Impaired glucose tolerance and type 2 diabetes mellitus: a new field for pediatrics in Europe

Transcription

1 (25) 29, S136 S142 & 25 Nature Publishing Group All rights reserved /5 $3. PAPER Impaired glucose tolerance and type 2 diabetes mellitus: a new field for pediatrics in Europe S Wiegand 1 *, A Dannemann 1, H Krude 1 and Annette Grüters 1 1 Paediatric Endocrinology, Charité Children s Hospital, Humboldt University, Berlin, Germany OBJECTIVE: Several studies have shown that obese children and adolescence seem to have an increased risk to develop a disturbed glucose metabolism as already known for obese adults. This might result in the same disastrous outcomes of cardiovascular diseases as it has been shown for adult obese patients. The most sensitive measurement for detecting changes in glucose metabolism in obese children seems to be an oral glucose tolerance test (OGGT) which is not practical for all daily outpatient clinics. DESIGN: Cross-sectional study. SUBJECTS AND MEASUREMENTS: We therefore made a preselection from a cohort of 491 subjects according to the American Diabetes Association (ADA) criteria for the diagnosis of diabetes. In the selected high-risk subgroup (n ¼ 12) of obese pediatric subjects, we measured the prevalence of impaired glucose tolerance (IGT) by OGTT. RESULTS: We diagnosed six patients with type 2 diabetes and 37 patients with impaired glucose tolerance. In addition, we found a close correlation of IGT to disturbances of triglyceride and cholesterol parameters. This prevalence was comparable to a similar study group that was screened without preselection. CONCLUSIONS: These prevalence data further underline the need to diagnose children with obesity-associated risk factors in terms of an insulin resistance syndrome. The preselection of a high-risk subgroup by ADA criteria might be a practical approach. (25) 29, S136 S142. doi:1.138/sj.ijo.8381 Keywords: juvenile type 2 diabetes mellitus; impaired glucose tolerance; childhood obesity; metabolic syndrome Aims Childhood obesity has reached epidemic dimensions and is present not only in the United States but even worldwide. 1 The prevalence of childhood obesity has more than doubled during the last two decades 2 in many regions of the world. Approximately 22 million children under 5 y of age are overweight. 3 In the multiethnic population of Berlin, in % of the preschool children were defined as obese, and in 21, 13%. 4 This epidemic of childhood obesity is associated with an increasing number of children and adolescents with type 2 diabetes, 5 originally reported predominantly in ethnic groups with a high basic risk for type 2 diabetes. 6 These data were confirmed in multiethnic groups in the United States and Italy. 7,8 First cases of type 2 diabetes in obese white children in Great Britain were reported recently. 9 In adults, the progression from normal to impaired glucose tolerance (IGT) and to type 2 diabetes is caused by insulin *Correspondence: Dr S Wiegand, Paediatric Endocrinology, Charité Children s Hospital, Humboldt University, Berlin, Germany. susanna.wiegand@chartie.de resistance followed by decreased B-cell insulin-secreting capacity. IGT harbours risk factors for cardiovascular disease. 1,11 Recently, it has been shown that with changes in lifestyle progression from IGT to type 2 diabetes can be prevented or delayed. 12 Severe obesity in childhood is a major risk factor in the pathogenesis of type 2 diabetes, but it is not yet clear which other clinical, biochemical or genetic factors determine manifestation. Therefore, we investigated the prevalence of type 2 diabetes, IGT and cardiovascular risk factors in a multiethnic European cohort of severely obese children and adolescents. Based on these results, we critically analyzed the recommended screening methods for the diagnosis of impaired glucose regulation. Methods Study population We examined a cohort of 491 obese children and adolescents (age 7 18 y) of a multiethnic European origin for endocrine and metabolic abnormalities. Patients with obesity syndromes and other illnesses were excluded from this study. All 489 patients underwent a standard baseline diagnostic

2 Table 1 Clinical characteristics according to sex and age group Characteristics Male (N ¼ 51) Female (N ¼ 51) Age (range) (y) (6 17) (6 17) BMI (kg/m 2 ) BMI-SDS Ethnic group (N) Caucasian Non-Caucasian 6 5 Plus minus values are means7s.e. Glucose and insulin concentrations were determined, and categorized using World Health Organization (WHO). 19 Impaired fasting glucose was defined as fasting plasma glucose of mg/dl ( mmol/l) (equals 1 19 mg/dl in venous whole blood). IGT was defined as a 2-h plasma glucose level of mg/dl (equals mg/dl in venous whole blood); type 2 diabetes was defined as a fasting glucose level of 126 mg/dl (7 mmol/l) (equals 11 mg/dl in venous whole blood) or higher or a 2-h plasma glucose level of 2 mg/dl (11 mmol) (18 mg/dl in venous whole blood) or higher. 2 S137 procedure including HDL-, LDL-, total cholesterol and triglyceride levels as well as baseline levels of T3, T4 and TSH. We selected a high-risk group from this cohort according to the criteria of the American Diabetes Association (ADA) for type 2 diabetes screening in children and adolescents. 13 The ADA recommends that children and adolescents should be tested for type 2 diabetes if they are obese and display at least one other risk factor, such as a positive family history of type 2 diabetes and signs of insulin resistance (acanthosis nigricans, dyslipidemia, hypertension, polycystic ovary syndrome or an ethnic predisposition). 13 These criteria were met by 51 boys and 51 girls (Table 1) who underwent a 2-h oral glucose tolerance test (OGTT). None of the other patients had an impaired fasting glucose or elevated HbA1c level. Body weight was measured with a digital scale (Soehnle, Germany) to the nearest.1 kg, and height was measured with a wall-mounted stadiometer (Keller, Germany). The body mass index (BMI) was calculated (the weight in kilograms divided by the square of the height in meters). All subjects had a BMI over the 97th percentile for age and sex and were thus classified as obese. The degree of obesity was quantified using Cole s least mean square method, which normalizes the BMI skewed distribution and expresses BMI as an standard deviation score (SDS-BMI). 14,15 Blood pressure was measured in a supine position after 5 min rest to the nearest 2 mmhg using a Dinamap s model and results were interpreted using percentiles for age and sex. 16 Pubertal status was assessed according to the criteria of Tanner. 17,18 None of the subjects was taking any medications. Oral glucose tolerance test The 12 subjects selected because of a high risk for type 2 diabetes followed a diet consisting of at least 25 g of carbohydrates per day for 3 days before the study. After a 12 h overnight fast, these subjects underwent an OGTT at 8 hours. Baseline samples were obtained for measurement of fasting (venous) glucose, fasting insulin, testosterone, thyroid hormones and lipids. Glucose was given orally (1.75 g/kg, up to a maximum of 75 g glucose). Blood samples were drawn after 3, 6, 9 and 12 min. Biochemical analysis Glucose was measured by the glucose oxidase method on venous whole blood immediately deproteinized with perchloric acid. The obtained values were transformed into plasma glucose values in order to calculate the insulin resistance index (HOMA) and the insulin sensitivity index (ISI) by adding 1% to the venous whole blood glucose level. The lipid levels were determined by an enzymatic colorimetric test (analyser 74/717 Roche-Hitachi). The thyroid hormones and testosterone were measured with a fluoroimmunoassay (AutoDELFIA s ). Insulin was measured with a radioimmunoassay (Pharmacia). The crossreactivity with C-peptide was o.18% and with proinsulin 41%. The intra-assay variation was 5.8% for insulin and the interassay variation was 7%. Calculations Two approaches were used to characterise insulin sensitivity: (1) The homeostasis model assessment for insulin resistance index (HOMA-IR) 21 was estimated using the formula: HOMA IR ¼ðfasting insulin ðmiu=lþ fasting glucose ðmmol=lþþ=22:5 Lower insulin resistance values indicate a higher insulin sensitivity, whereas higher values indicate a lower insulin sensitivity. The insulin resistance index correlates well with measures of insulin resistance obtained with the use of the euglycemic hyperinsulinemic clamp technique both from obese and nonobese children 7 and adults. 21,22 (2) Insulin sensitivity was determined by the ISI of Matsuda and De Fronzo. 23 ISI ¼1 =square root of ðfasting glucose ðmmol=lþ fasting insulin ðmiu=lþþðmean glucose ðmmol=lþ mean insulin ðmiu=lþ during OGTTÞ b-cell function was determined by insulinogenic index, 24 expressed as the ratio of the increment in the insulin level to that in the glucose level during the first 3 min after the

3 S138 ingestion of glucose. In children and adolescents, the insulinogenic index correlates well with the early insulin response obtained during a hyperglycemic clamp study. 7 A low insulinogenic index predicts the development of diabetes in adults. 25,26 Statistical analysis All values are expressed as means7s.e. Differences in the means of variables were tested using both parametric and nonparametric tests depending on the distribution of the variables. The OGTT data were analyzed using repeated measures analysis of variance. Time was treated as a repeated measures factor, and normal/pathologic glucose tolerance was used as a between-subject factor. Correlation analyses were conducted using Spearman s correlation coefficients or Pearson s correlation coefficients, depending once again on the distribution of the variables. A probability value of less than.5 was considered significant. SPSS version 11 (SPSS, Chicago, IL, USA) was used for analysis. The Ethic committee of the Humboldt University of Berlin approved the study. Written informed consent was obtained from the parents and the subjects. Glucose level (mg/dl) Insulin Level (uu/ml) Minutes 9 12 normal IGT diabetes normal IGT Results Prevalence of IGT and type 2 diabetes In all, 36.3% of the children and adolescents (N ¼ 37) had an IGT (Table 2). In 5.9% (N ¼ 6), type 2 diabetes was diagnosed. Among the children and adolescents with IGT, 86% were Caucasian and 14% were non-caucasian. Four Caucasian 3 6 Minutes 9 12 diabetes Figure 1 Mean (7s.e.) glucose and insulin responses during the oral glucose tolerance test in children and adolescents with normal glucose tolerance, impaired glucose tolerance and type 2 diabetes mellitus. Table 2 Clinical and metabolic phenotype of obese children and adolescents with normal glucose tolerance, IGT or type 2 diabetes Normal glucose tolerance (N ¼ 59) IGT (N ¼ 37) P-value Type 2 diabetes (N ¼ 6) Sex (N) Male Female Age (y) Caucasian (N) Non-caucasian (N) BMI (kg/m 2 ) P ¼ SDS-BMI P ¼ Insulin resistance index Po.1* ISI Po.1* Insulinogenic index P ¼ Fasting insulin level (mu/ml) Po.1* Cholesterol (mg/dl) P ¼ Triglycerides (mg/dl) P ¼.1* HDL-cholesterol (mg/dl) P ¼ LDL-cholesterol (mg/dl) P ¼ Plus minus values are means7s.e. All P-values are noted for the comparison with the group with normal glucose tolerance. The insulin resistance index was determined by homeostatic model assessment and calculated as the product of the fasting insulin level (in microunits per milliliter) and the fasting glucose level (in millimoles per liter), divided by *Po.5.

4 4. Table 3 Comparison of criteria for the diagnosis of impaired glucose regulation according to ADA and WHO S139 Mean Insulinogenic Index ADA criteria (fasting glucose) (N) WHO criteria (OGGT) (N) Impaired fasting glucose 2 12 IGT 37 Type 2 diabetes mellitus 2 6 Normal glucose regulation Total tolerance (P ¼.1). However, this increase in the glucose level at 3 min was not associated with a significant increase in insulin levels. Consequently, the calculated insulinogenic index was slightly but not significantly lower than that among subjects with normal and IGT (P ¼.1). Figure 2 Mean (7s.e.) of the change in insulin to the change in glucose (insulinogenic index) in obese children and adolescents with normal glucose tolerance (), impaired glucose tolerance (1) or diabetes mellitus (2). Values are in mg/dl, to convert to mmol/l multiply by adolescents and two non-caucasian adolescents had type 2 diabetes. Glucose and insulin responses to an oral glucose challenge Results of the OGTT in 12 children and adolescents are summarized for insulin and blood glucose (Figure 1). Insulin and glucose values changed significantly over time (Po.1). Fasting insulin levels (Table 2) were higher in children and adolescents with IGT or diabetes than in subjects with normal glucose tolerance. Furthermore, the insulin response to an oral glucose charge was elevated in children and adolescents with IGT as compared with the responses in those with normal glucose tolerance. The fasting blood glucose levels were higher in children and adolescents with IGT ( mg/dl) than in those with normal glucose tolerance ( mg/dl) (P ¼.2). Similarly, at 12 min, glucose levels were higher in children and adolescents with IGT ( mg/dl) than in those with normal glucose tolerance ( mg/dl) (Po.1) (Figure 1). Early-phase insulin secretion There were no significant differences in the early changes in the insulin level (within 3 min), glucose level or the insulinogenic index between children and adolescents with normal and IGT (Figure 2). In contrast, subjects with type 2 diabetes showed a significantly greater increase in the glucose level at 3 min compared with those that occurred in subjects either with IGT (P ¼.5) or normal glucose Fasting glucose level vs 2-h glucose level: comparison of WHO and American diagnostic criteria The ADA encourages the use of fasting glucose rather than the OGTT for the diagnosis of diabetes and impaired glucose regulation. 27 The WHO, however, recommends the OGTT as a screening procedure. 2 The differences in diagnostic criteria led to different results regarding prevalence rates. In this study, far more subjects were identified as having an impaired glucose regulation using WHO criteria compared to the ADA criteria. Four patients with type 2 diabetes would have been missed if only fasting blood glucose would have been employed as a diagnostic criteria (Table 3). Risk factors associated with insulin resistance and insulin sensitivity The correlation between HOMA-IR and ISI was r ¼.93 (Po.1). Both measures were strongly correlated with fasting insulin, with r ¼.98 (Po.1) and r ¼.9 (Po.1) for HOMA-IR and ISI, respectively. Both HOMA-IR and ISI were also correlated with the 2-h glucose level, with r ¼.44 (Po.1) for HOMA-IR and r ¼.55 (Po.1) for ISI. In addition HOMA-IR, ISI and fasting insulin were correlated to cardiovascular risk factors (Table 4). Nonparametric tests were used to compare differences in the means of HOMA-IR and ISI between different groups of children and adolescents. Puberty. The study population was divided into a prepubertal group (Tanner stage I) and a group approaching or at puberty (Tanner stages II V). Children at Tanner stage I (N ¼ 29) showed significantly lower values for HOMA-IR and higher values for ISI compared with children and adolescents at Tanner stage II V (P ¼.2 and P ¼.7).

5 S14 Table 4 Correlations of cardiovascular risk factors with insulin resistance, insulin sensitivity and fasting insulin levels (N ¼ 12) Variable Mean s.e. Correlation r with HOMA Correlation r with ISI Correlation r with fasting insulin Total cholesterol (mg/dl) LDL-cholesterol (mg/dl) HDL-cholesterol (mg/dl) (*).3**.2(*) Triglycerides (mg/dl) **.4**.33** Uric acid (mg/dl) (**).3(*).3(**) Systolic blood pressure (mmhg) (*).2*.2 Diastolic blood pressure (mmhg) SDS-BMI ** We calculated the Spearman correlations of HOMA, ISI and fasting insulin with fasting total cholesterol, LDL-cholesterol, HDL-cholesterol, triglycerides, uric acid, systolic and diastolic blood pressure and BMI-SDS. The correlation coefficients are presented in Table 4, along with means7s.e. *Po.5. **Po.1. Sex. There were no significant differences between girls and boys with concern to insulin resistance or insulin sensitivity. However, girls with increased levels of total testosterone showed significantly higher values for HOMA-IR and lower values for ISI compared with girls with normal hormone levels (P ¼.39 and P ¼.45). Acanthosis nigricans. The study population was divided into a group with and without acanthosis nigricans. Children and adolescents with this sign showed significantly higher values for HOMA-IR and lower values for ISI (P ¼.1 and Po.1). Ethnic origin. Caucasian and non-caucasian subjects showed similar HOMA-IR and ISI. The insulin resistance syndrome We evaluated the degree to which children and adolescents show early risk factors for diabetes type 2 described as the insulin resistance syndrome. 28 In our study population of 12 obese children and adolescents, four showed all signs of the insulin resistance syndrome, that is, insulin resistance, hypertension, dyslipoproteinemia and increased uric acid. In 8% of the children and adolescents, at least two of these signs were present. In all, 43 of the obese children and adolescents suffered from an impaired glucose regulation (N ¼ 37 with IGT þ N ¼ 6 with type 2 diabetes). This group was analyzed separately: in 27 of these subjects (63%), other signs of the insulin resistance syndrome, such as hyperlipidemia, hypertension or elevated uric acid, were present. Fasting total cholesterol and fasting triglycerides were significantly higher and HDL-cholesterol significantly lower among the children and adolescents with IGT and diabetes than among those with normal glucose tolerance (Table 2). No differences in systolic and diastolic blood pressure were observed between children and adolescents between the two groups. Conclusions Several studies have shown that obese children and adolescence seem to have an increased risk to develop a disturbed glucose metabolism as already known for obese adults. This might result in the same disastrous outcomes of cardiovascular diseases as it has been shown for adult obese patients. Therefore, obese pediatric patients need to be monitored for the occurrence of impaired glucose metabolism. The most sensitive measurement for detecting changes in glucose metabolism in obese children seems to be an OGGT since Sinha et al 7 has shown a low prevalence of impaired fasting glucose in obese children and adolescents with IGT defined by OGGT (less than.8 vs 25%). The same conclusion was drawn by the DECODE study group for the diagnosis of IGT in adult obese patients. 29 However, the actual high prevalence of obesity in childhood populations makes it difficult to screen every obese child by an OGGT; a more practical approach is therefore warranted. In order to test for a higher efficacy to diagnose impairments in glucose regulation, we estimated the prevalence of IGT by OGGT in a cohort of Caucasian obese children, which were preselected for the criteria suggested by the ADA as risk factors for type 2 diabetes, for example, a positive family history of type 2 diabetes and clinical signs of insulin resistance like acanthosis nigricans. The prevalence of IGT was compared to the prevalence defined by fasting glucose levels in the same individuals. From a large number of children and adolescents (n ¼ 491) with obesity attending the outpatient department, we selected a high-risk group of 12 individuals according to the ADA criteria for diabetes screening. In this group, a high prevalence of IGT was found based on OGGT (37%). In contrast, in the same group the fasting glucose level detected only 32% of the subjects with an IGT and only 33% of the patients with diabetes, which confirms the need to use the much more sensitive measurement of OGGT compared to fasting glucose. In a recent study, Wabitsch et al have estimated the prevalence of IGT in a comparable obese pediatric cohort

6 from Germany with the same mainly Caucasian background. In this study, OGGT was measured in 52 subjects without preselection revealing 6.7% of impaired glucose regulation. 3 Both cohorts are similar in size, ethnicity and BMI values. Consistent with these similar cohort characteristics, both measurements revealed a comparable number of subjects with impaired glucose regulation (35/52 vs 43/491). However, a much larger number of subjects were needed to investigate by the invasive OGGT to diagnose the patients in the cohort of Wabitsch et al. 3 This comparison might point to the fact that preselection of a cohort with increased risk for diabetes is more efficient to define obese pediatric patients with impaired glucose regulation and might be more practical for daily outpatient care. This needs now to be established by direct comparisons within the same cohort. Another interesting comparison results in the difference of IGT prevalence in the more multiethnic cohort of Sinha et al, 7 who found a much higher prevalence in their nonpreselected cohort (25%) compared to the non-preselected cohort of mainly Caucasian subjects of Wabitsch et al 3 (6.7%). These data further support the idea that the multiethnic pediatric population reflects a group of children with a higher vulnerability to develop IGT in a given range of obesity. 31 However, the possibility that duration of obesity before measurement of OGGT might also influence the likelihood to develop IGT and which might be longer in the US cohort cannot be excluded and might be one parameter that should be further included in prospect studies. Concerning the incidence of type 2 diabetes, the three studies revealed comparable results. Wabitsch et al 3 found eight patients (1.53%), Sinha et al 4 twelve patients (2.3%) and our study six patients (1.22% of the original cohort of 491 children). Four of these were of Caucasian and one of non-caucasian origin. All had a positive family history for type 2 diabetes. In contrast, the four adolescents with type 2 diabetes described by Sinha et al 7 belonged to ethnic risk groups for type 2 diabetes: two black Americans and two Hispanic Americans. These data indicate that type 2 diabetes in adolescents is a still low but emerging problem in Europe and the US and seems not only to occur in ethnic minority groups 32 but also in multiethnic mostly Caucasian European populations. In general, during the course of occurrence of type 2 diabetes in obese individuals, during earlier time points increasing glucose responses are compensated by equally increased insulin responses. Later in the presence of diabetes, however, insulin secretion is no longer sufficient to compensate for extremely high glucose levels. This is also demonstrated by a lower insulinogenic index in children and adolescents with diabetes compared to those with normal or IGT. Therefore, also our data with lower insulinogenic index in the type 2 diabetes subjects compared to the IGT subjects suggests that while insulin resistance and IGT are already present, regular b-cell function is still preserved. These results are in accordance to those reported by Sinha et al. 7 There are associations between different features of the insulin resistance syndrome that suggests that a clustering of risk factors is present in obese children and adolescents. As part of the combined appearance of obesity related risk factors in our study population, children and adolescents with IGT presented also significantly elevated total cholesterol and triglyceride levels and lower HDL-cholesterol levels compared to children and adolescents with normal glucose tolerance. Furthermore, insulin resistance index was closely correlated with hypertriglyceridemia and the ISI with reduced HDL-cholesterol levels. Arslanian et al 33 described correlations of basal insulin levels with VLDL and triglycerides in healthy lean children and adolescents. In her study, diastolic blood pressure was negatively correlated with insulin sensitivity. In our study, only the systolic blood pressure was negatively correlated with ISI. In a recent study including obese African-American children aged 5 1 y, insulin resistance was inversely correlated with increases in blood pressure and triglycerides, but it was not correlated with LDL- or HDL-cholesterol. 34 Together, these data from different studies underline the concern that also in the pediatric obese cohorts the presence of several signs of the insulin resistance syndrome might predict that they are at increased risk of developing severe cardiovascular complications in the adult age. 35 Therefore, these patients should be considered as high-risk patients who need a close follow-up and intervention. In summary, these data indicate that also in Europe impaired glucose regulation and other signs of the insulin resistance syndrome in children and adolescents are far more common than so far believed and that is not restricted to ethnic minority groups. To identify these subjects, the pre-selection of a risk group according to the ADA criteria followed by OGGT might be more efficient and still sensitive compared to the screening of all obese children by OGGT. Irrespective of the methods to identify these children, their most likely increased risk to develop later on cardiovascular complications force us to diagnose them early enough and to refer them more consequently to a lifestyle modification and therapeutic intervention. References 1 Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight prevalence and trends for children and adolescents. The National Health and Nutrition Examination Surveys, 1963 to Arch Pediatr Adolesc Med 1995; 149: Chinn S, Rona RJ. Can the increase in body mass index explain the rising trend in asthma in children? Thorax 21; 56: Deckelbaum RJ, Williams CL. Childhood obesity: the health issue. Obes Res 21; 9 (Suppl 4): 239S 243S. 4 Delekat D, Kis A. Gesundheitsberichterstattung Berlin - Zur gesundheitlichen Lage von Kindern in Berlin -. Berlin: Senatsverwaltung Arbeit Soziales Frauen 23; 2: Rosenbloom AL, Joe JR, Young RS, Winter WE. Emerging epidemic of type 2 diabetes in youth. Diabetes Care 1999; 22: S141

7 S142 6 Fagot-Campagna A. Emergence of type 2 diabetes mellitus in children: epidemiological evidence. J Pediatr Endocrinol Metab 2; 13: Sinha R, Fisch G, Teague B, Tamborlane WV, Banyas B, Allen K, Savoye M, Rieger V, Taksali S, Barbetta G, Sherwin RS, Caprio S. Prevalence of impaired glucose tolerance among children and adolescents with marked obesity. N Engl J Med 22; 346: Invitti C, Guzzaloni G, Gilardini L, Morabito F, Viberti G. Prevalence and concomitants of glucose intolerance in European obese children and adolescents. Diabetes Care 23; 26: Drake AJ, Smith A, Betts PR, Crowne EC, Shield JP. Type two diabetes in obese white children. Arch Dis Child 22; 86: Polonsky KS, Sturis J, Bell GI. Seminars in Medicine of the Beth Israel Hospital, Boston. Non-insulin-dependent diabetes mellitusfa genetically programmed failure of the beta cell to compensate for insulin resistance. N Engl J Med 1996; 334: Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK. Cardiovascular risk factors in confirmed prediabetic individuals. Does the clock for coronary heart disease start ticking before the onset of clinical diabetes? JAMA 199; 263: Tuomilehto J, Lindstrom J, Eriksson JG, Valle TT, Hamalainen H, Ilanne-Parikka P, Keinanen-Kiukaanniemi S, Laakso M, Louheranta A, Rastas M, Salminen V, Uusitupa M, Finnish Diabetes Prevention Study Group. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 21; 344: Type 2 diabetes in children and adolescents. American Diabetes Association. Pediatrics 2; 15: Cole F, Preece MA. The LMS method for constructing normalized growth standards. Eur J Clin Nutr 199; 44: Kromeyer-Hausschild W, Kunze D, Wabitsch M, Geller K. Perzentile für den Body-mass-Index für das Kindes-und Jugendalter unter Heranziehung verschiedener deutscher Stichproben. Monatsschr Kinderheilkd 21; 149: Institute NH. Task force on blood pressure control in children. Pediatrics 1987; 79: Marshall T. Variations in patterns of pubertal changes in girls. Arch Dis Child 1979; 44: Marshall T. Variations in patterns of pubertal changes in boys. Arch Dis Child 1979; 45: Alberti Z. New diagnostic criteria and classification of diabetesagain? Diabetes Med 1998; 15: Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: Diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetes Med 1998; 15: Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB, Monauni T, Muggeo M. Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 2; 23: Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999; 22: Phillips C, Hales CN, Osmond C. Understanding oral glucose tolerance: comparison of glucose or insulin measurements during the oral glucose tolerance test with specific measurements of insulin resistance and secretion. Diabetes Med 1993; 11: Efendic L. Aspects of the pathogenesis of type 2 diabetes. Endocr Rev 1984; 5: Haffner SM, Miettinen H, Gaskill SP, Stern MP. Decreased insulin secretion and increased insulin resistance are independently related to the 7-year risk of NIDDM in Mexican-Americans. Diabetes 1995; 44: Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997; 2: Reaven GM. Role of insulin resistance in human disease. Diabetes 1988; 37: Glucose tolerance and mortality: comparison of WHO and American Diabetes Association diagnostic criteria. The DECODE study group. European Diabetes Epidemiology Group. Diabetes Epidemiology: Collaborative analysis OF Diagnostic Criteria in Europe. Lancet 1999; 354: Wabitsch M, Hauner H, Hertrampf M, Muche R, Hay B, Mayer H, Kratzer W, Debatin k, Heinze E. Type II diabetes mellitus and impaired glucose regulation in Caucasian children and adolescents with obesity living in Germany. Int J Obes Relat Metab Disord 24; 28: Fagot-Campagna A, Saaddine JB, Engelgau MM. Is testing children for type 2 diabetes a lost battle? Diabetes Care 2; 23: Ehtisham B, Barret TG, Shaw NJ. Type 2 diabetes mellitus in UK childrenfan emerging problem. Diabetes Med 2; 17: Arslanian S, Suprasongsin C. Insulin sensitivity, lipids, and body composition in childhood: is syndrome present? J Clin Endocrinol Metab 1996; 81: Young-Hyman D, Schlundt DG, Herman L, De Luca F, Counts D. Evaluation of insulin resistance syndrome in 5- to 1-year old overweight/obese African-American children. Diabetes Care 21; 24: Freedman DS, Khan LK, Dietz WH, Srinivasan SR, Berenson GS. Relationship of childhood obesity to coronary heart disease risk factors in adulthood: the Bogalusa Heart Study. Pediatrics 21; 18: