Evaluation of the predictive accuracy of QUICKI and McAuley indices for insulin resistance in adolescents: Insights from a cross-sectional study

Abstract

Different indirect methods have been developed to assess insulin resistance (IR), though their validation has been limited to adult populations. In this sense, the study aim is to compare the predictive capacity of the McAuley, QUICKI, SPISE indices, and glucose-insulin ratio against insulin resistance (IR) in Spanish adolescents and to establish reliable cut-off values for these indices in this population. A cross-sectional study was conducted with 981 adolescents aged 11–16 years, from Southern Spain. Anthropometric measurements and fasting biochemical parameters, were assessed. IR indices, such as HOMA-IR, QUICKI, the McAuley index, SPISE, and the glucose-insulin ratio, were calculated. The ability of each index to predict IR was evaluated using multivariate regression analysis and receiver operating characteristic (ROC) curves. Boys exhibited higher waist circumference, triglyceride levels, and fasting insulin levels, while girls had a higher percentage of body fat (p < 0.05). HOMA-IR values increased in adolescents with obesity, whereas McAuley and QUICKI indices decreased significantly (p < 0.05). Similar results were found in adolescents with MetS in comparison to non-MetS adolescents. QUICKI showed the highest predictive accuracy (AUC: 1.00), followed by the McAuley index (AUC: 0.983 in boys, 0.977 in girls), and SPISE (AUC: 0.896 in boys, 0.852 in girls). The cut-off points for these indices were 5.794,0.316, and 7.8 respectively, both for boys and girls. The glucose-insulin ratio demonstrated poor predictive ability. The QUICKI, McAuley, and SPISE indices are effective in predicting IR among Spanish adolescents. Their use could aid in early detection of metabolic and cardiovascular disfunctions, though further validation in larger cohorts is required for clinical application.