Amsterdam 2013 Programme Satellite Meetings Registration Exhibition Virtual Exhibition Hotels Visa Letter Invitation
Search Abstracts by author or title
(results will display both Free Papers & Poster)
Back to Freepaper Session

A systematic review of safety and efficacy of epithelium removal and transepithelial corneal collagen cross-linking for keratoconus

Session Details

Session Title: Cross-linking

Session Date/Time: Monday 07/10/2013 | 14:30-16:30

Paper Time: 14:30

Venue: Elicium 2 (First Floor)

First Author: : Z.Shalchi UK

Co Author(s): :    M. Nanavaty              

Abstract Details


To review the safety and efficacy of epithelium removal (ER) and transepithelial (TE) corneal collagen crosslinking (CXL) for keratoconus.


East Surrey Hospital, Surrey & Sussex Healthcare NHS Trust, United Kingdom


We used MEDLINE to identify all ER and TE CXL studies in eyes with progressing keratoconus performed on a minimum of 20 eyes with at least 12 months of follow-up. We excluded animal and ex vivo studies, as well as studies where CXL was performed for any other indication or conducted in conjunction with other surgical treatments (e.g. intra-corneal rings, keratorefractive surgery). Data on uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), maximum keratometry (Kmax), spherical equivalent (SEQ) refraction and adverse events were recorded for all studies.


Thirty one ER (2,049 eyes) and 4 TE (185 eyes) studies were included. Following CXL, all ER studies but one showed an insignificant improvement in logMAR UDVA (median -0.14; range 0.00 to -0.37; p=0.12), as did all TE studies (median -0.26; range -0.23 to -0.30; p=0.18). All ER studies showed significant improvement in CDVA (median -0.10; range -0.01 to -0.19; p=0.01), whereas the TE group failed to reach significance (median -0.07; range -0.04 to -0.12; p=0.07). All ER studies showed significant reduction in Kmax (median -1.35D; range -0.08 to -4.34D; p=0.01), unlike the TE studies where Kmax did not decrease in all (median -0.03D; range -1.17 to +1.76D). All ER studies showed a significant reduction in myopic SEQ refraction, unlike TE studies (ER median +1.32D; range +0.81 to +2.15D; p=0.04 and TE median +0.55D; range +0.36 to +0.74D; p=0.18). ER studies reported stromal edema, haze, scar formation, and loss of CDVA in up to 70.0%, 12.7%, 9.5% and 18.9%, respectively. Failure, re-treatment and corneal transplantation occurred in up to 9.8%, 8.6% and 6.3% of eyes, respectively. In the TE group, haze occurred in up to 4% of patients but no other adverse events were reported.


Unlike TE, ER CXL is consistently associated with significant improvement in corrected visual acuity, flattening of cornea and reduction in myopic spherical equivalent. ER has greater efficacy than TE CXL for the treatment of keratoconus but also shows significantly higher adverse events. Techniques for TE CXL need modification and standardization to attain comparable efficacy. More TE studies are required with longer term follow-up.

Financial Interest:


loading Please wait while information is loading.