{"paper_id":"69e06971-7c87-42bb-b832-63a8d69a6add","body_text":"Endometriosis is a chronic, hormone-dependent neuroinflammatory disease where endometrial-like tissue grows outside the uterus. It affects about 10% of women and those assigned female at birth of reproductive age worldwide, and currently, there is no cure ( As-Sanie  et al.  2025 ). The four subtypes of endometriosis – superficial peritoneal (about 80% of cases), deep, ovarian (endometriomas), and extrapelvic – can occur alone or together. Superficial peritoneal endometriosis (SPE) is typically located on the surface of abdominal or pelvic organs and the pelvic wall. Deep endometriosis invades pelvic tissues or organs, such as the bowel or bladder. Ovarian endometriosis (OE), or endometriomas are cysts lined by endometrial tissue within the ovary. Extrapelvic endometriosis involves lesions outside the pelvis and can affect many organs, including the diaphragm, thoracic organs, or even the brain.\nA frequent symptom for up to half of women with endometriosis is infertility. Alone, infertility has detrimental impacts on mental health and quality of life for those wanting to conceive, which is further increased by endometriosis ( Mori  et al.  2024 ). Clinically, both endometriosis and infertility suffer delays with diagnosis and a lack of effective, accessible, treatment options.\nIn international guidelines for endometriosis, it is recommended that women with SPE consider surgery to remove SPE lesions based on evidence that this may improve spontaneous pregnancy rates within the first 12 months post-surgery ( NICE 2017 ,  Becker  et al.  2022 ). In many UK centres, the waitlist for a diagnostic laparoscopy for endometriosis is two or more years ( Endometriosis UK 2024 ,  Royal College of Obstetricians and Gynaecologists 2024 ). For these women, assisted reproductive technologies (e.g. IVF) provide one potential solution, although this is also subject to access to NHS provision (in the UK) and is costly. A 2023 study found that 3.2 million women of reproductive age in England had no or limited access to a fertility clinic in their area of residence, while the highest household income areas had the best access to fertility clinics ( Jones  et al.  2023 ). Importantly, there is a group of women who are only diagnosed with endometriosis when they attend fertility clinics when seeking to become pregnant and the majority of these are cases of SPE ( Van Gestel  et al.  2024 ). If these women are of advanced maternal age, it is unlikely that they will want to wait several years for a laparoscopy to remove the endometriosis in the hope that it may enable them to conceive spontaneously. To provide better and more equitable options for these women, we must better our understanding of the mechanisms behind how SPE contributes to infertility.\nThe endometriosis fertility index (EFI) predicts post-surgery pregnancy success for women with endometriosis based on the surgeon’s assessment of damage to Fallopian tubes, fimbria and ovaries, endometriosis staging score, patient age, and pregnancy history ( Adamson & Pasta 2010 ). Since its inception, it has been validated in a variety of settings and shown to accurately predict non-IVF pregnancy success for women with endometriosis post-surgery ( Tomassetti  et al.  2013 ,  Garavaglia  et al.  2015 ). It has also recently been shown that it can be accurately completed without surgical intervention ( Tomassetti  et al.  2021 a ). However, to date, its utility has been limited to testing on combined datasets containing all subtypes of endometriosis.\nIn clinics offering assisted reproductive technologies, circulating levels of anti-Müllerian hormone (AMH) are typically measured to predict the capacity of the ovary to respond to hormone stimulation and production of viable oocytes. Notably, AMH is an indirect marker of the quiescent pool of primordial follicles that make up the ovarian reserve and is produced by the proliferating granulosa cells surrounding an oocyte during development, prior to ovulation. AMH has previously been measured in cohorts of endometriosis patients, most commonly those with ovarian endometrioma. In these cases, AMH often decreases after surgical removal of an endometrioma via cystectomy ( Wang  et al.  2020 ,  Muraoka  et al.  2021 ,  Sarbazi  et al.  2021 ,  Fakehi  et al.  2022 ,  Mansouri  et al.  2022 ,  Shi  et al.  2022 ,  Tang & Li 2022 ,  Crestani  et al.  2023 ). Some reports suggest AMH levels increase again during follow-up periods; however, they never return to pre-surgery levels ( Kostrzewa  et al.  2019 ,  Sadullayev & Medvediev 2022 ). To date, only one study has reported AMH levels in a cohort of women with SPE, demonstrating no change in AMH compared to age-matched, population controls ( n  = 62 in each group) ( Lessans  et al.  2024 ). In another study, 40 expectantly managed women (no surgical intervention to treat their ovarian endometriosis) were reported to have AMH levels that declined faster than control women without endometriosis ( Kasapoglu  et al.  2018 ), providing the strongest evidence to date that there is a mechanism by which endometriosis drives reductions in ovarian reserve and contributes to infertility. In the current study, serum AMH was measured in a cohort of women with SPE, and levels were compared to both women without endometriosis (confirmed the absence of lesions by laparoscopy) and a group of women also with an endometrioma (SPE + OE), to investigate how AMH may be altered by SPE. Additionally, we measured systemic and peritoneal inflammation as potential mechanisms driving endometriosis-associated infertility and associated changes to AMH levels.\n\nParticipants were recruited to an ongoing observational study in South-East Scotland between October 2015 and July 2023 from the Royal Infirmary of Edinburgh (NHS Lothian, REC 20/LO/1298) and gave informed consent to be involved in the study and for biospecimens to be collected during surgery.\nAll study participants ( n  = 105) were selected from our Biobank resource ( n  = 621) and had undergone surgery to investigate symptoms suggestive of endometriosis (most commonly, pelvic pain) for the first time. All had undergone pre-operative ultrasound imaging and were carefully assessed at the time of surgery to exclude deep disease. Deep endometriosis was defined according to the International Terminology of Endometriosis, 2021 ( Tomassetti  et al.  2021 b ). A venous blood sample was taken prior to surgery, and where possible, a peritoneal fluid sample collected during surgery ( n  = 38).\nBiospecimens were processed in accordance with WERF EPHect protocols ( Rahmioglu  et al.  2014 ). Serum was collected from whole blood samples via centrifugation at 2500  g  for 10 min at 4°C and aliquots stored at −80°C. Peritoneal fluid was similarly processed by centrifugation at 900  g  for 5 min at 4°C and aliquots stored at −80°C.\nParticipants were categorised by an endometriosis surgeon according to surgical visual assessment and histological confirmation: no endometriosis (no lesions observed; no endo:  n  = 39), SPE (only superficial lesions present, SPE:  n  = 43), or SPE with ovarian endometriosis (SPE + OE:  n  = 23) ( Table 1 ). Exclusion criteria included participants currently pregnant or breastfeeding, known reproductive malignancies, previous history of PCOS, and previous history of endometriosis, specifically endometrioma, or visualisation of deep lesions. Current use of hormonal treatments to manage endometriosis symptoms was used to stratify samples. Hormone treatments included the combined oral contraceptive pill (COCP), progesterone-only pill (POP), Depo-ProveraTM, levonorgestrel intrauterine system (LNG-IUS), NexplanonTM implant, norethisterone, or a contraceptive patch. Their prevalence of use in this cohort is summarised in  Table 1 . The EFI was calculated for each participant as previously described ( Adamson & Pasta 2010 ), by the operating surgeon upon completion of the surgery.\nParticipant hormone treatment use. Data are presented as  n  (%).\nNo endo, no endometriosis lesions visualised at laparoscopy; SPE, superficial peritoneal endometriosis; SPE + OE, superficial peritoneal and ovarian endometriosis; COCP, combined oral contraceptive pill; POP, progesterone-only pill; LNG-IUS, levonorgestrel intrauterine system; others include norethisterone, patch, and Nexplanon implant.\nSerum samples were diluted 1:10 and assayed in duplicate using the human picoAMH ELISA (AL-124-r, Ansh Labs, USA) according to manufacturer’s directions. Absorbance was read using a Clariostar plate reader (BMG Labtech, Germany) at 450 nm.\nCirculating inflammatory cytokines were quantified using custom 13-plex Discovery Luminex (Bio-Techne, USA) to detect CD163, IL-1β, IL-8/CXCL8, IL-23, MIF, TGF-α, β-NGF, IL-1α, IL-6, IL-17A, IL-33, NRG1, and TNF-α according to manufacturer’s instructions. Serum samples were diluted 1:2, and peritoneal fluid diluted 1:10. All samples were assayed in duplicate. Luminex detection was completed using a Luminex xMAP INTELLIFLEX analyser using LX200 low-sensitivity settings.\nFor measurement of prostaglandin E2 levels, serum and peritoneal fluid were diluted 1:2 and assayed in duplicate according to manufacturer’s instructions (500141, Cayman Chemical, USA). Absorbance was read using a Clariostar plate reader (BMG Labtech, Germany) at 420 nm.\nSerum and peritoneal fluid samples were diluted 1:2 and assayed in duplicated according to manufacturer’s instructions (A75809,  antibodies.com , UK). Absorbance was read using a Clariostar plate reader (BMG Labtech) at 450 nm.\nSerum and peritoneal fluid samples were diluted 1:100 and assayed in duplicate according to manufacturer’s instruction (DCL180B, R&D Systems, USA). Absorbance was read using a Clariostar plate reader (BMG Labtech) at 450 nm.\nAnalyses were completed in GraphPad Prism 10. ELISA results were interpolated from the standard curve generated from each assay’s unique standards and corresponding blank-corrected absorbance values. Luminex analyses were completed using Quantist software (BioTechne, USA). Data were assessed for normality using Shapiro–Wilk normality test and the appropriate statistical test chosen accordingly. Linear regression modelling was performed using RStudio (version 2024.12.1 + 563) using the lm() function within the ‘stats’ package including a two-way interaction term to assess the relationship between endometriosis subtypes and hormone use, with age as a covariate. Comparisons of two groups used Welch’s  t -test for normal distribution or Mann–Whitney U test for non-normally distributed data. Comparison of three groups utilised one-way ANOVA with Tukey’s  post hoc  test for normal distribution or Kruskal–Wallis with Dunn’s post hoc test for non-normal distribution.\n\nDemographic information, including age and body mass index (BMI) of participants and their use of hormone treatments, is summarised in  Tables 1  and  2 . There was no difference in age between the no endometriosis and SPE groups, but the SPE + OE women were older than both no endometriosis and SPE women ( Table 2 ). BMI was comparable between all groups ( Table 2 ). Between 45 and 85% of women had no history of being diagnosed with infertility, and 43–60% of women in each group had been pregnant in the past ( Table 2 ). Mean EFI scores for women with SPE + OE were 6.95, which was significantly lower than the 8.63 for those with SPE ( Table 3 ). Serum AMH levels for SPE + OE were lower than no endometriosis women ( Table 3 ); however, linear regression modelling was performed to allow AMH levels between groups to be compared while accounting for confounding variables such as age and use of hormone treatments ( Table 4 ). For those not using hormone treatments, both SPE and SPE + OE groups had significantly lower AMH levels compared to the no endometriosis control when controlling for participant age (estimate: −0.64 and −0.77,  P  = 0.009 and  P  = 0.002, respectively;  Table 4 ). In contrast, when accounting for age in those participants using hormones, the observed changes in AMH were mitigated as demonstrated by positive estimate values (0.57 and 0.96, respectively;  Table 4 ), suggesting hormone use may influence differences observed in AMH.\nParticipant demographic information. Data are presented as  n  or as mean ± SD.\nNo endo, no endometriosis lesions visualised at laparoscopy; SPE, superficial peritoneal endometriosis; SPE + OE, superficial peritoneal and ovarian endometriosis; BMI, body mass index; SD, standard deviation.\nP  < 0.05 vs no endo.\nP  < 0.05 vs SPE.\nEndometriosis fertility index and levels of circulating AMH.\nNo endo, no endometriosis lesions visualised at laparoscopy; SPE, superficial peritoneal endometriosis; SPE + OE, superficial peritoneal and ovarian endometriosis; EFI, endometriosis fertility index; AMH, anti-Müllerian hormone; SD, standard deviation; IQR, interquartile range.\nP  < 0.001 vs SPE.\nP  < 0.05 vs No endo, no hormones.\nP  < 0.001 vs No endo, no hormones.\nLinear regression model of AMH. Relationship between AMH and SPE/SPE + OE vs control, with age or age and hormone use taken into consideration.\nSE, standard error; ns, not significant.\nP  < 0.05.\nP  < 0.01.\nTo assess differential inflammation in women with endometriosis and those without, inflammatory cytokines and proteins were quantified using multiplex Luminex and ELISA on the same serum samples used to measure AMH and are summarised in  Table 5 . When groups were separated by their use of hormones, the levels of circulating cytokines were not significantly changed between SPE and no endometriosis groups ( Table 5 ), for the majority of cytokines measured. The two exceptions were IL-17 ( Fig. 1A ) and TNF-α ( Fig. 1B ), both of which were significantly increased in women with SPE not using hormones ( n  = 9) compared to no endometriosis controls not using hormones ( n  = 6); notably, these differences were not significant for women using hormone therapies. Correlation analysis was performed to investigate the relationship between levels of these cytokines and respective AMH levels. No significant relationships were observed (Supplementary information (see section on  Supplementary materials  given at the end of the article)).\nCytokine levels in the circulation of women with and without superficial endometriosis. Data are presented as median (IQR).\nNo endo, no endometriosis lesions visualised at laparoscopy; SPE, superficial peritoneal endometriosis; IQR, interquartile range.\nP  < 0.05 vs no endo.\nWomen with SPE have elevated inflammatory cytokines, seen only in women not using hormone treatments. Levels of IL-17 (A) and TNF-α (B) in the circulation and IL-23 (C) in the peritoneal fluid are elevated in women with SPE not using hormones, compared to no endo. Data are mean ± SD,  n  = 6–13/group, * P  < 0.05, ** P  < 0.01. Shapiro–Wilk test for normality, one-way ANOVA with Tukey’s  post hoc  test (A and B), or Welch’s  t -test (C) for significance.\nIn a subset of women ( n  = 14), pelvic peritoneal fluid samples from those not using hormones were used to assess inflammation levels in the local pelvic peritoneal microenvironment. Concentrations of peritoneal fluid inflammatory markers are summarised in  Table 6 . While most factors were detected, variation in levels meant that most did not reach statistical significance, including those previously shown to be elevated in serum. Interestingly, one exception was IL-23 that was significantly elevated in women with SPE not using hormones compared to those without endometriosis ( Fig. 1C ).\nCytokine levels in the peritoneal fluid from women with and without superficial endometriosis. Data are presented as median (IQR).\nNo endo, no endometriosis lesions visualised at laparoscopy; SPE, superficial peritoneal endometriosis; IQR, interquartile range.\nP  < 0.05 vs no endo.\n\nThe mechanisms underlying endometriosis-associated infertility are poorly understood and rarely characterised by subtype of endometriosis. Here, we demonstrate for the first time that serum AMH is decreased in women who only have lesions identified as the SPE subtype, compared to women confirmed to have no endometriosis lesions detected during surgery. Interestingly, this finding was associated with elevated pro-inflammatory cytokine levels in the same biospecimens, suggesting that inflammation is one mechanism contributing to endometriosis-associated delays in conception in women with SPE.\nNotably, in our cohort, we found the EFI was lower for women with SPE + OE compared to women with SPE, suggesting a lower probability of future pregnancy success for women with SPE + OE even though some had already achieved a pregnancy. Since its inception, the EFI has been validated extensively and can be reliably and routinely calculated to predict pregnancy success for non-ART conceptions in endometriosis patients ( Adamson & Pasta 2010 ,  Tomassetti  et al.  2013 ,  Vesali  et al.  2020 ). A recent population study from Finland demonstrated that women with endometriosis have a lower overall fertility rate compared to women without endometriosis ( Tuominen  et al.  2025 ). Notably, their data were from a historical population-based cohort study (1998–2012) including 18,320 fertile-aged women with a surgical diagnosis of endometriosis, 5,786 of whom had a diagnosis of SPE alone. They reported lower fertility rates in the endometriosis group compared to a reference group. However, consistent with our data, 68% did achieve a pregnancy during follow-up. While registry studies provide large datasets, they are not able to provide insights into whether their lower fertility rate is due to an underlying biological reason or less sexual activity due to debilitating symptoms associated with endometriosis. Accurate comparison of infertility can be challenging due to the reliance on self-reported data in many cases. Our study starts to bridge this gap by assessing EFI and serum AMH in the same women with SPE alone, or SPE with OE. As EFI is not clinically applicable to patients with confirmed absence of endometriosis due to the design of the scoring system, a comparison cannot be made here between EFI scores for women with and without endometriosis, which is where the strength of our AMH data is highlighted. This, in conjunction with the cytokine data presented in this study, supports an impact on ovarian reserve and possibly fertility in the presence of SPE and its associated inflammation.\nIn this study, we focused on SPE as this is the most common subtype of endometriosis and one that receives less attention in fertility studies than ovarian or deep disease. We also focus on the impact on ovarian function and fertility, rather than the endometrium, as this has already been the subject of several previous studies (reviewed,  Griffiths  et al.  (2024) ). A strength of our cohort is the precise phenotyping of the participants and their disease subtypes. Previous studies have assessed AMH or fertility in groups of women with endometriosis, irrespective of the subtype (reviewed,  Griffiths  et al.  (2024) ). Our cohort was stratified into groups based on their surgical findings and medical history to confirm the absence of any previous diagnosis of an endometrioma or deep lesion that could confound the AMH results. Moreover, the control ‘no endo’ population has advantages over previously reported studies as women in this group were confirmed as having no observable endometriosis lesions after undergoing an identical diagnostic laparoscopy to those confirmed to have endometriosis lesions. However, our control group may also be a limitation as the impact of chronic pelvic pain without endometriosis on inflammation and AMH is not well characterised and the differences observed in the present study may actually underestimate the true impact of endometriosis on AMH and inflammatory cytokines. In contrast, previous studies investigating endometriosis-associated infertility have either utilised age-matched controls where the presence of asymptomatic and undiagnosed endometriosis has not been ruled out surgically, or women attending fertility clinics ( Lessans  et al.  2024 ). In both cases, there are likely to be other factors contributing to infertility, and as such, neither is an ideal control (‘fertile’) group to compare to the fertility in a group of women with endometriosis. This concern is also based on reports describing that almost half of women presenting to fertility clinics with unexplained infertility are subsequently found to have endometriosis, most of which have SPE ( Van Gestel  et al.  2024 ).\nInterestingly, changes in AMH levels were influenced by hormone use and a significant reduction in AMH levels for women with SPE was not seen when hormone use was taken into account in linear regression modelling. The ability of hormone treatments to alter AMH has been reported previously ( Bernardi  et al.  2021 ,  Hariton  et al.  2021 ,  Nelson  et al.  2023 ); however, the influence of hormone use on AMH levels for women who also have endometriosis has never been investigated to the best of our knowledge. It is possible the lack of change here is also driven by a smaller sample size once participants are stratified by their use of hormones. Therefore, studies in a larger sample size are warranted to confirm and validate these findings. There is evidence that women seeking IVF with AMH lower than 1.1 ng/mL have a lower chance of fertilisation and achieve a clinical pregnancy compared to those with AMH greater than 1.1 ng/mL ( Vijay  et al.  2022 ). Alternatively, data in otherwise healthy individuals suggest that AMH is not a good predictor of reproductive success ( Hawkins Bressler & Steiner 2018 ). Whether the reductions in AMH observed here have clinically meaningful implications in terms of chances of conception is unknown and would require dedicated studies in endometriosis patients to untangle.\nTo investigate a potential mechanism by which the pelvic microenvironment may contribute to depleted AMH, the associated inflammatory environment in the peritoneal fluid recovered at the time of surgery from the same women was characterised. From the panel of cytokines investigated here, which were selected based on previous works and evidence in the literature, only IL-17 and TNF-α and IL-23 were found to be elevated with SPE in the serum and peritoneal fluid, respectively. These results align with previous works demonstrating a dysregulated IL-17/IL-23 axis and elevated TNF-α levels with endometriosis ( Harada  et al.  1997 ,  Sisnett  et al.  2024 ). For example,  Sisnett  et al.  (2024)  showed elevated IL-23 in the plasma of patients with endometriosis ( n  = 13) compared to healthy, fertile controls ( n  = 19) ( Sisnett  et al.  2024 ). Elevated levels of TNF-α ( Harada  et al.  1997 ) and IL-17 ( Zhang  et al.  2005 ) in the peritoneal fluid of women with endometriosis have been previously reported. A strength of our study over previous work is that our subgroup analysis focussed on SPE or SPE + OE, rather than a general endometriosis versus control comparison. Interestingly,  Zhang  et al.  (2005)  also reported an even greater increase in IL-17 levels for patients with endometriosis and a diagnosis of infertility compared to endometriosis alone ( Zhang  et al.  2005 ). Due to limited sample size, we were unable to complete the same analyses. Taken together with our study showing IL-17 is elevated and AMH is reduced in women with SPE, it would be interesting for future studies to attempt to disentangle the relationship between IL-17 and endometriosis pathogenesis and endometriosis-associated reductions to ovarian reserve and infertility.\nInterestingly, differences in circulating IL-17 and TNF-α were unique to the participants not using hormone treatments. Interestingly, this appeared to be due to a similar elevation in these cytokines in the no endometriosis group using hormone treatments. While enhanced inflammatory responses to acute stressors have been reported previously in women using hormone treatments ( Larsen  et al.  2020 ,  Mengelkoch  et al.  2024 ), to the best of our knowledge, this has not been assessed in women with a more chronic inflammatory profile, as with endometriosis. Moreover, it is unclear what the consequences may be on levels of pro-inflammatory cytokines and subsequently AMH levels for women with SPE or SPE + OE if they stop taking hormone treatments to attempt to conceive. Here, our sample size was too limited to investigate the impacts of specific classes of hormone treatments on inflammation, although it would be interesting in future to determine differences in the impact of oestrogen-containing or progestin-only therapies. These are all relevant avenues for future work characterising how inflammation contributes to reduced ovarian reserve and endometriosis-associated infertility.\nSurprisingly, in our study, no differences were observed in well-established factors known to play a role in endometriosis-associated symptoms, including PGE2 ( Rakhila  et al.  2013 ) and IL-8 ( Jørgensen  et al.  2017 ). This may be due to differences in the biospecimens investigated with the  Rakhila  et al. ’s (2013)  study focused on levels in ectopic lesion tissue. The study from  Jørgensen  et al.  (2017)  measured levels of peritoneal fluid and reported significantly higher levels of IL-8 in 56 endometriosis patients compared to 38 without endometriosis attending a fertility clinic in biospecimens recovered during the luteal/secretory phase. In our study, we had only 8 samples from women not on hormones and the increase detected compared with controls was not statistically significant.\nIL-33 has been studied in endometriosis previously and shown to be elevated with deep endometriosis ( Santulli  et al.  2012 ,  Mbarik  et al.  2015 ,  Miller  et al.  2017 ). This may explain why it is unchanged in our study, as patients with deep endometriosis were not investigated.  Mbarik  et al.  (2015)  divided their cohort by the revised American Society for Reproductive Medicine staging and found stage I–II endometriosis (closest equivalent to the SPE group in the current study) had similar circulating IL-33 levels compared to control but elevated levels for the stage III–IV group in both the serum and peritoneal fluid ( Mbarik  et al.  2015 ). A limitation to our study is also one of its aforementioned strengths. While our control group are confirmed to have no endometriosis lesions visualised at the time of laparoscopy, they all experience pelvic pain justifying their investigation for suspected endometriosis. It is possible that alterations in the levels of these cytokines may be driven by inflammatory processes similarly contributing to pelvic pain, which may explain the lack of differences observed between groups here, particularly for the cytokines previously extensively published in relation to endometriosis.\nIt is also possible that these inflammatory peritoneal processes present for those with chronic pelvic pain contribute to infertility. Serum AMH levels in our ‘no endo’ control group were lower than those reported in the previous literature. The mean value for our ‘no endometriosis’ control group was 1.15 ng/mL (compared to 0.89 ng/mL in the SPE group). The previous study by  Lessans  et al.  (2024)  reported a mean AMH value of 3.0 ng/mL for their age-matched control group (2.8 ng/mL for peritoneal endometriosis), and a study combining all subtypes of endometriosis reported a mean AMH of 2.30 ng/mL for healthy controls (1.99 ng/mL for endometriosis) ( Lessans  et al.  2024 ,  Ramezani Tehrani  et al.  2025 ). Together, these data may suggest a mechanism where the pelvic microenvironment (including pelvic inflammation), irrespective of the presence of endometriosis lesions, contributes to diminished AMH and subfertility. Women with other pelvic inflammatory conditions, such as Crohn’s disease or inflammatory bowel disease, are known to impact fertility if they have a flare up of their condition during attempts to conceive and during pregnancy itself ( Nguyen  et al.  2016 ,  Mahadevan  et al.  2019 ,  Rosiou & Selinger 2023 ,  Torres  et al.  2023 ). While this requires further investigation in the settings of pelvic pain and endometriosis, and further studies may benefit from an additional asymptomatic, healthy control group, the notion of an inflammatory peritoneum compromising fertility is plausible.\n\nHere, for the first time, we demonstrate that AMH levels are reduced for women with SPE compared to women without endometriosis. These changes are accompanied by greater levels of IL-17 and TNF-α in the serum and IL-23 in the peritoneal fluid. Interestingly, these differences in cytokine levels were only observed in women not using hormone treatments. Our findings provide valuable information for clinicians to utilise when counselling women with endometriosis about their ovarian reserve and future fertility, with or without an endometrioma present.\n\nAW Horne is a co-Editor-in-Chief of  Reproduction & Fertility  and was not involved in the review or editorial process for this paper, on which he is listed as an author. PTKS’s institution (University of Edinburgh) receives payment for consultancy for Gesynta, Rathlin, Roche, and Gedeon Richter. AWH’s institution (University of Edinburgh) receives personal fees from Rathlin, Theramex, and Gedeon Richter.\n\nMJG secured an internal University of Edinburgh Deanery of Clinical Sciences Funding Challenge grant to fund part of this work. AWH received grants from the National Institute for Health and Care Research Health Technology Assessment, Chief Scientist Office, Wellbeing of Women, Roche Diagnostics, and European Union.\n\nMJG, DAG, PTKS, and AWH designed the study. MJG, MEB, and FC performed the experimental procedures and statistical analysis. MJG, DAG, PTKS, CED, and AWH interpreted the results. MJG, DAG, PTKS, and AWH wrote the manuscript. MJG, MEB, FC, DAG, CED, PTKS, and AWH edited the manuscript.","source_license":"CC0","license_restricted":false}