Effects of Different Doses of Virtual Reality Therapy in Chronic Stroke Patients

Study Design

This was a prospective, single-blind (assessor-blinded), randomized controlled trial conducted at Istanbul Physical Medicine Rehabilitation Training and Research Hospital. Ethical approval was obtained from the Istanbul University-Cerrahpasa Clinical Research Ethics Committee (Protocol No: 2025/28). Written informed consent was obtained from all participants prior to enrollment.

Participants

Patients with chronic stroke (onset >6 months) aged 20-75 years were screened for eligibility. Inclusion criteria: diagnosed with ischemic or hemorrhagic stroke, Mini-Mental State Examination (MMSE) score ≥18. Patients were excluded if they had an MMSE score below 18, a diagnosis of epilepsy, a diagnosis of intellectual disability, or visual and/or hearing impairment at a level that would affect participation in treatment.

Randomization

Eligible patients were randomized into three groups using a sealed envelope method: Control group (n=14), VR3 group (n=15), and VR5 group (n=14).

Interventions

All three groups received conventional physiotherapy and rehabilitation (45 min/session, 5 days/week, 4 weeks), including neurodevelopmental treatment, strengthening, and balance exercises. In addition to conventional therapy, the VR3 group received Wii Fit-based VR balance training 3 days/week (20 min/session, 12 sessions total) and the VR5 group received VR balance training 5 days/week (20 min/session, 20 sessions total). VR training was delivered using a Nintendo Wii Balance Board modified with four 16-bit pressure sensors (Becure Balance System). During the first two weeks, games were applied at beginner difficulty level with two repetitions per game; the final two weeks used advanced difficulty with the same repetition scheme. A physiatrist supervised all sessions for safety. Following the completion of the 4-week treatment period, all patients continued to receive conventional rehabilitation only during the 4-week follow-up interval between T1 and T2 assessments; no VR training was provided during this period.

Outcome Measures

Primary outcome: Berg Balance Scale (BBS; 0-56 points).

Secondary outcomes: Functional Independence Measure (FIM; 18-126 points), Stroke-Specific Quality of Life Scale (SS-QOL; 12 domains, 49 items), Becure Balance System center of pressure (COP) measurements under eyes-open and eyes-closed conditions (X, Y, D parameters), and static postural alignment via Becure Posture Mobile application (anterior, posterior, and lateral planes).

Assessment Timepoints

T0: Baseline (before treatment); T1: Post-treatment (immediately after 4 weeks); T2: Follow-up (4 weeks after treatment completion). All assessments were performed by an evaluator blinded to group allocation.

Statistical Analysis

Sample size was calculated based on the BBS with a medium effect size (partial eta squared=0.06), α=0.05, and power=0.85, yielding a minimum of 13 patients per group (39 total); accounting for 15% dropout, 45 patients were enrolled. Normality was assessed using the Shapiro-Wilk test. Between-group comparisons used One-Way ANOVA or Kruskal-Wallis test. Within-group changes over time were analyzed using repeated measures ANOVA or Friedman's test with Bonferroni correction. Group × time interactions were examined using two-way mixed ANOVA; when ANOVA assumptions were violated, the nparLD package (F1-LD-F1 design) in R was used. Effect sizes were reported as partial eta squared (η²p): small=0.01, medium=0.06, large=0.14. Baseline differences were controlled using ANCOVA. Statistical significance was set at p<0.05. Analyses were performed using IBM SPSS Statistics 26.0 and R/RStudio.