Internal Limiting Membrane Peel Size and Macular Hole Surgery Outcome: A Systematic Review and Individual Participant Data Study of Randomized Controlled Trials Publisher Pubmed



Teh BL¹ ; Li Y² ; Nanji K^{3, 4} ; Phillips M⁴ ; Chaudhary V^{3, 4} ; Steel DH^{1, 2} ; Anantharaman G¹² ; Modi A¹¹ ; Sadda S¹⁰ ; Yao Y⁹ ; Zhao M⁹ ; Qu J⁹ ; Khodabande A⁸ ; Khalili Pour E⁸ Show All Authors Show Affiliations
Source: Eye (Basingstoke) Published:2025

Abstract

Background: There is no consensus regarding the optimal internal limiting membrane (ILM) peel size during vitrectomy for idiopathic full thickness macular holes (iFTMH). Methods: A systematic review was performed to identify randomized controlled trials (RCTs) comparing vitrectomy with ILM peeling of differing sizes in adults with iFTMH. Individual participant data was obtained including relevant baseline variables. The effect of different ILM peel sizes, divided into “small” (1-disc diameter [DD] in radius or less) and “large” (>1-DD in radius) were analysed on primary hole closure and postoperative visual acuity (VA) at 6 months. A subgroup analysis analysing for the effect of macular hole size on the same outcomes was also performed. Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) was used to assess the certainty of evidence. Results: Five RCTs with 370 eyes were included. Primary closure was achieved in 74.7% in small peel group compared to 84.8% in large peel group (p = 0.016). Multilevel logistic regression showed that a peel radius of >1-DD probably improved hole closure slightly with odds ratio (OR) of 1.20 (95% CI: 1.11–1.31, p < 0.001) and a number needed to treat (NNT) to benefit of 31 (95% CI: 21–53). ILM peel size likely did not affect VA. The mean difference in postoperative VA was a -0.05 logMAR gain (2-3 ETDRS letters) (95% CI: −0.13 to 0.02, p = 0.155) in vision with a large ILM peel radius. The GRADE certainty of evidence was moderate for both outcomes. A significantly higher closure rate was found in the large peel group for iFTMH >400 microns with an OR of 1.24 (95% CI: 1.11–1.38, p < 0.001) and NNT to benefit of 21 (95% CI: 17–50), but not in holes <400 microns (OR 1.05 (95% CI: 0.93–1.18, p = 0.396)). Conclusions: Performing ILM peel of more than 1-DD in radius likely improves closure rates for iFTMH although the effect size is relatively small. The effect is greater in holes >400 microns. ILM peel size probably has no significant effect on postoperative VA. © The Author(s) 2025.