Early LV Recovery After TAVI Points to Better 5-Year Survival

Following TAVI for severe aortic stenosis with a left ventricular ejection fraction less than 50%, about one-third of patients will have an improved LVEF, and this is associated with greater survival at 5 years, according to new PARTNER data.

Additionally, researchers observed a strong interaction with sex, with mostly women seeing this mortality benefit.

“This study really seeks to increase awareness of the importance of cardiac function on clinical outcomes,” senior author Sammy Elmariah, MD, MPH (Massachusetts General Hospital, Boston), told TCTMD. “We often think that simply replacing the valve is all that these patients need, but what this study really supports is this concept that it goes beyond simply the valve, that we need to focus on the heart muscle in order to optimize our patients’ outcomes. That requires partnership with our heart failure colleagues and an intense focus on maximizing goal-directed medical therapy in order to provide patients with as much benefit from these procedures as possible.”

Previous studies have shown that LVEF improvement is associated with survival outcomes, including one looking at 1-year outcomes in patients at extreme and high surgical risk who underwent TAVI with CoreValve devices (Medtronic).

Lead author of that study, Harold Dauerman, MD (University of Vermont Medical Center, Burlington), commented to TCTMD that what the new data add is longevity, as they go out to 5 years, as well as the “provocative” finding that women are more likely to have a mortality benefit associated with LV recovery. This, he said, “points to some difference in pathophysiology or comorbidities of men versus women that I don't fully understand.”

LVEF Improvement Tied to Lower Risk of Death

For the study, published online this week in JAMA Cardiology, Dhaval Kolte, MD, PhD (Massachusetts General Hospital), Elmariah, and colleagues included data from 659 patients (mean age 82.4; 71% male) with a baseline LVEF ≤ 50% who were enrolled in the PARTNER 1, 2, and S3 trials and registries between July 2007 and April 2015.

Just under one-third of patients (32.8%) saw LVEF improvement within 30 days of TAVI with a mean improvement of 16.4%. Patients with prior MI, diabetes, cancer, higher baseline LVEF, larger LV end-diastolic diameter, and larger aortic valve area were less likely to report this improvement. Overall, 55.6% of the cohort died during the study period.

At 5 years, patients who had early recovery of LV function after TAVI had lower risks of all-cause mortality (50.0% vs 58.4%; P = 0.04) and cardiac death (29.5% vs 38.1%; P = 0.05) compared with those who did not. Multivariate analyses confirmed the association between LVEF improvement and 5-year all-cause mortality (adjusted HR per 5% increase in LVEF 0.94; 95% CI 0.88-1.00) and cardiac death (adjusted HR per 5% increase in LVEF 0.90; 95% CI 0.82-0.98).

An additional analysis demonstrated an inflection point above a 10% change in LVEF beyond which there was a steep decline in all-cause mortality with improved LVEF. There were no differences in rehospitalization, NYHA functional class, or Kansas City Cardiomyopathy Questionnaire score at 5 years between those who did or did not have early recovery of LV function.

In subgroup analyses, the link between 5-year all-cause mortality and LVEF improvement within 30 days remained regardless of whether or not the patient had CAD or prior MI. But when comparing the sexes, only women who had early LVEF recovery had lower rates of all-cause death, cardiac death, all-cause death or rehospitalization, and cardiac death or rehospitalization at 5 years; these relationships were not observed in men (P = 0.01 for interaction).

Elmariah said he was “a little bit” surprised by their findings. “The impact of a relatively small improvement in left ventricular ejection fraction is really quite significant and does remain durable for as long as 5 years, which is pretty impressive,” he said. “We're essentially talking about a change in ejection faction that happens within the first month after TAVR, and that has a 5-year impact on mortality risk.”

As for why women might be more likely to derive a survival benefit, Elmariah said this remains unknown but is likely related to the many biological differences in cardiac remodeling between the sexes. “There are differences in the extent of fibrosis [and] there's differences in the extent of hypertrophy in response to aortic stenosis,” he said. “I expect that there are some underlying biologic differences that, to be honest, we may not fully understand at this point.”

Some of it could be due to patient risk as well, he said, pointing out that STS score will be higher for a woman than a man with the same age and risk factors. “What that means is that independent of gender, the women are maybe a little bit healthier; they need less other medical conditions to qualify for an intermediate- and high-risk study,” Elmariah said. “And so because they're healthier, they actually may be less prone to some of the adverse clinical entities that can affect the myocardium and hinder its recovery.”

‘A Real Phenomenon’

Dauerman said that in the CoreValve study of patients at extreme and high surgical risk, nearly two-thirds saw early LVEF recovery. The lower percentage of patients seeing early LVEF recovery in this study likely has to do with the lower-risk patient population, with higher baseline ejection fractions included as well as the different definition of LV dysfunction. “If you define LV dysfunction as [an LVEF of] less than 40%, you have a much better chance of seeing a 10% improvement than if your EF is 48% and you're trying to get them to 58%,” he said.

But a common theme is that absence of prior MI is linked with early LVEF recovery. “Presumably, [prior MI is] a surrogate for a fixed scar in the ventricle, and those patients are less likely to get EF recovery,” Dauerman said.

This study “continues to tell a story that we've been trying to tell for about 6 years now that aortic stenosis is this high-afterload state that can lead to reduced cardiac outputs,” he continued. “When you relieve that high-afterload state and improve cardiac output and decrease vascular resistance, you benefit way more than just the valve itself. You help the LV remodel and recover.”

Dauerman’s take-home message “is that this is a real phenomenon—that it's persistent, that there's dynamic changes that occur in the ventricle, the cardiac output, the afterload state, and renal perfusion that benefit the patient far beyond just fixing the valve and their shortness of breath,” he said.

Elmariah agreed. “We have to think about what exactly is causing the patient's heart to fail,” he said, noting that if aortic stenosis is the main concern, the patient will likely “do well and survive for a long period of time” following TAVI. However, “if there are other clinical entities that have impaired the heart function before the aortic stenosis really came into play, those patients, unfortunately, while they're still better off for having undergone the TAVR, I think the magnitude of recovery and the clinical impact of that procedure is reduced in that patient subset.”

These are important concepts to bring into patient conversations and risk prediction, he said. “I think this observation really helps inform some of those conversations in regards to what we can tell patients about the likelihood that their heart function will improve.”

In the future, Elmariah said he would like to see more work done to improve the treatment of aortic stenosis “before heart injury progresses to a severe stage. What I mean by that is that we often in clinical practice wait too long to replace somebody's aortic valve, or it takes a long time for patients to actually seek the appropriate medical care and to have their valve replaced. And these data certainly suggest that if you wait [and] you are intervening too late, the cardiac injury is permanent and there are long-term consequences of that delay.”

Hence, additional studies that look at the timing of valve interventions and the effects on cardiac recovery, as well as on long-term survival, are warranted, he said.

It may also be time for sex-specific clinical trials in TAVI that eliminate any selection bias between women and men, Dauerman suggested. This “might be a way to address this question to determine the true amount of LV recovery. And then I would add other parameters that can help us understand what may be different. For example, cardiac MRI could give us parameters related to myocardial scar, fibrosis, and LV dimensions that may be different between men and women.”