Intra-uterine microbial colonization and occurrence of endometritis in women with endometriosis†

Is there any risk of intra-uterine bacterial colonization and concurrent occurrence of endometritis in women with endometriosis?

An increase in intra-uterine microbial colonization and concurrent endometritis occurred in women with endometriosis that was further increased after GnRH agonist (GnRHa) treatment.

Higher bacterial contamination of menstrual blood and increased endotoxin level in menstrual and peritoneal fluids have been found in women with endometriosis than in control women. However, information on intra-uterine microbial colonization across the phases of the menstrual cycle and possible occurrence of endometritis in women with endometriosis is still lacking.

This is a case-controlled study with prospective collection of vaginal smears/endometrial samples from women with and without endometriosis and retrospective evaluation.

Vaginal smears and endometrial smears were collected from 73 women with endometriosis and 55 control women. Twenty of the women with endometriosis and 19 controls had received GnRHa therapy for a period of 4-6 months. Vaginal pH was measured by intra-vaginal insertion of a pH paper strip. The bacterial vaginosis (BV) score was analyzed by Gram-staining of vaginal smears and based on a modified Nugent-BV scoring system. A panel of bacteria was analyzed by culture of endometrial samples from women treated with GnRHa or not treated. Immunohistochemcial analysis was performed using antibody against Syndecan-1 (CD138) and myeloperoxidase in endometrial biopsy specimens from women with and without endometriosis.

A significant shifting of intra-vaginal pH to ≥4.5 was observed in women with endometriosis compared with control women (79.3 versus 58.4%, P < 0.03). Compared with untreated women, use of GnRHa therapy also shifted vaginal pH to ≥4.5 in both control women (P = 0.004) and in women with endometriosis (P = 0.03). A higher risk of increasing intermediate flora (total score, 4-6) (P = 0.05) was observed in women with endometriosis who had GnRHa treatment versus untreated women. The number of colony forming units (CFU/ml) of Gardnerella, α-Streptococcus, Enterococci and Escherichia coli was significantly higher in endometrial samples from women with endometriosis than control women (P < 0.05 for each bacteria). GnRHa-treated women also showed significantly higher colony formation for some of these bacteria in endometrial samples than in untreated women (Gardnerella and E. coli for controls; Gardnerella, Enterococci and E. coli for women with endometriosis, P < 0.05 for all). Although there was no significant difference in the occurrence of acute endometritis between women with and without endometriosis, both GnRHa-treated controls and women with endometriosis had a significantly higher occurrence of acute endometritis (P = 0.003 for controls, P = 0.001 for endometriosis versus untreated women). Multiple analysis of covariance analysis revealed that an intra-vaginal pH of ≥4.5 (P = 0.03) and use of GnRHa (P = 0.04) were potential factors that were significantly and independently associated with intra-uterine microbial colonization and occurrence of endometritis in women with endometriosis. These findings indicated the occurrence of sub-clinical uterine infection and endometritis in women with endometriosis after GnRHa treatment.

We cannot exclude the introduction of bias from unknown previous treatment with immunosuppressing or anti-microbial agents. We have studied a limited range of bacterial species and used only culture-based methods. More sensitive molecular approaches would further delineate the similarities/differences between the vaginal cavity and uterine environment.

Our current findings may have epidemiological and biological implications and help in understanding the pathogenesis of endometriosis and related disease burden. The worsening of intra-uterine microbial colonization and higher occurrence of endometritis in women with endometriosis who were treated with GnRHa identifies some future therapeutic avenues for the management, as well as prevention of recurrence, of endometriosis. Further studies are needed to examine intra-uterine colonization of a broad range of common bacteria as well as different viruses and their role in the occurrence of endometritis.

This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Sports, Culture, Science and Technology of Japan. There is no conflict of interest related to this study.

Not applicable.