The impact of the microbiota-gut-brain axis on endometriosis-associated symptoms: mechanisms and opportunities for personalised management strategies

The influence of the MGB axis on the immune system has been investigated in several chronic inflammatory pain conditions. In this section we consider data from studies on conditions and symptoms relevant to endometriosis. Studies in animal models have provided evidence to support the two-way connection between pain pathways and the gut microbiota ( Table 3 ). For example, germ-free mice lacking gut bacteria had increased visceral hypersensitivity, which was normalised following microbial re-colonisation ( Luczynski et al. 2017 ). The phenotype was found to be transferable by faecal microbiota transfer (FMT) ( Yang et al. 2019 , Lucarini et al. 2022 ), and attenuated by treatment with antibiotics ( Aguilera et al. 2021 , Ding et al. 2021 ), potentially in an inflammasome-dependent manner ( Scuderi et al. 2020 , Aguilera et al. 2021 ). Taken together, these studies support a complex role involving the gut microbiota and immune interactions in pain responses. Table 3 Research studies in animal models investigating the relationship between the gut microbiota, pain, and inflammation. Model Findings Reference Mouse model of PI-IBS and PI-IBS treated with Bifidobacterium longum - VH was significantly lower in B. longum -treated mice compared to PI-IBS. Gu et al. (2016) - IL-18 and IL-1β expression were significantly lower in B. longum -treated mice compared to PI-IBS. - B. longum may inhibit NLRP3 inflammasome. MD rat model (male) - Significantly increased VH but no significant difference in total pain scores. Hoban et al. (2016) - More depressive-like behaviours. - Decreased gut microbiota β-diversity. Germ-free mice - Significantly increased VH, which normalised following microbial colonisation. Luczynski et al. (2017) - Altered volume of pain-processing brain structures: smaller anterior cingulate cortex and larger periaqueductal grey. MS mouse model vs TLR4 KO-MS mouse model - Increased VH in MS but not in TLR4 KO-MS mice. Tang et al. (2017) - LPS-treatment induced VH in MS via TLR4. - VH blocked by inhibition of TLR4 signalling. Mouse model of alcohol-induced neuroinflammation (female) - Neuroinflammation and increased intestinal proinflammatory cytokines attenuated with Ab treatment. Lowe et al. (2018) - Ab-treatment increased mRNA expression of some inflammasome components. Rat model of spared nerve injury (male) - Depression-susceptible rats had significantly decreased gut microbiota α-diversity. Yang et al. (2019) - FMT to MD-mice transferred a painful phenotype and depression-like behaviours. Rat model of IBS - Inflammasome inhibition reduced NF-ΚB expression and inflammation, and restored IBS-associated tight junction alterations. Scuderi et al. (2020) Mouse model of inflammasome inhibition* - Increased abundance of Bifidobacterium in Casp1 KO compared to WT. Aguilera et al. (2021) - Antibiotics reduced immune and inflammatory marker expression in Casp1 KO but not WT. - Antibiotics reduced pain in WT but not Casp1 KO. Mouse model of neuropathic pain † - Gut microbiota induced pain by influencing pro- and anti-inflammatory T cells. Ding et al. (2021) - Pain was attenuated by antibiotics. Rat model of colitis (male) - Unique bacterial profile and increased F/B ratio compared to controls. Lucarini et al. (2022) - FMT to healthy rats transferred VH. - FMT recipient rats had increased acetate but decreased butyrate. - FMT recipients had increased plasma IL-6 and TGF-β. * Casp1 knock-out; † Chronic-constriction injury of sciatic nerve. Ab, antibiotic; F/B, firmicutes/bacteroidetes ratio; FMT, faecal microbiota transfer; IBS, irritable bowel syndrome; MD, microbiota-depleted; MS, maternal separation; PI, postinfectious; VF, visceral hypersensitivity. Research studies in animal models investigating the relationship between the gut microbiota, pain, and inflammation. * Casp1 knock-out; † Chronic-constriction injury of sciatic nerve. Ab, antibiotic; F/B, firmicutes/bacteroidetes ratio; FMT, faecal microbiota transfer; IBS, irritable bowel syndrome; MD, microbiota-depleted; MS, maternal separation; PI, postinfectious; VF, visceral hypersensitivity. The potential impact of estrobolome-contributing microbial populations on pain sensitivity has also been demonstrated in a recent study: visceral sensitivity fluctuated throughout the estrous cycle in wild-type mice and increased following a reduction in ovarian steroids as a consequence of ovariectomy. Notably, neither of these effects were seen in germ-free mice, suggesting one mechanism of microbial influence on pain occurs in an oestrogen-dependent manner ( Tramullas et al. 2021 ). The potential role of the gut microbiota in the symptomology of IBS has been researched in some depth in both patients and animal models ( Table 4 ). IBS patients are reported to have significant differences in microbial diversity compared to healthy controls. The inverse association between Akkermansia muciniphila and pain intensity ( Cruz-Aguliar et al. 2019 ) is of interest due to its association with improved intestinal barrier function ( Cani & de Vos 2017 ). FMT has provided further supporting evidence, with transfer of the phenotype from patients to mice, and transfer from healthy donors to patients reducing symptoms and re-diversifying the gut microbiota. To date there are no reports of trials using FMT to treat endometriosis patients. Table 4 Research studies investigating the relationship between the gut microbiota and gastrointestinal symptoms. Model/cohort Findings Reference IBS patients - Microbial signatures clustered into two groups: normal microbiota vs. increased F/B ratio. Jeffery et al. (2012) - IBS with normal microbiota were more likely to have depression. - Suggests potential differing triggers for IBS-like symptoms. IBS patients and MD mouse model (male) - FMT from IBS patients to MD-mice transferred phenotype. Ge et al. (2017) - SCFAs and secondary bile acids were subsequently decreased in recipient mice. IBS patients and GF mouse model - FMT from IBS patients to GF-mice transferred phenotype including intestinal barrier dysfunction, innate immune activation, and anxiety-like behaviour. De Palma et al. (2017) IBS patients and healthy donor - FMT from healthy donor to IBS patients reduced abdominal pain symptoms and increased α- and β-diversities. Cruz-Aguliar et al. (2019) - Patients with microbiota mostly like the donor had the greatest reduction in pain. - Abundance of Akkermansia muciniphila inversely correlated with pain intensity. Review – multiple cohorts - Potential influence of Lactobacillaceae and Bacteroides metabolites on inflammation and bloating. Pittayanon et al. (2019) IBS patients - Significant differences in microbiome and metabolome profiles compared to controls. Jeffery et al. (2020) - Faecal metabolomes could differentiate IBS patients with and without bile acid malabsorption. - Decreased bacterial diversity. Meta-analysis of IBS patients - FMT from healthy donors significantly decreased IBS symptoms and improved QoL. Wang et al. (2023) Meta-analysis of IBS patients - FMT from healthy donors significantly decreased IBS symptoms but decreased QoL. Halkjær et al. (2023) Review of IBS patients - Increased F/B ratio in IBS. Shaikh et al. (2023) - No specific microbial signature. FMT, faecal microbiota transfer; IBS, irritable bowel syndrome; SCFA, short-chain fatty acid; QoL, quality of life. Research studies investigating the relationship between the gut microbiota and gastrointestinal symptoms. FMT, faecal microbiota transfer; IBS, irritable bowel syndrome; SCFA, short-chain fatty acid; QoL, quality of life. Comorbidity of GI issues and mood disorders is common. However, clinical evidence to support the role of the MGB axis remains limited. Similar to other areas of gut–brain research, differences in methodology and outcome measures have created challenges when comparing data. Although several studies have identified differences in both α- and β-diversities in the gut microbiota of people with depression and anxiety compared to controls, these findings are not consistent ( Simpson et al. 2021 ). Additionally, there were no uniform findings in the differing abundance of specific bacterial species associated with neither depression nor anxiety ( Simpson et al. 2021 ).

Article conception, FHY, AWH, and PTKS; literature survey and writing, FHY; editing and reviewing, FHY, AWH, PTKS, and SO; supervision, AWH, PTKS, and SO.

In the following section, to provide a framework for considering the role of the gut microbiome and its metabolites in endometriosis, we provide a brief overview and references to some recent relevant papers. The gut microbiota is the community of microorganisms, including bacteria, viruses, and archaea, residing in the GI tract. The bacterial community is dominated by the phyla Firmicutes and Bacteroidetes, which comprise approximately 90% of the total gut bacteria ( Sommer & Bäckhed 2013 ), with Fusobacteriota and Verrucomicrobiota present in low abundance ( Cryan et al. 2019 ). Human gut microbiota have been broadly grouped into separate enterotypes, depending on levels of three specific genera: Bacteroides, Prevotella, and Ruminococcus. Differing enterotypes are associated with the consumption of certain diets: the Bacteroides and Prevotella enterotypes are associated with high-fat/high-protein diets and high-carbohydrate diets, respectively ( Cryan et al. 2019 ). Bacterial diversity can be differentiated by α- and β-diversity indices, with the former focussed on diversity within a single sample and the latter comparing population diversity between different samples ( Wagner et al. 2018 ). Increased diversity is generally considered to be associated with improved health outcomes ( Valdes et al. 2018 ). The gut microbiota plays essential roles in promoting gut health, including maintaining intestinal barrier function, and priming and maintenance of the immune system. The bacterial community regulates and activates both peripheral and resident immune cells, either via direct contact or compounds secreted through the mucus layer and gut epithelium, as well as signalling to the brain via nerve stimulation. The MGB axis is a two-way communication pathway linking the CNS and gut bacteria. Gut bacteria produce an assortment of vital metabolites including: bile acids, short-chain fatty acids (SCFAs), hormones, and neurotransmitters; which can signal to the brain and regulate a range of functions throughout the body ( Liu et al. 2022 ). Signalling between cells in the gut and brain is mediated via neural pathways involving the vagus nerve and the enteric/parasympathetic nervous system, as well as immunological and hormonal factors, including those contributing to the HPA axis (see the comprehensive review by ( Cryan et al. 2019 )). Briefly, associations have been found between altered parasympathetic nerve activity, pain, and bacterial composition, including evidence from CNS disorders ( Wang & Kasper 2014 ). The production and utilisation of metabolic products, such as tryptophan and serotonin, by certain gut bacteria provide a secondary mechanism for their role in mood disorders via activation of the HPA axis ( O’Mahony et al. 2015 ). Gut microbiota also play a role in the maturation and maintenance of microglia, CNS-resident immune cells, which function in neuroinflammation and pain processing ( Erny et al. 2015 ). Activation of these neuroimmune cells is considered one of the key mechanisms in central sensitisation due to the production of proinflammatory mediators, including IL-1β, interferon-γ, and TNF-α ( Guo et al. 2019 ). These, amongst other cytokines and chemokines, disrupt the ratio of glutamate versus γ-aminobutyric acid (GABA) in synaptic transmission, leading to decreased pain thresholds ( Ustianowska et al. 2022 ). SCFAs are an important product of bacterial metabolism, produced by certain bacterial species as a by-product of dietary fibre fermentation. They modulate the inflammatory status of the gut by regulating the immune response ( Liu et al. 2022 ) and maintain the mucosal barrier by promoting the proliferation of intestinal epithelial cells ( Vinolo et al. 2011 ). SCFAs act via two primary mechanisms: activation of G-protein-coupled receptors (GPCRs), GPR41 and GPR43, expressed on neutrophils and monocytes, and throughout the GI tract; and inhibition of histone deacetylases ( Tan et al. 2014 ). SCFAs can promote peripheral Treg generation ( Arpaia et al. 2013 ) and have been found to regulate neuroinflammation via the GPCR HCAR2 ( Boccella et al. 2019 ), expressed during pain in the hypothalamus ( Li et al. 2020 ). The estrobolome is the collection of gut bacteria capable of altering the concentrations of bioactive steroids, including oestrogens, by enzymatic activities that cleave side chains from conjugated steroids ( Fig. 2A ). Examples include metabolism of oestrone-3-glucuronide and oestradiol-17-glucuronide, to oestrone (E1) and oestradiol (E2), respectively ( Ervin et al. 2019 ). This has been further evidenced by a correlation between microbial diversity and higher E2 levels ( Shin et al. 2019 ), with the bidirectionality of this relationship shown by the microbiota changes caused by ovariectomy ( O'Mahony et al. 2017 ). Figure 2 (A) The estrobolome.Certain gut bacteria, including Bacteroides, Bifidobacterium, Escherichia, and Lactobacillus, are capable of impacting circulating oestrogen concentrations. These bacterial genera have β-glucuronidase and β-glucosidase activity, enzymes that deconjugate endogenous oestrogen and exogenous phytoestrogens in the gut ( Kwa et al. , 2016 ). A higher prevalence of these bacteria, or increased activity of the enzymes, leads to increased concentration of biologically active oestrogen metabolites in circulation, which may influence cell proliferation or immune responses ( Symons et al. , 2018 ). On the other hand, decreased activity will diminish circulating free oestrogen leading to increased excretion of conjugated parent oestrogens, oestrone (E1) and oestradiol (E2) ( Spichak et al. , 2018 ). Created with BioRender.com. (B) Summary of potential symptom-management strategies via manipulation of the gut microbiome.A broad range of interventions have been trialled in endometriosis patients for their potential beneficial impact on symptoms. Different dietary modifications have been the most extensively researched strategies, with antibiotic and probiotic treatments currently in their infancy. *Trials conducted only in patients with IBS not endometriosis. Created with BioRender.com. (A) The estrobolome.Certain gut bacteria, including Bacteroides, Bifidobacterium, Escherichia, and Lactobacillus, are capable of impacting circulating oestrogen concentrations. These bacterial genera have β-glucuronidase and β-glucosidase activity, enzymes that deconjugate endogenous oestrogen and exogenous phytoestrogens in the gut ( Kwa et al. , 2016 ). A higher prevalence of these bacteria, or increased activity of the enzymes, leads to increased concentration of biologically active oestrogen metabolites in circulation, which may influence cell proliferation or immune responses ( Symons et al. , 2018 ). On the other hand, decreased activity will diminish circulating free oestrogen leading to increased excretion of conjugated parent oestrogens, oestrone (E1) and oestradiol (E2) ( Spichak et al. , 2018 ). Created with BioRender.com. (B) Summary of potential symptom-management strategies via manipulation of the gut microbiome.A broad range of interventions have been trialled in endometriosis patients for their potential beneficial impact on symptoms. Different dietary modifications have been the most extensively researched strategies, with antibiotic and probiotic treatments currently in their infancy. *Trials conducted only in patients with IBS not endometriosis. Created with BioRender.com. Gut dysbiosis, resulting from disturbances in normal microbiota communities, can be caused by many factors, such as stress, physical illness, antibiotics, and dietary changes ( Valdes et al. 2018 ). Dysbiosis and inflammation can increase permeability of the intestinal barrier via weakening of the tight junctions, allowing movement of bacteria and pathogenic-associated molecular patterns (PAMPs) into circulation ( Gieryńska et al. 2022 ). Subsequent recognition of PAMPs by TLRs triggers proinflammatory cytokine production, inducing activation of transcription factors, such as NF-κB ( Kawai & Akira 2010 ). The downstream effects of these pathways result in both local and systemic low-grade inflammation.

Clinical trials of dietary intervention for disease management are challenging to implement and standardise, with a plethora of variables likely to affect the outcomes ( Nap & De Roos 2022 ). Currently, most research into associations between diet and endometriosis is focussed on risk of disease development, rather than adapting diet for symptom-management ( Nap & De Roos 2022 ). However, there is anecdotal evidence within the endometriosis community for the benefit of dietary modifications as a self-management strategy and some preliminary clinical evidence to support these ideas. Diets (or specific foods) believed to increase bacterial diversity and growth of bacterial species associated with good health are often referred to as ‘prebiotics’, examples include diets rich in fibre and fermented foods ( Valdes et al. 2018 ). Health benefits of these diets have been described for various conditions, including IBS ( Salmeri et al. 2023 ); however, to date there has been no comprehensive randomised control trial (RCT) in endometriosis patients. More generally, other research into the impact of Western diet, including the increased consumption of ultra-processed foods, is gaining momentum, with reports of associations with increased low-grade inflammation ( Tristan Asensi et al. 2023 ). There is concern that diets high in ultra-processed foods may exacerbate symptoms in those with existing chronic inflammatory conditions, which may also include endometriosis. Recent surveys investigating the popularity of different diets used by people with endometriosis, and perceived effects on symptoms and QoL, have reported that, although no single dietary intervention appeared to be uniquely effective, many respondents found their chosen modification to be beneficial ( Krabbenborg et al. 2021 , Armour et al. 2021 ). In an Australian survey, 163 respondents had used dietary intervention, 69.0% of whom reported a reduction in the use of pharmaceutical medication ( Armour et al. 2021 ). In a Dutch study, 55.5% of the 157 respondents reported nutrition affecting their symptoms and 46.5% were currently following a diet ( Krabbenborg et al. 2021 ). These surveys were consistent in finding gluten -free, dairy -free/low lactose, and low fermentable oligosaccharides, disaccharides, monosaccharides and polyols (FODMAPs) as the most popular diets, although both studies are caveated by their relatively small participant numbers and limited geographical reach. The largest survey to date received 1385 responses, predominantly from the UK, of which 52.2% had tried adapting their diet to manage their endometriosis-associated gut symptoms ( Deepak Kumar et al. 2023 ). Again, gluten-free was one of the most popular diets; however, only 0.6% were following a low-FODMAP diet, highlighting the strength of a larger dataset. FODMAPs are fermentable oligosaccharides, disaccharides, monosaccharides and polyols, which in high doses can cause inflammation and visceral pain ( Zhou et al. 2017 ). A low-FODMAP diet is popular for management of IBS, with symptoms shown to improve after three weeks, alongside a reduction in serum levels of proinflammatory cytokines ( Hustoft et al. 2017 ). However, long-term use of the low-FODMAP diet may have a negative impact on the gut microbiome ( Staudacher 2017 ). One culprit for inflammatory responses to certain foods is histamine, released by mast cells present in the gut mucosa and further stimulated in a positive feedback loop with oestrogen ( Theoharides 2017 ). Three weeks on a low-FODMAP diet ( n = 19) was shown to reduce histamine levels eight-fold, in comparison to a high-FODMAP diet ( n = 18) ( McIntosh et al. 2017 ). Histamine mast cells express oestrogen receptors ( De Leo et al. 2017 ) and have been implicated in both pathogenesis and pain mechanisms of endometriosis ( Kirchhoff et al. 2012 , Mariuzzi et al. 2016 ), providing a mechanistic link between mast cell activation and intestinal inflammation. In a study of 160 women, those with both endometriosis and IBS were three-fold more likely to find a low-FODMAP diet effective for improving symptoms, compared to those with IBS alone ( Moore et al. 2017 ). This could suggest the cause of IBS-type symptoms frequently reported by endometriosis patients may differ from those with IBS alone and be more receptive to dietary intervention. A gluten-free diet is frequently adopted by people with endometriosis; however, there is currently no clinical evidence to support this practice. There has been one retrospective observational study of women with endometriosis who followed a gluten-free diet for 12 months, 75% of whom reported significant pain improvement ( Marziali et al. 2012 ). However, 88 of the original 295 participants withdrew within two weeks due to associated abdominal side effects. Anti-inflammatory diets, such as the typical Mediterranean diet, consisting of fruit, vegetables, whole grains, and oily fish, with low quantities of dairy and red meat, have been proven to decrease inflammatory markers including IL-6 and C-reactive protein ( Tristan Asensi et al. 2023 ). Five months on a Mediterranean diet was found to significantly improve pain in 68 women with endometriosis; however, the study had no control group ( Ott et al. 2012 ). A diet high in fermented foods has been shown to increase microbial diversity and decrease inflammatory markers in 18 healthy adults ( Wastyk et al. 2021 ). Signorile et al. compared 3 months of an anti-inflammatory dietary supplement, a linseed oil/calcium salt combination, or a placebo, with 30 endometriosis patients in each group ( Signorile et al. 2018 ). However, all participants also increased their fibre and omega-3 consumption and cut out soy, aloe, and oats. There was a significant decrease in reported pain symptoms associated with the anti-inflammatory supplement and a significant reduction in serum inflammatory markers (PGE2, CA-125). However, the potential impact of the dietary regime is unclear. A small double-blind RCT compared endometriosis patients taking an eight-week supplement of omega-3 ( n = 17) versus olive oil ( n = 16) ( Abokhrais et al. 2020 ). Improvements in pelvic pain and QoL scores were seen in both arms; however, there were no significant differences. The use of olive oil in the control arm may explain the results as it has endogenous anti-inflammatory properties ( Cicerale et al. 2012 ), ( Cicerale et al. 2012 ); therefore, a larger trial is now required with an alternative placebo. Another study also found no benefits of omega-3 over 6 months when comparing fish oil ( n = 20) to a placebo ( n = 22) ( Nodler et al. 2020 ). Taken together, these data make it difficult to say whether omega-3 supplementation is beneficial. An earlier study looked at postoperative pain in endometriosis comparing 6 months of hormonal therapy ( n = 77), dietary therapy ( n = 35), or placebo ( n = 110) ( Sesti et al. 2007 ). At the 12-month follow-up, all groups reported lower scores for menstrual pain compared to baseline, though these were significantly lower with hormonal, but not dietary, therapy when compared to placebo. On the other hand, following both therapies, non-menstrual pelvic pain was significantly lower than placebo. A broad range of vitamins, minerals, and nutritional supplements have been associated with inflammation and immunity, a few of which have been investigated for their potential benefits in endometriosis. Vitamins are important for the normal functioning of the immune system, as well as having antioxidant and anti-inflammatory properties ( Carr & Maggini 2017 , Lewis et al. 2019 ). The vitamin D receptor is expressed in reproductive tissues, leading to suggestions it may be involved in the aetiology of endometriosis ( Barnard et al. 2023 ). A recent online survey with 1385 respondents found 381 (27.5%) took a vitamin D supplement ( Deepak Kumar et al. 2023 ). One study compared 12 weeks of vitamin D ( n = 19) to a placebo ( n = 19) and found no difference in reported pelvic pain or dysmenorrhea ( Almassinokiani et al. 2016 ). Conversely, another paper reported 12 weeks of vitamin D treatment resulted in significantly decreased pelvic pain, compared to a placebo ( n = 25 each group) ( Mehdizadehkashi et al. 2021 ). A trial by Nodler et al. , comparing vitamin D ( n = 27), fish oil ( n = 20), and a placebo ( n = 22), found a significant reduction in ‘worst pain’ associated with vitamin D compared to the other groups ( Nodler et al. 2020 ). This study recruited adolescent girls with a mean age of 19.7 – lower than the mean ages of 29.9 and 35.2 in the other two studies. These methodological differences, alongside the small participant numbers, provide a potential explanation for their inconsistency. Other vitamins have also shown promise: women receiving a combination of vitamins C and E ( n = 30) had significantly lower pain scores for dysmenorrhea, dyspareunia, and CPP after 8 weeks, compared to a placebo ( n = 30) ( Amini et al. 2021 ). Curcumin is the active ingredient of turmeric with recognised anti-inflammatory properties ( Tabrizi et al. 2019 ). Studies into its use in several health conditions have found reductions in oestrogen concentrations and proinflammatory mediators, as well as inhibition of angiogenesis ( Piecuch et al. 2022 ). Following 2 months of daily curcumin supplementation, 33 women with endometriosis experienced significant improvements in pelvic pain, dysmenorrhea, and dyspareunia, with a 48% reduction in the number of participants using nonsteroidal anti-inflammatory drugs ( Fadin et al. 2020 ). Probiotic treatment, based on ingestion of specific strain(s) of ‘beneficial’ bacteria, is still a relatively new field, with only a few strains available due to culturing and shelf-life constraints. Furthermore, the complex and diverse nature of the gut microbiome means there is still doubt as to whether the added presence of select strains in the form of supplement probiotics can have a significant impact on dysbiosis, considering their unique and sometimes temporary effects on the gut microbiome ( Leeming et al. 2019 ). Two RCTs have been conducted using probiotics to treat endometriosis. For 8 weeks, 16 women with endometriosis were given a combination of four different Lactobacillus strains: Lactobacillus acidophilus; Lactobacillus plantarum; Lactobacillus fermentum; and Lactobacillus gasseri ; compared to a placebo ( n = 16) ( Khodaverdi et al. 2019 ). Both groups saw decreases in pain scores for CPP and dyspareunia, and for dysmenorrhea the change was significantly greater in the treatment arm. However, all pain scores had increased by the four-week follow-up which, though they had not reverted to baseline, suggests a potential lack of longevity for the probiotic combination. In the second study, 29 women were treated with L. gasseri for 12 weeks and experienced a significant reduction in pain scores compared to placebo ( n = 33) ( Itoh et al. 2011 b ). However, there were no follow-up data and therefore no indication of the long-term impact of the probiotic. The influence of L. gasseri on endometriosis has been investigated in rodent models, with an apparent reduction in lesion growth and activation of NK cells ( Itoh et al. 2011 a , Uchida and Kobayashi 2013 ). These data suggest the necessity for additional larger RCTs to investigate the use of L. gasseri as a treatment, with a focus on duration of response and the impact of repeated courses of supplementation. The use of probiotics as a treatment strategy for IBS has been well documented, with a variety of different strains improving symptom severity, including pain and bloating ( Cryan et al. 2019 , Francavilla et al. 2019 , Wilmes et al. 2021 ). Studies in animal models have demonstrated the alleviation of visceral pain following probiotic treatment ( Zhao et al. 2018 , Zhang et al. 2019 , Li et al. 2019 ), though this was not always replicated ( Huang et al. 2019 ). One study of 118 IBS patients showed treatment with L. gasseri reduced the mean abdominal pain score by 54.2% and attenuated symptoms in 85.0% of participants ( Ait Abdellah et al. 2022 ). However, this was not placebo controlled. Research into the use of probiotics to treat mood disorders provides inconsistent findings ( Taylor and Holscher 2020 ), with varying strains and methodologies used, meaning accurate comparisons are difficult. Lactobacillus rhamnosus reduced depression- and anxiety-like behaviour in mice ( Bravo et al. 2011 ). Interestingly, in a double-blind RCT, pregnant women treated with L. rhamnosus ( n = 212) reported significantly lower post-partum depression and anxiety scores compared to controls ( n = 211) ( Slykerman et al. 2017 ). Antibiotics have a strong influence on the gut, and their use in early life has been shown to have detrimental effects on the gut microbiota in adulthood ( O’Mahony et al. 2014 ). Nonetheless, the preliminary research discussed below raises the potential that symptoms of endometriosis might be treated with antibiotics. The theoretical benefits for treating endometriosis with antibiotics are multifaceted: perturbation of the gut microbiota could improve pain perception and mood disorders by altering signalling within the MGB axis and even influence disease progression if it blunted the immune response and/or reduced deconjugation of steroids by the estrobolome. Additionally, if bacterial infection is proven to have a causal role in lesion development ( Khan et al. 2018 ), then specific antibiotics could provide a defence mechanism against further lesion growth. Some of the studies reviewed below provide support for both these lines of enquiry. The effect of antibiotics on endometriosis was investigated using an endometriotic mouse model treated with a combination of vancomycin, neomycin, metronidazole, and ampicillin for 3 weeks. Antibiotic-treated mice had smaller lesions with fewer proliferative cells and lower concentrations of proinflammatory cytokines, compared to vehicle-treated controls. However, the gut microbiota in the antibiotic-treated mice had decreased α- and β-diversity, dominated by the phylum Proteobacteria, with negligible abundance of Bacteroidetes and Firmicutes ( Chadchan et al. 2019 ). Further analysis of individual treatments with metronidazole or neomycin identified only the former as able to reduce lesion growth ( Chadchan et al. 2019 ). The authors suggested this was due to susceptibility of the Bacteroides genus to metronidazole but not neomycin. Importantly, neomycin is a nonabsorbable antibiotic, meaning its influence is restricted to the gut, whereas metronidazole can move into circulation and even interact with the CNS. Therefore, the explanation for these differences could be a result of metronidazole having activities outside the gut. In a recent study of ovarian endometriosis patients, Muraoka et al. reported 64% had Fusobacterium nucleatum in their endometrium compared to 7% of controls ( n = 42 each group) ( Muraoka et al. 2023 ). They also used a mouse model of endometriosis combined with injection of F. nucleatum and tested the impact of both metronidazole and chloramphenicol. Presence of the bacteria increased lesion size, whereas treatment with antibiotics largely prevented lesion formation and reduced the size of established lesions ( Muraoka et al. 2023 ). This appeared to be due to activation of TGF-β1 signalling by the bacteria. Whilst these are new data that must be replicated by others, they do provide strong evidence that antibiotic treatment might be beneficial in some patients. To date, there has only been one clinical trial investigating the impact of antibiotics on endometriosis patients. In a double-blind RCT, women with stage III/IV endometriosis found 6 months of broad-spectrum antibiotic, clarithromycin ( n = 129), was no more effective than a placebo ( n = 160) for reducing pain after surgical removal of lesions ( Alborzi et al. 2019 ). Additionally, there was no difference in serum levels of inflammatory biomarkers, including TNF-α, between the two groups. Antibiotics, such as rifaximin, improve symptom severity in patients with IBS ( Vicari et al. 2017 ) and decrease visceral pain in animal models ( Aguilera et al. 2015 , Hoban et al. 2016 ). Research into the use of antibiotics for chronic pain with an unknown aetiology is limited, although a recent study into chronic lower back pain found no clinical effect following treatment with amoxicillin ( Bråten et al. 2019 ).

There is increasing, and robust, evidence that the gut microbiota and its metabolites play a key role in the bidirectional signalling pathway between the gut and brain, that can regulate pain, GI symptoms and mood disorders. As these symptoms are common in patients with endometriosis, there is increasing interest in exploring the contribution of the gut microbiota to the manifestation and exacerbation of symptoms, and subsequently whether the use of diet, supplements, probiotics, or antibiotics, all of which may alter the bacterial species in an individual’s microbiome, could be used to improve symptoms and QoL. Whilst data from endometriosis patients are limited, a large body of work on other chronic conditions has highlighted the impact of the microbiome on mechanisms known to be involved in aetiology, pathogenesis, and symptoms associated with endometriosis. These include immune education and regulation; biosynthesis of bacterial metabolites that interact with immune cells and nerves (enteric and CNS); and steroid metabolism/activation ( Cryan et al. 2019 , Shin et al. 2019 , Liu et al. 2022 , Gieryńska et al. 2022 ). If we are to realise the full potential of the MGB axis as a therapeutic target in endometriosis, it will be essential to develop standardised experimental methodology and to undertake large, well-controlled clinical trials, including careful phenotyping of patients regarding diet, symptoms, and disease stage, complemented by an in-depth analysis of microbial diversity, plus inflammatory and metabolic profiling, allowing comparisons to be made between international cohorts – a technique that has led to breakthroughs in the genetics of endometriosis ( Saunders and Horne 2023 ). The benefits of large-scale studies on the MGB would be two-fold. They could offer an opportunity to develop a microbial/biomarker profile that could be used to advise patients on personalised self-management strategies, such as the use of diet and probiotics, alongside pharmaceutical and surgical approaches ( Fig. 2B ). Secondly, if studies on the putative role of bacterial infection in disease progression can be replicated in diverse populations, this could provide a rationale for testing antibiotic treatments for endometriosis. However, these must be approached with caution as antibiotics may also upset the balance of beneficial versus dysbiotic resident gut microbiota. To summarise, the impact of the gut microbiota on both the aetiology and symptomology of endometriosis is a rapidly expanding field, with some promising avenues for future research focussed on its manipulation to improve patients’ QoL.

Endometriosis is a chronic inflammatory pain condition believed to impact the lives of one in ten women and those assigned female at birth ( Horne & Missmer 2022 ). Whilst endometriosis is defined by the presence of endometrial-like tissue growing as ‘lesions’ outside the uterus, patients can present with a range of seemingly unrelated symptoms, leading to new, sometimes controversial, reframing of endometriosis as a body-wide disorder ( Hickey et al. 2020 ). For example, whilst pain (cyclical or constant) and infertility are common symptoms, patients with endometriosis often present at clinics reporting a range of other problems. These include mood disorders (anxiety, depression) and symptoms affecting their digestive system, such as abdominal bloating and those mirroring irritable bowel syndrome (IBS) ( Saunders & Horne 2021 , Saunders & Horne 2023 ). The gut microbiota is the collection of bacteria, viruses, and archaea within the gastrointestinal (GI) tract which produce essential metabolites, hormones, and neurotransmitters. Evidence for the impact of diet on the gut microbiota and the importance of the microbiome to general health is rapidly expanding ( Cryan et al. 2019 ). Notably, metabolic products of the microbiota can affect the immune system and influence inflammation, leading to increased interest in how changes in the microbiota could impact on the severity of symptoms in disorders associated with aberrant immune responses, such as endometriosis ( Saunders & Horne 2021 ). A breakthrough in our understanding of the importance of the bidirectionality in signalling between the gut and brain has been informed by results of studies on symptoms including stress, pain, and mood disorders ( Rea et al. 2019 , Wilmes et al. 2021 ). Studies such as these have linked gut dysbiosis (an ‘imbalance’ in the gut microbial community) to the severity of symptoms and vice versa. Research into the relationship between the gut microbiota and endometriosis is still limited in scope, with a focus on endometriotic lesion development and disease progression, rather than its potential influence on symptomology ( Chadchan et al. 2023 , Wei et al. 2023 ). In this narrative review, we will provide a brief overview of the symptoms of endometriosis that may be altered by signalling within the microbiota–gut–brain (MGB) axis, briefly consider the existing primary research exploring the function of the gut microbiota in endometriosis lesion development, which has been explored in depth elsewhere ( Talwar et al. 2022 , Chadchan et al. 2023 ), before focussing on the potential role of dialogue between the gut microbiota, inflammatory response, and pain pathways in promoting/mitigating the body-wide symptoms associated with the disorder. To increase the range of evidence we will summarise findings from other chronic inflammatory pain conditions, to demonstrate the potential mechanisms of interaction between the gut microbiota and key symptoms of endometriosis: pain and inflammation; GI symptoms; and mood disorders. Finally, we will discuss promising therapeutic opportunities, such as dietary intervention, supplements, probiotics, and antibiotics, to alleviate symptoms via manipulation of the gut microbiota, providing exciting opportunities for future research with the priority of improving symptomology and patients’ quality of life (QoL), some of which have been conducted in cohorts of endometriosis patients. Notably, as our understanding of the role(s) of other microbiomes has increased, researchers have also begun to explore whether the vaginal, endometrial, oral, and peritoneal microbiomes are altered in endometriosis patients, but results to date are variable. For the purposes of the current narrative review, we have focussed on the evidence that the gut microbiome, acting as part of a gut–brain bidirectional signalling system, can impact on symptoms of endometriosis, as well as evidence from studies on endometriosis patients and other disorders often co-morbid with endometriosis, that regulation of the microbiome could be a target for symptom relief. A comprehensive literature review identified articles and reviews through PubMed by searching for specific keywords including endometriosis, (chronic) pain, gut microbiome/metabolites, diet, supplements, IBS, mood, and other relevant related terms.

FHY, PTKS, and SO have no conflicting interests. AWH is a Co-Editor-in-Chief of Reproduction and Fertility . AWH was not involved in the review or editorial process for this paper, on which he is listed as an author. AWH’s institution (The University of Edinburgh) has received payment for consultancy and grant funding from Roche Diagnostics to assist in the early development of a possible blood diagnostic biomarker for endometriosis. AWH’s institution has received payment for consultancy fees from Gesynta and Joii. AWH has received payment for a presentation from Theramex. AWH’s institution has received grant funding from the MRC, NIHR, CSO, and Wellbeing of Women for endometriosis research. AWH is listed as a co-inventor on a UK Patent Application (No. 2217921.2).

The exact cause of endometriosis is currently undetermined, although evidence shows genetic changes may increase the risk of developing the disorder ( Zondervan et al. 2018 , Saunders 2022 ). A defining hallmark of endometriosis is considered the presence of ‘lesions’ resembling endometrial tissue, most commonly detected in the peritoneal cavity ( Saunders & Horne 2021 ). Our understanding of the mechanisms resulting in the establishment and survival of lesions has evolved from the theory of retrograde menstruation – the concept of menstrual debris entering the pelvic cavity via the Fallopian tubes during menstruation and implanting into the peritoneum, complemented by other routes including transfer via the vasculature ( Yovich et al. 2020 ). In the past 20 years, evaluation of clinical samples and preclinical models have provided evidence to support a role for steroid hormone regulation of cell proliferation, inflammation, and neuroangiogenesis, with nerve projections connecting the lesions to the central nervous system (CNS), promoting the survival of lesion tissue and the development of pain symptoms (extensively reviewed elsewhere) ( Zondervan et al. 2018 , Zondervan et al. 2020 , Saunders & Horne 2021 ). Current therapeutic options for people with endometriosis are limited ( Saunders & Horne 2021 ). Surgical removal of lesions, hormonal therapies, and analgesics are the most common strategies, all of which can be associated with detrimental side effects, with over 50% of patients having repeat surgeries within 5 years ( Saraswat et al. 2018 ). Many endometriosis patients have reported trialling a selection of self-management strategies, including dietary interventions and dietary supplements, likely in response to the combination of diagnostic delays and the narrow range of therapeutic options currently available. An altered immune response and changes in immune-cell phenotype are reported in patients with endometriosis. Table 1 summarises endometriosis-associated changes in key immune cells. For example, several studies have shown increased infiltration of neutrophils and macrophages in the peritoneal fluid and lesions, with altered M1–M2 macrophage polarisation, alongside suppressed natural killer (NK) cell activity and increased numbers of T helper 17 cells (Th17) ( Symons et al. 2018 ). Importantly, these changes are associated with increased levels of proinflammatory cytokines in the lesion microenvironment, whose downstream effects include increased inflammation, angiogenesis, and cell proliferation, all of which contribute to survival/growth of lesions ( Herington et al. 2011 , Symons et al. 2018 ). Studies such as these have supported the argument that endometriosis should be considered as an inflammatory disorder, and therapies sought to blunt/normalise these responses (see Saunders & Horne 2021 ). Table 1 Summary of immune-cell changes in peritoneal fluid of endometriosis patients. Immune cell type Cellular changes Cytokine and chemokine production Downstream effects Reference Neutrophils ↑ PF infiltration; ↓ Apoptosis ↑ TNF-α; ↑ VEGF ↑ Inflammation; ↑ Angiogenesis Symons et al. (2018) Macrophages ↑ Infiltration in PF and lesion microenvironment; ↓ Phagocytosis; ↑ Activation of transcription factor NF- κ B; Co-localisation with nerve fibres; altered M1–M2 polarisation ↑ IL-1β; ↑ IL-6; ↑ IL-10; ↑ TNF-α; ↑ TGF-β; ↑VEGF ↑ Inflammation; ↑ Angiogenesis; ↑ Stromal cell proliferation and invasiveness Herington et al. (2011) , Symons et al. (2018) NK cells ↓ Activity ↑ TNF-α ↓ Cytotoxicity; ↑ Inflammation Herington et al. (2011) , Symons et al. (2018) T cells ↑ Th17:Treg ratio ↑ IL-17 leads to: ↑ IL-8 and ↑ COX-2 ↑ Inflammation; ↑ Angiogenesis; ↑ Stromal cell proliferation; attracts and activates neutrophils Symons et al. (2018) NK, natural killer. Summary of immune-cell changes in peritoneal fluid of endometriosis patients. NK, natural killer. Individuals with endometriosis report a variety of different types of pain ( Fig. 1A ). Mechanisms that initiate or promote endometriosis-associated pain symptoms remain the subject of intense research activity, with some evidence suggesting the role of nerve growth within lesions, which may occur in parallel with angiogenesis ( Asante & Taylor 2011 ). Many studies have reported a lack of correlation between pain intensity and the number, location, or type of lesions ( Vercellini et al. 2007 ), indicating other mechanisms also contribute to pain experience in addition to the extent/presence of lesion neurogenesis ( Fig. 1B ). Figure 1 Pain mechanisms in endometriosis . A global assessment of pain mechanisms that may differ in endometriosis patients has identified alternations in chronic pain pathways associated with hypersensitivity to stimuli. (A) Commonly reported pain symptoms in endometriosis include dysmenorrhea (pain during menstruation), dyspareunia (pain during sex), and pain on defecation and urination ( Saunders and Horne, 2021 ). (B) Inflammation and innervation of endometrial lesions contribute to pain experience via a connection with the CNS and recruitment of immune cells ( Tokushige et al. 2010 , Liu et al. 2012 ). (C) Differences have also been identified in the brains of endometriosis patients, with reduced grey matter volume in the left thalamus (pale yellow), right putamen (orange), left cingulate gyrus (dark yellow), and right insula of those with chronic pelvic pain (CPP) ( As-Sanie et al. 2012 ), as well as changes identified in the extent and location of activation in response to painful stimuli in those with dysmenorrhea ( Tu et al. 2009 ). (D) Central sensitisation occurs when pain pathways are persistently amplified, enabling them to be ‘learnt’ by the CNS, thus reducing the level of stimuli required to trigger pain, eventually leading to sensitivity to non-noxious stimuli and at sites distal to the inflammation ( Berkley et al. 2005 , Neziri et al. 2010 ). In CPP, this manifests as visceral hypersensitivity ( Vincent et al. 2011 , Kaya et al. 2013 ) and has been demonstrated in endometriosis ( Weiwei et al. 2010 , Aredo et al. 2017 , Zheng et al. 2019 ). The red arrow represents the bidirectional relationship between pain pathways triggered by local inflammation and immune responses in the peritoneum (or alternative lesion locations) and the structural and learnt changes in the brain which result in exacerbated pain experiences. Created with BioRender.com. Pain mechanisms in endometriosis . A global assessment of pain mechanisms that may differ in endometriosis patients has identified alternations in chronic pain pathways associated with hypersensitivity to stimuli. (A) Commonly reported pain symptoms in endometriosis include dysmenorrhea (pain during menstruation), dyspareunia (pain during sex), and pain on defecation and urination ( Saunders and Horne, 2021 ). (B) Inflammation and innervation of endometrial lesions contribute to pain experience via a connection with the CNS and recruitment of immune cells ( Tokushige et al. 2010 , Liu et al. 2012 ). (C) Differences have also been identified in the brains of endometriosis patients, with reduced grey matter volume in the left thalamus (pale yellow), right putamen (orange), left cingulate gyrus (dark yellow), and right insula of those with chronic pelvic pain (CPP) ( As-Sanie et al. 2012 ), as well as changes identified in the extent and location of activation in response to painful stimuli in those with dysmenorrhea ( Tu et al. 2009 ). (D) Central sensitisation occurs when pain pathways are persistently amplified, enabling them to be ‘learnt’ by the CNS, thus reducing the level of stimuli required to trigger pain, eventually leading to sensitivity to non-noxious stimuli and at sites distal to the inflammation ( Berkley et al. 2005 , Neziri et al. 2010 ). In CPP, this manifests as visceral hypersensitivity ( Vincent et al. 2011 , Kaya et al. 2013 ) and has been demonstrated in endometriosis ( Weiwei et al. 2010 , Aredo et al. 2017 , Zheng et al. 2019 ). The red arrow represents the bidirectional relationship between pain pathways triggered by local inflammation and immune responses in the peritoneum (or alternative lesion locations) and the structural and learnt changes in the brain which result in exacerbated pain experiences. Created with BioRender.com. Reduced pain thresholds, along with increased activity in several brain regions associated with pain perception, have been identified in both cynomolgus monkeys with naturally occurring endometriosis ( Yano et al. 2019 ) and a rat model of the disease ( Zheng et al. 2020 ). Coupled with the fact that the diagnostic delay experienced by individuals with endometriosis may increase the likelihood of developing chronic/persistent pain that is resistant to standard therapies, it is clear that new approaches to pain management are required. In women with chronic pelvic pain (CPP), including those with endometriosis, changes in brain structures have been detected that appear consistent with an amplified/abnormal response to stimuli ( Fig. 1C ) ( Brawn et al. 2014 ). Additional data have detected altered brain chemistry associated with these physical changes, consistent with amplification of pain signals and so-called central sensitisation augmenting signals from peripheral enteric nerves ( Fig. 1D ) ( As-Sanie et al. 2016 ). There is increased awareness of the impact of GI symptoms on the wellbeing of endometriosis patients ( Maroun et al. 2009 ). A large study found women with endometriosis had increased risk of inflammatory bowel disease, Crohn’s disease, and ulcerative colitis ( Jess et al. 2012 ). A two-fold higher incidence of IBS in individuals with endometriosis, compared to the general population ( Chiaffarino et al. 2021 , Aupetit et al. 2022 ), suggests overlapping mechanisms between the conditions, recently corroborated by evidence of shared genetic risk factors ( Yang et al. 2023 ). IBS is characterised by chronic gut inflammation, bloating, and visceral pain – symptoms common in patients with a diagnosis of endometriosis ( Saunders & Horne 2021 , Deepak Kumar et al. 2023 ). Crucially, patients highlight abdominal bloating as an important topic of unmet need ( Horne et al. 2017 ). Women with endometriosis have a higher incidence of psychiatric comorbidities and mood disorders, including depression and anxiety ( Gete et al. 2023 ). Mechanisms to explain these comorbidities are likely to involve the hypothalamic–pituitary–adrenal (HPA) axis, a neuroendocrine signalling pathway with a critical role in hormone regulation ( Oyola & Handa 2017 ), whose dysregulation is involved in mood disorders ( Bao & Swaab 2019 ). When activated, the HPA axis causes the adrenal cortex to release the glucocorticoid, cortisol, making it the primary coordinator of the stress response. Chronic pain can lead to dysregulation of the HPA axis, and this has been demonstrated in a range of inflammatory conditions ( Kuehl et al. 2010 ). A small study recently found an association between a dysregulated HPA axis and menstrual pain severity in endometriosis patients ( Ortiz et al. 2020 ). In a study of 26 women with endometriosis and CPP, physical and psychological therapy normalised cortisol levels, reduced perceived stress, and improved physical functioning ( Friggi Sebe Petrelluzzi et al. 2012 ). There is an increasing appreciation of the role of the stress response and cortisol in modulating the MGB axis at multiple levels, including gut function and composition of the gut microbiota. A recent comprehensive review on the connections between the HPA and MGB axis was published by Rusch et al. (2023) .