Gait analysis with wearables predicts conversion to parkinson disease

Abstract

Objective: Quantification of gait with wearable technology is promising; recent cross-sectional studies showed that gait characteristics are potential prodromal markers for Parkinson disease (PD). The aim of this longitudinal prospective observational study was to establish gait impairments and trajectories in the prodromal phase of PD, identifying which gait characteristics are potentially early diagnostic markers of PD.

Methods: The 696 healthy controls (mean age = 63 ± 7 years) recruited in the Tubingen Evaluation of Risk Factors for Early Detection of Neurodegeneration study were included. Assessments were performed longitudinally 4 times at 2-year intervals, and people who converted to PD were identified. Participants were asked to walk at different speeds under single and dual tasking, with a wearable device placed on the lower back; 14 validated clinically relevant gait characteristics were quantified. Cox regression was used to examine whether gait at first visit could predict time to PD conversion after controlling for age and sex. Random effects linear mixed models (RELMs) were used to establish longitudinal trajectories of gait and model the latency between impaired gait and PD diagnosis.

Results: Sixteen participants were diagnosed with PD on average 4.5 years after first visit (converters; PDC). Higher step time variability and asymmetry of all gait characteristics were associated with a shorter time to PD diagnosis. RELMs indicated that gait (lower pace) deviates from that of non-PDC approximately 4 years prior to diagnosis.

Interpretation: Together with other prodromal markers, quantitative gait characteristics can play an important role in identifying prodromal PD and progression within this phase. ANN NEUROL 2019;86:357-367.

Figure 1

Clinic assessment of gait characteristics: (A) Example of body worn monitor placement and setup for walking data collection. (B) Raw vertical acceleration processing: data extraction and segmentation (signal segments in dotted black lines) from walking conditions (in order: single task at fast speed, single task at usual speed, dual task 1, and dual task 2). (C) Example of gait characteristic evaluation from a walking bout. (D) Data output: conceptual model of gait representing 5 domains and 14 gait characteristics.

Figure 2

Kaplan–Meier curves illustrating shorter time to Parkinson disease (PD) diagnosis using different combinations of first visit gait variables derived from univariate analysis and forward and backward stepwise Cox regression. (A) Greater step length asymmetry at first visit (adjusted for age and sex). (B) Combination of greater step length and swing time asymmetry (adjusted for age and sex). (C) Combination of greater step length asymmetry, higher variability of step velocity and step time, and quicker step time (adjusted for age and sex). Low and high risk were determined as higher than (high risk) and lower than (low risk) the hazard models created from models described in Table 2.

Figure 3

The graph illustrates step velocity and step length at usual speed prior to diagnosis of Parkinson disease (Parkinson disease converters [PDC]; gray lines indicate individuals, solid black line indicates PDC group mean) compared to the mean walking speed of non‐PDC at first visit (indicated by the dotted line). The 2 lines intersect approximately 3.3 years prior to PD diagnosis for gait velocity and 4.1 years for step length, indicating the latency between early changes of gait and diagnosis.