Massie and colleagues41 questioned whether this increase in ejection fractions was associated with an improvement in the symptoms and/or physical findings of congestive heart failure or a reduction in the number of hospitalizations for congestive heart failure. In their Department of Veterans Affairs cooperative study of amiodarone in congestive heart failure, 674 patients with Heart Association class II through class IV symptoms and ejection fractions of 40% or lower were treated with amiodarone or placebo for a median of 45 months in a randomized, double-blind, placebo-controlled protocol. Clinical assessments and radionuclide ejection fraction testing were performed at baseline and after 6, 12, and 24 months. Compared with the placebo group, ejection fractions increased more in the amiodarone group at each time point (8.1% ± 10.2% [mean ± SD] vs 2.6% ± 7.9% at 6 months; 8.0% ± 10.9% vs 2.7% ± 8.0% at 12 months; and 8.8% ± 10.1% vs 1.9% ± 9.4% after 24 months; P<.001 for all). However, this difference was not associated with greater clinical improvement, lesser diuretic requirements, or fewer hospitalizations for congestive heart failure (11.1% for amiodarone and 13.6% for placebo group; overall relative risk [RR] in the amiodarone group, 0.81 [95% confidence interval (CI), 0.56-1.10], P=.18). Of note is the trend toward a reduction in the combined end point of hospitalizations and cardiac deaths (RR, 0.82 [95% CI, 0.65-1.03), P=.08), which was significant in patients with nonischemic etiology (RR, 0.56 [95% CI, 0.36-0.87], P=.01) and absent in the ischemic group (RR, 0.95). It was thought that although amiodarone therapy resulted in a substantial increase in left ventricular ejection fraction in patients with congestive heart failure, this was not associated with clinical benefit in the population as a whole. The substantial reduction in the combined end point of cardiac death plus hospitalizations for congestive heart failure in the nonischemic group suggested possible benefit in these patients.