Predictors of Left Main and Three-Vessel Coronary Artery Disease in Patients Undergoing Diagnostic Coronary Angiography: A Cross-Sectional Study from a Single Centre in Libya
Abstract
Left main and/or three-vessel coronary disease (LM/3VD) defines high-risk anatomy that shapes revascularisation decisions, yet its predictors in North African populations remain poorly characterised. This study examined the prevalence, pattern, clinical predictors and revascularisation strategies of LM/3VD in patients undergoing diagnostic coronary angiography at a Libyan centre. In this cross-sectional analytical study, 101 consecutive patients undergoing diagnostic coronary angiography at Albyida Medical Centre over one year were enrolled. Demographics, cardiovascular risk factors, clinical presentation, angiographic findings and management were recorded. The primary outcome was LM/3VD. Associations were tested using chi-square, Fisher's exact, Monte Carlo exact, Mann-Whitney U, Kruskal-Wallis and Spearman tests, and binary logistic regression (parsimonious primary model; sensitivity models). Mean age was 56.4 years (73.3% male); 53.5% had diabetes. LM/3VD affected 22/101 patients (21.8%; exact 95% CI 14.2-31.1). STEMI presentation was the strongest independent predictor (adjusted OR 4.16, 95% CI 1.42-12.21, p = 0.009); diabetes showed a borderline independent association (OR 2.57, 95% CI 0.86-7.69, p = 0.091). LM/3VD rose stepwise from 0% in unstable angina to 35.6% in STEMI (chi-square = 11.52, p = 0.009). All 22 patients with three-vessel disease were referred for coronary artery bypass grafting (100%). Among 54 diabetic patients, insulin therapy was strongly associated with LM/3VD (45.5% vs 4.8%; OR 16.67, 95% CI 2.00-139.14, p = 0.001). STEMI presentation and insulin-treated diabetes identified patients at highest risk of left main and three-vessel disease. These findings support targeted angiographic risk stratification of acute coronary syndrome patients and suggest insulin-treated diabetes marks a high-risk vascular phenotype.