Trends, associations and predictions of insulin resistance in prepubertal children (EarlyBird 29)

    loading  Checking for direct PDF access through Ovid

Abstract

Background

Rising obesity has been observed in all age groups. Anthropometric cut-points have been used to predict metabolic risk in children, although they are not based on known outcomes.

Aim

We examined the trends, associations and predictions of metabolic health from anthropometry in prepubertal children.

Method

Three hundred and seven healthy children were examined annually between 5 and 8 yr. Measures: height, weight, body mass index (BMI), sum of skinfold thickness at five sites (SSF) and waist circumference (WC). Outcome measures: homeostasis model assessment of insulin resistance (HOMA-IR), high-density lipoprotein cholesterol (HDL-C) and triglycerides (TG).

Results

Two hundred and thirty-one [131 boys (B) and 100 girls (G)] children had complete data sets at all four time points. (i) All measures of adiposity rose from 5 to 8 yr (BMI – B: +3.4%, G: +5.7%; WC – B: +10.4%, G: +11.8%; SSF – B: +23.3%, G: +30.7%, all p < 0.001). HOMA-IR unexpectedly fell (B: −16.6%, p = 0.01; G: −32.5%, p < 0.001). This fall was significant between 5 and 6 yr in both genders (5–6 yr – B: −17.8%, p < 0.001; G: −20.0%, p = 0.002) and between 6 and 7 yr in girls only (6–7 yr – B: −10.8%, p = 0.12; G: −19.2%, p = 0.001). HDL-C rose (B: +17.8%, G: +17.1%, both p < 0.001) and TG fell (B: −4.8%, p = 0.16; G: −11.6%, p = 0.006). (ii) Correlations between insulin resistance (IR) and anthropometry were poor at 5 yr but strengthened by 8 yr (BMI – B: r = 0.20/0.38, G: r = 0.28/0.49; WC – B: r = 0.25/0.40, G: r = 0.32/0.58; SSF – B: r = 0.11/0.36, G: r = 0.18/0.53). (iii) In girls, but not boys, adiposity at 5 yr predicted IR better at 8 yr (BMI – r2= 0.17; WC – r2= 0.28; SSF – r2= 0.17, all p < 0.001) than it did at 5 yr (BMI – r2= 0.08, p < 0.01; WC – r2= 0.10, p < 0.01; SSF – r2= 0.03, p = 0.07).

Conclusions

Cross-sectional association cannot indicate direction of trend or predict the future. Predicting metabolic health from anthropometric measures in prepubertal children requires longitudinal data, tracking variables from childhood into adulthood. Until the data set reaches adulthood, it is probably not safe to make recommendations on which children to ‘target’ or whether early intervention would be of benefit.

Related Topics

    loading  Loading Related Articles