We read with interest the article by Hamrah and associates, in which the authors aimed to measure and correlate epithelial and keratocyte densities with corneal sensation and subbasal nerve densities in patients with herpes zoster ophthalmicus (HZO). The rationale is sound, as subbasal nerves are often damaged in herpes simplex virus (HSV) keratitis as well as in HZO and in vivo confocal microscopy (IVCM) is a valuable tool that can quantitatively demonstrate nerve damage in these patients. However, we were surprised to observe that although the number of nerves and total length of nerves were identified as main outcome measures and subbasal corneal nerves were included among the measured parameters in the Methods, nerve densities and their correlative analyses were not presented in the Results section. Furthermore, nerve density data of the current study were also not mentioned in the Discussion. Instead, the authors commented on their findings from their previous works, in which bilateral nerve involvement had been found in cases of apparently unilateral clinical involvement of HZO and HSV keratitis. This missing correlation link between epithelial density and subbasal nerves would have been fundamental to elucidating the effects of subbasal nerve damage on epithelial cell health and turnover in these patients.
Another controversial issue is the utilization of superficial epithelial density in IVCM. Determination of superficial epithelial cell densities is problematic in that they cannot be reproducibly visualized with IVCM and their quantification is not as accurate as that of basal epithelial cells. In addition, the superficial epithelium may manifest different imaging characteristics with IVCM following the application of fluorescein dye, as was previously described by our group.
The authors also aimed to analyze keratocyte densities in HZO patients, but it is unclear why they chose to do so. The authors argue that subbasal nerve loss could have an impact on keratocyte density, but they do not present any scientific evidence from previous clinical or basic science studies to support their claim. Moreover, they do not provide their data on the relationship between subbasal nerves and keratocytes and they do not discuss the ramifications of their finding of unaltered keratocyte densities in HZO. The readers would be interested to know how the findings of unaltered keratocyte densities contribute to the disease pathogenesis. Since the authors analyzed the posterior corneal stroma, it is also puzzling not to see endothelial layer evaluation in this study.
A final question remains as to whether the data presented are sufficient to justify the concluding remarks, “An objective quantitative evaluation of epithelial cells can provide valuable clues to detect subclinical neurotrophic keratopathy before intractable clinical findings occur.” Although it is unclear from the data as to how many subjects with severe vs mild corneal sensation loss were analyzed in this study, the presence of severe sensation loss suggests the inclusion of cases with overt neurotrophic keratitis. Neurotrophic keratitis is associated with persistent and frequently generalized epithelial defects, and thus correlating the biomicroscopic findings with IVCM results in corneas at different stages of neurotrophic keratopathy may assist in understanding the progression of this debilitating disease.