Abstract
BACKGROUND: Increasing number of studies have demonstrated certain patterns of microbial changes in gynecological diseases; however, the interaction between them remains unclear. To evaluate the consistency or specificity across multiple studies on different gynecological diseases and microbial alterations at different sites of the body (gut and genital tract), we conducted a systematic review and meta-analysis. METHODS: We searched PubMed, Embase, Web of Science, and Cochrane Library up to December 5, 2022(PROSPERO: CRD42023400205). Eligible studies focused on gynecological diseases in adult women, applied next-generation sequencing on microbiome, and reported outcomes including alpha or beta diversity or relative abundance. The random-effects model on standardized mean difference (SMD) was conducted using the inverse-variance method for alpha diversity indices. RESULTS: Of 3327 unique articles, 87 eligible studies were included. Significant decreases were found in gut microbiome of patients versus controls (observed species SMD=-0.35; 95%CI, -0.62 to -0.09; Shannon index SMD=-0.23; 95%CI, -0.40 to -0.06), whereas significant increases were observed in vaginal microbiome (Chao1 SMD = 1.15; 95%CI, 0.74 to 1.56; Shannon index SMD = 0.51; 95%CI, 0.16 to 0.86). Most studies of different diagnostic categories showed no significant differences in beta diversity. Disease specificity was observed, but almost all the changes were only replicated in three studies, except for the increased Aerococcus in bacterial vaginosis (BV). Patients with major gynecological diseases shared the enrichment of Prevotella and depletion of Lactobacillus, and an overlap in microbes was implied between BV, cervical intraepithelial neoplasia, and cervical cancer. CONCLUSIONS: These findings demonstrated an association between alterations in gut and genital microbiota and gynecological diseases. The most observed results were shared alterations across diseases rather than disease-specific alterations. Therefore, further investigation is required to identify specific biomarkers for diagnosis and treatment in the future.