Type 2 Diabetes Mellitus in Children: Title and subTitle BreakPrimary Care and Public Health Considerations

Historically, diabetes mellitus has been classified as either juvenile-onset (now known as type 1) or adult-onset (type 2) due to distinct differences in the usual age of presentation of these 2 conditions. With the increasing prevalence of type 2 diabetes mellitus in children, these terms have become inaccurate. Recent estimates suggest that type 2 diabetes mellitus may now account for as many as half of all new cases of diabetes in certain pediatric populations. This apparent epidemic, attributable largely to the increased rates of obesity in children, carries enormous long-term public health implications. This article will consider these implications, examine the pathophysiology of type 2 diabetes mellitus in children, and review prevention and treatment strategies.

There are currently no nationwide epidemiological data focusing on type 2 diabetes mellitus in children, however, clinic-based reports and regional studies consistently emphasize the increasing prevalence of this clinical entity. Prevalence has been estimated at between 2 and 50 per 1000 in various populations, rates that have increased as much as 10-fold over the past 2 decades.^{1 - 2} In 2 studies conducted in the mid-1990s of persons aged 10 to 19 years, type 2 diabetes mellitus accounted for 33%² and 46%³ of all diabetes in that age group. These figures may underestimate the actual magnitude of the problem because type 2 diabetes mellitus is more likely to be misclassified, undiagnosed, or unreported than type 1 diabetes mellitus.¹

As with adults, type 2 diabetes mellitus in children is more strongly associated with obesity than with any other clinical condition. Virtually all published studies have found that mean body mass index (BMI) among children with type 2 diabetes mellitus is above the 95th reference percentile for age.¹ In 1 biracial cohort, 92% of children newly diagnosed with type 2 diabetes mellitus were characterized as obese.² There is broad consensus that the increasing prevalence of overweight in children, presently estimated to be 25% or more,⁴ has played a major role in the recent increase in pediatric type 2 diabetes mellitus.

In addition to obesity, other important risk factors include ethnicity, age, sex, sedentary lifestyle, family history, and perinatal influences. Type 2 diabetes mellitus is more common in American Indian, black, and Hispanic children than in the general population.^{1 ,5} In a study of 1 midwestern metropolitan area, 69% of children with type 2 diabetes mellitus were black, although blacks represented only 9.7% of patients with type 1 diabetes mellitus and 14.5% of the local population.² The peak age at diagnosis of type 2 diabetes mellitus in youth is between 12 and 16 years, corresponding to the midpubertal period.⁶ Most case series describe a high female-to-male ratio (range of up to 3:1).¹ Sedentary lifestyle is associated with significantly increased risk for type 2 diabetes mellitus in adults and this association is likely to exist among children as well.^{7 - 8} Family history constitutes another important risk factor. In 1 study, 80% of Mexican American children with type 2 diabetes mellitus had at least 1 affected first-degree relative.⁹

Perinatal factors also are thought to contribute to risk for type 2 diabetes mellitus. Risk appears to increase with either low or high birth weight, possibly because undernutrition or overnutrition in utero may cause permanent metabolic and hormonal changes that promote obesity, insulin resistance, and β-cell dysfunction later in life.^{10 - 11} In a study of Scandinavians, the odds ratio for developing type 2 diabetes mellitus in young adulthood increased by 1.38 for every 1-kg decrease in birth weight.¹² Conversely, maternal diabetes is associated with increased birth weight and greater risk for childhood type 2 diabetes mellitus.¹³ A remarkable 40% of type 2 diabetes mellitus cases among Pima Indian children has been attributed to the effects of exposure to diabetes in the intrauterine environment.¹⁴

The pathophysiology of type 2 diabetes mellitus differs fundamentally from that of type 1 and this difference has important implications for the design of strategies to reduce diabetes complications. Type 1 diabetes mellitus is usually caused by autoimmune destruction of pancreatic β-cells, resulting in absolute insulin deficiency. In contrast, type 2 diabetes mellitus is characterized by a relative insulin deficiency in the context of peripheral insulin resistance and increased hepatic glucose production.

A theoretical model of the natural history of type 2 diabetes mellitus is presented in Figure 1. Insulin resistance and compensatory hyperinsulinemia occur early in disease development, long before abnormalities of glucose homeostasis become evident.¹⁵ Most identified risk factors (eg, obesity, puberty, sedentary lifestyle, low birth weight) are thought to cause diabetes by promoting insulin resistance. This process places the β-cell under chronically increased stress. In genetically susceptible individuals, β-cell dysfunction may eventually occur, resulting in isolated postprandial hyperglycemia followed by impaired fasting blood glucose. Once frank diabetes develops, most individuals require increasingly aggressive therapy.¹⁶ It is important to note that disease progression is determined by a complex interplay between genes and environment, and that type 2 diabetes mellitus is reversible^{17 - 18} and potentially curable before permanent β-cell failure has occurred.

All types of diabetes increase the risk for microvascular and macrovascular disease, including myocardial infarction, stroke, renal failure, blindness, and neuropathy. However, risk for developing these complications may be substantially greater in type 2 diabetes mellitus than in type 1.

Elevated blood glucose, common to both types of diabetes, affects blood vessels through several mechanisms. Acutely, hyperglycemia increases blood pressure, promotes blood clot formation, and reduces endothelium-dependent blood flow, effects that appear to be mediated by oxidative stress.^{19 - 20} Long-term, hyperglycemia induces formation of advanced glycosylation end products that damage vascular endothelium.²¹ For these reasons, good glycemic control reduces diabetes complication rates.²²

However, unlike type 1, type 2 diabetes mellitus is strongly associated with 2 additional cardiovascular disease risk factors: obesity and hyperinsulinemia. The adverse effects of obesity on cardiovascular health have been widely recognized.²³ Recently, hyperinsulinemia has also been identified as an independent risk factor. One study compared 91 men aged 45 to 76 years who developed ischemic heart disease over a 5-year period with 105 disease-free controls matched for age, BMI, smoking habits, and alcohol consumption.²⁴ The study found that the odds ratio of developing heart disease increased by 70% for every single SD increase in fasting insulin level, after controlling for blood pressure and other potentially confounding factors.²⁴ Hyperinsulinemia has been postulated as a critical component of the insulin resistance syndrome (syndrome X), including insulin resistance, hypertension, hypertriglyceridemia, low high-density lipoprotein cholesterol, abnormal blood clotting tendency, and chronic inflammation.²⁵ This syndrome may be present in children before and during puberty.²⁶ Conventional diabetes therapy aimed at improving glycemic control does not address these important pathophysiological features of type 2 diabetes mellitus and may actually cause weight gain and increase insulin levels.¹⁶

Early diagnosis is imperative to identify comorbidities⁶ such as hypertension and dyslipidemia, and to initiate diabetes treatment. Early institution of treatment has the best chance of reversing progression to β-cell failure.

Because type 2 diabetes mellitus may remain asymptomatic for a prolonged period, the American Diabetes Association now recommends the screening of overweight children (BMI >85th percentile for age and sex) who have 2 or more of the following risk factors: family history of diabetes in a first- or second-degree relative; American Indian, black, Hispanic, or Asian/Pacific Islander ethnicity; signs of insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovary syndrome). Screening should be initiated at age 10 years (or at puberty) and performed every 2 years. Acceptable screening methods include fasting blood glucose or a 2-hour oral glucose tolerance test, but not glycosylated hemoglobin because of its low clinical sensitivity.⁶

If not identified in its early stages, type 2 diabetes mellitus may present with the classic signs and symptoms of diabetes, such as polyuria, polydipsia, dehydration, and sometimes ketoacidosis. Differentiation from type 1 diabetes mellitus may initially be problematic. Measurement of serum insulin, C-peptide, and islet antibodies related to autoimmune diabetes generally helps establish the specific diagnosis.⁶

Two central principles regarding the management of type 2 diabetes mellitus should be emphasized. First, frank diabetes represents a late metabolic decompensation of a chronic disease process (Figure 1). The increased risk of cardiovascular complications begins early in this disease process before formal diagnostic criteria for diabetes have been reached. Since no clear distinction exists between prevention and treatment, the importance of a healthful diet and a physically active lifestyle should be emphasized for all overweight children. Second, because type 2 diabetes mellitus is caused by relative insulin deficiency in the setting of insulin resistance, optimal therapy requires measures to decrease insulin resistance.

The cornerstone of treatment for type 2 diabetes mellitus is weight loss. Demonstrable improvement in insulin sensitivity occurs with modest weight loss. Significant weight loss can produce complete normalization of blood glucose levels and potentially permanent resolution of diabetes.^{17 - 18} Unfortunately, conventional treatment for obesity generally has poor long-term success.²⁷ While a discussion of current treatment options for pediatric obesity is beyond the scope of this review, a multidisciplinary approach, involving dietary modification, increased physical activity, decreased sedentary time, and behavioral therapy, offers the best hope for a successful outcome.^{28 - 29}

Several dietary factors may influence disease development and progression independent of body weight. Traditionally, low-fat and/or high-carbohydrate diets have been recommended for the prevention and treatment of type 2 diabetes mellitus. Recently, some investigators have argued that such diets promote insulin resistance and, for this reason, have advocated higher fat and lower carbohydrate intake.³⁰ Although the total amount of dietary fat appears to have little effect on disease risk, the type of fat may be important. Partially hydrogenated fat (trans-fat), which is commonly found in commercial food products, increases risk, whereas polyunsaturated fat from vegetable and marine sources decreases risk.³¹ Another potentially important factor may be the type of dietary carbohydrate. Habitual consumption of low glycemic index foods (ie, foods that cause a relatively small postprandial increase in blood glucose) may lower the risk of type 2 diabetes mellitus³² and improve metabolic control once the disease has developed.³³ A low-glycemic-index diet may also facilitate weight loss, though this possibility has not been examined in long-term clinical trials.³⁴

Independent of its effects on body weight and composition, physical activity improves insulin sensitivity and glucose tolerance. Even modest levels of physical activity, such as daily walking, have been shown to decrease risk for type 2 diabetes mellitus in adults.⁸ Conversely, sedentary activities, such as television viewing, have been shown to increase risk for obesity.^{35 - 36}

A total of 3 classes of medications for the treatment of diabetes act by improving insulin sensitivity or lowering insulin demand, though none have been approved for use in children by the Food and Drug Administration. The biguanides (eg, metformin) decrease hepatic glucose production and improve fasting hyperglycemia. α-Glucosidase inhibitors (eg, acarbose) delay the digestion and absorption of starchy food, and improve postprandial hyperglycemia. The thiazolidenediones (eg, rosiglitazone) increase peripheral insulin sensitivity. Metformin has been used safely in children and does not cause weight gain.³⁷

The 3 classes of medications in this category are sulfonylureas, meglitinides (eg, repaglinide), which stimulate endogenous insulin secretion, and insulin itself. In our opinion, these agents should not be considered first-line treatment for type 2 diabetes mellitus because they do not treat insulin resistance and because they have important adverse effects (weight gain and hypoglycemia). However, insulin may become necessary if other measures fail to establish acceptable glycemic control or in the event of an acute metabolic decompensation.

A variety of environmental factors including poor dietary quality, inadequate physical activity, and sedentary lifestyles have converged in the past 2 decades to produce an unprecedented epidemic of childhood obesity and its most serious complication, type 2 diabetes mellitus. As a result of this epidemic, we face the prospect of coronary heart disease becoming a disease of young adulthood. This impending crisis demands a concerted public health campaign calling on biomedical researchers to identify novel treatments for obesity and insulin resistance, public health agencies to devise community-based prevention strategies, public schools to promote physical activity and fitness, the commercial food industry to market healthful foods to children, and parents to model and support healthful lifestyle choices. The primary care clinician must play an active role in both the primary prevention of obesity and the timely diagnosis and treatment of type 2 diabetes mellitus in children.