Type 2 Diabetes mellitus (T2DM) in youth has been increasing over the last few decades paralleling the increased prevalence of obesity. Youth with T2DM tend to be obese with evidence of insulin resistance in the setting of insufficient insulin secretion. They may also present with other cardiovascular risk factors including hypertension or dyslipidemia. Commonly children who present with T2DM have a strong family history of T2DM. Certain ethnic groups are more at risk for T2DM, most notably children of native Americans and Pacific Islanders. Presentations can be acute, like in patients with Type 1 diabetes mellitus (T1DM) who often present with diabetic ketoacidosis, symptomatic with polyuria and polydipsia. Children with T2DM may also present like many adults with T2DM who are diagnosed based on screening studies and are asymptomatic at the time of diagnosis. Obese youth with phenotypic T2DM should have pancreatic antibodies measured to rule out T1DM. Other monogenic causes of diabetes may also be considered in patients with a strong family history who present atypically. Approved first line therapies includes lifestyle modification, Metformin and insulin. Other antihyperglycemic medications are being used in older adolescents for T2DM, but are not yet approved in children < 18 years.

The diagnosis of diabetes mellitus is made by demonstrating a random blood glucose level at or above 200 mg/dL plus the classic symptoms of polyuria and polydipsia, a fasting blood glucose level of greater than or equal to 126 mg/dL, an elevation of greater than or equal to 200 mg/dL of glucose 2 hours after a standard oral glucose load (75 grams). In addition, the ADA has added the elevation of HgbA1C (hemoglobin A1C, glycated hemoglobin, glycosylated hemoglobin) to greater than or equal to 6.5% done by standard assay (defined as National Glycohemoglobin Standardization Program certified and standardized to the Diabetes Control and Complications Trial assay), and this diagnostic testing is becoming more widely accepted (See Table I below). At least two abnormal studies, inclusive of those listed above, in the absence of symptoms are required to make the diagnosis.

Table I.
Criteria: Overweight (BMI >85th percentile for age and sex, weight for height >85th percentile, or weight >120% of ideal for height)
Plus any two of the following risk factors:

Family history of type 2 diabetes in first- or second-degree relative

Continue Reading

Race/ethnicity (Native American, African American, Latino, Asian American, Pacific Islander)

Signs of insulin resistance or conditions associated with insulin resistance (acanthosis nigricans, hypertension, dyslipidemia, polycystic ovary syndrome, or small for gestational age birthweight)

Maternal history of diabetes or GDM during the child’s gestation

Age of initiation: Age 10 years or at onset of puberty, if puberty occurs at a younger age
Frequency: Every 3 years

The use of HgbA1C as a primary diagnostic tool remains somewhat controversial. The American Association of Clinical Endocrinologists (AACE) and the American College of Endocrinologists (ACE) suggested that it not be used as a primary diagnostic tool, and not used when suspecting type 1 diabetes mellitus or in a high risk group for hemoglobinopathies or other hematologic, hepatic or renal diseases that might alter the reliability of the test. There may also be other ethnic differences in this measurement. This controversy is discussed in more detail below.

The determination of type 1 (autoantibody mediated) versus T2DM may be difficult. Children who present with T2DM tend to have the additional characteristics of obesity, acanthosis nigricans (velvety skin hyperpigmentation), a family history of T2DM, and lack of pancreatic autoantibodies.

The difficulty in distinguishing obese T1DM from T2DM is illustrated in the TODAY study of adolescents with phenotypic type 2 diabetes mellitus.Klingensmith et al demonstrated that 9.8% of participants diagnosed as having T2DM based on clinical criteria had pancreatic autoantibodies, with 5.9% positive for a single antibody and 3.9% for more than one, suggesting they were obese children with antibodies consistent with T1DM. The adolescents with positive antibodies were more likely to be Caucasian and male with lower BMI and fewer metabolic abnormalities. Thus pancreatic autoantibodies should be obtained at the time of diagnosis to help distinguish the type of diabetes mellitus in an obese child. If the antibodies are negative and the presentation is consistent with type 2 diabetes in an obese adolescent with a strong family history, for example, this is supportive evidence for the diagnosis of type 2 diabetes. However, as noted in the TODAY study, almost 10% of children determined to have type 2 diabetes with adequate or generous insulin reserves had positive pancreatic autoantibodies, and thus these children likely have a combination of type 1 diabetes complicated by obesity with some degree of insulin resistance in the setting preserved insulin reserves.

Symptoms and signs of the disease:

  • polyuria, polydipsia, weight loss

  • obesity, acanthosis nigricans, other features of Metabolic Syndrome

  • a strong family history of T2DM

  • certain ethnic groups are at higher risk (African-American, Native American, Hispanic)


The presentation of type 2 diabetes mellitus may be acute with symptoms of polyuria and polydipsia. It may be asymptomatic with the incidental finding of hyperglycemia or glycosuria, or may be critical with the presentation of diabetic ketoacidosis or hyperosmolar coma. In rare cases, the presentation may be a malignant hyperthermia syndrome with rhabdomyolysis.


The current American Diabetes Association guidelines suggest that high risk children be screened for type 2 diabetes mellitus or increased risk states. The criteria are outlined in Table I.

The recommendations suggest that the appropriate testing may be either a HgbA1C level, a fasting plasma glucose or a 2 hour 75 gram OGTT. In clinical practice, the fasting glucose level has historically been the most commonly used test. However, with the addition of HgbA1C to the criteria for diagnosing diabetes, more practitioners are using this as an easy, nonfasting initial screening modality. Thus, the same tests are used for screening and for diagnosis, although diagnosis adds the additional requirement of a repeat measure to confirm the diagnosis.

In the current environment, in which obesity is highly prevalent, the initial determination of type of diabetes can be very difficult and sometimes impossible to determine. Children with type 1 and type 2 diabetes, as well as some forms of monogenic diabetes can present with similar electrolyte and acid/base abnormalities and degree of hyperglycemia.

The acute management is the restoration of normal fluid balance, resolution of acidosis, improvement in electrolyte imbalance and reduction of hyperglycemia in a safe manner while preventing cerebral edema and hyperosmolar coma. This is done with careful fluid management and insulin administration if T1DM is being considered, or if acidosis or ketosis is present, or the patient is significantly hyperglycemic regardless of type of diabetes suspected. In addition, some children who appear to have T1DM with DKA at presentation and negative classic pancreatic autoantibodies do not have T1DM and may have a form of antibody-negative diabetes such as Type 1b, T2DM or even Maturity-onset Diabetes of Youth (MODY), also known as monogenic diabetes.

These rare forms of diabetes mellitus, (as outlined by the American Diabetes Association Standards of Medical Care in Diabetes, 2010) include forms of diabetes due to: “genetic defects in beta cell function, genetic defects of insulin action, diseases of the exocrine pancreas (such as cystic fibrosis), and drug or chemical-induced diabetes (such as in the treatment of AIDS or after organ transplantation).” The monogenic causes of diabetes (MODY) may be misdiagnosed as type 1 or type 2 diabetes and should be considered in individuals with a multi-generation history of diabetes and those who present in an atypical manner. The SEARCH trial (Pihoker, et al, JCEM, 2013) identified 586 participants who were pancreatic autoantibody negative and had measurable C-peptide levels and found MODY mutations in 8% of those tested, representing 1.2% of the entire pediatric diabetes population studied.

Type 2 diabetes mellitus, a diagnosis previously uncommon in childhood, is a growing problem among children and adolescents. Recent estimates suggest that up to 30% of the population will develop diabetes in the current generation. One in every 3 children born in the year 2000 are predicted to develop diabetes.

The growing obesity epidemic makes the diagnosis of diabetes mellitus in childhood more complex as providers are faced with the challenge of determining whether a child has type 1 or type 2 diabetes mellitus in order to direct care and provide prognostic information. However, the specific diagnosis may not be clear at the time of presentation. The classic presentation of hyperglycemia, polyuria, polydipsia and weight loss can be common to both type 1 and type 2 diabetes presentations in childhood. In addition, both diagnoses can present with ketosis, although severe diabetic ketoacidosis is more common in the presentation of type 1, autoimmune-mediated diabetes mellitus.

Contributing factors to the development of T2DM include genetic predisposition, high risk ethnicity, obesity, and inactivity. As indicated in the screening criteria outlined above, individuals who are overweight, defined as BMI greater than the 85th percentile for age and sex based on CDC standards; with a family history of diabetes; of Native American, African American; Latino; Asian American or Pacific Islander ethnicity; with polycystic ovarian syndrome; born small for gestational age or to a mother with gestational diabetes are at increased risk.

There are several high risk ethnicities and predisposing genetic factors. Some large genome wide association studies (GWAS) have begun to identify gene variants that may increase risk in the population at large. However, screening for increased genetic risk is currently done clinically by a detailed family history. There are also epigenetic factors that are likely to play a role, including alterations in the prenatal environment caused by maternal overnutrition or undernutrition and/or gestational diabetes that may increase risk.

The timing of presentation of T2DM is usually during or after puberty. Diabetes evolves on a continuum from normal glucose levels to impaired glucose levels and subsequently to frank diabetes mellitus (see laboratory studies section). Thus, the presentation of T2DM around puberty suggests that the physiologic increase of insulin resistance during puberty in an already obese, genetically predisposed child leads to frank diabetes. T2DM can be present well before the diagnosis is made and come to medical attention through screening paradigms or through an acute symptomatic presentation.

In practice, it is often difficult to determine the type of diabetes (obese T1DM versus T2DM) at the time of presentation. A diagnosis of diabetes is made by demonstrating a random blood glucose level at or above 200 mg/dL plus the classic symptoms of polyuria or polydipsia, OR a fasting blood glucose level of greater than or equal to 126 mg/dL on two occasions, OR an elevation greater than or equal to 200 mg/dL glucose 2 hours after a standard oral glucose load (75 grams), OR elevation of HgbA1C (hemoglobin A1C, glycated hemoglobin, glycosylated hemoglobin) to greater than or equal to 6.5% done by standard assay (defined as National Glycohemoglobin Standardization Program certified and standardized to the Diabetes Control and Complications Trial assay) on two occasions.

Obese children thought to have T2DM based on other risk factors should have pancreatic autoantibodies measured to rule out autoimmune pancreatic destruction as a contributor.

A genetic evaluation for monogenic causes of diabetes (MODY) should be considered in children with a multi-generation family history of non-T1DM, negative pancreatic autoantibodies and lacking classic risk factors or presentation for T2DM (Table II).

Table II.
1. A1C ≥6.5%. The test should be performed in a laboratory using a method that is NGSP certified and standardized to the DCCT assay*
2. FPG ≥126 mg/dl (7.0 mmol/l). Fasting is defined as no caloric intake for at least 8 h*
3. Two-hour plasma glucose ≥200 mg/dl (11.1 mmol/l) during an OGTT. The test should be performed as described by the World Health Organization, using a glucose load containing the equivalent of 75 g anhydrous glucose dissolved in water.*
4. In a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose ≥200 mg/dl (11.1 mmol/l).

*In the absence of unequivocal hyperglycemia, criteria 1–3 should be confirmed by repeat testing. (ADA, Diabetes Care , 2010)

There are limited relevant imaging studies for the diagnosis and management of type 2 diabetes mellitus in children. However, the presentation of diabetes can often be complicated by or confused with other causes of a presentation of acute abdominal symptoms or alteration in consciousness. Thus, abdominal imaging as appropriate to rule out other potential diagnoses, such as appendicitis, cholecystitis, or obstruction should be considered. In addition, DKA can present with elevations in pancreatic enzymes and therefore pancreatic imaging is sometimes considered. Neurologic imaging, such as a head computed tomography (CT) is appropriate when the cause of the alteration of consciousness is not known or to evaluate worsening mental status in DKA.

The diagnosis is confirmed with one of the criteria outlined above including classic symptoms of hyperglycemia (polyuria, polydipsia and weight loss) with a random plasma glucose of greater than or equal to 200 mg/dL, a fasting plasma glucose of greater than or equal to 126 mg/dL, and elevation in blood glucose to greater than or equal to 200 mg/dL when 2 hours after an oral glucose challenge or an elevated HgbA1C. As noted, these are consensus guidelines that establish specific cut-offs to define a disease state that is actually a progressive process. The change in the most recently updated diagnostic criteria include the use of a HgbA1C of greater than or equal to 6.5%. This was done utilizing the NHANES databases for prevalence of complications, specifically retinopathy, in adults.

Determining the type of diabetes in the obese adolescent after the diagnosis of diabetes is confirmed is done by measuring pancreatic autoantibodies as detailed above.

Type 2 diabetes in childhood, as in adulthood, is often an outpatient diagnosis. If this is the case, a referral to a comprehensive diabetes management team is preferred. This may include physicians/nurse practitioners/physician’s assistants/nurses specializing in the care of T2DM in youth; dieticians with expertise in obesity treatment; and mental health professionals with an interest in diabetes and obesity. The outpatient treatment plan must include lifestyle modification goals, including improved dietary intake and increased physical activity. In addition, an important component of initial management is the teaching and implementation of home blood glucose monitoring and ketone monitoring.

Lifestyle modification: Lifestyle modification therapy implemented by a multidisciplinary team is essential for successful treatment. This includes a family based intervention to modify nutrition, physical activity and behaviors related to weight gain. However, diabetes is a progressive process of insulin resistance and beta cell failure and thus over time, patients require medical therapy. Recent studies demonstrate that children may progress even faster than adults to worsening glycemic control and complications. Thus, the correct time for medication implementation in the progression of diabetes remains individualized in children and is often necessary at the time of diagnosis.

In 2013, the American Academy of Pediatrics (AAP) published a clinical practice guideline for the management of newly diagnosed type 2 diabetes mellitus in children and adolescents ages 10 to 18 years. The specific guidelines addressed when to initiate insulin therapy and/or metformin therapy with lifestyle management as primary therapy. In addition, the committee provided monitoring guidance for finger stick blood glucose and HgbA1C measurements.

The AAP guidelines suggest that insulin therapy should be started for newly diagnosed children and adolescents with type 2 diabetes who have ketones detected in urine or blood, are in diabetic ketoacidosis (ph < 7.30, hyperglycemia and ketosis) and in those in whom the type of diabetes (type 1 vs. type 2) is unclear as the clinician will need to default to the care recommended for type 1 diabetes which would mandate insulin initiation. In addition, the guidelines recommend initiating insulin in those children and adolescents who present with significant hyperglycemia, defined by a plasma blood glucose level over 250 mg/dL or HgbA1C > 9 %. Metformin therapy and lifestyle modification, including dietary counselling and emphasis on a goal of 60 minutes daily of moderate to vigorous physical activity is recommended for all other newly diagnosed children and adolescents with type 2 diabetes.

In children without ketosis or symptoms and with HgbA1C of greater than 6.5%, treatment with metformin oral therapy is recommended. The medication is prescribed in an escalating fashion to allow for the assessment of tolerance. For example, 500 mg daily for one week and then increasing to 500 mg bid as tolerated, to a maximum total daily dose of 2000 mg divided twice daily. Metformin is best tolerated when taken with morning and evening meals. An extended release form is also available. The mechanism of action of metformin is to reduce hepatic glucose production as well as increase insulin sensitivity in peripheral tissues.

There are a number of insulin regimens possible including long acting insulin (such as insulin detemir or insulin glargine), intermediate acting insulin (such as NPH) and/or short acting insulin (such as Regular insulin, insulin aspart, insulin lispro or insulin glulisine). The AACE/ACE consensus panel on glycemic control in T2DM in adults outlines the specific insulin types and the reasons for the recommendation of the use of long acting and as needed rapid acting insulin therapy and this is often extrapolated for use in children as well as adults.

In children who present with significant elevations in HgbA1C, ketosis and/or acidosis, an inpatient admission may be required to provide rehydration, resolution of ketosis and acidosis and initiation of insulin therapy. The initial diagnosis of diabetes in a hospitalized patient may occur in the setting of hospitalization for another disease process. Thus, the context of the diagnosis, such as the use of glucocorticoids and/or critical illness must be considered when utilizing the standard diagnostic criteria.

Most pediatric patients with T2DM are managed initially with metformin or insulin and then transitioned to metformin. Adult studies demonstrate that 3 years after diagnosis, half of all patients require a second agent. Data from the TODAY study indicates that some adolescents with T2DM may progress even faster to needing a second agent. When insulin is used as a secondary medication with metformin, the initiation of a long-acting insulin at bedtime may be sufficient to obtain adequate glycemic control for a period of time and may reduce the risk of hypoglycemia. In addition, often insulin is used at diagnosis until glycemia is improved and acidosis is resolved and then metformin is added and increased as tolerated. Utilizing frequent blood glucose monitoring and outpatient follow up, insulin may be reduced and/or discontinued once the patient has been stabilized and is tolerating the metformin therapy.

Thus, most children are treated with metformin and/or insulin plus metformin. There is limited evidence regarding combination therapy with the other available medications that are frequently used in adults. These medications include thiazolidinediones, particularly pioglitazone, and sulfonylureas. Additional agents are beginning to be utilized in older adolescents and are currently under study in this population such as meglitindies, incretin mimetics (exenatide), DPP-IV inhibitors (sitagliptin), amylin receptor agonists (pramlintide), and alpha glucosidase inhibitors (acarbose and miglitol). Of note, only metformin and insulin therapy are FDA approved for use in children. Metformin is the only oral agent approved for use in children ages 10 and older.

Chronic treatment of T2DM in children includes a focus on lifestyle modification, specifically altering dietary content and increasing physical activity to reduce hyperglycemia, improve insulin resistance and reduce the need for antihyperglycemic agents. The specific changes that induce these health improvements remain an area of scientific research. A recent Cochrane review detailed the studies to date investigating lifestyle modification and determined that a comprehensive, multidisciplinary, family centered program was likely to be the most successful, but the specific dietary, physical activity and behavioral components were not conclusively defined. As noted above, the recent AAP consensus recommendations suggested dietary counselling and emphasized a goal of 60 minutes daily of moderate to vigorous physical activity.

Metformin therapy has gastrointestinal side effects in almost 50% of patients taking it, although this does not usually lead to discontinuation of therapy. The gradual escalation of dosage described above usually serves to reduce acute symptoms and discontinuation of therapy. A rare but very serious side effect is lactic acidosis and thus prior to initiation of therapy one must evaluate for renal, hepatic or cardiac dysfunction to reduce this risk by determining when it may not be appropriate to initiate therapy. Recent studies have supported potentially relaxing the restrictions on the use of metformin in renal impairment, but this remains an area of controversy.

In addition, patients are advised to stop metformin during periods of acute illness with dehydration, when instructed to fast, or when having a radiographic study with iodinated contrast. Additional potential side effects include vitamin deficiencies, particularly B12 deficiency. Thus, evaluation with liver function tests, renal function tests and a complete blood count are recommended monitoring before initiating metformin therapy. Furthermore, supplementation with a multivitamin is recommended when prescribing metformin to potentially prevent vitamin deficiencies.

Insulin is often used in children with T2DM as initial therapy or when HgbA1c goals are not reached with metformin alone. The primary side effects of insulin are hypoglycemia and weight gain. Frequent blood glucose monitoring is utilized to reduce hypoglycemia. An intensive nutritional plan can be utilized to limit weight gain and reduce the insulin required to maintain euglycemia.

T2DM is a progressive, lifelong illness often requiring additional therapy over a lifetime. If diagnosed early in the progression to beta cell failure, some children will be able to be maintained on lifestyle modification alone with adequate glucose control. However, inevitably the child will require medication to maintain the goals of therapy, including a HgbA1C of less than 7%. Despite the need for medication, a healthy lifestyle continues to play a major role in the therapy of this condition.

Complications that develop in adults with T2DM are beginning to be documented in children and there may even be more rapid progression to the development of complications in those who are diagnosed with T2DM at a younger age. These complications include retinopathy, albuminuria and nephropathy, peripheral neuropathy and autonomic neuropathy.

In addition, children with T2DM are at increased risk for other components of the metabolic syndrome including hyperlipidemia and hypertension. This combination of risk factors significantly increases the likelihood of early cardiovascular disease states. Furthermore, the metabolic syndrome, and insulin resistance in particular increases the risk for alterations in reproductive function, specifically polycystic ovary syndrome, the combination of clinical and/or biochemical hyperandrogenism and altered menstrual cycling.

Lifestyle modification is an essential component of treatment and remains with limited side effects. The two approved medication therapies in children are metformin and insulin therapy. The risks and benefits of these medications are outlined above.

T2DM in children has been increasing in frequency over the last few decades. In a UK study on the prevalence of T2DM in children less than 17 years of age 0.21 per 100,000 or 1 per 500,000 children had this diagnosis.

The SEARCH for Diabetes in Youth Study (SEARCH study) is a multicenter study of the prevalence in 2001 and the incidence in 2002-2005 of diabetes in children ages 0-19 years of age in representative areas of the United States, including four American Indian populations. The study, which began accruing data in 2000 has found the prevalence of T2DM among 10 to 19 year olds ranged from 0.18/1000 (approximately 1 in 5600) for white children to 1.45/1000 (approximately 1 in 700) for Native American children. Prevalence rates for Black, Hispanic, Asian and Pacific Islander children are intermediate between these two groups. The incidence of T2DM was determined by the SEARCH study to range from 3.7/100,000/year for white children to 27.7/100,000/year for Native American children.

Of note, the prevalence of T1DM remains higher in Caucasian children of all ages. The prevalence of T1DM in white children was 2/1000 and the incidence was 23.6/100,000. Thus, in white children, the large majority of diabetes cases in children less than 20 years of age remains T1DM. This is in contrast to specific Native American populations. For example, in the Navajo component of the SEARCH study over 80% of the children with diabetes were determined to have T2DM. Thus, these studies served to emphasize the increased risk of T2DM in certain ethnic groups.

Family based linkage analyses identified several monogenic causes of diabetes that are distinct from T1DM and T2DM. These include MODY (Maturity-onset diabetes of youth) diabetes; mitochondrial diabetes and neonatal diabetes. The specific genes and the defects are reviewed by McCarthy, NEJM, 2010.

Subsequent association studies have identified several potential candidate genes with common variants that have an impact on the risk for the development of T2DM, (for example PPARG and KCNJ11 which encode drug targets in known therapies). Genome wide association studies (GWAS) have proved to be the most productive tools as this search enables an unbiased discovery process. More recent studies done in larger cohorts have demonstrated multiple genes with small effect sizes. Thus the overall effect size for known common variants is approximately 5-10% for T2DM. Although these findings are driving research into novel mechanisms in the pathophysiologic mechanisms of diabetes development and progression, they each have a modest effect size and thus reconfirm the underlying complexity of this disease process.

The genetic common variants found to effect T2DM demonstrate only a modest effect size. Thus, this remains a multifactorial disease state influenced by genetic risk as well as environmental factors such as the perinatal environment, nutritional intake, physical activity level and other as yet unknown influences that remain areas of active investigation.

Children with T2DM are at an increased risk of mental illness, particularly depression, and report a worse quality of life assessment even when compared to children with T1DM. Depression may be related to poor glycemic control in diabetes as well as reduced function in school and home environments. Therefore, a mental health provider is an essential component of the treatment team for T2DM to evaluate and treat factors that influence disease management.

Microvascular and macrovascular complications may be both more prevalent and more aggressive in young patients, with those who are diagnosed at less than 40 years of age having a 14 fold higher incidence of cardiac disease after 20 years of diabetes. Thus, retinopathy, neuropathy, renal disease or cardiovascular disease may be noted at presentation in young adults with early onset disease. Furthermore, other elements of the metabolic syndrome such as hypertension and hyperlipidemia, as well as sleep apnea, orthopedic problems and psychological problems are frequent comorbidities and should be assessed and potentially treated simultaneously. Thus, early detection, adequate glycemic control and screening for complications are essential to prevent comorbidities and treat early complications following diagnosis.

The increased prevalence of T2DM in childhood has led to the initial investigation of complications of the disease and treatment. In 2007, Pinchas-Hamiel published a review in Lancet of the limited relevant data that has accrued over the last decade. The acute complications of T2DM can include presentation with diabetic ketoacidosis (DKA), and hyperglycemic hyperosmolar coma (glucose level greater than 600 mg/dL, serum osmolality over 330 mOSm/L with mild acidosis and ketosis).

The mortality associated with DKA is known to be approximately 0.15% in the United States and is likely to occur in both type 1 and type 2 diabetes. There have been several case series of mortality from hyperosmolar hyperglycemia in children determined to have T2DM, some from hypercoagulability and others from a very rare progression to rhabdomyolysis. The case reports suggest an increased risk in African-American males.

It is important to note that some children with T2DM and negative pancreatic autoantibodies remain at risk for DKA. These individuals are difficult to distinguish from those with more typical T2DM and thus instruction on the monitoring of ketones during times of illness and/or hyperglycemia is important in all children diagnosed with diabetes.

As noted above, it is difficult to differentiate the risk from secondary comorbidities that are attributable to the obesity that underlies early T2DM such as hypertension, hyperlipidemia, hepatic steatosis, and reproductive abnormalities such as polycystic ovary syndrome (PCOS). These additional comorbidities also contribute to an increased risk of cardiovascular disease in adults studied. There appears to be some evidence that individuals diagnosed at an early age with T2DM may have an increased risk of early complications, especially retinopathy and renal disease.

In addition to the treatment for T2DM specifically, physicians should be aware of the screening and treatment recommendations for common comorbidities. The recommendations include annual screening of fasting lipids, evaluation and treatment of hypertension with monitoring every 3 months, annual dilated retinal examination, annual foot examination, as well as annual liver function studies and urine for microalbumin.

Genetic testing for the monogenic causes of diabetes (MODY) are now commercially available. MODY is often misdiagnosed as the more common type 1 or type 2 diabetes. These monogenic causes of diabetes are thought to comprise approximately 5% of all diabetes cases in the United States, mostly in Caucasians. As noted above, the SEARCH trial found that in youth with diabetes diagnosed at less than 20 years of age, MODY genes were discovered in an estimated 1.2% of the group, when a subset was screened. The inheritance pattern is autosomal dominant but can be sporadic in a given individual. However, a family history of early onset diabetes in three generations should raise the question of MODY in a child diagnosed with diabetes, especially if antibody negative and atypical in presentation.

Although the testing is now commercially available, insurance carriers do not consistently cover the testing and it remains expensive. Thus, the decision to test for MODY can be complicated by these limitations.

The reasons to consider testing are that specific mutation may drive prognosis and therapy. For example, MODY 2 is caused by a loss of function mutation in the gene for the enzyme glucokinase (GCK) which is involved in glycolysis. The mutation serves to reset the glucose level that triggers insulin release resulting in mild hyperglycemia and may be treated with lifestyle modification alone. Recent studies demonstrate that no treatment is necessary and there does not appear to be any long term complications related to the resultant hyperglycemia. In contrast, MODY 1, 3 and 4 (mutations in HNF4A, TCF1 and IPF1, respectively) cause beta cell abnormalities and result in progressive beta cell failure over time. However, these types of MODY respond well to sulfonylurea medications, at least in early stages.

In addition, specific mutations also increase the risk for related complications or abnormalities in multiple organs. For example, MODY 5 (TCF2 mutation) has associated renal abnormalities and increased risk of kidney failure. Thus, MODY should be considered in children without a typical presentation and/or course of diabetes in whom a multi-generation family history of diabetes is present. Specifically a lean or normal weight child, without acanthosis nigricans who presents with mild hyperglycemia and no evidence of insulin deficiency or pancreatic autoimmunity may be a candidate for consideration of MODY testing.

The Diabetes Prevention Program (DPP) demonstrated that both lifestyle modification and metformin therapy may reduce the risk of progression to diabetes mellitus in adults with impaired glucose tolerance. However, there are limited data in children on prevention methods utilizing diabetes as an endpoint. Freemark et al. demonstrated in a 6 month double blinded randomized controlled trial in obese children with an elevated fasting insulin and a family history of T2DM, that metformin given as 500 mg twice daily improved fasting glucose and insulin levels, although it did not improve insulin sensitivity by minimal model analysis.

Atabek et al. utilized a similar model of high risk children using metformin 500 mg twice daily for 6 months and found an improvement in fasting insulin, 120 minute insulin after glucose challenge and insulin resistance by HOMA-IR. Srinivasan et al. utilized 1 gram of metformin twice daily in a crossover study design of 12 months and demonstrated improved fasting glucose and insulin with no significant difference in insulin sensitivity by minimal model analysis.

Thus, the current data support some efficacy of metformin therapy in improving fasting insulin and glucose, but there is limited evidence regarding use in prevention of T2DM in childhood. Furthermore, the lifestyle modification arm of the DPP study, consisting of intensive lifestyle modification with nutritional changes and physical activity supports, demonstrated an impressive risk reduction, suggesting that this should be the primary prevention model in adults. Bell et al. evaluated an 8 week intensive exercise training program in obese adolescents and demonstrated improved insulin sensitivity by hyperinsulinemic clamp.

Other studies have demonstrated some improvements in fasting parameters with exercise interventions. However, the data for supporting a specific intervention for prevention of T2DM in childhood is not adequate. Due to the known benefits and limited side effects of a comprehensive lifestyle modification including alterations in nutritional content, physical activity and behavioral changes, these are currently recommended as the best preventative measures for T2DM in childhood.

Although, GWAS, discussed above, have demonstrated genetic factors that may contribute to this disease process, it is a multifactorial disease and thus genetic counseling does not have a role in prevention, unless a monogenic cause has been determined. However, there is increased risk of T2DM in obese children with a strong family history of T2DM and thus screening and prevention should be initiated at an early age in these families, as noted in the ADA recommendations. There are limited additional predictors in childhood of progression to T2DM other than an increased risk during puberty.

Table III. Studies of type 2 diabetes interventions in children and adolescents

Table III.
References Size and subjects Study design Intervention Duration Primary outcome
Jones et alPMID: 11772907 82 children with type 2 diabetes, BMI > 50th %, 10–16 years old Randomized, placebo-controlled trial Metformin up to 1000 mg po twice daily Up to 16 weeks Change in fasting plasma glucose greater for metformin group: –2.4 mmol/L (–42.9 mg/dL) vs +1.2 mmol/L (+21.4 mg/dL) (P < 0.001)
Sellers et alPMID: 15570994 8 children with type 2 diabetes, 10–18 years old Open, single arm Pre-mixed 30/70 insulin twice daily starting at 0.5 units/kg/day Up to 16 weeks treatment and 12 months follow-up Mean HbA1c at the end of treatment: 7.59% (95% CI, 6.54, 8.64) and at 12 months 7.46% (95% CI, 5.68, 9.24) were significantly different (P < 0.001) from baseline HbA1c 12.81%
Zuhri-Yafi et alPMID: 12017229 25 children with type 2 diabetes, 8–15 years old, BMI > 85th percentile Retrospective chart review Metformin alone, Insulin alone, metformin and insulin Variable, mean treatment duration 27.5 months 72% started on insulin alone at diagnosis and only 28% of these weaned to metformin alone. Mean change in HbA1c was –2.9% for insulin only, –2.3% for metformin + insulin and –4.4% for metformin alone
Kadmon et alPMID: 15506677 18 children with type 2 diabetes, age 14.0 ± 1.9 years Retrospective chart review Metformin alone, insulin alone followed by metformin alone Variable Glycemic control deteriorated when patients changed from insulin to metformin (5.0 ± 2.6 to 8.4 ± 2.9%, P < 0.05). At follow-up, no difference in mean HbA1c between metformin alone and insulin followed by metformin
Gottschalk et alPMID: 17392540 285 children with type 2 diabetes, HbA1c 7.1%–12.0%, and 8–17 years old Single-blind, active-controlled, randomized trial Metformin 500–1000 mg twice daily vs glimepiride 1–8 mg once daily 24 weeks Significant within group reductions in HbA1c but no between group differences. Metformin group –0.54%, P = 0.001; glimepiride group –0.71%, P = 0.0002
Zeitler et alTODAY StudyPMID: 17448130 Approximately 800 children with type 2 diabetes < 2 years’ duration, 10–17 years old Randomized parallel group trial Metformin 1000 mg twice daily alone, metformin 1000 mg twice daily plus intensive lifestyle, metformin 1000 mg twice daily and rosiglitazone 4 mg twice daily 2- to 6-month single blind run-in and treatment up to 5 years Primary outcome is time to treatment failure, defined as either HbA1c > 8% for 6 months or inability to wean from temporary insulin therapy within 3 months following an acute metabolic decompensation (Results expected in 2011).

The use of HgbA1C for the diagnosis of diabetes, the use of bariatric surgery in children to prevent and or treat T2DM and the cardiovascular benefit of tight glycemic control are ongoing controversies.

The debate over the use of HgbA1C for the diagnosis of diabetes mellitus stems from the 2009 addition by the International Expert Committee of the American Diabetes Association to the criteria for diagnosis of diabetes of a HgbA1C of greater than or equal to 6.5%. This cut-off was derived from the use of large epidemiologic studies indicating that a HgbA1C of 6.5% represented an increase in the diabetes retinopathy prevalence in adults. There are several limitations to this definition.

Lee and colleagues utilizing the National Health and Nutrition Examination Surveys (NHANES 1999-2006), concluded that the use of HgbA1C for the diagnosis of diabetes in adolescents may be less sensitive than in adult populations. However, the small number of children with diabetes makes these studies difficult to interpret. Several centers have published data that even with standardized A1C assays, testing on the same patient with different methods can vary over 0.5%, and thus complicate diagnosis. Clinical factors may also play a role in reduced reliability of A1C assays, such as anemia and differential results in various ethnic groups. The added cost of the HgbA1C testing compared to a glucose level has also been raised as a concern in the introduction of this screening option. As noted by the expert panel and subsequent discussants on the topic, the goal was not to introduce HgbA1C as a measure to identify identical prevalences compared to the previous definitions using fasting and post glucose load measurements as these are measuring different elements of glucose metabolism.

HbA1c is now included in the definition of diabetes in both adults and children, but the exclusive use remains an area of controversy. Practically, this measure is useful in clinical screening and monitoring due to the ease of testing using a one-time nonfasting sample.

Bariatric surgery has been found to treat T2DM in obese adults. There is evidence to support that bariatric surgery in adolescents can also improve glycemia without medication therapy. The current American Diabetes Association guidelines for adult bariatric surgery includes consideration in adults with a BMI greater than or equal to 35 kg/m2 and diabetes, especially if diabetes remains in poor control despite attempts at lifestyle and medication therapy. NIH guidelines suggest adolescents should have failed a 6-month weight loss program prior to enrollment, have a BMI > 40 kg/m2, be at final adult stature and have related comorbidities such as type 2 diabetes or sleep apnea (adapted from Inge, et al Archives of Pediatric and Adolescent medicine, 2007). The mental health and support structure in the family must also be considered prior to proceeding to operative therapy.

The use of these procedures in adolescents has expanded and there is recent data published in the NEJM regarding the 3 year outcomes. Inge et al reported that T2DM was in remission in 95% of adolescents who had the diagnosis of type 2 diabetes at the time of surgery (n=29; 19 of 20 who had available data were in remission with normal HgbA1C and fasting blood glucose values). In assessing adverse outcomes, at 3 years 57% of subjects had low ferritin levels, and the frequency of deficiencies in vitamins B12 and A also increased significantly. In addition, 30 participants required additional abdominal surgeries.

Thus, the use of bariatric surgery in adolescents is growing and the resolution of diabetes appearing promising with relatively low frequency of severe adverse events, similar to those outcomes demonstrated in adults.

The outpatient glycemic control controversy involves the debate about whether cardiovascular mortality might be increased by tight glycemic control in adults and how this relates to our goals for control during childhood. Several large scale studies in adults have been completed in an attempt to address this important issue. The first such study was the ACCORD (Action to Control Cardiovascular Disease in Diabetes) study, which was NIH sponsored and included more than 10,000 subjects. This study was stopped at 3 years due to the interim analysis revealing increased mortality in the intensive treatment group.

This caused significant concern about the goal of intensive diabetes management with a goal of reduction in HgbA1C to the normal range. However, the subsequent discussions surrounding the data have revealed that it may in fact be the increased risk of hypoglycemia or other patient specific factors that lead to the increase in mortality. Subsequent studies, the ADVANCE (Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation) study. UKPDS (United Kingdom Prospective Diabetes Study) and the VADT (Veterans Affairs Diabetes Trial) also continued the controversy.

However, many experts in the field continue to support the use of intensive therapy for diabetes and the goal of tight glycemic control (defined in adults as less than or equal to 6.5% HgbA1C by the AACE and less than 7% by the ADA), with the caveat that the plan to obtain these goals is individualized and that the risk of hypoglycemia is minimized. These studies have not been repeated in pediatric populations. Thus, the data in adults is often extrapolated to the pediatric population. However, in children the increased risk to a developing neurologic system that occurs with hypoglycemia makes these controversies even more relevant.

Future studies in children are needed to determine the ideal targets in this population for both inpatient and outpatient glycemia control. Some relevant studies are currently ongoing. Meanwhile, clinically, we continue to utilize the adult standard of 7% as an ideal goal for glycemic control in children. However, we seek to individualize care to limit hypoglycemia and glycemic excursions in pediatric patients.