Adverse Events Drop With Fuller Accounting of Evolut Low Risk Trial

Transcatheter aortic valve replacement (TAVR) with CoreValve and Evolut devices maintained good results in the Evolut Low Risk Trial through 2 years, with investigators now boasting complete follow-up and no more need to rely on Bayesian analysis.

In 2019, a prespecified interim analysis of the trial found that the primary endpoint — all-cause mortality or disabling stroke at 24 months — occurred in 5.3% of TAVR recipients compared with 6.7% of the surgical group under Bayesian methods, demonstrating that TAVR was noninferior for efficacy.

Now, with complete follow-up, 2-year event rates were 4.3% and 6.3%, respectively (log-rank P=0.084), such that TAVR looks, at least numerically, even better than before, said John Forrest, MD, of Yale University School of Medicine in New Haven, Connecticut, during a presentation at this year’s virtual European Association of Percutaneous Cardiovascular Interventions (EuroPCR) meeting.

It’s “not surprising” that Kaplan-Meier event rates were generally lower than Bayesian predicted rates, Forrest said, citing device evolution from the CoreValve classic at the beginning of the trial to Evolut R to Evolut PRO by the end. Adaptive Bayesian design allows for better prediction as more data are accrued.

There was no convergence in Kaplan-Meier curves for death or disabling stroke between 1 and 2 years, Forrest noted. The two individual endpoints each showed no difference between groups.

However, surgical aortic valve replacement (SAVR) continued to be superior to TAVR with regards to the incidence of permanent pacemakers at 2 years (21.1% vs 7.9%).

“We’ve certainly made improvements [with pacemakers] since the study was completed … in particular using the RAO [right anterior oblique] cusp overlap, [which] has been shown to significantly reduce the incidence of pacemaker with the Evolut valve. I think what we’ve seen is pacemaker rates drop in the 2 years since the study was completed,” Forrest said.

Greater-than-mild paravalvular leak also favored SAVR, whereas TAVR beat surgery in terms of valve hemodynamics and prosthesis-patient mismatch.

Trial participants had severe aortic stenosis with a 30-day surgical mortality risk of <3% per local heart team assessment.

A central clinical events committee adjudicated all adverse events and a central lab assessed all echocardiograms.

The present as-treated analysis included 730 TAVR and 684 SAVR recipients, 97.3% and 92.3% of whom, respectively, had complete follow-up data at 2 years.

Notably, when the interim Bayesian analysis was published in 2019, only 137 patients had gotten their full 24-month data into the manuscript.

Forrest acknowledged that longer follow-up is needed to assess differences between TAVR and SAVR groups and their potential impact on patient outcomes. Follow-up for this trial is being planned out to 10 years, he said.

In PARTNER 3, another low-risk trial, TAVR with the Sapien 3 valve lost some of its early advantages over surgery and saw a concerning elevation in valve thrombosis at 2 years.