Abstract
OBJECTIVE: Insulin resistance (IR) is a well-established major risk factor for type 2 diabetes mellitus, nonalcoholic fatty liver disease, and atherosclerotic cardiovascular disease. Previous studies have shown an association between increased serum albumin (ALB) levels and the risk of IR. However, there is a lack of studies simultaneously evaluating the association of total protein (TP), ALB, and globulin (GLB) with IR. METHODS: A total of 14,828 individuals (average age 49 ± 18 years) with complete data from the National Health and Nutrition Examination Survey (NHANES) were enrolled and divided into two groups (non-IR group, n = 8,653 and IR group, n = 6,175). Spearman's correlation analysis, multivariable logistic regression models, restricted cubic spline curves, and subgroup analysis were performed to explore those associations. RESULTS: After adjustment for potential confounders, multivariable logistic regression analysis revealed that scaled per 10g/L increment, the fully adjusted odds ratios (ORs) (95% confidence interval (CI)) for IR prevalence were 1.54 (95% CI 1.41-1.69, P < 0.0001), 1.09 (95% CI 0.95-1.25), P = 0.1995), and 1.62 (95% CI 1.47-1.79, P < 0.0001) for TP, ALB, and GLB respectively. Compared to those in the lowest quantiles, the prevalence of IR in subjects in the highest TP and GLB quantiles was 2.06 and 1.91 times, respectively. Furthermore, restrictive cubic curves confirmed that the relationship of TP, ALB, and GLB with IR prevalence was a linear relationship. CONCLUSIONS: The present cross-sectional study, for the first time, provided supportive evidence of positive associations of TP and GLB with IR, but not ALB, and demonstrated that TP and GLB might be useful markers for IR prevalence.