\r\nMethods: In this retrospective study, total of 25 eyes in only ERM group and 26 eyes in ILM group were evaluated. Best corrected visual acuity (BCVA), central macular thickness (CMT) were compared. In optical coherence tomography angiography (OCTA), foveal vessel density (VD), parafoveal VD and perifoveal VD of superficial capillary plexus (SCP) and deep capillary plexus (DCP) and foveal avascular zone were compared. Foveal herniation was also evaluated.

\r\nResults: There was no different in terms of BCVA, CMT, OCTA parameter change and recurrence rate, between ILM and only ERM groups. In OCTA parameters, DCP parafoveal VD of ILM group was significantly increased, postoperatively. (44.91±6.34, 49.52±5.74, p=0.024) SCP perifoveal VD and DCP foveal VD of only ERM group was significantly decreased, postoperatively. (SCP perifoveal VD: 44.67±4.89, 42.4±5.07, p=0.041; DCP foveal VD: 43.61±6.09, 39.44±5.99, p=0.045) According to having foveal herniation (FH), increase in SCP and DCP perifoveal VD was significant in ILM group. (SCP perifoveal VD change: FH(+): 4.86±9.46, FH(-): -1.8±5.48, p=0.036; DCP perifoveal VD change: FH(+): 12,26±9,39, FH(-): 0,3±13.3, p=0.048)

\r\nConclusion: There is no difference between two technique in terms of BCVA, CMT and any OCTA parameters at 1 year follow-up. However, in a small subgroup analysis including FH, significant changes were observed in OCTA parameters. Keywords : epiretinal membrane, pars plana vitrectomy, internal limiting membrane peeling, foveal herniation, optical coherence tomography angiography