Digital Defocus Vision Training Using Virtual Reality for Myopia Control

Importance: Virtual reality-based digital defocus vision training (DDVT) may offer a home-based strategy to slow myopia progression in children, but evidence is limited. Objective: To evaluate efficacy and safety of a head-mounted virtual reality-based DDVT system on myopia progression. Design, Setting, and Participants: This randomized clinical trial was conducted among children aged 6 to 12 years with myopia enrolled in July 2024, with follow-up completed in September 2025. The trial was conducted at a single center in Guangzhou, China. Data were analyzed from September 2025 to October 2025. Intervention: Participants were randomized to DDVT plus single-vision spectacles (SVS) or SVS alone. Participants in the DDVT group used a head-mounted virtual reality device at home for 15 minutes per day under parental supervision after an initial in-hospital training session. Main Outcomes and Measures: The primary outcome was the between-group difference in axial length (AL) change from baseline to 6 months. The secondary outcome was between-group difference in spherical equivalent refraction (SER) change. Results: Of 120 randomized participants, 57 of 60 participants in the DDVT group (95%) and 59 of 60 participants in the SVS group (98%) completed the study. At baseline, mean (SD) ages in the DDVT and SVS groups were 9.42 (1.52) years and 9.31 (1.13) years, respectively, and 53 of 116 participants (45.7%) were female. Mean (SD) AL and spherical equivalent in the DDVT and SVS groups were 24.35 (0.81) mm and 24.40 (0.81) mm and -1.91 (1.11) diopter (D) and -2.04 (1.06) D, respectively. Mean best-corrected visual acuity was -0.04 logMAR (Snellen equivalent, 20/18) and -0.03 logMAR (20/19), respectively. At 6 months, AL increased by 0.15 mm and 0.25 mm in the DDVT and SVS groups, respectively (difference = 0.11 mm; 95% CI, 0.07-0.15 mm; P < .001). SER changed by -0.23 D and -0.46 D (difference = -0.23 D; 95% CI, -0.35 to -0.11 D; P = .003). Visual acuity change at 6 months was -0.02 logMAR (Snellen equivalent, 20/18) and 0.01 logMAR (20/20) in the DDVT and SVS groups, respectively (difference = 0.03 logMAR [Snellen equivalent, 20/21]; 95% CI, 0.02-0.05 [20/21-20/22]; P < .001). No training-related adverse events were observed. Conclusions and Relevance: In this randomized clinical trial, the DDVT group had a small absolute reduction in AL and SER myopic progression compared with SVS alone over 6 months in 120 children with myopia aged 6 to 12 years. The clinical relevance of these differences over the short and long term cannot be determined from this trial but support additional studies to assess this intervention as a promising adjunctive approach for pediatric myopia control. Trial Registration: ClinicalTrials.gov Identifier: NCT07042022.