Determinants of exercise intolerance in ambulatory heart failure: insights from cardiopulmonary exercise testing
L Lorenzo Alves, J Silva Cardoso, R Rodrigues, S AmorimAbstract
Background
Cardiopulmonary exercise testing (CPET) provides an integrated assessment of exercise capacity in heart failure (HF). Peak oxygen uptake (peak VO₂) remains the reference parameter for functional evaluation, while circulatory power, combining peak VO₂ and peak systolic blood pressure, reflects haemodynamic reserve. However, the clinical and biological determinants of exercise intolerance assessed by these CPET-derived indices are not fully characterised in ambulatory HF.
Purpose
To identify independent clinical and biological determinants of exercise intolerance in ambulatory HF patients using peak VO₂ and circulatory power as complementary measures of functional capacity.
Methods
We conducted a single-centre retrospective study including ambulatory HF patients (NYHA class I–IV) who underwent CPET between January 2023 and September 2025. Clinical, laboratory and echocardiographic data were collected from the closest evaluation to CPET. Exercise intolerance was assessed using peak VO₂ (mL/kg/min) and circulatory power. We selected peak VO₂ and circulatory power as complementary CPET-derived measures that directly quantify exercise capacity and hemodynamic reserve. Associations were explored using Spearman correlations and linear regression analyses. Multivariable models included clinically relevant variables with low collinearity.
Results
Eighty-one patients were included (mean age 51.5 ± 13.0 years; 74% male). Mean peak VO₂ was 17.1 ± 5.2 mL/kg/min and mean circulatory power was 2415 ± 954 units. Peak VO₂ correlated with NYHA functional class (ρ = –0.48, p < 0.001), left ventricular ejection fraction (ρ = 0.43, p < 0.001), serum uric acid (ρ = –0.38, p < 0.001) and NT-proBNP (ρ = –0.52, p < 0.001). In multivariable analysis, NYHA functional class (β = –2.67 mL/kg/min per class, p < 0.001) and serum uric acid (β = –0.52 mL/kg/min per mg/dL, p = 0.035) emerged as independent determinants of peak VO₂, whereas left ventricular ejection fraction was not independently associated. Circulatory power showed similar univariable associations; however, in multivariable analysis, NYHA functional class remained the sole independent determinant (β = –547 units per class, p < 0.001).
Conclusion
In ambulatory heart failure, exercise intolerance reflects different underlying mechanisms depending on the CPET-derived parameter used. Peak VO₂ captures the combined impact of functional status and systemic metabolic dysfunction, whereas circulatory power predominantly reflects global clinical severity. These findings underscore the multifactorial nature of exercise intolerance in HF and support the complementary use of CPET-derived indices beyond left ventricular systolic function alone.TableFor image description, please refer to the figure legend and surrounding text.BlotFor image description, please refer to the figure legend and surrounding text.