Macular, papillary and peripapillary perfusion densities measured with optical coherence tomography angiography in primary open angle glaucoma and pseudoexfoliation glaucoma

Purpose To compare the blood flow situation in primary open-angle glaucoma (POAG) and pseudoexfoliation glaucoma (PXG) using optical coherence tomography angiography (OCTA). Methods In this prospective study a total of 26 POAG and 23 PXG eyes were included. All patients underwent a complete ophthalmological examination including standard automated perimetry, stereoscopic photographs of the optic disc, peripapillary retinal nerve fibre layer analysis and examination of vascular parameters of the optic nerve head (ONH), the peripapillary region and macula using OCTA. In addition to the vascular parameters recorded by the device, the vascular images were graphically evaluated using Image J. All recorded vascular parameters were compared between both groups and correlated to structural and functional parameters. Results The mean superficial perifoveal plexus perfusion density (PD) was significantly lower in PXG eyes than compared to POAG eyes using OCTA (32.57% ± 3.57% vs. 34.92% ± 2.11%, p = 0.007). The mean PD parameters for the superficial peripapillary plexus (40.98% ± 3.04% vs. 42.09% ± 2.29%, p = 0.152) as well as the size of the foveal avascular zone (FAZ) (0.23 mm 2 ± 0.1 mm 2 vs. 0.23 mm 2 ± 0.09 mm 2 ) did not differ between both groups. Additional graphic evaluation using Image J showed no significant difference for superficial perifoveal plexus PD (32.97% ± 1.11% vs. 33.35% ± 0.95%, p = 0.194) and peripapillary plexus PD (46.65% ± 0.83% vs. 46.95% ± 0.5%, p = 0.127) between the groups. Retinal nerve fibre layer (RNFL) thickness correlated significantly with peripapillary plexus PD for both OCTA data and Image J data ( p < 0.001, p = 0.032). Conclusion The severity of the glaucoma seems to be crucial for peripapillary and macular perfusion densities, and not the form of glaucoma. An additional graphic evaluation is a possible step that could be implemented to improve the comparability of OCTA scans and to optimize the possibility of quantitative perfusion analysis in the case of deviating quality criteria.