Abstract
This cross-sectional study analyzed National Health and Nutrition Examination Survey data from 2011 to 2018, focusing on individuals aged ≥20 years. The association between metabolic score for insulin resistance (METS-IR) and sarcopenia was examined using weighted multivariable logistic regression, with dose-response relationships characterized by restricted cubic spline analysis. Subgroup and sensitivity analyses were performed, and receiver operating characteristic curve analysis assessed METS-IR's ability to detect sarcopenia, with the area under the curve used for evaluation. The study included 4553 participants (mean age, 40 years; 49.4% male and 50.6% female). In the descriptive analysis, METS-IR levels in sarcopenia (mean, 52.39) were significantly higher than METS-IR levels in nonsarcopenia (mean, 41.94), indicating an association with sarcopenia. A univariate logistic regression analysis showed that sarcopenia and METS-IR were positively correlated. Even after accounting for all variables, METS-IR maintained a stable positive correlation with the prevalence of sarcopenia (odds ratio, 1.06 [95% CI, 1.06-1.08]). The results remained stable when METS-IR was categorized into quartiles. METS-IR was found to positively correlate with sarcopenia prevalence using restricted cubic spline analysis. According to subgroup analysis, there is a consistent and stable positive correlation between the prevalence of sarcopenia and METS-IR. Sensitivity analysis showed that METS-IR and sarcopenia continued to have a significant positive connection even after excluding extreme findings. The area under the curve value of METS-IR in the receiver operating characteristic curve analysis was 0.7217, suggesting that METS-IR could be a useful predictor of sarcopenia.