Association between single-point insulin sensitivity estimator and three-vessel coronary disease: a cross-sectional...

The single-point insulin sensitivity estimator (SPISE) is a novel surrogate marker for assessing insulin resistance in humans. Three-vessel coronary disease (TVD) remains a major challenge in coronary interventional therapy. However, the association between the SPISE index and TVD remains unclear. Therefore, this study aimed to evaluate the relationship between SPISE and TVD in an elderly population.

Generalized additive models (GAM) and random forest variable importance scores were applied for feature selection to identify key predictors for constructing a TVD risk model. Multivariable regression and restricted cubic spline (RCS) analyses were performed to investigate the association between SPISE and TVD. Subgroup analyses were further conducted to assess potential effect modification.

GAM and random forest variable importance scores identified age and SPISE as the most important predictors. After adjustment for all covariates, SPISE was inversely associated with TVD (OR = 0.75, 95% CI: 0.68-0.83, P < 0.001). RCS and segmented regression analyses revealed a significant nonlinear relationship with a distinct inflection point. Receiver operating characteristic (ROC) analysis demonstrated that SPISE showed superior performance compared with the TyG index in predicting TVD risk (AUC: 0.718 vs. 0.701). Subgroup analyses confirmed a consistent inverse association between SPISE and TVD, with significant interactions observed across strata of sex, alcohol consumption, age, and BMI.

A higher SPISE index is independently and inversely associated with TVD and may serve as a reliable predictor of disease risk, highlighting its potential utility in cardiovascular risk assessment.