Clinical, hemodynamic, and cardiopulmonary exercise test determinants of survival in patients referred for evaluation of heart failure

Background: Accurate prognosis in chronic heart failure has become increasingly important in assessing the efficacy of treatment and in appropriately allocating scarce resources for transplantation. Previous studies of severe heart failure have been limited by short follow-up periods and few deaths. Objective: To establish clinical, hemodynamic, and cardiopulmonary exercise test determinants of survival in patients with heart failure. Design: Retrospective study. Setting: Hospital-based outpatient heart failure clinic. Participants: 644 patients referred for evaluation of heart failure over 10 years. Measurements: Age, cause of heart failure, body surface area, cardiac index, ejection fraction, pulmonary capillary wedge pressure, left ventricular dimensions, watts achieved during exercise, heart rate, maximum systolic blood pressure, a nd oxygen uptake ((V) over dot O-2) at the ventilatory threshold and at peak exercise were measured at baseline. Univariate and multivariate analyses were done for clinical, hemodynamic, and exercise test predictors of death. A Cox hazards model was developed for time of death. Results: During a mean follow-up period of 4 years, 187 patients (29%) died and 101 underwent transplantation. Actuarial 1-year and 5-year survival rates were 90.5% and 73.4%, respectively. Resting systolic blood pressure, watts achieved, peak (V) over dot O-2, (V) over dot O-2 at the ventilatory threshold, and peak heart rate were greater among survivors than among nonsurvivors. Cause of heart failure (coronary artery disease or cardiomyopathy) was a strong determinant of death (relative risk for coronary artery disease, 1.73; P < 0.01). By multivariate analysis, only peak (V) over dot O-2 was a significant predictor of death. Stratification of peak (V) over dot O-2 above and below 12, 14, and 16 mL/kg per minute demonstrated significant differences in risk for death, but each cut-point predicted risk to a similar degree. Conclusions: Peak (V) over dot O-2 outperforms clinical variables, right-heart catheterization data, exercise time, and other exercise test variables in predicting outcome in severe chronic heart failure. Direct measurement of (V) over dot O-2 should be included when clinical or surgical decisions are being made in patients referred for evaluation of heart failure or those considered for transplantation.