Dr Dille raises an interesting question about the role of eye rubbing as a potential confounder in the associations between elevated serum immunoglobulin E (IgE) levels to cat and cockroach antigens and glaucoma that were reported in our study. While eye rubbing is an interesting characteristic that may potentially mediate an association between allergies and glaucoma, there were no questionnaire or examination data collected in the National Health and Nutrition Examination Survey (NHANES) database used for our study that evaluated participant eye rubbing habits or use of common over-the-counter allergy medications. Furthermore, there are currently only a few case reports in the literature that hypothesize an association between eye rubbing and optic neuropathy, with no studies conducted in a manner to suggest a causal association between eye rubbing and optic nerve damage. Thus, even if eye rubbing data were available in NHANES, its inclusion in our analyses would be insufficient to prove that our findings were explained by an eye rubbing mechanism rather than by sensitization to antigens. Finally, our study examined serum IgE levels rather than allergy symptoms, and serum IgE may be elevated in the absence of allergic symptoms and also in other conditions such as atopy and parasitic disease. Therefore, we cannot conclude that most people with the exposure of interest in our study (ie, elevated serum IgE) were necessarily rubbing their eyes and consequently cannot infer from our findings that someone who is prone to eye rubbing is at higher risk of developing optic neuropathy, as suggested by Dr Dille.
Regardless of the points mentioned above, we found in our study that only elevated serum IgE to cat and cockroach antigens were associated with increased glaucoma risk, while elevated IgE to dog antigens was associated with decreased glaucoma risk, and elevated IgE to several other indoor allergens and elevated total IgE levels had no association with glaucoma risk. To our knowledge, eye rubbing is a generalized allergy symptom that is not known to specifically occur with sensitization to one type of allergen. Therefore, if an association between sensitization to antigens and increased glaucoma risk were actually explained by eye rubbing, we would anticipate that we would find associations between elevated IgE levels to most or all of the antigens examined in our study and increased glaucoma risk, rather than only with the cat and cockroach antigens. Specifically, participants of NHANES are most commonly sensitized to the American and European dust mite antigens, and we would especially anticipate strong positive associations between sensitization to these antigens and glaucoma risk. Given that these phenomena were not observed, we find it doubtful that our findings were explained entirely by eye rubbing without involvement of a potential autoimmune mechanism.
We thank Dr Dille for his interesting insights and believe his suggestions raise important questions for future study in the associations between allergic symptoms, autoimmunity, and glaucoma.