The aberrant expression of microRNAs (miRNAs) is associated with a variety of diseases, including cancer. In our study, we examined the miRNA expression profile of meningiomas, which is a common type of benign intracranial tumor derived from the protective meninges membranes that surround the brain and spinal cord. To define a typical human meningioma miRNA profile, the expression of 200 miRNAs in a training sample set were screened using quantitative reverse transcription polymerase chain reaction analysis, and then significantly altered miRNAs were validated in a secondary independent sample set. Kaplan–Meier and univariate/multivariate Cox proportional hazard regression analyses were performed to assess whether miRNA expression could predict the recurrence of meningioma after tumor resection. After a two‐phase selection and validation process, 14 miRNAs were found to exhibit significantly different expression profiles in meningioma samples compared to normal adjacent tissue (NAT) samples. Unsupervised clustering analysis indicated that the 14‐miRNA profile differed between tumor and NAT samples. Downregulation of miR‐29c‐3p and miR‐219‐5p were found to be associated with advanced clinical stages of meningioma. Kaplan–Meier analysis showed that high expression of miR‐190a and low expression of miR‐29c‐3p and miR‐219‐5p correlated significantly with higher recurrence rates in meningioma patients. Cox proportional hazard regression analysis revealed that miR‐190a expression level is an important prognostic predictor that is independent of other clinicopathological factors. Our results suggest that the use of miRNA profiling has significant potential as an effective diagnostic and prognostic marker in defining the expression signature of meningiomas and in predicting postsurgical outcomes. What's new? The aberrant expression of microRNAs is associated with a variety of diseases, including cancer. In the present study, we examined the miRNA expression profile of meningiomas, and 14 miRNAs were found to exhibit significantly different expression profiles in meningioma samples compared with NAT samples. High expression of miR‐190a and low expression of miR‐29c‐3p and miR‐219‐5p correlated significantly with higher recurrence rates in meningioma patients, which indicated the potential use of miRNAs in predicting meningioma recurrence.