The relationship between diuretic use or change in diuretic use and outcomes in chronic heart failure (HF) remains poorly defined. We evaluated the association between diuretic use and changes in health status, exercise capacity, and clinical events in a large randomized trial of subjects with HF.

HF-ACTION randomized 2,331 outpatients with HF and ejection fraction ≤35% to aerobic exercise training versus usual care. We grouped patients according to loop diuretic use from baseline through 6 months: continued use, never use, initiated, and discontinued. The association between diuretic use and changes in health status, exercise capacity, and clinical outcomes (all-cause mortality/hospitalization, cardiovascular mortality, and HF hospitalization) through 12 months was assessed using Cox proportional hazards models and generalized linear regression models, respectively.

A total of 2,004 (86%) patients had complete data on diuretic use. There was no association between diuretic status and Kansas City Cardiomyopathy Questionnaire, 6-minute walk distance, or peak Vo₂ in adjusted analyses (all P > .05). A dose increase was associated with decrease in 6-minute walk distance (−4.25 m, SE 1.12 m, P < .001) and change in Kansas City Cardiomyopathy Questionnaire overall score (−0.56 m, SE 0.24 m, P = .02). There were no between-group differences for all-cause death or hospitalization comparing continuous use versus never use (adjusted HR 0.91; 95% CI 0.72-1.15; P = .432).

The initiation or discontinuation of diuretics over a 6-month time frame was not associated with a difference in mortality, hospitalizations, exercise, or health status outcomes, but a dose increase in HF patients was associated with worse exercise and health status outcomes.