Objective To estimate the diagnostic value of mesothelin in ovarian cancer. Methods PubMed, The Cochrane Library, CBM, CNKI and WanFang Data databases were searched from inception to October 2016 to collect relevant diagnostic accuracy studies of mesothelin in ovarian cancer. Two reviewers independently screened literature, extracted data and assessed the risk of bias of included studies. Statistical analysis was performed using Meta-Disc 1.4, Stata 12.0 and RevMan 5.2 softwares. The pooled sensitivity, specificity and diagnostic odds ratio were calculated, the summary receiver operating characteristic curve (SROC) was drawn and the area under the curve (AUC) was calculated. Results Seventeen studies involving 2 052 patients were included. The pooled sensitivity, specificity, DOR were 0.63 (95%CI 0.60 to 0.67), 0.92 (95%CI 0.90 to 0.93) and 26.62 (95%CI 14.96 to 47.38), respectively. The AUC and Q index were 0.915 1 and 0.847 8, respectively. Conclusion The current evidence indicates that mesothelin has high specificity and low sensitivity, which can’t be used alone as a biomarker for the detection of ovarian cancer, but should be combined with other biomarkers.