1 Research Unit of Ophthalmology, Department of Clinical Research, Det Sundhedsvidenskabelige Fakultet, SDU2 Research Unit of Ophthalmology, Department of Clinical Research, Det Sundhedsvidenskabelige Fakultet, SDU
Incidence of diabetic macular edema and associated risk factors in a cohort of patients with type 1 diabetes in Denmark. Rasmussen M.L.1, Pedersen R.B. 1, Sjølie A.K. 1, Grauslund J1 1University of Southern Denmark, Department of ophthalmology, Odense University Hospital, Denmark Purpose: To evaluate the 16-year incidence of diabetic macular edema (DME) in a cohort of type 1 diabetic patients and to investigate possible risk factors of developing DME. Methods: This was a prospective cohort study of Danish type 1 diabetic patients. A total of 131 patients were examined at baseline in 1995 and at follow-up in 2011. At baseline, median age and duration of diabetes was 21.8 and 12.3 years, respectively. Two field fundus photos were taken, retinopathy was graded in accordance to ETDRS and macular microaneurysms (MA) were counted. Baseline measurements included HbA1c, albuminuria, blood pressure and BMI. At follow-up SD-OCT were performed. Incident DME was defined as former central laser treatment or current DME on OCT. Excluded were patients with DME at baseline (n=1) and patients without fundus photos and OCT at follow-up (n=2). Incidence of DME was calculated and logistic regression was made for potential baseline risk factors: Sex, age, duration of diabetes, retinopathy (worst eye), blood pressure, nephropathy, MA and BMI. Results At baseline no retinopathy (ETDRS 10) was found in 32.8%, mild non-proliferative retinopathy (NPDR) (ETDRS 20-35) in 60.9% moderate NPDR (ETDRS 43-61) in 4.7% and proliferative retinopathy (ETDRS < 61) in 0.8%. The cumulated incidence of DME was 17.2% (n=22). In a multivariate logistic regression model incident DME was predicted by moderate NPDR at baseline (odds ratio [OR]: 17.6 versus patients with no retinopathy, 95% CI 1.19-261), macroalbuminuria (OR: 24.2 versus patients without albuminuria, 95% CI, 1.25-468) and diastolic blood pressure (OR: 2.5 per 10mmHg, 95% CI, 1.11-5.5). There were no association between incident DME and other baseline variables. Conclusion: In this population of Danish type 1 diabetic patients, almost one in five developed DME in 16 years. Patients with moderate NPDR, macroalbuminuria and diastolic hypertension were more likely to develop DME.