Effect of bilateral internal mammary artery grafts on long-term survival: a meta-analysis approach.
Selected in Circulation by G.G. Stefanini
G. Yi, B. Shine, SM Rehman, DG Altman, DP Taggart
Circulation. 2014 Aug 12;130(7):539-45
LinkAccess the Abstract
What is known
Several studies have indicated a potential survival benefit of bilateral internal mammary artery (BIMA) grafting compared with single internal mammary artery (SIMA) grafting. This potential benefit is likely due to the >90% patency rates of BIMA grafts during long term.
Aim of the study
To assess whether the use of BIMA might provide long-term survival benefits into the second decade of follow-up after coronary artery bypass grafting.
The authors performed a systematic review and meta-analysis of studies comparing BIMA with SIMA including at least 100 patients in each group with a mean follow-up of at least 9 years.
- 9 studies were identified, none of which was a randomized controlled trial.
- 15,583 patients were included in these 9 observational studies, treated with BIMA (N=7313) or SIMA (N=8270).
- Mean follow-up ranged from 9.8 to 20 years.
- Random-effect meta-analysis show that the use of BIMA significantly reduces the risk of mortality as compared to SIMA (HR 0.79, 95% CI 0.75-0.84) at longest available follow-up, with no evidence of significant heterogeneity across trials (I-squared=0%).
- In a sensitivity analysis results were stratified according to study design showing a consistent mortality risk reduction favoring BIMA in unmatched studies (HR 0.81, 95% CI 0.69-0.94), quintile-based propensity score analysis (HR 0.81, 95% CI 0.75-0.87), and propensity score-based exact patient matching (HR 0.75, 95% CI 0.65-0.85).
According to this meta-analysis of 9 observational studies including 15,583 patients, BIMA is associated with a hazard ratio of 0.81 as compared to SIMA with a narrow 95% confidence interval (0.75-0.84). These findings indicate that the potential survival benefit with BIMA over SIMA is maintained during very long-term follow-up (i.e., >9 years). Of note, BIMA grafting should be used in patients <70 years of age according to the most recent joint guidelines on myocardial revascularization of the European Society of Cardiology and European Association for Cardio-Thoracic Surgery (class of recommendation IIa, level of evidence B). Notwithstanding, the use of BIMA is limited to around 10% of patients undergoing coronary artery bypass grafting in Europe. The contrast between available evidence, guidelines recommendations, and current clinical practice is partly explainable by the risk of wound complications, the longer operation time, and the increased technical challenge. Moreover, it must be acknowledged that the observational nature of the studies included in the meta-analysis represents an important limitation. Meta-analyses of observational studies may lead to very precise but equally spurious results due to confounders and selection bias. In this view, the primary endpoint findings of the ongoing Arterial Revascularization Trial (ART) randomized trial will provide definitive conclusions on the potential survival advantage of BIMA during long-term follow-up to 10 years. However, the primary endpoint of the ART trial will not be reported until 2018. At this point in time – in the absence of randomized trials with long-term follow-up – this meta-analysis of 9 observational studies represents the strongest available evidence supporting the use of BIMA in patients undergoing coronary artery bypass grafting.