Exercise training in chronic heart failure: is it really safe?

There is little doubt that tailored moderate exercise training programmes using aerobic exercises and calistenics can result in improved exercise capacity, ventilatory efficiency, symptomology and quality of life, and reduced neuroendocrine activation in patients with chronic heart failure. The CHANGE (Chronic Heart Failure and Graded Exercise) study, as reported in this issue has once again supported this concept, which has been demonstrated from other research and from experience. Because there were no adverse cardiac events in this study, exercise training has been deemed safe in chronic heart failure patients categorized as New York Heart Association class II and III. The relatively large sample size of 80 patients was considered unique, and it was hoped that this would provide stronger evidence for the safety of exercise training in chronic heart failure than was previously available. Should we agree that the evidence from the CHANGE study is sufficient to conclude that exercise training in most chronic heart failure patients is safe? In terms of the question, ‘Is exercise training in chronic heart failure patients really safe?’ we should consider only randomized controlled trials that involve patients with defined congestive heart failure, meaning chronically low cardiac output with compensatory sodium and water retention, neurohumoral activation, and thoracic congestion. Additionally, according to standard aerobic training methods, only studies in which large muscle groups were trained should be considered. In some studies, patients with left ventricular dysfunction, with and without signs of heart failure, were lumped together in one pot, and thus not properly categorized. Based on these arguments, which studies have been conducted which fit the above criteria? Including the CHANGE study and the not-yet-published EXERT study (McKelvie), there are 10 randomized controlled trials involving 431 chronic heart failure patients; this is still a relatively small number, statistically considered. The mean age of patients was 56 years, the majority under 60. This relatively young age does not take into consideration the fact that the incidence