Department of Gastroenterology, Second Medical Clinic, Ippokration Hospital, Aristotle University of Thessaloniki, Thessaloniki, Greece
We read the paper written by Razeghinejad et al. (1), who conducted a study on anti- Helicobacter pylori (H. pylori) IgG antibody levels in an Iranian glaucoma cohort, with considerable interest and based on our original concept (2-5). The authors concluded that a relation between H. pylori infection and primary open-angle glaucoma is not supported by their results, because the aqueous anti- H. pylori IgG antibody concentration did not differ significantly from the cataract control population. However, in the discussion there is little attention to the limitations of their work. Specifically, there is no discussion of: (a) the relatively small sample size, (b) the limited power of the study, (c) the possibility that the control group represents a selected group resulting in bias, (d) the absence of normalization of aqueous antibody titers to another serum protein to serve as control, such as IgG or albumin, and (e) the limited accuracy of using the commercial ELISA technique in the aqueous humor, originally manufactured for serum samples. It has been reported that the prevalence of H. pylori infection in the city of Shiraz, where the study by Razeghinejad et al. was conducted, is very high (6,7) regardless of the socioeconomic status, an already established significant factor affecting H. pylori prevalence in the European countries. This means that to prove a difference in H. pylori prevalence between any two groups in Shiraz, several hundreds or even a few thousands of participants are required. Instead, Razeghinejad et al. presented their results based on a small number of patients, and therefore the power of their study was too low. The authors failed to comment on the prevalence of H. pylori infection they found in their study groups, which should be very high in both groups according to previous reports and increase with increasing age (6,7).