Abstract: To quantitatively evaluate the diagnostic accuracy of the serum Epididymal Secretory Protein E4, for human epithelial ovarian cancer ( OEC ). Methods: A systematic search for relevant studies published between January 1997 and March 2011 was performed using the Medline, Embase, and China National Knowledge Infrastructure ( CNKI ) databases, among others. The criteria for inclusion were established based on the validity criteria for diagnostic research published by the Cochrane Methods Group on Screening and Diagnostic Test. Two reviewers independently assessed the methodological quality of each study by QUADAS ( Quality Assessment of Diagnostic Accuracy Studies ). Statistical analysis was performed employing MATLAB Review Manager 4.2 and Meta-Disc1.4. A meta-analysis of the reported sensitivity and specificity of each study and the Summary Receiver Operating Characteristic ( SROC ) curve was performed. Results: A total of 14 studies with benign control, and 14 with benign and healthy control were included. The area under the curve ( AUC ) of SROC was 0.8882 ± 0.0150, with Q index 0.8188 ± 0.0155 for benign control, and AUC was 0.8903 ± 0.0346 with Q index 0.8210 ± 0.0359 for benign and healthy control.  Detection assay and population groups were the causes of heterogeneity by meta regression analysis. Conclusion: The assay method of race and country was the important source of heterogeneity. With the relative high AUC and Q index of HE4 antibody in patients with ovarian cancer, the positive HE4 antibody might be a useful parameter for the clinical diagnosis of patients with epithelial ovarian cancer.