A literature search without filters was conducted in PubMed, Embase, Scopus, Cochrane, and Web of Science databases up to September 2025 via Boolean logic and Medical Subject Heading terms with the following search strategy: (“LHEP” OR “Lamellar hole-associated epiretinal proliferation” OR “degenerative lamellar macular hole” OR “lamellar hole epiretinal proliferation”) AND (“vitrectomy” OR “surgery” OR “surgical treatment” OR “observation” OR “conservative management”) AND (“retinal morphology” OR “visual acuity” OR “anatomic” OR “anatomy” OR “progression” OR “full-thickness macular hole” OR “FTMH” OR “outcome” OR “OCT” OR “Optical Coherence Tomography”). The results were exported to Mendeley (Elsevier©, Netherlands), and duplicates were excluded.

The literature search strategy and research question were designed using the patient–intervention–comparator–outcome (PICO) framework. This study was specifically designed to evaluate LHEP as a distinct OCT-defined phenotype. Eligible studies included adult patients with LMH presenting LHEP who underwent vitrectomy, regardless of surgical technique. Patients with LMH without LHEP or with other associated retinal conditions were excluded. Patients with LMH and LHEP who were managed conservatively were included as comparators.

The primary outcome was change in visual acuity, and the secondary outcome was the rate of conversion to full-thickness macular hole (FTMH). Both prospective and retrospective studies, including randomized and nonrandomized designs, were eligible given the limited number of available studies. No restrictions were applied regarding publication date or geographic location; however, only studies with full text available in English were included. Studies were excluded if they were case reports, conference abstracts or proceedings, systematic reviews, meta-analyses, or theses.

Studies were excluded from quantitative synthesis if outcome data could not be extracted in a format compatible with meta-analysis (eg, absence of baseline and final visual acuity values, lack of discrete FTMH event reporting, or nonstandardized outcome reporting formats). To ensure methodological consistency and validity of pooled estimates, only studies with extractable and standardized quantitative outcomes were included, which may have led to the exclusion of studies reporting results in noncomparable formats.

Four independent reviewers comprehensively reviewed the titles and abstracts of all the publications. Full-text articles that met the inclusion criteria were retrieved and assessed for eligibility. Discrepancies in the inclusion of the studies were discussed with a fifth reviewer to reach a consensus. The following information was collected from the included studies and added to a Microsoft Excel (Microsoft, USA) spreadsheet: first author, year of publication, location of the study, vitrectomy technique, number of patients who underwent vitrectomy, vitrectomy technique, number of patients observed, patients’ mean age and SD (SD) (vitrectomy group and observation group), mean and SD of the baseline visual acuity in logMAR (vitrectomy group and observation group), mean and SD of the final visual acuity in logMAR (vitrectomy group and observation group), number of patients who developed FTMH at the end of the follow-up period (vitrectomy group and observation group), and mean and SD follow-up time (vitrectomy group and observation group).

The risk of bias for individual studies was independently assessed by 2 reviewers, with the Newcastle–Ottawa Scale for case‒control studies, prospective longitudinal studies, and cohort studies. RoB2 was used to assess risk of bias in randomized clinical trials. ^, Article bias was also assessed by visual analysis of the funnel plot for evidence asymmetry and Egger’s test P value. To explore the potential impact of funnel plot asymmetry on the pooled estimate, the trim-and-fill method was employed.

Statistical analysis was performed via IBM SPSS version 30.0 (IBM, Inc., Chicago, IL). Continuous outcomes (visual acuity) were reported as the mean ± SD, mean difference (final– baseline), and SD of change according to the formula $S{D}_{change}=\sqrt{S{D}_{baseline}^2+S{D}_{final}^2-\left(2xcorrxS{D}_{baseline}xS{D}_{final}\right)}$ , and the correlation factor was considered 0.7, according to previously described methods. ^, Binary outcomes (progression to FTMH) were reported as the total number of events. Individual effect size and SE (SE) were estimated via logit transformation with continuity correction to lead with zeros ( $effectsize=\text{ln}\left(\frac{p}{1-p}\right)$ ; $SE=\sqrt{\frac{1}{np}+\frac{1}{n\left(1-p\right)}}$ , where P = the proportion of patients progressing to FTMH, n = the total number of patients, and 0.5 was used for continuity correction), according to previous methods. The pooled rate was calculated by the overall effect size ( $pooledrate=\frac{e^{\left(Logitvalue\right)}}{1+e^{\left(Logitvalue\right)}}$ ), according to previously described methods. A 95% CI (CI) was also calculated for effect size. Homogeneity between the included studies was assessed via Q statistics and its P value, and heterogeneity was assessed via I ². The results are illustrated in forest plots and were considered significant if the P value <.05.