The Prognostic Value and Predictors of Responding to Exercise Training in Heart Failure Patients

Abstract

Improvement of physical fitness by exercise training during cardiac rehabilitation (CR) is an important target for secondary prevention in heart failure (HF). However, not all HF patients improve physical fitness following CR, i.e. non-responders, which may impact the risk of future morbidity and mortality. PURPOSE: We compared the risk of death and unplanned hospitalization in HF responders versus non-responders to CR. Secondly, we identified predictors of improvement in physical fitness following CR in HF patients. METHODS: HF patients performed an incremental cycle exercise test combined with gas analysis to assess peak oxygen consumption (VO2peak), before and after completion of 8-, 12-, or 26-week CR. Responders showed ≥6% improvement in VO2peak, which is chosen to compensate for inter-test variability, whereas non-responders showed a decrease, no change, or improve in VO2peak <6%. Patient characteristics (age, sex, BMI, smoking), co-morbidities (hypertension, hypercholesterolemia, diabetes mellitus, COPD), and HF features (HF etiology, ejection fraction, NYHA class, use of medical devices and medication) were collected from electronic patient files. Main end point was all-cause mortality and secondary end point all-cause mortality or unplanned hospitalization. RESULTS: 45% of our study population (n=155) responded to exercise training. After adjustment for age, sex, BMI, smoking and baseline VO2peak, responders had a hazard ratio (HR) for all-cause mortality of 0.41 (95% CI 0.12; 1.35) and a HR for all-cause mortality or unplanned hospitalization of 0.54 (95% CI 0.29; 0.98) compared to non-responders. The prediction model included lower age, BMI and baseline VO2peak as significant predictors for exercise response. The performance of the model, assessed by ROC analysis, had an AUC of 0.68 (95% CI 0.60; 0.77). CONCLUSIONS: The finding that a minority of HF patients responded to CR is of clinical relevance since this group showed a 46% reduction of the risk for death and/or unplanned hospitalization compared with non-responders. HF patients who are older, have a higher BMI and a higher baseline VO2peak have a larger chance to become a non-responder. These observations emphasize the importance to improve CR based exercise interventions in HF patients to improve survival and prevent hospitalization.