It is well known that peak oxygen consumption and heart rate (HR) recovery after exercise obtained from the cardiopulmonary exercise test are prognostic parameters in patients with chronic heart failure (CHF). However, it is unclear whether exercise-induced parameters obtained from the routine exercise stress test predict mortality in patients with CHF. We studied 136 patients (93 males/43 females) with CHF. All patients underwent symptom-limited exercise stress testing. Exercise parameters included exercise duration, exercise-induced HR and systolic blood pressure (SBP), and metabolic equivalents (METs). During the follow-up period (mean 6.2 years), 34 patients died. Survival rates at the 3rd and 5th years were 90% and 83%, respectively. Body mass index was significantly smaller in the nonsurvival group than in the survival group (P < 0.01). The incidence of patients with New York Heart Association III class was higher in the nonsurvival group than in the survival group (P < 0.05). In univariate analysis, predictors of mortality included peak HR and SBP, increases in HR and SBP during exercise, HR and SBP at the 1st minute after exercise, HR at the 3rd minute after exercise, and METs. The use of beta-adrenergic blocking agents was not associated with prognosis. In Cox hazard model analysis, the increase in SBP (P < 0.002), HR at the 3rd minute after exercise (P < 0.05), and METs (P < 0.05) were independent predictors of mortality. SBP response to exercise, HR recovery after exercise, and METs obtained from the routine exercise test predicted mortality in patients with CHF irrespective of the use of beta-adrenergic blocking agents.