Background In recent years, many studies focused on the association between the microRNAs (miRNAs) and the risk of systemic lupus erythematosus (SLE), especially miRNA-21 (miR-21). We aimed to investigate the role of circulating miRNAs, especially the miR-21, as a biomarker in detecting SLE. Methods We searched PubMed, EMBASE, the Cochrane Central Register of Controlled Trials, and China National Knowledge Infrastructure through Mar 3th, 2019. We performed this meta-analysis in a fixed/random-effect model using Meta-disc 1.4 and STATA 15.1. Results A total of 17 relevant studies were eligible to analyze pooled accuracy. The overall performance of total mixed miRNAs (TmiRs) detection was: pooled sensitivity, 0.71 (95% confidence interval [CI], 0.69 to 0.72); pooled specificity, 0.81 (95%CI, 0.79 to 0.83); and area under the summary receiver operating characteristic curves value (SROC), 0.8797. The miR-21 detection was: pooled sensitivity, 0.68 (95%CI, 0.62 to 0.74); pooled specificity, 0.77 (95%CI, 0.69 to 0.84); and SROC, 0.8281. The meta-regression analysis showed that the type of samples was the sources of heterogeneity. The subgroup analysis suggested that detection in plasma group had the largest AUC of SROC in all the subgroups: pooled sensitivity, 0.8 (95%CI, 0.78 to 0.82); pooled specificity, 0.83 (95%CI, 0.8 to 0.86); and SROC, 0.9068. Conclusions Our meta-analysis demonstrated that circulating miRNAs might be potential novel biomarkers for detecting SLE, especially miR-21. Moreover, plasma is recommended as the clinical specimen for diagnostic detection.