Retinal function as assessed by multifocal electroretinography and central perimetry before and after vitrectomy with conventional versus fovea-sparing internal limiting membrane peeling for idiopathic macular hole

Purpose: To perform multifocal electroretinography (mfERG)- and central perimetry-based evaluation of the function of the macula before and after vitrectomy with conventional internal limiting membrane (ILM) peeling versus fovea-sparing ILM peeling for idiopathic macular hole (IMH). Material and Methods: This study included 70 patients (71 eyes) who received 25-G vitrectomy with conventional or fovea-sparing ILM peeling and gas tamponade with 20% SF6 or 15% С3F8 for stage-2 to stage-4 holes as per the classification by Gass. Eyes of study patients underwent optical coherence tomography angiography (OCTA) evaluation of IMH diameter and choriocapillaris perfusion density, ten-degree static perimetry and 20-degree 5-ring mfERG before and 1 month after surgery. Results: Before surgery, eyes with IMH showed significantly reduced foveal light sensitivity and overall parafoveal sensitivity, increased Pattern Standard Deviation (PSD), and reduced retinal response density in mfERG rings 1 and 2 compared to fellow eyes. The foveal threshold sensitivity in the affected eyes was found to be correlated with minimal diameter of IMH (r = -0.77; р < 0.05) and the postoperative BCVA (r = 0.66; р < 0.05), whereas the overall retinal sensitivity, with the maximal diameter of IMH (r = -0.56), preoperative BCVA (r = 0.6) and postoperative BCVA (r = 0.7). MfERG retinal response density in ring 1 was significantly reduced (р = 0.00001) and correlated with the preoperative foveal threshold sensitivity (r = 0.6) and choriocapillaris perfusion density (r = 0.39). After macular hole closure, median BCVA (interquartile range) in the fovea-sparing ILM peeling group and the conventional ILM peeling group improved to 0.55 (0.35–0.7) and 0.43 (0.35–0.6), respectively. In addition, the foveal threshold sensitivity within 10-degree area in the former and latter groups improved, but was 13.6% (р = 0.009) and 15% (р = 0.0001), respectively, lower than in the fellow eyes (34.5 ± 2.9 dB). The overall retinal sensitivity in the fovea-sparing ILM peeling group improved more substantially, to 509.6 ± 13.9 dB, and almost reached the fellow-eye value (528.0 ± 25.8 dB). Moreover, the retinal response density in the conventional ILM peeling group improved in rings 1-5, whereas that in the fovea-sparing ILM peeling group, in rings 2-4, but not in ring 1. Conclusion: In eyes with IMH, retinal photoreceptor function as assessed by perimetry and mfERG was found to be impaired at baseline and improved after macular hole closure. In the fovea-sparing ILM peeling group, the overall retinal sensitivity in the affected eyes improved more substantially than in the conventional ILM peeling group.