Prognostic usefulness of the functional aerobic reserve in patients with heart failure

Background: Peak oxygen consumption derived from cardiopulmonary exercise (CPX) testing provides important prognostic information in patients with heart failure (HF). The oxygen consumption at the ventilatory threshold (VT) has also been shown to be prognostic. However, the VT cannot always be detected in patients with HF. Other variables such as the difference between peak oxygen consumption and oxygen consumption at the VT (termed the functional aerobic reserve [FAR]) may also provide prognostic information. The purpose of this study was to determine the prognostic value of an undetectable VT and FAR. Methods: Eight hundred seventy-four patients with chronic, systolic HF (70% male, age 54 ± 14 years, ejection fraction 29% ± 12%) underwent CPX and were tracked for 2 years for major events (death, transplant, and left ventricular assist device implantation). Results: Patients were divided into 2 subgroups based on whether VT could be detected or not. There were 141 major events during the 2-year follow-up. Kaplan-Meier analysis for the 2 VT subgroups demonstrated worse prognoses for patients with a nondetectable VT versus those with a detectable VT (P < .001). Based on receiver operating characteristic curve analysis (FAR = 0 mlO2 kg-1 min-1 for patients with undetectable VT), the optimal cut-point for FAR was ≤/>3 mlO2 kg-1 min-1 (sensitivity/specificity 69%/60%). Cox regression analysis identified the FAR as a significant univariate predictor of risk and was retained in multivariate analysis. Conclusion: In conclusion, these data reveal that an undetectable VT and the FAR during CPX testing can provide useful prognostic information in patients with HF.