1168 Quantifying Neurological Aging in Parkinson’s Disease Using Multidimensional Sleep EEG Biomarkers

Abstract Introduction Sleep microstructure shows age-related decline, including alterations in sleep spindle characteristics and reduced slow wave activity, which reflect diminished synaptic integrity and increased cognitive vulnerability (Tsapanou et al., 2020). Sleep EEG models can estimate brain age, and deviations from chronological age predict inflammation, cognitive impairment, and dementia risk (Miao et al., 2025; Sun et al., 2019, 2025). Parkinson’s disease is marked by pronounced sleep pathology, including elevated REM sleep without atonia and abnormal REM theta activity, both linked to neurodegeneration and cognitive decline(Galbiati et al., 2018; Iranzo et al., 2010) . This study aims to integrate sleep microstructure features into a unified Neurological Age Index (NAI) and evaluate its association with cognitive and clinical measures as an indicator of neurological aging. Methods The cohort included individuals with drug naïve Parkinson’s disease and age matched healthy controls who completed overnight laboratory polysomnography with high density EEG. Clinical measures included Hoehn and Yahr staging, UPDRS III motor scores, MoCA, and a motor learning task. Sleep microstructure metrics were derived from standardized EEG preprocessing and included spindle characteristics, slow wave activity, spectral ratios, slow oscillation and spindle coupling, and the RSWA index. Associations between sleep physiology, cognitive performance, and clinical severity were evaluated using nonparametric correlations and integrated into a unified NAI for multivariate modeling. Results Preliminary analyses showed a disease specific coupling between sleep microstructure and clinical severity in Parkinson’s disease. Elevated REM theta activity was the strongest marker, correlating with both Hoehn and Yahr stage (r = 0.558, p = 0.005) and UPDRS III scores (r = 0.425, p = 0.038). NREM spectral features showed additional but variable associations, while RSWA demonstrated modest relationships with disease stage. Additional features are being incorporated into the unified NAI model. Conclusion These findings suggest that specific elements of sleep microstructure provide sensitive markers of brain pathology. Integrating these features into a unified NAI may offer a novel physiological measure of neurodegeneration. This approach holds promise for improving early detection and tracking disease progression. Support (if any) This work was supported by the Aufzien Family Center for the Prevention and Treatment of Parkinson’s Disease (APPD).