A meta-analysis of ≥5-year mortality in randomized controlled trials of off-pump versus on-pump coronary artery bypass grafting

Abstract

We sought to determine whether off-pump coronary artery bypass grafting (CABG) increases long-term mortality, by performing a meta-analysis of randomized controlled trials (RCTs) of off-pump versus on-pump CABG with ≥5-year follow-up. MEDLINE and EMBASE, and the Cochrane Central Register of Controlled Trials were searched through July 2018. Hazard, risk, or odds ratios (HRs, RRs, or ORs) of long-term (≥5-year) mortality for off-pump versus on-pump CABG were extracted from each individual trial. Study-specific estimates were combined using inverse variance-weighted averages of logarithmic HRs in the random-effects model. Our search identified eight medium- to large-size RCTs at low risk of bias with long-term follow-up of off-pump versus on-pump CABG enrolling a total of 8780 patients. Combining four RCTs reporting actual HRs generated a statistically significant result favoring on-pump CABG (HR, 1.21; P = 0.02). A pooled analysis of all eight RCTs demonstrated a statistically significant increase in mortality with off-pump CABG (HR/RR, 1.19; P = 0.01). There was no evidence of significant publication bias in the meta-analysis of all eight RCTs. In a sensitivity analysis, extracting RRs or ORs from all eight RCTs and pooling them demonstrated a statistically significant increase in mortality with off-pump CABG (RR, 1.17; P = 0.01; OR, 1.20; P = 0.007). Eliminating 2 RRs and combining six HRs still generated a statistically significant result favoring on-pump CABG (HR, 1.19; P = 0.05). Off-pump CABG increases long-term (≥5-year) mortality compared with on-pump CABG.