Exercise training has recently become an accepted therapeutic modality in chronic heart failure after myocardial infarction. Because the therapeutic mechanism behind it is controversial and not well understood, we analyzed the influence of exercise training on blood viscosity. Twenty-five patients with chronic heart failure (ejection fraction < 40%) after myocardial infarction were randomly assigned to either an 8-week intensive exercise program at a residential rehabilitation center or 8 weeks of sedentary life at home. Exercise consisted of two 1-hour walking sessions per day and four intensive bicycle ergometer training sessions of 40 minutes at 70% to 80% peak exercise capacity per week. Whole blood viscosity, viscosity at standardized hematocrit of 45% (P45) at high and low shear rates, and plasma viscosity were measured in a Couette-type viscometer before, during, and at the end of the study period. Exercise training, which significantly increased maximal cardiac output and oxygen uptake, did not change plasma viscosity, whole blood viscosity, and P45 significantly. Sedentary controls, however, had a higher whole blood viscosity and P45 after 8 weeks. No statistical difference was found, however, between the two groups. We conclude that blood rheology remains unaffected by exercise training in patients with chronic heart failure. The improvement of blood viscosity remains an interesting therapeutic option for the symptoms of these patients, which must be achieved by methods other than exercise training.