TAVR Matches Open-Heart Surgery for Valve Durability at 7 Years

For decades, open-heart surgery was the gold standard for replacing a failing aortic valve, and concerns about long-term durability kept transcatheter aortic valve replacement (TAVR) reserved largely for older, higher-risk patients. As TAVR technology matured and indications expanded to younger, lower-risk populations, the critical unanswered question became: how long do these minimally invasive valves actually last?

This ad hoc analysis from the PARTNER 3 randomized clinical trial provides the most robust long-term answer to date. Between 2016 and 2017, 1,000 low-risk patients with symptomatic severe aortic stenosis were randomized at 71 centers across the US and Canada to receive either balloon-expandable TAVR using the SAPIEN 3 valve or conventional surgical aortic valve replacement. The analysis tracked valve-specific outcomes through March 2025, with echocardiographic data available in 80% of surviving enrolled patients at the 7-year mark.

The headline result is striking in its equivalence. Cumulative incidence of structural valve deterioration-related dysfunction was 7.3% for TAVR versus 7.6% for surgery. All-cause bioprosthetic valve failure stood at 6.9% versus 7.5%, and reintervention rates were 6.0% versus 5.5% — none of these differences approaching statistical significance. Roughly three-quarters of patients in both groups remained alive and free of valve failure at seven years.

One meaningful difference emerged: TAVR valves experienced higher rates of stage 2 or 3 valve thrombosis (5.2% vs 0.9%), though most events clustered within the first three years and rarely progressed to clinical valve failure — suggesting this is a manageable rather than catastrophic risk.

These findings carry significant implications for clinical decision-making. Clinicians counseling patients in their 60s or early 70s can now cite robust randomized data showing comparable 7-year valve performance. Caveats include the abstract-only basis for this summary, the ad hoc nature of the analysis, and the fact that 10- and 15-year data remain unavailable.