MicroRNAs (miRNAs) have been proposed as promising diagnostic biomarkers for many diseases, particularly in the field of cancer research. Numerous studies have explored the use of miRNAs in the detection of hepatocellular carcinoma (HCC), with some reporting inconsistent results. Thus, we conducted this meta-analysis to evaluate the potential diagnostic value of miRNAs in HCC. All relevant literature was collected from the PubMed and other databases before June 3, 2014. The summary receiver operator characteristic (SROC) curve and other parameters were used to estimate overall predictive performance. Fourteen studies involving 1,848 cases with HCC and 1,187 controls (576 healthy controls and 611 individuals with chronic liver diseases) were included in this meta-analysis. SROC analyses for the diagnostic power of miRNAs yielded an area under the curve (AUC) of 0.93 with 86 % sensitivity and 86 % specificity in discriminating patients with HCC from healthy subjects and an AUC of 0.88 with 79 % sensitivity and 83 % specificity in differentiating patients with HCC from those with chronic liver diseases (CLDs). Furthermore, subgroup analyses showed that miRNA panels yielded excellent diagnostic characteristics, with an AUC of 0.99 (96 % sensitivity and 96 % specificity) for detection of HCC from healthy controls and an AUC of 0.93 (85 % sensitivity and 88 % specificity) for HCC from those with CLDs. MiRNAs might be novel potential biomarkers for the diagnosis of HCC, and a combination of multiple miRNAs could significantly improve the diagnostic accuracy.