Low-Risk TAVR at 2 Years: Thrombosis Concerns Creep In

Transcatheter aortic valve replacement (TAVR) maintained good results but lost some of its early advantages over surgery 2 years into the PARTNER 3 trial, with valve thrombosis emerging as a potential point of concern.

In low-risk patients with symptomatic severe aortic stenosis, the trial’s primary endpoint still favored those randomized to TAVR instead of surgical aortic valve replacement (SAVR) at 2 years (11.5% vs 17.4%, HR 0.63, 95% CI 0.45-0.88). However, the difference at that point was driven solely by rehospitalization among the components:

• Death: 2.4% after TAVR vs 3.2% after SAVR (P=0.47)
• Stroke: 2.4% vs 3.6% (P=0.28)
• Rehospitalization: 8.5% vs 12.5% (P=0.046)

The TAVR group had had more deaths and strokes between 1 and 2 years compared with the SAVR group, according to the PARTNER 3 investigators led by Martin Leon, MD, of Columbia University Irving Medical Center in New York City. Their report was published in the March 9 issue of the Journal of the American College of Cardiology.

“Clearly, ongoing assessment of clinical and echocardiographic findings is needed in younger and low-risk patients and planned follow-up in PARTNER 3 will continue through at least 10 years,” Leon and colleagues wrote.

The convergence in death and stroke came as no surprise to cardiac surgeon Joseph Bavaria, MD, of the Hospital of the University of Pennsylvania in Philadelphia, who participated in the trial.

Bavaria noted that TAVR and SAVR both offer excellent results in general. However, “at this rate, there will be a crossing of the curves by 3 years, which actually happened in PARTNER 2 as well … By 5 years, there could be a benefit for SAVR,” he said.

Leon’s group reported that valve thrombosis rates continued to diverge through 2 years such that TAVR’s numerically higher rate at 1 year (1.0% vs 0.2%, P=0.13) grew to a significant excess by 2 years (2.6% vs 0.7%, P=0.02).

However, three out of every four patients with valve thrombosis had it discovered from imaging and experienced no symptoms.

“Nevertheless, these findings were frequently associated with elevated valve gradients and need for anticoagulation (with increased risk of bleeding), and disabling stroke in two patients,” commented a group led by Bernard Prendergast, MD, of St. Thomas’ Hospital in London, in an accompanying editorial.

It is possible that valve thrombosis was behind TAVR’s small increase in stroke, according to Bavaria.

Suzanne Arnold, MD, of Saint Luke’s Mid America Heart Institute in Kansas City, Missouri, thought differently.

“I don’t think this can be explained by the slight increase in valve thrombosis. I think the mortality and stroke rates are so low in this low-risk cohort that a few events in one direction or the other can impact the results,” she commented.

“Although the incidence of clinically relevant valve thrombosis is low and consequences of asymptomatic minor increases in aortic valve gradients are uncertain, further work to identify those at highest risk will be needed before use of TAVR expands further to even younger cohorts,” Prendergast’s group wrote.

Other echocardiographic findings from PARTNER 3 showed similarly infrequent VARC-3 defined hemodynamic valve deterioration and bioprosthetic valve failure between low-risk TAVR and SAVR groups at 2 years. Device durability is not expected to be an issue this early on, however.

There was no difference in moderate or severe paravalvular regurgitation, but surgery maintained reduced rates of mild regurgitation.

TAVR held a small but statistically significant advantage over surgery for 2-year disease-specific health status per the Kansas City Cardiomyopathy Questionnaire.

“I think, overall, the story remains what has been already thought. In patients at low surgical risk, TAVR results in quicker recovery and rapid improvement in health status, which translates to fewer hospitalizations during that recovery period,” Arnold concluded.

American College of Cardiology/American Heart Association guidelines for valvular heart disease recommend SAVR for people under 65 years of age and TAVR for those older than 80. People in the gray area by age have a choice of intervention that is ideally informed by a shared decision-making process.

In 2019, on the basis of 1-year data from the PARTNER 3 and Evolut Low Risk trials, FDA expanded the indication for TAVR to include people at low surgical risk, making people at all risk categories eligible for this procedure.

PARTNER 3 included 1,000 people randomly assigned to transfemoral TAVR with the Sapien 3 valve or SAVR. Average STS score was 1.9%, and mean age 73 years.

TAVR and SAVR groups differed in the proportion of patients with available 2-year follow-up data for the present analysis (99.0% vs 93.8%).

“Trial logistic issues, including disproportionate study withdrawal in the surgery cohort and missing follow-up data or lack of formal adjudication of some secondary endpoints are study limitations being managed with sensitivity analyses and other statistical adjustments,” according to Leon’s group.

“It should be emphasized that because PARTNER 3 excluded patients with specific anatomic features suboptimal for TAVR, bicuspid aortic valve disease, and patients without acceptable transfemoral access, the reported results cannot be generalized to all patients and apply only to the enrolled study population,” the authors cautioned.