The prognostic significance of metabolic score for insulin resistance in evaluating all-cause and cardiovascular mortality risk among individuals across stages 1-4 of cardiovascular-kidney-metabolic syndrome: a nationwide prospective cohort study.

BACKGROUND AND AIM: Cardiovascular-kidney-metabolic syndrome (CKM) embodies the intricate interaction among metabolic, kidney, and cardiovascular dysfunctions. Insulin resistance (IR) is frequently observed in patients with CKM. The metabolic score for insulin resistance (METS-IR) offers a more accessible approach to assessing IR. The relationship between the METS-IR and mortality is unclear in CKM stage 1-4 patients. METHODS AND RESULTS: Using data from the NHANES, we included 18,295 participants with CKM stages 1-4 from 1999 to 2018. The association between METS-IR and all-cause mortality as well as cardiovascular disease (CVD)-cause mortality was evaluated using multivariable Cox regression, restricted cubic spline models, two-stage Cox models, and subgroup analyses. Four machine learning algorithms were employed to evaluate the predictive value of METS-IR for all-cause mortality. SHAP values were used to model interpretation. During a median follow-up of 100 months, a total of 901 all-cause and 267 CVD-cause deaths were recorded. After adjusting for confounders, each 1-unit rise in METS-IR corresponds to a HR of 1.560 for all-cause mortality (95 % CI, 1.120-2.190) and a HR of 2.309 for CVD mortality (95 % CI, 1.314-4.060) among CKM stage 1-4. RCS curve and two-stage Cox models revealed a nonlinear positive correlation between METS-IR and all-cause as well as CVD mortality, with thresholds of 9.836 for both. Subgroup analyses suggest a significant interaction with alcohol consumption (P for interaction = 0.006). Penalty COX showed the best performance in AUC of 0.849. METS-IR showed high SHAP values. CONCLUSIONS: METS-IR≥9.836 is positive association with all-cause mortality and CVD mortality in CKM stage 1-4 patients. Incorporating it into prognostic models may facilitate the early identification of high-risk individuals.