[Predictive ability of anthropometric and body composition indicators in the identification of insulin resistance]

To assess the ability of anthropometric and body composition indicators in identifying insulin resistance (IR), determining cut-off points for those showing the best efficacy.

138 men were evaluated. Waist perimeter (WP), sagittal abdominal diameter (SAD), conicity index, body mass index (BMI), body fat percent, sagittal index, and the waist-to-height, waist-to-hip and waist-to-thigh ratios were determined. IR was assessed by the HOMA-IR index. Statistical analysis consisted of Spearman correlation coefficient and ROC (receiver operating characteristic) curves, calculating the area under the curve (AUC).

SAD (r=0.482, AUC=0.746) and WP (r=0.464, AUC=0.739) showed stronger correlations with the HOMA-IR and greater ability to identify IR (p<0.001), being 89.3 cm and 20.0 cm the best cut-offs, respectively.

The anthropometric indicators of central obesity, WP and SAD, have shown greater ability to identify IR in men. We encourage studies in women and elderly people in search of the best cut-off points for the entire population.