Thyroid cancer (TC) is the most common endocrine system tumor. Several studies had revealed the potential of circulating microRNAs (miRNAs) as novel biomarkers for the diagnosis of TC. The purpose of this meta-analysis is to summarize published studies and evaluate the diagnostic accuracy of circulating miRNAs in TC detection.In this meta-analysis we systematically searched three databases: PubMed, EMBASE and Cochrane Library. We used the bivariate mixed-effects regression model to calculate the pooled diagnostic parameters and conduct the summary receiver operator characteristic curve (SROC). All calculations were performed using STATA software.Thirty-five studies from 9 articles, including 663 TC patients, 519 patients with benign thyroid nodules (BTNs), and 84 healthy controls were included in this meta-analysis. The pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR) and area under the SROC curve(AUC) were 0.81 (95%CI 0.75-0.86), 0.81 (95%CI 0.75-0.86), 4.3 (95%CI 3.2-5.6), 0.24 (95%CI 0.18-0.31), 18 (95%CI 12-28), and 0.88 (95%CI 0.85-0.90) respectively in BTN controls, and 0.81(95%CI 0.75-0.86), 0.85(95%CI 0.75-0.91), 5.3(95%CI 3.3-8.7), 0.23(95%CI 0.18-0.29), 24(95%CI 14-39), 0.89 (95%CI 0.86-0.91) in healthy controls. The sub-group analysis found that multiple miRNA assays had higher diagnostic accuracy than single miRNA assays with sensitivity of 0.88, specificity of 0.89 and AUC of 0.94.Circulating miRNAs have good values to diagnose TC and distinguish TC patients from BTN patients. MiRNAs can assist in the diagnosis of malignancy and avoid unnecessary surgery. In summary, circulating miRNAs should be added to our current clinical tools.