The microbiome and endometriosis

The objective of this study was to systematically review the literature on the human microbiome in association with endometriosis. PubMed/Medline, Cochrane, and Embase databases were searched for literature published from 1986 to August 2021. All human studies that assessed the microbiome using 16S rRNA sequencing or shotgun sequencing in women with endometriosis were included. Two reviewers independently abstracted data from the selected articles into tables. To assess the quality of included studies, the National Institutes of Health Study Quality Assessment Tools were utilized. This review included 12 case–control studies. Included studies compared the microbiome from various anatomical sources (fecal, vaginal, cervical, peritoneal, endometrial, and intra-lesional) between patients with endometriosis and a heterogeneous set of control patients. Study quality ranged from poor to good, with 8 of 12 studies rated fair. Multiple studies reported a different distribution of bacteria among women with endometriosis across anatomical sites, but the results were highly heterogeneous. Pseudomonas was overrepresented in peritoneal fluid among women with endometriosis across multiple studies but was also observed to be increased in vaginal, endometrial, and intra-lesional samples. Among bacteria noted across different anatomical samples, Gardnerella was found to be increased in cervical but decreased in endometrial, fecal, and vaginal samples of patients with endometriosis, while Atopium was found to be decreased in vaginal and cervical samples from patients with endometriosis. Sphingobium was found to be increased in vagina, endometrium, and peritoneal fluid from patients with endometriosis. Streptococcus was found to be increased in peritoneal, endometrial, and cervical samples from women with endometriosis. Microbiomal comparisons stratified by endometriosis stage or site of endometriosis involvement were limited and highly heterogeneous. Lay summary The microbiome, a group of bacteria found in a particular place in the body, has been shown to vary when patients have some diseases, such as cancer or inflammatory bowel disease. Less is known about the microbiome in patients with endometriosis. This review looked at existing studies comparing the bacteria found in patients with endometriosis and others without. Twelve studies were found that assessed the bacteria from swabs collected from different places, including the vagina, cervix, endometrium, peritoneum, feces, and endometriosis lesions themselves. Most of the studies found higher or lower levels of specific bacteria at each of these places, but the findings were often inconsistent. The findings were probably limited by the small numbers of patients involved and variations in the groups studied. More research is needed to find out which bacteria are over- and underrepresented in patients with endometriosis and where they are found.


Introduction
Endometriosis is a multi-factorial disease defined by the presence of endometrial stroma or glands outside the uterine cavity. Patients with endometriosis, representing approximately 10-15% of reproductive-aged women, commonly experience dysmenorrhea, dyspareunia, and chronic pelvic pain, although there is a wide range in symptom prevalence as well as disease severity (Dunselman et al. 2014).
The most accepted theory on endometriosis pathogenesis is retrograde menstruation, in which reflux of menstrual blood through the fallopian tubes during menstrual cycles associated with an abnormal peritoneal environment permits the implantation and growth of ectopic endometrial tissue (Burney & Giudice 2019).
The advent of genomics technologies has greatly facilitated the characterization of the bacterial environment from clinical specimens with granular specieslevel detail. Previous studies have demonstrated that the microbiome may affect the development and progression of various diseases associated with an abnormal immune/ inflammatory response, including inflammatory bowel diseases , autoimmune diseases (Tsai et al. 2021), and cancer (Lim et al. 2021, Pothuraju et al. 2021. It is unknown whether an altered microbiome at any anatomical site can cause the development or progression of endometriosis. Similarly, it is not known whether endometriosis can directly induce an altered microbiome. Khan et al. (Khan et al. 2018) proposed a 'bacterial contamination hypothesis' for endometriosis, whereby bacterial endotoxins activate a peritoneal proinflammatory response, increase cell-to-cell adhesion, and facilitate the growth of ectopic endometrial implants.
Endometriosis most commonly occurs at sites such as the peritoneal cavity that are traditionally assumed to be sterile. However, microbiomal studies have also investigated swabs collected from sites known to have significant bacterial colonization, such as the vagina or rectum (Chen et al. 2017. Previous reviews (Leonardi et al. 2020, D'Alterio et al. 2021) have sought to investigate the association between endometriosis and the microbiome from different locations. For this systematic review, we considered microbiome analyses of swabs collected from all potential anatomical sites, regardless of whether the site was locally affected by endometriosis, and also sought to comprehensively collect data on endometriosis stage, menstrual phase, hormonal intake, and endometriosis symptoms.

Objective
The primary objective was to systematically review the association between endometriosis and an altered microbiome across various anatomical sites. Secondary objectives were to evaluate the association between the endometriosis stage or pain symptoms and the microbiome.

Search strategy
Briefly, a literature search was performed on PubMed/ Medline, Cochrane, and Embase databases from 1986 to August 2, 2021, using a combination of the following keywords: (microbiome OR microbial OR microbiota) AND (endometriosis OR endometrioma), and only articles published in English were considered. The full search strategy, including a dictionary of synonyms for the above keywords, is described in Supplementary Appendix 1 (see section on supplementary materials given at the end of this article).

Selection criteria
All studies utilizing human subjects that assessed a bacterial microbiome in association with patients with endometriosis were included. Only case-control studies using semi-quantitative methodologies (such as 16s rRNA amplification or shotgun sequencing) capable of quantifying the relative bacterial prevalence between groups were included. Case reports, reviews, conference abstracts, animal studies, and unpublished studies were excluded from this review.

Study selection
Two reviewers (FRO and CHM) independently screened the studies. Conflicts regarding study inclusion were resolved after a discussion between the two reviewers with a third author (MPA) and a senior author (MSA). Reviewers were not blinded to author names, institutional affiliations, or journal identities.

Data abstraction
Two reviewers (FRO and CHM) independently abstracted data from the selected articles into tables. The following data were extracted for each study: author, year of publication, study design, comparison, sample size, endometriosis type (superficial, ovarian, and deep endometriosis), American Society of Reproductive Medicine (ASRM) stage (Revised American Society for Reproductive Medicine classification of endometriosis: 1996 1997), and results. When data were missing from the manuscript, efforts were made by two of the authors (CHM and FRO) to contact the corresponding authors to obtain complete data.

Risk of bias
To assess the quality of included studies, publicly available study quality assessment tools provided by the National Institutes of Health (NIH) National Heart, Lung and Blood Institute were utilized, with specific forms for case-control and prospective non-randomized studies (https ://ww w.nhl bi.ni h.gov /heal th-to pics/ study -qual ity-a ssess menttools ). Conflicts regarding study quality were resolved with the senior authors (MPA and MSA). Studies that fulfilled 70% or more criteria were classified as good, 30-70% as fair, and less than 30% as poor quality.

Statistical analysis
Studies were summarized and described qualitatively. Due to the heterogeneity of included studies, meta-analysis was not performed.

Study selection
Using the search strategy described above, the initial search identified 209 studies. After excluding 65 duplicates, 122 of the remaining 144 studies were excluded following title and abstract review. Full-text screening of 23 studies to evaluate for inclusion and exclusion criteria according to the study design, type of publication, methods, and results was performed by two authors (CHM and FOR), yielding 12 articles, all case-control studies (Khan et al. 2016, Xu et al. 2017, Wang et al. 2018, Akiyama et al. 2019, Ata et al. 2019, Chen et al. 2020, Hernandes et al. 2020, Perrotta et al. 2020, Wei et al. 2020, Chao et al. 2021, Lee et al. 2021, Svensson et al. 2021, meeting study inclusion criteria for data abstraction and qualitative analysis (Fig. 1). Key design characteristics of included studies are summarized in Table 1.

Study quality assessment and risk of bias
Eight studies were rated as fair (Khan et al. 2016, Xu et al. 2017, Wang et al. 2018, Akiyama et al. 2019, Ata et al. 2019, Chen et al. 2020, Wei et al. 2020, Lee et al. 2021, three as good (Hernandes et al. 2020, Perrotta et al. 2020, Svensson et al. 2021, and one as poor quality (Chao et al. 2021). Only one study (Perrotta et al. 2020) included a sample size justification and only one (Svensson et al. 2021) included concurrent controls. Researchers were not blinded in any of included studies (Tables 2 and 3). Owing to their case-control designs, no studies provided more than limited evidence (level 3b according to Oxford Center for Evidence-Based Medicine) for their findings.

Control cohorts utilized for microbiome analysis
Eleven studies (Khan et al. 2016, Wang et al. 2018, Ata et al. 2019, Akiyama et al. 2019, Chen et al. 2020, Hernandes et al. 2020, Perrotta et al. 2020, Wei et al. 2020, Chao et al. 2021, Lee et al. 2021, Svensson et al. 2021) compared the microbiome between patients with and without endometriosis and one (Xu et al. 2017) compared endometriotic patients with and without chronic stress. In these studies, the control groups comprised patients who underwent surgery for other benign gynecological conditions (Khan et al. 2016, Akiyama et al. 2019, Chen et al. 2020, Hernandes et al. 2020, Lee et al. 2021, Perrotta et al. 2020, Wei et al. 2020, infertility (Wang et al. 2018), or chronic pelvic pain (CPP) (Chao et al. 2021) or asymptomatic patients who presented for routine gynecologic (Ata et al. 2019, Chao et al. 2021 or general visits (Svensson et al. 2021). The relative expression of bacteria across anatomical sites in patients with endometriosis compared to those without endometriosis is summarized in Table 3.

Female reproductive tract microbiome and endometriosis
Seven studies evaluated the microbiome in vaginal and cervical samples (Akiyama et al. 2019, Ata et al. 2019, Chen et al. 2020, Hernandes et al. 2020, Perrotta et al. 2020 1. Was the research question or objective in this paper clearly stated and appropriate?
Was the study population clearly specified and defined?
3. Did the authors include a sample size justification?
Were controls selected or recruited from the same or similar population that gave rise to the cases (including the same timeframe)?
Were the definitions, inclusion and exclusion criteria, and algorithms or processes used to identify or select cases and controls valid, reliable, and implemented consistently across all Were the cases clearly defined and differentiated from controls?
If less than 100% of eligible cases and/ or controls were selected for the study, were the cases and/or controls randomly selected from those eligible?
Were the investigators able to confirm that the exposure/risk occurred prior to the development of the condition or event that defined a participant as a case?
10. Were the measures of exposure/risk clearly defined, valid, reliable, and implemented consistently (including the same time period) across all study participants? Case-control studies were assessed using NIH study quality assessment tools. The replies represents if the study fulfilled each criteria (Y, yes; N, no: N/A, not applicable, not reported, or cannot determine). Overall study quality is summarized in the final row. Table 3 Relative expression of bacterial loads in patients with endometriosis compared to patients without endometriosis.

3:3
Summary of studies that evaluated the microbiome at different sites in patients with and without endometriosis. All studies compared relative frequencies of all bacteria reads performed by 16S RNA next generation sequencing. C H Miyashira et al.

3:3
Wei et al. 2020, Chao et al. 2021) and three (Khan et al. 2016, Hernandes et al. 2020, Wei et al. 2020) in endometrial samples. Akiyama et al. (2019) performed a case-control study comparing 39 moderate-to-severe endometriosis patients against 30 patients with benign gynecological conditions undergoing surgery and found that the cervical microbiota was similar between the two groups. Lactobacilli species were predominant in both groups whereas Enterobacteriaceae and Streptococcus were more prevalent in women with endometriosis (P < 0.05). Chen et al. (2020) compared the cervical and vaginal microbiome in 68 Chinese women stratified by the presence of endometriosis and adenomyosis and defined 4 groups: no endometriosis or adenomyosis, endometriosis only, adenomyosis only, and both adenomyosis and endometriosis (n = 36, 13, 12 and 7, respectively). Lactobacillus was the most prevalent genus in the vagina in all groups, but the genus Atopobium was more commonly identified in women with both endometriosis and adenomyosis. Campylobacter, Ezakiella, and Faecalibaterium were also more abundant among patients with both endometriosis and adenomyosis. Ata et al. (2019) studied the cervical and vaginal microbiome of 28 Caucasian women (14 with endometriosis ASRM stages III-IV and 14 asymptomatic patients without endometriosis who presented for a routine gynecological visit). They found that women with endometriosis were more likely to harbor Alloprevotella in the cervix, while Atopobium and Sneathia were only identified in the controls. Gemella and Atopobium were not detected in the vaginal microbiomes of endometriosis patients. When excluding Lactobacillus from the analysis, the relative abundance of Gardnerella, Streptococcus, Escherichia/Shigella, and Ureaplasma was found to be increased in endometriosis patients.
Hernandes et al. (2020) compared vaginal fluid and endometrial samples between 10 women with deep endometriosis and 11 without endometriosis undergoing benign gynecological surgery. While Lactobacillus predominated in the vaginal fluid of both endometriosis and control patients, Gardnerella and Prevotella were in lower relative abundance in samples of vaginal fluid and endometrium from endometriosis patients. Perrotta et al. (2020) conducted an observational study comparing 35 Brazilian women with endometriosis stages I-IV against 24 without endometriosis undergoing surgery for benign gynecological diseases. The authors found no significant differences in the vaginal and rectal microbiome between endometriosis and control patients.  Chao et al. (2021) compared 128 samples from the posterior vaginal fornix of Chinese women and divided them into 3 groups: 37 women with CPP plus endometriosis or adenomyosis, 25 women with CPP without endometriosis/adenomyosis, and 66 without CPP with endometriosis/adenomyosis who presented for a routine gynecologic visit. The group with endometriosis/ adenomyosis and associated CPP was associated with a greater relative abundance of bacteria of the genera Clostridium, Alloscardovia, Veillonella, and Stenotrophomonas and a lower abundance of Megasphaera, Lactobacillus, and Shuttleworthia compared to those without endometriosis. Khan et al. (2016) identified 32 women with endometriosis stages I-IV and 32 without endometriosis who underwent benign gynecological surgery and compared the presence of 5 bacterial families in endometrial samples: Lactobacillacae, Streptococcaceae, Staphylococaceae, Enterobacteriaceae, and Moraxellaceae. In women with endometriosis, there was an increase in Streptococcaceae and Moraxellaceae.

Peritoneal fluid microbiome
Three studies (Wang et al. 2018, Wei et al. 2020, Lee et al. 2021 analyzed the relationship between endometriosis and the microbiome within the peritoneal fluid, one of which (Wei et al. 2020) also collected samples from other sites. Lee et al. (2021), compared 45 women with stages III and IV endometriosis (mean age: 36.2 ± 1.3 years old) against 45 controls who underwent laparoscopy, 31 for myomas and 14 for benign ovarian cysts (mean age: 39.4 ± 1.1 years old). At a genus level, there was a significant increase in Acinetobacter, Pseudomonas, Streptococcus, and Enhydrobacter in the endometriosis group compared to the control group (P < 0.05), as well as a significant reduction in the genera Propionibacterium, Actinomyces, and Rothia (P < 0.05). Wang et al. (2018) compared 55 individuals with endometriosis and infertility (mean age: 37.2 ± 8.2 years old) against 30 controls with infertility without endometriosis (mean age: 37.7 ± 7.4 years old). The main bacteria detected in the peritoneal fluid were Proteobacteria and Firmicutes, followed by Actinobacteria, Bacteroides, Fusobacterium, and Tenericutes. There was no statistically significant difference between endometriosis and control groups (P > 0.05). Wei et al. (2020) compared peritoneal fluid samples of 50 Chinese women, 36 with pelvic endometriosis and 14 who underwent laparoscopy for ovarian teratoma, serous cystadenoma, or uterine fibroids. They found a significant increase in Pseudomonas and Sphingobium in the peritoneal fluid of women with endometriosis.

Fecal microbiome
Two case-control studies compared the fecal microbiome of women with and without endometriosis (Ata et al. 2019, Svensson et al. 2021. Svensson et al. (2021) included 264 patients, comparing 66 women with endometriosis and with 198 matched controls from a cohort of descendants participating in the Malmö Diet and Cancer Cardiovascular Cohort (MDC-CC). The analysis showed only three bacteria with a significant difference with higher abundance between endometriosis and control groups: Lachnospira, Oscillospira, and a genus in the order Bacterioidales (P < 0.05). Ata et al. (2019) compared 14 women with endometriosis against 14 asymptomatic reproductive-aged women who presented for a routine well-woman visit or preconception counseling. They found that the relative abundance of bacteria in the genera Sneathia, Barnesella, and Gardnerella from stool samples of the endometriosis group was significantly decreased (P < 0.001).

Endometriosis stage or type and microbiome
Three studies (Khan et al. 2016, Perrotta et al. 2020, Svensson et al. 2021) compared the microbiome between patients across different endometriosis types or stages (Table 4). Perrotta et al. (2020) showed that the vaginal microbiome during the menstrual phase was significantly different between patients with ASRM stages III-IV compared to stages I-II (P = 0.019), which was not significantly different from the vaginal microbiome of control patients. Patients with ASRM stage III-IV endometriosis had vaginal microbiomes enriched for Anaerococcus compared with lower-stage patients.
Two studies (Khan et al. 2016, Svensson et al. 2021) compared the fecal or ovarian cyst microbiome among different types of endometriosis without using the ASRM staging system. Svensson et al. (2021) found no significant difference in the stool microbiome between ovarian and deep endometriosis. Khan et al. (2016) found a significantly higher percentage of Streptococcaceae and Staphylococaceae and a significant reduction in Lactobacillacae in the ovarian endometrioma cystic fluid in comparison with nonendometriotic cysts.

Microbiome and menstrual cycle phase
While six microbiomal studies among endometriosis patients (Khan et al. 2016, Akiyama et al. 2019, Ata et al. 2019, Wei et al. 2020, Perrotta et al. 2020, Chao et al. 2021 reported on the menstrual cycle phase, only two compared the microbiome during different menstrual phases (Table  4). Akiyama et al. (2019) found no significant differences in the cervical microbiome across different menstrual phases of either endometriosis or control patients. Perrotta et al. (2020) observed an increase in vaginal Lactobacillus species in the proliferative phase compared to the secretory and menstrual phases. The authors (Perrotta et al. 2020) also observed an increase in anaerobic bacteria in the endometrium or peritoneal fluid during the proliferative and secretory phases compared to the menstrual phase.

Hormonal intake and microbiomal variation among endometriosis patients
Four studies (Khan et al. 2016, Hernandes et al. 2020, Chao et al. 2021, Svensson et al. 2021 included women possibly taking hormonal agents, while seven (Wang et al. 2018, Akiyama et al. 2019, Ata et al. 2019, Chen et al. 2020, Perrotta et al. 2020, Wei et al. 2020, Lee et al. 2021 were restricted to patients without current hormonal intake (Table 4). Khan et al. 2016 evaluated the effect of the use of a gonadotropin-releasing hormone agonist (GnRHa) on women with endometriosis and showed that Lactobacillacae was significantly decreased (P < 0.01), while Streptococcaceae, Staphylococaceae, and Enterobacteriaceae were significantly increased (P < 0.05 for each) in vaginal swabs from GnRHatreated women with endometriosis compared with GnRHa-untreated women. In contrast, vaginal samples from GnRHa-treated control women showed significantly higher colonization with Staphylococaceae (P < 0.05) and insignificant colonization with Enterobacteriaceae (P = 0.071) compared with samples from GnRHa-untreated control women. Svensson et al. (2021) examined the fecal microbiome among women with endometriosis and identified a higher abundance of Blautia, Ruminococcus, and Butyricimonas among those taking hormonal medications, including estrogen, combined oral contraceptives, progestin, or gonadotropin-releasing hormone analogs. The remaining studies did not report on changes in the microbiome in association with hormonal intake.

Endometriosis symptoms and microbiome
Three studies (Xu et al. 2017, Chao et al. 2021, Svensson et al. 2021) compared the association between endometriosis symptoms and the microbiome. Svensson et al. (2021) compared 66 patients with endometriosis and 198 asymptomatic women without endometriosis from the MDC-CC cohort described above. In a subanalysis of the 66 endometriosis patients, they reported no significant association of their stool microbiome with the intensity of pain symptoms or digestive complaints, including abdominal pain, constipation, diarrhea, bloating, and vomiting. Xu et al. (2017) studied the fecal microbiome of ten subjects with endometriosis, five reporting chronic stress and five not reporting chronic stress. They found significantly decreased levels of Paraprevotella, Odoribacter, Veillonella and Ruminococcus in chronically stressed endometriosis patients, while Prevotella was significantly increased among the chronically stressed endometriosis patients. Chao et al. (2021) compared the fecal microbiome of 37 patients with endometriosis or adenomyosis plus (CPP, 25 patients without endometriosis but reporting CPP, and 66 without endometriosis or CPP. Patients with endometriosis and CPP were found to have the lowest relative abundance of Lactobacillus jensenii and the highest abundance of Clostridium butyricum compared to the other two groups. Endometriosis patients with CPP also had significantly lower Lactobacillus and Shuttleworthia and significantly higher Clostridiales and Alloscardovia abundance compared with no endometriosis patients without CPP, but no difference compared to patients with CPP and without endometriosis.

Discussion
This review identified multiple microbiome studies on patients with endometriosis. This systematic review highlighted many of the limitations of such studies, including heterogeneous methods for identifying and typing bacteria, various anatomical sources for microbiomal sample collection (fecal, vaginal, cervical, peritoneal, endometrial, and intra-lesional), significant heterogeneity among patients both with endometriosis and the so-called controls (including heterogeneity in menstrual cycle timing, use of hormonal medications, symptomatology, disease severity, and the presence of comorbid conditions such as adenomyosis), and inherent publication bias. Such limitations collectively have precluded completing a metaanalysis of the underlying studies.
Nonetheless, several trends appear to stand out from these imperfect, heterogeneous studies. Several studies suggest that peritoneal fluid appears to contain a different distribution of bacteria among women with endometriosis, though only Pseudomonas (Wei et al. 2020, Lee et al. 2021 was found to be overrepresented among patients with endometriosis in multiple studies. Fecal microbiome studies (Xu et al. 2017, Ata et al. 2019, Svensson et al. 2021 appear to be conflicting in the reported prevalence of various bacteria. The one study that reported an association between chronic stress in endometriosis and an altered fecal microbiome (Xu et al. 2017) is yet to be validated.
While the association between the fecal microbiome and endometriosis remains inconclusive, the topic remains biologically plausible. The gut microbiome interacts with immune and metabolic systems and is associated with various disease states, including inflammatory bowel syndrome, arthritis, psoriasis, and cancer (Smet et al. 2021, Wertman et al. 2021. The dysbiosis of the gastrointestinal tract can lead to higher gut permeability, a higher concentration of macrophages in peritoneal fluid, secretion of interleukin IL-1 and IL-10, and modulation of local immune response to the clearance of menstrual debris and thus potentiate endometriosis development (D'Alterio et al. 2021). Also, it has been suggested that dysbiosis of the gut microbiome may alter the so-called estrobolome and lead to enhanced estrogen deconjugation and increased free circulating levels, potentially contributing to endometriosis progression (García-Peñarrubia et al. 2020).
The inferior female reproductive tract is a major source of human microbiota, urogenital microbiota being responsible for 9% of all bacterial species in the human body (Cani 2018). Cervicovaginal lactobacilli deficiency is correlated with higher genital pro-inflammatory cytokines and activation of antigen-presenting cells through lipopolysaccharide (LPS) pathways (Cani 2018). Also, studies have shown that the fecal and vaginal microbiota are correlated and that the use of probiotics can impact both the fecal and vaginal environments, suppressing pro-inflammatory cytokine production (Melis et al. 2018).
While the diversity of the vaginal and fecal microbiome is well-recognized, the presence of meaningful bacterial colonization at other sites such as the endometrium or within endometriosis biopsies remains controversial.
Identification of bacteria at supposedly sterile sites may suggest contamination or another infectious process rather than evidence of endometriosis (Chen et al. 2017). The upper genital tract may become colonized via the bloodstream, mesenteric lymph nodes, or through the retrograde progression of cervical and vaginal bacteria, though its role in modulating uterine health in unclear (Baker et al. 2018.
Previous studies suggested that the microbiome in the vaginal tract may be influenced by hormonal treatments and the menstrual cycle phase. Despite this, only two studies (Akiyama et al. 2019, Perrotta et al. 2020 attempted to address confounding from the menstrual phase. The lack of such standardization and correction for clear confounding variables is a significant limitation that should be addressed in future studies. Similarly, most studies did not attempt to control for the endometriosis stage, thus limiting the generalizability of observed results. For example, patients with endometriosis infiltrating the bowel have a much more plausible and direct connection to developing an altered fecal microbiome than patients with endometriosis without bowel involvement. Future prospective studies with larger samples and stricter methodology combined with patient standardization are needed to clarify the role of the microbiome in endometriosis pathogenesis and clinical features and allow for a precise measurement of the effect of any interventions.

Conclusion
Clear differences have been reported from studies of the fecal, vaginal, cervical, endometrial, and peritoneal microbiomes of women with and without endometriosis. An association of the microbiome with the hormonal intake, menstrual cycle phase, and pain symptoms in patients with endometriosis was reported by a few studies. However, studies are limited due to a lack of standardization and small samples, and the cause-effect relationship between the microbiome and endometriosis is yet to be established.

Supplementary materials
This is linked to the online version of the paper at https ://do i.org /10.1 530/ R AF-21 -0113 .

Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Author contribution statement
Carlos H Miyashira: study design, data collection, data analysis and interpretation, manuscript preparation. Fernanda Reali Oliveira: study design, data collection, data analysis and interpretation, manuscript preparation. Marina Paula Andres: study design, data analysis and interpretation, manuscript preparation. Julian A Gingold: manuscript preparation. Mauricio Simões Abrão: study design, manuscript preparation.