{"paper_id":"1b18c603-6423-4629-9b2a-1774863a0c92","body_text":"R E S E A R C H A R T I C L E Open Access\nGastrointestinal symptoms among\nendometriosis patients —A case-cohort\nstudy\nMalin Ek 1*, Bodil Roth 1, Per Ekström 2, Lil Valentin 2, Mariette Bengtsson 3 and Bodil Ohlsson 1\nAbstract\nBackground: Women with endometriosis often experience gastrointestinal symptoms. Gonadotropin-releasing\nhormone (GnRH) analogs are used to treat endometriosis; however, some patients develop gastrointestinal\ndysmotility following this treatment. The aims of the present study were to investigate gastrointestinal symptoms\namong patients with endometriosis and to examine whether symptoms were associated with menstruation,\nlocalization of endometriosis lesions, or treatment with either opioids or GnRH analogs, and if hormonal treatment\naffected the symptoms.\nMethods: All patients with diagnosed endometriosis at the Department of Gynecology were invited to participate\nin the study. Gastrointestinal symptoms were registered using the Visual Analogue Scale for Irritable Bowel\nSyndrome (VAS-IBS); socioeconomic and medical histories were compiled using a clinical data survey. Data were\ncompared to a control group from the general population.\nResults: A total of 109 patients and 65 controls were investigated. Compared to controls, patients with endometriosis\nexperienced significantly aggravated abdominal pain (P = 0.001), constipation (P = 0.009), bloating and flatulence\n(P = 0.000), defecation urgency (P = 0.010), and sensation of incomplete evacuation (P = 0.050), with impaired\npsychological well-being (P = 0.005) and greater intestinal symptom influence on their daily lives ( P = 0.001). The\nsymptoms were not associated with menstruation or localization of endometriosis lesions, except increased nausea\na n dv o m i t i n g(P = 0.010) in patients with bowel-associated lesions. Half of the patients were able to differentiate\nbetween abdominal pain from endometriosis and from the gastrointestinal tract. Patients using opioids experienced\nmore severe symptoms than patients not using opioids, and patients with current or previous use of GnRH analogs\nhad more severe abdominal pain than the other patients ( P = 0.024). Initiation of either combined oral contraceptives\nor progesterone for endometriosis had no effect on gastrointestinal symptoms when the patients were followed\nprospectively.\nConclusions: The majority of endometriosis patients experience more severe gastrointestinal symptoms than controls.\nA poor association between symptoms and lesion localization was found, indicating existing comorbidity between\nendometriosis and irritable bowel syndrome (IBS). Treatment with opioids or GnRH analogs is associated with\naggravated gastrointestinal symptoms.\nKeywords: Abdominal pain, Endometriosis, Gastrointestinal symptoms, Gonadotropin-releasing hormone,\nMenstruation, Opioids\n* Correspondence: malin.ek@med.lu.se\n1Department of Clinical Sciences, Division of Internal Medicine, Skåne\nUniversity Hospital, Lund University, 205 02, Malmö, Sweden\nFull list of author information is available at the end of the article\n© 2015 Ek et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0\nInternational License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and\nreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a\nlink to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain\nDedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this\narticle, unless otherwise stated.\nEk et al. BMC Women's Health  (2015) 15:59 \nDOI 10.1186/s12905-015-0213-2\n\nBackground\nEndometriosis is a benign, gynecological disease associated\nwith the primary symptoms of chronic pelvic pain, deep\ndyspareunia, and dysmenorrhea [1]. The prevalence of\nendometriosis differs in the literature, but is estimated to\naffect approximately 7 –10 % of women [2]. Clinically,\nwomen with endometriosis commonly experience\ngastrointestinal symptoms, and one study has shown\nthat gastrointestinal symptoms are almost as common\nas gynecological symptoms in these patients [3].\nGastrointestinal symptoms among patients with endo-\nmetriosis described in the literature include abdominal\npain, bloating, nausea, constipation, vomiting, painful bowel\nmovements, and diarrhea [3 –5]. However, reported symp-\ntoms differ between studies. Aggravated symptoms during\nmenstruation have been reported [4, 6, 7] such as cyclic-\nrelated bloating and constipation [4]. Fauconnier et al. [7]\nconcluded that symptoms including diarrhea, constipa-\ntion, and colic rectal pain were more frequent among\npatients with endometriosis lesions within or close to\nthe bowel. In contrast, Maroun et al. [3] reported gastro-\nintestinal symptoms to be primarily independent of\nlocalization of endometrios is lesions in relation to the\nbowel. Different explanationsconcerning the occurrence of\nthese symptoms include: endom etriosis lesions cause in-\nflammatory activity and local prostaglandin release, which\ncan alter bowel function [8]; endometriosis lesions within\nthe bowel cause symptoms due to mechanical obstruction\nor cyclic micro-hemorrhages [9]; or there is an existing co-\nmorbidity between endometr iosis and irritable bowel\nsyndrome (IBS) [8].\nGonadotropin-releasing hormone (GnRH) is a hypothal-\namic hormone [10], which has also been shown to be\npresent in neurons in the human enteric nervous system\n(ENS) [11]. Recent studies have suggested a link between\nGnRH and gastrointestinal function [12, 13], and some\npatients develop severe dysmotility in the form of chronic\nintestinal pseudo-obstruction(CIPO) or enteric dysmotility\n(ED) after treatment with GnRH analogs in relation to in\nvitro fertilization (IVF) or endometriosis [14]. Full-\nthickness biopsies of the bowel wall have shown enteric\nneurodegeneration with almost total absence of GnRH-\ncontaining neurons [11, 14, 15]. Hammar et al. [13] investi-\ngated 124 patients before and after treatment with GnRH\nanalogs and concluded that there was a significant exacer-\nbation of gastrointestinal symptoms after treatment, and\nabdominal pain was still exacerbated at 5-year follow-up.\nThe aim of the present study was to investigate the se-\nverity of gastrointestinal symptoms among patients with\nendometriosis compared to a control group from the\ngeneral population. Furthermore, an association between\nsymptoms and menstruation, localization of endometriosis\nlesions, and treatment with opioids or GnRH analogs was\ninvestigated. The final aim was to investigate women\ninitiating hormonal treatment to examine whether this\ntreatment had an impact on gastrointestinal symptoms over\ntime.\nMethods\nThis study was approved by the Ethics Review Board of\nLund University, Dnr 2012/564, and performed in ac-\ncordance with the declaration of Helsinki. All subjects\ngave their written, informed consent before inclusion in\nthe study.\nPatients\nPatients who had sought treatment for endometriosis in\nthe past 5 years were recruited from the Department of\nGynecology at Skåne University Hospital in Malmö.\nThe patients were identified with the International\nStatistical Classification of D iseases and Related Health\nProblems, ICD-10, N80.1, 80.4, 80.5, 80.8, and 80.9\nfrom Skåne University Hospital medical records. The\nclinic ’s catchment area is the southernmost districts of\nSweden. The recruitment was conducted continuously\nfrom March 2013 through July 2014. Inclusion criteria\nwere a diagnosis of endometriosis made by laparos-\ncopy, ultrasonography, or a definite clinical diagnosis\nmade by a gynecologist. The included patients were also\nrequired to comprehend the Swedish or English language.\nExclusion criteria were an uncertain diagnosis of endo-\nmetriosis, patients living too far from the geographical\narea of Skåne University Hospital, multi-morbidity,\ncurrent pregnancy, or a diagnosis of cancer, inflamma-\ntory bowel disease, multiple sclerosis, psychiatric dis-\nease, or rheumatoid arthritis.\nControls\nData regarding control subjects was obtained from a\nprevious study conducted by Hammar et al. [13]. In this\nstudy, control subjects were randomly acquired from the\nSwedish Population Registry, and 248 subjects were con-\ntacted. In total, after one reminder, 29 questionnaires\nwere returned. Because of the low response rate, further\ncontrols were recruited amongst female hospital staff. In\ntotal, 65 women representing the general population,\nwith a mean age of 40 ± 9 years, were recruited and\ncompleted the Visual Analogue Scale for Irritable Bowel\nSyndrome (VAS-IBS).\nStudy design\nThe patients were contacted via mail, and within a week,\nthey were also contacted via telephone. After agreement\nto participate in the study, an appointment for an inter-\nview and a blood draw was done 1 –4 weeks after inclu-\nsion in the study. The questionnaire, Visual Analogue\nScale for Irritable Bowel Syndrome (VAS-IBS), and a\nclinical data survey were sent via mail, with instructions\nEk et al. BMC Women's Health  (2015) 15:59 Page 2 of 10\n\nto complete these questionnaires at a maximum of 1\nweek before the appointment. At the hospital visit, all\npatients were interviewed regarding previous treatment\nfor endometriosis and their position in the menstrual\ncycle at the time of completing the VAS-IBS. The pa-\ntients who were included at the first contact with the\nDepartment of Gynecology also completed the VAS-IBS\nquestionnaire and the clinical data survey, were inter-\nviewed and had blood samples drawn at 3 and 6 months\nafter their first visit.\nA review of the patients ’ medical journals was con-\nducted to investigate the localization of endometriosis\nlesions, current hormonal treatment, and whether they\nhad undergone diagnostic or operative laparoscopy. A\nlesion was confirmed when it was seen macroscopically\nduring laparoscopy, visualized by ultrasonography, or in\na few cases by palpation conducted by a gynecologist.\nQuestionnaires\nClinical data survey\nThe clinical data survey addresses socioeconomic fac-\ntors, physical exercise, nicotine- and alcohol habits,\ncurrent diseases, and medication, as well as questions\nabout gastrointestinal symptoms such as onset and trig-\ngers, whether the subject can differentiate between the\nabdominal pain from endometriosis and the gastrointes-\ntinal tract, and whether pharmacological treatment was\nused because of the complaints.\nThree questions were added after the study was initi-\nated: which year endometriosis-associated symptoms\nbegan; which year gastrointestinal symptoms began; and\nwhether the patient could differentiate symptoms from\nendometriosis and symptoms from the gastrointestinal\ntract. Therefore, an additional survey with these ques-\ntions was sent by mail to the 21 first participants already\nincluded in the study.\nVisual Analogue Scale for Irritable Bowel Syndrome\nThe VAS-IBS was used to investigate gastrointestinal com-\nplaints in the study groups. It is a validated questionnaire\nfor estimation of the most co mmon gastrointestinal com-\nplaints in patients with non-organic, functional bowel dis-\nease [16]. This scale has also b een validated for estimation\nof symptoms over time [17]. The seven items measured in\nthe VAS-IBS address the symp toms abdominal pain, diar-\nrhea, constipation, bloating and flatulence, nausea and\nvomiting, psychological well- being, and intestinal symp-\ntoms’ influence on daily life. These items were measured\non a scale from 0 to 100, where 0 represents severe prob-\nlems and 100 represents a complete lack of problems. An\nadditional two questions, i f the subject experienced\ndefecation urgency and had a sensation of incomplete\nevacuation, were answered with yes or no.\nStatistical methods\nThe data was analyzed using the statistical software\npackage SPSS for Windows (release 22.0; IBM). Because\nthe controls were slightly older than the endometriosis\npatients, variables were age-standardized using a linear\nregression model into which age was added as a covari-\nate and the variables were then expressed as z-scores.\nWhen comparing VAS scores between groups, the age-\nstandardized values were used. All variables were analyzed\nfor normal distribution using the Kolomogorov-Smirnov\ntest. As normality was rejected, except for age, the Mann-\nWhitney U-test and the Fisher ’s exact test were used to\ncompare different groups. Student t-test was used to per-\nform a failure analysis. To calculate differences within the\ngroup, Friedman’s test and Wilcoxon’s signed rank test were\nused. Spearman’s correlation test was used for correlations\nbetween age and symptoms. Values were expressed as\nmean ± standard deviation (SD), median [interquartile\nrange (IQR)], or number (n) and percent (%). P ≤ 0.05 was\nconsidered statistically significant.\nResults\nPatient characteristics\nA total of 627 patients with suspected endometriosis\nwere identified. Of these, 320 were excluded because\nthey either did not fulfill the inclusion criteria or they\nfulfilled the exclusion criteria. The most common reason\nfor exclusion was uncertainty about the diagnosis. Then,\n307 patients remained who fulfilled the inclusion cri-\nteria. Of these, 198 patients were excluded since 144\nwere not willing to participate, 49 had moved from the\nregion, and 4 denied the diagnosis when contacted.\nThus, 109 could finally be included in the study. Eighteen\nof these patients were followed prospectively regarding\ntheir gastrointestinal symptoms, and 14 of these initiated a\nnew hormonal treatment for endometriosis during the\nstudy period.\nThe mean age of the 109 included patients was 37 ±\n7 years. The median duration of endometriosis was 8.5\n(4.0–16.0) years. Patients who were unwilling to participate\nin the study were younger (35 ± 6 years) than those who\nparticipated (P = 0.014). Almost two-thirds of the patients\nhad a degree from a university or a college. The vast major-\nity had never smoked and consumed less than one standard\nglass of alcohol a week (T able 1). Aside from endometriosis\nt h em a j o r i t yo fp a t i e n t sw e r eh e a l t h y(n =5 7 ) ,a n dt h e m o s t\ncommon diseases reported were allergies or bronchial\nasthma ( n = 18) followed by migraine ( n =7 ) a n d i r r i t a b l e\nbowel syndrome (IBS) (n =6 ) .\nAlmost all patients had received hormonal treatment\nfor endometriosis, most commonly combined oral con-\ntraceptives (Table 2). Of the 109 patients, 87 had\nlaparoscopically-verified endometriosis, 21 had received\ntheir diagnosis by ultrasonography, and only one was\nEk et al. BMC Women's Health  (2015) 15:59 Page 3 of 10\n\ndiagnosed clinically. The most common localization of\nendometriosis lesions was in the ovaries ( n = 83),\nfollowed by the peritoneum ( n = 21), the bowel ( n = 18),\nand the Pouch of Douglas ( n = 16).\nPatients’ gastrointestinal complaints\nOf the patients, 85 % reported gastrointestinal complaints\nduring the past year. The onset of symptoms had been\ngradual for a majority. The median duration of gastro-\nintestinal symptoms was 5 months shorter than the me-\ndian duration of endometriosis. One-third of the patients ’\ncomplaints had been diagnosed as IBS or endometriosis,\nbut one out of five reported never having received any\ndiagnosis for the symptoms. However, almost half of the\npatients reported having the ability to differentiate be-\ntween pain symptoms from endometriosis and from the\ngastrointestinal tract. A variety of different triggers for the\nsymptoms were described (T able 3). Nearly half of the pa-\ntients had received pharmacological treatment for their\nTable 1 Patient characteristics\nPatients with\nendometriosis n = 109\nAge (years) 36.78 ± 7.39\nBMI (kg/m2), missing value = 4 24.00 (22.00 –26.00)\nMarital status (n, %), missing value = 1\nLiving alone 36, 33.0\nMarried or living with a partner 70, 64.2\nLiving with parents 2, 1.8\nEducation (n, %)\nNot completed elementary school 1, 0.9\nCompleted elementary school 2, 1.8\nCompleted secondary school 20, 18.3\nMore than a year of further education\nafter secondary school\n24, 22.0\nDegree from university or college 62, 56.9\nOccupation (n, %), missing value = 1\nWork full time 62, 56.9\nWork 99–51 % 22, 20.2\nWork 1–50 % 8, 7.3\nOn sick-leave 5, 4.6\nRetired 1, 0.9\nUnemployed 1, 0.9\nStudent 9, 8.3\nSmoking habits (n, %)\nSmoke regularly 8, 7.3\nSmoke occasionally 10, 9.2\nDo not smoke, but have been a smoker 21, 19.3\nHave never smoked 70, 64.2\nSnuff habits (n, %)\nSnuff regularly 6, 5.5\nDo not snuff but have been a regular snuff\nuser\n3, 2.8\nHave never used snuff 100, 91.7\nAlcohol use (n, %)\nLess than one standard glass a week 68, 62.4\n1–4 standard glasses a week 31, 28.4\n5–9 standard glasses a week 10, 9.2\nPhysical exercise (n, %)\nNo time 18, 16.5\nLess than 30 minutes a week 16, 14.7\n30–90 minutes a week 37, 33.9\nMore than 90 minutes a week 38, 34.9\nBMI = body mass index; n, % = number and percentage of patients. One\nstandard glass of alcohol contains 12 grams of pure alcohol. Physical exercise\nwas defined as physical activity that results in shortness of breath. Values are\ngiven as mean ± standard deviation, median [interquartile range (IQR)] and\nnumber, %\nTable 2 Hormonal treatment of endometriosis\nPatients with\nendometriosis\nn = 109\nPrevious or current hormonal treatment (n)\nAny type of hormonal treatment 106\nCombined oral contraceptives 90\nGnRH analogs 55\nProgesterone 55\nEstrogen 2\nCurrent hormonal treatment (n), missing value = 23\nCombined oral contraceptives 27\nProgesterone 24\nNo hormonal treatment 24\nGnRH analogs 9\nEstrogen 2\nOnset of gastrointestinal symptoms\nafter hormonal treatment (Yes/No/Unknown)\n23/27/59\nType of hormonal treatment causing symptoms (n)\nAfter/during IVF 6\nAfter cessation of combined oral\ncontraceptives\n5\nAfter progesterone medication 3\nAfter or during usage of an IUD 2\nAfter GnRH treatment 2\nAfter Pergotime (Klomifen) treatment 1\nIVF = in vitro fertilization; GnRH = gonadotropin-releasing hormone;\nIUD = intrauterine device; n = number of patients. Patients presented can have\nhad more than one hormonal treatment and have stated more than one\nhormonal treatment as the onset trigger for gastrointestinal symptoms.\nPatients assigned to the group “GnRH analogs ” are those who reported GnRH\nanalogs when asked about hormonal treatment and those who reported\nhaving undergone IVF. Values are given as number\nEk et al. BMC Women's Health  (2015) 15:59 Page 4 of 10\n\ngastrointestinal complaints; slightly less than one out of five\nhad been treated with opioids, and almost half of the pa-\ntients had been treated with other drugs than opioids, for\nexample nonsteroidal anti-in flammatory drugs (NSAID),\nparacetamol, or anti-depressant medication. However, in\nmany cases, it was not obvious whether the anti-depressant\ndrugs were given for treatment of abdominal pain or de-\npression (T able 4).\nGastrointestinal symptoms measured by the Visual\nAnalogue Scale for Irritable Bowel Syndrome\nPatients with endometriosis experienced aggravated symp-\ntoms compared to controls regarding abdominal pain,\nconstipation, bloating and flatulence, and impaired psy-\nchological well-being and greater influence of intestinal\nsymptoms on daily life. The endometriosis patients also ex-\nperienced defecation urgency and a sensation of incomplete\nevacuation significantly more often (T able 5). Increasing pa-\ntient age correlated with less diarrhea ( rs =0 . 2 8 1 ,P = 0.03),\nless bloating and flatulence (rs = 0.239, P =0 . 0 1 3 ) ,l e s sn a u -\nsea and vomiting (rs =0 . 2 8 6 ,P = 0.003), and better psycho-\nlogical well-being ( rs =0 . 2 9 5 ,P = 0.002), whereas age in\ncontrols did not correlate with symptoms (data not shown).\nThe patients with lesions within the bowel, or in close\nproximity to the bowel (the pouch of Douglas, the posterior\nwall of the vagina, or the recto-vaginal septum) ( n = 34),\nscored more severe symptoms regarding nausea and vomit-\ning than the other patients [69.71 (33.75 –100.00) vs. 84.36\n(75.00–100.00); P = 0.010]. None of the other symptoms\ndiffered depending on localization of the lesions (data not\nshown).\nThe patients currently using opioids scored more se-\nvere symptoms than the other patients regarding all\nsymptoms except sensation of incomplete evacuation\n(Table 6). The patients who were not currently using opi-\noids scored more severe symptoms than controls regarding\nabdominal pain [66.12 (45.00 –100.00) vs. 81.40 (73.00 –\n98.00); P = 0.013], bloating and flatulence [53.96 (20.00 –\n90.00) vs. 71.72 (41.75 –96.00); P = 0.005], intestinal symp-\ntoms’ influence on daily life [66.01 (35.00–100.00) vs. 81.28\n(75.00–98.00); P = 0.011], and sensation of incomplete\nevacuation [45 (47.4 %) vs. 17 (26.2 %); P = 0.016]. The fol-\nlowing parameters all showed tendency towards significant\ndifferences between patients not using opioids and controls:\nconstipation [68.88 (45.00–100.00) vs. 84.47 (83.00–98.00);\nP = 0.051], psychological well-being [68.74 (50.00 –100.00)\nvs. 82.07 (74.00–96.00); P = 0.064], and defecation urgency\n[ 2 9( 3 0 . 5% )v s .1 1( 1 6 . 9% ) ;P = 0.066].\nThe patients who currently received hormonal treat-\nment for endometriosis (combined oral contraceptives,\nprogesterone, estrogen, or GnRH analogs) ( n =6 2 )d i dn o t\nexperience more severe gastrointestinal symptoms than\nthe other patients ( n = 24) (data not shown). However,\nwhen comparing patients with current or previous GnRH\nTable 3 Characterization of gastrointestinal symptoms\nPatients with\nendometriosis\nn = 109\nOnset of gastrointestinal symptoms (n, %),\nmissing value = 19\nGradual 70, 64.2\nRapid 20, 18.3\nDuration of gastrointestinal complaints (years) 8.00 (4.00 –15.50)\nThe evolution of symptoms over time (n, %),\nmissing value = 17\nConstant complaints 14, 12.8\nIntermittent complaints 45, 41.3\nProgressively worse 33, 30.3\nGastrointestinal complaints during the last year\n(Yes/No)\n93/16\nAbility to differentiate between pain due to\nendometriosis and from the gastrointestinal tract\n(Yes/No/Unknown)\n53/32/24\nHas changed occupation because of complaints\n(Yes/No/Unknown)\n5/101/3\nRelatives with similar complaints (Yes/No/Unknown) 46/47/16\nExperience a trigger for the complaints\n(Yes/No/Unknown)\n35/55/19\nTriggers for gastrointestinal complaints (n)\nMenstruation 13\nStress 6\nAbdominal surgery 4\nCessation of combined oral contraceptives 4\nEndometriosis 3\nSide-effects from medications 2\nFood 1\nSexual activity 1\nPregnancy 1\nGastrointestinal infection 1\nDiagnosis (n)\nHave not been given any name 21\nEndometriosis 16\nIBS 16\nGastritis 6\nSensitive stomach/bowels 3\nChronic constipation 2\nReflux, dyspepsia 2\nStress 1\nAllergy 1\nBowel spasm 1\nStricture 1\nIBS = irritable bowel syndrome, n, % = number and percentage of subjects.\nPatients presented can have more than one trigger for gastrointestinal\nsymptoms and more than one diagnosis ascribed to the symptom. Values are\ngiven as median [interquartile range (IQR)] and number, %\nEk et al. BMC Women's Health  (2015) 15:59 Page 5 of 10\n\ntreatment ( n = 55) towards the other patients ( n =5 1 ) ,\nthere was a higher degree of abdominal pain among the\nfirst group [55.35 (25.00 –91.25) vs. 69.51 (50.00 –100.00);\nP = 0.024]. Among those who had no hormonal treatment,\nthe patients who were currently menstruating ( n =3 ) ,\nwere compared to those who did not menstruate ( n =2 1 ) ,\nand no significant differences regarding symptoms be-\ntween the groups were found (data not shown).\nProspective follow-up of gastrointestinal symptoms\nduring treatment\nOf the 14 patients initiating a new hormonal treatment\nfor endometriosis, the only significant effect was im-\nproved psychological well-being from 0 to 3 months,\nfollowed by impaired well-being from 3 to 6 months\n(Friedman’s test; P = 0.006). When performing subgroup\nanalyses, the patients who had initiated a new treatment\nwith combined oral contraceptives ( n = 5) did not ex-\nperience any significant reduction in any of the symp-\ntoms over time (data not shown). Subgroup analyses in\npatients who had initiated treatment with progesterone\n(n = 6) showed a significant effect regarding psycho-\nlogical well-being, which had improved at the 3-month\nfollow-up but was impaired at the 6-month follow-up\n(Friedman’s test; P = 0.016).\nDiscussion\nCompared to controls, endometriosis patients experi-\nence more abdominal pain, constipation, bloating and\nflatulence, influence of intestinal symptoms on daily life,\ndefecation urgency, sensation of incomplete evacuation,\nand decreased psychological well-being. Localization of\nendometriosis lesions showed no association with symp-\ntoms, except lesions within or close to the bowel, which\nwere associated with more nausea and vomiting. Patients\nusing opioids had more severe symptoms than patients\nnot using opioids, and patients treated with GnRH ana-\nlogs had more abdominal pain than the other patients. A\nprospective analysis of patients initiating a new hormo-\nnal treatment of endometriosis during the study showed\nno impact from treatment of gastrointestinal symptoms\nover time. In contrast, prior studies have shown that\nprogestins relief gastrointestinal symptoms related to the\nmenstrual cycle, as well as overall diarrhea and intestinal\ncramping [18]. However, the latter study could not de-\ntect any effect of progestins on overall constipation, ab-\ndominal bloating and feeling of incomplete evacuation\n[18]. The difference between studies is that the present\nTable 4 Examinations, medical treatment, and surgery due to\ngastrointestinal symptoms\nPatients with\nendometriosis\nn = 109\nExaminations conducted because of gastrointestinal\nsymptoms (n)\nDiagnostic laparoscopy (without surgery) 30\nNo examinations conducted 23\nUltrasonography 17\nColonoscopy 13\nGastroscopy 12\nUnspecified examinations (including x-ray and\nendoscopy)\n9\nRectoscopy 8\nPrevious or current pharmacological treatment\nbecause of gastrointestinal complaints\n(Yes/No/Unknown)\nOpioids 20/88/1\nOther analgesic treatment than opioids 21/78/10\nOther pharmacological treatment than analgesics 29/70/10\nTypes of pharmacological treatment other than opioids\nfor gastrointestinal complaints (n)\nNSAIDs 16\nBulking agents or laxatives 11\nPPIs 10\nParacetamol 9\nLoperamide 2\nGabapentin or pregabalin 2\nNatural remedies or probiotics 2\nCyanocobalamin 1\nDiazepam 1\nPapaverine 1\nTypes of current medication (n)\nCombined oral contraceptives 23\nNSAIDs 20\nAntidepressants (including SSRIs, venlafaxin,\nand mirtazapin)\n19\nOpioids 14\nParacetamol 14\nAllergy and asthma medication 11\nHave undergone abdominal surgery (Yes/No) 98/11\nType of abdominal surgery (n)\nLaproscopic surgery (because of endometriosis) 71\nUnspecified 43\nSectio 11\nAppendectomy 8\nHysterectomy 8\nRemoval of ovaries and/or oviducts 6\nNSAID = non-steroidal anti-inflammatory drugs; SSRIs = selective serotonin\nre-uptake inhibitors; PPIs = proton pump inhibitors; GnRH = gonadotropin\nreleasing hormone. The six most common currently used medications and\nabdominal surgery are reported. Patients could have undergone more than\none examination, used more than one medication, and had more than one\ntype of abdominal surgery conducted. Values are given as number\nEk et al. BMC Women's Health  (2015) 15:59 Page 6 of 10\n\npatients did not experience diarrhea. Thus, hormonal\ntreatment may have its main effect on endometriosis\nand endometriosis-related symptoms, and less effect on\noverall gastrointestinal symptoms.\nTo the best of our knowledge, this is the first study\nthat has investigated gastrointestinal symptoms among\npatients with endometriosis using a continuous scale,\nthe VAS-IBS. Also, to the best of our knowledge, no\nstudy has investigated the symptoms in respect to use of\nopioids and GnRH analogs.\nThe findings of the present study partially support the\nfindings of previous studies in regard to bloating, ab-\ndominal pain, and constipation among endometriosis\npatients [3, 5]. However, in other studies, diarrhea, nausea,\nand vomiting were reported to be common [3, 4]; however,\nthese symptoms were not more frequent among patients\nTable 5 Gastrointestinal symptoms measured by the Visual Analogue Scale for Irritable Bowel Syndrome\nPatients with\nendometriosis\nn = 109\nControls\nn =6 5\nP value\nAge (years) 36.78 ± 7.39 40.20 ± 8.75 0.036\nAbdominal pain\nZ-score −0.68 (−2.12–0.64) 0.30 ( −0.43–0.70) 0.001\nAbsolute score 65.00 (30.00 –100.00) 93.00 (73.00 –98.00)\nMissing value 2 8\nDiarrhea\nZ-score 0.25 ( −1.18–0.69) 0.44 ( −0.02–0.65) 0.617\nAbsolute score 87.00 (45.00 –100.00) 95.00 (80.00 –98.00)\nMissing value 2 10\nConstipation\nZ-score −0.29 (−2.11–0.57) 0.43 (0.03 –0.58) 0.009\nAbsolute score 80.00 (30.00 –100.00) 95.00 (83.00 –98.00)\nMissing value 2 8\nBloating and flatulence\nZ-score −0.94 (−1.8–0.46) 0.41 ( −1.08–0.85) 0.000\nAbsolute score 45.00 (20.00 –85.00) 81.50 (41.75 –96.00)\nMissing value 2 7\nNausea and vomiting\nZ-score 0.33 ( −0.95–0.50) 0.37 (0.11 –0.48) 0.284\nAbsolute score 95.00 (65.00 –100.00) 98.00 (92.00 –99.00)\nMissing value 2 10\nPsychological\nwell-being\nZ-score −0.52 (−2.54–0.91) 0.19 ( −0.57–0.86) 0.005\nAbsolute score 75.00 (40.00 –95.00) 85.00 (74.00 –96.00)\nMissing value 2 4\nInfluence on daily life\nZ-score −0.42 (−1.83–0.57) 0.35 ( −0.27–0.67) 0.001\nAbsolute score 70.00 (30.00 –100.00) 96.00 (75.00 –98.00)\nMissing value 2 7\nDefecation urgency\n(Yes/No/Unknown) 37/66/6 11/52/2 0.010\nSensation of incomplete\nevacuation\n(Yes/No/Unknown) 53/50/6 17/46/2 0.050\nZ-score = standard score. The z-scores were used for calculations with the Mann –Whitney U-test. Fisher ’s exact test was used to calculate dichotomous variables.\nP ≤ 0.05 was considered statistically significant. Values are given as mean ± standard deviation, median [interquartile range (IQR)], and number\nEk et al. BMC Women's Health  (2015) 15:59 Page 7 of 10\n\nthan controls in the current study. Previous studies have\nreported exacerbation of symptoms during menstru-\nation [4, 6, 7]. This phenomenon does not differentiate be-\ntween endometriosis and IBS, since not only abdominal\npain due to endometriosis, but also gastrointestinal symp-\ntoms due to functional bowel diseases have been shown to\nincrease during menstruation [19]. Symptoms were not\nf o u n dt ob em o r es e v e r ed u r i n gm e n s t r u a t i o ni nt h i ss t u d y ,\nwhich can depend on few menstruating ( n = 3) vs. non-\nmenstruating ( n = 21) patients in the group of non-\nhormonal treatment, since some patients experienced men-\nstruation as a trigger factor for gastrointestinal symptoms.\nSymptoms including diarrhea, constipation, and colic rec-\ntal pain have been reported to be more common among\npatients with endometriosis lesions within or close to the\nbowel than among patients with distant lesions [7, 9]. How-\never, other studies have not found this association [3]. In\nthe current study, nausea and vomiting were associated\nwith bowel-related lesions, but no other symptom associ-\nation was detected. These findings, and the fact that hor-\nmonal endometriosis treat ment did not influence\ngastrointestinal symptoms over time, indicate an exist-\ning comorbidity between endometriosis and IBS.\nOne explanation for this possible comorbidity is an\nimmunological link, involving increased mast cell activa-\ntion. In IBS, colonic mast cell activation and mediator\nrelease close to mucosal innervation have been reported\n[20], and in deep infiltrating endometriosis, the presence\nof an increased number of activated mast cells close to\nnerves have been described [21]. Issa et al. [22] suggested\nthat the visceral hypersensitivity found in both patients with\nIBS and endometriosis could amplify gastrointestinal symp-\ntoms in patients with endometriosis. A hormonal link is\nalso plausible; GnRH-containing neurons [11] and recep-\ntors for LH [23] have been shown to be present in the\nhuman gastrointestinal tract as well as in the pelvic organs\n[24, 25]. IBS is predominantly diagnosed in women, with a\nfemale to male ratio ranging from 2:1 to 4:1, depending on\nt h ed i a g n o s t i cp r o c e d u r e[26]. Female sex hormones have\nbeen suggested to be involved in IBS-associated pain, since\nf l u c t u a t i o ni nI B Ss y m p t o m sh a sb e e nr e p o r t e dw i t he x -\nacerbation of abdominal pain during menstruation [19].\nHowever, the mechanisms underlying these findings are un-\nclear [26]. Exacerbation of ga strointestinal symptoms in\nendometriosis patients during menstruation has also been\nreported [4, 6, 7]. Another explanation for the symptoms\ncould be a yet undefined gastrointestinal disease or undis-\ncovered endometriosisin the intestinal wall.\nUse of opioids is known to cause gastrointestinal\nsymptoms such as constipation, nausea, and abdominal\npain, in severe cases referred to as narcotic bowel syn-\ndrome [27]. Almost 20 % of the patients had been pre-\nscribed opioids because of abdominal pain, and patients\nwith current opioid use had more severe symptoms on\nTable 6 Gastrointestinal complaints among patients with endometriosis in relation to opioid use\nPatients with current\nopioid use n =1 3\nPatients with no\ncurrent opioid\nuse n =9 4\nP value\nAbdominal pain\nAbsolute score 30.00 (10.00 –57.50) 70.00 (45.00 –100.00) 0.002\nDiarrhea\nAbsolute score 75.00 (19.00 –93.50) 90.00 (60.00 –100.00) 0.046\nConstipation\nAbsolute score 25.00 (10.00 –48.00) 80.00 (45.00 –100.00) 0.002\nBloating and flatulence\nAbsolute score 25.00 (15.50 –40.00) 50.00 (20.00 –90.00) 0.012\nNausea and vomiting\nAbsolute score 50.00 (20.00 –65.00) 100.00 (75.00 –100.00) 0.000\nPsychological well-being\nAbsolute score 30.00 (5.00 –57.50) 75.00 (50.00 –100.00) 0.000\nInfluence on daily life\nAbsolute score 25.00 (7.50 –60.00) 75.00 (35.00 –100.00) 0.002\nDefecation urgency\n(Yes/No/Unknown)\n8/4/2 29/62/4 0.015\nSensation of incomplete\nevacuation (Yes/No)\n8/4/2 45/46/4 0.145\nCalculations were made using the Mann Whitney U-test. Fisher ’s exact test was used to calculate dichotomous variables. P ≤ 0.05 was considered statistically\nsignificant. Values are given as median [interquartile range (IQR)] and number\nEk et al. BMC Women's Health  (2015) 15:59 Page 8 of 10\n\nalmost all parameters measured by the VAS-IBS than pa-\ntients not using opioids. One explanation to these find-\nings is that the patients with severe symptoms were\nprescribed opioids, but another explanation is that opi-\noids could cause or aggravate symptoms. A conclusion\nto be drawn is that many endometriosis patients may\nnot benefit from opioid treatment. Perhaps, opioids have\nno or weak positive impact on gastrointestinal symp-\ntoms; they possibly may even exacerbate them. Thus, all\ninitiating of opioid treatment in this patient group\nmust be carefully evaluated before continued. When\nexcluding current opioid-users, the endometriosis pa-\ntients still had more severe symptoms than controls,\neven though the P values were slightly higher than 0.05\nin some parameters.\nGnRH has been shown to be present in the ENS [11]\nand has been linked to gastrointestinal function [12, 13],\nbut its role in gut physiology and pathophysiology has\nnot been completely elucidated. Hammar et al. [13] re-\nported increased symptoms of constipation, as well as\nnausea and vomiting among women after treatment with\nGnRH analogs, with exacerbation of abdominal pain\nshowing a tendency towards significance. At 5-year\nfollow-up, the patients had more abdominal pain than they\nexperienced prior to GnRH treatment [13], and some pa-\ntients have developed severe dysmotility secondary to re-\npeated or prolonged GnRH treatment [11, 14, 15]. In the\npresent study, patients treated with GnRH analogs had\nmore abdominal pain than the other patients. This could\nhave several explanations; one is that patients with more\nsevere symptoms received GnRH treatment, but another\nplausible explanation is that the symptoms are due to side\neffects evoked by GnRH on the ENS. Further research on\nGnRH and its role in gastrointestinal function and dysfunc-\ntion is needed.\nThe current study has several limitations. Approxi-\nmately half of the patients identified as fulfilling inclusion\ncriteria declined to participate. One can hypothesize that\nthe patients who agreed to participate experienced more\nsevere symptoms, making them more prone to report\nthem. Another weakness is that the patient group and\ncontrol group were not matched regarding age. To\nminimize the impact of this factor, z-scores were calcu-\nlated and analyzed. However, age only correlated with\nsymptoms in patients and not in controls, and the patients\nwho declined to participate were slightly younger, which\ncorrelated with more severe symptoms. The patients re-\nported pharmacological treat ment of endometriosis and\ngastrointestinal symptoms, and as the majority of patients\nmost likely had no medical education, one can assume that\nthey less accurately reported their symptoms. To minimize\ninaccuracy, the pharmacological treatment was discussed at\nthe interview. When asked to measure abdominal pain on\nthe VAS-IBS, it is not clear whether patients responded\nregarding pain related to endometriosis or gastrointestinal\nsymptoms. However, almost half of the patients reported\nhaving the ability to differentiate pain from endometriosis\nfrom that of the gastrointesti nal tract. The patients appear\nto be a selected group with high education and a healthy\nlifestyle; they reported low consumption of alcohol and\ntobacco. However, these habits could also be due to the fact\nthat alcohol and smoking exacerbate their symptoms.\nThis study has several clinical impacts. Endometri-\nosis is a disease with a considerable diagnostic delay\n[28], and an extended knowledge of symptoms associ-\nated with the disease could p otentially contribute to\nreducing this delay. It is also of importance to acknow-\nledge these symptoms in or der to provide adequate\npharmacological treatment, especially since hormonal\ntreatment of endometriosis appeared to have no effect\non gastrointestinal symptoms. The findings also indicate\nthe importance of conservativeness in prescribing opioids\nto this group of patients, especially if the indication is\ngastrointestinal symptoms, because opioids appear to have\nno or weak effect on these symptoms. Physicians must\ncarefully evaluate the effect of opioids and should withdraw\nopioid treatment when the effect is uncertain.\nConclusions\nA large proportion of patients with endometriosis suffer\nfrom gastrointestinal symptoms. Compared to controls,\npatients with endometriosis suffer from more severe ab-\ndominal pain, constipation, bloating and flatulence, im-\npaired psychological well-being, influence of symptoms\non daily life, defecation urgency, and sensation of incom-\nplete evacuation. The location of the endometriosis lesions\nis not associated with symptoms, except increased nausea\nand vomiting among patients with lesions within or close\nto the bowel. Patients treated with opioids have more\nsevere symptoms than patients not treated with opioids,\nand current or previous treatment with GnRH analogs is\nassociated with increased abdominal pain. Initiating a new\nhormonal treatment of endometriosis has no impact on\ngastrointestinal symptoms over time.\nAbbreviations\nCIPO: Chronic intestinal pseudo-obstruction; GnRH: Gonadotropin-releasing\nhormone; IVF: In vitro fertilization; LH: Luteinizing hormone; VAS-IBS: Visual\nanalogue scale for irritable bowel syndrome.\nCompeting interests\nThe authors declare that they have no competing interests.\nAuthors’ contributions\nBO, MB, PE, and LV designed and planned the study. BR enrolled the\npatients and collected blood samples. ME performed the statistical\ncalculations and wrote the manuscript. BO financed the study. All authors\ncontributed to revision and finalization of the manuscript as well as approval\nof the final version.\nEk et al. BMC Women's Health  (2015) 15:59 Page 9 of 10\n\nAcknowledgements\nThis study was sponsored by grants from King Gustaf V and Queen Victoria\nFree Maison ’s Foundation, the Bengt Ihre Foundation, the Development\nFoundation from Region Skåne, and the Foundation of Skåne University\nHospital.\nAuthor details\n1Department of Clinical Sciences, Division of Internal Medicine, Skåne\nUniversity Hospital, Lund University, 205 02, Malmö, Sweden. 2Department of\nClinical Sciences, Division of Gynecology, Skåne University Hospital, Lund\nUniversity, 205 02 Malmö, Sweden. 3Faculty of Health and Society, Institution\nof Care Science, Malmö University, Malmö, Sweden.\nReceived: 1 April 2015 Accepted: 22 July 2015\nReferences\n1. Giudice LC, Kao LC. Endometriosis Lancet. 2004;364(9447):1789 –99.\n2. Ozkan S, Arici A. Advances in treatment options of endometriosis. Gynecol\nObstet Invest. 2009;67(2):81 –91.\n3. Maroun P, Cooper MJ, Reid GD, Keirse MJ. Relevance of gastrointestinal\nsymptoms in endometriosis. Aust N Z J Obstet Gynaecol. 2009;49(4):411 –4.\n4. Luscombe GM, Markham R, Judio M, Grigoriu A, Fraser IS. Abdominal\nbloating: an under-recognized endometriosis symptom. J Obstet Gynaecol\nCan. 2009;31(12):1159 –71.\n5. Ballweg ML. Impact of endometriosis on women ’s health: comparative\nhistorical data show that the earlier the onset, the more severe the disease.\nBest Pract Res Clin Obstet Gynaecol. 2004;18(2):201 –18.\n6. Meurs-Szojda MM, Mijatovic V, Felt-Bersma RJ, Hompes PG. Irritable bowel\nsyndrome and chronic constipation in patients with endometriosis.\nColorectal Dis. 2011;13(1):67 –71.\n7. Fauconnier A, Chapron C, Dubuisson JB, Vieira M, Dousset B, Breart G.\nRelation between pain symptoms and the anatomic location of deep\ninfiltrating endometriosis. Fertil Steril. 2002;78(4):719 –26.\n8. Seaman HE, Ballard KD, Wright JT, de Vries CS. Endometriosis and its\ncoexistence with irritable bowel syndrome and pelvic inflammatory disease:\nfindings from a national case-control study-Part 2. BJOG. 2008;115(11):1392–6.\n9. Roman H, Ness J, Suciu N, Bridoux V, Gourcerol G, Leroi AM, et al. Are\ndigestive symptoms in women presenting with pelvic endometriosis\nspecific to lesion localizations? A preliminary prospective study. Hum\nReprod. 2012;27(12):3440–9.\n10. Conn PM, Crowley Jr WF. Gonadotropin-releasing hormone and its analogs.\nAnnu Rev Med. 1994;45:391 –405.\n11. Hammar O, Ohlsson B, Veress B, Alm R, Fredrikson GN, Montgomery A.\nDepletion of enteric gonadotropin-releasing hormone is found in a few\npatients suffering from severe gastrointestinal dysmotility. Scand J\nGastroenterol. 2012;47(10):1165 –73.\n12. Mathias JR, Clench MH, Abell TL, Koch KL, Lehman G, Robinson M, et al.\nEffect of leuprolide acetate in treatment of abdominal pain and nausea in\npremenopausal women with functional bowel disease: a double-blind,\nplacebo-controlled, randomized study. Dig Dis Sci. 1998;43(6):1347 –55.\n13. Hammar O, Roth B, Bengtsson M, Mandl T, Ohlsson B. Autoantibodies and\ngastrointestinal symptoms in infertile women in relation to in vitro\nfertilization. BMC Pregnancy Childbirth. 2013;13:201.\n14. Cordeddu L, Bergvall M, Sand E, Roth B, Papadaki E, Li L, et al. Severe\ngastrointestinal dysmotility developed after treatment with\ngonadotropin-releasing hormone analogs. Scand J Gastroenterol.\n2015;50(3):291 –9.\n15. Ohlsson B, Veress B, Janciauskiene S, Montgomery A, Haglund M, Wallmark\nA. Chronic intestinal pseudo-obstruction due to buserelin-induced\nformation of anti-GnRH antibodies. Gastroenterology. 2007;132(1):45 –51.\n16. Bengtsson M, Ohlsson B, Ulander K. Development and psychometric testing\nof the Visual Analogue Scale for Irritable Bowel Syndrome (VAS-IBS). BMC\nGastroenterol. 2007;7:16.\n17. Bengtsson M, Persson J, Sjolund K, Ohlsson B. Further validation of the visual\nanalogue scale for irritable bowel syndrome after use in clinical practice.\nGastroenterol Nurs. 2013;36(3):188–98.\n18. Ferrero S, Camerini G, Ragni N, Venturini PL, Biscaldi E, Remorgida V.\nNorethisterone acetate in the treatment of colorectal endometriosis: a pilot\nstudy. Hum Reprod. 2010;25(1):94 –100.\n19. Moore J, Barlow D, Jewell D, Kennedy S. Do gastrointestinal symptoms vary\nwith the menstrual cycle? Br J Obstet Gynaecol. 1998;105(12):1322 –5.\n20. Barbara G, Stanghellini V, De Giorgio R, Cremon C, Cottrell GS, Santini D,\net al. Activated mast cells in proximity to colonic nerves correlate with\nabdominal pain in irritable bowel syndrome. Gastroenterology.\n2004;126(3):693–702.\n21. Anaf V, Chapron C, El Nakadi I, De Moor V, Simonart T, Noel JC. Pain, mast\ncells, and nerves in peritoneal, ovarian, and deep infiltrating endometriosis.\nFertil Steril. 2006;86(5):1336 –43.\n22. Issa B, Onon TS, Agrawal A, Shekhar C, Morris J, Hamdy S, et al. Visceral\nhypersensitivity in endometriosis: a new target for treatment? Gut.\n2012;61(3):367–72.\n23. Sand E, Bergvall M, Ekblad E, D ’Amato M, Ohlsson B. Expression and\ndistribution of GnRH, LH, and FSH and their receptors in gastrointestinal\ntract of man and rat. Regul Pept. 2013;187:24 –8.\n24. Rispoli LA, Nett TM. Pituitary gonadotropin-releasing hormone (GnRH)\nreceptor: structure, distribution and regulation of expression. Anim Reprod\nSci. 2005;88(1–2):57–74.\n25. Yung Y, Aviel-Ronen S, Maman E, Rubinstein N, Avivi C, Orvieto R, et al.\nLocalization of luteinizing hormone receptor protein in the human ovary.\nMol Hum Reprod. 2014;20(9):844 –9.\n26. Meleine M, Matricon J. Gender-related differences in irritable bowel\nsyndrome: potential mechanisms of sex hormones. World J Gastroenterol.\n2014;20(22):6725–43.\n27. Azizi Z, Javid Anbardan S, Ebrahimi DN. A review of the clinical manifestations,\npathophysiology and management of opioid bowel dysfunction and narcotic\nbowel syndrome. Middle East J Dig Dis. 2014;6(1):5–12.\n28. Husby GK, Haugen RS, Moen MH. Diagnostic delay in women with pain and\nendometriosis. Acta Obstet Gynecol Scand. 2003;82(7):649 –53.\nSubmit your next manuscript to BioMed Central\nand take full advantage of: \n• Convenient online submission\n• Thorough peer review\n• No space constraints or color figure charges\n• Immediate publication on acceptance\n• Inclusion in PubMed, CAS, Scopus and Google Scholar\n• Research which is freely available for redistribution\nSubmit your manuscript at \nwww.biomedcentral.com/submit\nEk et al. BMC Women's Health  (2015) 15:59 Page 10 of 10","source_license":"CC0","license_restricted":false}