Clinical Determinants and Predictive Modeling of Fibrinolytic Success in Non-Diabetic STEMI Patients Using Logistic Regression and CART Analysis

Abstract

Introduction: Fibrinolytic therapy remains an essential reperfusion strategy for ST-segment elevation myocardial infarction (STEMI) in settings where primary percutaneous coronary intervention is not readily available. However, the determinants of successful fibrinolysis and the role of predictive modeling in non-diabetic populations remain incompletely understood.

Objectives: This study aimed to identify clinical determinants and develop a predictive model of fibrinolytic success in non-diabetic STEMI patients using logistic regression and classification and regression tree (CART) analysis.

Methods: A retrospective cohort study was conducted among non-diabetic STEMI patients who received fibrinolytic therapy at a tertiary referral hospital between 2018 and 2025. The primary outcome was successful fibrinolysis, defined as ≥50% ST-segment resolution within 60–90 minutes with clinical improvement. Bivariate and multivariable logistic regression analyses were performed to identify independent predictors. CART analysis was used to explore hierarchical relationships and predictive performance.

Results: A total of 182 patients were included, with a fibrinolytic success rate of 73.6%. Early symptom-to-treatment time (<6 hours) was independently associated with higher success (adjusted OR 3.98; 95% CI 1.45–10.92; p=0.007). Clopidogrel loading therapy was also a significant predictor (adjusted OR 3.21; 95% CI 1.12–9.18; p=0.029). CART analysis identified symptom-to-treatment time as the primary determinant, followed by antiplatelet therapy, with moderate predictive performance.

Conclusions: Fibrinolytic success in non-diabetic STEMI patients is primarily determined by timely treatment and adjunctive antiplatelet therapy. Predictive modeling using CART provides additional insights into variable interactions and may support clinical decision-making in resource-limited settings.