Benefits of Early SAVR for Asymptomatic Aortic Stenosis Persist a Decade On

Patients with asymptomatic severe aortic stenosis (AS) who undergo immediate SAVR have better outcomes than those who opt for a more conservative approach when followed for at least 10 years, according to final results of the RECOVERY trial.

The primary endpoint, a composite of operative mortality or death from CV causes, occurred in just two patients in the surgery group (3%) compared with 17 patients managed conservatively (24%; HR 0.10; 95% CI 0.02-0.43), lead author Duk-Hyun Kang, MD, PhD (Asan Medical Center, Seoul, South Korea), and colleagues report in a paper published online this week in the New England Journal of Medicine. Kang previously presented the results at the American Heart Association (AHA) Scientific Sessions in November.

There were no cases of operative mortality—ie, during surgery or in the 30 days following the operation—either among patients randomized to immediate SAVR or among those in the control group who underwent delayed AVR. The rate of all-cause death was 15% in the early SAVR group and 32% in the conservative group (HR 0.42; 95% CI 0.21-0.86).

These longer-term results are consistent with what was seen in the trial at a median follow-up of about 6 years, as presented by Kang at the AHA 2019 Scientific Sessions and published in NEJM.

“The lack of convergence of the curves for death from cardiovascular causes and death from any causes over this prolonged period of follow-up underscores the sustained benefits of early surgery,” Kang told TCTMD via email. “Accordingly, we believe that early aortic valve replacement may be the preferred strategy for asymptomatic patients with severe aortic stenosis (peak aortic jet velocity ≥ 4.5 m/s).”

Robert Bonow, MD (Northwestern University Feinberg School of Medicine, Chicago, IL), commented to TCTMD that aside from now having the longest follow-up, RECOVERY also stands out from the other trials evaluating early AVR in patients with asymptomatic severe AS—including AVATAR, EARLY TAVR, and EVOLVED—in that patients were younger and had fewer cardiovascular comorbidities (including a low rate of diabetes), a high prevalence of bicuspid aortic valves, and greater severity of AS indicated by an increased peak aortic jet velocity and transaortic pressure gradient.

Those features could help explain why there was such a low mortality rate seen in the early-surgery arm of RECOVERY and why there was a significant difference in mortality when compared with conservative management, he said. That was something that was not seen in EARLY TAVR or EVOLVED and only emerged in extended follow-up of AVATAR.

These longer-term data from RECOVERY help resolve concerns raised after the initial results were presented about an apparent uptick in CV mortality in the control arm after 4 years, when the frequency of surveillance may have dropped off, Bonow said. Now, with all patients having attained at least 10 years of follow-up, the mortality curves look similar to those from EARLY TAVR, which has very careful follow-up of patients, he said.

“That’s reassuring because it suggests that, indeed, these patients [in RECOVERY] were being followed carefully,” he noted.

Early Intervention vs Watchful Waiting

Longer-term follow-up is critical to understand how the effects of early AVR in asymptomatic patients with severe AS may shift over time, “as the long-term risk related to prosthetic valves, including bioprosthetic valve degeneration, thromboembolic complications, and anticoagulation-related bleeding, may exceed the risk of irreversible myocardial damage associated with more-prolonged exposure to pressure overload imposed by severe aortic stenosis,” Kang said.

The RECOVERY investigators followed patients for at least 10 years to explore that question. The trial included 145 asymptomatic patients (mean age 64.2 years; 49% men) who had very severe AS, defined by an aortic valve area ≤ 0.75 cm² plus a peak aortic jet velocity ≥ 4.5 m/s or a mean transaortic gradient ≥ 50 mm Hg. They were randomized to early surgery (within 2 months of randomization) or conservative care, which included referral for AVR if symptoms developed, if LVEF dropped below 50%, or if peak aortic jet velocity increased by more than 0.5 m/s per year.

Through the final follow-up, at a median of 12 years, 85% of the conservatively managed patients eventually underwent AVR—59 underwent SAVR and two TAVI. The development of symptoms led to the intervention in 80% of those cases.

In a Kaplan-Meier analysis, the cumulative incidence of operative mortality or CV death at long-term follow-up was 1% after early SAVR and 19% in the control arm. The number needed to treat (NNT) to prevent one CV death was 6 and to prevent one all-cause death was 7.

In the conservative arm, the cumulative incidence of all-cause death or AVR was 74% at 5 years and 97% at 10 years. There were no hospitalizations for heart failure in the early-surgery arm and 14 (19%) in the patients who were managed conservatively.

Shared Decision-making Is Key

Bonow noted that many patients in RECOVERY already would fall under class IIa recommendations in current European and US guidelines to consider prompt AVR if the peak aortic jet velocity is over 5 m/s (the average was about 5.1 m/s in the trial). The trial reinforces that in patients with such severe AS, regardless of symptoms, “we should be leaning toward an early intervention,” he said.

The choice of whether asymptomatic patients with severe AS will undergo early AVR or wait for symptoms to develop or for the stenosis to progress comes down to shared decision-making with an individual patient, Bonow said.

“We are having really in-depth conversations with our patients who have asymptomatic, progressively severe aortic stenosis [about how] it’s inevitable that they’re going to require a valve replacement,” with consideration of doing it immediately or waiting, he said.

Many patients choose to wait because they feel fine, while others want to get their valve replaced right away.

Bonow said that if patients understand that they will need to be followed very carefully if they forgo an immediate intervention and will need to promptly alert the healthcare team if symptoms develop, then watchful waiting remains a reasonable option.

“I think in the current era, where we can do TAVR in these asymptomatic patients who have severe aortic stenosis, it really becomes an important piece of [information to share] with the patients so that they know what we know and what we don’t know, and that they will be part of the decision-making process,” he said.

Kang noted that among older patients with more comorbidities, there are more competing risks that can potentially reduce the long-term benefits of early AVR.

“On the other hand, considering substantial undertreatment of severe AS in elderly patients, early AVR might provide a much greater benefit than conservative treatment in real-world practice,” he said, adding that larger, ongoing trials, like EASY-AS, will provide more insights.