Abstract Introduction Cerebrospinal fluid (CSF) influx, a key process of the glymphatic system, is driven by cardiac pulsations which move cerebral arterial walls and push CSF into brain parenchyma. In mice, CSF influx shows circadian rhythmicity, peaking during rest periods, and dipping during active periods. In humans, pulsatile CSF waves are larger during sleep-deprived wakefulness, resembling CSF waves during sleep. This study examined both homeostatic and circadian modulation of pulsatile CSF during sleep deprivation in humans. Pulsatile CSF waves were expected to be larger over time, peaking during the sleep-deprived night. Methods Twenty-one healthy adults (i.e., 10 women; mean age = 23.6) were kept awake for 39 hours and scanned every six hours. Dynamic diffusion-weighted imaging (b = 150 s/mm2; TR = 2035 ms; TE = 67.20 ms) was used to model CSF cardiac pulsatility (pCSF) in perivascular spaces (sPVS) surrounding cerebral arteries (Wen et al., 2022). The amplitude of pCSF was quantified as the trough-to-peak range of its waveform. A cosinor model was fit to the data with a fixed 24-hour period using the GLMMcosinor package in R. Rhythmic parameters included the Midline Estimating Statistic of Rhythm (MESOR), amplitude, and acrophase (timing of peak). A linear term for hours awake was added to account for homeostatic sleep pressure. Results The MESOR of pCSF amplitude surrounding large arteries was 2.392 mm2/s (β = 2.392, SE = 0.130, p 0.001) with a 24-hour amplitude of 0.223-3 mm2/s (β = 0.223, SE = 0.067, p 0.001). Acrophase occurred at 20.90 hours of wakefulness at 4:23 am (β = -0.815, SE = 0.295, p = 0.006). There was a positive linear trend of time awake on the pCSF amplitude signal (β = 0.009, SE = 0.004, p = 0.024). Conclusion Pulsatile CSF, an index of glymphatic influx, displayed circadian rhythmicity that peaked during the typical rest phase (~4 a.m.). During this peak, pulsatile CSF waves appeared larger and more sleep-like in perivascular spaces surrounding major arteries. Consistent with predictions, pulsatile CSF amplitude also increased with accumulating homeostatic sleep pressure. These findings support dual circadian and homeostatic regulation of CSF pulsatility. Support (if any) Army Research Office award W911NF2210223
Huskey et al. (Fri,) studied this question.