Excessive daytime sleepiness increases dementia risk in cognitively unimpaired Parkinson’s disease patients over 12 years

Problem Statement
Parkinson's disease (PD) is characterized by a complex trajectory of motor and non-motor symptoms, with cognitive impairment affecting approximately 80% of patients over the disease course. This impairment ranges from cognitively normal (PD-CN) to mild cognitive impairment (PD-MCI) and finally to Parkinson's dementia (PDD). While excessive daytime sleepiness (EDS) is a prevalent non-motor symptom that often precedes motor diagnosis, its specific role as a predictor of cognitive decline remains inconsistent in the literature. Previous cross-sectional and longitudinal studies have yielded conflicting results regarding whether EDS predicts deficits in global cognition, executive function, or processing speed, potentially due to methodological variations, small sample sizes, and uncontrolled confounding variables. There is a critical need to clarify whether EDS serves as an independent marker for the conversion from PD-CN to PD-MCI or PDD, particularly in early-stage, de novo cohorts.

Methodology
This study utilized a longitudinal, multicenter dataset from the Parkinson's Precision Medicine Initiative (PPMI), tracking 1,147 initially cognitively normal PD patients and 266 cognitively normal (CN) older adults over a 12-year period. The PD cohort consisted of de novo, untreated participants diagnosed within two years of screening, while CN participants were matched by age, sex, and education.

• Assessments: EDS was measured using the Epworth Sleepiness Scale (ESS). Cognitive performance was evaluated annually using the Montreal Cognitive Assessment (MoCA) for global cognition, and specific tests for executive function and processing speed: Trail Making Test A and B (TMT-A/B), Verbal Fluency, Symbol Digit Modalities Test (SDMT), and Letter-Number Sequencing (LNS).
• Covariates: Analyses controlled for demographics (age, sex, education), disease severity (UPDRS-III), medication (Levodopa Equivalent Daily Dose or LEDD), neuropsychiatric symptoms (anxiety, depression), and REM sleep behavior disorder (RBD).
• Statistical Analysis:
  • Conversion Risk: Cox proportional hazards models were employed to assess the risk of three transitions: PD-CN to PD-MCI, PD-CN to PDD, and PD-MCI to PDD.
  • Cognitive Trajectories: Mixed-effect linear models (MELM) were used to examine the longitudinal association between ESS scores and cognitive performance over time in both PD and CN groups.
  • Correction: All models applied False Discovery Rate (FDR) correction to account for multiple comparisons.

Key Results

• Conversion to Dementia (PD-CN to PDD): Higher baseline ESS scores were significantly associated with an increased risk of converting from cognitively normal to dementia (Hazard Ratio [HR] = 1.41, p = 0.003). This association remained significant after FDR correction. Other significant predictors included older age, higher motor symptom severity, anxiety, and depression.
• Conversion to MCI (PD-CN to PD-MCI): In contrast, EDS was not a significant predictor for conversion to MCI. Instead, older age, greater RBD symptoms, and higher anxiety levels were the primary risk factors. Notably, higher LEDD was associated with a reduced risk of conversion to MCI.
• Conversion from MCI to PDD: Age was the only significant predictor for the transition from PD-MCI to PDD; EDS, RBD, and other clinical markers did not show significant associations in this specific transition.
• Cognitive Trajectories: Longitudinally, higher ESS scores in the PD group were associated with a decline in processing speed (SDMT performance) over time (FDR-corrected p = 0.011) and a trend toward lower global cognition (MoCA), though the latter did not survive FDR correction. No significant associations between EDS and cognitive decline were found in the CN group after FDR correction.

Key Contributions
This study provides the first longitudinal evidence, using a large de novo cohort with up to 12 years of follow-up, that EDS is an independent predictor of conversion to dementia in PD, distinct from its relationship with MCI. The research distinguishes the temporal roles of different sleep disturbances: while RBD appears to predict earlier cognitive decline (MCI), EDS emerges as a salient marker for progression to dementia. Furthermore, the study demonstrates that EDS is longitudinally associated with declining processing speed in PD patients, extending previous cross-sectional findings.

Significance and Claims
The authors conclude that EDS may serve as a clinically useful, early marker of cognitive vulnerability in Parkinson's disease, specifically identifying patients at risk for rapid progression to dementia. The findings suggest that EDS reflects underlying neurodegeneration of sleep-wake promoting nuclei that overlap with circuits implicated in PDD. Consequently, the paper advocates for the integration of ESS screening into cognitive monitoring protocols for PD patients. The study posits that combined screening of sleep disturbances (EDS and RBD) and neuropsychiatric symptoms could improve the predictive accuracy for different cognitive outcomes, aiding in better clinical prognosis and treatment planning. The authors remain cautious, noting that their interpretations regarding distinct neural mechanisms are speculative and require future neuroimaging validation.