REHAB-HF: Tailored Cardiac Rehab Promotes Physical Function After Hospital Stay

For frail, elderly patients with decompensated heart failure (HF), a hospital stay can mean further impairment and worsened quality of life. But data from the randomized REHAB-HF trial provide a ray of hope: a comprehensive, individually tailored cardiac rehabilitation program that starts in the hospital and continues to escalate after discharge can greatly improve physical functioning.

The trial’s lead investigator, however, cautioned TCTMD that this is a phase II study, with a lot left to learn before its strategy can be implemented on a wider basis.

“We were clearly not powered for clinical events, and it’s usually clinical events that are required to change clinical practice guidelines as well as reimbursement. This is not a trivial cost,” Dalane W. Kitzman, MD (Wake Forest University School of Medicine, Winston-Salem, NC), told TCTMD, referring to the program’s one-on-one rehab sessions.

On the one hand, he continued, “we had robust improvements. We’re proud of it.” On the other, ”we also have a signal that there were numerically more deaths in the rehab group. The number of events were small—it was not significant. But the investigators believe that this suggests that we have more work to be done.”

Kitzman presented the results this morning at the virtual American College of Cardiology (ACC) 2021 Scientific Session; they were simultaneously published in the New England Journal of Medicine.

Better Endurance, QoL

For REHAB-HF, 349 patients hospitalized with acute decompensated HF were randomly assigned to rehabilitation or usual care; seven hospitals participated, of which four were community-based, Kitzman said. The rehabilitation intervention spanned four functional domains: strength, balance, mobility, and endurance. It progressed through four prespecified functional levels within each domain, with the intensity and specific exercises tailored to each patient’s performance. The training began during or shortly after hospitalization and then continued for 36 outpatient sesions after discharge.

Mean age of the study participants was 72.7 years, 52% were women, and 49% were nonwhite (94% of whom were Black). For 35%, ischemic heart disease was the cause of HF, and 53% had preserved ejection fraction. One-quarter had previously been hospitalized for HF. At baseline, 97% were considered frail or prefrail according to Fried criteria, which encompasses weight loss, exhaustion, low physical activity, slowness, and weakness. Many suffered from urinary incontinence, falls, and depression. Overall, they had “severely imparied physical function, poor quality of life, and at least mild cognitive dysfunction,” Kitzman et al report in NEJM.

Excluding 12 patients who died, fully 82% of patients assigned to the rehab intervention completed the study, with a median of 24.3 outpatient sessions. Mean adherence to the exercise program was 67%; when accounting for sessions missed due to medical appointments and illness, it rose to 78%.

Our patients had dramatic improvements. There are some patients that were barely able to walk who returned to normal activity. Dalane W. Kitzman

“We had good safety: three serious adverse events, possibly related to the intervention but all were self-limited. We had very good progression across the 36 sessions,” Kitzman said, with performance levels increasing over time for all of the four domains. First came strength, balance, and mobility, and only after that—in the interest of safety—came endurance.

“Our goal, [which] we achieved, was to increase these and address these before addressing endurance, which was our final goal,” he explained. Endurance, quantified as walking duration, doubled over the course of the program, from a mean of 10.7 minutes to 22.0 minutes between the first and last sessions.

The primary outcome was based on the Short Physical Performance Battery (SPPB) at 3 months. Baseline scores were 6.0 ± 2.8 in the rehab group and 6.1 ± 2.6 in the control group. After adjustment for these and other baseline characteristics, the least-squares mean for the physical function measure had improved to 8.3 ± 0.2 with rehabilitation and 6.9 ± 0.2 with usual care, amounting to a mean between-group difference of 1.4 (95% CI 0.9-2.0; P < 0.001).

For controls, the more-modest gain in function on SPPB indicates that after hospitalization, these patients “had long-term disability that does not improve on its own,” Kitzman said. “However, those randomized to the rehab intervention group had a strongly statistically significant and large 1.5-unit improvement; this effect size was three times the clinically meaningful change. This affect of the intervention on the SPPB was relatively uniform across a wide range of prespecified subgroups, including women and those with preserved ejection fraction.”

Additionally, quality of life as measured by the Kansas City Cardiomyopathy Questionnaire (KCCQ) rose from an overall mean score of 40 at baseline to 69 at 3 months in the rehab patients—it increased from 42 to 62 in controls. This translated into a 7.1-point difference in KCCQ score between the two groups.

At 6-month follow-up, hospitalization for any cause occurred in 1.18% of the rehab group and 1.28% of controls. Death rates were 0.13% and 0.10%, respectively.

There was a piece of good news at 6 months, when fully “83% of intervention participants reported regular home exercise, which suggests behavioral change, a prerequisite for durable behavioral change,” Kitzman observed.

How It Was Done

Vera Bittner, MD (The University of Alabama at Birmingham), discussant for the late-breaking presentation, described REHAB-HF as an “exciting trial to hear about” and questioned how the researchers came up with the specific training regimen.

With care, Kitzman responded, taking into account mobility and strength, both at baseline and throughout the trial. “As you might guess, if a patient has significant impairments in balance, which they did, . . . it would be hazardous to subject them to traditional endurance training such as walking-based treadmill or even bicycle,” Kitzman noted. “So we addressed that first. Mobility means the capability to locomotor movement. . . . Strength as well: if you’re not able to stand with confidence, and you’re not able to walk on a treadmill.”

Adherence, he added, was a pleasant surprise, and may have stemmed from the fact that the home environment was assessed at the outset, as well as the impact of regular meetings of the Sustaining Patient Engagement Committee, or SPEC, where the investigators and others with firsthand involvement in the rehab intervention met biweekly to brainstorm ideas and discuss individual patients’ progress.

“But I think there’s another aspect, and that is we gave good attention to social determinants of health—we engaged the patient’s family, their caregivers, all significant others,” Kitzman stressed, adding, “Our patients had dramatic improvements. There are some patients that were barely able to walk who returned to normal activity. We believe that they appreciated what the intervention was doing for them and wanted to continue that.”

In a press briefing after the ACC presentation, discussant Eileen Handberg, PhD (UF Health, Gainesville, FL), observed that “this work is extremely important, because the measures of outcomes that they evaluated are the ones that matter to patients the most—this is about being able to get out of a chair without assistance and not falling, walking farther, and feeling better,” as opposed to the more-traditional metric of treadmill exercise, which she said may be beyond these patients’ capacities.

But What Comes Next?

Cardiac rehab for HF patients is hardly a new idea, Stefan D. Anker, MD, PhD (Charité – Universitätsmedizin Berlin, Germany), and Andrew J.S. Coats, DM (University of Warwick, Coventry, England), point out in an accompanying editorial. Yet it’s not a single entity—the specifics matter.

“Exercise is not like a drug. There is no standard dose, formulation, or frequency of administration, and it has many variations, all of which depend crucially on patient motivation to elicit beneficial effects,” they say. “To reduce variability, trials in heart failure frequently exclude elderly patients with multiple coexisting conditions, unstable clinical status, or frailty. Thus, the patients with the greatest need are rarely studied.”

This is not the case with REHAB-HF, the editorialists comment. “We congratulate the authors for having developed a program that is pragmatic, widely applicable, and able to be duplicated relatively easily in practice.”

They also cite the quality of life gains on KCCQ that came with rehab. “We hypothesize that part of the benefit may be due to better physical performance and that part may be due to improvements in psychosocial factors and mood,” Anker and Coats note. “It appears that exercise also resulted in patients becoming happier—or at least less depressed.”

Taken together, the REHAB-HF results provide a “compelling argument” that such programs should be adopted, they conclude.

Kitzman, in the press conference, announced that a subgroup analysis of clinical events in REHAB-HF is now in press. These show “quite an interesting trend, considerable disparity, where patients with heart failure with preserved ejection fraction (HFpEF) had the appearance of a uniformly better response. The interaction P values were not significant” for the primary endpoint, he acknowledged, “but for clinical events they’re actually close.” For death, the interaction between cardiac rehab and ejection fraction was significant, he reported, with a 35% reduction seen for HFpEF and a 250% increase for HF patients with reduced ejection fraction.

The fact that there were only 92 patients with HFpEF assigned to rehab is something to keep in mind “before we rush and say this is good news for [this subgroup],” Kitzman urged. These analyses, however, have inspired the researchers to work towards a HFpEF-focused trial.