Purchase this article with an account.
Vivian Schreur, Freekje Van Asten, Heijan Ng, Jack Weeda, Cees Tack, Carel C B Hoyng, B. Jeroen Klevering, Eiko de Jong; Risk Factors for Diabetic Retinopathy in Type 1 Diabetes Mellitus Patients with Poor Glycemic Control. Invest. Ophthalmol. Vis. Sci. 2016;57(12):1598.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate risk factors for the development and progression of diabetic retinopathy (DR) in type 1 Diabetes Mellitus (DM1) patients with poor glycemic control.
We conducted a retrospective cohort study among 288 DM1 patients with poor glycemic control (defined as HbA1c level ≥53 mmol/mol). Subjects with DR (n=200) were compared to subjects without DR (n=88). A variety of variables were analyzed: age, gender, median HbA1c, HbA1c variability (defined as coefficient of variation of five separate measurements), duration of diabetes, age of onset of DM1, mean arterial blood pressure (MAP), body mass index (BMI), glomerular filtration rate, albuminuria, lipid profile, visual acuity, history of cigarette smoking and family history of DM1 or DM2. We used multivariable binary logistic regression models to analyze the data, and accuracy of prediction was determined through a Receiver Operating Characteristic (ROC)-curve. The association between these variables and the progression of DR into Vision Threatening DR (VTDR) was determined using Cox regression analyses.
Median HbA1c (OR 1.024 [1.002-1.046], p=0.029), HbA1c variability (OR 1.087 [1.012-1.167], p=0.029), MAP (OR 1.063 [1.030-1.097], p=1.4x10-4) and BMI (OR 1.009 [1.014-1.190], p=0.021) were independently associated with the development of DR. The ROC-curve showed an area under the curve of 0.711. Median HbA1c (HR 1.033 [1.021-1.046], p=1.5x10-7) was associated with progression of DR to VTDR.
Median HbA1c, HbA1c variability, MAP and BMI were associated with the development of DR in DM1 patients with poor glycemic control. Median HbA1c was associated with progression of DR. The discriminative ability of the prediction model was 71.1%. Further research is warranted to identify other risk factors.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only