Native-Coronary versus Bypass-Graft PCI after CABG: A Design- and Time-Stratified Systematic Review and Meta-analysis
Abstract Background Observational evidence has historically favored percutaneous coronary intervention (PCI) of the bypassed native coronary artery rather than the bypass graft after coronary artery bypass grafting (CABG). PROCTOR, the first randomized target-strategy trial, reported a directionally different one-year estimate. We evaluated whether these evidence blocks could be reconciled by study design, effect-measure family, follow-up horizon, anatomy, and registry overlap. Methods We systematically reviewed randomized and comparative observational studies of native-vessel versus bypass-graft PCI after CABG. Randomized and observational evidence was analyzed separately. Unadjusted binary outcomes were pooled as risk ratios (RRs) using Paule-Mandel random-effects models with Hartung-Knapp confidence intervals; adjusted hazard ratios (HRs) and odds ratios (ORs) were synthesized only within the same effect-measure family. Follow-up was classified as early, 9–12 months, or > 12 months. Risk of bias was assessed using RoB 2 and ROBINS-I, and certainty using GRADE. Results Thirty-five original comparative reports were identified (34 observational and one randomized trial). Twelve observational studies contributed to the primary long-term MACE synthesis, which favored native-vessel PCI (RR 0.64, 95% CI 0.53–0.77; I2 = 47.6%). At one year, PROCTOR favored graft PCI (HR 2.14, 95% CI 1.25–3.65), whereas adjusted observational estimates favored native PCI (HR 0.58, 95% CI 0.35–0.95; OR 0.79, 95% CI 0.69–0.91). Four adjusted observational mortality HRs yielded HR 0.81 (95% CI 0.69–0.95). These estimates were not pooled across design or metric. Conclusions Randomized and observational evidence represent clinically different target-selection scenarios and should not be combined into a single causal estimate. Current data support an individualized, anatomy- and time-dependent strategy rather than a universal native-first or graft-first rule; longer randomized follow-up is needed to determine comparative durability.
Authors
- Hamzah Pratama Megantara (ORCID: https://orcid.org/0000-0002-1926-3535)
- Joeristanti Soelistyaningroem (ORCID: https://orcid.org/0009-0006-8851-1877)
- Naura Anjanie (ORCID: https://orcid.org/0009-0004-4668-3916)
Institutions
- Universitas Airlangga (ID)
- University of Indonesia (ID)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-12
- DOI
- https://doi.org/10.21203/rs.3.rs-10689689/v1
- Primary Topic
- Coronary Interventions and Diagnostics
- Type
- preprint