To characterize neuropathological diagnoses in patients who underwent sleep evaluation.

Sleep disorders occur frequently in neurodegenerative disease. The range of autopsy-confirmed neuropathological diagnoses in patients with well-defined sleep disorders is poorly understood.

Participants from the Mayo Clinic Study of Aging (MCSA) and Alzheimer’s disease research center (ADRC) who had antemortem comprehensive sleep medicine evaluation including polysomnography (PSG) and underwent autopsy between January 2020 and May 2024 were included. Proportion of neuropathologic diagnosis within each sleep disorder was compared to the overall cohort using Fisher’s exact test. Multiple logistic regression using death age, sleep visit age, sex, and sleep diagnoses confirmed relationships between sleep disorders and neuropathologies. 

Ninety-five participants were identified with male predominance (64%) and average ages of 73.5 years at sleep evaluation and 84.8 years at death. Neuropathologic diagnoses included moderate/severe arteriolosclerosis (87.7%); intermediate/high likelihood Alzheimer’s disease neuropathologic change (64.2%); aging-related tau astrogliopathy (59.0%); limbic-predominant age-related TDP-43 encephalopathy neuropathic change, LATE-NC (33.7%); Lewy body disease, LBD (32.6%); primary age-related tauopathy (14.7%); and argyrophilic grain disease (12.6%) with 72.6% of participants having ≥2 pathologies. OSA was the most common diagnosis (78.9%) with equally distributed neuropathologies. Restless legs syndrome (RLS) occurred in 13.7% and most had LATE-NC (69%, p=0.029). RBD was diagnosed in 12.6%, with LBD pathology in the majority (83.3%, p<0.01) and both non-LBD cases were on serotonergic medications at PSG. Both associations remained significant on multiple logistic regression, with an additional association between LBD and RLS (p<0.01). Periodic limb movement disorder was present in 20% and insomnia in 9%. At PSG 21 patients had a clinical neurological diagnosis: MCI (n=8), Lewy body dementia (n=5), Alzheimer’s disease (n=4), unspecified dementia (n=3), and ALS (n=1). 

Our study reinforces the association between RBD and LBD and a potential relationship between RLS and LBD. The association between LATE-NC and RLS requires further evaluation.