Systemic Arterial Oxygen Levels Differentiate Pre- and Post-capillary Predominant Hemodynamic Abnormalities During...

Whether systemic oxygen levels (SaO₂) during exercise can provide a window into invasively derived exercise hemodynamic profiles in patients with undifferentiated dyspnea on exertion is unknown.

We performed cardiopulmonary exercise testing with invasive hemodynamic monitoring and arterial blood gas sampling in individuals referred for dyspnea on exertion. Receiver operator analysis was performed to distinguish heart failure with preserved ejection fraction from pulmonary arterial hypertension.

Among 253 patients (mean ± SD, age 63 ± 14 years, 55% female, arterial O₂ [PaO₂] 87 ± 14 mmHg, SaO₂ 96% ± 4%, resting pulmonary capillary wedge pressure [PCWP] 18 ± 4mmHg, and pulmonary vascular resistance [PVR] 2.7 ± 1.2 Wood units), there was no exercise PCWP threshold, measured up to 49 mmHg, above which hypoxemia was consistently observed. Exercise PaO₂ was not correlated with exercise PCWP (rho = 0.04; P = 0.51) but did relate to exercise PVR (rho = -0.46; P < 0.001). Exercise PaO₂ and SaO₂ levels distinguished left-heart-predominant dysfunction from pulmonary-vascular-predominant dysfunction with an area under the curve of 0.89 and 0.89, respectively.

Systemic O₂ levels during exercise distinguish relative pre- and post-capillary pulmonary hemodynamic abnormalities in patients with undifferentiated dyspnea. Hypoxemia during upright exercise should not be attributed to isolated elevation in left heart filling pressures and should prompt consideration of pulmonary vascular dysfunction.