Corneal collagen cross-linking for keratoconic cornea pretreated with PRK.

To the Editor: We read the recent article by Tuwairqi and Sinjab,1 which appeared in the May 2012 issue of the Journal of Refractive Surgery, and would like to offer some comments. Tuwairqi and Sinjab, in this prospective and noncontrolled study, performed corneal collagen cross-linking (CXL) in patients with stage 1 or 2 keratoconus pretreated with topography-guided photorefractive keratectomy (PRK) and demonstrated the improvements in visual acuity and refraction during 1-year follow-up. The aim of this combination, as the authors stated, was to remodel the irregular cornea and decrease the irregular astigmatism to enhance the fi nal visual and refractive outcomes. In this context, the application of PRK is to reshape the irregular cornea while CXL is applied to arrest the progression of keratoconus. Questions therefore arise about whether PRK substantially infl uences the biomechanics of the keratoconic cornea, because both the Bowman layer and central anterior stroma, which are thought to be the strongest and the second strongest region of the cornea,2 respectively, are ablated in this combined procedure. The biomechanical change in the keratoconic cornea is assumed to occur in the anterior rather than posterior cornea,3 which is evidenced by the achieved optimal outcomes of CXL as a result of the corneal stiffening in the anterior stroma in keratconic eyes. This assumption can be further supported by modifi ed transepithelial CXL where the formation of cross-links occurs in the upper third of the corneal stroma, 20 to 30 μm beneath Bowman layer, and the 20to 30-μm-thick cross-links can serve to halt the progression of ectasia.4 Absence of the Bowman membrane and anterior corneal stroma may not elicit corneal ectasia in the normal cornea as seen in most PRK cases; however, there is no evidence that this is the case in the keratoconic cornea. Although signifi cantly decreased keratometry reading was observed in this study, studies with larger sample size and longer follow-up are needed to further assess this combined procedure. Additionally, it is noted that mitomycin C was used to avoid haze formation in this study; its infl uence on the biomechanics of a keratoconic cornea, however, also needs evaluation. For the treatment of keratoconus, what should be preferentially considered is no elevated risk for disease progression rather than improvement in visual acuity. Zhen-Yong Zhang, MD Xing-Ru Zhang, MS Shanghai, China