Abstract P179: Abdominal Skin-fold Thickness Improves Anthropometric Prediction of Insulin Resistance in Prepubescent Colombian Children
Introduction: Body mass index (BMI), waist circumference (WC), and waist-to-height ratio (WHtR), have been considered poor measures of cardiometabolic risk in children, in whom increases in these measures may reflect increases in lean mass more so than fat mass. Hypothesis: We examined the hypothesis that abdominal skinfolds (ASF) have better predictive value than BMI and other anthropometric measures for identifying insulin resistance (IR) in prepubescent Colombian children.
Methods: We used data from a population-based cross-sectional study of 1,262 children, aged 6-10 y, in Bucaramanga, Colombia. Logistic regression stepwise variable selection (P<0.05 for entry and retention) was performed to identify anthropometric predictors of IR, as determined by homeostatic model assessment (HOMA). Receiver operating characteristic (ROC) curves were used to compare area under the curve (AUC).
Results: There were 57 boys and 70 girls classified as IR (HOMA >90th percentile). Among anthropometric measures compared - including WC, WHtR, and four skinfolds - only age- and sex-specific ASF and BMI z-scores were retained as independent anthropometric predictors of IR. Prediction of IR was marginally better using ASF z-scores than BMI z-scores (P for contrast= 0.13). However, the final Model, which included both ASF and BMI, significantly increased the AUC from 0.80 (95% CI: 0.75-0.84) to 0.82 (95% CI: 0.78-0.86) - an improvement of 0.02 (95% CI: 0.004-0.04; P for contrast= 0.01)(see table). In the final Model, after adjusting for BMI z, a 1-SD increment in ASF z was associated with 2.38 (95% CI: 1.74-3.26) greater odds of IR; whereas a 1-SD increment in BMI z, after adjustment for ASF z, was associated with 1.46 (95% CI: 1.11-1.91) greater odds of IR.
Conclusions: ASF was independent from and marginally better than BMI in predicting IR in Colombian children. Furthermore, integration of ASF with BMI improved IR risk stratification compared to BMI alone, opening new perspectives in the prediction of cardiometabolic risk in children.
|ROC Curve Contrasts||Difference in AUC||95% CI||P for Contrast|
|WCz - BMIz||0.006||(-0.016, 0.028)||0.622|
|WHtRz - BMIz||−0.045||(-0.074, -0.016)||0.002|
|ASFz - BMIz||0.018||(-0.005, 0.040)||0.126|
|Model (ASFz + BMIz) - BMIz||0.021||(0.004, 0.038)||0.014|
BMI z-score (BMIz) was used as reference in all contrast estimates and tests against waist circumference z-score (WCz), waist-to-height z-score (WHtRz), abdominal skinfolds z-score (ASFz), and the best stepwise-selection Model (includes ASFz + BMIz)
- © 2012 by American Heart Association, Inc.