Q&A

What is a reasonable interval for retinopathy screening in patients with diabetes?

J Fam Pract. 2003 May;52(5):349-364
By
Seth Miller, MD
Erik J. Lindbloom, MD, MSPH
Author and Disclosure Information

  • BACKGROUND: Diabetes mellitus is the leading cause of blindness among adults in the US. The onset of diabetic retinopathy is often asymptomatic, and treatments for reducing visual loss are most effective in the earlier stages. The most widely accepted guidelines for intervals between screening exams are based on opinion rather than direct evidence.
  • POPULATION STUDIED: The Liverpool Diabetic Eye Study screening program identified 9890 patients from general practices in western England with diabetes mellitus type 2. Of these patients, 7615 underwent baseline ophthalmologic screening, which identified 7265 individuals who did not have sight-threatening diabetic retinopathy. The study reports on the 4770 patients without sight-threatening retinopathy who received at least 1 follow-up examination.
  • STUDY DESIGN AND VALIDITY: In this prospective cohort study, patients underwent visual acuity testing and retinal photography. Retinal images were graded using a standardized algorithm based on the Early Treatment Diabetic Retinopathy Study protocol. Graded images were divided into 4 groups: no retinopathy, background retinopathy, mild preproliferative retinopathy, and sight-threatening retinopathy (defined as moderate preproliferative retinopathy or worse, or clinically significant maculopathy).
  • OUTCOMES MEASURED: The investigators report the presence, incidence, and severity of retinopathy at baseline screening and at follow-up examinations.
  • RESULTS: One year after baseline, 5.3% (95% confidence interval [CI], 4.6%–6.0%) of patients with normal initial retinal examinations developed at least some degree of retinopathy. The incidence of sight-threatening retinopathy in patients with no retinopathy at baseline was 0.3% (95% CI, 0.1%–0.5%) in the first year, and the yearly incidence steadily increased to reach a cumulative incidence of sight-threatening retinopathy at 5 years of 3.9% (95% CI, 2.8%–5.0%).
PDF Download

 

PRACTICE RECOMMENDATIONS

Assuming that a given patient is reliable for follow-up and that a clinical system is in place to handle a more individualized screening protocol, the investigators suggest the following approach: 3-year intervals for patients with no retinopathy and no risk factors (risk factors being diabetes for longer than 20 years or use of insulin); annual screening for patients with no retinopathy and 1 or both risk factors or for patients with background retinopathy; and 4-month intervals for patients with mild preproliferative retinopathy.

These intervals provide at least a 95% probability of remaining free of sight-threatening diabetic retinopathy between screenings.

Recommended Reading

What is the relative cardiovascular benefit of lowering cholesterol, blood pressure, and glucose levels in patients with type 2 diabetes?
MDedge Family Medicine
Which oral antihyperglycemics are most efficacious in reducing hemoglobin A1C in diabetic patients?
MDedge Family Medicine
Outcomes of audit-enhanced monitoring of patients with type 2 diabetes
MDedge Family Medicine
Continuity and quality of care in type 2 diabetes
MDedge Family Medicine
Are thiazolidinediones first-line agents?
MDedge Family Medicine
Is an elevated serum transferrin saturation associated with the development of diabetes?
MDedge Family Medicine
Does microalbuminuria screening in diabetes prevent complications?
MDedge Family Medicine
How beneficial are thiazolidinediones for diabetes mellitus?
MDedge Family Medicine
Acarbose delays onset of type 2 diabetes mellitus
MDedge Family Medicine
Screening for and treating asymptomatic bacteriuria not useful in women with diabetes
MDedge Family Medicine