Abstract: This study was intended to combine previous results of different independent researches and judge them integrally to renew our knowledge of the corelationship between selenium and the activity of GPx from the aspect of evidencebased medicine. A Metaanalysis was facilitated, by searching multidatabases and related resources, and performed by combining and quantitative previous multiresults of different independent researches. The different characteristics of each research were checked. The results of Metaanalysis showed that there was a significant difference of GPx activity in chicks′ hepatic tissue between experimental and control groups (P<0.01) and there was a positive corelationship between seleniumand the activity of GPx. But the result of Metaanalysis also indicated that there were the facts that the effect of selenium on GPx activity had been overestimated or underestimated in the past, and an extra 26.79 μg·mg1 of GPx activity in comparison with control group in chick′s hepatic tissue had statistic importance when 02 mg·kg1 selenium was supplied in diets. The results of the corelationship between selenium and the activity of GPx generated by classical methods were credible, while it was unprecious to predict the degree of the corelationship and contrary to each other sometime. This historical phenomena just right reflected the necessary of Metaanalysis and this is in keeping with the development of veterinary medicine.