Background: Abnormal expression levels of microRNAs (miRNAs) were observed in ankylosing spondylitis (AS) in recent articles, suggesting that miRNAs may be used as biomarkers for AS diagnoses. In this paper, we conducted a meta-analysis to identify the overall diagnostic accuracy of miRNA biomarkers in AS patients. Methods: An extensive search was undertaken in PubMed, Embase, Cochrane databases, and Wan Fang database up to 30 December 2020 using the following key words: ("microRNAs" or "microRNA" or "miRNA" or "miR" or "RNA, Micro" or "Primary MicroRNA") and ("Spondylitis Ankylosing" or "Spondyloarthritis Ankylopoietica" or "Ankylosing Spondylarthritis" or "Ankylosing Spondylarthritides" or "Spondylarthritides Ankylosing" or "Ankylosing Spondylitis") and ("blood" or "serum" or "plasma"). Statistical evaluation of dysregulated miRNAs using the sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), and the area under the curve (AUC). Results: Twenty-nine articles reporting on the miRNAs of AS were included. A total of 42 miRNAs were observed to be up-regulated and 45 miRNAs were down-regulated in the AS cases compared with the controls. Besides, 29 studies from nine articles were included in our meta-analysis. The pooled sensitivity, specificity, PLR, NLR, DOR, and AUC were 0. 76 (95% CI, 0.70-0.81), 0.80 (95% CI, 0.74-0.85), 3.75 (95% CI, 2.82-5.01), 0.30 (95% CI, 0.24-0.39), 12.32 (95% CI, 7.65-19.83), 0.85 (95% CI, 0.81-0.88), respectively, suggesting a good diagnostic accuracy of miRNAs for AS. Conclusions: Circulating miRNAs are deregulated in AS patients. miRNAs may be used as a relatively non-invasive biomarkers for the detection of AS.