Long-Term Clinical Outcomes of PCI Versus Medical Therapy in NSTEMI Patients with Prior CABG

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Background and Objectives: Patients with a prior history of coronary artery bypass grafting (CABG) who present with non-ST-segment elevation myocardial infarction (NSTEMI) represent a complex, high-risk subgroup due to advanced comorbidity burden and challenging coronary anatomy. Whether an invasive strategy offers meaningful benefit over conservative management in this population remains unclear. Therefore, this study aimed to compare long-term outcomes of percutaneous coronary intervention (PCI) versus medical therapy in NSTEMI patients with previous CABG and to identify independent predictors of major adverse cardiovascular events (MACE) and all-cause mortality. Materials and Methods: This retrospective cohort study included 286 NSTEMI patients with prior CABG (PCI: 112; medical therapy: 174). Baseline demographic, clinical, laboratory, and angiographic characteristics were assessed. The primary endpoint was MACE, while the secondary endpoint was all-cause mortality. Kaplan–Meier analysis evaluated survival differences, and multivariable Cox regression identified independent predictors. Results: During follow-up, MACE rates were comparable between PCI and medical therapy (14.3% vs. 18.9%; p = 0.305). All-cause mortality was likewise similar (9.8% vs. 10.3%; p = 0.541). Kaplan–Meier analysis showed no survival benefit with PCI (log-rank p = 0.334). Hypoalbuminemia independently predicted both MACE and mortality, while CKD and HF were major determinants of long-term mortality. Conclusions: In NSTEMI patients with prior CABG, no long-term superiority of PCI over medical therapy was observed with respect to MACE or mortality. Prognosis appears more closely linked to hypoalbuminemia, CKD, and HF than to the chosen management strategy. These findings underscore the importance of individualized and risk-adapted clinical decision-making in this complex population.

This publication has 34 references indexed in Scilit: