Colour vision losses in diabetes in the absence of proliferative retinopathy

Abstract

AbstractPurpose There are over 30 published studies evaluating colour vision in diabetes. The evidence is, however, not consistent – the predictive accuracy of the studies does not support the use of colour vision as a screening test for diabetic retinopathy (Health Technol Assess, 13 (60), 2009). The CAD colour vision test (Expert Rev.Ophthalmol. 6(4):409‐420, 2011) addresses some of the major flaws in clinical colour testing. This poster reports the application of the CAD test to 214 diabetic patients.Methods A consecutive series of diabetics attending a diabetes center in Abu Dhabi were invited to participate in the study. Inclusion criteria were a best corrected visual acuity of 6/18 or better, no more than moderate non‐proliferative retinopathy or moderate maculopathy and no co‐existing glaucoma. Patients had a full eye examination (including colour photography and macular OCT), a full assessment of their diabetes and the CAD colour test.Results Age‐corrected binocular colour threshold limits were employed to screen for normal colour vision. Abnormal colour vision was noted in 80% of Arab and 67% of non Arabs. This did not correlate with duration of diabetes, age, HBA1C, body mass index, visual acuity or central retinal thickness.Conclusion Colour vision is abnormal in the large majority of the study population. The severity of colour vision loss is not correlated with age or gender of the patients, duration or severity of diabetes, current diabetic control, or central macular thickness. These findings suggest the existence of early retinal changes in diabetes that affect specifically colour vision and occur independently of factors such as those causing macula edema.