Quantification of the association between selenium and risk of oesophageal adenocarcinoma (OAC) is still conflicting. The purpose of this meta-analysis is to explore the relationship between selenium levels and OAC risk. PubMed and Web of Knowledge were searched for the related articles. Pooled relative risks (RRs) with 95% confidence intervals (CIs) from random effects models were calculated. Sensitivity analysis and publication bias were conducted. Dose-response relationship was assessed by restricted cubic spline and variance-weighted least squares regression analysis. Five articles involving 748 OAC cases were included in this meta-analysis. Pooled results suggest that higher selenium level was not significantly associated with the risk of OAC (summary RRs=1.08, 95% CIs=0.84-1.39, I(2)=0%). Besides, no significant association was found in case-control studies (summary RRs=1.13, 95% CIs=0.84-1.52, I(2)=0%) or cohort studies (summary RRs=0.99, 95% CIs=0.55-1.78, I(2)=32.6%). A linear dose-response relationship was attested that an increase in dietary selenium intake of 10 μg/day is marginally associated with 1% increase in the risk of developing OAC (summary RRs=1.01, 95% CIs=0.99-1.03), but not statistically significant. No publication bias was found. In conclusion, our analysis indicated that a higher selenium level was not significantly associated with the risk of OAC. The relevant further studies are warranted.