File
Lubetzky A.1, Kary E.1, Harel D.2, Hujsak B.3, Perlin K.4
1New York University, Physical Therapy, New York, United States, 2New York University, Humanities and Social Scienes, New York, United States, 3The Ear Institute, Hearing, and Balance Center Mount Sinai Health Sytem, Physical Therapy, New York, United States, 4New York University, Computer Science, New York, United States
Background: Falls are a major public health problem and can lead to devastating consequences and decreased quality of life.1 Falls typically occur within changing environments, as a result of unexpected external forces, and when multitasking. Physical therapy balance rehabilitation programs need to address the specific underlying impairments and environmental constraints leading to balance deficits, such as sensory deficits associated with vestibular dysfunction. For that, sensitive instruments are required. Nevertheless, Clinical assessments of sensory integration for balance, such as the 'foam/eyes closed' paradigm,2-4 are crude and primarily focus on single/dual-task assessment in static and predictable environments. Recent advances in virtual reality (VR) technology have created access to domains that were not accessible before. Using novel VR technology (Unity for Oculus), we created dynamic environments that could be used to measure balance reactions to various types of visual cues (e.g., different lighting, speed, distraction etc.).
Purpose: The purpose of this preliminary study was to test the feasibility of our new VR platform in healthy young adults (HEALTHY), describe their responses based on traditional measures of postural control, and explore their test- retest reliability.
Methods: Twenty-one HEALTHY volunteers viewed 6 different scenes, each repeated 3 times when standing on one of two surfaces: floor (forceplate) or two blue soft Thera Band® stability trainers. Their postural sway was recorded in response to: 4 different 60-second displays of moving spheres (medio-lateral 4mm amplitude at 0.48 or 0.56 Hz; anterior-posterior 5 or 32mm amplitude at 2Hz, performed on floor only); a ball flying towards their head every 6 seconds for 2 minutes with changing background environment every 30 seconds (Park scene); and standing in the middle of a busy street for 2 minutes (City) with changes in lighting or speed of movement every 30 seconds. We quantified participants center-of-pressure displacement and velocity in the anterior-posterior and medio-lateral planes and calculated intra-class correlations (ICC) between performances on two testing sessions performed 1-2 weeks apart.
Results: Participants reported slight to no side effects with a few moderate transient effects. Overall, postural sway was minimal on the floor with some increase on the stability trainers. ICC2,1 for the 4 park scene segments ranged between 0.66 to 0.91 on the floor and between 0.82 to 0.95 on the stability trainers. The range for spheres scenes was 0.66 to 0.89 (medio-lateral) and 0.42 to 0.75 (anterior-posterior). ICCs for the city scene segments on either surface and direction were 0.5-0.6.
Conclusion(s): Our VR platform was found to be feasible and reliable in HEALTHY. To enhance reliability, we chose to eliminate one of the spheres scenes found to be less reliable and to provide more specific instructions to the participants. We are currently testing this assessment in patients with vestibular dysfunction.
Implications: Our novel VR platform is portable, affordable and taps into domains currently not accessible in the clinic. If shown to be valid and reliable in adults with vestibular dysfunction, our VR platform could become a clinical assessment of sensory integration for balance and be further developed into a specific balance intervention.
Funding acknowledgements: We thank Helene Darmanine& Dipti Wani, (testing); Shahank Garg, (software) and David Lobser (VR software). This work was not funded.
Topic: Disability & rehabilitation
Ethics approval: This work was approved by the New York University Committee on Activities Involving Human Subjects (#IRB-FY2016-155)
All authors, affiliations and abstracts have been published as submitted.
