Abstract
Sleep disturbances are increasingly recognized as early and clinically meaningful features in the pathophysiology of neurodegenerative diseases. Polysomnography (PSG), i.e., the gold standard for objectively characterizing sleep architecture, may provide non-invasive and scalable biomarkers for both early detection and differential diagnosis of dementia. This systematic review and network meta-analysis aims to synthesize existing evidence on PSG-derived sleep alterations across the neurodegenerative continuum, including subjective cognitive impairment, mild cognitive impairment, Alzheimer's disease, frontotemporal dementia, Lewy body dementia, and Parkinson's disease dementia, compared to healthy controls. It will adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines for reporting systematic reviews that include network meta-analyses, and it has been registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251114418. We will include peer-reviewed studies with nocturnal PSG data from adult participants, categorized using validated diagnostic criteria, and scored according to the most recent American Academy of Sleep Medicine guidelines. Both pairwise and network meta-analyses will be conducted using standardized mean differences to quantify group-level effects. Additional analyses will explore the correlations between PSG parameters, severity of cognitive impairment, behavioral symptoms, treatments, and relevant comorbidities. Longitudinal data, where available, will be analyzed separately to evaluate prognostic value. This study will provide a comprehensive synthesis of PSG alterations across neurodegenerative disorders, offering insights into their diagnostic utility and potential as early markers for stratification in clinical trials of disease-modifying therapies.