Coronary angiography- or fractional flow reserve-guided complete revascularization in multivessel disease STEMI: A Bayesian hierarchical network meta-analysis

Aims: To identify the best strategy to achieve complete revascularization (CR) in patients with ST-elevation myocardial infarction (STEMI) and multi-vessel disease (MVD). Methods and results: We systematically reviewed the literature for randomized controlled trials (RCTs) comparing IRA-only PCI and CR guided by angiography or fractional flow reserve (FFR) in MVD-STEMI. Both frequentist (classical) and Bayesian network meta-analysis were performed, including a comparative hierarchy estimation of the probability to reduce the primary composite endpoint of all-cause death and new myocardial infarction (MI). We identified 11 RCTs, including 8193 STEMI patients. Compared with IRA-only strategy, CR significantly reduced the primary endpoint (OR: 0.73; 95%CI0.55–0.97). We observed non-significant difference between angiography and FFR guidance in reducing the primary endpoint (OR: 0.73, 95% CI 0.35–1.57). The Bayesian probability analysis ranked angio-guided CR as the best intervention yielding lowest risk of all-cause death or new MI (SUCRA92%). Conclusions: In patients with MVD-STEMI, CR is associated with a reduction in all-cause mortality and new MI compared with IRA-only PCI. Angio-guided CR is associated with the lowest risk of all-cause death or new MI, therefore the role of FFR-guidance in this setting is questionable. Condensed abstract: Both frequentist and Bayesian network meta-analysis were performed to compare infarct-related artery (IRA)-only percutaneous coronary intervention (PCI) and complete revascularization (CR) guided by angiography or fractional flow reserve (FFR) in multivessel disease (MVD) and acute ST-elevation myocardial infarction (STEMI). Eleven randomized controlled trials were identified, including 8193 STEMI patients. Compared with IRA-only strategy, CR significantly reduced the incidence of the composite endpoint of all-cause death and new myocardial infarction without significant difference in angio-guided and FFR-guided CR. The Bayesian probability analysis ranked angio-guided CR as the best intervention yielding lowest risk of the composite endpoint and, therefore the role of FFR-guidance in this setting is questionable.