The impact of coronary artery bypass grafting added to aortic valve replacement on long-term outcomes in octogenarian patients. A reconstructed time-to-event meta-analysis

Interact Cardiovasc Thorac Surg. 2022 Jun 20;ivac164. doi: 10.1093/icvts/ivac164. Online ahead of print.


Objectives: The long-term results in studies comparing octogenarian patients who received either isolated surgical aortic valve replacement (i-SAVR) or coronary artery bypass grafting (CABG) in addition to SAVR are still debated. We performed a reconstructed time-to-event data meta-analysis of studies comparing i-SAVR and CABG+SAVR to evaluate the impact of CABG and to analyse the time-varying effects on long-term outcome.

Methods: We performed a systematic review of the literature from January 2000 through November 2021, including studies comparing i-SAVR and CABG+SAVR which reported at least 3-years follow-up and that plotted Kaplan-Meier curves of overall survival. The primary endpoint was overall long-term survival; secondary endpoints were in-hospital/30-day mortality and postoperative outcomes. The pooled hazard ratio (HR) and odd ratio (OR) with 95% confidence interval (CI) were calculated for primary and secondary endpoints, respectively. Random-effect model was used in all analyses.

Results: Sixteen retrospective studies were included (5382 patients, i-SAVR=2,568 and CABG+SAVR=2,814). I-SAVR showed a lower incidence of in-hospital mortality compared to CABG+SAVR (OR = 0.73; 95% CI = 0.60-0.89; p = 0.002). Landmark analyses showed a significantly higher all-cause mortality within 1-year from surgery in CABG+SAVR (HR = 1.17; 95% CI = 1.01-1.36; p = 0.03); after 1-year, no significant difference was observed (HR = 0.95; 95% CI = 0.87-1.04; p = 0.35). Landmark analysis was confirmed by time-varying-trend of hazard ratio.

Conclusions: Late survival of octogenarians did not differ significantly between the two interventions. Interestingly, CABG added to SAVR was associated with both higher in-hospital and within 1-year mortality after surgery, whereas this difference was statistically non-significant at long-term follow-up.