Is a History of Diabetes Mellitus Protective Against Developing Primary Open-angle Glaucoma?

In 2002, the Ocular Hypertension Treatment Study¹ (OHTS) reported baseline factors that increased the risk for developing primary open-angle glaucoma (POAG): older age, a larger vertical cup-disc ratio, higher intraocular pressure, greater Humphrey visual field pattern standard deviation, and a thinner central corneal measurement. A history of diabetes mellitus at baseline appeared to be protective against developing POAG. Six of the 191 participants (3.1%) who reported diabetes mellitus at baseline developed POAG compared with 119 of the 1427 participants (8.3%) who did not (multivariate hazard ratio, 0.37; 95% confidence interval, 0.15-0.90). This finding was unexpected and contradicted most of the literature on risk factors for POAG.^{2 - 4}

The 2002 article acknowledged methodological issues in the OHTS that might account for this finding. Participants were classified at baseline as having a positive history of diabetes mellitus if they responded yes to the question, “Has a doctor ever told you that you have . . . diabetes, or sugar in the blood?” A positive history of diabetes mellitus was not corroborated by record review, blood tests, or medication use. Underascertainment of diabetes mellitus could result from underdiagnosis as well as underreporting by participants. In addition, OHTS entry criteria excluded diabetic patients with any retinopathy so that participants with diabetes mellitus were likely to be atypical.⁵

Starting in February 2003, after approval by the Data and Safety Monitoring Committee and local institutional review boards of participating clinics, the OHTS began collecting more detailed information about whether participants had been diagnosed with diabetes mellitus and what treatments they were receiving. We reran the same Cox proportional hazards prediction model for POAG using the same data set of baseline predictors and outcomes and varying only the definition of diabetes mellitus. No blood tests or corroborating record reviews were performed. The different definitions of diabetes mellitus, all of which were based on participant self-report, varied in sensitivity and specificity as follows:

During follow-up, 409 participants responded that a doctor had told them they had diabetes or sugar in the blood, as compared with 191 participants who answered yes to the same question at baseline (Table). All of the 191 participants who had reported a positive history of diabetes mellitus at baseline also reported a positive history at 1 or more follow-up visits. In updated univariate and multivariate analyses, a history of diabetes mellitus was not statistically significantly predictive for the development of POAG for all of the 3 definitions tested.

The protective effect of a history of diabetes mellitus for the development of POAG reported in the 2002 OHTS prediction model¹ was not supported in univariate or multivariate analyses using updated self-reported data on diabetes mellitus history and its treatment. We believe the difference from the 2002 article reflects more complete ascertainment of diabetes mellitus. Many more participants reported a positive history of diabetes mellitus during follow-up than at baseline. The results of these reanalyses are consistent with several previous studies^{2 - 3} reporting that diabetes mellitus either increased the risk of developing POAG or had no effect.

Correspondence: Dr Gordon, Department of Ophthalmology and Visual Sciences, Washington University School of Medicine, 660 S Euclid, Box 8096, St Louis, MO 63110 (mae@vrcc.wustl.edu).

Group Information: A complete list of Ocular Hypertension Treatment Study Group investigators appears at https://vrcc.wustl.edu.

Financial Disclosure: None reported.

Funding/Support: This work was supported by grants from the National Eye Institute, the National Center on Minority Health and Health Disparities, and the National Institutes of Health (grants EY09341 and EY09307) and by unrestricted grants from Merck Research Laboratories and Research to Prevent Blindness, Inc.