REHAB-HF Lays the Groundwork for Recovery of Function After Acute HF

This transcript has been edited for clarity.

Ileana Piña, MD, MPH: Hello. I'm Ileana Piña from Central Michigan University, and this is my blog. We are reporting on the American College of Cardiology national meeting which, once again this year, is virtual. The late-breaking sessions have been fabulous and I've been watching most of them from early in the morning. One that was presented on Sunday morning was from my good friend, Dalane Kitzman, who is professor of medicine and the Phillips Chair of Cardiovascular Medicine at Wake Forest School of Medicine. For all the time I've known Dalane, he has been consumed with heart failure with preserved ejection fraction (HFpEF). I think he knows more about the syndrome than anyone else in the country.

His other love is geriatric cardiology. Today he presented results of REHAB-HF, a fascinating study for those of us who are involved with cardiac rehab.

Welcome, Dalane. Thank you for taking the time to join us. First, give us an overview of the trial. What were you trying to accomplish, and what did you accomplish?

Dalane Kitzman, MD: Thank you, Ileana. I'm honored to be here. REHAB-HF grew out of the observation that older persons hospitalized for acute decompensated heart failure have very poor outcomes, frequent rehospitalizations, poor quality of life, and high death rates, and a range of interventions aimed at improving their outcomes have been neutral. This suggested to us that something has been overlooked, and we hypothesized that perhaps it was severe physical dysfunction. Indeed, our preliminary data suggested that these patients had severe physical dysfunction and high rates of frailty in domains we had not anticipated — poor balance, strength, and mobility — in addition to the endurance deficits that we did anticipate. So we assembled a group of rehabilitation experts who helped us design a novel rehab intervention specifically to meet the needs of these patients. It addressed all four domains. We decided to tackle balance, strength, and mobility before we undertook endurance. That was important because geriatric studies have shown that if you take frail patients like this and subject them directly to endurance training, you have poor efficacy and frequently have injuries and falls.

The intervention was one-on-one, it was tailored, and it started in the hospital and transitioned to the outpatient setting for 12 weeks, three times per week. We also had a home-based assessment to make sure that the home was safe for home exercise. After 12 weeks we transitioned the patients to independent home exercise for another 3 months. We had excellent retention and adherence. We believe that may be because patients appreciated how frail they were and how much this was helping. Fully 83% patients reported continued home exercise at the end of the study.

Our primary outcome measure may be unfamiliar to cardiologists, but it's very familiar to geriatricians: the short physical performance battery (SPPB), which was developed by the National Institute on Aging (NIA), has been used in many large trials and is currently being inserted into clinical practice. The SPPB is a validated, brief, 5- to 6-minute measure of global physical function. At baseline, our patients had SPPB scores of 6.1. This is the average you see in a nursing home population and is on the threshold of not being able to take care of yourself.

At the end of the 3-month, outpatient intervention, those who were in the attention control group (usual care, could include physical therapy or cardiac/pulmonary rehab) had increased SPPB scores, just mildly to 7.1, confirming our hypothesis. These patients have long-term disability and don't get better on their own. Those randomly assigned to the rehab intervention group saw a large, statistically significant 1.5-unit improvement in SPPB, threefold higher than that associated with a minimal clinically meaningful difference. We also found large improvements in the 6-minute walk distance: 34 meters more than those in the control group. At baseline, these patients only walked about 196 meters in 6 minutes, meaning that proportionally it was a large, 20% improvement. They had a large improvement in the Kansas City Cardiomyopathy Questionnaire (KCCQ) quality-of-life score, 7.1 units when 5 is significant, and also improvements in depression and frailty.

However, we didn't find any improvements in our secondary outcome of all-cause rehospitalization, heart failure hospitalization, or death. In fact, I need to point out that numerically, there were more deaths in the rehab group. This was not statistically significant and the number of events was small. But we believe that is a signal that needs to be explored further.

Piña: Were you powered for mortality?

Kitzman: We weren't powered for mortality. We knew going in that, in reality, this was a phase 2 clinical trial. Our senior leader, Chris O'Connor, repeatedly told us that with 349 patients, even though your pilot data may give you the confidence that you have a chance to get clinical outcomes, be prepared that if this trial is positive, we will need to have a phase 3 trial. Of course, Chris was right.

Piña: I would strongly support that because I think there's more here than meets the eye. I have several observations. First, everyone thinks that rehab in the hospital is just getting the patient out of bed and pushing them down the hallway, making them walk. This is something I have argued against for a long time. Patients need to go to a skilled nursing facility. Looking at our numbers in heart failure, the number of patients being sent to skilled nursing facilities rather than home after an acute heart failure hospitalization has tripled, which is rather amazing. You would think that the skilled nursing facility would follow the same exercise practices that you used in the study. But that's not the case.

I have to compliment you on your high adherence rate. You and I agonized over our low adherence in the HF-ACTION trial. What do you think was so significantly different from the HF-ACTION population that you had such wonderful adherence?

Kitzman: We were gratified by that, and of course, the Achilles heel of exercise interventions was always adherence. I think there were three factors that influenced adherence. The first was that the home environment assessment wasn't just looking around at the home; it also took into account the community and the neighborhood. We met with important social support persons so we engaged with family, friends, and neighbors.

Piña: Did the investigator do that?

Kitzman: The interventionalists went out to the homes. As you remember in HF-ACTION, we did go to the home but it was just to set up equipment. This was a full-fledged assessment with what Amy Pastva calls the "built environment." It included everything that could influence the patient's behavior, willingness, and safety. So right from the beginning, we encouraged them and taught them how to do home exercise safely twice a week. Then, during the 12 weeks of the outpatient session, we reinforced this. We asked them how they were doing at home, and we helped them overcome barriers so that by the time they finished their 12 weeks and went into their 3 months of independent exercise, they were well prepared.

We also included something we called a Patient Engagement Committee. This was the brainchild of Pam Duncan. Every 2 weeks, we got together all of the interventionists, the study coordinators, and as many of the principal investigators who could join, and we did troubleshooting on each and every patient. We tracked them, and we shared success stories and challenges.

The third factor was that our patients reliably told us how big a difference this made to them. Some who were barely able to get up were later able to resume what was normal activity for them. They valued this, and they wanted to be able to continue it. I think those were three key ingredients that helped us get such a good retention.

Piña: Your centers were at Wake Forest and in Philadelphia?

Kitzman: Duke University was the second hub and Thomas Jefferson was our third hub. Similar to HF-ACTION, we also had four satellites — two associated with Duke in Durham, North Carolina, and two associated with Jefferson in Philadelphia. That means that four out of seven sites were community hospitals in small to medium-sized towns. But I believe that helps boost our generalizability and the ability — if we confirm this in a phase 3 trial — to roll it out and make it a really practical intervention.

Piña: Do you think Medicare will reimburse for this? Your population was primarily those with HFpEF; your mean ejection fraction was in the 50% range. So it's not the low-EF population. Do you think Medicare will pay?

Kitzman: We don't think so, and that is not our goal. About 53% did have HFpEF. As you know, in the HF-ACTION trial with 2133 patients and a positive signal for events, you and I and other HF-ACTION investigators had to campaign for 5 years and meet repeatedly with the Centers for Medicare & Medicaid Services, and provide numerous additional analyses. We don't have that for REHAB-HF.

As encouraged as we are with these results, we believe that in order to truly, convincingly change clinical care guidelines, and certainly reimbursement guidelines policy for what is a pretty expensive and intensive intervention, we're going to need clinical events data. We can get that, and here's why. I have been authorized by Chris O'Connor to announce to you the results of our subgroup analysis on EF. As you know, The New England Journal of Medicine does not publish subgroup analyses of secondary outcomes. But in press, and soon to be released electronically in JACC: Heart Failure, our subgroup analysis shows the following surprising, encouraging trends: The HFpEF patients had substantially more severe physical dysfunction, frailty, quality-of-life problems, and depression, and when exposed to the intervention, they appeared to have consistently greater benefit, sometimes up to 50% more benefit in SPPB or 6-minute walk. When we looked at clinical events, for hospitalization there was a not-even-move-the-needle 1% lower hospitalization rate in patients with HF with reduced EF (HFrEF) but a 17% lower rate in patients with HFpEF. We've got to be cautious here because we were not powered for these.

What about death? There was a 35% lower death rate in patients with HFpEF. It was 250% higher in those with HFrEF. The interaction P value was significant for death. For most of the others, the interaction P value was not significant, although it got pretty close for some. So, overall we saw very encouraging trends. But before we rush out and take all hospitalized patients with HFpEF and say let's do this, I have to caution that this was an exploratory analysis; it only included 92 patients with HFpEF who were exposed to the rehab intervention. So as encouraging as these results are, we don't believe that we can claim that this is ready for clinical practice. But the investigators are working hard to plan and seek funding for a subsequent phase 3 trial. We hope it will be positive, and we can bring this into routine clinical practice the right way with definitive evidence and with compelling evidence that would compel payers to coverage.

Piña: The NIA should be quite interested in this, as they funded the REHAB-HF study for you.

Kitzman: A shout-out to the NIA; they were visionary. The National Heart, Lung, and Blood Institute (NHLBI) was also supportive. Again, a quick shout-out to Jerry Fleg at NHLBI and to many others who supported HF-ACTION. But the NIA really embraced REHAB-HF and saw that this was well aligned with their mission. We hope the NIA will be enthusiastic about a subsequent trial but we're just in the planning stages.

Piña: Just to remind our audience, very often the cause of mortality for patients with HFpEF is one of their comorbidities, rather than cardiac or sudden death. And then if you want to tie that to the inability to care for yourself because you can't get out of a chair, the 6-minute walk distance here should scare the bejeepers out of everyone. It's one of the lowest I have seen. We know that anything below 300 meters is severely impaired, and yours were less than 200. Yet you had high pro B-type natriuretic peptides (proBNPs). So the patients had been pretty sick, with higher proBNPs.

To the clinicians, when you're sending these patients home, at least get them physical therapy in the home and talk to the families about doing it together. I know it's very hard to get patients to do it, but dangling the carrot — telling them they may not need long-term care if we can get them moving and they can take care of themselves — that should be a huge carrot to promote independence. A lot of things are linked here.

I can't thank you enough for having done this trial and talking to us about it. It's revealing for all of us. We'll look forward to Dr O'Connor's JACC: Heart Failure paper, and we look forward to reading the New England Journal paper. Congratulations, Dalane. I hope you'll come back after the JACC: Heart Failure article is published and we can have an even longer discussion.

To our audience, I hope we left you with some good pointers. This is very interesting and applicable to your own care. These patients are not going away. The numbers are going up, even as we speak.

Ileana L. Piña, MD, MPH, is a heart failure and cardiac transplantation expert. She serves as an advisor/consultant to the FDA's Center for Devices and Radiological Health and has been a volunteer for the American Heart Association since 1982. Originally from Havana, Cuba, she is passionate about enrolling more women and minorities in clinical trials. She also enjoys cooking and taking spin classes.

Follow theheart.org | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube