Background: Exercise intolerance is considered a cardinal symptom in the heart failure (HF) syndrome, reflected by reduced peak oxygen consumption (peak VO₂) in cardiopulmonary exercise testing (CPET). Beyond peak VO₂ as established prognostic marker, VE/VCO₂-slope and O₂ pulse have been reported to predict outcome in HF, mirroring distinct parts of the oxygen chain. However, data comparing the predictive value of the measures in chronic HF phenotypes is still limited.

Methods: Data from the MyoVasc study (NCT04064450), a prospective cohort study on HF, were analyzed. Participants underwent deep clinical phenotyping at baseline, including CPET performed on a cycle ergometer using the MasterScreen CPX system (Vyaire Medical, Höchberg, Germany) with a modified WHO exercise protocol. Peak VO₂ and respiratory exchange ratio (RER) were defined as highest 30s-averages within the last 30 seconds of exercise, VE/VCO₂-slope was assessed throughout exercise. HF phenotypes were defined according to the 2021 universal definition and classification of HF. The study end point was worsening of HF, defined as the combination of cardiac death and HF hospitalization. The association of single CPET parameters as independent variables (i.e., peak VO₂, VE/VECO₂-slope, and O₂-pulse) with worsening of HF was investigated by subdistribution hazard models with adjustment for potential effect mediators and confounders.

Results: In total, the analysis sample comprised N=926 individuals with stage C or D HF and available CPET data, of whom N=369 were classified as HF with preserved EF (HFpEF), N=228 as HF with mid-range EF (HFmrEF), and N=177 as HF with reduced EF (HFrEF). Median age was 69 years (interquartile range (IQR), 61; 75), 30.5% (N=282) were women. The overall sample had a median RER of 1.02 (IQR 0.91; 1.13), peak VO₂ of 15.2 (IQR 12.5; 19.1) ml/min/kg, O₂-pulse of 14 (IQR 11; 16) ml/beat, and VE/VCO₂-slope of 33.8 (IQR 30.4/38.7), respectively. Across the spectrum of chronic HF, peak VO₂ (Hazard ratio [HR] per one standard deviation [SD] decrease 1.73, 95% confidence interval [95%CI] 1.43;2.08, p<0.0001), O₂-pulse (HR_{per -1SD} 1.23 [1.01; 1.41], p=0.07), and VE/VCO₂-slope (HR_{per 1SD} 1.36 [1.21; 1.52], p<0.0001) were identified as strong predictors for worsening of HF under adjustment for age, sex, and CPET protocol. After additional adjustment for clinical profile, the relationship of peak VO₂ and VE/VCO₂-slope was confirmed, but no longer detectable for O₂-pulse. O₂-Pulse predicted worsening of HF only in HFpEF (HR 1.41 [1.02; 1.95], p=0.04), but not in phenotypes with reduced EF. For VE/VCO₂ slope, a predictive value was detected for both individuals with HFpEF (HR 1.3 [1.03; 1.62], p=0.024) and HFrEF (HR 1.33 [1.06; 1.67], p=0.013), but not HFmrEF (HR 1.04 [0.81; 1.33], p=0.77). Of clinical importance, the predictive value of peak VO₂ for worsening of HF was strongest in subjects with HFpEF (HR 2.29 [1.61; 3.25], p<0.001), whereas only a trend towards higher risk was observed for subjects with HFrEF (HR 1.37 [0.95; 1.95], p=0.089) and no relationship to outcome was identified in individuals with HFmrEF (HR 1.08 [0.72; 1.61], p=0.72)