Purchase this article with an account.
S. Maguluri, U.S. Rao, B. Barahimi, L.M. Merin, C.C. Recchia, A. Chomsky, F.M. Recchia; Digital Fundus Photography versus Dilated Fundus Exam in Proliferative Diabetic Retinopathy Screening . Invest. Ophthalmol. Vis. Sci. 2006;47(13):970.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To study the efficacy of digital fundus photography using the Vanderbilt Ophthalmic Imaging Center (VOIC) protocol in screening proliferative diabetic retinopathy (PDR).
Diabetic patients in the Tennessee Valley Veterans Affairs Health System were screened using an established VOIC protocol. Two monoscopic fundus photos (one centered on the nerve and one centerd on the fovea) were taken of each eye through a dilated pupil using a Canon CR6–45NM (Canon Medical Products, Irvine, CA). Digitized images were evaluated by trained non–ophthalmologist graders. A photographic diagnosis of PDR was made by the observation of media haze suggestive of vitreous hemorrhage (VH) or preretinal neovascularization Patients suspected of having PDR were recommended to have a complete ophthalmic examination within two weeks of screening. Records of all referred patients were then reviewed to determine the incidence of PDR on clinical exam.
3024 diabetic patients were screened. Of these, 68 (2.2%) were given a photographic diagnosis of PDR. Of the 68 patients, records of 39 patients were available for review. A total of 39 eyes were diagnosed by the VOIC protocol as having PDR. Of these 39 eyes, 23 eyes had been sent for evaluation of PDR only, 13 eyes for PDR plus macular edema and 3 eyes had PDR plus miscellaneous eye disease. Of the 39 eyes only 6 (15%) had clinical PDR. Of these 6, 2 (33%) were managed with observation, while 4 (67%) were managed with panretinal photocoagulation, None of the PDR patients in this study needed surgery as a primary intervention. Of the 17 eyes referred for CSME, 4 eyes (23%) had CSME on clinical exam. 10.3% of the patients had no diabetic retinopathy, while 48.7% had non–proliferative retinopathy. False positive rate for PDR was 85%.
The current photographic screening protocol for PDR produces a high number of false–positives New diagnostic criteria are necessary to improve the accuracy photographic screening for PDR.
This PDF is available to Subscribers Only