Epiretinal Membrane Vitrectomy With and Without Intraoperative Intravitreal Dexamethasone Implant: A Systematic Review With Meta-Analysis

Abstract

Purpose: To evaluate the efficacy of vitrectomy combined with intravitreal dexamethasone implant vs. vitrectomy without the implant in patients with epiretinal membrane (ERM) by conducting a systematic review and meta-analysis.Methods: Studies that compared ERM vitrectomy with and without intraoperative dexamethasone implant with a follow-up ≥3 months were included. The primary outcome was mean best corrected visual acuity (BCVA) change between eyes undergoing ERM vitrectomy combined with dexamethasone implant (DEX group) and eyes undergoing ERM vitrectomy alone (control group) at 3 months. Secondary outcomes included mean BCVA change at 6 months and mean optical coherence tomography central macular thickness (CMT) change at both 3-months and 6-months follow-up. Mean differences (MDs) with their 95% confidence interval (95%CI) were calculated. Meta-analyses were based either on random effect model or fixed effect model according to heterogeneity.Results: Four studies were included. At 3 months, ERM vitrectomy combined with dexamethasone implant yielded a greater visual gain compared to vitrectomy alone (MD = 9.7; 95%CI = 2.6–16.8; p = 0.01). However, significant heterogeneity was found. A sensitivity analysis excluding the only retrospective non-randomized study confirmed a greater visual gain in the DEX group (MD = 7.1; 95%CI = 2.7–11.6; p < 0.01), with no heterogeneity. At 6 months, a non-significant but borderline difference in visual gain was shown between in the two groups (MD = 5.1; 95%CI = −0.3–10.5; p = 0.06), with no heterogeneity. Three-month analysis of CMT revealed a greater reduction in the DEX group (MD = −80.2; 95%CI =−149.1–11.2; p = 0.02), but with significant heterogeneity. A sensitivity analysis excluding the only retrospective non-randomized study allowed to reduce heterogeneity, but no difference in 3-months CMT change was found between the two groups (MD = −50.0; 95%CI = −106.2–6.2; p = 0.08). At 6 months, no difference in CMT change was shown between the two groups (MD = −48.5; 95%CI = −120.5–23.5; p = 0.19), with significant heterogeneity.Conclusions: Intraoperative dexamethasone implant in eyes undergoing vitrectomy for ERM provided a better visual outcome at 3 months compared to ERM vitrectomy without the implant, with limited evidence of better anatomic outcome as well. Further studies are needed to ascertain whether dexamethasone implant would ensure a significant long-term visual benefit as a result of a faster reduction of macular thickening.