Summary

Several small case-control studies have investigated whether factor V Leiden (FVL) is a risk factor for retinal vein occlusion (RVO) and generated conflicting data. To clarify this question we performed a large two-centre case-control study and a metaanalysis of published studies. Two hundred seven consecutive patients with RVO and a control group of 150 subjects were screened between 1996 and 2006. A systematic meta-analysis was done combining our study with further 17 published European case-control studies. APC resistance was detected in 16 out of 207 (7.7%) patients and eight out of 150 (5.3%) controls. The odds ratio (OR) estimated was 1.49 with a (non-significant) 95% confidence interval (CI) of 0.62–3.57. The meta-analysis including 18 studies with a total of 1,748 patients and 2,716 controls showed a significantly higher prevalence of FVL in patients with RVO compared to healthy controls (combined OR 1.66; 95% CI 1.19–2.32). All single studies combined in the metaanalysis were too small to reliably detect the effect individually. This explains the seemingly contradictory data in the literature. In conclusion, the prevalence of APC resistance (and FVL) is increased in patients with RVO compared to controls, but the effect is only moderate. Therefore, there is no indication for general screening of factor V mutation in all patients with RVO. We recommend this test to be performed in patients older than 50 years with an additional history of thromboembolic event and in younger patients without general risk factors like hypertension.