European Journal of Internal Medicine 25 (2014) 415–421

Contents lists available at ScienceDirect

European Journal of Internal Medicine

journal homepage: www.elsevier.com/locate/ejim

Review Article

Approach to the patient with spontaneous hypoglycemia

Pieter Martens a,⁎, Jos Tits b
a
Department of Internal Medicine, University Hospital Gasthuisberg Leuven, KU Leuven, Herestraat 49, B-3000 Leuven, Belgium
b
Department of Endocrinology, Hospital Ziekenhuis Oost-Limburg ZOL, Genk, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Hypoglycemia is common in daily clinical practice and often occurs during the treatment of diabetes mellitus.
Received 29 December 2013 However, a small minority of hypoglycemia encountered in clinical practice is spontaneous and thus not induced
Received in revised form 27 February 2014 by glycemic lowering agents. These spontaneous hypoglycemic events confront the clinician with a diagnostic
Accepted 28 February 2014
enigma. Although the trained clinician can recognize the autonomic and neuroglycopenic symptoms of hypogly-
Available online 16 March 2014
cemia even in a patient not on insulin, it remains challenging to decipher the etiology of a spontaneous hypogly-
Keywords:
cemic event. A logical and stepwise approach to the spontaneous hypoglycemic event allows for a conclusive
Spontaneous hypoglycemia diagnosis. This diagnostic process consists of adequately diagnosing hypoglycemia by fulfilling Whipple's triad,
Diagnostic work-up stratifying patients according to their clinical status and analyzing a full hypoglycemic blood panel. A complete
Insulinoma hypoglycemic blood panel should include the analysis of glucose, insulin, C-peptide, pro-insulin, insulin antibodies
Noninsulinoma pancreatogenous and the presence of oral hypoglycemic agents. For patients with episodes of hypoglycemia induced by excessive
hypoglycemia syndrome endogenous insulin, additional imaging is often required to detect the presence of an underlying insulinoma. By
Insulin auto-immunity and treatment diagnosing the underlying cause of the spontaneous hypoglycemia, the physician also diagnosis the mechanism
by which the hypoglycemic event occurs. Allowing for a problem orientated therapeutic approach.
Methodology: The present review is based upon a comprehensive PubMed search between 1985 and 2013. This
uses search terms of spontaneous hypoglycemia, insulinoma, nesidioblastosis, insulin auto-immunity,
noninsulinoma pancreatogenous hypoglycemia syndrome, hormone deficiency, pro-IGF II, and pro-insulin
growth factor II, and cross reference searching of pivotal articles in the subject.
© 2014 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.

1. Introduction
In daily clinical practice, hypoglycemia is most commonly iatrogenic,
caused by insulin or insulinsecretagogue used to treat diabetes mellitus
[1]. The diagnosis and treatment of a hypoglycemic event in a patient
using medication to lower plasma glucose level are therefore straight-
forward. But spontaneous hypoglycemia in a non-diabetic patient
confronts the clinician with a diagnostic enigma. Hypoglycemia is
confirmed when Whipple's triad is present [2,3]: (1) symptoms or
signs consistent with hypoglycemia, (2) a plasma glucose level less
than 55 mg/dl, measured with a precise method such as a venous
blood sample, and (3) resolution of symptoms after raising plasma
glucose level [4]. Symptoms consistent with hypoglycemia are broadly
categorized as autonomic or neuroglycopenic. Symptoms that arise first
are the autonomic symptoms, mediated by the adrenergic and cholinergic
axes of the sympathetic nervous system. Adrenergic symptoms consist of
palpitations, tremor, and anxiety and are mediated by an up-regulation of
norepinephrine and epinephrine. Cholinergic symptoms include hunger,
⁎ Corresponding author at: Department of Internal Medicine, University Hospital
Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium. Tel.: +32 16 34 42 75; fax: +32
16 34 42 30.
sweating, and paresthesia and are derived from the acetylcholine released
by postganglionic sympathetic neurons. This sympathetic response is part
of the physiological counter regulatory mechanism, directed against a
decrease in plasma glucose level [4,5]. Although the sympathetic re-
sponse generates the first type of symptoms, it is not the first counter
regulatory mechanism against hypoglycemia. The first physiological re-
sponse to a decreasing plasma glucose level is a down-regulation of in-
sulin secretion, followed by a second defense of heightened glucagon
secretion. Only when these fail to call a halt to the decreasing plasma
glucose, is a sympathetic response apparent [6,7]. A second type of
symptoms is the neuroglycopenic symptoms. These symptoms arise
due to central nervous system glucose deprivation. Neuroglycopenic
symptoms range from confusion to amnesia, blurred vision, diplopia,
dysarthria, seizure and if sufficiently profound loss of consciousness
[8]. Prolonged hypoglycemia can cause brain death and hypoglycemia
has shown to increase all-cause mortality in cardiac patients [9,10].
Mortality is especially higher for the spontaneous hypoglycemia in
non-diabetics [11]. Probably because the underlying cause heralds a
more ominous prognosis and because these hypoglycemic events are
a marker for vulnerability. Recurrence of hypoglycemia is a great source
of morbidity and generates a great burden to the patient. The following
is a discussion on the management of the non-diabetic patient with
hypoglycemia.

http://dx.doi.org/10.1016/j.ejim.2014.02.011
0953-6205/© 2014 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
416 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421
2. Diagnosis of hypoglycemia
Hypoglycemia is uncommon in non-diabetics because of the effec-
tiveness of redundant counter regulatory mechanisms. A rapid decline
in insulin secretion combined with an augmented glucagon secretion
and heightened sympathetic outflow allows for a rapid response against
decreasing plasma glucose levels. Growth hormone and cortisol are
implicated in the defense against prolonged hypoglycemia. Only when
these defense mechanism fail to substantially increase endogenous
glucose mobilization to counter-act the falling glucose levels, does
hypoglycemia develop [5,6]. Given the extensive redundancy of these
defense mechanisms, one should consider hypoglycemia in non-
diabetics a red flag. A meticulous evaluation is thus warranted. The diag-
nostic process starts by the recognition of hypoglycemia as a cause of
the presenting symptoms such as confusion, altered level of conscious-
ness, seizure or any of the other autonomic and neuroglycopenic symp-
toms. This can be difficult because the symptoms are not exclusive for
hypoglycemia. Conformation is done by documenting Whipple's triad
[4]. After documenting Whipple's triad, two elements are key in the
approach to the non-diabetic with hypoglycemia: (1) diagnosis of the
hypoglycemic mechanism, and (2) management of the low blood
glucose level [4].
3. Mechanism diagnosis
Hypoglycemic disorders used to be classified as being postabsorptive
hypoglycemia (fasting hypoglycemia) versus postprandial hypoglyce-
mia (re-active hypoglycemia) [6]. This classification was based on the
assumption that the former is caused by organic pathologies presenting
mostly with neuroglycopenic symptoms and the latter arises from
functional disorder presenting mostly with autonomic features. This
classification is not very useful because it neither expedites diagnosis
nor facilitates an understanding of the pathophysiology of the disorders
causing hypoglycemia [6]. An insulinoma, for example, can present post-
prandial or a post-absorptive state [12]. A more useful classification for
the clinician is to establish whether the patient is seemingly well or has
a concurrent illness [4]. At all times drugs should be suspected, even in
the non-diabetic, because insulin or an insulin secretagogue could be
administered in an accidental or even malicious fashion. Also some
non-antidiabetic drugs can cause an iatrogenic hypoglycemia [13]. The
approach of dividing hypoglycemia in categories of seemingly well,
concurrent illness or iatrogenic allows for a pathophysiological and ther-
apeutically relevant approach. The different causes of hypoglycemia are
listed in Table 1. An initial history, physical examination and careful
review of available laboratory information will allow broad categorizing
of the different causes of hypoglycemia. Fig. 1 illustrates a proposed
diagnostic work-up.
Table 1
Causes of hypoglycemia.
Iatrogenic
Insulin or insulinsecretagogue
Alcohol
Others:
Moderate quality of evidence
Quinine,
Gatifloxacin, pentamidine,
Indomethacin,
Glucagon (during endoscopy)
Low quality of evidence
Lithium, IGF-I,
Chloroquinoxaline sulfonamide,
propoxyphene
Dextropropoxyphene, Artesunate
IGF denotes insulin-like growth factor; NIPHS denotes noninsulinoma pancreatogenous hypoglycemia.
PGBH denotes post-gastric bypass hypoglycemia.
3.1. Concurrent illness
The subgroup concurrent illness defines a group of patients in which
the clinical findings during history or physical examination indicate a
primary pathology which is held responsible for the hypoglycemic
event. Hereby delineating patients at increased risk of mortality in
which the hypoglycemic event is a sign of increased vulnerability. Con-
current illnesses that present with hypoglycemia are: critical illnesses,
hormone deficiencies and some non-beta cell tumors [4–6].
3.1.1. Critical illnesses
Among hospitalized patients, serious illnesses such as renal, hepatic
or cardiac failure, inanition and sepsis are frequent causes of hypoglyce-
mia. Acute and massive hepatocellular injury (e.g. shock liver, toxic hep-
atitis) abolishes the hepatic ability to increase plasma glucose level by
means of glyconeogenesis and induces hypoglycemia typically in a
fasting state. Renal failure reduces insulin clearance and diminishes
the mobilization of gluconeogenic precursors. The exact mechanism of
hypoglycemia in heart failure is unknown. In sepsis there is an increased
glucose utilization induced by cytokines [4].
3.1.2. Hormone deficiencies
Growth hormone (GH) and cortisol are implicated in the defense
against hypoglycemia during prolonged fasting. Hypopituitarism and
primary adrenal insufficiency (Addison's disease) can cause hypoglyce-
mia during prolonged fasts. Chronic cortisol deficiency generates hypo-
glycemia by inducing a state of precursor deprived gluconeogenesis.
Leading to a hypoglycemic event when glycogen storage is depleted.
GH deficiency can cause hypoglycemia in children but is uncommon
in adults. Both GH and cortisol deficiency lead to a number of symptoms
beside hypoglycemia, assisting in the diagnostic process [4,5].
3.1.3. Non-beta cell tumors
Several mesenchymal and epithelial tumors such as hepatomas, gas-
tric tumors or sarcomas can produce an incompletely processed form of
insulin-like growth factor II (pro-IGF II). These tumors are mostly large
tumors (N10 cm) and generate symptoms not only by means of space
occupation, but also by inducing hypoglycemia. Hypoglycemia typically
occurs in a fasting state with a typical suppressed insulin concentration
by pro-IGF II. This suppressed insulin concentration allows it to be dif-
ferentiated from hyperinsulinemic states of hypoglycemia. The concen-
tration of IGF II is typically normal, but the ratio of pro-IGF II/IGF II is
increased [4,5,14,15].
3.2. Seemingly well
When a patient presents with a spontaneous hypoglycemia, without
any apparent stigmata of a causative underlying illness such as critical
Concurrent illness Seemingly well
Critical illness Endogenous hyperinsulinism
Hepatic, renal, cardiac failure Insulinoma
Sepsis NIPHS/PGBH
Inanition Insulin autoimmunity
Hormone deficiency Exogenous hyperinsulinism
Cortisol Accidental
Growth hormone Factitious
Glucagon Glucagon
Non-beta cell tumors
 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421 417

Fig. 1. Abbreviations: GH denotes growth hormone, IGF II denotes insulin-like growth factor II, HH denotes hyperinsulinemic hypoglycemia, Ab denotes antibody, SPACI denotes
SELECTIVE pancreatic arterial calcium injective and NIPHS denotes noninsulinoma pancreatogenous hypoglycemia syndrome. Words in red preceded by a dash indicate the main diag-
nostic clue. *Complete hypoglycemic panel consists of glucose, insulin, C-peptide, pro-insulin, beta-hydroxybutyrate, and glucose concentration after glucagon, and screen for circulating
oral hypoglycemic agents and for insulin antibodies. The cutoffs of these texts are a listed in Table 2. °NIPHS and PGBH show similar features in the work-up such as an endogenous HH
after stimulation with a mixed meal test, a negative work-up for insulinoma and the absence of insulin antibodies. The difference between these two syndromes is the severity of the
clinical picture predominating in NIPHS. SPACI and tissue analysis revealing nesidioblastosis can clearly differentiate NIPHS from PGBH.

illness, hormone deficiency or large neoplasms and thus seems other-
wise healthy, an expeditious but detailed blood analysis is warranted.
The blood analysis should be done without delay because persisting
hypoglycemia carries a risk [9–11]. Although persisting hypoglycemia
carries a risk, treatment with glucose should be deferred until a plasma
sample is obtained. This is because the directions of glucose, insulin,
C-peptide and pro-insulin during a hypoglycemic event facilitate
diagnosis. When a euglycemic patient presents with a history indicative
of hypoglycemic spells, a stress test can be done to induce hypoglyce-
mia. The history of spells determines what type of stress test can be
used. A history of fasting hypoglycemia requires a 72-hour fast, whilst
postprandial hypoglycemia requires a mixed meal test to induce hypo-
glycemia [4,6,7]. An oral glucose tolerance test (OGTT) to diagnose post-
prandial hypoglycemia has been discouraged by the literature [4,16,17].
Because at least 10% of healthy individuals develop a plasma glucose
nadir of less than 50 mg/dl during OGTT [18]. This effect is not seen in
healthy individuals following a mixed meal test [16].
Plasma analysis of a hypoglycemic state (spontaneous hypoglycemia
or induced after a stress test) allows to differentiate between an exoge-
nous and a endogenous source of hypoglycemia, and allows to break
down the differential diagnosis even further. Promptly treating the
hypoglycemia will distort the diagnostic value of a detailed blood anal-
ysis, necessitating a stress test to induce a second hypoglycemia. The
initial blood analysis should cover plasma glucose, insulin, C-peptide,
pro-insulin, and β-hydroxybutyrate concentration and should screen
for oral hypoglycemic agents and insulin antibodies. Hereby allowing
a differentiation between different causes. The diagnostic results of
these tests are a listed in Table 2.
In the seemingly well patient exogenous hyperinsulinism is caused
by factiously or even maliciously administering insulin or an oral hypo-
glycemic agent. This situation will be denied during history taking, and
requires from a legal standpoint of view a methodological correct
approach, clearly documenting exogenous hypoglycemia [19–21]. A
state of endogenous hyperinsulinism is a rare condition that can be
caused by several pathologies, such as an insulinoma, noninsulinoma
pancreatogenous hypoglycemia syndrome (NIPHS), post-gastric bypass
hypoglycemia and insulin autoimmunity. Pathophysiologically they all
share the fundamental feature that insulin concentration fails to
decrease in appropriateness with the low plasma glucose concentration.
The endogenous hyperinsulinism is confirmed by: plasma insulin con-
centration of at least 3 μU/ml, plasma C-peptide concentration of at
least 0.2 nmol/l, and a plasma pro-insulin concentration of at least
418 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421

Table 2
Patterns of finding during a spontaneous hypoglycemia, during fasting hypoglycemia or hypoglycemia after a mixed meal. This is in a normal individual with no signs or
symptoms despite apparent low plasma glucose (no documentation of Whipple's triad), in hyperinsulinemic causes of hypoglycemia and in a person with a hypo-insulinemic
cause of hypoglycemia (IGF-mediated).

Symptoms Glucose Insulin C-peptide Pro insulin Beta hydroxybutyrate Glucose increase Circulating oral Insulin Ab Diagnostic interpretation
(mg/dl) (μU/ml) (nmol/l) (pmol/l) (mmol/l) after glucagon hypoglycemic
(mg/dl) agents

No b55 b3 b0.2 b5 N2.7 b25 a No No Normal

Yes b55 ≫3 b0.2 b5 ≤2.7 N25 No No b Exogenous insulin
Yes b55 ≥3 ≥0.2 ≥5 ≤2.7 N25 No No Insulinoma, NIPHS, PGBH
(endogenous hyperinsulinism)
Yes b55 ≥3 ≥0.2 ≥5 ≤2.7 N25 Yes No Oral hypoglycemic agent
Yes b55 ≫3 ≫0.2c ≫5c ≤2.7 N25 No Yes Insulin autoimmunity
Yes b55 b3 b0.2 b5 ≤2.7 N25 No No IGF-mediated d

IGF denotes insulin-like growth factor, NIPHS denotes noninsulinoma pancreatogenous hypoglycemia.
PGBH denotes post-gastric bypass hypoglycemia, NA denotes not available.
a
Result during a prolonged fast.
b
Insulin antibodies can be present when taking exogenous insulin, the direction of C-peptide combined with pro-insulin (both track in the same direction) versus insulin allows to
differentiate from insulin autoimmunity.
c
Very high concentrations due to antibody interaction, but the free fraction C-peptide and pro-insulin are low.
d
Increased pro-IGF II/IGF II ratio.

5.0 pmol/l. This in a state of apparent hypoglycemia, confirmed by a
plasma glucose of b 55 mg/dl [4]. After documenting the endogenous
hyperinsulinism efforts should be made to diagnose the causative
disease, these include the following. Both endogenous states of hyperin-
sulinism and exogenous states of hyperinsulinism (excessive insulin ad-
ministration) share the finding of hypoglycemia in the face of an excess
of insulin. But these two states can be differentiated from each other be-
cause only endogenous states of hyperinsulinism have a simultaneous
rise in insulin, pro-insulin and C-peptide. Because exogenous insulin is
administered devoid of C-peptide and pro-insulin, it will present as a
hyperinsulinemic hypoglycemia with low c-peptide and pro-insulin.
An important caveat in this diagnostic tactic is the recent finding that
some newer insulin detecting assays only measure endogenous insulin
and lack the affinity towards epitopes on exogenous insulin [22,23].
Using such assays would jeopardize the diagnosis of exogenous
hyperinsulinemia. So before excluding the diagnosis of exogenous
hyperinsulinemic hypoglycemia one must make sure the laboratory is
using an assay capable of detecting epitopes on exogenous adminis-
tered insulin.
3.2.1. Insulinoma
An insulinoma is an endocrine tumor of the pancreas derived from
β-cells that ectopically secrete insulin, causing spells of hypoglycemia.
These hypoglycemic spells are typically in a fasting state, but can also
occur postprandial. As illustrated by a retrospective cohort study
reporting fasting hypoglycemia in 73% of insulinomas, postprandial
hypoglycemia in 6% and both in the remaining 21% [12]. Insulinoma
often presents in the 4th or 5th decade of life and it is a rare disease,
with an estimated incidence of 4 per million person-years [8]. When
biochemical evidence of an endogenous hyperinsulinemic hypoglyce-
mia is present, attempts should be made to localize an insulinoma, by
a non-invasive method as possible. The localization can be difficult
since 90% of insulinomas are smaller than 2 cm [8,24]. Standard imaging
techniques such as CT and MRI have detection rates of 70% and 85% re-
spectively [24]. Endoscopic pancreatic ultrasonography combined with
fine needle aspiration detects 57% to 94% of insulinomas depending on
the location in the pancreas [24,25]. Combination of CT with endoscopic
pancreatic ultrasonography allows for a detection rate approaching the
100% [8,24]. The relative importance of positron emission tomography
remains to be determined, but a small prospective study with 18F-
DOPA shows promising results [26]. Nuclear imaging with 68-Ga-
DOTATOC has major potential in locating the insulinoma [27].
3.2.2. Noninsulinoma pancreatogenous hypoglycemia syndrome
Noninsulinoma pancreatogenous hypoglycemia syndrome (NIPHS)
is a clinical entity first described in 1999, in a case-series of five patients
presenting with severe hypoglycemic attacks with neuroglycopenic
symptoms (fulfillment of Whipple's triad) [28]. These attacks all
presented postprandially and not in the fasting state, leading to the hy-
pothesis that a structural pathology other than insulinoma generated
these severe hypoglycemic attacks. Histopathological evaluation of the
pancreases showed nesidioblastosis. Nesdioblastosis is a histological
finding first termed in 1938 to describe a state of widespread β-cell
hyperplasia combined with diffuse proliferation and hypertrophy of
islet cells from pancreatic ducts [29]. Nesidioblastosis as a structural pa-
thology (causing hyperfunction of β-cells) has long been thought to be
the cause of endogenous hyperinsulinemic hypoglycemia in children.
The identification of disease causing mutations in the Kir6.2 and SUR1
genes has casted doubt on the hypothesis that merely the nesidioblastic
state of the β-cells causes the hyperinsulinemic hypoglycemia in these
infants [30]. These genetic mutations were absent in the five adult
patients first described with NIPHS [28]. Imaging studies of the pancreas
were unable to demonstrate an insulinoma. A selective pancreatic
arterial calcium injection (SPACI) showed increased insulin concentra-
tion after stimulating multiple vascular territories, indicating a diffuse
hyperfunction of the pancreas. The histopathological status of the
nesidioblastic β-cells generating an exaggerated insulin response was
postulated as the culprit of the severe postprandial hypoglycemic at-
tacks. These patients did not have any gastro-intestinal surgery such
as Roux and Y, indicating a form of idiopathic nesidioblastosis. Later
case-series report similar results of idiopathic nesidioblastosis as cause
of NIPHS [31–33]. Heighted attention to the entity of NIPHS led to the
recognition of similar clinical findings (severe neuroglycopenic symp-
toms) in patients with a history of gastric bypass surgery [34–37].
These patients had attacks of hyperinsulinemic hypoglycemia with
neuroglycopenic symptoms. A 72-h fast and imaging studies were
unable to detect an insulinoma. Diffuse pancreatic hyperfunction was
found after a SPACI, and tissue analysis revealed nesidioblastosis,
concluding to the existence NIPHS after gastric bypass. Later reports
show similar findings after a Nissen-fundoplication [38]. Whether
NIPHS in the situation of gastric bypass is caused by an organic pathol-
ogy (the nesidioblastic state of the β-cells) or by a functional disorder
(exaggerated incretin response, like GLP-1, secondary to altered nutrient
presentation [39] or an out of proportion response of the pre-surgical
hyperfunctioning β-cells [40]) is a point of great discussion [17,39–42].
But irrespective of the exact pathogenesis of NIPHS, it should not be
confused with post-gastric bypass hypoglycemia (PGBH). PGBH is an
attack of postprandial hyperinsulinemic hypoglycemia, constituting
a part of the entity described as late dumping (seen after bariatric
or esophagogastric surgery) [43]. The hypoglycemic attacks in PGBH
are normally relatively mild, and do not often present with severe
neuroglycopenic symptoms. PGBH presents most often in women and
 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421
the incidence exceeds that of insulinoma by a considerable degree
[4,41,42]. PGBH most often responds well to diet modification and
treatment with acarbosis and somatostatin analogs [43]. This clearly
contrasts NIPHS, which presents most in males and does present with
severe neuroglycopenic symptoms.
The diagnosis of NIPHS is suspected when a diagnostic work-up for
an endogenous hyperinsulinemic hypoglycemia fails to document an
insulinoma, by means of a negative 72 hour fast and negative imaging
studies [37]. The next appropriate step for documenting NIPHS would
be a SPACI. SPACI can conclude hyperfunction of β-cells when there is
a doubling of the baseline insulin concentration in the draining right
hepatic vein in response to an intra-arterial calcium injection (calcium
is an insulinsecretagogue). Because the pancreas has a threefold arterial
supply, consisting of the splenic-artery (supplying the body and tail of
the pancreas), gastroduodenal-artery (supplying the head and to a lesser
extend the uncinate process) and the superior mesenteric artery
(supplying the uncinate process and less the head), the extent of pancre-
as hyperfunction can be anatomically delineated. Diffuse hyperfunction
indicates nesidioblastosis, whilst a single artery response could indicate
an insulinoma missed on screening imaging, although this situation is
deemed very unlikely [37]. SPACI can also guide surgical resection,
even though the theoretical benefit of this approach has no effect on
clinical outcome [44]. After surgery the aberrant pancreatic tissue can
be confirmed to be nesidioblastosis by pathological analysis, but current-
ly different histological criteria are postulated [36,45].
3.2.3. Insulin autoimmunity
Hyperinsulinemic hypoglycemia associated with insulin antibodies
is a rare condition. This disease has a predominance for persons from
Japanese descent. In Japan it is the third leading cause of spontaneous
hypoglycemia, but the disease still remains uncommon with only 244
cases reported between 1970 and 1997 [46]. Among non-Asians insulin
autoimmunity is even a rarer phenomenon [47]. Predisposing condi-
tions are reported for the development of this disorder, including:
systemic lupus erythematosus, multiple myeloma, chronic ulcerative
colitis and the use of sulfhydryl containing drugs such as propylthiouracil
[48–58]. Patients with insulin autoimmunity present with neuro-
glycopenic symptoms mostly postprandial or sometimes during fasting.
The antibodies initially bind the postprandial released insulin to release
it again later but out of synchrony with the low ambient glucose level,
causing a hyperinsulinemic hypoglycemia. An initial evaluation shows
hypoglycemia with very high levels of insulin but also very high levels
of C-peptide and proinsulin. These dramatically elevated levels of insulin,
proinsulin and C-peptide should function as a clue for the clinician that
insulin autoantibodies could be present. Documentation of the insulin
antibody is all that is required to confirm the diagnosis, in a state of
endogenous hyperinsulinemic hypoglycemia [47]. Insulin antibodies
could also be seen in patients who use insulin for diabetes. Making the
evaluation of the hypoglycemic event (direction of C-peptide versus
insulin) primordial to differentiate between true insulin autoimmunity
and factious insulin mediated hypoglycemia, see Table 2.
4. Therapeutic management
The therapeutic management of hypoglycemia in non-diabetics con-
sists of two elements. First, an urgent treatment is warranted because
prolonged hypoglycemia carries a risk. Secondly, the underlying condi-
tion should be treated to prevent recurrence of hypoglycemic spells.
4.1. Urgent treatment
After venipuncture has obtained a diagnostic plasma sample, treat-
ment of the hypoglycemia is commenced. Unlike diabetic patients
where the goal of treatment is euglycemia, over-treating of hypoglycemia
in non-diabetics has no ill effects. The route of treatment depends on the
clinical status of the patient. Patients with advanced neuroglycopenic
419
symptoms might be unable to take oral glucose, necessitating intravenous
administration. Administering glucagon is efficient in most patients, but
not all patients. Glucagon functions as a stimulator of glycogenolysis, an
ineffective response illustrates a glycogen-depleted individual, such as
those with alcohol-induced hypoglycemia. Because of the higher cost
price, possible ineffectiveness and limited diagnostic value of glucagon
it should be reserved for hypoglycemia with severe neuroglycopenic
symptoms secondary to insulin. When the patient is awake and willing,
oral treatment with 15–20 g of glucose is a reasonable initial treatment.
In comatose patients IV glucose is preferred, an initial bolus of 25 g follow-
ed by continuous infusion of glucose 5 or 10% is a feasible treatment. Fre-
quent monitoring of plasma glucose response is required [5].
4.2. Prevention of recurrence
The prevention of recurrence requires an understanding of the hy-
poglycemic mechanism. Identification of the hypoglycemic mechanism
allows establishing of a diagnosis in a stepwise fashion, hereby focusing
therapy. Offending drugs can be discontinued or the dose can be re-
duced. One must keep the half-life of the drug in mind when initially
treating the hypoglycemia. But half-life's can be significantly influenced
by certain co-morbidities, e.g. sulfonylurea's half-life is increased during
kidney failure. In patients with a concurrent illness, the goal is to abolish
the primary disease. Critical illnesses can be managed with intensive
care. Non-beta cell tumors require an optimal oncologic treatment to
reduce tumor mass, although curative treatment is often difficult
[4,14,15]. GH and cortisol can be replaced when deficient. The therapy
for a hyperinsulinemic hypoglycemia in the seemingly well patient
differs according to the cause, this justifies the elaborated diagnostic
work-up. Surgical resection of an insulinoma resolves clinical symptoms
and when successfully removed, the patient life-expectancy is not
affected [6,8]. Treatment of autoimmune hyperinsulinemic hypoglyce-
mia by means of corticosteroids or immunosuppressive agents is diffi-
cult, making therapy often symptomatic by means of frequent feeding
and late night feeding to prevent fasting hypoglycemia. The disease is
often self-limiting in Asians [4].
The appropriate treatment choice for NIPHS is difficult. A balance
should be sought between the risk of the therapy and the degree of
debilitation caused by the hypoglycemic spells. Different therapeutic
algorithms have been proposed in the literature, but the common ratio-
nal is that the non-invasive therapies should be tried first. When these
fail and symptoms persist being debilitating, surgical treatment can be
considered. The medical treatment of NIPHS consists of diet modifica-
tion, and medications such as acarbose, diazoxide, verapamil and
somatostatin. Most evidence for success of medical treatment originates
from case-reports or small case-series [32,41,44,59,60]. These reports
indicate that in mildly symptomatic patients pharmacotherapy can
ameliorate symptoms. But little is known about how long improvement
can be sustained, and NIPHS was not always methodologically correct
documented in these studies [32,41,44,59,60]. Diet measures consist
of a modified low-carbohydrate diet with frequent feeding [43,61].
When dietary alterations fail a trial of pharmacotherapy with diazoxide
(agonist of KATP channel in β-cells, preventing β-cell depolarization
hereby reducing insulin secretion), acarbose (α-glucosidase inhibitor,
reducing carbohydrate digestion to monosaccharides), verapamil
(calcium-channel blocker) and octreotide (secretion inhibitor) can be
initiated. When medical management fails, and symptoms are debilitat-
ing, surgical partial pancreatectomy is the recommended treatment.
The balance between too much and too little resection is difficult. A
total pancreatectomy generates a state of severe apancreatic (type 3c)
diabetes [41]. A recent retrospective cohort study shows that patients
who underwent a partial pancreatectomy reported a significant
improvement in quality of life. The average follow-up period in the
study was 4.5 years. Patients were initially relieved from hypoglycemic
spells, but the vast majority of patients reported recurrence of symptoms,
albeit less debilitating and less severe, making partial pancreatectomy
420 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421
only a symptomatic therapy. Therefore it should only be considered in
patient who fails medical therapy and who suffers from severe debil-
itating neuroglycopenic symptoms [44].
5. Conclusion
Hypoglycemia is most often iatrogenic, caused by medications used
to lower plasma glucose (insulin and insulinsecretagogue). Spontane-
ous hypoglycemia in a non-diabetic patient is a less common clinical
syndrome, sometimes confronting the clinician with a diagnostic
enigma. Patients with spontaneous hypoglycemia suffer from greater
mortality-rates as compared with patients suffering from iatrogenic
hypoglycemia, indicating the greater vulnerability of this patient-
population. This hereby underscores the importance of a correct clinical
approach. Given the aspecificity of autonomic and neuroglycopenic
symptoms and signs during hypoglycemia, one should use Whipple's
triad to document hypoglycemia. Signs and symptoms consistent with
hypoglycemia, during a state of low plasma glucose (b 55 mg/dl), that
resolves after plasma glucose is raised, indicating fulfillment of this
triad. Once hypoglycemia is documented clinical practice guidelines
propose an evaluation, initiated by a thorough history and physical
examination. A review of medication can implicate offending drugs. A
physical examination identifies patients in distress with stigmata of
concurrent illnesses such as critical illnesses, hormone deficiencies
and non-β-cell tumors. Absence of these clinical findings narrows the
differential diagnosis to a subgroup of seemingly well patients. A
complete hypoglycemic evaluation during a hypoglycemic spell
(spontaneous or induced by an appropriate stress test) generates
the clues for a final diagnosis. The complete hypoglycemic evalua-
tion should comprise plasma glucose, insulin, C-peptide, pro-insulin,
β-hydroxybutyrate, and glucose increase after glucagon administration
and should screen for circulating oral hypoglycemic agents and insulin
antibodies. This approach allows identifying of exogenous sources of
hyperinsulinism, insulin autoimmunity and the biochemical entity of
endogenous hyperinsulinemic hypoglycemia. Failure to document a
focal cause of endogenous hyperinsulinemic hypoglycemia suggests
the diagnosis of NIPHS. Requiring a SPACI to confirm the diagnosis.
NIPHS is a relatively novel syndrome, for which the pathophysiology
is a topic of fierce debate. Irrespective of the exact cause of the sponta-
neous hypoglycemia, treatment consists of correcting the glycemic state
and preventing recurrence. Prevention of recurrence requires a funda-
mental understanding of the hypoglycemic mechanism. This justifies
an elaborate diagnostic work-up, enabling the clinician to problem-
orientate his therapy.
Conflict of interests
Both authors declare no conflict of interest.
Both authors declare that they are not receiving any financial
support, posing a potential conflict of interest.
References
[1] Budnitz DS, Lovegrove MC, Shehab N, Richards CL. Emergency hospitalizations for
adverse drug events in older Americans. New England Journal of Medicine
2011;365:2002–12.
[2] Whipple AO, Frantz VK. Adenoma if islet cells with hyperinsulinism: a review.
Annals of Surgery 1935;101:1299–335.
[3] WHIPPLE AO. The surgical therapy of hyperinsulinism. Journal International de
Chirurgie 1938;3:237–76.
[4] Cryer PE, Axelrod L, Grossman AB, Heller SR, Montori VM, Seaquist ER, et al.
Evaluation and management of adult hypoglycemic disorders: an Endocrine Society
Clinical Practice Guideline. Journal of Clinical Endocrinology and Metabolism
2009;94:709–28.
[5] Longo, Fauci, Kasper, Hauser, Jameson, Loscalzo. Harrison's: principles of internal
medicine18 ed. ; 2012 3003–9.
[6] Service FJ. Hypoglycemic disorders. New England Journal of Medicine 1995;332:
1144–52.
[7] Service FJ. Diagnostic approach to adults with hypoglycemic disorders. Endocrinology
and Metabolism Clinics of North America 1999;28:519–32 [vi].
[8] Vaidya A, Kaiser UB, Levy BD, Loscalzo J. Clinical problem-solving. Lying low. New
England Journal of Medicine 2011;364:871–5.
[9] Svensson AM, McGuire DK, Abrahamsson P, Dellborg M. Association between hyper-
and hypoglycaemia and 2 year all-cause mortality risk in diabetic patients with
acute coronary events. European Heart Journal 2005;26:1255–61.
[10] Pinto DS, Skolnick AH, Kirtane AJ, Murphy SA, Barron HV, Giugliano RP, et al. U-
shaped relationship of blood glucose with adverse outcomes among patients with
ST-segment elevation myocardial infarction. Journal of the American College of
Cardiology 2005;46:178–80.
[11] Kosiborod M, Inzucchi SE, Goyal A, Krumholz HM, Masoudi FA, Xiao L, et al. Relation-
ship between spontaneous and iatrogenic hypoglycemia and mortality in patients
hospitalized with acute myocardial infarction. JAMA 2009;301:1556–64.
[12] Placzkowski KA, Vella A, Thompson GB, Grant CS, Reading CC, Charboneau JW, et al.
Secular trends in the presentation and management of functioning insulinoma at the
Mayo Clinic, 1987–2007. Journal of Clinical Endocrinology and Metabolism
2009;94:1069–73.
[13] Murad MH, Coto-Yglesias F, Wang AT, Sheidaee N, Mullan RJ, Elamin MB, et al.
Clinical review: drug-induced hypoglycemia: a systematic review. Journal of Clinical
Endocrinology and Metabolism 2009;94:741–5.
[14] Fukuda I, Hizuka N, Ishikawa Y, Yasumoto K, Murakami Y, Sata A, et al. Clinical fea-
tures of insulin-like growth factor-II producing non-islet-cell tumor hypoglycemia.
Growth Hormone & IGF Research 2006;16:211–6.
[15] Tietge UJ, Schofl C, Ocran KW, Wagner S, Boker KH, Brabant G, et al. Hepatoma with
severe non-islet cell tumor hypoglycemia. American Journal of Gastroenterology
1998;93:997–1000.
[16] Hogan MJ, Service FJ, Sharbrough FW, Gerich JE. Oral glucose tolerance test com-
pared with a mixed meal in the diagnosis of reactive hypoglycemia. A caveat on
stimulation. Mayo Clinic Proceedings 1983;58:491–6.
[17] Vella A, Service FJ. Incretin hypersecretion in post-gastric bypass hypoglycemia—
primary problem or red herring? Journal of Clinical Endocrinology and Metabolism
2007;92:4563–5.
[18] Lev-Ran A, Anderson RW. The diagnosis of postprandial hypoglycemia. Diabetes
1981;30:996–9.
[19] Giurgea I, Ulinski T, Touati G, Sempoux C, Mochel F, Brunelle F, et al. Factitious hy-
perinsulinism leading to pancreatectomy: severe forms of Munchausen syndrome
by proxy. Pediatrics 2005;116:e145–8.
[20] Manning PJ, Espiner EA, Yoon K, Drury PL, Holdaway IM, Bowers A. An unusual
cause of hyperinsulinaemic hypoglycaemia syndrome. Diabetic Medicine
2003;20:772–6.
[21] Marks V, Teale JD. Hypoglycemia: factitious and felonious. Endocrinology and
Metabolism Clinics of North America 1999;28:579–601.
[22] Jaeger C, Winter S, Eckhard M, Hardt P, Brendel MD, Bretzel RG. Binding
characteristics and crossreactivity of insulin autoantibodies and insulin anti-
bodies directed to three different insulin molecules. Acta Diabetologica
2008;45:191–4.
[23] Yoshida R, Ohkubo K, Akehi Y, Harada S, Shinohara K, Kawashima H, et al. Clinical
application of the different cross-reactivities of anti-insulin antibodies to insulin
lispro to evaluate endogenous insulin secretion. Clinica Chimica Acta 2013;415:
250–4.
[24] Noone TC, Hosey J, Firat Z, Semelka RC. Imaging and localization of islet-cell tumours
of the pancreas on CT and MRI. Best Practice & Research Clinical Endocrinology &
Metabolism 2005;19:195–211.
[25] McLean A. Endoscopic ultrasound in the detection of pancreatic islet cell tumours.
Cancer Imaging 2004;4:84–91.
[26] Kauhanen S, Seppanen M, Minn H, Gullichsen R, Salonen A, Alanen K, et al. Fluorine-
18-L-dihydroxyphenylalanine (18F-DOPA) positron emission tomography as a tool
to localize an insulinoma or beta-cell hyperplasia in adult patients. Journal of Clinical
Endocrinology and Metabolism 2007;92:1237–44.
[27] Breeman WA, de BE, Sze CH, Konijnenberg M, Kwekkeboom DJ, Krenning EP. (68)
Ga-labeled DOTA-peptides and (68)Ga-labeled radiopharmaceuticals for positron
emission tomography: current status of research, clinical applications, and future
perspectives. Seminars in Nuclear Medicine 2011;41:314–21.
[28] Service FJ, Natt N, Thompson GB, Grant CS, van Heerden JA, Andrews JC,
et al. Noninsulinoma pancreatogenous hypoglycemia: a novel syndrome of
hyperinsulinemic hypoglycemia in adults independent of mutations in Kir6.2 and
SUR1 genes. Journal of Clinical Endocrinology and Metabolism 1999;84:1582–9.
[29] Laidlaw GF. Nesidioblastoma, the islet tumor of the pancreas. American Journal of
Pathology 1938;14:125–34.
[30] Palladino AA, Stanley CA. Nesidioblastosis no longer! It's all about genetics. Journal
of Clinical Endocrinology and Metabolism 2011;96:617–9.
[31] Thompson GB, Service FJ, Andrews JC, Lloyd RV, Natt N, van Heerden JA, et al.
Noninsulinoma pancreatogenous hypoglycemia syndrome: an update in 10 surgically
treated patients. Surgery 2000;128:937–44.
[32] Won JG, Tseng HS, Yang AH, Tang KT, Jap TS, Lee CH, et al. Clinical features
and morphological characterization of 10 patients with noninsulinoma
pancreatogenous hypoglycaemia syndrome (NIPHS). Clinical Endocrinology
2006;65:566–78.
[33] Witteles RM, Straus II FH, Sugg SL, Koka MR, Costa EA, Kaplan EL. Adult-onset
nesidioblastosis causing hypoglycemia: an important clinical entity and continuing
treatment dilemma. Archives of Surgery 2001;136:656–63.
[34] Anlauf M, Wieben D, Perren A, Sipos B, Komminoth P, Raffel A, et al. Persistent
hyperinsulinemic hypoglycemia in 15 adults with diffuse nesidioblastosis: diagnos-
tic criteria, incidence, and characterization of beta-cell changes. American Journal of
Surgical Pathology 2005;29:524–33.
 P. Martens, J. Tits / European Journal of Internal Medicine 25 (2014) 415–421
[35] Patti ME, McMahon G, Mun EC, Bitton A, Holst JJ, Goldsmith J, et al. Severe
hypoglycaemia post-gastric bypass requiring partial pancreatectomy: evidence for
inappropriate insulin secretion and pancreatic islet hyperplasia. Diabetologia
2005;48:2236–40.
[36] Raffel A, Krausch MM, Anlauf M, Wieben D, Braunstein S, Kloppel G, et al. Diffuse
nesidioblastosis as a cause of hyperinsulinemic hypoglycemia in adults: a diagnostic
and therapeutic challenge. Surgery 2007;141:179–84.
[37] Service GJ, Thompson GB, Service FJ, Andrews JC, Collazo-Clavell ML, Lloyd RV.
Hyperinsulinemic hypoglycemia with nesidioblastosis after gastric-bypass surgery.
New England Journal of Medicine 2005;353:249–54.
[38] Bernard B, Kline GA, Service FJ. Hypoglycaemia following upper gastrointestinal
surgery: case report and review of the literature. BMC Gastroenterology 2010;10:77.
[39] McLaughlin T, Peck M, Holst J, Deacon C. Reversible hyperinsulinemic hypoglycemia
after gastric bypass: a consequence of altered nutrient delivery. Journal of Clinical
Endocrinology and Metabolism 2010;95:1851–5.
[40] Meier JJ, Butler AE, Galasso R, Butler PC. Hyperinsulinemic hypoglycemia after gas-
tric bypass surgery is not accompanied by islet hyperplasia or increased beta-cell
turnover. Diabetes Care 2006;29:1554–9.
[41] Cui Y, Elahi D, Andersen DK. Advances in the etiology and management of
hyperinsulinemic hypoglycemia after Roux-en-Y gastric bypass. Journal of Gastroin-
testinal Surgery 2011;15:1879–88.
[42] Patti ME, Goldfine AB. Hypoglycaemia following gastric bypass surgery—diabetes
remission in the extreme? Diabetologia 2010;53:2276–9.
[43] Tack J, Arts J, Caenepeel P, De WD, Bisschops R. Pathophysiology, diagnosis and man-
agement of postoperative dumping syndrome. Nature Reviews Gastroenterology &
Hepatology 2009;6:583–90.
[44] Vanderveen KA, Grant CS, Thompson GB, Farley DR, Richards ML, Vella A, et al.
Outcomes and quality of life after partial pancreatectomy for noninsulinoma
pancreatogenous hypoglycemia from diffuse islet cell disease. Surgery 2010;
148:1237–45.
[45] Kloppel G, Anlauf M, Raffel A, Perren A, Knoefel WT. Adult diffuse nesidioblastosis:
genetically or environmentally induced? Human Pathology 2008;39:3–8.
[46] Uchigata Y, Hirata Y. Insulin autoimmune syndrome (IAS, Hirata disease). Annales
de Médecine Interne (Paris) 1999;150:245–53.
[47] Basu A, Service FJ, Yu L, Heser D, Ferries LM, Eisenbarth G. Insulin autoimmunity and
hypoglycemia in seven white patients. Endocrine Practice 2005;11:97–103.
[48] Archambeaud-Mouveroux F, Canivet B, Fressinaud C, de BB, Treves R, Laubie B. Auto-
immune hypoglycemia: the fault of pyritinol? Presse Médicale 1988;17:1733–6.
421
[49] Benson EA, Healey LA, Barron EJ. Insulin antibodies in patients receiving penicillamine.
American Journal of Medicine 1985;78:857–60.
[50] Blackshear PJ, Rotner HE, Kriauciunas KA, Kahn CR. Reactive hypoglycemia and
insulin autoantibodies in drug-induced lupus erythematosus. Annals of Internal
Medicine 1983;99:182–4.
[51] Cho BY, Lee HK, Koh CS, Min HK. Spontaneous hypoglycemia and insulin autoanti-
bodies in a patient with Graves' disease. Diabetes Research and Clinical Practice
1987;3:119–24.
[52] de Faguer MB, Burgard M, Boitard C, Desplanque N, Fanjoux J, Tchobroutsky G. Auto-
immune hypoglycemic syndrome induced by pyritinol. Diabète & Métabolisme
1988;14:423–9.
[53] Goswami R, Jayasuryan N, Jaleel A, Tandon N, Kochupillai N. Insulin autoantibodies
before and after carbimazole therapy in Asian Indian patients with Graves' disease.
Diabetes Research and Clinical Practice 1998;40:201–6.
[54] Lee YJ, Shin SJ, Torng JK, Liu WJ, Tsai JH. Insulin autoimmune syndrome in a
methimazole-treated Graves' patient with polyclonal anti-insulin autoantibodies:
report of a case. Taiwan Yi Xue Hui Za Zhi 1987;86:164–70.
[55] Moller DE, Ratner RE, Borenstein DG, Taylor SI. Autoantibodies to the insulin receptor
as a cause of autoimmune hypoglycemia in systemic lupus erythematosus. American
Journal of Medicine 1988;84:334–8.
[56] Redmon B, Pyzdrowski KL, Elson MK, Kay NE, Dalmasso AP, Nuttall FQ. Hypoglyce-
mia due to an insulin-binding monoclonal antibody in multiple myeloma. New
England Journal of Medicine 1992;326:994–8.
[57] Rouabhia S, Ramanoelina J, Godmer P, Reach G, Dutel JL, Guillevin L. Insulin autoim-
mune syndrome revealing systemic lupus erythematosus. Annales de Médecine
Interne (Paris) 2003;154:59–60.
[58] Varga J, Lopatin M, Boden G. Hypoglycemia due to antiinsulin receptor
antibodies in systemic lupus erythematosus. Journal of Rheumatology
1990;17:1226–9.
[59] Kapoor RR, James C, Hussain K. Advances in the diagnosis and management of
hyperinsulinemic hypoglycemia. Nature Clinical Practice Endocrinology & Metabolism
2009;5:101–12.
[60] Moreira RO, Moreira RB, Machado NA, Goncalves TB, Coutinho WF. Post-prandial
hypoglycemia after bariatric surgery: pharmacological treatment with verapamil
and acarbose. Obesity Surgery 2008;18:1618–21.
[61] Kellogg TA, Bantle JP, Leslie DB, Redmond JB, Slusarek B, Swan T, et al. Postgastric by-
pass hyperinsulinemic hypoglycemia syndrome: characterization and response to a
modified diet. Surgery for Obesity and Related Diseases 2008;4:492–9.