Abstract
AIM: To determine the discriminating performance of the macular ganglion cell-inner plexiform layer (GC-IPL) parameters between all the consecutive stages of glaucoma (from healthy to moderate-to-severe glaucoma), and to compare it with the discriminating performances of the peripapillary retinal nerve fiber layer (RNFL) parameters and optic nerve head (ONH) parameters. METHODS: Totally 147 eyes (40 healthy, 40 glaucoma suspects, 40 early glaucoma, and 27 moderate-to-severe glaucoma) of 133 subjects were included. Optical coherence tomography (OCT) was obtained using Cirrus HD-OCT 5000. The diagnostic performances of GC-IPL, RNFL, and ONH parameters were evaluated by determining the area under the curve (AUC) of the receiver operating characteristics. RESULTS: All GC-IPL parameters discriminated glaucoma suspect patients from subjects with healthy eyes and moderate-to-severe glaucoma from early glaucoma patients (P<0.017, for all). Also, minimum, inferotemporal and inferonasal GC-IPL parameters discriminated early glaucoma patients from glaucoma suspects, whereas no RNFL or ONH parameter could discriminate between the two. The best parameters to discriminate glaucoma suspects from subjects with healthy eyes were superonasal GC-IPL, superior RNFL and average c/d ratio (AUC=0.746, 0.810 and 0.746, respectively). Discriminating performances of all the parameters for early glaucoma vs glaucoma suspect comparison were lower than that of the other consecutive group comparisons, with the best GC-IPL parameters being minimum and inferotemporal (AUC=0.669 and 0.662, respectively). Moreover, minimum GC-IPL, average RNFL, and rim area (AUC=0.900, 0.858, 0.768, respectively) were the best parameters for discriminating moderate-to-severe glaucoma patients from early glaucoma patients. CONCLUSION: GC-IPL parameters can discriminate glaucoma suspect patients from subjects with healthy eyes, and also all the consecutive stages of glaucoma from each other (from glaucoma suspect to moderate-to-severe glaucoma). Further, the discriminating performance of GC-IPL thicknesses is comparable to that.