Effect of intensive diabetes therapy on the progression of diabetic retinopathy in patients with type 1 diabetes: 18 years of follow-up in the DCCT/EDIC

Abstract

The Diabetes Control and Complications Trial (DCCT) demonstrated that a mean of 6.5 years of intensive therapy aimed at near-normal glucose levels reduced the risk of development and progression of retinopathy by as much as 76% compared with conventional therapy. The Epidemiology of Diabetes Interventions and Complications study (EDIC) observational follow-up showed that the risk of further progression of retinopathy 4 years after the DCCT ended was also greatly reduced in the former intensive group, despite nearly equivalent levels of HbA1c, a phenomenon termed metabolic memory. Metabolic memory was shown to persist through 10 years of follow-up. We now describe the risk of further progression of retinopathy, progression to proliferative diabetic retinopathy, clinically significant macular edema, and the need for intervention (photocoagulation or anti-VEGF) over 18 years of follow-up in EDIC. The cumulative incidence of each retinal outcome continues to be lower in the former intensive group. However, the year-to-year incidence of these outcomes is now similar, owing in large part to a reduction in risk in the former conventional treatment group.

Trial registration: ClinicalTrials.gov NCT00360815 NCT00360893.

Figure 1

Box plots of the distribution of glycosylated hemoglobin (HbA_1c) values in the DCCT treatment group at the end of the DCCT and at each of the first 18 years of EDIC. Each box shows the quartiles, + denotes the mean, and whiskers show the range.

Figure 2

Estimated cumulative incidence of further progression of retinopathy from DCCT closeout to EDIC year 18 within the former DCCT INT and CONV. A: Further three-step progression from the level at DCCT closeout (n = 1,358). B: New onset of PDR (n = 1,318). C: New onset of CSME (n = 1,277). D: New photocoagulation (pan-retinal or focal laser or anti-VEGF use) based on fundus photography grading and/or patient reporting (n = 1,335). Estimated cumulative incidence was based on the Weibull regression models adjusted for the level of retinopathy at the end of the DCCT, primary vs. secondary cohort, glycated hemoglobin value on entry to the DCCT, and diabetes duration at DCCT baseline. Subjects who had prior scatter photocoagulation during the DCCT (n = 36), who died during the DCCT (n = 11), or who had no EDIC measurements (n = 36) were excluded from all the analyses. Subjects with prior PDR during the DCCT (n = 78) excluded from B, prior CSME during the DCCT (n = 120) excluded from C, and prior treatment during the DCCT (n = 74) excluded from D.

Figure 3

Estimated hazard rate (incidence) function of further progression of retinopathy from DCCT closeout to EDIC year 18 within the former DCCT INT and CONV. A: Further three-step progression from the level at DCCT closeout. B: New onset of PDR. C: New onset of CSME. D: New photocoagulation (pan-retinal or focal laser or anti-VEGF use) based on fundus photography grading and/or patient reporting. Estimated hazard rate was based on a smoothed Turnbull nonparametric estimate of the survival function, without adjustment for other factors. Risk reductions from DCCT closeout to EDIC year 10 and from EDIC year 10 to EDIC year 18 were obtained from separate Weibull regression models of those at risk during each period.