OCT-defined Foveal Restoration Is Associated With Visual Recovery After Vitrectomy for Lamellar Macular Hole: A Multicenter Study LMH Report No. 1

Purpose Anatomic restoration after vitrectomy for lamellar macular hole (LMH) is not always achieved. However, the impact of unresolved LMH on visual outcomes remains unclear. We evaluated postoperative outcomes in eyes with unresolved LMH and postoperative full-thickness macular hole (FTMH) to reassess the meaning of anatomic success after LMH surgery. Design Multicenter, retrospective cohort study. Subjects One hundred sixty-six consecutive eyes of 166 patients who underwent vitrectomy for LMH. Methods Anatomic success was defined as restoration of foveal configuration, whereby the postoperative OCT findings no longer met the diagnostic criteria for LMH. Postoperative anatomic outcomes were classified as anatomic success, unresolved LMH (including both persistent or recurrent LMH), or postoperative FTMH. Main Outcome Measures Incidence and risk factors of unresolved LMH and postoperative FTMH, and best-corrected visual acuity (BCVA). Results Anatomic success was achieved in 141 eyes (84.9%), whereas 14 eyes (8.4%) had unresolved LMH and 11 eyes (6.6%) progressed to FTMH during follow-up (mean 17.8 months). At 12 months, BCVA was significantly worse in eyes with unresolved LMH than in those with anatomic success (0.24 ± 0.24 vs. 0.09 ± 0.19 logarithm of the minimum angle of resolution [logMAR]; P=0.041). Internal limiting membrane (ILM) peeling was significantly associated with a reduced risk of unresolved LMH (odds ratio [OR], 0.011; 95% CI, 0.002–0.062; P<0.001). Postoperative FTMH was significantly associated with epiretinal proliferation (P=0.018), thinner preoperative central foveal thickness (P=0.017), and no ILM peeling (P=0.001). Multivariable analysis confirmed that ILM peeling was an independent factor preventing postoperative FTMH (OR, 0.080; 95% CI, 0.010–0.620; P=0.001). Postoperative FTMH achieved closure in 9 eyes (82%) with additional treatment or observation, while the final BCVA was significantly worse in eyes that developed FTMH than in those with anatomic success (0.50 ± 0.32 vs. 0.12 ± 0.20 logMAR; P<0.001). Conclusions While most eyes achieved anatomic success after vitrectomy for LMH, approximately 15% developed unresolved LMH or postoperative FTMH, both of which were associated with the absence of ILM peeling. Because anatomic failure was associated with poorer visual outcomes, foveal restoration on OCT may represent a more clinically meaningful anatomic endpoint after LMH surgery.