Tommi Väyrynen1, Heta Helakari1, and Vesa Kiviniemi1
1OFNI/Diagnostic Imaging, Oulu University Hospital, Oulu, Finland
Synopsis
Keywords: fMRI Analysis, Neurofluids, Consciousness
Motivation: Vasomotor waves increase in sleep but their causal impact on brain electrical activity and hydrodynamics is not understood.
Goal(s): We wanted to find out what is driving sleep related glymphatic solute transport.
Approach: We used multimodal 3D whole brain MREG, dcEEG and waterNIRS simultaneously to find out what VLF < 0.1 Hz pulsation is driving the other(s).
Results: In sleep the normal causal electro/hydrodynamic drive of the vasomotor BOLD waves reverses locally near upper parasagital and deeper structures.
Impact: The driver of the increase brain solute transport in human sleep has been unknown. In out ultrafast multimodal MREG study, we show significant reversals in the causal drive of vasomotor waves locally.
Background.
Background.
Sleep
increases brain solute convection, and this has been related to increased power
of physiological brain pulsations1,2,3. In
this study we investigated how the causal connection between the brain water
concentrations (wNIRS) and
dcEEG
derived blood-brain-barrier (BBBEEG)
potential dynamics4 is
altered with respect to venous vasomotor BOLD fluctuations. We hypothesized
that the increased brain solute transport is driven by increased vasomotor BOLD
waves. To
test the hypothesis, we statistically compared the causal relations of the
brain water
wNIRS and
electrodynamic BBBEEG signals during normal awake and
increased glymphatic convection in NREM sleep.
Methods.
After
approved informed consent, healthy subjects (n=21 awake, n=19 EEG verified
sleep) were scanned with: a)
water sensitized near-infrared spectroscopy (wNIRS)
to measure brain water concentration5 in
synchrony with b) ultrafast MREG scanning (TR 100 ms,
3D whole brain, flip angle 5°) reflecting brain non-aliased venous BOLD
fluctuations6,and
c) 256 ch
dcEEG
recording (EGI) for BBBEEG.
Standard FSL pre-processing was used on MREG BOLD data, dcEEG
was corrected also with standard fMRI artifact removal. Water concentration wNIRS
was calculated using Beer-Lambert law from 3 NIR wavelengths (690 nm, 830 nm
and water sensitive 980 nm). The simultaneous dcEEG,
wNIRS
and MREG data were segmented into 2 min EEG-verified sleep segments and
bandpass filtered to very low frequency (VLF, 0.01-0.08 Hz). Hilbert
transformation was applied for both VLF traces to extract instantaneous
phase time-series. Phase transfer entropy (PTE)7 was
then used to study the causal directional interactions with differential form (dPTE).
FSL randomise (TFCE) with 5k permutations p < 0.05 was used for statistical
inferences in brain images of altered dPTE
causal drive.
Results. In
awake state the brain H2O-signal and dcEEG
reflecting BBBEEG potential causally precede and
drive VLF BOLD signal wave amplitude and phase transitions over the whole
brain, blue color c.f. Fig 1-2. However, in N1 and N2 sleep this interaction
inverts; the VLF BOLD waves starts to
causally drive both H2O signal and BBB potential
regionally, c.f. Fig’s 1-2 hot
colors in cortex. In especially the BBBEEG
the
spatial extent in N2 > N1 and co-localizes with sensory and parasagittal
areas known to have increased glymphatic slow delta activity1,2. Deeper areas of the brain seem to
maintain the normal causal drive of BOLD in even in N1-N2 sleep.
Conclusion: Vasomotor waves causal
relation to electro/hydrodynamic signals inverts in NREM sleep in parasagittal
areas of slow delta sleep. Vasomotor waves start to drive glymphatic water and
electrolyte convection over BBB.
Acknowledgements
JAES-Foundation, Aivosäätiö, VTR & Research Council of Finland are cordially acknowledged for their grant support.References
1. Xie
et al., 2013. Science 342:373–377.
2. Helakari et al. 2022. J. Neurosci. 42(12):2503–2515.
3. Nedergaard et al., 2013. Science, 340(6140), 1529-1530.
4. Korhonen et al.,
2014. Brain Connect. 4, 677–689.
5. Myllylä et al., 2018. Journal of biophotonics, 11(8), e201700123.
6. Kiviniemi et al., 2016. J Cereb
Blood Flow Metab
36:1033–1045.
7. Lobier et al., 2014. NeuroImage. Jan 15;85 Pt 2:853-72.