How do different indicators of cardiac pump function impact upon the long-term prognosis of patients with chronic heart failure?

Background The prognosis of patients with mild-moderate chronic heart failure (CHF) over a long-term follow-up period is more difficult to predict than for patients with more severe CHF in the short term. This study assessed the prognostic value of various indicators of cardiac pump function to gain insight into how different aspects of organ function impact upon prognosis. Methods Unselected, consecutive patients with CHF (n = 219, 166 men, mean [±SD] age 56 ± 13 years) who underwent symptom limited cardiopulmonary treadmill exercise testing with noninvasive estimation of cardiac output using carbon dioxide rebreathing techniques were followed up for a median period of 8.6 ± 1.0 years in survivors. Cardiac power output (CPO) was calculated from the product of cardiac output and mean arterial pressure and cardiac reserve was estimated by subtracting resting from peak exercise CPO or cardiac output (CO). Results All-cause mortality was 36% (78 deaths). Survivors had a significantly greater cardiac pumping reserve with the greatest difference seen in CPO reserve (+57%) and CO reserve (+49%) (both P < .001). Although various direct and indirect indicators of cardiac function were predictive of outcome on univariate analyses, multivariate analysis using the Cox proportional hazards model identified CO reserve to be the independent variable predictive of all-cause mortality, with a hazard ratio (95% CI) of 0.682 (0.612-0.757, P < .001) for each L/min increase in cardiac output reserve. Survival at 10 years in patients with tertiles of good, moderate, or poor cardiac output reserve was 89%, 63%, and 36.1%, respectively (P < .001). Conclusion In this long-term follow-up study involving a cohort of unselected ambulatory patients with mild-moderate CHF, cardiac pumping reserve measured noninvasively by cardiopulmonary exercise testing was found to be the strongest independent predictor of prognosis.