Game Based Vestibular Exercise for Home Rehabilitation

Protocol

Design \& Methodology:

This study will adopt a randomized clinical trial design. Recruitment and screening (including diagnostics) will be coordinated by co-applicant, Dr. Jordon Hochman, department of Otolaryngology, Karen Reimer, and Andrea Giacobbo,; Physical Therapists.

Inclusion criteria:

Consist of (a) age 18 to 65 , (b) diagnosed with unilateral peripheral vestibular hypo function based on a detailed neuro-otological and neuro-orthoptic analysis to include binocular electro-oculography with caloric testing, and (c) have a home computer running Windows or Mac OS.

Exclusion criteria:

Consist of those with CNS disorders for example CVA, Multiple Sclerosis, epilepsy, migraines, concussion recent fractures of the lower extremities or vertebra, advanced hip/knee OA , and cardiac disease. Participants will be randomly assigned to either the Control group (active comparator) or Experimental group (therapeutic gaming). Participants will be blinded to the treatment. The treating therapist will know who is receiving the home or out-patient clinical program.

Control group will receive; (a) Herdman gaze stabilization exercises, (b) optokinetic exercises, and (c) balance exercises using sponge pads. These exercises are presently a standard of vestibular care. Participants will attend an out-patient physical therapy clinic once a week for 8-10 weeks. The program also includes a 20 minute home exercise program prescribed four times per week.

Experimental group: Will receive the game-based rehabilitation program delivered at home. The program for gaze will include provision of the motion mouse (Gyration). This approach provides a highly flexible treatment tool applied to train gaze control, and allows incorporation of graded balance demands. The use of this HID compliant computer input device allows many different therapeutic exercises to be coupled to a wide range of inexpensive commercial computer games. Though initially designed for clinical use, with the easy to use and inexpensive motion sense mouse and computer games this intervention approach can be extended to home settings. Many therapeutic exercises and activities (singly or in combination) can be used to practice a variety of gaze and balance skills, and, importantly, while playing fun computer games.

Participants will attend three clinical sessions during which time the home therapy programs will be established and training in the use of the computer, motion mouse and games will be provided. Based on clinical experience and pilot data, three sessions is sufficient before initiating the home program.16 The treating physiotherapist will attend the participant's home to ensure proper set-up and operation of the TRP equipment and computer applications. Participants will be asked to perform their respective home programs four times per week for 20 minutes per session. They will receive a phone call at one week to assess their ability to manage their home program. Thereafter participants will attend for clinical follow-up once per month.

Recording and Data Analysis The following information will be collected at baseline prior to start of the interventions; age, gender, work history, occupation and (history of disease/injury process, and current medications. Assessor will be blinded to participant assignments.

Primary Outcome Measures:

1. Clinical Test of Sensory Interaction in Balance (CTSIB); performance-based measure of balance under altered sensory conditions).
2. Dizziness Handicap Inventory (DHI), is a self-rating questionnaire based on 25 questions used to quantify a client's perception of his or her dizziness and its impact on their life.

Secondary outcome measures:

1. Computerized Visual tracking test (CVT), involves tracking a bright visual target moving horizontally left and right on a computer display in a sinusoidal fashion for several cycles, and at predetermined speeds. The application records (80 Hz) coordinates of both reference and head cursors. Coefficient of determination (COD) will be computed based on total and average residual difference between trajectories of the reference and head cursor motions. A value approaching one equates to excellent gaze performance. This test will be performed in standing and while walking on a treadmill at 0.9 m/s.
2. Participants in the experimental group will be given the visual tracking game application for home use. The game software automatically logs reference and head cursor coordinates (gaze performance) during the visual tracking task and saves the data to a coded and time stamped computer file (anatomized). Participants will be asked to play the tracking task (in standing at the beginning of every training session. This will involve three short tests of 45 seconds at three tracking speeds. The application automatically saves the logged data file, and the clients can either e-mail the files to the investigator, or save on a flash drive provided and return at a follow-up visit.
3. Spatiotemporal gait variables and walking stability measures. This is included to examine whether the exercise program transfers to walking. A treadmill instrumented with a pressure mat (Vista Medical, CA) will be used to record vertical foot forces for each step during walking trials of 1 minute at 0.9 m/s, and thus data for 40 consecutive steps.
4. Participants will be ask to complete a log of dates and duration of each exercise session, and submit either a paper version or by e-mail.

Statistical Analysis \& Power Assessment: Descriptive statistics, including means, standard deviations, frequencies, and percentages, will be used to describe the experimental and control groups on the baseline demographic variables. Differences between the groups on the outcomes will be tested using t-tests and tests of independence.

To achieve Objective 1, we will test the difference between experimental and control groups on continuous and normally distributed outcome measures using analysis of covariance, with the dependent variable being the post-intervention measurement of the outcome, the covariate being the pre-intervention measurement, and group membership as the between-subjects effect. Descriptive statistics, including measures of skewness and kurtosis, will be used to assess departures from the assumptions of a normal distribution of responses. If the distribution contains extreme observations, a robust ANCOVA statistic will be adopted. To test differences between experimental and control groups on discrete outcomes, such as counts of number of times participants feel dizzy during their daily activities, we will use a Poisson regression model, with the dependent variable being a count of the number of relevant events, the covariate being the pre-intervention measurement, and group membership as the between-subjects effect.

To analyze the longitudinal data associated with Objective 2, we will use a mixed-effects regression model with time as the random effect and group as the fixed effect. This model will be adopted because it accounts for clustering of repeated measurements within study participants and uses all available longitudinal data (i.e., it does not result in case-wise deletion of study participants due to missing observations.

Based on pilot data for the primary outcome (i.e., CTSIB), we conducted a power analysis of the required sample size to test the difference between the experimental and controls groups at the post-intervention measurement occasion (i.e., objective 1). The pilot data showed a standardized effect size of 0.80. Assuming the number of model covariates to be 3, the proportion of variance explained by these covariates to be 10%, and a two-tailed test of the null hypothesis of no group difference at p = .05, a sample size of 46 is required. We expect an attrition rate of 20% over the study observation period for objective 2. Given this, we propose to recruit a sample of N = 56 individuals to participate, with equal numbers to be recruited to the treatment and control groups.