• Written By:
    Cornea/External Disease

    In this 7-year follow-up cohort study, O’Bart and colleagues report positive outcomes on the long-term effectiveness of corneal collagen cross-linking (CXL) to halt progression of keratoconus. The results are valuable for cornea specialists as only 1 other paper has a comparably long follow-up period (10 years).

    Study design

    The 36 enrolled eyes underwent CXL for keratoconus or post-refractive surgery ectasia using the original epithelium-off treatment protocol described by Wollensak: a 10-minute loading period for the 0.1% riboflavin and 20% dextran, followed by a 30-minute, 370-nm UVA exposure at 3 mW/cm2.


    The authors demonstrated a statistically significant reduction in spherical equivalent, mean cylindrical error and UDVA at 1 year compared with baseline, with results remaining stable to the 5- and 7-year follow-up marks. Mean CDVA improved significantly from baseline at year 1 (P<0.04) and was maintained to year 5. At year 7, CDVA improved further compared with values at years 1 (P<0.05) and 5 (P<0.03). Vision in the initially untreated fellow eyes, which served as controls, remained unchanged compared with baseline levels over a mean 68-month follow-up.

    Study eyes also showed significant decreases in mean maximum keratometry values (Kmax), simulated keratometry (SimK), root mean square corneal higher order aberrations and coma at 7 years. In addition, corneal thickness remained stable and no patients had corneal scarring, loss of transparency or change to the ocular surface. 

    No CXL eyes showed evidence of keratoconic or ectatic progression, and none required re-treatment. Meanwhile, untreated control eyes showed increased refractive cylinder, mean simK and mean KMax over the study period.


    The authors used a protocol with minimal riboflavin loading time and without slit-lamp confirmation of anterior chamber flare after loading.  This method may have limited the extent of treatment effect. Also, 2 patients were excluded from analysis due to intrastromal ring segment implantation subsequent to CXL. While no patients in the study group progressed, it is unknown whether perhaps poorer CXL outcomes necessitated further surgery in these 2 patients.

    Clinical significance

    Though the mid-term effectiveness of CXL is apparent, this study provides essential long-term data. O’Bart’s results demonstrate that CXL maintains refractive benefits, and they confirm a low incidence of significant complications by comparing with a contralateral control eye group. This paper bolsters support for the use of CXL in keratoconic and post-refractive surgery ectatic eyes.