Annual case volume on mortality after coronary artery bypass grafting: a dose–response meta-analysis

OBJECTIVES: This study evaluated the effect of both hospital and surgeon annual case volumes on patient mortality following coronary artery bypass grafting (CABG). METHODS: PubMed and Embase databases were searched for clinical studies on CABG. The outcome was mortality, including operative mortality, in-hospital mortality and 30-day mortality. RESULTS: Twenty-five studies involving 3 492 101 participants and 143 951 deaths were included for hospital volume, and 4 studies involving 108 356 participants and 2811 deaths were included for surgeon volume. The pooled estimate revealed that both hospital and surgeon annual case volumes were inversely associated with mortality in patients after CABG [odds ratio (OR) for hospital: 0.62, 95% confidence interval (CI) 0.56-0.69; P < 0.001; OR for surgeon: 0.51, 95% CI 0.31- 0.83; P < 0.001] with high heterogeneity (hospital: I2 = 90.6%, Pheterogeneity < 0.001; surgeon: I2 = 86.8%, Pheterogeneity < 0.001). The relationship remained consistent and robust in most subgroup and sensitivity analyses. Our meta-regression analysis of time suggested that the strength of the negative associations between volume and mortality for both hospitals and surgeons remained unattenuated over time even though the CABG mortality gradually decreased over time. The dose-response analysis suggested a non-linear relationship between both hospital and surgeon annual case volumes and mortality (both Pnon-linearity = 0.001). CONCLUSIONS: Both higher hospital and surgeon annual case volumes are associated with lower mortality in patients undergoing CABG, and the negative associations remain unattenuated over time. CLINICAL REGISTRATION NUMBER: The study was registered at PROSPERO as CRD42017067912.