**Background:** Insulin resistance (IR) is a critical risk factor for cardiovascular disease. The triglyceride glucose-body mass index (TyG-BMI) has been proposed as a reliable surrogate measure of IR, but its ability to predict cardiovascular outcomes in patients with coronary artery disease (CAD) undergoing percutaneous coronary intervention (PCI) remained uncertain. This study aimed to evaluate the association between the TyG-BMI index and major adverse cardiac and cerebrovascular events (MACCEs) in patients who underwent PCI with drug-eluting stent (DES) implantation.
**Methods:** The study analyzed data from 1438 consecutive patients who underwent PCI with DES implantation between July 2009 and August 2011, drawn from a larger cohort of 2533 participants after excluding those with missing data. The TyG-BMI index was calculated as ln[fasting triglyceride (mg/dL) × fasting blood glucose (mg/dL)/2] × BMI. Patients were divided into tertiles based on TyG-BMI levels: T1 (<195.26, n=479), T2 (195.26–226.26, n=479), and T3 (≥226.26, n=480). The primary endpoint was MACCEs, a composite of acute myocardial infarction, repeat revascularization, stroke, and all-cause mortality at a median follow-up of 29.8 months (quartiles 25.6–34 months). Statistical analyses included univariate and multivariate logistic regression, subgroup analyses, restricted cubic spline modeling, and receiver operating characteristic (ROC) curve analysis with area under the curve (AUC) comparisons using DeLong's test, as well as net reclassification improvement (NRI) and integrated discrimination improvement (IDI).
**Key Results:** Among the 1438 participants (mean age 60.1±11.1 years, male/female ratio 2.18, mean TyG-BMI 213.7), 195 patients (13.6%) experienced at least one MACCE during follow-up. The incidence of MACCEs showed no statistically significant difference across TyG-BMI tertiles in the overall population (T1: 12.5%, T2: 13.4%, T3: 14.8%; p=0.584). In the overall population, multivariate logistic regression showed no significant association between TyG-BMI (per 1 SD increase) and MACCEs (OR=1.11, 95% CI 0.91–1.37, p=0.304). However, subgroup analysis revealed a significant interaction by age (p for interaction=0.019). In elderly patients (≥60 years, n=791, 137 MACCEs), a higher TyG-BMI index (per 1 SD) was significantly associated with MACCEs in both univariate (OR=1.21, 95% CI 1.02–1.44, p=0.034) and multivariate analyses (OR=1.22, 95% CI 1.011–1.467, p=0.038), with a linear relationship confirmed by restricted cubic splines (p for non-linearity=0.407). In female patients (n=452, 64 MACCEs), the TyG-BMI index (per 1 SD) was significantly associated with MACCEs in univariate (OR=1.39, 95% CI 1.08–1.78, p=0.010) and multivariate analyses (OR=1.33, 95% CI 1.004–1.764, p=0.047), also showing a linear relationship (p for non-linearity=0.938). No significant association was found in male patients or non-elderly patients. When added to traditional risk factor models, the TyG-BMI index did not significantly improve risk prediction in elderly patients (baseline model AUC 0.687 vs. model + TyG-BMI AUC 0.694, p=0.313; NRI 0.221, p=0.020; IDI 0.006, p=0.059) or in female patients (baseline model AUC 0.759 vs. model + TyG-BMI AUC 0.765, p=0.508; NRI 0.222, p=0.106; IDI 0.012, p=0.102).
**Clinical Implications:** The TyG-BMI index may serve as an independent prognostic factor for MACCEs specifically in elderly and female patients after DES implantation, with higher values proportionally related to increased risk. However, its limited incremental predictive value over traditional risk factors suggests it may not substantially improve clinical risk stratification in these populations. The findings highlight potential sex- and age-specific differences in the relationship between insulin resistance and cardiovascular outcomes. The study's limitations include its single-center retrospective design, exclusion of many patients due to missing data, low event rates for individual outcomes, lack of time-to-event analysis, and inability to compare TyG-BMI with other IR measurement methods. Further large-scale, multicenter prospective studies are needed to validate these findings.