LAUSR

Background The vaginal microbiota can be altered by uropathogenic bacteria associated with recurrent cystitis (RC), and the vaginal administration of Lactobacillus have suggested certain effects to prevent RC. The relationship between vaginal microbiota and the development of RC has not been elucidated. We aimed to clarify the etiology of RC from vaginal microbiota and importance of vaginal Lactobacillus. Methods Vaginal samples obtained from 39 postmenopausal women were classified into four groups: healthy controls; uncomplicated cystitis; RC; and prevention (prevented RC by Lactobacillus crispatus-containing vaginal suppositories). Principal coordinate analysis and beta-diversity analysis was used to assess 16S rRNA gene sequencing data from the vaginal microbiome. Results Cluster analysis divided the vaginal bacterial communities among 129 vaginal samples into three clusters (A, B, and C). Fourteen of 14 (100%) samples from the RC group and 51 of 53 (96%) samples from the prevention group were in clusters B and C, while 29 of 38 (76%) samples from the healthy group and 14 of 24 (58%) samples from the uncomplicated cystitis group were in cluster A. The principal coordinate analysis showed that plots in the uncomplicated cystitis group were similar to the healthy group, indicating a large separation between the RC group and the uncomplicated cystitis group. On beta-diversity analysis, there were significant differences between the healthy group and the uncomplicated cystitis group (p = 0.045), and between the RC group and the uncomplicated cystitis group or the healthy group (p = 0.001, p = 0.001, respectively). There were no significant differences between the RC group and the prevention group (p = 0.446). The top six taxa were as follows: Prevotella, Lactobacillus, Streptococcus, Enterobacteriaceae, Anaerococcus, and Bifidobacterium. Among patients with RC, Lactobacillus was undetectable before administration of suppositories, while the median relative abundance of Lactobacillus was 19% during administration of suppositories (p = 0.0211), reducing the average cystitis episodes per year (6.3 vs. 2.4, p = 0.0015). Conclusion The vaginal microbiota of postmenopausal women with RC is differed from healthy controls and uncomplicated cystitis in terms of lack of Lactobacillus and relatively dominant of Enterobacteriaceae. Vaginal administration of Lactobacillus-containing suppositories can prevent RC by stabilizing vaginal dysbiosis and causing a loss of pathogenic bacteria virulence.