Corneal crosslinking: riboflavin concentration in corneal stroma exposed with and without epithelium.

Purpose: To evaluate intrastromal concentrations of riboflavin with and without epithelium to ensure the efficacy and safety of corneal crosslinking (CXL) by the standard and transepithelial procedures.

Setting: Department of Ophthalmology and Department of Pharmacology G. Segre, Siena University, Siena, Italy.

Methods: This study comprised keratoconic patients enrolled for penetrating keratoplasty (PKP) and warm-stored sclerocorneal rings unsuitable for transplantation. Half the PKP specimens were debrided, and half were left with the epithelium in situ. One of the latter and 1 debrided sample were not exposed to riboflavin (controls). Samples in both groups were soaked with 0.1% riboflavin-dextran 20% solution instilled every 2 minutes for 5, 15, and 30 minutes. Riboflavin concentrations were determined by high-performance liquid chromatography (HPLC).

Results: The study evaluated 14 PKP specimens and 16 sclerocorneal rings. Control samples did not show a riboflavin emission peak. In exposed samples with epithelium, the mean riboflavin concentration was 91.88 ng/g after 5 minutes of exposure, 95.60 ng/g after 15 minutes, and 94.92 ng/g after 30 minutes. In the debrided samples, the mean riboflavin concentration was 14.42 microg/g, 20.92 microg/g, and 24.06 microg/g, respectively. No differences were seen between the in vivo samples and the ex vivo samples.

Conclusions: The HPLC quantitative study showed that stromal concentrations of riboflavin increased with exposure time only if the epithelium was removed. A theoretically safe and effective riboflavin concentration of 15 microg/g was obtained for ultraviolet A-induced CXL only after the epithelium was removed and after at least 10 minutes of riboflavin application every 2 minutes.