Although deregulation of miR-146a has been reported in type 2 diabetes repeatedly, the direction of deregulation events (up or down) remained to be inconsistent in literatures.  Therefore, in this study, we performed a meta-analysis on the possible association between miR-146a expression levels and type 2 diabetes. A systematic literature searching of PubMed, ISI Web of Science and Google Scholar was performed up to the end of September 2016. Finally, a total of 12 studies including 344 diabetic patients and 316 controls were selected for meta-analysis. All statistical analysis was performed using the meta for package with R software. Moreover, publication bias was assessed by Egger’s and sensitivity analysis was applied on the meta-analysis. The results are presented as log10 odds ratios (logORs), 95% confidence intervals (CI) with relevant P values. The results revealed that miR-146a was downregulated in type 2 diabetes cases compared with normal subjects (P=0.01, logOR:-4.76, 95% CI:-8.41, -1.11). Furthermore, sub-group analysis showed that the association between miR-146a expression levels and type 2 diabetes in whole blood (P<0.001) and PBMCs (P<0.001) samples were significant. However, this association was not significant in the serum (P=0.67) and plasma (P=0.90) samples. Our finding suggests that miR-146a downregulation could be associated with type 2 diabetes susceptibility. Further investigations with larger sample size are required to evaluate this association in the type 2 diabetes pathogenesis.