RESUMO
BACKGROUND: Increasing evidences have shown that long noncoding RNAs (lncRNAs) are involved in cancer diagnosis and prognosis. However, the overall diagnostic accuracy of lncRNAs for hepatocellular carcinoma (HCC) remains unclear. Herein, we perform a meta-analysis to assess diagnostic value of lncRNAs for HCC. METHODS: The online PubMed, Cochrane, Web of Science, and Embase database were searched for eligible studies published until October 5, 2016. Study quality was evaluated with the Quality Assessment for Studies of Diagnostic Accuracy (QUADAS). All statistical analyses were conducted with Stata 12.0 and Meta-Disc 1.4. RESULTS: We included 19 studies from 10 articles with 1454 patients with HCC and 1300 controls. The pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and AUC for lncRNAs in the diagnosis of HCC were 0.83 (95% confidence interval [CI]: 0.76-0.88), 0.80 (95% CI: 0.73-0.86), 4.2 (95% CI: 3.00-5.80), 0.21 (95% CI: 0.15-0.31), 20 (95% CI: 11-34), and 0.88 (95% CI: 0.85-0.91), respectively. Additionally, the diagnostic value of lncRNAs varied based on sex ratio of cases and characteristics of methods (specimen type and reference gen). CONCLUSION: This meta-analysis suggests lncRNAs show a moderate diagnostic accuracy for HCC. However, prospective studies are required to confirm its diagnostic value.