Rose Bengal Electromagnetic Activation With Green Light for Infection Reduction: Follow-Up of a Randomized Clinical...

Infectious keratitis results in poor outcomes despite maximal medical therapy. While rose bengal photodynamic therapy (RB-PDT) has not been shown to be beneficial at improving best spectacle-corrected visual acuity (BSCVA) or cure rate, decreasing scar size, or complication rate at a shorter period of 6 months, long-term results are essential to evaluate clinically relevant scarring and visual recovery.

To evaluate 12-month outcomes from the clinical trial (REAGIR) comparing adjunctive rose bengal photodynamic therapy (RB-PDT) vs sham in treatment of fungal, acanthamoeba, and smear/culture-negative infectious keratitis.

This international multicenter, randomized, double-masked, sham-controlled clinical trial took place at Aravind Eye Hospitals in India (Madurai, Coimbatore, and Pondicherry) and the Federal University of São Paulo in Brazil. The study included a total of 330 participants with corneal ulcers who were randomized to the above groups. These data were analyzed from June 1, 2025, to July 15, 2025.

Participants were randomized to receive (1) a standardized loading dose of topical 0.1% rose bengal, followed by 15 minutes of green light irradiation (RB-PDT) or (2) identical procedure without activating the green light source (sham). All participants received standard antimicrobial therapy.

Prespecified primary outcome was BSCVA (logarithm of the minimum angle of resolution) at 6 months. Prespecified secondary outcomes included BSCVA, infiltrate and/or scar size at 12 months, corneal perforation (CP) and/or the rate of therapeutic penetrating keratoplasty (TPK), and microbiological cure rate at 12 months.

Of 330 enrolled patients (mean [SD] age, 50 [13] years; 213 male [65%] and 117 female [35%]), 282 (85%) had BSCVA measurements and 250 (75%) had 12-month infiltrate/scar measurements. There was no evidence of benefit of RB-PDT vs sham for BSCVA at 12 months (mean difference, 0.01; 95% CI -0.13 to 0.14; P = .91). Scar size was not different at 12 months (mean difference, 0.006 mm; 95% CI, -0.32 to 0.33; P = .97). CP/TPK rates were 31 vs 34 events at 12 months (hazard ratio, 1.21; 95% CI, 0.74-1.98; P = .44), indicating no difference. There was no difference in outcomes by organism subgroup.

In this study, at 12 months, RB-PDT did not confer a benefit over sham therapy for BSCVA, infiltrate/scar size, or rates of CP/TPK. These findings support 6-month REAGIR results, supporting the likelihood that there is no benefit to adjunctive RB-PDT for infectious keratitis at 1 year. These findings do not rule out the possibility that alternative photosensitizers or treatment algorithms might be beneficial.

ClinicalTrials.gov Identifier: NCT05110001.