We read with great interest the recent article from Kanellopoulos and Asimellis. They concluded that corneal epithelial thickness measured using novel anterior segment optical coherence tomography (AS-OCT) was significantly thicker in dry eyes in comparison with age- and gender-matched control eyes. Even if the results are very interesting, we have a few comments regarding this study.
The most important limitation of this study, which the authors did not discuss, is the resolution of the device they used to measure corneal epithelial thickness. One can ask if it is clinically relevant to find a difference of 6 μm between 2 groups (53 μm in the control group vs 59 μm in the dry-eye group) when we know that this difference is approximately equal to the maximal resolution of the device (ie, 5 μm for the Fourier-domain AS-OCT system RTVue-100). We face this problem every day in our corneal department when we use this 3-dimensional epithelial thickness mapping for keratoconic and non-keratoconic eyes. To be applicable in clinical practice, we feel that a difference of more than 1.5 or 2 times the resolution should be considered. This limit in resolution may also explain the overlap in the corneal epithelial thickness between control and dry eyes.
In regard to the methodology and the results of the study, we believe that valuable information is lacking and that more details should be provided by the authors to permit other corneal specialists to reproduce their results.
In the abstract, the authors said that they conducted an age-matched case-control study; however, this matching does not appear in the Materials and Methods section of the article. Moreover, in the Results section, we observe that the dry-eye group was 3 years older than the control group (50.8 vs 47.5). Did the authors really conduct a matching on age? Or was the matching on age performed plus or minus 2 years? The authors should explain this discrepancy.
In the Materials and Methods section, the authors state that 4 individual acquisitions were performed in each case on the same day. Did the authors use, for their analysis, the mean value of the 4 acquisitions or the mean value of part of them? Or did they use only 1 acquisition? In that case, which one was selected, and why? Finally, did they observe a good repeatability among the 4 acquisitions?
In regard to the receiver operating characteristic analyses that were performed for the average and the central epithelium thickness, the most useful value in clinical practice, which the authors did not provide, is the best cut-off that maximizes sensitivity and specificity for the diagnosis of dry eye. This threshold value could be a new clinical indicator for dry eye syndrome and then be used in further investigation to distinguish between patients with dry eye and patients without dry eye. Can the authors provide this best cut-off for the average and the central epithelium thickness?
We commend the authors for this article, and we hope these remarks will help them to improve the accuracy of their analysis.