Study suggests less frequent retinopathy screening in low-risk patients with type 1 diabetes

Annual screening led to an average of 18.4 exams up to time of detection of proliferative diabetic retinopathy and clinically significant macular edema, compared to 7.7 exams with an individualized schedule, for an average reduction of 10.7 exams per patient (58% fewer).


Researchers recently proposed extending the standard one-year screening interval for retinopathy to three or four years in low-risk patients with type 1 diabetes.

Using data from the Diabetes Control and Complications Trial and its follow-up study, they assessed regularly scheduled fundus photographs obtained from about 1,400 patients from1983 through 2012. They graded the photographs using the Early Treatment Diabetic Retinopathy Study grading scale.

State 1 equaled no retinopathy, state 2 included microaneurysms only (mild nonproliferative diabetic retinopathy), state 3 was moderate nonproliferative diabetic retinopathy, and state 4 meant severe nonproliferative diabetic retinopathy. State 5 included proliferative diabetic retinopathy and clinically significant macular edema.

Researchers used a longitudinal Markov model to estimate the cumulative incidence of transitions between states. Based on current retinal status and mean glycated hemoglobin level, they developed an evidence-based individualized screening schedule so that the risk of developing state 5 retinopathy before the next exam was low (about 5%). Results were published online on April 20 by the New England Journal of Medicine.

On average, patients remained in state 2 for the longest period (4.1 years) before transitioning to a different state and remained in state 4 for the shortest period (0.39 year). For patients in states 1 through 4, the researchers suggested screening every four years, three years, six months, and three months, respectively. Corresponding cumulative incidence of progression to state 5 among the respective groups would be 2.9%, 3.7%, 6.6%, and 14.4% if these screening intervals were followed.

Annual screening led to an average of 18.4 exams up to time of detection of state 5 retinopathy or the end of a 20-year period, compared to 7.7 exams with the individualized schedule, for an average reduction of 10.7 exams per patient (58% fewer). Considering a cost of about $200 per photograph and the 1 million U.S. patients with type 1 diabetes, the study authors estimated a cost savings of about $1 billion over 20 years (a 43.4% reduction compared to annual screening).

“The personalized schedules have the potential to reduce both the undetected time and the number of negative examinations at substantially lower cost,” the authors wrote. They noted limitations of the study, such as its reliance on seven-field fundus photography versus other exams.

An accompanying editorial, written by two ophthalmologists, voiced concerns about the proposed schedule, such as how the only variables used to calculate risk of transitioning to state 5 retinopathy were the patient's current retinopathy status and glycated hemoglobin levels.

Determining frequencies of screening based on these variables would be more complicated for physicians and patients than annual screening, and longer screening intervals could decrease rates of screening adherence in the low-risk population, the editorialists noted. “If the concerns discussed here can be addressed, targeted screening intervals could be a step forward in detecting diabetic retinopathy,” they wrote.