This meta-analysis compared 1-year outcomes for standard vs transepithelial epithelium-off corneal collagen crosslinking (CXL).

Study design

The analysis included 8 studies from a review of the following databases: PubMed, EMBASE, the Cochrane Library, ClinicalTrials.gov, VIP Database, Wanfang Database, and China National Knowledge Infrastructure. The studies encompassed 228 eyes that underwent standard epi-on CXL and 227 eyes that had epi-off CXL.

Outcomes

At 1 year, there was a statistically significant difference in reduction of mean K favoring the epi-off CXL group (mean difference of 0.28 D; P=0.03). Subgroup analysis revealed that standard CXL had a similar effect on reducing mean K with epi-on CXL protocols when using chemical enhancers (P=0.82). However, standard CXL was still superior to epi-on CXL with iontophoresis in reducing mean K (P=0.01).

There were no statistically significant differences seen in Kmax, corrected and uncorrected distance visual acuities, mean spherical equivalent, central corneal thickness, and endothelial cell density.

Complications occurring only in the standard CXL group included various grades of stromal haze and transient corneal edema; none of these adverse effects were seen in the epi-on CXL group.

Limitations

Results were only compared at 1-year post-treatment, a relatively short time to definitively comment on long-term comparisons between CXL treatment methods. Iontophoresis protocols have changed and improved significantly since the time of these studies; it is difficult to make any forgone conclusions about its efficacy without further investigation.

Clinical significance

This meta-analysis suggests that standard and transepithelial CXL have similar results in many parameters when evaluated 1-year post-treatment. There was, however, a statistically significant improvement in mean K when comparing standard CXL vs iontophoresis-assisted transepithelial CXL. These differences were not seen when comparing against chemically enhanced transepithelial CXL. It is noteworthy that iontophoresis protocols have continuously improved and future studies may result in different findings. Notably, corneal haze was only seen in the epi-off group, a known complication of any procedure involving epithelial debridement.