Synopsis

Keywords: Data Analysis, Metabolism

Sleep is fundamental to human health and function. Cerebral metabolism and blood supply are key physiological parameters of brain function, but the manner in which they change during different sleep stages is still largely unknown. In this study, we collected wakefulness and sleep data with concurrent EEG-MRI. We measured CBF, SvO₂ and CMRO₂ with radial OxFlow MRI. Our data show that CMRO₂ is lower during non-REM sleep than during wakefulness and declines progressively as sleep stages become deeper. CBF decreases during non-REM sleep compared with wakefulness, while SvO₂ gradually increases from wakefulness to slow wave sleep.

Introduction

Sleep is important for various aspects of brain function. The human sleep cycle consists of three non-rapid eye movement (non-REM) sleep stages defined as N1, N2, slow wave sleep (SWS) and rapid eye movement sleep. Each sleep stage has a distinct function. Cerebral blood flow (CBF), venous oxygen saturation (SvO₂) and cerebral metabolic rate of oxygen (CMRO₂) are key physiological parameters of brain function. A recent study has demonstrated feasibility of quantifying changes in cerebral metabolism associated with sleep, specifically to uncover the changes in brain oxygen metabolism during SWS¹. In distinction to this prior work, the purpose of the present study was to investigate the changes in these physiological parameters during various sleep stages.

Method

The study was approved by the Institutional Review Board of the University of Pennsylvania. Eight subjects (6 female, mean age: 27±4.1 y) were recruited. EEG data were acquired with a 15-channel MR-compatible EEG system from Brain Products with a sampling rate of 5000 Hz. EEG data were processed offline with Brain Vision Analyzer. First, gradient artifacts were removed using a sliding window average of 21 intervals. Pulse artifacts were removed by subtracting an average ECG artifact template. Subsequently, a passband filter encompassing the range from 0.5 to 24 Hz was applied, with a notch filter at 60 Hz. Independent component analysis was applied to remove residual noise. The data were further down-sampled to 250 Hz. Sleep stage was scored for every 30-s frame of preprocessed EEG data according to AASM criteria.Five discrete sleep stages were identified and scored: wakefulness, non-REM stage 1 (N1), non-REM stage 2 (N2), slow wave sleep (SWS), and REM.
Imaging was performed on a 3T MR scanner (Siemens Prisma, Erlangen, Germany) using a 64-channel head-neck coil. Figure 1 shows the study protocol. Scan parameters for the OxFlow sequence were as follows: FOV=240*240 mm², slice thickness=5 mm, flip angle=20°, TR=40 ms, inter-echo spacing=5 ms, velocity encoding (VENC)=76.4 cm/s, first echo time=6.03 ms, duration=80 min. The OxFlow sequence uses golden angle radial k-space sampling^1-2. The temporal footprint is 48.8 s and the effective temporal resolution is 2.72 s. CBF was quantified in the superior sagittal sinus (SSS), and upscaled to yield total CBF; SvO₂ was also quantified in the SSS via MRI susceptometry³. CMRO₂ was calculated via Fick’s principle. Mean and standard deviation of the parameters were calculated for each sleep stage. One-way ANOVA with Tukey’s post hoc test was applied to compare the differences in the three parameters under different sleep stages.

Results

Three out of eight subjects achieved slow wave sleep and one of them was excluded due to head motion, thus two subjects (S1: female, 29y; S2: female, 27y) were included in this abstract. Figure 2 shows CBF, SvO₂ and CMRO₂ time-series for the two subjects who successfully achieved slow-wave sleep. Subject S1 exhibits a typical non-REM sleep pattern. CMRO₂ decreases gradually during the first non-REM sleep cycle (from t=0 to t=32 min). The subject achieved slow-wave sleep at t=15 min, and CMRO₂ reached a nadir at t=25 min. Subsequently, the subject awakens from SWS at t=32 min, coincident with a sharp increase in CMRO₂. Subject S2 displays more arousals compared with the first subject. The subject achieved slow-wave sleep at t=45 min, with the mean CMRO₂ during SWS clearly lower than during wakefulness and light sleep (N1). Figure 3 shows the three parameters during different sleep stages. Multiple comparison results show that CBF during wakefulness, N1 and SWS are significantly higher than N2 for the first subject. For the second subject, CBF during wakefulness is higher than during non-REM sleep, reaching a minimum during SWS. Conversely, SvO₂ gradually increases from wakefulness toward SWS, reaching a maximum during SWS for both subjects. CMRO₂ is lower than during wakefulness during non-REM sleep, declining progressively toward greater sleep stage for both subjects.

Discussion

We investigated the changes in cerebral metabolism during wakefulness and non-REM sleep with radial OxFlow MRI. The data suggest SvO₂ to be greater than during wakefulness during all non-REM sleep stages, and increasing progressively toward deeper sleep, while CMRO₂ during non-REM sleep is lower than during wakefulness, reaching a minimum during SWS as found previously¹. CBF for the second subject was found to decrease from wakefulness to non-REM sleep, consistent with recent finding in the authors’ laboratory¹. However, CBF during wakefulness was not significantly higher than during N1 and SWS for the first subject. The reason might be that the subject was very sleepy at beginning of the exam, as suggested by the short sleep onset latency of 3 minutes. Further, the total wake time of 6 minutes was found to be unusually short. There are two limitations in the study. First, five out of eight subjects did not achieve SWS, which we attribute to the discomfort of the scanner environment (EEG cap, scanner noise, space confinement), which could, at least in part, be mitigated through active sleep deprivation. Further, while minor, we did not distinguish eyes open from eyes closed conditions. Previous studies found that global CBF differs between eyes-open and eyes-closed resting state^4,5. In future work, we plan to determine the metabolic implications of these two states of wakefulness.

Acknowledgements

NIH R21 AG065816