[Is fundus photography useful in screening for diabetic retinopathy in patients with type II diabetes mellitus?].

OBJECTIVE: To determine the value of a screening programme with fundus photography for diabetic retinopathy in diabetes mellitus type II patients. DESIGN: Prospective. SETTING: General practices in the region Zwolle, The Netherlands. METHOD: 650 Patients from 50 general practices were photographed after dilating both eyes. Of these 215 were examined by an ophthalmologist. Data of patients who did not join the study were analysed in 13 general practices. Any symptom of retinopathy and/or photographs of poor quality implied referral to an ophthalmologist. RESULTS: Of a total of 1300 photographs 208 (16%) could hardly or not at all be assessed. The agreement in the group of 215 patients between photography and ophthalmoscopy was statistically significant at classification level (Cohen's kappa 0.41). The advice given to patients after photography did not differ from that after ophthalmoscopy (kappa: 0.50); photography did not miss any high-risk characteristics, it yielded more warnings, and underestimated the level of retinopathy in 8 out of 215 cases. With current ophthalmologic rates in the Netherlands fundus photography was not financially advantageous as 71 (33%) out of 215 patients needed to be referred to an ophthalmologist. However, in the other 144 (67%) patients photography sufficed. This may offer a solution where ophthalmic care is in short supply. Of 168/420 patients who did not apply for photography 116 (69%) were either under ophthalmic supervision already or too disabled to be screened. CONCLUSION: The use of a fundus camera is equivalent to funduscopy by an ophthalmologist in screening for retinopathy of type II diabetic patients. There are no financial benefits, but it reduces the work load of ophthalmologists.