Ultra-Widefield Imaging Can Identify Lesions Missed by ETDRS 7-Field Assessment

Ultra-widefield imaging can identify neovascularization, preretinal hemorrhage, neovascular disease, and scatter laser scars — conditions that may be missed during evaluation with Early Treatment of Diabetic Retinopathy Study (ETDRS) 7-field assessment, according to research published in Eye. While the 2 imaging methods demonstrated discrepancies in less than 5% of patients evaluated, the report highlights the importance of evaluating the entire ultra-widefield region to avoid missing lesions that may potentially threaten vision.

Researchers conducted a retrospective, cross-sectional study to compare diabetic retinopathy (DR) severity levels determined by ETDRS 7-field region vs ultra-widefield imaging. The team used 2 different protocols (International Classification of Diabetic Retinopathy [ICDR] scale and the DRCR.net Protocol AA grading scale) for each imaging method and included a total of 250 eyes from 157 patients with diabetes.

When using the ICDR protocol, the investigators noted a severity level discrepancy among 6 eyes (2.4%) between the ETDRS 7-field region and ultra-widefield imaging methods. These discrepancies were the result of 2 intraretinal hemorrhages, and 4 cases of neovascular disease detected in the peripheral field with ultra-widefield imaging that were not detected with ETDRS 7-field assessment.

After using the ETDRS DRSS Protocol AA grading scale, the researchers observed a severity level discrepancy in 14 eyes between the 2 imaging methods. Overall, ultra-widefield imaging detected intraretinal hemorrhage (n = 4), neovascularization (n = 4), preretinal hemorrhage (n = 2), or scatter laser scars (n = 4) in the peripheral fields of 5.6% of eyes that were missed by ETDRS 7-field assessment.

“Although the number of discrepant cases was low, severe potentially sight-threatening lesions such as neovascularization were missed,” according to the study authors. “This would seem to highlight the importance of assessing the entire retina including the periphery when evaluating eyes with DR.”

Study limitations include possible ascertainment bias due to its retrospective nature, the strict use of only monoscopic images, and a single center design.