Obstructive Sleep Apnea is a Risk Factor for Parkinson’s Disease and CPAP Mitigates Risk of PD: An EHR-based Cohort Study in Military Veterans
Isabella Montano1, Jasmine L. May1, Lee Neilson1, Yeilim Cho2, Jeffrey Iliff2, Jonathan E. Elliott1, Gregory Scott1, Miranda Lim1
1VA Portland Health Care System, 2VA Puget Sound Health Care System
Objective:
To measure the risk of Parkinson’s Disease (PD) in those with obstructive sleep apnea (OSA) and evaluate the impact of early versus late Continuous Positive Airway Pressure (CPAP) treatment.    
Background:
Previous studies have shown an association between OSA and various synucleinopathies, but a causal link between OSA and PD is unknown. This study evaluated OSA as a potential risk factor for PD by leveraging the VA Corporate Data Warehouse (CDW) spanning 20+ years and 20+ million Veterans.   
Design/Methods:
OSA was defined by ICD-10 code G47.33 (OSA+). Outcomes included PD and death from any cause. We validated PD diagnosis previously using 2 case definitions (PPV 76% and 90%; PMID 37309872). CPAP use was determined using HealthFactor field, a semi-structured field containing data from medical interviews. Only 9.9% of Veterans had mentions of CPAP in this field (CPAP+); therefore, individuals missing HealthFactor data were excluded from the secondary CPAP analysis. “CPAP+ Early” was defined as mentions of CPAP within 2 years of OSA diagnosis; “CPAP+ Late” indicated >2 years since OSA diagnosis.  
Results:
1,552,505 OSA+ and 9,759,246 OSA- Veterans were identified. After Inverted Probability Treatment Weighting analysis, balancing of birthyear/age, sex, smoking status, race, ethnicity, pseudo-randomization by covariates, and adjustment for competing risk of death, OSA+ was associated with a significantly increased incidence of PD with 1.8 [1.4, 2.3] 95% confidence intervals, p < 0.001) extra cases of PD per 1000 people at 5 years after OSA onset. “CPAP+ Late” had a similar incidence of PD to CPAP-In contrast, CPAP+ Early” had significantly lower incidence of PD, with a reduction of 2.3 cases of PD (p < 0.001) 5 years after OSA.
Conclusions:
Results indicate that OSA may be an important, modifiable risk factor for the development of PD and potentially other synucleinopathies.
10.1212/WNL.0000000000210388
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.