Background The results of human observational studies on the correlation between gut microbiota perturbations and polycystic ovary syndrome (PCOS) have been contradictory. This study aimed to perform the first systematic review and meta-analysis to evaluate the specificity of the gut microbiota in PCOS patients compared to healthy women. Methods Literature through May 22, 2023, was searched on PubMed, Web of Science, Medline, Embase, Cochrane Library, and Wiley Online Library databases. Unreported data in diversity indices were filled by downloading and processing raw sequencing data. Systematic review inclusion: original studies were eligible if they applied an observational case-control design, performed gut microbiota analysis and reported diversity or abundance measures, sampled general pre-menopausal women with PCOS, and are longitudinal studies with baseline comparison between PCOS patients and healthy females. Systematic review exclusion: studies that conducted interventional or longitudinal comparisons in the absence of a control group. Two researchers made abstract, full-text, and data extraction decisions, independently. The Joanna Briggs Institute Critical Appraisal Checklist was used to assess the methodologic quality. Hedge's g standardized mean difference (SMD), confidence intervals (CIs), and heterogeneity (I-2) for alpha diversity were calculated. Qualitative syntheses of beta-diversity and microbe alterations were performed. Results Twenty-eight studies (n = 1022 patients, n = 928 control) that investigated gut microbiota by collecting stool samples were included, with 26 and 27 studies having provided alpha-diversity and beta- diversity results respectively. A significant decrease in microbial evenness and phylogenetic diversity was observed in PCOS patients when compared with control participants (Shannon index: SMD = - 0.27; 95% CI, - 0.37 to - 0.16; phylogenetic diversity: SMD = - 0.39; 95% CI, - 0.74 to - 0.03). We also found that reported beta-diversity was inconsistent between studies. Despite heterogeneity in bacterial relative abundance, we observed depletion of Lachnospira and Prevotella and enrichment of Bacteroides, Parabacteroides, Lactobacillus, Fusobacterium, and Escherichia/Shigella in PCOS. Gut dysbiosis in PCOS, which might be characterized by the reduction of short-chain fatty acid (SCFA)-producing and bile-acid-metabolizing bacteria, suggests a shift in balance to favor pro-inflammatory rather than anti-inflammatory bacteria. Conclusions Gut dysbiosis in PCOS is associated with decreased diversity and alterations in bacteria involved in microbiota-host crosstalk.