Large idiopathic macular hole surgery: remodeling of outer retinal layers after traditional internal limiting membrane peeling or inverted flap technique.

Objective:To investigate the influence of the inverted flap technique compared with traditional internal limiting membrane (ILM) peeling in the postoperative remodelling of outer retinal layers of idiopathic macular holes (MHs) >450 mu m.Methods:We analyzed medical records and imaging studies of all patients with an idiopathic MH >450 mu m who underwent vitrectomy at the Sacco University Hospital, Milan, and the Sacro Cuore Don Calabria Hospital, Verona, Italy, between January 2008 and December 2017. Out of 41 eyes evaluated, 17 were treated with traditional ILM peeling and 24 with the inverted ILM flap technique. All patients underwent follow-up examinations every 3 months and all of them completed a final visit 12 months after surgery. Best-corrected visual acuity (BCVA) and optical coherence tomography (OCT) parameters were evaluated at each visit. The main outcome measures were the postoperative recovery rate of the external limiting membrane (ELM) and ellipsoid zone (EZ), and postoperative BCVA. Correlations between OCT measurements and visual outcome were analyzed.Results:The ELM recovery rate in the ILM peeling group (15/17 eyes, 88%) was higher than in the ILM flap group (14/24 eyes, 58%) (p= 0.079). The EZ recovery rate was similar in the 2 groups, 7/17 eyes (41%) in the ILM peeling and 8/24 eyes (33%) in the ILM flap group (p= 0.744). Eyes without a persistent hyper-reflective "plug" at the edges of the MH showed a significantly higher EZ recovery rate (11/18, 61%) compared with eyes showing a persistent plug (4/23, 17%) (p= 0.008). The mean BCVA improved significantly in both groups: from 0.93 logMAR (20/170) to 0.26 logMAR (20/36) in the ILM peeling and from 0.98 logMAR (20/190) to 0.37 logMAR (20/46) in the ILM flap group. The final BCVA tended to be better in the ILM peeling group (p= 0.085).Conclusions:Given the limited information about the influence of ILM flap versus traditional ILM peeling in the postoperative remodelling of large idiopathic MHs, our data provides some new insights into the healing process of MHs >450 mu m. This should be considered as part of the decision process about whether to perform an ILM flap in these patients.