Neuroinflammation is associated with non‐REM sleep reduction in individuals without dementia

Abstract Background Sleep disturbances and especially reduction of non‐REM (NREM) sleep are common in aging with accumulating evidence showing it might contribute to cognitive decline as well as increase the risk of developing dementia due to Alzheimer’s disease (AD). Lately, sleep disturbances and AD separately have been linked to increase in systemic inflammation. NREM sleep is thought to allow clearance of extracellular Aβ, which led to studies suggesting NREM could be an early biomarker of AD risk. The main objective of this study was to see the relationship between NREM sleep as assessed with polysomnography and neuroinflammation in vivo . Method 25 individuals (18 cognitively unimpaired (CU) and 7 mild cognitive impairment (MCI)) underwent a [ 11 C]PBR28 neuroinflammation‐PET scan, an MRI a full neuropsychological evaluation as well as a polysomnography. [ 11 C]PBR28 standardized uptake value ratios (SUVRs) used the cerebellum grey matter as the reference region and was calculated between 0‐90 min post‐injection. A voxel based regression model evaluated the relationship between neuroinflammation as assessed by PET, and the total minutes of NREM sleep. The model’s covariates were age and diagnosis. Result We found a negative correlation between neuroinflammation and the total minutes of non‐REM sleep measured by a polysomnography. The regions showing neuroinflammation were the reticular formation, the arcuate nucleus of the medulla, the frontal cortex and especially the vmPFC as well as the insula and temporal poles. When we corrected for sex and years of education, results were similar. Conclusion These preliminary results show that less NREM sleep correlates with neuroinflammation, in the brain of CU and MCI individuals. The regions impacted, such as the reticular formation, the vmPFC and the insula are needed for arousal and alertness. The arcuate nucleus of the medulla is required for proper breathing. Finally, temporal poles are part of the first regions affected in AD, and later on in the disease, the medial frontal cortex also starts degrading. Our study adds to previous research showing a link between sleep disturbances and neuroinflammation in vivo and identifies anatomical correlates.