Weekly Home Monitoring of Glaucoma with Virtual Reality Perimetry: A 2-Year Prospective Longitudinal Study.

PURPOSE: To evaluate the feasibility, and adherence of unsupervised high-frequency home virtual reality (VR) perimetry and its agreement with standard automated perimetry in open-angle glaucoma and to compare the precision of progression estimates. DESIGN: Prospective, longitudinal cohort study. PARTICIPANTS: Fifty-three patients (106 eyes) with open-angle glaucoma. METHODS: Participants performed weekly home testing with the Olleyes VisuALL VR perimeter (HomeVF) during longitudinal follow-up. In-office Humphrey Field Analyzer (HVF) 24-2 SITA Standard testing was performed every 6 months. Agreement in mean deviation (MD) and pattern standard deviation (PSD) was assessed using repeated-measures correlation and Bland-Altman analysis. Baseline estimate variability and standard errors (SEs) of the slopes were compared between modalities. MAIN OUTCOME MEASURES: Primary outcome was HomeVF-HVF agreement for MD and PSD. Secondary outcomes included adherence, reliability, test duration. Exploratory outcomes included test-retest variability and slope-estimate precision. RESULTS: Fifty-three participants were enrolled; 51 contributed follow-up data, and 2 contributed baseline data only. The mean follow-up was 16.6 ± 8.3 months (range, 2.1-30.5 months), and a total of 4,817 HomeVF tests were collected, with a median of 77 tests per participant (interquartile range, 36-130). At 24 months, 86.7% of eligible participants remained active, with a mean frequency of >2 tests per month. Mean HomeVF duration was shorter than HVF (4.93 ± 0.66 vs 5.80 ± 1.09 minutes; P < 0.001). Overall, the median percentage of reliable tests/eye was 98.4%. HomeVF correlated strongly with HVF (MD r = 0.83-0.94; PSD r = 0.81-0.86) across all time points. In paired HVF-aligned comparisons through 12 months, HomeVF values were consistently higher than HVF values (mean difference [HomeVF - HVF]: MD +1.42 to +1.46 dB; PSD +0.78 to +1.34 dB). Intertest variability was similar between the two modalities. The MD slope SEs were significantly lower with HomeVF (0.40 vs 0.53 dB/year; P = 0.049). CONCLUSIONS: Unsupervised home VR perimetry is feasible over long-term follow-up, demonstrating sustained adherence, high reliability, comparable intertest repeatability, and strong agreement with office HVF with predictable offsets. Higher-frequency home testing improves the precision of MD progression estimates, supporting home VR perimetry as a practical adjunct for longitudinal glaucoma monitoring.