Effect of mechanically expanded vs self-expanding transcatheter aortic valve replacement on mortality and major adverse clinical events in high-risk patients with aortic stenosis

Selected in the JAMA by R.L.J. Osnabrugge

The REPRISE III randomized clinical trial



Feldman TE, Reardon MJ, Rajagopal V, Makkar RR, Bajwa TK, Kleiman NS, Linke A, Kereiakes DJ, Waksman R, Thourani VH, Stoler RC, Mishkel GJ, Rizik DG, Iyer VS, Gleason TG, Tchétché D, Rovin JD, Buchbinder M, Meredith IT, Götberg M, Bjursten H, Meduri C, Salinger MH, Allocco DJ, Dawkins KD


JAMA. 2018 Jan 2;319(1):27-37


January 2018


Read the abstract

My Comment

Why this study – the rationale/objective?

Different designs of transcatheter aortic valve systems have distinct advantages and disadvantages.

This study was designed to test the non-inferiority of a mechanically expanded transcatheter valve (MEV) compared to a self-expanding transcatheter valve (SEV).

How was it executed – the methodology?

In total 912 patients with high or extreme risk and severe aortic stenosis were enrolled at 55 North American and European centres. The REPRISE III trial randomized patients in a 2:1 ratio to either MEV (Lotus Valve System, Boston Scientific) or SEV (Corevalve or Corevalve Evolut R, Medtronic).

The primary safety endpoint was the 30-day composite of all-cause mortality, stroke, life-threatening bleeding, stage 2/3 kidney injury and major vascular complications. The non-inferiority margin was set at 10.5%.
The primary effectiveness endpoint was the composite of all-cause mortality, disabling stroke and moderate or greater paravalvular leak at 1 year. The non-inferiority margin for this endpoint was 9.5%.
If the non-inferiority criteria were met, the secondary endpoint of 1-year moderate or greater paravalvular leak was tested for superiority.

What is the main result?

In total, 874 (96%) with a mean STS-PROM of 6.8% were available for evaluation at 1 year. The primary 30-day safety endpoint occurred in 20.3% in the MEV group compared to 17.2% in the SEV group (3.1% difference, [95% CI -2.3 to 8.5], p=0.003 for non-inferiority).

The 1-year primary effectiveness endpoint occurred 15.4% in the MEV group and 25.5% in the SEV group (-10.1, 95% CI -16.2 to -3.9, p<0.001 for non-inferiority).

Superiority testing of the secondary endpoint of moderate or severe PVL showed a lower rate in the MEV group versus the SEV group (-6.1%, 95% CI -9.6 to -2.6%, p<0.001 for superiority).

Prespecified superiority testing of the 1-year primary effectiveness endpoint (all-cause mortality, disabling stroke and moderate or greater PVL) showed a between-group difference of -10.2% in favour of the MEV (95% CI -16.3% to -4.0%).

Exploratory additional analyses showed a lower rate in disabling strokes after MEV (difference -3.5, 95% CI -6.8 to -0.2), new pacemaker implantation rates (difference 18.4%, 95% CI 11.5% to 25.3%) and valve thrombosis (difference 1.5%, 95% CI 0.5% to 2.5%).

Critical reading and the relevance for clinical practice

The results of this study suggest that the use of the MEV (Lotus System, Boston Scientific) may be a useful addition for treating high-risk patients with severe aortic stenosis as compared with a self-expandable valve system.

There are however several limitations of this study.
First of all the non-inferiority margins were based on available data at the time of protocol writing. The lower than expected event rates mean that the confidence intervals of the primary endpoints should carefully be taken into account.
Second, the MEV was compared to an earlier generation SEV, which in the meantime has been replaced by a newer generation SEV. Also, there was no comparison to balloon-expandable transcatheter valves, which are an important option in the armamentarium of the interventionalist. Importantly, the long-term durability of the devices was not assessed.

No comments yet!