Variation in Recent Onset Parkinson's Disease: Implications for Prodromal Detection.

Background: The detection of prodromal Parkinson's disease (PD) is desirable to test drugs with neuroprotective potential, but will be affected by known disease variations. Objective: To assess the prevalence of four key non-motor prodromal PD markers, and evaluate the sensitivity of case detection when non-motor screening tools for prodromal PD are implemented in an early clinical PD cohort. Methods: Hyposmia (University of Pennsylvania smell identification test <= 15th centile or Sniffin' Sticks at or <= 10th centile corrected for age and sex), rapid-eye movement sleep behaviour disorder (RBD questionnaire >4), constipation (<1 daily spontaneous bowel motion) and depression (Leeds >6) were recorded in recent onset PD cases, and proposed non-motor screening criteria applied. Results: In 1,719 PD cases, mean age 68.6 years (SD 8.1), 65.5% male, mean disease duration 1.3 years (SD 0.9), 72.2% were hyposmic, 43.3% had RBD, 22.1% depression, and 21.5% constipation. 11.6% of cases had no key non-motor features, 38.8% one, 32.1% two, 15.5% three, and 2.0% all four. Increasing numbers of non-motor features were associated with younger age (p = 0.019), higher motor scores (p <0.001), more postural instability gait difficulty (PIGD) (p <0.001), greater cognitive impairment (p <0.001) and higher total non-motor burden (p <0.001). Cases with hyposmia alone were younger (p <0.001), had less severe cognitive (p = 0.006) and other non-motor features (p <0.001). All screening criteria selected younger patients (p = 0.001, p <0.001), three of four greater overall non-motor burden (p = 0.005, p <0.001), and inclusion of RBD more cognitive impairment (p = 0.003, p = 0.001) and PIGD (p = 0.004, p = 0.001). Conclusions: Varying sensitivity levels, and age and phenotype selectivity, are found when different non-motor screening methods to detect prodromal PD are applied to an early clinical PD cohort.