Evidence-Based Screening for DR in Type 1 Diabetes
New England Journal of Medicine
Nathan et al. used retinal photographs from 2 long-running studies of type 1 diabetes to develop an evidence-based fundus photography screening program for diabetic retinopathy (DR). They found that their program, which takes patients’ retinopathy status and glycated hemoglobin level into account, could reduce the frequency of eye examinations without delaying the diagnosis of clinically significant disease.
For this study, the authors used data and fundus photographs from the Diabetes Control and Complications Trial (DCCT), which ran from 1983 to 1993, and the follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) study, which started in 1994 and is still ongoing. All told, the researchers were able to draw upon approximately 24,000 ophthalmologic assessments with 7-field fundus photography that were performed at intervals of 6-48 months during almost 30 years of follow-up. They then employed Markov modeling to determine the likelihood of progression to proliferative DR or clinically significant macular edema (ME).
The Markov models provided estimates of 5 states of DR progression, from preclinical (state 1) to proliferative DR or vision-threatening ME (state 5). State 1 patients or those with mild DR (state 2) were unlikely to progress to state 5 disease over a period of ≥ 4 years. However, those who had moderate (state 3) or severe (state 4) DR were highly likely to experience progression within a short time frame.
Overall, the data suggest that a practical, evidence-based screening schedule for time to next examination would be 4 years for state 1, 3 years for state 2, 6 months for state 3, and 4 months for state 4 disease. More frequent screening would be required for patients with higher glycated hemoglobin levels, as elevated mean glycated hemoglobin was strongly associated with a higher risk of progression. For instance, state 1 patients with a glycated hemoglobin level of 6% had a 1% risk of progressing to state 5 disease over a 5-year period. In contrast, the risk of progression from state 1 to state 5 for those with a glycated hemoglobin level of 10% was 4.3% over 3 years.
The original article can be found here.