Decrease in head sway as a measure of sensory integration following vestibular rehabilitation: A randomized controlled trial

Objective: The purpose of this study was to determine the extent to which sensory integration strategies via head sway, derived from a Head-Mounted Display (HMD), change in people with vestibular disorders following vestibular rehabilitation. Design: Randomized Controlled Trial Setting: Vestibular Rehabilitation Clinic Participants: Thirty participants with vestibular dysfunction and 21 age-matched controls. Main Outcome Measures: Participants experienced two levels of visual surround (static or moving 'stars', front to back at 0.2Hz, 32mm) and white noise (none or rhythmic) while their head sway was recorded via the HTC Vive. We quantified head sway via Directional Path (DP) and Root Mean Square Velocity (RMSV) in 5 directions: anterior-posterior, medio-lateral, pitch, yaw, and roll and Power Spectral Density in low (PSD 1), medium (PSD 2) and high (PSD 3) frequencies in the anterior-posterior direction. Interventions: Participants performed the assessment prior to being randomized into 8-weeks of contextual sensory integration training in virtual reality or traditional vestibular rehabilitation and once again following completion of the intervention. Controls performed the assessment once. Twelve participants dropped out, half due to covid lock-down. We applied an intention to treat analysis. Results: We observed significant increases in AP DP, RMSV and all PSDs with change in visual level. Both intervention groups significantly decreased medio-lateral, pitch and roll DP and RMSV and anterior-posterior PSD 2 with no group differences. Vestibular participants were significantly higher than controls on all outcomes pre rehabilitation. Post rehabilitation they were only significantly higher on PSD 2. Sound was not a significant predictor of head sway in this protocol. Conclusions: Head sway decreased following vestibular rehabilitation regardless of visual load or type of intervention applied.