Background and purpose

The restorative function of sleep is well known¹ as are the adverse effects of persistent insufficient and disrupted sleep on cognitive function and cardiovascular health^2-4. During sleep, reduced synaptic transmissions lead to a slowdown in the cerebral metabolic rate of oxygen (CMRO₂) and glucose utilization⁵. Early work from Madsen et al. using invasive technology based on the Kety-Schmidt technique using xenon tracer kinetics, showed CMRO₂ to decrease from 5% to 25%, during light and deep non-REM sleep, respectively^6,7. However, in order to be clinically practical, the method must be noninvasive, robust and be able to yield data at a rate on the order of a few seconds. Susceptometry-based MRI oximetry satisfies these requirements^8,9. The technique, referred to as OxFlow, relies on a joint measurement of total cerebral blood flow (tCBF) and venous oxygen saturation (SvO₂) of a major cerebral vein, typically the superior sagittal sinus (SSS) draining the cortex. Here we used a golden-angle radial OxFlow (rOxFlow) to follow CMRO₂ dynamically with high temporal resolution¹⁰. We had previously measured the sleep-driven CMRO₂ reduction in healthy volunteers, assessing their sleep state empirically¹¹. However, to definitively assess whether the subject is asleep and possibly correlate the level of CMRO₂ reduction with sleep stage, electroencephalography (EEG) is required. Here, we adapted rOxFlow for concurrent EEG monitoring and were able to evaluate the evolution of CMRO₂ during the transition from wakefulness to sleep at 3.0 T.

Methods

Three healthy male volunteers (ages 22, 35 and 40 years), were scanned in late evening following dinner. None were sleep deprived, but they all had indicated previously that they were able to sleep during MRI data collection in the scanner. EEG data were recorded in the MR scanner (Siemens Prisma, 3.0 T) using a 15-channel customized MR-compatible sleep cap and a 32-channel amplifier (BrainAmp MR Plus). Scalp electrodes were placed according to the international 10-20 system. The EEG+rOxFlow protocol lasted 45 minutes: during the first 5 and the last 10 minutes the subjects were in eyes open condition, whereas in the central 30 minutes whey were at liberty to sleep (Fig. 1). CMRO₂ was computed off-line via the Fick’s principle: $$$CMRO_2=C_a\cdot tCBF\cdot(SaO_2-SvO_2)$$$, with $$$C_a$$$ being the O₂ carrying capacity of hemoglobin (1.39 ml O₂/g[Hb]), in which [Hb] is the hemoglobin concentration measured from a finger stick blood sample (Hemocue Hb 201+). rOxFlow technique yields the total cerebral blood flow through PC-MRI, and the venous oxygen saturation through susceptometry-based oxymetry. tCBF was obtained by upscaling the flow measured in the superior sagittal sinus (SSS_BF) by a calibration factor corresponding to the ratio between the inflow from the internal carotid and vertebral arteries and the outflow from SSS⁹. was measured by pulse oximetry (Veris Medrad 8400). CMRO₂ and tCBF were normalized to brain mass. The rOxFlow protocol was slightly adapted with respect to the one used previously¹¹, based on preliminary tests for simultaneous EEG + rOxFlow acquisition in eyes open closed conditions. TR=50ms (frequency=20Hz) avoids signal contamination of α-waves, characteristic of eyes-closed condition. rOxFlow resolution (and thus of CMRO₂) was 3.4s. EEG data were digitally filtered to attenuate gradient induced and cardioballistic artifacts, and Fourier transformed to yield power spectral densities of characteristic wavebands. The onset of non-REM sleep was inferred from delta power density increase¹².

Results and Discussion

The stability of the protocol when performed during wakefulness is illustrated in Fig. 2. Time-courses of the measured parameters during transition from wakefulness to sleep are plotted in Fig. 3, together with heart rate (HR). The subjects were inferred to be asleep based on the characteristic increase and subsequent decrease in delta-wave power density (Fig. 4). CMRO₂ was found to decrease following onset of sleep with respect to pre-sleep wakefulness in all three subjects (P<0.0001), corresponding to an average relative change of about -15% in subjects 1 and 2 and -12% in subject 3 (see Table 1). The observed CMRO₂ depression was paralleled by a HR decrease of 6-7 bpm, consistent with previous studies¹³. An increase in HR could be observed following forced awakening (Fig. 3). Of note, however, is that post-sleep CMRO₂ did not recover to pre-sleep values during the 15-minute post-sleep period. Future studies will require inclusion of prolonged post-sleep periods, to allow monitoring of the return to baseline values.

Conclusions

These preliminary data illustrate the feasibility of in-scanner monitoring of the effect of sleep on CMRO₂ via temporally resolved whole-brain oximetry and concurrent EEG. The method should be suited for longer-term studies in sleep-deprived subjects to quantify the sleep-stage specific effects on brain oxygen metabolism.

Acknowledgements

NIH Grants R01 HL122754 and R21 EB022687

References

1. Benington JH, Heller HC. Restoration of brain energy metabolism as the function of sleep. Progress in neurobiology 1995;45(4):347-360. 2. Goel N, Rao H, Durmer JS, Dinges DF. Neurocognitive consequences of sleep deprivation. Semin Neurol 2009;29(4):320-339. 3. Mullington JM, Haack M, Toth M, Serrador JM, Meier-Ewert HK. Cardiovascular, inflammatory, and metabolic consequences of sleep deprivation. Prog Cardiovasc Dis 2009;51(4):294-302. 4. Pack AI, Gislason T. Obstructive sleep apnea and cardiovascular disease: a perspective and future directions. Prog Cardiovasc Dis 2009;51(5):434-451. 5. Boyle PJ, Scott JC, Krentz AJ, Nagy RJ, Comstock E, Hoffman C. Diminished brain glucose metabolism is a significant determinant for falling rates of systemic glucose utilization during sleep in normal humans. J Clin Invest 1994;93(2):529-535. 6. Madsen PL, Schmidt JF, Holm S, Vorstrup S, Lassen NA, Wildschiodtz G. Cerebral oxygen metabolism and cerebral blood flow in man during light sleep (stage 2). Brain Research 1991;557(1-2):217-220. 7. Madsen PL, Schmidt JF, Wildschiodtz G, Friberg L, Holm S, Vorstrup S, Lassen NA. Cerebral O2 metabolism and cerebral blood flow in humans during deep and rapid-eye-movement sleep. Journal of Applied Physiology 1991;70(6):2597-2601. 8. Jain V, Langham MC, Wehrli FW. MRI estimation of global brain oxygen consumption rate. J Cereb Blood Flow Metab 2010;30(9):1598-1607. 9. Rodgers ZB, Jain V, Englund EK, Langham MC, Wehrli FW. High temporal resolution MRI quantification of global cerebral metabolic rate of oxygen consumption in response to apneic challenge. J Cereb Blood Flow Metab 2013;33(10):1514-1522. 10. Cao W, Chang YV, Englund EK, Song HK, Barhoum S, Rodgers ZB, Langham MC, Wehrli FW. High-speed whole-brain oximetry by golden-angle radial MRI. Magn Reson Med 2018;79(1):217-223. 11. Lee H, Langham MC, Rodriguez-Soto AE, Detre JA, Schwab R, Wehrli FW. CMRO2 Changes During Sleep in Humans. 2018 June 16-21; Paris, France. 12. Campbell IG. EEG recording and analysis for sleep research. Curr Protoc Neurosci 2009;Chapter 10:Unit10.2. 13. Khatri, I. M., & Freis, E. D. Hemodynamic changes during sleep. Journal of Applied Physiology 1967;22(5),:867-873.