Surgical management, use and efficacy of adjuvant dyes in idiopathic epiretinal membranes: a systemic review with network meta-analysis

Background: The epiretinal membrane (ERM) is a nonvascular fibrocellular tissue formed by cellular metaplasia and proliferation at the vitreoretinal surface and is generally treated by pars plana vitrectomy (PPV) with or without internal limiting membrane (ILM) peeling. This network meta-analysis aimed to compare the efficacy of all available ERM removal interventions and assessed the use and efficacy of surgical dyes in managing idiopathic ERMs.

Methods: MEDLINE, EMBASE, Cochrane CENTRAL, and the US National Library of Medicine were searched (June 28, 2023). Clinical studies that included patients with ERMs were included. Randomized controlled trials (RCTs) were also appraised using Cochrane risk of bias (ROB).

Results: Ten RCTs and ten non-RCTs were included in this study. A pairwise meta-analysis between ERM removal and combined ERM and ILM removal showed no significant difference in visual outcome (change in BCVA) 1 year postintervention (MD = - 0.0034, SE = 0.16, p = 0.832). Similarly, there was no significant difference in the central macular thickness postoperatively between the two groups (MD = - 4.95, SE = 11.11, p = 0.656) (Q = 4.85, df = 3, p = 0.182, I² = 41.21%). The difference in ERM recurrence between the groups was also not statistically significant (OR = 4.64, p = 0.062, I² = 0). In a network meta-analysis, there was no significant difference in visual outcomes between ERM removal only and other treatment modalities: combined ILM and ERM removal (MD = 0.039, p = 0.837) or watchful waiting (MD = 0.020, p = 0.550). In a network meta-analysis, there was no significant difference in the visual outcomes between ERM removal alone and dye-stained combined ERM and ILM peeling (MD = 0.122, p = 0.742 for brilliant blue G; BBG and MD = 0.00, p = 1.00 for membrane blue-dual; MBD). The probability of being a better surgical dye for better visual outcomes was 0.539 for the MBD group and 0.396 for the BBG group. The recurrence of ERM was not significantly different when the ILM was stained with any of the dyes. No study was judged on ROB assessment as having low ROB in all seven domains.

Conclusion: The two types of surgical modalities provided comparable efficacy, with no significant differences between the outcomes. Among the dye-assisted ILM peeling methods, the membrane blue-dual dye was the most effective in providing better structural and functional outcomes.