Heart failure classification based on resting ejection fraction does not display a unique exercise response pattern

Background: Heart failure with preserved (HFpEF), mildly reduced (HFmrEF) and reduced (HFrEF) ejection fraction (EF) remains a controversial categorization. Whether these three categories reflect a distinct pattern of exercise limitation in cardiopulmonary exercise testing (CPET) needs to be investigated. We aimed to analyze whether CPET variables differ between all heart failure categories (HF).

Methods: We analyzed CPET variables of stable HFpEF (n = 123), HFmrEF (n = 31), and HFrEF (n = 153; 74 patients with and 79 patients without left ventricular assist device, LVAD) patients. The association between HF and peak oxygen consumption (VO_2peak) was used as a primary outcome, while the association between HF, oxygen uptake efficiency slope (OUES), and increase of O₂ pulse (ΔO₂ pulse) were analyzed as secondary outcomes.

Results: VO_2peak displayed a consistent decline across all HF categories (19.8 ml ± 6.2/kg/min vs. 17.5 ± 7.9 ml/kg/min vs. 13.7 ± 4.0 ml/kg/min, p < 0.001). OUES only showed differences between HFpEF and HFrEF (1.8 ± 0.6 vs. 1.4 ± 0.5, p < 0.001) as well as HFmrEF and HFrEF (1.9 ± 0.9 vs. 1.4 ± 0.5, p = 0.004). ΔO₂ pulse differed between HFpEF and HFrEF (7.7 ± 3.5 ml/beat/kg*100 vs. 5.5 ± 3.0 ml/beat/kg*100, p < 0.001) as well as HFpEF and HFmrEF (7.7 ± 3.5 ml/beat/kg*100 vs. 6.3 ± 4.1 ml/beat/kg*100, p = 0.049). Outcome variables did not differ between HFrEF with and without LVAD support (VO_2peak: p = 0.364, OUES: p = 0.129, ΔO₂ pulse: p = 0.564).

Conclusions: HF did not display a distinct CPET profile. Thus, EF-based categorization does not entirely reflect exercise limitations. CPET variables could contribute to better characterize HF phenotypes.