Pooling individual off-pump and on-pump coronary artery bypass patient records from three contemporary multicenter, randomized, controlled trials: an individual participant data meta-analysis protocol

Abstract Background For over three decades, the comparative effectiveness of off-pump versus on-pump coronary artery bypass grafting (CABG) has been debated. Although randomized controlled trials (RCTs) generally report modest long-term advantages for on-pump CABG, potential off-pump benefits remain uncertain particularly among high-risk patients. To address this knowledge gap, this CABG individual participant data meta-analysis (CABG-IPD-MA) will analyze de-identified patient records from the only three multicenter RCTs reporting 5-year follow-up (ROOBY, GOPCABE, and CORONARY) and test five protocol-driven hypotheses. Methods Across short-term, intermediate, and longer-term periods, this PRISMA-IPD-compliant study will use a one-stage IPD meta-analysis framework to compare off-pump versus on-pump CABG for two co-primary endpoints [i.e., all-cause mortality and a composite metric comprised all-cause death, nonfatal acute myocardial infarction, and repeat revascularization (i.e., repeat CABG or percutaneous coronary intervention)] at 1 and 5 years; at 30 days/in-hospital, perioperative stroke will also be included in the composite. Secondary endpoints include individual sub-components of the composite endpoint and resource-utilization; resource-utilization analyses will be exploratory-only. Eligible studies included all post-2001 multicenter RCTs reporting off-pump versus on-pump post-CABG 5-year outcomes. Re-verified June 2025, the original 2019 PRISMA-IPD-guided systematic review identified only three eligible RCTs; prior publications reported these RCTs as having low Cochrane risk-of-bias. All three RCTs’ participants ( n = 9494) will be included. This CABG-IPD-MA study’s hierarchical hypotheses will evaluate: (1) overall unadjusted treatment effects; (2) overall risk-adjusted treatment effects; (3) prespecified high-risk subgroups’ risk-adjusted treatment effects; (4) the impact of complete revascularization; and (5) the concordance between this CABG-IPD-MA study’s findings with an aggregate-data meta-analysis findings. All primary CABG-IPD-MA analyses will utilize an intention-to-treat framework. Prespecified sensitivity analyses will evaluate findings under alternative treatment-conversion and high-risk restrictions. Discussion By pooling and harmonizing individual patient-level records from the only three contemporary multicenter RCTs reporting long-term follow-up, this study was designed to provide precise, clinically relevant assessments of off-pump versus on-pump benefits. These CABG-IPD-MA analyses will evaluate short-, intermediate-, and long-term outcomes exploring whether specific patient subgroups may derive any differential treatment benefits. Findings are expected to generate high-quality, clinically actionable evidence to support future patient-centered decision-making, clinical guideline refinements, and health policy decision-making.