Exercise Training Improves Depressive Symptoms in Heart Failure: HF-ACTION

July 31, 2012

July 31, 2012 (Chicago, Illinois) — Measures of depression improved over several months in patients with heart failure who started a supervised exercise training program as part of a randomized trial, although less than a third of them had clinical depression at the outset [1].

That subgroup initially with depression showed an especially pronounced drop in depressive symptoms with exercise, but that didn't entirely account for the significant benefit seen in the population as a whole, say researchers, on the basis of their prospectively planned secondary analysis from the Heart Failure--A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) study.

Exercise is known to improve symptoms of depression, but "to our knowledge, this is the first randomized trial to show that exercise resulted in a small but statistically significant reduction in depressive symptoms in patients with heart failure," write the authors, led by Dr James A Blumenthal (Duke University Medical Center, Durham, NC), in the August 1, 2012 issue of the Journal of the American Medical Association.

Moreover, Blumenthal said for heartwire , the magnitude of benefit was tied to how much the patient exercised. "It didn't require marathon training. It looks like about 90 minutes a week, or three 30-minute sessions, was really sufficient to reduce depressive symptoms. I think that's good news for cardiac patients and really consistent with cardiac rehabilitation guidelines."

HF-ACTION had randomized >2300 stable patients with systolic NYHA 2–4 heart failure despite optimal therapy to follow an initially supervised aerobic exercise program on top of evidence-based standard care or to receive the evidence-based standard care on its own. The trial was conducted in North America and France.

Those assigned to the exercise program showed significantly improved measures of quality of life and overall health. In the primary analysis, they didn't show clinical improvement, but there were significant survival and hospitalization gains in some secondary adjusted analyses, as heartwire has previously reported.

In the current ancillary analysis, depressive symptoms were self-rated according to the Beck Depression Inventory II (BDI-II); 28% of the trial population had BDI-II scores of >14, considered to indicate clinically important depression.

Assignment to the exercise training program (n=1159) was associated with significantly lower BDI-II scores compared with usual care alone (n=1172) at both three and 12 months for the entire study population as well as for those with depression based on BDI-II scores. But the scores dropped over time in both randomization groups. "Everyone's symptoms did improve, but the exercise group improved to a greater extent," Blumenthal said.

Adjusted Mean BDI-II Scores at Three and 12 Months by Randomization Group, Overall Population and Subgroup Initially With Depression

Patient population 3 mo 12 mo
Exercise intervention 8.95 8.86
Usual care 9.70 9.54
p 0.002 0.01
Baseline BDI-II score >14    
Exercise intervention 16.66 15.85
Usual care 17.98 17.34
p 0.04 0.02

BDI-II=Beck Depression Inventory II

That there was a steeper reduction in depressive symptoms in the initially depressed subgroup than in the overall population suggests that the subgroup effect might account for all or most of the overall benefit. But that wasn't the case, according to Blumenthal. "We didn't observe a depression-severity-by-treatment-group interaction," which would have suggested different treatment effects for patients more vs less depressed at baseline.

Also, in multivariate analyses, assignment to the exercise group was associated with an 11% reduction in the composite end point of all-cause mortality or first all-cause hospitalization (p=0.03) and a 15% drop in heart-failure hospitalizations or death (p=0.03) over the median follow-up of 30 months. Those with more depression at baseline compared with less had significantly higher risks of both composite end points.

In their heart-failure patients, said Blumenthal, "I think it's very important for clinicians to monitor depressive symptoms, and if there's a worsening of symptoms, then it's important to know that those patients are particularly vulnerable to worse outcomes."

Blumenthal and his colleagues acknowledge that the reduction in depression seen in the exercise group at both three and 12 months was small in absolute terms. But "it was a robust finding in that the difference was maintained throughout the year, suggesting it wasn't simply a chance occurrence, that it was a real effect."

He proposes that the observed benefit would probably have been greater had there been fewer "crossovers" in the control group. "Almost half of the people who received [only] the evidence-based usual care indicated that they exercised several times a week over that one-year period, on their own. They weren't prohibited from exercising; it's part of the guidelines."

Also, about a fourth of the control group were on antidepressants, he observed, "and they received biweekly phone calls from the staff, which provided them with a level of social support [comparable to that in the intervention group] and a feeling of engagement in the study," he said. "My suspicion is that [otherwise] we would have seen a larger difference between the groups."

Blumenthal discloses that he received compensation for his work by a grant from the National Heart, Lung, and Blood Institute, which funded HF-ACTION. Disclosures for the coauthors are listed in the paper.