Type 2 diabetes mellitus in children and adolescents: An emerging problem

Type 2 diabetes mellitus in children and adolescents: An emerging problem

Boney, Charlotte

Type 2 diabetes mellitus (T2DM) used to be considered a disease of adults. However, childhood obesity is now epidemic in North America, and parallel to this increase in obesity is an emerging epidemic of T2DM in children and adolescents. This article will review T2DM in childhood, the clinical features that distinguish it from the more common form of diabetes in children, type 1 (autoimmune) diabetes mellitus, and the therapeutic management of T2DM in children.

T2DM has a strong hereditary component that is thought to reflect the contributions of multiple genes. It results from insulin resistance followed by the decline of sufficient insulin secretion to overcome the insulin resistance causing hyperglycemia. A generation ago, it was recognized that a small proportion of children with non-type I diabetes presented with an insidious onset and had a slowly progressive clinical course not unlike the adult-form of T2DM. This childhood form of T2DM is distinct from the rare but well described form of non-type 1 diabetes in children called Maturity Onset Diabetes of Youth (MODY), a group of six autosomal dominant forms of diabetes. MODY is not associated with obesity, and its incidence has not changed despite the increase in obesity. Prior estimates of children (under age 20) having non-type 1 diabetes mellitus were 1-2%, but recent reports indicate 8-45% of children with newly diagnosed diabetes have T2DM.(1) The wide variation in percentages depends on the populations that are sampled and the prevalence of obesity in those populations.

There has been a remarkable increase in the prevalence of overweight and obese children since 1980. Currently, greater than 20% of children of all ethnic groups between the ages of 6-18 years have a body mass index (BMI) greater than the 85th percentile (%tile) for age (overweight), and more that 10% have a BMI greater than the 95th %tile for age (obese). The prevalence of obesity (BMI >95th %tile) doubled between 1963 and 1991.(2) Data from the National Health Nutrition Exam Survey 1988-94 reveal significant increases in obesity in children age 12-17 years, but there were differences in prevalence among racial and ethnic groups. For example, between 1991 and 1994, the incidence of obesity in Caucasian girls rose from 8% to 10% compared to African American and Hispanic girls, where the incidence increased from 10% to 15%.(3)

The highest prevalence of obesity is in the Pima Indians of Arizona, where 68% of children age 5-19 years are obese. This population, known to have the highest adult prevalence oFT2DM of any population in the world, also had the earliest reports of T2DM in their children. Recent data reveal that 2% of Pima children ages 10-14 years and 5% aged 15-19 years have T2DM.4 Multiple logistic regression models suggest that the increase in T2DM in Pima children is explained by increases in gestational diabetes and obesity.

Additional reports ofT2DM in children from different areas of the United States show a greater incidence among African American and Hispanic American children, which parallels higher obesity rates in these populations. Eight studies describing the increase in T2DM in children were published between 1996 and 1998. These reports described 461 cases of T2DM in children with a mean age of 13.5 years and a female-to-male predominance that varied from 1.6:1 to 3:1. The ethnicity was mostly African American and Hispanic American, with Caucasians making up 5-35%. Over 95% of patients were overweight (BMI >85th %tile for age), and over 85% of them had a family history ofT2DM in a first-degree relative. These studies were mostly case reports from inner city populations seen at tertiary care centers, and in all of them, the incidence of T2DM paralleled the incidence of obesity in the particular racial group. One of the eight studies from the University of Cincinnati compiled all newly diagnosed cases of diabetes from birth to age 19 from 1982 to 1994.(5) New-onset diabetes patients with T2DM rose from 4% of all new cases in 1992 to 16% in 1994. Paralleling this increase was a huge increase in the incidence of obesity in their population.

Collectively, these published reports suggest that 30-40% of newly diagnosed cases of diabetes in teenagers is T2DM, which represents at least a three-fold increase since 1980. Our experience at Hasbro Children’s Hospital is consistent with these data. Between 1998 and 2000, the Division of Pediatric Endocrinology and Metabolism in the Department of Pediatrics at Hasbro Childrenis Hospital had 108 new diagnoses of diabetes: 27 were teenagers and 8 of those teenagers had T2DM. The emergence of T2DM in teenagers is not limited to the United States. An increased incidence has been reported in Japan, Europe, Australia and Canada, coinciding with increased rates of obesity.

Regular exercise is a key component of cardiovascular health, caloric expenditure, and weight management. Modest increases in exercise have been shown to delay the onset of T2DM in adults with glucose intolerance and reduce the risk of diabetes independent of obesity.6,7 Physical activity decreases insulin resistance, so its role in the management of children and adults with all types of diabetes cannot be overemphasized. Lifestyle changes that decrease exercise have played a role in the increased incidence of obesity in children. A prospective study of 2100 adolescents showed that the prevalence of obesity increased 2% for each additional hour of television.8 A more recent study demonstrated that time spent on sedentary activities was proportional to fat mass.9 These studies, and the increasing trend in public schools to decrease or eliminate physical education as a routine part of the school day, suggest that decreased exercise is a major factor in the obesity epidemic and likely the increased prevalence of T2DM.

Children with T2DM present with a broad spectrum of signs and symptoms, so that it is not possible to reliably distinguish between type I and type 2 diabetes. (Table 1) Some patients may be totally asymptomatic and found to have glucosuria or mild hyperglycemia incidentally. Other patients present with features typical of type 1 diabetes, including polyuria, polydipsia, weight loss, dehydration, ketonuria, even ketoacidosis. Information from many large pediatric diabetes centers, although anecdotal, suggests that compared to patients with type I diabetes, children with T2DM tend to have milder hyperglycemia, higher levels of insulin and C-peptide, less ketonuria and milder DKA, and no serologic evidence of beta cell autoimmunity. The autoantibodies present in type I diabetes are well described, and measurements by radioligand-binding assays are now commercially available. Over 80% of type I diabetics will have measurable levels of at least two of three specific antibodies to insulin, glutamic acid decarboxylase (GAD)-65, and protein tyrosine phosphatase IA-2.(10) Recent studies, although limited, have shown that unlike type I patients, most patients with T2DM have absent titers and a small proportion have low titers of not more than one of these antibodies.11 Therefore, measuring insulin, GAD-65 and IA-2 antibodies at presentation in children who might have T2DM is helpful in distinguishing type 1 from T2DM. We, like most other pediatric diabetes clinics, routinely order all three antibody tests on new-onset diabetics over the age of 10 years who are obese and have a family history of T2DM.

Obesity is pre ent in the overwhelming majority of children with T2DM; however, igns and symptoms at presentation are not helpful in distinguishing type 1 from T2DM. In a study published in 1997 from the University of Arkansas, signs and symptoms of teenagers presenting with new-onset type 1 versus T2DM mere compared.12 (Figure 1) Almost all (96%) of the T2DM patients had a BMI greater than the 85th percentile, but surprisingly, 24% of the type 1 patients also had a BMI greater than the 85th percentile. Although the obesity in patients with T2DM certainly is not surprising, the high incidence of obesity in the type 1 patients highlights the epidemic of obesity in our children and points to the difficulty of distinguishing type1 versus T2DM patients based on BMI alone. Even more surprising was the observation that presenting symptoms were not different between the type 1 and T2DM groups. The only significant difference in symptoms was that 70% of type 1 compared to 40% of T2DM patients had experienced recent weight loss. The groups did have different biochemical characteristics at presentation. Although hemoglobin A1c levels were similar, a higher percentage of type 1 patients had ketonuria, DKA and marked hyperglycemia (mean of 544 mg/dl in type 1 patients versus 339 mg/dl in T2DM). These data suggest that weight loss and DKA make it more likely that an obese patient has type 1 diabetes, but do not rule out T2DM.

Many studies also indicate that signs of insulin resistance, such as the cutaneous marker acanthosis nigricans, are often present in childhood T2DM. Acanthosis nigricans refers to areas of thickened skin with hyperpigmentation and a velvety texture. It is particularly prominent in skin folds such as the base of the neck and axillae. Acanthosis nigricans is believed to reflect hyperinsulinemia, and thus, the severity of insulin resistance. While not a specific marker for T2DM, its increased occurrence in obese children further emphasizes the link between insulin resistance and the development of T2DM.

It has been observed that glycemic control in children with type 1 diabetes often deteriorates during puberty. This is attributed in part to insulin resistance resulting from increased growth hormone, a major insulin counter regulatory hormone, In children with T2DM, the average age of diagnosis is 13-14 years. Interestingly, this age coincides with completion of the pubertal growth spurt following peak growth hormone secretion, suggesting that the “normal” insulin resistance of puberty may exacerbate the insulin resistance of obesity and precipitate T2DM in genetically susceptible individuals. However, this does not mean that prepubertal, obese children are not at risk for developing T2DM. Several studies have shown that body fat is closely correlated to insulin resistance, irrespective of pubertal status.13 In addition, rates of obesity are similar before and after puberty. Although the growth hormone– mediated insulin resistance at puberty may play a role, there likely are environmental factors, such as diet and exercise, that also play important roles in T2DM onset after puberty.

There is a growing literature on the clustering of other signs of insulin resistance in obese children and adolescents, including hypertension and dyslipidemia. In adults, the association of obesity, glucose intolerance or T2DM, hypertension and dyslipidemia is referred to as Syndrome X, the Metabolic Syndrome or the Insulin Resistance Syndrome, and it is strongly associated with cardiovascular disease even in the absence of overt diabetes.14 This raises great concern about the potential occurrence of cardiovascular disease developing early in individuals who develop Syndrome X in childhood. Hypertension in Syndrome X can be mild or severe and can be characterized by elevated diastolic and/or systolic blood pressures. In children, blood pressure measurements at or above the 95th percentile for age are considered hypertensive, and careful follow-up is merited for blood pressures at the 90th %tile. Age-specific percentiles can be found from the Task Force on Blood Pressure Control in Children.15

The characteristic dyslipidemia of insulin resistance includes hypertriglyceridemia and decreased HDL cholesterol, but may also include increased LDL or VLDL cholesterol. Total blood cholesterol levels in children and adolescents should be less than 170 mg/dl, and are considered elevated if greater than 200 mg/dl. Fasting triglycerides should be less than 150 mg/dl and HDL cholesterol should be greater than 35 mg/dl.16 The association between insulin resistance and hypertriglyceridemia/ depressed HDL cholesterol, which has been well described in adults, occurs in children.” There is accumulating evidence that Syndrome X, originally described in adults, also occurs in obese children. The best data come from the Bogalusa Heart Study, a 25 year, community based longitudinal study that includes Caucasian and African-American children. Obesity was associated with one or more cardiac risk factors (dyslipidemia, hypertension, hyperinsulinemia) in 58% of obese children.18 Therefore, the presence of acanthosis nigricans, hypertension or dyslipidemia in a patient presenting with new-onset diabetes indicates insulin resistance and should prompt a consideration of the diagnosis ofT2DM.

A family history ofT2DM in a first or second degree relative, regardless of race or ethnic background, is strongly associated with T2DM in children. In some studies, close to 100% of children with T2DM have a family history of T2DM. This suggests that genetic susceptibility plays a role in the development of T2DM. Although accumulating epidemiological data suggest a higher prevalence of T2DM in children that are African American, Hispanic American or Native American,19 these racial and ethnic groups have higher rates of obesity. Thus, genetic susceptibility cannot be equated with race or ethnic group per se. Overwhelming evidence indicates that, given an underlying genetic susceptibility, T2DM is largely determined by environmental factors including obesity, physical inactivity, and a diet high in fat and refined carbohydrates and low in fiber, all of which contribute to insulin resistance. Two recent longitudinal, large cohort studies indicate that obesity is the single most important determinant of T2DM, and that lifestyle factors such as diet and exercise are also important.6,7

Rigorous clinical guidelines have not been established to determine which children presenting with new-onset diabetes have T2DM. The distinction between type 1 and T2DM is important because it helps to establish the necessity of long-term insulin therapy and the need to monitor for co-morbid conditions that are associated with obesity and insulin resistance. Children with type 1 diabetes should be treated with insulin immediately in order to prevent metabolic crises and restore normal growth. However, long term insulin therapy in children with T2DM may predispose them to unnecessary hypoglycemia and difficulty losing weight. Complications of diabetes such as retinopathy and nephropathy now have been clearly associated with glycemic control, regardless of whether the patient has type 1 or T2DM. However, other co-morbid conditions independent of these microvascular complications are different in type 1 versus T2DM. Children with type 1 diabetes are at increased risk for other autoimmune diseases such as acquired hypothyroidism and celiac disease, whereas children with T2DM are at risk for conditions associated with insulin resistance, including hypertension, dyslipidemia and polycystic ovary disease.

The practical approach to diagnosis used by most pediatric endocrinologists is based on clinical presentation and the level of suspicion forT2DM. Clearly, normal weight children and adolescents presenting with classic signs and symptoms of diabetes (polyuria, polydipsia, weight loss and a random plasma glucose >200 mg/dl) have type 1 diabetes. Because obesity is now present in more than 20% of our youth, most practitioners would also agree that any child, regardless of weight status, who presents with severe DKA has type 1 diabetes until proven otherwise. However, the challenge is distinguishing type 1 from T2DM in the obese child over the age of 10 years with a family history ofT2DM. Our division now routinely obtains serology for all three autoimmune markers of type 1 diabetes (insulin, GAD-65 and IA-2 antibodies) in these children, and we work closely with pediatric dieticians to implement weight loss and diet management. Patients with negative antibody testing and a family history ofT2DM are monitored closely for evidence of insulin reserve and are considered candidates for glucose-lowering oral agents. However, we make it very clear to patients and their families that discontinuing insulin therapy does not mean they have a “milder” or less serious form of diabetes.

Initial treatment of children presenting with new onset diabetes, even if they are suspected of having T2DM because of obesity and a positive family history, usually includes a period of insulin administration. Insulin therapy remains the cornerstone of diabetes treatment in the pediatric population. At present, it is still the only drug approved by the Food and Drug Administration for the treatment of diabetes in children. The majority of children with new onset diabetes will have type 1 diabetes, so insulin therapy must be instituted promptly to reverse catabolism and restore glycemia. Any child, regardless of obesity or family history ofT2DM who presents with dehydration, ketonuria or acidosis must be treated with insulin. Diabetic ketoacidosis and hyperglycemia with hyperosmolality are associated with high morbidity and mortality in children. Obese children presenting with classic symptoms of diabetes who are not ill (absence of dehydration, significant ketonuria or DKA) are also treated with insulin initially. Soon after starting insulin therapy, insulin requirements often decline rapidly and near-normal glycemic control may follow. This is probably due to resolution of glucose toxicity, a phenomenon of peripheral insulin resistance and decreased endogenous insulin secretion mechanistically linked to chronic hyperglycemia. While a period of improved endogenous insulin secretion known as the !honeymoon periodi occurs in patients with type I diabetes, these patients manifest a need for more insulin over time, usually within the first 6 to 12 months after diagnosis. By contrast, patients with T2DM may do well for longer periods of time on small insulin doses and will have biochemical evidence of insulin reserve with normal or elevated C peptide levels. Strict dietary management, exercise and slow but sustained weight loss facilitate this phenomenon of excellent glycemic control on small insulin doses in adolescents with T2DM.

In asymptomatic children with obesity and a positive family history, hyperglycemia may be found incidentally. In these patients, levels of insulin, GAD-65 and IA-2 antibodies are obtained, and initial treatment with diet, exercise and an oral agent may be appropriate. In addition, patients receiving insulin who have excellent glycemic control but who are suspected of having T2DM may be transitioned to oral therapy. Despite a wealth of data about the therapeutic efficacy of glucose-lowering oral agents in adults with T2DM, there are very limited data in children. Although use of oral agents is increasing in the childhood population with T2DM, any discussion of therapy is largely anecdotal. A recent survey of 130 pediatric endocrinology practices in the United States and Canada indicated that 48% of children with presumed T2DM were treated with insulin and 44% were treated with one or more oral agents.20 The most commonly used oral agents were metformin (71%) and sulfonylureas (46%). Metformin is now the most commonly prescribed oral agent for T2DM at all ages in the United Sates. One randomized double-blind placebocontrolled trial of metformin therapy in pediatric patients has been published. In a small cohort of 82 children aged 10-16 years with new onset T2DM and plasma glucose levels 126-240 mg/dl, metformin was compared to diet and exercise for 16 weeks (21). Metformin was shown to be safe and efficacious based on fasting glucose levels and hemoglobin Alc levels. Only 10% of subjects who received metformin (compared to 65% of subjects in the diet and exercise group) needed additional therapy because of poor glycemic control by the end of the 16 week study period. In spite of the encouraging response to metformin in these children, it is important to emphasize the need for ongoing follow-up. As demonstrated in the United Kingdom Prospective Diabetes Study of adults, T2DM is a progressive disease even in patients who initially respond to a single oral agent.22

At Hasbro Children’s Hospital, we use metformin therapy in the majority of our patients with T2DM who are being treated with an oral agent, either after initial therapy with insulin or in addition to diet and exercise at diagnosis. Advantages of oral agents in children with T2DM include potentially better compliance and fewer side effects such as weight gain and hypoglycemia. We have had more limited experience with other oral agents, including the sulfonylureas and the thiazolidinediones. The newer sulfonylureas are well tolerated and hypoglycemia appears not to be a serious problem, although weight gain is clearly a disadvantage. The newer thiazolidinediones are also well tolerated and seem to be quite effective in some morbidly obese, severely insulin resistant patients. However, we remain concerned about potential hepatic toxicity, and the frequent monitoring of liver enzymes generally is not well received by teenaged patients. We have no experience with alpha-glucosidase inhibitors, mostly because teenagers will not agree to medication at every meal.

The most important treatment strategy in addition to medical therapy is comprehensive education that includes dietary and lifestyle changes and routine blood glucose monitoring. Dietary changes and weight management require long-term involvement of a pediatric nutritionist who can establish a relationship with the teenager and make realistic recommendations. The goal of treatment, regardless of the type of diabetes, is glycemic control adequate to support normal growth and development and to prevent long– term complications. We set goals for our teenagers of hemoglobin AMc levels at or below 8% and fasting blood glucose levels 80-140 mg/dl. We see most T2DM patients every three months, just as frequently as type 1 patients. We monitor them annually for early evidence of nephropathy by using a screening test for microalbuminuria. We recommend yearly dilated eye exams for evidence of retinopathy. We insist on regular blood glucose monitoring in T2DM patients, although not as frequently as in patients taking insulin. Our experience at Hasbro Children’s Hospital with adolescent T2DM patients indicates they are just as likely as their type 1 peers to be poorly compliant with their medical regimen, even when it is as simple as once-a-day pills. Involvement of parents in the diabetic care regimen and attention to psychosocial risk factors such as depression, substance abuse, and eating disorders are just as important in patients with T2DM as they are in insulin-requiring patients with type 1 diabetes.

In summary, T2DM is no longer just a disease of adults. The epidemic of obesity in children has led to an increase in a number of obesity-related diseases during childhood, including T2DM. Primary prevention efforts to reduce obesity and prevent T2DM should be emphasized. Lifestyle interventions focusing on weight reduction and increased physical activity should be promoted in all children. Given the increasing incidence and prevalence of T2DM in children and adolescents, The American Diabetes Association (ADA) consensus statement (March 2000) included guidelines for testing children for T2DM.” (Table 2) They recommend screening obese children over the age of 10 years who have a family history of T2DM and evidence of insulin resistance, such as acanthosis nigricans. We encourage all primary care providers who care for children meeting these criteria to vigorously educate families about diet, exercise, weight reduction, and the devastating long term consequences of diabetes. Optimal care for the majority of children with T2DM usually requires the resources of a pediatric diabetes center where specially trained endocrinologists, diabetes nurse educators, pediatric nutritionists and social workers work as a team. The burden of this chronic disease is enormous, so we as practitioners must aggressively try to reverse this epidemic of obesity and T2DM in our youth.


1. Center for Disease Control, National Center for Chronic Disease Prevention and Health Promotion, Diabetes Public Health Resource Web Site, 2001

2. Styne DM. Childhood and Adolescent Obesity: Prevalence and Significance. Pediatr Clinics NAmer 2001; 48:823-54.

3. Morbidity Mortality Weekly Reports 1997;46:199-202.

4. Dabelea D, Hanson RL, Bennett PH, et al. Increasing prevalence of type 2 diabetes in American Indian children. Diabetologia 1998;41:904-10.

5. Pinhas-Hamiel 0, Dolan LM, Daniels SR, et al. Increased incidence of non-insulin dependent diabetes among adolescents. J Pediatr 1996;128:608-15.

6. Toumilehto J, Lindstorm J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. NEJM 2001;344:1343-50.

7. Hu FB, Manson JE, Stampfer MJ, et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. NEJM 2001;345:790-7.

8. Dietz WH, Gortmaker SL. Do we fatten our children at the television set? Obesity and television viewing in children and adolescents. Pediatrics 1985;75:807-12.

9. Maffeis C, Zaffanello M, Schatz Y. Relationship between physical activity and adiposity in prepubertal boys. J Pediatr 1997;131:288-92.

10. Leslie RD, Atkinson MA, Notkins Al. Autoantigens IA-2 and GAD in type 1 (insulin-dependent) diabetes. Diabetologia 1999;42:3-14.

11. Grasso YZ, Reddy SK, Rosenfeld CR, et al. Autoantibodies to IA-2 and GAD65 in patients with type 2 diabetes mellitus of varied duration: Prevalence and correlation with clinical features. Endocr Pract 2001;7:339-45.

12. Scott CR, Smith JM, Cradock MM, Pihoker C. Characteristics of youth-onset noninsulin-dependent diabetes mellitus and insulin-dependent diabetes mellitus at diagnosis. Pediatr 1997;100:84-91.

13. Arslanian S, Suprasongsin C. Insulin sensitivity, lipids and body composition in childhood: is iSyndrome Xi present?JClin Endocrinol Metab 1996;81:1058-62.

14. Hansen BC. The Metabolic Syndrome X. Annals NYAcad Sciences 1999; 892:1-24. 15. Task Force on Blood Pressure Control in

Children: Report of the Second Task Force on Blood Pressure Control in Children. Pediatrics 1987; 79:1-25.

16. Report of the Expert Panel on Blood Cholesterol Levels in Children and Adolescents. Pediatrics 1992;89:525-84.

17. Steinberger J, Moorehead C, Katch V, Rocchini AP Relationship between insulin resistance and abnormal lipid profile in obese adolescents. J Pediatr 1995;126:690-5.

18. Chen W, Bao W, Begum S, et at. Age-related patterns of the clustering of cardiovascular risk variables of syndrome X from childhood to young adulthood in a population made up of black and white subjects: the Bogalusa Heart Study. Diabetes 2000;49:1042-8.

19. Fagot-Campaagna A, Pettitt DJ, Engelgau MM, et al. Type 2 diabetes among North American children and adolescents: an epidemiologic review and a public health perspective. JPediatr 2000;136:664-72.

20. Silverstein JH, Rosenbloom AL. Treatment of type 2 diabetes mellitus in children and adolescents. J Pediatr Endocrinol Metab 2000;13 Suppl 6:1403-9.

21. Jones KL, Arslanian S, Peterokova VA, et al. Effect of metformin in pediatric patients with type 2 diabetes. Diabetes Care 2002;25:89-94.

22. Turner RC, Cull CA, Frighi V, Holman RR Glycemic control with diet, sulfonylurea, metformin or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies. JAMA 1999;281:2005-12.

23. American Diabetes Association. Type 2 diabetes in children and adolescents. Pediatr 2000; 105:671-80.

Charlotte Boney, MD

Charlotte Bones MD, is a pediatric endocrinologist at Hasbro Children’s Hospital


Charlotte Boney, MD

Hasbro Children’s Hospital

593 Eddy St.

Providence, RI 02903

phone: (401) 444-4300

fax: (401) 444-2534

e-mail: CBoney@lifspan.org

Copyright Rhode Island Medical Society Apr 2003

Provided by ProQuest Information and Learning Company. All rights Reserved