Expert Reviews ajog.org

Bacterial vaginosis and its association with , , and pelvic inflammatory disease Jacques Ravel, PhD; Inmaculada Moreno, PhD; Carlos Simo´n, MD, PhD

Introduction In healthy women of reproductive age, , pelvic inflammatory disease, and endometritis are infections of the typical vaginal microbiota include aero- genital tract that can lead to many adverse health outcomes, including infertility. Bacterial bic, facultative anaerobic, and obligate vaginosis is characterized by a lower prevalence of lactobacilli and a higher prevalence of anaerobic . Most women have anaerobic , including , Megasphaera spp., and Atopobium vaginal microbiota that are predomi- vaginae. Endometritis and pelvic inflammatory disease are caused by the ascension of nated by lactobacilli.1,2 Optimal vaginal to the , although the mechanisms by which they do so are microbiota tend to exist symbiotically unclear. Bacterial vaginosis, chronic endometritis, and pelvic inflammatory disease have and are believed to protect against been linked to infertility in retrospective and prospective trials. Similarly, the causes of pathogenic bacterial colonization and bacterial vaginosis and endometritis-related infertility are likely multifactorial and stem infection through the production of from inflammation, immune targeting of sperm antigens, the presence of bacterial toxins, lactic acid and antimicrobial byproducts and increased risk of sexually transmitted infections. Diagnosis and treatment of bacterial and by low-level immune system acti- vaginosis, chronic endometritis, and pelvic inflammatory disease before attempting vation.3 Disruption of the predominance conception may be important components of preconceptional care for symptomatic of lactobacilli has been shown to increase women to improve outcomes of natural and assisted reproduction. the risk of sexually transmitted in- Key words: fections (STIs) and upper genital tract bacterial vaginosis, endometritis, infertility, lactobacilli, pelvic inflammatory infections through the ascension of disease bacterial pathogens and other anaerobic bacteria.4 Bacterial vaginosis (BV) is a common acideproducing bacteria and increased Although BV is frequently asymp- lower genital tract infection that affects proportions of a wide array of strict and tomatic, women with BV are more e approximately 29% of women of repro- facultative anaerobes.1,6 8 Bacteria likely than those without BV to report ductive age in the United States, commonly associated with BV include vaginal odor, itching, and discharge.9 although variations in prevalence exist Gardnerella vaginalis, Megasphaera spp., Serious adverse health outcomes have among different races and ethnicities.5 Atopobium vaginae, Dialister spp., been associated with BV, including BV is associated with the disruption of spp., Sneathia amnii, Snea- increased risk of infertility; adverse optimal vaginal microbiota character- thia sanguinegens, Porphyromonas spp., outcomes; STIs, including ized by decreased proportions of lactic and spp.6,8 chlamydia, gonorrhea, human

From the Institute for Genome Sciences, University of Maryland School of Medicine, Baltimore, MD (Dr Ravel); Department of Microbiology and Immunology, University of Maryland School of Medicine, Baltimore, MD (Dr Ravel); Igenomix Foundation-Instituto de Investigación Sanitaria Hospital Clínico (INCLIVA), Valencia, Spain (Drs Moreno and Simón); Research and Development Department, Igenomix-Ferring Preconceptional InnoHub, Boston, MA (Drs Moreno and Simón); Department of Pediatrics, Obstetrics and Gynaecology, School of Medicine, University of Valencia, Valencia, Spain (Dr Simón); Department of Obstetrics and Gynecology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA (Dr Simón); Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX (Dr Simón). Received Aug. 12, 2020; revised Oct. 9, 2020; accepted Oct. 15, 2020. J.R. is a cofounder of LUCA Biologics, Inc, a biotechnology company focusing on translating microbiome research into live biotherapeutics drugs for women’s health. I.M. reports receiving personal fees as a part-time employee of Igenomix research and development outside the submitted work. C.S. reports receiving personal fees from Igenomix SL and invited lectures from Ferring Pharmaceuticals, Merck Serono, Merck Sharp & Dohme, Teva Pharmaceutical Industries Ltd, and Theramex outside the submitted work. The authors were responsible for all content and editorial decisions and received no honoraria related to the development of this article. All of the authors contributed to the research, writing, and reviewing of all drafts of this article and approved the final version. Corresponding author: Jacques Ravel, PhD. [email protected] 0002-9378 ª 2020 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.ajog.2020.10.019

MARCH 2021 American Journal of Obstetrics & Gynecology 251 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. Expert Reviews ajog.org papilloma virus (HPV), and human examination. High Nugent scores reflect A substantial percentage of women immunodeficiency virus (HIV); and the absence of and presence have asymptomatic BV (ie, pelvic inflammatory disease (PID), of the strict anaerobes Gardnerella and between 4 and 10 but no symptoms). e including endometritis.10 13 Mobiluncus spp. or BV-associated bac- Some patients may also have 3 of 4 PID and endometritis are upper gen- teria.24 In contrast, low Nugent scores symptoms but do not report symptoms ital tract infections with a range of clin- represent a high abundance of Lactoba- on direct questioning. Guidelines issued ical presentations and manifestations.4 cillus spp. and relative absence of anaer- by the Centers for Disease Control and Acute PID is caused by the ascension of obes. Molecular BV testing, such as Prevention do not recommend treat- strict or facultative anaerobes from the genetic sequencing or polymerase chain ment for these women, as there is a lack to the and adnexa reaction, may also be used in research of evidence that treatment for asymp- for 30 days. Chronic endometritis is an studies to complement Nugent tomatic BV decreases adverse outcomes, infection that lasts 30 days.4 Greater scoring.25 although recurrence and associated costs than 85% of PID cases are caused by BV- Patients with symptomatic BV are are high in asymptomatic women.31 related bacteria and/or STIs.4,14 Of those treated with either oral or intravaginal Advances in BV treatment include anti- cases, fewer than half are caused by Neis- . In the 2020 American biotics that require less frequent dosing, seria gonorrhoeae or Chlamydia trachoma- College of Obstetricians and Gynecol- such as secnidazole, and treatment that tis, suggesting an important role for ogists Practice Bulletin on in combines antimicrobials and Lactoba- ascension of BV-associated anaerobic bac- nonpregnant patients, recommended cillus crispatusecontaining probiotics to teria and other non-BVerelated pathogens treatments included oral metronida- address recurrence.32 In a recent double- (eg, Mycoplasma genitalium)inendome- zole, intravaginal gel, or blind, placebo-controlled trial to eval- e tritis and PID pathophysiology.15 17 PID intravaginal cream uate the ability of L crispatus to prevent and endometritis are associated with (Table 2).22 Single-dose oral secnida- recurrence, women aged 18 to 45 years adverse health outcomes, such as chronic zole was approved by the US Food and with a diagnosis of BV who had pain, ectopic pregnancy, tubo-ovarian ab- Drug Administration for the treatment completed a course of vaginal metroni- scess, and infertility.18,19 of BV in 201726 and was reported to dazole gel were randomly assigned to In this review article, we will describe provide a cure rate that was compara- receive vaginally administered L crispatus the current evidence for the associations ble with a 7-day oral metronidazole or placebo for 11 weeks, with follow-up among BV, PID, and endometritis. regimen in a research setting in which through week 24. The use of L crispatus Moreover, the impact of untreated BV patients were at least 80% adherent to after treatment with vaginal metronida- and PID on infertility will be reviewed. treatment.27 Because these treatments zole resulted in a significantly lower have comparable safety and efficacy incidence of BV recurrence vs placebo at Bacterial Vaginosis profiles, the choice of therapy should 12 weeks (30% vs 45%; P¼.01).33 Diagnosis and treatment of bacterial be individualized on the basis of fac- vaginosis tors, such as patient preferences, cost, Bacterial vaginosis and fertility Patients showing symptoms of BV typi- convenience, adherence, ease of use, BV has been linked to increased risk cally present with increased levels of and history of response to previous of infertility, particularly tubal e associated with a treatments or adverse reactions.22 Un- infertility.34 39 In a study of women strong fishy odor. When women present fortunately, although treatment efficacy undergoing oocyte recovery for in vitro to a healthcare provider with symptoms, is high at 3 to 4 weeks after treatment, fertilization (IVF), seropositivity for BV is usually diagnosed using the Amsel BV is highly recurrent, and 58% of Chlamydia species and the presence of criteria, which evaluate the presence of 4 women recur within a year.28 It has BV were both strongly and indepen- e signs and symptoms (Table 1).20 22 The been hypothesized that nonadherence dently associated with tubal infertility. presence of at least 3 of these signs and to multidose therapy could contribute However, there was no difference in symptoms must be met to fulfill the to the development of recurrent BV, pregnancy rates in any of the groups, diagnosis of BV.20,21 Although the Amsel although this association has not yet regardless of serologic status for chla- criteria are easy to assess and are asso- been tested in clinical trials.29 Addi- mydia or current BV.34 In a sample of ciated with good predictive values, tional putative reasons for recurrence patients seeking fertility treatment, commercial molecular tests have also include the persistence of residual Nugent-BV was present in 31.5% of been developed to detect BV, which may infection. For example, biofilms that patients with tubal infertility and 19.7% be useful in cases where microscopy is protect BV-associated bacteria from of patients with nontubal infertility.35 In not available.20 In research settings, BV is antimicrobial drugs foster persistence. a separate study, an intermediate Nugent diagnosed using the Nugent scoring Resistance to antimicrobial drugs and score was reported in 12.1% of women system,23 which uses a 0 to 10 score to reinfection from partners of either sex presenting for fertility treatment, and estimate the presence of vaginal bacterial may play a role. Nevertheless, the un- Nugent-BV was reported in 24.3%, with morphotypes that are characteristic of derlying mechanisms of recurrent eti- a higher prevalence among women with BVusing Gram staining and microscopic ology of BV are not fully understood.30 tubal infertility (34.6%).36 Furthermore,

252 American Journal of Obstetrics & Gynecology MARCH 2021 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. ajog.org Expert Reviews

TABLE 1 Amsel criteria for the diagnosis of BV 1. Homogenous, thin, grayish-white vaginal discharge that smoothly coats the vaginal walls 2. Presence of 20% clue cells on saline wet mount 3. Vaginal pH of >4.5 4. Positive whiff-amine test result BV, bacterial vaginosis. Ravel. Bacterial vaginosis and association with infertility, endometritis, and pelvic inflammatory disease. Am J Obstet Gynecol 2021.22 idiopathic infertility has been linked to a underwent IVF and who were positive determined; further research that in- unique vaginal bacterial signature that for BV, suggesting a shared pathogenesis cludes large-scale longitudinal and includes bacteria related to BV.37,38 (P¼.001).40 mechanistic studies are needed. Because of the heterogeneity seen in In the setting of IVF, BV has been Although the cause of infertility studies reporting BV prevalence in implicated in difficulty conceiving. among patients with BV is unclear, infertile populations, it may also be of Women with a lower prevalence of several mechanisms have been proposed. use to look at metaanalysis results. In a vaginal lactobacilli were less likely to One possibility is the association be- systematic review and metaanalysis of have successful embryo implantation tween BV microbiota and subsequent studies assessing BV and infertility, BV than those with a higher prevalence of inflammation, which may lead to was 3.3 times more likely to be identified lactobacilli.41 Furthermore, women with reduced fertility. BV-related bacteria in infertile women than in antenatal lower microbial diversity and those with have been shown to induce immune women within the same population.39 In a higher proportion of abnormal vaginal activation through dendritic cell matu- a systematic review and metaanalysis to microbiota were more likely to have poor ration and to increase levels of proin- evaluate the risks associated with BV in reproductive outcomes following IVF.42 flammatory cytokines, resulting in patients who underwent IVF, 16% BV Nevertheless, the metaanalysis of 12 mucosal inflammation of the genital prevalence was observed. However, the studies in the IVF setting found that BV tract.43,44 Higher levels of cervical prevalence ranged from 4% to 38%, did not significantly impact the live birth interleukin (IL)-1b, IL-6, and IL-8 cy- indicating a large heterogeneity in the rate (relative risk [RR], 1.47; 95% con- tokines have been reported in women studies examined, which may be fidence interval [CI], 0.96e1.57) or the with infertility and BV.45 Restoration of explained by different diagnostic clinical pregnancy rate (RR, 0.93; 95% normal vaginal microbiota with use of a methods, ethnicities, and types of infer- CI, 0.75e1.15).40 Although there is a probiotic vaginal tablet containing tility. was signifi- clear association between BV and infer- lactobacilli has been shown to reduce cantly prevalent among patients who tility, causality has not been conclusively levels of proinflammatory cytokines,

TABLE 2 ACOG treatment recommendations for the management of BV in nonpregnant patients Drug Formulation Dosage Duration Recommended treatment regimens Metronidazole Oral 500 mg, twice daily 7 d Metronidazole Intravaginal gel 0.75% 5 g, once daily 5 d Clindamycin Intravaginal cream 2% 5 g, once daily at bedtime 7 d Alternative treatment regimens Secnidazole Oral 2 g, single dose 1 d Oral 2 g, once daily 2 d Tinidazole Oral 1 g, once daily 5 d Clindamycin Oral 300 mg, twice daily 7 d Clindamycin Intravaginal ovules 100 mg, once daily at bedtime 3 d ACOG, American College of Obstetricians and Gynecologists; BV, bacterial vaginosis. Ravel. Bacterial vaginosis associated with infertility, endometritis, and pelvic inflammatory disease. Am J Obstet Gynecol 2021.22

MARCH 2021 American Journal of Obstetrics & Gynecology 253 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. Expert Reviews ajog.org supporting the hypothesis that BV- acquisition.49 D-lactic acid, which is with histopathologic evidence of endo- associated bacteria may increase produced by L crispatus, L gasseri, and L metritis or laparoscopic findings inflammation.46 The species of Lactoba- jensenii—but not L iners—was shown to consistent with PID can also support the cillus used in the vaginal tablet were L prevent infection by C trachomatis diagnosis with high specificity.20 brevis, L salivarius subsp. salicinius, and L in vitro by directly affecting the function In contrast with PID and acute plantarum, not the more typical L crisp- of the cervicovaginal .51 This endometritis, subclinical or chronic atus, L gasseri, and L jensenii seen in an is important because STIs—and C tra- endometritis (CE) persists for a longer optimal vaginal environment. More chomatis and N gonorrhoeae in partic- period and is either asymptomatic or research introducing these typical ular—have been linked to an increased associated with more subtle or nonde- vaginal lactobacilli species into a vaginal risk of upper genital tract infection, PID, script symptoms such as pelvic discom- tablet is warranted.46 The presence of and infertility.52,53 fort, spotting, and . These immune activation and inflammation at Finally, BV increases the risk of upper symptoms are more difficult to diagnose, the vaginal mucosa may lead to immune genital tract infection and PID, which and there are no universally accepted targeting of seminal fluid components, have been linked to infertility. BV-related diagnostic criteria for CE.55 Studies have which are highly antigenic. Seminal vaginal microbial signatures have shown that the presence of endometrial components binding to the acrosomal been associated with increased risk of stromal plasmacytes is a specific and sperm to protect it are carried into the PID, whereas Lactobacillus-dominated sensitive finding for CE.55,56 Common upper female genital tract. Various microbiota did not increase the risk.54 pathogens that have been detected in CE pathologic processes may occur at this Women with acute endometritis were include Staphylococcus spp., Strepto- juncture.47 Ongoing research and anal- 90% less likely to have typical ratios of coccus spp., Escherichia coli, Enterococcus ysis of these processes are warranted. lactobacilli and were 2.4-fold more likely faecalis, C trachomatis, Mycoplasma spp., Another BV-related mechanism that to have Nugent-BV.16 BV has also been Klebsiella pneumoniae, and Candida may contribute to infertility is the effect associated with subclinical PID, which is spp.57,58 of sialidase and other mucinases on marked by asymptomatic ascension of cervical mucus integrity. In the female infectious agents to the upper genital Endometritis, pelvic inflammatory reproductive tract, a primary function of tract and is also associated with chla- disease, and fertility cervical mucus is the defense of the up- mydia or gonorrhea.4,13 Subclinical PID PID and endometritis have been associ- per reproductive tract from microbial is 2.7-fold more common in women ated with infertility in past studies. The invasion. To overcome the mucus bar- with Nugent-BV.13 The presence of BV- presence of BV-associated bacteria in the rier, microorganisms may produce a associated bacteria in the endometrium endometrium has been linked to a 3.4- range of hydrolyzing enzymes, including has also been associated with recurrent fold increased risk of infertility.17 In a mucinases, that are capable of degrading PID and persistent endometritis after study of women with Nugent-BV, mucins. These enzymes may also work to recommended treatment with cefoxitin gonorrhea, or chlamydia or at risk of enhance bacterial and subse- and doxycycline.17 infections, such as gonorrhea or chla- quent colonization in the upper repro- These data provide support for the mydia, researchers prospectively evalu- ductive tract by generating attachment role of BV in infertility through a variety ated pregnancy outcomes after a biopsy sites on the mucosal surfaces and pro- of mechanisms, including immune was performed to identify endometritis. ducing nutrition for bacteria from the activation, inflammation, toxin produc- Participants were treated for BV and mucin breakdown products,48 fostering tion, STI susceptibility, and PID other infections. After a median of 2.1 colonization with further propensity for susceptibility. years of follow-up, women with sub- upper reproductive tract disease, clinical PID at diagnosis had a 40% including infertility. Endometritis and Pelvic decreased likelihood of pregnancy Women with BV are at increased risk Inflammatory Disease compared with those without subclinical for acquiring STIs, which are known to Diagnosis and treatment of PID. A study limitation is that the contribute to infertility. BV has been endometritis and pelvic inflammatory women enrolled were not known to be shown to increase susceptibility to C disease specifically trying to get pregnant nor trachomatis and N gonorrhoeae by 3.4- PID presents with a variety of signs and was fertility intent queried during and 4.1-fold, respectively.49 Other inci- symptoms that are frequently nonspe- follow-up, which could confound the dent infections linked to BV include cific and include cervical motion assessment of an association between BV vaginalis, herpes simplex tenderness, uterine tenderness, or and infertility because of endometritis.59 virus, HPV, and HIV.10,12,50 Vaginal adnexal tenderness on pelvic examina- Similar results, however, were found in a colonization with lactobacilli has been tion. Symptoms of acute endometritis large population-based study of women shown to be protective from chlamydial are similar to those of PID, and outside who underwent diagnostic laparoscopy or gonorrheal infections, suggesting a of pregnancy, providers often use the for PID. Tubal infertility was found in role for optimal Lactobacillus-dominated terms endometritis and PID inter- 10.8% of patients diagnosed with PID vaginal microbiota in preventing STI changeably. Positive endometrial biopsy compared with 0% of those who tested

254 American Journal of Obstetrics & Gynecology MARCH 2021 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. ajog.org Expert Reviews negative.18 In addition, in the US Na- expression levels of chemokines and true for women suffering from infertility tional Health and Nutrition Examina- adhesion molecules in patients with CE, or tubal infertility. tion Survey (2013e2016), 2626 women which may impact trophoblast implan- Previous studies have shown that of reproductive age self-reported infer- tation.65 Taken together, it is possible treatment of genital infections may tility and treatment for PID. Infertility that these immunologic and inflamma- improve fertility outcomes. Successful was reported by 24.2% of women with tory changes disrupt endometrial func- treatment of women with CE resulted in past PID treatment compared with tion and decrease receptiveness of the a significantly higher pregnancy rate 13.3% of women without PID endometrium to embryo implantation than those with persistent disease and treatment.60 and development.66 those without endometritis diagnosis Further support is provided for the PID is most frequently linked to tubal (76.3% vs 20% vs 9.5%; P<.0001).61 For association between endometritis and infertility, which may be explained by women undergoing IVF, those with PID and infertility based on data from pathologic tubal inflammation, fibrosis, cured CE had a 6.8-fold higher ongoing women treated with assisted reproduc- and subsequent scarring.52 This has been pregnancy and live birth rate and a 4.0- tion. CE has been shown to be highly most frequently studied in the context of fold higher clinical pregnancy rate than prevalent among patients with unex- C trachomatis infectioneassociated PID, those with persistent disease.71 In addi- plained infertility (34%e66%).61 In which appears to lead to an innate im- tion to CE treatment, some fertility women undergoing IVF, those with mune response mediated by infected specialists have suggested that coloniza- endometrial microbiota dominated by epithelial cells and an adaptive T-cell tion of the catheter transfer tip with nonlactobacilli were significantly less response.67,68 In a macaque model, beneficial lactobacilli at the time of em- likely to have successful implantation, recurrent C trachomatis infections led to bryo transfer may improve implantation pregnancy, or ongoing pregnancy than mononuclear infiltration (primarily rates.72 Although BV treatment and its those with microbiota dominated by CD8 T cells), fibroblast proliferation, association with successful reproductive lactobacilli (>90%) (P<.05).62 However, and connective tissue deposition, outcomes through natural or assisted it is important to note that although culminating in fibrosis of the fallopian means have not been evaluated, the lower prevalence of lactobacilli have tubes.67 Evidence has also demonstrated benefits of preventing progression to been associated with BV and endome- the effect of N gonorrhoeae as a pathogen PID and CE are likely to positively affect tritis, there is no standard definition of involved in reproductive tract morbid- the chances of natural and assisted abnormal and normal endometrial ities, including tubal factor infertility conception. Clinical trials assessing the microbiota, and the abundance of these and PID. Moreover, limited evidence impact of BV treatment on successful bacteria in the endometrium is un- suggests that other organisms, such as M conception may be difficult to perform known.63 Nonetheless, CE cure with genitalium and T vaginalis, and varia- because of the prevalence of recurrent antimicrobial treatment has been shown tions in the overall vaginal microbiome, BV and challenges associated with to improve outcomes in women under- such as those that occur in BV, may treating this condition. going IVF.61 For example, compared contribute to conditions that interfere Treatment is recommended for with women with cured CE, those with with female fertility.52 symptomatic women. The benefits of persistent CE had a lower pregnancy rate therapy in nonpregnant women are to (33.0% vs 65.2%; P¼.039) and lower live Managing Bacterial Vaginosis, Pelvic relieve symptoms and signs of infection birth rate (60.8% vs 13.3%; P¼.02) after Inflammatory Disease, and and reduce the risk of acquiring C tra- IVF.61 Endometritis Before Pregnancy chomatis, N gonorrhoeae, T vaginalis, Although the precise etiology of Diagnosis of BV, PID, and endometritis HIV, and herpes simplex type 2.20 endometritis and PID-associated infer- may be complicated, because symptoms Treatment of PID with metronidazole tility is unclear, several pathophysiolog- can vary and be mild, nonspecific, or and ceftriaxone and doxycycline was ical contributors have been proposed. As absent. There can be difficulty in iden- shown to eradicate anaerobic bacteria with BV, endometritis has been associ- tifying pathogens in the endometrium from the endometrium and decrease ated with disrupted inflammatory and by means of microbial culture, with only pelvic tenderness at 30 days.73 Amore immunologic signatures. The endome- a 20% concordance among histology, recent trial by this group showed trium contains many immunocompe- hysteroscopy, and microbial culture re- further evidence of the benefitof tent cells that contribute to regulation of sults.69 There is also the potential for metronidazole with ceftriaxone and inflammation, immune response, and contamination of endometrial samples doxycycline for treating PID; it also trophoblast implantation and growth.64 with vaginal bacteria.63 Therefore, a low reduced the detection of M genitalium Many of these immune cell populations level of clinical suspicion for BV, CE, or at 30 days.74 For symptomatic patients, have been shown to be altered among PID should be sufficient for initiating testing for BV or endometritis is rela- women with endometritis, which can testing in women with risk factors for tively cost-effective and can be per- lead to uterine immune cell infiltration. these infections, such as a history of STIs formed during a routine preconception The altered immune environment has or sexual behaviors that could lead to pelvic examination, which may already been shown to lead to abnormal uterine transmission.70 This may be particularly be a component of the preconceptional

MARCH 2021 American Journal of Obstetrics & Gynecology 255 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. Expert Reviews ajog.org appointment. On laboratory confirma- Nutrition Examination Survey data. Obstet transmitted diseases treatment guidelines, tion of BV and CE, treatment with an- Gynecol 2007;109:114–20. 2015. MMWR Recomm Rep 2015;64:1–137. 6. Fredricks DN, Fiedler TL, Marrazzo JM. Mo- 21. Schwebe JR, Hillier SL, Sobel JD, tibiotics is warranted and may improve lecular identification of bacteria associated with McGregor JA, Sweet RL. Validity of the vaginal fertility outcomes. For asymptomatic bacterial vaginosis. N Engl J Med 2005;353: for the diagnosis of bacterial vagi- patients with suspicion of CE because of 1899–911. nosis. Obstet Gynecol 1996;88:573–6. previous subclinical infections and with 7. Jung HS, Ehlers MM, Lombaard H, 22. Committee on Practice Bulletins—Gyne- an unknown cause of infertility, Redelinghuys MJ, Kock MM. Etiology of bacte- cology. Vaginitis in nonpregnant patients: ACOG rial vaginosis and polymicrobial biofilm forma- Practice Bulletin, number 215. Obstet Gynecol repeated implantation failures, repeated tion. Crit Rev Microbiol 2017;43:651–67. 2020;135:e1–17. miscarriages, or previous 8. Muzny CA, Blanchard E, Taylor CM, et al. 23. Nugent RP, Krohn MA, Hillier SL. Reliability with intrauterine infections, such as Identification of key bacteria involved in the in- of diagnosing bacterial vaginosis is improved by or deciduitis, duction of incident bacterial vaginosis: a pro- a standardized method of Gram stain interpre- screening for CE should also be spective study. J Infect Dis 2018;218:966–78. tation. J Clin Microbiol 1991;29:297–301. 9. Klebanoff MA, Schwebke JR, Zhang J, 24. Srinivasan S, Morgan MT, Liu C, et al. More considered. Nansel TR, Yu KF, Andrews WW. Vulvovaginal than meets the eye: associations of vaginal bac- symptoms in women with bacterial vaginosis. teria with gram stain morphotypes using molecular Conclusions Obstet Gynecol 2004;104:267–72. phylogenetic analysis. PLoS One 2013;8:e78633. BV, endometritis, PID, and infertility are 10. Atashili J, Poole C, Ndumbe PM, 25. Coleman JS, Gaydos CA. Molecular diag- related to interconnected pathophysio- Adimora AA, Smith JS. Bacterial vaginosis and nosis of bacterial vaginosis: an update. J Clin fl HIV acquisition: a meta-analysis of published Microbiol 2018;56:e00342-18. logical pathways. Immunity, in amma- studies. AIDS 2008;22:1493–501. 26. National Institutes of Health. US National Li- tion, cervicovaginal microbiota, and 11. Leitich H, Bodner-Adler B, Brunbauer M, brary of Medicine. Solosec (secnidazole) oral fibrotic pathways all play a role in Kaider A, Egarter C, Husslein P. Bacterial vagi- granules prescribing information. Solosec. 2019. contributing to infertility; however, nosis as a risk factor for preterm delivery: a meta- Available at: https://dailymed.nlm.nih.gov/ additional large, prospective, longitudi- analysis. Am J Obstet Gynecol 2003;189: dailymed/drugInfo.cfm?setid¼551e43d5-f700- 139–47. 4d6e-8029-026f8a8932ff.AccessedAugust2, nal studies are needed to conclusively 12. Bautista CT, Wurapa EK, Sateren WB, 2020. determine the link among BV, PID, Morris SM, Hollingsworth BP, Sanchez JL. As- 27. Bohbot JM, Vicaut E, Fagnen D, endometritis, and infertility. Until then, sociation of bacterial vaginosis with chlamydia Brauman M. Treatment of bacterial vaginosis: a symptomatic BV (and possibly asymp- and gonorrhea among women in the U.S. Army. multicenter, double-blind, double-dummy, tomatic BV) should be urgently treated Am J Prev Med 2017;52:632–9. randomised phase III study comparing secni- 13. Wiesenfeld HC, Hillier SL, Krohn MA, et al. dazole and metronidazole. Infect Dis Obstet to prevent BV sequelae, including STIs, Lower genital tract infection and endometritis: Gynecol 2010;2010:705692. PID, and endometritis, which all seem to insight into subclinical pelvic inflammatory dis- 28. Bradshaw CS, Morton AN, Hocking J, et al. contribute to an increased risk of ease. Obstet Gynecol 2002;100:456–63. High recurrence rates of bacterial vaginosis over infertility. - 14. Simms I, Eastick K, Mallinson H, et al. As- the course of 12 months after oral metronidazole sociations between Mycoplasma genitalium, therapy and factors associated with recurrence. Chlamydia trachomatis and pelvic inflammatory J Infect Dis 2006;193:1478–86. ACKNOWLEDGMENTS disease. J Clin Pathol 2003;56:616–8. 29. Muzny CA, Kardas P. A narrative review of The authors wish to acknowledge the contribu- 15. Hillier SL, Kiviat NB, Hawes SE, et al. Role of current challenges in the diagnosis and man- tions of Shawana Taylor, MD, of Lupin Phar- bacterial vaginosis-associated microorganisms agement of bacterial vaginosis. Sex Transm Dis maceuticals, Inc, who provided invaluable in endometritis. Am J Obstet Gynecol 1996;175: 2020;47:441–6. insights in the research and development of this 435–41. 30. Faught BM, Reyes S. Characterization and article. Editorial support in the preparation of this 16. Haggerty CL, Hillier SL, Bass DC, Ness RB; treatment of recurrent bacterial vaginosis. article was provided by Phase Five Communi- PID Evaluation and Clinical Health Study In- J Womens Health (Larchmt) 2019;28:1218–26. cations, funded by Symbiomix Therapeutics, vestigators. Bacterial vaginosis and anaerobic 31. Peebles K, Velloza J, Balkus JE, LLC, a Lupin Pharmaceuticals, Inc, company. bacteria are associated with endometritis. Clin McClelland RS, Barnabas RV. High global Infect Dis 2004;39:990–5. burden and costs of bacterial vaginosis: a sys- 17. Haggerty CL, Totten PA, Tang G, et al. tematic review and meta-analysis. Sex Transm REFERENCES Identification of novel microbes associated with Dis 2019;46:304–11. 1. Ravel J, Gajer P, Abdo Z, et al. Vaginal pelvic inflammatory disease and infertility. Sex 32. Bradshaw CS, Brotman RM. Making in- microbiome of reproductive-age women. Proc Transm Infect 2016;92:441–6. roads into improving treatment of bacterial Natl Acad Sci U S A 2011;108(Suppl1):4680–7. 18. Weström L, Joesoef R, Reynolds G, vaginosis - striving for long-term cure. BMC 2. Zhou X, Brown CJ, Abdo Z, et al. Differences Hagdu A, Thompson SE. Pelvic inflammatory Infect Dis 2015;15:292. in the composition of vaginal microbial commu- disease and fertility. A cohort study of 1,844 33. Cohen CR, Wierzbicki MR, French AL, et al. nities found in healthy Caucasian and black women with laparoscopically verified disease Randomized trial of lactin-V to prevent recur- women. ISME J 2007;1:121–33. and 657 control women with normal laparo- rence of bacterial vaginosis. N Engl J Med 3. Witkin SS, Linhares IM, Giraldo P. Bacterial scopic results. Sex Transm Dis 1992;19: 2020;382:1906–15. flora of the female genital tract: function and 185–92. 34. Gaudoin M, Rekha P, Morris A, Lynch J, immune regulation. Best Pract Res Clin Obstet 19. Haggerty CL, Peipert JF, Weitzen S, et al. Acharya U. Bacterial vaginosis and past chla- Gynaecol 2007;21:347–54. Predictors of chronic in an urban mydial infection are strongly and independently 4. Brunham RC, Gottlieb SL, Paavonen J. Pelvic population of women with symptoms and signs associated with tubal infertility but do not affect inflammatory disease. N Engl J Med 2015;372: of pelvic inflammatory disease. Sex Transm Dis in vitro fertilization success rates. Fertil Steril 2039–48. 2005;32:293–9. 1999;72:730–2. 5. Allsworth JE, Peipert JF. Prevalence of bac- 20. Workowski KA, Bolan GA; Centers for Dis- 35. Liversedge NH, Turner A, Horner PJ, terial vaginosis: 2001e2004 National Health and ease Control and Prevention. Sexually Keay SD, Jenkins JM, Hull MG. The influence of

256 American Journal of Obstetrics & Gynecology MARCH 2021 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados. ajog.org Expert Reviews

bacterial vaginosis on in-vitro fertilization and 47. Brazdova A, Senechal H, Peltre G, unexplained implantation failure and the IVF embryo implantation during assisted repro- Poncet P. Immune aspects of . success rate after therapy. Hum duction treatment. Hum Reprod 1999;14: Int J Fertil Steril 2016;10:1–10. Reprod 2015;30:323–30. 2411–5. 48. Wiggins R, Hicks SJ, Soothill PW, Millar MR, 62. Moreno I, Codoñer FM, Vilella F, et al. Evi- 36. Wilson JD, Ralph SG, Rutherford AJ. Rates Corfield AP. Mucinases and sialidases: their role dence that the endometrial microbiota has an of bacterial vaginosis in women undergoing in the pathogenesis of sexually transmitted in- effect on implantation success or failure. Am J in vitro fertilisation for different types of infertility. fections in the female genital tract. Sex Transm Obstet Gynecol 2016;215:684–703. BJOG 2002;109:714–7. Infect 2001;77:402–8. 63. Winters AD, Romero R, Gervasi MT, et al. 37. Campisciano G, Florian F, D’Eustacchio A, 49. Wiesenfeld HC, Hillier SL, Krohn MA, Does the endometrial cavity have a molecular et al. Subclinical alteration of the cervical-vaginal Landers DV, Sweet RL. Bacterial vaginosis is a microbial signature? Sci Rep 2019;9:9905. microbiome in women with idiopathic infertility. strong predictor of Neisseria gonorrhoeae and 64. Du MR, Wang SC, Li DJ. The integrative J Cell Physiol 2017;232:1681–8. Chlamydia trachomatis infection. Clin Infect Dis roles of chemokines at the maternal-fetal inter- 38. Wee BA, Thomas M, Sweeney EL, et al. 2003;36:663–8. face in early pregnancy. Cell Mol Immunol A retrospective pilot study to determine whether 50. Cherpes TL, Meyn LA, Krohn MA, Lurie JG, 2014;11:438–48. the reproductive tract microbiota differs be- Hillier SL. Association between acquisition of 65. Kitaya K, Yasuo T. Aberrant expression of tween women with a history of infertility and herpes simplex virus type 2 in women and selectin E, CXCL1, and CXCL13 in chronic fertile women. Aust N Z J Obstet Gynaecol bacterial vaginosis. Clin Infect Dis 2003;37: endometritis. Mod Pathol 2010;23:1136–46. 2018;58:341–8. 319–25. 66. Park HJ, Kim YS, Yoon TK, Lee WS. Chronic 39. van Oostrum N, De Sutter P, Meys J, 51. Edwards VL, Smith SB, McComb EJ, et al. endometritis and infertility. Clin Exp Reprod Med Verstraelen H. Risks associated with bacterial The cervicovaginal microbiota-host interaction 2016;43:185–92. vaginosis in infertility patients: a systematic re- modulates Chlamydia trachomatis infection. 67. Van Voorhis WC, Barrett LK, Sweeney YT, view and meta-analysis. Hum Reprod 2013;28: mBio 2019;10:e01548-19. Kuo CC, Patton DL. Repeated Chlamydia tra- 1809–15. 52. Tsevat DG, Wiesenfeld HC, Parks C, chomatis infection of Macaca nemestrina fallo- 40. Haahr T, Zacho J, Bräuner M, Peipert JF. Sexually transmitted diseases and pian tubes produces a Th1-like cytokine Shathmigha K, Skov Jensen J, Humaidan P. infertility. Am J Obstet Gynecol 2017;216:1–9. response associated with fibrosis and scarring. Reproductive outcome of patients undergoing 53. Wiringa AE, Ness RB, Darville T, Beigi RH, Infect Immun 1997;65:2175–82. in vitro fertilisation treatment and diagnosed with Haggerty CL. , endome- 68. Mukura LR, Hickey DK, Rodriguez- bacterial vaginosis or abnormal vaginal micro- tritis and sequelae among women with clinically Garcia M, Fahey JV, Wira CR. Chlamydia tra- biota: a systematic PRISMA review and meta- suspected pelvic inflammatory disease. Sex chomatis regulates innate immune barrier analysis. BJOG 2019;126:200–7. Transm Infect 2020;96:436–8. integrity and mediates cytokine and antimicro- 41. Koedooder R, Singer M, Schoenmakers S, 54. Ness RB, Kip KE, Hillier SL, et al. A cluster bial responses in human uterine ECC-1 epithelial et al. The vaginal microbiome as a predictor for analysis of bacterial vaginosis-associated cells. Am J Reprod Immunol 2017;78. outcome of in vitro fertilization with or without microflora and pelvic inflammatory disease. Am 69. Moreno I, Cicinelli E, Garcia-Grau I, et al. The intracytoplasmic sperm injection: a prospective J Epidemiol 2005;162:585–90. diagnosis of chronic endometritis in infertile study. Hum Reprod 2019;34:1042–54. 55. Kitaya K, Takeuchi T, Mizuta S, asymptomatic women: a comparative study of 42. Haahr T, Humaidan P, Elbaek HO, et al. Matsubayashi H, Ishikawa T. Endometritis: new histology, microbial cultures, hysteroscopy, and Vaginal microbiota and in vitro fertilization time, new concepts. Fertil Steril 2018;110: molecular microbiology. Am J Obstet Gynecol outcomes: development of a simple diag- 344–50. 2018;218:602.e1–16. nostic tool to predict patients at risk of a 56. Kitaya K, Matsubayashi H, Yamaguchi K, 70. Gallo MF, Macaluso M, Warner L, et al. poor reproductive outcome. J Infect Dis et al. Chronic endometritis: potential cause of Bacterial vaginosis, gonorrhea, and chlamydial 2019;219:1809–17. infertility and obstetric and neonatal complica- infection among women attending a sexually 43. van Teijlingen NH, Helgers LC, Zijlstra- tions. Am J Reprod Immunol 2016;75:13–22. transmitted disease clinic: a longitudinal analysis Willems EM, et al. Vaginal dysbiosis associated- 57. Cicinelli E, De Ziegler D, Nicoletti R, et al. of possible causal links. Ann Epidemiol 2012;22: bacteria Megasphaera elsdenii and Prevotella Chronic endometritis: correlation among hyster- 213–20. timonensis induce immune activation via den- oscopic, histologic, and bacteriologic findings in 71. Vitagliano A, Saccardi C, Noventa M, et al. dritic cells. J Reprod Immunol 2020;138: a prospective trial with 2190 consecutive office Effects of chronic endometritis therapy on in vitro 103085. hysteroscopies. Fertil Steril 2008;89:677–84. fertilization outcome in women with repeated 44. Lennard K, Dabee S, Barnabas SL, et al. 58. Cicinelli E, De Ziegler D, Nicoletti R, et al. implantation failure: a systematic review and Microbial composition predicts genital tract Poor reliability of vaginal and endocervical cul- meta-analysis. Fertil Steril 2018;110:103–12.e1. inflammation and persistent bacterial vaginosis tures for evaluating microbiology of endometrial 72. Sirota I, Zarek SM, Segars JH. Potential in- in South African adolescent females. Infect cavity in women with chronic endometritis. fluence of the microbiome on infertility and Immun 2017;86:e00410–7. Gynecol Obstet Investig 2009;68:108–15. assisted reproductive technology. Semin 45. Spandorfer SD, Neuer A, Giraldo PC, 59. Wiesenfeld HC, Hillier SL, Meyn LA, Reprod Med 2014;32:35–42. Rosenwaks Z, Witkin SS. Relationship of Amortegui AJ, Sweet RL. Subclinical pelvic in- 73. Wiesenfeld HC, Hillier SL, Meyn L, et al. abnormal vaginal flora, proinflammatory cyto- flammatory disease and infertility. Obstet Gyne- Impact of metronidazole on clearance of anaer- kines and idiopathic infertility in women under- col 2012;120:37–43. obes in women with acute pelvic inflammatory going IVF. J Reprod Med 2001;46:806–10. 60. Anyalechi GE, Hong J, Kreisel K, et al. Self- disease: the ACE trial. Am J Obstet Gynecol 46. Hemalatha R, Mastromarino P, reported infertility and associated pelvic inflam- 2017;217:714. Ramalaxmi BA, Balakrishna NV, Sesikeran B. matory disease among women of reproductive 74. Wiesenfeld HC, Meyn LA, Darville T, Effectiveness of vaginal tablets containing lac- age-National Health and Nutrition Examination Macio IS, Hillier SL. A randomized controlled trial tobacilli versus pH tablets on vaginal health and Survey, United States, 2013-2016. Sex Transm of ceftriaxone and doxycycline, with or without inflammatory cytokines: a randomized, double- Dis 2019;46:446–51. metronidazole, for the treatment of acute pelvic blind study. Eur J Clin Microbiol Infect Dis 61. Cicinelli E, Matteo M, Tinelli R, et al. Preva- inflammatory disease. Clin Infect Dis 2020 [Epub 2012;31:3097–105. lence of chronic endometritis in repeated ahead of print].

MARCH 2021 American Journal of Obstetrics & Gynecology 257 Descargado para Anonymous User (n/a) en National Library of Health and Social Security de ClinicalKey.es por Elsevier en junio 04, 2021. Para uso personal exclusivamente. No se permiten otros usos sin autorización. Copyright ©2021. Elsevier Inc. Todos los derechos reservados.