Low Risk Patients: TAVR with Self-Expanding Valves Offers Similar Outcomes to Surgery Based on a Bayesian Analysis?

When compared against surgical valve replacement (SAVR) transcatheter aortic valve replacement (TAVR) has shown benefits across different risk groups, not only in events as death and stroke, but also in quality of life (especially in the first months post procedure) and improved symptoms.

Pacientes de bajo riesgo: ¿El TAVI con válvulas autoexpandibles ofrece resultados similares a la cirugía utilizando un análisis Bayesiano?

However, a catch-up phenomenon was observed in low-risk patients between the first and second year followup, which was associated to an early benefit of TAVR in death or disabling stroke.

This phenomenon has not yet been looked at in TAVR with self-expanding valves.

The present study carried out a Bayesian analysis of the Evolute Low Risk at 2 years to assess non-inferiority of TAVR with self-expanding valves vs. SAVR for all-cause mortality and disabling stroke.

It included 1414 patients, 730 undergoing TAVR and 684 SAVR.

There were no differences in population characteristics. Mean age was 74, 35% were women, , 30% diabetic, they were mostly hypertensive, 15% had atrial fibrillation, 3.5% had prior pacemaker. Ejection fraction was 61% and STS Score was 2% for TAVR and 1.9% for SAVR.

There were no differences in death or disabling stroke at 2 years between both groups (4.3% TAVR vs. 6.3% SAVR, P=0.084), neither were there divergencies in these events between first and second year after procedure.

At 2 years, all-cause mortality rate was 3.5% for TAVR vs 4.4% for SAVR patients (P=0.366) and disabling stroke was 1.5% vs. 2.7% (P=0.119) respectively. The presence of valve thrombosis was similar.

TAVR presented lower gradient, lower prosthesis mismatch and larger effective aortic area. However, it was associated with more minor paravalvular leak. As regards quality of life, it was superior with TAVR at 20 days, with not differences after 2 years.

Bayesian Analysis rendered comparable outcomes: 5.3% for TAVR and 6.7% for SAVR (difference: -1.4%; 95% Bayesian credible interval: -4.9% to 2.1%). Between the first and second year there was no convergence of the primary outcome curves.

Conclusion

At 2 year followup, the Evolut Low Risk showed TAVR is non-inferior vs. SAVR for the primary end points of mortality or disabling stroke, with event rates slightly better to those predicted based on Bayesian analysis.