Evaluating The Safety And Efficacy Of Pack-Cxl As An Adjunctive Treatment In Infectious Corneal Ulcers

Infectious keratitis remains a vision-threatening ophthalmic emergency with substantial morbidity despite intensive antimicrobial therapy. Photoactivated Chromophore for Keratitis–Corneal Cross-Linking (PACK-CXL) has emerged as a potential adjunctive therapy for infectious keratitis. However, real-world outcomes vary, and evidence is limited by heterogeneity in ulcer severity, microbiological spectrum, and procedural parameters. This clinical audit aimed to evaluate the safety and efficacy of an accelerated high-fluence PACK-CXL protocol used as adjunctive therapy for infectious corneal ulcers at a UK tertiary centre.

James Cook University Hospital, UK, Ophthalmology Directorate (Cornea service). Patients undergoing PACK-CXL for infectious keratitis between January 2016 and December 2024 were included. Cases were identified through the Medisight audit tool, and data were extracted from electronic patient records (Evolve, Medisight, Heidelberg Imaging Platform) and physical case notes.

Design: Single-centre clinical audit evaluating outcomes after PACK-CXL performed for infectious keratitis.

Eligibility: Patients who underwent corneal cross-linking specifically for management of infectious keratitis/corneal ulceration.

PACK-CXL protocol: All cases received the same accelerated protocol: riboflavin soak for 10 minutes, followed by UVA irradiation at 30 mW/cm² for 4 minutes continuous (total energy 7.2 J/cm²). 

Outcome measures: Primary efficiacy endpoints were time to full re-epithelialisation (days from PACK-CXL to intact epithelial surface) and Ulcer resolution (complete stromal quiescence with no active infiltrate). Primary safety endpoints were occurance of complications or surgical escalation. 

Nine patients were included (mean age 63 ± 20 years). Ulcers were predominantly central (56%) with a mean surface area of 17.4 ± 11.5 mm² and mean ulcer size was 3.47 × 5.17 mm (largest diameter 8.3 mm); microbiological confirmation was obtained in 67%, including bacterial and fungal pathogens. Resolution occurred in 6/9 eyes (67%). Mean time to epithelial healing was 59 ± 43 days and mean time to resolution was 88 ± 81 days. Four eyes (44%) had no complications. Surgical escalation was required in three eyes. Two patients developed recurrent microbial infection. All healed eyes demonstrated visual improvement or stability.

In this cohort of complex infectious keratitis, accelerated PACK CXL achieved a two-thirds resolution rate with an acceptable safety profile, comparable to published refractory series. These findings support PACK CXL as a useful adjunctive therapy in selected cases and suggest benefit from earlier use in the treatment pathway.