Best Evidence Topic Presentation

Neurology Clerkship
NAMES
August 2, 2013
Title: Etiology, Manifestation and Management of Diabetic Amyotrophy in patients with NIDDM
Clinical Scenario:
Patient is a 51 year old African American male with Type II Diabetes who presented to the emergency
department with complaint of weakness and numbness in the lower extremities for over a month. Patient was
admitted to Neurology for further work up. Upon physical examination, weakness and numbness were exhibited
bilaterally (4/5) in the lower limbs. Decreased sensation and numbness were particularly localized to the lateral
portion of the patients left lower leg. Weakness and numbness were also elicited in the upper extremities, with
the right side exhibiting greater loss of sensation, and decreased strength (4/5) as compared to the left side.
Patient denies pain. Muscle wasting was evident in the right hand, and most pronounced in the thenar,
interossei, and abductor digiti minimi muscles. Patient has to concentrate on walking but physical exam
reveals no cerebellar defect. Bicep, tricep, brachioradialis, patellar, and achilles reflexes are intact and
responsive (5/5). Babinski sign is negative. MRI positive for central canal narrowing (T11-T12). MRI positive
for compression in the lumbosacral spine (L5-S1), with greater severity on the right side.
Search Strategy:
MEDLINE 1993-08/13 using the OVID interface, Google scholar, Pubmed. American Association of
Neuromuscular & Electrodiagnostic Medicine. UpToDate.com.

Author, date and

country

Kelkar, Praful MD; Masood, Moeen MD; Parry, Gareth J. MD. July 12, 2000.
USA

Patient group

15 patients with Proximal Diabetic Neuropathy (PDN) also known as diabetic

amyotrophy and two diabetic control groups

Study type (level of

evidence)

Case-control Study

Outcomes

Support for humoral factors comes from findings of Ig and complement

deposition in peripheral nerves in diabetes.
Noted small-vessel neutrophilic vasculitis with PMN infiltration in
postcapillary venules along with IgM deposition in endoneurium and in
affected vessel walls, and activated complement deposition in small
vessels.
Deposition of IgM in the endoneurium and in the walls of affected vessels.
We also found deposition of the activated complement along the
endothelium in small vessels with mAb directed against C5b-9
Humoral as well as T cell-mediated processes seem to be involved.

Diagnosis of PDN based on presence of subacute illness with progressive,

asymmetric proximal lower extremity weakness and wasting, with or
without pain, presenting in patients with a diagnosis of diabetes mellitus;
1

reduced or absent lower extremity deep tendon reflexes; and

electrodiagnostic studies showing fibrillation potentials or positive sharp
waves in at least two proximal lower extremity muscles innervated by two
different nerves (may include paraspinal muscles), with or without
evidence of peripheral neuropathy.
Diagnosis of PDN required absence of upper motor neuron signs
(spasticity, hyperreflexia, + Babinski), CNS disorder (excluding diabetic
neuropathy; presence of severe peripheral vascular disease as manifested
by femoral bruits or absent lower extremity pulses; and structural spinal
cord or pelvic lesions such as lumbar canal stenosis, carcinomatous
meningitis, radiation to spinal cord or lumbosacral plexus, and malignant
infiltration of the lumbosacral plexus.

Key results

Four patients showed the distinctive findings of polymorphonuclear smallvessel vasculitis affecting epineurial vessels with transmural infiltration of
postcapillary venules with polymorphonuclear leukocytes.
Immunoglobulin M (IgM) deposits were found along the endothelium and
intramurally in affected vessels. IgM staining was seen in the
subperineurial space and in the endoneurium.
Activated complement deposition was seen along endothelium of small
vessels.
Six patients showed "perivasculitis" with mononuclear cell infiltrates
around small epineurial vessels without vasculitis (fibrinoid necrosis or
transmural inflammation).
One patient showed recanalized vessels with transmural lymphocytes
without fibrinoid necrosis, possibly suggesting healed vasculitis

Study Weaknesses

Patient population (cases) was not completely uniform, and thus they may
introduce unseen biases to the study
Because of the progressive nature of (diabetic) neuropathy
functional impairment varied from mild difficulty in ambulation to
being wheelchair bound (3 wheelchair-bound, 4 walker for
ambulation, and 8 required minimum support with a cane)
Results may be affected by Lead-time Bias
Thirteen patients experienced significant weight loss, which ranged from
10 to 43 pounds; muscle weakness can be associated with malnutrition or
sudden weight loss. Therefore this finding may introduce a Confounding
bias into the study
Study was not blinded
Pygmalion effect- occurs when a researcher's belief in the efficacy
of a treatment changes the outcome of that treatment

Author, date and country

Howard W. Sander, M.D., and Sudhansu Chokroverty, M.D.

Patient group

Journal article on the current conceptual understanding of diabetic

amyotrophy distinguishing it from other forms of neuropathies.

Study type (level of evidence) Meta-analysis

Electrodiagnostic studies:
2

Outcomes

Nerve conduction and needle electromyography examinations assist in

distinguishing diabetic amyotrophy from other possible disorders.
Femoral motor nerve conduction studies usually reveal prolonged femoral
nerve latency and reduced amplitude of the femoral nerve compound
muscle action potential (CMAP). Conduction velocity and distal latencies
in the peroneal and tibial nerves are normal or mildly decreased and the
CMAP amplitude may be reduced. These studies indicate a
predominantly proximal axonopathy in the lower limbs, with or without a
superimposed mild distal sensorimotor polyneuropathy.

Diabetic amyotrophy is a distinct clinical entity that can be

differentiated from other diabetic neuropathies. Electrodiagnostic
and pathologic studies reveal evidence of a neurogenic lesion
affecting predominantly proximal musculature. A concomitant distal
sensorimotor polyneuropathy may be present, in which case there is
often a slower progression and more symmetrical involvement
While recovering from diabetic amyotrophy strength may improves
but the muscle wasting may persist. Muscle stretch reflexes often do
not recover. Sanders cites a study by Coppack and Watkins of 26
patients with diabetic amyotrophy, no additional recovery occurred
following 18 months. Mild residual functional weakness (difficulty
ambulating and climbing stairs) was present in seven patients, and
decreased thigh circumference with depressed knee reflexes
persisted in 13 of the patient. He also cites a prior study by Casey
and Harrison of 12 patients only 3 recovered full function and 5 had
significant residual disability. Improvement has been correlated with
better glycemic control, although Improvement also occurs in some
patients without concomitant control of hyperglycemia."
The severe pain associated with diabetic amyotrophy begins to abate
several months after onset, but residual pain may continue for
several years. Analgesics are helpful for pain control in most of the
cases. Amitriptyline and desipramine have been shown to be
efficacious for treatment of pain in the more common distal
sensorimotor neuropathy of diabetes, and phenytoin and
carbamazepine have been used as well. Transcutaneous nerve
stimulation has been beneficial in an occasional patient.

Key results

Study of 13 patients with either diabetic amyotrophy or mononeuritis

multiplex (Krendel et al)
10 patients (biopsy of leg muscle or nerve performed)
reported small vessel disease
7 of these 10 had perivascular chronic inflammatory cells and
3 had arterial occlusion with inflammation.
Sural nerve biopsy study in 12 patients with proximal diabetic
neuropathy (among 20 patients with diabetic neuropathy) (Younger
et al)
6 had a T-cell microvasculitis
6 had perivascular lymphocytic infiltration
4 had mononuclear cell infiltration of nerve
3

 2 had focal changes suggestive of ischemia

All patients showed a predominantly CD8+ T-cell infiltration
of nerve
Neurogenic changes were seen in four of five proximal
diabetic neuropathy patients who underwent muscle biopsy
Myopathic changes were present in one patient.
All (21) patients (including the 13 patients with either diabetic
amyotrophy or mononeuritis multiplex) who were treated with a
variety of immunomodulating agents stopped worsening and began
to improve.
All (8) patients (including 5 with proximal diabetic neuropathy)
whose treatment with intravenous immunoglobulin (IVIG) was
based on vasculitic sural nerve biopsy findings had decreased pain
and significant improvement in strength.

Dose of IVIG was 2g/kg total body weight over 5 days, monthly for 3
months

Study Weaknesses

No additional primary evidence

Author, date and country

Bruce Taylor, and John Dunne. May 7, 2004. USA

Patient group

Two patients with definite diabetic amyotrophy progressing subacutely to

severe quadriparesis are discuss extensively within the article.

Study type (level of

evidence)

Case Report
Case 1. A 76-year-old woman presented with a 6-week history of
bilateral but asymmetric (right left) knee pain followed by
progressive difficulty with walking, because of the knees giving
way.
Case 2. A 56-year-old woman presented with 2 weeks of increasing
asymmetric medial thigh pain and paresthesias (right > left), and
proximal leg weakness.

Outcomes

Both cases presented with asymmetric proximal lower-limb pain

followed by weakness, as has been previously described as typical
for diabetic amyotrophy.
Case 1 then had a stepwise progression of asymmetric lowerthen upper-limb weakness that occurred over 3 months
resulting in quadriparesis, with severe residual disabilities.
Case 2 had a gradual progression to generalized weakness
over 6 weeks that resulted in severe quadriparesis, but
ultimately had an excellent although incomplete recovery.
Presence of distal sensory peripheral neuropathy in both cases
supports the view that the various forms of diabetic peripheral
nervous system disease are not necessarily mutually exclusive
conditions.
An argument can be made for distinct diabetic neuropathies
However, the clinical descriptions of diabetic amyotrophy
4

and subacute diabetic proximal neuropathy make the

distinction far from clear.
Most authors would now favor a distinction between
principally metabolic/microangiopathic neuropathy as in
diabetic length-dependent peripheral neuropathy and immune
vasculopathy as seen in diabetic amyotrophy.
Key results

Case 1
Findings were consistent with an evolving and severe
polyradiculoneuropathy
Stepwise progression of patchy and asymmetric lower- and then
upper-limb weakness and sensory loss continued over a total of 20
weeks from onset, leading to severe quadriparesis (mostly grades 0
3),
Respiratory function tests indicative of respiratory muscle weakness.
Patient became totally bedbound
Slow recovery then commenced, however after 42 months she had at
least moderate weakness (grades 34) of all limbs, and remained
chairbound and dependent
Sural nerve biopsy on light microscopy showed prominent axonal
loss mainly involving large myelinated (1012 m) fibers, patchy
perivascular mononuclearcell infiltrates without involvement of the
vessel wall, and no demyelination, consistent with diabetic
neuropathy
Electron microscopy showed loss of myelinated fibers
Case 2
Initial examination showed grade 3 hip-flexor weakness only and
absent knee jerks, with minor and patchy impairment of pinprick
sensation distal to the mid-calves.
Gradual progression over 4 weeks led to asymmetric and severe
quadriparesis (mostly grades 12), the weakest muscles being
proximal and in the lower limbs. All deep tendon reflexes were lost.
Patchy sensory impairment to all modalities was present in the
fingers and toes. Patient was bed-bound and totally dependent
Sural nerve biopsy showed prominent axon loss mainly affecting
large-diameter (1216 m) myelinated fibers, with increased
endoneurial collagen, some areas of segmental myelin breakdown
and remyelination, and thickening of the media and intima of
epineural blood vessels.
Slow recovery commenced 8 weeks after onset
After an additional 3 months patient was able to walk with a stick
and was independent
Moderate diffuse weakness and stocking sensory impairment
persisted to the mid-calf.
After 12 months, only residual signs were mild truncal and proximal
lower-limb weakness affecting particularly the iliopsoas, hip
adductors, and abdominal muscles, and diffuse hyporeflexia.
Overview
The two cases we present indicate the potential for patients
presenting with diabetic amyotrophy to progress to severe
5

quadriparesis
Clinical spectrum of diabetic neuropathy, particularly diabetic
amyotrophy, includes progression to severe quadriparesis,
and major residual disabilities
In both patients, neurophysiological findings were consistent with a
severe and diffuse polyradiculoneuropathy, predominantly axonal in
type.
Sural nerve biopsies showed primary changes of axon loss
with predominant loss of large myelinated fibers
Perivascular inflammatory changes were prominent in one
case but absent in the other.
These findings are consistent with those that have been previously
reported in diabetes mellitusassociated neuropathies
Study Weaknesses

Power of the study is weak (only two patients)

Author, date and country

D. Kilfoyle, P. Kelkar, and G. J. Parry. 2003. NZ

Patient group

9 patients (5 men and 4 women; mean age 65 years (range, 4388 y)) with
diabetic amiotrophy occurring between 2000 and 2002

Study type (level of evidence)

Outcomes Research/ Retrospective Cohort Study

Retrospectively reviewed the case records of 10 episodes of
diabetic amyotrophy in 9 patients treated with pulsed oral or
intravenous methylprednisolone.

Outcomes

Pulsed methylprednisolone (MP) appears to be a safe and effective

treatment for diabetic amyotrophy

Key results

Patients were considered to have benefited from the treatment if pain

had improved by 3 months after symptom onset and strength or function
had improved by 6 months.
MP at a dosage of 500 mg was administered either
intravenously or orally on 2 consecutive days every 2 weeks
for up to 3 months, depending on response.

6 episodes there was marked improvement in pain within days of

starting treatment. Strength improved more slowly but faster than
the natural history of the disease.
Treatment started within 2 months of symptom onset was
associated with rapid improvement in pain
Very early treatment, started within 4 weeks of symptom onset,
resulted in rapid improvement of both strength and pain.
Blood glucose increased on treatment days but no patient required
lasting changes in diabetic treatment as the result of this therapy
and no other serious adverse effects were seen.
Each MP pulse was associated with transient asymptomatic
hyperglycemia.
In one patient, temporary insulin treatment was instituted
but in no other patient was the diabetes treatment altered.
One 84-year-old patient had a mild transient elevation of
6

blood pressure during some treatments and experienced

visual hallucinations with her final treatment.
Insomnia and irritability on the treatment days were the
only other adverse effects.

Study Weaknesses

Patients were considered to have benefited from the treatment if

pain had improved by 3 months after symptom onset and strength
or function had improved by 6 months.
Pain and Strength are often relative markers; lack of
empiricism may affect interpretation of results
Recurrence of severe pain and weakness in the opposite leg that
occurred 2 months after completion of a 3-month course of
treatment of initial symptoms in one patient was treated within 2
weeks of onset, and was regarded by the authors as a second
episode
This relapse may actually point to failure of the treatment,
which did not seem to be accounted for in the authors
assessment.
Patients level of weakness, strength, and immobility varied
significantly
This may introduce lead-time bias into interpretation of the
success of the treatment plan and outcome
Although there was a clear trend toward a more rapid improvement
(in patients with a shorter pretreatment duration of symptoms), the
numbers are too small for statistical analysis; Power of the study is
weak

Author, date and country

Pascoe MK, Low PA, Windebank AJ, Litchy WJ. 1997. USA

Patient group

44 patients with subacute diabetic proximal neuropathy

Study type (level of evidence)

A retrospective review of medical records of Mayo Clinic patients with

diabetes who had subacute onset and progression of proximal weakness.
The responses of treated versus untreated patients were compared
statistically.

Outcomes

During the designated study period, 44 patients with subacute

diabetic proximal neuropathy were encountered. Most patients
7

were middle-aged or elderly, and no sex preponderance was noted.

The proximal muscle weakness often was associated with reduced
or absent lower extremity reflexes. Associated weight loss was a
common finding. Frequently, patients had some evidence of
demyelination on nerve conduction studies, but it invariably was
accompanied by concomitant axonal degeneration. The
cerebrospinal fluid protein concentration was usually increased.
Diffuse and substantial autonomic failure was generally present.
In most cases, a sural nerve biopsy specimen suggested
demyelination, although evidence of an inflammatory infiltrate was
less common. Of 12 patients who received treatment (with
prednisone, intravenous immunoglobulin, or plasma exchange), 9
had improvement of their conditions, but 17 of 29 untreated
patients (59%) with follow-up also eventually had improvement,
albeit at a much slower rate.
Key results

the entity of subacute diabetic proximal neuropathy is an extensive

and severe variant of bilateral lumbosacral radiculoplexopathy,
with some features suggestive of an immune-mediated cause.
It differs from chronic inflammatory demyelinating
polyradiculoneuropathy in that most cases have a more restricted
distribution and seem to be monophasic and self-limiting.
The efficacy of immunotherapy is unproved, but such intervention
may be considered in the severe and progressive cases or ones
associated with severe neuropathic pain.

Study Weaknesses

Power of the study is weak (sample size 44)

Author, date and country

Barohn RJ, Sahenk Z, Warmolts JR, Mendell JR. 1991. USA

Patient group

Clinical, laboratory, and biopsy results were reviewed from 17 consecutive

patients (14 women and three men) who were examined between January
1984 and July 1988 at the Neuromuscular Service at Ohio State University,
Columbus. The diagnosis of proximal diabetic neuropathy, also referred to
as diabetic lumbosacral radiculoplexopathy, was made on the basis of the
following criteria: (1) onset of diabetes mellitus 2 either before or at the
onset of neurologic symptoms; (2) abrupt onset of hip, back, or leg pain,
either unilateral or bilateral; (3) weakness of the lower extremities
involving proximal muscles or proximal and distal muscles, unilaterally or
bilaterally (patients with weakness confined to a peripheral-nerve
distribution were excluded); (4) results of electrophysiologic studies
excluding myopathy; and (5) lumbosacral magnetic resonance images,
computed tomograms, or myelograms excluding neoplastic disease,
multiple compressive radiculopathy, and spinal stenosis.

Study type (level of evidence)

Retrospective Study

Outcomes

Fourteen patients had unilateral onset that later involved the other
extremity in 3 days to 8 months.
8

 All patients reported stepwise or steady progression during 2 to 18

months that was documented during serial examinations. In 16
patients, both proximal and distal muscles were involved.
Sural nerve biopsy specimens demonstrated multifocal variability
in nerve fiber density manifesting as nonrandom fiber loss between
and within fascicles compared with age-matched controls. These
findings demonstrate that patients may have a rapidly evolving
course of proximal diabetic neuropathy followed by continued
progression for many months and emphasize the overlap between
proximal diabetic neuropathies of presumed different types.
Key results

Dr. Barohn states that their cases, as well as others, cast doubt on notions
supporting two distinct types of proximal diabetic neuropathies represented
by the rapid evolution of asymmetric weakness on an ischemic basis, in
contrast to a more slowly progressive condition of metabolic pathogenesis.

Study Weaknesses

Power of the study is weak (sample size 17)

Conclusion:
The etiology of Diabetic Amyotrophy (DA) is not very well understood. There is a strong argument for immune
involvement, and increasing evidence points to a cause by microvasculitis of the vasa nervorum. There may be
nerve ischemia secondary to immune-mediated vasculopathy rather than diabetic microangiopathy. It is further
argued that neuropathy seen in DA has an immune-mediated inflammatory basis and suggest that
polymorphonuclear (PMN) vasculitis with immune complex and complement deposition may be the primary
impetus.
Diabetic amyotrophy often occurs in patients with a concomitant, preexisting distal sensorimotor
polyneuropathy. Incidence in diabetics, and concomitance with other neuropathies cause DA to be difficult to
identify. However the localization of muscle weakness and loss of sensation may assist in recognizing this
condition. Presentation of DA is often subacute and progressive with involvement often clinically first noted in
the proximal lower limbs particularly quadriceps femoris. However simultaneous involvement of the gluteal,
hamstring, adductor, and iliopsoas muscles often occurs. The predominantly proximal motor involvement is in
clear contrast to the distal, predominantly sensory "stocking"distribution typical of the common diabetic
polyneuropathy. In addition to proximal muscle weakness, wasting in muscles of the proximal lower extremity
favors a diagnosis of DA.
The Clinical Bottom Line:
Diabetic Amyotrophy is difficult to identify and treat. It can be recognized by proximal muscle weakness and
loss of sensation in the lower limbs, particularly in the pelvifemoral muscles. There may be associated pain and
loss of sensation in the affected muscles and limbs. Upper limb involvement is rare. Etiology of the condition
may have an autoimmune component, and thus further research needs to be conducted into management with
IVIG and steroids.
Reference Images

10