Objective. Insulin sensitivity is impaired among some obese children, reflecting an atherogenic risk factor profile for the affected subjects. This study was performed to examine the amount of weight reduction required to improve insulin sensitivity.
Methods. We studied changes in insulin sensitivity indices (ISIs) for glucose metabolism (homeostasis model assessment and quantitative insulin sensitivity check index) and fat metabolism (free fatty acids) during a 1-year period among obese children who attended an obesity intervention program. The children were divided into 4 groups according to their changes in body mass index (BMI) SD score (SDS), as follows: group I, decrease in SDS-BMI of ≥0.5; group II, decrease in SDS-BMI of ≥0.25 to <0.5; group III, decrease in SDS-BMI of <0.25; group IV, increase in SDS-BMI.
Results. Fifty-seven obese children (age range: 6–14 years; median age: 10 years; 46% boys) were included in the study. The 4 groups did not differ with respect to age, gender, degree of overweight (SDS-BMI), or ISI values at baseline. An increase in SDS-BMI (group IV, n = 12) was followed by a significant decrease in ISI values. The ISIs improved for group I (n = 9), whereas there were no significant changes in these parameters for group II (n = 21) and group III (n = 15).
Conclusions. During a 1-year period, an increase in weight among obese children was associated with a decrease in insulin sensitivity. Weight loss was followed by significant improvement in insulin sensitivity for glucose and fat metabolism but only if the SDS-BMI decreased by ≥0.5 during the 1-year period.
Obesity is related to an impairment in insulin sensitivity not only in adulthood1,2 but even in childhood.3 Cardiovascular comorbidity of obesity is associated with risk factors such as dyslipidemia, hypertension, and impaired glucose metabolism (metabolic syndrome), which are attributed to disturbances in insulin sensitivity.1,2
Weight loss is the appropriate approach to increasing insulin sensitivity and thus treating cardiovascular risk factors in obesity.4 Studies addressing the amount of weight reduction required to improve insulin sensitivity have not yet been performed among obese children and adolescents. Therefore, we studied insulin sensitivity index (ISI) values of glucose and fat metabolism among obese children and adolescents during a 1-year period, in relation to the degree of weight loss of the patients.
We studied 57 obese children and adolescents (age range: 6–14 years; median age: 10 years; 46% boys) who attended the Obeldicks intervention program for obese children. Children with endocrine disorders or syndromal obesity were excluded from the study. Obesity was defined according to the body mass index (BMI) 97th percentile, reaching BMI values of 30 kg/m2 at 18 years of age, with population-specific data.5
The 1-year outpatient Obeldicks program is based on physical exercise, nutrition education (high-carbohydrate, fat-reduced diet), and behavioral therapy. including individual psychologic care of the child and his or her family.6,7 An interdisciplinary team of pediatricians, dietitians, psychologists, and exercise physiologists are responsible for the training.
Fasting serum insulin, free fatty acid (FFA), and blood glucose concentrations were measured at baseline and 1 year later. The children and their parents were precisely instructed that the children were to fast for a period of ≥14 hours. The children were not observed overnight in a clinical study unit. Insulin concentrations were measured with a microparticle enzyme assay (Abboth, Bau Nauheim, Germany). Blood glucose and FFA concentrations were determined with a colorimetric test (Vitros GLU-Analyseplättchen, Neckar-Gmünd, Germany; Wako Freie Fettsäure, Neuss, Germany). Intraassay and interassay coefficients of variation were <10% for all methods.
Homeostasis model assessment (HOMA)8 and the quantitative insulin sensitivity check index (QUICKI)9,10 were used to detect the degree of insulin sensitivity in glucose metabolism. The sensitivity was calculated from the fasting glucose and insulin concentrations with the following formulas: ISI-HOMA = 22.5/(insulin [mU/L] × glucose [mmol/L]); ISI-QUICKI = 1/[log(insulin [mU/L]) + log(glucose [mg/dL])].
ISI-FFA was calculated with the following formula: 2/(insulin [mU/L] × FFA [mmol/L]) + 1.11 The lower limit of the interquartile range for ISI-HOMA among healthy children of normal weight depends on age and gender (prepubertal boys: 0.87; pubertal boys: 0.49; prepubertal girls: 0.78; pubertal girls: 0.44).12
Height and weight were measured at baseline and 1 year later. The weight status was recorded as BMI and BMI SD score (SDS). Because BMI is not normally distributed, we used the LMS method to calculate SDS-BMI, which summarizes the data in terms of 3 smooth age-specific curves called lambda (L), mu (M), and sigma (S).5,13 The M and S curves correspond to the median and coefficient of variation of BMI, respectively, for German children of each age and gender, whereas the L curve allows for the substantial age-dependent skew in the BMI distribution. The assumption underlying the least mean squares method is that, after Box-Cox power transformation, the data for each age are normally distributed.13
The pubertal developmental stage was determined according to the method of Marshall and Tanner,14,15 and subjects were categorized into 2 groups (prepubertal: boys in pubic hair and gonadal stage I and girls in pubic hair and breast stage I; pubertal: boys in pubic hair and gonadal stage ≥II and girls in pubic hair and breast stage ≥II). The children were divided into 4 groups according to their changes in SDS-BMI at 1 year, as follows: group I, decrease in SDS-BMI of ≥0.5; group II, decrease in SDS-BMI of ≥0.25 to <0.5; group III, decrease in SDS-BMI of <0.25; group IV, increase in SDS-BMI. This division was used because, in a previous study, we demonstrated improvement of cardiovascular risk factors such as hypertension and dyslipidemia only if the SDS-BMI decreased by ≥0.5 during a 1-year period.16 We also studied ISI values during a 3-month period among 10 healthy, normal-weight children (age range: 5-14 years; median age: 11 years; median SDS-BMI: −0.54), without significant weight changes.
Statistical analyses were performed with Winstat software (Fitch Software, Staufen, Germany). P < .05 was considered statistically significant. Values are expressed as median and interquartile range. Statistically significant differences were tested with the nonparametric Kruskal-Wallis test for quantitative items and with the nonparametric Wilcoxon test for paired observations. Changes in insulin indices of glucose metabolism (ISI-HOMA and ISI-QUICKI) during the 1-year period were correlated with changes in insulin sensitivity of fat metabolism (ISI-FFA) and with changes in degree of overweight (SDS-BMI) with Spearman rank correlations. ISI values for the obese and normal-weight children at baseline were compared with the nonparametric Kruskal-Wallis test. Two coefficients of variation were calculated for ISI-HOMA, ISI-QUICKI, and ISI-FFA on the basis of the data for the 10 normal-weight children, representing between-subject variation at baseline and within-subject variation between the 2 measurements during the 3-month period. The study was approved by the local ethics committee of the University of Witten-Herdecke.
The ISI values for the obese and normal-weight children are presented in Table 1. The obese children did not differ significantly, in terms of age (P = .958) and gender (P = .160), from the normal-weight children. The obese children demonstrated significantly decreased insulin sensitivity in both glucose and fat metabolism (Table 1). At baseline, 50 children (88%) had ISI-HOMA values below the interquartile range for healthy children.
ISI values and weight status for the normal-weight children did not change significantly during the 3-month period (change in median ISI-HOMA: 0.09, P = .284; change in median ISI-QUICKI: 0.00, P = .241; change in median ISI-FFA: 0.01, P = .646; change in SDS-BMI: −0.09, P = .114). None of these children changed their pubertal status during this time period. The between-subject variabilities in ISI values were 34% for ISI-HOMA, 6% for ISI-QUICKI, and 34% for ISI-FFA. The within-subject variabilities of ISI values between the 2 measurements were 22% for ISI-HOMA, 20% for ISI-QUICKI, and 7% for ISI-FFA.
None of the 57 obese children discontinued the Obeldicks intervention program. Forty-five participants (79%) in the outpatient training program reduced their weight (SDS-BMI). Age, gender, pubertal stage, and SDS-BMI of the children according to their weight changes are presented in Table 2. The 4 groups did not differ significantly in terms of age (P = .069), gender (P = .790), degree of overweight (SDS-BMI) (P = .435), or ISI values at baseline (ISI-HOMA: P = .091; ISI-QUICKI: P = .092; ISI-FFA: P = .511).
The changes in ISI values are shown in Table 3 and Figs 1 to 3⇓⇓⇓. A decrease in SDS-BMI of ≥0.5 (group I) was associated with significant increases in ISI values (ISI-HOMA, ISI-QUICKI, and ISI-FFA) (Table 3). All 9 children with SDS-BMI decreases of ≥0.5 improved their ISI values for glucose metabolism (ISI-HOMA and ISI-QUICKI) (Figs 1 and 2). Eight of 9 children with SDS-BMI decreases of ≥0.5 improved their ISI values for fat metabolism (ISI-FFA) (Fig 3).
There was no significant improvement in the insulin sensitivity parameters studied for groups II and III. SDS-BMI decreases of <0.5 were associated with both increases and decreases in ISI values (ISI-HOMA, ISI-QUICKI, and ISI-FFA) (Figs 1–3).
When the 3 children with onset of puberty during the 1-year study period were omitted from group II, there were still no significant changes in ISI-HOMA (P = .420), ISI-QUICKI (P = .528), and ISI-FFA (P = .306). When the 4 children with onset of puberty during the 1-year study period were omitted from group III, there were no significant changes in ISI-HOMA (P = .505), ISI-QUICKI (P = .618), and ISI-FFA (P = .286).
Increases in SDS-BMI (group IV) were associated with significant decreases in ISI values (ISI-HOMA, ISI-QUICKI, and ISI-FFA) for 10 of the 12 children (Table 3). The improvement in ISI values for the other 2 children with increasing overweight was minimal (improvement in ISI-HOMA: <0.05; improvement in ISI-QUICKI: <0.01; improvement in ISI-FFA: <0.1) (Figs 1–3). The changes in ISI-FFA were significantly correlated with changes in ISI-HOMA (r = 0.76; P < .001) and ISI-QUICKI (r = 0.76; P < .001).
This is the first study among obese children and adolescents that addresses changes in ISI values in relation to the degree of weight reduction. Obese children demonstrated decreased ISI values, compared with normal-weight children of the same age, according to other studies.17 Significant improvement in all studied ISIs for obese children was demonstrated for the children with the greatest change in BMI (reduction of SDS-BMI of ≥0.5) during the 1-year period, whereas reduction of SDS-BMI of <0.5 showed no significant improvement.
Because of its moderate sample size, our study is not sufficient to define a clear cutoff point for SDS-BMI change that leads to decreases in ISI values. In larger cohorts, insulin sensitivity may significantly improve at a lower level of weight reduction.
A reduction in SDS-BMI of 0.5 can be achieved with nearly stable weight loss during a 1-year period among growing obese children in multidisciplinary outpatient intervention programs.18 In an ongoing study, we demonstrated a mean decrease in SDS-BMI during a 1-year period of ∼0.5 among 132 obese participants in the Obeldicks intervention program.19 In an intention-to-treat analysis, 55% of all participants decreased their overweight by at least 0.5 SDS-BMI.19
Corresponding to our results, SDS-BMI reductions of ≥0.5 among obese children led to improvement of cardiovascular risk factors (reductions in hypertension, triglyceride concentrations, and LDL cholesterol concentrations and an increase in HDL cholesterol concentrations).16 Impairment of insulin sensitivity is the main cause of hypertriglyceridemia, decreases in HDL cholesterol concentrations, and increased blood pressure in obesity and is correlated with the degree of overweight.2,20
The changes in ISI values were attributable to significant changes in serum insulin and FFA levels, whereas fasting blood glucose levels remained stable during the 1-year period. This is in accordance with other studies that demonstrated decreasing FFA and insulin levels during weight loss among children and adults.21,22
The observed changes in the insulin sensitivity in our sample represented the effects of weight loss based on reduced fat intake and/or increased physical activity attributable to an outpatient training program.6,7 There has been controversy regarding the relative effects of energy restriction versus weight loss, with some studies suggesting a dominant role for reduced energy intake23,24 and others for weight loss per se.25,26 Both energy restriction and weight loss have beneficial effects on insulin action.27 Furthermore, physical activity by itself improves insulin sensitivity.28
In our sample, there were significant decreases in ISI values among children with increases in weight (group IV). In addition to the effect of increasing overweight, this could probably be explained in part on the basis of puberty progression during a 1-year period.29 However, no child in this group entered puberty during the study period.
Changes in ISIs for glucose metabolism (ISI-HOMA and ISI-QUICKI) were correlated significantly with changes in ISIs for fat metabolism (ISI-FFA). Furthermore, significant improvement in ISIs for glucose and fat metabolism were observed only among children with SDS-BMI decreases of ≥0.5. Therefore, insulin sensitivity for glucose metabolism seems to be related to insulin sensitivity for fat metabolism.
In our study of an outpatient obesity treatment strategy, obese children's failure to achieve weight loss was associated with decreases in insulin sensitivity after 1 year. More importantly, however, improvements in insulin sensitivity can be expected with reductions in SDS-BMI of ≥0.5. This finding suggests that comprehensive treatment strategies for childhood obesity are able to reduce risk factors for cardiovascular disease such as insulin resistance.
- Accepted May 26, 2004.
- Reprint requests to (T.R.) Vestische Youth Hospital, University of Witten-Herdecke, Dr F. Steiner Strasse 5, 45711 Datteln, Germany. E-mail:
No conflict of interest declared.
- ↵Bonora E, Kiechl S, Willeit J, et al. Prevalence of insulin resistance in metabolic disorders: the Bruneck Study. Diabetes.1998;47 :1643– 1649
- ↵Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care.2001;24 :683– 689
- ↵Allard P, Delvin EE, Paradis G, et al. Distribution of fasting plasma insulin, free fatty acids, and glucose concentrations and of homeostasis model assessment of insulin resistance in a representative sample of Quebec children and adolescents. Clin Chem.2003;49 :644– 649
- ↵Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls. Arch Dis Child.1969;44 :291– 303
- ↵Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in boys. Arch Dis Child.1970;45 :13– 23
- ↵Reinehr T, Andler W. Changes in the atherogenic risk-factor profile according to degree of weight loss. Arch Dis Child.2004;89 :419– 422
- ↵Sarioglu B, Ozerkan E, Can S, Yaprak I, Topcuoglu R. Insulin secretion and insulin resistance determined by euglycemic clamp. J Pediatr Endocrinol Metab.1988;11 :27– 33
- ↵Reinehr T, Kersting M, Wollenhaupt A, et al. Evaluation of the training program “OBELDICKS” for obese children and adolescents. Klin Padiatr. In press
- ↵Nicklas BJ, Rogus EM, Berman DM, Dennis KE, Goldberg AP. Responses of adipose tissue lipoprotein lipase to weight loss affect lipid levels and weight regain in women. Am J Physiol Endocrinol Metab.2000;279 :E1012– E1019
- ↵Hughes TA, Gwynne JT, Switzer BR, et al. Effects of caloric restriction and weight loss on glycemic control, insulin release and resistance, and atherosclerotic risk in obese patients with type II diabetes mellitus. JAMA.1984;77 :7– 17
- ↵Wing RR, Blair EH, Bononi P, Marcus MD, Watanabe R, Bergman RN. Caloric restriction per se is a significant factor in improvements in glycemic control and insulin sensitivity during weight loss in obese NIDDM patients. Diabetes Care.1994;17 :30– 36
- ↵Markovic TP, Jenkins AB, Campbell LV, Furler SM, Kraegen EW, Chisholm DJ. The determinants of glycemic responses to diet restriction and weight loss in obesity and NIDDM. Diabetes Care.1998;21 :687– 694
- ↵Short KR, Vittone JL, Bigelow ML, et al. Impact of aerobic exercise training on age-related changes in insulin sensitivity and muscle oxidative capacity. Diabetes.2003;52 :1888– 1896
- Copyright © 2004 by the American Academy of Pediatrics