Purpose To investigate the structure-function relationship between optical coherence tomography (OCT) macular retinal and peripapillary nerve fiber layer (NFL) thickness and automated visual field (VF) findings. Design Cross-sectional observational study. Methods Retrospective institutional study where 150 consecutive eyes (101 subjects) from a glaucoma service were included. All the participants had full ophthalmic evaluation, VF testing and prototype OCT scanning at the same visit. Orthogonal OCT macular analysis was obtained to maximize the sampling of the area of interest. Pearson age-adjusted correlation was determined between macular retinal thickness and peripapillary NFL thickness. Area under the receiver operator characteristics (AROC) curves for the association between macular retinal thickness and peripapillary NFL thickness and VF findings were calculated in a subgroup of eyes without VF defect and eyes with VF defect confined to one hemifield. Results The correlation between macular retinal and peripapillary NFL measurements ranged between r = .27 to .54 for quadrants, .44 to .55 for hemiretina, and .52 for the overall mean. Areas under the receiver operator characteristics for macular thickness were higher in areas corresponding to the VF defect location than the noncorresponding locations. Areas under the receiver operator characteristics for peripapillary NFL thickness were higher than for the macular retinal thickness. Including both macular retinal thickness and peripapillary NFL thickness measurements in the logistic regression model yielded AROCs (range: .69 - .77) similar to those found for the peripapillary NFL alone. Conclusion Macular retinal thickness, as measured by OCT, was capable of detecting glaucomatous damage and corresponded with peripapillary NFL thickness; however, peripapillary NFL thickness had higher sensitivity and specificity for the detection of VF abnormalities.
ASJC Scopus subject areas