The alarming increase in the prevalence of obesity in children in the United States and globally raises major concerns about its future adverse impact on public health. One outcome of this disturbing trend that is already evident is the rapidly increasing incidence of type 2 diabetes at all ages. This disease, once thought to be nonexistent in children, is increasing coincident with obesity. This article addresses the role that obesity plays in type 2 diabetes and also explores its effects on other types of diabetes that occur in childhood. The new challenges for physicians who formulate a differential diagnosis of diabetes in children are discussed. Also examined are modifications of traditional diabetes treatment that can be helpful in combating the insulin resistance associated with obesity and that use medications that are not traditionally used in this age group. Cases are presented to illustrate certain points. An underlying thesis suggests that specific classification may not be as important to the clinician as the understanding of pathophysiologic factors that contribute to hyperglycemia in individual patients. Recommendations are offered to the practitioner for diagnosing and treating the obese child or adolescent with diabetes.
The United States and much of the Westernized and developing world are experiencing a marked increase in the prevalence of obesity.1 This increase is affecting all age groups but is, for many reasons, especially troublesome in the young.2–4 The potential public health and fiscal impacts of this phenomenon have been extensively discussed.5–7 The goal of this article is to address specifically how obesity can change the clinical presentation and course of diabetes in children and adolescents and how it has complicated the differential diagnosis and therapy.
PATHOPHYSIOLOGY AND DIAGNOSIS
Diabetes, for all of its varied clinical and metabolic manifestations, is primarily a disorder of glucose metabolism, and the diagnostic criteria are purely glycemic. The guidelines established by the American Diabetes Association8 and the World Health Organization9 are alike and based on (1) a fasting plasma glucose level of ≥126 mg/dL or 7.0 mmol/L, (2) an oral glucose tolerance test result of a 2-hour plasma glucose level of ≥200 mg/dL or 11.1 mmol/L, or (3) symptoms of diabetes combined with a plasma glucose level of ≥200 mg/dL or 11.1 mmol/L.
Classification systems have been proposed to subdivide the various types of diabetes, and these have expanded as more causes, both genetic and environmental, have been described. The relatively similar systems of the World Health Organization and the American Diabetes Association are presented in Table 1. The various disorders are discussed extensively in the references, but a few relevant comments are appropriate here. The general pediatrician will probably see individuals with cystic fibrosis and insulinopenic diabetes secondary to the pancreatic destruction that occurs in this disease. Their practices may also contain children who develop diabetes as a result of the insulin resistance that occurs during high-dosage glucocorticoid therapy for asthma or in the therapy of certain malignant disorders. Other medications can inhibit insulin secretion. In all of the categories, hyperglycemia occurs as a result of insufficient secretion of insulin, diminished response to insulin (insulin resistance), or a combination of both, as in type 2 diabetes.10 The majority of causes of diabetes in children fit into the former, or insulinopenic category, the most common being autoimmune type 1 diabetes.
Some of the recently characterized genetic forms of diabetes that affect children also deserve a brief special comment. This is especially true of maturity-onset diabetes of the young (MODY), because our understanding of it has undergone significant change since its initial description. This change has been in our molecular understanding of its causes and in our approach to therapy. This disorder is less common than type 1 but also results from insufficient insulin secretion. It was originally defined by Fajans11 as a dominantly inherited form of diabetes in young adults that would initially respond to sulfonylurea therapy. At first considered 1 disease, it has now been divided into 6 different disorders resulting from mutations of transcription factors in 5 and of an enzyme, glucokinase, in 1.12 The relatively newly characterized transient and permanent types of neonatal diabetes13 and mitochondrial diabetes14 also have insufficient insulin secretion as their primary etiologic defects. Diabetes that presents before 6 months of age with a persistent course (permanent neonatal diabetes) or with a course that is characterized by remission and later relapse (transient neonatal diabetes) has been well described15,16 and is, at least in their early stages, unaffected by obesity. As is discussed, the clinical course of type 1 diabetes or of 1 of the various types of MODY in children can be affected by the presence of obesity.
METABOLIC EFFECTS OF OBESITY
That obesity exerts a major effect on glucose metabolism is well established.17,18 The mechanism of obesity-induced insulin resistance, whether attributable to adipokine secretion,19,20 to extra-adipose lipid deposition,21 or to other processes, is not within the scope of this discussion. The focus here is rather on the impact of this reduction in insulin sensitivity on the age of onset and other clinical manifestations of diabetes in young individuals. Obesity and its associated insulin resistance have profound effects on glucose metabolism, requiring hypersecretion of insulin and chronic hyperinsulinemia to maintain normoglycemia in obese individuals without diabetes22,23 and frequently leading to the development of type 2 diabetes.10
OBESITY AND TYPE 2 DIABETES IN CHILDREN AND ADOLESCENTS
General awareness of type 2 diabetes in children has developed only during the past 10 to 15 years. Although documentation of this disease in children in the United States can be traced to 1968,24 there were very few reports before 1995 and the first review of the subject in the pediatric literature appeared in 1996.25 Prevalence data were available at that time for few populations, most of them being native inhabitants of North America.26 Prevalence was especially well documented in the Pima Indians at 22.3 in 1000 in 10- to 14-year-olds and 50.9 in 1000 in those from 10 to 19 years.27 In the second half of the decade of the 1990s, case report series from many US centers appeared and were collected and summarized in 2000 by Fagot-Campagna et al.28 By 2004, the increase in type 2 diabetes in children was documented as a global phenomenon.29 This increase has clearly mirrored the increase in obesity prevalence in this age group.3
DIFFERENTIAL DIAGNOSIS OF TYPE 2 DIABETES IN CHILDREN AND ADOLESCENTS
The documentation of type 2 diabetes as an increasingly frequent diagnosis in children created a dilemma for physicians who treat diabetes in children and adolescents. Until late in the last century, diabetes in this age group was considered to be type 1 diabetes, with an occasional case of MODY or of neonatal diabetes. With type 2 diabetes currently accounting for between 40% and 50% of new diabetes cases in adolescents in some regions of the United States,30,31 it is now a common diagnosis and 1 that should be differentiated from other forms of diabetes. Guidelines for distinguishing type 1 diabetes from type 2 diabetes have been published,32 and some useful differentiating features are summarized in Table 2, but the challenge to clinicians of making this differential diagnosis in some patients has also been recognized.33 Although obesity and race/ethnicity are proposed as helpful clinical discriminators, when patients have a typical type 2 diabetes clinical phenotype, neither of these criteria clearly separates those who have positive immune markers and would be considered to have type 1 from those who do not.34 Figure 1A presents data from children who were referred with clinical diagnoses of type 2 diabetes for inclusion in a drug trial of metformin, an oral antidiabetic agent. When the children who were found not to have markers of β-cell immunity and met the inclusion requirements for the study were compared with those who were found to have immune markers and were excluded from the study, there were no significant differences in the BMI percentiles for age and gender. Figure 1B, using the same subject population, compares the racial/ethnic composition by percentage of antibody-positive and -negative individuals. There is a slightly higher percentage of white individuals in the antibody-positive cohort compared with that in the antibody-negative group, The reverse was seen in the black and “other” populations, with each having slightly higher percentages of the antibody-negative group. This is as would be suspected from the epidemiologic data described previously, but the differences were so small that race/ethnicity cannot be used as a reliable distinguishing feature when obesity and other clinical features suggest a clinical diagnosis of type 2 diabetes. The difficulty of clearly separating type 2 diabetes from type 1 diabetes results from several factors, 1 of which is the lack of a specific test to confirm the diagnosis of type 2 diabetes, so it continues to be a clinical diagnosis. The presence of immune markers of β-cell disease is the most widely used criterion for confirming type 1 diabetes,36 and there is a strong correlation of these immune markers with predisposing genetic factors for type 1 diabetes and with its typical clinical course37; however the antibodies that are associated with type 1 diabetes are also present in the normal population without diabetes, in ∼10% of adults who received a clinical diagnosis of type 2 diabetes,38,39 and in 7% of 1 population selected for a characteristic type 2 phenotype.40
An argument has been made that the presence of positive immune markers in the young individual with diabetes confirms the diagnosis of type 1 diabetes, but this point of view is not universally accepted because obese adolescents with diabetes and positive immune markers are still being reported as having type 2 diabetes.41 Those who believe that the presence of an immune marker means that the individual has type 1 diabetes do not usually make this argument on the basis of the specific level of the titer but rather on its being “positive,” or above the laboratory reference range. The usual practice in adult diabetes care is to make the differential diagnosis from the clinical picture. An exception is the subdivision of latent autoimmune diabetes of adults for which specifically defining criteria have been suggested.42 The relative novelty of a type 2 phenotype with positive antibodies in some children has given rise to suggestions for an innovative nomenclature variously describing the condition as type 1.5,43 as type 3 (type 1 + type 2), or as double diabetes.44 These do not describe new forms of diabetes but rather are attempts to incorporate the blending of clinical and laboratory features that traditionally are associated with type 1 diabetes and type 2 diabetes in the same individual.
Another question raised by the concurrence of obesity and diabetes in young individuals is whether the added insulin resistance of obesity will cause type 1 diabetes to occur at an earlier age—the accelerator hypothesis.45,46 Insulin resistance that occurs in individuals with type 1 diabetes47 seems to influence the progression of antibody-positive individuals to type 1 diabetes,48 but this has not yet been specifically demonstrated with obesity. The SEARCH study did not find increased BMI to be associated with earlier onset of type 1 diabetes, except in children who already had compromised β-cell function as evidenced by diminished C-peptide levels.49 This finding does not eliminate the possibility that in individual cases of type 1 diabetes, as β-cells fail as a result of the autoimmune process, clinical diabetes will occur earlier in an obese, insulin-resistant state than in a lean, insulin-sensitive state. Indeed, curves representing the relationship among insulin sensitivity, insulin secretion, and diabetes predict that this should occur.50
Given the uncertainty about the definitive differential of type 1 diabetes versus type 2 diabetes in adolescents at the time of diagnosis, one can ask, “Is this distinction necessary for appropriate clinical care?” It could be important because of the risk for acute decompensation in individuals who have type 1 diabetes, and it has been argued that anyone who tests positive for immune markers has type 1 diabetes and should be treated as such. The data in adults suggest that individuals with diabetes and the type 2 diabetes phenotype can be treated for this disorder until endogenous insulin secretion is lost, regardless of the cause of the β-cell failure.
IMPACT OF OBESITY ON THERAPY OF DIABETES IN CHILDREN
Individuals who have type 1 diabetes and present with acute decompensation and diabetic ketoacidosis require appropriate therapy for the acute illness and insulin replacement from the time of diagnosis. The latter is necessary because those affected are not able to recover sufficient endogenous insulin secretion to reestablish euglycemia.
Traditionally, oral antidiabetic agents have had no role in the therapy of type 1 diabetes, but studies have reported that metformin use in conjunction with insulin in adolescents with poorly controlled diabetes improved their hemoglobin A1c (HbA1c) levels, also reducing their insulin requirement51 and increasing insulin-induced glucose uptake.52 The authors of 1 study also mentioned that 3 female patients who were overweight by BMI (>28 kg/m2) responded “particularly well.”52 Metformin, with its primary effect of diminishing hepatic glucose output, has been used effectively in adults with type 1 diabetes.53,54
The insulin-sensitizing thiazolidinediones are another consideration for reducing insulin resistance in obese patients. Rosiglitazone has been used in combination with insulin in overweight adult patients with type 1 diabetes.55 The benefit to the group that received rosiglitazone rather than placebo in conjunction with insulin was greater in individuals with BMI ≥30 kg/m2, with greater mean reductions in HbA1c level and in insulin dosage. One study that evaluated thiazolidinediones as adjunctive therapy in adolescents with type 1 diabetes and insulin requirements >0.9 U/kg per day found that pioglitazone did not improve glycemic control and increased the BMI in those who received it.56 Recent concerns about increased cardiovascular risk associated with thiazolidinediones in adult patients with type 2 diabetes57 and their potential adverse effects on bone mineralization suggest that any use in children must be cautious and carefully monitored.58
“HYBRID” THERAPY IN “HYBRID” DISEASE
In treating individuals who have diabetes that results from this “hybrid” disease—obesity and insulin resistance combined with another condition that compromises insulin secretion, such as MODY or autoimmune β-cell disease—is it better for physicians to attempt to place them in a specific diagnostic classification and use the treatment regimens proscribed for that disease or to treat the underlying causes of the hyperglycemia? This question can be directly addressed, and a case can be made in the context of several examples.
This boy presented at age 12 years 4 months with several months of nocturia, polydipsia, and weight loss. He was seen by his pediatrician, who documented glucosuria and referred him with a diagnosis of diabetes. He had been born in Cambodia and had come to the United States with his mother early in his first year of life. At the time of diagnosis, his mother was working and he was attending school but otherwise spending much of his time alone, eating, and watching television. He had been well but had gained a significant amount of weight before the onset of his symptoms.
The pregnancy that resulted in his birth had been unremarkable, and his birth weight was unknown. There was no diabetes in his mother's family, but little was known about his father's health or his father's family. The remainder of his medical history contributed little relevant information.
When first seen, he was a pleasant and cooperative Asian boy who was obese with a BMI of 36 kg/m2, >95th percentile for his age and gender. He had marked acanthosis nigricans involving his neck, axillae, antecubital areas, belt line, and inguinal regions. His genitalia and pubic hair were Tanner stage 2. His BP and the remainder of his physical examination were normal. His admission blood glucose level was 326 mg/dL, and blood chemistry results including a lipid panel were normal. His admission HbA1c level was 12%. Immune markers were negative for islet markers ICA512 and insulin autoantibodies but positive for glutamic acid decarboxylase. His C-peptide level on admission was 4.2 ng/mL (reference range: 0.5–2.0 ng/mL).
At this point, one could ask which type of diabetes was affecting this young man. Certainly the clinical picture is of type 2 diabetes with obesity, nonacute onset occurring during puberty, marked acanthosis nigricans, and evidence of increased insulin secretion. There is, however, the complication of his positive glutamic acid decarboxylase antibodies. If he is classified as having type 1 diabetes, then should he be treated with insulin? If we suggest instead that he has type 2 diabetes, then should we begin therapy that is appropriate for that disease? If his diabetes were to be classified by 1 of the “hybrid” names, then there are no guidelines for therapy.
Given his initial clinical presentation and with other data not immediately available, the decision was made to treat him with insulin to return him quickly to relative normoglycemia. When his elevated C-peptide level was reported, metformin was started with gradual reduction and elimination of his insulin replacement. Over several months, with his receiving 2 g of metformin daily, his HbA1c level dropped into the range of 8% to 9% and stabilized. The next change was in lifestyle, when he received a much requested bicycle as a present and his activity level increased significantly. According to his mother, he was constantly cycling, and his HbA1c level dropped into the 6.5% to 7.5% range. This pattern continued for 1 year until he irreparably damaged his bicycle and returned to his previous, less active lifestyle. Gradually his clinic visits became more irregular, and he was lost to follow-up until age 15, when he returned much thinner with a BMI of 24 kg/m2. His puberty was staged at Tanner 5, and the previously described acanthosis had disappeared. His HbA1c level was 11%, and a repeat test of his C-peptide level, after mixed-meal (Boost, Novartis Medical Nutrition, Minneapolis, MN) stimulation, revealed 0.4 ng/mL. He was restarted on multiple daily insulin injection therapy with frequent home glucose monitoring.
What is his diagnosis? The initial clinical picture was that of obesity with insulin resistance and type 2 diabetes, and he was treated with oral antidiabetic medications. Over several years, his insulin secretion diminished as did his obesity and his physical signs of insulin resistance. Was the diminished insulin secretion attributable to autoimmune destruction of his β-cells or to evolution of β-cell loss characteristic of type 2 diabetes? With the answer to this question unknown, the argument can be made that treating the major cause of his hyperglycemia was appropriate. Because his initial disease seemed to be the result of insulin resistance, he was treated as though he had type 2 diabetes and responded well. As his disease evolved and he lost his endogenous insulin secretion, he was treated with insulin replacement regardless of the cause of his insulinopenia—immune destruction of β-cells or progression of type 2 diabetes.
This case does not involve autoimmune diabetes but does illustrate the usefulness of combination therapy when a disorder that usually is treated for inadequate insulin secretion with an insulin secretogogue is complicated by obesity that requires additional treatment for insulin resistance. This white girl was referred at 10 years of age for evaluation of short stature and early sexual development. In eliciting the family history, it was discovered that her father, paternal grandfather, and paternal great grandmother had diabetes. Her father, treated with oral agents, had been found to have diabetic retinopathy at the age of 37 years. Given this 3-generation history of diabetes, even in the absence of symptoms, routine urinalysis was ordered and she was found to have glucosuria. A fasting blood glucose level was measured at 240 mg/dL, hyperglycemia was confirmed with a second test, and diabetes was diagnosed. Her physical examination at the initial visit was unremarkable, and her BMI was 27.2 kg/m2. There was no organic explanation for her mild short stature, and her pubertal development was not precocious.
She had genetic testing, which confirmed a diagnosis of MODY 3. She was initially started on glyburide and did well, with HbA1c levels in the 6.0% to 6.5% range. As she progressed through puberty, she began to gain weight and her BMI increased from its initial value to 41.5 kg/m2. As her weight increased and her glycemic control worsened, metformin was added to her regimen with reduction in her HbA1c level. She began a change in lifestyle, which resulted in a loss of 10 kg, and she was able to discontinue the metformin and maintain good glycemic control again with glyburide alone. She later regained weight and increased her BMI to 44 kg/m2. She was then restarted on metformin and has maintained HbA1c levels in the 8% to 10% range.
This girl's course illustrates the possible influence of obesity on the age of onset of her primary diagnosis of diabetes as a result of a MODY3 defect and a definite impact on its therapy. At the time of diagnosis, the primary cause of her disease was the deficient insulin secretion characteristic of MODY, although it could be argued that her slightly increased BMI contributed to the relatively early onset. She responded well to the sulfonylurea therapy recommended for individuals with MODY.59 Her subsequent weight gain added a component of insulin resistance that worsened her glycemic control and required the addition of another medication with a different mechanism, metformin. This medication, effective in type 2 diabetes with obesity, improved her glycemic control but was safely withdrawn when she lost weight.
Her clinical condition would not be called by any of the names mentioned previously and suggested the more common combination of a type 2 phenotype and positive immune markers, although the combination of obesity and insulin resistance with another process compromising insulin secretion was present. This makes the point that obesity can occur with and complicate the diagnosis, course, and therapy of different forms of diabetes whose primary defect is in insulin secretion.
The final case is similar to case 2 but adds the feature of a much earlier presentation with diabetes than is typically seen with either of the conditions present. This young boy presented at 5 years 11 months with a history of polyuria, polydipsia, and enuresis. His mother volunteered that he had drunk excessively and had a voracious appetite since early in his life. His symptoms and the documentation of hyperglycemia confirmed diabetes. His physical examination revealed a markedly obese (BMI: 36.4 kg/m2, >95th percentile for age and gender) blond boy who was hyperactive but otherwise well appearing. He had acanthosis nigricans in his axillae and at the base of his neck with no other abnormal physical findings.
His family history was remarkable for diabetes. His mother, relatively lean and from a family of Northern European heritage, had received a diagnosis of diabetes at 17 years of age. She had been started on a single injection of intermediate insulin and had remained on this regimen until the time of her son's diagnosis. Her mother and her maternal grandmother both had diabetes diagnosed in midlife and were being treated with oral agents that the mother believed was a sulfonylurea. None of these maternal family members was known to have microvascular disease. The patient's father was obese and of Mexican American heritage. Adult-onset diabetes and obesity were present in many members of his family.
The patient was initially placed on insulin, starting on 0.5 U/kg per day but requiring increases to >1.0 U/kg per day to control his hyperglycemia. When his admission laboratory data revealed no immune markers for type 1 diabetes and a C-peptide level of 3.7 ng/dL (reference range: 0.5–2.0 ng/dL), he was started on glyburide with reductions in his insulin dosage. Metformin was added after insulin was discontinued. Glyburide was stopped by the family when they ran out of the medication. His visits became erratic, because the family moved frequently, and his HbA1c level rose while he was receiving metformin alone. He was then lost to the clinic until 10 years of age. At this return visit, glyburide was restarted and his HbA1c level decreased.
The course of this child's diabetes is another illustration of the effect of obesity on insulinopenic diabetes. Although neither he nor his mother has been tested, the family history suggestive of dominant inheritance and the clinical courses suggest a form of MODY. The maternal passage of the diabetes could also indicate mitochondrial diabetes,60 but, in either case, obesity certainly contributed to the early onset. As in patient 2, obesity in combination with limited insulin secretion required combination therapy using an insulin secretagogue and a medication used in type 2 diabetes.
The increasing prevalence of obesity in children in the United States and throughout the world is adding many new health concerns for providers of pediatric care. One of the major concerns is its impact on diabetes. Physicians who care for children have to be concerned about the increasing prevalence of type 2 diabetes and its treatment and comorbidities, but they also have to be aware of the effect of obesity on the presentation, course, and treatment of other forms of diabetes that typically result from diminished or absent insulin secretion.
The major consequence of obesity on glucose metabolism is increased insulin resistance. When severe enough, this can lead to impaired glucose tolerance and then to diabetes as it does in type 2 diabetes and in syndromes of severe insulin resistance, but obesity and its associated insulin resistance can also complicate the differential diagnosis of new-onset diabetes and its subsequent treatment. As illustrated in the cases described, individuals can present with the clinical phenotype of type 2 diabetes, have positive antibodies, respond to medications that are used to treat type 2 diabetes, and then evolve to a type 1 diabetes picture requiring insulin replacement. Similarly, individuals with a confirmed diagnosis of MODY can respond appropriately to a sulfonylurea and then develop obesity that requires addition of another agent to treat the associated insulin resistance.
Extensive discussion has been devoted to the differential diagnosis of diabetes in obese children and adolescents. How does one classify this combination, or what does one call it? A different question that may be asked is whether classification matters to the treating physician. This is not to discount the importance of differentiating and classifying diabetes in children, but, for the treating physician, it may be more important to understand the pathophysiology underlying the hyperglycemia in individual patients and to treat it appropriately. If a child presents with insulin deficiency and there is no suspicion that he or she has 1 of the forms that may respond to an insulin secretagogue, then insulin replacement is important and should be initiated. If the family history suggests a genetic form of diabetes that responds to sulfonylureas, then a trial is indicated. If a patient presents with a phenotype indicative of type 2 diabetes and has evidence of normal insulin secretion or hypersecretion, then that patient can be treated with type 2 therapy until such time as there is insufficient insulin secretion, regardless of the cause of the insulin deficiency. Conversely, if a patient who is being treated for a form of diabetes as a result of insulin deficiency becomes obese and develops difficulty with control or requires large insulin doses, then therapy to reduce the effects of the insulin resistance produced by obesity can be successfully used.
As molecular medicine increases the number of recognized causes of diabetes, specific diagnosis and classification becomes more complex, especially for the pediatrician who must incorporate this complexity into an already complex practice. Obesity complicates this task further by adding type 2 diabetes to the diagnosis list and by superimposing insulin resistance onto other clinical pictures. The guideposts suggested here for diagnosis and treatment in primary care pediatrics include the following:
recognizing uncomplicated type 1 diabetes and treating it appropriately;
recognizing the hereditary forms of MODY and neonatal diabetes that respond to sulfonylureas and implementing this therapy;
recognizing that obese patients who have the phenotype of type 2 diabetes and have hyperinsulinemia can be treated for type 2 diabetes regardless of whether they have immune β-cell markers; and
recognizing that children who have diabetes that is produced by insulinopenia and are or become obese may benefit from drugs that reduce hepatic glucose output (metformin).
I express my appreciation to Dr Michael Gottschalk and Dr Martin Stein for review of the manuscript and helpful suggestions.
- Accepted July 25, 2007.
- Address correspondence to Kenneth Lee Jones, MD, MC 5103, 3020 Children's Way, San Diego, CA 92123-4282. E-mail:
The author has indicated he has no financial relationships relevant to this article to disclose.
- ↵World Health Organization. Obesity: Preventing and Managing the Global Epidemic. Geneva, Switzerland: World Health Organization; 2000. WHO technical report series 894
- ↵Strauss RS, Pollack HA. Epidemic increase in childhood overweight, 1986–1998. JAMA.2001;286(22) :2845– 2848
- ↵Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States. 1999–2004. JAMA.2006;295(13) :1549– 1555
- ↵Allison DB, Fontaine KR, Manson JE, Stevens J, VanItallie TB. Annual deaths attributable to obesity in the United States. JAMA.1999;282(16) :1530– 1538
- National Heart Lung and Blood Institute Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults: the evidence report. National Institutes of Health. Obes Res.1998;6(suppl 2) :51S– 209S
- ↵Thorpe KE, Florence CS, Howard DH, Joski P. Impact of obesity on rising medical spending: Health Tracking 2004 American Diabetes Association Position Statement. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care.2006;29(suppl 1) :S43– S48
- ↵World Health Organization: Diabetes Mellitus: Report of a WHO Study Group. Geneva, Switzerland: World Health Organization; 1985. WHO technical report series 727
- ↵Frayling TM, Lindgren CM, Chevre JC, et al. A genome-wide scan in families with maturity-onset diabetes of the young: evidence for further genetic heterogeneity. Diabetes.2003;52 (3):872– 881
- ↵Maassen JA, 'T Hart LM, van Essen E, et al. Mitochondrial diabetes: molecular mechanisms and clinical presentation. Diabetes.2004;53(suppl 1) :S103– S109
- ↵Sperling MA. The genetic basis of neonatal diabetes mellitus. Pediatr Endocrinol Rev.2006;(4 suppl 1) :71– 75
- ↵Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature.2006;444(7121) :840– 846
- ↵Shuldiner AR, Yang R, Gong DW. Resistin, obesity, and insulin resistance: the emerging role of the adipocyte as an endocrine organ. N Engl J Med.2001;345(18) :1345– 1346
- ↵Chaldakov GN, Stankulov IS, Hristova M, Ghenev PI. Adipobiology of disease: adipokines and adipokine-targeted pharmacology. Curr Pharm Des.2003;9(12) :1023– 1031
- ↵Yki-Jarvinen H. Ectopic fat accumulation: an important cause of insulin resistance in humans. J R Soc Med.2002;95(suppl 42) :39– 45
- ↵Li C, Ford ES, McGuire LC, Mokdad AH, Little RR, Reaven GM. Trends in hyperinsulinemia among nondiabetic adults in the U.S. Diabetes Care.2006;29(11) :396– 402
- ↵Paulson EP, Richenderfer L, Ginsberg-Fellner F. Plasma glucose, free fatty acids, and immunoreactive insulin in sixty-six obese children. Diabetes.1968;17 (5):261– 269
- ↵Glaser N, Jones KL. Non-insulin-dependent diabetes in children and adolescents. Adv Pediatr.1996;43 :359– 396
- ↵Dean H. NIDDM-Y in First Nation children in Canada. Clin Pediatr (Phila).1998;37 (2):89– 96
- ↵Savage PJ, Bennett PH, Senter RG, Miller M. High prevalence of diabetes in young Pima Indians. Diabetes.1979;28(10) :937– 942
- ↵Oeltmann JE, Liese AD, Heinze HJ, Addy CL, Mayer-Davis EJ. Prevalence of diagnosed diabetes among African-American and non-Hispanic white youth. Diabetes Care.2003;26 (9):2531– 2535
- ↵Phillips SA, Jones KL. Type 2 diabetes in children and adolescents. In: Daneman D, Hamilton J, eds. Insulin Resistance in Children and Adolescents. Huppauge, NY: Nova Science Publishers, Inc.; 2005:121– 162
- ↵Jones KL, Eisenbarth G, Klingensmith G, Fiedorek FT. Comparison of type 1 marker positive and negative children referred with presumptive type 2 diabetes. In: Proceedings of the 84th Annual Endocrine Society Meeting. P2-357, 2002:404
- ↵Vähäsalo P, Knip M, Karjalainen J, Tuomilehto-Wolf E, Lounamaa R, Akerblom HK. The Childhood Diabetes in Finland Study Group: islet cell-specific autoantibodies in children with insulin-dependent diabetes mellitus and their siblings at clinical manifestation of the disease. Eur J Endocrinol.1996;135 (6):689– 696
- ↵Jones KL, Arslanian S, Peterokova VA, Park J-S, Tomlinson MJ. Effect of metformin in pediatric patients with type 2 diabetes: a randomized controlled trial. Diabetes Care.2002;25 (1):89– 94
- ↵Sabbah E, Savola K, Ebeling T, et al. Genetic, autoimmune and clinical characteristics of childhood- and adult-onset type 1 diabetes. Diabetes Care.2000;23 (9):1326– 1332
- ↵Turner R, Stratton I, Horton V, et al. UKPDS 25: autoantibodies to islet-cell cytoplasm and glutamic acid decarboxylase for prediction of insulin requirement in type 2 diabetes. UK Prospective Diabetes Study Group. Lancet.1997;350(9087) :1288– 1293
- ↵Hathout EH, Thomas W, El-Shahaway M, et al. Diabetic autoimmune markers in children and adolescents with type 2 diabetes. Pediatrics.2001;107 (6). Available at: www.pediatrics.org/cgi/content/full/107/6/e102
- ↵Palmer JP, Hirsch IB. What's in a name: latent autoimmune diabetes of adults, type 1.5, adult-onset, and type 1 diabetes. Diabetes Care.2003;26 (2):536– 538
- ↵Kibirige M, Metcalf B, Renuja R, Wilkin TJ. Testing the accelerator hypothesis: the relationship between body mass and age at diagnosis of type 1 diabetes. Diabetes Care.2003;26(10) :2865– 2870
- ↵Fourlanos S, Narendran P, Byrnes GB, Colman PG, Harrison LC. Insulin resistance is a risk factor for progression to type 1 diabetes. Diabetologia.2004;47(10) :1661– 1667
- ↵Dabelea D, D'Agostino RB Jr, Mayer-Davis EJ, et al. Testing the accelerator hypothesis: body size, beta-cell function, and age of onset of type 1 (autoimmune) diabetes. Diabetes Care.2006;29 (2):290– 294
- ↵Kahn SE, Prigeon RL, McCulloch DK. Quantification of the relationship between insulin sensitivity and beta-cell function in human subjects: evidence for a hyperbolic function. Diabetes.1993;42(11) :1663– 1672
- ↵Hamilton J, Cummings E, Zdravkovic V, Finegood D, Daneman D. Metformin as an adjunct therapy in adolescents with type 1 diabetes and insulin resistance. Diabetes Care.2003;26 (1):138– 143
- ↵Särnblad S, Kroon M, Åman J. Metformin as additional therapy in adolescents with poorly controlled type 1 diabetes: randomized placebo-controlled trial with aspects on insulin sensitivity. Eur J Endocrinol.2003;149 (4):323– 329
- ↵Coscelli C, Palmari V, Saccardi F, Alpi O, Bonora E. Evidence that metformin addition to insulin induces an amelioration of glycaemic profile in type 1 (insulin-dependent) diabetes mellitus. Curr Ther Res.1984;35 (6):1058– 1064
- ↵Strowig SM, Raskin P. The effect of rosiglitazone on overweight subjects with type 1 diabetes. Diabetes Care.2005;28 (7):1562– 1567
- ↵Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med.2007;356(24) :2457– 2471
- ↵Maassen JA, 'T Hart LM, van Essen E, et al. Mitochondrial diabetes: molecular mechanisms and clinical presentation. Diabetes.2004;53(suppl 1) :S103– S109
- ↵Clemons D, Gottschalk ME, Jones KL. Basal and post prandial C-peptide values in a racially mixed population of adolescents with type 2 diabetes mellitus. Diabetes.1999;48(suppl 1) :80A
- Copyright © 2008 by the American Academy of Pediatrics