Biomarkers

Sleep was associated with an increased extracellular volume fraction (ECVF) in mice and enhanced cerebrospinal fluid (CSF) clearance of neurotoxic amyloid-beta (Aβ) proteins - a hallmark of Alzheimer's disease (AD). Such a beneficiary impact of sleep is however difficult to study on humans due to the lack of non-invasive imaging techniques. Last year (2024), we were, with unique sodium MRI and MRI-compatible EEG, able to study a cohort of healthy subjects and found a decrease, instead of increase, in ECVF during sleep. To confirm such an unexpected finding, here we report a study on a different cohort of healthy subjects using the same technologies as in the last-year study, i.e., simultaneous measurements of ECVF by sodium (²³Na) MRI and sleep by MRI-compatible EEG.

This study (Figure 1) was performed on 30 cognitively normal human subjects (age 25-87 years, 22 females, 8 males), with approved IRB and signed consent. Each subject underwent a 90-min sodium MRI (Siemens Prisma, 3T) with a dual-tuned (¹H-²³Na) birdcage head coil (QED, Cleveland, OH) and a continuous recording of MR-compatible EEG (Brain Vision, 32 channels). MRI scans consist three segments, each of 16min long with a 2-min gap in between for artifact-free EEG recording. Sleep was scored to five stages (wake, N1, N2, N3, and REM) according to AASM standards (V2.6, 2020). The pulse sequence was custom-developed twisted projection imaging (TPI). ECVF was quantified voxel-by-voxel using a two-compartment model of intra- and extra-cellular spaces.

Figure 1 (P2) shows representatives of our EEG waveforms from our study subjects at different sleep stages. Figure 2 presents typic maps of ECVF from an individual subject, while Figure 3 summarizes ECVF from all of the subjects studied. Overall, ECVF changed during sleep in both white and gray matter regions of the brain; decreasing statistically significant in gray matter regions during N3 (p = 0.005) but not in N2 (p = 0.464) and not in white matter regions during N3 (p = 0.067).

We found ECVF to decrease in slow wave sleep (N3), confirming our previous results in a different cohort of subjects but contradicting previous animal studies. This prompts further investigation of physiological importance of sleep.