263 * Influence of Previous Percutaneous Coronary Intervention on Clinical Outcome of Coronary Artery Bypass Grafting: A Meta-Analysis of Comparative Studies

Objectives: The prognostic significance of previous percutaneous coronary intervention (PCI) in patients undergoing coronary artery bypass grafting (CABG) is still unclear. We conducted this first meta-analysis to assess whether previous PCI increases postoperative mortality in CABG. Methods: MEDLINE and Embase were searched through April 2014. Studies satisfying the following criteria were included in this meta-analysis: (1) comparing CABG patients with previous PCI versus without previous PCI and (2) reporting hospital mortality. Results: Our search identified 19 comparative studies including 159,337 patients: 141,048 without previous PCI, 18,289 with previous PCI. Pooled analysis demonstrated that previous PCI had an adverse effect on hospital mortality: odds ratio (OR) 1.48, 95% confidence interval (CI) 1.14–1.92 (Fig. 1). Further, the subgroup analysis divided by the proportion of multiple previous PCI (= number of patients with multiple previous PCI/number of patients with single or multiple previous PCI) was performed. In the subgroup of studies with the proportion <40%, the adverse effect was not significant: OR 0.89 (95% CI 0.56–1.40). However, in the subgroup of studies with the proportion ≥40%, the adverse effect of previous PCI was significant: OR 2.26 (95% CI 1.60–3.20). A meta-regression coefficient was significantly positive for the proportion of patients with multiple PCI history (coefficient 0.833; 95% CI 0.023–1.640; P = 0.044). Conclusion: This meta-analysis would suggest that as the proportion of patients with multiple previous PCI in CABG cohort increases, postoperative mortality also increases. This result re-emphasises the importance of the heart team approach to coronary revascularisation.