{"paper_id":"87131f61-6958-4bbf-b35f-a43f772d03aa","body_text":"Central European Journal of Immunology 2013; 38(4)\n500\nClinical immunology\nSerum cytokine and growth factor levels  \nin patients with endometriosis\nAyCuruk kubAtovA 1, Ahmet erdem 1, mehmet erdem 1, mehmet FirAtmutlu 2,  \numit koruCuoglu 3 \n1Department of Obstetrics & Gynaecology, Gazi University School of Medicine, Ankara, Turkey \n2Department of Obstetrics & Gynaecology, HRS Women Hospital, Ankara, Turkey \n3Department of Obstetrics & Gynaecology, Koru Hospital, Ankara, Turkey\nAbstract\nthe aim of this prospective study is to define markers that can be used in the diagnosis and fol-\nlow-up of patients with endometriosis by determining serum CA 125, transforming growth factor β1 \n(tg F-β1), interleukin 6 (il -6), and il-12 levels. the study group consisted of 14 patients diagnosed \nto have mild to moderate peritoneal endometriosis, 47 patients with ovarian endometrioma during \nlaparoscopy and 12 healthy patients as a control group. Serum CA 125, tgF-β1, il-6, and il-12 levels \nwere evaluated using eliSA kits. Serum CA 125 levels were significantly higher in the endometrioma \ngroup when compared to the control group (p < 0.001) and mild to moderate peritoneal endometriosis \ngroup (p < 0.001). mean serum tgF-\nβ1 levels were significantly higher in the endometrioma group \nwhen compared to the mild to moderate peritoneal endometriosis group (p < 0.001). in addition, serum \ntgF- β1 levels were found to be significantly different with regard to the stage of the endometriosis. \nSerum il-6 levels were significantly higher in the endometrioma group compared to the endometriosis \ngroup (p = 0.024) and control group (p = 0.05). mean serum il-12 levels were lower in patients with \nendometriosis. in conclusion; tgF- β1 and il-6 measurements might be a promising alternative in \nadjunct to CA 125 for the non-invasive diagnosis of endometrioma. however CA 125, tgF- β1, il-6 \nand il-12 seem not to have the diagnostic value in the diagnosis of early stage endometriosis. of all \nthe serum markers studied, only tgF- β1 was found to be correlated with the stage of endometriosis. \nKey words: endometriosis, serum marker, CA 125, tgF- β1, il-6, and il-12.\n \n(Centr eur J immunol 2013; 38 (4):  500-504)\nDOI: 10.5114/ceji.2013.39768\nCorrespondence: Mehmet Firat Mutlu, MD, Department of Obstetrics & Gynaecology, HRS Women Hospital, Güneş Sokak No. 14 \nKavaklldere Ankara, Turkey, tel. 532 325 77 90, e-mail: firatmutlu78@hotmail.com\nIntroduction \nEndometriosis is an oestrogen-dependent clinical entity \ncharacterized by the presence of endometrial glands and \nstroma outside the uterus. It is estimated to occur in 3-10% \nof reproductive age women [1-3]. Endometriosis interferes \nwith the quality of life by causing dysmenorrhoea, dyspa-\nreunia, chronic pelvic pain and infertility.\nUp to now, several mechanisms such as ectopic trans-\nplantation of endometrial tissue, coelomic metaplasia, in-\nduction theory and loss of apoptosis have been proposed \nto explain the etiopathogenesis of endometriosis. The im -\nmune system has been thought to play a  role in the com -\nplex events of endometriosis secondary to the fact that ret-\nrograde menstruation was a common event in women, but \nnot all women who had retrograde menstruation developed \nendometriosis. It has been hypothesized that the disease \nmight develop as a result of altered immunologic clear-\nance of viable endometrial cells [4]. Substantial evidence \nsuggested that endometriosis was associated with a state of \nsubclinical peritoneal inflammation, marked by increased \ninflammatory cytokines and growth factors. Cytokines and \ngrowth factors are soluble low-molecular-weight proteins \nwith autocrine and paracrine effects that play a role in mi-\ntosis, angiogenesis and chemotaxis [5, 6].\nThe gold standard in the diagnosis of endometriosis \nis the histological examination of the surgically removed \ntissue. A  serum marker that is equally valuable for the \ndiagnosis of endometriosis has long been researched to \nprevent any invasive procedure for the accurate diagnosis. \nThe most widely used serum marker for this purpose is CA \n125 [7, 8]. Several studies have been performed to detect \n\nCentral European Journal of Immunology 2013; 38(4)\n501\nSerum cytokine and growth factor levels in patients with endometriosis\na serum marker equally or more valuable in the diagnosis \nof endometriosis, but the results were conflicting [9-14]. \nMost of these data pointed to a possible use of interleukin 6 \n(IL-6) and transforming growth factor (TGF-β1) as serum \nmarkers of endometriosis. Our prospective study aimed to \ndetermine serum CA 125, TGF-β1, IL-6, and IL-12 levels \nin patients with endometriosis. The study was expected to \ndefine new serum markers that can be used in the diagnosis \nand follow-up of patients with endometriosis.\nMaterial and methods\nStudy design\nThis was a prospective study of 73 women referred to \nthe gynaecology clinic of the Gazi University School of \nMedicine with a clinical or sonographic suspicion of endo-\nmetriosis and endometrioma, who underwent laparoscopy. \nThe control group consisted of 12 healthy patients who un-\nderwent laparoscopy for suspected endometriosis or tubal \nligation and were confirmed to be free of endometriosis \nand any ovarian cysts during surgery. The patients were at \nthe age of 18-40 years. The approval of the ethics commit-\ntee of the Gazi University and informed consent from all \nparticipants were obtained prior to the study.\nThe clinical suspicion of endometriosis was made ac-\ncording to the symptoms of dysmenorrhoea, chronic pelvic \npain, dyspareunia, infertility, urinary and rectal symptoms. \nThe sonographic suspicion was made by the typical ultra-\nsound characteristics, the presence of cystic ovarian mass-\nes with homogeneous low-level internal echoes, punctuate \nperipheral echogenic foci and thick cystic walls. Ovarian \nendometriomas were detected before surgery by transvag-\ninal ultrasound and then confirmed by histological ex-\namination. All ovarian endometriomas were ≥  3 cm. No \nperitoneal implants or deeply infiltrating endometriosis \nwere observed during laparoscopy in the ovarian endome-\ntrioma group. In all patients with stage 1-2 endometriosis, \nmultiple biopsies were taken from suspected appearance \nof endometriosis and were confirmed by histological ex-\namination. \nPatients were divided into three groups: women with \nstage 1-2 endometriosis confirmed by laparoscopy ac-\ncording to the revised “American Fertility Society Clas-\nsification” American Society for Reproductive Medicine \n(1997) (n  = 14); women with unilateral or bilateral en-\ndometrioma and no history of previous ovarian surgery  \n(n = 47) and the control group (n = 12). Also to eliminate \nthe possible effect of basal sonographic cystic structures \non the results, five patients from the control group and one \npatient from the endometriosis group were excluded from \nthe study due to the presence of ovarian cysts.\nPatients with myoma uteri, dermoid cysts, ovarian \ncystic structures > 3 cm other than endometrioma, pelvic \ninflammatory disease, any malignancy and current users of \noral contraceptives, GnRH analogues, progestin, danazol \nor patients using any hormonal therapy, were excluded \nfrom the study. None of the patients had taken anti-inflam-\nmatory medications or had been diagnosed with an inflam-\nmation or infection in previous 6 months before the study.\nLaparoscopic surgery was performed in the prolif-\nerative phase of the menstrual cycle by insertion of a 10 \nmm umbilical trocar and two ancillary trocars in the lower \nabdomen. All interventions were carried out by two ex-\nperienced laparoscopists (AE, ME). The same protocol \nwas used during the diagnostic phase of laparoscopy. \nThis included an inspection of pelvic and peritoneal or-\ngans, peritoneal washings, staging of endometriosis. All \npatients with suspected endometriosis and endometrioma \nhad a routine histological confirmation of the condition. \nPrior to anaesthesia induction, 10 ml of venous blood \nwere drawn from each patient. Blood samples were cen -\ntrifuged at 3000 rpm for 15 minutes and stored at –50\noC \nuntil assay. Serum CA 125 levels were measured by che-\nmiluminescence using IMMULITE 2000 hormone an-\nalyzer (Diagnostic Products Corporation, Los Angeles, \nCA, USA). Serum TGF- β1, IL-6, and IL-12 levels were \nmeasured by using solid phase sandwich enzyme linked \nimmunosorbent assay (ELISA) kits (Biosource Interna-\ntional, California, USA). Intra- and inter-assay coeffi-\ncients of variation were < 10% for all assays.\nStatistical analysis\nResults were expressed as mean + standard deviation. \nKruskal Wallis univariate analysis was used to compare \nthe means of serum CA 125, TGF-β1, IL-6 and IL-12 lev-\nels between groups. Nonparametric Mann-Whitney u test \nwas used to compare the means of serum CA 125, TGF-β1, \nIL-6, and IL-12 levels between two independent groups. \nSPSS for Windows version 10.0 was used for all statistical \nanalysis; p < 0.05 was set as the significance criterion. \nResults\nBaseline characteristics of patients are summarized \nin Table 1. One way Anova test revealed no significant \ndifference between groups with respect to mean ages  \n(p = 0.29). The preoperative ultrasonographic examination \nof patients revealed ovarian cysts in one patient of the peri-\ntoneal endometriosis group (6.7%), in all of the patients of \nthe endometrioma group and in 5 patients of the control \ngroup (29.4%). These were excluded from the study.\nPreoperative serum CA 125, TGF-β 1, IL-6 and IL-12 \nlevels in all three groups are shown in Table 2. Kruskal \nWallis univariate analysis demonstrated that three groups \nwere significantly different from each other with respect \nto serum CA 125 levels (p = 0.001). The Post Hoc Schef-\nfe test failed to show any difference between pelvic en-\ndometriosis and control groups, but CA 125 levels were \n\nCentral European Journal of Immunology 2013; 38(4)\n502\nAycuruk Kubatova et al.\nsignificantly higher in the endometrioma group than \nboth the pelvic endometriosis group (p < 0.001) and con-\ntrol group (p  < 0.001). Three groups were also different \nfrom each other with respect to serum TGF-β 1 levels  \n(p = 0.004). The Post Hoc Scheffe test showed no differ-\nence between control and pelvic endometriosis groups  \n(p = 0.17) and between control and endometrioma groups \n(p = 0.15). But the endometrioma group was signifi-\ncantly different from the pelvic endometriosis group \nwith respect to serum TGF-β 1 levels (p  = 0.001). Kru-\nskal Wallis univariate analysis demonstrated that three \ngroups were significantly different from each other with \nrespect to serum IL-6 levels ( p = 0.02). Serum IL-6 lev -\nels in the endometrioma group were significantly different \nfrom both the control group (p  = 0.05) and endometriosis  \n(p = 0.024). The three groups were similar to each other \nwith respect to serum IL-12 levels. \nNone of the serum markers studied (TGF-β1, IL-6, and \nIL-12) had a positive correlation with serum levels of CA \n125 with respect to the diagnosis and follow-up of endome-\ntriosis. As a correlation of serum markers among themselves \nwas studied, only a weak correlation between IL-6 and IL-\n12 was detected (r = –0.256; p = 0.006). Of all the serum \nmarkers studied, only TGF-β1 was found to be correlated \nwith the stage of endometriosis (r = –0.424; p = 0.001).\nThe association of serum CA 125 levels with clinical \nfindings is summarized in Table 3. In both pelvic endome-\ntriosis and endometrioma groups, the serum CA 125 level \nwas not found to be associated with any clinical finding. \nSimilarly, in both pelvic endometriosis and endometrio-\nma groups, serum TGF-β1 and IL-6 levels were not found \nto be associated with any clinical finding. In the pelvic \nendometriosis group, the IL-12 level failed to show any \ncorrelation with any clinical finding. \nDiscussion\nThe immune system has been thought to play a role \nin the complex events of endometriosis. Most studies \nperformed up to now focused on detection of cytokines \nand growth factors in the peritoneal fluid and endome-\ntriotic tissues. The only serum marker that proved to \nbe of potential use for the diagnosis and follow-up of \nendometriosis has been CA 125. CA 125 has been cor-\nrelated with the stage of endometriosis and its level has \nbeen found to increase with higher stages [7]. Our study \nsimilarly detected that CA 125 levels were significant-\nly higher in the endometrioma group than both the pel-\nvic endometriosis group (p  < 0.001) and control group  \n(p = 0.001). However, a single serum CA 125 level has \nbeen reported to have a quite low sensitivity for the di-\nagnosis of endometriosis [8]. Our study supported this \nfinding as all of the patients with mild disease and 32% of \npatients with advanced disease had serum CA 125 levels \nTable 1. Baseline characteristics of patients\nPelvic endometriosis \n(n = 14)\nEndometrioma \n(n = 47)\nControl group \n(n = 12)\nMean age + SD 30.6 ±5.3 32.6 ±7.4 31.6 ±7.2\nMarital status \nsinglemarried/divorced\n1 (7.1%) \n13 (92.9%)\n8 (17%) \n39 (83%)\n4 (33.3%) \n8 (66.7%)\nChronic pelvic pain 5 (35.7%) 24 (51.1%) 2 (16.7%)\nDysmenorrhoea 8 (57.1%) 34 (72.3%) 2 (16.7%)\nDyspareunia 6 (42.9%) 16 (34%) –\nInfertility 12 (85.7%) 8 (20.5%) 4 (33.3%)\nTable 2.  Preoperative serum CA 125, TGF-β1, IL-6, and IL-12 levels in all three groups\nPelvic endometriosis \n(n = 14)\nEndometrioma \n(n = 47)\nControl group \n(n = 12)\np\nCA 125 \n(U/ml) 14.9 ±7.9 75.3 ±110.8 14.7 ±8.7 < 0.001\nTGF-β1 \n(pg/ml) 11 089 ±4981 17 849 ±6856 14 934 ±6734 0.004\nIL-6 \n(pg/ml) 57.1 ±123.5 98.1 ±152.7 44.6 ±72.8 0.02\nIL-12 \n(pg/ml) 80.1 ±49.6 76.5 ±32.1 69.6 ±35.4 0.74\n\nCentral European Journal of Immunology 2013; 38(4)\n503\nSerum cytokine and growth factor levels in patients with endometriosis\nbelow 35 U/ml, which is considered as a normal upper lim-\nit. As a consequence, the serum CA 125 level is inadequate \nfor the diagnosis and follow-up of endometriosis patients \nand new serum markers with a high sensitivity and speci-\nficity would prove useful. \nCytokines and growth factors secreted from peritoneal \nsurface cells or from endometriotic tissues are known to be \ncrucial for the development of ectopic endometrial cells. \nThis is the reason why this study included serum TGF-β1, \nIL-6, and IL-12 levels. The TGF-β1 secreted from ectopic \nendometrial cells is thought to play the major role in the \npathogenesis of endometriosis by increasing adhesion of \nendometrial stromal cells to peritoneum, suppressing local \nimmune response and protecting ectopic endometrial cells \nfrom clearing. \nTransforming growth factor β1 expressions by eutopic \nand ectopic endometrial tissues have been studied immu-\nnohistochemically [15, 16]. Up to date there have been two \npublished studies in the literature regarding the association \nof TGF-β1 with endometriosis [16, 17]. Pizzo et al. found \na significantly higher level of TGF-β 1 in both peritoneal \nfluid and serum of patients with endometriosis as com-\npared to the control group in the study investigating the \nserum TGF-β1 levels in patients with endometriosis [16]. \nThe same study showed that peritoneal fluid and serum \nTGF-β1 levels increased with the increased stage of en-\ndometriosis. In contrast, D’Hooghe et al. did not observe \nany significant difference in peripheral blood cytokine lev-\nels between deep infiltrating endometriosis compared to \nsuperficial disease [17]. In our study, the endometrioma \ngroup was significantly different from the pelvic endo-\nmetriosis group with respect to the serum TGF-β 1 levels  \n(p < 0.001). In addition, the serum TGF-β1 level was found \nto be the only serum marker correlated positively with the \nstage of endometriosis (r  = –0.424; p = 0.001).Thus, the \nserum TGF-β 1 level might be considered as a valuable \nmarker of endometriosis. But because of different assays, \na lack of standard measurement is a significant obstacle. \nTwo proinflammatory cytokines thought to play a role \nin the etiopathogenesis of endometriosis are IL-6 and IL-12. \nResults of studies investigating peritoneal fluid and serum \nIL-6 levels in patients with endometriosis are conflicting \n[5, 11, 18-20]. Some studies report an increase in perito-\nneal fluid and serum IL-6 levels in patients with endome-\ntriosis and even a positive correlation with an increased \nstage of disease, however, some fail to prove this asso-\nciation. In some studies investigating serum IL-6 levels \nin patients with endometriosis, it has been reported that \nserum IL-6 levels were higher in patients with endometri-\nosis as compared to the control group [9, 11, 12, 21, 22]. \nHowever, this correlation was not verified by other authors \n[14, 17, 23-25]. In a relatively recent study, serum IL-6 \nlevels were increased in all women with endometriosis and \nin those with minimal-mild endometriosis, compared with \ncontrols [26]. The authors argued that the use of IL-6 as \na serum marker for the diagnosis of endometriosis is use-\nful. Our study demonstrated that serum IL-6 levels in the \nendometrioma group was significantly different from the \ncontrol group (p = 0.05). Although the highest serum IL-6 \nlevel was detected in patients with stage 4 disease (156.3 \n±263.3), three groups were not significantly different from \neach other. On the other hand, Martinez et al. revealed that \nIL-6 levels appear higher in patients with mild to moder-\nate endometriosis, which is in contrast with our findings \n[21]. It is a well-established fact that the ovary is the major \nsource of the peritoneal fluid. Therefore, patients with en-\ndometrioma may have higher concentrations of cytokines \nin the peritoneal fluid [27]. Consequently, this may explain \nwhy IL-6 levels were higher in the endometrioma group in \ncomparison to patients with an early stage endometriosis \nin our study.\nDifferences that are mentioned in the studies may be \na result of several factors, such as differences in control \ngroups, differences in compared pathologies (i.e. ovarian \ncysts), differences in assay sensitivities and differences in \nthe inclusion criteria of studies. Our study demonstrated \nthat serum IL-6 levels increased especially in advanced \nstages of endometriosis. Previous findings showing that se-\nrum IL-6 levels decrease after use of GnRH analogues and \nafter laparoscopic treatment render IL-6 a valuable marker \nfor follow-up of patients with endometriosis [12].\nInterleukin 12 is required for clearance of ectopic en-\ndometrial cells. Its decreased levels may be associated with \ndefective clearing of ectopic endometrial cells shed to the \nabdominal cavity via retrograde menstruation or formed \nvia metaplasia. In their endometriosis model, Somigliana  \nTable 3. Association of serum CA 125 levels with clinical findings\nPelvic endometrios Endometrioma\n(–) \n(n = 9)\n(+) \n(n = 5) p (–) \n(n = 23)\n(+) \n(n = 24) p\nChronic pelvic pain 12.7 ±5.3 17.2 ±11.6 0.46 55.4 ±35.1 94.5 ±150.4 0.68\nDysmenorrhoea 12.4 ±6.3 15.5 ±8.8 0.48 59.3 ±40.9 81.5 ±128.0 0.53\nDyspareunia 13.1 ±5.8 16.3 ±9.8 0.56 49.7 ±37.2 82.2 ±106.0 0.41\nInfertility 16.4 ±4.0 14.5 ±8.4 0.58 64.3 ±81.3 55.8 ±48.4 0.87\n\nCentral European Journal of Immunology 2013; 38(4)\n504\nAycuruk Kubatova et al.\net al. demonstrated that IL-12 application prevented implan-\ntation of ectopic endometrial cells [14]. Also, p40 subunit of \nIL-12 has been previously claimed to inhibit natural killer \ncell activity and predispose to the development of endome-\ntriosis. Similar to IL-6, studies researching IL-12 levels in \nendometriosis have conflicting results [11, 28, 29]. In the \nonly study investigating serum IL-12 level in patients with \nendometriosis reported in the literature, authors claimed that \nserum IL-12 levels were not different in patients with endo-\nmetriosis as compared to the control group [11]. Similarly, \nour study found that three groups were similar to each other \nwith respect to serum IL-12 levels.\nAs a result, in comparison to CA125, serum TGF-β 1, \nIL-6 and IL12 levels may be not sufficient enough to de-\ntect early stages of endometriosis. However, TGF-β 1 and \nIL-6 measurements might be a promising alternative in \nadjunct to CA 125 for the non-invasive diagnosis of endo-\nmetrioma. Of all the serum markers studied, only TGF-β1 \nwas found to be correlated with the stage of endometriosis. \nHowever, more prospective studies are required to further \nclarify these findings. \n \nthe authors declare no conflict of interest. \nReferences\n 1.  Missmer SA, Cramer DW (2003): The epidemiology of en-\ndometriosis. Obstet Gynecol Clin North Am 30: 1-19, vii.\n 2.  D’Hooghe TM, Debrock S, Hill JA, Meuleman C (2003): \nEndometriosis and subfertility: is the relationship resolved? \nSemin Reprod Med 21: 243-254.\n 3.  Giudice LC, Kao LC (2004): Endometriosis. Lancet 364: \n1789-1799.\n 4.  Dmowski WP, Steele RW, Baker GF (1981): Deficient cel-\nlular immunity in endometriosis. Am J Obstet Gynecol 141: \n377-383.\n 5.  Harada T, Yoshioka H, Yoshida S, et al. (1997): Increased \ninterleukin-6 levels in peritoneal fluid of infertile patients \nwith active endometriosis. Am J Obstet Gynecol 176:  \n593-597.\n 6.  Seli E, Arici A (2003): Endometriosis: interaction of immune \nand endocrine systems. Semin Reprod Med 21: 135-144.\n 7.  Barbieri RL, Niloff JM, Bast RC Jr, et al. (1986): Elevated \nserum concentrations of CA-125 in patients with advanced \nendometriosis. Fertil Steril 45: 630-634.\n 8.  Mol BW, Bayram N, Lijmer JG, et al. (1998): The perfor-\nmance of CA-125 measurement in the detection of endome -\ntriosis: a meta-analysis. Fertil Steril 70: 1101-1108.\n 9.  Pellicer A, Albert C, Mercader A, et al. (1998): The follicular \nand endocrine environment in women with endometriosis: lo-\ncal and systemic cytokine production. Fertil Steril 70: 425-431.\n10.  Fasciani A, D’Ambrogio G, Bocci G, et al. (2001): Vascular \nendothelial growth factor and interleukin-8 in ovarian cystic \npathology. Fertil Steril 75: 1218-1221.\n11.  Bedaiwy MA, Falcone T, Sharma RK, et al. (2002): Predic -\ntion of endometriosis with serum and peritoneal fluid mark-\ners: a prospective controlled trial. Hum Reprod 17: 426-431.\n12.  Cheong YC, Shelton JB, Laird SM, et al. (2002): IL-1, IL-6 \nand TNF-alpha concentrations in the peritoneal fluid of wom-\nen with pelvic adhesions. Hum Reprod 17: 69-75.\n13.  Gagné D, Pagé M, Robitaille G, et al. (2003): Levels of vas-\ncular endothelial growth factor (VEGF) in serum of patients \nwith endometriosis. Hum Reprod 18: 1674-1680.\n14.  Somigliana E, Vigano P, Tirelli AS, et al. (2004): Use of the \nconcomitant serum dosage of CA 125, CA 19-9 and interleu-\nkin-6 to detect the presence of endometriosis. Results from \na series of reproductive age women undergoing laparoscopic \nsurgery for benign gynaecological conditions. Hum Reprod \n19: 1871-1876.\n15.  Oosterlynck DJ , Cornillie FJ, Waer M, et al. (1991): Wom -\nen with endometriosis show a defect in natural killer activity \nresulting in a decreased cytotoxicity to autologous endome-\ntrium. Fertil Steril 56: 45-51.\n16.  Pizzo A, Salmeri FM, Ardita FV, et al. (2002): Behaviour of \ncytokine levels in serum and peritoneal fluid of women with \nendometriosis. Gynecol Obstet Invest 54: 82-87.\n17.  D’Hooghe TM, Xiao L, Hill JA (2001): Cytokine profiles in \nautologous peritoneal fluid and peripheral blood of women \nwith deep and superficial endometriosis. Arch Gynecol Obstet \n265: 40-44.\n18.  Koyama N, Matsuura K, Okamura H (1993): Cytokines in the \nperitoneal fluid of patients with endometriosis. Int J Gynaecol \nObstet 43: 45-50.\n19.  Keenan JA, Chen TT, Chadwell NL, et al. (1994): Interfer-\non-gamma (IFN-gamma) and interleukin-6 (IL-6) in perito-\nneal fluid and macrophage-conditioned media of women with \nendometriosis. Am J Reprod Immunol 32: 180-183.\n20.  Taylor RN, Ryan IP, Moore ES, et al. (1997): Angiogenesis \nand macrophage activation in endometriosis. Ann N Y Acad \nSci 828: 194-207.\n21.  Martínez S, Garrido N, Coperias JL, et al. (2007): Serum in-\nterleukin-6 levels are elevated in women with minimal-mild \nendometriosis. Hum Reprod 22: 836-842.\n22.  Othman Eel-D, Hornung D, Salem HT, et al. (2008): Serum \ncytokines as biomarkers for nonsurgical prediction of endo-\nmetriosis. Eur J Obstet Gynecol Reprod Biol 137: 240-246.\n23.  Kalu E, Sumar N, Giannopoulos T, et al. (2007): Cytokine \nprofiles in serum and peritoneal fluid from infertile women \nwith and without endometriosis. J Obstet Gynaecol Res  33: \n490-495.\n24.  Jee BC, Suh CS, Kim SH, Moon SY (2008): Serum soluble \nCD163 and interleukin-6 levels in women with ovarian endo-\nmetriomas. Gynecol Obstet Invest 66: 47-52.\n25.  Seeber B, Sammel MD, Fan X, et al. (2008): Panel of markers \ncan accurately predict endometriosis in a subset of patients. \nFertil Steril 89: 1073-1081.\n26.  Mihalyi A, Gevaert O, Kyama CM, et al. 2010. Non-invasive \ndiagnosis of endometriosis based on a combined analysis of \nsix plasma biomarkers. Hum Reprod 25: 654-664.\n27.  Carmona F, Chapron C, Martínez-Zamora MÁ, et al. (2012): \nOvarian endometrioma but not deep infiltrating endometriosis \nis associated with increased serum levels of interleukin-8 and \ninterleukin-6. J Reprod Immunol 95: 80-86.\n28.  Gazvani R, Bates M, Vince G, et al. (2001): Concentration of \ninterleukin-12 in the peritoneal fluid is not influenced by the \npresence of endometriosis, its stage or the phase of the men-\nstrual cycle. Acta Obstet Gynecol Scand 80: 175-178.\n29.  Zeyneloglu HB, Senturk LM, Seli E, et al. (1998): The peri-\ntoneal fluid levels of interleukin-12 in women with endome-\ntriosis. Am J Reprod Immunol 39: 152-156.","source_license":"CC0","license_restricted":false}