We are grateful for the interest shown by Dr Shoaib in our study. In our series, none of the cases had been treated with steroids—appropriately or otherwise—before seeking treatment at our center. As such, the severity of corneal thinning or perforation at the time of initial presentation clearly is unlikely to be the result of excessive steroid use.

With regard to a standardized treatment approach or criteria for treatment, in view of the significant range of disease severity of corneal involvement, a standardized approach toward frequency of steroids would not be realistic, but we would like to bring the author’s attention to the general overview and principles of therapy, as laid out in our “Methods” section.

We respectfully defend the use of the adjective severity in the title of our article, simply because our clinical data revealed that corneal involvement was disproportionately more severe in our series of Asian patients as compared with other published studies, and this was extensively addressed in our “Discussion” section. Because corneal involvement may be more immediately sight-threatening than lid involvement, we also believed it appropriate to grade overall disease primarily on the more sight-threatening corneal changes. Because our data also revealed that few, if any, of the relatively mild lid changes seen were sight threatening, it was evident that lid changes were less important in our series.

It is well established in the literature that inflammatory lid margin disease with accompanying secondary corneal changes represents a conglomeration of symptoms and signs with presumed similar etiologic and pathophysiologic mechanisms, most of which we detailed in Table 1. Some clinical entities such as angular blepharitis and seborrheic dermatitis were omitted because we did not encounter them. We avoided the term interstitial keratitis because it suggests inappropriate causes such as luetic disease. There is considerable overlap in the presumed inflammatory pathophysiologic features of these individual entities, and it is unlikely that the treatment options would differ among them. We agree that perhaps we specifically should have mentioned that inflammatory responses to microbial organisms may be an etiologic factor in some of these conditions. However, because this was a retrospective study, cultures were not available for most of the subjects, and the treatment in any case hinges on the inflammatory, rather than the infective, component. We do not believe that dermatologic evaluation is required routinely in subjects with limited facial or eyelid disease.

We thank Dr Shoaib for his comments and interest in our study.