Routine versus deferred percutaneous coronary intervention in patients undergoing transcatheter aortic valve implantation: a frequentist and Bayesian meta-analysis of randomized trials with trial sequential analysis

Background The optimal management of concomitant coronary artery disease (CAD) in patients undergoing transcatheter aortic valve implantation (TAVI) remains debated. Therefore, we performed this meta-analysis of randomized controlled trials (RCTs) to compare routine percutaneous coronary intervention (PCI) versus deferred PCI in patients with severe aortic stenosis and significant coronary artery disease undergoing TAVI. Methods We systematically searched major electronic databases for eligible RCTs. The primary efficacy outcome was spontaneous myocardial infarction (MI); the primary safety outcome was major bleeding. Bayesian random-effects meta-analysis models were fitted using weakly informative prior distributions. Effect estimates are reported as posterior median risk ratios (RR) with 95% credible intervals (CrI). The posterior probability of benefit (PPB) for efficacy outcomes or harm (PPH) for safety outcomes was calculated (RR < 1 or RR > 1, respectively). Results Three RCTs comprising 1156 patients were included. Routine PCI did not demonstrate a convincing posterior probability of benefit for spontaneous MI (posterior median RR 0.84, 95% CrI 0.43–1.73; PPB: 70.2%) or all-cause mortality (posterior median RR 0.91, 95% CrI 0.60–1.43; PPB: 69.6%). However, routine PCI was associated with a reduced risk of urgent revascularization (posterior median RR 0.37, 95% CrI 0.15–0.91; PPB: 98.1%). For safety outcomes, routine PCI was associated with an increased posterior probability of harm for major bleeding (posterior median RR 1.56, 95% CrI 0.91–2.71; PPH: 95.6%) and life-threatening or disabling bleeding (posterior median RR 1.74, 95% CrI 1.00–3.07; PPH: 97.4%). Conclusions Among patients undergoing TAVI with concomitant CAD, routine PCI did not reduce ischemic events or mortality but increased bleeding complications. These findings suggest that a selective revascularization strategy, rather than a routine one, may be preferable.