooth mus-
cle contractility, psychological function, CNS
function, and androgen secretion (Fig.  4). The
efficacy on ED of lifestyle interventions (based on
Mediterranean diet, weight loss, and exercise) or
statin treatment [123, 124] and, at least in type 1
diabetes, intensive glycemic control [110] con-
firms the pathogenetic role of mechanisms trig-
gered by metabolic and glycemic changes.
Inflammation is gaining increasing pathogenetic
relevance also in female sexual dysfunction [124].
Hypogonadism
Hypogonadotropic hypogonadism, character-
ized by low serum testosterone, reduced sper-
matogenesis, and inappropriately low or normal
gonadotropins levels, is associated with obesity,
metabolic syndrome, and diabetes [125]. The
prevalence of hypogonadism at the onset of dia-
betes is 17.2% [83] but can reach 50% in the
later stages [125]. This relationship is bidirec-
tional and probably based on a vicious circle
which involves either central or peripheral
mechanisms. Most probably, obesity represents
the major player. The increased aromatase activ-

Cofactors/comorbidities Diabetes Depression

Obesity
Hyperglycaemia
Oxidative
Hypertension stress Micro Neuropathy
Dyslipidemia angiopathy
Psychological
Physical inactivity
Inflammation Macro distress
angiopathy
Hypogonadism
mic

Endothelial dysfunction
otor
ono

orim
Aut

↓ Penile sensaon
Sens

↓ Cavernosal vasodilaon
↑ Cavernosal vasoconstricon
↓ Smooth muscle relaxaon

Erectile dysfunction
Fig. 4  Multifactorial pathogenesis of diabetic erectile leading to endothelial dysfunction and an impaired erec-
dysfunction. Cardiovascular risk factors and other meta- tion. Moreover, diabetic complications such as atheroscle-
bolic syndrome components exacerbate the effects of rotic disease, microangiopathy and neuropathy,
hyperglycemia-driven pathways such as AGEs formation diabetes-related hypogonadism, antihypertensive drugs
and PKC activation, thus promoting oxidative stress and (β-blockers apart from nebivolol, thiazide diuretics, and
low-grade inflammation. These changes impair the NO aldosterone antagonists), and psychological factors are
bioavailability and synthesis and favor an imbalance further pathogenetic factors
between vasodilating and vasoconstrictive mechanisms,
508
ity in adipose tissue enhances testosterone con-
version into estradiol, leading to a suppression
of central GnRH production [125]. Insulin resis-
tance reduces the central secretion of GnRH and
elevates inflammatory mediators [126] and
these later may inhibit LHRH stimulated LH
release. Moreover, obesity related high leptin
concentrations play a role. Leptin has inhibitory
effects on Leydig cells, as suggested by human
models [127] as well as leptin resistance in the
hypothalamic-­ pituitary axis reducing LH
release, as suggested in animal models [128].
Further, SHBG levels are reduced in men with
obesity and/or type 2 diabetes and are nega-
tively associated with the risk of metabolic dis-
orders [129]. Conversely, hypogonadism
increases lipoprotein lipase activity, free fatty
acids uptake, and triglyceride content within
adipocytes and stimulates adipocyte prolifera-
tion and visceral adipose tissue deposition, con-
tributing to visceral obesity [125]. Low
testosterone levels are associated with an
adverse metabolic profile, promote insulin resis-
tance in men by impairing mitochondrial func-
tion [125], and may impair endothelial integrity
and promote inflammation [97].
3.4 Diagnosis of Diabetic Erectile
Dysfunction
The diagnostic pathway of ED does not differ
between diabetes and the general population
[130, 131].
3.4.1 History and Questionnaires
A detailed family, medical and sexual history, is
the primary approach, including the timing of
onset, severity, duration of ED, presence of noc-
turnal erections, and impact on the patient’s sex-
ual and general QoL [132]. A complete
medication list, including supplements, should be
V. Spallone et al.
obtained as well as information on other, poten-
tially reversible, causes (smoking, drug, or
alcohol abuse). An objective score of erectile
­
function can be achieved through self-­
administered validated questionnaires, which
provide a measure of ED severity also useful for
cardiovascular risk stratification. The short five-
question form of the International Index of
Erectile Function (IIEF-5) and the Sexual Health
Inventory for Men (SHIM) are useful tools for
both diagnosis and assessment of treatment
response [133]. Nevertheless, structured inter-
views are generally considered a more reliable
instrument than questionnaires as they tend to
build a stronger patient–physician relationship
and reduce the risk of missing relevant informa-
tion. So far, the only validated interview on ED
that has proven useful in several clinical studies
is the Structured Interview on Erectile
Dysfunction (SIEDY), composed of 13 items
exploring male sexual function [134]. A psycho-
logical assessment is recommended in patients
with psychiatric disorders or young men with
short diabetes duration. It should be taken into
account that psychogenic aspects are present in
different degrees in all patients independently of
the disease stage.
3.4.2 Physical Examination
An accurate physical examination in addition to
heart rate, BP, waist circumference, and BMI
should be focused on the general body habitus,
including the chest (body hair distribution and
presence of gynecomastia) and genitalia (testic-
ular size, structure, consistency and secondary
sexual habits), seeking for signs of hypogonad-
ism or other comorbidities [135]. The cremas-
teric reflex can also be evaluated. A normal reflex
can be elicited if the thoracolumbar erection cen-
ter is intact. Evaluation of the penis needs to
exclude Peyronie’s disease, phimosis or frenu-
lum breve, balanitis, and other diabetes-related
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
conditions affecting erectile response. In selected
high-­risk patients or patients with LUTS, a digi-
tal rectal examination and prostate-specific anti-
gen test are recommended [136]. Neurological
evaluation in patients with diabetes is appropri-
ate for frequent and early coexistence of diabetic
neuropathy: clinical diagnosis includes assess-
ment of neuropathic and autonomic symptoms
and signs using easy-to-use questionnaires and
hand-­held devices, while CARTs allow a con-
firmed diagnosis of CAN [60, 137, 138]. Vice
versa annual screening for erectile dysfunction
in men with other forms of diabetic neuropathy
is recommended by guidelines [60] and is also
advisable in any patient with type 2 diabetes or
long-­standing type 1 diabetes.
3.4.3 Laboratory Investigations
They are driven by the patient’s history and
physical examination findings and include gly-
cemic control, lipids profile, renal and hepatic
function for cardiovascular and metabolic risk
stratification [131]. Morning fasting total testos-
terone assay and SHBG to calculate free testos-
terone are always indicated in men with diabetes
and ED as well as in men with ED and signs of
hypogonadism, hypoactive sexual desire, or
incomplete response to PDE5Is [139]. If total
testosterone results <12 nmol/L (<346 ng/dL), a
second-morning measurement after at least a
week, together with serum LH and PRL levels,
is required in order to both confirm hypogonad-
ism and determine its etiology (central or
peripheral) [139].
3.4.4 Instrumental Procedures
A role for rigidity testing with RigiScan is still
useful for medical-legal cases and in investiga-
tive pharmacological studies to assess treat-
ment impact [140]. Penile duplex Doppler
ultrasound before and after an intracorporal
injection of a vasoactive drug provides infor-
509
mation on penile hemodynamics to distinguish
arterial insufficiency and venous-occlusive dys-
function from other causes of ED and improve
cardiovascular risk stratification [141].
Moreover, it is useful in patients with ED who
failed to respond to oral agents. Peak systolic
velocities after PGE-1 injection >35  cm/s are
considered normal and values <25 cm/s sugges-
tive of arterial ­insufficiency (92% of accuracy).
Cavernous venous-­ occlusive dysfunction is
demonstrated on Doppler ultrasounds by a
rapid detumescence despite normal peak sys-
tolic and end-diastolic velocities (>5  cm/s).
Vascular resistive index can be assessed as the
difference between peak systolic and end-dia-
stolic velocities divided by peak systolic veloc-
ity (normal values <0.75) [142]. Penile
arteriography is reserved to young men with
arterial trauma and abnormal duplex hemody-
namics or to identify candidates for vascular
reconstructive surgery.
Generally, most men with ED can be managed
in the primary care setting. In the general popula-
tion, ED requires further cardiovascular workup
according to the Princeton III Consensus Panel
[131], and in patients with diabetes tight control
of cardiovascular risk factors should also be
considered.
In summary, a cost-effective diagnostic
approach to patients with ED (Fig. 5) relies on
a careful medical history and physical exami-
nation. Routine laboratory testing is limited
and aimed at a proper cardiovascular and
metabolic risk stratification. Considering the
relationship between diabetes and hypogonad-
ism, serum testosterone levels, SHBG, LH,
PRL are more important steps than in men
without diabetes. The vascular investigation
with penile duplex Doppler ultrasound might
be more appropriate in patients with diabetes,
especially for those with high cardiovascular
risk.
510 V. Spallone et al.

Diagnosis Medications list and psychological assessment

Medical and sexual history
1st Step Validated self-questionnaires (e.g., IIEF-5, SHIM, SIEDY)

General habitus, genitalia, signs of hypogonadism,

Physical examination
anthropometric measures, BP, neurological examination*

Metabolic, liver and kidney markers, hormonal status

2nd Step Laboratory tests
Assess global cardiovascular risk, exclude hypogonadism

Selected Penile duplex doppler ultrasound before and after an

3rd Step Instrumental procedures
patients intracorporal injection of a vasoactive drug

Treatment
General Measures Lifestyle interventions Physical activity, diet, weight loss

Glucometabolic control Optimization of metabolic indices and comorbidities

Specific Measures
Psychosexual strategies Psychological counseling, partner involvement
1st Step
Oral treatment with PDE5i
1st line
In non-responders, switch to another PDE5i, dose escalation
or chronic daily treatment with tadalafil
Pharmacological strategies 2nd line
IUS or ICI (e.g., PGE1)
Hypogonadism
TRT in monotherapy or in combination

Selected
2nd Step Mechanical devices External vacuum devices
patients
Selected Penile prosthesis
Surgical Interventions
patients Vascular reconstructive surgery

Fig. 5  An algorithm for a diagnostic and therapeutic
approach to diabetic erectile dysfunction. A preliminary
assessment is based on medical history, validated self-­
questionnaires, and physical examination. A second step
with routine laboratory testing for a proper cardiovascular
and metabolic risk stratification. Hormonal assessment is
an important diagnostic step. The vascular investigation,
like penile duplex Doppler ultrasound, is limited to
selected patients [130, 131]. Therapeutic management is
based on lifestyle changes and the optimization of gluco-
metabolic control and comorbidities. The first-line oral
therapy with PDE5Is is the treatment of choice. In event
of inefficacy, in hypogonadal men, TRT should be started,
while in eugonadal men, escalation of the PDE5I dose, a
change of PDE5I, the use of daily dosing tadalafil could
be attempted. Psychological counseling is useful for an
underlying psychosexual dysfunction (including the part-
ner). IUS or ICI should be attempted in second-line,
whereas external vacuum device and surgery remain lim-
ited to a small number of selected individuals [139, 143–
146]. PDE5Is type 5 phosphodiesterase inhibitors, TRT
testosterone replacement therapy, IUS intraurethral
alprostadil suppositories, ICI intracavernosal injections

3.5 Treatment of Diabetic Erectile
Dysfunction
The stepwise management of ED includes gen-
eral and specific measures.
3.5.1 Lifestyle Interventions
Lifestyle modifications target ED and cardio-
vascular risk by promoting physical activity
[147], weight loss, Mediterranean diet [148],
avoidance of smoking, drugs and alcohol abuse,
and glycemic and lipidic control. The Look
AHEAD study showed that weight loss inter-
vention was mildly helpful in maintaining erec-
tile function in older overweight/obese men
with type 2 diabetes [149], while intensive gly-
cemic control was able to reduce the incidence
of ED in the DCCT/EDIC study in type 1 diabe-
tes [110].
3.5.2 Psychosexual Strategies
In patients with an evident psychological involve-
ment, as is possible with young men with new-­
onset diabetes, psychosexual counseling should
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
be encouraged besides pharmacological therapy.
The counseling should also be extended to the
partner, as it has shown better outcomes in resolv-
ing relationship conflicts and improving sexual
function [150].
3.5.3 Pharmacological Treatment:
PDE5Is
Oral PDE5Is are effective, safe, and well-­tolerated
therapies for the treatment of men with ED and
are for most men first-line treatment [143].
Sildenafil, first in class, and later vardenafil,
tadalafil, avanafil, and other available PDE5Is
have shown efficacy in a wide range of etiologies,
including cardiovascular diseases and diabetes,
with an overall success rate of up to 76% depend-
ing on study populations [143, 151]. Their com-
petitive inhibition of PDE5 increases cGMP
concentrations after NO release. Consequently,
PDE5Is cannot induce an erection without proper
sexual stimulation. An adequate time before sex-
ual intercourse is also required to allow peak
absorption of the drugs. Patients should be
instructed on optimal conditions for effective
medications function. Sildenafil and vardenafil
reach a median Cmax (maximum observed
plasma concentration) values 1 h after administra-
tion with a serum half-life of 3–5 h. Both drugs
showed better efficacy if taken with an empty
stomach because lipids in foods decrease and
delay absorption. Food impact on pharmacoki-
netic properties is less evident with tadalafil and
avanafil [152] (Table 1). A network meta-analysis
of 15 studies confirmed that PDE5Is are effective,
safe, and well tolerated in men with diabetes and
ED, found better mean differences in the domain
of erectile function of IIEF compared to placebo
for vardenafil, mirodenafil, and sildenafil, and
suggested an overall possible superiority of varde-
nafil and mirodenafil, while no significant differ-
ences between regular and on-demand regimens
of PDE5Is were observed [154]. In a previous
clinical trial of 54 men with type 2 diabetes a
6-month 20  mg daily treatment with vardenafil
was able to improve erectile function and in addi-
tion serum IL-6 levels and serum testosterone,
which returned within the eugonadal range in 13
511
hypogonadal men [107]. Moreover, the beneficial
effects of PDE5Is in patients with diabetes could
go beyond the treatment of ED with cardioprotec-
tive effects, an improvement in markers of dia-
betic cardiomyopathy, and endothelial dysfunction
[107, 155]. The side effects of PDE5Is are usually
mild and occur in less than 40% of patients like
headache, nasal congestion or facial flushing,
nausea, and mild reversible visual disturbances.
Although without a definitely proven association,
deafness and non-arteritic ischemic optic neurop-
athy are uncommon adverse events of PGE5Is use
requiring urgent medical assessment and with-
drawal [156, 157]. Some caution should be used
with concomitant antihypertensives drugs and
α-blockers, whereas co-administration with
nitrates is prohibited [158]. In the event of initial
unresponsiveness to the treatment, switching to at
least one other PDE5I class member should be
attempted as up to 50% of patients has shown to
be respondent [159].
3.5.4 Pharmacological Treatment:
Testosterone
Men with ED and hypogonadism often fail to
achieve initial clinical success with PDE5Is and
thus testosterone replacement therapy (TRT) is
indicated. Clinical trials with TRT in men with
type 2 diabetes and hypogonadism documented
an improvement in sexual function and erection,
beneficial effects on various aspects of sexual
function like sexual desire, a possible delay of
some weeks for a benefit in erectile function
[160], and a lower efficacy compared to men
without diabetes. A combination of TRT and a
PDE5I can further improve erectogenic outcomes
[161]. TRT requires monitoring of prostate-­
specific antigen, hematocrit, and cardiovascular
performances [162]. A meta-analysis of six stud-
ies in men with type 2 diabetes with/without
hypogonadism showed a moderate improvement
of sexual desire and erectile function with TRT
but emphasized the lack of information on the
long-term risks of treatment and the need for a
preliminary risk/benefit assessment [144]. An
improvement in metabolic parameters with TRT
has been also documented, mostly in patients
512
with type 2 diabetes [125], with less data in type
1 diabetes [163], but there is not definitive evi-
dence on TRT’s beneficial effects on hypogonad-
otropic hypogonadism-related metabolic
features.
3.5.5 Pharmacological Treatment:
Intracavernosal
and Intraurethral Therapy
Second-line treatment for PDE5Is-non-­responders
includes intraurethral vasoactive PGE1 (alprosta-
dil) suppositories (IUS) or intracavernosal injec-
tions (ICI) with vasoactive drugs (Table  1).
Intraurethral alprostadil absorption causes an
immediate increase in intracellular cyclic AMP,
smooth muscle cells relaxation in the corpora cav-
ernosa and a consequent erection. IUS efficacy has
been described in up to 56% of cases, with patients
with organic ED responding better, including
those with diabetes [164]. Dosages, lag time, and
common side effects are described in Table  1
[145]. ICIs use vasoactive substances self-injected
directly into the corpora cavernosa via a small
needle to obtain an immediate erection. Alprostadil
ICI is the single agent most widely used, with a
good efficacy rate and fewer systemic side effects
than other agents [145]. The erection occurs after
10 min and may be enhanced by sexual stimula-
tion. In an open-label study in patients with diabe-
tes a satisfactory erectile response was achieved
after 99% of injections without differences
between type 1 and type 2 diabetes and with a fair
safety profile [165]. Other vasoactive medications
for ICI include papaverine, phentolamine, and
atropine for both mono- or multiple combinations
and possible dose titration. Priapism is a major
problem with IUS and ICI, and patients should be
informed to require urgent medical intervention if
the condition emerges. Fibrosis, pain, bruising,
patient’s reluctance to penile injections, and costs
related drop-out are other important issues to be
considered [145] (Table 1). Training in self-injec-
tion technique as well as counseling might favor
treatment adherence. Combined therapy with ICIs
and PDE5Is can also be effective in the event of
single-­class failures [166].
V. Spallone et al.
3.5.6 Mechanical Devices
Vacuum erectile devices are long-term, inexpen-
sive non-surgical tools for patients with ED refrac-
tory to first-line therapy. They consist of a
cylindrical mechanical pump that creates negative
pressure that artificially draws blood into the penis.
Erection is maintained by an elastic band placed
around the base of the penis but application should
not exceed 30 min. Even though the efficacy rate is
reportedly high (above 70%), the long-term drop-
out rates are also high. Petechiae and hematomas
are the most common mild adverse events and can
cause discontinuation [138]. Caution is needed
when taking anticoagulants [167].
3.5.7 Surgery
Surgical interventions are indicated for patients who
are refractory to first- and second-line treatments, or
with contraindications to other treatments, or with
adverse reactions or acquired neural or vascular
penile dysfunctions after pelvic trauma. Penile
implants prostheses divide into two main types: mal-
leable and inflatable. Both are surgically inserted into
the corpora cavernosa in order to supply a physiolog-
ical erection. The surgical approach can be infra-
pubic, penoscrotal, or subcoronal [146]. Malleable
prostheses rely on the patient’s physical manipulation
for changing from a straight to a bent position.
Although easy to handle, they often can cause social
embarrassment and discomfort due to their mechani-
cal rigidity. The two-piece inflatable prosthesis con-
sists of two cylinders inserted in the corpora cavernosa
and a small scrotal pump, which drives fluid move-
ment from a small reservoir inside the patient’s lower
abdomen when an erection is desired. Although it
provides similar rigidity to a malleable device, it also
enables better flaccidity when deflated. Skin erosion,
fluid leakage, genitourinary infections, and the pros-
thesis’s mechanical failure are the most documented
complications, albeit reduced in frequency with the
latest devices [168].
Penile revascularization surgery is currently
indicated for young patients (<55  years of age)
with penile vascular Doppler ultrasound demon-
strated and isolated injuries. A microvascular
anastomosis between the inferior epigastric
Table 1  Comparison of the main oral type 5 phosphodiesterase inhibitors (PDE5Is) and intraurethral vasoactive PGE1 suppositories (IUS) or intracavernosal injections (ICI)
with vasoactive drugs [108, 143, 153]
PDE5i compound Dose range Tmaxa (h) T1/2a IC50 for Main adverse events Comment
(h) PDE5a (%)
(nmol/L)
Sildenafil 50–100 mg 0.95 3.98 3.7 Headache (19%), First in class; delayed
flushing (14%), absorption after high fat
dyspepsia (9%) meals
Tadalafil 10–20 mg 2 17.5 1.8 Headache (21%), Efficacy for up to 36 h;
5 mg daily dyspepsia (17%), back efficacy not reduced by high
pain (6%) fat meals
Vardenafil 10–20 mg 0.66 3.9 0.091 Headache (16%), Delayed absorption after
flushing (12%), nasal high fat meals
congestion (10%)
Avanafil 5–200 mg 0.75 5.1 5.2 Headache (7–8%), Highly selective; efficacy
flushing (3–5%) not reduced by high fat
meals
IUS or ICI compound Administration Dose range Time to onset Main adverse events Comment
(%)
Alprostadil IUS 125–1000 μg 10 min Pain (30%), urethral Easy and available;
bleeding (5%), priapism uncommon systemic side
(<1%) effects
Alprostadil ICI 5–20 μg 5–15 min Pain (25%), priapism Most used; uncommon
(<2%), fibrosis (<5%) systemic side effects
Papaverine + phentolamine (Bimix) ICI 30 mg/mL + 0.5 mg/ 5–10 min Pain (14%), priapism Not licensed for ED
mL (5%), fibrosis (13%) treatment; only available as
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes

galenic preparation
Papaverine + phentolamine + alprostadil ICI 30 mg/mL + 1 mg/ 5–10 min Pain (15%), priapism Not licensed for ED
(Trimix) mL + 10 μg/mL (3%), fibrosis (<5%) treatment; only available as
galenic preparation
VIP + Phentolamine ICI 25 μg + 1–2 mg Not recorded Flushing, headache, Approved in Denmark, the
priapism (0.06%) UK, and in New Zealand
a
Referred to fasting state, higher recommended dose. Tmax: time-to-maximum plasma concentration; T1/2: plasma elimination halftime; Cmax: maximum plasma concentration;
IC50: half maximal inhibitory concentration; VIP: Vasoactive intestinal polypeptide
513
514 V. Spallone et al.

artery and the penile dorsal artery is performed to inflammatory properties of GLP-1RA and DPP-4i
ensure arterial blood supply to corpora cavernosa [153] or on DPP-4i ability to promote vascular
[169]. Due to high procedure related risks and repair and endothelial function [173], while a
often limited long-term results [170] penile vas- 4-week treatment with empagliflozin increased
cular surgery has little application in diabetes. nitrergic relaxation of erectile tissues in a type 2
diabetes animal model [174]. From a clinical
3.5.8 Future Treatments standpoint, in a 12-week observational study a
Future therapies should address the need of patients combined treatment with exenatide and metfor-
nonresponders to first-line PDE5 inhibition. Low- min induced a greater improvement in total tes-
intensity extracorporeal shockwave therapy has tosterone levels and sexual function than
preliminarily shown some efficacy. Its proposed glimepiride-metformin combination in obese men
action mechanism ranges from improvement in with type 2 diabetes [175], with, as expectedly, an
cavernous hemodynamics and increased angio- increase in testosterone and sexual function being
genic factors to restoration of smooth muscle cell strongly related to weight and waist circumfer-
activities [171]. Biotherapies to regenerate cavern- ence loss, which are the known effects of exena-
ous endothelial cells and nerves and to inhibit cav- tide treatment. The two- to threefold increase in
ernous fibrosis have also been investigated using risk of genital infections associated with SGLT2i
growth factors, gene or protein therapy, and human should be taken into account for appropriate pre-
umbilical cord stem cells [167], with the latter vention and timely treatment [176].
showing beneficial effects on erectile function in a In summary, very few animal studies and
few men with severe type 2 diabetes when admin- observational clinical studies have investigated
istered locally [172]. Phase II–III studies are the impact of DPP-4i, GLP-1RA, and SGLT-2i
required for these regenerative therapies. on erectile and sexual function. The limited
In summary, a therapeutic algorithm for ED in data with GLP-1RA in different subsets of
men with diabetes (Fig. 5) is based on patient’s patients do not provide comparable results
clinical and laboratory profile. First-line oral [77]. Nevertheless, novel drugs have shown
therapy with PDE5i is the treatment of choice, promising metabolic effects and weight loss
but patients with diabetes could be non-­ might be a way, albeit indirect, to improve
responders. In this case, in hypogonadal men, gonadal and sexual function in patients with
TRT should be started, while in eugonadal men, diabetes and sexual dysfunction [77].
an escalation in the PDE5i dose, a change of
PDE5i, or the use of daily dosing tadalafil could
be attempted. Psychological counseling to 4 Ejaculation Dysfunction
improve patients’ sexual health and overall QoL in Diabetes
is a useful tool for an underlying psychosexual
dysfunction and should include the partner. More 4.1 Epidemiology of Ejaculation
invasive treatments, like IUS and ICI, should be and Orgasmic Dysfunction
chosen in second-line, whereas vacuum device in Diabetes
and surgical interventions remain often limited to
a small number of selected individuals. Premature ejaculation (PE) is defined according
to the International Society for Sexual Medicine
[177] as characterized by ejaculation that nearly
3.6 Novel Antihyperglycemic always occurs prior to or within about 1 min of
Drugs and Male Sexual vaginal penetration from the first sexual experi-
Function ence (lifelong PE), or a clinically significant
reduction in latency time, often to about 3 min or
Indirect evidence from animal models suggests a less (acquired PE); the inability to delay ejacula-
possible benefit of the latest antihyperglycemic tion in nearly all vaginal penetrations; and nega-
drugs based on anti-atherogenic and anti-­ tive personal consequences and/or the avoidance
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
of sexual intimacy [78, 177]. In real-life clinical
practice, the same definition could be applied to
other sexual stimuli and non-heterosexual set-
tings [178]. Prevalence of PE in the general pop-
ulation ranges from 8% to 30% [179]. A few
studies described an association between PE and
diabetes with a prevalence up to 40.2% [180] and
of 28.3% in the SUBITO-DE study in men with
recently diagnosed type 2 diabetes [83]. In young
men with type 1 diabetes, the prevalence of PE
was 24% and both HbA1c and glycemic excur-
sion in the hypoglycemic range were positively
related to PE severity [181]. ED and PE can co-­
occur in the same subjects in up to 50% of cases,
thus creating a vicious cycle [178] in both the
general and diabetic population with PE being
associated with an odds ratio for ED of 3.68 and
4.41, respectively [83, 180].
There is a paucity of reliable prevalence data
on delayed ejaculation (defined as a distressing
lengthening of ejaculatory latency) with rates
varying from 1% to 10% in the general popula-
tion [179, 182]. In SUBITO-DE study, delayed
ejaculation was present in 32.9% [83].
Retrograde ejaculation is defined as the expul-
sion of seminal fluid into the bladder because of
bladder neck dysfunction in the presence of oth-
erwise normal emission and expulsion [179].
Prevalence of retrograde ejaculation ranges from
0.3% to 2% in patients attending fertility clinics.
Given the main autonomic control of anterograde
ejaculation, as expected, diabetic autonomic neu-
ropathy was found to be associated with retro-
grade ejaculation in case reports [182]. True
prevalence might be higher than actually docu-
mented, as suggested by a small study where
34.6% of men with diabetes had retrograde ejac-
ulation compared to 0% of control subjects [182].
Ejaculatory dysfunction was the most common
cause of infertility among patients with diabetes
attending an infertility clinic [183].
The prevalence of decreased sexual interest or
desire in the general population ranges from 15%
to 25% until 60 years of age, after which preva-
lence sharply increases [179]. The prevalence of
orgasmic dysfunction is difficult to assess and
ranges from 1.8% to 19.4% with most values
being ≥10% [179]. In a population-based study,
515
26.1% and 37.6% of participants, aged
57–84 years, with known diabetes referred inabil-
ity to climax and a lack of interest in sex com-
pared to 15.9% and 24.8% of those without
diabetes, respectively [184]. In the SUBITO-DE
study, hypoactive sexual desire was present in
58.4% [83]. In the UroEDIC study, low sexual
desire and orgasmic dysfunction were present in
40% and 14%, respectively, and disorders of
orgasm, including retrograde ejaculation and
anejaculation closely overlapped with ED [10].
In the same cohort, men reporting only orgasmic
dysfunction presented a predominantly psycho-
genic phenotype, while men with orgasmic dys-
function and concomitant ED showed a mixed
phenotype with a physiological predominance
and more severe diabetic and metabolic altera-
tions [185].
In summary, albeit with evidence based on
only a few studies with an inhomogeneous defini-
tion of conditions, ejaculation and orgasmic dys-
function appear to be more prevalent in diabetes
than in the general population, with rates from
24% to 40% for PE, of 33% for delayed ejacula-
tion, up to 35% for retrograde ejaculation, from
25% to 58% for low sex desire/interest, and from
14% to 26% for orgasmic dysfunction. There is
close overlapping between ejaculation dysfunc-
tion, orgasmic dysfunction, and ED, suggesting
common and interactive pathogenesis and the
need for screening for the other two conditions in
patients referred for one of them. Furthermore,
ejaculation dysfunction might also impair
fertility.
4.2 Premature Ejaculation
in Diabetes
4.2.1 Pathophysiology of Premature
Ejaculation
Ejaculation depends on the integrity of the auto-
nomic nervous system and its central and periph-
eral neurotransmitters [177]. It consists of two
main phases—i.e., emission and expulsion—and
orgasm is generally associated with ejaculation
although the two processes are physiologically
different [186]. PE has been associated with
516
hypersensitivity of the glans penis, abnormali-
ties of serotonergic neurotransmission, hyper-
thyroidism, chronic prostatitis, diabetes, high
testosterone and low prolactin levels, sexual
desire disorder, and ED [177, 178]. Moreover,
sexual performance anxiety and psychological
or relationship problems often coexist with PE
as either a cause or a consequence of PE [187].
Such comorbidities should be considered as pos-
sible causes of acquired PE and worsening fac-
tors of lifelong PE for which a genetic
pathogenesis has been hypothesized with incon-
clusive evidence [177]. Regarding a PE and dia-
betes association, possible underlying
mechanisms might involve neurologic, neu-
rotransmitter, and psychologic changes without
any supporting findings apart from changes in
metabolism of NO that in male rats inhibits sem-
inal emission, probably by decreasing sympa-
thetic nervous system activity [188].
4.2.2 Management of Premature
Ejaculation
Medical and psycho-sexological history should
establish the PE diagnosis, differentiate lifelong
PE and acquired PE, and assess the erectile func-
tion and the impact on relationships and
QoL. Partner and patient self-estimation of ejacu-
latory latency are currently the preferred assess-
ment of PE as a proxy measure of objective
intravaginal ejaculatory latency time assessment,
whereas stopwatch measures have been consid-
ered disruptive of sexual spontaneity [177]. In
this context, questionnaires such as the Premature
Ejaculation Profile (PEP) or the Index of
Premature Ejaculation (IPE) are useful tools
[177]. A physical examination is highly advisable
and associated/causal diseases such as ED, thy-
roid dysfunction, or prostatitis should be assessed.
The only currently on-label pharmacotherapy
for both lifelong PE and acquired PE is oral
dapoxetine, a short-acting selective serotonin
reuptake inhibitor (SSRI). Local anesthetic (lido-
caine/prilocaine) spray represents the second
officially approved drug for this indication [189].
Off-label therapies include a combination
V. Spallone et al.
between dapoxetine and lidocaine/prilocaine,
PDE5i, and other SSRIs, after screening for
depression. Their off-label use requires patient’s
informed consent [178]. Psychological/behav-
ioral interventions are indicated to manage the
psychosocial aspects of PE, and a pharmacologi-
cal and psychological combined approach is sug-
gested [178].
4.3 Retrograde Ejaculation
in Diabetes
4.3.1 Pathophysiology of Retrograde
Ejaculation
Any factor that invalidates bladder neck contrac-
tion may lead to retrograde ejaculation. These
factors can be mechanical, from anatomical dis-
ruption during transurethral prostate resection,
pharmacological, due to α-blockers use, or neu-
rological, from impaired sympathetic control of
the internal vesical sphincter [189] as can occur
due to diabetic autonomic neuropathy in long-­
standing diabetes.
4.3.2 Management of Retrograde
Ejaculation
Retrograde ejaculation should be suspected in any
patient with aspermia or hypospermia and diag-
nosis is based on the demonstration of 10–15
sperm per high powered field in post-ejaculation
urine [190]. In the case of pharmacological form,
the offending drug should be, if possible, stopped.
Medical treatments include α-agonists such as
ephedrine, pseudoephedrine, and phenylpropa-
nolamine, and imipramine, a tricyclic antidepres-
sant with α-adrenergic action, used alone or in
combination with an α-agonist. Their efficacy has
also been demonstrated in patients with diabetes
[191]. Moreover, endourethral injection of colla-
gen type 2 showed promising results in patients
with type 1 diabetes [192]. In a patient wishing to
conceive and refractory to drugs, post-­ejaculation
bladder harvest of sperm or testicular sperm
extraction can allow for the performance of medi-
cally assisted reproductive techniques [189].
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes 517

5 Female Sexual Dysfunction 68.6% in women with type 2 diabetes, ranging

in Diabetes from 17% in Italy to 94.4% in Iran [195]. In the
UroEDIC I in sexually active women with type 1
5.1 Definition and Epidemiology diabetes, FSD was reported by 35.4%, including
of Diabetic Female Sexual decreased desire, problems with orgasm, lubrica-
Dysfunction tion, and arousal, pain, and low overall sexual
satisfaction by 25% [196].
Female sexual dysfunction (FSD) is a complex BMI was the only factor associated with FSD
condition affecting the physical and emotional in the cited meta-analysis in both types of diabe-
well-being of women of all ages and ethnicities. tes [194], while depression and marital status
According to the Diagnostic and Statistical were the only independent predictors in type 1
Manual of Mental Disorders (DSM)-5, FSD diabetes in the UroEDIC I study, with odds ratios
includes four primary diagnoses: female sexual of 2.08 and 2.49, respectively [196]. Depression
interest and arousal disorder, female orgasmic was also independently associated with the
disorder, genitopelvic pain, and penetration dis- Female Sexual Function Index (FSFI) score in
order. In contrast, the International Classification both type 1 [197] and type 2 diabetes [198]. In
of Diseases and Statistics (ICD)-10 distinguishes young women with type 1 diabetes those on insu-
between nonorganic and organic sexual dysfunc- lin pumps had a prevalence of FSD equal to con-
tions. This approach did not consider the multi- trol women and lower than women on multiple
factorial etiology of FSD and has been surpassed daily injections [197]. In the UroEDIC II study,
by the newest ICD-11 classification that includes at EDIC year 17, CAN and CARTs were inde-
both the components formerly classified in men- pendently associated with FSD (present in 41%
tal and behavioral disorders and those located of sexually active women) and predicted the sub-
among genitourinary system diseases [193]. sequent development of FSD [20], with the diag-
Regardless of several methodological discrep- nosis of CAN having an odds ratio of 1.67 for
ancies and an inconsistency in epidemiological FSD [10]. In the UroEDIC II study, among
data, the prevalence of women in the general women with FSD, 52% showed FSD in isolation
population who report at least one sexual dys- and 48% in association with one or more among
function among desire, arousal, orgasmic, and urinary incontinence, LUTS, and urinary tract
genitopelvic pain dysfunction is around 40–50%, infections [10]. This co-occurrence might docu-
irrespective of age [179]. Risk factors are aging, ment both shared pathogenesis and mutual inter-
diabetes, hypertension, genitourinary disease, action between diabetes, urinary, and sexual
psychiatric/psychological disorders, cardiovas- function also in women.
cular disease, cancer, and other chronic diseases. In summary, epidemiological data on FSD
Sociocultural factors include social isolation, are limited and of poor consistency in the gen-
financial and employment status, religious eral and diabetic population. Women with dia-
beliefs, education level, and physical activity betes are less sexually active than those without
[148]. The NSHAP study indicated that 33% of diabetes. FSD appears to be more common in
women aged 57–85  years with diabetes were women with diabetes with even doubled the
sexually active compared to 45.5% of women associated risk. Psychosocial aspects such as
without diabetes [184]. A meta-analysis of 26 depression were the main clinical correlates,
studies showed that diabetes was associated with especially in type 1 diabetes, with additional
an odds ratio for FSD of 2.27 in type 1 and 2.49 in variables as BMI mainly in type 2 diabetes. A
type 2 diabetes [194]. A more recent meta-­ predictive value for FSD was shown for CAN in
analysis showed an overall prevalence of FSD of DCCT/EDIC study.
518
5.2 Physiology
and Pathophysiology
of Female Sexual Function
The physiology of female sexual desire and
arousal involves a combination of psychological,
environmental, nervous, and hormonal factors.
As a consequence of sexual arousal and erotic
stimulation, vasodilators, such as NO and vasoin-
testinal polypeptide, are released, thank to both
parasympathetic and sympathetic nerves, leading
to vaginal vasocongestion, smooth muscle relax-
ation, and arteriolar dilation, which increases
transudation of interstitial fluid, promoting lubri-
cation [199]. Estrogens are involved in increasing
vasoprotective and vasodilatory effects through-
out the nervous system [200]. Similarly, andro-
gens play a role, despite the lack of relationships
between any FSD and having a low serum testos-
terone or androstenedione level in a large
Australian study [201].
5.3 Pathogenesis of Diabetic
Female Sexual Dysfunction
Although FSD appears to be mainly related to
social and psychological factors, like depression
[196, 198], diabetes can affect the mechanisms of
female sexual function at many levels. Animal
models documented diabetes-related changes in
contractile and relaxant capacity of vaginal mus-
culature, reduction in clitoral and vaginal blood
flow, and fibrosis development [202].
Theoretically, diabetes-related endothelial
­dysfunction might cause vascular insufficiency in
genital tissues. Higher clitoral resistance to blood
flow was associated with both metabolic syn-
drome and impaired sexual arousal [202, 203].
Although FSD is more common in metabolic
syndrome and also linked with inflammation
markers, there is no definitive evidence of an
association between cardiovascular risk factors
and FSD nor the prognostic cardiovascular value
of FSD [148, 200]. Moreover, given the neuro-
regulation of female sexual function, the expected
data of a role of diabetic neuropathy in diabetic
V. Spallone et al.
FSD were lacking until the DCCT/EDIC study
that reported an association between CAN and
FSD [10, 20]. Finally, diabetes-related modifica-
tions in steroid hormones and comorbidities,
such as polycystic ovarian syndrome and thyroid
diseases might intervene in FSD [202].
5.4 Diagnosis of Diabetic Female
Sexual Dysfunction
Assessing FSD can be very challenging given the
multiple involved factors to be examined. The
International Society for the Study of Women’s
Sexual Health has proposed a care process with
practical recommendations to screen for and
detect sexual problems in women, focusing on
patient-centered communication [193].
5.4.1 Questionnaires and Clinical
Assessment
The FSFI questionnaire, one of the most used,
includes 19 items in six domains (desire, arousal,
lubrication, orgasm, satisfaction, and pain)
addressing sexual function over 4 weeks: a total
score ≤26 identifies a risk of FSD [193]. A vali-
dated abridged FSFI version, the FSFI-6, includes
six questions and takes 3 min, with a score <19
requiring further investigations [204]. Once a
FSD has been detected, a clinical assessment
based on general medical history (including med-
ications) and a physical examination, searching
for contributing factors (infectious, inflamma-
tory, neoplastic, neurologic) and vaginal atrophy,
skin lesions, or pain, is essential [193]. A com-
plete diabetes history should be collected. The
patient’s psychological state should also be
investigated, as this is a key player in the whole
process of her sexual enjoyment [193].
5.4.2 Laboratory Testing
This may be helpful (e.g., vaginal pH) but usually
unnecessary. Testing for sexually transmitted
infections and other vaginal pathogens may be of
help in presence of pain. Other laboratory tests,
such as TSH and PRL levels, may be indicated
based on symptoms or signs [193].
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes 519

5.4.3 Instrumental Procedures
Somatosensory evoked potentials and clitoral
electromyography could be used to assess the
muscular and neural system [205]. Different
diagnostic systems evaluate female genital blood
flow, whose increase is the main marker of sexual
arousal, including indirect heat dissipation mea-
sures (vaginal thermistors, labial thermistors,
heated oxygen electrode), clitoral and vaginal
photoplethysmography, Doppler ultrasonogra-
phy, dynamic contrast and non-contrast MRI, and
pudendal arteriogram [205]. Vaginal photople-
thysmography is low cost and easy to use [205].
Doppler ultrasonography is a quick, relatively
easy, non-invasive technique that provides a real-­
time assessment of anatomy and blood flow of
genitalia [205]. Women with type 1 diabetes
showed lower clitoral peak systolic velocity and
end-diastolic velocity and higher resistance index
compared with healthy controls [206].
Furthermore, the clitoral pulsatility index was
correlated with metabolic syndrome and with a
subjective decrease in sexual arousal [203].
However, all these objective measures are rarely
performed in clinical practice and lack of valida-
tion, normative values, and correlation with sub-
jective response [205].
In summary, the preliminary assessment of a
woman with diabetes and FSD is based on vali-
dated self-questionnaires, careful medical h­ istory
and physical examination, and routine laboratory
testing and hormonal status. Instrumental proce-
dures, including several non-­validated methods,
remain very limited in clinical practice (Fig. 6).

Diagnosis Medications list and psychological assessment

Medical and sexual history
1st Step Validated self-questionnaires (e.g., FSFI-19; FSFI-6)

General habitus and genitalia, anthropometric measures and

Physical examination BP, search for infection, inflammation, neoplasm, vaginal
atrophy, lesions

2nd Step Laboratory tests Metabolic, liver and kidney markers, hormonal status

Non-validated
methods Vaginal photoplethysmography, Doppler ultrasonography,
3rd Step Instrumental procedures
Selected techniques for neuromuscular system and genital blood flow
patients

Treatment
General Measures Lifestyle interventions Physical activity, diet, weight loss

Glucometabolic control Optimization of metabolic indices and comorbidities

Specific Measures Psychosexual strategies Psychological counseling, partner involvement

Post-menopausal
women Hormone therapy

Pharmacological strategies
Non-FDA PDE5I
approved Flibanserin
Ospemifene

Fig. 6  An algorithm for a diagnostic and therapeutic
approach to female sexual dysfunction. A preliminary
assessment is based on validated self-questionnaires,
medical history, and physical examination. A second step
concerns routine laboratory testing, while instrumental
procedures have limited application in clinical practice
[193, 205]. Therapeutic management is based on lifestyle
changes and the optimization of glucometabolic control
and comorbidities. Psychological counseling is a useful
tool for an underlying psychosexual dysfunction and
should include the partner. Hormone therapy is approved
for postmenopausal women, whereas other pharmacologi-
cal options are not FDA approved [193, 207]. FSFI female
sexual function index, BP blood pressure, PDE5I type 5
phosphodiesterase inhibitors
520 V. Spallone et al.

5.5 Treatment of Diabetic Female the available studies have precluded the approval
Sexual Dysfunction of testosterone therapy for FSD. PDE5i-induced
vasodilatation might theoretically improve vagi-
Treatment of FSD in women with diabetes nal lubrication and vulvar engorgement, but
requires a multidisciplinary approach involving available results are limited and discordant: stud-
an endocrinologist, gynecologist, and psycho- ies using self-reported measures of sexual func-
therapeutic specialist. Psychological issues tioning showed mixed results, whereas major
should be identified and treated [193]. effects were reported by studies examining phys-
iological effects on genital vasocongestion [212].
5.5.1 Lifestyle Intervention The limited relevance of PDE5 vascular mecha-
Lifestyle changes, including a balanced diet, reg- nisms in FSD and low PDE5 levels in the female
ular exercise, and weight loss, should be the first reproductive system [194] are possible reasons
step for treating FSD in women with diabetes. for this. Other pharmacological options are sup-
The open MEDITA trial showed a lower decrease ported by promising data but not FDA approved.
in the FSFI score with a Mediterranean diet com- Flibanserin, a 5-HT1A agonist/5-HT2A antago-
pared to low-fat diet in women with newly diag- nist, showed an improvement in sexual desire and
nosed type 2 diabetes followed for 8 years [124]. reduction in distress associated with sexual dys-
Similarly, in the Look AHEAD trial, 1-year function in premenopausal women with hypoac-
intensive lifestyle intervention in women with tive sexual desire disorder [193]. Ospemifene, a
type 2 diabetes and FSD at baseline was associ- selective estrogen receptor modulator, was effec-
ated with greater weight loss and higher odds to tive in the treatment of vulvovaginal atrophy in
experience FSD remission than control manage- postmenopausal women with vaginal dryness
ment [208]. [193]. Additional strategies include pelvic phys-
iotherapy and counseling for sexual pain, and
5.5.2 Antihyperglycemic Treatment mindfulness and sex therapy for orgasmic dys-
Glycemic control by reducing the risk of diabetes-­ function [207].
related complications (including genitourinary In summary, therapeutic management is based
infections) is crucial [148]. Metformin is known on general measures including lifestyle changes
to provide a beneficial effect on all female sexual and optimization of glucometabolic control and
function domains, with this effect’s strength comorbidities. Psychological counseling is a use-
depending on the degree of insulin resistance ful tool for an underlying psychosexual dysfunc-
[209]. Given the positive impact of novel GLP-­ tion and should include the partner. Hormone
1RA and SGLT2i on body weight, it might be therapy is approved for postmenopausal women,
expected that these drugs would result in better whereas other pharmacological options are not
sexual outcomes, but no studies have evaluated FDA approved. A multidisciplinary approach to
this point in women and the increased risk of uro- FSD favors better communication and outcome
genital infections with SGLT2i should be taken (Fig. 6).
into account [176].

5.5.3 Pharmacological Options 6 Conclusions on Diabetic

Hormone therapy is approved for postmeno-
pausal women mainly in the form of vaginal
estrogen [193]. Although testosterone showed a
beneficial effect on sexual function [210] and sci-
entific societies suggest its evidence-based use
for postmenopausal women with hypoactive sex-
ual desire disorder with the exclusion of women
at high cardiometabolic risk [211], limitations in
Genitourinary Dysfunction
Bladder and sexual dysfunction in diabetes are
present in more than 50% of patients with diabe-
tes. They represent a significant burden on
patients’ mental health, QoL, and prognosis.
Their multifactorial pathophysiology and multi-
dimensional characteristics call for more research
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
and an integrated multidisciplinary approach.
They have been on the fringes of diabetes
research and clinical practice even more than dia-
betic neuropathy. Their clinical relevance, the
ease of screening, and the availability of effective
therapeutic options no longer justify them being
left on the margins of the diabetic field.
References
1. Frimodt-Møller C.  Diabetic cystopathy: epide-
miology and related disorders. Ann Intern Med.
1980;92(2 Pt 2):318–21.
2. Brown JS, Wessells H, Chancellor MB, Howards
SS, Stamm WE, Stapleton AE, Steers WD, Van Den
Eeden SK, McVary KT.  Urologic complications of
diabetes. Diabetes Care. 2005;28(1):177–85.
3. Kaplan SA, Te AE, Blaivas JG.  Urodynamic find-
ings in patients with diabetic cystopathy. J Urol.
1995;153(2):342–4.
4. Thüroff JW, Abrams P, Andersson KE, Artibani
W, Chapple CR, Drake MJ, Hampel C, Neisius A,
Schröder A, Tubaro A.  EAU guidelines on urinary
incontinence. Eur Urol. 2011;59(3):387–400.
5. Martin S, Vincent A, Taylor AW, Atlantis E,
Jenkins A, Januszewski A, O’Loughlin P, Wittert
G.  Lower urinary tract symptoms, depression,
anxiety and systemic inflammatory factors in
men: a population-based cohort study. PLoS One.
2015;10(10):e0137903.
6. Gacci M, Corona G, Sebastianelli A, Serni S, De
Nunzio C, Maggi M, Vignozzi L, Novara G, McVary
KT, Kaplan SA, Gravas S, Chapple C.  Male lower
urinary tract symptoms and cardiovascular events:
a systematic review and meta-analysis. Eur Urol.
2016;70(5):788–96.
7. Joseph MA, Harlow SD, Wei JT, Sarma AV, Dunn
RL, Taylor JM, James SA, Cooney KA, Doerr KM,
Montie JE, Schottenfeld D.  Risk factors for lower
urinary tract symptoms in a population-based sam-
ple of African-American men. Am J Epidemiol.
2003;157(10):906–14.
8. Liu RT, Chung MS, Lee WC, Chang SW, Huang ST,
Yang KD, Chancellor MB, Chuang YC. Prevalence
of overactive bladder and associated risk factors
in 1359 patients with type 2 diabetes. Urology.
2011;78(5):1040–5.
9. Pop-Busui R, Hotaling J, Braffett BH, Cleary PA,
Dunn RL, Martin CL, Jacobson AM, Wessells
H, Sarma AV.  DCCT/EDIC Research Group.
Cardiovascular autonomic neuropathy, erectile dys-
function and lower urinary tract symptoms in men
with type 1 diabetes: findings from the DCCT/
EDIC. J Urol. 2015;193(6):2045–51.
10. Wessells H, Braffett BH, Holt SK, Jacobson AM,
Kusek JW, Cowie C, Dunn RL, Sarma AV. DCCT/
521
EDIC Study Group. Burden of urological com-
plications in men and women with long-standing
type 1 diabetes in the diabetes control and com-
plications trial/epidemiology of diabetes interven-
tions and complications cohort. Diabetes Care.
2018;41(10):2170–7.
11. Michel MC, Mehlburger L, Schumacher H, Bressel
HU, Goepel M. Effect of diabetes on lower urinary
tract symptoms in patients with benign prostatic
hyperplasia. J Urol. 2000;163(6):1725–9.
12. Lee WC, Wu HP, Tai TY, Liu SP, Chen J, Yu
HJ. Effects of diabetes on female voiding behavior. J
Urol. 2004;172(3):989–92.
13. Yu HJ, Lee WC, Liu SP, Tai TY, Wu HP, Chen
J.  Unrecognized voiding difficulty in female
type 2 diabetic patients in the diabetes clinic: a
prospective case-control study. Diabetes Care.
2004;27(4):988–9.
14. Jackson RA, Vittinghoff E, Kanaya AM, Miles
TP, Resnick HE, Kritchevsky SB, Simonsick EM,
Brown JS, Health, Aging, and Body Composition
Study. Urinary incontinence in elderly women: find-
ings from the Health, Aging, and Body Composition
Study. Obstet Gynecol. 2004;104(2):301–7.
15. Lifford KL, Curhan GC, Hu FB, Barbieri RL,
Grodstein F.  Type 2 diabetes mellitus and risk of
developing urinary incontinence. J Am Geriatr Soc.
2005;53(11):1851–7.
16. Danforth KN, Townsend MK, Curhan GC, Resnick
NM, Grodstein F. Type 2 diabetes mellitus and risk
of stress, urge and mixed urinary incontinence. J
Urol. 2009;181(1):193–7.
17. Ebbesen MH, Hannestad YS, Midthjell K, Hunskaar
S.  Diabetes and urinary incontinence  – prevalence
data from Norway. Acta Obstet Gynecol Scand.
2007;86(10):1256–62.
18. Phelan S, Kanaya AM, Subak LL, Hogan PE,
Espeland MA, Wing RR, Burgio KL, Dilillo V,
Gorin AA, West DS, Brown JS.  Action for Health
in Diabetes (Look AHEAD) Research Group.
Prevalence and risk factors for urinary incontinence
in overweight and obese diabetic women: action for
health in diabetes (Look AHEAD) study. Diabetes
Care. 2009;32(8):1391–7.
19. Appa AA, Brown JS, Creasman J, Van Den Eeden
SK, Subak LL, Thom DH, Ferrara A, Huang
AJ. Clinical predictors and significance of postvoid
residual volume in women with diabetes. Diabetes
Res Clin Pract. 2013;101(2):164–9.
20. Hotaling JM, Sarma AV, Patel DP, Braffett BH, Cleary
PA, Feldman E, Herman WH, Martin CL, Jacobson
AM, Wessells H, Pop-Busui R.  Diabetes Control
and Complications Trial/Epidemiology of Diabetes
Interventions and Complications Research Group.
Cardiovascular autonomic neuropathy, sexual dys-
function, and urinary incontinence in women with
type 1 diabetes. Diabetes Care. 2016;39(9):1587–93.
21. Lenherr SM, Clemens JQ, Braffett BH, Dunn
RL, Cleary PA, Kim C, Herman WH, Hotaling
JM, Jacobson AM, Brown JS, Wessells H, Sarma
522
AV.  DCCT/EDIC Research Group. Glycaemic
control and risk of incident urinary incontinence
in women with Type 1 diabetes: results from the
Diabetes Control and Complications Trial and
Epidemiology of Diabetes Interventions and
Complications (DCCT/EDIC) study. Diabet Med.
2016;33(11):1528–35.
22. Wilke T, Boettger B, Berg B, Groth A, Mueller
S, Botteman M, Yu S, Fuchs A, Maywald
U. Epidemiology of urinary tract infections in type
2 diabetes mellitus patients: an analysis based on a
large sample of 456,586 German T2DM patients. J
Diabetes Complicat. 2015;29(8):1015–23.
23. Renko M, Tapanainen P, Tossavainen P, Pokka
T, Uhari M.  Meta-analysis of the significance of
asymptomatic bacteriuria in diabetes. Diabetes Care.
2011;34(1):230–5.
24. Boyko EJ, Fihn SD, Scholes D, Chen CL, Normand
EH, Yarbro P.  Diabetes and the risk of acute uri-
nary tract infection among postmenopausal women.
Diabetes Care. 2002;25(10):1778–83.
25. de Groat WC, Yoshimura N. Anatomy and physiol-
ogy of the lower urinary tract. Handb Clin Neurol.
2015;130:61–108.
26. Daneshgari F, Liu G, Birder L, Hanna-Mitchell
AT, Chacko S.  Diabetic bladder dysfunction:
­current translational knowledge. J Urol. 2009;182(6
Suppl):S18–26.
27. Esteghamati A, Rashidi A, Nikfallah A,
Yousefizadeh A.  The association between urody-
namic findings and microvascular complications in
patients with long-term type 2 diabetes but with-
out voiding symptoms. Diabetes Res Clin Pract.
2007;78(1):42–50.
28. Rapidi CA, Karandreas N, Katsifotis C, Benroubi
M, Petropoulou K, Theodorou C. A combined uro-
dynamic and electrophysiological study of diabetic
cystopathy. Neurourol Urodyn. 2006;25(1):32–8.
29. Kempler P, Amarenco G, Freeman R, Frontoni
S, Horowitz M, Stevens M, Low P, Pop-Busui R,
Tahrani AA, Tesfaye S, Várkonyi T, Ziegler D,
Valensi P.  Toronto Consensus Panel on Diabetic
Neuropathy. Management strategies for gastrointes-
tinal, erectile, bladder, and sudomotor dysfunction
in patients with diabetes. Diabetes Metab Res Rev.
2011;27(7):665–77.
30. Van Poppel H, Stessens R, Van Damme B, Carton
H, Baert L. Diabetic cystopathy: neuropathological
examination of urinary bladder biopsies. Eur Urol.
1988;15(1–2):128–31.
31. Lee WC, Wu HP, Tai TY, Yu HJ, Chiang
PH.  Investigation of urodynamic characteristics
and bladder sensory function in the early stages of
diabetic bladder dysfunction in women with type 2
diabetes. J Urol. 2009;181(1):198–203.
32. de Groat WC, Griffiths D, Yoshimura N.  Neural
control of the lower urinary tract. Compr Physiol.
2015;5(1):327–96.
33. Yoshimura N, Chancellor MB, Andersson KE, Christ
GJ. Recent advances in understanding the biology of
V. Spallone et al.
diabetes-associated bladder complications and novel
therapy. BJU Int. 2005;95(6):733–8.
34. Mossa AH, Galan A, Cammisotto PG, Velasquez
Flores M, Shamout S, Barcelona P, Saragovi HU,
Campeau L. Antagonism of proNGF or its receptor
p75NTR reverses remodelling and improves bladder
function in a mouse model of diabetic voiding dys-
function. Diabetologia. 2020;63(9):1932–46.
35. Daneshgari F, Liu G, Hanna-Mitchell AT.  Path of
translational discovery of urological complications
of obesity and diabetes. Am J Physiol Ren Physiol.
2017;312(5):F887–96.
36. de Oliveira MG, de Medeiros ML, Tavares EBG,
Mónica FZ, Antunes E.  Methylglyoxal, a reactive
glucose metabolite, induces bladder overactivity
in addition to inflammation in mice. Front Physiol.
2020;11:290.
37. Klee NS, McCarthy CG, Lewis S, McKenzie JL,
Vincent JE, Webb RC. Urothelial senescence in the
pathophysiology of diabetic bladder dysfunction-a
novel hypothesis. Front Surg. 2018;5:72.
38. Geerlings SE, Meiland R, van Lith EC, Brouwer
EC, Gaastra W, Hoepelman AI.  Adherence of type
1-­fimbriated Escherichia coli to uroepithelial cells:
more in diabetic women than in control subjects.
Diabetes Care. 2002;25(8):1405–9.
39. Yang Z, Dolber PC, Fraser MO. Diabetic urethropa-
thy compounds the effects of diabetic cystopathy. J
Urol. 2007;178(5):2213–9.
40. Cao N, Gu B, Gotoh D, Yoshimura N.  Time-­
dependent changes of urethral function in diabetes
mellitus: a review. Int Neurourol J. 2019;23(2):91–9.
41. Beshay E, Carrier S. Oxidative stress plays a role in
diabetes-induced bladder dysfunction in a rat model.
Urology. 2004;64(5):1062–7.
42. Yuan Z, Tang Z, He C, Tang W. Diabetic cystopathy:
a review. J Diabetes. 2015;7(4):442–7.
43. Harding GK, Zhanel GG, Nicolle LE, Cheang
M. Manitoba Diabetes Urinary Tract Infection Study
Group. Antimicrobial treatment in diabetic women
with asymptomatic bacteriuria. N Engl J Med.
2002;347(20):1576–83.
44. Nicolle LE, Gupta K, Bradley SF, Colgan R, DeMuri
GP, Drekonja D, Eckert LO, Geerlings SE, Köves B,
Hooton TM, Juthani-Mehta M, Knight SL, Saint
S, Schaeffer AJ, Trautner B, Wullt B, Siemieniuk
R.  Clinical practice guideline for the management
of asymptomatic bacteriuria: 2019 update by the
Infectious Diseases Society of America. Clin Infect
Dis. 2019;68(10):1611–5.
45. Tang H, Cui W, Li D, Wang T, Zhang J, Zhai S,
Song Y.  Sodium-glucose co-transporter 2 inhibi-
tors in addition to insulin therapy for management
of type 2 diabetes mellitus: a meta-analysis of ran-
domized controlled trials. Diabetes Obes Metab.
2017;19(1):142–7.
46. Donnan JR, Grandy CA, Chibrikov E, Pharm DCM,
Aubrey-Bassler K, Johnston K, Swab M, Hache J,
Curnew D, Nguyen H, Gamble JM. Dose response
of sodium glucose cotransporter-2 inhibitors in rela-
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
tion to urinary tract infections: a systematic review
and network meta-analysis of randomized controlled
trials. CMAJ Open. 2018;6(4):E594–602.
47. U.S.  Food and Drug Administration. Drug safety
communication: FDA revises labels of SGLT2
inhibitors for diabetes to include warnings about
too much acid in the blood and serious uri-
nary tract infections. https://www.fda.gov/drugs/
drug-­s afety-­a nd-­availability/fda-­r evises-­l abels-­
sglt2-­inhibitors-­diabetes-­include-­warnings-­about-­
too-­much-­acid-­blood-­and-­serious. Accessed 23 Feb
2021.
48. Yap TL, Cromwell DC, Emberton M. A systematic
review of the reliability of frequency-volume charts
in urological research and its implications for the
optimum chart duration. BJU Int. 2007;99(1):9–16.
49. van Haarst EP, Bosch JL.  The optimal duration of
frequency-volume charts related to compliance and
reliability. Neurourol Urodyn. 2014;33(3):296–301.
50. Panicker JN, Fowler CJ, Kessler TM. Lower urinary
tract dysfunction in the neurological patient: clini-
cal assessment and management. Lancet Neurol.
2015;14(7):720–32.
51. Rosier PFWM, Schaefer W, Lose G, Goldman
HB, Guralnick M, Eustice S, Dickinson T,
Hashim H.  International Continence Society Good
Urodynamic Practices and Terms 2016: urodynam-
ics, uroflowmetry, cystometry, and pressure-flow
study. Neurourol Urodyn. 2017;36(5):1243–60.
52. Groen J, Pannek J, Castro Diaz D, Del Popolo
G, Gross T, Hamid R, Karsenty G, Kessler TM,
Schneider M, ‘t Hoen L, Blok B.  Summary of
European Association of Urology (EAU) guidelines
on neuro-urology. Eur Urol. 2016;69(2):324–33.
53. Stöhrer M, Blok B, Castro-Diaz D, Chartier-Kastler
E, Del Popolo G, Kramer G, Pannek J, Radziszewski
P, Wyndaele JJ. EAU guidelines on neurogenic lower
urinary tract dysfunction. Eur Urol. 2009;56(1):81–8.
54. Dumoulin C, Adewuyi T, Booth J, Bradley C, Burgio
K, Hagen S, Hunter K, Imamura M, Morin M,
Morkved S, Thakar R, Wallace S, Williams K. Adult
conservative management. In: Abrams P, Cardozo
L, Wagg A, Wein A, editors. Incontinence: 6th
International consultation on incontinence, Tokyo,
September 2016. Bristol: International Continence
Society (ICS) and International Consultation on
Urological Diseases (ICUD); 2017. p. 1443–628.
55. Coolen RL, Groen J, Scheepe JR, Blok
BFM.  Transcutaneous electrical nerve stimulation
and percutaneous tibial nerve stimulation to treat
idiopathic nonobstructive urinary retention: a sys-
tematic review. Eur Urol Focus. 2020;7(5):1184–94.
56. Wittig L, Carlson KV, Andrews JM, Crump RT,
Baverstock RJ.  Diabetic bladder dysfunction: a
review. Urology. 2019;123:1–6.
57. Coyne KS, Thompson CL, Lai JS, Sexton CC.  An
overactive bladder symptom and health-related
quality of life short-form: validation of the OAB-q
SF. Neurourol Urodyn. 2015;34(3):255–63.
523
58. Abrams P, Avery K, Gardener N, Donovan J. ICIQ
Advisory Board. The international consultation
on incontinence modular questionnaire. J Urol.
2006;175(3 Pt 1):1063–6; www.iciq.net.
59. Greco C, Di Gennaro F, D’Amato C, Morganti R,
Corradini D, Sun A, Longo S, Lauro D, Pierangeli G,
Cortelli P, Spallone V. Validation of the Composite
Autonomic Symptom Score 31 (COMPASS 31)
for the assessment of symptoms of autonomic
neuropathy in people with diabetes. Diabet Med.
2017;34(6):834–8.
60. Pop-Busui R, Boulton AJ, Feldman EL, Bril
V, Freeman R, Malik RA, Sosenko JM, Ziegler
D.  Diabetic neuropathy: a position statement by
the American Diabetes Association. Diabetes Care.
2017;40(1):136–54.
61. Phelan S, Kanaya AM, Ma Y, Vittinghoff E, Barrett-­
Connor E, Wing R, Kusek JW, Orchard TJ, Crandall
JP, Montez MG, Brown JS.  Diabetes Prevention
Program Research Group. Long-term prevalence and
predictors of urinary incontinence among women in
the Diabetes Prevention Program Outcomes Study.
Int J Urol. 2015;22(2):206–12.
62. Breyer BN, Phelan S, Hogan PE, Rosen RC, Kitabchi
AE, Wing RR, Brown JS.  Look AHEAD Research
Group. Intensive lifestyle intervention reduces uri-
nary incontinence in overweight/obese men with
type 2 diabetes: results from the Look AHEAD trial.
J Urol. 2014;192(1):144–9.
63. Corona G, Tirabassi G, Santi D, Maseroli E, Gacci
M, Dicuio M, Sforza A, Mannucci E, Maggi
M. Sexual dysfunction in subjects treated with inhib-
itors of 5α-reductase for benign prostatic hyperpla-
sia: a comprehensive review and meta-analysis.
Andrology. 2017;5(4):671–8.
64. Han HH, Ko WJ, Yoo TK, Oh TH, Kim DY, Kwon
DD, Byun SS, Kim SI, Jung TY. Factors associated
with continuing medical therapy after transurethral
resection of prostate. Urology. 2014;84(3):675–80.
65. Schneider T, Marschall-Kehrel D, Hanisch JU,
Michel MC. Does concomitant diabetes affect treat-
ment responses in overactive bladder patients? Int J
Clin Pract. 2013;67(11):1138–43.
66. Wagg A, Nitti VW, Kelleher C, Castro-Diaz D,
Siddiqui E, Berner T.  Oral pharmacotherapy for
overactive bladder in older patients: mirabegron as
a potential alternative to antimuscarinics. Curr Med
Res Opin. 2016;32(4):621–38.
67. Finazzi-Agrò E, Petta F, Sciobica F, Pasqualetti
P, Musco S, Bove P.  Percutaneous tibial nerve
stimulation effects on detrusor overactivity incon-
tinence are not due to a placebo effect: a random-
ized, double-blind, placebo controlled trial. J Urol.
2010;184(5):2001–6.
68. Peters KM, Macdiarmid SA, Wooldridge LS,
Leong FC, Shobeiri SA, Rovner ES, Siegel SW,
Tate SB, Jarnagin BK, Rosenblatt PL, Feagins
BA.  Randomized trial of percutaneous tibial nerve
stimulation versus extended-release tolterodine:
524
results from the overactive bladder innovative ther-
apy trial. J Urol. 2009;182(3):1055–61.
69. Gaziev G, Topazio L, Iacovelli V, Asimakopoulos
A, Di Santo A, De Nunzio C, Finazzi-Agrò
E.  Percutaneous Tibial Nerve Stimulation (PTNS)
efficacy in the treatment of lower urinary tract
dysfunctions: a systematic review. BMC Urol.
2013;13:61.
70. Daniels DH, Powell CR, Braasch MR, Kreder
KJ.  Sacral neuromodulation in diabetic patients:
success and complications in the treatment
of voiding dysfunction. Neurourol Urodyn.
2010;29(4):578–81.
71. Haraway AM, Clemens JQ, He C, Stroup C, Atiemo
HO, Cameron AP.  Differences in sacral neuro-
modulation device infection rates based on pre-
operative antibiotic selection. Int Urogynecol J.
2013;24(12):2081–5.
72. Wang CC, Liao CH, Kuo HC.  Diabetes mellitus
does not affect the efficacy and safety of intravesi-
cal onabotulinumtoxinA injection in patients with
refractory detrusor overactivity. Neurourol Urodyn.
2014;33(8):1235–9.
73. Lamin E, Newman DK.  Clean intermittent
catheterization revisited. Int Urol Nephrol.
2016;48(6):931–9.
74. Pistolesi D, Barsotti M, Giannarini G, Mantovani
M, Boggi U, Kessler TM, Selli C.  Sacral neuro-
modulation enabling simultaneous pancreas and
kidney transplantation: a first case. BMJ Case Rep.
2009;2009:bcr08.2008.0823.
75. Bump RC, Voss S, Beardsworth A, Manning M,
Zhao YD, Chen W. Long-term efficacy of duloxetine
in women with stress urinary incontinence. BJU Int.
2008;102(2):214–8.
76. Kokanali MK, Doğanay M, Aksakal O, Cavkaytar
S, Topçu HO, Özer İ. Risk factors for mesh ero-
sion after vaginal sling procedures for urinary
incontinence. Eur J Obstet Gynecol Reprod Biol.
2014;177:146–50.
77. Corona G, Isidori AM, Aversa A, Bonomi M, Ferlin
A, Foresta C, La Vignera S, Maggi M, Pivonello R,
Vignozzi L, Lombardo F.  Male and female sexual
dysfunction in diabetic subjects: focus on new anti-
hyperglycemic drugs. Rev Endocr Metab Disord.
2020;21(1):57–65.
78. McCabe MP, Sharlip ID, Atalla E, Balon R, Fisher
AD, Laumann E, Lee SW, Lewis R, Segraves
RT.  Definitions of sexual dysfunctions in women
and men: a consensus statement from the Fourth
International Consultation on Sexual Medicine
2015. J Sex Med. 2016;13(2):135–43.
79. Feldman HA, Goldstein I, Hatzichristou DG, Krane
RJ, McKinlay JB.  Impotence and its medical and
psychosocial correlates: results of the Massachusetts
Male Aging Study. J Urol. 1994;151:54–61.
80. Kouidrat Y, Pizzol D, Cosco T, Thompson T,
Carnaghi M, Bertoldo A, Solmi M, Stubbs B,
Veronese N.  High prevalence of erectile dys-
function in diabetes: a systematic review and
V. Spallone et al.
meta-analysis of 145 studies. Diabet Med.
2017;34(9):1185–92.
81. Fedele D, Coscelli C, Santeusanio F, Bortolotti
A, Chatenoud L, Colli E, Landoni M, Parazzini
F. Erectile dysfunction in diabetic subjects in Italy.
Gruppo Italiano Studio Deficit Erettile nei Diabetici.
Diabetes Care. 1998;21(11):1973–7.
82. Parazzini F, Menchini Fabris F, Bortolotti A, Calabrò
A, Chatenoud L, Colli E, Landoni M, Lavezzari M,
Turchi P, Sessa A, Mirone V. Frequency and deter-
minants of erectile dysfunction in Italy. Eur Urol.
2000;37(1):43–9.
83. Corona G, Giorda CB, Cucinotta D, Guida P, Nada
E. Gruppo di studio SUBITO-DE. Sexual dysfunc-
tion at the onset of type 2 diabetes: the interplay of
depression, hormonal and cardiovascular factors. J
Sex Med. 2014;11(8):2065–73.
84. Skeldon SC, Detsky AS, Goldenberg SL, Law
MR. Erectile dysfunction and undiagnosed diabetes,
hypertension, and hypercholesterolemia. Ann Fam
Med. 2015;13(4):331–5.
85. Johannes CB, Araujo AB, Feldman HA, Derby CA,
Kleinman KP, McKinlay JB.  Incidence of erectile
dysfunction in men 40 to 69 years old: longitudinal
results from the Massachusetts male aging study. J
Urol. 2000;163(2):460–3.
86. Fedele D, Coscelli C, Cucinotta D, Forti G,
Santeusanio F, Viaggi S, Fiori G, Velonà T,
Lavezzari M. Diade Study Group. Incidence of erec-
tile dysfunction in Italian men with diabetes. J Urol.
2001;166(4):1368–71.
87. Braun M, Wassmer G, Klotz T, Reifenrath B,
Mathers M, Engelmann U. Epidemiology of erectile
dysfunction: results of the ‘Cologne Male Survey’.
Int J Impot Res. 2000;12(6):305–11.
88. Nicolosi A, Moreira ED Jr, Shirai M, Bin Mohd
Tambi MI, Glasser DB.  Epidemiology of erectile
dysfunction in four countries: cross-national study
of the prevalence and correlates of erectile dysfunc-
tion. Urology. 2003;61(1):201–6.
89. Malavige LS, Levy JC. Erectile dysfunction in dia-
betes mellitus. J Sex Med. 2009;6(5):1232–47.
90. Sarma AV, Hotaling JM, de Boer IH, Dunn RL,
Oerline MK, Singh K, Goldberg J, Jacobson A,
Braffett B, Herman WH, Pop-Busui R, Wessells
H.  DCCT/EDIC Research Group. Blood pressure,
antihypertensive medication use, and risk of erectile
dysfunction in men with type I diabetes. J Hypertens.
2019;37(5):1070–6.
91. Rosen RC, Wing RR, Schneider S, Wadden TA,
Foster GD, West DS, Kitabchi AE, Brancati FL,
Maschak-Carey BJ, Bahnson JL, Lewis CE,
Gendrano Iii IN. Erectile dysfunction in type 2 dia-
betic men: relationship to exercise fitness and car-
diovascular risk factors in the Look AHEAD trial. J
Sex Med. 2009;6(5):1414–22.
92. Liu Q, Zhang Y, Wang J, Li S, Cheng Y, Guo J, Tang
Y, Zeng H, Zhu Z. Erectile dysfunction and depres-
sion: a systematic review and meta-analysis. J Sex
Med. 2018;15(8):1073–82.
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
93. Pouwer F. Depression: a common and burdensome
complication of diabetes that warrants the continued
attention of clinicians, researchers and healthcare
policy makers. Diabetologia. 2017;60(1):30–4.
94. De Berardis G, Franciosi M, Belfiglio M, Di Nardo
B, Greenfield S, Kaplan SH, Pellegrini F, Sacco
M, Tognoni G, Valentini M, Nicolucci A.  Quality
of Care and Outcomes in Type 2 Diabetes (QuED)
Study Group. Erectile dysfunction and qual-
ity of life in type 2 diabetic patients: a serious
problem too often overlooked. Diabetes Care.
2002;25(2):284–91.
95. De Berardis G, Pellegrini F, Franciosi M, Belfiglio
M, Di Nardo B, Greenfield S, Kaplan SH, Rossi
MC, Sacco M, Tognoni G, Valentini M, Nicolucci
A.  QuED (Quality of Care and Outcomes in Type
2 Diabetes) Study Group. Longitudinal assessment
of quality of life in patients with type 2 diabetes and
self-reported erectile dysfunction. Diabetes Care.
2005;28(11):2637–43.
96. Jacobson AM, Braffett BH, Cleary PA, Dunn
RL, Larkin ME, Wessells H, Sarma AV.  DCCT/
EDIC Research Group. Relationship of urologic
complications with health-related quality of life
and perceived value of health in men and women
with type 1 diabetes: the Diabetes Control and
Complications Trial/Epidemiology of Interventions
and Complications (DCCT/EDIC) cohort. Diabetes
Care. 2015;38(10):1904–12.
97. Corona G, Rastrelli G, Isidori AM, Pivonello
R, Bettocchi C, Reisman Y, Sforza A, Maggi
M.  Erectile dysfunction and cardiovascular risk: a
review of current findings. Expert Rev Cardiovasc
Ther. 2020;18(3):155–64.
98. Besiroglu H, Otunctemur A, Ozbek E. The relation-
ship between metabolic syndrome, its components,
and erectile dysfunction: a systematic review and a
meta-analysis of observational studies. J Sex Med.
2015;12(6):1309–18.
99. Ning L, Yang L.  Hypertension might be a risk
factor for erectile dysfunction: a meta-analysis.
Andrologia. 2017;49(4)
100. Wang XM, Bai YJ, Yang YB, Li JH, Tang Y, Han
P. Alcohol intake and risk of erectile dysfunction: a
dose-response meta-analysis of observational stud-
ies. Int J Impot Res. 2018;30(6):342–51.
101. Allen MS, Walter EE.  Erectile dysfunction: an
umbrella review of meta-analyses of risk-factors,
treatment, and prevalence outcomes. J Sex Med.
2019;16(4):531–41.
102. Zhao B, Hong Z, Wei Y, Yu D, Xu J, Zhang W. Erectile
dysfunction predicts cardiovascular events as an
independent risk factor: a systematic review and
meta-analysis. J Sex Med. 2019;16(7):1005–17.
103. Vlachopoulos CV, Terentes-Printzios DG,
Ioakeimidis NK, Aznaouridis KA, Stefanadis
CI. Prediction of cardiovascular events and all-cause
mortality with erectile dysfunction: a systematic
review and meta-analysis of cohort studies. Circ
Cardiovasc Qual Outc. 2013;6(1):99–109.
525
104. Montorsi P, Ravagnani PM, Galli S, Rotatori F,
Briganti A, Salonia A, Rigatti P, Montorsi F.  The
artery size hypothesis: a macrovascular link between
erectile dysfunction and coronary artery disease. Am
J Cardiol. 2005;96(12B):19M–23M.
105. Frantzen J, Speel TG, Kiemeney LA, Meuleman
EJ.  Cardiovascular risk among men seeking
help for erectile dysfunction. Ann Epidemiol.
2006;16(2):85–90.
106. Gazzaruso C, Solerte SB, Pujia A, Coppola A,
Vezzoli M, Salvucci F, Valenti C, Giustina A,
Garzaniti A.  Erectile dysfunction as a predictor of
cardiovascular events and death in diabetic patients
with angiographically proven asymptomatic coro-
nary artery disease: a potential protective role for
statins and 5-phosphodiesterase inhibitors. J Am
Coll Cardiol. 2008;51(21):2040–4.
107. Santi D, Granata AR, Guidi A, Pignatti E, Trenti
T, Roli L, Bozic R, Zaza S, Pacchioni C, Romano
S, Nofer JR, Rochira V, Carani C, Simoni M.  Six
months of daily treatment with vardenafil improves
parameters of endothelial inflammation and of
hypogonadism in male patients with type 2 dia-
betes and erectile dysfunction: a randomized,
double-blind, prospective trial. Eur J Endocrinol.
2016;174(4):513–22.
108. Tzoumas N, Farrah TE, Dhaun N, Webb
DJ.  Established and emerging therapeutic uses of
PDE type 5 inhibitors in cardiovascular disease. Br J
Pharmacol. 2020;177(24):5467–88.
109. Pavy-Le Traon A, Fontaine S, Tap G, Guidolin B,
Senard JM, Hanaire H.  Cardiovascular autonomic
neuropathy and other complications in type 1 diabe-
tes. Clin Auton Res. 2010;20(3):153–60.
110. Wessells H, Penson DF, Cleary P, Rutledge BN,
Lachin JM, McVary KT, Schade DS, Sarma
AV.  Diabetes Control and Complications Trial/
Epidemiology of Diabetes Interventions and
Complications Research Group. Effect of intensive
glycemic therapy on erectile function in men with
type 1 diabetes. J Urol. 2011;185(5):1828–34.
111. Azmi S, Ferdousi M, Alam U, Petropoulos IN,
Ponirakis G, Marshall A, Asghar O, Fadavi H, Jones
W, Tavakoli M, Boulton AJM, Jeziorska M, Soran H,
Efron N, Malik RA. Small-fibre neuropathy in men
with type 1 diabetes and erectile dysfunction: a cross-­
sectional study. Diabetologia. 2017;60(6):1094–101.
112. Dhage S, Ho JH, Ferdousi M, Kalteniece A, Azmi S,
Adam S, Marshall A, Jeziorska M, Donn R, Soran H,
Malik RA. Small fibre pathology is associated with
erectile dysfunction in men with type 2 diabetes.
Diabetes Metab Res Rev. 2020;36(3):e3263.
113. Wellmer A, Sharief MK, Knowles CH, Misra VP,
Kopelman P, Ralph D, Anand P. Quantitative sensory
and autonomic testing in male diabetic patients with
erectile dysfunction. BJU Int. 1999;83(1):66–70.
114. Sáenz de Tejada I, Angulo J, Cellek S, González-­
Cadavid N, Heaton J, Pickard R, Simonsen
U.  Pathophysiology of erectile dysfunction. J Sex
Med. 2005;2(1):26–39.
526
115. Saenz de Tejada I, Goldstein I, Azadzoi K, Krane RJ,
Cohen RA.  Impaired neurogenic and endothelium-­
mediated relaxation of penile smooth muscle
from diabetic men with impotence. N Engl J Med.
1989;320(16):1025–30.
116. Ganz P.  Erectile dysfunction: pathophysiologic
mechanisms pointing to underlying cardiovascular
disease. Am J Cardiol. 2005;96(12B):8M–12M.
117. Andersson KE, Wagner G.  Physiology of penile
erection. Physiol Rev. 1995;75(1):191–236.
118. Dean RC, Lue TF. Physiology of penile erection and
pathophysiology of erectile dysfunction. Urol Clin
North Am. 2005;32(4):379–95.
119. Isidori AM, Buvat J, Corona G, Goldstein I, Jannini
EA, Lenzi A, Porst H, Salonia A, Traish AM, Maggi
M. A critical analysis of the role of testosterone in
erectile function: from pathophysiology to treatment-
­a systematic review. Eur Urol. 2014;65(1):99–112.
120. Sopko NA, Hannan JL, Bivalacqua TJ. Understanding
and targeting the Rho kinase pathway in erectile dys-
function. Nat Rev Urol. 2014;11(11):622–8.
121. Ludwig W, Phillips M. Organic causes of erectile dys-
function in men under 40. Urol Int. 2014;92(1):1–6.
122. Jannini EA, McCabe MP, Salonia A, Montorsi
F, Sachs BD.  Organic vs. psychogenic? The
Manichean diagnosis in sexual medicine. J Sex Med.
2010;7(5):1726–33.
123. Gupta BP, Murad MH, Clifton MM, Prokop L, Nehra
A, Kopecky SL. The effect of lifestyle modification
and cardiovascular risk factor reduction on erectile
dysfunction: a systematic review and meta-analysis.
Arch Intern Med. 2011;171(20):1797–803.
124. Maiorino MI, Bellastella G, Giugliano D, Esposito
K. From inflammation to sexual dysfunctions: a jour-
ney through diabetes, obesity, and metabolic syn-
drome. J Endocrinol Investig. 2018;41(11):1249–58.
125. Dhindsa S, Ghanim H, Batra M, Dandona
P.  Hypogonadotropic hypogonadism in men with
diabesity. Diabetes Care. 2018;41(7):1516–25.
126. Brüning JC, Gautam D, Burks DJ, Gillette J,
Schubert M, Orban PC, Klein R, Krone W, Müller-­
Wieland D, Kahn CR. Role of brain insulin receptor
in control of body weight and reproduction. Science.
2000;289(5487):2122–5.
127. Isidori AM, Caprio M, Strollo F, Moretti C, Frajese
G, Isidori A, Fabbri A.  Leptin and androgens in
male obesity: evidence for leptin contribution to
reduced androgen levels. J Clin Endocrinol Metab.
1999;84(10):3673–80.
128. Zhai L, Zhao J, Zhu Y, Liu Q, Niu W, Liu C, Wang
Y. Downregulation of leptin receptor and kisspeptin/
GPR54  in the murine hypothalamus contributes to
male hypogonadism caused by high-fat diet-induced
obesity. Endocrine. 2018;62(1):195–206.
129. Simó R, Sáez-López C, Barbosa-Desongles A,
Hernández C, Selva DM.  Novel insights in SHBG
regulation and clinical implications. Trends
Endocrinol Metab. 2015;26(7):376–83.
130. Burnett AL, Nehra A, Breau RH, Culkin DJ, Faraday
MM, Hakim LS, Heidelbaugh J, Khera M, McVary
V. Spallone et al.
KT, Miner MM, Nelson CJ, Sadeghi-Nejad H, Seftel
AD, Shindel AW. Erectile dysfunction: AUA guide-
line. J Urol. 2018;200(3):633–41.
131. Nehra A, Jackson G, Miner M, Billups KL, Burnett
AL, Buvat J, Carson CC, Cunningham GR, Ganz P,
Goldstein I, Guay AT, Hackett G, Kloner RA, Kostis
J, Montorsi P, Ramsey M, Rosen R, Sadovsky R,
Seftel AD, Shabsigh R, Vlachopoulos C, Wu FC. The
Princeton III Consensus recommendations for the
management of erectile dysfunction and cardiovas-
cular disease. Mayo Clin Proc. 2012;87(8):766–78.
132. Althof SE, Rosen RC, Perelman MA, Rubio-­
Aurioles E.  Standard operating procedures for tak-
ing a sexual history. J Sex Med. 2013;10(1):26–35.
133. Cappelleri JC, Rosen RC.  The Sexual Health
Inventory for Men (SHIM): a 5-year review of
research and clinical experience. Int J Impot Res.
2005;17(4):307–19.
134. Petrone L, Mannucci E, Corona G, Bartolini M,
Forti G, Giommi R, Maggi M. Structured interview
on erectile dysfunction (SIEDY): a new, multidi-
mensional instrument for quantification of pathoge-
netic issues on erectile dysfunction. Int J Impot Res.
2003;15(3):210–20.
135. Ghanem HM, Salonia A, Martin-Morales A.  SOP:
physical examination and laboratory testing
for men with erectile dysfunction. J Sex Med.
2013;10(1):108–10.
136. Gratzke C, Bachmann A, Descazeaud A, Drake MJ,
Madersbacher S, Mamoulakis C, Oelke M, Tikkinen
KAO, Gravas S. EAU guidelines on the assessment
of non-neurogenic male lower urinary tract symp-
toms including benign prostatic obstruction. Eur
Urol. 2015;67(6):1099–109.
137. Spallone V, Ziegler D, Freeman R, Bernardi L,
Frontoni S, Pop-Busui R, Stevens M, Kempler P,
Hilsted J, Tesfaye S, Low P, Valensi P.  Toronto
Consensus Panel on Diabetic Neuropathy.
Cardiovascular autonomic neuropathy in diabetes:
clinical impact, assessment, diagnosis, and manage-
ment. Diabetes Metab Res Rev. 2011;27(7):639–53.
138. Kamenov ZA.  A comprehensive review of erec-
tile dysfunction in men with diabetes. Exp Clin
Endocrinol Diabetes. 2015;123(3):141–58.
139. Corona G, Goulis DG, Huhtaniemi I, Zitzmann
M, Toppari J, Forti G, Vanderschueren D, Wu
FC. European Academy of Andrology (EAA) guide-
lines on investigation, treatment and monitoring
of functional hypogonadism in males: endorsing
organization: European Society of Endocrinology.
Andrology. 2020;8(5):970–87.
140. Bella AJ, Lee JC, Carrier S, Bénard F, Brock GB.
2015 CUA Practice guidelines for erectile dysfunc-
tion. Can Urol Assoc J. 2015;9(1-2):23–9.
141. Corona G, Fagioli G, Mannucci E, Romeo A,
Rossi M, Lotti F, Sforza A, Morittu S, Chiarini V,
Casella G, Di Pasquale G, Bandini E, Forti G, Maggi
M. Penile Doppler ultrasound in patients with erec-
tile dysfunction (ED): role of peak systolic velocity
measured in the flaccid state in predicting arterio-
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
genic ED and silent coronary artery disease. J Sex
Med. 2008;5(11):2623–34.
142. Jung DC, Park SY, Lee JY.  Penile Doppler
ultrasonography revisited. Ultrasonography.
2018;37(1):16–24.
143. Hatzimouratidis K, Salonia A, Adaikan G, Buvat J,
Carrier S, El-Meliegy A, McCullough A, Torres LO,
Khera M. Pharmacotherapy for erectile dysfunction:
recommendations from the Fourth International
Consultation for Sexual Medicine (ICSM 2015). J
Sex Med. 2016;13(4):465–88.
144. Algeffari M, Jayasena CN, MacKeith P, Thapar
A, Dhillo WS, Oliver N.  Testosterone therapy
for sexual dysfunction in men with Type 2 dia-
betes: a systematic review and meta-analysis
of randomized controlled trials. Diabet Med.
2018;35(2):195–202.
145. Duncan C, Omran GJ, Teh J, Davis NF, Bolton DM,
Lawrentschuk N.  Erectile dysfunction: a global
review of intracavernosal injectables. World J Urol.
2019;37(6):1007–14.
146. Levine LA, Becher EF, Bella AJ, Brant WO,
Kohler TS, Martinez-Salamanca JI, Trost L, Morey
AF. Penile prosthesis surgery: current recommenda-
tions from the international consultation on sexual
medicine. J Sex Med. 2016;13(4):489–518.
147. Silva AB, Sousa N, Azevedo LF, Martins C. Physical
activity and exercise for erectile dysfunction: sys-
tematic review and meta-analysis. Br J Sports Med.
2017;51(19):1419–24.
148. Maiorino MI, Bellastella G, Esposito K.  Diabetes
and sexual dysfunction: current perspectives.
Diabetes Metab Syndr Obes. 2014;7:95–105.
149. Look AHEAD Research Group, Wing RR.  Long-­
term effects of a lifestyle intervention on weight
and cardiovascular risk factors in individuals
with type 2 diabetes mellitus: four-year results
of the Look AHEAD trial. Arch Intern Med.
2010;170(17):1566–75.
150. Boddi V, Castellini G, Casale H, Rastrelli G, Boni L,
Corona G, Maggi M.  An integrated approach with
vardenafil orodispersible tablet and cognitive behav-
ioral sex therapy for treatment of erectile dysfunc-
tion: a randomized controlled pilot study. Andrology.
2015;3(5):909–18.
151. Chen L, Staubli SE, Schneider MP, Kessels AG, Ivic
S, Bachmann LM, Kessler TM.  Phosphodiesterase
5 inhibitors for the treatment of erectile dysfunc-
tion: a trade-off network meta-analysis. Eur Urol.
2015;68(4):674–80.
152. Zucchi A, Costantini E, Scroppo FI, Silvani M,
Kopa Z, Illiano E, Petrillo MG, Cari L, Nocentini
G.  The first-generation phosphodiesterase 5 inhibi-
tors and their pharmacokinetic issue. Andrology.
2019;7(6):804–17.
153. Hirano T, Mori Y.  Anti-atherogenic and anti-­
inflammatory properties of glucagon-like peptide-1,
glucose-dependent insulinotropic polypeptide, and
dipeptidyl peptidase-4 inhibitors in experimental
animals. J Diabetes Investig. 2016;7(Suppl 1):80–6.
527
154. Liao X, Qiu S, Bao Y, Wang W, Yang L, Wei
Q.  Comparative efficacy and safety of phosphodi-
esterase type 5 inhibitors for erectile dysfunction
in diabetic men: a Bayesian network meta-­analysis
of randomized controlled trials. World J Urol.
2019;37(6):1061–74.
155. Giannetta E, Feola T, Gianfrilli D, Pofi R, Dall’Armi
V, Badagliacca R, Barbagallo F, Lenzi A, Isidori
AM. Is chronic inhibition of phosphodiesterase type
5 cardioprotective and safe? A meta-analysis of ran-
domized controlled trials. BMC Med. 2014;12:185.
156. Liu B, Zhu L, Zhong J, Zeng G, Deng T. The asso-
ciation between phosphodiesterase type 5 inhibitor
use and risk of non-arteritic anterior ischemic optic
neuropathy: a systematic review and meta-analysis.
Sex Med. 2018;6(3):185–92.
157. Manna S, Gray ML, Kaul VF, Wanna
G.  Phosphodiesterase-5 (PDE-5) Inhibitors and
ototoxicity: a systematic review. Otol Neurotol.
2019;40(3):276–83.
158. Giuliano F, Jackson G, Montorsi F, Martin-Morales
A, Raillard P. Safety of sildenafil citrate: review of
67 double-blind placebo-controlled trials and the
postmarketing safety database. Int J Clin Pract.
2010;64(2):240–55.
159. Lee M, Sharifi R.  Non-invasive management
options for erectile dysfunction when a phospho-
diesterase type 5 inhibitor fails. Drugs Aging.
2018;35(3):175–87.
160. Hackett G, Cole N, Saghir A, Jones P, Strange RC,
Ramachandran S.  Testosterone replacement ther-
apy: improved sexual desire and erectile function
in men with type 2 diabetes following a 30-week
randomized placebo-controlled study. Andrology.
2017;5(5):905–13.
161. Spitzer M, Basaria S, Travison TG, Davda MN,
Paley A, Cohen B, Mazer NA, Knapp PE, Hanka S,
Lakshman KM, Ulloor J, Zhang A, Orwoll K, Eder
R, Collins L, Mohammed N, Rosen RC, DeRogatis
L, Bhasin S.  Effect of testosterone replacement on
response to sildenafil citrate in men with erectile
dysfunction: a parallel, randomized trial. Ann Intern
Med. 2012;157(10):681–91.
162. Bhasin S, Brito JP, Cunningham GR, Hayes FJ,
Hodis HN, Matsumoto AM, Snyder PJ, Swerdloff
RS, Wu FC, Yialamas MA.  Testosterone therapy
in men with hypogonadism: an endocrine society
clinical practice guideline. J Clin Endocrinol Metab.
2018;103(5):1715–44.
163. Saad F, Yassin A, Almehmadi Y, Doros G, Gooren
L.  Effects of long-term testosterone replacement
therapy, with a temporary intermission, on glycemic
control of nine hypogonadal men with type 1 diabe-
tes mellitus  - a series of case reports. Aging Male.
2015;18(3):164–8.
164. Padma-Nathan H, Hellstrom WJ, Kaiser FE, Labasky
RF, Lue TF, Nolten WE, Norwood PC, Peterson
CA, Shabsigh R, Tam PY, Place VA, Gesundheit
N. Treatment of men with erectile dysfunction with
transurethral alprostadil. Medicated Urethral System
528
for Erection (MUSE) Study Group. N Engl J Med.
1997;336(1):1–7.
165. Heaton JP, Lording D, Liu SN, Litonjua AD,
Guangwei L, Kim SC, Kim JJ, Zhi-Zhou S, Israr D,
Niazi D, Rajatanavin R, Suyono S, Benard F, Casey
R, Brock G, Belanger A. Intracavernosal alprostadil
is effective for the treatment of erectile dysfunction
in diabetic men. Int J Impot Res. 2001;13(6):317–21.
166. Moncada I, Martinez-Salamanca J, Ruiz-Castañe E,
Romero J. Combination therapy for erectile dysfunc-
tion involving a PDE5 inhibitor and alprostadil. Int J
Impot Res. 2018;30(5):203–8.
167. Kim S, Cho MC, Cho SY, Chung H, Rajasekaran
MR. Novel emerging therapies for erectile dysfunc-
tion. World J Mens Health. 2021;39(1):48–64.
168. Scherzer ND, Dick B, Gabrielson AT, Alzweri LM,
Hellstrom WJG.  Penile prosthesis complications:
planning, prevention, and decision making. Sex Med
Rev. 2019;7(2):349–59.
169. Dicks B, Bastuba M, Goldstein I.  Penile revascu-
larization--contemporary update. Asian J Androl.
2013;15(1):5–9.
170. Hsieh CH, Hsu GL, Chang SJ, Yang SS, Liu SP,
Hsieh JT. Surgical niche for the treatment of erectile
dysfunction. Int J Urol. 2020;27(2):117–33.
171. Dong L, Chang D, Zhang X, Li J, Yang F, Tan K,
Yang Y, Yong S, Yu X. Effect of low-intensity extra-
corporeal shock wave on the treatment of erectile
dysfunction: a systematic review and meta-analysis.
Am J Mens Health. 2019;13(2):1557988319846749.
172. Bahk JY, Jung JH, Han H, Min SK, Lee
YS.  Treatment of diabetic impotence with umbili-
cal cord blood stem cell intracavernosal transplant:
preliminary report of 7 cases. Exp Clin Transplant.
2010;8(2):150–60.
173. Huang CY, Shih CM, Tsao NW, Lin YW, Huang PH,
Wu SC, Lee AW, Kao YT, Chang NC, Nakagami
H, Morishita R, Ou KL, Hou WC, Lin CY, Shyu
KG, Lin FY.  Dipeptidyl peptidase-4 inhibitor
improves neovascularization by increasing circu-
lating endothelial progenitor cells. Br J Pharmacol.
2012;167(7):1506–19.
174. Assaly R, Gorny D, Compagnie S, Mayoux E,
Bernabe J, Alexandre L, Giuliano F, Behr-Roussel
D. The favorable effect of empagliflozin on erectile
function in an experimental model of type 2 diabe-
tes. J Sex Med. 2018;15(9):1224–34.
175. Shao N, Yu XY, Yu YM, Li BW, Pan J, Wu WH,
Zhang HJ, Ma XF, Hao M, Kuang HY.  Short-term
combined treatment with exenatide and metformin
is superior to glimepiride combined metformin
in improvement of serum testosterone levels in
type 2 diabetic patients with obesity. Andrologia.
2018;50(7):e13039.
176. Caparrotta TM, Greenhalgh AM, Osinski K, Gifford
RM, Moser S, Wild SH, Reynolds RM, Webb DJ,
Colhoun HM.  Sodium-glucose co-transporter 2
inhibitors (SGLT2i) exposure and outcomes in type
2 diabetes: a systematic review of population-based
observational studies. Diabetes Ther. 2021;12:991.
V. Spallone et al.
177. Althof SE, McMahon CG, Waldinger MD, Serefoglu
EC, Shindel AW, Adaikan PG, Becher E, Dean
J, Giuliano F, Hellstrom WJ, Giraldi A, Glina S,
Incrocci L, Jannini E, McCabe M, Parish S, Rowland
D, Segraves RT, Sharlip I, Torres LO. An update of
the International Society of Sexual Medicine’s guide-
lines for the diagnosis and treatment of premature
ejaculation (PE). J Sex Med. 2014;11(6):1392–422.
178. Sansone A, Aversa A, Corona G, Fisher AD, Isidori
AM, La Vignera S, Limoncin E, Maggi M, Merico
M, Jannini EA. Management of premature ejacula-
tion: a clinical guideline from the Italian Society
of Andrology and Sexual Medicine (SIAMS). J
Endocrinol Investig. 2020;44:1103.
179. McCabe MP, Sharlip ID, Lewis R, Atalla E, Balon
R, Fisher AD, Laumann E, Lee SW, Segraves
RT.  Incidence and prevalence of sexual dysfunc-
tion in women and men: a consensus statement from
the Fourth International Consultation on Sexual
Medicine 2015. J Sex Med. 2016;13(2):144–52.
180. Malavige LS, Jayaratne SD, Kathriarachchi ST,
Sivayogan S, Fernando DJ, Levy JC.  Erectile dys-
function among men with diabetes is strongly
associated with premature ejaculation and reduced
libido. J Sex Med. 2008;5(9):2125–34.
181. Bellastella G, Maiorino MI, Olita L, Della Volpe E,
Giugliano D, Esposito K.  Premature ejaculation is
associated with glycemic control in Type 1 diabetes.
J Sex Med. 2015;12(1):93–9.
182. Di Sante S, Mollaioli D, Gravina GL, Ciocca
G, Limoncin E, Carosa E, Lenzi A, Jannini
EA.  Epidemiology of delayed ejaculation. Transl
Androl Urol. 2016;5(4):541–8.
183. Sexton WJ, Jarow JP.  Effect of diabetes melli-
tus upon male reproductive function. Urology.
1997;49(4):508–13.
184. Lindau ST, Tang H, Gomero A, Vable A, Huang ES,
Drum ML, Qato DM, Chin MH.  Sexuality among
middle-aged and older adults with diagnosed and
undiagnosed diabetes: a national, population-based
study. Diabetes Care. 2010;33(10):2202–10.
185. Agochukwu-Mmonu N, Malaeb BS, Hotaling JM,
Braffett BH, Holt SK, Dunn RL, Palmer MR, Martin
CL, Jacobson AM, Herman WH, Wessells H. Sarma
AV; and the DCCT/EDIC Research Group. Risk fac-
tors for orgasmic and concomitant erectile dysfunc-
tion in men with type 1 diabetes: a cross-sectional
study. Int J Impot Res. 2021;33(1):59–66.
186. Alwaal A, Breyer BN, Lue TF. Normal male sexual
function: emphasis on orgasm and ejaculation. Fertil
Steril. 2015;104(5):1051–60.
187. Kempeneers P, Andrianne R, Cuddy M, Blairy
S.  Sexual cognitions, trait anxiety, sexual anxi-
ety, and distress in men with different subtypes of
premature ejaculation and in their partners. J Sex
Marital Ther. 2018;44(4):319–32.
188. Hull EM, Lumley LA, Matuszewich L, Dominguez
J, Moses J, Lorrain DS. The roles of nitric oxide in
sexual function of male rats. Neuropharmacology.
1994;33(11):1499–504.
Diabetic Neuropathy: Clinical Management—Genitourinary Dysfunction in Diabetes
189. Gray M, Zillioux J, Khourdaji I, Smith
RP. Contemporary management of ejaculatory dys-
function. Transl Androl Urol. 2018;7(4):686–702.
190. Althof SE, McMahon CG.  Contemporary manage-
ment of disorders of male orgasm and ejaculation.
Urology. 2016;93:9–21.
191. Arafa M, El Tabie O. Medical treatment of retrograde
ejaculation in diabetic patients: a hope for spontane-
ous pregnancy. J Sex Med. 2008;5(1):194–8.
192. Kurbatov D, Russo GI, Galstyan GR, Rozhivanov
R, Lepetukhin A, Dubsky S, Shwartz YG, Cimino
S, Morgia G, Sansalone S. Correction of retrograde
ejaculation in patients with diabetes mellitus using
endourethral collagen injection: preliminary results.
J Sex Med. 2015;12(11):2126–9.
193. Parish SJ, Hahn SR, Goldstein SW, Giraldi A,
Kingsberg SA, Larkin L, Minkin MJ, Brown V,
Christiansen K, Hartzell-Cushanick R, Kelly-­
Jones A, Rullo J, Sadovsky R, Faubion SS.  The
International Society for the Study of women’s
sexual health process of care for the identification of
sexual concerns and problems in women. Mayo Clin
Proc. 2019;94(5):842–56.
194. Pontiroli AE, Cortelazzi D, Morabito A.  Female
sexual dysfunction and diabetes: a systematic review
and meta-analysis. J Sex Med. 2013;10(4):1044–51.
195. Rahmanian E, Salari N, Mohammadi M, Jalali
R.  Evaluation of sexual dysfunction and female
sexual dysfunction indicators in women with type
2 diabetes: a systematic review and meta-analysis.
Diabetol Metab Syndr. 2019;11:73.
196. Enzlin P, Rosen R, Wiegel M, Brown J, Wessells
H, Gatcomb P, Rutledge B, Chan KL, Cleary
PA. DCCT/EDIC Research Group. Sexual dysfunc-
tion in women with type 1 diabetes: long-term find-
ings from the DCCT/EDIC study cohort. Diabetes
Care. 2009;32(5):780–5.
197. Maiorino MI, Bellastella G, Castaldo F, Petrizzo M,
Giugliano D, Esposito K. Sexual function in young
women with type 1 diabetes: the METRO study. J
Endocrinol Investig. 2017;40(2):169–77.
198. Bąk E, Marcisz C, Krzemińska S, Dobrzyn-Matusiak
D, Foltyn A, Drosdzol-Cop A.  Relationships of
sexual dysfunction with depression and accep-
tance of illness in women and men with type 2
diabetes mellitus. Int J Environ Res Public Health.
2017;14(9):1073.
199. Basson R.  Clinical practice. Sexual desire and
arousal disorders in women. N Engl J Med.
2006;354(14):1497–506.
200. Maseroli E, Scavello I, Vignozzi L. Cardiometabolic
risk and female sexuality-Part II.  Understanding
(and overcoming) gender differences: the key role
of an adequate methodological approach. Sex Med
Rev. 2018;6(4):525–34.
201. Davis SR, Davison SL, Donath S, Bell RJ. Circulating
androgen levels and self-reported sexual function in
women. JAMA. 2005;294(1):91–6.
529
202. Barbagallo F, Mongioì LM, Cannarella R, La
Vignera S, Condorelli RA, Calogero AE.  Sexual
dysfunction in diabetic women: an update on current
knowledge. Diabetology. 2020;1(1):11–21.
203. Maseroli E, Fanni E, Cipriani S, Scavello I,
Pampaloni F, Battaglia C, Fambrini M, Mannucci E,
Jannini EA, Maggi M, Vignozzi L. Cardiometabolic
risk and female sexuality: focus on clitoral vascular
resistance. J Sex Med. 2016;13(11):1651–61.
204. Isidori AM, Pozza C, Esposito K, Giugliano D,
Morano S, Vignozzi L, Corona G, Lenzi A, Jannini
EA.  Development and validation of a 6-item ver-
sion of the female sexual function index (FSFI) as a
diagnostic tool for female sexual dysfunction. J Sex
Med. 2010;7(3):1139–46.
205. Woodard TL, Diamond MP.  Physiologic measures
of sexual function in women: a review. Fertil Steril.
2009;92(1):19–34.
206. Caruso S, Rugolo S, Mirabella D, Intelisano G, Di
Mari L, Cianci A. Changes in clitoral blood flow in
premenopausal women affected by type 1 diabetes
after single 100-mg administration of sildenafil.
Urology. 2006;68(1):161–5.
207. Krakowsky Y, Grober ED.  A practical guide to
female sexual dysfunction: an evidence-based
review for physicians in Canada. Can Urol Assoc J.
2018;12(6):211–6.
208. Wing RR, Bond DS, Gendrano IN III, Wadden
T, Bahnson J, Lewis CE, Brancati F, Schneider S,
Kitabchi AE, Van Dorsten B, Rosen RC.  Sexual
Dysfunction Subgroup of the Look AHEAD
Research Group. Effect of intensive lifestyle inter-
vention on sexual dysfunction in women with type
2 diabetes: results from an ancillary Look AHEAD
study. Diabetes Care. 2013;36(10):2937–44.
209. Krysiak R, Drosdzol-Cop A, Skrzypulec-Plinta
V, Okopień B.  Sexual functioning and depressive
symptoms in women with diabetes and prediabetes
receiving metformin therapy: a pilot study. Exp Clin
Endocrinol Diabetes. 2017;125(1):42–8.
210. Islam RM, Bell RJ, Green S, Page MJ, Davis
SR. Safety and efficacy of testosterone for women: a
systematic review and meta-analysis of randomised
controlled trial data. Lancet Diabetes Endocrinol.
2019;7(10):754–66.
211. Davis SR, Baber R, Panay N, Bitzer J, Cerdas
Perez S, Islam RM, Kaunitz AM, Kingsberg SA,
Lambrinoudaki I, Liu J, Parish SJ, Pinkerton
J, Rymer J, Simon JA, Vignozzi L, Wierman
ME.  Global consensus position statement on the
use of testosterone therapy for women. Climacteric.
2019;22(5):429–34.
212. Chivers ML, Rosen RC.  Phosphodiesterase type
5 inhibitors and female sexual response: faulty
protocols or paradigms? J Sex Med. 2010;7(2 Pt
2):858–72.
 Index

A B
Abnormal mitochondrial morphology, 186 Baroreflex sensitivity testing, 213
Acceptance and commitment therapy (ACT), 231, 448 BDNF-mediated spinal disinhibition, 131
Actinobacteria, 405 BDNF, see Brain-derived neurotrophic factor (BDNF)
Action to control cardiovascular risk in diabetes Behavioural therapy (BT), 231
(ACCORD), 244 Benfotiamine, 320, 321
Adrenergic receptors, 368 β common receptor (βcR) subunits, 140
Advanced glycation end-products (AGEs), 258, 286, β-hydroxybutyrate (β-OHB), 402
288, 313, 353, 389, 406 Bioenergetics, 280, 285, 286
Age-related muscle weakness, 184 “Black-box” approach, 212
Age-related sarcopenia, 188 Bladder dysfunction, 28, 491–493
Aldose reductase inhibitors (ARIs), 254, 337 diagnosis, 496, 498
Allodynia, 74 epidemiology, 492, 493
Alpha lipoic acid (ALA), 319, 444 pathophysiology, 493–494, 496
Alzheimer’s disease, 136, 381, 383 treatment, 499–501
Ambulatory BP monitoring (ABPM), 212 Blood oxygen level dependent, 431
American Academy of Neurology (AAN), 95 Brachial plexus involvement, 171
American Academy of Physical Medicine and Brain-derived neurotrophic factor (BDNF), 130, 132,
Rehabilitation (AAPM&R), 95 402, 407, 420
American Association of Electrodiagnostic Medicine Bypass angioplasty revascularization investigation in
(AAEM), 95 type 2 diabetes (BARI2D) trial, 244
American Diabetes Association (ADA), 97, 230
American Urological Association Symptom Index
(AUASI), 210 C
AMP-activated protein kinase (AMPK), 317 Calcitonin gene-related peptide (CGRP), 356
Amputation, 70 Calcium homeostasis, 132
Angiotensin receptor blockers, 264 Cardiac sympathetic imaging, 213
Angiotensin-converting enzyme (ACE), 243, 264 Cardiovascular autonomic neuropathy (CAN), 5, 21
Anterior cingulate cortex, 430 clinical care, 210, 212
Anti-inflammatory properties, 266 clinical impact and implications, 208–210
Antioxidant response element (ARE), 312 diagnosis, 210, 212, 213
Anti-viral agent, 177 diagnostic algorithm for, 211
Atypical diabetic neuropathies, 184–185 epidemiology, 203–205
Autonomic dysfunction, 4 pathophysiology, 205–207
Autonomic neuropathy, 4, 5 treatment, 213
cardiovascular autonomic neuropathy, 44 Central nervous system, 428
gastrointestinal autonomic neuropathy, 44 Charcot-Marie-Tooth disease, 244, 245
sudomotor dysfunction, 45 Chronic hyperglycemia, 185
urogenital autonomic neuropathy, 45 Clinical medicine, 4
Axo-glial interaction, 293 Clinical sensory function, 74
Axon regeneration, 339 CNTF depletion, 134
Axon terminals, 327, 341 Coenzyme Q (CoQ), 308
Axotomy, 339 Cognitive dysfunction, 381, 382

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature 531
Switzerland AG 2023
S. Tesfaye et al. (eds.), Diabetic Neuropathy, Contemporary Diabetes,
https://doi.org/10.1007/978-3-031-15613-7
532 Index

implications, 389, 391 Diabetic distal symmetric polyneuropathy (diabetic DSP)

structural brain changes, 384–386, 388 asymptomatic clinical presentation, 71, 72
type 1 diabetes, 383 clinical manifestations, 70
type 2 diabetes, 383, 384 clinical risk factors, 69
COMBO‐DPN study, 224 complications, 21
The Committee for Medicinal Products for Human Use composite symptom scale, sign scales and combined
(CHMP) of the European Medicines Agency scales, 75
(EMA), 120 diagnostic considerations, 68–69
Common diabetic cranial mononeuropathy is oculomotor differential diagnosis, 86, 88
(CNIII) mononeuropathy, 175 epidemiological studies of, 7
Compound muscle action potentials (CMAPs), 94, 177 foot complications, 69
Concurrent validity, 72 impaired protective sensation, 71
Coronary artery risk development in young adults incidence of, 19, 20
(CARDIA) study, 214 Michigan Neuropathy Screening Instrument, 76, 77
C-peptide, 136, 137 modified TCNS, 79
Cranial mononeuropathies, 175 muscle strength in, 188, 189
Cranial neuropathies, 28 neuropathic signs, 74, 75
Cyclooxygenase-2 pathway, 264 neuropathic symptoms, 73, 74
Cystopathy, 491 neuropathy symptom score, 82, 83
NIS-LL, 79
painful diabetic neuropathy, 16, 19
D prevalence of, 7, 15, 17–18
Dementia, 381, 382 revised neuropathy disability score, 83
Depression, 221 risk factors for, 20
Detection of ischemia in asymptomatic diabetics staging of, 68
(DIAD), 209 Toronto clinical neuropathy score, 77–79
DFNS protocol in type 1 and 2 diabetes, 8–14
clustering approach, 118, 120 Utah early neuropathy scale, 80, 82
of quantitative sensory testing, 114 Diabetic erectile dysfunction, 502–509, 511, 512, 514
sensory profiling, 117, 118, 122, 123 Diabetic lumbosacral radiculoplexus neuropathy
sensory testing, 121 (DLRPN), 168
thermal thresholds, 115, 116 clinical presentation, 167
Z-transformation and reference values, 116, 117 diagnosis, 169
Diabetes, 279, 282, 287, 290, 351 epidemiology, 167
Diabetes control and complications trial (DCCT), 253, historical perspective, 166
318 management, 169
Diabetes distress, 221 outcome and prognosis, 170
Diabetes mellitus, 174, 327 pathophysiology, 168, 169
Diabetes neuropathies Diabetic mononeuropathies, 28, 165
acute painful-distal sensorimotor polyneuropathies, Diabetic myopathy, 186
42 Diabetic neurons, 337
classification, 38, 39 Diabetic neuropathy, 4, 166, 354, 416, 444, 494
co-morbidities of, 41 benfotiamine, 320
diabetic peripheral neuropathy, 39, 40 central mechanisms, 419–421
differential diagnosis of, 42 chaperones, 314
epidemiology, 37, 38 clinical manifestations of, 165
hyperglycaemia induced acute painful distal early history of, 3
sensorimotor polyneuropathy, 42 genomics of
painful diabetic peripheral neuropathy, 40, 41 with Charcot-Marie-Tooth, 244, 245
treatment, 43 in human, 240, 243
Diabetes prevention program (DPP), 319 neuropathic pain in, 246
Diabetic autonomic neuropathy (DAN), 453 sodium channel variants and painful diabetic
cardiovascular autonomic neuropathy, 21 neuropathy, 246–248
incidence of, 27 transcriptional profiling, 245
prevalence of, 22–26 in type 2 diabetes patients, 241–242
risk factors for, 28 incidence of, 5
prevalence of, 25 mGluRs, 314
Diabetic cervical radiculoplexus neuropathy (DCRPN), mitochondrial dysfunction, 308–311
170 neuropathic pain, 415
Diabetic cranial neuropathies, 165 NF-kappaB, 312, 313
Index 533

nitrosative stress, 313, 318 E

oxidative stress, 310, 312 Electrochemical sweat conductance (ESC), 104
peripheral mechanisms, 416–419 Electromyography, 188
PGC-1α, 309 Endothelial dysfunction, 419
risk factors, 415 Endothelial nitric oxide synthase (eNOS), 354
SIRT1, 309 Endothelium, 353
TFAM, 310 Endothelium dysfunction, 355
Diabetic peripheral neuropathy (DPN), 51, 255, 397 Endothelium-derived hyperpolarizing factor (EDHF),
clinical features of, 53–55 261
clinical trials in, 58, 60–62 Entrapment neuropathy, 172
corneal confocal microscopy, 104–106 Epalrestat, 257
dietary interventions, 399–405 Epidemiology of Diabetes Interventions and
exercise and physical activity, 406, 407 Complications (EDIC) study, 203
high fat diet, 398, 399 Epidermal innervation, 99
microbiome, 405, 406 Epineurial arterioles, 259
nerve conduction studies, 94–97 Epo receptor (EpoR) homodimer, 140
obesity, 398 Erectile dysfunction (ED), 28, 210
pharmacologic treatment, 397, 398 Erythropoietin (Epo), 140
quantitative sensory testing, 101–103 Estimated glomerular filtration rate (eGFR), 209
risk factors for, 55, 57 Eubacterium rectale, 405
skin biopsy for IENFD measurement, 98–104 European Insulin-Dependent Diabetes Mellitus
sudomotor testing, 103–105 Prospective Complications Study
Diabetic polyneuropathy (DPN), 94, 327 (EURODIAB IDDM), 203–204
Diabetic radiculoplexus neuropathy (DRPN), 29, 165, Exercise training, 195
166
Diabetic thoracic radiculoneuropathy (DTRN), 171, 172
Diabetic truncal mononeuropathy, 171 F
Diagnostic accuracy, 78, 83 Face validity, 82, 83
Diagnostic tests, 93 Facial paralysis, 177
Diffusion tensor imaging (DTI), 388 Facial weakness, 177
Dihydrolipoic acid (DHLA) metabolism, 258 Female sexual dysfunction (FSD), 517–520
Dipeptidyl peptidase IV (DPP-IV), 264 Female sexual function index (FSFI-R), 210
Distal symmetric polyneuropathy (DSPN), 5, 51, 183, Fibular mononeuropathy, 174
221, 232 Fludrocortisone, 460
in clinical care, 230, 231 Foot complications, 70
clinical presentation, 184, 185 Foot dorsiflexion, 75
depression and anxiety, 226, 227 Functional magnetic resonance imaging (fMRI), 431
diagnostic approach, 186
generalized anxiety vs. specific fears, 227, 228
generic health status/QoL Instruments, 228, 229 G
generic to DPN-specific QoL assessments, 229, 230 GABAA receptor, 131
interventions, 194–196 Gabapentin, 444, 447
motor function, 194 Gastric emptying, 482
neurogenic muscle weakness, 186 Gastrointestinal autonomic neuropathy, 28
pathophysiology of, 185 Gastrointestinal neuropathy, 471
on physical and mental functioning, 222, 223 autonomic function, 471, 472
psychological treatment options for, 231, 232 esophagus, 472, 473
psychosocial factors, 224–226 gallbladder, 480
sleep and mental health, 224 gastric emptying, 474, 481–483
Distal vessels, 358 postprandial hypotension, 483, 484
Dorsal root ganglia (DRGs), 280, 328, 329 prokinetic agents, 479
diabetes mellitus, 331 small and large intestine, 480–482
injury, 329, 330 stomach, 473–478, 480
insulin, 337, 338 Gastroparesis, 28
sensory neurons, 335–337 GDNF family ligands, 133
Droxidopa, 461 Gene transfer, 141
Dynamic mechanic allodynia (DMA), 116, 443 Genome wide association study (GWAS) approach, 240
Dynamometry, 190 Glabrous skin, 98
Dysesthesia, 441 Glucagon-like peptide 1 (GLP-1), 338
Dyslipidemia, 279, 283, 287 Glucose transporters (GLUTs), 283
534 Index

Glutamate carboxypeptidase II, 315 M

Glutamatergic receptors, 368 Magnetic resonance imaging (MRI), 430
Glycemia, 473 Male sexual dysfunction, 502
Glycemic control, 4, 72 Manual muscle testing (MMT), 190
Grey matter, 430 Mechanical detection threshold (MDT), 116
Mechanical pain sensitivity (MPS), 116
Mechanical pain threshold (MPT), 116
H Median mononeuropathy across the wrist (MNW), 172
Heart rate (HR), 204 Mediterranean diet, 399
Heat shock proteins (HSPs), 314 Meissner’s corpuscles, 98
Hedgehog pathway agonism, 139 Mendelian disorders, 240
Hereditary sensorimotor neuropathy, 430 Metabotropic glutamate receptors (mGluRs), 314
Hexosamine pathway, 260 Michigan Neuropathy Screening Instrument (MNSI), 76,
High fat diet (HFD), 398 77
Histone deacetylases (HDACs), 405 Microcirculation, 352
Hospital Anxiety and Depression Scale, 221 Microvascular changes, 160
Hyperalgesia, 74 Microvascular disease, 351–353
Hypercholesterolemia, 185 autonomic denervation, 355, 356
Hyperglycemia, 352 diabetic neuropathy, 354, 355
Hyperglycemic rodents, 130 nerve axon reflex, 356–358
Hypertriglyceridemia, 185 Microvascular ischemia, 176
Hypogonadism, 504 Microvessel wall, 168
Midodrine, 460
Mindfulness-based cognitive therapy, 233
I Mitochondria, 307
IGF-1 receptor (IGF-1R), 137 Mitochondrial function, 366, 367
IL-6 cytokine family, 133–135 Modified TCNS (mTCNS), 79
IL-6 receptor sub-unit (IL-6R), 135 Monocarboxylic acid transporter 1 (MCT1), 402
Inflammation, 282, 363–364, 419 Mononeuropathies, 5, 28
Insulin, 4, 136 entrapment mononeuropathies, 46
Insulin-like growth factors, 137 microvasculitic diabetic mononeuropathies, 45
International Index of Erectile Function (IIEF), 210 Morphogens, 138
Intraepidermal fibers (IENFs), 99 MOS sleep problems index, 222
Intraepidermal nerve fiber density (IENFD), 93 Motor and sensory nerves, 183
Intramuscular fat-accumulations, 188–189 Motor function, 75, 194
Intravenous immunoglobulin (IVIG), 170 Motor involvement and diabetes, 196
Intravenous methylprednisolone (IVMP), 169 Motor nerve dysfunction, 187
Ipswich touch test, 75 Motor neuropathy, 183–185, 190, 191, 196
Ischemic nerve injury, 171 MTHFR gene, 243
Ischemic reperfusion injury (IR), 207 Multi-ethnic study of atherosclerosis (MESA), 204
Multiple regression analysis, 225
Muscle size, 190
J Muscle strength
JAK/STAT signaling, 134 assessment of, 189, 190
diabetic distal symmetric polyneuropathy, 188, 189
Muscle weakness, 191
K Muscular atrophy, 186, 190, 191
KCC2 expression, 131 Myopathy, 186
Ketogenic diets, 400, 401

N
L NAD-dependent sorbitol dehydrogenase, 254
Langerhans cells, 142 Native peptide sequence, 142, 143
Lateral femoral cutaneous mononeuropathy, 174 Nephropathy, 358
Linear regression models, 228 Neprilysin, 265
Lipoic acid (LA), 258 Nerve axon reflex, 356, 357
Lipoprotein lipase (LPL), 293 Nerve conduction studies (NCS), 93, 96, 183
Lumbosacral DRPN, 29 Neural impairment, 263
Index 535

Neuropad, 104 disease-modifying therapies, 297–299

Neuropathic pain, 113, 415, 429, 446 dyslipidemia, 290–293, 296
Neuropathy, 4, 358 inflammation, 282
Neuropathy symptom score, 82, 83 lipid metabolism, 283, 291, 292
Neurostimulation, 448 metabolic syndrome, 279, 281–282
Neuroticism, 225 mitochondria, 283
Neurotrophic signaling, 143 neuron, 280, 283, 285
Neurotrophic support, 127 prediabetes, 279
Neurotrophin family, 128, 494 schwann cells, 280
BDNF, 130, 132 Perivascular inflammatory, 169
GDNF family ligands, 133 Peroxynitrite, 261
IL-6 cytokine family, 133–135 Physical tests, 194
NGF, 129 Pilot longitudinal study, 78
NGF injection therapy, 129 Pinprick hyperalgesia, 116
Nitrosative stress, 313 Plantarflexion, 75
Non-contractile tissue infiltration, 186 Polyol pathway, 254, 255
Non-enzymatic glycation, 258–260 Polyradiculopathies, 5
Nuclear factor-2 erythroid related factor-2 (Nrf2), 264 Positron emission tomography (PET), 433
Postprandial hypotension, 483
Posturography, 194
O Prediction of Arrhythmic Events with Positron
Ocular motor mononeuropathies (OMM), 175, 176 Emission Tomography (PAREPET)
Olmsted County Minnesota, 175 study, 208
Omega-3 polyunsaturated fatty acids (PUFA), 264, 265, Pregabalin, 444, 447
267 Premature ejaculation (PE), 514–516
Orthostatic hypotension, 214, 453, 454 Progressive resistance training, 195
clinical presentation, 464 Proinflammatory peptides, 365
diagnostics, 465–467 Proliferator-activated receptor-gamma co-activator 1α
evaluation, 456 (PGC-1α), 309
initial steps, 455 Proof-of-concept study, 267
neurogenic supine hypertension, 462 Protein kinase C (PKC), 260, 286
nonpharmacological interventions, 458–462 Psychological flexibility, 225
postprandial hypotension, 463
sudomotor dysfunction, 463
symptoms, 455 Q
treatment, 458, 467 Quantitative sensory testing (QST), 320, 428
valsalva maneuver, 457, 458 application, in clinical trials, 102
Outer mitochondrial membrane (OMM), 284 point of care, 101, 102
Oxidative phosphorylation (OXPHOS), 289 Quantitative sudomotor axon reflex testing (QSART),
Oxidative stress, 260, 262, 282, 311 103

P R
Pain management, 113
Painful diabetic polyneuropathy (P-DPN), 113, 427, 428,
441
mononeuropathies, 441
non pharmacological treatment, 448
pharmacological treatment, 444–447
radiculoplexopathy, 441
Painful neuropathy, 364, 367
Paradoxical heat sensations (PHS), 115
PARP inhibitor 3-aminobenzamide, 263
Patient reported outcomes (PROs), 229
Periaqueductal grey, 431
Peripheral arterial disease (PAD), 358
Peripheral neuropathy (PN), 279, 280, 282
bioenergetics, 283, 285, 288
Radiculopathies, 28
Radiculoplexus neuropathies, 46, 47
Randomized controlled trials (RCTs), 231
Ranirestat, 257
Rate-dependent depression (RDD), 131
Reactive oxygen species (ROS), 308, 369
Rearranged during transfection, 133
Receptor for advanced glycation end products (RAGE),
258
Renin-angiotensin system (RAS), 264
Reproducibility, 82
Retinopathy, 358
Retrograde ejaculation, 516
Rho kinase (ROCK), 367
Roux-Y gastric bypass (RYGB), 313
536
S T
Sandvik Severity Index, 210
Saturated fatty acids (SFA), 291
Scale face validity, 76, 80, 82
Schwann cells, 127, 130, 133, 134, 254, 258, 280
Sensory modalities, 83
Sensory nerve action potentials (SNAPs), 94
Sensory organs, 98
Serotonin, 444, 446
Serotonin/noradrenaline reuptake inhibitors (SSRI/
SNRI), 370
Severe acute respiratory syndrome coronavirus 2
(SARS-CoV-2) pandemic, 86
Severity, 71
Short Form (SF)-36, 224
Short-form McGill Pain Questionnaire (SF-MPQ-2), 231
Silent mating type information regulation 2 homolog1,
309
Single nucleotide polymorphisms (SNPs), 243
Skin biopsy, 99, 100, 188
Sleep problem index, 222
Small and large fiber function, 94
Small fiber dysfunction, 74
Small vessel diseases (SVDs), 385
Small-fiber neuropathy (SFN), 93
Smooth muscle actin (SMACTIN) staining, 169
Sonic hedgehog protein (Shh), 340
Sorbinil, 255
Sorbitol dehydrogenase (SORD) gene, 244
Spinal cord, diabetic neuropathy, 363
inflammation, 364, 366
molecular mechanisms, 366, 367
proinflammatory peptides, 365, 366
Spinal cord stimulation (SCS), 373, 374, 448
Spinal disinhibition, 370–373
Standard ENG examination, 187
Stem cells and secretomes, 143, 144
Sudomotor testing, 453, 456
electrochemical sweat conductance, 104
neuropad, 104
quantitative sudomotor axon reflex testing, 103
thermoregulatory sweat testing, 104
Superoxide, 261
Survival motor neuron (SMN), 336
Symmetric polyneuropathy, 5
Sympathetic skin response (SSR), 466
Synaptic plasticity, 367
Index
Tanezumab, 129
Test quality and reliability, 76, 83
Thermal thresholds, 115, 116
Thermoregulatory sweat testing (TST), 104
Tibial nerve stimulation, 500
Tingling, 441
Toronto clinical neuropathy score (TCNS), 77–79
Toronto consensus criteria, 94
Toronto diabetic neuropathy expert consensus, 243
Transient receptor potential (TRP), 368
Treatment induced neuropathy (TIN), 29
Treatment induced neuropathy of diabetes (TIND)
clinical features of, 158
historical background, 157, 158
Tricarboxylic acid (TCA) cycle, 207
Tricyclic antidepressants (TCAs), 444, 445
Tumor necrosis factor alpha (TNFα), 366
U
Ulnar mononeuropathy across the elbow (UNE), 173
United Kingdom Prospective Diabetes Study (UKPDS),
253, 352
Urethra dysfunction, 495
Urinary incontinence, 28
Urothelial dysfunction, 494
Utah early neuropathy scale (UENS)
examination, 80–82
V
Valsalva maneuver, 456
Vascular disease, 351
Vascular endothelial growth factor (VEGF), 139, 140
Vasopeptidase inhibitors, 265
Visual Analog Scale (VAS), 443
Voltage-gated calcium channels, 369
Voltage-gated sodium channels, 369
W
Ward, John D, 4
White matter, 430
White matter hyperintensities (WMH), 386
Wide dynamic range (WDR), 367
Williamson, RT, 4