Cardiorespiratory Fitness and Risk of Heart Failure with Preserved Ejection Fraction.

Document Type


Publication Date


Publication Title

European journal of heart failure : journal of the Working Group on Heart Failure of the European Society of Cardiology


AIMS: Preventive strategies for heart failure (HF) with preserved ejection fraction (HFpEF) include pharmacotherapies and lifestyle modifications. However, the association between cardiorespiratory fitness (CRF) assessed objectively by a standardized exercise treadmill test (ETT) and the risk of HFpEF has not been evaluated. Thus, we evaluated the association between CRF and HFpEF incidence.

METHODS AND RESULTS: We assessed CRF in US Veterans (624,551 men; mean age 61.2 ± 9.7 years and 43,179 women; mean age 55.0±8.9 years) by a standardized ETT performed between 1999-2020 across US Veterans Affairs Medical Centers. All had no evidence of HF or myocardial infarction prior to completion of the ETT. We assigned participants to one of five age-and-gender-specific CRF categories (quintiles) based on peak metabolic equivalents (METs) achieved during the ETT and four categories based on CRF changes in those with two ETT evaluations (n=139,434) ≥1.0 year apart. During the median follow-up of 10.1 years (IQR 6.0-14.3 years), providing 6,879,229 person-years, there were 16,493 HFpEF events with an average annual rate of 2.4 events per 1,000 person-years. The adjusted risk of HFpEF decreased across CRF categories as CRF increased, independent of comorbidities. For fit individuals (≥10.5 METs) the hazard ratio (HR) was 0.48 (95% CI 0.46-0.51) compared with least fit (≤ 4.9 METs; referent). Being unfit carried the highest risk (HR, 2.88; 95% CI, 2.67-3.11) of any other comorbidity. The risk of unfit individuals who became fit was 37% lower (HR 0.63, 95% CI 0.57-0.71), compared to those who remained unfit.

CONCLUSIONS AND RELEVANCE: Higher CRF levels are independently associated with lower HRpEF in a dose-response manner. Changes in CRF reflected proportional changes in HFpEF risk, suggesting that the HFpEF risk was modulated by CRF. This article is protected by copyright. All rights reserved.





PubMed ID