Abstract
Purpose :
Although type 2 diabetes mellitus (DM2) has been proposed as a risk factor for glaucoma, objective evidence of the association between poorly controlled DM2 and glaucoma progression is yet to be determined. The purpose of this study was to investigate the association between longitudinal levels of hemoglobin A1c (HbA1c) and rates of visual field loss over time in primary open angle glaucoma (POAG) and glaucoma suspect patients.
Methods :
This study included 491 eyes of 302 patients with diagnosis of DM2 who had either glaucoma (53%) or were suspected of glaucoma (47%), extracted from the Duke Glaucoma Registry (DGR). All patients had at least two standard automated perimetry (SAP) tests and two HbA1c measures over time, and a minimum follow-up of 6 months. Patients with history of panretinalphotocoagulation were censored before the procedure. Mean, fluctuation and peak HbA1c levels over time were calculated for each patient. HbA1c fluctuation was represented by the standard deviation (SD) of values over time. Multivariable linear mixed models were used to determine the effect of HbA1c on rates of change in SAP mean deviation (MD) over time, while adjusting for gender, race, age at baseline, peak intraocular pressure, baseline MD, and time of follow-up.
Results :
Mean follow-up time was 6.3 ± 4.3 years. Subjects had an average of 4.4 (range: 2 to 24) SAP tests over time and 8.1 (range 2 to 21) HbA1c tests over time. Mean ± SD HbA1c over time was 7.06 ± 1.1%, ranging from 5.4 to 12.6%. Mean HbA1c peak over time was 8.1 ± 1.9% (range: 5.5 to 15.6%) and mean fluctuation was 0.6 ± 0.6% (range 0.0 to 4.4). Mean rate of SAP MD change was -0.16 ± 0.29 dB/y, ranging from -1.28 to 0.56 dB/y. After adjustment for confounding factors, there was no statistically significant association between mean HbA1c levels and rates of MD change over time (P=0.840). There was also no significant association of HbA1c peak (P=0.621) and fluctuations (P=0.887) with rates of SAP change over time. No significant differences were seen after stratification for baseline diagnosis (glaucoma versus suspect).
Conclusions :
We were not able to find a significant association between levels of HbA1c over time and rates of visual field loss in glaucoma patients or individuals suspected of disease. These results would argue against an association between poor diabetes control and glaucoma progression.
This is a 2020 ARVO Annual Meeting abstract.