EP 90. Combining motion capture and accelerometers for the assessment of rotation and acting forces in bilateral vestibulopathy patients during walking tasks

Abstract

Background The vestibular system is essential for orientation, eye coordination and balance. Disorders of the vestibular system are more common in older people and present with symptoms such as imbalance which is usually compensated with an increase in sensory loading to the visual and proprioceptive systems. This compensation is often insufficient in cases of increasing demands during turning, different gait speeds and cognitive loading. In this study, we investigated the relationship of the position and forces acting on the semicircular canal (Reid’s plane) during such gait tasks and its effects on gait trajectory and stability. Furthermore, we attempted to quantify the effects of these forces and positions on the gait cycle. Methods 10 young healthy subjects (20–40 years), 10 older healthy subjects (40–60 years), and 10 bilateral vestibulopathy patients (BVP) were recruited for the study. Motion capturing was achieved using Qualisys® with 8 Oqus 100 cameras sampling at 128 Hz. 11 retro-reflective markers were attached to the head and trunk (4 for the Reid’s plane) of each subject ( Fig. 1 a). 2 Synchronized Opal® sensors (tri-axial accelerometers, gyroscopes, magnetometers: Mobility lab® – APDM Devices) were attached to the head and trunk of the subjects sampling at 128 Hz. Subjects were asked to walk on along an elongated circle twice (circumference: 12.6 m) at three different speeds: preferred; slow and fast, tandem gait, and cognitive dual task (serial subtraction). Additionally a Timed-up-and-Go test was performed along the long axis. Data analysis was performed using MATLAB® R2012a. Results Mean deviations from the optimal path in BVP patients (slow speed: 5.23 ± 0.96; preferred speed: 4.48 ± 0.78 ;maximal speed: 3.19 ± 7.36, meters) was larger than that of younger healthy subjects (slow speed: 3.78 ± 0.76; preferred speed: 3.12 ± 0.37; maximal speed: 2.72 ± 0.38, meters, P = 0.037; 0.002;0.267; respectively) and older healthy subjects (slow speed: 2.95 ± 0.37 ; preferred speed: 2.95 ± 0.08; maximal speed: 2.83 ± 0.36, meters, P = 0.006; 0.002; 0.600 respectively) for all gait speeds except maximal speed gait. Phase difference between head and trunk angles revealed significant differences between groups during slow speed gait (F = 5.306, P = 0.013) and maximal speed gait (F = 3.455, P = 0.048) but not preferred speed gait (F = 2.871, P = 0.077). While post-hoc analysis was not significant within groups, BVP had higher deviations during slow speed and preferred speed (7.81°) compared to maximal speed gait (2.83°). In comparison older healthy subjects had greater deviations during maximal speed gait (8.00°) compared to slow (0.5°) and preferred speed gait (3.37°). Conclusions This study not only confirms the speed dependent nature of sensory processing during gait but also demonstrates its relationship to gait trajectory. Trajectory deviations are higher during slow and preferred speed gait in BVP compared to maximal speed gait due to the decreased sensory inputs during this mode. Aging leads to a loss of dynamic range in trajectory correction. Significant differences between head and trunk phase angles is seen during non-preferred gait signifying requirement of additional visual sensory information to maintain stability during these modes.