Obstructive sleep apnea is associated with systemic inflammatory network reorganization and motor impairment in Parkinson’s disease
TL;DR
OSA is highly prevalent in Parkinson’s disease and is associated with worse motor performance and significant polysomnographic disruption, particularly in moderate-to-severe cases, suggesting a mechanistic connection between intermittent hypoxia and immune dysregulation in PD.
Abstract
Background: Obstructive sleep apnea (OSA) is highly prevalent in Parkinson’s disease (PD) and may exacerbate neurodegeneration through intermittent hypoxia, sleep fragmentation, and systemic inflammation. However, the extent to which OSA modulates inflammatory networks and clinical severity in PD remains incompletely understood. Objective: To investigate the association between OSA, polysomnographic parameters, motor impairment, and systemic cytokine/chemokine profiles in patients with PD. Methods: This cross-sectional study included 48 patients with PD (Hoehn & Yahr stages 1–3) evaluated at a tertiary center in Brazil. It was approved by the local Research Ethics Committee No. 58684622.2.0000.0007. Participants underwent a comprehensive neurological assessment using the Unified Parkinson’s Disease Rating Scale (UPDRS I–IV), sleep questionnaires, full-night video polysomnography, and peripheral blood sampling. OSA was defined according to ICSD-3-TR criteria. Plasma cytokines and chemokines were quantified using a 20-plex multiplex immunoassay. Multivariate statistics, network correlation analysis (Spearman), Louvain community detection, principal component analysis, and linear discriminant analysis were applied to characterize inflammatory signatures. Results: OSA was diagnosed in 54.2% of patients. The PD+OSA+ group was older and exhibited worse functional (UPDRS-II, p=0.009) and motor impairment (UPDRS-III, p=0.02). Polysomnography revealed reduced total sleep time (p=0.008), decreased REM sleep (p=0.03), higher arousal index (p=0.003), and increased oxygen desaturation (p<0.001). Univariate cytokine analysis showed limited individual differences (IL-18 higher in PD+OSA−, p=0.003). However, multivariate and network approaches revealed marked immune reorganization in PD+OSA+. Correlation networks demonstrated a shift from a fragmented inflammatory architecture in PD+OSA− to a dense, highly integrated cytokine–chemokine network in PD+OSA+, characterized by coordinated IL-1β/IL-6/TNF-α signaling and chemokine-driven leukocyte recruitment (CCL2, CXCL1, CXCL8). Louvain analysis confirmed consolidation into larger inflammatory modules. Stratification by OSA severity showed that moderate-to-severe OSA was associated with a mixed Th1/Th2 axis (IL-6, TNF-α, IFN-γ, IL-13, IL-4, CCL2), whereas mild OSA displayed a profile suggestive of innate activation and chemotaxis (IL-1β, IL-18, CXCL10, GM-CSF). Latent variable analysis demonstrated good discrimination between severity groups (AUC=0.87). Conclusion: OSA is highly prevalent in PD and is associated with worse motor performance and significant polysomnographic disruption. Beyond sleep fragmentation, OSA is linked to systemic inflammatory network reorganization, particularly in moderate-to-severe cases, suggesting a mechanistic connection between intermittent hypoxia and immune dysregulation in PD. Early detection and treatment of OSA may represent a modifiable strategy to mitigate inflammatory burden and potentially influence disease progression.