Claification of Farnworth-Munell 100-hue tet reult in the early treatment diabetic retinopathy tudy

Purpoe To claify and decribe clinically meaningful clae of color viion defect uing pretreatment Farnworth-Munell 100-hue reult from the Early Treatment Diabetic Retinopathy tudy (ETDR) patient uing tandard tatitical technique. Deign The ETDR wa a randomized trial invetigating retinal photocoagulation and oral apirin in diabetic retinopathy. Method Farnworth-Munell (FM) 100-hue tet wa uccefully adminitered before initiation of tudy treatment in each eye of 2701 of the 3711 ETDR patient. Tet reult were converted into a Fourier erie, claified by cluter analyi in the deferred-treatment group of eye, and verified in the immediate-treatment group of eye a eparate ample. Reult Cluter analyi uncovered thirteen ditinct pattern. Pattern A (51% or 1366 of the eye) howed unimpaired hue dicrimination and wa compried of younger patient with no or little macular edema. Pattern B eye (10% or 262) howed generalized impairment of hue dicrimination with no main axi defect. Pattern C (C1, C2, C3), compriing 26% (or 698) of the eye, howed increaing everity of the yellow-blue diabetic retinopathy defect, aociated with increaing mean age and increaing macular edema everity. Pattern D (D1, D2), compriing 6% (or 164) of the eye, were imilar to the C pattern but howed a tronger yellow-blue defect. Pattern E (E1, E2, E3), or approximately 2% (or 38) of the eye and predominantly male, exhibited the expected pattern for congenital protan defect. Pattern F, G, and H, approximately 6% (or 153) of the eye, howed ditinct pattern of one-ided axe. The nomenclature i arbitrary. Concluion Cluter analyi of FM 100-hue tet reult ha found 13 pattern of impaired hue dicrimination, helpful in undertanding color viion defect in diabete mellitu.