The effect of isocapnic hyperoxia on neurophysiology as measured with MRI and MEG

Croal, Paula L. and Hall, Emma L. and Driver, Ian D. and Brookes, Matthew Jon and Gowland, Penny A. and Francis, Susan T. (2015) The effect of isocapnic hyperoxia on neurophysiology as measured with MRI and MEG. NeuroImage, 105 . pp. 323-331. ISSN 1053-8119

PDF (pdf file) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution Non-commercial No Derivatives.
Download (249kB) | Preview


The physiological effect of hyperoxia has been poorly characterised, with studies reporting conflicting results on the role of hyperoxia as a vasoconstrictor. It is not clear whether hyperoxia is the primary contributor to vasoconstriction or whether induced changes in CO2 that commonly accompany hyperoxia are a factor. As calibrated BOLD fMRI based on hyperoxia becomes more widely used, it is essential to understand the effects of oxygen on resting cerebral physiology. This study used a RespirActTM system to deliver a repeatable isocapnic hyperoxia stimulus to investigate the independent effect of O2 on cerebral physiology, removing any potential confounds related to altered CO2. T1-independent Phase Contrast MRI was used to demonstrate that isocapnic hyperoxia has no significant effect on carotid blood flow (normoxia 201 ± 11 ml/min, -0.3 ± 0.8 % change during hyperoxia, p = 0.8), whilst Look Locker ASL was used to demonstrate that there is no significant change in arterial cerebral blood volume (normoxia 1.3 ± 0.4 %, -0.5 ± 5 % change during hyperoxia). These are in contrast to significant changes in blood flow observed for hypercapnia (6.8 ± 1.5 %/mmHg CO2). In addition, magnetoencephalography provided a method to monitor the effect of isocapnic hyperoxia on neuronal oscillatory power. In response to hyperoxia, a significant focal decrease in oscillatory power was observed across the alpha, beta and low gamma bands in the occipital lobe, compared to a more global significant decrease on hypercapnia. This work suggests that isocapnic hyperoxia provides a more reliable stimulus than hypercapnia for calibrated BOLD, and that previous reports of vasoconstriction during hyperoxia probably reflect the effects of hyperoxia-induced changes in CO2. However, hyperoxia does induce changes in oscillatory power consistent with an increase in vigilance, but these changes are smaller than those observed under hypercapnia. The effect of this change in neural activity on calibrated BOLD using hyperoxia or combined hyperoxia and hypercapnia needs further investigation.

Item Type: Article
Keywords: Hyperoxia, Magnetoencephalography, BOLD, fMRI, Cerebral Blood Flow, Cerebral Blood Volume, Neural oscillations
Schools/Departments: University of Nottingham UK Campus > Faculty of Science > School of Physics and Astronomy
Identification Number:
Depositing User: Francis, Susan
Date Deposited: 22 Jul 2015 14:01
Last Modified: 17 Sep 2016 20:17

Actions (Archive Staff Only)

Edit View Edit View