Philadelphia bureau chief Mitchel L. Zoler contributed to this report.
NEW ORLEANS — In the largest study of exercise training as part of the management of heart failure to date, a guided exercise program was safe and modestly effective, but researchers acknowledged that patients found it hard to keep up the routine.
The safety of exercise training in heart failure patients, outside of a supervised environment, has been a concern. This study proved benefits could be obtained without excess risk, said Dr. Christopher M. O'Connor, presenting results of the Heart Failure and A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) study at the annual scientific sessions of the American Heart Association.
“Over 30 randomized trials have shown increased exercise capacity and possibly improved survival with exercise training, but these were largely single-center studies that were underpowered or lacked adequate controls and produced limited data on safety,” he noted at a press conference.
HF-ACTION, a randomized, phase III trial sponsored by the National Heart, Lung, and Blood Institute, followed 2,331 heart failure patients at 82 international sites for an average of 2.5 years. The relatively young population, median age 59 years, had an average left ventricular ejection fraction (LVEF) of 25%, indicating moderate HF. History of coronary occlusion and prior myocardial infarction was common.
Patients were randomized to an exercise training program aimed at increasing workout intensity and duration or to usual care, where they were encouraged to exercise, based on the American College of Cardiology/AHA recommendations of 30 minutes of moderate exercise most days of the week. Both groups received optimized medical treatment, patient education, and follow-up telephone calls.
The exercise training followed the cardiac rehabilitation model. Patients were prescribed a multistage, guided workout of 36 supervised training sessions of 30 minutes of exercise three times a week. At the 18th session, patients received a treadmill or exercise bicycle for home use, learned how to monitor their heart rate during exercise, and were encouraged to complete five weekly sessions of similar intensity and 40 minutes' duration.
At 4–6 weeks, patients were exercising a median of 95 minutes per week, short of the goal of 120 minutes. This was consistent for the first year and then diminished further.
After 3 years, people were exercising for about 50 minutes. We had wanted them to exercise for 120 minutes. So adherence is extremely difficult,” reported Dr. O'Connor, professor of medicine and director of the heart center at Duke University Medical Center, Durham, N.C.
Exercise training was not associated with a significant reduction in the primary end point, all-cause mortality and hospitalization, or in secondary composite end points: cardiovascular (CV) mortality plus CV, and CV mortality plus HF, he reported.
But improvements in outcomes emerged in the prespecified adjusted analysis that accounted for additional key prognostic variables related to heart failure outcomes. These included exercise duration, LVEF, Beck Depression Inventory score, and history of atrial fibrillation/flutter.
In the adjusted analysis, the primary end point was significantly reduced by 11%, and CV mortality plus heart failure hospitalization was significantly reduced by 15%. The reduction in CV mortality plus CV hospitalization remained a nonsignificant 8%.
“The prespecified adjusted analysis is a fair analysis of these data and is probably closest to the truth,” Dr. O'Connor maintained. “Prognostic factors are most important.”
The study found no excess risk for CV events or fractures with intensive exercise. “Perhaps the most important finding is that exercise training of this degree was safe,” Dr. O'Connor added.
Health status was assessed by the Kansas City Cardiomyopathy Questionnaire (KCCQ). In the exercise group, the average KCCQ score rose by 5 points after the first 3 months of training, a statistically and clinically significant increase that continued for the remainder of the study. In the usual-care group, the average KCCQ score rose by 3 points, which was also maintained. The 3-point increase was statistically significant but fell short of clinical significance. The increase in the usual-care group was significantly less than in the exercise group, reported Kathryn E. Flynn, Ph.D., of the Center for Clinical and Genetic Economics at Duke University, Durham, N.C.
The improvements in outcomes were obtained in a setting of excellent overall cardiac care, as more than 90% of the patients received evidence-based medical therapy for their disease. “We achieved an 11% to 15% meaningful reduction in clinical end points above that, with a safe intervention,” Dr. O'Connor emphasized.