Diabetes Mellitus and the risk of Primary open angle glaucoma

Association of Diabetes Medication With Open-Angle Glaucoma, Age-Related Macular Degeneration, and Cataract in the Rotterdam Study

Importance

Recent studies suggest that the diabetes drug metformin has a protective effect on open-angle glaucoma (OAG) and age-related macular degeneration (AMD). However, studies have not addressed the critical issue of confounding by indication, and associations have not been evaluated in a large prospective cohort.

Objective

To determine the association between diabetes medication and the common eye diseases OAG, AMD, and cataract and to evaluate their cumulative lifetime risks in a large cohort study.

Design, Setting, and Participants

This cohort study included participants from 3 independent cohorts from the prospective, population-based Rotterdam Study between April 23, 1990, and June 25, 2014. Participants were monitored for incident eye diseases (OAG, AMD, cataract) and had baseline measurements of serum glucose. Data on diabetes medication use and data from ophthalmologic examinations were gathered.

Exposures

Type 2 diabetes (T2D) and the diabetes medications metformin, insulin, and sulfonylurea derivatives.

Main Outcomes and Measures

Diagnosis and cumulative lifetime risk of OAG, AMD, and cataract.

Results

This study included 11 260 participants (mean [SD] age, 65.1 [9.8]; 6610 women [58.7%]). T2D was diagnosed in 2406 participants (28.4%), OAG was diagnosed in 324 of 7394 participants (4.4%), AMD was diagnosed in 1935 of 10 993 participants (17.6%), and cataract was diagnosed in 4203 of 11 260 participants (37.3%). Untreated T2D was associated with a higher risk of OAG (odds ratio [OR], 1.50; 95% CI, 1.06-2.13; P = .02), AMD (OR, 1.35; 95% CI, 1.11-1.64; P = .003), and cataract (OR, 1.63; 95% CI, 1.39-1.92; P < .001). T2D treated with metformin was associated with a lower risk of OAG (OR, 0.18; 95% CI, 0.08-0.41; P < .001). Other diabetes medication (ie, insulin, sulfonylurea derivates) was associated with a lower risk of AMD (combined OR, 0.32; 95% CI, 0.18 to 0.55; P < .001). The cumulative lifetime risk of OAG was lower for individuals taking metformin (1.5%; 95% CI, 0.01%-3.1%) than for individuals without T2D (7.2%; 95% CI, 5.7%-8.7%); the lifetime risk of AMD was lower for individuals taking other diabetes medication (17.0%; 95% CI, 5.8%-26.8% vs 33.1%; 95% CI, 30.6%-35.6%).

Conclusions and Relevance

Results of this cohort study suggest that, although diabetes was clearly associated with cataract, diabetes medication was not. Treatment with metformin was associated with a lower risk of OAG, and other diabetes medication was associated with a lower risk of AMD. Proof of benefit would require interventional clinical trials.

Intraocular pressure measurements in diabetes mellitus

PURPOSE

To investigate the impact of diabetes mellitus-induced changes on intraocular pressure measurements using Goldmann applanation tonometry, Ocular Response Analyzer, and Corvis ST.

METHODS

Measurements were done using Goldmann applanation tonometry, Ocular Response Analyzer, and Corvis ST in 69 diabetic patients. Biomechanical-corrected intraocular pressure values by Ocular Response Analyzer (IOPcc) and Corvis ST (bIOP) were used. In addition, biometry and tomography were performed and information on diabetes mellitus specific factors was collected. Results were compared to an age-matched group of 68 healthy subjects.

RESULTS

In diabetes mellitus, Goldmann applanation tonometry intraocular pressure (P = 0.193) and central corneal thickness (P = 0.184) were slightly increased. Also, IOPcc (P = 0.075) and bIOP (P = 0.542) showed no significant group difference. In both groups, IOPcc was higher than Goldmann applanation tonometry intraocular pressure (P = 0.002, P < 0.001), while bIOP was nearly equal to Goldmann applanation tonometry intraocular pressure (P = 0.795, P = 0.323). Central corneal thickness showed a tendency to higher values in poorly controlled than in controlled diabetes mellitus (P = 0.059). Goldmann applanation tonometry intraocular pressure correlated to central corneal thickness, while IOPcc and bIOP were independent from central corneal thickness in both groups. All intraocular pressure values showed significant associations to corneal biomechanical parameters. Only in diabetes mellitus, bIOP was correlated to Pachy slope (P = 0.023).

CONCLUSION

In diabetes mellitus, Goldmann applanation tonometry intraocular pressure was slightly, but not significantly, increased, which might be caused by a higher central corneal thickness and changes in corneal biomechanical properties. However, intraocular pressure values measured by Ocular Response Analyzer and Corvis ST were not significantly different between diabetes mellitus patients and healthy subjects. The bIOP showed a higher agreement with Goldmann applanation tonometry than IOPcc and was independent from central corneal thickness.

Corneal Vibrations during Intraocular Pressure Measurement with an Air-Puff Method

Introduction

The paper presents a commentary on the method of analysis of corneal vibrations occurring during eye pressure measurements with air-puff tonometers, for example, Corvis. The presented definition and measurement method allow for the analysis of image sequences of eye responses—cornea deformation. In particular, the outer corneal contour and sclera fragments are analysed, and 3D reconstruction is performed.

Methods

On this basis, well-known parameters such as eyeball reaction or corneal response are determined. The next steps of analysis allow for automatic and reproducible separation of four different corneal vibrations. These vibrations are associated with (1) the location of the maximum of cornea deformation; (2) the cutoff area measured in relation to the cornea in a steady state; (3) the maximum of peaks occurring between applanations; and (4) the other characteristic points of the corneal contour.

Results

The results obtained enable (1) automatic determination of the amplitude of vibrations; (2) determination of the frequency of vibrations; and (3) determination of the correlation between the selected types of vibrations.

Conclusions

These are diagnostic features that can be directly applied clinically for new and archived data.