Abstract
BACKGROUND: To evaluate macular microvascular changes measured by optical coherence tomography angiography (OCTA) in patients with inactive Graves’ ophthalmopathy (GO) compared with healthy controls by performing a meta-analysis of published studies. METHODS: A systematic literature search of PubMed, Scopus, and Web of Science was performed through August 2025 according to PRISMA guidelines. The inclusion criteria were original peer-reviewed studies comparing OCTA macular vessel density (VD) parameters or foveal avascular zone (FAZ) area between inactive GO patients (clinical activity score < 3) and healthy controls, with means and standard deviations reported. Extracted outcomes included foveal VD, parafoveal VD, parafoveal temporal/superior/nasal/inferior VD, and FAZ area. Effect sizes were calculated as Hedges’ g with 95% confidence intervals (CIs). Heterogeneity was assessed with Cochran’s Q and I(2); fixed or random effects models were applied accordingly. Publication bias was evaluated using Egger’s test. RESULTS: Six studies met inclusion criteria, comprising 433 eyes from inactive GO patients and 380 eyes from healthy controls. Meta-analysis found no significant differences between inactive GO patients and controls for: foveal VD (Hedges’ g = 0.040, 95% CI=[− 0.121 to 0.201], p = 0.628, I(2) = 0%), parafoveal VD (Hedges’ g = 0.044, 95% CI=[− 0.357 to 0.445], p = 0.830, I(2) = 73.5%), parafoveal temporal VD (Hedges’ g = 0.063, 95% CI=[− 0.212 to 0.337], p = 0.656, I(2) = 70%), parafoveal superior VD (Hedges’ g = 0.043, 95% CI=[− 0.293 to 0.380], p = 0.801, I(2) = 80%), parafoveal inferior VD (Hedges’ g = 0.176, 95% CI=[− 0.132 to 0.585], p = 0.263, I(2) = 76%), parafoveal nasal VD (Hedges’ g = − 0.043, 95% CI=[− 0.334 to 0.246], p = 0.775, I(2) = 73.3%), and FAZ area (Hedges’ g = 0.248, 95% CI=[− 0.207 to 0.703], p = 0.286, I(2) = 70.2%). Egger’s tests did not indicate significant publication bias for the analyzed outcomes. Moderate to high heterogeneity was observed for most parameters except foveal VD. CONCLUSION: This meta-analysis revealed no significant differences in macular VD and FAZ area between inactive GO patients and healthy controls. The absence of differences suggests that macular microvascular alterations in GO may be associated primarily with active inflammatory disease and may normalize in inactive stages, or that macular VD is not a sensitive biomarker for inactive disease.