Until recently, diabetes mellitus in childhood and adolescence was synonymous with diabetes mellitus type 1 (DM1). However, this concept has changed in recent years with the onset of type 2 diabetes mellitus (DM2) in this population group. DM2 has ceased being a rare disease in the pediatric population. It is currently estimated that between 15% and 45% of new cases of diabetes in childhood and adolescence correspond to DM2,1,2 disproportionately affecting certain ethnic and racial minorities and disadvantaged societal environments.3 In a study conducted in the United States between 2002 and 2012, the relative annual increase in the incidence rate was 1.8% for DM1 and 4.8% for DM2.4

The epidemic of DM2 in childhood is the result of a variety of factors, the most important of which is the increase in pediatric obesity rates, which started in the 1960s, although it appears to be reaching a plateau.5 Another determinant is that DM2 in young patients occurs in complex cultural and psychosocial environments, which makes it difficult to implement lifestyle changes and ensure compliance with medical recommendations. These complexities also hinder the success and completion of research programs and clinical trials,6 leaving large gaps in our understanding of the treatment pathophysiology and optimization.

In this review, we highlight the most relevant epidemiological, pathophysiological, clinical and therapeutic aspects of DM2 in children and adolescents.

Over the past 20 years, we have witnessed a true epidemic of DM2 in children and adolescents.7–9 At the beginning of the 1990s, DM2 represented 3% of cases of pediatric diabetes in the US.10 By 2003, however, DM2 represented 20% of cases of pediatric diabetes, and, depending on the geographical area, almost 50% of cases of diabetes among adolescents aged 15–19 years.11 This increase in DM2 has been confirmed both in developed and developing countries and parallels the increased prevalence of obesity in childhood and adolescence.12,13

The prevalence of DM2 in children and adolescents is estimated at 12/100,000 in the US and 2.5/100,000 in Europe.14 Screening studies on adolescents with obesity 12 years of age or older have observed a DM2 prevalence of 0.4–1%. Another relevant issue is that most cases of DM2 in the young occur in ethnic subgroups such as African Americans, Hispanics, Asian Americans and Native Americans (22.3/1000 in children 10–14 years of age).15

Obesity is the main risk factor for developing DM2. As with adults, central obesity increases insulin resistance and promotes the development of DM2. Excess weight and obesity in children 2–19 years of age is classified according to the percentile of the body mass index (BMI, kg/m2) by age and sex as follows: low weight (BMI<5 percentile), normal (BMI within 5–85 percentile), excess weight (BMI within 85–95 percentile), obesity (BMI≥95 percentile) and severe obesity (BMI>120% of the 95 percentile or BMI>99 percentile).

The increase in excess weight and obesity rates in school-age children has been occurring worldwide in recent decades, reaching 34% and 10%, respectively, in America.16 A third of adolescents with DM2 had not been previously diagnosed, and these patients had more obesity (BMI, 35.7kg/m2) than those who did know the diagnosis of DM2 (BMI, 29.1kg/m2).17

According to the cutoffs of the International Obesity Task Force, the highest excess weight and obesity rates in Europe among 7 to 11-year-old children are in Malta (35%) and Spain (34%).18

In Spain, obesity has grown exponentially in recent decades. According to data from the 2012 National Survey of Health in Spain, the prevalence of excess weight and obesity among 2–17 year olds is currently 27.6%, with the autonomous community of the Canary Islands in first place with 34.5%.19

Pediatric obesity is intimately tied to changes in eating habits, physical activity, physical inactivity, socioeconomic level and length of sleep. A study conducted with Spanish schoolchildren showed that BMI was inversely proportional to how often they performed physical activity.20 Likewise, the study observed an increase in BMI in the schoolchildren who spent more than 2h a day watching television and in those who slept less than 8h daily. Other factors that influenced BMI were of a financial and educational nature in the family environment, such that families with higher incomes and education levels had a lower rate of excess weight and obesity.21 It has also been reported that adverse experiences during childhood, such as threats to the child's physical, family or social safety and childhood violence and abuse and school bullying were associated with an increased risk of DM2 in adulthood.22

DM2 is a complex metabolic disorder in which various societal, behavioral and environmental risk factors act on a foundation of genetic susceptibility. The disease has a strong hereditary component (probably polygenic), which is responsible for the differences in the prevalence of DM2 in different racial groups.23

DM2 is characterized not only by hyperglycemia and insulin resistance but also by a relative reduction in insulin secretion. Both characteristics are influenced by genetic and environmental factors. Insulin resistance explains the clinical association of DM2 with obesity. Patients have a combination of differing degrees of insulin resistance and reduced insulin concentrations. Hyperglycemia by itself can change the function of pancreatic beta cells and exacerbate insulin resistance, resulting in a vicious cycle that worsens the metabolic condition. Adolescents and adults typically have lost 80% of the function of pancreatic beta cells before the diagnosis of DM2.1

The progression of a normal blood glucose condition to that of impaired glucose tolerance (IGT) is associated with a worsening of insulin resistance. IGT is an intermediate stage in the natural history of DM2 and is a predictor of the risk of developing DM and cardiovascular diseases.24,25 Likewise, it has been proposed that hyperglycemia can worsen both insulin resistance and insulin secretion, facilitating the transition of IGT to diabetes.25 However, a high rate of spontaneous conversion to normal glucose tolerance has been observed in children and adolescents with IGT within 3–5 years.25–27

Puberty appears to play an important role in the development of DM2 in children.23 It has been postulated that an increase in growth hormone secretion could be responsible for insulin resistance during this period of development.28 This hypothesis could explain why the presentation of DM2 in children coincides with the mean age of puberty.25–29

The adverse effect of obesity on glucose metabolism is already apparent in childhood. Children with obesity have hyperinsulinemia and an approximately 40% lower insulin-stimulated glucose metabolization (assessed by blood glucose under fasting conditions and 2h after eating) than children who have no obesity.30 The inverse relationship between insulin sensitivity and abdominal fat distribution is stronger for visceral fat than for subcutaneous fat.30,31

The adipose tissue that increases with obesity synthesizes and secretes metabolites and signaling proteins such as leptin, adiponectin and tumor necrosis factor alpha.32 These factors change insulin secretion and sensitivity and can even cause insulin resistance in experimental and clinical conditions.33

The racial differences in insulin sensitivity are well-established. African-American children 7–11 years of age have significantly higher insulin concentrations than white children of the same age.34,35 These data suggest that certain ethnic groups have a genetic predisposition to insulin resistance, which can increase their risk of developing DM2. These racial differences have also been observed in Europe, where Swedish children with obesity have higher glucose concentrations under fasting conditions than German children with obesity.35

It has also been observed that a low pancreatic reserve and a high glycated hemoglobin (HbA1c) level are independent predictors for the loss of glycemic control in patients undergoing drug treatment.36,37 The deterioration in pancreatic beta cell function is also observed in adults with DM2, although it is not as accelerated as in the young.38–41

DM2 in the young occurs predominantly in the female sex. The mean age at diagnosis is 13.5 years, although DM2 almost always starts at the age of 10 years, unlike DM1, which starts before the age of 10 years in 50% of cases.29 In most cases, DM2 is associated with obesity as the main determinant.29

The clinical presentation in children can be diverse and varies from asymptomatic hyperglycemia to ketoacidosis (6% of cases in patients between 10–19 years of age).42 Unlike the onset of DM1, the cardinal symptoms of DM2 are unclear. Thus, children with DM2 usually present glycosuria without ketonuria, mild or nonexistent polyuria, polydipsia and little or no weight loss. Although it is uncommon, children with DM2 are at risk of presenting nonketotic hyperosmolar hyperglycemic decompensation, which is associated with high mortality.43 Differing clinical characteristics, depending on the patients’ ethnicity and race, have also been observed, with a predominance of DM1 in whites and DM2 in at-risk ethnic groups29 (African Americans, American Hispanics, Asians and Native Americans). The differential characteristics between the two types of diabetes are shown in Table 1.44

One of the characteristics of adult-onset DM2 is the lack of beta cell antibodies. However, a number of studies have observed the presence of these antibodies in almost 30% of children and adolescents with DM2 in Europe.45 This group of adolescents does not usually require insulin therapy, at least during the first year. It has been postulated that this clinical form is an early presentation of latent autoimmune diabetes mellitus in adulthood.46 In adolescents, the term “latent autoimmune diabetes mellitus in youth” has been proposed.47

Maturity-onset diabetes of the young (MODY) is a form of maturity diabetes that occurs in children. MODY is an uncommon form of diabetes in children that includes several forms caused by monogenic defects in beta cell function (the most common of which are the hepatocyte nuclear factor 4 alpha gene mutation and the glucokinase gene mutation). Patients with MODY have dominant genetic traits, do not have obesity and have low insulin concentrations under fasting conditions. These genetic anomalies are believed to be rare and require molecular diagnostic techniques. Recent studies have suggested that the clinical presentation spectrum of MODY is broad and ranges from asymptomatic hyperglycemia to severe acute onset.48 This variety of diabetes has been reported in all races and ethnicities. The main characteristics of MODY are listed in Table 1.

Diabetes mellitus can be classified reliably, in most patients, based on the clinical presentation and progression.24 In the unusual situation in which a more precise classification is needed, other specific measurements might be required, such as insulin and peptide C concentrations under fasting conditions and, occasionally, beta cell antibody levels.14 The diagnostic criteria for DM in children and adolescents does not differ from those established by the American Diabetes Association (ADA) for adults.49Fig. 1 shows a flow diagram for classifying DM in children and adolescents.

Figure 1.

Flow diagram for the classification of diabetes mellitus in children and adolescents.

Abbreviations: DM1: type 1 diabetes mellitus; DM2: type 2 diabetes mellitus; MODY: maturity onset of diabetes mellitus in youth.

Source: Reinehr.14

(0.14MB).

Most white European children and adolescents with DM2 and a third of children in the US were asymptomatic at the time of diagnosis.29 Accordingly, DM2 screening studies appear to be necessary, given that hyperglycemia can go unnoticed and contribute to long-term microvascular and macrovascular damage. A DM2 screening for the entire young population is currently not cost effective; it is therefore only advisable to study patients at greater risk. Due to the close direct relationship with obesity, the ADA recommends DM2 screening for children and adolescents with excess weight at the onset of puberty who also have other risk factors (Table 2).49 The screening test of choice is measuring fasting plasma glucose and the oral glucose overload test (OGTT). However, these recommendations have a number of drawbacks. Fasting plasma glucose is a comfortable, simple and inexpensive procedure but failed in a quarter of patients in a European study.29 The OGTT can be a better screening method but costs more, is more labor intense and at times has little reproducibility.50 We also do not know if the cutoffs for OGTT established for adulthood are applicable for childhood. With regard to the use of HbA1c for diagnosing DM2, the ADA continues to recommend it despite its limitations for children and adolescents due to its low sensitivity and high cost.51–53

The American Academy of Pediatrics has published guidelines on treating DM2 in children and adolescents.54 Most of these recommendations have been extrapolated from experience with adults.55 The 3 main treatment objectives are lifestyle changes, blood glucose normalization and control of associated comorbidities.24 The final treatment objective is to decrease the risk of acute and chronic complications.

Pharmacotherapy is indicated if the treatment objective (HbA1c <7.5% and blood glucose under fasting conditions <130mg/dL) is not achieved through nutritional education and exercise. There are many available drugs for treating diabetes, although only metformin and insulin have been approved by regulatory agencies for use in patients younger than 18 years.24 Most pediatric diabetologists use oral agents as the first therapeutic approach because they facilitate therapeutic compliance.

Metformin, a biguanide, is unquestionably the most appropriate starting point for the pharmacotherapy of DM2 in children, as it is for adults.69 The drug reduces hepatic glucose production and increases hepatic and muscle sensitivity to insulin without a direct effect on beta cell function. Metformin has the advantage of slightly reducing weight, improving the lipid profile without increasing the risk of hypoglycemia and is easy to use and inexpensive. The initial dose is 500mg/d and is increased progressively to 1000g/12h within 4 weeks, unless there are adverse gastrointestinal effects. The drug's most common adverse effects are gastrointestinal abnormalities (bloating and diarrhea) and, although it has a good safety profile, should not be administered in cases of hypoxemia, severe infection, severe liver or kidney disease or alcohol abuse, due to the risk of lactic acidosis. Metformin should be used with caution and its dosage adjusted if the patient has impaired renal function. If monotherapy with metformin is ineffective after 3–6 months, combining another drug should be considered.

The efficacy of metformin in adolescents has been tested in 2 clinical trials.34,70 In the Treatment Options for type 2 Diabetes in Adolescents and Youth (TODAY) trial,34 children and adolescents were distributed to the 3 treatment arms (metformin, metformin plus rosiglitazone and metformin plus lifestyle changes). The results suggest that in standard clinical practice a large portion of young patients with DM2 could require combined therapy with insulin a few years after the diagnosis. In this study,34 monotherapy with metformin was associated with adequate blood glucose control in approximately half of the patients with DM2. The marketing of rosiglitazone was restricted by the Food and Drug Administration in 2010 after a slight increase in cardiovascular risk was observed. Another study that compared metformin versus placebo observed a 1–2% reduction in HbA1c after 4 weeks of therapy, with a similar tolerance to that reported in adults.70

The use of insulin is preferred for patients with ketosis or intense hyperglycemia and who have mixed characteristics of DM1 and DM2. The new guidelines recommend insulin therapy when the plasma glucose concentrations are ≥250mg/dL or HbA1c levels are >9, with monitoring every 3 months to adjust the dosage or withdraw it if the control is optimal.71

There is no specific contraindication for insulin therapy in children, and the most appropriate administration regimen for each patient's lifestyle should be selected. The most common adverse effects are weight gain and hypoglycemia. NPH insulin once a day or basal insulin (glargine, detemir or degludec) at an initial dose of 0.25–0.5units/kg is often effective for metabolic control.72

Although studies have been conducted on the use of other hypoglycemic agents (thiazolidinediones, sulfonylureas, meglitinides, alpha-glucosidase inhibitors, dipeptidyl peptidase-4 inhibitors, glucagon-like peptide-1 analogs and amylin analogs) in this population group, their use has still not been approved.73

The comorbidities that accompany DM2 in the young are very common and need to be properly approached to prevent the development of vascular complications.

The SEARCH study observed that 92% of the adolescents with DM2 met the criteria of metabolic syndrome by presenting 2 risk factors for cardiovascular disease, as well as DM, compared with 14% of the patients with DM1.77 In addition to obesity, the presence of other comorbidities such as arterial hypertension, dyslipidemia, retinopathy, nephropathy and depression should be ruled out. The recommendations for their screening and treatment are shown in Table 3.78

In a study conducted in the US that included children and adolescents with DM1 or DM2 with a mean disease duration of 7.9 years, the patients with DM2 had a higher prevalence (adjusted for age) of diabetic nephropathy, retinopathy, peripheral neuropathy, arterial stiffness and hypertension than those with DM1.79

We therefore need to pay attention to preventing excess weight and obesity in this population group through public education in healthy life habits and early detection in consultations.

The authors declare that they have no conflicts of interest.