Optimizing exercise testing‐based risk stratification to predict poor prognosis after acute heart failure

AIMS: The timely selection of severe heart failure (HF) patients for cardiac transplantation and advanced HF therapy is challenging. Peak oxygen consumption (VO(2)) values obtained by the cardiopulmonary exercise testing are used to determine the transplant recipient list. This study reassessed the prognostic predictability of peak VO(2) and compared it with the Heart Failure Survival Score (HFSS) in the modern optimized guideline‐directed medical therapy (GDMT) era. METHODS AND RESULTS: We retrospectively selected 377 acute HF patients discharged from the hospital. The primary outcome was a composite of all‐cause mortality, or urgent cardiac transplantation. We divided these patients into the more GDMT (two or more types of GDMT) and less GDMT groups (fewer than two types of GDMT) and compared the performance of their peak VO(2) and HFSS in predicting primary outcomes. The median follow‐up period was 3.3 years. The primary outcome occurred in 57 participants. Peak VO(2) outperformed HFSS when predicting 1 year (0.81 vs. 0.61; P = 0.017) and 2 year (0.78 vs. 0.58; P < 0.001) major outcomes. The cutoff peak VO(2) for predicting a 20% risk of a major outcome within 2 years was 10.2 (11.8–7.0) for the total cohort. Multivariate Cox regression analyses showed that peak VO(2), sodium, previous implantable cardioverter defibrillator (ICD) implantation, and estimated glomerular filtration rate were significant predictors of major outcomes. CONCLUSIONS: Optimizing the cutoff value of peak VO(2) is required in the current GDMT era for advanced HF therapy. Other clinical factors such as ICD use, hyponatraemia, and chronic kidney disease could also be used to predict poor prognosis. The improvement of resource allocation and patient outcomes could be achieved by careful selection of appropriate patients for advanced HF therapies, such as cardiac transplantation.