Discrimination of glaucomatous from non-glaucomatous optic neuropathy with swept-source optical coherence tomography.

Objective: To assess the ability of optic nerve head (ONH) parameters, peripapillary retinal nerve fiber layer (pRNFL), and macular ganglion cell layer (GCL) thickness measurements with swept-source optical coherence tomography (SS-OCT), to discriminate between glaucomatous and non-glaucomatous optic neuropathy (GON and NGON).

Methods: This retrospective cross-sectional study involved 189 eyes of 189 patients, 133 with GON and 56 with NGON. The NGON group included ischemic optic neuropathy, previous optic neuritis, and compressive, toxic-nutritional, and traumatic optic neuropathy. Bivariate analyses of SS-OCT pRNFL and GCL thickness and ONH parameters were performed. Multivariable logistic regression analysis was employed to obtain predictor variables from OCT values, and the area under the receiver operating characteristic curve (AUROC) was calculated to differentiate between NGON and GON.

Results: Bivariate analyses showed that the overall and inferior quadrant of the pNRFL was thinner in the GON group (P=0.044 and P<0.01), while patients with NGON had thinner temporal quadrants (P=0.044). Significant differences between the GON and NGON groups were identified in almost all the ONH topographic parameters. Patients with NGON had thinner superior GCL (P=0.015), but there were no significant differences in GCL overall and inferior thickness. Multivariate logistic regression analysis demonstrated that vertical cup-to-disc ratio (CDR), cup volume, and superior GCL provided independent predictive value for differentiating GON from NGON. The predictive model of these variables along with disc area and age achieved an AUROC=0.944 (95% CI 0.898-0.991).

Conclusions: SS-OCT is useful in discriminating GON from NGON. Vertical CDR, cup volume, and superior GCL thickness show the highest predictive value.