MRI measurement of the BOLD-specific flow–volume relationship during hypercapnia and hypocapnia in humans

It is widely assumed in fMRI that the relationship between cerebral blood flow (CBF) and volume (CBV) changes observed during end-tidal CO2 (PETCO2) perturbations is equivalent to that elicited by neuronal activation. This assumption has been validated in PET studies insofar as relating total flow to total CBV changes, but remains unconfirmed for venous CBV changes, which pertains to the primary vascular compartment modulating the BOLD signal. In this study, we measured CBF and venous CBV changes in healthy subjects in response to graded hypercapnia and hypocapnia, induced using computerized end-tidal CO2 targeting, with a ΔPETCO2 range of between − 6 and + 9 mm Hg. Hypercapnia was found to elicit robust increases in CBF and venous CBV, while hypocapnia produced decreases in both. We used steady-state flow and volume changes to estimate the power–law relationship for cortical and subcortical brain regions, and did not observe significant difference between the two. The combined fit resulted in a power coefficient of 0.18 ± 0.02, substantially lower than Grubb's coefficient of 0.38, but comparable to previous observations during neuronal activation. These results confirm that the BOLD-specific flow–volume relationship during CO2 challenges is similar to that characterizing neuronal activation.

Research highlights► The first MRI measurement of changes in cerebral blood flow and venous blood volume in grey matter during graded hypercapnic and hypocapnic challenges in humans, and doing so using computerized end-tidal targeting. ► The BOLD-specific capnic flow–volume relationship differs substantially from Grubb's law. ► Similar BOLD-specific flow-volume relationships apply to both capnia-induced and neuronal activation-induced hyperemia. ► Minimal difference across cortex, and between cortical and subcortical grey matter.