Study Summary

In this study from the France-TAVI registry, balloon-expandable (BE) transcatheter aortic valves were compared with self-expandable (SE) valves for paravalvular regurgitation (PVR) and mortality. Between 2013 and 2015, the nationwide registry included 12,141 patients who received a BE valve or a SE valve for aortic stenosis. Using propensity scores (25 clinical, anatomical, and procedural variables) and the date of the procedure (within 3 months), patients treated with a BE valve (Edwards; n = 3,910) were matched 1:1 with patients treated with a SE valve (Medtronic; n = 3,910). The first co-primary outcome was the occurrence of moderate-to-severe PVR or in-hospital mortality. The second co-primary outcome was all-cause mortality at 2 years.

In matched propensity-score analysis, the incidence of the first co-primary outcome was higher with SE than with BE valves (19.8% vs 11.9%; RR 1.68; 95% CI 1.46–1.91; P <0.0001). Each component of the outcome was more frequent in patients who received a SE valve. During follow-up (median 20 months), all-cause mortality was higher for SE than for BE valves (2-year Kaplan–Meier estimate, 29.8% vs 26.6%; HR 1.17; 95% CI 1.06–1.29; P = 0.003). The impact of PVR on short-term mortality remains unclear. Although excess mortality was observed in the first few months after the procedure, the mortality risk persisted after adjustment for PVR, suggesting that PVR is acting as a marker rather than being the main driver of the mortality difference between the valve designs.


Study Strengths: To date, this is the largest observational comparison of the effect of SE and BE valves on PVR and 2-year mortality. The prespecified falsification endpoints demonstrated no differences between valve types, suggesting adequate control of unmeasured confounders. Similar methodology has previously been highly predictive of the results of randomized studies.

Study Limitations: This is not a randomized trial and potential differences in unmeasured variables might remain. Specifically, extensive valve/root calcification and small femoral size were not measured and could be more frequent in patients receiving a SE than a BE valve, explaining all or part of the difference in mortality. Furthermore, severity of PVR and clinical events were site-reported and not centrally adjudicated, potentially resulting in heterogeneity in PVR grading and under-reporting of events. The newest SE valve (Evolut-Pro) was not available during the study period and, therefore, the results of this study may now have less relevance.

Next Steps/Clinical Perspective: This study from the FRANCE-TAVI registry suggests important differences in clinical outcomes relating to valve design. However, propensity-score matching cannot rule out residual confounders, and the most-recent valve technologies were not investigated. A contemporary randomized clinical trial powered to compare SE and BE valves would overcome these limitations but is unlikely to occur.

Trial Reference

Van Belle,  E. et al. Balloon-expandable versus self-expanding transcatheter aortic valve eplacement: a propensity-matched comparison from the France-TAVI registry. Circulation doi:10.1161/CIRCULATIONAHA.119.043785AA (2019).