Like pediatric obesity, the prevalence of type 2 diabetes in childhood is increasing in epidemic proportions. Until recently, type 2 diabetes mellitus (type 2 DM) was considered an ‘adult’ disease and, as of 10 years ago, constituted less than 2% of the new cases of DM in children. Currently, the burden of diabetes, and of obesity the prevalence of which is increasing at a rate of 20-30% per decade, falls disproportionately on African- and Hispanic- Americans, in whom between 25% and 50 % of new-onset childhood diabetics are type 2 [19, 21, 30, 41-43]. Thus, along with the increasing prevalence of obesity among children, type 2 DM has become a “pediatric” disease, producing the same renal, ophthalmological, neurological and cardiovascular morbidities in children with type 2 DM as in type 2 diabetic adults [44]. This is more worrisome because there is increasing evidence that morbidities accrue faster in pediatric-onset compare to adult-onset type 2 diabetes [45].

Obesity is the major risk factor for type 2 DM in adolescents [46, 47] and adiposity accounts for approximately 55% of the variance in insulin sensitivity in children [48]. As in adults, 50-90% of children with type 2 DM, and especially those of African-American or Latino descent, have a BMI>85%ile [19,21,30,42,43,46,47]. After controlling for adiposity and body fat distribtuion, insulin sensitivity has been reported to be approximately 40% higher in African-American than Caucasian pre-pubertal children but about 35% lower in adolescents [49-52], suggesting that the African-American teenager tends to become insulin resistant at a lower level of body fatness.

The pathophysiology of type 2 DM is discussed elsewhere in this textbook 9SEE CHAPTERS 9 AND 27, Diabetes Section). Type 2 DM is a complex metabolic disorder reflecting, in most instances, interactions among genes that influence an individual’s susceptibility to diabetes and an environment which favors the expression of that susceptibility by providing easy access to calorically dense foods and opportunities for a sedentary lifestyle [53]. In studies of adults with a strong family history of type 2 DM, it appears that impaired pancreatic islet-cell function is the first identifiable metabolic abnormality in some subjects who subsequently develop type 2 DM, while in other populations, insulin resistance is the first identifiable phenotype [54, 55], These data, along with the observation that subjects may be insulin-resistant but not diabetic, and that many of the individuals with impaired β-cell function may not go on to develop type 2 DM [56, 57], suggest that type 2 DM is due to a combination of insulin-resistance and impaired β-cell ability to respond to that state of insulin-resistance. In this sense, a state of relative insulin resistance, or the expression of an underlying tendency towards conditions associated with insulin resistance the major causes of which in adolescence would be pubertal hormonal changes and/or obesity, may act to “unmask” a pre-diabetic state of impaired insulin secretion in some individuals. Available evidence suggests that the incidence of type 2 DM in children peaks around puberty, as do the ethnic differences in the prevalence of pediatric obesity [21, 43]. This is not unexpected in view of the known decline in insulin-sensitivity and increase in adiposity in the peripubertal period [47,58,59] and is fully consistent with a model of type 2 DM in children in which increased insulin-resistance unmasks and exacerbates an underlying, perhaps longstanding, impairment of insulin secretory capacity.

Body fat distribution, usually defined on the basis of waist circumference or the ratio of waist-to-hip circumference, is an independent predictor of adiposity-related insulin insensitivity in adolescents and adults [47 Frerichs, 1979 #546,60]. There appear to be effects of ethnicity on the relative impact of body fat distribution on insulin sensitivity. In Caucasian-American children visceral adiposity is the best correlate of hyperinsulinism and insulin secretion during OGTT and of glucose disposal during glucose clamp studies [47]. In African American, but not Caucasian, pre-pubertal children, intraabdominal adipose tissue volume measured by magnetic resonance imaging was significantly correlated with fasting insulin concentrations and with insulin sensitivity as measured by area under the curve (AUC) during oral glucose tolerance testing [50, 51, 61]. Other studies of prepubertal children have found that fasting insulin concentrations and insulin sensitivity are significantly correlated with subcutaneous, but not visceral, adipose tissue volume in African-American prepubertal girls [62]. Because of the increasing frequency of type 2 DM among obese adolescents, and the worsening of diabetes-related morbidities that may result from delayed diagnosis, the clinician should be alert to the possible of type 2 DM in all obese adolescents, and especially those with a family history of early-onset (< 40 years of age) type 2 diabetes [63].

The storage of excess calories as fat would have been highly advantageous to our progenitors and would have conferred a reproductive advantage by increasing survival during periods of prolonged caloric restriction, as well as increasing the fertility of women and enhancing their ability to breastfeed their offspring. The opportunities for our distant forebears to consume calories to the point of morbid obesity and the frequency of their survival to an age at which such morbidities (type 2 diabetes, hypertension, hyperlipidemia) would be evident were both very low. Thus, it is likely through natural selection that the human genome would be enriched with genes favoring the storage of calories as adipose tissue [64, 65] Conversely, there would be very few, if any, evolutionary pressures to discourage obesity and ‘defend’ body thinness.

With the possible exceptions of the rare cases of obesity due to single gene mutations (see below) or specific anatomic/endocrine lesions (see above), body fatness is a continuous quantitative trait reflecting the interaction of development and environment with genotype. Twin and adoption studies indicate that the heritability of body fatness and of body fat distribution in adulthood is 65 to 80%, (approximately equal to the heritiability of height and greater than the heritability of schizophrenia (68%) or breast cancer (45%)) [66]. Recent studies have also identified significant genetic influences (heritability greater than 30%) on resting metabolic rate, feeding behavior, food preferences, and on changes in energy expenditure which occur in response to overfeeding [40,67-74]. Genetic influences on resting energy expenditure (REE) are evidenced by studies demonstrating that African-American children tend to have lower REE than Caucasian-American children, even when adjusted for body composition, gender, age, and pubertal status [75].

The calculation of heritability in twin studies is based on the assumption that each member of a monozygotic or dizygotic pair is reared in the same environment, and that the degree to which body fatness is more similar within mono- than dizygotic twin pairs is due to the greater genetic similarity of identical vs. non-identical twins. Studies comparing adopted children with their adoptive and their biological parents assume that each child shares little or none of the immediate environment with each biological parent, and that the degree to which body fatness is more similar between children and their biologic vs. adoptive parents is due to the 50% of their genotype that each child shares with each biological parent.

There are rare instances of single gene/locus disorders which result in human obesity (e.g. Prader-Willi, Bardet-Biedl, Ahlstrom, Cohen), in association with other often dysmorphic phenotypes [40, 76] (see Table 8). The pivotal role of genetics in the control of body weight is confirmed by the existence of single gene mutations capable of producing profound increases in body fat content. The fact that mutations in different genes can produce obesity suggests that these genes may be part of a control system for the regulation of body weight, i.e., that feeding behavior and energy expenditure are integrated in a system with complex control mechanisms which can be disrupted at many loci. As shown in Table 9, there are human orthologs of known rodent obesity mutations that affect specific aspects of the energy homeostastic signaling system described in chapter 6, Obesity Section.

Similarly, a number of single gene mutations associated with obesity have been identified in rodents. Human orthologs are known for all rodent genes that are associated with obesity and, in the case of Lep and Lepr, human mutations associated with obesity have been reported (see Table 9).

Studies that examine the prevalence of obesity in children conceived during periods of natural or man-made famine such as the Nazi-imposed Dutch famine of 1944-45 (the “Winter Hunger”) [78] report a small but statistically significant increase in the prevalence of obesity (defined as weight for height greater than 120% of WHO standards for 1948) in 19 year old male military recruits whose mothers were malnourished only during the first trimester of pregnancy (2.77% prevalence if mother was in famine area vs. 1.45% if mother was outside of famine area during pregnancy) and a decrease in in the prevalence of obesity among recruits whose mothers were malnourished during the child’s immediate post-natal period (0.82 % if mother was in famine area vs. 1.32% if mother was outside of famine area during pregnancy). It has been hypothesized that early intrauterine malnutrition might affect hypothalamic ("appetite center") development while the anti-obesity effects of early post-natal malnutrition might be due to suppression of adipocyte formation.

Long-term tracking studies of children who are small for gestational age, and therefore demonstrating clinical evidence of prenatal undernutrition, have reported that, even when corrected for adult adiposity, birthweight is negatively correlated with the incidence of adiposity-related morbidities, including type 2 diabetes mellitus, hypertension, stroke, and cardiovascular disease, in adulthood, even when corrected for adult adiposity [79-84]. This association implies an interaction between the prenatal environment and development/function of pancreatic beta-cells or other organs, e.g., hypothalmus, liver, etc.,. that are involved in the regulation of adult energy homeostasis and cardiovascular function. As hypothesized by Barker [85-87] the metabolic, cardiovascular, and endocrine bases for adult adiposity-related morbidities may originate through adaptations that the fetus makes in response to undernourishment. Therefore, the small-for-gestational-age baby should be considered to be at increased risk for adult moribidities that are exacerbated by increased adiposity [64].

A “thrifty genotype” hypothesis and a “catch-up growth” hypothesis have been proposed as mechanistic explanations for the association between low birth weight and adiposity-related co-morbidities. In the 'thrifty genotype' view, intrauterine undernutrition invokes insulin resistance and islet cell dysfunction that favor delivery of nutrients to vital organs over somatic growth [88]. Postnatally, in the nutrient-rich extrauterine environment, this tendency to optimize the efficiency of storing calories in the setting of insulin resistance would favor the development of obesity, diabetes, and dyslipidemia. In the 'catch-up growth' view intrauterine undernutriton provokes decreased insulin and insulin-like growth factor production [89]. When the production of these molecules increases during the calorically-rich post-natal catch up growth phase, tissues respond by becoming insulin and insulin-like growth factor resistant to protect against hypoglycemia [90]. In either view, resistance to the auxotrophic actions of insulin and other growth-promoting factors (e.g., insulin-like growth factors) act to maximize energy storage at the expense of growth in the setting of an unstable food supply and in which genes that reduce insulin secretion or increase insulin resistance also predispose infants to low birthweights. Once adequate nutrients are available, these phenotypes that minimized energy expenditure in growth now predispose to obesity and type 2 diabetes [91, 92]. In addition, it is certainly possible – even likely – that intrauterine dysnutrition also influences brain development in a manner that could also predispose to neuroendocrine and behavioral phenotypes favoring increased adiposity and its co-morbidities. More recently, maternal smoking in the first trimester of pregnancy has been associated with increased body fatness and blood pressure in childhood [93], perhaps via effects on blood flow to the developing fetus.

The infant of the diabetic mother (IDM) is a model for the influences of fetal overnutrition on post-natal adiposity . The high ambient glucose concentrations of the prenatal environmnt stimulates fetal hyperinsulinemia, increased lipogenesis, and macrosomia. Since gestational diabetic women are often obese, it is difficult to separate the metabolic effects of gestational diabetes on subsequent adiposity of the IDM from the possibility that the mother has transmitted a genetic tendency towards obesity to her offspring. In studies controlled for the effects of maternal adiposity, being an IDM is still associated with an increased risk of obesity, independent of the degree of maternal obesity [94-97].

Accurate assessment of the effects of early infant feeding practices on subsequent adiposity must control for possible effects of maternal adiposity as well as socioeconomic status and other factors that may affect the ability to breastfeed [98]. A number of recent well-designed studies have suggested that predominantly breastfeeding for at least 6 months is associated with an approximately 20-30% reduction in the prevalence of obesity (defined as BMI > 95%ile for age and sex) through early adolescence [99, 100] but not into adulthood [101], even when controlled for other adiposity-risk variables. Neither the age at which specific foods are introduced into the diet nor the proportions of fat, carbohydrate or protein in the diet significantly influence subsequent adult adiposity[102, 103]. However, the institution of a well-balanced diet in childhood may form the basis for long-term healthy dietary habits that will significantly lower cardiovascular disease risk even if the diet composition does not substantially affect body composition.

Significant negative correlations have been reported between physical activity and body fatness in pre-school children while positive correlations have been noted between adiposity and the amount of time spent watching television in adolescence [104-108]. Over 60% of television commericals during children’s programs are food-related, and television watching promotes both inactivity and increased caloric intake [109].

Before prescribing any type of treatment for obesity, weight reduction, health personnel should assess the risk/benefit ratio for any treatment in the particular patient. In the older and otherwise healthy overweight child without family history of adiposity-related morbidity, the fact that adolescent obesity may constitute an independent risk factor for adult mortality and morbidity must be weighed against the possible morbidities (poor statural growth, precipitation of eating disorders, etc.) associated with therapeutic weight reduction. Long-term studies of weight-reduced children and adults have shown that 75-95% return to their previous weight percentiles. Obese children and their families must recognize that maintenance of a reduced degree of body fatness will probably require a lifetime of attention to energy intake and expenditure. The cautions emphasized above in deciding who should undergo a therapeutic weight reduction, and the relatively slow rate at which weight reduction of slowing down of weight gain should be prescribed, reflect the significant morbidities associated with these processes. Diets extremely low in caloric content or with unusual distribution of calories as fat, protein, and carbohydrate may precipitate cardiac arrhythmias, severe electrolyte disturbances, or other morbidities. As many as 80% of children using unsupervised diets obtained from popular magazines have been found to suffer from weakness, headaches, fatigue, nausea, constipation, nervousness, dizziness, poor concentration, dysmenorrhea and/or fainting. Children on a supervised diet must also be closely monitored for treatment-associated psychological morbidities (stigmatization of the child, precipitation of anorexia nervosa or bulimia).

Therapeutic intervention should emphasize the need for participation of the entire family and lifelong attention to, and benefits of, a healthy lifestyle, as well as positive reinforcement for even small degrees of compliance. Preparation of the family and child for therapeutic intervention is as important as the intervention itself (see Figures 4 and 5).

Dietary restriction should never be presented in a punitive manner and, if possible, the obese child and the entire family should adhere to a similar diet to minimize feelings of isolation by the obese child. Family members, clinicians, and patients (especially adolescents) will be frustrated by the need for prolonged attention to diet and exercise that is required to achieve and maintain a reduced level of body fatness. Encouragement can be provided by examining growth and growth velocity curves with patients and their families to illustrate progress. If appropriate, the significance of any evident reduction in morbidity (e.g., lowering of blood pressure or cholesterol) can be reinforced. Reasonable goals in the form of a "target" body weight at the next visit should be set at each office visit so that the patient and parents are aware of what is expected. These goals should be modest and attainable even if patients are only moderately compliant with their diet and exercise regimens since achievement of an interval "target weight" will also encourage the patient.

The prescribed diet should initially provide 300 to 400 kcal/day below weight-maintenance requirements as assessed by dietary history or as calculated based upon formula relating anthrompometry to energy expenditure, e.g., the Harris-Benedict Equation [131]. Self-reported caloric intake is generally very inaccurate. The child's ad libitum diet should be directly observed and recorded by the parents for a minimum of 5 consecutive days. A 300-400 kcal per diem energy deficit should result in weight loss of approximately 1 pound per week.. Note that since weight reduction per se causes decreased energy expenditure (both from decreased metabolic mass and whatever hypometabolic state is invoked by losing weight [14, 40, 111] and during weight loss, periodic downard adjustments of energy intake will be necessary to sustain ongoing weight reduction. The family should be instructed in long-term monitoring of caloric intake within- and outside of- the home and cautioned not to become overly critical or punitive towards the child if weight loss is slow or compliance is suboptimal.

The composition of the diet should be in accordance with the American Heart Association “Heart Healthy” recommendations and contain at least the minimal recommend amounts of protein, essential fatty acids, vitamins, and minerals and be low in saturated fats (less than 30% of calories as fat and less than 10% of calories as saturated fat). (The American Academy of Pediatric Statement on Cholesterol in Children describes a healthful, cardiovascular-disease, risk-lowering diet [132].) Diets consisting of drastically altered proportions of nutrients may be dangerous and yield no better results than a limited intake of a nutritionally balanced diet. Nutritional counseling should encourage decreasing the use of calorically dense (high fat or fried) foods, and adding more fruits and vegetables to the daily diet. The substitution of water for non-nutritious high calorie sugar containing drinks (juices, iced teas and soda pop) may be very helpful [130]. In some cases, reductions in calorically dense foods and sugar-containing drinks through substitution and/or elimination alone, can decrease calories and weight without changing the general pattern of food consumption in the family. When families eat at restaurants and fast food places, they have less control over food choices than they do at home. Thus, reduction in the number of meals prepared outside the home may also be an effective weight-loss strategy. Parents and adult caregivers should understand the important role they play in the development of proper eating habits in their young children. The parents’ food preferences, the quantities and variety of foods in the home, the parents’ eating behavior and physical activity patterns all determine how supportive the home environment is to the obese child.

Regular aerobic exercise will allow the patient to ingest more calories and hopefully encourage the long-term continuation of such a regimen. Exercise will promote increased muscle mass, thereby raising total metabolic rate, and the putative effects of exercise to reduce visceral adipose tissue mass independently lower the risk of hyperlipidemia and diabetes mellitus [133-135]. However, the energy cost of even vigorous exercise is low when compared to the caloric content of many "fast foods" or other "snacks", and exercise should not be viewed as a "license to eat". For example, walking at 3 miles per hour for 1 hour consumes about 200 kilocalories, about the same number of calories contained in a 1 ¾ ounce bag of potato chips. Obviously, "treats", such as ice cream, potato chips, etc., should not be used as incentives to exercise.

While no specific aspect of the sedentary lifestyle has been shown to directly cause obesity, behaviors such as television viewing, reading, working at a computer, driving a car or commuting do exert effects on health. Television viewing appears to be directly associated with the incidence of obesity, and inversely associated with the remission of obesity. The impact of television viewing on obesity seems to be due to both displacing more vigorous activities and its effect on diet. Not only is television viewing a sedentary behavior, but also food has constituted the most heavily advertised product on children’s television in the United States. In Mexican-American children, adiposity was significantly correlated with time spent watching television but not with time spent watching videos [136], suggesting that the bulk of the positive association of television watching and adIposity is due to the approximately 60% of advertising that is devoted to food [109]. Children and adolescents should be encouraged to view as little television as possible. Limitation of television, video games, and interrnet surfing will encourage greater participation in physical activity. Clinicians should encourage children to participate in organized or individual sports (participate, not watch from the bench) and advocate for better community- and school-based- activity programs.

If the patient is unable to lose weight and/or co-morbid conditions persist, consideration should be given to referral of the child to a physician specializing in the treatment of pediatric obesity. Weight-loss programs, weight-reduction camps, etc. are often not covered by medical insurance and should be considered for the morbidly-obese child with some caution. Enrollment in a highly supervised environment may demonstrate to an overweight child that weight loss is possible and encourage them to continue. However, rapid weight loss may precipitate cholelithiasis [137] or eating disorders. A child may become overly pre-occupied with his/her weight and, even if a moderate degree of weight-loss is achieved, lose self-esteem. Obsession with weight on the part of the child or their family may lead to serious deterioration of intra-family relationships.