Characterization of the fast interactions between mean cerebral blood flow and brain activity in healthy subjects

Cerebral vasomotion is modulated according to brain demands via neurovascular coupling (NVC). NVC can be evaluated noninvasively via the simultaneous recording of electroencephalogram (EEG) and mean cerebral blood velocity (MCBv) monitored via transcranial Doppler (TCD) device. We propose an approach for NVC estimation grounded on the assessment of the cross-correlation at zero lag between the beat-to-beat variability series of MCBv and EEG power assessed over typical frequency bands utilized in EEG analysis, namely δ, θ, α and β bands. A surrogate data approach was applied to test the null hypothesis of the absence of NVC. This methodology was tested in 20 right-handed healthy volunteers (9 males, 11 females; age: 31±4 yrs) in which MCBv was monitored via TCD device along with a 19-channel EEG. Recordings were performed at rest in supine position (REST) and during active standing (STAND). We found that the strength of NVC varies with frequency band, experimental condition and brain region. More specifically, NVC strength was significant at REST especially in the δ and α bands and in the occipital and parietal regions, while it was much weaker during STAND, especially in those brain areas where NVC was particularly strong at REST. This novel approach to the NVC characterization highlights the complexity of the interactions between cerebral vasomotion and brain activity even in absence of a definite cognitive task.