Rendimiento diagnóstico de once indicadores para resistencia a la insulina en una muestra de pobladores peruanos

Abstract

Introduction: Insulin resistance (IR) is one of the main causes of chronic disease. Early detection is essential, which is why it is important to study more affordable and less expensive methods, such as biomarkers. Objective: To determine the diagnostic accuracy of 11 biomarkers of IR in a sample of Peruvian residents. Method: diagnostic tests. Secondary Database Analysis of the PERU-MIGRANT Study. To measure RI, a homeostatic model evaluation (HOMA-IR) ≥ 2.8 was used as a reference. Biomarkers were based on the lipid ratio, visceral lipid indicators, indicators of triglycerides and glucose (TyG), and indicators of abdominal waist. For precision, the receiver operating characteristic curve and area under the curve (AUC) with their respective 95% confidence intervals (95%CI) were used. Results: A total of 938 participants were studied. The prevalence of IR was 9.91%. In relation to the ROC analysis, the TyG index – body mass index (TyG – BMI) had the highest AUC, both in men: AUC=0.85 (0.81-0.90), cut-off=241.55; sens=92.5 (79.6-98.4) and sp=78.3 (73.9-82.2); as in women: AUC=0.81 (0.76-0.85), cut-off=258.77; sens=79.2 (70.3-86.5) and esp= 82.1 (78.0-85.8). Discussion: According to the data analyzed, the TyG-IMC index is the best indicator for measuring IR. It is a simple index that can be routinely used in clinical practice. Future prospective studies are needed to confirm its predictive capacity.