Effect of bilateral internal mammary artery grafts on long-term survival: a meta-analysis approach.

Abstract

Although the potential survival benefit of bilateral internal mammary artery (BIMA) grafting in comparison with single internal mammary artery (SIMA) grafting has been emphasized by many investigators, the use of BIMA is still low in clinical practice in the absence of randomized trials and long-term results. In the current study, we aimed to assess if there is a long-term survival benefit of BIMA up to 10 years after coronary bypass surgery. We selected published articles comparing survival between SIMA and BIMA patients with follow-up duration of more than a mean of 9 years. We evaluated the log hazard ratio with 95% confidence interval for included studies by using a random-effects meta-analysis. Nine eligible observational studies provided 15 583 patients (8270 SIMA and 7313 BIMA) for meta-analysis. Five studies used propensity score methods for statistical adjustment, 2 with a propensity score-based patient-matching method and 3 with quintile-based stratification. A significant reduction in mortality by using BIMA was observed (hazard ratio, 0.79; 95% confidence interval, 0.75-0.84); no study showed any significantly harmful effect of BIMA on survival. Subgroups of studies using different statistical approaches-unmatched, quintile-based propensity score analysis, and propensity score-based exact patient matching-all showed the survival benefit of BIMA grafting. BIMA grafting appears to have better survival with up to 10 years follow-up in comparison with SIMA grafting. Long-term survival benefit of BIMA seems to continue in the second decade after surgery. An ongoing randomized trial comparing SIMA and BIMA groups will add evidence on this issue.