Mechanisms of hypoxia‐induced intrapulmonary arteriovenous shunting in healthy humans at rest: arterial oxygen saturation or pulmonary artery systolic pressure?

Abstract

Acute, brief (5–10 min) exposure to hypoxia (fraction of inspired oxygen (FIO2) = 0.12) has been shown to open intrapulmonary arteriovenous (IPAV) shunt pathways in some subjects, but not all, at rest. We set out to determine the threshold FIO2 necessary to open large diameter IPAV pathways in all humans at rest and determine if the detection of IPAV shunt was correlated with arterial oxygen saturation (SpO2) or pulmonary artery systolic pressure (PASP). Ten healthy subjects (5 female) breathed four levels of hypoxia (FIO2=0.16, 0.14, 0.12, 0.10) at rest in ascending or descending order for 30 min, with each FIO2 separated by a 15 min normoxic break. Saline contrast echocardiography was used to detect the passage of bubbles through IPAV shunts into the left heart. The number of bubbles appearing in the left heart was scored on a 0–5 scale with a bubble score ≥2 defined as IPAV shunt. PASP was determined from the peak velocity of the tricuspid regurgitation. IPAV shunt occurred after 30 min of breathing an FIO2=0.16, 0.14, 0.12, and 0.10 in 1/10, 6/10, 7/10, and 10/10 subjects, respectively. Shunt score and SpO2 were well correlated (Spearman correlation, rs=−0.77, p<0.01) whereas shunt score and PASP were poorly correlated (rs=0.14, p=0.397). Our results demonstrate that hypoxia‐induced IPAV shunt can occur in all subjects at rest and that it is primarily mediated by arterial O2 saturation rather than PASP. Funding: OHSU MRF 0820