graft patency of off-pump and conventional coronary surgery A systematic review of randomized trials comparing revascularization rate and

Objective: Although many trials have been conducted to evaluate the benefits of off-pump coronary surgery, few have concentrated on graft patency. We sought to evaluate the impact of off-pump surgery on completeness of revascularization and graft patency compared with conventional surgery. Methods: A systematic literature search was undertaken of all randomized trials of off-pump coronary surgery in MEDLINE, EMBASE, the Cochrane Library Controlled Trials Register, the National Research Register, and abstracts from major conferences. Results: In total, 132 publications were identified. From this number, we excluded 43 without a conventional surgery arm and 80 that did not evaluate graft patency. One trial was excluded for selective angiography and one abstract was excluded because of insufficient information. A total of 7 trials were eligible for overview. On initial analysis, the relative risk of graft patency in off-pump coronary surgery compared with conventional surgery was 0.959 (95% confidence interval 0.936-0.983; P = .001). The analysis was repeated after excluding one specific trial because of clinical and statistical heterogeneity (x(6)(2) = 27.78; P < .001), and a relative risk of 0.953 (95% confidence interval 0.927-0.980; P = .001) was obtained with no further evidence of heterogeneity (x(5)(2) = 5.35; P = .374). In 5 trials that included the mean number of grafts performed per arm, the standardized mean difference in revascularization comparing off-pump with conventional surgery was -0.164 (-0.286 to -0.043; P = .008). Conclusion: In a meta-analysis of randomized trials, patients undergoing off-pump coronary surgery had a lower rate of revascularization and lower graft patency than did patients undergoing conventional coronary surgery.