Background Observational studies have demonstrated a dose-dependent association between vaginal douching and bacterial vaginosis. We sought to estimate the effect of douching cessation on the vaginal microbiota in a pilot crossover study.

Methods Thirty-two women self-collected vaginal swabs twice-weekly (n=950) during a douching observational phase (“D”, 4 weeks), followed by douching cessation (“DC”, 12 weeks). Vaginal microbiota were characterized by 16S rRNA gene sequencing (V3-V4) and clustered into community state types (CSTs). A conditional logistic regression model, adjusted for menstruation and sexual behaviors, allowed each woman to serve as her own control. Wilcoxon signed-rank tests were used to evaluate paired changes in microbiota between phases. Broad-range qPCR assays provided estimates of bacterial absolute abundance per swab. A piecewise linear mixed effects model was used to assess differences in rates of change in bacterial absolute abundance before and after douching.

Results There was not a statistically significant change in the odds of Lactobacillus-dominated CSTs comparing DC to D (aOR 0.54, 95% CI: 0.27–1.11). There were no significant changes for four individual Lactobacillus spp. and no meaningful changes in other taxa investigated. The rates of change in bacterial absolute abundance was not significantly different in samples collected 3 days before and after douching (p=0.46). Women who had a Lactobacillus-dominated CST at baseline experienced shifts to low-Lactobacillus CST in DC, and vice versa for women who had a low-Lactobacillus CST at baseline (interaction on entry CST, p-value <0.02), however, these findings were driven by changes occurring in the final weeks.

Conclusion In this pilot study, douching cessation was not associated with major changes in vaginal microbiota. Shifts in Lactobacillus-dominance may represent regression to the mean as the shifts occurred late in DC, giving ample time for fluctuations. Disparate findings between this study and prior analyses using Nugent score may be related to low-Lactobacillus CSTs receiving low/intermediate Nugent scores.

Disclosure No significant relationships.