{"paper_id":"d4ebd31a-f448-4594-9eba-ece9fe7291c3","body_text":"Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162  \nhttps://doi.org/10.1186/s12884-024-06347-9\nRESEARCH Open Access\n© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which \npermits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the \noriginal author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or \nother third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line \nto the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory \nregulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this \nlicence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom-\nmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.\nBMC Pregnancy and Childbirth\nCausal effects of endometriosis on SLE, \nRA and SS risk: evidence from meta-analysis \nand Mendelian randomization\nTianyou Tang1*, Yi Zhong1, Sipei Xu2 and Huilin Yu3 \nBackground Endometriosis is an underdiagnosed disorder that affects an estimated 6-10% of women of reproduc-\ntive age. Endometriosis has been reported in epidemiological studies to be associated with autoimmune diseases. \nHowever, the relationship remains controversial.\nMethods A meta-analysis of observational studies was undertaken to evaluate the risk of autoimmune diseases \nin patients with endometriosis. The relevant studies were retrieved via the databases Medline, Embase and Web of Sci-\nence until July 20, 2023. Mendelian randomization (MR) was subsequently utilized to scrutinize the causal influence \nof genetic predisposition toward endometriosis on three autoimmune diseases.\nResults The meta-analysis findings revealed a relationship between endometriosis and the onset of SLE (cohort \nstudies: RR = 1.77, 95% confidence interval (CI): 1.47–2.13,  I2 = 0%; Case-control and cross-sectional studies: OR = 5.23, \n95% CI: 0.74–36.98,  I2 = 98%), RA (cohort studies: RR = 2.18, 95% CI: 1.85–2.55,  I2 = 92%; Case-control and cross-sectional \nstudies: OR = 1.40, 95% CI: 1.19–1.64,  I2 = 0%) and SS (cohort studies: RR = 1.49, 95% CI: 1.34–1.66,  I2 = 0%). Similarly, \nin our MR study, the results of the inverse-variance-weighted (IVW) model suggested that genetic predisposition \nto endometriosis was causally associated with an increased risk for SLE (OR = 1.915, 95% CI: 1.204–3.045, p = 0.006) \nand RA (OR = 1.005, 95% CI: 1.001–1.009, p = 0.014).\nConclusions Both our meta-analysis and MR study indicate that endometriosis increases the risk of autoimmune \ndiseases. These findings not only broaden our understanding of the genetic mechanisms underlying the comorbidity \nof endometriosis and autoimmune diseases, but also offer a new strategy for autoimmune disease prevention.\nKeywords Endometriosis, Systemic lupus erythematosus (SLE), Rheumatoid arthritis (RA), Sjögren’s syndrome (SS), \nMeta-analysis, Mendelian randomization\nIntroduction\nEndometriosis is an estrogen-dependent chronic inflam -\nmatory disease that affects approximately 7–10% of \nwomen worldwide. It is characterized by two main symp-\ntoms: pelvic pain and infertility [1]. Serval studies have \nindicated that endometriosis is associated with numer -\nous diseases, including gastrointestinal diseases, malig -\nnancies, cardiovascular diseases, mental disorders and \nautoimmune diseases [2, 3]. The pathogenesis of endo -\nmetriosis remains unclear, but the retrograde menstrua -\ntion theory is currently widely accepted [4].\n*Correspondence:\nTianyou Tang\n2020210155@stu.cqmu.edu.cn\n1 Children’s Hospital of Chongqing Medical University, Chongqing, China\n2 The First Medicine College, Chongqing Medical University, Chongqing, \nChina\n3 The Second Medicine College, Chongqing Medical University, \nChongqing, China\n\nPage 2 of 11Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \nAutoimmune diseases affect 3–5% of the popula -\ntion, with some being organ-specific, like RA, and oth -\ners involving multiple organs, such as SLE [5]. There \nhave been many review articles discussing the immuno -\nlogical aspects of endometriosis [6, 7]. They believe that \nthe changes in cell-mediated and humoral immunity in \npatients with endometriosis may be the reason for the \nincreased risk of autoimmune diseases. However, there \nare few articles based on population-based observational \nstudies. In an attempt to understand more about the risk \nof autoimmune diseases in endometriosis, we embarked \non a meta-analysis by including cohort studies, cross-sec-\ntional studies and case–control studies.\nMendelian randomization (MR) analysis employs \ngenetic variation as an instrumental variable, enabling \nthe evaluation of relationships between an exposure \nand an outcome. By leveraging the random distribu -\ntion of genetic variation, MR helps eliminate confound -\ning factors and reverse causation, thus simulating the \nrandomization process seen in a randomized controlled \nexperiment [8–10]. The degree of the connection and the \ndirection of causality between endometriosis and auto -\nimmune diseases were evaluated in this study using MR. \nIn this study, we employed a meta-analysis in conjunc -\ntion with MR analysis to elucidate the causal relation -\nship, strength of association, and direction of causality \nbetween endometriosis and three autoimmune diseases.\nMethods\nMeta‑analysis\nGeneral information\nWe performed this meta-analysis following the Preferred \nReporting Items for Systematic Reviews and Meta-anal -\nyses (PRISMA) guidelines [11]. The study protocol was \nregistered at PROSPERO (CRD42023444650). (https:// \nwww. crd. york. ac. uk/ PROSP ERO/).\nSearch strategy\nA search strategy was developed as presented in (Data -\nsheet1: Table S1). Two researchers (TYT and YZ) con -\nducted a comprehensive electronic literature search \nof the PubMed, Web of Science, and Embase databases \nfrom their inception until July 2023. No restrictions were \napplied regarding geographic area, language, or pub -\nlication status. Additionally, the researchers manually \nreviewed the reference lists of relevant articles to identify \nany additional studies that may have been missed in the \ninitial search.\nStudy selection\nTwo of the authors (TYT and YZ) initially screened the \ntitles and abstracts of the studies to exclude those that \nappeared irrelevant. Then, they thoroughly read the full \ntexts of the remaining studies to further exclude any \nstudies that did not meet the eligibility criteria. Any \ndisagreements between the two authors were resolved \nthrough discussion.\nEligible articles for this study had to satisfy the fol -\nlowing criteria: 1) having a cohort, case–control or \ncross-sectional study design and published in English, \n2) comparing the risk of autoimmune disease among \nwomen with/without endometriosis, 3) providing data \non odds ratio (OR), risk ratio (RR), hazard ratio (HR), \nstandardized incidence ratio (SIR), incidence rate ratio \n(IRR) for autoimmune disease. (Table 1 ).\nData extraction\nTwo authors (TYT and YZ) independently extracted \ndata and a consensus was reached in case of any \ninconsistency.\nUsing a pre-designed data extraction form, the fol -\nlowing information was meticulously recorded: title, \nthe name of the primary author, publication year, coun -\ntry, average age, duration of follow-up, sample size, \noutcome assessment, risk estimate, corresponding 95% \nconfidence intervals.\nAssessing the Risk of Bias.\nThe Newcastle–Ottawa quality assessment scale \n(NOS) was used to evaluate the methodological qual -\nity of cohort study and case–control study included in \nthe analysis [12]. In the absence of established stand -\nard criteria, we categorized studies with 0–3 stars, 4–6 \nstars, or 7–9 stars as low-quality, moderate-quality, or \nhigh-quality, respectively. To evaluate the methodologi -\ncal quality of cross-sectional studies, we used the cri -\nteria provided by the Agency for Healthcare Research \nand Quality (AHRQ) [13]. Each item in the assessment \nwas assigned a score of ’0’ if it was answered as ’NO’ \nor ’UNCLEAR’ , and a score of ’1’ if it was answered as \n’YES’ . The total score for each study was then calcu -\nlated. Based on the total score, the article quality was \ncategorized as low-quality (0–3), moderate-quality \n(4–7), or high-quality (8–11). Disagreements were \nresolved through discussion.\nTable 1 PICOS criteria for inclusion of studies\nParticipants The general population\nIntervention/exposure Endometriosis\nComparison People without endometriosis\nOutcome Autoimmune diseases risk\nStudy design Cohort, case–control \nand cross-sectional study\n\nPage 3 of 11\nTang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n \nStatistical analysis\nThe meta-analysis was conducted using Review Manager \n5.4. For cross-sectional and case–control studies, raw \ndata were extracted to compute a odds ratio (OR) accom-\npanied by 95% confidence intervals (CIs). For cohort \nstudies, SIR, IRR and HR were treated as the relative risk \n(RR), and the pooled RR with a 95% confidence interval \nwas calculated [14, 15]. To ensure a more accurate assess-\nment of the relationships between endometriosis and \nSLE, RA and SS, categorical meta-analyses were con -\nducted. The I2 statistic was used to evaluate the degree of \nheterogeneity among the included studies. If the I2 value \nexceeded 50% or the p-value was less than 0.05, indicat -\ning a high level of heterogeneity, a random-effects model \nwas employed. Conversely, if the I2 value was below \n50% or the p-value was greater than or equal to 0.05, a \nfixed-effect model was used in the meta-analysis. This \napproach helps to account for heterogeneity and provide \nmore reliable results.\nMendelian randomization\nStudy design\nThe Mendelian randomization (MR) method is based on \nthree key assumptions, which are summarized in Fig.  1. \nFirst, the selected SNPs must be significantly correlated \nwith the exposure factor. Second, SNPs must be inde -\npendent of potential confounding factors. Third, SNPs \nshould not have a direct association with outcome.\nData source\nFor the study of endometriosis, we retrieved genome-\nwide association study (GWAS) summary data from \nNilufer R et  al. (PMID: 36,914,876) [16]. In this GWAS, \na total of 60,694 patients with endometriosis and 701,926 \ncontrols of European and East Asian descent were \nincluded by the authors. Summary statistics of SLE are \nfrom Wang YF et  al. (PMID: 33,536,424), including up \nto 12,653 participants (4,222 cases and 8,431 controls) \nof East Asian ancestry [17]. Summary statistics of RA \nare from Neale Lab, including up to 337,159 participants \n(3,730 cases and 333,429 controls) of European ances -\ntry. The summary statistics of SS were obtained from \nthe FinnGen consortium release data (1,290 cases and \n213,145 controls).\nStatistical analysis\nCochran’s Q test was used in this MR study to determine \nwhether there was variability in estimates of specific \ngenetic variants [18]. Inverse variance weighted (IVW) \nanalysis was the main technique employed [19]. In addi -\ntion to IVW, further analyses were carried out utiliz -\ning the weighted median method [20], simple mode, \nweighted mode, and MR-egger regression method [21]. \nFinally, to guarantee the accuracy of the results, we tested \nand calibrated horizontal pleiotropic outliers in the IVW \nmodel using MR pleiotropy residual sum and outlier \n(MR-PRESSO) [22].\nFig. 1 Diagram for key assumptions of MR analyses. Exposure SNPs were used as the genetic instruments to investigate the causal effect \nof endometriosis on outcome. The directional arrows indicates that the genetic instruments (SNPs) are associated with the exposure and can \nonly influence the outcome through the exposure. Dashed lines represent that the genetic instruments (SNPS) are independent of any \nconfounding variables affecting the results\n\nPage 4 of 11Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \nSensitivity analysis\nTo detect potential pleiotropy, we conducted the MR-\nEgger test and interpreted a P-value greater than 0.05 for \nthe MR-Egger intercept as an absence of horizontal plei -\notropy [23]. To assess the stability of the results, leave-\none-out sensitivity analyses were conducted, wherein a \nsingle SNP was excluded in each iteration. This analysis \nhelps determine if any single SNP is driving the observed \nassociations. Funnel plots and forest plots were gen -\nerated to visually explore the existence of pleiotropy, \nwhich is when a genetic variant affects multiple traits \nor outcomes. A two-sided p-value of less than 0.05 was \nconsidered as suggestive of significance. All the analyses \nwere performed using the \"Two-Sample-MR\" and \"MR-\nPRESSO\" packages in R software, specifically Version \n4.2.3.\nResults\nMeta‑analysis\nStudy selection and characteristics\nA flowchart of the process of choosing the specific \nliterature is shown in Fig.  2. Tables  2 and 3  provide a \ndetailed summary of key features for the 13 included \nresearch. In brief, 3 (23.1%) investigations were con -\nducted in North America, 4 (30.8%) in Europe, and 6 \n(46.2%) in Asia. In terms of study design, one (7.8%) \nstudy was cross-sectional, four (30.8%) were case–\ncontrol studies, and eight (61.5%) were cohort studies \n(four prospective cohort studies and four retrospective \ncohort studies).\nQuality assessment\nAccording to the information provided, the researchers \nused the Agency for Healthcare Research and Quality \n(AHRQ) rating criteria to assess the quality of cross-\nsectional studies included in their analysis. Addition -\nally, the Newcastle–Ottawa Scale (NOS) was used to \nevaluate the quality of case–control and cohort studies. \nThe results of this assessment can be found in Tables  4 \nand 5, which presumably shows that all the publications \nincluded in the study were rated as high or moderate \nquality based on the use of AHRQ and NOS criteria.\nFig. 2 The flowchart of meta-analysis\n\nPage 5 of 11\nTang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n \nData synthesis\nEndometriosis and SLE Eight studies were included in \na meta-analysis comparing the risk of SLE in individu -\nals with endometriosis and those without endometriosis, \ninvolving 477,501 individuals and 113,318 endometrio -\nsis cases [24–31]. For case–control and cross-sectional \nstudies, we did not find a significant association between \nthe two conditions. For cohort studies, with low hetero -\ngeneity, the pooled risk of SLE was greater in individuals \nwith endometriosis than in those without endometriosis \n(RR = 1.77, 95% CI 1.47–2.13; I2 = 0%) (Fig. 3).\nEndometriosis and RA Six studies were included in a \nmeta-analysis comparing the risk of RA in individuals \nwith endometriosis and those without endometriosis, \ninvolving 434,377 individuals and 73,006 endometrio -\nsis cases [25, 30–34]. Case–control and cross-sectional \nstudies have described the association between endome -\ntriosis and RA (OR = 1.40, 95% CI 1.19–1.64; I2 = 0%). \nFor cohort studies, with high heterogeneity, we observed \nsimilar trends in the results (RR = 1.89, 95% CI 1.04–3.42; \nI2 = 92%) (Fig. 3).\nEndometriosis and SS Three studies were included in \na meta-analysis comparing the risk of SS in individuals \nwith endometriosis and those without endometriosis, \ninvolving 215,006 individuals and 56,074 endometriosis \ncases [28, 30, 35]. For cohort studies, with low hetero -\ngeneity, the pooled risk of SS was greater in individuals \nwith endometriosis than in those without endometriosis \n(RR = 1.49, 95% CI 1.34–1.66; I2 = 0%) (Fig. 3).\nMendelian randomization study\nGenetic associations between endometriosis and SLE, RA, \nand SS risk\nUsing the random-model IVW, we discovered a link \nbetween each standard rise in endometriosis risk and a \nfaster development to SLE (OR = 1.915, 95% CI: 1.204–\n3.045, p = 0.006) and RA (OR = 1.005, 95% CI: 1.001–\n1.009, p = 0.014). However, no causal relationship was \nfound between endometriosis and SS. (Fig. 4).\nSensitivity analysis\nCochran’s Q test findings revealed that these SNPs \nexhibited no heterogeneity. We found no evidence of \nTable 2 Characteristics of case–control and cross-sectional studies\nCitation Country Study period Study design Effect estimates Study \npopulation (n)\nDiagnosis criteria\nEndometriosis Autoimmune \ndisease\nMatorras et al. \n(2007) [27]\nSpain 1990-\n2004\nCase–control \nstudy\nSLE: OR = 2.9\n(95%CI:0.27–32.57)\nP > 0.05\nSS: \nOR、95%CI:not \ncalculable\nP > 0.05\nCase: 342\nControl: 501\nHistology Clinical interview \nand medical \nrecords according \nto the ACR criteria\nYoshii et al. (2021) \n[31]\nJapan 2011-\n2018\nCase–control \nstudy\nSLE: IRR = 1.35 \n(95%CI:0.99–1.84)\nRA: IRR = 1.31\n(95%CI:1.05–1.64)\nCase: 30,516\nControl: 120,976\nICD-10-CM ICD-10-CM\nPorpora et al. \n(2019)\nItaly 2014-\n2017\nRetrospective \ncase–control study\nSLE:\nOR = 8.63\n(95%CI:1.07–69.91)\nP = 0.01\nCase: 148\nControl: 150\nLaparoscopy \nand histology\nAntinuclear anti-\nbodies, extractable \nnuclear antigen, \nanti-cardiolipin anti-\nbodies, antiphos-\npholipid antibodies, \nand lupus antico-\nagulant\nSinaii et al. (2002) \n[30]\nUSA,\nCanada\n1988 Cross-sectional \nstudy\nSLE: OR = 20.7 \n(95%CI:14.3–29.9)\nP < 0.0001\nRA: OR = 1.5\n(95%CI:1.2–1.9)\nP = 0.001\nSS: OR = 23.9\n(95%CI:15.5–36.5)\nP < 0.0001\nStudy sample: \n3680\nSelf-reports \nof laparoscopy/\nlaparotomy\nSelf-reported physi-\ncian diagnosis\n\nPage 6 of 11Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \ndirectional pleiotropy using MR Egger intercepts and \nMR-PRESSO (Datasheet 2). The leave-one-out analysis \nproduced a stable result when each SNP was eliminated, \nas shown in Datasheet 3.\nDiscussion\nMany previous studies have found a link between endo -\nmetriosis and autoimmune disorders, a literature review \nwas compiled on the topic (Datasheet1:Table S2 ). We \nemploy meta-analysis and MR to investigate causative \nassociations between endometriosis and SLE, RA, and \nSS risk. To our knowledge, the latest and probably only \nmeta-analysis was published in 2019 [36]. Due to the \ninability of cross-sectional and case–control studies to \nresolve the issue of the temporal relationship between \nendometriosis and autoimmune diseases, their research \nfaced challenges in determining the sequence of disease \ndevelopment and manifestation, as well as potential \ncausal relationships. Compared to this study, our study \ncomprises more studies in patients with endometriosis, \nincluding newer and larger cohort studies. We aimed to \ninvestigate the causal effect of endometriosis on SLE, \nRA and SS. As the first report to employ MR in inves -\ntigating the causal association between endometriosis \nand the risk of SLE, RA, and SS, our study eliminates \nTable 3 Characteristics of cohort studies\nCitation Country Study period Study design Effect estimates Study \npopulation (n)\nDiagnosis criteria\nEndometriosis Autoimmune \ndisease\nNielsen et al. \n(2011) [28]\nDenmark 1977-\n2007\nRetrospective cohort \nstudy;\n12.1 years follow-up\nSLE:\nSIR:1.6\n(95%CI:1.2–2.1)\nSS:\nSIR:1.6\n(95%CI:1.3–2.0)\nExposure: 37,661 ICD8:codes \n62,530–62539\nICD10: code \ngroup N80\nMedical records \naccording \nto the ACR criteria\nHarris et al. \n(2016a) [25]\nUSA 1989-\n2011\nProspective cohort study;\n22 years follow-up\nSLE:\nHR = 1.61\n(95%CI:0.88–2.92)\nP < 0.05\nRA:\nHR = 1.16\n(95%CI:0.84–1.59)\nP < 0.05\nExposure: 6434\nControl: 108,019\nSelf-reported \nlaparoscopy\nMedical records \naccording to the \nACR criteria\nLin et al.  \n(2020) [26]\nTaiwan 2000-\n2012\nRetrospective cohort \nstudy;\npatients were followed \nuntil diagnosed with SLE, \ndeath(means = 8.1 years)\nSLE:\nHR = 1.86\n(95%CI:1.36–2.53)\nP < 0.0001\nExposure: 17,779\nControl: 17,779\nICD-9-CM Medical records \naccording to \nboard-certified \nrheumatologists\nFan et al.  \n(2021) [24]\nTaiwan 2000-\n2011\nRetrospective cohort \nstudy;\n12 years follow-upar\nSLE:\nHR = 2.37\n(95%CI:1.35–4.14)\nExposure: 16,758\nControl: 16,758\nICD-9-CM ICD-9-CM\nMerlino et al. \n(2003) [33]\nUSA 1986-\n1997\nProspective cohort study;\n11 years follow-up\nRA:\nRR = 1.59\n(95%CI:0.82–3.08)\n/ Self-reported \nphysician diag-\nnosis\nSelf-reported phy-\nsician diagnosis\nChen et al. (2020) Taiwan 2000-\n2012\nProspective cohort study;\npatients was followed \nuntil the appearance \nof RA, their removal \nfrom the NHIP , death, \nor the end of 2013 \n(means = 8.1 years)\nRA:\nHR = 3.71\n(95%CI: \n2.91–5.73)\nP = 0.77\nExposure: 17,913\nControl: 17,913\nICD-9-CM ICD-9-CM\nXue et al. (2020) Taiwan 2000-\n2013\nProspective cohort study;\n13 years follow-up\nRA:\nHR = 1.75\n(95%CI: \n1.27–2.41)\nP < 0.05\nExposure: 14,463\nControl: 14,463\nICD-9-CM ICD-9-CM\nChao et al.  \n(2022) [35]\nTaiwan 2000-\n2012\nRetrospective cohort \nstudy\nSS:\nHR = 1.45\n(95%CI:1.27–1.65)\nP < 0.001\nExposure: 14,733\nControl: 58,932\nICD-9-CM ICD-9-CM\n\nPage 7 of 11\nTang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n \nconfounding factors and reverse causality effects, which \nmay yield more reliable results.\nThe meta-analysis findings reveal that endometriosis \npatients are more likely to develop SLE and SS, which \nis consistent with the results of the meta-analysis from \n2019. However, in contrast to previous studies, we found \nthat endometriosis also increases the risk of RA. The \ndiscrepancy in findings between the two studies may be \ndue to the fact that the 2019 study only had two cohort \nstudies available for inclusion. We observed that only a \nlimited number of studies accounted for the effects of \nconfounding factors, which may influence the level of the \nrisk. Smoking, alcohol usage, caffeine consumption, and \na lack of exercise have all been linked to an increased risk \nof endometriosis [37]. These lifestyle and environmental \nfactors have also been linked to an increased susceptibil -\nity to cancer occurrence [38]. At the same time, the num-\nber of included studies was relatively small, which poses \na risk of generating spurious associations. Therefore, \nwe encourage more researchers to further investigate \nthe association by employing prospective cohort study \ndesigns.\nWe also utilize a two-sample MR approach to fur -\nther investigate the impact of endometriosis on the risk \nof autoimmune diseases. We employ independent loci \nassociated with endometriosis identified from the larg -\nest available genome-wide association studies (GWAS) to \ndate. All SNPs were identified in the PhenoScanner data -\nbase ( http:// www. pheno scann er. medsc hl. cam. ac. uk/) to \nexclude SNPs related to confounding factors. With MR, \nwe found that endometriosis is linked with a higher risk \nof SLE and RA, which supports prior meta-analyses.\nThe imbalance of the immune system may explain \nthe the observed impact of endometriosis on autoim -\nmune diseases in clinical practice [39–41]. Previous \nstudies have demonstrated that elevated expression of \nIL-6, IL-15, and TGF-β1 in patients with endometriosis \ncan reduce the activity of NK cells [42–44]. It has been \nobserved that patients with endometriosis often have an \nincrease in neutrophils and macrophages in their peri -\ntoneal fluid [45, 46]. In the latest meta-analysis to date, \nRiccio et al. suggested that there is an increase in B lym -\nphocytes and excessive production of autoantibodies in \nendometriosis [47]. These alterations play an important  \nrole in mediating the pathogenesis of autoimmune  \ndiseases [48–50].\nEstrogen also plays an important role in the develop -\nment of endometriosis and autoimmune diseases. Endo -\nmetriosis is an estrogen-dependent disease, and the \ndisruption of estrogen signaling leads to hormonal imbal-\nance, which causes its symptoms [51]. Targeting estrogen \nis still considered the optimal approach for controlling \nthe progression and inflammation of endometriosis [52]. \nEstrogen has also been found to regulate the immune \nsystem and contribute to the transduction pathways of \nautoimmunity by activating its nuclear receptor AhR \n[53]. Estrogen raises the risk of autoimmune diseases by  \nraising the generation of type 1 interferon and promoting  \nthe survival of B cells that create pathogenic IgG \nautoantibodies [54].\nFirstly, pleiotropy has always been an important \nissue in Mendelian randomization. However, neither \nTable 4 NOS assessment for case–control and cross-sectional studies\nNEWCASTLE—OTTAWA QUALITY ASSESSMENT SCALE\nAuthor Selection Comparability Exposure Total score Quality grade\nNielsen et al. (2011) [28] 2 1 3 6 moderate\nHarris et al. (2016a) [25] 1 2 3 7 high\nLin et al. (2020) [26] 4 1 2 7 high\nFan et al. (2021) [24] 4 2 3 9 high\nMerlino et al. (2003) [33] 4 2 3 9 high\nChen et al. (2020) 4 1 2 7 high\nXue et al. (2020) 4 2 3 9 high\nChao et al. (2022) [35] 3 2 2 7 high\nMatorras et al. (2007) [27] 3 1 3 7 high\nPorpora et al. (2019) 2 1 3 6 moderate\nYoshii et al. (2021) 2 2 2 6 moderate\nTable 5 AHRQ assessment for cross-sectional studies\nAgency for Healthcare Research and Quality\nAuthor Total score Quality grade\nSinaii et al. (2002) [30] 6 moderate\nShafrir et al. (2021) 8 high\n\nPage 8 of 11Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \nFig. 3 Forest plots of studies. (A) the association of endometriosis with SLE risk; (B) the association of endometriosis with RA risk; (C) the association \nof endometriosis with SS risk\n\nPage 9 of 11\nTang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n \nthe MR-Egger nor the MR-PRESSO analyses revealed \nany indication of horizontal pleiotropy, indicating a \nvery low level of pleiotropic bias. Secondly, the obser -\nvational studies collected in our meta-analysis did not \nconsider the influence of mediation effects. For exam -\nple, patients with endometriosis exhibited an increased \nsusceptibility to sedentary behavior [3 ]. Sedentary \nbehavior is also acknowledged as a risk factor for auto -\nimmune diseases [55]. Finally, our study did not exam -\nine the effect of endometriosis on the prognosis of \nautoimmune disorders due to a lack of data. Therefore, \nmore studies are warranted to elucidate the possible \nrelation between the two conditions. \nAbbreviations\nFig  Figure\nGWAS  Genome-wide association study\nIVW  Inverse-variance-weighted\nMR  Mendelian randomization\nSupplementary Information\nThe online version contains supplementary material available at https:// doi. \norg/ 10. 1186/ s12884- 024- 06347-9.\nSupplementary Material 1. \nSupplementary Material 2. \nSupplementary Material 3. \nAuthors’ contributions\nConceptualization: T.Y.T. Software: T.Y.T. Data curation: H.L.Y. and Y.Z. Writing \n– original draft: T.Y.T. Visualization: S.P .X. The work reported in the paper has \nbeen performed by the authors, unless clearly specified in the text. All authors \nreviewed the manuscript.\nFunding\nThis work was supported by National innovation and entrepreneurship train-\ning program [SRIEP202104].\nAvailability of data and materials\nThe 42 SNPs selected for endometriosis are provided in Datasheet2 Table S1. \nThe data that support the findings of this study are openly available in an \nopen website (https:// gwas. mrcieu. ac. uk/). All data generated or analysed dur-\ning this study are included in this published article and datasheet.\nDeclarations\nEthics approval statement and consent to participate\nThe data we used were obtained from published studies approved by the cor-\nresponding ethics committee, thus no further ethical approval was required \nfor this study.\nConsent for publication\nNot applicable.\nCompeting interests\nThe authors declare no competing interests.\nReceived: 20 September 2023   Accepted: 13 February 2024\nReferences\n 1. Giudice LC, Kao LC. Endometriosis Lancet. 2004;364:1789–99. https:// doi. \norg/ 10. 1016/ S0140- 6736(04) 17403-5.\n 2. Koller D, Pathak GA, Wendt FR, Tylee DS, Levey DF, Overstreet C, et al. \nEpidemiologic and Genetic Associations of Endometriosis With Depres-\nsion, Anxiety, and Eating Disorders. JAMA Netw Open. 2023;6: e2251214. \nhttps:// doi. org/ 10. 1001/ jaman etwor kopen. 2022. 51214.\n 3. Parazzini F, Esposito G, Tozzi L, Noli S, Bianchi S. Epidemiology of \nendometriosis and its comorbidities. Eur J Obstet Gynecol Reprod Biol. \n2017;209:3–07. https:// doi. org/ 10. 1016/j. ejogrb. 2016. 04. 021.\n 4. Burney RO, Giudice LC. Pathogenesis and pathophysiology of endometri-\nosis. Fertil Steril. 2012;98:511–9. https:// doi. org/ 10. 1016/j. fertn stert. 2012. \n06. 029.\n 5. Wang L, Wang FS, Gershwin ME. Human autoimmune diseases: a com-\nprehensive update. J Intern Med. 2015;278:369–95. https:// doi. org/ 10. \n1111/ joim. 12395.\n 6. Paul DW, Braun DP . Immunology of endometriosis. Best Pract Res Clin \nObstet Gynaecol. 2004;18:245–63. https:// doi. org/ 10. 1016/j. bpobg yn. \n2004. 02. 001.\nFig. 4 Associations of genetic predisposition to endometriosis with SLE, RA, and SS risk\n\nPage 10 of 11Tang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n 7. Symons LK, Miller JE, Kay VR, Marks RM, Liblik K, Koti M, et al. The Immu-\nnopathophysiology of Endometriosis. Trends Mol Med. 2018;24:748–62. \nhttps:// doi. org/ 10. 1016/j. molmed. 2018. 07. 004.\n 8. Burgess S, Butterworth A, Thompson SG. Mendelian randomization \nanalysis with multiple genetic variants using summarized data. Genet \nEpidemiol. 2013;37:658–65. https:// doi. org/ 10. 1002/ gepi. 21758.\n 9. Lawlor DA, Harbord RM, Sterne JA, Timpson N, Davey SG. Mendelian \nrandomization: using genes as instruments for making causal inferences \nin epidemiology. Stat Med. 2008;27:1133–63. https:// doi. org/ 10. 1002/ sim. \n3034.\n 10. Smith GD, Ebrahim S. “Mendelian randomization”: can genetic epidemiol-\nogy contribute to understanding environmental determinants of dis-\nease? Int J Epidemiol. 2003;32:1–22. https:// doi. org/ 10. 1093/ ije/ dyg070.\n 11. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for \nsystematic reviews and meta-analyses: the PRISMA statement. Plos Med. \n2009;6: e1000097. https:// doi. org/ 10. 1371/ journ al. pmed. 10000 97.\n 12. Norris JM, Simpson BS, Ball R, Freeman A, Kirkham A, Parry MA, et al. A \nModified Newcastle-Ottawa Scale for Assessment of Study Quality in \nGenetic Urological Research. Eur Urol. 2021;79:325–6. https:// doi. org/ 10. \n1016/j. eururo. 2020. 12. 017.\n 13. Soffin EM, Gibbons MM, Ko CY, Kates SL, Wick E, Cannesson M, et al. \nEvidence Review Conducted for the Agency for Healthcare Research \nand Quality Safety Program for Improving Surgical Care and Recovery: \nFocus on Anesthesiology for Total Knee Arthroplasty. Anesth Analg. \n2019;128:441–53. https:// doi. org/ 10. 1213/ ANE. 00000 00000 003564.\n 14. Greenland S. Quantitative methods in the review of epidemiologic litera-\nture. Epidemiol Rev. 1987;9:1–30. https:// doi. org/ 10. 1093/ oxfor djour nals. \nepirev. a0362 98.\n 15. Siristatidis C, Sergentanis TN, Kanavidis P , Trivella M, Sotiraki M, Mavroma-\ntis I, et al. Controlled ovarian hyperstimulation for IVF: impact on ovarian, \nendometrial and cervical cancer–a systematic review and meta-analysis. \nHum Reprod Update. 2013;19:105–23. https:// doi. org/ 10. 1093/ humupd/ \ndms051.\n 16. Rahmioglu N, Mortlock S, Ghiasi M, Møller PL, Stefansdottir L, Galarneau \nG, et al. The genetic basis of endometriosis and comorbidity with other \npain and inflammatory conditions. Nat Genet. 2023;55:423–36. https:// \ndoi. org/ 10. 1038/ s41588- 023- 01323-z.\n 17. Wang YF, Zhang Y, Lin Z, Zhang H, Wang TY, Cao Y, et al. Identification of \n38 novel loci for systemic lupus erythematosus and genetic heterogene-\nity between ancestral groups. Nat Commun. 2021;12:772. https:// doi. org/ \n10. 1038/ s41467- 021- 21049-y.\n 18. Bowden J, Del GMF, Minelli C, Davey SG, Sheehan N, Thompson J. A \nframework for the investigation of pleiotropy in two-sample summary \ndata Mendelian randomization. Stat Med. 2017;36:1783–802. https:// doi. \norg/ 10. 1002/ sim. 7221.\n 19. Yang J, Ferreira T, Morris AP , Medland SE, Madden PA, Heath AC, et al. \nConditional and joint multiple-SNP analysis of GWAS summary statistics \nidentifies additional variants influencing complex traits. Nat Genet. \n2012;44(369–75):S1-03. https:// doi. org/ 10. 1038/ ng. 2213.\n 20. Bowden J, Davey SG, Haycock PC, Burgess S. Consistent Estimation \nin Mendelian Randomization with Some Invalid Instruments Using a \nWeighted Median Estimator. Genet Epidemiol. 2016;40:304–14. https:// \ndoi. org/ 10. 1002/ gepi. 21965.\n 21. Bowden J, Davey SG, Burgess S. Mendelian randomization with invalid \ninstruments: effect estimation and bias detection through Egger regres-\nsion. Int J Epidemiol. 2015;44:512–25. https:// doi. org/ 10. 1093/ ije/ dyv080.\n 22. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal \npleiotropy in causal relationships inferred from Mendelian randomization \nbetween complex traits and diseases. Nat Genet. 2018;50:693–8. https:// \ndoi. org/ 10. 1038/ s41588- 018- 0099-7.\n 23. Burgess S, Thompson SG. Interpreting findings from Mendelian rand-\nomization using the MR-Egger method. Eur J Epidemiol. 2017;32:377–89. \nhttps:// doi. org/ 10. 1007/ s10654- 017- 0255-x.\n 24. Fan YH, Leong PY, Chiou JY, Wang YH, Ku MH, Wei JC. Association \nbetween endometriosis and risk of systemic lupus erythematosus. Sci \nRep. 2021;11:532. https:// doi. org/ 10. 1038/ s41598- 020- 79954-z.\n 25. Harris HR, Costenbader KH, Mu F, Kvaskoff M, Malspeis S, Karlson EW, \net al. Endometriosis and the risks of systemic lupus erythematosus \nand rheumatoid arthritis in the Nurses’ Health Study II. Ann Rheum Dis. \n2016;75:1279–84. https:// doi. org/ 10. 1136/ annrh eumdis- 2015- 207704.\n 26. Lin YH, Yang YC, Chen SF, Hsu CY, Shen YC. Risk of systemic lupus erythe-\nmatosus in patients with endometriosis: A nationwide population-based \ncohort study. Arch Gynecol Obstet. 2020;302:1197–203. https:// doi. org/ \n10. 1007/ s00404- 020- 05726-9.\n 27. Matorras R, Ocerin I, Unamuno M, Nieto A, Peiró E, Burgos J, et al. Preva-\nlence of endometriosis in women with systemic lupus erythematosus \nand Sjogren’s syndrome. Lupus. 2007;16:736–40. https:// doi. org/ 10. 1177/ \n09612 03307 081339.\n 28. Nielsen NM, Jørgensen KT, Pedersen BV, Rostgaard K, Frisch M. The \nco-occurrence of endometriosis with multiple sclerosis, systemic lupus \nerythematosus and Sjogren syndrome. Hum Reprod. 2011;26:1555–9. \nhttps:// doi. org/ 10. 1093/ humrep/ der105.\n 29. Porpora MG, Scaramuzzino S, Sangiuliano C, Piacenti I, Bonanni V, Piccioni \nMG, et al. High prevalence of autoimmune diseases in women with \nendometriosis: a case-control study. Gynecol Endocrinol. 2020;36:356–9. \nhttps:// doi. org/ 10. 1080/ 09513 590. 2019. 16557 27.\n 30. Sinaii N, Cleary SD, Ballweg ML, Nieman LK, Stratton P . High rates of \nautoimmune and endocrine disorders, fibromyalgia, chronic fatigue syn-\ndrome and atopic diseases among women with endometriosis: a survey \nanalysis. Hum Reprod. 2002;17:2715–24. https:// doi. org/ 10. 1093/ humrep/ \n17. 10. 2715.\n 31. Yoshii E, Yamana H, Ono S, Matsui H, Yasunaga H. Association between \nallergic or autoimmune diseases and incidence of endometriosis: A \nnested case-control study using a health insurance claims database. Am J \nReprod Immunol. 2021;86: e13486. https:// doi. org/ 10. 1111/ aji. 13486.\n 32. Chen SF, Yang YC, Hsu CY, Shen YC. Risk of Rheumatoid Arthritis in \nPatients with Endometriosis: A Nationwide Population-Based Cohort \nStudy. J Womens Health (Larchmt). 2021;30:1160–4. https:// doi. org/ 10. \n1089/ jwh. 2020. 8431.\n 33. Merlino LA, Cerhan JR, Criswell LA, Mikuls TR, Saag KG. Estrogen and other \nfemale reproductive risk factors are not strongly associated with the \ndevelopment of rheumatoid arthritis in elderly women. Semin Arthritis \nRheum. 2003;33:72–82. https:// doi. org/ 10. 1016/ s0049- 0172(03) 00084-2.\n 34. Xue YH, You LT, Ting HF, Chen YW, Sheng ZY, Xie YD, et al. Increased risk of \nrheumatoid arthritis among patients with endometriosis: a nationwide \npopulation-based cohort study. Rheumatology (Oxford). 2021;60:3326–33. \nhttps:// doi. org/ 10. 1093/ rheum atolo gy/ keaa7 84.\n 35. Chao YH, Liu CH, Pan YA, Yen FS, Chiou JY, Wei JC. Association Between \nEndometriosis and Subsequent Risk of Sjögren’s Syndrome: A Nationwide \nPopulation-Based Cohort Study. Front Immunol. 2022;13: 845944.  \nhttps:// doi. org/ 10. 3389/ fimmu. 2022. 845944.\n 36. Shigesi N, Kvaskoff M, Kirtley S, Feng Q, Fang H, Knight JC, et al. The asso-\nciation between endometriosis and autoimmune diseases: a systematic \nreview and meta-analysis. Hum Reprod Update. 2019;25:486–503.  \nhttps:// doi. org/ 10. 1093/ humupd/ dmz014.\n 37. McLeod BS, Retzloff MG. Epidemiology of endometriosis: an assessment \nof risk factors. Clin Obstet Gynecol. 2010;53:389–96. https:// doi. org/ 10. \n1097/ GRF. 0b013 e3181 db7bde.\n 38. Xiao XY, Chen Q, Shi YZ, Li LW, Hua C, Zheng H. Risk factors of systemic \nlupus erythematosus: an overview of systematic reviews and Mendelian \nrandomization studies. Adv Rheumatol. 2023;63:42. https:// doi. org/ 10. \n1186/ s42358- 023- 00323-1.\n 39. Kolanska K, Alijotas-Reig J, Cohen J, Cheloufi M, Selleret L, D’Argent E, \net al. Endometriosis with infertility: A comprehensive review on the role \nof immune deregulation and immunomodulation therapy. Am J Reprod \nImmunol. 2021;85: e13384. https:// doi. org/ 10. 1111/ aji. 13384.\n 40. Vanni VS, Villanacci R, Salmeri N, Papaleo E, Delprato D, Ottolina J, et al. \nConcomitant autoimmunity may be a predictor of more severe stages \nof endometriosis. Sci Rep. 2021;11:15372. https:// doi. org/ 10. 1038/ \ns41598- 021- 94877-z.\n 41. Khizroeva J, Nalli C, Bitsadze V, Lojacono A, Zatti S, Andreoli L, et al. Infertil-\nity in women with systemic autoimmune diseases. Best Pract Res Clin \nEndocrinol Metab. 2019;33: 101369. https:// doi. org/ 10. 1016/j. beem. 2019. \n101369.\n 42. Guo SW, Du Y, Liu X. Platelet-derived TGF-β1 mediates the down-modu-\nlation of NKG2D expression and may be responsible for impaired natural \nkiller (NK) cytotoxicity in women with endometriosis. Hum Reprod. \n2016;31:1462–74. https:// doi. org/ 10. 1093/ humrep/ dew057.\n 43. Kang YJ, Jeung IC, Park A, Park YJ, Jung H, Kim TD, et al. An increased \nlevel of IL-6 suppresses NK cell activity in peritoneal fluid of patients \n\nPage 11 of 11\nTang et al. BMC Pregnancy and Childbirth          (2024) 24:162 \n \nwith endometriosis via regulation of SHP-2 expression. Hum Reprod. \n2014;29:2176–89. https:// doi. org/ 10. 1093/ humrep/ deu172.\n 44. Yu JJ, Sun HT, Zhang ZF, Shi RX, Liu LB, Shang WQ, et al. IL15 promotes \ngrowth and invasion of endometrial stromal cells and inhibits killing \nactivity of NK cells in endometriosis. Reproduction. 2016;152:151–60. \nhttps:// doi. org/ 10. 1530/ REP- 16- 0089.\n 45. Berbic M, Schulke L, Markham R, Tokushige N, Russell P , Fraser IS. \nMacrophage expression in endometrium of women with and without \nendometriosis. Hum Reprod. 2009;24:325–32. https:// doi. org/ 10. 1093/ \nhumrep/ den393.\n 46. Milewski Ł, Dziunycz P , Barcz E, Radomski D, Roszkowski PI, Korczak-\nKowalska G, et al. Increased levels of human neutrophil peptides 1, 2, \nand 3 in peritoneal fluid of patients with endometriosis: association with \nneutrophils, T cells and IL-8. J Reprod Immunol. 2011;91:64–70. https:// \ndoi. org/ 10. 1016/j. jri. 2011. 05. 008.\n 47. Riccio L, Baracat EC, Chapron C, Batteux F, Abrão MS. The role of the B \nlymphocytes in endometriosis: A systematic review. J Reprod Immunol. \n2017;123:29–34. https:// doi. org/ 10. 1016/j. jri. 2017. 09. 001.\n 48. Dang VD, Stefanski AL, Lino AC, Dörner T. B- and Plasma Cell Subsets in \nAutoimmune Diseases: Translational Perspectives. J Invest Dermatol. \n2022;142:811–22. https:// doi. org/ 10. 1016/j. jid. 2021. 05. 038.\n 49. Fresneda AM, McLaren Z, Wright HL. Neutrophils in the Pathogenesis \nof Rheumatoid Arthritis and Systemic Lupus Erythematosus: Same Foe \nDifferent M.O. Front Immunol 2021;12:649693. https:// doi. org/ 10. 3389/ \nfimmu. 2021. 649693\n 50. Zhang C, Tian Z. NK cell subsets in autoimmune diseases. J Autoimmun. \n2017;83:22–30. https:// doi. org/ 10. 1016/j. jaut. 2017. 02. 005.\n 51. Marquardt RM, Kim TH, Shin JH, Jeong JW. Progesterone and Estrogen \nSignaling in the Endometrium: What Goes Wrong in Endometriosis? Int J \nMol Sci 2019;20. https:// doi. org/ 10. 3390/ ijms2 01538 22\n 52. Dolmans MM, Donnez J. Emerging Drug Targets for Endometriosis. \nBiomolecules 2022;12. https:// doi. org/ 10. 3390/ biom1 21116 54\n 53. Merrheim J, Villegas J, Van Wassenhove J, Khansa R, Berrih-Aknin S, le \nPanse R, et al. Estrogen, estrogen-like molecules and autoimmune dis-\neases. Autoimmun Rev. 2020;19: 102468. https:// doi. org/ 10. 1016/j. autrev. \n2020. 102468.\n 54. Hughes GC, Choubey D. Modulation of autoimmune rheumatic diseases \nby oestrogen and progesterone. Nat Rev Rheumatol. 2014;10:740–51. \nhttps:// doi. org/ 10. 1038/ nrrhe um. 2014. 144.\n 55. Pinto AJ, Roschel H, de Sá PA, Lima FR, Pereira R, Silva CA, et al. Physical \ninactivity and sedentary behavior: Overlooked risk factors in autoimmune \nrheumatic diseases? Autoimmun Rev. 2017;16:667–74. https:// doi. org/ 10. \n1016/j. autrev. 2017. 05. 001.\nPublisher’s Note\nSpringer Nature remains neutral with regard to jurisdictional claims in pub-\nlished maps and institutional affiliations.","source_license":"CC0","license_restricted":false}