Unraveling causal pathways in retinal vein occlusion: a systematic review of Mendelian randomization studies

Retinal vein occlusion (RVO) is a leading cause of vision loss, yet there are inconsistent risk estimates related to risk factors. Mendelian randomization (MR) uses genetic variants as proxies for lifelong exposure and can clarify causal pathways for RVO. We aimed to systematically review MR studies to identify causally supported systemic and ocular risk factors for RVO.

Four databases (PubMed, Embase, Scopus, and Web of Science) were searched from inception to June 2025 for peer-reviewed MR studies evaluating modifiable systemic or ocular risk factors in relation to any form of RVO utilizing GWAS data. Narrative synthesis was undertaken as methodological heterogeneity precluded meta-analysis. All effect estimates (ORs) were extracted directly from individual studies and robustness of evidence for each exposure across studies was assessed as robust, probable, suggestive, insufficient, and non-evaluable based on significance and direction of evidence.

Twelve two-sample MR studies, all conducted in European cohorts, met inclusion criteria. Ocular traits showed the most consistent signals: higher intraocular pressure (RVO (OR = 1.53, 95% CI: 1.0402.26) and glaucoma liability (OR = 1.31, 95% CI: 1.18-1.45) were robustly associated with greater risk of RVO. Among cardiovascular factors, elevated blood pressure/hypertension liability demonstrated probable evidence of increased RVO risk (OR = 1.58, 95% CI: 1.34-1.85), whereas lipid profiles yielded mixed signals, with some support for higher LDL (OR = 1.23, 95% CI: 1.05-1.44) and total cholesterol (OR = 1.44, 95% CI: 1.08-1.92) effects. For metabolic factors, glycemic traits showed probable to robust evidence with fasting glucose (OR = 5.01, 95% CI: 2.00-12.55) and two-hour glucose (OR = 3.17, 95% CI: 1.63-6.18) associated with higher RVO risk. Similarly, type 2 diabetes liability showed probable evidence (OR = 2.82, 95% CI: 2.07-3.85); anthropometric measures offered probable to robust support with body mass index (OR = 1.94, 95% CI: 1.23-3.08) and waist circumference (OR = 2.40, 95% CI: 1.36-4.24) associated with RVO. In other domains, selected coagulation and platelet traits showed probable-robust signals, vitamin D evidence was insufficient, and gut microbiota instruments provided preliminary robust evidence for Bacilli and Family XIII AD3011 association with RVO.

Genetic evidence supports a multifactorial vascular-metabolic model for RVO in which elevated IOP, glaucoma, hypertension, adiposity, and acute hyperglycemia are genetically supported risk factors. These findings highlight blood-pressure control, weight management, and glycemic regulation as important prevention targets and underscore the need for ancestry-diverse MR studies with refined phenotyping.