Bicuspid Anatomy: Wrench Thrown in TAVR for Younger, Low-Risk Patients?

PARIS -- Patients with questionable suitability for transcatheter aortic valve replacement (TAVR) in fact fared worse in the short term after the catheter-based procedure than after surgery, the NOTION-2 randomized trial showed.

Unlike previous trials, this study of exclusively younger, low-risk patients with severe symptomatic aortic stenosis (AS) could not demonstrate noninferiority of TAVR relative to surgical aortic valve replacement (SAVR) in terms of combined death, stroke, or hospitalization (procedure-, valve-, or heart failure-related) at 1 year (10.2% vs 7.1%, HR 1.4, 95% CI 0.7-2.9).

The defining characteristic of the trial was its inclusion of low-risk individuals under age 75 -- and no exclusion of those with bicuspid valve anatomy, according to Ole De Backer, MD, PhD, of Rigshospitalet, University of Copenhagen in Denmark, at the EuroPCR conference. The full study manuscript was simultaneously published in the European Heart Journal.

Whereas the tricuspid cohort had similar outcomes between TAVR and SAVR groups regarding the primary endpoint (8.7% vs 8.3%), it appeared that the small subgroup of patients with bicuspid anatomy were the ones who had disproportionately bad outcomes if assigned to TAVR instead of SAVR (14.3% vs 3.9%, HR 3.8, 95% CI 0.8-18.5), De Backer reported.

Similarly, when considering death or disabling stoke, the tricuspid cohort in NOTION-2 had "truly excellent outcomes" after TAVR (2.2% vs 1.5% with surgery), he said, while the bicuspid patients again suggested harm (6.1% vs 2.0%, HR 3.1, 95% CI 0.3-30.0).

However, he pointed to the relatively wide confidence interval of the bicuspid findings and noted that these patients represented just 100 people in this study (most having a type 1 bicuspid phenotype), accounting for roughly 27% of the total NOTION-2 cohort.

De Backer said the experience from this "very exploratory" trial taught the investigators that type 2 bicuspid phenotypes in particular may be just too risky for TAVR, and use of cerebral embolic protection could be used more in the bicuspid population.

Cardiac surgeon Sabine Bleiziffer, MD, PhD, of Heart and Diabetes Center Nordrhein-Westfalen in Bad Oeynhausen, Germany, cited the ongoing NAVIGATE Bicuspid randomized trial as being expected to provide a lot of answers regarding the choice of TAVR versus surgery across phenotypes of bicuspid anatomy.

For now, she told the audience that TAVR for bicuspid AS patients be reserved for nonsurgical patients.

On the other hand, session discussant Davide Capodanno, MD, PhD, interventional cardiologist of University of Catania, Italy, stressed that the NOTION-2 study is "inconclusive" as the TAVR results straddle both lines of equivalence and noninferiority relative to SAVR. Part of the issue, he explained, was the fewer than expected adverse events in the surgical arm.

The mix of valves allowed in the trial was another concern raised by Martin Leon, MD, of NewYork-Presbyterian Hospital/Columbia University Irving Medical Center in New York City, another panelist at the session. He cautioned that not all TAVR valves perform the same.

De Backer agreed and acknowledged that the NOTION-2 results should not be extrapolated to all valve types. The TAVR arm was roughly split into 70% self-expanding valves and 30% balloon-expandable ones, he reported.

Trial participants were 370 people randomized to TAVR or SAVR. The group was age 71 years on average, with over 60% men. The cohort's "extremely low risk" group was evidenced by an STS-PROM score averaging just 1.1%.

The main driver of the between-group difference for the primary endpoint was the signal of more nondisabling strokes with TAVR (3.7% vs 0.5%, HR 7.0, 95% CI 0.9-56.5).

New permanent pacemaker implants (15.1% vs 8.0%) and moderate or greater paravalvular regurgitation (4.7% vs 0%) suggested other disadvantages to TAVR.

In contrast, the secondary endpoints favoring TAVR were major or life-threatening bleeding (4.8% vs 17.5%), new-onset atrial fibrillation (3.2% vs 41.7%), severe patient-prosthesis mismatch (10.1% vs 19.4%), and length of hospital stay (3 vs 7 days).

De Backer reported faster functional recovery in TAVR patients at 1 month compared with surgery. This difference disappeared by 12 months, however.

Comment