Modulation of the human brain oxygen extraction fraction and metabolic rate in response to hyperoxia: A consequence of hypocapnia or reduced oxygen delivery?

Background: Exposure to supplementary oxygen (hyperoxia) alters respiratory patterns, causing shifts in carbon dioxide partial pressure (PaCO 2 ) levels – hypocapnia – and creating a cascade that disturbs cerebral blood perfusion. Although hyperoxia-induced hypocapnia decreases cerebral perfusion, its impact on brain oxygen extraction fraction and metabolism remains inconclusive as likely changes may also be evoked by reductions in cerebral oxygen delivery. To address this gap, we investigated how hyperoxia-driven hypocapnia affects brain oxygen delivery (CDO 2 ), oxygen extraction fraction (OEF), and metabolic rate for oxygen (CMRO 2 ) in healthy young humans. Methods: 9 healthy men (24 ± 4 years, 24 ± 2 Kg.m -2 ) participated in two randomized experimental sessions. In one session, participants underwent a 5-minute normoxic trial (NX), followed by a 10-minute isocapnic hyperoxia (IH) trial (100% O 2 with controlled PaCO 2 via rebreathing system). In the other session, the procedure was identical except for a 10-minute poikilocapnic hyperoxia (PKH) trial (100% O 2 without PaCO 2 clamp). Throughout the protocol, heart rate (HR, via ECG), beat-by-beat blood pressure (BP, using photoplethysmography), cardiac output (CO, photoplethysmography), and ventilation (VE, gas analyzer) were continuously monitored. Global cerebral blood flow (gCBF) was calculated using blood flow (BF) measurements from the internal carotid artery (ICA) and vertebral artery (VA) obtained via Doppler ultrasound, with gCBF determined as gCBF = (ICA flow + VA flow) x 2. Measurements were performed during each phase (NX and hyperoxia) of both visits. Radial artery and right jugular venous blood samples were collected to quantify PaCO 2 , CDO 2 , OEF, and CMRO 2 . Results: HR, BP, and CO did not change throughout the protocol of both sessions. PaCO 2 and VE remained unchanged during the IH, while PKH provoked hyperventilation and hypocapnia. gCBF was reduced in both IH (NX: 913.3 ± 153.1 vs. IH: 767.1 ± 130.1 mL.min -1 , P < 0.0001) and PKH (NX: 974.3 ± 212.6 vs. PKH: 716.4 ± 128.4 mL.min -1 , P = 0.0007). This reduction in gCBF was more significant in response to PKH (P = 0.006). CDO 2 was also reduced during IH and PKH, but no significant changes were observed when conditions were compared (P = 0.079). OEF remained unaltered in response to IH (NX: 31.9 ± 7.7 vs. IH: 33.2 ± 4.5%, P = 0.483). However, OEF increased during PKH (NX: 28.1 ± 6.2 vs. PKH: 35.5 ± 9.3%, P = 0.050). CMRO 2 did not change in response to PKH (NX: 49.9 ± 11.8 vs. PKH: 50.6 ± 13.2 mL.min -1 , P = 0.824), but it was reduced during IH (NX: 53.5 ± 15.1 vs. IH: 45.7 ± 8.6 mL.min -1 , P = 0.013). Conclusion: Findings reveal that hyperoxia-driven reductions in PaCO 2 significantly modulate cerebral oxygen dynamics. The increase in OEF and stability of CMRO 2 , even in the face of a reduction in CDO 2 in response to PKH, suggests that the decrease in PaCO 2 plays a role in the modulation not only of cerebral hemodynamics but also oxidative metabolism. Conversely, OEF did not change during IH while CDO 2 and CMRO 2 were reduced, highlighting the critical role of PaCO 2 regulation in gCBF and metabolic responses to hyperoxia. These results emphasize that the cerebral vascular and metabolic responses to hyperoxia are modulated by the fluctuations of PaCO 2 . CNPq, CAPES, FAPERJ, and FINEP. This abstract was presented at the American Physiology Summit 2025 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.