Small idiopathic full-thickness macular holes: a systematic review and meta-analysis

To evaluate the anatomical and visual outcomes of three vitrectomy techniques—internal limiting membrane (ILM) peeling, inverted flap, and no peeling—for the treatment of idiopathic small full-thickness macular holes (FTMH; ≤250 μm). A systematic review and meta-analysis was conducted according to PRISMA guidelines. PubMed, Embase, and Medline were searched up to April 2025. Eligible studies reported outcomes of small FTMH treated with vitrectomy using one of the three specified techniques, with a minimum follow-up of 3 months. Primary endpoints were FTMH closure rate and change in best-corrected visual acuity (BCVA, logMAR). Subgroup, meta-regression and sensitivity analyses were performed to evaluate technique-specific estimates and the influence of baseline BCVA. Eleven non-randomized studies comprising 957 eyes were included. The overall FTMH closure rate was 96.6% (95% CI: 93.2–98.3%; I² = 61.6%, τ² = 0.681, p = 0.0036). Closure rates were 98.4% for ILM peeling, 97.6% for inverted flap, and 86.5% for no peeling (test for differences between pooled technique subgroups: p < 0.0001; I² = 0% in all subgroups). The pooled BCVA improvement was 0.40 logMAR (95% CI: 0.36–0.44; I² = 64.4%, τ² = 0.0027, p = 0.002). Mean gains were 0.45 logMAR for no peeling, 0.41 for ILM peeling, and 0.39 for inverted flap (test for differences between pooled technique subgroups: p = 0.37). Exploratory meta-regression suggested that poorer baseline BCVA may be associated with a lower likelihood of closure. Sensitivity analyses confirmed the robustness of the main findings. ILM peeling showed the highest pooled closure estimate, while visual improvement did not differ significantly across technique subgroups. Given the predominantly observational and indirect evidence, these findings should be interpreted cautiously and do not establish comparative superiority.