Jan Friberg M.D., Ph.D.
Studies on the bacteriology of the female genital tract was for a long time focused on the vaginal and cervical areas. This portion of the body is easily accessible and has the advantage of having an abundance of bacteria present. Until the second half of the last century, the endometrium was considered sterile but we now know that this is not the case. The vagina has been estimated to harbor an estimated 10 10 bacteria, but the upper female genital tract contains much fewer bacteria estimated to be in the magnitude of 10 6 bacteria. Women with the vaginal/cervical colonization of mostly Lactobacillus bacteria have a better prognosis in infertility treatment than women colonized with organisms associated with bacterial vaginosis and/or women with many different types of bacteria present (so called dysbiosis).
New molecular techniques for isolation of bacteria using NGS (next generation screening) of the 16S ribosomal RNA (16SrRNA), gene sequencing and whole metagenome sequencing has made it possible to investigate the microbiome in areas with very low bacterial content such as the uterus, tubes and ovaries, areas that previously were considered sterile. The microbiome of the uterus is also of special interest since it is there that implantation with embryo development occurs and the presence of different bacteria appears to influence the implantation.
The vaginal area and endometrial tissue usually contain the same type of bacteria but one out of five patients has very different types of bacteria in the endometrium compared to the vagina. Therefore, vaginal cultures cannot be substituted for endometrial cultures. In the majority of the patient the bacteria have likely ascended into the endometrial cavity from the vaginal/cervical pool, but in a few patients bacterial types are present that are similar to bacteria normally seen in the oral cavity or the gut indicating that haematogenous spread of bacteria to the uterus may sometimes occur. In normal healthy patients the great majority of the bacteria present are of the Lactobacillicus type and the most common species are L.crispatus, L.gasseri, L.iners and L. jensenii. In fact the genus Lactobacillus contains >130 species of which 20 occur in the vagina. Different types and strains of Lactobacillius bacteria carry different functional properties such as lactic acid and hydrogen peroxide (H2O2) production, bacteriocidal effect against other bacteria, different adherence capacity to vaginal and uterine epithelium, competitive use of available nutrients and tolerance to tissue iron content that make one strain of Lactobacillus to have a survival advantage over another in different areas of the genital tract. The Lactobacillus bacteria also appears to function as a protective shield against other more harmful bacteria. A small proportion of healthy women carry bacteria associated with bacterial vaginosis without seeming to have any inflammation or noticeable disease conditions, an enigma that is still poorly understood!
Uterine infection is a known risk factor for infertility. In the 1990’s the growth of bacteria other than Lactobacillus from the IVF transfer catheter tip showed an association with implantation failure and an increased risk for miscarriage. Particularly bacteria such as Enterococcus, Streptococcus, Staphylococcus, E. coli, Klebsiella, and other Gram-negative bacteria seem to be negatively affecting the pregnancy rate. During these and other studies two different groups of patients were identified. One group with more than 90 % Lactobacillus present had a normal fertility rate and another group of patients with less than 90 % Lactobacillus and dysbiotic bacteria had a low pregnancy rate in IVF and also an increased risk for spontaneous abortions. More studies followed that demonstrated the poor pregnancy rate when endometrial pathogenic bacteria were demonstrated but also that an improved pregnancy rate was seen after selective antibiotic treatment. The impact of the endometrial microbiome on reproductive outcome showed poor success when abnormal bacteria or low numbers of Lactobacillus types were present in the endometrial tissue even if the endometrium appeared to be in a receptive phase for implantation.
Chronic endometritis has been recognized from histological exams of hysterectomy specimens and endometrial biopsies, but for many years no major illnesses were associated with the observation of inflammatory cells in the uterine endometrium and stroma. The frequent absence of symptoms and the need for an invasive procedure to get the diagnosis also made the condition less interesting for the average gynecologist. The cause of the chronic endometritis was also unclear as routine bacterial cultures were only positive in little more than half of the patients.
Over the past 10 years however the picture has changed. Multiple extensive studies have shown the clinical significance and involvement of chronic endometritis in infertility, embryo implantation failure in IVF, low IVF pregnancy rate, pregnancy complications and recurrent pregnancy loss.
As routine bacterial cultures are not very helpful in the diagnosis of chronic endometritis immunohistochemical staining of CD138 plasma cells in the uterine tissue has become the routine way to secure the diagnosis. The technical performance of the test has a number of challenges and pitfalls, but a recent quantification of the test has made the diagnosis more precise although no stringent criteria for the diagnosis of chronic endometritis has been established. More recent evaluations of the uterine microbiome with molecular techniques using DNA targeted sequencing and microarray studies of areas with very low presence of bacteria has shown that bacteria are essentially always present in the uterine cavity. The microbiota shows that mainly Lactobacillus type of bacteria are present (>90%) and only <10% other bacteria are detected in “normal” patients whereas in chronic endometritis patients more pathogenic bacteria are present.
In chronic endometritis large numbers of B cell lymphocytes invade both the endometrial basal layer and the functional layer (the layer that is lost at menstruation) and invade further into the glandular lumen. Together with the invading cells, inflammatory interleukins and other pro-inflammatory molecules are increased in the endometrial tissue and is also elevated in the menstrual blood from patients with endometritis. The plasma cells that invade the endometrium and stroma in chronic endometritis also secrete high levels of immunoglobulins in the endometrial tissue. This local immune response is likely to explain the abnormal implantation environment in the midluteal phase endometrial morphology, often referred to as “out of phase endometrium” that is often encountered in patients with chronic endometritis. Multiple antibiotic treatment protocols have been shown to clear CD 138 cells from the endometrial stroma of patients with chronic endometritis.
Endometrial bacteria have also been associated with endometriosis. Pathogenic bacteria have frequently been cultured from the endometrium in endometriosis patients and from menstrual fluids obtained from patients with endometriosis. Initial impact of pathogenic bacteria on the uterine epithelium may create an immune reaction to this bacterial invasion and this would then propagate the endometriotic response. Patients with and without endometriosis have different bacterial types and “endometriosis” bacteria may affect the contractility of the uterus and facilitate the retrograde menstrual flow that is characteristics of endometriosis.
A role for endometrial bacteria in the development of endometrial hyperplasia and endometrial adeno carcinoma has also been proposed. The presence of pathogenic bacteria in the uterus has also been associates with a persistent infection during pregnancy and the development of obstetric complications such as premature rupture of membranes, placental dysfunction, chorioamnionitis, pre-eclampsia, poor fetal growth and post-partum endometritis.
The reproductive tract is subject to many genetic variations and personal background changes and it is therefore important not only to identify specific bacteria but also to study the combined biological effect and function of the bacteriome with the host. All procedures performed must be interpreted within the limits of the technology used. Quantification of bacteria using DNA sequencing techniques is difficult and there is a small risk that these molecular based methods sometimes may identify genetic material from non-viable bacteria or bacteria that are present in so low numbers that they may be clinically insignificant.
The identification of dysbiosis as a new cause for human infertility is clear. The detection of the abnormal bacteria, identification of antibiotics susceptibility and finding the correct treatment for the restoration of the endometrial function will be a new way to further personalize the infertility treatment for our patients.