Division of Endocrinology, Beth Israel Deaconess Medical Center and Joslin Diabetes Center, Boston, Massachusetts 02215
Address all correspondence and requests for reprints to: Christos S. Mantzoros, M.D., D.Sc., Division of Endocrinology, RN 325, 99 Brookline Avenue, Boston, Massachusetts 02215. E-mail: cmantzor{at}bidmc.harvard.edu
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
![]() |
Definition and in vivo assessment of insulin resistance |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although assessment of either fasting or peak insulinemia after OGTT provides a convenient, readily available means of classifying individuals into normal, mild to moderate, and severe insulin resistant (5), the results of this test must be interpreted in the context of plasma glucose levels, because the presence of any degree of hyperglycemia suggests the presence of defects in insulin secretion, further exacerbates insulin resistance, and invalidates the degree of insulinemia as an index of insulin resistance (5). Fasting insulin levels above 5070 µU/mL or peak (post-OGTT) insulin levels above 350 µU/mL suggest severe insulin resistance, in contrast to the fasting serum insulin levels below 20 µU/mL or peak (post-OGTT) insulin levels below 150 µU/mL observed in normal individuals (5). Similarly, the rate and degree of plasma glucose fall in response to ITT are dependent not only on insulin sensitivity, but also on the presence and magnitude of the counterregulatory hormone response (including epinephrine, glucagon, cortisol, and GH) (5), thus decreasing the value of ITT in assessing insulin sensitivity per se.
In contrast, the assessment of an index of insulin sensitivity (Si) by employing the minimal model kinetic analysis to data obtained from the FSIVGTT appears to represent a more accurate means of quantifying insulin sensitivity (6). In this test, an iv injection of a fixed amount of glucose is followed by frequent blood sampling over 180 min and subsequent modeling of the relevant plasma glucose and insulin data to derive the indexes of insulin sensitivity (Si) and glucose effectiveness (Sg) (6). Si index values below 2 x 104 µU/mL·min typically occur in the presence of severe insulin resistance, whereas values above 5 x 104 µU/mL·min are observed in normal subjects (4). The Si correlates well with the insulin-mediated glucose disposal rate (M), as determined by the euglycemic hyperinsulinemic clamp (7). The latter, considered to be the gold standard in the assessment of insulin resistance, involves the concurrent iv infusion of insulin at a fixed rate (usually raising plasma insulin levels to either 100 or 1000 µU/mL) and glucose at a variable rate, as necessary to maintain normoglycemia (8). Upon reaching steady state, the glucose disposal rate (M) is proportional to the exogenous glucose infusion rate (8). Patients with severe insulin resistance have M rates below 2 mg/kg·min, compared with M rates above 6 mg/kg·min in normal individuals (5). Intermediate values typify mild to moderate insulin resistance.
![]() |
Clinical phenotypes |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
All syndromes of severe insulin resistance share a number of laboratory findings. Among these, hyperinsulinemia, resulting from increased insulin secretion to compensate for the peripheral insulin resistance and (in many cases) reduced insulin clearance, is by far the most consistent finding (1, 3, 4). Additionally, impaired glucose tolerance or frank diabetes mellitus commonly, but not universally, occur at a later stage in the natural course of these syndromes (1, 3, 4). These manifestations depend on the ability of the pancreas to compensate for the peripheral insulin resistance by increasing insulin secretion (1, 3, 4).
Unique features associated with each syndrome have been recognized as a result of extensive studies and have led to the classification of patients with severe insulin resistance into several distinct phenotypes. The term type A syndrome was originally applied to lean adolescent female patients with severe insulin resistance, acanthosis nigricans, severe ovarian hyperandrogenism, and decreased insulin binding to circulating leukocytes (4) and is currently used for both female and male patients with severe inherited insulin resistance and acanthosis nigricans in the absence of autoantibodies to the insulin receptor (IR) (3, 5). Postpubertal females also have evidence of mild to severe androgen excess of ovarian origin, ranging from hirsutism, acne, oligoamenorrhea, and infertility to frank virilism with markedly elevated testosterone levels (1, 3, 4, 5). Additional, but not invariable, features of the syndrome include short stature, acral hypertrophy, accelerated linear growth, muscle cramps, and retinitis pigmentosa (1, 3, 4, 5).
In contrast to the typically early onset of the type A syndrome, patients afflicted with the type B syndrome are commonly middle-aged at presentation and, in addition to the common features of severe insulin resistance (i.e. abnormal glucose homeostasis, acanthosis nigricans, and ovarian hyperandrogenism), often demonstrate features associated with autoimmunity, including vitiligo, alopecia areata, arthritis, nephritis, and primary biliary cirrhosis, as well as Hodgkins disease and ataxia-telangiectasia (1, 3, 4, 5). These patients may present with fasting hypoglycemia (with or without postprandial hyperglycemia) or may develop hypoglycemia during the course of their disease, even subsequent to a period of hyperglycemia and diabetes (1, 3, 4, 5, 10). In addition to nonspecific laboratory findings, including elevated erythrocyte sedimentation rate, leukopenia, hypergammaglobulinemia, serum antinuclear antibodies, and proteinuria, these patients demonstrate the presence of anti-IR antibodies in the plasma (1, 3, 4). Commonly, anti-IR antibody titers are in proportion to the magnitude of insulin resistance (1, 3, 4). These anti-IR antibodies are the diagnostic hallmark of the type B syndrome and explain several of its manifestations, as will be discussed later. The type B syndrome is quite distinct from the resistance to exogenous (usually of animal origin) insulin, which occurs as a result of developing antiinsulin antibodies that bind insulin and prevent its interaction with IRs (3).
In addition to the type A and B syndromes, the term HAIR-AN (hyperandrogenism, insulin resistance, and acanthosis nigricans) has been applied to women with all of the above features, often in association with obesity (5). However, it has not yet been fully clarified whether this syndrome represents a distinct entity from other syndromes of severe insulin resistance, such as the type A and B syndromes, or PCOS (5), and it will not be discussed in detail.
Another very rare syndrome associated with severe insulin resistance was initially described by Mendenhall (3, 11) and is currently known as the Rabson-Mendenhall syndrome. These patients present in childhood with severe insulin resistance and diabetes mellitus (commonly refractory to large doses of insulin), acanthosis nigricans, abnormal nails and dentition, accelerated linear growth, precocious pseudopuberty, and, ostensibly, pineal hyperplasia (3, 11). Prognosis is generally poor, mainly due to the development of severe microvascular complications of diabetes (3).
Leprechaunism was first recognized in 1954 and is characterized by severe intrauterine and postnatal growth retardation and failure to thrive, lipoatrophy, dysmorphic features (globular eyes, large ears, and micrognathia), and acanthosis nigricans (3, 12). These infants have massive hyperinsulinemia, often associated with glucose intolerance or frank diabetes mellitus, in addition to fasting hypoglycemia (3, 12). Additionally, affected female infants commonly have hirsutism and clitoromegaly, whereas affected males commonly present with penile enlargement (3, 12). Other features of this syndrome include dysmorphic lungs, renal disease, and breast hyperplasia (3). Few of these infants live beyond the first year of life, although a few may survive until adolescence (3, 12).
The lipodystrophy syndromes represent a diverse group of disorders characterized by severe insulin resistance and associated with severe hypertriglyceridemia leading to pancreatitis, and fatty infiltration of the liver leading to cirrhosis (3, 5). These syndromes have been conveniently subclassified according to the extent and the location of the lipodystrophy and the age of onset (3, 5). Specifically, newborns or infants with congenital generalized lipodystrophy (Berardinelli-Seip syndrome), an autosomal recessive condition, lack adipose tissue completely in both sc and visceral locations and commonly manifest impaired glucose tolerance or diabetes, accelerated linear growth, precocious puberty, muscular hypertrophy, and hypertriglyceridemia (3, 13). In contrast to the Berardinelli-Seip syndrome, patients with acquired total lipodystrophy (Lawrence syndrome) appear normal at birth, but develop lipoatrophy over days to weeks, sometimes after an infectious prodrome (3). Histological evidence of panniculitis has suggested an inflammatory etiology for this syndrome, although this remains to be demonstrated (3). In addition to the above variants of generalized lipodystrophy, several forms of partial lipodystrophy have been recognized and affect specific body areas. Thus, face-sparing lipodystrophy (Kobberling-Dunnigan syndrome), an X-linked (or, rarely, autosomal dominant) condition, spares the face, which is typically full, in contrast to the lipoatrophic trunk and extremities (3, 14). Another form of partial lipodystrophy occurs in association with mandibuloacral dysplasia and joint contractures and is termed lipodystrophy with other dysmorphic features (3). Additionally, a sporadic form of partial lipodystrophy, named cephalothoracic lipodystrophy, has been described predominantly in women and occurs in association with messangiocapillary glomerulonephritis, presumably as a result of complement activation (3).
Another rare syndrome of severe insulin resistance that was recently described and characterized is insulin resistance in association with acromegaloidism (15). In addition to severe insulin resistance, these patients have features reminiscent of acromegaly, including coarse facies and bone thickening, despite a GH-IGF-I axis that appears to be normal (3, 15). However, whether these physical findings result from high insulin levels signaling through the IGF-I receptor or, alternatively, the IR per se remains to be established (3, 15).
Finally, a number of rare genetic syndromes are associated with severe insulin resistance. Among them, Alstrom syndrome, an autosomal recessive disorder, presents with retinitis pigmentosa, sensorineural deafness, hypogonadism, and obesity and is commonly associated with severe insulin resistance and acanthosis nigricans (3). Myotonic dystrophy, an autosomal dominant condition that presents with progressive muscular dystrophy, myotonia, mild mental retardation, baldness, cataracts, and postpubertal testicular atrophy, has been associated with severe insulin resistance (3). Werners syndrome, a progeria syndrome, presents with bird-like facies, gray hair, cataract formation, slender extremities, and severe insulin resistance (16).
![]() |
Pathogenesis of severe insulin resistance |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Family studies indicate an autosomal dominant or autosomal recessive pattern of transmission of the type A syndrome, with variable penetrance (1, 3, 4, 5). Furthermore, in vivo studies of such patients show increased hepatic glucose output and diminished insulin-mediated glucose disposal rates (1, 3, 4, 5). Such findings are further corroborated by in vitro evidence of impaired insulin binding and action at the cellular level (1, 3, 4, 5, 19, 20). Although several IR mutations have been previously associated with the type A syndrome (18, 21), it currently appears that most patients with this syndrome do not possess such mutations, implying the presence of other critical primary defects in insulin signaling (22, 23), as seems to be the case in patients with NIDDM (24). Moreover, it appears that the correlation between genotype and phenotype of patients with severe insulin resistance is imprecise, as suggested by the presence of the same (Leu 193 Pro) mutation of the IR gene in a patient with the type A syndrome and in another with the Rabson-Mendenhall syndrome (24).
In addition to mutations of the IR gene, a transmembrane protein named PC-1 has been proposed as the cause of the type A syndrome in one patient (18, 25), possibly by interfering with IR tyrosine kinase activity (18). However, its significance in the pathogenesis of severe insulin resistance has not been conclusively demonstrated (18). Additionally, excessive IR serine phosphorylation has been implicated as a potential mechanism for insulin resistance in a subset of PCOS patients (2).
As mentioned above, anti-IR antibodies occur in association with the type B syndrome, presumably as a result of either loss of immune tolerance or generation of an immune response to an exogenous antigen and autoantibody formation through molecular mimicry (3). These antibodies can lead to insulin resistance by sterically interfering with insulin binding (3), although some anti-IR antibodies appear to lead to IR activation, explaining the fasting hypoglycemia that may occur in these patients (1, 3, 4, 5, 10). Additionally, defects of signaling intermediates distal to the IR are increasingly being demonstrated in a minority of patients with severe insulin resistance, including the presence of an IRS-1 mutation in such a patient, although its etiological significance remains unclear (18). More recently, selective impairment of insulin-stimulated phosphoinositide 3-kinase activity was demonstrated in three patients with severe insulin resistance and pseudoacromegaly (26).
Although an autosomal recessive mode of transmission has been suggested
for the Berardinelli-Seip syndrome (3, 5), the pathogenesis of
associated insulin resistance is poorly understood, and it remains
unclear whether insulin resistance is primary or occurs secondary to
lipodystrophy. Linkage analysis in 10 families with congenital
lipodystrophy failed to implicate 14 candidate genes, the IR, IRS-1,
and IGF-I genes among them (27). According to Randles hypothesis (or
cycle), excessive plasma FFA may lead to insulin resistance by
decreasing peripheral glucose utilization and increasing hepatic
gluconeogenesis (3). In addition, the recent demonstration of insulin
resistance in white and brown adipose tissue-diphtheria toxin A-ablated
(BP2-DTA) mice, whose adipose tissue was completely absent (28),
suggests that insulin resistance may indeed be secondary to the lack of
adipose tissue and raises the hope that the etiologies for human
lipodystrophy may be elucidated soon.
The pathogenesis of severe insulin resistance in patients with the previously mentioned rare genetic syndromes is unclear, although impaired insulin binding has been demonstrated in insulin target tissues from patients with myotonic dystrophy (3), and impaired postreceptor insulin signaling has been shown in patients with Werners syndrome (16).
![]() |
Treatment |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Moreover, drug therapy for patients with severe insulin resistance syndromes is currently unsatisfactory. Insulin is often administered in very high doses, but usually fails to provide adequate glycemic control (3, 29). Similarly, administration of sulfonylureas to patients with severe insulin resistance has failed to show significant benefits, probably because these agents act primarily by augmenting insulin secretion, which is already increased in patients with severe insulin resistance (29).
Administration of IGF-I, which may act by binding to either the IGF-I receptor or a functioning IR, has been attempted in patients with several syndromes of severe insulin resistance, including the type A or B syndromes, the Rabson-Mendenhall syndrome, leprechaunism, and lipodystrophy (29, 30, 31) (Mantzoros, C. S., and A. C. Moses, unpublished observations) and has led to improvement in glycemic control and decrease in fasting insulin levels in short term studies (29) (Mantzoros, C. S., and A. C. Moses, unpublished observations). However, some of these beneficial effects were not maintained in a 10-week trial (30). Moreover, IGF-I administration is occasionally associated with acute side-effects, such as fluid retention, carpal tunnel syndrome, and jaw pain. In addition, there is concern that IGF-I administration may exacerbate the development of microvascular complications, particularly retinopathy, in patients with diabetes (29), and endogenous IGF-I has recently been associated with breast and prostate cancer (32, 33). Thus, its efficacy-safety profile in patients with severe insulin resistance remains unclear.
Agents that improve insulin sensitivity present attractive candidates
for the treatment of individuals with severe insulin resistance.
Specifically, metformin, a biguanide that suppresses hepatic glucose
output and increases insulin-mediated glucose disposal, has been
shown to improve glycemia in patients with the type B syndrome or
lipoatrophic diabetes (29), but did not improve the insulin resistance
in patients with myotonic dystrophy (29). More recently, troglitazone,
a thiazolidinedione that improves insulin sensitivity in NIDDM, was
shown to improve insulin resistance in patients with Werners syndrome
(34) and is currently being studied in individuals with other syndromes
of severe insulin resistance, including the HAIR-AN syndrome.
Furthermore, administration of vanadate or vanadium salts to patients
with NIDDM has led to improvement in glycemic profile and peripheral
insulin resistance (29, 35), although the roles of these compounds in
patients with severe insulin resistance remain unclear. Limited data
suggest an improvement in insulin sensitivity in response to
administration of phenytoin to patients with the type A syndrome (29).
Additionally, functional activation of a mutant IR, obtained from a
patient with the Rabson-Mendenhall syndrome, by a monoclonal antibody
in vitro led to improved IR autophosphorylation and glycogen
synthesis in vitro, raising hopes that such therapy may
benefit patients with severe insulin resistance (36). Small studies
suggest improvement in insulin sensitivity of patients with
lipodystrophic diabetes in response to administration of bezafibrate
(37) or dietary supplementation with -3 fatty acid-rich fish oil
(29), possibly by interfering with Randles cycle (3). Finally,
immunosuppressants and plasmapheresis have been tried in some patients
with the type B syndrome with beneficial results (3).
![]() |
Future directions |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Footnotes |
---|
Received May 4, 1998.
Revised June 5, 1998.
Accepted June 16, 1998.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|