The 1985 edition of a major textbook of pediatrics described two forms of childhood diabetes mellitus: the autoimmune form of juvenile (now called type 1) diabetes, and a monogenic form of diabetes called maturity onset diabetes of the young (MODY). There was no mention of adult onset or noninsulin dependent type 2 diabetes; the condition was thought to affect only mature adults.
During the past 20 years, the pediatric community has witnessed a worldwide surge in the prevalence of type 2 diabetes among children, adolescents, and young adults. The causes of this mini-epidemic are manifold, encompassing genetic, environmental, and nutritional factors, as well as changes in lifestyle that modulate energy expenditure. The objectives of this chapter are: (a) to discuss controversies related to the diagnosis of type 2 diabetes in childhood; (b) to delineate factors that play important roles in disease pathogenesis; (c) to describe potential complications that may arise in the short and long terms; and (d) to outline approaches to disease prevention and treatment.
Many children and adolescents with type 2 diabetes are asymptomatic at the time of diagnosis and for long periods of time thereafter. Consequently, the condition is defined by measures of blood glucose that lie above the normal range. With increasing recognition of the long-term risks of mild elevations of blood-glucose concentrations in adults, the definition of diabetes has changed during the past generation. The diagnosis is established if fasting blood-glucose concentrations exceed 125 mg% and if postprandial or postdextrose challenge glucose concentrations exceed 200 mg%. Isolated fasting hyperglycemia must be confirmed on a subsequent day, while postprandial or postchallenge hyperglycemia must be accompanied by diabetic symptoms, such as polyuria or polydipsia. The transient hyperglycemia caused by acute stress, illness, pancreatitis, or certain medications (e.g., glucocorticoids, thi-azides) must be excluded.
Initially, it seemed that differentiation of type 2 diabetes from type 1 diabetes and MODY would be straightforward. Children with type 1 diabetes often presented emer-gently at an early age with ketoacidosis, and the great majority had evidence of islet autoimmunity, with seropositivity to pancreatic islet cells and islet antigens, including glutamic acid decarboxylase (GAD65), tyrosine phosphatase IA-2, and insulin. In contrast, adults with type 2 diabetes had a more protracted course, mild symptomatology, and no evidence of islet autoimmunity. Children and adults with MODY were also mildly symptomatic but had a strong family history of early-onset disease.
Recent studies and clinical experience, however, have clouded the picture. For example, many children with type 1 diabetes have mild hyperglycemia and limited symptoms for weeks to months prior to diagnosis, and most now present without ketoacidosis. Although progressive beta-cell failure is a constant feature of the illness, the initial plasma C-peptide concentrations may fall within the normal range (though below those which might be expected given the coexisting hyperglycemia), and endogenous insulin production may persist for months and, in some cases, years after diagnosis. Like many patients with type 2 diabetes, some children and
Risk Factors for the Development of Type 2 Diabetes Mellitus in Pediatric Patients
SGA, small for gestational age; IDM, infant of diabetic mother; LGA, large for gestational age; PCOS, polycystic ovary syndrome; GH, growth hormone. Genetic Background and Genetic Disorders Positive family history High risk ethnic group Girls > boys Genetic syndromes
Prader Willi, Klinefelter's, Alstrom
Defects in insulin signaling (e.g., leprechaunism)
Mitochondrial disorders (e.g., Friedrich's ataxia, myotonic dystrophy) Prenatal/Perinatal Growth and Weight Gain
SGA/early malnutrition with rapid catch up weight gain IDM/LGA
Postnatal Weight Gain/Obesity (Especially Abdominal) Hormonal Disorders
Ovarian hyperandrogenism and PCOS Cushing's syndrome GH excess Hyperprolactinemia
Glucocorticoids Atypical antipsychotics Tricyclic antidepressants Lithium
Anticonvulsants (e.g., valproate, carbamazepine, vigabatrin, gabapentin) Lifestyle
Diet: high in saturated fat, high glycemic load low in fiber, Vitamin D, calcium and magnesium Energy expenditure: sedentary adolescents with type 1 diabetes are obese, and a minority (particularly Asians) may fail to show evidence of islet autoimmunity .
Previous assumptions notwithstanding, the development of ketoacidosis (DKA) does not confirm that the patient has type 1 diabetes. DKA may occur at presentation in children and adolescents later found to have type 2 diabetes. In the experience of the author, those with severe acidosis (arterial pH < 7.1) and less-severe hypergly-cemia are more likely to have type 1 diabetes, while those with severe hyperglycemia (blood glucose > 1000 mg%) and mild acidosis are more likely to have type 2 diabetes. Age of onset may be helpful but is not diagnostic; type 2 diabetes can occur as early as 3 years of age but presents more commonly after age 10 years. Type 1 diabetes can develop at any age but is far more likely than type 2 diabetes in children under the age of 8 years.
Most confusing is the evidence of islet autoimmunity in a significant percentage of children and adolescents with type 2 diabetes. Like children with type 1 diabetes, children and adolescents with type 2 diabetes may be seropositive for antibodies to the 65 kDa form of glutamic acid decarboxylase (GAD65), to insulinoma-associated protein-2 (IA-2), to insulin, and to islet cells in cryostat sections of human pancreas. A recent investigation  of 156 racially-diverse patients with new-onset diabetes mellitus demonstrated seropositivity to at least one islet antigen in 74 percent of children whose subsequent clinical course suggested type 2 diabetes. Nevertheless, seropositivity to three or more islet antigens was detected in 41 percent to 46 percent of children with type 1 diabetes but only 11 percent of children with type 2 diabetes. The diabetes-related HLA types DR-3 and DR-4 were found in 89 percent of children with type 1 diabetes, 67 percent of children with type 2 diabetes, and 44 percent of the general nondiabetic population.
Thus, the differentiation of type 2 diabetes from type 1 diabetes can be challenging and is in some cases impossible at the present time. Certain historical and clinical findings may, however, be informative. First, children and adolescents with type 2 diabetes are very likely to have first- and second-degree family members with adult-onset type 2 diabetes. Second, peripubertal children and adolescents with type 2 diabetes, in contrast to children with type 1 diabetes or MODY, commonly have acanthosis nigricans, a marker of insulin resistance. Nevertheless, acanthosis is less common in Asians with type 2 diabetes. Third, fasting C-peptide concentrations in children with type 2 diabetes, in contrast to those in children with type 1 diabetes or MODY, commonly (but not always) exceed 1 ng/ml by one year after diagnosis. Finally, adolescent girls with hirsutism, anovulatory menses, or the polycystic ovary syndrome (PCOS) are more likely to have type 2 diabetes than type 1 diabetes. Nevertheless, poor glycemic control may cause menstrual irregularity in any form of diabetes, and mild hirsutism is common in obese adolescents, even those with type 1 diabetes.
III. RISK FACTORS AND THE PATHOGENESIS OF INSULIN RESISTANCE AND TYPE 2 DIABETES
Type 2 diabetes is the endpoint of a process of metabolic decompensation in which genetic background, environmental determinants, and changes in body composition conspire to induce abnormalities in insulin production and action. Although some studies suggest that hypersecretion of insulin may induce obesity and secondarily limit peripheral insulin action, the preponderance of evidence suggests that insulin resistance is a primary event in the evolution of the disease. The resistance to insulin action is accompanied by hyperinsulinemia and (extrapolating from animal studies) an increase in islet size and beta-cell mass. Progression from insulin resistance to impaired fasting glucose (IFG) and impaired glucose tolerance (IGT or prediabetes) is associated with dysregulation of basal insulin secretion, loss of first-phase glucose-dependent insulin secretion, and altered insulin processing, revealed as an increase in the circulating ratio of proinsulin to insulin . The phenotype of longstanding type 2 diabetes is characterized by a decline in total insulin production, relative or absolute hypoinsulinemia, a reduction in beta-cell mass, and deposition of amyloid in the pancreatic islets. Thus, type 2 diabetes reflects a progressive loss of beta-cell function superimposed upon a loss of insulin sensitivity.
The risk of developing type 2 diabetes is molded on a crucible of genetic inheritance and modulated by developmental and nutritional factors and energy expenditure (Table 11.1). The disease occurs more commonly among African Americans, Hispanic Americans, Native Americans, Pacific Islanders, and (possibly) Asian Americans than among those of Caucasian background and is far more prevalent among subjects with a family history of the disease . The higher concordance rate among monozygotic than among dizygotic twins supports the strong influence of (thus far poorly defined) genetic factors. The prevalence of type 2 diabetes increases with age, but risk is modified by events that transpire before birth: The disease occurs more frequently in those born to diabetic mothers and those born small for gestational age (SGA), particularly if there is rapid catch-up growth in early childhood [5-7]. The heightened risk of type 2 diabetes in children born SGA may reflect reductions in skeletal-muscle insulin sensitivity and diminished pancreatic beta-cell mass. Breast-feeding of newborn infants appears to reduce diabetes risk, at least among Native American children .
Rates of type 2 diabetes are higher in girls (1.7-fold) than boys. The prevalence of impaired glucose tolerance and type 2 diabetes in teenage girls with ovarian hyperandrogenism or PCOS approximate 35 percent and 6 percent, respectively; by virtue of a heightened risk of PCOS in adolescence, prepubertal girls with adrenarche also appear to be vulnerable .
However, the most important modifiable risk factor for type 2 diabetes is obesity. The critical role of obesity in the development of type 2 diabetes in adults was established in the Nurses Health Study ; among nearly 17,000 adult women followed prospectively for 16 years, the risk of developing type 2 diabetes was nearly fortyfold higher among those with the highest body mass index (BMI) than in those with lowest BMI; in contrast, smoking, intake of saturated fats, and a sedentary lifestyle increased diabetes risk by 1.8-fold to 2.4-fold.
Obesity plays a similarly important role in the pathogenesis of type 2 diabetes in children; with the exception of some teenage girls with PCOS, the overwhelming majority of pediatric patients with type 2 diabetes are obese. Insulin sensitivity in prepubertal and pubertal children correlates inversely with BMI and percent body fat [9, 11, 12], and severe obesity in prepubertal American children and adolescents is commonly associated with IGT (21 percent to 25 percent) and in some cases (4 percent of teenagers) with unsuspected type 2 diabetes [11, 12]. BMI in childhood (age 7-13 years) correlates with the clustering of obesity, hypertension, hyperin-sulinemia, and dyslipidemia in adulthood (age 22-25 years), and obesity and hyper-insulinemia in Pima and African American children predict the development of type 2 diabetes in adolescence and adulthood [13-19]. Finally, obesity is a common feature of genetic and hormonal conditions associated with IGT and type 2 diabetes such as the Prader-Willi syndrome.
Still, factors other than the simple accumulation of body fat modulate the risk of glucose intolerance. For example, the distribution of body fat appears to be of critical importance. Accumulation of upper body (visceral [intraperitoneal] and
Was this article helpful?