To the Editor.
—We read the article by Cartwright and coworkers,1 published in the April 1992 issue of the Archives, with much interest. The investigators compared the prevalence of immune-related disease among patients with normal (low)-tension glaucoma (NTG) with that among controls with ocular hypertension (OHT) based on a retrospective analysis of medical records, concluding that there is an association between NTG and "immunoreactive tendencies." We believe that there are design flaws in the study that may render such a conclusion premature.The study's foundation, according to its authors, is based on the presumptive differences in susceptibility to intraocular pressure in the development of glaucomatous optic neuropathy between the NTG cases and OHT controls. This contention is not necessarily valid. Furthermore, its implicit inclusion in the study design only serves to complicate the analysis from an epidemiologic standpoint. If the hypothesis being tested is that immune-related disease is associated