Five-Year Incidence of Proliferative Diabetic Retinopathy and Associated Risk Factors in a Nationwide Cohort of 201 945 Danish Patients with Diabetes

Abstract

Purpose: To evaluate the proliferative diabetic retinopathy (PDR) progression rates and identify the demographic and clinical characteristics of patients who later developed PDR compared with patients who did not progress to that state.

Design: A national 5-year register-based cohort study including 201 945 patients with diabetes.

Subjects: Patients with diabetes who had attended the Danish national screening program (2013-2018) for diabetic retinopathy (DR).

Methods: We used the first screening episode as the index date and included both eyes of patients with and without subsequent progression of PDR. Data were linked with various national health registries to investigate relevant clinical and demographic parameters. The International Clinical Retinopathy Disease Scale was used to classify DR, with no DR as level 0, mild DR as level 1, moderate DR as level 2, severe DR as level 3, and PDR as level 4.

Main outcome measures: Hazard ratios (HRs) for incident PDR for all relevant demographic and clinical parameters and 1-, 3-, and 5-year incidence rates of PDR according to baseline DR level.

Results: Progression to PDR within 5 years was identified in 2384 eyes of 1780 patients. Proliferative diabetic retinopathy progression rates from baseline DR level 3 at 1, 3 and 5 years were 3.6%, 10.9%, and 14.7%, respectively. The median number of visits was 3 (interquartile range, 1-4). Progression to PDR was predicted in a multivariable model by duration of diabetes (HR, 4.66 per 10 years; 95% confidence interval [CI], 4.05-5.37), type 1 diabetes (HR, 9.61; 95% CI, 8.01-11.53), a Charlson Comorbidity Index score of > 0 (score 1: HR, 4.62; 95% CI, 4.14-5.15; score 2: HR, 2.28; 95% CI, 1.90-2.74; score ≥ 3: HR, 4.28; 95% CI, 3.54-5.17), use of insulin (HR, 5.33; 95% CI, 4.49-6.33), and use of antihypertensive medications (HR, 2.23; 95% CI, 1.90-2.61).

Conclusions: In a 5-year longitudinal study of an entire screening nation, we found increased risk of PDR with increasing baseline DR levels, longer duration of diabetes, type 1 diabetes, systemic comorbidity, use of insulin, and blood pressure-lowering medications. Most interestingly, we found lower risk of progression from DR level 3 to PDR compared with that in previous studies.

Financial disclosures: Proprietary or commercial disclosure may be found after the references.

Keywords: Epidemiology; Proliferative diabetic retinopathy; risk factor.

Figure 1

A flowchart displaying the selected study population. DiaBase = The Danish Registry of Diabetic Retinopathy; DR = diabetic retinopathy; PDR = proliferative diabetic retinopathy.

Figure 2

Forrest plots of the hazard ratio for subsequent progression to proliferative diabetic retinopathy according to baseline characteristics. The figure shows a semiadjusted (sex and age) and a multivariable Cox proportional hazard model. Cholesterol lowering refers to all types of cholesterol-lowering medications. Glucose lowering refers to glucose-lowering medications exclusive of insulins. CCI = Charlson Comorbidity Index; HR = hazard ratio; ref. = reference.