Evaluation of retinal nerve fiber layer, optic nerve head, and macular thickne meaurement for glaucoma detection uing optical coherence tomography

Purpoe To compare the ability of optical coherence tomography retinal nerve fiber layer (RNFL), optic nerve head, and macular thickne parameter to differentiate between healthy eye and eye with glaucomatou viual field lo. Deign Obervational cae-control tudy. Method Eighty-eight patient with glaucoma and 78 healthy ubject were included. All patient underwent ONH, RNFL thickne, and macular thickne can with tratu OCT during the ame viit. ROC curve and enitivitie at fixed pecificitie were calculated for each parameter. A dicriminant analyi wa performed to develop a linear dicriminant function deigned to identify and combine the bet parameter. Thi LDF wa ubequently teted on an independent ample coniting of 63 eye of 63 ubject (27 glaucomatou and 36 healthy individual) from a different geographic area. Reult No tatitically ignificant difference wa found between the area under the ROC curve (AUC) for the RNFL thickne parameter with the larget AUC (inferior thickne, AUC = 0.91) and the ONH parameter with larget AUC (cup/dik area ratio, AUC = 0.88) (P = .28). The RNFL parameter inferior thickne had a ignificantly larger AUC than the macular thickne parameter with larget AUC (inferior outer macular thickne, AUC = 0.81) (P = .004). A combination of elected RNFL and ONH parameter reulted in the bet claification function for glaucoma detection with an AUC of 0.97 when applied to the independent ample. Concluion RNFL and ONH meaurement had the bet dicriminating performance among the everal tratu OCT parameter. A combination of ONH and RNFL parameter improved the diagnotic accuracy for glaucoma detection uing thi intrument.