We read with great interest the manuscript of Alessio and associates titled “Confocal microscopy analysis of corneal changes after photorefractive keratectomy plus cross-linking for keratoconus: 4-year follow up.” This publication is important, as it examines the long-term structural alterations of the corneal tissue after combined treatment of photoablation and corneal cross-linking. In addition, the nature of the study (prospective) and the methodology followed make it objective and scientifically sound.

The mechanisms of action of cross-linking are not yet fully understood, and studies that provide information of corneal behavior after this treatment are important for the scientific community. The first reports of corneal confocal microscopy concerning the original cross-linking procedure, which did not include corneal photoablation, demonstrated a series of structural alterations, with the most important being the absence of keratocytes from the anterior 300 microns of the corneal stroma ; evidence of keratocyte repopulation was reported after 6–12 months post cross-linking. Most studies seem to agree that the cornea “macro-structurally” returns to preoperative state 1 year after cross-linking ; this does not seem to be the case when cross-linking is combined with photoablation of the cornea. The prominent finding after this combined treatment is the absence of keratocytes from the anterior and mid corneal stroma in more than 50% of the patients, as Alessio and associates report, even 4 years after cross-linking.

The failure of corneal tissue to restore its structural form is significant and yet not understood; furthermore, we may not predict the possible implications on corneal homeostasis in the absence of stromal keratocytes in the long term. The authors in their study present their outcomes after combined treatment, while they choose not to present the fellow eyes of their patients, which received the standard cross-linking treatment. Even though we understand their reasoning, a prospective, comparative 4-year follow-up study would offer significant information to the scientific community, as confocal microscopy evaluation is not yet fully objective (it is dependent on the experience of the users and the software variations between instruments) and a direct comparison between studies may not be established. The authors here have a great opportunity to present their comparative findings and provide significant information about corneal behavior after cross-linking treatment and cross-linking with photoablation.

The main corneal structural difference between the treatment modalities in question is the absence of the Bowman membrane (owing to ablation); as the authors mention in their discussion, the ablation of the Bowman may lead to a deeper penetration of riboflavin into the corneal stroma, resulting in a different treatment outcome. We would like to add here another implication that the absence of the Bowman membrane might have on cross-linking. It has been demonstrated that the Bowman membrane plays a significant role in ultraviolet (UV) light absorption; therefore its absence may be a factor facilitating higher UV-A absorption from the corneal stroma, leading to a more intense cross-linking treatment (higher absorption of UV-A would induce a stronger photochemical reaction between UV-A and intrastromal riboflavin ). A final variable on which we would comment is whether the authors used mitomycin C after photoablation, which may also have contributed to their finding (keratocyte absence).

We would like to congratulate the authors for their great work and we hope that our scientific interaction will further improve our understanding of cross-linking treatment.