Parkinson's disease subtypes in the Oxford Parkinson disease centre (OPDC) discovery cohort

摘要: Background: Within Parkinson’s there is a spectrum of clinical features at presentation which may represent sub-types the disease. However no widely accepted consensus how best to group patients. Objective: Use data-driven approach unravel any heterogeneity in phenotype well-characterised, population-based incidence cohort. Methods: 769 consecutive patients, with mean disease duration 1.3 years, were assessed using broad range motor, cognitive and non-motor metrics. Multiple imputation was carried out chained equations deal missing data. We used an exploratory then confirmatory factor analysis determine suitable domains include within our cluster analysis. K-means scores all variables not loading into phenotypic subgroups. Results: Our found three important factors that characterised by: psychological well-being features; nontremor motor features, such as posture rigidity; features. subsequent five model identified groups by (1) mild (25.4%), (2) poor cognition (23.3%), (3) severe tremor (20.8%), (4) well-being, RBD sleep (18.9%), (5) (11.7%). Conclusion: several sub-groups driven largely dopaminergic-resistant (RBD, impaired posture, well-being) that, addition dopaminergic-responsive be for studying aetiology, progression, medication response early Parkinson’s.