sign in sign up
<<
Home , Childhood obesity

SECTION III

Metabolic Complications of Childhood Identifying and mitigating the risk

1 RAM WEISS, MD, PHD the clinician is faced with the challenge 2 FRANCINE RATNER KAUFMAN, MD of identifying individuals at greatest risk for morbidity. Interventions to halt and promote in The growing number of obese children and adolescents across the world creates a diagnostic children are of limited success and de- challenge to caregivers. The early clinical manifestations of abnormalities related to childhood mand significant resources and contin- obesity, attributed to obesity-driven resistance, are impaired glucose and uous follow-up and monitoring (5). nonalcoholic fatty disease. Both have no symptoms and demand a high index of suspicion This puts the caregiver in the dilemma and the proper choice of tests for establishing the diagnosis. The clinician should gather infor- of where to allocate the limited available mation derived from thoroughly taken history and a focused physical examination to stratify resources and who among the continu- patients by their risk. Focused lifestyle modification–aimed interventions are showing promising ous flux of obese children will benefit results in improving the metabolic profile of obese children. Early diagnosis may help allocate resources for intensive interventions that may benefit individuals at greatest risk for early obesity- most from a focused therapeutic and/or related morbidity. behavioral intervention.

Diabetes Care 31 (Suppl. 2):S310–S316, 2008 IMPACT OF GENETIC, INTRAUTERINE, AND he rise in the prevalence of obesity plications of obesity have their origins CHILDHOOD FACTORS — Theim- in children and adolescents is one during childhood and are closely re- portance of obtaining a meticulous his- of the most alarming lated to the presence of insulin resis- T tory of the obese ’s prenatal period, issues facing the world today. Although tance/hyperinsulinemia, with the most early childhood, and family history can- in the U.S. the problem of childhood common abnormality associated with not be overemphasized. It is well estab- obesity affects all racial and ethnic obesity. The obesity-related morbidities lished that traits and components of the groups, African-American and Hispanic that emerge early in childhood are an tend to cluster in have shown the greatest increases alteration in glucose metabolism and families. In comparison to children whose in the prevalence of obesity over the last fatty infiltration of the liver (nonalco- parents do not meet the criteria for the decade (1). As the prevalence of child- holic [NAFLD]). Al- metabolic syndrome, individuals with at hood obesity increases, its health impli- though an accelerated atherogenic least one parent who meets the criteria for cations are becoming more evident (2). process is present, the clinical manifes- the syndrome have a significantly in- The earliest alterations are abnormali- tations of do not creased odds ratio for having abdominal ties of glucose metabolism that can lead appear in the pediatric age-group. The obesity, for having high triglycerides, and to type 2 . Obesity is associated problematic aspect of diagnosis and risk of meeting the pediatric criteria for the with significant health problems in chil- assessment arises because both im- syndrome (6). Whereas the heritability of dren and is an early risk factor for much paired glucose tolerance (pre-diabetes) the syndrome itself has been reported to of adult morbidity and mortality (3). and NAFLD are conditions with no clin- be in the range of ϳ25%, heritability of Importantly, childhood obesity tends to ical manifestations, and their diagnosis some of its individual components may be track to adulthood and thus represents depends on the right choice of screen- as high as 60% (7). Positive family history an early beginning of a potentially life- ing and diagnostic tests. of has been associated time pathological process (4). Many of As and obese children with early alterations in glucose metabo- the metabolic and cardiovascular com- and adolescents are so commonly seen, ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● lism—mainly reduced insulin sensitivity and a low disposition index (8). Youth 1 From the Department of and Metabolism and the Diabetes Center, Hadassah Hebrew with type 2 diabetes have been reported University School of Medicine, Jerusalem, Israel; and the 2Department of , Keck School of Medi- cine, and the University of Southern California and the Division of and Metabolism, Chil- to have a first- or second-degree relative dren’s Hospital of Los Angeles, Los Angeles, California. with diabetes 75–100% of the time (9). In Address correspondence and reprint requests to Ram Weiss, MD, PhD, Department of Pediatrics and the the National Heart, Lung and Blood Insti- Diabetes Center, Hadassah Hebrew University School of Medicine, PO Box 12000, Jerusalem 91120, Israel. tute Family Heart Study of 445 families, E-mail: [email protected]. genetic correlations between BMI, waist The authors of this article have no relevant duality of interest to declare. This article is based on a presentation at the 1st World Congress of Controversies in Diabetes, Obesity and circumference, HDL , triglyc- (CODHy). The Congress and the publication of this article were made possible by unrestricted erides, insulin, and plasminogen activa- educational grants from MSD, Roche, sanofi-aventis, Novo Nordisk, Medtronic, LifeScan, World Wide, Eli tor-1 antigen were found (10). A positive Lilly, Keryx, Abbott, Novartis, Pfizer, Generx Biotechnology, Schering, and Johnson & Johnson. family history of cardiovascular disease at Abbreviations: ALT, alanine transaminase; IGT, impaired glucose tolerance; IMCL, intra-myocellular lipid; NAFLD, nonalcoholic fatty liver disease. an early age, type 2 diabetes, hyperten- DOI: 10.2337/dc08-s273 sion, or dyslipidemia, specifically in a par- © 2008 by the American Diabetes Association. ent that is not necessarily obese, may

S310 DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 Weiss and Kaufman suggest that the child already is at risk for betes (13). The absence of a similar phe- and a seemingly lower prevalence of dys- the development of adverse outcomes nomenon in offspring of diabetic fathers lipidemia (26). These differences remain and that the addition of obesity may add a suggests a “programming effect” charac- after adjustment for differences in dietary major metabolic burden and thus pro- teristic of the diabetic intrauterine milieu. factors (27). The higher triglyceride levels mote and accelerate pathological pro- Another factor that has to be taken observed in Caucasian youth seem to re- cesses even more. into consideration when assessing chil- sult from differences in VLDL subclasses, A history of being born small for dren and adolescents is their stage of pu- specifically greater large VLDL particles gestational age has been shown to be an bertal development. Mid- is (28). African Americans have also been independent risk factor for the develop- characterized by a reduction of peripheral shown to have larger LDL particles in ment of insulin resistance and the met- insulin sensitivity by ϳ30% (14), thus comparison to their Caucasian peers (29). abolic syndrome in adulthood (11). adding further demand on the ␤-. Similarly, African Americans have a lower Insulin resistance seems to be the key el- prevalence of NAFLD in comparison to ement driving the development of the IMPACT OF RACE AND Caucasians (30). In contrast to their pathophysiological processes leading to ETHNICITY — The rise in the preva- seemingly favorable lipid profile, African- the development of altered glucose me- lence of obesity in children and adoles- American children and adolescents have a tabolism, dyslipidemia, and hyperten- cents in the U.S. is significantly more greater systolic blood pressure per given sion. Newborns who are born small for pronounced in youth of ethnic minorities, degree of obesity and display differences gestational age suffered from a period of specifically Hispanics and African Ameri- in renal handling of potassium (31). limited nutritional resources and/or of the cans. In parallel to the rise of the preva- Recent studies indicate that the rise in effect of other stressogens during the in- lence of obesity, the emergence of type 2 the prevalence of the metabolic syndrome trauterine period. The period of exposure diabetes in the pediatric age-group is alongside the rise in obesity is affecting to limited energy resources is followed by more common in ethnic minorities (15). the Asian pediatric population as well a postnatal period of “catch-up growth” One of the putative explanations for the (32). As Asians differ in their body com- resulting from a practically limitless sup- effect of ethnicity is that youth of ethnic position and lipid partitioning pattern, ply of calories. These intrauterine and minorities are more obese and more in- their risk for the development of adverse postnatal growth patterns possibly cause sulin resistant in comparison to their cardiovascular outcomes begins to rise at a programming of specific in mul- Caucasian peers (16). Obese African- lower BMI thresholds (33). These obser- tiple tissues that promotes efficient energy American and Hispanic children and ad- vations imply that obese children and ad- storage, which in the context of energy olescents with normal glucose tolerance olescents of different ethnic and racial surplus, may lead to early development of have been shown to have an increased background cannot be assessed for meta- insulin resistance. Of note, not all individ- acute insulin response and reduced insu- bolic risk by using a single risk assessment uals born small for gestational age will de- lin clearance per given degree of insulin tool. The clinician taking care of obese velop the metabolic syndrome in sensitivity in comparison to Caucasians children and adolescents should thus take adulthood; thus, a complex interaction of (17–19). This suggests that youth of eth- into account their specific vulnerabilities early intrauterine exposures, specific ge- nic minorities have greater ␤-cell demand and typical clinical manifestations, based netic susceptibility, and several environ- to secrete adequate amounts of insulin as on their race and ethnicity. Whereas se- mental factors may all have independent well as greater circulating insulin concen- vere obesity may be a universal risk factor and cumulative contributions to the de- trations in the face of equal degrees of in- in all children, lower degrees of obesity velopment of the metabolic syndrome in sulin sensitivity in comparison to may put specific children, such as those of adulthood. Caucasians. While the former may facili- Asian ancestry, at risk earlier than others A history of maternal gestational dia- tate increased and earlier ␤-cell fail- and should thus be dealt with earlier. betes also has significant implications for ure, the latter may be the culprit of several Dyslipidemia should similarly be seen in the offspring. Independently of the genet- of the components of the metabolic syn- the context of ethnicity, and African- ics that predispose individuals to develop drome and enhance a vicious cycle that American thresholds that confer meta- type 2 diabetes, the intrauterine exposure promotes further weight gain (20). Dys- bolic risk may be lower than those used to hyperglycemia and hyperinsulinemia lipidemia related to insulin resistance is for the other ethnicities. Further studies probably has a genetic “programming ef- typically characterized by low HDL cho- are needed to evaluate the utility of lipid fect” that affects later extra-uterine energy lesterol and high triglyceride concentra- particle subclasses in the assessment of balance. Maternal is tions (21,22) and is known to increase obese children and adolescents. associated with adiposity and higher glu- cardiovascular disease risk (23). de Fer- cose and insulin concentrations in off- ranti et al. (24) found that low HDL and IMPACT OF THE DEGREE OF spring girls, as early as at the age of 5 years hypertriglyceridemia are the most preva- OBESITY — Classification of the de- (12). There appears to be a U-shaped re- lent metabolic abnormalities present in gree of obesity in adults considers a BMI lationship of birth weight and type 2 dia- adolescents from a population-based co- Ͼ30 and Ͻ35 kg/m2 as class 1 obesity, betes, with children at either end of the hort. As in adults, hypertriglyceridemia BMI 35–39.9 kg/m2 as class 2 obesity, and spectrum at higher risk. In the Pima In- and low HDL were most common among BMI Ն40 kg/m2 as class 3 obesity (34). dian population, it has been shown that Caucasians and least common among Af- Numerous reports have shown that in offspring born after their mother devel- rican Americans (25). Components of the adults, as the degree of obesity increases, oped diabetes were more obese as chil- metabolic syndrome seem to differ be- so does the risk for the development of dren and more likely to have diabetes in tween children and adults of different eth- cardiovascular disease, type 2 diabetes, their 20s than their siblings who were nic background, since African Americans overall mortality, and even cognitive de- born before their mothers developed dia- have a greater prevalence of hypertension terioration. At present, no classifications

DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 S311 Childhood obesity: identifying and mitigating the risk for the degree of obesity exist for children who continued to gain weight rapidly de- viduals to greater insulin resistance, while and adolescents, except for the definition veloped type 2 diabetes (37). Moreover, obesity with low IMCL deposition seems of individuals whose BMI is between the baseline degree of obesity was a strong to be more “metabolically benign”. A par- 85th and 95th percentile as “at risk for predictor of deteriorating glucose toler- adoxical observation in this context is overweight” (equivalent to overweight in ance over time in both individuals with that an increased amount of lipid drop- adults) and those at Ͼ95th percentile as normal or impaired glucose tolerance. lets is present in the myocytes of trained “overweight” (equivalent to obese in This longitudinal assessment was held in athletes, who have very high peripheral adults). Similar to adults, several studies a standard care clinic setting in which pa- insulin sensitivity (43). Possibly, the have recently shown that the degree of tients were seen biannually and received availability of lipotoxic fatty acid deriv- obesity has a similar adverse impact on dietary and physical activity guidance and atives may be influenced by the size of the metabolic profile of obese youth, al- recommendations. Thus, the implications the lipid droplets and their location though no sub-categorization of the de- of these studies is that, among obese chil- within the cell in relation to other cyto- grees of obesity within the upper five dren and adolescents, individuals at the solic structures, such as the mitochon- percentiles exists. Weiss et al. (35) di- 99th percentile in the “severely obese” dria and nucleus (44). vided obese children and adolescents to category are an extremely high-risk group Increased visceral adiposity has also moderately (BMI z score of 2–2.5, corre- for the presence of components of the been shown to be related to a greater athero- sponding to the 97th to the 99.5 percen- metabolic syndrome, adverse biochemi- genic metabolic profile in childhood (45). tile) and severely (BMI z score Ͼ2.5, cal biomarkers, future class 2–3 obesity in Visceral has been shown to be related to corresponding to the 99.5 percentile) adulthood, and progression to diabetes. greater insulin resistance and lower insulin obese and compared them with over- secretory response in obese children and weight and nonobese youth. In that IMPACT OF LIPID adolescents (46). Adiponectin levels are study, increasing obesity categories in PARTITIONING — Although obe- lower in obese children with increased vis- children and adolescents were associated sity is the most common cause of insulin ceral fat deposition (47), even when the with worsening of all components of the resistance in children and adolescents, comparison is made between individuals metabolic syndrome. Specifically, an in- some obese youth may be very insulin with similar overall adiposity. crease in fasting glucose, fasting insulin, sensitive and thus be at reduced risk for The assessment of IMCL is not feasi- triglycerides, systolic blood pressure, and the development of the adverse cardio- ble on clinical grounds; however, an esti- the prevalence of impaired glucose toler- vascular and metabolic outcomes driven mation of visceral fat can be performed ance and a decrease of HDL cholesterol by insulin resistance. In a study aimed at using measurement of waist circumfer- were observed as the degree of obesity discovering the underlying pathophysiol- ence. Although waist circumference pedi- rose. The prevalence of the metabolic syn- ogy of altered glucose metabolism in atric reference charts are not readily drome was ϳ30% in the moderately obese children and adolescents, it was available, use of National Health and Nu- obese participants and nearly 50% in se- clearly demonstrated that individuals trition Examination Survey (NHANES) verely obese participants. When C-reac- with IGT were significantly more insulin derived data has been shown to be clini- tive and adiponectin, used as resistant than individuals with normal cally useful in identifying youth at risk for adverse and protective biomarkers, re- glucose tolerance, despite having an over- the metabolic syndrome (48). The assess- spectively, were tested, C-reactive protein all equal degree of adiposity (38). The dif- ment of waist circumference as well as was positively and adiponectin was nega- ference in insulin sensitivity was BMI should both be performed in obese tively related with the degree of obesity. attributed to different patterns of lipid children and adolescents, since the cumu- C-reactive protein serves as a surrogate of partitioning. Individuals with severe in- lative data may provide a better tool for the subclinical inflammation typical of sulin resistance were characterized by in- risk assessment. obesity. Low adiponectin is known to be creased deposition of lipid in the visceral associated with reduced insulin sensitiv- and intra-myocellular compartments. ALTERED GLUCOSE ity and adverse cardiovascular outcomes. Increased intra-myocellular lipid METABOLISM IN OBESE Thus, adverse biomarkers were already (IMCL) deposition has been shown to oc- CHILDREN — The rise in the preva- present in obese children and adoles- cur early in childhood obesity and be di- lence and severity of childhood obesity cents, and their degree worsened with in- rectly associated with peripheral insulin has been accompanied by the appearance creasing obesity. Freedman et al. (36) sensitivity (39). Importantly, not all obese of a new pediatric disease: type 2 diabetes stratified the Bogalusa cohort participants children have increased IMCL levels, and (2). The prevalence of type 2 diabetes in according to discrete percentiles above those who do not are much more insulin children and adolescents is increasing in the 90th for BMI. They found that indi- sensitive (40). Offspring of diabetic par- both developed and developing countries viduals in the 99th percentile for age and ents have been shown to have lower mito- (49,50). Although the diagnosis of clini- sex had a much greater prevalence of bio- chondrial content in skeletal muscle and cally overt diabetes is not complicated, chemical abnormalities associated with that is postulated to predispose them to the diagnosis of IGT depends on the per- the metabolic syndrome and had a very increased lipid accumulation within the formance of an oral glucose tolerance test high predictive value for adult BMI of myocyte (41). Fatty acid derivates of the in asymptomatic individuals. Although Ͼ35 kg/m2. The only longitudinal study stored IMCL cause a disturbance of the no studies were published regarding the published thus far on the natural history insulin signal transduction pathway, effectiveness of interventions in children of normal and impaired glucose tolerance eventually leading to reduced glucose up- with IGT, with appropriate changes in (IGT) in children and adolescents showed take (42). Thus, a tendency for increased lifestyle and/or pharmacologic interven- that in children with IGT who had greater IMCL deposition, which is partially ge- tions, progression from IGT to frank dia- degrees of obesity at baseline and those netically determined, predisposes indi- betes in adults can be delayed or

S312 DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 Weiss and Kaufman prevented (51,52). The prevalence of IGT amination Survey to be most prevalent in NAFLD can be present without elevated in obese children and adolescents is re- obese African-American and Hispanic ALT levels (71), the clinician should seek ported to be between 10 and 30% in var- males, with type 2 diabetes, hypertension, other clues to suspect this diagnosis. ious studies performed in different and (64). These associa- countries (48,53–55). tions have led to the hypothesis that LIFESTYLE INTERVENTIONS It should be emphasized that the ma- NAFLD may precede the onset of type 2 FOR OBESE CHILDREN — Obe- jority of children and adolescents with diabetes in some individuals. Although sity is perhaps the most critical, and po- IGT and, even some with silent type 2 the natural history of NAFLD in children tentially modifiable, factor in the diabetes, have normal fasting glucose lev- is unknown, it may progress to cirrhosis development of insulin resistance and els (56). This is probably because obese and related complications (65). type 2 diabetes, and adiposity accounts youth with IGT have marked peripheral NAFLD is associated with increased for ϳ50% of the variance seen in insulin insulin resistance, which is mainly at the visceral fat deposition in adults and chil- sensitivity. Adiposity may best be as- level of the muscle. The sensitivity of the dren (66). The association between ab- sessed with BMI or measurement of waist liver to insulin is relatively preserved at dominal obesity and fatty liver may be circumference. Visceral adiposity appears this stage. The American Diabetes Associ- partially explained by sustained exposure to be more correlated with basal and stim- ation published screening guidelines for of the liver to an increased flux of free fatty ulated insulin levels and inversely with type 2 diabetes in children and adoles- acids from the visceral depot (67). insulin sensitivity (72). Physical inactivity cents (57) that recommend performing a NAFLD may represent an early manifes- promotes obesity, insulin resistance, and fasting glucose sample in individuals who tation of ectopic lipid deposition in the diabetes, whereas physical activity and meet criteria for obesity and two addi- liver. It presents a challenge to the clini- reduce these risks. Studies per- tional risk factors (positive family history, cian because it is asymptomatic for the formed in the school setting have shown specific ethnic background, and presence most part but associated with potential the beneficial effects of exercise in chil- of insulin resistance, as evidenced by ac- serious long-term outcomes. As the imag- dren and youth. It has been shown that anthosis nigricans, hypertension, dyslipi- ing modalities of the liver are improving, weight loss in obese adolescents improves demia, or polycystic ovary syndrome). In future noninvasive quantification of liver insulin sensitivity and lowers glucose val- contrast, the World Health Organization lipid deposition may enable detection of ues (73) and that increased aerobic activ- recommends performing an oral glucose early fatty infiltration of the liver. ity lowers insulin levels regardless of tolerance test in individuals at risk. Be- Screening the obese child for fatty weight loss (74). Because the beneficial cause the performance of an oral glucose liver infiltration may be accomplished by effects of both aerobic exercise and resis- tolerance test is costly and labor intensive, measuring alanine transaminase (ALT) tance training can be short lived, exercise it should be performed on individuals at and ␥-glutamyl transferase. However, the and physical activity must be sustained to greatest risk to find IGT (pre-diabetes) sensitivity and specificity of these liver- optimize health and maintain weight loss. and diabetes. derived enzymes is limited. To establish These data support that lifestyle inter- Pre-diabetes is a very dynamic condi- the diagnosis, the “gold standard” diagno- ventions should be targeted to racial/ tion. In a study to determine the outcome sis of NAFLD is based on performance of ethnic minorities, already burdened with of IGT with a limited sample size (37), a liver biopsy, an invasive and not rou- diabetes and obesity. Girls, pubertal or 45% of obese youth with IGT converted tinely used procedure. Thus, the majority prepubertal children, and individuals to normal glucose tolerance and ϳ25% of epidemiological studies on NAFLD in who are overweight or at risk for being progressed to type 2 diabetes over a fol- children are based on surrogate measures overweight, unfit, and sedentary are the low-up period of Ͻ2 years. Individuals such as ALT levels, alongside ruling out likely targets for such programs and the who converted to normal glucose toler- other potential diagnoses. The Third Na- individuals most likely to benefit if long- ance were less obese at baseline and tional Health and Nutrition Examination lasting behavior change occurs. gained minimal weight, whereas individ- Survey data have shown that 6% of over- uals who progressed to overt type 2 dia- weight and 10% of obese adolescents THE GOALS OF BEHAVIOR betes were more obese at baseline and have elevated ALT levels (68), whereas a CHANGE PROGRAMS — The gained on average ϳ27 kg over the fol- population-based study of autopsies re- goals of treating overweight or at-risk for low-up time period. vealed a prevalence rate of 38% in obese overweight children are to decrease body children. In an obesity clinic–based co- weight, optimize , im- NAFLD IN OBESE CHILDREN — hort, 14% of participants had ALT lev- prove well-being and lifestyle, and pre- NAFLD represents fatty infiltration of the els Ͼ 35 U/l, but when liver fat content vent or reverse insulin resistance, liver without excessive alcohol consump- was assessed using magnetic resonance metabolic syndrome, and diabetes and tion (58). The spectrum of NAFLD ranges imaging, 32% of participants had elevated other related comorbidities. These pro- from isolated fatty infiltration (steatosis) liver fat, of whom only ϳ50% had ele- grams must be comprehensive, focusing to inflammation (steatohepatitis, also vated ALT (66). NAFLD should be sus- on optimizing nutrition and weight loss, known as NASH) to fibrosis and even cir- pected in obese children who present in increasing physical activity, and inducing rhosis (59). NAFLD is by no means con- early adolescence with elevated ALT lev- behavior change for the child and family. fined to adults, but is now the most els and no specific complaints. Physical It has been shown that participants in a common liver disease among obese ado- findings may include hepatomegaly, yet child and parent lescents in North America (60,61), with the presence of acanthosis nigricans (69) program have significantly greater de- similar reports coming from other coun- (in up to 50% of cases) and a positive creases in percent overweight 5 and 10 tries (62,63). NAFLD was found in the family history of fatty liver should raise years postintervention (Ϫ11.2 and Third National Health and Nutrition Ex- the index of suspicion (70). Because Ϫ7.5%, respectively) than a child-only

DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 S313 Childhood obesity: identifying and mitigating the risk group, or a group with variable family While lifestyle interventions are safe References participation (75). and effective, pharmacotherapy must be 1. Ogden CL, Flegal KM, Carroll MD, John- Encouraging examples of successful viewed as an adjunct to lifestyle modifica- son C: Prevalence and trends in over- lifestyle modification interventions tar- tion therapy. The use of medication to weight among US children and geted at obese children and adolescents treat obesity, insulin resistance, and the adolescents, 1999–2000. JAMA 288: have recently been described in the liter- metabolic syndrome must be done with 1728–1732, 2002 ature. In an attempt to treat children at caution for the following reasons: few 2. Rosenbloom AL, Joe JR, Young RS, Winter risk for overweight (BMI Ͼ85th percen- drugs are currently approved by the Food WE: Emerging epidemic of T2DM in tile for age) and who are overweight (BMI and Drug Administration for children, youth. Diabetes Care 22:345–354, 1999 Ͼ95th percentile for age), the Center at there is no long-term data, lifestyle has 3. Dietz WH: Health consequences of obe- Children’s Hospital Los Angeles estab- been proven more efficacious for diabetes sity in youth: childhood predictors of lished a family-centered weight manage- prevention (in adults), and there are few adulthood disease. Pediatrics 101 (Suppl. 3):518–525, 1998 ment program called KidsNFitness in well-controlled scientific studies of safety 4. Whitaker RC, Wright JA, Pepe MS, Seidel 1998. The program involves nutritional and efficacy of pharmacological interven- KD, Dietz WH: Predicting obesity in education, family therapy, and exercise tions in children. The relative risk for se- young adulthood from childhood and pa- activities. A recent evaluation of the pro- vere adverse events must be weighed rental obesity. N Engl J Med 337: gram (76) demonstrated that at baseline, against the long-term potential for reduc- 869–873, 1997 49.5% of subjects had multiple risk fac- tion in obesity-related morbidity and 5. Summerbell CD, Waters E, Edmunds tors associated with the metabolic syn- mortality. One must keep in mind that LD, Kelly S, Brown T, Campbell KJ: In- drome that highly correlated with their many drugs previously used for the treat- terventions for preventing obesity in degree of insulin resistance and 10% had ment of adult obesity have resulted in side children. Cochrane Database Syst Rev impaired fasting glucose and/or impaired effects and complications. (3):CD001871, 2005 glucose tolerance. Upon completion of 6. Park HS, Park JY, Cho SI: Familial ag- gregation of the metabolic syndrome in the intervention, 43 subject evaluations Korean families with adolescents. Ath- showed a statistically significant improve- SUMMARY — The clinician who Ϯ erosclerosis 186:215–221, 2006 ment in BMI (pre: 33.65 1.15 vs. post: sees growing numbers of obese children 7. Lin HF, Boden-Albala B, Juo SH, Park N, 2 33.19 Ϯ 1.12 kg/m , P Ͻ 0.005), systolic and adolescents should attempt to iden- Rundek T, Sacco RL: Heritabilities of the blood pressure (pre: 118.3 Ϯ 2.8 vs. post: tify those at greatest risk for the develop- metabolic syndrome and its components 113.3 Ϯ 2.8 mmHg, P Ͻ 0.05), lipids ment of early morbidity. The multiple in the Northern Manhattan Family Study. (total cholesterol pre: 183.0 Ϯ 5.9 vs. factors that affect “baseline vulnerability” Diabetologia 48:2006–2012, 2005 post: 171.8 Ϯ 5.3 mg/dl, P Ͻ 0.005, LDL and additional factors determined by his- 8. Arslanian SA, Bacha F, Saad R, Gungor N: cholesterol pre: 109.9 Ϯ 4.7 vs. post: tory and physical examination should Family history of T2DM is associated with 103.3 Ϯ 4.9 mg/dl, P Ͻ 0.05, and triglyc- guide the clinician in the risk assessment decreased insulin sensitivity and an im- erides pre: 148.1 Ϯ 11.5 vs. post: process. A thoroughly taken pregnancy, paired balance between insulin sensitivity Ϯ Ͻ and insulin secretion in white youth. Di- 120.8 8.7 mg/dl, P 0.05), 2-h post- postnatal, and family history provides abetes Care 28:115–119, 2005 glucose load (1.75 g/kg, max dose 75 g) clues to identify individuals predisposed Ϯ 9. Silverstein JH, Rosenbloom AL: Type 2 glucose (pre: 111.5 4.2 vs. post: to insulin resistance. Clinical judgment diabetes in children. Curr Diab Rep 1:19– 102.5 Ϯ 2.5 mg/dl, P Ͻ 0.05), and leptin should be used to assess the degree of 27, 2001 levels (pre: 32.0 Ϯ 3.8 vs. post: 26.2 Ϯ obesity and to identify greater visceral ad- 10. Tang W, Hong Y, Province MA, Rich SS, 3.0, P Ͻ 0.05) (18). Similarly, the Yale iposity, both of which are strongly associ- Hopkins PN, Arnett DK, Pankow JS, “Bright Bodies” program is comprised of ated with increased metabolic risk. Miller MD, Eckfeldt JH: Familial cluster- an intensive family-based program in- Importantly, anthropometric data should ing for features of the metabolic syn- cluding exercise, nutrition, and behavior be interpreted in the context of the appro- drome: the National Heart, Lung and modification, bi-weekly for 6 months and priate racial or ethnic references, since, Blood Institute Family Heart Study. Dia- bi-monthly thereafter. When comparing for instance, similar degrees of obesity betes Care 29:631–636, 2006 Ͼ 11. Levy-Marchal C, Jaquet D: Long-term 100 obese children and adolescents may confer different metabolic risks in metabolic consequences of being born who completed the program to a group children of different races. Screening for small for gestational age. Pediatr Diabetes who received standard clinical care, clinically silent conditions such as im- 5:147–153, 2004 6-month improvements were sustained at paired glucose tolerance and NAFLD 12. Krishnaveni GV, Hill JC, Leary SD, Veena 12 months, including changes in weight should be based on a high index of suspi- SR, Saperia J, Saroja A, Karat SC, Fall CH: (ϩ0.3 [95% CI Ϫ1.4 to 2.0] vs. ϩ7.7 kg cion, using information gathered from Anthropometry, glucose tolerance, and [5.3 to 10.0]), BMI (Ϫ1.7 [Ϫ2.3 to Ϫ1.1] history taking and anthropometric pa- insulin concentrations in Indian children: vs. ϩ1.6 kg/m2 [0.8 to 2.3]), body fat rameters. It is imperative that lifestyle relationships to maternal glucose and in- (Ϫ3.7 [Ϫ5.4 to Ϫ2.1] vs. ϩ5.5 kg [3.2 to interventions focus primarily on indi- sulin concentrations during pregnancy. 7.8]), and HOMA-IR (Ϫ1.52 [Ϫ1.93 to viduals at greatest risk. From a public Diabetes Care 28:2919–2925, 2005 Ϫ1.01] vs. ϩ0.90 [Ϫ0.07 to 2.05]). health perspective, these programs 13. Franks PW, Looker HC, Kobes S, Touger L, Tataranni PA, Hanson RL, Knowler Thus, a focused intensive lifestyle inter- need to be part of the school curriculum WC: Gestational glucose tolerance and vention program for obese children has and available in communities as well as risk of type 2 diabetes in young Pima In- beneficial effects on body composition in the clinical setting. The challenging dian offspring. Diabetes 55:460–465, and insulin resistance in overweight chil- task of risk determination should be a 2006 dren that are sustained up to 12 months critical focus of primary care and the 14. Goran MI, Gower BA: Longitudinal study (77). pediatric endocrinologist. on pubertal insulin resistance. Diabetes

S314 DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 Weiss and Kaufman

50:2444–2450, 2001 lar risk in a nationally representative sam- 39. Sinha R, Dufour S, Petersen KF, LeBon V, 15. Dabelea D, Pettitt DJ, Jones KL, Arslanian ple. Nutrition 21:718–725, 2005 Enoksson S, Ma YZ, Savoye M, Rothman SA: T2DM mellitus in minority children 28. Freedman DS, Bowman BA, Otvos JD, DL, Shulman GI, Caprio S: Assessment of and adolescents: an emerging problem. Srinivasan SR, Berenson GS: Differences skeletal muscle triglyceride content by Endocrinol Metab Clin North Am 28:709– in the relation of obesity to serum triacyl- (1)H nuclear magnetic resonance spec- 729, 1999 glycerol and VLDL subclass concentra- troscopy in lean and obese adolescents: 16. Arslanian SA: Metabolic differences be- tions between black and white children: relationships to insulin sensitivity, total tween Caucasian and African-American the Bogalusa Heart Study. Am J Clin Nutr body fat, and central adiposity. Diabetes children and the relationship to T2DM 75:827–833, 2002 51:1022–1027, 2002 mellitus. J Pediatr Endocrinol Metab 15 29. Freedman DS, Bowman BA, Otvos JD, 40. Weiss R, Taksali SE, Dufour S, Yeckel (Suppl. 1):509–517, 2002 Srinivasan SR, Berenson GS: Levels and CW, Papademetris X, Cline G, Tambor- 17. Weiss R, Dziura JD, Burgert TS, Taksali correlates of LDL and VLDL particle sizes lane WV, Dziura J, Shulman GI, Caprio S: SE, Tamborlane WV, Caprio S: Ethnic dif- among children: the Bogalusa heart study. The “obese insulin-sensitive” adolescent: ferences in beta cell adaptation to insulin 152:441–449, 2000 importance of adiponectin and lipid par- resistance in obese children and adoles- 30. Louthan MV, Theriot JA, Zimmerman E, titioning. J Clin Endocrinol Metab 90: cents. Diabetologia 49:571–579, 2006 Stutts JT, McClain CJ: Decreased preva- 3731–3737, 2005 18. Arslanian SA, Saad R, Lewy V, Danadian lence of nonalcoholic fatty liver disease in 41. Morino K, Petersen KF, Dufour S, Befroy K, Janosky J: Hyperinsulinemia in Afri- black obese children. J Pediatr Gastroen- D, Frattini J, Shatzkes N, Neschen S, can-American children: decreased insulin terol Nutr 41:426–429, 2005 White MF, Bilz S, Sono S, Pypaert M, clearance and increased insulin secretion 31. Berenson G, Srinivasan S, Chen W, Li S, Shulman GI: Reduced mitochondrial and its relationship to insulin sensitivity. Patel D, Bogalusa Heart Study Group: density and increased IRS-1 serine phos- Diabetes 51:3014–3019, 2002 Racial (black-white) contrasts of risk for phorylation in muscle of insulin-resistant 19. Goran MI, Bergman RN, Cruz ML, Wa- hypertensive disease in youth have im- offspring of type 2 diabetic parents. J Clin tanabe R: Insulin resistance and associ- plications for preventive care: the Boga- Invest 115:3587–3593, 2005 ated compensatory responses in African- lusa Heart Study. Ethn Dis 16 (3 Suppl. 42. Shulman GI: Cellular mechanisms of in- American and Hispanic children. Diabetes 4):S4–2–S4–9, 2006 sulin resistance. J Clin Invest 106:171– Care 25:2184–2190, 2002 32. Yoshinaga M, Tanaka S, Shimago A, 176, 2000 20. Lustig RH: Childhood obesity: behavioral Sameshima K, Nishi J, Nomura Y, 43. Goodpaster BH, He J, Watkins S, Kelley aberration or biochemical drive? Reinter- Kawano Y, Hashiguchi J, Ichiki T, DE: Skeletal muscle lipid content and in- preting the First Law of Thermodynam- Shimizu S: Metabolic syndrome in over- sulin resistance: evidence for a paradox in ics. Nat Clin Pract Endocrinol Metab 2: weight and obese Japanese children. Obes endurance-trained athletes. J Clin Endocri- 447–458, 2006 Res 13:1135–1140, 2005 nol Metab 86:5755–5761, 2001 21. Krauss RM: Insulin resistance syndrome 33. WHO Expert Consultation: Appropriate 44. Cooney GJ, Thompson AL, Furler SM, Ye and dyslipidemia. Endocrine Practice 9 body-mass index for Asian populations J, Kraegen EW: Muscle long-chain acyl (Suppl. 2):67–72, 2003 and its implications for policy and inter- CoA esters and insulin resistance. Ann N Y 22. Reaven GM: Banting lecture 1988: Role of vention strategies. Lancet 363:157–156, Acad Sci 967:196–207, 2002 insulin resistance in human disease. Dia- 2004 45. Bacha F, Saad R, Gungor N, Janosky J, betes 37:1595–1607, 1988 34. Kuczmarski RJ, Flegal KM: Criteria for Arslanian SA: Obesity, regional fat distri- 23. Manninen V, Tenkanen L, Koskinen P, definition of overweight in transition: bution, and syndrome X in obese black Huttunen JK, Manttari M, Heinonen OP, background and recommendations for versus white adolescents: race differential Frick MH: Joint effects of serum triglycer- the . Am J Clin Nutr 72: in diabetogenic and atherogenic risk fac- ide and LDL cholesterol and HDL choles- 1074–1081, 2000 tors. J Clin Endocrinol Metab 88:2534– terol concentrations on coronary heart 35. Weiss R, Dziura J, Burgert TS, Tambor- 2540, 2003 disease risk in the Helsinki Heart Study: lane WV, Taksali SE, Yeckel CW, Allen K, 46. Cruz ML, Bergman RN, Goran MI: implications for treatment. Circulation 85: Lopes M, Savoye M, Morrison J, Sherwin Unique effect of visceral fat on insulin 37–45, 1992 RS, Caprio S: Obesity and the metabolic sensitivity in obese Hispanic children 24. de Ferranti SD, Gauvreau K, Ludwig DS, syndrome in children and adolescents. with a family history of T2DM. Diabetes Neufeld EJ, Newburger JW, Rifai N: Prev- N Engl J Med 350:2362–2374, 2004 Care 25:1631–1636, 2002 alence of the metabolic syndrome in 36. Freedman DS, Mei Z, Srinivasan SR, Be- 47. Lee S, Bacha F, Gungor N, Arslanian SA: American adolescents: findings from the renson GS, Dietz WH: Cardiovascular Racial differences in adiponectin in youth: Third National Health and Nutrition Ex- risk factors and excess adiposity among relationship to visceral fat and insulin amination Survey. Circulation 110:2494– overweight children and adolescents: the sensitivity. Diabetes Care 29:51–56, 2006 2497, 2004 Bogalusa Heart Study. J Pediatr 150:12– 48. Cruz ML, Weigensberg MJ, Huang TT, 25. Srinivasan SR, Frerichs RR, Webber LS, 17, 2007 Ball G, Shaibi GQ, Goran MI: The meta- Berenson GS: Serum lipoprotein profile in 37. Weiss R, Taksali SE, Tamborlane WV, bolic syndrome in overweight Hispanic children from a biracial community: the Burgert TS, Savoye M, Caprio S: Predic- youth and the role of insulin sensitivity. Bogalusa Heart Study. Circulation 54: tors of changes in glucose tolerance status J Clin Endocrinol Metab 89:108–113, 2004 309–318, 1976 in obese youth. Diabetes Care 28:902– 49. Kaufman FR: T2DM in children and 26. Jago R, Harrell JS, McMurray RG, Edel- 909, 2005 youth. Rev Endocr Metab Disord 4:33–42, stein S, El Ghormli L, Bassin S: Prevalence 38. Weiss R, Dufour S, Taksali SE, Tambor- 2003 of abnormal lipid and blood pressure val- lane WV, Petersen KF, Bonadonna RC, 50. Fagot-Campagna A, Pettitt DJ, Engelgau ues among an ethnically diverse popula- Boselli L, Barbetta G, Allen K, Rife F, Sa- MM, Burrows NR, Geiss LS, Valdez R, tion of eighth-grade adolescents and voye M, Dziura J, Sherwin R, Shulman GI, Beckles GL, Saaddine J, Gregg EW, Wil- screening implications. Pediatrics 117: Caprio S: in obese youth: a liamson DF, Narayan KM: T2DM among 2065–2073, 2006 syndrome of impaired glucose tolerance, North American children and adoles- 27. Diaz VA, Mainous AG 3rd, Koopman RJ, severe insulin resistance, and altered cents: an epidemiologic review and a pub- Carek PJ, Geesey ME: Race and in the myocellular and abdominal fat partition- lic health perspective. J Pediatr 136:664– overweight: association with cardiovascu- ing. Lancet 362:951–957, 2003 672, 2000

DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008 S315 Childhood obesity: identifying and mitigating the risk

51. Knowler WC, Barrett-Connor E, Fowler children. Curr Gastroenterol Rep 5:253– insulin resistance, and other clinicopath- SE, Hamman RF, Lachin JM, Walker EA, 259, 2003 ological correlates of pediatric nonalco- Nathan DM: Diabetes Prevention Pro- 61. Lavine JE, Schwimmer JB: Nonalcoholic holic fatty liver disease. J Pediatr 143: gram Research Group: Reduction in the fatty liver disease in the pediatric popula- 500–505, 2003 incidence of T2DM with lifestyle inter- tion. Clin Liver Dis 8:549–558, 2004 70. Willner IR, Waters B, Patil SR, Reuben A, vention or . N Engl J Med 346: 62. Tominaga K, Kurata JH, Chen YK, Fuji- Morelli J, Riely CA: Ninety patients with 393–403, 2002 moto E, Miyagawa S, Abe I, Kusano Y: nonalcoholic steatohepatitis: insulin re- 52. Tuomilehto J, Lindstrom J, Eriksson JG, Prevalence of fatty liver in Japanese chil- sistance, familial tendency, and severity of Valle TT, Hamalainen H, Ilanne-Parikka dren and relationship to obesity: an epi- disease. Am J Gastroenterol 96:2957– P, Keinanen-Kiukaanniemi S, Laakso M, demiological ultrasonographic survey. 2961, 2001 Louheranta A, Rastas M, Salminen V, Dig Dis Sci 40:2002–2009, 1995 71. Schwimmer JB, Behling C, Newbury R, Uusitupa M: Finnish Diabetes Prevention 63. Franzese A, Vajro P, Argenziano A, Puzz- Deutsch R, Nievergelt C, Schork NJ, La- Study Group: Prevention of T2DM mel- iello A, Iannucci MP, Saviano MC, Bru- vine JE: Histopathology of pediatric non- litus by changes in lifestyle among sub- netti F, Rubino A: Liver involvement in alcoholic fatty liver disease. Hepatology jects with impaired glucose tolerance. obese children: ultrasonography and liver 42:641–649, 2005 N Engl J Med 344:1343–1350, 2001 enzyme levels at diagnosis and during fol- 72. Caprio S: Insulin resistance in childhood 53. Sinha R, Fisch G, Teague B, Tamborlane low-up in an Italian population. Dig Dis obesity. J Pediatr Endocrinol Metab 15: WV, Banyas B, Allen K, Savoye M, Rieger Sci 42:1428–1432, 1997 487–492, 2002 V, Taksali S, Barbetta G, Sherwin RS, 64. Meltzer AA, Everhart JE: Association be- 73. Kahle EB, Zipf WB, Lamb DR, Horswill Caprio S: Prevalence of impaired glucose tween diabetes and elevated serum ala- CA, Ward KM: Association between mild, tolerance among children and adolescents nine aminotransferase activity among routine exercise and improved insulin dy- with marked obesity. N Engl J Med 346: Mexican Americans. Am J Epidemiol 146: namics and glucose control in obese ado- 802–810, 2002 565–571, 1997 lescents. Int J Sports Med 17:1–6, 1996 54. Wiegand S, Maikowski U, Blankenstein 65. Feldstein AE, Canbay A, Angulo P, Taniai 74. McMurray RG, Bauman M, Harrell JS, O, Biebermann H, Tarnow P, Gruters A: M, Burgart LJ, Lindor KD, Gores GJ: He- Brown S, Banqdiwala S: Effects of im- T2DM and impaired glucose tolerance in patocyte apoptosis and fas expression are provement in aerobic power on resting European children and adolescents with prominent features of human nonalco- insulin and glucose concentrations in obesity: a problem that is no longer re- holicsteatohepatitis.Gastroenterology125: children. Eur J Appl Physiol 81:132–139, stricted to minority groups. Eur J Endocri- 437–443, 2003 2000 nol 151:199–206, 2004 66. Burgert TS, Taksali SE, Dziura J, Good- 75. Epstein LH, Valoski A, Wing RR, Mc- 55. Shalitin S, Abrahami M, Lilos P, Phillip M: man TR, Yeckel CW, Papademetris X, Curley J: Ten-year outcomes of behavioral Insulin resistance and impaired glucose Constable RT, Weiss R, Tamborlane WV, family-based treatment for childhood tolerance in obese children and adoles- Savoye M, Seyal AA, Caprio S: Alanine obesity. Health Psychol 13:373–383, 1994 cents referred to a tertiary-care center in aminotransferase levels and fatty liver in 76. Monzavi R, Dreimane D, Geffner M, Israel. Int J Obes 29:571–578, 2005 childhood obesity: associations with insu- Braun S, Conrad B, Klier M, Kaufman FR: 56. Libman IM, Arslanian SA: Prevention and lin resistance, adiponectin, and visceral Improvement in risk factors for metabolic treatment of T2DM in youth. Horm Res fat. J Clin Endocrinol Metab 91:4287– syndrome and insulin resistance in over- 67:22–34, 2006 4294, 2006 weight children treated with lifestyle in- 57. American Diabetes Association: T2DM in 67. Montague CT, O’Rahilly S: The perils of tervention. Pediatrics 117:1111–1118, children and adolescents. Diabetes Care portliness: causes and consequences of 2006 23:381–389, 2000 visceral adiposity. Diabetes 49:883–888, 77. Savoye M, Shaw M, Dziura J, Tamborlane 58. Angulo P: Nonalcoholic fatty liver dis- 2000 WV, Rose P, Guandalini C, Goldberg-Gell ease. N Engl J Med 346:1221–1231, 2002 68. Strauss RS, Barlow SE, Dietz WH: Preva- R, Burgert TS, Cali AMG, Weiss R, Caprio 59. Ludwig J, Viggiano TR, McGill DB, Ott BJ: lence of abnormal serum aminotransfer- S: Effects of a weight management pro- Nonalcoholic steatohepatitis: Mayo clinic ase values in overweight and obese gram on body composition and metabolic experience with a hitherto unnamed dis- adolescents. J Pediatr 136:727–733, 2000 parameters in overweight children: a ran- ease. Mayo Clin Proc 55:434–438, 1980 69. Schwimmer JB, Deutsch R, Rauch JB, Beh- domized controlled trial. JAMA 297: 60. Roberts E: Nonalcoholic steatohepatitis in ling C, Newbury R, Lavine JE: Obesity, 2697–2704, 2007

S316 DIABETES CARE, VOLUME 31, SUPPLEMENT 2, FEBRUARY 2008