Objective: The aim of this study was to investigate if free-living gait monitoring can discriminate individuals with REM sleep behaviour disorder (RBD) from controls.

Background: Parkinson’s disease (PD) is a progressive multisystem disease. A prodromal period precedes the onset of definitive motor/non-motor symptoms. RBD is closely associated with abnormal accumulations of α-synuclein and is a risk factor for the subsequent development of a synucleinopathy including PD. Recently RBD has been associated with subtle changes in gait [1]; however these data were obtained in a clinical setting. We were interested to see if gait measured in free-living/habitual conditions in RBD using a body worn sensor could identify early subtle differences that may have potential as clinical markers of prodromal PD and discriminate individuals with RBD from controls.

Methods: 63 individuals with RBD (66±10 years) and 35 age-matched controls (CL) recruited in the Oxford Parkinson’s Disease Centre Discovery Study were assessed. Data were collected continuously for 7 days with a device placed on the lower back. We measured Macro gait characteristics (volume, pattern, and variability of free-living gait), and 14 Micro gait characteristics previously validated [2]. ANCOVA was used to examine between-group differences, and we explored the predictive validity to identify RBD using AUC and stepwise, forward, logistic regression analyses.

Results: No significant differences were found for Macro characteristics; individuals with RBD tended to reduce volume of walking, decrease variability, and walk with a higher number of shorter bouts compared to CL. Individuals with RBD walked with significantly slower gait, reduced step velocity variability and cadence compared to CL (p≤0.044). Step velocity, step velocity variability and rhythm (step, swing and stance time) significantly predicted RBD (AUC≥0.620). Swing time was the best predictor (sensitivity 92%, specificity 20%, accuracy 66%).

Conclusions: Our pilot results show early differences in gait under free-living conditions and suggest that discrete gait characteristics may have utility as markers of prodromal PD. Gait assessment in this way is low cost and potentially easy to implement. We see this as a useful basis for further analyses with the existing dataset and for validation purposes in independent cohort studies.

References: [1] McDade EM et al., Mov Disord, 2013; 28(13):1847-53. [2] Del Din et al., J Gerontol (A Biol Sci Med Sci), 2017; glx254.

« Back to 2018 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/free-living-gait-in-rem-sleep-behaviour-disorder-measures-of-prodromal-parkinsons-disease/