Tear Function and Ocular Surface Alterations After Accelerated Corneal Collagen Crosslinking in Progressive Keratoconus
We read with interest the study by Akcay et al.1 on tear function and ocular surface alterations after accelerated corneal crosslinking (CXL), which has recently been published in your journal. Authors did not identify any significant differences in the levels of tear osmolarity, tear break-up time, Schirmer test, Ocular Surface Disease Index scoring, and fluorescein and rose bengal ocular surface staining between pretreatment and 3 months postoperatively. However, conjunctival impression cytology showed metaplastic changes and a reduction in the density of the goblet cells in the superior and temporal bulbar conjunctiva, which, according to the authors, is probably induced by toxicity of ultraviolet-A irradiation 3 months after accelerated CXL.
Hereby, we would like to highlight two aspects, which merit further consideration. Authors express their concern about accidental irradiation and possible collateral effect of scattered ultraviolet-A light on limbal epithelial stem cells and conjunctival goblet cells, despite the absence of direct intentional irradiation of the limbal or conjunctival region during CXL. The authors erroneously interpret the decreased depth of demarcation line after accelerated CXL and the suggestion by Kymionis and Tsoulnaras2 to increase total energy in accelerated CXL, as proof of energy loss during accelerated CXL with subsequent reduction in efficacy, which they attribute to ultraviolet-A light scattering to adjacent tissues. It is well established that accelerated CXL with 9 mW/cm2 for 10 min, as used in this study, is as effective in halting the progression of keratoconus as standard Dresden protocol–based CXL without any requirement for increase of total energy, despite the more superficial demarcation line observed in accelerated CXL.3 However, the amount of CXL is less when accelerated CXL is applied, as shown in vitro by the group of Professor Meek in United Kingdom.4 The inherent factor in accelerated CXL which accounts for the superficial demarcation line is possibly associated to limitations of the Bunsen-Roscoe reciprocity law validity with increasing CXL intensity.5 There are controversial data in the literature regarding the potential effect of CXL on limbus, with all in vivo studies showing no negative impact on limbal stem cells, as also indicated by the authors.6 However, there is no evidence that ultraviolet-A light may scatter during CXL and therefore we consider the author's assumption unjustified.
Finally, we would like to comment on the authors' statement, that tear osmolarity is the most sensitive and specific diagnostic method for diagnosis and classification of dry eye disease. Their statement is followed by the citation of an irrelevant publication, which does not support in any evidence-based manner the validity of their argument. We strongly challenge this statement because we are not aware of any published study on the sensitivity and specificity of tear osmolarity in the diagnosis of dry eye disease. On the contrary, tear osmolarity as measured by TearLab is characterized by light variability, and its changes do not correlate significantly with changes in patient symptoms or corneal fluorescein staining in dry eye disease.7,8 As far as we know, there is also no official guideline by the International Dry Eyes Workshop or the Tear Film and Ocular Surface society on the clinical implication of tear osmolarity measurement in the diagnosis or classification of dry eye disease, confirming the plausibility of the authors' statement.