Background/Objectives: Central retinal artery occlusion (CRAO) is a vision-threatening condition with limited evidence-based treatment options. Hyperbaric oxygen therapy (HBOT) has emerged as a potential intervention, but its efficacy remains debated. This systematic review and meta-analysis evaluated the therapeutic efficacy and safety of HBOT in CRAO. Methods: Relevant studies were identified across seven databases using optimized Boolean and MeSH-based strategies. Eligible studies evaluated HBOT in CRAO and reported visual or safety outcomes. Extracted data included demographics, intervention details, treatment timing, visual acuity outcomes, and adverse events. Risk of bias was assessed using ROBINS-I. Visual acuity outcomes were standardized to logMAR whenever directly reported or convertible, and subgroup analyses were stratified by HBOT initiation time (12 h), study type, and baseline visual severity when reported. A random-effects model was used, and pooled estimates were expressed as odds ratios (ORs) with 95% confidence intervals (CIs). Results: Twelve studies were included. The pooled efficacy analysis favored HBOT (OR = 0.47; 95% CI: 0.26–0.87; p = 0.02), although heterogeneity was substantial (Tau2 = 0.64; I2 = 78%). Stratified synthesis showed that studies in which HBOT was initiated within 12 h consistently reported greater visual improvement, whereas delayed or variably timed treatment showed attenuated and inconsistent benefits. After outcome harmonization, studies reporting logMAR-compatible data generally demonstrated clinically relevant visual improvement, while adverse-event rates did not differ significantly between HBOT and non-HBOT groups (OR = 0.70; 95% CI: 0.43–1.16; p = 0.17; I2 = 0%). Conclusions: HBOT appears most beneficial when initiated early, particularly within the first 12 h. However, heterogeneity in treatment timing, study design, and baseline severity reporting limits the certainty of these results and supports the need for standardized outcome reporting and protocol-driven prospective studies.
ALBalawi et al. (Tue,) studied this question.