Long-term valve durability and clinical outcomes after transcatheter versus surgical aortic valve replacement in low-risk patients: a systematic review and meta-analysis of randomized trials

Transcatheter aortic valve replacement (TAVR) has extended its use to low-surgical-risk populations, yet long-term comparative data against surgical aortic valve replacement (SAVR) are still limited, particularly concerning durability and the need for reintervention. We conducted a search of PubMed/MEDLINE, Embase, CENTRAL, and ClinicalTrials.gov through March 2026 for randomized trials comparing TAVR with SAVR in patients with low surgical risk and severe aortic stenosis, specifically those with at least three years of follow-up. Outcomes were pooled as risk ratios (RR) using Mantel-Haenszel random-effects models, and durability and mortality outcomes were analyzed as time-to-event hazard ratios (HR) to accommodate different follow-up durations. Prespecified subgroup analyses by valve platform were performed. Four trials involving 3,014 patients (follow-up ranging from 3 to 10 years) were included. Cardiovascular mortality (HR 1.11, 95% CI 0.86–1.42; P = 0.43) and all-cause stroke (HR 1.09, 95% CI 0.73–1.64; P = 0.67) did not differ significantly between TAVR and SAVR. Aortic valve reintervention was higher with TAVR in risk-ratio analysis (RR 1.58, 95% CI 1.16–2.16) but did not reach statistical significance in the prespecified time-to-event analysis (HR 1.29, 95% CI 0.81–2.05). The excess reintervention risk was confined to the self-expanding valve platform. Permanent pacemaker implantation was significantly more frequent after TAVR (RR 2.08, 95% CI 1.48–2.93; P < 0.0001), with a significant interaction by valve platform (self-expanding RR 2.57 vs balloon-expandable RR 1.40; P = 0.02 for interaction). Bioprosthetic valve failure showed a nonsignificant trend favoring TAVR (HR 0.80, 95% CI 0.54–1.17). The only available 10-year randomized data (NOTION trial) showed significantly less severe structural valve deterioration with TAVR than with SAVR. The results suggest that TAVR and SAVR are equivalent in terms of cardiovascular mortality and stroke outcomes. However, pacemaker implantation rates were higher with TAVR and were strongly dependent on the valve platform used. The observed excess in reinterventions was confined to self-expanding valves and did not show robust support in the time-to-event analysis. The unique 10-year data from the self-expanding platform indicate less severe structural valve deterioration compared to SAVR, suggesting that durability should not be generalized across different device families. These findings advocate for individualized, platform-aware decision-making in clinical practice.