{"paper_id":"d1892793-f7ab-4fb0-ae89-d847db67318c","body_text":"Cochrane\nLibrary\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of\nendometriosis (Review)\n/uni00A0\n/uni00A0Nisenblat V, Prentice L, Bossuyt PMM, Farquhar C, Hull ML, Johnson N /uni00A0\n/uni00A0 Nisenblat/uni00A0V, Prentice/uni00A0L, Bossuyt/uni00A0PMM, Farquhar/uni00A0C, Hull/uni00A0ML, Johnson/uni00A0N. \nCombination of the non-invasive tests for the diagnosis of endometriosis. \nCochrane Database of Systematic Reviews 2016, Issue 7. Art. No.: CD012281. \nDOI: 10.1002/14651858.CD012281.\n/uni00A0\n/uni00A0 www.cochranelibrary.com /uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)/uni00A0\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nT A B L E /uni00A0 O F /uni00A0 C O N T E N T S\nABSTRACT..................................................................................................................................................................................................... 1\nPLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2\nSUMMARY OF FINDINGS .............................................................................................................................................................................. 4\nBACKGROUND.............................................................................................................................................................................................. 12\nFigure 1.................................................................................................................................................................................................. 15\nOBJECTIVES.................................................................................................................................................................................................. 16\nMETHODS..................................................................................................................................................................................................... 16\nRESULTS........................................................................................................................................................................................................ 21\nFigure 2.................................................................................................................................................................................................. 22\nFigure 3.................................................................................................................................................................................................. 24\nFigure 4.................................................................................................................................................................................................. 25\nFigure 5.................................................................................................................................................................................................. 26\nFigure 6.................................................................................................................................................................................................. 27\nFigure 7.................................................................................................................................................................................................. 29\nFigure 8.................................................................................................................................................................................................. 30\nFigure 9.................................................................................................................................................................................................. 31\nFigure 10................................................................................................................................................................................................ 32\nFigure 11................................................................................................................................................................................................ 33\nFigure 12................................................................................................................................................................................................ 34\nDISCUSSION.................................................................................................................................................................................................. 35\nAUTHORS' CONCLUSIONS........................................................................................................................................................................... 37\nACKNOWLEDGEMENTS................................................................................................................................................................................ 39\nREFERENCES................................................................................................................................................................................................ 40\nCHARACTERISTICS OF STUDIES.................................................................................................................................................................. 45\nDATA.............................................................................................................................................................................................................. 75\nTest 1. IL-6 (>15.4 pg/ml) [serum] + PGP 9.5 [endometrium]............................................................................................................. 76\nTest 2. CA-125 [serum] (>35 U/ml) + P450 aromatase [endometrium].............................................................................................. 77\nTest 3. VDBP-Cr [urine] x CA-125 [serum] (>2755)............................................................................................................................... 77\nTest 4. NNE_Cr [urine] + CA-125 [serum] (>27.23).............................................................................................................................. 77\nTest 5. Hx (dysmenorrhoea, dyspareunia) + PV examination + TVUS (fixed ovary).......................................................................... 77\nTest 6. Hx (length of menses) + CA-125 [serum] (>35 U/ml) + leukocytes [endometrium]............................................................... 77\nTest 7. Hx (parity, past IUD, past endometriosis, alcohol intake, dyspareunia) + CA-125 [serum]................................................... 77\nTest 8. PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for DIE, endometrioma or severe adhesions........ 78\nTest 9. PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for DIE, endometrioma or severe adhesions...... 78\nTest 10. PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for DIE................................................................. 78\nTest 11. PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for DIE.............................................................. 78\nTest 12. PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for endometrioma.............................................. 78\nTest 13. PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for endometrioma........................................... 79\nTest 14. TVUS + CA-125 [serum] (≥25 U/ml) + CA-19.9 [serum] (≥12 U/ml)........................................................................................ 79\nTest 15. TVUS + (CA-125 [serum] (≥25 U/ml) OR CA-19.9 [serum] (≥12 U/ml))................................................................................... 79\nTest 16. TVUS + CA-19.9 [serum] (≥12 U/ml)....................................................................................................................................... 79\nTest 17. TVUS OR CA-19.9 [serum] (≥12 U/ml).................................................................................................................................... 79\nTest 18. TVUS + CA-125 [serum] (≥20 U/ml)........................................................................................................................................ 79\nTest 19. TVUS OR CA-125 [serum] (≥20 U/ml)..................................................................................................................................... 80\nTest 20. TVUS + CA-125 [serum] (≥25 U/ml)........................................................................................................................................ 80\nTest 21. TVUS OR CA-125 [serum] (≥25 U/ml)..................................................................................................................................... 80\nTest 22. TVUS + CA-125 [serum] (≥35 U/ml)........................................................................................................................................ 80\nTest 23. TVUS OR CA-125 [serum] (≥35 U/ml)..................................................................................................................................... 80\nTest 24. PV examination + TVUS for POD obliteration........................................................................................................................ 81\nTest 25. PV examination + TVUS for vaginal endometriosis............................................................................................................... 81\nTest 26. PV examination + TVUS for RVS endometriosis.................................................................................................................... 81\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\ni\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTest 27. PV examination + TVUS for rectal endometriosis................................................................................................................. 81\nADDITIONAL TABLES.................................................................................................................................................................................... 81\nAPPENDICES................................................................................................................................................................................................. 88\nCONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 115\nDECLARATIONS OF INTEREST..................................................................................................................................................................... 115\nSOURCES OF SUPPORT............................................................................................................................................................................... 115\nDIFFERENCES BETWEEN PROTOCOL AND REVIEW.................................................................................................................................... 115\nNOTES........................................................................................................................................................................................................... 116\nINDEX TERMS............................................................................................................................................................................................... 116\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\nii\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n[Diagnostic Test Accuracy Review]\nCombination of the non-invasive tests for the diagnosis of endometriosis\nVicki Nisenblat1, Lucy Prentice2, Patrick MM Bossuyt3, Cindy Farquhar4, M Louise Hull1, Neil Johnson1\n1Discipline of Obstetrics and Gynaecology, School of Medicine, Robinson Research Institute, The University of Adelaide, Adelaide,\nAustralia. 2Obstetrics and Gynaecology, Tauranga Hospital, Bay of Plenty DHB, Tauranga, New Zealand. 3Department of Clinical\nEpidemiology, Biostatistics and Bioinformatics, Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands.\n4Department of Obstetrics and Gynaecology, University of Auckland, Auckland, New Zealand\nContact: Vicki Nisenblat, Discipline of Obstetrics and Gynaecology, School of Medicine, Robinson Research Institute, The University of\nAdelaide, Level 6, Medical School North,, Frome Rd, Adelaide, SA, 5005, Australia. vnisenblat@gmail.com.\nEditorial group: Cochrane Gynaecology and Fertility Group.\nPublication status and date: New, published in Issue 7, 2016.\nCitation:  Nisenblat/uni00A0V, Prentice/uni00A0L, Bossuyt/uni00A0PMM, Farquhar/uni00A0C, Hull/uni00A0ML, Johnson/uni00A0N. Combination of the non-invasive tests for the diagnosis\nof endometriosis. Cochrane Database of Systematic Reviews 2016, Issue 7. Art. No.: CD012281. DOI: 10.1002/14651858.CD012281.\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\nA B S T R A C T\nBackground\nAbout 10% of women of reproductive age suffer from endometriosis, a costly chronic disease causing pelvic pain and subfertility.\nLaparoscopy is the gold standard diagnostic test for endometriosis, but is expensive and carries surgical risks. Currently, there are no\nnon-invasive tests available in clinical practice to accurately diagnose endometriosis. This review assessed the diagnostic accuracy of\ncombinations of different non-invasive testing modalities for endometriosis and provided a summary of all the reviews in the non-invasive\ntests for endometriosis series.\nObjectives\nTo estimate the diagnostic accuracy of any combination of non-invasive tests for the diagnosis of pelvic endometriosis (peritoneal and/or\novarian or deep infiltrating) compared to surgical diagnosis as a reference standard. The combined tests were evaluated as replacement\ntests for diagnostic surgery and triage tests to assist decision-making to undertake diagnostic surgery for endometriosis.\nSearch methods\nWe did not restrict the searches to particular study designs, language or publication dates. We searched CENTRAL to July 2015, MEDLINE\nand EMBASE to May 2015, as well as the following databases to April 2015: CINAHL, PsycINFO, Web of Science, LILACS, OAIster, TRIP,\nClinicalTrials.gov, DARE and PubMed.\nSelection criteria\nWe considered published, peer-reviewed, randomised controlled or cross-sectional studies of any size, including prospectively collected\nsamples from any population of women of reproductive age suspected of having one or more of the following target conditions: ovarian,\nperitoneal or deep infiltrating endometriosis (DIE). We included studies comparing the diagnostic test accuracy of a combination of several\ntesting modalities with the findings of surgical visualisation of endometriotic lesions.\nData collection and analysis\nThree review authors independently collected and performed a quality assessment of the data from each study by using the QUADAS-2\ntool. For each test, the data were classified as positive or negative for the surgical detection of endometriosis and sensitivity and specificity\nestimates were calculated. The bivariate model was planned to obtain pooled estimates of sensitivity and specificity whenever sufficient\ndata were available. The predetermined criteria for a clinically useful test to replace diagnostic surgery were a sensitivity of 0.94 and a\nspecificity of 0.79 to detect endometriosis. We set the criteria for triage tests at a sensitivity of 0.95 and above and a specificity of 0.50 and\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n1\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nabove, which 'rules out' the diagnosis with high accuracy if there is a negative test result (SnOUT test), or a sensitivity of 0.50 and above\nand a specificity of 0.95 and above, which 'rules in' the diagnosis with high accuracy if there is a positive result (SpIN test).\nMain results\nEleven eligible studies included 1339 participants. All the studies were of poor methodological quality. Seven studies evaluated pelvic\nendometriosis, one study considered DIE and/or ovarian endometrioma, two studies differentiated endometrioma from other ovarian cysts\nand one study addressed mapping DIE at specific anatomical sites. Fi/f_teen different diagnostic combinations were assessed, including\nblood, urinary or endometrial biomarkers, transvaginal ultrasound (TVUS) and clinical history or examination. We did not pool estimates\nof sensitivity and specificity, as each study analysed independent combinations of the non-invasive tests.\nTests that met the criteria for a replacement test were: a combination of serum IL-6 (cut-off >15.4 pg/ml) and endometrial PGP 9.5 for\npelvic endometriosis (sensitivity 1.00 (95% confidence interval (CI) 0.91 to 1.00), specificity 0.93 (95% CI, 0.80, 0.98) and the combination\nof vaginal examination and transvaginal ultrasound (TVUS) for rectal endometriosis (sensitivity 0.96 (95% CI 0.86 to 0.99), specificity 0.98\n(95% CI 0.94 to 1.00)). Tests that met the criteria for SpIN triage tests for pelvic endometriosis were: 1. a multiplication of urine vitamin-D-\nbinding protein (VDBP) and serum CA-125 (cut-off >2755) (sensitivity 0.74 (95% CI 0.60 to 0.84), specificity 0.97 (95% CI 0.86 to 1.00)) and\n2. a combination of history (length of menses), serum CA-125 (cut-off >35 U/ml) and endometrial leukocytes (sensitivity 0.61 (95% CI 0.54\nto 0.69), specificity 0.95 (95% CI 0.91 to 0.98)). For endometrioma, the following combinations qualified as SpIN test: 1. TVUS and either\nserum CA-125 (cut-off ≥25 U/ml) or CA 19.9 (cut-off ≥12 U/ml) (sensitivity 0.79 (95% CI 0.64 to 0.91), specificity 0.97 (95% CI 0.91 to 1.00)); 2.\nTVUS and serum CA 19.9 (cut-off ≥12 U/ml) (sensitivity 0.54 (95% CI 0.37 to 0.70), specificity 0.97 (95% CI 0.91 to 1.0)); 3-4. TVUS and serum\nCA-125 (cut-off ≥20 U/ml or cut-off ≥25 U/ml) (sensitivity 0.69 (95% CI 0.49 to 0.85), specificity 0.96 (95% CI 0.88 to 0.99)); 5. TVUS and serum\nCA-125 (cut-off ≥35 U/ml) (sensitivity 0.52 (95% CI 0.33 to 0.71), specificity 0.97 (95% CI 0.90 to 1.00)). A combination of vaginal examination\nand TVUS reached the threshold for a SpIN test for obliterated pouch of Douglas (sensitivity 0.87 (95% CI 0.69 to 0.96), specificity 0.98 (95%\nCI 0.95 to 1.00)), vaginal wall endometriosis (sensitivity 0.82 (95% CI 0.60 to 0.95), specificity 0.99 (95% CI 0.97 to 1.0)) and rectovaginal\nseptum endometriosis (sensitivity 0.88 (95% CI 0.47 to 1.00), specificity 0.99 (95% CI 0.96 to 1.00)).\nAll the tests were evaluated in individual studies and displayed wide CIs. Due to the heterogeneity and high risk of bias of the included\nstudies, the clinical utility of the studied combination diagnostic tests for endometriosis remains unclear.\nAuthors' conclusions\nNone of the biomarkers evaluated in this review could be evaluated in a meaningful way and there was insufficient or poor-quality evidence.\nLaparoscopy remains the gold standard for the diagnosis of endometriosis and using any non-invasive tests should only be undertaken\nin a research setting.\nP L A I N /uni00A0 L A N G U A G E /uni00A0 S U M M A R Y\nCombination of different types of tests for the non-invasive diagnosis of endometriosis\nReview Question\nCan any combination of non-invasive tests be accurate enough to replace or reduce the need for surgery in the diagnosis of endometriosis?\nBackground\nWomen with endometriosis have endometrial tissue (the tissue that lines the womb and is shed during menstruation) growing outside\nthe womb within the pelvic cavity. This tissue responds to reproductive hormones, causing painful periods, chronic lower abdominal pain\nand difficulty conceiving. Currently, the only reliable way of diagnosing endometriosis is to perform keyhole surgery and visualise the\nendometrial deposits inside the abdomen. Because surgery is risky and expensive, combinations of various tests have been evaluated for\ntheir ability to detect endometriosis non-invasively. An accurate test could lead to the diagnosis of endometriosis without the need for\nsurgery or it could reduce the need for diagnostic surgery so only women who were most likely to have endometriosis would require it.\nStudy characteristics\nThe evidence included in this review is current to April 2015. We included 11 studies on combinations of several testing methods involving\n1339 participants. All studies evaluated women of reproductive age who were undertaking diagnostic surgery to investigate symptoms\nof endometriosis or for other indications. Fi/f_teen combinations of different blood, endometrial and urinary biomarkers were studied,\nincorporating ultrasound, clinical history and examination. Each combination of tests was assessed in small individual studies.\nKey results and quality of evidence\nSeveral studies identified the combined tests that might be of value in diagnosing endometriosis, but there are too few reports to be sure\nof their diagnostic benefit.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n2\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nThe reports were of low methodological quality, which is why these results cannot be considered reliable unless confirmed in large high-\nquality studies. Overall, there is not enough evidence to demonstrate benefit of any combined non-invasive test for use in clinical practice\nfor the diagnosis of endometriosis over the current ‘gold standard’ of diagnostic laparoscopy.\nFuture research\nMore high-quality research studies are needed to accurately assess the diagnostic potential of any type of non-invasive tests or their\ncombinations that were identified in only a few studies as possibly having value in the detection of endometriosis.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n3\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n4\nS U M M A R Y /uni00A0 O F /uni00A0 F I N D I N G S\n/uni00A0\nSummary of findings 1. /uni00A0 Summary of findings table\nReview question What is the diagnostic accuracy of the combined test of different testing modalities with or without clinical history or exami-\nnation in detecting pelvic endometriosis [peritoneal endometriosis, endometrioma, DIE]?\nImportance A simple and reliable non-invasive test for endometriosis with the potential to either replace laparoscopy or to triage women\nin order to reduce surgery, would minimise surgical risk and reduce diagnostic delay\nPatients Women of reproductive age: 1) with suspected endometriosis, or 2) with persistent ovarian mass, or 3) undergoing infertility\nworkup/gynaecological laparoscopy\nSettings Hospitals (public or private of any level): outpatient clinics (general gynaecology, reproductive medicine, pelvic pain) or re-\nsearch laboratories\nReference standard Visualisation of endometriosis at surgery (laparoscopy or laparotomy), with or without histological confirmation\nStudy design Cross sectional studies with a 'single-gate' design (n = 10) or a 'two-gate' design (n = 1); prospective enrolment; a single study\ncould assess more than one test\nRisk of bias and applicability concerns Overall judgement: Poor quality of most of the studies ( no study had a 'low risk' assessment in all four domains)\nPatient selection bias High risk: 1 study; Unclear risk: 5 studies; Low risk 5 studies\nIndex test interpretation bias High risk: 9 studies; Unclear risk: 1 studies; Low risk 1 study\nReference standard interpretation bias High risk: 0 studies; Unclear risk: 3 studies; Low risk 8 studies\nFlow and timing selection bias High risk: 3 studies; Unclear risk: 0 studies; Low risk 8 studies\nApplicability concerns Concerns regarding patient selection: high concern - 6 studies, unclear concern - 0 studies; low concern 5 studies;\nConcerns regarding index test: high concern - 0 studies, unclear concern - 1 study, low concern - 10 studies;\nConcerns regarding reference standard: high concern - 0 studies; unclear concern - 0 studies; low concern - 11 studies\nDiagnostic criteria Replacement test: sensitivity ≥ 94% and specificity ≥ 79%\nSnOUT triage test: sensitivity ≥ 95% and specificity ≥ 50%\nSpIN triage test: sensitivity ≥ 50% and specificity ≥ 95%\nTest with the diagnostic estimates within 5% of the set threshold were considered as approaching the criteria\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n5\nOutcomes /uni00A0Biomarker N of studies;\nN of women\nTrue positives\n(endometrio-\nsis)\nFalse posi-\ntives (incor-\nrectly clas-\nsified as en-\ndometriosis)\nFalse nega-\ntives (incor-\nrectly\nclassified as\ndisease-free)\nTrue nega-\ntives (dis-\nease-free)\nDiagnostic\nestimates\n[95% CI]\nImplications\n1. Tests for diagnosis of overall pelvic endometriosis\n1. IL-6 [serum] + PGP 9.5 [endometrium] for\npelvic endometriosis, rASRM I-II\n-----------------------------------------------\ncut-off IL-6 >15.4 pg/ml; PGP 9.5 - present;\nboth tests positive\n1; 78 38 3 0 37 Sens 1.00\n[0.91, 1.00]\nSpec 0.93\n[0.80, 0.98]\nMeets criteria for a\nreplacement and\nSnOUT triage test;\napproaches crite-\nria for a SpIN triage\ntest\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n2. CA-125 [serum] + aromatase P450 [en-\ndometrium] for pelvic endometriosis,\nrASRM I-IV\n-----------------------------------------------\ncut-off CA-125 >35 U/ml; aromatase -\npresent; both tests positive\n1; 58 33 7 3 15 Sens 0.92\n[0.78, 0.98]\nSpec 0.68\n[0.45, 0.86]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n3. VDBP-Cr [urine] x CA-125 [serum] for\npelvic endometriosis, rASRM I-IV\n-----------------------------------------------\ncut-off > 2755; multiplication of both tests\n1; 95 42 1 15 37 Sens 0.74\n[0.60, 0.84]\nSpec 0.97\n[0.86, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n4. NNE_Cr [urine] + CA-125 [serum] for\npelvic endometriosis, rASRM III-IV\n-----------------------------------------------\ncut-off > 27.23; sum of both tests\n1; 59 30 3 9 17 Sens 0.77\n[0.61, 0.89]\nSpec 0.85\n[0.62, 0.97]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n6\n5. History + PV examination + TVUS for\npelvic endometriosis, rASRM I-IV [focus on\nendometriosis with para-ovarian adhesions]\n-----------------------------------------------\nHx - dysmenorrhoea, dyspareunia; PV\n- presence of at least one of the follow-\ning: pelvic tenderness, a fixed retroverted\nuterus, tender USL, deeply infiltrating nod-\nules on USL or in POD; TVUS - fixed ovaries\n(ovaries did not move freely over the ipsilat-\neral internal iliac vessels or pelvic sidewall\nor uterus with the gentle pressure); all tests\npositive\n1; 106 34 27 3 42 Sens 0.92\n[0.78, 0.98]\nSpec 0.61\n[0.48, 0.72]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n6. History + CA-125 [serum] + leukocytes\n[endometrium] for pelvic endometriosis,\nrASRM I-IV\n-----------------------------------------------\ncut-off Hx - length of menses; CA-125 >35 U/\nml; leukocytes - different cut-offs for each of\nthe 8 leukocyte subsets; all tests positive\n1; 368 106 10 67 185 Sens 0.61\n[0.54, 0.69]\nSpec 0.95\n[0.91, 0.98]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n7. History + CA-125 [serum] for pelvic en-\ndometriosis, rASRM I-IV\n-----------------------------------------------\ncut-off Hx - parity, past IUD, past en-\ndometriosis, alcohol intake, dyspareunia;\nCA-125 - not reported; both tests positive\n1; 101 64 12 5 20 Sens 0.93\n[0.84, 0.98]\nSpec 0.63\n[0.44, 0.79]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n2. Tests for diagnosis of DIE or ovarian endometriosis\n1. PV examination + CA-125 [serum] for DIE,\nendometrioma or severe adhesions\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; both tests positive\n1; 41 10 0 14 17 Sens 0.42\n[0.22, 0.63]\nSpec 1.00\n[0.80, 1.00]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n2. PV examination OR CA-125 [serum] for\nDIE, endometrioma or severe adhesions\n1; 41 21 3 3 14 Sens 0.88\n[0.68, 0.97]\nInsufficient ev-\nidence to draw\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n7\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; either test positive\nSpec 0.82\n[0.57, 0.96]\nmeaningful conclu-\nsions\n3. PV examination + CA125 [serum] for DIE\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; both tests positive\n1; 30 5 2 8 15 Sens 0.38\n[0.14, 0.68]\nSpec 0.88\n[0.64, 0.99]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n4. PV examination OR CA-125 [serum] for\nDIE\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; either test positive\n1; 30 11 5 2 12 Sens 0.85\n[0.55, 0.98]\nSpec 0.71\n[0.44, 0.90]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n5. PV examination + CA-125 [serum] for en-\ndometrioma\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; both tests positive\n1; 26 5 2 4 15 Sens 0.56\n[0.21, 0.86]\nSpec 0.88\n[0.64, 0.99]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n6. PV examination OR CA-125 [serum] for\nendometrioma\n-----------------------------------------------\ncut-off PV - menstrual nodularities present;\nCA-125 ≥35 U/ml; either test positive\n1; 26 8 6 1 11 Sens 0.89\n[0.51, 1.00]\nSpec 0.65\n[0.38, 0.86]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n3. Tests for differentiating ovarian endometriosis versus other benign ovarian cysts in women of reproductive age\n1. TVUS + CA-125 [serum] + CA-19.9 [serum]\nfor endometrioma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥ 25 U/ml; CA-19.9 ≥12 U/\nml; all tests positive\n1; 118 19 1 20 78 Sens 0.49\n[0.32, 0.65]\nSpec 0.99\n[0.93, 1.00]\nApproaches crite-\nria for a SpIN triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n8\n2. TVUS + (CA-125 [serum] OR CA-19.9\n[serum]) for endometrioma vs other ovarian\ncysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥ 25 U/ml; CA-19.9 ≥ 12 U/\nml; either blood test positive\n1; 118 31 2 8 77 Sens 0.79\n[0.64, 0.91]\nSpec 0.97\n[0.91, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n3. TVUS + CA-19.9 [serum] for endometri-\noma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-19.9 ≥ 12 U/ml; both tests pos-\nitive\n1; 118 21 2 18 77 Sens 0.54\n[0.37, 0.70]\nSpec 0.97\n[0.91, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n4. TVUS OR CA-19.9 [serum] for endometri-\noma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-19.9 ≥ 12 U/ml; either test pos-\nitive\n1; 118 36 24 3 55 Sens 0.92\n[0.79, 0.98]\nSpec 0.70\n[0.58, 0.79]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n5. TVUS + CA-125 [serum] for endometrioma\nvs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥ 20 U/ml; both tests pos-\nitive\n1; 101 20 3 9 69 Sens 0.69\n[0.49, 0.85]\nSpec 0.96\n[0.88, 0.99]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n6. TVUS OR CA-125 [serum] for endometri-\noma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\n1; 101 27 34 2 38 Sens 0.93\n[0.77, 0.99]\nSpec 0.53\n[0.41, 0.65]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n9\nthe ovary; CA-125 ≥20 U/ml; either test posi-\ntive\nmeaningful conclu-\nsions\n7. TVUS + CA-125 [serum] for endometrioma\nvs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥ 25 U/ml; both tests pos-\nitive\n1; 101 20 3 9 69 Sens 0.69\n[0.49, 0.85]\nSpec 0.96\n[0.88, 0.99]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n8. TVUS OR CA-125 [serum] for endometri-\noma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥ 25 U/ml; either test pos-\nitive\n1; 101 26 27 3 45 Sens 0.90\n[0.73, 0.98]\nSpec 0.63\n[0.50, 0.74]\nApproaches criteria\nfor a SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n9. TVUS + CA-125 [serum] for endometrioma\nvs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥35 U/ml; both tests posi-\ntive\n1; 101 15 2 14 70 Sens 0.52\n[0.33, 0.71]\nSpec 0.97\n[0.90, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n10. TVUS OR CA-125 [serum] for endometri-\noma vs other ovarian cysts\n-----------------------------------------------\ncut-off TVUS - presence of a round shaped\nhomogeneous hypoechoic 'tissue' within\nthe ovary; CA-125 ≥35 U/ml; either test posi-\ntive\n1; 101 26 18 3 54 Sens 0.90\n[0.73, 0.98]\nSpec 0.75\n[0.63, 0.84]\nApproaches criteria\nfor a replacement\nand SnOUT triage\ntest;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n4. Tests for mapping of DIE at specific anatomical locations\n1. PV examination + TVUS for POD oblitera-\ntion\n1; 200 26 3 4 167 Sens 0.87\n[0.69, 0.96]\nMeets criteria for a\nSpIN triage test;\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n10\n-----------------------------------------------\nPV - nodularity or stiffened or thickened\narea or a palpable cystic expansion in POD;\nTVUS - a. uterus, adnexa and rectosigmoid\ncolon fixed to each other with disappear-\nance of the peritoneal structure (complete\nPOD obliteration); b. peritoneal limits par-\ntially identified with the presence or ab-\nsence of suspended or lateralised fluid col-\nlection (incomplete POD obliteration); both\ntests positive\nSpec 0.98\n[0.95, 1.00]\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n2. PV examination + TVUS for vaginal en-\ndometriosis\n-----------------------------------------------\ncut-off PV - nodularity or stiffened or thick-\nened area or a palpable cystic expansion in\nvaginal wall; TVUS - thickening or the pres-\nence of a hypoechogenic cystic or non-cys-\ntic nodularity within the posterior vaginal\nwall\n1; 200 18 1 4 177 Sens 0.82\n[0.60, 0.95]\nSpec 0.99\n[0.97, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n3. PV examination + TVUS for RVS en-\ndometriosis\n-----------------------------------------------\nPV - nodularity or stiffened or thickened\narea or a palpable cystic expansion in RVS;\nTVUS - presence of a hypoechogenic nodu-\nlarity or cystic mass within RVS (area be-\ntween rectum and posterior vaginal wall\nfrom the level of introitus up to a level de-\nfined by the lower border of posterior lip of\ncervix); both tests positive\n1; 200 7 2 1 190 Sens 0.88\n[0.47, 1.00]\nSpec 0.99\n[0.96, 1.00]\nMeets criteria for a\nSpIN triage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\n4. PV examination + TVUS for rectal en-\ndometriosis\n-----------------------------------------------\nPV - nodularity or stiffened or thickened\narea or a palpable cystic expansion in rec-\ntosigmoid; TVUS - presence of a regular or\nirregular hypoechogenic mass distorting\n1; 200 46 3 2 149 Sens 0.96\n[0.86, 0.99]\nSpec 0.98\n[0.94, 1.00]\nMeets criteria for\na SnOUT and SpIN\ntriage test;\nInsufficient ev-\nidence to draw\nmeaningful conclu-\nsions\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n11\nand replacing the normal appearance of the\nmuscular layer of the rectal wall; both tests\npositive\n(r)ASRM: (revised) American Society for Reproductive Medicine; CA-125: cancer antigen; DIE: deep infiltrating endometriosis; IL: interleukin; IUD: intrauterine device; POD: pouch\nof Douglas; PV: per vaginam; TVUS: transvaginal ultrasound; USL: uterosacral ligament; VDBPCr: vitamin-D-binding protein level corrected for creatinine.\n/uni00A0\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nB A C K G R O U N D\nTarget condition being diagnosed\nEndometriosis\nEndometriosis is defined as an inflammatory condition\ncharacterised by endometrial-like tissue at sites outside of the\nuterus (Johnson 2013). Endometriotic lesions can occur at different\nlocations, including the pelvic peritoneum and the ovary, or\npenetrate pelvic structures below the surface of peritoneum,\nas deeply infiltrating endometriosis. Each of these types of\nendometriosis are thought to represent a separate clinical entity,\nbut also can coexist in the same woman. Rarely, endometriotic\nimplants can be found at more distant sites, including lung,\nliver, pancreas and operative scars, with consequent variations in\npresenting symptoms.\nEndometriosis afflicts 10% of women of reproductive age\ncausing dysmenorrhoea (painful periods), dyspareunia (painful\nintercourse), chronic pelvic pain and infertility (Vigano 2004). The\nclinical presentation can vary from asymptomatic and unexplained\ninfertility to severe dysmenorrhoea and chronic pain. These\nsymptoms can occur with bowel or urinary symptoms, an abnormal\npelvic examination or the presence of a pelvic mass, however\nno symptom is specific to endometriosis. The prevalence of\nendometriosis in a symptomatic population is reported as 35% to\n50% (Giudice 2004).\nWomen with endometriosis are also at increased risk of developing\nseveral cancers (Somigliana 2006 ) and autoimmune disorders\n(Sinaii 2002 ). The presence of disease is associated with\nchanges in the immune response, vascularisation, neural function,\nthe peritoneal environment and the eutopic endometrium,\nsuggesting that endometriosis is a systemic, rather than localised,\ncondition (Giudice 2004). Endometriosis has a profound effect on\npsychological and social well-being and imposes a substantial\neconomic burden on society. Women with endometriosis incur\nsignificant direct medical costs from diagnostic and therapeutic\nsurgeries, hospital admissions and fertility treatments, however\nthese costs are superceded by the indirect costs of endometriosis\nincluding absenteeism from work and loss of productivity (Gao\n2006; Simoens 2012 ). In the USA, the financial burden of\nendometriosis is estimated at US $12,419 per woman (Simoens\n2012).\nAlthough the pathogenesis of endometriosis has not been fully\nelucidated, it is commonly thought that endometriosis occurs\nwhen endometrial tissue contained within the menstrual fluid\nflows retrogradely through the fallopian tubes and implants\nat an ectopic site within the pelvic cavity (Sampson 1927 ).\nHowever, this theory does not explain the fact that although\nretrograde menstruation is seen in up to 90% of women, only\n10% of women develop endometriosis. There is evidence that a\nvariety of environmental, immunological and hormonal factors are\nassociated with endometriosis (Vigano 2004), and genetic loci that\nconfer a risk of endometriosis have been identified (Nyholt 2012).\nThe relative contribution of these and other causal factors remains\nto be elucidated.\nAlthough it is impossible to time the onset of disease, on average,\nwomen have a six- to 12-year history of symptoms before obtaining\na surgical diagnosis of endometriosis, indicative of considerable\ndiagnostic delay (Matsuzaki 2006). Untreated endometriosis is\nassociated with reduced quality of life and contributes to outcomes\nsuch as depression, inability to work, sexual dysfunction and\nmissed opportunity for motherhood (Gao 2006).\nTreatment of endometriosis\nThere is no cure for endometriosis. Treatment options include\nexpectant management, pharmacological (hormonal) therapy and\nsurgery (Johnson 2013 ). Treatment is individualised, taking into\nconsideration the therapeutic goal (pain relief or conception), and\nthe location of the disease. Current pharmacological therapies\nsuch as the combined oral contraceptive pill, progestogens,\nweak androgens and gonadotropin-releasing hormone (GnRH)\nagonists and antagonists act to reduce the effect of oestrogen\non endometrial tissues and suppress menstruation. These drugs\ncan ameliorate the symptoms of dysmenorrhoea and chronic\npelvic pain, but are associated with side effects such as breast\ndiscomfort, irritability, androgenic symptoms and bone loss.\nSurgical excision of endometriotic lesions can reduce pain\nsymptoms, however is associated with high recurrence rates\nof 40% to 50% at five years post-surgery (Guo 2009 ). Early\ntreatment of endometriosis improves pain levels and physical\nand psychological functioning. Furthermore, improvements in\nmenstrual management (the use of the Mirena coil and\nthe continuous use of the combined contraceptive pill) and\nfertility preservation (oocyte vitrification) raise the possibility of\nsuppressing the progression of endometriosis and prospectively\nmanaging subfertility in endometriosis sufferers. The potential\nsuccess of these preventative strategies is dependent on an\naccurate and early diagnosis. A major impediment to earlier and\nmore efficacious treatment of this disease is diagnostic delay due\nto the invasive nature of standard diagnostic tests (Dmowski 1997).\nClinical history and pelvic examination can raise the possibility\nof a diagnosis of endometriosis, but the heterogeneity in clinical\npresentation, the high prevalence of asymptomatic endometriosis\n(2% to 50%), and the poor association between presenting\nsymptoms and severity of the disease contribute to the difficulty\nin obtaining a reliable diagnosis of endometriosis based solely on\npresenting symptoms (Ballard 2008; Fauconnier 2005; Spaczynski\n2003). Although an abnormal pelvic examination correlates with\nthe presence of endometriosis on laparoscopy in 70% to 90%\nof cases (Ling 1999 ), there is a wide differential diagnosis for\nmost positive physical findings. Furthermore, a normal clinical\nexamination does not exclude endometriosis, as laparoscopically-\nproven disease has been diagnosed in more than 50% women\nwith a clinically normal pelvic examination (Eskenazi 2001). A\nvariety of tests utilising pelvic imaging, blood markers, eutopic\nendometrium characteristics, urinary markers or peritoneal fluid\ncomponents have been suggested as diagnostic measures for\nendometriosis. Although large numbers of the reported markers\ndistinguish women with and without endometriosis in small pilot\nstudies, many do not show convincing potential as a diagnostic\ntest when they are evaluated in larger studies by different research\ngroups. The diagnostic value of these tests has not previously been\nfully systematically evaluated and summarised using Cochrane\nmethods. Currently, there is no simple non-invasive test for the\ndiagnosis of endometriosis that is routinely implemented in clinical\npractice.\nSurgical diagnostic procedures for endometriosis include\nlaparoscopy (minimal access, or keyhole surgery) or laparotomy\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n12\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n(open surgery via an abdominal incision). In the last several\ndecades, laparoscopy has become an increasingly common\nprocedure and has largely replaced traditional open surgery\nin women suspected of having endometriosis (Yeung 2009).\nLaparoscopy has significant advantages over laparotomy creating\nfewer complications and shorter recovery times. Furthermore, a\nmagnified view at laparoscopy allows better visualisation of the\nperitoneal cavity. Despite continuing controversy in the literature\nwith regard to the superiority of one surgical modality over\nanother in treating pelvic pathology, laparoscopy is the preferred\ntechnique to evaluate the pelvis and abdomen and to treat\nbenign conditions such as ovarian endometriomas (Medeiros\n2009). Surgery is currently also the only accepted way to determine\nthe extent and severity of endometriosis. Several classification\nsystems have been suggested for endometriosis (Adamson 2008 ;\nBatt 2003; Chapron 2003a; Martin 2006 ), but most researchers\nand clinicians use the revised American Society for Reproductive\nMedicine (rASRM) classification, which is internationally accepted\nas a respected tool for the objective assessment of the disease\n(American Society for Reproductive Medicine 1997). The rASRM\nclassification system considers appearance, size and depth of\nperitoneal or ovarian implants and adhesions visualised during\nlaparoscopy (Table 1) and allows uniform documentation of the\nextent of disease. Unfortunately this classification system has little\nvalue in clinical practice due to the lack of correlation between\nlaparoscopic staging, the severity of symptoms and response to\ntreatment (Chapron 2003b; Guzick 1997; Vercellini 1996). A recent\nendeavour to attain consensus around the optimal classification for\nendometriosis has been undertaken by the World Endometriosis\nSociety (Johnson 2015).\nThe European Society of Human Reproduction and Embryology\n(ESHRE) Special Interest Group for Endometriosis stated in their\nguidelines for the diagnosis and treatment of endometriosis that\nfor women presenting with symptoms suggestive of endometriosis,\na definitive diagnosis of most forms of endometriosis requires\nvisual inspection of the pelvis at laparoscopy as the 'gold standard'\ninvestigation (Kennedy 2005). Currently the visual or histological\nidentification of endometriotic tissue in the pelvic cavity during\nsurgery is not just the best available but the only diagnostic test for\nendometriosis in clinical practice.\nThe disadvantages of laparoscopic surgery include, but are not\nlimited to, the high cost, the need for general anaesthesia and the\npotential for adhesion formation post procedure. Laparoscopy has\nbeen associated with a 2% risk of injury to pelvic organs, a 0.001%\nrisk of damaging a major blood vessel and a mortality rate of\n0.0001% (Chapron 2003c). Even though the major complications of\nlaparoscopy are rare, it is difficult to determine the exact incidence\nof complications, and delayed recognition adds to surgical\nmorbidity and mortality. Only one third of women who undertake\na laparoscopic procedure will receive a diagnosis of endometriosis;\ntherefore many disease-free women are unnecessarily exposed to\nsurgical risk (Frishman 2006).\nThe validity of laparoscopy as a reference test for endometriosis\nhas been assessed as being highly dependent on the skills of the\nsurgeon. The diagnostic accuracy of laparoscopic visualisation has\nbeen compared with histological confirmation in a sole systematic\nreview and it was estimated as having a sensitivity of 0.94 and\nspecificity of 0.79 ( Wykes 2004). Subsequent studies suggested\nthat incorporation of histological verification in the diagnosis\nof endometriosis may improve diagnostic accuracy (Almeida\nFilho 2008 ; Marchino 2005; Stegmann 2008), but these papers\nhave not been systematically reviewed. The clinical significance\nof histological verification remains debatable, and a diagnosis\nbased on visual findings can be considered reliable with an\naccurate inspection of the abdominal cavity by properly trained\nand experienced surgeons (Redwine 2003). Furthermore, excised\npotential endometriotic tissues are rarely serially sectioned in\nclinical practice and small lesions can be missed by pathologists\nin mild disease. Thus sampling inconsistencies are also likely to\ninfluence the accuracy of histological reporting.\nSummary\nA diagnostic test in place of surgery would reduce associated\nsurgical risks, increase diagnostic accessibility and improve\ntreatment outcomes. The need for an accurate and non-\ninvasive diagnostic test for endometriosis continues to encourage\nextensive research in the field and was endorsed at the\ninternational consensus workshop at the 10th World Congress of\nEndometriosis in 2008 (Rogers 2009). Although multiple markers\nand imaging techniques have been explored as diagnostic tests\nfor endometriosis, none of them have been implemented routinely\nin clinical practice and many have not been subject to systematic\nreview.\nIndex test(s)\nThis review assesses combinations of tests, including blood,\nurine and endometrial biomarkers and imaging modalities that\nhave been proposed as non-invasive tests for the diagnosis of\nendometriosis (Table 2). This review is part of the review series\non non-invasive diagnostic tests for endometriosis. The other\nreviews from this series are: 'Blood biomarkers for the non-\ninvasive diagnosis of endometriosis' (Nisenblat 2016a), 'Endometrial\nbiomarkers for the non-invasive diagnosis of endometriosis' (Gupta\n2016), ' Urinary biomarkers for the non-invasive diagnosis of\nendometriosis' ( Liu 2015 ) and ' Imaging modalities for the non-\ninvasive diagnosis of endometriosis' (Nisenblat 2016b).\nThe definition of ‘non-invasive’ varies between medical dictionaries\nbut refers to a procedure that does not involve penetration of\nskin or physical entrance to the body (McGraw-Hill Dictionary\nof Medicine 2006 ; The Gale Encyclopedia of Medicine 2008).\nAlthough intracavity imaging and tests involving venipuncture\nor endometrial sampling are invasive by this definition, when\ncompared to diagnostic surgery for endometriosis, these tests are\ngenerally considered to be 'non-invasive' or 'minimally invasive'.\nFor the purpose of these reviews, we will define all tests that do not\ninvolve anaesthesia and surgery as non-invasive.\nThe potential advantages of using imaging modalities, blood\nbiomarkers, endometrial biomarkers, urinary biomarkers, clinical\nparameters that include examination findings and clinical history,\nor a combination of them to diagnose endometriosis, include\ntheir less invasive nature, lower cost and increased availability\nwhen compared to surgery. These tests are more acceptable to\nwomen, and usually provide a rapid result. However, the testing\nis dependant on the reliability of laboratory techniques and\nquality control protocols for the biomarker assays, on the skills\nof the operators performing imaging tests or examination and on\nwomen's access to appropriate radiology services.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n13\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nThe cellular and molecular processes that have been identified to\ncharacterise ectopic endometrium and peritoneal fluid in human\nand animal models ( D'Hooghe 2001 ; Hull 2008 ; Kao 2003 ) have\ninspired the use of markers of these pathophysiological processes\npresent in blood, urine and endometrium samples as a single test\nor a combination of several biomarkers. Of these tests, urinary\nbiomarker discovery is a new and rapidly expanding field with most\nstudies published in the last five years. Several large systematic\nreviews of all proposed biomarkers for endometriosis identified\nmultiple putative biomarkers, but none of these biomarkers could\nbe recommended for use in clinical practice (May 2010 ; May\n2011), which was supported by a more recent narrative review\n(Fassbender 2015). The biomarker research in endometriosis\ntends to shi/f_t towards diagnostic panels which include one or\nseveral testing modalities such as blood, endometrial or imaging\ntests. Systematic reviews on imaging in endometriosis (Guerriero\n2015; Hudelist 2011a;Medeiros 2015; Moore 2002) and narrative\nreviews on the topic primarily addressed diagnostic performance\nof imaging methods and not as a part of a diagnostic panel.\nIn line with general consensus, clinical parameters (history and\nexamination) have low reliability in the diagnosis of endometriosis,\nhowever they may improve the diagnostic performance of other\nnon-invasive tests when incorporated in a diagnostic model. So\nfar, combinations of non-invasive tests have only been assessed\nin a limited number of small studies, which vary in the type of\nmethodology and tests used and type of endometriosis evaluated.\nThere is a current need to evaluate the diagnotic test accuracy\nof the combination of different testing modalities and diagnostic\nalgorithms for endometriosis using Cochrane methods.\nClinical pathway\nWomen presenting with symptoms of endometriosis\n(dysmenorrhoea, dyspareunia, chronic pelvic pain or difficulty\nconceiving) generally are investigated with a pelvic ultrasound scan\nto exclude other pathologies, which is in line with international\nguidelines (Dunselman 2014; SOGC 2010; ACOG 2010). There are no\nother standard investigative tests, and although evidence suggests\nthat magnetic resonance imaging (MRI) is superior to ultrasound,\nit is used conservatively because of its cost. If women seek pain\nmanagement rather than conception, physicians generally initiate\nempirical treatment with progestogens or the combined oral\ncontraceptive pill. Diagnostic laparoscopy is considered if empirical\ntreatment fails or if women decline or do not tolerate empirical\ntreatment. In women who have difficulty conceiving, laparoscopy\ncan be undertaken before fertility treatment (particularly if severe\npelvic pain or endometrioma are present) or a/f_ter failed assisted\nreproductive technology (ART) treatments. Endometriosis can be\nalso diagnosed during fertility investigations in women who have\nminimal or no pain symptomatology.\nOn average there is a delay of between six to 12 years from\nonset of symptoms to definitive diagnosis at surgery (Dunselman\n2014). Early referral to a gynaecologist with the capability to\nperform diagnostic surgery is expected to reduce time to diagnosis.\nCollectively, young women, women in remote and rural locations\nand women of lower socioeconomic status have reduced access\nto surgery, and are less likely to obtain a prompt diagnosis of\nendometriosis.\nPrior test(s)\nMost women presenting with symptoms suggestive of\nendometriosis have a full history and examination and a routine\ngynaecological ultrasound before a decision is made to have\ndiagnostic surgery. However, there is no consensus on whether or\nnot ultrasound or any other test should be routinely used as part of\na standardised approach.\nRole of index test(s)\nA new diagnostic test can fulfil one of three roles.\n1. Replacement: replacing an existing test by having more\naccuracy, or a similar accuracy with other advantages.\n2. Triage: used as an initial step in a diagnostic pathway to identify\nthe group of women who need further testing with an existing\ntest. Although ideally a triage test has a high sensitivity and\nspecificity, it may have a lower sensitivity but higher specificity\nthan the current test or vice versa. The triage test does not aim\nto improve the diagnostic accuracy of the existing test but rather\nto reduce the number of individuals having an unnecessary\ndiagnostic test.\n3. Add-on: used in addition to existing testing to improve\ndiagnostic performance (Bossuyt 2008).\nIdeally, a diagnostic test is expected to correctly identify all women\nwith a disease and to exclude all women without that disease, in\nother words it should have a sensitivity and specificity of 1.00. A\nhigh sensitivity indicates that there are a low number of women\nwho have a negative test and do have the disease (i.e. a low\nnumber of false-negative results). High specificity corresponds\nto a low number of women who have a positive test but do\nnot have the disease (i.e. low false-positive results). In practice,\nhowever, it is extremely rare to find a test with equally high\nsensitivity and specificity. An acceptable replacement test would\nneed to have a similar or higher sensitivity and specificity than\nthe current gold standard of laparoscopy. The only systematic\nreview that determines the accuracy of laparoscopy in diagnosing\nendometriosis reported a sensitivity of 0.94, and a specificity of 0.79\n(Wykes 2004) and we have taken this as a cut-off for a replacement\ntest.\nThe purpose of triage tests can vary depending on the clinical\ncontext and a woman’s priorities. One reasonable approach is to\nexclude the diagnosis to avoid further unnecessary and expensive\ndiagnostic investigation. High-sensitivity tests have few false\nnegative results and act to rule conditions out (SnOUT). A negative\nresult from a test with high sensitivity will exclude the disease with\nhigh certainty independent of the specificity. As women without\ndisease would be assured of having a negative test, unnecessary\ninvasive interventions can be avoided. However, a positive result\nhas less diagnostic value particularly when the specificity is low.\nWe predetermined that a clinically useful SnOUT triage test should\nhave a sensitivity of 0.95 or more and a specificity of 0.50 and\nabove. We set the sensitivity cut-off for a SnOUT triage test at 0.95\nand above, assuming that a 0.05 false negative rate is statistically\nand clinically acceptable. We set the specificity cut-off at 0.50 and\nabove, to avoid diagnostic uncertainty in more than 50% of the\npopulation with a positive result.\nAn alternative approach would be to avoid a missed diagnosis.\nHigh-specificity tests have few false positive results and act to\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n14\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nrule conditions “in” (SpIN). A positive result for a highly specific\ntriage test indicates a high likelihood of having endometriosis.\nThis information could be used to prioritise these women for\nsurgical treatment. A positive SpIN test could also provide a clinical\nrationale to start targeted disease-specific medical management in\na woman without a surgical diagnosis, under the assumption that\ndisease is present. Surgical management could then be reserved\nfor cases when conservative treatment fails. This is particularly\nrelevant in some populations where the therapeutic benefits of\nsurgery for endometriosis have to be carefully balanced with the\ndisadvantages (e.g. young women, women with medical conditions\nor pain-free women with a history of infertility). In this scenario we\nconsidered a sensitivity of 0.50 and above and a specificity of 0.95\nand higher as suitable cut-offs for a SpIN triage test.\nWe evaluated combinations of tests for their potential to replace\nsurgery (replacement test) or to improve the selection of women for\nsurgery (triage test to rule out (SnOUT) or rule in (SpIN) the disease).\nBoth types of triage test are clinically useful, minimising the\nnumber of unnecessary interventions. Sequential implementation\nof SnOUT and SpIN tests can also optimise a diagnostic algorithm\n(Figure 1). We did not assess any test as an add-on test, as we sought\ntests that reduce the need for surgery and not tests that improve\nthe accuracy of the currently available surgical diagnosis.\n/uni00A0\nFigure 1. /uni00A0 Sequential approach to non-invasive testing of endometriosis\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n15\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nAlternative test(s)\nThere are no alternative tests for the diagnosis of endometriosis\nthat are available in routine clinical practice.\nRationale\nMany women with endometriosis suffer long-standing pelvic pain\nand infertility prior to a diagnosis. Surgery is the only current\nmethod of diagnosing endometriosis, but it is associated with\nhigh costs and surgical risks. Simple and reliable non-invasive\ntests for endometriosis, with the potential to either replace\nlaparoscopy or to triage women in order to reduce surgery, would\nminimise surgical risk and reduce diagnostic delay. Physicians\ncould then detect endometriosis at less advanced stages and\ninstitute earlier interventions. Early diagnosis would provide the\nopportunity for a preventive approach for this debilitating disease,\npotentially reducing healthcare-related costs and favouring more\ncost-effective and efficient treatments. Furthermore, identifying\nthe tests that do not pertain to endometriotic disease would help\nclinicians and researchers focus on clinically relevant biomarker\ndetection.\nO B J E C T I V E S\nPrimary objectives\nTo estimate the diagnostic accuracy of any combination of non-\ninvasive tests for the diagnosis of pelvic endometriosis (peritoneal\nand/ or ovarian or deep infiltrating) compared to surgical diagnosis\nas a reference standard. The combined tests were evaluated\nas replacement tests for diagnostic surgery as well as triage\ntests which would assist decision-making to undertake diagnostic\nsurgery for endometriosis.\nSecondary objectives\nTo investigate the influence of heterogeneity on the diagnostic\naccuracy of combined non-invasive test for endometriosis.\nPotiential sources of heterogeneity include:\n1. characteristics of the study population: age (adolescents versus\nlater reproductive years); clinical presentation (subfertility,\npelvic pain, ovarian mass, asymptomatic women); stage of\ndisease (rASRM classification system); geographic location of\nstudy;\n2. histological confirmation in conjunction with laparoscopic\nvisualisation compared to laparoscopic visualisation alone;\n3. changes in technology over time: year of publication;\nmodifications applied to conventional laboratory techniques;\n4. methodological quality: differences in the QUADAS-2 (Quality\nAssessment of Diagnostic Accuracy Studies-2) evaluation (Table\n3), including a) low versus unclear or high risk; b) consecutive\nversus non-consecutive enrolment; c) blinding of surgeons to\nthe results of index tests;\n5. study design ('single-gate design' versus 'two-gate design'\nstudies).\nM E T H O D S\nCriteria for considering studies for this review\nTypes of studies\nPublished peer-reviewed studies that compared the results of a\ncombination of several testing modalities with the results obtained\nfrom a surgical diagnosis of endometriosis.\nStudies were included if they included the following study designs.\n1. Randomised controlled trials (RCTs).\n2. Observational studies with the following designs.\na. Single-gate design (studies with a single set of inclusion\ncriteria defined by clinical presentation). All participants had\nclinically suspected endometriosis.\nb. Two-gate design (studies where participants are sampled\nfrom distinct populations with respect to clinical\npresentation). The same study includes participants with a\nclinical suspicion of having the target condition (e.g. women\nwith pelvic pain) and also participants in whom the target\ncondition is not suspected (e.g. women admitted for tubal\nligation). Two-gate studies were eligible only where all cases\nand controls belonged to the same population with respect\nto the reference standard (i.e. all the participants were\nscheduled for laparoscopy) (Rutjes 2005).\n3. For studies on biological samples - performed on prospectively\ncollected samples, irrespective of the actual time of the test\nassay. The timing of sample collection relative to surgery\nis important because the surgical excision of endometriotic\nlesions could influence biomarker expression and hence bias\nthe results. Therefore, we only included studies where the\nbiological sample was collected before the surgical procedure,\ni.e. prospectively collected. We considered to be eligible\nthe studies performed on tissue bank samples collected\nfrom prospectively recruited, well-defined populations, which\nprevented the omission of valuable data from adequately\ndesigned studies. The time interval between sample collection\nand laboratory testing may influence test outcomes, which\ncould be dependent on sample storage conditions and the\nstability of each individual biomarker during storage and freeze-\nthawing. This information was not readily available for most\nmolecules, and we did not address it in this review, but we will\nconsider it in future updates if more evidence emerges.\n4. For studies on clinical or imaging examination - performed\non prospectively recruited women with the index test being\ncompleted prior to the reference standard.\nWe did not impose limits on eligibility related to the healthcare\nsettings where the study took place, the language of publication,\nthe number of participants in the included studies or the number of\nstudies that evaluated each index test.\nThe following studies were excluded.\n1. Narrative or systematic reviews.\n2. Studies of retrospective design where the sample collection,\nclinical or imaging examination were performed a/f_ter execution\nof reference test.\n3. Studies of retrospective design where the participants were\nselected from retrospective review of the case notes/ archived\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n16\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nsamples and information on recruitment methods or study\npopulation was not available.\n4. Case reports or case series.\n5. Studies reported only in abstract form or in conference\nproceedings where the full text was not available. We applied\nthis limitation a/f_ter facing substantial difficulty in obtaining\nthe information from the abstracts, which precluded a reliable\nassessment of eligibility and methodological quality.\nParticipants\nStudy participants included women of reproductive age (puberty\nto menopause) with suspected endometriosis based on clinical\nsymptoms or pelvic examination, who undertook both the index\ntest and reference standard.\nThe participants were selected from populations of women\nundergoing abdominal surgery for the following indications:\n1) clinically suspected endometriosis (pelvic pain, infertility,\nabnormal pelvic examination, or a combination of the above); 2)\novarian mass, regardless of symptoms; 3) a mixed group, which\nconsists of women with suspected endometriosis/ovarian mass or\nwomen with other benign gynaecological conditions (e.g. surgical\nsterilisation, fibroid uterus, etc). Asymptomatic women who had\nan incidental finding of endometriosis at surgery performed for\nanother indication were also included.\nStudies that included participants of postmenopausal age were\neligible when the data for the reproductive age group was\navailable in isolation. We excluded studies with participants that\nclearly would not undergo the index test in the relevant clinical\nsituation or would not benefit from the test (e.g. women with\nectopic pregnancies or acute pelvic inflammatory disease). We\nalso excluded publications that only analysed participants with a\npositive index test or reference standard and did not provide data\nfor the whole cohort.\nIndex tests\nWe assessed any combination of non-invasive tests for\nendometriosis comprising of more than one test modality. This\nincluded the combinations of blood, endometrial, urine and\nimaging tests with or without clinical parameters, such as pre-\ndefined examination findings, specific symptoms or characteristics\n(e.g. length of menstrual cycle). The assessed index tests are\npresented in Table 2.\nThe panel of biomarkers from the same single category (e.g.\nseveral blood biomarkers or combination of imaging methods) was\nassessed in the relevant review on the topic and are presented\nseparately in other reviews from this series. The studies that\nsolely assessed specific technical aspects, qualitative description\nof lesion appearance or interobserver variability of the index\ntests without reporting the data on diagnostic performance were\nexcluded from the review. When the evaluated biomarker(s)\nshowed differential expression between the groups of women with\nand without endometriosis, the publication was considered only if\nthe data were reported with sufficient detail for the construction\nof 2 x 2 contingency tables. However, when the contingency tables\nwere not available because the expression level of index test did\nnot significantly differ between the groups and the inclusion criteria\nwere otherwise met, we made a critical appraisal and presented the\nstudy in the descriptive part of the review. Thus, we evaluated the\nadequately designed studies that identified biomarkers without\ndiagnostic value, as they provide information that is likely to focus\nfuture research on other more clinically useful biomarkers.This\nmethodology also identified biomarkers that were associated with\nendometriosis in some but not other studies. We did not include\nevaluations of screening or predictive accuracy tests in this review.\nWe considered the diagnostic performance of an index test to be\nhigh when the test reached the criteria for a replacement test\n(sensitivity of equal or greater than 0.94 with specificity of equal or\ngreater than 0.79) or triage test (sensitivity of equal or greater than\n0.95 with specificity of equal or greater than 0.50 or vice versa) or\napproached these criteria (diagnostic estimates within 0.05 of the\nset thresholds). We considered all other diagnostic estimates to be\nlow.\nTarget conditions\nPelvic endometriosis, defined as endometrial tissue located in the\npelvic cavity: involving any of the pelvic organs, peritoneum and\npouch of Douglas (POD).\nThree types of pelvic endometriosis were assessed.\n1. Peritoneal endometriosis, defined as endometrial deposits\ndetected on the peritoneum covering pelvic organs, pelvic side\nwalls or POD.\n2. Ovarian endometriosis (endometrioma), defined as an ovarian\ncyst lined by endometrial tissue, appearing as an ovarian mass\nof varying size.\n3. Deep infiltrating endometriosis (DIE), defined as subperitoneal\ninfiltration of endometrial implants, i.e. when the endometriotic\nimplants penetrate the retroperitoneal space at a distance\nof 5 mm or more (Koninckx 1991). DIE may be present in\nmultiple locations, involving either the anterior or posterior\npelvic compartments, or both.\nWe did not include certain rare types of endometriosis such\nas extrapelvic, bladder and ureteric endometriosis because the\nmajority were reported in case reports or case series, and\nlaparoscopy or laparotomy are not reliable reference standards for\nthese conditions.\nWe excluded the studies where diagnosis of endometriosis was\nnot the primary outcome (e.g. malignant versus benign masses\nor normal versus abnormal pelvis) and the separate data for\nendometriosis were not available.\nWe also excluded the studies where the findings of the index test\nformed the basis of selection for the reference standard, because\nthis was likely to distort an assessment of the diagnostic value of\nthe index test.\nWe did include studies that recruited selected populations\nof women with endometriosis (i.e. those with specific rASRM\nstages), because there is a poor correlation between the rASRM\nclassification and infertility or pain symptoms. Exclusion of these\nstudies could result in a loss of potentially important diagnostic\ninformation from otherwise eligible publications. Where possible,\nthe impact of these studies was addressed in the assessments\nof heterogeneity. When a study analysed a large population with\na wide spectrum of endometriosis and additionally reported a\nsubgroup analysis of the different stages of disease severity, we\nonly considered estimates for the entire population. This is because\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n17\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\na subgroup analysis would not directly address the review question\nregarding the clinical utility of the biomarker in disease detection.\nReference standards\nThe reference standard was visualisation of endometriosis at\nsurgery (laparoscopy or laparotomy) with or without histological\nconfirmation, as this is currently the best available test for\nendometriosis. Information regarding the inter- and intra-observer\ncorrelation of the reference standard was reviewed if reported.\nWe only included studies in which the reference test was performed\nwithin 12 months of the sample collection or imaging test, on the\nassumption that the disease status could change within a period\nof one year or longer, either naturally or as a result of treatment.\nWe excluded studies in which the participants did not undergo the\nreference standard or where the findings of the index test formed\nthe basis of selection for undertaking the reference standard, as this\nwas likely to distort an assessment of the diagnostic value of the\nindex test.\nSummary of inclusion/exclusion criteria\nInclusion criteria\n1. Types of studies\na. Published and peer-reviewed\nb. RCTs\nc. Observational designs, including:\ni. single-gate design (single set of inclusion criteria defined\nby clinical presentation): all the participants had clinically\nsuspected endometriosis;\nii. two-gate design (two sets of inclusion criteria with respect\nto clinical presentation and one set of inclusion criteria\nwith respect to reference standard): the participants with\nor without a clinical suspicion of endometriosis scheduled\nfor abdominal surgery.\nd. Performed on prospectively collected samples, including the\ntissue bank samples collected from a prospectively recruited\nwell-defined population; for clinical/imaging testing -\nperformed on prospectively recruited participants when\nindex test performed before reference standard\ne. Published in any language\nf. Performed in any healthcare setting\ng. Any sample size\n2. Participants\na. Women of reproductive age\nb. Clinically suspected endometriosis, this also included:\ni. women who underwent abdominal surgery for other\nbenign gynaecological conditions and had a surgical\nassessment for presence/absence of endometriosis;\nii. asymptomatic women who have an incidental finding\nof endometriosis at surgery performed for another\nindication.\nc. Undertook both the index test and reference standard\n3. Index tests\n1. a. Combined non-invasive tests for endometriosis comprising\nof several testing modalities, including the combinations of\nblood, endometrial, urine and imaging tests with or without\nclinical parameters\nb. Data reported in sufficient detail for the construction of 2\nx 2 tables for the tests that showed differential expression\nbetween the groups\nc. Tests where a 2 x 2 table could not be constructed because\nthe results did not differ between women with and without\nendometriosis, but all other inclusion criteria were met\n2. Target condition\na. Pelvic endometriosis\ni. peritoneal endometriosis;\nii. ovarian endometrioma;\niii. DIE;\niv. combinations of the above.\n3. Reference standard\na. Surgical visualisation of lesions for the diagnosis of\nendometriosis (laparoscopy or laparotomy) with or without\nhistological verification\nb. Performed within 12 months of the endometrial sample\ncollection\nExclusion criteria\n1. Types of studies\na. Narrative or systematic reviews\nb. Retrospective design where biological samples were\ncollected or clinical/ imaging index test was performed a/f_ter\nexecution of reference test\nc. Prospectively collected samples that were selected from the\narchived material, but information on the study population\nor the selection process was unclear\nd. Case reports or case series\ne. Conference proceeding\n2. Participants\na. Included cohort was not representative of the target\npopulation that would benefit from the test (e.g. women with\nknown genital tract malignancy, ectopic pregnancies or acute\npelvic inflammatory disease)\nb. Study included participants of postmenopausal age and the\ndata for the reproductive age group were not available in\nisolation\nc. Analysis only included participants with positive index test or\npositive reference standard\n3. Index tests\na. Biomarkers presented as a single test or a panel of several\nmarkers from the same category (e.g. only blood biomarkers)\nb. Study presented only specific technical aspects of an\nindex test or focused on the biological events, rather than\ndiagnostic performance of the test\nc. Study assessed screening or predictive test accuracy\n4. Target condition\na. Endometriosis was not the primary outcome of the trial (e.g.\nmalignant versus benign masses or normal versus abnormal\npelvis)\nb. Atypical, rare sites of endometriosis\n5. Reference standard\na. Reference standard performed only in a subset of the study/\ncontrol group\nb. Findings of the index test formed the basis of selection for the\nreference standard\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n18\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nc. Rather than specified in inclusion criteria\nSearch methods for identification of studies\nWe developed the search strategy in collaboration with the Trials\nSearch Co-ordinator of the Gynaecology and Fertility Review\nGroup, following recommendations of the Cochrane Handbook for\nSystematic Reviews of Diagnostic Test Accuracy (de Vet 2008). We did\nnot limit the searches to particular types of study design or impose\nlanguage or publication date restrictions. The search strategy used\na combination of both free text words and index terms. We initially\ncreated the search for one broad review looking at all diagnostic\nmarkers for endometriosis, but due to complexity, the review team\nsplit the originally planned review into five separate reviews. We\ndesigned two separate search strategies: one for all the biomarkers-\nbased tests, and another for the imaging tests; both strategies were\nutilised in this review. We searched CENTRAL to July 2015 and\nperformed all other searches from database inception to April 2015.\nWe present the search strategies for each database and the number\nof hits per search in Appendix 1; Appendix 2; Appendix 3; Appendix\n4; Appendix 5 ; Appendix 6 ; Appendix 7 ; Appendix 8 ; Appendix 9 ;\nAppendix 10. The summary of the results is presented in Results of\nthe search.\nElectronic searches\nWe searched the following databases to identify the published\narticles that assessed the diagnostic value of non-invasive tests for\nendometriosis.\na. CENTRAL (2015, July).\nb. MEDLINE (inception to May 2015).\nc. EMBASE (inception to May 2015).\nd. CINAHL (inception to April 2015).\ne. PsycINFO (inception to April 2015).\nf. Web of Science (inception to April 2015).\ng. LILACS (inception to April 2015).\nh. OAIster (inception to April 2015).\ni. TRIP (inception to April 2015).\nj. Databases of the trial registers:\ni. ClinicalTrials.gov (inception to April 2015);\nii. World Health Organization (WHO) International Clinical\nTrials Registry Platform (ICTRP) (inception to April 2015).\nk. Databases to identify reviews and guidelines as sources of\nreferences to potentially relevant studies.\ni. MEDION (inception to January 2014, the last available\ndate);\nii. DARE (inception to April 2015);\niii. PubMed, a 'Systematic Review' search under the 'Clinical\nQueries' link (inception to April 2015).\nl. Searches for papers recently published and not yet indexed\nin the major databases:\ni. PubMed (simple search for the six months to April 2015).\nSearching other resources\nWe handsearched the reference list of all relevant publications\n(retrieved full texts of the key articles and identified reviews).\nWe abandoned an initial attempt to locate the grey literature\n(unpublished studies and conference proceedings), as we faced\nsubstantial difficulty in obtaining full-text publications or further\ndetails of studies reported in an abstract form.\nData collection and analysis\nSelection of studies\nTwo authors of this review (LP, VN) and four other authors or\ncontributors of the other reviews from this series (Devashana\nGupta, Emily Liu, Rabia Shaikh and Deepika Arora) scanned the\ntitles of studies identified by our search to remove any clearly\nirrelevant articles. The titles and abstracts of the remaining\nstudies were reviewed to select potentially relevant publications.\nThe relevant articles were then divided into four categories\nof endometriosis biomarkers: serum, endometrial, urinary and\ncombined tests (imaging had already been completed in a separate\nsearch). Three of the combined biomarker review authors (LP,\nNJ, VN) independently reviewed each of the full-text versions of\nthe articles selected by title and abstract and assessed them for\neligibility for inclusion, based on the criteria listed above under\nCriteria for considering studies for this review. A single failed\neligibility criterion was sufficient for a study to be excluded from the\nreview.\nThe review authors who assessed the relevance of the studies and\neligibility for inclusion were not blind to the information about each\narticle, including the publishing journal, the names of authors, the\ninstitution and the results. Any disagreements were resolved by\ndiscussion.\nWhen papers updated previous publications and were performed\non the same study population at different recruitment points, the\nmost complete data set that superseded previous publications\nwas used to avoid double counting participants or studies. Missing\ndata were retrieved by directly contacting authors to clarify\nstudy eligibility. When potentially relevant studies were found in\nlanguages other than English, a translation was undertaken. For\nexcluded studies, the reasons for exclusion and details of which\ncriteria were not met were documented. The characteristics of\nincluded and excluded studies are presented under Characteristics\nof included studies and Characteristics of excluded studies,\nrespectively.\nData extraction and management\nData were independently extracted from eligible studies by three\nreview authors (LP, NJ, VN) and any disagreement was resolved by\nconsensus. If required, we contacted study investigators to resolve\nany questions regarding the data.\nTo collect details from included studies, a data extraction form\nwas specifically designed for this review and pilot-tested on\nthree studies of diagnostic accuracy tests for endometriosis. The\nfollowing information was recorded for each study.\n1. General information and study design: first author, year of\npublication, country, language, setting, objectives, inclusion/\nexclusion criteria, type of enrolment.\n2. Characteristics of the study participants: age, symptoms/\nhistory/previous tests, type of target condition and its\nprevalence in the study population, number of participants\nenrolled and available for analysis, reasons for withdrawal.\n3. Features of the index test and reference standard: type,\ndiagnostic criteria, number and experience of the operators,\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n19\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nblinding of the operators to other tests or clinical data,\ninterobserver variability, time interval between index test and\nreference standard.\n4. The reported number of true positives (TP), false negatives (FN),\ntrue negatives (TN) and false positives (FP) was used to construct\na two-by-two (2 x 2) table for each index test. If these values were\nnot reported, we attempted to reconstruct the 2 x 2 tables from\nthe summary estimates presented in the article.\nData were extracted into Review Manager® (RevMan) so/f_tware,\nwhich was used to display graphically the quality assessment, the\ndiagnostic estimates data, and the descriptive analyses.\nAssessment of methodological quality\nTo assess the quality of each included study, we used QUADAS-2,\na modified version of the QUADAS tool for systematic reviews of\ndiagnostic accuracy studies (Whiting 2011).\nThe review-specific QUADAS-2 tool and explanatory document are\npresented in Table 3. Each paper was judged as having a 'low',\n'high' or 'unclear' risk for each of four domains and concerns\nabout applicability were assessed in three domains. We considered\nstudies as having low methodological quality when they were\nat high or unclear risk of bias or when we had a high concern\nregarding applicability at least in one domain. The assessment\nof each included study was performed independently by three\nreview authors (LP, NJ, VN) and disagreements were settled by\nconsensus. Two review authors (LP, NJ) independently piloted\nthe topic-specific tool to rate four of the included studies with a\nhigh level of agreement. Modifications specific to the combined\nbiomarkers review were made to the signalling questions of the\noriginal QUADAS-2 tool and were as following.\nDomain 1\nWe rephrased an original signalling question, 'Was a case-control\ndesign avoided?' as 'Was a two-gate design avoided?'. The\ndiagnostic studies are cross-sectional in nature, aiming to compare\nthe result of an index test with the result of the reference standard\nin the same group of participants. Study investigators measure the\nparameters at a single point in time and classify the groups by\nthe outcome of the reference standard, albeit they perform the\nanalysis retrospectively. Therefore, unlike epidemiological studies,\nthe terminology 'cohort' and 'case-control' is less informative for\ndiagnostic test trials, so we substituted them for 'single-gate' and\n'two-gate' designs. We included this question because a two-gate\ndesign has more potential to introduce selection bias.\nDomain 2\nFor the biomarker studies\n2.1. We introduced an additional signalling question, 'Was the\nphase of the menstrual cycle considered in interpreting the index\ntest?' to assess bias in the interpretation of the test results. Some\nbiochemical markers are sensitive to fluctuation in steroid sex\nhormone levels across a menstrual cycle, which could result in\nthe differential expression of endometriosis biomarkers at different\ncycle phases.\nFor the studies on clinical/imaging tests\n2.2 We introduced an additional signalling question 'Was the index\ntest performed by a single operator?' to assess interobserver\nvariation bias.\n2.3 We introduced an additional signalling question 'Were the same\nclinical data available when the index test results were interpreted\nas that which would be available when the test is used in practice?'\nto assess a bias in clinical applicability.\n2.4 We rephrased an original signalling question, 'If a threshold was\nused, was it pre-specified?' as 'Did the study provide a clear pre-\nspecified definition of what was considered to be a positive index\ntest result' because this question was more applicable to imaging\nmodalities.\nWe undertook the assessment of methodological quality for each\ndomain, but we did not calculate a summary score to estimate the\noverall quality of studies (Whiting 2005).\nStatistical analysis and data synthesis\nThe estimates of sensitivity and specificity were generated in forest\nplots and plotted in the receiver operating characteristic (ROC)\nspace for each index test using Review Manager 5 so/f_tware (RevMan\n2014). The diagnostic performance of each test was investigated\nand inter-study variation in the performance of each index test\nwas visually explored in relation to woman characteristics, study\ndesign, and study quality factors. Two or more tests evaluated in\nthe same cohort were included as separate data sets, since the unit\nof analysis was the test result, not the woman.\nFor studies that reported subgroup analyses per phase of the\nmenstrual cycle, we presented the data in a clinically relevant\nway. For instance, we presented pooled estimates when there\nwas no statistically significant difference in biomarker expression\nbetween cycle phases. Alternatively, where putative biomarkers\ndemonstrated cycle-dependent expression or were noted to be\nmodulated by ovarian hormones, we reported the test performance\neither at several time points across the menstrual cycle or in the\nphase that demonstrated the most distinct difference between\ngroups.\nWe planned to perform the bivariate logit normal random-effects\nmodel for all meta-analyses with four studies or more and a\nfixed-effect meta-analysis of sensitivity and specificity for smaller\ngroups of studies (two or three) in the absence of substantial\nheterogeneity. The meta-analyses were planned to be performed\nusing SAS NLMIXED so/f_tware (Cary, NC: SAS Institute Inc) in order\nto provide plots of the estimated summary points of sensitivity and\nspecificity and confidence regions. In this review a meta-analysis\nwas not performed due to the paucity of data for each combination\nof non-invasive tests.\nThe comparative accuracy of index tests was assessed in two ways.\nIn direct, fully-paired comparisons where all the study participants\nreceived more than one index test as well as the reference standard,\nthe estimates were plotted in RevMan. If a meta-analysis was\npossible, test-level covariates in the bivariate logit normal model\nwere used to identify statistically significant differences. Otherwise,\nthe available comparative data were reported in a narrative way\nand illustrated using forest and ROC plots.\nWhen test performance was judged against the predetermined\ndiagnostic criteria, the point estimates of sensitivity and specificity\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n20\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nwere considered as the most informative presentation of test\nperformance. We acknowledge that tests with point estimates that\ndid not reach the predetermined criteria but confidence intervals\n(CIs) which contained values above the threshold, could have\ndiagnostic value. Furthermore, tests with point estimates that\nreached the criteria but CIs which contained values below the\nthreshold, could have an overestimated diagnostic value. If the\nrange of the CIs rather than the point estimates of the data are used,\nthe predetermined cut-off becomes meaningless. Therefore, we did\nnot consider CIs in qualifying the test performance, but utilised this\ninformation in interpreting the reliability of the obtained data.\nDealing with missing data\nMissing data were defined as any information on the study\npopulation, index tests or reference standard that was not available\nin the publication which was required to determine the eligibility of\nthe study for inclusion, the methodological quality or to construct\nthe results table. If missing data were identified, we contacted the\nauthors in an attempt to obtain this information. If missing data\nprevented a clear judgment regarding applicability for inclusion\nor the construction of accurate 2 x 2 tables and the data were\nnot available from the primary investigators (for example, we were\nunable to locate the contact details of the authors or there was no\nreply from the authors or the authors replied that the requested\ninformation was unavailable), we excluded the study from the\nreview.\nInvestigations of heterogeneity\nWe planned to assess heterogeneity by visually examining the\nforest plots of sensitivities and specificities and the ROC plots for\neach index test. The potential sources of heterogeneity are stated\nin the Secondary objectives. For diagnostic tests with more than\n10 eligible studies, we planned to formally explore heterogeneity\nby using study-level covariates. We were unable to assess sources\nof heterogeneity in this review because there was only one study\nfor each test. We also planned to assess the sensitivity of results\nto the inclusion and exclusion of outlying studies in all analyses,\nbut refrained from doing so, again because of the small number of\nstudies for most analyses.\nSensitivity analyses\nWe planned to conduct sensitivity analyses to assess the impact\nof the methodological quality of included studies on the results\nof any meta-analyses if sufficient data were available. Low-quality\nstudies were defined by the identification of a high risk of bias\nfor one or more QUADAS-2 domains. We also planned to use the\n'leave-one-out’ procedure to assess the impact of each study on\nthe meta-analysis results (leading study effect). In this review we\nwere unable to undertake sensitivity analyses due to the paucity of\nstudies evaluating each biomarker.\nAssessment of reporting bias\nA comprehensive search of multiple sources for eligible studies, a\nsearch of trial registers and no language restrictions minimised the\nrisk of reporting bias. However, publication bias generally arises\nwhen studies have a higher chance of being published if their\nresults are positive. Therefore unpublished and published study\ndatabases and conference proceedings were initially searched and\nevaluated. During the process of qualifying the studies for inclusion\nin this review, we faced substantial difficulty in obtaining full-\ntext publications or further details of studies published in an\nabstract form. This precluded a reliable assessment of eligibility\nand methodological quality and it was decided not to include these\npublication sources in this review.\nR E S U L T S\nResults of the search\nThe literature search identified 33,438 references for the biomarker-\nbased tests in the following databases: CENTRAL (n = 226), MEDLINE\n(n = 10,328), EMBASE (n = 10,313), CINAHL (n = 1131), PsycINFO (n\n= 174), Web of Science (n = 7425), LILACS (n = 420), OAIster (n =\n446), Trip (n = 1648), trial registers for ongoing and registered trials\n(n = 523), MEDION (n = 2), DARE (n = 99), PubMed, a ‘systematic\nreview’ search (n = 418) and simple search PubMed (n = 267). For the\nimaging tests, the search identified 32,275 references as following:\nCENTRAL (n = 445), MEDLINE (n = 7391), EMBASE (n = 12,161),\nCINAHL (n = 668), PsycINFO (n = 174), Web of Science (n = 7425),\nLILACS (n = 420), OAIster (n = 446), TRIP (n = 1648), trial registers\nfor ongoing and registered trials (n = 523), MEDION (n = 190), DARE\n(n = 99), PubMed, a ‘systematic review’ search (n = 418) and simple\nsearch PubMed (n = 267). These databases were searched from\ninception to 20 April - 31 July 2015.\nThe flow of the selection process is presented in Figure 2. Titles\nwere screened to exclude duplicates (n = 20,017) and clearly\nirrelevant studies (n = 40,723). A further 4941 references were\neliminated a/f_ter the abstracts were reviewed because either they\ndid not address the research question or they clearly did not meet\nthe inclusion criteria. The full texts of the remaining 32 references\nwere retrieved and assessed for eligibility. Data from five studies\nrequired additional clarification from the authors and two non-\nEnglish publications were translated. Ultimately, 11 studies were\neligible and provided data for the review and 21 studies were\nexcluded.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n21\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 2. /uni00A0 Flow of the studies identified in literature search for systematic review on combination of non-invasive\ntests for diagnosis of endometriosis.\n/uni00A0\nBasic features of the included studies\nThe list and details of the included studies are presented in\nCharacteristics of included studies. The 11 eligible studies included\n1339 participants, with a median of 101 women per study (range\n55 to 368). Of these studies, four were conducted in Europe, four in\nAsia, one in Australia, one in North America and one in the Middle\nEast. Ten studies were performed at University Hospitals, two of\nwhich were tertiary endometriosis centres and one study was\nperformed at a biotechnology firm. Three studies were published\nin 1996, three studies were published between 2003 and 2009 and\nthe remaining five studies were published between 2012 and 2014.\nAll the included studies evaluated women of reproductive age.\nThere were no randomised controlled trials and all the studies were\nobservational, mainly of cross-sectional design. Ten studies were\n'single-gate', where both cases and controls were sampled from\nthe same population and one study was of a 'two-gate design',\nincluding a wider group of participants who were undergoing\nsurgery for various indications. Laparoscopy was used for diagnosis\nin all studies, laparotomy was co-utilised in four studies and seven\nstudies used histopathology to confirm the surgical diagnosis.\nSeven studies evaluated any pelvic endometriosis, of which one\nstudy included only participants with minimal-mild endometriosis\n(rASRM stage I-II), one study included only participants with\nmoderate-severe endometriosis (rASRM stage III-IV) and one study\nconcentrated on endometriosis with peri-ovarian adhesions. Two\nother studies addressed only ovarian endometriosis, one study\nfocused on a combination of ovarian endometriosis and deep\ninfiltrating endometriosis (DIE), and one study addressed mapping\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n22\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nDIE at specific anatomical sites. The reported prevalence of\nendometriosis varied from 29% to 69%. Four studies received\nfinancial support, of which one study reported commercial funding\nand most authors of that publication worked in the biotechnology\nindustry. Four other groups of authors declared no conflict of\ninterest and no information was available from the remaining\nstudies.\nBasic features of the excluded studies\nThe list and descriptions of the excluded studies are presented\nin Characteristics of excluded studies. Based on a full-text\nassessment, 21 publications were excluded, of which 15 studies\nevaluated several testing modalities but did not present diagnostic\nestimates for the combined test. A further two studies reported\nstatistically significant differences in biomarker levels between the\nstudy and control groups, but contained insufficient diagnostic\naccuracy information for the construction of 2 x 2 contingency\ntables. One study was of retrospective design where the\nparticipants were recruited a/f_ter the surgical procedure and\nenrolled postmenopausal women. In one excluded paper the target\ncondition was outside the inclusion criteria and normal versus\nabnormal pelvis comparison was made without any independent\ndata for endometriosis. One study incorporated imaging evaluation\ninto the combined test but reported only 'lesion-level' analysis and\none study was excluded because it was a review article.\nMethodological quality of included studies\nThe quality of the included studies is illustrated in the QUADAS-2\nresults summary (Figure 3 and Figure 4). Overall, the studies were of\npoor methodological quality and all studies had an unclear or high\nrisk of bias in at least one domain.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n23\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 3. /uni00A0 Risk of bias and applicability concerns summary: review authors' judgements about each domain for each\nincluded study\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n24\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 4. /uni00A0 Risk of bias and applicability concerns graph: review authors' judgements about each domain presented\nas percentages across included studies\n/uni00A0\nFive studies presented a low risk of patient selection bias (Guerriero\n1996a; Guerriero 1996b; Hudelist 2009; Koninckx 1996; Marasinghe\n2014), five studies demonstrated an unclear risk and one study\nwas assessed at high risk for this domain. Non-consecutive or\nnon-random selection of participants, utilisation of a two-gate\ndesign for participant selection, the absence of a clear definition\nof inclusion/exclusion criteria and using a highly selected group of\nwomen were the main reasons for a high risk assessment of bias.\nOne study demonstrated a low risk of index test interpretation bias\n(Marasinghe 2014), one study demonstrated an unclear risk and\nnine studies carried a high risk. A lack of clear pre-specified criteria\nfor a positive diagnosis and index test operators not being blind\nto the results of reference standard were the main reasons for a\nhigh risk assessment. The skill level of a test operator and the\ninterobserver variability, both of which directly affect performance\nof the tests, were rarely reported.\nEight studies were at low risk of bias in the 'reference standard'\ndomain (Cho 2012; el Sharkwy 2013; Gagné 2003; Guerriero 1996a;\nGuerriero 1996b; Marasinghe 2014; Paiva 2014; Yun 2014), three\nstudies were classified as unclear risk and no studies demonstrated\na high risk. An unclear risk of bias was assigned if there was not\nenough information to determine how likely the reference standard\nwas to have correctly classified the target condition. Specifially,\nsurgical procedures were not well-described, the criteria for a\npositive reference standard were not stated, it was unclear if\nhistology was utilised to confirm surgical diagnosis, or there was\nno information regarding the experience of the surgeons or the\npathologists involved.\nEight studies presented a low risk of bias in the 'flow and timing'\ndomain (Cho 2012; Gagné 2003; Guerriero 1996a; Guerriero 1996b;\nHudelist 2009; Paiva 2014; Yun 2014; Zeng 2005), no studies\ndemonstrated an unclear risk and three studies carried a high\nrisk. In every study all participants received the same reference\nstandard. The time interval between the index test and the\nreference standard was placed as 12 months or less and the\nmost commonly reported time interval was immediately before\nsurgery. A high risk of bias was assigned if there were unexplained\nwithdrawals that exceeded 5% of the enrolled population or if the\nreason for withdrawal could introduce selection bias regarding the\nsamples analysed.\nFive studies presented a low concern for patient selection\napplicability, no studies demonstrated an unclear concern and\nsix were of high concern. A high concern in patient selection\napplicability was assigned if the study utilised two-gate selection\nfor cases and controls or if only a limited spectrum of disease\nwas evaluated. In our view, any sampling deviation from a\nrepresentative group of the entire clinically relevant population\ncould skew the estimates of diagnostic accuracy in any direction.\nIn 10 studies there was a low concern of index test applicability,\nwhereas in one study the concern was unclear and none of the\nstudies presented a high concern. An unclear concern was assigned\nwhen the study did not present sufficient information regarding the\nconduct of the tests, such as the laboratory methods or reagents\nused or the level of expertise of the test operators.\nAll 11 studies were of low concern for applicability in regards to\nthe reference standard and none of the studies had high or unclear\nconcern. All the included studies implemented pelvic surgery\n(laparoscopy or laparotomy) as a reference standard, which could\nbe relied upon to match the review question.\nFindings\nA total of 15 diagnostic combinations of several testing modalities\nwere evaluated in the 11 included studies (Summary of findings\n1). Of these, seven were assessed for their value in detecting\npelvic endometriosis, two tests were appraised in a context of DIE\nor endometrioma and 10 tests were evaluated for their accuracy\nto differentiate endometrioma from other benign ovarian cysts.\nOne additional test looked at specific anatomical sites of DIE and\ntherefore may be considered for a preoperative mapping rather\nfor a primary diagnosis of the disease. The ways the tests were\ncombined in a diagnostic panel varied between the studies and\nincluded the following: 1. all the tests of the panel are positive\nconsidering a specific cut-off for each constituent; 2. either of\nthe tests included in a diagnostic panel is positive; 3. sum or\nmultiplication of values of all the tests comprising diagnostic panel\nutilising distinct cut-off value; 4. multivariate logistic regression\nmodel.\n1. Tests for the diagnosis of any pelvic endometriosis\n1) IL-6 (> 15.4 pg/ml) [serum] + PGP 9.5 [endometrium]\nThe diagnostic performance of serum IL-6 in combination with\nendometrial PGP 9.5 was evaluated in one study with a total\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n25\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nof 78 women (el Sharkwy 2013 ). The test was performed in the\nfollicular phase of the menstrual cycle and was evaluated for\nonly minimal-mild endometriosis, rASRM I-II. The definition of\npositive test was a cut-off value > 15.4 pg/ml for IL-6 and positive\nimmunohistochemistry (IHC) staining for PGP 9.5 in the functional\nlayer of endometrium. When both components of the test were\npositive, the sensitivity was 1.00 (95% CI 0.91 to 1.00), and the\nspecificity 0.93 (95% CI 0.80 to 0.98) (Figure 5; Figure 6). The point\nestimates met the criteria for a replacement and SnOUT triage test\nand approached the criteria for a SpIN triage test. The diagnostic\nestimates of the combined test were higher than for each individual\ntests assessed in this study: for serum IL-6 with a cut-off above\n15.4 pg/ml the sensitivity and specificity were 0.89 (95% CI 0.75\nto 0.97) and 0.82 (95% CI 0.67, 0.93), respectively; for PGP 9.5 the\nsensitivity and specificity were 0.92 (95% CI 0.79 to 0.98) and 0.80\n(95% CI 0.64 to 0.91), respectively. The CIs were broad for each of\nthe included tests, which was particularly prominent for individual\ntests. Further testing in larger studies including participants with a\nwider spectrum of endometriosis is needed to confirm the role of\nthe above test in detecting endometriosis.\n/uni00A0\nFigure 5. /uni00A0 Forest plot of the combined tests for detection of pelvic endometriosis. Plot shows the estimates of\nsensitivity and specificity (squares) with 95% CI (black line) specific for each evaluation (each evaluation was\nderived from a single study), country in which the study was conducted and severity of the disease assessed by each\nstudy, reported as rASRM stage. FN: false negative; FP: false positive; TN: true negative; TP: true positive.\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n26\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 6. /uni00A0 Study specific estimates of the diagnostic accuracy of the combined tests for detection of pelvic\nendometriosis plotted in ROC space. Each point represents the pair of sensitivity and specificity from each\nevaluation (each evaluation was derived from a single study). The size of each point is proportional to the sample\nsize the shape designates different tests. The bars correspond to 95% CIs of each individual evaluation.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n27\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n2) CA-125 [serum] (> 35 U/ml) + aromatase P450 [endometrium].\nOne study comprising 58 women evaluated the role of serum\nCA-125 combined with endometrial aromatase P450 in diagnosing\nendometriosis (Zeng 2005). The test was performed in the follicular\nor luteal phases of the menstrual cycle, but the effect of the\ncycle phase on the test performance was not assessed. The study\naddressed pelvic endometriosis, rASRM I-IV. The test was positive\nwhen the CA-125 level was above 35 U/ml and endometrial IHC was\npositive for aromatase. Considering both positive components, the\ntest had a sensitivity of 0.92 (95% CI 0.78 to 0.98) and a specificity\nof 0.68 (95% CI 0.45 to 0.86) (Figure 5; Figure 6), approaching\nthe criteria for a SnOUT triage test. Direct comparison between\nthe combination and each individual test assessed in this study\nrevealed that the combined test had higher sensitivity but lower\nspecificity than each individual test: CA-125 sensitivity 0.44 (95%\nCI 0.28 to 0.62), specificity 0.82 (95% CI 0.60 to 0.95); aromatase\nP450 sensitivity 0.83 (95% CI 0.67 to 0.94), specificity 0.86 (95% CI\n0.65 to 0.97) with wide CIs for each evaluation. This result requires\nfurther validation in large well-defined populations, accounting for\na menstrual cycle phase of testing.\n3) VDBP-Cr [urine] x CA-125 [serum] (> 2755)\nThe diagnostic performance of the combination of urinary VDBP\n(vitamin-D-binding protein) and serum CA-125 was evaluated in\none study, which included 95 women (Cho 2012 ). The test was\nperformed in the follicular or luteal cycle phase. Even though the\nstudy included endometriosis of varying severity (rASRM I-IV), more\nthan 90% of women with endometriosis had moderate - severe\ndisease (52/57). Urinary VDBP levels were significantly higher in\nendometriosis only in luteal cycle phase, however the performance\nof the combined test was not stratified by a cycle phase. The test\nwas considered positive when a multiplication of urinary VDBP\nlevel corrected for creatinine (VDBP-Cr) by serum CA-125 level was\nabove 2755. The test demonstrated a sensitivity of 0.74 (95% CI\n0.60 to 0.84) and a specificity of 0.97 (95% CI 0.86 to 1.00) (Figure\n5; Figure 6) and met the criteria for a SpIN triage test. In direct\ncomparison this combination had higher diagnostic estimates than\nVDBP only (sensitivity 0.58 (95% CI 0.44 to 0.71), specificity 0.55\n(95% CI 0.38 to 0.71). Both individual urinary VDBP and combined\nVDBP + CA-125 test manifested wide CIs; separate data for CA-125\nonly were not available from this study. Further evaluation of\nVDBP - CA-125 combination across the spectrum of endometriosis\nparticularly in the luteal phase may help to clarify the diagnostic\nrole of this tests in endometriosis.\n4) NNE-Cr [urine] + CA-125 [serum] (> 27.23)\nThe role of combining urinary NNE (enolase I) and serum CA-125\nwas assessed in one study with a total of 59 participants (Yun\n2014). The test was performed in the follicular or luteal cycle phase\nand assessed only moderate - severe endometriosis, rASRM III-IV.\nUrinary NNE expression was not influenced by cycle phase and was\nsignificantly greater (P = 0.026) in women with endometriosis only\na/f_ter correction for creatine excretion. A positive test was defined as\nan arithmetical sum of the urinary NNE-Cr and serum CA-125 level\nabove 27.23. This test exhibited a sensitivity of 0.77 (95% CI 0.61 to\n0.89) and a specificity 0.85 (95% CI 0.62 to 0.97) (Figure 5; Figure\n6). Although the diagnostic estimates of the combination were\nsuperior to those of NNE only (sensitivity 0.56 (95% CI 0.40 to 0.72),\nspecificity 0.70 (95% CI 0.46 to 0.88); the diagnostic estimated for\nCA-125 only were not available), the criteria for either replacement\nor triage test were not met. Considering a single study, there is\nnot enough information on diagnostic utility of NNE + CA-125\ncombination in detecting endometriosis.\n5) History (dysmenorrhoea and dyspareunia) + PV examination\n+ TVUS\nOne study comprising 106 participants evaluated the combination\nof history, gynaecological examination and transvaginal ultrasound\n(TVUS) for detecting pelvic endometriosis, rASRM I-IV (Marasinghe\n2014). The authors did not specify the menstrual cycle phase\nof the testing. The test was considered as positive when: 1. the\nclinical history was positive for dysmenorrhoea and dyspareunia.\nPain severity was assessed using a visual analogue scale ranging\none to 10 with a score of one considered as 'no pain'; 2.\nbimanual pelvic vaginal examination (PV) was used to detect\nthe presence of pelvic tenderness, a fixed retroverted uterus,\ntender uterosacral ligaments and deeply infiltrating nodules on the\nuterosacral ligaments or in the cul-de-sac; 3. TVUS demonstrated\n'fixed ovaries', defined when the ovaries did not move freely over\nthe ipsilateral internal iliac vessels or pelvic sidewall or uterus with\nthe gentle pressure; any of these findings resulted in a 'positive' test\nresult. Sonographic criterion suggested that the authors focused\non endometriosis with per-ovarian adhesions. The combination\nof positive history, examination findings and TVUS findings had\na sensitivity of 0.92 (95% CI 0.78 to 0.98) and a specificity of\n0.61 (95% CI 0.48 to 0.72) (Figure 5; Figure 6). The diagnostic\nestimates approached the criteria for a SnOUT triage test, although\ncontained wide CIs. The reported sensitivity and specificity for\neach component of the test in this study were 0.46 and 0.77 for\ndyspareunia, 0.76 and 0.70 for dysmenorrhoea, 0.73 and 0.88 for\npositive vaginal examination, and 0.78 and 0.94 for fixed ovaries on\nTVUS.\n6) History (length of menses) + CA-125 [serum] (> 35 U/ml) +\nleukocytes [endometrium]\nOne study with a total of 368 participants assessed the performance\nof history, serum CA-125 and endometrial leukocyte subsets in\ndiagnosing endometriosis (Gagné 2003 ). The test was performed\nin the luteal cycle phase and was utilised for detecting a wide\nspectrum of pelvic endometriosis, rASRM I-IV. The diagnostic test\nfor endometriosis included the following: 1. clinical history (length\nof menses) 2. serum CA-125 with a cut-off value above 12.8 U/ml; 3.\nendometrial leukocytes (CD3+, CD16+, CD3-HLADR-, CD3-CD45RA-,\nCD3+CD16-, CD3+CD56-, CD56-CD16+, CD16b+) with specific cut-off\nfor each leukocyte subset. The test parameters were selected by\nunivariate analysis and then included in the predictive model by\nutilising a multiple logistic regression with subsequent bootstrap\nmethod validation. The model adjusted for gravidity and histologic\ndating (early, mid or late luteal phase) demonstrated a sensitivity\nof 0.61 (95% CI 0.54 to 0.69) and a specificity of 0.95 (95% CI 0.91 to\n0.98) (Figure 5; Figure 6). In this study, the combined test performed\nbetter than CA-125 only (sensitivity 0.20 (95% CI 0.15 to 0.27),\nspecificity 0.92 (95% CI 0.87 to 0.95)) and met the criteria for a SpIN\ntriage test. The diagnostic estimates for endometrial leukocytes\nonly were not available.\n7) History (parity, past use of IUD, past endometriosis, alcohol\nintake, dyspareunia) + CA-125 [serum]\nThe combination of the clinical and demographical parameters\nwith serum CA-125 for discriminating women with and without\nendometriosis was evaluated in one study comprising 101\nparticipants (Paiva 2014). The test was performed at different\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n28\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nphases of menstrual cycle (menstrual, follicular or luteal) and\nassessed the full spectrum of endometriosis, rASRM I-IV. The\nlevel of CA-125 did not vary across the menstrual cycle. Fourteen\nparameters identified by univariate analysis were used with logistic\nregression to produce the diagnostic model, which included 1.\nclinical data: parity, ever had an intrauterine device (IUD), history\nof endometriosis, alcohol intake, dyspareunia); 2. serum CA-125,\ncut-off value not specified. The diagnostic model demonstrated\na sensitivity of 0.93 (95% CI 0.84 to 0.98) and a specificity\nof 0.63 (95% CI 0.44 to 0.79) (Figure 5; Figure 6). The test\napproached the criteria for a SnOUT triage test, although exhibited\nwide CIs for both sensitivity and specificity. The information on\ndiagnostic performance of individual components of the test was\nnot available.\n2. Tests for diagnosis of DIE or ovarian endometriosis\n1) PV examination (menstrual nodularities) + CA-125 [serum] (>\n35 IU/L)\n2) PV examination (menstrual nodularities) OR CA-125 [serum]\n(> 35 IU/L)\nOne study comprising 55 participants, evaluated the role of\ngynaecological examination in adjunct with serum CA-125 for\ndetecting one of the following target conditions: 1. DIE or ovarian\nendometrioma or severe pelvic adhesions; 2. only DIE; 3. only\novarian endometrioma (Koninckx 1996). The test included 1.\nbimanual PV examination during menstruation, which was scored\nas positive when an induration or painful nodularities was felt; 2.\nserum CA-125 measured in mid-follicular cycle phase with a cut-off\nvalue above 35 IU/L. Two variations of the test were assessed for\neach target condition: 1. both components of the test were positive\nor 2. either of the two tests was positive.\nFor detecting DIE, endometrioma or severe adhesions the test\nexhibited a sensitivity of 0.42 (95% CI 0.22 to 0.63) with a specificity\nof 1.00 (95% CI 0.80 to 1.00) when both examination and serum\nCA-125 were positive. The test achieved a sensitivity of 0.88 (95%\nCI 0.68 to 0.97) with a specificity of 0.82 (95% CI 0.57 to 0.96)\nwhen either component of the combined test was positive (Figure\n7; Figure 8). For DIE only, the test demonstrated a sensitivity of\n0.38 (95% CI 0.14 to 0.68) with a specificity of 0.88 (95% CI 0.64\nto 0.99) when both examination and serum CA-125 were positive,\nand a sensitivity of 0.85 (95% CI 0.55 to 0.98) with a specificity\nof 0.71 (95% CI 0.44 to 0.90) when either component of the test\nwas positive (Figure 7; Figure 8). For endometrioma, the test had\na sensitivity of 0.56 (95% CI 0.21 to 0.86) with a specificity of 0.88\n(95% CI 0.64 to 0.99) when both examination and serum CA-125\nwere positive, and a sensitivity of 0.89 (95% CI 0.51 to 1.00) with\na specificity of 0.65 (95% CI 0.38 to 0.86) when either component\nof the composite test was considered (Figure 7; Figure 8). None of\nthese tests met the criteria for either a replacement or of the triage\ntests and all evaluations were featured by wide CIs. The reported\ndiagnostic parameters for CA-125 on its own were sensitivity 0.5\nwith specificity 0.88 for DIE, endometrioma or severe adhesions;\nsensitivity 0.47 with specificity 0.81 for DIE and sensitivity 0.67 with\nspecificity 0.81 for endometrioma.\n/uni00A0\nFigure 7. /uni00A0 Forest plot of the combined tests for detection of DIE or ovarian endometriosis. Plot shows the estimates\nof sensitivity and specificity (squares) with 95% CI (black line) specific for each evaluation (each evaluation was\nderived from a single study), country in which the study was conducted and type of target condition assessed by\neach study. FN: false negative; FP: false positive; TN: true negative; TP: true positive.\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n29\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 8. /uni00A0 Study specific estimates of the diagnostic accuracy of the combined tests for detection of DIE or ovarian\nendometriosis plotted in ROC space. Each point represents the pair of sensitivity and specificity from each\nevaluation (each evaluation was derived from a single study). The size of each point is proportional to the sample\nsize the shape designates different tests. The bars correspond to 95% CIs of each individual evaluation.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n30\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n3. Tests for differentiating ovarian endometriosis versus other\nbenign ovarian cysts in women of reproductive age\n1) TVUS + CA-125 [serum] (≥ 25 U/ml) + CA-19.9 [serum] (≥12 U/\nml)\n2) TVUS + (CA-125 [serum] (≥ 25 U/ml) OR CA-19.9 [serum] (≥12 U/\nml))\n3) TVUS + CA-19.9 [serum] (≥ 12 U/ml)\n4) TVUS OR CA-19.9 [serum] (≥ 12 U/ml)\nOne study comprising 118 participants evaluated a composite test\nof TVUS and serum tumour markers CA-125 and/ or CA-19.9 for\ndiscriminating ovarian endometrioma from other benign cysts in\nwomen of reproductive age (Guerriero 1996a). All the participants\nwere tested in the follicular phase of the menstrual cycle. Positive\nTVUS test (presence of endometrioma) was described as a presence\nof a round shaped homogeneous hypoechoic 'tissue' within the\novary with clear demarcation from the parenchyma and without\npapillary proliferations. A cut-off for a positive serum biomarker\nwas above 25 U/ml for CA-125 and above 12 U/ml for CA-19.9.\nThe test demonstrated a sensitivity of 0.49 (95% CI 0.32 to 0.65) and\na specificity of 0.99 (95% CI 0.93 to 1.00) when all three components\nof the test were positive (Figure 9; Figure 10), approaching the\ncriteria for a SpIN triage test. The test had a higher sensitivity (0.79\n(95% CI 0.64 to 0.91)) and slightly lower specificity (0.97 (95% CI 0.91\nto 1.00)) when either positive blood test was considered in adjunct\nwith TVUS (Figure 9; Figure 10), meeting the criteria for a SpIN triage\ntest.\n/uni00A0\nFigure 9. /uni00A0 Forest plot of the combined tests (TVUS and/or CA-125 and/or CA-19.9) for differentiation of ovarian\nendometriosis vs. other benign ovarian cysts. Plot shows the estimates of sensitivity and specificity (squares) with\n95% CI (black line) specific for each evaluation (each evaluation was derived from a single study Guerriero 1996a),\ncountry in which the study was conducted and target condition assessed by each study. FN: false negative; FP: false\npositive; TN: true negative; TP: true positive.\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n31\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nFigure 10. /uni00A0 Study specific estimates of the diagnostic accuracy of the combined tests (TVUS and/or CA-125 and/or\nCA-19.9) for differentiation of ovarian endometriosis vs. other benign ovarian cysts plotted in ROC space. Each point\nrepresents the pair of sensitivity and specificity from each evaluation (each evaluation was derived from a single\nstudy Guerriero 1996a). The size of each point is proportional to the sample size the shape designates different\ntests. The bars correspond to 95% CIs of each individual evaluation.\n/uni00A0\nWhen only TVUS and CA-19.9 were considered, a sensitivity was\n0.54 (95% CI 0.37 to 0.70) and a specificity was 0.97 (95% CI 0.91\nto 1.00) for both positive components (Figure 9; Figure 10), which\ncould qualify as a SpIN triage test. When either TVUS or CA-19.9 was\npositive, the test demonstrated a sensitivity of 0.92 (95% CI 0.79 to\n0.98) and a specificity of 0.70 (95% CI 0.58 to 0.78) (Figure 9; Figure\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n32\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n10), approaching the criteria for a SnOUT triage test. Considering\nthe data reported by a single study and wide overlapping CIs, we\nsuggest caution in interpretation of the presented findings.\nIn a head-to-head direct comparison, the test based on a\ncombination of ether positive blood biomarker with TVUS and\na combination of CA-19.9 with TVUS performed better than the\ntest including both positive CA-19.9 and CA-125. The study-specific\ndiagnostic estimates for TVUS only were sensitivity 0.85 (95%\nCI 0.69 to 0.94) and specificity 0.97 (95% CI 0.91 to 1.00). This\nsuggested that addition of biomarkers to ultrasound examination\ndid not improve diagnostic performance of the test in this study,\nresulting in lower sensitivity and only marginally higher or similar\nspecificity for most combinations. This was particularly noticeable\nfor the combinations with CA-125. Further, no blood biomarker\ncombinations from this study without TVUS met the criteria of\neither replacement or triage test (Nisenblat 2016a).\n5) TVUS + CA-125 [serum] (≥ 20 U/ml)\n6) TVUS OR CA-125 [serum] (≥ 20 U/ml)\n7) TVUS + CA-125 [serum] (≥ 25 U/ml)\n8) TVUS OR CA-125 [serum] (≥ 25 U/ml)\n9) TVUS + CA-125 [serum] (≥ 35 U/ml)\n10) TVUS OR CA-125 [serum] (≥ 35 U/ml)\nOne study with a total of 101 women evaluated the combination of\nTVUS and serum CA-125 in diagnosing ovarian endometrioma when\ncompared with other benign ovarian cysts (Guerriero 1996b). The\nstudy was performed by the same group that evaluated the above\nmentioned combined testing for endometrioma (TVUS or CA-125\nor CA-19.9), utilising similar sonographic criteria and similar testing\ntime (follicular cycle phase). Two variations of the test included 1.\nboth TVUS and CA-125 positive and 2. either test is positive.Three\ndifferent cut-off thresholds for CA-125 were used for each pair of the\ncombined test (≥ 20 U/ml; ≥ 25 U/ml; ≥ 35 U/ml).\nFor TVUS and CA-125 with a cut-off value ≥ 20 U/ml, the diagnostic\nestimates met the criteria for a SpIN triage test, when both tests\nwere positive (sensitivity 0.69 (95% CI 0.49 to 0.85), specificity of\n0.96 (95% CI 0.88 to 0.99)) and approached the criteria for a SnOUT\ntriage test when either positive test was considered (sensitivity 0.93\n(95% CI 0.77 to 0.99), specificity of 0.53 (95% CI 0.41 to 0.65)) (Figure\n11; Figure 12).\n/uni00A0\nFigure 11. /uni00A0 Forest plot of the combined tests (TVUS and/or CA-125 at varying thresholds) for differentiation of\novarian endometriosis vs. other benign ovarian cysts. Plot shows the estimates of sensitivity and specificity\n(squares) with 95% CI (black line) specific for each evaluation (each evaluation was derived from a single study\nGuerriero 1996b), country in which the study was conducted and target condition assessed by each study. FN: false\nnegative; FP: false positive; TN: true negative; TP: true positive.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n33\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0\nFigure 12. /uni00A0 Study specific estimates of the diagnostic accuracy of the combined tests (TVUS and/or CA-125 at varying\nthresholds) for differentiation of ovarian endometriosis vs. other benign ovarian cysts plotted in ROC space. Each\npoint represents the pair of sensitivity and specificity from each evaluation (each evaluation was derived from\na single study Guerriero 1996b). The size of each point is proportional to the sample size the shape designates\ndifferent tests. The bars correspond to 95% CIs of each individual evaluation.\n/uni00A0\nFor TVUS and CA-125 with a cut-off value ≥ 25 U/ml, the diagnostic\nestimates met the criteria for a SpIN triage test when both tests\nwere positive (sensitivity 0.69 (95% CI 0.49 to 0.85), specificity of\n0.96 (95% CI 0.88 to 0.99)) and approached the criteria for a SnOUT\ntriage test when either positive test was considered (sensitivity 0.90\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n34\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n(95% CI 0.73 to 0.98), specificity of 0.63 (95% CI 0.50 to 0.74)) (Figure\n11; Figure 12).\nFor CA-125 with a cut-off value ≥ 35 U/ml, the diagnostic estimates\nmet the criteria for a SpIN triage test when both tests were positive\n(sensitivity 0.52 (95% CI 0.33 to 0.71), specificity of 0.97 (95% CI 0.90\nto1.00)) and approached the criteria for a replacement and SnOUT\ntriage test when either positive test was considered (sensitivity 0.90\n(95% CI 0.73 to 0.98), specificity of 0.75 (95% CI 0.63 to 0.84)) (Figure\n11; Figure 12).\nIn direct comparison, TVUS and CA-125 at a cut-off value ≥ 25 U/\nml and ≥ 35 U/ml performed better than at a cut-off value ≥20\nU/ml only when either positive test was considered, but did not\nimprove diagnostic estimates of the combination when both tests\nwere positive. In this study TVUS alone (sensitivity 0.83 (95% CI 0.64\nto 0.94), specificity 0.93 (95% CI 0.85 to 0.98)), was more sensitive\nthan the combination TVUS + CA-125 and more specific than\nthe combination TVUS OR CA-125. None of the blood biomarkers\nwithout TVUS could qualify as useful diagnostic test (Nisenblat\n2016a).\n4. Tests for mapping of DIE at specific anatomical locations\n1) PV examination + TVUS\nOne study comprising 200 participants evaluated the combination\nof gynaecological examination and TVUS for detecting DIE\nat specific anatomical locations: 1. pouch of Douglas (POD)\nobliteration; 2. vaginal wall; 3. rectovaginal septum (RVS); 4.\nrectum. The cycle phase of the testing was not reported (Hudelist\n2009). The gynaecological bimanual examination was considered\npositive when nodularity or stiffened or thickened area or\na palpable cystic expansion were detected at the evaluated\nanatomical sites. TVUS criteria were defined for each anatomical\nlocation, specifically 1. uterus, adnexa and rectosigmoid colon\nfixed to each other with disappearance of the peritoneal structure\n(complete POD obliteration); peritoneal limits partially identified\nwith the presence or absence of suspended or lateralised fluid\ncollection (incomplete POD obliteration); 2. thickening or the\npresence of a hypoechogenic cystic or non-cystic nodularity\nwithin the posterior vaginal wall (vaginal DIE); 3. presence of a\nhypoechogenic nodularity or cystic mass within RVS - area between\nrectum and posterior vaginal wall from the level of introitus up\nto a level defined by the lower border of posterior lip of cervix\n(RVS DIE); 4. presence of a regular or irregular hypoechogenic mass\ndistorting and replacing the normal appearance of the muscular\nlayer of the rectal wall (rectal DIE). Considering that only specific\nsites of endometriosis were assessed, these tests can not be\nconsidered diagnostic but can be utilised for preoperative mapping\nof the disease and more careful planning of endometriosis surgery.\nTherefore, the tests for preoperative mapping of the disease were\nonly evaluated as triage tests to inform decisions to undertake\nsurgery for endometriosis.\nThe combination of PV examination and TVUS demonstrated the\nfollowing diagnostic estimates:\n1. for POD obliteration: a sensitivity of 0.87 (95% CI 0.69 to 0.96)\nand a specificity of 0.98 (95% CI 0.95 to 1.00), meeting the criteria\nfor a SpIN triage test;\n2. for vaginal wall DIE: a sensitivity of 0.82 (95% CI 0.60 to 0.95) and\na specificity of 0.99 (95% CI 0.97 to 1.00), meeting the criteria for\na SpIN triage test;\n3. for RVS endometriosis: a sensitivity of 0.88 (95% CI 0.47 to 1.00)\nand a specificity of 0.99 (95% CI 0.96 to 1.00), meeting the criteria\nfor a SpIN triage test; sensitivity demonstrated wide CIs;\n4. for rectal endometriosis: a sensitivity of 0.96 (95% CI 0.86 to 0.99)\nand a specificity of 0.98 (95% CI 0.94 to 1.00), meeting the criteria\nfor a replacement and both a SnOUT and SpIN triage tests.\nSeparate diagnostic estimates for TVUS only were not reported in\nthis study.\nInvestigations of heterogeneity and sensitivity analyses\nThe potential sources of heterogeneity are outlined in Secondary\nobjectives. There was only one study evaluating each test, therefore\ninvestigations of heterogeneity and sensitivity analyses were not\npossible in this review.\nD I S C U S S I O N\nSummary of main results\nSummary of main results presented in this review\nFi/f_teen combinations of several non-invasive methods for the\ndiagnosis of endometriosis were evaluated in 11 included\nstudies published between 1996 and 2014 and comprising 1339\nparticipants. The composite tests have been assessed in small\nindividual studies, providing insufficient data to perform a meta-\nanalysis. None of the included studies were of high methodological\nquality. There were too few studies to perform a meaningful\nevaluation for any of the combination tests. Although some tests\nwere sensitive and specific enough to qualify as a replacement or\ntriage test for detecting endometriosis, each was explored in only\none study and warrant further validation.\nCombinations of several testing methods that met the criteria for a\nreplacement test.\n1. IL-6 >15.4 pg/ml [serum] + PGP 9.5 [endometrium] - for pelvic\nendometriosis\n2. PV examination + TVUS - for rectal endometriosis\nCombinations of several testing methods that met the criteria for a\nSpIN triage test.\n1. VDBP-Cr [urine] x CA-125 [serum] >2755 - for pelvic\nendometriosis\n2. History (length of menses) + CA-125 [serum] >35 U/ml +\nleukocytes [endometrium] - for pelvic endometriosis\n3. TVUS + (CA-125 [serum] ≥25 U/ml OR CA-19.9 [serum] ≥12 U/ml)\n- for ovarian endometrioma\n4. TVUS + CA-19.9 [serum] ≥12 U/ml - for ovarian endometrioma\n5. TVUS + CA-125 [serum] ≥20 U/ml - for ovarian endometrioma\n6. TVUS + CA-125 [serum] ≥25 U/ml - for ovarian endometrioma\n7. TVUS + CA-125 [serum] ≥35 U/ml - for ovarian endometrioma\n8. PV examination + TVUS - for the following anatomic locations:\na. POD obliteration;\nb. vaginal wall;\nc. RVS.\nIn all the included studies, combinations of the biomarkers had\nhigher diagnostic estimates than those reported for each individual\ncomponent of the combined test. However, addition of CA-125\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n35\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nhad only a small contribution to the diagnostic performance of\nthe test. We also observed that combinations of biomarkers with\nTVUS for detecting ovarian endometrioma had lower sensitivity\nand largely comparable specificity than that presented for TVUS\nalone in the review on imaging tests (Nisenblat 2016b). Considering\nthe results of meta-analysis from the imaging tests review, addition\nof vaginal examination to TVUS improved diagnostic performance\nof ultrasound in detecting vaginal and rectal endometriosis, but\ndid not seem to be superior to TVUS alone for detecting of POD\nobliteration and RVS endometriosis.\nThe findings of this 'combination of the tests' review need\nto be interpreted with caution. Considering both the level of\nheterogeneity and the high/unclear risk of bias of included studies,\nthe results do not appear sufficiently reliable to inform clinical\npractice.\nStrengths and weaknesses of the review\nThis review is part of a comprehensive review series of minimally\ninvasive biomarkers for the diagnosis of endometriosis, including\n1) imaging tests (Nisenblat 2016b ), 2) urinary biomarkers (Liu\n2015), 3) blood biomarkers (Nisenblat 2016a ), 4) endometrial\nbiomarkers (Gupta 2016), and 5) combination of several testing\nmodalities, presented in this review. The main strength of the\nreview series is its attempt to systematically review the vast\nnumber of widely heterogeneous studies in the literature while\napplying similar methods of study selection, data extraction and\nquality appraisal. It therefore should allow the most accurate\npicture of diagnostic test accuracy of non-invasive tests for\nendometriosis. The review series provide systematisation of the\ncurrent evidence on all the non-invasive tests for endometriosis,\nidentifying pitfalls in all the imaging and biomarker research areas\nand provides practical suggestions on further directions for high-\nquality diagnostic research in the field of endometriosis.\nThe following are the main strong points of this review.\n1. A very thorough search of the current literature was undertaken\nand included studies written in languages other than English.\n2. Data extraction by three independent reviewers and use of a\nmodified QUADAS-2 tool to perform quality assessments.\n3. Stringent selection criteria ensured that eligible studies utilised\nprospectively collected samples for the biomarker-based tests\nand prospectively enrolled and tested women for the imaging-\nbased tests.\n4. Inclusion of only clinically relevant population limited to women\nof reproductive age, which minimised the risk of bias in\ninterpreting the reference standard and index test.\n5. The authors of the studies were approached in attempt to\nobtain any missing information required to assess eligibility and\ncritically appraise the studies.\n6. The combinations including examination and imaging tests also\nprovided information on detecting of specific anatomical sites\nof DIE, which aids in preoperative mapping of the disease.\n7. In this review only one study (9%) was of a 'two-gate design' (a\npoorer quality design feature than ‘single-gate design’).\nTherefore, the majority of the included studies comprised a\nclinically relevant population that would have undergone tests\nin practice and were at low risk of misinterpretation of the test\nresults secondary to bias in selecting an adequate control group.\nThese strengths permit the conclusion that this is the most robust\nreview on the topic currently available to inform improvements in\nthe care of women being considered for diagnosis of endometriosis.\nThe main limitation of the review is that there was a single study for\neach evaluated index test and no meta-analysis was possible. The\nstudies varied with respect to the included populations, severity\nof endometriosis, menstrual cycle phase at testing, or radiological\nprotocols for index tests. Sources of heterogeneity were unable to\nbe explored for any test due to a single study in all evaluations.\nAll the included studies were of high or unclear risk of bias, which\ncontributed to the low quality of evidence presented in this review.\nAdditional weaknesses of this review include the following.\n1. Most of the biomarker studies determined the diagnostic cut-\noff thresholds using a ROC analysis without any subsequent\nvalidation in an independent cohort.\n2. The variation in the selection of the case and control\ngroups with inclusion of participants that may not reflect a\nclinically representative population. The reported prevalence of\nendometriosis in this review (up to 69%) was generally higher\nthan the previously reported prevalence for endometriosis (6%\nto 10% in the general female population and 35% to 50% in\nsymptomatic women) (Giudice 2004). This may reflect a high\nrisk of patient selection bias in tertiary referral centres, where\nmost of the studies were conducted. Selection bias appeared\nto be reduced, but not eliminated by consecutively enrolling\nparticipants, however the information on method of enrolment\nwas missing in most of the included studies. In this review,\n44% of the studies included women with a limited spectrum or\nspecific type of endometriosis. These studies were included to\navoid omission of potentially valuable diagnostic information,\nbut could skew the diagnostic estimates in either direction and\nsubsequently interfere with the interpretation of the index test\nresults. It was not possible to evaluate population and disease\nspectrum effects because of the paucity of suitable data.\n3. Inappropriate assignation to the endometriosis and control\ngroups could not be excluded in many studies and is\nanother weakness of the review. Surgical misdiagnosis is\na potential cause of bias as the number and experience\nof the surgical team, the surgical diagnostic criteria and\nthe surgical methods were poorly described in most of the\nincluded studies. We now have a standardised technique\nfor performing laparoscopy and we recommend that any\nfuture studies use this standardised method of undertaking\nlaparoscopy (Becker 2014). Additionally, we did not confine\nthe studies included in this review to those that reported\nhistological confirmation of endometriotic lesions. In this\nreview, 36% of the included studies relied on surgical diagnosis\nwithout histological confirmation. Although a recent ESHRE\nguideline stated that evidence is lacking to support laparoscopy\nwithout histology to confirm endometriosis (Dunselman 2014),\nthe clinical significance of histological verification remains\ndebatable. Diagnosis by surgical visualisation only, remains a\ncommon clinical practice and can be considered reliable when\nan accurate inspection of the abdominal cavity is performed\nby experienced surgeons. We chose to include the studies that\nused surgery alone to diagnose endometriosis as we did not\nwish to lose this potentially valuable information, however this\ncould impact the accuracy of assignment to the case and control\ngroups.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n36\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n4. Furthermore, excluding unpublished data could potentially\neliminate valuable information on the tests that were not\naltered by endometriosis. The decision to exclude unpublished\nstudies was made due to difficulties in reliable assessment of\neligibility and methodological quality of the studies reported\nonly in abstract form. However, this contributed to high risk of\npublication bias.\n5. The optimal methodology of systematic reviews of diagnostic\ntest accuracy is still emerging. This includes assessment\nand assignment of methodological quality, approach to data\nanalysis and interpretation of the results. There are no well-\nestablished criteria for replacement or triage diagnostic tests,\ntherefore we chose criteria that were both realistic and clinically\napplicable to assist in the interpretation of the complex results.\nFor a replacement test, we considered the threshold reported\nby the only systematic review on accuracy of the reference\nstandard (laparoscopy) in detecting endometriosis (Wykes 2004)\nto be the most objective. The meta-analysis was published in\n2004 and included four eligible studies comprising 433 women.\nWe acknowledge the limitations associated with emphasising a\nsingle review, particularly if it does not present the latest and\npossibly more accurate data that reflect advances in surgical\nexpertise and technology. A further systematic analysis of\nmore recently published studies to determine the accuracy of\nlaparoscopy was beyond the scope of this review. The criteria\nfor triage tests utilised the common concepts of SnOUT and\nSpIN in medical statistics and the cut-offs were set at levels we\nconsidered to be clinically relevant (see Role of index test(s)). We\nencourage the readers to apply independent interpretations of\nthe presented diagnostic estimates with using thresholds that\nmay be more applicable to specific populations and clinical\ncircumstances.\nApplicability of findings to the review question\nQUADAS-2 assigned a low rank to clinical applicability with respect\nto patient selection in 55% of the studies (6/11), summarised as\na high concern. This occurred when the set of women in the\nstudy was broader than that seen in clinical practice or when the\nspectrum of the target condition was limited and the findings may\nnot be applicable to the review question and to clinical practice.\nApplicability of the index test and reference standard was judged\nto be satisfactory using the QUADAS-2 tool for all studies. However,\nthe majority of included studies were conducted in academic\ninstitutions with a high level of expertise in laboratory techniques\nor in gynaecological imaging and the index test outcome measures\nmay not be able to be reproduced in all institutions or extrapolated\nto general gynaecological practice.\nSome potentially relevant well-designed studies were excluded as\nthey did not directly address the review question. For example,\nwe did exclude studies that compared endometrioma with other\novarian masses as they either did not meet our inclusion criteria\nfor reproductive age or assessed the numbers of cysts rather\nthan the number of women. Therefore the review question on\nnon-invasive diagnosis of ovarian endometriosis could not be\nfully addressed. Some forms of endometriosis, such as bladder,\nureteric or those involving the extra-pelvic sites (e.g. umbilicus,\nhernia sacs, abdominal wall, lung, kidney, etc.) were also excluded\nfrom the review as they are informed predominantly by case\nreports or small case series and diagnostic laparoscopy is not\nan applicable reference test for these conditions. Although these\ntarget conditions are rare, from a clinical perspective the diagnostic\noptions for these forms of endometriosis remains unclear.\nA U T H O R S ' /uni00A0 C O N C L U S I O N S\nImplications for practice\nAlthough several combinations of tests reached the threshold of\ndiagnostic accuracy to be considered as a replacement test for\ndiagnostic laparoscopy or a triage to improve selection for surgery,\nthese results hinged on only one study in each case, so would need\nto be confirmed prior to widespread implementation.\nOne of the combination tests that qualified for a replacement\ntest for detecting endometriosis included endometrial PGP 9.5.\nIt must be noted that PGP 9.5 has not yet reached the criteria\nfor routine use as a low-invasive diagnostic test in clinical\npractice, as demonstrated in the endometrial biomarkers review\n(Gupta 2016). Its utility is dependent on its consistency of\ndetection in the endometrium and its consistency of accuracy\nin diagnosing endometriosis. More work on establishing the best\nway of endometrial sampling and universal laboratory methods\nis needed. Besides, in-office sampling of the endometrium may\nnot be applicable to the group of adolescent girls, for whom\nearly diagnosis and abstaining from the diagnostic surgery are\nparticularly important.\nSerum CA-125 showed disappointing results and appeared to have\nno value in diagnosing endometriosis as a single test (Nisenblat\n2016a). This is consistent with international guidelines which\ndo not recommend CA-125 testing in women with suspected\nendometriosis (ACOG 2010; Dunselman 2014; SOGC 2010). CA-125\nwas incorporated in the diagnostic panels that showed high\ndiagnostic performance, however its value as a part of a combined\npanel has to be established.\nCombination of transvaginal ultrasound (TVUS) with blood\nbiomarkers (CA-125 or CA 19.9) could establish the diagnosis of\novarian endometrioma with high certainty, whereas negative test\ncould not confirm that participants are disease-free. Scrutiny of\nthe diagnostic test accuracy statistics reveals that, in fact, addition\nof any of these biomarkers does not substantially add to the\naccuracy of diagnosing endometrioma provided by TVUS alone,\nas demonstrated in the imaging review from this series (Nisenblat\n2016b). Combination of TVUS with vaginal examination was\naccurate enough to detect endometriosis in the pouch of Douglas\n(POD), vaginal wall and rectovaginal septum (RVS), but a normal\nexamination could not exclude endometriosis. This is consistent\nwith international guidelines which recommend TVUS as a first line\ninvestigation in conjunction with a history and pelvic examination\nin women with suspected endometriosis, but does not recommend\nits use as a replacement test for diagnostic surgery (ACOG 2010;\nDunselman 2014 ; SOGC 2010 ). Considering the findings of the\nimaging tests review from this series, several imaging methods\ndisplayed high accuracy in detecting pelvic, ovarian or deep\ninfiltrating endometriosis (DIE), demonstrating estimates superior\nto those for imaging and biomarkers combinations (Nisenblat\n2016b). These tests included TVUS with bowel preparation (TVUS-\nBP) and rectal water contrast (RWC-TVS) and MRI, but none of these\ntests was included in any of the combined test panels.\nRectal endometriosis was the only site that could be accurately\ndetected by using TVUS and pelvic examination. This is particularly\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n37\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nimportant for detecting rectosigmoid endometriosis as presurgical\nbowel preparation and surgeries that combine the expertise of\ngynaecologists and colorectal surgeons (or involve gynaecological\nsurgeons with the expertise to undertake bowel surgery) can be\nplanned preoperatively when rectosigmoid lesions are relatively\nreliably detected.\nTherefore, the evidence on combinations of the tests to be\nused in clinical practice as a replacement test to supplant\nlaparoscopic diagnosis or a triage test to reduce the requirement\nfor laparoscopic surgical diagnosis remains insufficient. Although\ndiagnostic potential was demonstrated for a number of tests,\nthe level of heterogeneity, wide confidence intervals and high/\nunclear risk of bias in most studies included in this review series\nundermines the reliability of the presented results and hence, these\ndata cannot be used confidently to inform clinical practice.\nIf the findings of large high-quality studies confirm that any of\nthese tests are suitable replacement tests, this would be strong\ngrounds to consider these as an alternative low invasive diagnostic\napproach instead of the current gold standard laparoscopic surgical\ndiagnosis. An accurate 'negative' non-invasive test is expected to\nreduce the need for diagnostic surgery in 50% to 70% of women\nwith chronic pelvic pain or infertility (Giudice 2004), although it\nis likely that some women with a negative test would still require\nsurgery to explore other pathologies. An accurate 'positive' non-\ninvasive test for endometriosis is likely to increase the need for\nsurgery in women with mild symptoms or subfertility (D'Hooghe\n2006). The ability to diagnose endometriosis in an outpatient\nsetting would see the diagnosis being made sooner, with fewer\nfollow-up visits, earlier institution of effective treatment and likely\na cost-saving benefit for the woman and the health service (both\nin direct medical costs and in time off work). Other potential\nadvantages of a non-invasive test over a surgical diagnosis include\nreduced discomfort, shorter recovery times and a reduction in the\nrare but serious complications of anaesthesia and surgery.\nIf the findings of the included studies suggesting any of these tests\nas a triage test can be replicated in other settings, this would\nbe strong grounds to consider these tests to improve selection\nfor more invasive surgical diagnosis. The triage process can be\nfurther improved by utilising a sequential approach with both\nSpIN and SnOUT types of tests (Figure 1). This would reduce the\ncomplications and costs of surgery and is expected to cut down long\nsurgical wait lists making surgery more accessible for the women\nwho are likely to benefit from it.\nAlthough guidelines from multiple authorities suggest medical\nmanagement as a first-line treatment for pelvic pain, most women\nwould prefer having a definite diagnosis before commencing\npotentially long-term therapy. Also, the indications for fertility\nmanagement are less clear in 'undiagnosed' women with\nsuspected endometriosis. If therapeutic surgery is considered,\nreliable detection of ovarian endometriomas potentially enables\nsurgeons to assess ovarian reserve and counsel women about\nfertility preservation before operating on ovarian tissue and risking\na reduction in their future fertility. Reliably detecting DIE/posterior\nDIE could add weight to a decision to prioritise surgery and could\nimprove preoperative informed consent. Until an accurate non-\ninvasive diagnostic test is developed and tested in large clinical\npopulations, it is impossible to predict accurately its impact on\nsurgical uptake and the number of women that would benefit from\nperforming the test.\nIt is important to emphasise that in the absence of well-established\ncriteria for an adequate diagnostic test, the diagnostic criteria for\nreplacement and triage tests were determined by the authors of this\nreview series in a way that we believe will aid the interpretation\nfor clinically active readers. However, we encourage readers to\napply different criteria according to each clinical population and\nsituation.\nImplications for research\nCurrently randomised controlled trials of treatment require women\nwith and without endometriosis to have had diagnostic surgery\nfor accurate group allocation. For ethical reasons, therapeutic\nsurgery is usually performed at the same time potentially biasing\ntreatment trial outcomes. Thus our current inability to diagnose\nand assess the progression of endometriosis in a non-invasive way\nis a significant limitation in the advancement of clinical research in\nendometriosis.\nIt does appear that combinations of diagnostic tests hold\npromise for the future, based purely on the number of studies\nwhose diagnostic test accuracy results reached or approached\nthe threshold for a replacement or a triage test, compared to\nthe paucity of studies in which single tests approached these\nthresholds.\nThe QUADAS quality assessment of the included studies identified\nseveral weaknesses in study design that can impede an objective\nevaluation of the findings. We recommend that future researchers\nundertaking studies for endometriosis diagnostic test accuracy\nconsider: 1) including large cohorts a/f_ter predefining the sample\nsize via a power calculation (Liu 2005 ); 2) focusing on a 'single-\ngate' design that only includes a clinically relevant population\n(Rutjes 2005 ); 3) utilising a diagnostic accuracy study design\nthat adheres to the recommendations of the Standards for\nReporting of Diagnostic Accuracy (STARD) initiative (Bossuyt\n2003); 4) incorporating the QUADAS checklist into the study\ndesign ( Whiting 2011 ); 5) formally assessing inter-and intra-\nobserver variability of the laboratory methods and imaging tests;\n6) establishing universally acceptable laboratory methodologies\n(Rahimoglu 2014), radiological protocols and diagnostic criteria\nfor a positive test; 7) utilising universally acceptable methods of\nperforming laparoscopy (Becker 2014) as the reference standard\ntest; 8) implementing validation techniques to assess how the\nresults of a statistical analysis will generalise to an independent\ndata set; 9) undertaking direct comparisons of promising tests in\nconjunction with a cost effectiveness analysis, 10) applying testing\nto different clinical phenotypes (Vitonis 2014) rather than to women\nclassified according to rASRM staging; and 11) assessing the long-\nterm outcomes and lifetime healthcare costs of women who have\nparticipated in diagnostic test accuracy trials of specific diagnostic\ntests.\nEvaluation of the strongest candidates for possible replacement\nand triage non-invasive diagnostic tests should continue. Specific\nopportunities for further research identified by this review include\nthe following.\n1. Assessing the diagnostic potential of the tests identified in this\nand other reviews from the series as promising replacement or\ntriage test for endometriosis in larger, high-quality studies.\na. Developing a simplified and improved detection\ntechnique for endometrial neuronal immuno-histochemical\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n38\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nbiomarkers, such as PGP9.5 utilising digitally enhanced\nassessment of differences in immunohistochemistry\nappearances; an innovative clinical research to improve\ntechniques of endometrial biopsy, including utilisation of\nendocervical analgesia, further improvements of the biopsy\ncannulas and possibly development of endometrial brushes\nfor superficial sampling of the endometrium (although this\nmight not be suited for markers that are only expressed\nwithin the stroma).\nb. Further development of the composite tests with\nincorporation of different testing modalities.\nc. Attempt to develop a diagnostic algorithm for diagnosis\nand accurate topographical mapping of endometriosis by\nutilising several imaging or combined testing methods.\nd. Establishing of universal radiological diagnostic criteria and\nstudy protocols.\n2. Incorporation of clinical history or pelvic examination in a\ndiagnostic model.\n3. Exploring the value of sequential testing, implementing SnOUT\nand SpIN triage tests in diagnosing endometriosis, including\nclinical parameters in the decision tree algorithm in conjunction\nwith the cost-effectiveness of such testing.\n4. Direct comparison of several promising tests in well-designed\ndiagnostic accuracy studies.\n5. Evaluation of the whole spectrum of disease across all phases\nof the menstrual cycle, aiming to identify the most appropriate\ntarget population and the best time of testing.\n6. Attempting testing in the populations that differ by clinical\nphenotype rather than by rASRM staging in view of poor\ncorrelation of this classification with clinical presentations and\ntreatment outcomes.\n7. To add separate evaluations of the biomarker and imaging tests\nfor differentiating ovarian endometrioma from other ovarian\nmasses, including malignant and borderline tumours in women\nof reproductive age.\n8. Assessing the long-term outcomes and lifetime healthcare costs\nin diagnostic test accuracy trials that have evaluated specific\ndiagnostic blood tests.\nA C K N O W L E D G E M E N T S\nWe would like to thank Associate Professor Petra Macaskill\nfor her valuable comments and substantial contribution to the\ndevelopment of the statistical methods for the review. Sincere\nthanks to the late Professor Ali Akoum and Professor Ian Fraser\nfor an intellectual input and help with dra/f_ting of the protocol. We\nare grateful to Marian Showell, the Information Specialist of the\nCochrane Gynaecology and Fertility Group, for her help in designing\nand conducting the literature search and in locating the full texts\nof the relevant studies. We also thank the authors and contributors\nof the review series Devashana Gupta, Emily Liu, Rabia Shaikh and\nDeepika Arora, for their dedicated assistance in studies' selection\nprocess. Finally, we thank all contacted authors who contributed\ninformation to this review.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n39\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nREFERENCES\n/uni00A0\nReferences to studies included in this review\nCho 2012 {published data only}\nCho/uni00A0S, Choi/uni00A0YS, Yim/uni00A0SY, Yang/uni00A0HI, Jeon/uni00A0YE, Lee/uni00A0KE, et al. Urinary\nvitamin D-binding protein is elevated in patients with\nendometriosis. Human Reproduction 2012;27(2):515-22.\nel Sharkwy 2013 {published data only}\nel/uni00A0Sharkwy/uni00A0IAE, Evers/uni00A0JLH, Dunselman/uni00A0GAJ, Vanderlinden/uni00A0PJQ.\nCombination of non-invasive and semi-invasive tests for\ndiagnosis of minimal to mild endometriosis. Archives of\nGynecology & Obstetrics 2013;288(4):793-7.\nGagné 2003 {published data only}\nGagné/uni00A0D, Rivard/uni00A0M, Pagé/uni00A0M, Lépine/uni00A0M, Platon/uni00A0C, Shazand/uni00A0K,\net al. Development of a non surgical diagnostic tool for\nendometriosis based on the detection of endometrial\nleukocyte subsets and serum CA-125 levels. Fertility & Sterility\n2003;80(4):876-85.\nGuerriero 1996a {published data only}\nGuerriero/uni00A0S, Ajossa/uni00A0S, Paoletti/uni00A0AM, Mais/uni00A0V, Angiolucci/uni00A0M,\nMelis/uni00A0GB. Tumour marker and transvaginal ultrasonography\nin the diagnosis of endometrioma. Obstetrics & Gynaecology\n1996;88(3):403-7.\nGuerriero 1996b {published data only}\nGuerriero/uni00A0S, Mais/uni00A0V, Ajossa/uni00A0S, Paoletti/uni00A0AM, Angiolucci/uni00A0M,\nMelis/uni00A0GB. Transvaginal ultrasonography combined with CA-125\nplasma levels in the diagnosis of endometrioma. Fertility &\nSterility 1996;65(2):293-8.\nHudelist 2009 {published data only}\nHudelist/uni00A0G, Oberwinkler/uni00A0KH, Singer/uni00A0CF, Tuttlies/uni00A0F, Rauter/uni00A0G,\nRitter/uni00A0O, et al. Combination of transvaginal sonography and\nclinical examination for preoperative diagnosis of pelvic\nendometriosis. Human Reproduction 2009;24(5):1018-24.\nKoninckx 1996 {published data only}\nKoninckx/uni00A0P, Meuleman/uni00A0C, Oosterlynck/uni00A0D, Cornillie/uni00A0F. Diagnosis\nof deep endometriosis by clinical examination during\nmenstruation and plasma CA-125 concentration. Fertility &\nSterility 1996;65(2):280-7.\nMarasinghe 2014 {published data only}\nMarasinghe/uni00A0J, Senanayake/uni00A0H, Saravanabhava/uni00A0N, Arambepola/uni00A0C,\nCondous/uni00A0G, Greenwood/uni00A0P. History, pelvic examination findings\nand mobility of ovaries as a sonographic marker to detect\npelvic adhesions with fixed ovaries. Journal of Obstetrics and\nGynaecology Research 2014;40(3):785-90.\nPaiva 2014 {published data only}\nPaiva/uni00A0P, Lappas/uni00A0M, Barker/uni00A0G, Healey/uni00A0M. Using symptom scores,\nlifestyle measures and biochemical markers to create a test for\nendometriosis. Journal of Endometriosis 2014;6(3):135-43.\nYun 2014 {published data only}\nYun/uni00A0BH, Lee/uni00A0YS, Chon/uni00A0SJ, Jung/uni00A0YS, Yim/uni00A0SY, Kim/uni00A0HY, et al.\nEvaluation of elevated urinary enolase I levels in patients with\nendometriosis. Biomarkers 2014;19(1):16-21.\nZeng 2005 {published data only}\nZeng/uni00A0F, Xue/uni00A0M, Zevallos/uni00A0HBV, Lai/uni00A0D, Arthur/uni00A0J, Ng/uni00A0C, et al.\nDiagnostic value of the detection of aromatase cytochrome\nP450 and CA125 for endometriosis [芳⾹化酶细胞⾊素P450及\nCA125 联合检测对⼦宫内膜异位症的诊断价值]. Journal of\nCentral South University (Medical Sciences) 2005;30(6):682-5.\n/uni00A0\nReferences to studies excluded from this review\nAbrao 2007 {published data only}\nAbrao/uni00A0MS, Goncalves/uni00A0MODC, Dias/uni00A0JA/uni00A0Jr, Podgaec/uni00A0S, Chamie/uni00A0LP,\nBlasbalg/uni00A0R. Comparison between clinical examination,\ntransvaginal sonography and magnetic resonance imaging\nfor the diagnosis of deep endometriosis. Human Reproduction\n2007;22(12):3092-7.\nAdamyan 1993 {published data only}\nAdamyan/uni00A0L, Fanchenko/uni00A0N, Alexeyeva/uni00A0M, Andreyeva/uni00A0Y, Novikov/uni00A0Y,\nJahan/uni00A0I. Hormonal and immunologic methods in the diagnosis\nand treatment of patients with benign ovarian tumors\nand endometriotic cysts. International Journal of Fertility\n1993;38(2):92-8.\nAlcazar 2011 {published data only}\nAlcazar/uni00A0J, Guerriero/uni00A0S, Minguez/uni00A0J, Ajossa/uni00A0S, Paoletti/uni00A0A, Ruiz-\nZambrana/uni00A0A, et al. Adding cancer antigen 125 screening\nto gray scale sonography for predicting specific diagnosis\nof benignadnexal masses in premenopausal women:\nis it worthwhile?. Journal of Ultrasound in Medicine\n2011;30(10):1381-6.\nBadawy 1984 {published data only}\nBadawy/uni00A0SZ, Cuenca/uni00A0V, Stitzel/uni00A0A, Jacobs/uni00A0RD, Tomar/uni00A0RH.\nAutoimmune phenomena in infertile patients with\nendometriosis. Obstetrics & Gynecology 1984;63(3):271-5.\nBazot 2009 {published data only}\nBazot/uni00A0M, Lafont/uni00A0C, Rouzier/uni00A0R, Roseau/uni00A0G, Thomassin-Naggara/uni00A0I,\nDaraii/uni00A0E. Diagnostic accuracy of physical examination,\ntransvaginal sonography, rectal endoscopic sonography, and\nmagnetic resonance imaging to diagnose deep infiltrating\nendometriosis. Fertility & Sterility 2009;92(6):1825-33.\nBorboletto 1995 {published data only}\nBorboletto/uni00A0C, Goncalves/uni00A0W, Giusa/uni00A0M, de/uni00A0Freitas/uni00A0V, Baracat/uni00A0E,\nde/uni00A0Lima/uni00A0G. Transvaginal ultrasound, color Doppler\nvelocimetry and CA -125 in the diagnosis and follow-up\nof pelvic endometriosis [Ultra-sonografia transvaginal,\ndoplervelocimetria colorida e dosagem de CA-125 no\ndiagnóstico e no seguimento da endometriose pélvica]. Femina\n1995;23(8):723-6.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n40\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nCho 2007 {published data only}\nCho/uni00A0S, Oh/uni00A0Y, Nam/uni00A0A, Kim/uni00A0H, Park/uni00A0J, Kim/uni00A0J, et al. Evaluation\nof serum and urinary angiogenic factors in patients with\nendometriosis. American Journal of Reproductive Immunology\n2007;58(6):497-504.\nda Silva 2014 {published data only}\nda/uni00A0Silva/uni00A0C, Belo/uni00A0A, Andrade/uni00A0S, Campos/uni00A0P, Ferreira/uni00A0M,\nda/uni00A0Silva-Filho/uni00A0A, et al. Identification of local angiogenic and\ninflammatory markers in the menstrual blood of women\nwith endometriosis. Biomedicine and Pharmacotherapy\n2014;68(7):899-904.\nDias 2012 {published data only}\nDias/uni00A0J, Podgaec/uni00A0S, de/uni00A0Oliveira/uni00A0R, Marin/uni00A0M, Baracat/uni00A0E, Abrao/uni00A0M.\nPatients with endometriosis of the rectosigmoid have a higher\npercentage of natural killer cells in peripheralblood. Journal of\nMinimally Invasive Gynecology 2012;19(3):317-24.\nEskenazi 2001 {published data only}\nEskenazi/uni00A0B, Warner/uni00A0M, Bonsignore/uni00A0L, Olive/uni00A0D, Samuels/uni00A0S,\nVercellini/uni00A0P. Validation study of nonsurgical diagnosis of\nendometriosis. Fertility & Sterility 2001;76(5):929-35.\nFedele 1988 {published data only}\nFedele/uni00A0L, Vercellini/uni00A0P, Arcaini/uni00A0L, Grazia da Dalt/uni00A0M, Candiani/uni00A0G. CA\n125 in serum, peritoneal fluid, active lesions, and endometrium\nof patients with endometriosis. American Journal of Obstetrics &\nGynecology 1988;158(1):166-70.\nGuerriero 1997 {published data only}\nGuerriero/uni00A0S, Mallarini/uni00A0G, Ajossa/uni00A0S, Risalvato/uni00A0A, Satta/uni00A0R, Mais/uni00A0V,\net al. Transvaginal ultrasound and computed tomography\ncombined with clinical parameters and CA-125 determinations\nin the differential diagnosis of persistent ovarian cysts in\npremenopausal women. Ultrasound in Obstetrics & Gynecology\n1997;9(5):339-43.\nHudelist 2011b {published data only}\nHudelist/uni00A0G, Ballard/uni00A0K, English/uni00A0J, Wright/uni00A0J, Banerjee/uni00A0S,\nMastoroudes/uni00A0H, et al. Transvaginal sonography vs. clinical\nexamination in the preoperative diagnosis of deep infiltrating\nendometriosis. Ultrasound in Obstetrics & Gynecology\n2011;37(4):480-7.\nKocbek 2014 {published data only}\nKocbek/uni00A0V, Bersinger/uni00A0N, Brglez/uni00A0V, Mueller/uni00A0M, Petan/uni00A0T, Rizner/uni00A0T.\nPhospholipase A2 group IIA is elevated in endometriomas but\nnot in peritoneal fluid and serum of ovarian endometriosis\npatients. Gynecological Endocrinology 2015;31(3):214-8.\nKuessel 2014 {published data only}\nKuessel/uni00A0L, Jaeger-Lansky/uni00A0A, Pateisky/uni00A0P, Rossberg/uni00A0N, Schulz/uni00A0A,\nSchmitz/uni00A0A, et al. Cytokeratin-19 as a biomarker in urine and in\nserum for the diagnosis ofendometriosis – a prospective study.\nGynecological Endocrinology 2014;30(1):38-41.\nLee 2014 {published data only}\nLee/uni00A0Y, Tan/uni00A0CW, Venkatratnam/uni00A0A, Tan/uni00A0CS, Liang/uni00A0C,\nLoh/uni00A0S, et al. Dysregulated sphingolipid metabolism in\nendometriosis. Journal of Clinical Endocrinology & Metabolism\n2014;99(10):1913-21.\nNezhat 1994 {published data only}\nNezhat/uni00A0C, Santolaya/uni00A0J, Nezhat/uni00A0F, Nezhat/uni00A0C. Comparison of\ntransvaginal sonography and bimanual pelvic examination in\npatients with laparoscopicallyconfirmed endometriosis. Journal\nof the American Association of Gynecologic Laparoscopists\n1994;1(2):127-30.\nSzubert 2014 {published data only}\nSzubert/uni00A0M, Suzin/uni00A0J, Duechler/uni00A0M, Szuławska/uni00A0A, Czyż/uni00A0M,\nKowalczyk-Amico/uni00A0K. Evaluation of selected angiogenic and\ninflammatory markers in endometriosis before and a/f_ter\ndanazol treatment. Reproduction, Fertility and Development\n2014;26(3):414-20.\nWeerakiet 2000 {published data only}\nWeerakiet/uni00A0S, Wongkularb/uni00A0A, Rochanawutanon/uni00A0M, Rojanasakul/uni00A0A.\nTransvaginal ultrasonography combined with pelvic\nexamination in the diagnosis of ovarian endometrioma. Journal\nof the Medical Association of Thailand 2000;83(5):523-8.\nWolfler 2005 {published data only}\nWolfler/uni00A0M, Nagele/uni00A0F, Kolbus/uni00A0A, Seidl/uni00A0S, Schneider/uni00A0B, Huber/uni00A0J.\nA predictive model for endometriosis. Human Reproduction\n2005;20(6):1702-8.\nYong 2013 {published data only}\nYong/uni00A0P, Sutton/uni00A0C, Suen/uni00A0M, Williams/uni00A0C. Endovaginal ultrasound-\nassisted pain mapping in endometriosis and chronic pelvic\npain. Journal of Obstetrics & Gynaecology 2013;33:715-9.\n/uni00A0\nAdditional references\nACOG 2010\nThe American College of Obstetricians and Gynecologists.\nPractice Bulletin No. 114: Management of Endometriosis.\nObstetrics and Gynecology 2010;116(1):223-36.\nAdamson 2008\nAdamson/uni00A0GD, Pasta/uni00A0DJ. Endometriosis Fertility Index (EFI):\nthe new validated endometriosis staging system. Fertility and\nSterility 2010;94(5):1609-15.\nAlmeida Filho 2008\nAlmeida Filho/uni00A0DP, Oliveira/uni00A0LJ, Amaral/uni00A0VF. Accuracy of\nlaparoscopy for assessing patients with endometriosis. Sao\nPaulo Medical Journal 2008;126:305-8.\nAmerican Society for Reproductive Medicine 1997\nAmerican Society for Reproductive Medicine. Revised\nAmerican Society for Reproductive Medicine classification of\nendometriosis: 1996. Fertility and Sterility 1997;67(5):817-21.\nBallard 2008\nBallard/uni00A0KD, Seaman/uni00A0HE, de/uni00A0Vries/uni00A0CS, Wright/uni00A0JT. Can\nsymptomatology help in the diagnosis of endometriosis?\nFindings from a national case-control study - Part 1. British\nJournal of Obstetrics and Gynaecology 2008;115(11):1382-91.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n41\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nBatt 2003\nBatt/uni00A0R, Mitwally/uni00A0MF. Endometriosis from thelarche to midteens:\npathogenesis and prognosis, prevention and pedagogy. Journal\nof Pediatric and Adolescent Gynecology 2003;16:333-47.\nBecker 2014\nBecker/uni00A0CM, Laufer/uni00A0MR, Stratton/uni00A0P, Hummelshoj/uni00A0l, Missmer/uni00A0SA,\nZondervan/uni00A0KT, et al. World Endometriosis Research Foundation\nEndometriosis Phenome and biobanking harmonization\nproject: I. Surgical phenotype data collection in endometriosis\nresearch. Fertility and Sterility 2014;102(5):1213-22.\nBossuyt 2003\nBossuyt/uni00A0PM, Reitsma/uni00A0JB, Bruns/uni00A0DE, Gatsonis/uni00A0CA, Glasziou/uni00A0PP,\nIrwig/uni00A0LM, et al. Towards complete and accurate reporting\nof studies of diagnostic accuracy: the STARD initiative. BMJ\n2003;326(7379):41-4.\nBossuyt 2008\nBossuyt/uni00A0PM, Leeflang/uni00A0MM. Chapter 6: Developing Criteria for\nIncluding Studies. Cochrane Handbook for Systematic Reviews\nof Diagnostic Test Accuracy Version 0.4 [updated September\n2008]. The Cochrane Collaboration, 2008.\nChapron 2003a\nChapron/uni00A0C, Fauconnier/uni00A0A, Vieira/uni00A0M, Barakat/uni00A0H, Dousset/uni00A0B,\nPansini/uni00A0V, et al. Anatomical distribution of deeply infiltrating\nendometriosis: surgical implications and proposition for a\nclassification. Human Reproduction 2003;18(1):157-61.\nChapron 2003b\nChapron/uni00A0C, Fauconnier/uni00A0A, Dubuisson/uni00A0JB, Barakat/uni00A0H, Vieira/uni00A0M,\nBréart/uni00A0G. Deep infiltrating endometriosis: relation between\nseverity of dysmenorrhea and extent of disease. Human\nReproduction 2003;18:760-6.\nChapron 2003c\nChapron/uni00A0C, Cravello/uni00A0L, Chopin/uni00A0N, Kreiker/uni00A0G, Blanc/uni00A0B,\nDubuisson/uni00A0JB. Complications during set-up procedures for\nlaparoscopy in gynecology: open laparoscopy does not reduce\nthe risk of major complications. Acta Obstetrica Gynecologica\nScandinavica 2003;82:1125-9.\nD'Hooghe 2001\nD'Hooghe/uni00A0TM, Bambra/uni00A0CS, Xiao/uni00A0L, Peixe/uni00A0K, Hill/uni00A0JA. Effect of\nmenstruation and intrapelvic injection of endometrium on\ninflammatory parameters of peritoneal fluid in the baboon\n(Papio anubis and Papio cynocephalus). American Journal of\nObstetrics and Gynecology 2001;184(5):917-25.\nD'Hooghe 2006\nD’Hooghe/uni00A0TM, Mihalyi/uni00A0AM, Simsa/uni00A0P, Kyama/uni00A0CK, Peeraer/uni00A0K,\nDe/uni00A0Loecker/uni00A0P, et al. Why we need a non-invasive diagnostic\ntest for minimal to mild endometriosis with a high sensitivity.\nGynecologic and Obstetric Investigation 2006;62(3):136-8.\nde Vet 2008\nde/uni00A0Vet/uni00A0HCW, Eisinga/uni00A0A, Riphagen/uni00A0II, Aertgeerts/uni00A0B, Pewsner/uni00A0D.\nChapter 7: Searching for studies. Cochrane Handbook for\nSystematic Reviews of Diagnostic Test Accuracy Version 0.4\n[updated September 2008]. The Cochrane Collaboration, 2008.\nDmowski 1997\nDmowski/uni00A0WP, Lesniewicz/uni00A0R, Rana/uni00A0N, Pepping/uni00A0P, Noursalehi/uni00A0M.\nChanging trends in the diagnosis of endometriosis: a\ncomparative study of women with pelvic endometriosis\npresenting with chronic pelvic pain or infertility. Fertility and\nSterility 1997;67:238-43.\nDunselman 2014\nDunselman/uni00A0GA, Vermeulen/uni00A0N, Becker/uni00A0C, Calhaz-Jorge/uni00A0C,\nD'Hooghe/uni00A0T, De/uni00A0Bie/uni00A0B, et al. ESHRE guideline: management\nof women with endometriosis. Human Reproduction\n2014;29(3):400-12.\nFassbender 2015\nFassbender/uni00A0A, Burney/uni00A0RO, Dorien/uni00A0FO, D’Hooghe/uni00A0T, Giudice/uni00A0L.\nUpdate on biomarkers for the detection of endometriosis.\nBioMed Research International 2015;Epub:1-14.\nFauconnier 2005\nFauconnier/uni00A0A, Chapron/uni00A0C. Endometriosis and pelvic pain:\nepidemiological evidence of the relationship and implications.\nHuman Reproduction Update 2005;11:595-606.\nFrishman 2006\nFrishman/uni00A0GN, Salak/uni00A0JR. Conservative surgical management of\nendometriosis in women with pelvic pain. Journal of Minimally\nInvasive Gynecology 2006;13:546-58.\nGao 2006\nGao/uni00A0X, Yeh/uni00A0YC, Outley/uni00A0J, Simon/uni00A0J, Botteman/uni00A0M, Spalding/uni00A0J.\nHealth-related quality of life burden of women with\nendometriosis: a literature review. Current Medical Research and\nOpinion 2006;22:1787-97.\nGiudice 2004\nGiudice/uni00A0LC, Kao/uni00A0LC. Endometriosis. Lancet 2004;364:1789-99.\nGuerriero 2015\nGuerriero/uni00A0S, Ajossa/uni00A0S, Orozco/uni00A0R, Perniciano/uni00A0M, Jurado/uni00A0M,\nMelis/uni00A0GB, et al. Diagnostic accuracy of transvaginal ultrasound\nfor diagnosis of deep endometriosis in the recto-sigmoid:\na meta-analysis. Ultrasound in Obstetrics and Gynecology\n2015;Epub ahead of print:1.\nGuo 2009\nGuo/uni00A0SW. Recurrence of endometriosis and its control. Human\nReproduction Update 2009;15(4):441-61.\nGupta 2016\nGupta/uni00A0D, Hull/uni00A0ML, Fraser/uni00A0I, Miller/uni00A0L, Bossuyt/uni00A0PMM, Johnson/uni00A0N, et\nal. Endometrial biomarkers for the non-invasive diagnosis of\nendometriosis. Cochrane Database of Systematic Reviews 2016,\nIssue 4. [DOI: 10.1002/14651858.CD012165]\nGuzick 1997\nGuzick/uni00A0DS, Silliman/uni00A0NP, Adamson/uni00A0GD, Buttram/uni00A0VC/uni00A0Jr, Canis/uni00A0M,\nMalinak/uni00A0LR, et al. Prediction of pregnancy in infertile women\nbased on the American Society for/uni00A0 Reproductive Medicines\nrevised classification of endometriosis. Fertility and Sterility\n1997;67(5):822-9.\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n42\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nHudelist 2011a\nHudelist/uni00A0G, English/uni00A0J, Thomas/uni00A0AE, Tinelli/uni00A0A, Singer/uni00A0CF,\nKeckstein/uni00A0J. Diagnostic accuracy of transvaginal ultrasound\nfor non-invasive diagnosis of bowel endometriosis: systematic\nreview and meta-analysis. Ultrasound in Obstetrics and\nGynecology 2011;37(3):257-63.\nHull 2008\nHull/uni00A0ML, Escareno/uni00A0CR, Godsland/uni00A0JM, Doig/uni00A0JR, Johnson/uni00A0CM,\nPhillips/uni00A0SC, et al. Endometrial-peritoneal interactions during\nendometriotic lesion establishment. American Journal of\nPathology 2008;173(3):700-15.\nJohnson 2013\nJohnson/uni00A0NP, Hummelshoj/uni00A0L. Consensus on current\nmanagement of endometriosis. Human Reproduction\n2013;28(6):1552-68.\nJohnson 2015\nJohnson/uni00A0NP, Hummelshoj/uni00A0L, Adamson/uni00A0GD, Kecstein/uni00A0J,\nTaylor/uni00A0H, Abrao/uni00A0MS, et al. Consensus on the classification\nof endometriosis. Human Reproduction 2016 in press;in\npreparation/press:1.\nKao 2003\nKao/uni00A0LC, Germeyer/uni00A0A, Tulac/uni00A0S, Lobo/uni00A0S, Yang/uni00A0JP, Taylor/uni00A0RN,\net al. Expression profiling of endometrium from women\nwith endometriosis reveals candidate genes for disease-\nbased implantation failure and infertility. Endocrinology\n2003;144:2870-81.\nKennedy 2005\nKennedy/uni00A0S, Bergqvist/uni00A0A, Chapron/uni00A0C, D'Hooghe/uni00A0T, Dunselman/uni00A0G,\nGreb/uni00A0R, et al. ESHRE guideline for the diagnosis and treatment\nof endometriosis. Human Reproduction 2005;20:2698-704.\nKoninckx 1991\nKoninckx/uni00A0PR, Meuleman/uni00A0C, Demeyere/uni00A0S, Lesaffre/uni00A0E, Cornillie/uni00A0FJ.\nSuggestive evidence that pelvic endometriosis is a progressive\ndisease, whereas deeply infiltrating endometriosis is associated\nwith pelvic pain. Fertility and Sterility 1991;55(4):759-65.\nLing 1999\nLing/uni00A0F. Randomized controlled trial of depot leuprolide in\npatients with chronic pelvic pain and clinically suspected\nendometriosis. Obstetrics and Gynecology 1999;93:51-8.\nLiu 2005\nLiu/uni00A0A, Schisterman/uni00A0EF, Mazumdar/uni00A0M, Hu/uni00A0J. Power and sample\nsize calculation of comparative diagnostic accuracy studies\nwith multiple correlated test results. Biometrical Journal\n2005;47(2):140-50.\nLiu 2015\nLiu/uni00A0E, Nisenblat/uni00A0V, Farquhar/uni00A0C, Fraser/uni00A0I, Bossuyt/uni00A0PMM,\nJohnson/uni00A0N, et al. Urinary biomarkers for the non-invasive\ndiagnosis of endometriosis. Cochrane Database of Systematic\nReviews 2015, Issue 12. [DOI: 10.1002/14651858.CD012019]\nMarchino 2005\nMarchino/uni00A0GL, Gennarelli/uni00A0G, Enria/uni00A0R, Bongioanni/uni00A0F, Lipari/uni00A0G,\nMassobrio/uni00A0M. Diagnosis of pelvic endometriosis with use of\nmacroscopic versus histologic findings. Fertility and Sterility\n2005;84:12-5.\nMartin 2006\nMartin/uni00A0DC. Applying STARD criteria to the laparoscopic\nidentification of endometriosis (abstract)/uni00A0. Fertility and Sterility.\n2006; Vol. 86 Suppl 2:270.\nMatsuzaki 2006\nMatsuzaki/uni00A0S, Canis/uni00A0M, Pouly/uni00A0JL, Rabischong/uni00A0B, Botchorishvili/uni00A0R,\nMage/uni00A0G. Relationship between delay of surgical diagnosis\nand severity of disease in patients with symptomatic deep\ninfiltrating endometriosis. Fertility and Sterility 2006;86:1314-6.\nMay 2010\nMay/uni00A0KE, Conduit-Hulbert/uni00A0SA, Villar/uni00A0J, Kirtley/uni00A0S, Kennedy/uni00A0SH,\nBecker/uni00A0CM. Peripheral biomarkers of endometriosis:\nA systematic review. Human Reproduction Update\n2010;16(6):651-74.\nMay 2011\nMay/uni00A0KE, Villar/uni00A0J, Kirtley/uni00A0S, Kennedy/uni00A0SH, Becker/uni00A0CM. Endometrial\nalterations in endometriosis: a systematic review of putative\nbiomarkers. Human Reproduction Update 2011;17(5):637-53.\nMcGraw-Hill Dictionary of Medicine 2006\nSegen JC/uni00A0(author). McGraw-Hill Concise Dictionary of Modern\nMedicine. The McGraw-Hill Companies, Inc, 2006.\nMedeiros 2009\nMedeiros/uni00A0LRF, Rosa/uni00A0DD, Bozzetti/uni00A0MC, Fachel/uni00A0JMG, Furness/uni00A0S,\nGarry/uni00A0R, et al. Laparoscopy versus laparotomy for benign\novarian tumour. Cochrane Database of Systematic Reviews 2009,\nIssue 2. [DOI: 10.1002/14651858.CD004751.pub3]\nMedeiros 2015\nMedeiros/uni00A0LR, Rosa/uni00A0MI, Silva/uni00A0BR, Reis/uni00A0ME, Simon/uni00A0CS,\nDondossola/uni00A0ER, et al. Accuracy of magnetic resonance\nin deeply infiltrating endometriosis: a systematic review\nand meta-analysis. Archives of Gynecology and Obstetrics\n2015;291(3):611-21.\nMoore 2002\nMoore/uni00A0J, Copley/uni00A0S, Morris/uni00A0J, LIndsell/uni00A0D, Golding/uni00A0S, Kennedy/uni00A0S. A\nsystematic review of the accuracy of ultrasound in the diagnosis\nof endometriosis. Ultrasound in Obstetrics & Gynecology\n2002;20:630-4.\nNisenblat 2016a\nNisenblat/uni00A0V, Bossuyt/uni00A0PMM, Shaikh/uni00A0R, Arora/uni00A0D, Farquhar/uni00A0C,\nJordan/uni00A0V, et al. Blood biomarkers for the non-invasive diagnosis\nof endometriosis. Cochrane Database of Systematic Reviews\n2016, Issue 5. [DOI: 10.1002/14651858.CD012179]\nNisenblat 2016b\nNisenblat/uni00A0V, Bossuyt/uni00A0PMM, Farquhar/uni00A0C, Johnson/uni00A0N, Hull/uni00A0ML.\nImaging modalities for the non-invasive diagnosis of\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n43\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nendometriosis. Cochrane Database of Systematic Reviews 2016,\nIssue 2. [DOI: 10.1002/14651858.CD009591.pub2]\nNyholt 2012\nNyholt/uni00A0DR, Low/uni00A0SK, Anderson/uni00A0CA, Painter/uni00A0JN, Uno/uni00A0S, Morris/uni00A0AP,\net al. Genome-wide association meta-analysis identifies new\nendometriosis risk loci. Nature Genetics 2012;44(12):1355-9.\nRahimoglu 2014\nRahmioglu/uni00A0N, Fassbender/uni00A0A, Vitonis/uni00A0AF, Tworoger/uni00A0SS,\nHummelshoj/uni00A0L, D’Hooghe/uni00A0TM, et al. World Endometriosis\nResearch Foundation Endometriosis Phenome and biobanking\nharmonization project: III. Fluid biospecimen collection,\nprocessing, and storage in endometriosis research. Fertility and\nSterility 2014;102(5):1233-43.\nRedwine 2003\nRedwine/uni00A0DB. Invisible' microscopic endometriosis: a review.\nGynecologic and Obstetric Investigation 2003;55:63-7.\nRogers 2009\nRogers/uni00A0PA, D'Hooghe/uni00A0TM, Fazleabas/uni00A0A, Gargett/uni00A0CE, Giudice/uni00A0LC,\nMontgomery/uni00A0GW, et al. Priorities for Endometriosis Research:\nrecommendations from an international consensus workshop.\nReproductive Sciences 2009;16(4):335-46.\nRutjes 2005\nRutjes/uni00A0AWS, Reitsma/uni00A0JB, Vandenbroucke/uni00A0JP, Glas/uni00A0AS,\nBossuyt/uni00A0PMM. Case–control and two-gate designs in diagnostic\naccuracy studies. Clinical Chemistry 2005;51(8):1335-41.\nSampson 1927\nSampson/uni00A0JA. Peritoneal endometriosis due to menstrual\ndissemination of endometrial tissue into the peritoneal cavity.\nAmerican Journal of Obstetrics and Gynecology 1927;14:442-69.\nSimoens 2012\nSimoens/uni00A0S, Dunselman/uni00A0G, Dirksen/uni00A0C, Hummelshoj/uni00A0L, Bokor/uni00A0A,\nBrandes/uni00A0I, et al. The burden of endometriosis: costs and quality\nof life of women with endometriosis and treated in referral\ncentres. Human Reproduction 2012;27(5):1292-9.\nSinaii 2002\nSinaii/uni00A0N, Cleary/uni00A0SD, Ballweg/uni00A0ML, Nieman/uni00A0LK, Stratton/uni00A0P. High\nrates of autoimmune and endocrine disorders, fibromyalgia,\nchronic fatigue syndrome and atopic diseases among women\nwith endometriosis: A survey analysis. Human Reproduction\n2002;17(10):2715-24.\nSOGC 2010\nSociety of Obstetricians Gynaecologists of Canada.\nEndometriosis: diagnosis and management. SOGC clinical\npractice guideline no. 244. Journal of Obstetrics and\nGynaecology Canada 2010;32:S1-S28.\nSomigliana 2006\nSomigliana/uni00A0E, Vigano/uni00A0P, Parazzini/uni00A0F, Stoppelli/uni00A0S, Giambattista/uni00A0E,\nVercellini/uni00A0P. Association between endometriosis and cancer:\nA comprehensive review and a critical analysis of clinical\nand epidemiological evidence. Gynecologic Oncology\n2006;101(2):331-41.\nSpaczynski 2003\nSpaczynski/uni00A0RZ, Duleba/uni00A0AJ. Diagnosis of endometriosis. Seminars\nin Reproductive Medicine 2003;21:193-208.\nStegmann 2008\nStegmann/uni00A0BJ, Sinaii/uni00A0N, Liu/uni00A0S, Segars/uni00A0J, Merino/uni00A0M, Nieman/uni00A0LK,\net al. Using location, color, size, and depth to characterize\nand identify endometriosis lesions in a cohort of 133 women.\nFertility and Sterility 2008;89:1632-6.\nThe Gale Encyclopedia of Medicine 2008\nDonna Olendorf/uni00A0(Editor), Christine Jeryan/uni00A0(Editor), Karen\nBoyden/uni00A0(Editor), Gale Group (Corporate Author). The Gale\nEncyclopedia of Medicine (5 volume set). The Gale Group, Inc,\n2008.\nVercellini 1996\nVercellini/uni00A0P, Trespidi/uni00A0L, De/uni00A0Giorgi/uni00A0O, Cortesi/uni00A0I, Parazzini/uni00A0F,\nCrosignani/uni00A0GP. Endometriosis and pelvic pain: relation\nto disease stage and localization. Fertility and Sterility\n1996;65:299-304.\nVigano 2004\nVigano/uni00A0P, Parazzini/uni00A0F, Somigliana/uni00A0E, Vercellini/uni00A0P. Endometriosis:\nepidemiology and aetiological factors. Best Practice & Research.\nClinical Obstetrics and Gynaecology 2004;18:177-200.\nVitonis 2014\nVitonis/uni00A0AF, Vincent/uni00A0K, Rahmioglu/uni00A0N, Fassbender/uni00A0A, Buck\nLouis/uni00A0GM, Hummelshoj/uni00A0L, et al. WERF EPHect Working Group.\nWorld Endometriosis Research Foundation Endometriosis\nPhenome and biobanking harmonization project: II. Clinical and\ncovariate phenotype data collection in endometriosis research.\nFertility and Sterility 2014;102(5):1223-32.\nWhiting 2005\nWhiting/uni00A0PF, Harbord/uni00A0R, Kleijnen/uni00A0J. No role for quality scores in\nsystematic reviews of diagnostic accuracy studies. BMC Medical\nResearch Methodology 2005;5:19.\nWhiting 2011\nWhiting/uni00A0PF, Rutjes/uni00A0AW, Westwood/uni00A0ME, Mallett/uni00A0S, Deeks/uni00A0JJ,\nReitsma/uni00A0JB, et al. the QUADAS-2 Group. QUADAS-2: A Revised\nTool for the Quality Assessment of Diagnostic Accuracy Studies.\nAnnals of Internal Medicine 2011;155(8):529-36.\nWykes 2004\nWykes/uni00A0CB, Clark/uni00A0TJ, Khan/uni00A0KS. Accuracy of laparoscopy in the\ndiagnosis of endometriosis: a systematic quantitative review.\nBJOG - an International Journal of Obstetrics and Gynaecology\n2004;111:1204-12.\nYeung 2009\nYeung/uni00A0PP/uni00A0Jr, Shwayder/uni00A0J, Pasic/uni00A0RP. Laparoscopic management\nof endometriosis: comprehensive review of best evidence.\nJournal of Minimally Invasive Gynecology 2009;16:269-81.\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n44\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nC H A R A C T E R I S T I C S /uni00A0 O F /uni00A0 S T U D I E S\nCharacteristics of included studies [ordered by study ID]\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To investigate proteins secreted in urine of women with endometriosis us-\ning proteomic techniques in order to identify potential markers for the clinical diagnosis of en-\ndometriosis; to evaluate urinary VDBP in women with endometriosis\nStudy population Women who underwent laparoscopy for various indications including pelvic\nmasses, pelvic pain, suspicious endometriosis, infertility and diagnostic evaluation\nSelection criteria Inclusion criteria: pre-menopausal age; exclusion criteria: previous hormone\nor GnRH agonist use, adenomyosis, endometrial cancer, hyperplasia or endometrial polyps, in-\nfectious diseases, chronic or acute inflammatory diseases, malignancy, autoimmune disease\nand cardiovascular disease\nStudy design Cross-sectional, single-gate design, prospective collection of samples\nPatient characteristics and setting Clinical presentation Pelvic masses, pelvic pain, suspicious endometriosis, infertility\nAg: Mean age 34.22 ± 6.88 years (endometriosis group), 32.76 ± 10.26 years (control group)\nNumber of participants enrolled 95 women\nNumber of participants available for analysis 95 women (in follicular or luteal cycle phase,\nnumbers not specified)\nSetting Gangnam Severance Hospital, Yonsei University College of Medicine\nPlace of study Seoul, Korea\nPeriod of study January 2008 - October 2010\nLanguage English\nIndex tests Index test Urinary VDBP-Cr x serum CA-125\nDetails of the index test procedure as stated Urinary VDBP was measured using specific com-\nmercial sandwich ELISA assays according to manufacturer's protocols (ALPCO Diagnostics,\nSalem, NH, USA); urine VDBP values were normalized to urine Cr concentrations; serum CA-125\nwere measured using CA-125 II electro chemiluminescence immunoassay with the Roche/Hi-\ntachi Modular Analytics E170 system (Roche Diagnostics, Tokyo, Japan); sample handling de-\nscribed\nThreshold for positive result Cut-off value > 2755, not pre-specified\nExaminers: No information provided; unclear if blinded to the result of reference standard\nInterobserver variability Not reported\nTarget condition and reference\nstandard(s)\nTarget condition: Endometriosis\nPrevalence of target condition in the sample: n = 57/95 (60%): stage I-II 5, stage III-IV 52; con-\ntrols n = 38\nReference standard: Laparoscopy and histology\nDescription of positive case definition by reference standard as reported: Visual inspec-\ntion, confirmed by histopathology; staging according to the rASRM classification\nCho 2012/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n45\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nExaminers: No information provided\nFlow and timing Time interval between index test and reference standard: Blood was collected preopera-\ntively, urine was collected after induction of anaesthesia\nWithdrawals: None reported\nComparative /uni00A0\nKey conclusions by the authors Urinary VDBP levels are elevated in women with endometriosis, but they have limited value as\na potential diagnostic biomarker for endometriosis (sensitivity 58%, specificity 76%)\nConflict of interest The authors reported no conflict of interests; supported by the Basic Science Research Pro-\ngram of NRF of Korea by the Ministry of Education, Science and Technology (2010-0023323)\nNotes The reported diagnostic estimates for urinary or blood test alone are presented in other re-\nviews from this series\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random sam-\nple of patients enrolled?\nUnclear /uni00A0 /uni00A0\nDid the study avoid inappropriate\nexclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 2: Index Test All tests\nWere the index test results inter-\npreted without knowledge of the\nresults of the reference standard?\nYes /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nNo /uni00A0 /uni00A0\nWas a cycle phase considered in in-\nterpretation of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was\nconsidered to be a positive result\nof index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test performed by a\nsingle operator or interpreted by\nconsensus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data avail-\nable when the index test results\n/uni00A0 /uni00A0 /uni00A0\nCho 2012/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n46\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nwere interpreted as would be\navailable when the test is used in\npractice?\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards likely to\ncorrectly classify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference standard re-\nsults interpreted without knowl-\nedge of the results of the index\ntests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate interval\nbetween index test and reference\nstandard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same\nreference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nCho 2012/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To evaluate the diagnostic value of serum measurement of IL-6 combined\nwith the presence of nerve fibres in the functional layer of endometrium for diagnosis of mini-\nmal-mild endometriosis\nStudy population Women undergoing laparoscopy for evaluation of infertility or pelvic pain at\nthe authors' institution\nSelection criteria Inclusion criteria: reproductive age (18 - 36 years), follicular cycle phase, reg-\nular menstrual cycle; exclusion criteria: any current infection (genital or systemic), any medica-\ntion within 1/12 prior to laparoscopy, previous surgery for endometriosis, smoking or drinking\nalcohol\nStudy Design Cross-sectional, single-gate design, prospective recruitment and collection of\nsamples\nPatient characteristics and setting Clinical presentation Dysmenorrhoea - 64/114, dyspareunia - 17/114, dyschezia - 6/114, pelvic\npin - 35/114, infertility - 91/114\nAge Mean age 31 ± 1.1 years (endometriosis group), 29 ± 0.6 years (controls)\nNumber of participants enrolled 114 women\nel Sharkwy 2013/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n47\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nNumber of participants available for analysis 78 women (only minimal-mild endometriosis\nincluded; all in follicular cycle phase)\nSetting Department of O&G, Zagazig University Hospital\nPlace of study Zagazig, Egypt\nPeriod of study December 2010 - April 2012\nLanguage English\nIndex tests Index test IL-6 in serum + PGP 9.5 in endometrium\nDetails of the index test procedure as stated Serum IL-6 level assessed using a commercially\navailable ELISA (DRG, Germany); endometrial PGP 9.5 assessed using immunohistochemistry\n(assessment using Olympus microscope, normal skin as positive control, rabbit immunoglobu-\nlin fraction as a negative control); sample processing described\nThreshold for positive result IL-6 > 15.4 pg/ml, PGP 9.5 - presence of nerve fibres in functional\nlayer of endometrium; not pre-specified\nExaminers IL-6 - no information provided; endometrial nerve fibres - experienced gynaecolo-\ngist and two experienced pathologists; unclear if blinded to the result of reference standard\nInterobserver variability Nerve fibre counting - 96% correlations between the two patholo-\ngists\nTarget condition and reference\nstandard(s)\nTarget condition Endometriosis\nPrevalence of target condition in the sample n = 74/114 (65%): stage I-II 38, stage III-IV 36;\ncontrols n = 40\nReference standard Laparoscopy n = 114 (100%)\nDescription of positive case definition by reference standard test as reported Visual in-\nspection; staging according to the rASRM classification\nExaminers Three experienced gynaecologists in endometriosis\nFlow and timing Time interval between index test and reference standard Blood and endometrial samples\nwere obtained on the day of surgery\nWithdrawals 36 participants with moderate-severe disease were not included in final analysis\nComparative /uni00A0\nKey conclusions by the authors Combination of both serum IL-6 and presence of nerve fibres in the endometrium is more reli-\nable method for diagnosis of minimal-mild endometriosis than in single test\nConflict of interest The authors declared no conflict of interest\nNotes The reported separate data for endometrial or blood biomarkers alone are presented in other\nreviews from this series\nThe evaluations were performed only for minimal-mild disease\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nel Sharkwy 2013/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n48\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a consecutive or random sam-\nple of patients enrolled?\nUnclear /uni00A0 /uni00A0\nDid the study avoid inappropriate\nexclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear High\nDOMAIN 2: Index Test All tests\nWere the index test results inter-\npreted without knowledge of the\nresults of the reference standard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nNo /uni00A0 /uni00A0\nWas a cycle phase considered in in-\nterpretation of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was\nconsidered to be a positive result\nof index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test performed by a\nsingle operator or interpreted by\nconsensus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data avail-\nable when the index test results\nwere interpreted as would be\navailable when the test is used in\npractice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards likely to\ncorrectly classify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference standard re-\nsults interpreted without knowl-\nedge of the results of the index\ntests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nel Sharkwy 2013/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n49\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas there an appropriate interval\nbetween index test and reference\nstandard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same\nreference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nNo /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High /uni00A0\nel Sharkwy 2013/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To determine whether the proportion of several leukocyte subsets is modulated in\nthe endometrium of women with endometriosis and, if yes, whether it can be used for diagnostic pur-\nposes\nStudy population Women scheduled to undergo laparoscopy or laparotomy at one of the eight clinical\ninstitutions in Montreal\nSelection criteria Inclusion criteria: women of premenopausal age who had never been pregnant, luteal\nphase of the menstrual cycle (based on the last period and further confirmed by histology), regular cy-\ncles (21 - 35 days), not acute salpingitis, no hormonal treatment or intrauterine device in previous three\nmonths.\nStudy Design Multi-centre study of two-gate design, prospective recruitment, random sample of women\n[participation rate 94%]\nPatient characteristics\nand setting\nClinical presentation Infertility (7% controls, 16% cases); pain (19% controls, 33% cases); pelvic mass\n(8% controls, 13% cases); fibroids (9% controls, 15% cases); menorrhagia (2% controls, 4% cases); tubal\nligation (60% controls, 25% cases); hysterectomy (19% controls, 32% cases); diagnostic laparoscopy\n(20% controls, 43% cases); history of endometriosis (3% controls, 16% cases)\nAge Random sampling from a population with mean age of 37.3 ± 6.4 years\nNumber of participants enrolled 368 women\nNumber of participants available for analysis 368 women (in luteal phase of menstrual cycle)\nSetting Biotech firm - MetrioGene BioSciences (a subsidiary of PROCREA BioSciences)\nPlace of study Montreal, Canada\nPeriod of study July 1997 - May 2001\nLanguage English\nIndex tests Index test: Predictive model including: clinical history (length of menses) + serum CA-125 level + en-\ndometrial leukocytes (CD3+, CD16+, CD3-HLADR-, CD3-CD45RA-, CD3+CD16-, CD3+CD56-, CD56-CD16+,\nCD16b+)\nDetails of the index test procedure as stated Clinical history was collected using a questionnaire in\nwhich a clinical profile as well as information concerning personal habits, menstrual characteristics,\ncrude evaluation of the intensity of pain, and contraception and parity were obtained by the investiga-\ntors; serum CA-125 level determined by using a one step-sandwich radioimmunoassay (Fujirebio Amer-\nica Inc.) with assay sensitivity 0.4 U/ml; endometrial leukocyte subsets determined by Coulter EPICS XL\nGagné 2003/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n50\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nflow cytometer (Beckman/Coulter) with an argon laser operating at 488 nm and detectors at 525, 575,\n610, and 675 nm to measure FITC, PE, ECD, and PerCP emission, respectively; the percentage of cells\nwith the markers of interest was evaluated within the CD45+ cells; sample handling and laboratory pro-\ncedure described in details; predictive model by a multiple logistic regression with subsequent boot-\nstrap method validation by drawing 200 replicate samples with replacement from the original data set\nThreshold for positive result Predictive model - predictive probability 0.61, not pre-specified; CA-125 >\n12.8 U/ml and >35 U/ml; endometrial leukocytes: cut-offs different for each subset selected by ROC, not\npre-specified\nExaminers No information provided; unclear if were blinded to the result of reference standard\nInterobserver variability CA-125: Inter- and intra-assay variations < 5%\nTarget condition and ref-\nerence standard(s)\nTarget condition Endometriosis\nPrevalence of target condition in the sample n = 173/368 (47%): stage I-II 78%, stage III-IV 22%; con-\ntrols n = 195\nReference standard Laparoscopy/Laparotomy n = 368 (100%)\nDescription of positive case definition by reference standard test as reported Cases were defined by\nthe presence of endometriotic lesions confirmed at the time of surgical examination; staging according\nto the ASRM system\nExaminers Gynaecologists collaborating in the study were trained surgeons experienced with the man-\nagement of endometriosis who were skilled in detecting and identifying all forms of endometriotic le-\nsions\nFlow and timing Time interval between index test and reference standard Blood and endometrial samples were ob-\ntained on the day of surgery, clinical data were obtained preoperatively\nWithdrawals None\nComparative /uni00A0\nKey conclusions by the\nauthors\nThe predictive model represents a novel diagnostic tool to identify women with a high likelihood of suf-\nfering from endometriosis\nConflict of interest All the authors except RM are (or were) employees of PROCREA BioSciences; supported by the Industri-\nal Research Assistance Program (IRAP) from NSERC grant #15453Q and internal resources at PROCREA\nBioSciences\nNotes The reported data for blood biomarkers or endometrial test alone are presented in other reviews from\nthis series\nThe reported diagnostic estimates per severity of endometriosis are not presented in this review\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or ran-\ndom sample of patients\nenrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inap-\npropriate exclusions?\nYes /uni00A0 /uni00A0\nGagné 2003/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n51\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a 'two-gate' design\navoided?\nNo /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High High\nDOMAIN 2: Index Test All tests\nWere the index test re-\nsults interpreted without\nknowledge of the results\nof the reference stan-\ndard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used,\nwas it pre-specified?\nNo /uni00A0 /uni00A0\nWas a cycle phase con-\nsidered in interpreta-\ntion of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a\nclear pre-specified defin-\nition of what was consid-\nered to be a positive re-\nsult of index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test per-\nformed by a single opera-\ntor or interpreted by con-\nsensus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical\ndata available when the\nindex test results were\ninterpreted as would be\navailable when the test is\nused in practice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference stan-\ndards likely to correctly\nclassify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference stan-\ndard results interpreted\nwithout knowledge of\nthe results of the index\ntests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nGagné 2003/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n52\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas there an appropriate\ninterval between index\ntest and reference stan-\ndard?\nYes /uni00A0 /uni00A0\nDid all patients receive\nthe same reference stan-\ndard?\nYes /uni00A0 /uni00A0\nWere all patients includ-\ned in the analysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nGagné 2003/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To evaluate the accuracy of CA-19.9 plasma levels (with or without\nCA-125 levels) combined with transvaginal ultrasonography in the differential diagnosis of en-\ndometriosis\nStudy population Women undergoing laparoscopy or laparotomy for persistent adnexal\nmass at the authors' institution\nSelection criteria Inclusion criteria: premenopausal, non-pregnant (only moderate-severe\nendometriosis included)\nStudy Design Cross-sectional, single-gate design, prospective recruitment and collection of\nsamples, consecutive series\nPatient characteristics and setting Clinical presentation Pelvic mass - 100%, infertility - 53%\nAge Mean age 33.3 ± 9.6 years\nNumber of participants enrolled 118 women\nNumber of participants available for analysis 118 women (only moderate-severe en-\ndometriosis included; all in follicular cycle phase)\nSetting Department of O&G, University of Cagliari\nPlace of study Cagliari, Italy\nPeriod of study November 1994 - November 1995\nLanguage English\nIndex tests Index test: CA-19.9 ± CA-125 in serum + Transvaginal Ultrasonography (TVUS)\nDetails of the index test procedure as stated Serum CA-125 levels assessed by immuno-\nradiometric assay (CIS Bio International, Gif sur Yvette, France), limit of detection 0.5 U/ml;\nserum CA-19.9 levels assessed by immunoradiometric assay (CIS Bio International, Gif sur\nYvette, France), limit of detection 1.5 U/ml; TVUS performed with a 5 MHz endovaginal probe\n(Acuson XP/10 OB ultrasound system), procedure described in details\nThreshold for positive result CA-125 ≥ 25 IU/ml, pre-specified; CA-19.9 ≥ 12 U/ml, not pre-\nspecified; TVUS - presence of round, intra-ovarian homogenous, hypoechoic \"tissue,\" with a\nGuerriero 1996a/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n53\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nclear demarcation from the parenchyma and without papillary proliferations - referenced to\nthe primary source, pre-specified\nExaminers For blood assays - no information provided, unclear if were blinded to the results\nof reference standard; for TVUS - same physician blinded to reference standard\nInterobserver variability Intra- and inter-assay CV for CA-125 3.9% and 4.2%; for CA-19.9\n4.6% and 5.3%\nTarget condition and reference\nstandard(s)\nTarget condition Ovarian endometriosis\nPrevalence of target condition in the sample n = 39/118 (33%): all stage III-IV; controls - 79\nReference standard Laparoscopy n = 99/ Laparotomy n = 19 (n = 118, 100%) + histology\nDescription of positive case definition by reference standard test as reported Visual in-\nspection with careful assessment of the ovaries, followed by histopathological diagnosis; sur-\ngical staging according to the rAFS classification\nExaminers No information provided\nFlow and timing Time interval between index test and reference standard Blood was collected on the day\nof surgery, TVUS was performed several days prior\nWithdrawals None\nComparative /uni00A0\nKey conclusions by the authors Transvaginal ultrasonography used alone is the most cost-effective method in the preopera-\ntive differential diagnosis of endometrioma\nConflict of interest Not reported\nNotes The reported diagnostic estimates for blood biomarkers or TVUS alone are presented in other\nreviews from this series\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random sam-\nple of patients enrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inappropriate\nexclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low High\nDOMAIN 2: Index Test All tests\nWere the index test results interpret-\ned without knowledge of the results\nof the reference standard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nNo /uni00A0 /uni00A0\nGuerriero 1996a/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n54\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a cycle phase considered in in-\nterpretation of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was\nconsidered to be a positive result of\nindex test?\nYes /uni00A0 /uni00A0\nWas the index test performed by a\nsingle operator or interpreted by\nconsensus in a joint session?\nYes /uni00A0 /uni00A0\nWere the same clinical data avail-\nable when the index test results\nwere interpreted as would be avail-\nable when the test is used in prac-\ntice?\nUnclear /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards likely to\ncorrectly classify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference standard results\ninterpreted without knowledge of\nthe results of the index tests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate interval\nbetween index test and reference\nstandard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same ref-\nerence standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nGuerriero 1996a/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To assess the role of transvaginal ultrasonography in combination with\nCA-125 plasma levels in diagnosis of endometrioma\nStudy population Women who were submitted to laparoscopy or laparotomy because of\nthe presence of a persistent adnexal mass\nGuerriero 1996b/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n55\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nSelection criteria Inclusion criteria: premenopausal, non-pregnant women\nStudy Design Cross-sectional, single-gate design, prospective recruitment and collection of\nsamples, consecutive series\nPatient characteristics and setting Clinical presentation Adnexal mass\nAge: Age range 20-49 years, mean age not provided\nNumber of enrolled 101 women\nNumber of available for analysis 101 women (only moderate-severe endometriosis includ-\ned; all in follicular cycle phase)\nSetting University Hospital, University of Cagliari\nPlace of study Cagliari, Italy\nPeriod of study November 1993 - October 1994\nLanguage English\nIndex tests Index test CA-125 in serum + Transvaginal Ultrasonography (TVUS)\nDetails of the index test procedure as stated Serum CA-125 levels assessed by immuno-\nradiometric assay (CIS Bio International, Gif sur Yvette, France), limit of detection 0.5 U/ml;\nTVUS performed with a 5 MHz endovaginal probe (Acuson XP/10 OB ultrasound system), pro-\ncedure described in details\nThreshold for positive result CA-125: ≥20 IU/ml ≥25 IU/ml, ≥35 IU/ml pre-specified; TVUS -\npresence of round, homogenous, hypoechoic \"tissue,\" within the ovary - referenced to the\nprimary source, pre-specified\nExaminers For blood assays - no information provided; unclear if were blinded to the results\nof reference standard; for TVUS - same physician blinded to reference standard\nInterobserver variability Intra- and inter-assay CV for CA-125 3.9% and 4.2%\nTarget condition and reference stan-\ndard(s)\nTarget condition Ovarian endometriosis\nPrevalence of target condition in the sample n = 29/101 (29%): all stage III-IV; controls n =\n72\nReference standard Laparoscopy/ Laparotomy (number for each group not reported) +\nhistopathology\nDescription of positive case definition by reference test as reported: Visual inspection\nconfirmed on histopathology; histological criteria reported; surgical procedure described;\nsurgical staging according to the rAFS classification\nExaminers No information provided\nFlow and timing Time interval between index test and reference standard Blood was collected on the day\nof surgery, TVUS was performed several days prior\nWithdrawals None\nComparative /uni00A0\nKey conclusions by the authors Transvaginal ultrasonography used alone has a better predictive capacity in differentiating\nendometrioma from other adnexal masses than combined methods.\nConflict of interest Not reported\nGuerriero 1996b/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n56\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nNotes The reported diagnostic estimates for blood biomarker or TVUS alone are presented in other\nreviews from this series\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random sample\nof patients enrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inappropriate ex-\nclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low High\nDOMAIN 2: Index Test All tests\nWere the index test results interpret-\ned without knowledge of the results\nof the reference standard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nYes /uni00A0 /uni00A0\nWas a cycle phase considered in in-\nterpretation of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was con-\nsidered to be a positive result of in-\ndex test?\nYes /uni00A0 /uni00A0\nWas the index test performed by a\nsingle operator or interpreted by con-\nsensus in a joint session?\nYes /uni00A0 /uni00A0\nWere the same clinical data available\nwhen the index test results were in-\nterpreted as would be available when\nthe test is used in practice?\nUnclear /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 3: Reference Standard\nIs the reference standards likely to\ncorrectly classify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference standard results\ninterpreted without knowledge of the\nresults of the index tests?\nYes /uni00A0 /uni00A0\nGuerriero 1996b/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n57\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate interval be-\ntween index test and reference stan-\ndard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same ref-\nerence standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nGuerriero 1996b/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To evaluate the accuracy of routine clinical examination (per vaginam, PV) com-\nbined with transvaginal sonography (TVS) for presurgical, non-invasive diagnosis of endometriosis\nStudy population Women with suspected endometriosis, who were either referred to the pelvic pain\nclinic or self-referred\nSelection criteria Inclusion criteria: availability of the complete past medical, social, obstetrical and\ngynaecological history and the woman’s consent; exclusion criteria: history of gynaecological cancer,\nprevious surgery for DIE involving rectal surgery or dissection of the POD or RVS or other disease enti-\nties including resection of the bladder or anterior rectal wall and virginity of the woman\nStudy design Cross-sectional, single-gate design, prospective recruitment and collection of samples,\nconsecutive series\nPatient characteristics and\nsetting\nClinical presentation Suspected endometriosis: dysmenorrhoea - 77.5%, dyspareunia - 34.5%, chronic\npelvic pain - 22%, dyschezia - 14.5% or subfertility - 17%\nAge Median age 33 years, range 16–45 years\nNumber of enrolled 223 women\nNumber of available for analysis 200 women (cycle phase not reported)\nSetting Tertiary referral service Villach General Hospital (endometriosis centre)\nPlace of study Villach, Austria; Worthing and Chertsey, UK\nPeriod of study September 2007 - June 2008\nLanguage English\nIndex tests Index test Clinical examination + TVUS\nDetails of the index test procedure as stated Clinical examination included speculum and bimanual\nPV examination; TVUS performed with either a Logic 9 (GE) or a Accuvix XQ (Accuvix) scanner using a 5–\n9 MHz endovaginal transducers, procedure described in details\nDescription of positive case definition by index test as reported Clinical examination - a palpable\nnodularity or stiffened or thickened area or a palpable cystic expansion with topographic-anatomical\nHudelist 2009/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n58\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\ncorrelation to the following sites: le/f_t and right USLs, vagina, RVS, POD, the rectosigmoid and the uri-\nnary bladder (posterior wall); TVUS - endometrioma: presence of a cyst or multiple cysts containing dif-\nfuse low-level echoes, DIE: regular or irregular hypoechogenic nodular structure, cystic mass or hypoe-\nchogenic linear thickening with regular or irregular margins, described for each site (USL, vaginal wall,\nRVS, bladder, rectosigmoid colon); POD complete obliteration: uterus, adnexa and rectosigmoid colon\nadherent, with disappearance of the peritoneal structure, POD incomplete obliteration: peritoneal lim-\nits partially identified with the presence or absence of fluid collection\nExaminers: All combined PV examination and TVS (PV followed by TVS) were performed by one of the\nthree examiners with extensive experience in TVS for DIE (> five years)\nInterobserver variability: Not provided\nTarget condition and ref-\nerence standard(s)\nTarget condition: DIE - separate anatomical sites; ovarian endometriosis\nPrevalence of target condition in the sample Pelvic endometriosis n = 135/200 (67.5%); DIE n =\n64/200 (32%); ovarian endometriosis n = 49/200 (24.5%)\nReference standard Laparoscopy n = 200 (100%) + histopathology\nDescription of positive case definition by reference test as reported Visual inspection confirmed on\nhistopathology, histological criteria specified and referenced to primary source; surgical procedure de-\nscribed in details\nExaminers: No information provided; surgery performed in endometriosis referral centre\nFlow and timing Time interval between index test and reference standard: Both tests were performed within two\nmonths before surgery\nWithdrawals: 23 women were excluded because they did not meet the inclusion criteria\nComparative /uni00A0\nKey conclusions by the au-\nthors\nThe combination of PV and TVS accurately predicts the presence of endometriosis affecting the ovaries,\nvagina, rectum, USL, RVS and POD in women with suspected endometriosis. We suggest the routine\ncombination of PV and TVS as an essential part of the standard primary assessment of pelvic pain\nwomen with suspected endometriosis\nConflict of interest Not reported\nNotes The accuracy estimates for endometrioma and USL are reported by the authors but not presented in\nthe review, because this was a lesion-specific analysis\nThe reported diagnostic performance of vaginal examination alone is not included in this review\nThe diagnostic estimates for bladder endometriosis are reported but not included in the review\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or ran-\ndom sample of patients\nenrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inap-\npropriate exclusions?\nYes /uni00A0 /uni00A0\nHudelist 2009/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n59\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a 'two-gate' design\navoided?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 2: Index Test All tests\nWere the index test re-\nsults interpreted without\nknowledge of the results\nof the reference standard?\nYes /uni00A0 /uni00A0\nIf a threshold was used,\nwas it pre-specified?\nYes /uni00A0 /uni00A0\nWas a cycle phase consid-\nered in interpretation of\nthe result of index test?\nUnclear /uni00A0 /uni00A0\nDid the study provide a\nclear pre-specified defin-\nition of what was consid-\nered to be a positive result\nof index test?\nYes /uni00A0 /uni00A0\nWas the index test per-\nformed by a single opera-\ntor or interpreted by con-\nsensus in a joint session?\nNo /uni00A0 /uni00A0\nWere the same clinical da-\nta available when the in-\ndex test results were inter-\npreted as would be avail-\nable when the test is used\nin practice?\nNo /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards\nlikely to correctly classify\nthe target condition?\nYes /uni00A0 /uni00A0\nWere the reference stan-\ndard results interpreted\nwithout knowledge of the\nresults of the index tests?\nUnclear /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate\ninterval between index\ntest and reference stan-\ndard?\nYes /uni00A0 /uni00A0\nHudelist 2009/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n60\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nDid all patients receive the\nsame reference standard?\nYes /uni00A0 /uni00A0\nWere all patients included\nin the analysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nHudelist 2009/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To evaluate clinical examination during menstruation and plasma CA-125 con-\ncentration to diagnose endometriosis\nStudy population Women scheduled for laparoscopy for suspected endometriosis\nSelection criteria Exclusion criteria: hormonal treatment or medical treatment for endometriosis\nin the three months preceding laparoscopy, refusal for a clinical examination during menstruation\n(only DIE considered)\nStudy design Cross-sectional single-gate, prospective recruitment and collection of samples, con-\nsecutive series\nPatient characteristics and\nsetting\nClinical presentation Infertility (n = 33), pain (n = 13), infertility + pain (n = 6), hydrosalpinx (n = 1),\novarian cyst (n= 2)\nAge Range 20 - 45 years (personal communication with the author)\nNumber of participants enrolled 61 women\nNumber of participants available for analysis 55 women (only DIE or endometrioma or severe\npelvic adhesions included; all in menstrual, follicular and early luteal phase of menstrual cycle)\nSetting Division of endoscopic surgery, University Hospital Gasthiusberg, University of Leuven\nPlace of study Leuven, Belgium\nPeriod of study Not stated\nLanguage English\nIndex tests Index test Clinical examination in menstrual phase (menstrual nodularities) + CA-125 in mid follic-\nular phase\nDetails of the index test procedure as stated Clinical examination included pelvic bimanual ex-\namination with careful assessment of pelvic nodularities and their tenderness; CA-125 assay using\na second generation IRMA kit (CA-125 II, Centocor, Malvern, Pa; all kits from the same production\nbatch)\nThreshold for positive result Clinical examination - presence of induration (with one or more\nsmall nodules) or painful nodularities (larger spherical nodule), pre-specified and described in de-\ntails; CA-125 >35 U/ml, not pre-specified\nExaminers Clinical examination: one of the two authors, always preoperatively and hence blinded\nto the results of reference standard; CA-125 - no information provided; unclear if were blinded to\nreference standard\nInterobserver variability CA-125: intra-and inter-assay variation < 5% and < 8%\nKoninckx 1996/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n61\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTarget condition and refer-\nence standard(s)\nTarget condition DIE, ovarian endometrioma or severe pelvic adhesions\nPrevalence of target condition in the sample n = 38/55 (69%): stage I-II 29, stage III-IV 9; DIE 13,\nendometrioma 9, deep endometriosis + severe cul de sac adhesions + endometrioma 24; controls n\n= 17)\nReference standard Laparoscopy n = 55 (100%)\nDescription of positive case definition by reference standard test as reported Visual inspection,\ndeep endometriosis classified as type I - III, reference to the source with diagnostic criteria and de-\nscribed; staging according to the rAFS classification .\nExaminers Not stated; unclear if were blinded to the result of pelvic examination\nFlow and timing Time interval between index test and reference standard The tests were performed within four\nmonths before surgery (personal communication with the author)\nWithdrawals In six women (11%) the surgery was cancelled for various reasons\nComparative /uni00A0\nKey conclusions by the au-\nthors\nClinical examination during menstruation can diagnose reliably deep endometriosis, cystic ovari-\nan endometriosis or cul de sac adhesions. This test, preferentially combined with a follicular phase\nCA-125 assay, should be used to decide whether a preparation for bowel surgery should be given\nConflict of interest Not reported\nNotes The reported diagnostic estimates for blood biomarker alone are presented in other reviews from\nthis series\nThe presented diagnostic estimates are for DIE or ovarian endometrioma or severe cul de sac adhe-\nsions\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random\nsample of patients enrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inappro-\npriate exclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoid-\ned?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low High\nDOMAIN 2: Index Test All tests\nWere the index test results in-\nterpreted without knowledge\nof the results of the reference\nstandard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it\npre-specified?\nNo /uni00A0 /uni00A0\nKoninckx 1996/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n62\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a cycle phase considered\nin interpretation of the result\nof index test?\nYes /uni00A0 /uni00A0\nDid the study provide a clear\npre-specified definition of\nwhat was considered to be a\npositive result of index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test performed\nby a single operator or inter-\npreted by consensus in a joint\nsession?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data\navailable when the index test\nresults were interpreted as\nwould be available when the\ntest is used in practice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards like-\nly to correctly classify the tar-\nget condition?\nYes /uni00A0 /uni00A0\nWere the reference standard\nresults interpreted without\nknowledge of the results of the\nindex tests?\nUnclear /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate in-\nterval between index test and\nreference standard?\nYes /uni00A0 /uni00A0\nDid all patients receive the\nsame reference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in\nthe analysis?\nNo /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High /uni00A0\nKoninckx 1996/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To compare the performance of history and examination findings combined\nwith transvaginal ultrasound 'so/f_t marker' evaluation of ovarian mobility, for the prediction of fixed\novaries secondary to endometriosis at laparoscopy\nMarasinghe 2014/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n63\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nStudy population Women scheduled for laparoscopy for chronic pelvic pain and or subfertility\nSelection criteria Exclusion criteria: previous surgical diagnosis of endometriosis or pelvic adhe-\nsions, confirmed diagnosis of genital abnormalities, those who did not consent to vaginal examina-\ntions\nStudy design Cross-sectional single-gate, prospective recruitment and collection of samples, con-\nsecutive series\nPatient characteristics and\nsetting\nClinical presentation Infertility 83%, chronic pelvic pain 17%, dysmenorrhoea 46%, dyspareunia\n31%\nAge Mean age 33.3 ± 5.1 years, range 22 - 48 years\nNumber of participants enrolled 110 women\nNumber of participants available for analysis 106 women\nSetting University Gynecology unit, National Hospital of Sri Lanka, tertiary referral centre\nPlace of study Colombo, Sri Lanka\nPeriod of study December 2009 - March 2010\nLanguage English\nIndex tests Index test Clinical history + examination + TVUS\nDetails of the index test procedure as stated Clinical history - interview with history of dysmen-\norrhoea and dyspareunia, severity was assessed using a visual analogue scale ranging 1-10 with\na score of one considered as 'no pain'; examination included pelvic bimanual examination to de-\ntect the presence of pelvic tenderness, a fixed retroverted uterus, tender uterosacral ligaments and\ndeeply infiltrating nodules on the uterosacral ligaments or in the cul-de-sac; TVUS - Assessment\nwas with gentle pressure on the ovary with the transvaginal probe, fixed or mobile ovaries were di-\nagnosed by assessing the ovaries in relation to the uterus and ipsilateral internal iliac vessels\nThreshold for positive result Clinical history - dysmenorrhoea or dyspareunia; examination - at\nleast one of the abovementioned examination features; TVUS: at least one 'fixed' ovary\nExaminers Clinical history and examination: one clinician with > four years experience in gynaecol-\nogy; TVUS - single examiner with > four years experience in ultrasound blinded to the clinical data\nInterobserver variability Each tests was performed by a single operator\nTarget condition and refer-\nence standard(s)\nTarget condition Endometriosis and other peri-ovarian adhesions\nPrevalence of target condition in the sample n = 32/106 (30%): stage I-II 19, stage III-IV 13; other\nperi-ovarian adhesions n = 5; controls n = 17)\nReference standard Laparoscopy n = 106 (100%) ± histology\nDescription of positive case definition by reference standard test as reported Visual inspection,\nendometriotic adhesions causing a fixed ovary - defined as ovaries fixed to the uterus or ovarian\nfossa and when it could not be elevated from the ovarian fossa by using a blunt probe; histology\nwas considered in selected cases, histological criteria provided; staging according to the rAFS clas-\nsification .\nExaminers Single examiner with 15 years of laparoscopic experience, blinded to the results of in-\ndex test\nFlow and timing Time interval between index test and reference standard The tests were performed within two\nweeks before surgery\nMarasinghe 2014/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n64\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWithdrawals In four women at least one ovary could not be visualised on transvaginal scanning\nand they were excluded\nComparative /uni00A0\nKey conclusions by the au-\nthors\nA combination of clinical and transvaginal ultrasound based 'so/f_t marker' of ovarian mobility pro-\nvides a valid method for identifying fixed ovaries secondary to endometriosis\nConflict of interest Not reported\nNotes The primary outcome of the study was endometriosis with fixed ovaries or other peri-ovarian adhe-\nsions\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random\nsample of patients enrolled?\nYes /uni00A0 /uni00A0\nDid the study avoid inappro-\npriate exclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoid-\ned?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 2: Index Test All tests\nWere the index test results in-\nterpreted without knowledge\nof the results of the reference\nstandard?\nYes /uni00A0 /uni00A0\nIf a threshold was used, was it\npre-specified?\nYes /uni00A0 /uni00A0\nWas a cycle phase considered\nin interpretation of the result\nof index test?\n/uni00A0 /uni00A0 /uni00A0\nDid the study provide a clear\npre-specified definition of\nwhat was considered to be a\npositive result of index test?\nYes /uni00A0 /uni00A0\nWas the index test performed\nby a single operator or inter-\npreted by consensus in a joint\nsession?\nYes /uni00A0 /uni00A0\nWere the same clinical data\navailable when the index test\nresults were interpreted as\nYes /uni00A0 /uni00A0\nMarasinghe 2014/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n65\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nwould be available when the\ntest is used in practice?\n/uni00A0 /uni00A0 Low Low\nDOMAIN 3: Reference Standard\nIs the reference standards like-\nly to correctly classify the tar-\nget condition?\nYes /uni00A0 /uni00A0\nWere the reference standard\nresults interpreted without\nknowledge of the results of the\nindex tests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate in-\nterval between index test and\nreference standard?\nYes /uni00A0 /uni00A0\nDid all patients receive the\nsame reference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in\nthe analysis?\nNo /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High /uni00A0\nMarasinghe 2014/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To develop a test to discriminate between women suffering from pelvic pain\nassociated with presence or absence of endometriosis, using symptom visual analogue scale\n(VAS) scores, demographic and lifestyle factors and known and novel plasma biomarkers\nStudy population Women undergoing laparoscopy for evaluation of chronic pelvic pain, dys-\nmenorrhoea, or dyspareunia\nSelection criteria Exclusion criteria: women on current hormonal therapy, failure to complete\nquestionnaire,\nStudy design Cross-sectional single-gate, prospective collection of samples\nPatient characteristics and set-\nting\nClinical presentation Pelvic pain, dysmenorrhoea, dyspareunia\nAge Mean age 27 years, range 18 - 44 years (endometriosis group) and 30 years, range 19 - 43\nyears (controls)\nNumber of participants enrolled 172 women\nNumber of participants available for analysis 101 women (in menstrual, proliferative or secre-\ntory cycle phase)\nPaiva 2014/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n66\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nSetting Department of O&G, Royal Women's Hospital, University of Melbourne\nPlace of study Melbourne, Australia\nPeriod of study May 2006 - February 2009\nLanguage English\nIndex tests Index test Predictive model including: clinical history (parity, ever had IUD, hx of endometriosis,\nalcohol intake, dyspareunia) + serum CA-125\nDetails of the index test procedure as stated Clinical history obtained with a questionnaire in-\ncluding demographic details, LMP, risk factors for endometriosis and symptom severity on form\nof visual analogue scale (VAS); serum CA-125 was measured using 2-plex magnetic human circu-\nlating cancer biomarker panel kit (Millipore, USA), detection limit 0.26 pg/ml; laboratory meth-\nods and sample processing described; predictive model was constructed using multivariate lo-\ngistic regression model (initial selection of 14 parameters)\nThreshold for positive result Not reported\nExaminers No information provided; unclear if blinded to the results of reference standard\nInterobserver variability CA-125: the intra- and inter-assay CV < 10%\nTarget condition and reference\nstandard(s)\nTarget condition Endometriosis\nPrevalence of target condition in the sample n = 69/101 (68%): stage I-II 45, stage III-IV 24; con-\ntrols n = 32\nReference standard Laparoscopy n = 101 (100%) + histopathology\nDescription of positive case definition by reference standard test as reported Visual inspec-\ntion confirmed by histological demonstration of endometrial glands and stroma; staging ac-\ncording to the rASRM classification\nExaminers No information provided\nFlow and timing Time interval between index test and reference standard Blood samples and questionnaire\nwere obtained preoperatively on the same day\nWithdrawals 71 participants were excluded: 16 due to current hormone treatment, 31 - not\ncompleted questionnaire, 24 - no samples available due to laboratory freezer failure\nComparative /uni00A0\nKey conclusions by the authors Combining symptom scores, historical measures and CA-125 provides a reasonable means\nto discriminate between women with pelvic pain associated with presence or absence of en-\ndometriosis, but greater specificity is needed before such a model could replace laparoscopy\nConflict of interest The authors declared no conflict of interests; the study was supported by several research grants\nNotes The predictive model included history of endometriosis, hence can be used for primary diagno-\nsis as well as for assessment of recurrence\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nPaiva 2014/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n67\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a consecutive or random\nsample of patients enrolled?\nUnclear /uni00A0 /uni00A0\nDid the study avoid inappropriate\nexclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 2: Index Test All tests\nWere the index test results inter-\npreted without knowledge of the\nresults of the reference standard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it\npre-specified?\nNo /uni00A0 /uni00A0\nWas a cycle phase considered in\ninterpretation of the result of in-\ndex test?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was\nconsidered to be a positive result\nof index test?\nYes /uni00A0 /uni00A0\nWas the index test performed by\na single operator or interpreted\nby consensus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data avail-\nable when the index test results\nwere interpreted as would be\navailable when the test is used in\npractice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards like-\nly to correctly classify the target\ncondition?\nYes /uni00A0 /uni00A0\nWere the reference standard re-\nsults interpreted without knowl-\nedge of the results of the index\ntests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nPaiva 2014/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n68\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas there an appropriate interval\nbetween index test and reference\nstandard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same\nreference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nPaiva 2014/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To validate and investigate the clinical value of urinary enolase I (NNE) in\nwomen with endometriosis.\nStudy population Women who underwent laparoscopy for diagnostic evaluation of pelvic\nmasses, pelvic pain, suspicious endometriosis and infertility\nSelection criteria Exclusion criteria: postmenopausal status, previous use of hormone or go-\nnadotropin-releasing hormone agonist, adenomyosis, endometrial cancer or hyperplasia, en-\ndometrial polyps, infectious diseases, chronic or acute inflammatory diseases, malignancy, au-\ntoimmune disease and cardiovascular disease\nStudy design Cross-sectional, single-gate design, prospective collection of samples\nPatient characteristics and setting Clinical presentation: Pelvic masses, pelvic pain, suspicious endometriosis and infertility\nAge: Mean age 31.48 ± 6.30 for endometriosis group, 29.35 ± 6.87 for control group\nNumber of enrolled: 59 women\nNumber of available for analysis: 59 women (in follicular or luteal cycle phase, numbers not\nspecified; only moderate/severe endometriosis)\nSetting: Gongnam Severance Hospital, Yonsei University College of Medicine\nPlace of study: Seoul, Republic of Korea\nPeriod of study: January 2009 - December 2011\nLanguage: English\nIndex tests Index test Urinary enolase I (NNE-Cr) + serum CA-125\nDescription of positive case definition by index test as reported Urinary enolase I was mea-\nsured with commercial ELISA according to the manufacturer’s protocols (USCN Life Science\n& Technology Company, TX) with minimal detectable concentration of 0.312 ng/ml; urinary\nNNE values were normalised to urinary Cr concentrations; serum CA-125 was measured using\nCA-125 II electro chemiluminescence immunoassay with Roche/Hitachi Modular Analytics E170\nsystem (Roche Diagnostics, Japan); sample handling described\nThreshold for positive result Cut-off value > 27.23, not pre-specified\nExaminers: No information provided, unclear if were blinded to the results of reference stan-\ndard\nYun 2014/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n69\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nInterobserver variability Not provided\nTarget condition and reference\nstandard(s)\nTarget condition Endometriosis\nPrevalence of target condition in the sample n = 39/59 (55%): all stage III-IV; controls n = 20\nReference standard: Laparoscopy and histology\nDescription of positive case definition by reference test as reported Visual inspection con-\nfirmed on histopathology; staging according to the rASRM classification\nExaminers Not information provided\nFlow and timing Time interval between index test and reference standard Blood was collected preoperative-\nly, urine was collected after induction of anaesthesia\nWithdrawals No withdrawals reported\nComparative /uni00A0\nKey conclusions by the authors Urinary enolase I was significantly increased in women with endometriosis, though the find-\nings undermine its capacity as a diagnostic marker. May have potential as one of the combined\nmarkers.\nConflict of interest The authors declared no conflict of interest; supported by the Basic Science Research Program\nthrough the National Research Foundation of Korea (NRF) funded by the Ministry of Education,\nScience and Technology (2010-0023323)\nNotes The reported diagnostic estimates for urine or blood test only are presented in other reviews\nfrom this series\nOnly information for severe disease available\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random sam-\nple of patients enrolled?\nUnclear /uni00A0 /uni00A0\nDid the study avoid inappropriate\nexclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear High\nDOMAIN 2: Index Test All tests\nWere the index test results inter-\npreted without knowledge of the\nresults of the reference standard?\nYes /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nNo /uni00A0 /uni00A0\nYun 2014/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n70\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWas a cycle phase considered in in-\nterpretation of the result of index\ntest?\nYes /uni00A0 /uni00A0\nDid the study provide a clear pre-\nspecified definition of what was\nconsidered to be a positive result\nof index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test performed by a\nsingle operator or interpreted by\nconsensus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data avail-\nable when the index test results\nwere interpreted as would be\navailable when the test is used in\npractice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Low\nDOMAIN 3: Reference Standard\nIs the reference standards likely to\ncorrectly classify the target condi-\ntion?\nYes /uni00A0 /uni00A0\nWere the reference standard re-\nsults interpreted without knowl-\nedge of the results of the index\ntests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate interval\nbetween index test and reference\nstandard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same\nreference standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the\nanalysis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nYun 2014/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nStudy characteristics\nPatient sampling Primary objectives To evaluate the diagnostic value of examining endometrial biopsy\nspecimens for aromatase cytochrome P450 and CA-125 for endometriosis\nZeng 2005/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n71\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nStudy population Women undergoing laparoscopy or laparotomy for pelvic pain or infer-\ntility\nSelection criteria Inclusion criteria: reproductive age regular menstrual cycle; exclusion\ncriteria: hormonal treatment for 3/12 prior reproductive age, preoperative diagnosis of\nuterine fibroids, adenomyosis.\nStudy design Cross-sectional single-gate, prospective collection of samples\nPatient characteristics and setting Clinical presentation Infertility or pelvic pain\nAge Mean age 33 ± 4 years, range 26 - 40 years (endometriosis), 32 ± 4 years, range 25-39\nyears (controls)\nNumber of participants enrolled 58 women\nNumber of participants available for analysis 58 women (31 women in follicular and 27\nwomen in luteal cycle phase)\nSetting Department of O&G, Third Xiangya Hospital, Central South University\nPlace of study Changsha, China\nPeriod of study March 2003 - February 2004\nLanguage Chinese\nIndex tests Index test CA-125 in serum + P450 aromatase in endometrium\nDetails of the index test procedure as stated Serum CA-125 was determined by chemilu-\nminescence assay; endometrial aromatase protein was evaluated by immunohistochem-\nistry, 2nd generation assay (ElivisionTM plus kit), positive IHC staining indicated by pres-\nence of brown particles within the cytoplasm; sample handling or laboratory technique not\ndescribed\nThreshold for positive result CA-125 >35 U/ml, not pre-specified; P450 aromatase: posi-\ntive or negative test\nExaminers No information provided, unclear if blinded to the result of reference standard\nInterobserver variability Not stated\nTarget condition and reference stan-\ndard(s)\nTarget condition Endometriosis\nPrevalence of target condition in the sample n = 36/58 (62%): stage I-II 20, stage III-IV 16;\ncontrols n = 22\nReference standard Laparoscopy/Laparotomy n = 58 (100%)\nDescription of positive case definition by reference standard test as reported Visual in-\nspection; staging according to rAFS classification\nExaminers Not stated\nFlow and timing Time interval between index test and reference standard Blood and endometrial sam-\nples were collected on the day of surgery\nWithdrawals None\nComparative /uni00A0\nKey conclusions by the authors The combination assay of aromatase cytochrome P450 in eutopic endometrium and\nCA-125 can be used as a diagnostic test for endometriosis, especially for the early stage of\nendometriosis, which is superior to the assay of CA-125\nZeng 2005/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n72\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nConflict of interest Not reported\nNotes Translated from Chinese, some information may have been misinterpreted\nThe reported diagnostic estimates for endometrial or blood markers alone are presented\nin other reviews from this series\nMethodological quality\nItem Authors' judgement Risk of bias Applicability concerns\nDOMAIN 1: Patient Selection\nWas a consecutive or random sample\nof patients enrolled?\nUnclear /uni00A0 /uni00A0\nDid the study avoid inappropriate ex-\nclusions?\nYes /uni00A0 /uni00A0\nWas a 'two-gate' design avoided? Yes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 2: Index Test All tests\nWere the index test results interpret-\ned without knowledge of the results of\nthe reference standard?\nUnclear /uni00A0 /uni00A0\nIf a threshold was used, was it pre-\nspecified?\nNo /uni00A0 /uni00A0\nWas a cycle phase considered in inter-\npretation of the result of index test?\nUnclear /uni00A0 /uni00A0\nDid the study provide a clear pre-speci-\nfied definition of what was considered\nto be a positive result of index test?\n/uni00A0 /uni00A0 /uni00A0\nWas the index test performed by a sin-\ngle operator or interpreted by consen-\nsus in a joint session?\n/uni00A0 /uni00A0 /uni00A0\nWere the same clinical data available\nwhen the index test results were inter-\npreted as would be available when the\ntest is used in practice?\n/uni00A0 /uni00A0 /uni00A0\n/uni00A0 /uni00A0 High Unclear\nDOMAIN 3: Reference Standard\nIs the reference standards likely to cor-\nrectly classify the target condition?\nUnclear /uni00A0 /uni00A0\nZeng 2005/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n73\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nWere the reference standard results in-\nterpreted without knowledge of the re-\nsults of the index tests?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Unclear Low\nDOMAIN 4: Flow and Timing\nWas there an appropriate interval be-\ntween index test and reference stan-\ndard?\nYes /uni00A0 /uni00A0\nDid all patients receive the same refer-\nence standard?\nYes /uni00A0 /uni00A0\nWere all patients included in the analy-\nsis?\nYes /uni00A0 /uni00A0\n/uni00A0 /uni00A0 Low /uni00A0\nZeng 2005/uni00A0/uni00A0(Continued)\n(r)AFS: (revised) American Fertility Society; (r)ASRM: (revised) American Society for Reproductive Medicine; CA-125: cancer antigen; CV:\ncoefficient of variation; DIE: deep infiltrating endometriosis; ECD: Electron Coupled Dye; ELISA: enzyme-linked immunosorbent assay;\nFITC: fluorescein isothiocyanate; GnRH: gonadotropin-releasing hormone; IL: interleukin; IUD: intrauterine device; LMP: last menstrual\nperiod; NNE: non neuronal enolase;PE: phycoerythrin; PerCP: peridinin; chlorophyll protein; PGP: permeability glycoprotein; POD: pouch\nof Douglas; PV: per vaginam; RVS: rectovaginal septum; TVS: transvaginal sonography; TVUS: transvaginal ultrasound; USL: uterosacral\nligament; VAS: visual analogue scale; VDBP: vitamin-D-binding protein; VDBPCr: VDBP level corrected for creatinine.\n/uni00A0\nCharacteristics of excluded studies [ordered by study ID]\n/uni00A0\nStudy Reason for exclusion\nAbrao 2007 Index test outside inclusion criteria (no combined diagnostic estimates available)\nAdamyan 1993 Index test outside inclusion criteria (no combined diagnostic estimates available)\nAlcazar 2011 Index test outside inclusion criteria (no combined diagnostic estimates available)\nBadawy 1984 Index test outside inclusion criteria (no combined diagnostic estimates available)\nBazot 2009 Index test outside inclusion criteria (no combined diagnostic estimates available)\nBorboletto 1995 Review article\nCho 2007 Index test outside inclusion criteria (no combined diagnostic estimates available)\nda Silva 2014 Index test outside inclusion criteria (no combined diagnostic estimates available)\nDias 2012 Index test outside inclusion criteria (no combined diagnostic estimates available)\nEskenazi 2001 Index test outside inclusion criteria (no combined diagnostic estimates available)\nFedele 1988 Index test outside inclusion criteria (no combined diagnostic estimates available)\nGuerriero 1997 Index test outside inclusion criteria (only 'lesion-level' analysis for imaging component of the test)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n74\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nStudy Reason for exclusion\nHudelist 2011b Index test outside inclusion criteria (no combined diagnostic estimates available)\nKocbek 2014 Index test outside inclusion criteria (no combined diagnostic estimates available)\nKuessel 2014 Index test outside inclusion criteria (no combined diagnostic estimates available)\nLee 2014 Index test outside inclusion criteria (no combined diagnostic estimates available)\nNezhat 1994 Index test outside inclusion criteria (no combined diagnostic estimates available)\nSzubert 2014 Insufficient diagnostic test accuracy information (unable to construct 2 x 2 tables)\nWeerakiet 2000 Study design, population and index test outside inclusion criteria (retrospective selection of cases;\npostmenopausal women were included; only 'lesion-level' analysis)\nWolfler 2005 Insufficient diagnostic test accuracy information (unable to construct 2 x 2 tables)\nYong 2013 Target condition outside inclusion criteria (not endometriosis but \"abnormal laparoscopy in\nwomen with pelvic pain\")\n/uni00A0\n/uni00A0\nD A T A\nPresented below are all the data for all of the tests entered into the review.\n/uni00A0\nTable Tests. /uni00A0 Data tables by test\nTest No. of studies No. of participants\n1 IL-6 (>15.4 pg/ml) [serum] + PGP 9.5 [endometrium] 1 78\n2 CA-125 [serum] (>35 U/ml) + P450 aromatase [endometrium] 1 58\n3 VDBP-Cr [urine] x CA-125 [serum] (>2755) 1 95\n4 NNE_Cr [urine] + CA-125 [serum] (>27.23) 1 59\n5 Hx (dysmenorrhoea, dyspareunia) + PV examination + TVUS (fixed ovary) 1 106\n6 Hx (length of menses) + CA-125 [serum] (>35 U/ml) + leukocytes [endometri-\num]\n1 368\n7 Hx (parity, past IUD, past endometriosis, alcohol intake, dyspareunia) +\nCA-125 [serum]\n1 101\n8 PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for\nDIE, endometrioma or severe adhesions\n1 41\n9 PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for\nDIE, endometrioma or severe adhesions\n1 41\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n75\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTest No. of studies No. of participants\n10 PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for\nDIE\n1 30\n11 PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for\nDIE\n1 30\n12 PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for\nendometrioma\n1 26\n13 PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for\nendometrioma\n1 26\n14 TVUS + CA-125 [serum] (≥25 U/ml) + CA-19.9 [serum] (≥12 U/ml) 1 118\n15 TVUS + (CA-125 [serum] (≥25 U/ml) OR CA-19.9 [serum] (≥12 U/ml)) 1 118\n16 TVUS + CA-19.9 [serum] (≥12 U/ml) 1 118\n17 TVUS OR CA-19.9 [serum] (≥12 U/ml) 1 118\n18 TVUS + CA-125 [serum] (≥20 U/ml) 1 101\n19 TVUS OR CA-125 [serum] (≥20 U/ml) 1 101\n20 TVUS + CA-125 [serum] (≥25 U/ml) 1 101\n21 TVUS OR CA-125 [serum] (≥25 U/ml) 1 101\n22 TVUS + CA-125 [serum] (≥35 U/ml) 1 101\n23 TVUS OR CA-125 [serum] (≥35 U/ml) 1 101\n24 PV examination + TVUS for POD obliteration 1 200\n25 PV examination + TVUS for vaginal endometriosis 1 200\n26 PV examination + TVUS for RVS endometriosis 1 200\n27 PV examination + TVUS for rectal endometriosis 1 200\n/uni00A0\n/uni00A0\nTest 1. /uni00A0 IL-6 (>15.4 pg/ml) [serum] + PGP 9.5 [endometrium].\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n76\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTest 2. /uni00A0 CA-125 [serum] (>35 U/ml) + P450 aromatase [endometrium].\n/uni00A0\n/uni00A0\nTest 3. /uni00A0 VDBP-Cr [urine] x CA-125 [serum] (>2755).\n/uni00A0\n/uni00A0\nTest 4. /uni00A0 NNE_Cr [urine] + CA-125 [serum] (>27.23).\n/uni00A0\n/uni00A0\nTest 5. /uni00A0 Hx (dysmenorrhoea, dyspareunia) + PV examination + TVUS (fixed ovary).\n/uni00A0\n/uni00A0\nTest 6. /uni00A0 Hx (length of menses) + CA-125 [serum] (>35 U/ml) + leukocytes [endometrium].\n/uni00A0\n/uni00A0\nTest 7. /uni00A0 Hx (parity, past IUD, past endometriosis, alcohol intake, dyspareunia) + CA-125 [serum].\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n77\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0\n/uni00A0\nTest 8. /uni00A0 PV examination (menstrual nodularities) + CA-125\n[serum] (>35 IU/ml) for DIE, endometrioma or severe adhesions.\n/uni00A0\n/uni00A0\nTest 9. /uni00A0 PV examination (menstrual nodularities) OR CA-125\n[serum] (>35 IU/ml) for DIE, endometrioma or severe adhesions.\n/uni00A0\n/uni00A0\nTest 10. /uni00A0 PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for DIE.\n/uni00A0\n/uni00A0\nTest 11. /uni00A0 PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for DIE.\n/uni00A0\n/uni00A0\nTest 12. /uni00A0 PV examination (menstrual nodularities) + CA-125 [serum] (>35 IU/ml) for endometrioma.\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n78\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTest 13. /uni00A0 PV examination (menstrual nodularities) OR CA-125 [serum] (>35 IU/ml) for endometrioma.\n/uni00A0\n/uni00A0\nTest 14. /uni00A0 TVUS + CA-125 [serum] (≥25 U/ml) + CA-19.9 [serum] (≥12 U/ml).\n/uni00A0\n/uni00A0\nTest 15. /uni00A0 TVUS + (CA-125 [serum] (≥25 U/ml) OR CA-19.9 [serum] (≥12 U/ml)).\n/uni00A0\n/uni00A0\nTest 16. /uni00A0 TVUS + CA-19.9 [serum] (≥12 U/ml).\n/uni00A0\n/uni00A0\nTest 17. /uni00A0 TVUS OR CA-19.9 [serum] (≥12 U/ml).\n/uni00A0\n/uni00A0\nTest 18. /uni00A0 TVUS + CA-125 [serum] (≥20 U/ml).\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n79\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0\n/uni00A0\nTest 19. /uni00A0 TVUS OR CA-125 [serum] (≥20 U/ml).\n/uni00A0\n/uni00A0\nTest 20. /uni00A0 TVUS + CA-125 [serum] (≥25 U/ml).\n/uni00A0\n/uni00A0\nTest 21. /uni00A0 TVUS OR CA-125 [serum] (≥25 U/ml).\n/uni00A0\n/uni00A0\nTest 22. /uni00A0 TVUS + CA-125 [serum] (≥35 U/ml).\n/uni00A0\n/uni00A0\nTest 23. /uni00A0 TVUS OR CA-125 [serum] (≥35 U/ml).\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n80\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nTest 24. /uni00A0 PV examination + TVUS for POD obliteration.\n/uni00A0\n/uni00A0\nTest 25. /uni00A0 PV examination + TVUS for vaginal endometriosis.\n/uni00A0\n/uni00A0\nTest 26. /uni00A0 PV examination + TVUS for RVS endometriosis.\n/uni00A0\n/uni00A0\nTest 27. /uni00A0 PV examination + TVUS for rectal endometriosis.\n/uni00A0\n/uni00A0\nA D D I T I O N A L /uni00A0 T A B L E S\n/uni00A0\nDepthLocation of \nendometriosis\nExtent\n< 1 cm 1-3 cm > 3 cm\nSuperficial 1 2 4Peritoneum\nDeep 2 4 6\nR Superficial 1 2 4\nDeep 4 16 20\nOvary\nL Superficial 1 2 4\nTable 1. /uni00A0 Staging of endometriosis, rASRM classification/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n81\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nDeep 4 16 20\nPartial CompletePosterior Cul de sac Obliteration\n4 40\nAdhesions < 1/3 Enclosure 1/3-2/3 Enclo-\nsure\n> 2/3 Enclosure\nR Filmy 1 2 4\nDense 4 8 16\nL Filmy 1 2 4\nOvary\nDense 4 8 16\nR Filmy 1 2 4\nDense 4* 8* 16\nL Filmy 1 2 4\nTube\nDense 4* 8* 16\n* If the fimbriated end of the fallopian tube is completely enclosed, change the point assignment to 16\nAmerican Society for Reproductive Medicine 1997\nTable 1. /uni00A0 Staging of endometriosis, rASRM classification/uni00A0/uni00A0(Continued)\n/uni00A0\n/uni00A0\nN Test\n1 IL-6 (>15.4 pg/ml) [serum] + PGP 9.5 [endometrium]\n2 CA-125 [serum] (>35 U/ml) + P450 aromatase [endometrium]\n3 VDBP-Cr [urine] x CA-125 [serum] (>2755)\n4 NNE_Cr [urine] + CA-125 [serum] (>27.23)\n5 History (dysmenorrhoea, dyspareunia) + PV examination + TVUS (fixed ovary)\n6 History (length of menses) + CA-125 [serum] (>35 U/ml) + leukocytes [endometrium]\n7 History (parity, past IUD, past endometriosis, alcohol intake, dyspareunia) + CA-125 [serum]\n8 PV examination (menstrual nodularities) + CA125 (>35 U/ml) [serum]\n9 PV examination (menstrual nodularities) OR CA125 (>35 U/ml) [serum]\n10 TVUS + CA-125 [serum] (≥25 U/ml) + CA-19.9 [serum] (≥12 U/ml)\n11 TVUS + (CA-125 [serum] (≥25 U/ml) OR CA-19.9 [serum] (≥12 U/ml))\nTable 2. /uni00A0 Combination of the non-invasive tests for endometriosis evaluated in this review/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n82\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n12 TVUS + CA-19.9 [serum] (≥12 U/ml))\n13 TVUS OR CA-19.9 [serum] (≥12 U/ml))\n14 TVUS + CA-125 [serum] (≥20 U/ml; ≥25 U/ml; ≥35 U/ml)\n15 PV examination + TVUS\nTable 2. /uni00A0 Combination of the non-invasive tests for endometriosis evaluated in this review/uni00A0/uni00A0(Continued)\nCA-125: cancer antigen; IL: interleukin; IUD: intrauterine device;NNE: non neuronal enolase;PV: per vaginam; TVUS: transvaginal\nultrasound; VDBP: vitamin-D-binding protein\n/uni00A0\n/uni00A0\nDomain 1 - Patient selection\nDescription Describe methods of patient selection and included women\nType of bias assessed Selection bias, spectrum bias\nReview Question Women of reproductive age with clinically suspected endometriosis (symptoms, clinical examina-\ntion ± presence of pelvic mass), scheduled for surgical exploration of pelvic/abdominal cavity for\nconfirmation of the diagnosis ± treatment\nInformaton collected Study objectives, study population, selection (inclusion/ exclusion criteria), study design, clinical\npresentation, age, number of enrolled and number of available for analysis, setting, place and peri-\nod of the study\nSignalling question 1 Was a consecutive or random sample of patients enrolled?\nYes If a consecutive sample or a random sample of the eligible participants was included in the study\nNo If non-consecutive sample or non-random sample of the eligible participants was included in the\nstudy\nUnclear If this information was unclear\nSignalling question 2 Did the study avoid inappropriate exclusions?\nYes If inclusion/exclusion criteria were presented and all women with suspected endometriosis were\nincluded, with an exception for those who a) had a history of medical conditions or were on med-\nical therapy that would have potentially interfered with interpretation of index test (e.g. malig-\nnancy, pregnancy, autoimmune disorders, infectious diseases, treatment with hormonal or im-\nmunomodulator substances); b) refused to participate in the study; or c) were unfit for surgery\nNo If the study excluded the participants based on education level, psychosocial factors, genetic test-\ning or phenotype or excluded participants with any co-morbidities commonly present in general\npopulation, including a population that could have undergone a testing for endometriosis in clini-\ncal setting (hypertension, asthma, obesity, benign gastro-intestinal or renal disease, etc)\nUnclear If the study did not provide clear definition of the selection (inclusion/exclusion) criteria and 'no'\njudgement was not applicable\nSignalling question 3 Was a 'two-gate' design avoided?\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n83\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nYes If the study had a single set of inclusion criteria, defined by the clinical presentation (i.e. only par-\nticipants in whom the target condition is suspected) - a ‘single-gate design’.\nNo If the study had more than one set of inclusion criteria in respect to clinical presentation (i.e. partic-\nipants suspected of target condition and participants with alternative diagnosis in whom the target\ncondition would not be suspected in clinical practice) - a 'two-gate' study design\nUnclear If it was unclear whether a 'two-gate design' was avoided or not\nRisk of bias Could the selection of patients have introduced bias?\nHigh If 'no' classification for any of the above three questions\nLow If 'yes' classification for all the above three questions\nUnclear If 'unclear' classification for three of the above questions\nConcerns about applicability Are there concerns that the included patients do not match the review question?\nHigh If the study population differed from the population defined in the review question in terms of de-\nmographic features and co-morbidity (e.g. studies with multiple sets of inclusion criteria with re-\nspect to clinical presentation including either healthy controls or alternative diagnosis controls\nthat would not have undergone index test in real practice). Further, if target condition diagnosed\nin the study population was not representative of the entire spectrum of disease, such as limited\nspectrum of severity (e.g. only mild forms) or limited type of endometriosis (e.g. only DIE)\nLow If the study includes only clinically relevant population that would have undergone index test in re-\nal practice and includes representative form of target condition\nUnclear If this information was unclear (e.g. severity of endometriosis was not reported)\nDomain 2 - Index test\nDescription Describe the index test, how it was conducted and interpreted\nType of bias assessed Test review bias, clinical review bias, interobserver variation bias\nReview Question Any type of test that combines several different testing modalities with and without clinical history\nor examination\nInformaton collected Index test name, description of positive case definition by index test as reported, threshold for pos-\nitive result, examiners (number, level of expertise, blinding), interobserver variability, conflict of in-\nterests\nSignalling question 1 Were the index test results interpreted without knowledge of the results of the reference stan-\ndard?\nYes If the operators performing/interpreting index test were unaware of the results of reference stan-\ndard\nNo If the operators performing/interpreting index test were not blinded to the results of reference\nstandard\nUnclear If this information was unclear\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n84\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nSignalling question 2 If a threshold was used, was it pre-specified? [only for the studies that included biomarker test-\ning]\nYes If study clearly provided a threshold for positive result and was defined before execution/interpre-\ntation of index test\nNo If a threshold for positive result was not provided or not defined prior to test execution\nUnclear If it was unclear whether a threshold was pre-specified or not\nSignalling question 3 Was a menstrual cycle phase considered in interpreting the index test? [only for the studies that\nincluded biomarker testing]\nYes If all the included participants were in the same phase of menstrual cycle or if the study reported\nsubgroup analyses per cycle phase or if study reported the pooled estimates after impact of the cy-\ncle phase on biomarker expression was not detected\nNo If study included participants in different phases of menstrual cycle, but effect of cycle phase on in-\ndex test was not assessed\nUnclear If the cycle phase was not reported\nSignalling question 4 Did the study provide a clear pre-specified definition of what was considered to be a “positive”\nresult of index test? [only for the studies that included imaging modalities]\nYes If study provided clear definition of positive findings and this was defined before execution/inter-\npretation of index test\nNo If definition of the positive result was not provided or if study described the findings derived from\nthe index test and not defined prior to its execution\nUnclear If it was unclear whether the criteria were pre-specified or not\nSignalling question 5 Was the index test performed by a single operator or interpreted by consensus in a joint session?\n[only for the studies that included imaging modalities]\nYes If test was performed/interpreted either by single operator or interpreted after collegial discussion\nof the case\nNo If test was performed/interpreted by various operators in different participants\nUnclear If this information was unclear\nSignalling question 6 Were the same clinical data available when the index test results were interpreted as would be\navailable when the test is used in practice? [only for the studies that included imaging modali-\nties]\nYes If operators performing/interpreting the test were aware of suspected endometriosis or of the clin-\nical history, but were not aware of the results of other imaging tests or of previous diagnosis of en-\ndometriosis, including the results of previous surgeries\nNo If operators performing/interpreting the test were informed on previously or recently surgically di-\nagnosed endometriosis or were not blinded to the results of other imaging tests or tests raising sus-\npicion for endometriosis\nUnclear If this information was unclear\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n85\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nRisk of bias Could the conduct or interpretation of the index test have introduced bias?\nHigh If 'no' classification for any of the first three questions [for the studies that included biomarker test-\ning] or if 'no' classification for any of the following: signalling questions 1, 4, 5, 6 [for studies that in-\ncluded imaging modalities]\nLow If 'yes' classification for all the relevant questions: signalling questions 1 - 3 [for the studies that\nincluded biomarker testing] or signalling questions 1, 4, 5, 6 [for studies that included imaging\nmodalities]\nUnclear If 'unclear' classification for any of the relevant questions\nConcerns about applicability Are there concerns that the index test, its conduct, or interpretation differ from the review\nquestion?\nHigh We did not consider the studies where index tests other than combinations of different testing\nmodalities were included or where index test looked at other target conditions not specified in the\nreview (e.g. studies aimed at classifying pelvic masses as benign and malignant), therefore none of\nthe included studies was classified as 'high concern'\nLow We considered all types of combinations of different testing modalities as eligible, therefore all the\nincluded studies were classified as 'low concern', unless 'unclear' judgement was applicable\nUnclear If study reported, but did not present sufficient information on any of the following: laboratory\nmethod, sample handling, reagents used, radiological protocol or equipment (where applicable),\nexperience of the test operators\nDomain 3 - Reference standard\nDescription Describe the reference standard, how it was conducted and interpreted\nType of bias assessed Verification bias, bias in estimation of diagnostic accuracy due to inadequate reference standard\nReview Question Target condition - pelvic endometriosis, ovarian endometriosis, DIE. Reference standard - visuali-\nsation of endometriosis at surgery (laparoscopy or laparotomy) with or without histological confir-\nmation\nInformaton collected Target condition, prevalence of target condition in the sample, reference standard, description of\npositive case definition by reference test as reported, examiners (number, level of expertise, blind-\ning)\nSignalling question 1 Is the reference standards likely to correctly classify the target condition?\nYes If the study reported at least one of the following: surgical procedure was described in sufficient\ndetails or criteria for positive reference standard were stated or diagnosis was confirmed by\nhistopathology or the procedure was performed by the team with high level of expertise in diagno-\nsis/surgical treatment of target condition, including tertiary referral centres for endometriosis\nNo If reference standard did not classify target condition correctly; considering the inclusion criteria\nand a nature of the reference standard, none of the studies were classified as 'no' for this item\nUnclear If information on execution of the reference standard, its interpretation or operators was unclear\nSignalling question 2 Were the reference standard results interpreted without knowledge of the results of the index\ntests?\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n86\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nYes If operators performing the reference test were unaware of the results of index test\nNo If operators performing the reference test were aware of the results of index test\nUnclear If this information was unclear\nRisk of bias Could the reference standard, its conduct, or its interpretation have introduced bias?\nHigh If 'no' classification for any of the above two questions\nLow If 'yes' classification for all the above two questions The Robinson Institute, University of Adelaide\nUnclear If 'unclear' classification for any of the above two questions and 'high risk' judgement was not ap-\nplicable\nConcerns about applicability Are there concerns that the target condition as defined by the reference standard does not\nmatch the question?\nHigh We excluded the studies where participants did not undergo surgery for diagnosis of endometrio-\nsis, therefore none of the included studies were classified as 'high concern'\nLow Considering the inclusion criteria, all the studies were classified as 'low concern', therefore all the\nincluded studies were classified as 'low concern', unless 'unclear' judgement was applicable\nUnclear Only studies were laparoscopy/ laparotomy served as a reference test were included; therefore\nnone of the included studies was classified as 'unclear concern'\nDomain 4 - Flow and timing\nDescription Describe any participants who did not receive the index tests or reference standard or who were ex-\ncluded from the 2 x 2 table, describe the interval and any interventions between index tests and the\nreference standard\nType of bias assessed Disease progression bias, bias of diagnostic performance due to missing data\nReview Question Less than 12 months interval between index test and reference standard - endometriosis may\nprogress over the time, so we had chosen an arbitrary time interval of '12' months as an acceptable\ntime interval between the index test and surgical confirmation of diagnosis\nInformaton collected Time interval between index test and reference standard, withdrawals (overall number of reported\nand if were explained)\nSignalling question 1 Was there an appropriate interval between index test (sample collection) and reference stan-\ndard?\nYes If time interval was reported and was less than 12 months\nNo We excluded all the studies where time interval was longer than 12 months, therefore none of the\nincluded studies were classified as 'no' for this item\nUnclear if time interval was not stated clearly, but authors description allowed to assume that the interval\nwas reasonably short\nSignalling question 2 Did all patients receive the same reference standard?\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n87\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nYes If all participants underwent laparoscopy/laparotomy as a reference standard. Considering the in-\nclusion criteria, all the studies were classified as 'yes' for this item, as anticipated\nNo If all participants did not undergo surgery or had alternative reference standard or if only a subset\nof participants had surgery as reference standard, but the information on this population was not\navailable in isolation\nUnclear If this information was unclear. Considering the inclusion criteria, none of the included studies\nwere classified as 'unclear' for this item\nSignalling question 3 Were all patients included in the analysis?\nYes If all the participants were included in the analysis or if the participants were excluded because\nthey did not meet inclusion criteria prior to execution of index test or if the withdrawals were less\nthan 5% of the enrolled population (arbitrary selected cut-off)\nNo If any participants were excluded from the analysis because of un interpretable results, inability to\nundergo either index test or reference standard or unclear reasons\nUnclear If this information was unclear\nRisk of bias Could the patient flow have introduced bias?\nHigh If 'no' classification for any of the above three questions\nLow If 'yes' classification for all the above three questions\nUnclear If 'unclear' classification for any of the above three questions and 'high risk' judgement was not ap-\nplicable\nTable 3. /uni00A0 Risk of bias and applicability judgments for the quality assessment of diagnostic accuracy studies\n(QUADAS-2)/uni00A0/uni00A0(Continued)\nDIE: deep infiltrating endometriosis\n/uni00A0\n/uni00A0\nA P P E N D I C E S\nAppendix 1. Biomarkers search strategy for CENTRAL (OVID platform)\nDatabase: EBM Reviews - Cochrane Central Register of Controlled Trials <July 2015 (3.09.2015)>\n1 (biomarker$ or marker$).tw. (23692)\n2 Laboratory Test$.tw. (2793)\n3 growth factor$.tw. (5448)\n4 scatter factor$.tw. (8)\n5 cytokine$.tw. (6264)\n6 hepatocyte growth factor.tw. (111)\n7 (FGF or fibroblast growth factor$).tw. (433)\n8 (PDGF or platelet derived growth factor$).tw. (250)\n9 (EGF or epidermal growth factor$).tw. (1077)\n10 (IGF-I or insulin-like growth factor$ or IGF1).tw. (2132)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n88\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n11 (TGF-a or transforming growth factor alfa or TGFa).tw. (519)\n12 (TGF-b or transforming growth factor beta or TGFb).tw. (236)\n13 (EGFR or epidermal growth factor receptor$).tw. (1905)\n14 (VEGF or vascular endothelial growth factor$).tw. (1532)\n15 exp Luteinizing Hormone/bl [Blood] (151)\n16 leptin$.tw. (1399)\n17 exp Progesterone/bl [Blood] (58)\n18 Proteolytic enzyme$.tw. (136)\n19 exp matrix metalloproteinase 1/ or exp matrix metalloproteinase 2/ or exp matrix metalloproteinase 3/ or exp matrix metalloproteinase\n9/ (292)\n20 matrix metalloproteinase$.tw. (676)\n21 MMP$.tw. (905)\n22 TIMP$.tw. (229)\n23 exp \"tissue inhibitor of metalloproteinase-1\"/ or exp \"tissue inhibitor of metalloproteinase-2\"/ (101)\n24 exp Glycoproteins/ (10108)\n25 (Ca-125 or Ca125 or cancer antigen 125).tw. (305)\n26 (Ca-19-9 or Ca19-9 or cancer antigen 19-9).tw. (71)\n27 (PP 14 or PP14).tw. (23)\n28 serum placental protein$.tw. (6)\n29 exp Follistatin/ (13)\n30 Osteopontin$.tw. (80)\n31 exp intercellular adhesion molecule-1/ or exp selectins/ (929)\n32 soluble intercellular adhesion.tw. (256)\n33 Soluble adhesion molecule$.tw. (89)\n34 sICAM.tw. (319)\n35 sVCAM$.tw. (223)\n36 (sEcadherin or soluble E-cadherin).tw. (4)\n37 (sEselectin or soluble E-selectin).tw. (99)\n38 exp t-lymphocytes/ or exp natural killer t-cells/ (2645)\n39 Immune cells alteration$.tw. (1)\n40 (T helper$ or T supressor$ or T helper$ T supressor$ ratio).tw. (445)\n41 Total complement level$.tw. (0)\n42 Autoantibodies.tw. (428)\n43 exp Antibodies, Antiphospholipid/ (85)\n44 Anti-endometrial.tw. (0)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n89\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n45 Antiphospholipid$.tw. (152)\n46 exp hla antigens/ or exp hla-a1 antigen/ or exp hla-a2 antigen/ (563)\n47 (HLA or human leucocyte antigen$).tw. (1724)\n48 Anti-laminin-1.tw. (0)\n49 Anti-thyroid.tw. (49)\n50 Anti-Thomsen Friedenreich antigen$.tw. (0)\n51 Anti-transferrin.tw. (0)\n52 Anti-LDL.tw. (3)\n53 (Anti-2HSG or Heremans-Schmidt glycoprotein).tw. (0)\n54 interleukin$.tw. (7276)\n55 (MCP-I or monocyte chemoattractant protein-I).tw. (0)\n56 (MIF or migration inhibitory factor$).tw. (75)\n57 (TNF-a or tumour necrosis factor$ alfa).tw. (3923)\n58 Fas ligand$.tw. (47)\n59 Endometrial marker$.tw. (2)\n60 CAMs.tw. (53)\n61 cell adhesion molecule$.tw. (568)\n62 exp Integrins/ (781)\n63 Integrin$.tw. (248)\n64 Selectin$.tw. (2183)\n65 Cadherin$.tw. (71)\n66 Aromatase P450.tw. (3)\n67 estrogen receptor$.tw. (1252)\n68 progesterone receptor$.tw. (531)\n69 MTMMP$.tw. (0)\n70 cyr61.tw. (1)\n71 exp Cysteine-Rich Protein 61/ (1)\n72 cysteine-rich heparin-binding protein$.tw. (0)\n73 (ANXA 1 or ANXA1).tw. (3)\n74 (Annexin 1 or Annexin1).tw. (2)\n75 (PGP 9?5 or PGP9?5 or protein gene product$).tw. (18)\n76 serum marker$.tw. (411)\n77 neural marker$.tw. (9)\n78 cell surface marker$.tw. (46)\n79 inflammatory marker$.tw. (1739)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n90\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n80 microarray$.tw. (501)\n81 microRNA$.tw. (103)\n82 proteomic$.tw. (176)\n83 genomic$.tw. (526)\n84 (endometri$ adj2 biops$).tw. (464)\n85 Follistatin$.tw. (26)\n86 Vascular Endothelial Growth Factor A/ (560)\n87 Vitamin D-Binding Protein/ (18)\n88 exp Cytokines/ (13960)\n89 exp interleukins/ or exp interleukin-1/ or exp interleukin-6/ or exp interleukin-8/ or exp interleukin-12/ or exp interleukin-13/ (4413)\n90 exp Epidermal Growth Factor/ (91)\n91 exp Fibroblast Growth Factors/ (197)\n92 Platelet-Derived Growth Factor/ (99)\n93 Keratin-19/ (19)\n94 exp Clinical Laboratory Techniques/ (35164)\n95 (Luteinizing Hormone$ or LH).tw. (2935)\n96 cytokeratin-19.tw. (25)\n97 (VDBP or vitamin D-binding protein$).tw. (44)\n98 urinary peptide$.tw. (8)\n99 VDBP-Cr.tw. (0)\n100 urinary VDBP corrected for creatinine expression.tw. (0)\n101 urinary marker$.tw. (67)\n102 or/1-101 (90390)\n103 Endometriosis/di [Diagnosis] (6)\n104 102 or 103 (90394)\n105 exp Endometriosis/ (469)\n106 Endometrio$.tw. (1026)\n107 105 or 106 (1067)\n108 104 and 107 (226)\n109 (animals not (humans and animals)).sh. (1)\n110 108 not 109 (226)\nAppendix 2. Biomarkers search strategy for MEDLINE (OVID platform)\nDatabase: MEDLINE (Ovid) <1946 to February, week 2 2015 (16.2.2015)>\n1 (biomarker$ or marker$).tw. (605002)\n2 Laboratory Test$.tw. (29839)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n91\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n3 growth factor$.tw. (272049)\n4 scatter factor$.tw. (1287)\n5 cytokine$.tw. (250618)\n6 hepatocyte growth factor.tw. (8053)\n7 (FGF or fibroblast growth factor$).tw. (31798)\n8 (PDGF or platelet derived growth factor$).tw. (19864)\n9 (EGF or epidermal growth factor$).tw. (58069)\n10 (IGF-I or insulin-like growth factor$ or IGF1).tw. (43539)\n11 (TGF-a or transforming growth factor alfa or TGFa).tw. (281)\n12 (TGF-b or transforming growth factor beta or TGFb).tw. (28842)\n13 (EGFR or epidermal growth factor receptor$).tw. (41719)\n14 (VEGF or vascular endothelial growth factor$).tw. (53588)\n15 exp Luteinizing Hormone/bl [Blood] (24587)\n16 leptin$.tw. (24994)\n17 exp Progesterone/bl [Blood] (18412)\n18 Proteolytic enzyme$.tw. (9768)\n19 exp matrix metalloproteinase 1/ or exp matrix metalloproteinase 2/ or exp matrix metalloproteinase 3/ or exp matrix metalloproteinase\n9/ (22968)\n20 matrix metalloproteinase$.tw. (34522)\n21 MMP$.tw. (44439)\n22 TIMP$.tw. (10777)\n23 exp \"tissue inhibitor of metalloproteinase-1\"/ or exp \"tissue inhibitor of metalloproteinase-2\"/ (6146)\n24 exp Glycoproteins/ (637149)\n25 (Ca-125 or Ca125 or cancer antigen 125).tw. (6761)\n26 (Ca-19-9 or Ca19-9 or cancer antigen 19-9).tw. (4194)\n27 (PP 14 or PP14).tw. (229)\n28 serum placental protein$.tw. (33)\n29 exp Follistatin/ (1134)\n30 Osteopontin$.tw. (6769)\n31 exp intercellular adhesion molecule-1/ or exp selectins/ (25302)\n32 soluble intercellular adhesion.tw. (1588)\n33 Soluble adhesion molecule$.tw. (779)\n34 sICAM.tw. (2258)\n35 sVCAM$.tw. (1277)\n36 (sEcadherin or soluble E-cadherin).tw. (95)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n92\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n37 (sEselectin or soluble E-selectin).tw. (689)\n38 exp t-lymphocytes/ or exp natural killer t-cells/ (272580)\n39 Immune cells alteration$.tw. (1)\n40 (T helper$ or T supressor$ or T helper$ T supressor$ ratio).tw. (21275)\n41 Total complement level$.tw. (23)\n42 Autoantibodies.tw. (33457)\n43 exp Antibodies, Antiphospholipid/ (7522)\n44 Anti-endometrial.tw. (23)\n45 Antiphospholipid$.tw. (9974)\n46 exp hla antigens/ or exp hla-a1 antigen/ or exp hla-a2 antigen/ (64462)\n47 (HLA or human leucocyte antigen$).tw. (80501)\n48 Anti-laminin-1.tw. (33)\n49 Anti-thyroid.tw. (1414)\n50 Anti-Thomsen Friedenreich antigen$.tw. (6)\n51 Anti-transferrin.tw. (275)\n52 Anti-LDL.tw. (181)\n53 (Anti-2HSG or Heremans-Schmidt glycoprotein).tw. (3)\n54 interleukin$.tw. (175195)\n55 (MCP-I or monocyte chemoattractant protein-I).tw. (44)\n56 (MIF or migration inhibitory factor$).tw. (4479)\n57 (TNF-a or tumour necrosis factor$ alfa).tw. (1344)\n58 Fas ligand$.tw. (6032)\n59 Endometrial marker$.tw. (11)\n60 CAMs.tw. (1756)\n61 cell adhesion molecule$.tw. (20903)\n62 exp Integrins/ (44414)\n63 Integrin$.tw. (39960)\n64 Selectin$.tw. (55426)\n65 Cadherin$.tw. (20780)\n66 Aromatase P450.tw. (180)\n67 estrogen receptor$.tw. (38819)\n68 progesterone receptor$.tw. (16623)\n69 MTMMP$.tw. (7)\n70 cyr61.tw. (559)\n71 exp Cysteine-Rich Protein 61/ (386)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n93\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n72 cysteine-rich heparin-binding protein$.tw. (9)\n73 (ANXA 1 or ANXA1).tw. (313)\n74 (Annexin 1 or Annexin1).tw. (339)\n75 (PGP 9?5 or PGP9?5 or protein gene product$).tw. (2096)\n76 serum marker$.tw. (5429)\n77 neural marker$.tw. (925)\n78 cell surface marker$.tw. (4456)\n79 inflammatory marker$.tw. (10916)\n80 microarray$.tw. (75404)\n81 microRNA$.tw. (29731)\n82 proteomic$.tw. (45292)\n83 genomic$.tw. (190985)\n84 (endometri$ adj2 biops$).tw. (3411)\n85 Follistatin$.tw. (1663)\n86 Vascular Endothelial Growth Factor A/ (35738)\n87 Vitamin D-Binding Protein/ (1282)\n88 exp Cytokines/ (547522)\n89 exp interleukins/ or exp interleukin-1/ or exp interleukin-6/ or exp interleukin-8/ or exp interleukin-12/ or exp interleukin-13/ (188479)\n90 exp Epidermal Growth Factor/ (21298)\n91 exp Fibroblast Growth Factors/ (25075)\n92 Platelet-Derived Growth Factor/ (11030)\n93 Keratin-19/ (1090)\n94 exp Clinical Laboratory Techniques/ (2132820)\n95 (Luteinizing Hormone$ or LH).tw. (56679)\n96 cytokeratin-19.tw. (1469)\n97 (VDBP or vitamin D-binding protein$).tw. (1158)\n98 urinary peptide$.tw. (137)\n99 VDBP-Cr.tw. (1)\n100 urinary VDBP corrected for creatinine expression.tw. (1)\n101 urinary marker$.tw. (638)\n102 or/1-101 (4086291)\n103 Endometriosis/di [Diagnosis] (3354)\n104 102 or 103 (4088946)\n105 exp Endometriosis/ (17244)\n106 Endometrio$.tw. (21492)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n94\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n107 105 or 106 (24940)\n108 104 and 107 (10490)\n109 (animals not (humans and animals)).sh. (3892900)\n110 108 not 109 (10113)\nAdditional search February 2015 - May 2015\nOvid MEDLINE(R) In-Process & Other Non-Indexed Citations, Ovid MEDLINE(R) Daily and Ovid MEDLINE(R) <1946 to Present (3.9.2015)>\n1 (biomarker$ or marker$).tw. (652345)\n2 Laboratory Test$.tw. (31389)\n3 growth factor$.tw. (287701)\n4 scatter factor$.tw. (1326)\n5 cytokine$.tw. (267766)\n6 hepatocyte growth factor.tw. (8585)\n7 (FGF or fibroblast growth factor$).tw. (33674)\n8 (PDGF or platelet derived growth factor$).tw. (20842)\n9 (EGF or epidermal growth factor$).tw. (61625)\n10 (IGF-I or insulin-like growth factor$ or IGF1).tw. (45386)\n11 (TGF-a or transforming growth factor alfa or TGFa).tw. (306)\n12 (TGF-b or transforming growth factor beta or TGFb).tw. (30559)\n13 (EGFR or epidermal growth factor receptor$).tw. (46446)\n14 (VEGF or vascular endothelial growth factor$).tw. (58203)\n15 exp Luteinizing Hormone/bl [Blood] (24870)\n16 leptin$.tw. (26783)\n17 exp Progesterone/bl [Blood] (18699)\n18 Proteolytic enzyme$.tw. (9992)\n19 exp matrix metalloproteinase 1/ or exp matrix metalloproteinase 2/ or exp matrix metalloproteinase 3/ or exp matrix metalloproteinase\n9/ (24504)\n20 matrix metalloproteinase$.tw. (37055)\n21 MMP$.tw. (47849)\n22 TIMP$.tw. (11419)\n23 exp \"tissue inhibitor of metalloproteinase-1\"/ or exp \"tissue inhibitor of metalloproteinase-2\"/ (6447)\n24 exp Glycoproteins/ (662211)\n25 (Ca-125 or Ca125 or cancer antigen 125).tw. (7058)\n26 (Ca-19-9 or Ca19-9 or cancer antigen 19-9).tw. (4399)\n27 (PP 14 or PP14).tw. (232)\n28 serum placental protein$.tw. (34)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n95\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n29 exp Follistatin/ (1180)\n30 Osteopontin$.tw. (7267)\n31 exp intercellular adhesion molecule-1/ or exp selectins/ (26225)\n32 soluble intercellular adhesion.tw. (1663)\n33 Soluble adhesion molecule$.tw. (795)\n34 sICAM.tw. (2374)\n35 sVCAM$.tw. (1360)\n36 (sEcadherin or soluble E-cadherin).tw. (97)\n37 (sEselectin or soluble E-selectin).tw. (713)\n38 exp t-lymphocytes/ or exp natural killer t-cells/ (284378)\n39 Immune cells alteration$.tw. (1)\n40 (T helper$ or T supressor$ or T helper$ T supressor$ ratio).tw. (22494)\n41 Total complement level$.tw. (24)\n42 Autoantibodies.tw. (35161)\n43 exp Antibodies, Antiphospholipid/ (7759)\n44 Anti-endometrial.tw. (22)\n45 Antiphospholipid$.tw. (10351)\n46 exp hla antigens/ or exp hla-a1 antigen/ or exp hla-a2 antigen/ (66724)\n47 (HLA or human leucocyte antigen$).tw. (83856)\n48 Anti-laminin-1.tw. (33)\n49 Anti-thyroid.tw. (1478)\n50 Anti-Thomsen Friedenreich antigen$.tw. (8)\n51 Anti-transferrin.tw. (284)\n52 Anti-LDL.tw. (183)\n53 (Anti-2HSG or Heremans-Schmidt glycoprotein).tw. (3)\n54 interleukin$.tw. (184697)\n55 (MCP-I or monocyte chemoattractant protein-I).tw. (46)\n56 (MIF or migration inhibitory factor$).tw. (4718)\n57 (TNF-a or tumour necrosis factor$ alfa).tw. (1428)\n58 Fas ligand$.tw. (6204)\n59 Endometrial marker$.tw. (11)\n60 CAMs.tw. (1823)\n61 cell adhesion molecule$.tw. (22033)\n62 exp Integrins/ (46487)\n63 Integrin$.tw. (42447)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n96\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n64 Selectin$.tw. (58540)\n65 Cadherin$.tw. (22688)\n66 Aromatase P450.tw. (182)\n67 estrogen receptor$.tw. (41210)\n68 progesterone receptor$.tw. (17437)\n69 MTMMP$.tw. (7)\n70 cyr61.tw. (620)\n71 exp Cysteine-Rich Protein 61/ (425)\n72 cysteine-rich heparin-binding protein$.tw. (9)\n73 (ANXA 1 or ANXA1).tw. (355)\n74 (Annexin 1 or Annexin1).tw. (358)\n75 (PGP 9?5 or PGP9?5 or protein gene product$).tw. (2190)\n76 serum marker$.tw. (5721)\n77 neural marker$.tw. (1026)\n78 cell surface marker$.tw. (4751)\n79 inflammatory marker$.tw. (12244)\n80 microarray$.tw. (81764)\n81 microRNA$.tw. (35967)\n82 proteomic$.tw. (49911)\n83 genomic$.tw. (205064)\n84 (endometri$ adj2 biops$).tw. (3518)\n85 Follistatin$.tw. (1762)\n86 Vascular Endothelial Growth Factor A/ (38477)\n87 Vitamin D-Binding Protein/ (1356)\n88 exp Cytokines/ (575020)\n89 exp interleukins/ or exp interleukin-1/ or exp interleukin-6/ or exp interleukin-8/ or exp interleukin-12/ or exp interleukin-13/ (197567)\n90 exp Epidermal Growth Factor/ (21875)\n91 exp Fibroblast Growth Factors/ (26259)\n92 Platelet-Derived Growth Factor/ (11355)\n93 Keratin-19/ (1179)\n94 exp Clinical Laboratory Techniques/ (2203416)\n95 (Luteinizing Hormone$ or LH).tw. (57796)\n96 cytokeratin-19.tw. (1538)\n97 (VDBP or vitamin D-binding protein$).tw. (1262)\n98 urinary peptide$.tw. (148)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n97\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n99 VDBP-Cr.tw. (1)\n100 urinary VDBP corrected for creatinine expression.tw. (1)\n101 urinary marker$.tw. (679)\n102 or/1-101 (4283825)\n103 Endometriosis/di [Diagnosis] (3449)\n104 102 or 103 (4286552)\n105 exp Endometriosis/ (17833)\n106 Endometrio$.tw. (22478)\n107 105 or 106 (26003)\n108 104 and 107 (10936)\n109 (animals not (humans and animals)).sh. (4004321)\n110 108 not 109 (10539)\n111 (201501$ or 201502$ or 201503$ or 201504$).ed. (322721)\n112 110 and 111 (215)\nAppendix 3. Biomarkers search strategy for EMBASE (OVID platform)\nDatabase: EMBASE (Ovid) <1980 to 2015 Week 07 (16.02.2015)>\n1 Laboratory Test$.tw. (41662)\n2 growth factor$.tw. (318593)\n3 scatter factor$.tw. (1388)\n4 cytokine$.tw. (322134)\n5 hepatocyte growth factor.tw. (9594)\n6 (FGF or fibroblast growth factor$).tw. (37191)\n7 (PDGF or platelet derived growth factor$).tw. (23530)\n8 (EGF or epidermal growth factor$).tw. (69553)\n9 (IGF-I or insulin-like growth factor$ or IGF1).tw. (49806)\n10 (TGF-a or transforming growth factor alfa or TGFa).tw. (542)\n11 (TGF-b or transforming growth factor beta or TGFb).tw. (30820)\n12 (EGFR or epidermal growth factor receptor$).tw. (64664)\n13 (VEGF or vascular endothelial growth factor$).tw. (73191)\n14 exp luteinizing hormone/ec [Endogenous Compound] (21924)\n15 leptin$.tw. (32576)\n16 exp progesterone blood level/ or exp progesterone urine level/ (6285)\n17 Proteolytic enzyme$.tw. (9643)\n18 exp matrix metalloproteinase/ (19364)\n19 matrix metalloproteinase$.tw. (41445)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n98\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n20 MMP$.tw. (58466)\n21 TIMP$.tw. (14174)\n22 exp \"tissue inhibitor of metalloproteinase 2\"/ (4824)\n23 exp \"tissue inhibitor of metalloproteinase 1\"/ (8779)\n24 exp glycoprotein/ec [Endogenous Compound] (246077)\n25 (Ca-125 or Ca125 or cancer antigen 125).tw. (9536)\n26 (Ca-19-9 or Ca19-9 or cancer antigen 19-9).tw. (6054)\n27 (PP 14 or PP14).tw. (244)\n28 serum placental protein$.tw. (43)\n29 exp follistatin/ (2148)\n30 Osteopontin$.tw. (8475)\n31 exp intercellular adhesion molecule 1/ (32066)\n32 exp selectin/ (3082)\n33 soluble intercellular adhesion.tw. (1788)\n34 Soluble adhesion molecule$.tw. (919)\n35 sICAM.tw. (2888)\n36 sVCAM$.tw. (1793)\n37 (sEcadherin or soluble E-cadherin).tw. (120)\n38 (sEselectin or soluble E-selectin).tw. (822)\n39 exp T lymphocyte/ (374675)\n40 exp natural killer T cell/ (5800)\n41 Immune cells alteration$.tw. (6)\n42 (T helper$ or T supressor$ or T helper$ T supressor$ ratio).tw. (24786)\n43 Total complement level$.tw. (20)\n44 Autoantibodies.tw. (42037)\n45 exp phospholipid antibody/ (9920)\n46 Anti-endometrial.tw. (23)\n47 Antiphospholipid$.tw. (13777)\n48 exp HLA antigen/ (81011)\n49 exp HLA A1 antigen/ (597)\n50 exp HLA A2 antigen/ (3288)\n51 (HLA or human leucocyte antigen$).tw. (104497)\n52 Anti-laminin-1.tw. (43)\n53 Anti-thyroid.tw. (1873)\n54 Anti-Thomsen Friedenreich antigen$.tw. (5)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n99\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n55 Anti-transferrin.tw. (290)\n56 Anti-LDL.tw. (186)\n57 (Anti-2HSG or Heremans-Schmidt glycoprotein).tw. (4)\n58 interleukin$.tw. (199692)\n59 (MCP-I or monocyte chemoattractant protein-I).tw. (112)\n60 (MIF or migration inhibitory factor$).tw. (5063)\n61 (TNF-a or tumour necrosis factor$ alfa).tw. (5998)\n62 Fas ligand$.tw. (6708)\n63 Endometrial marker$.tw. (18)\n64 CAMs.tw. (2100)\n65 cell adhesion molecule$.tw. (24039)\n66 exp integrin/ (29036)\n67 Integrin$.tw. (48293)\n68 Selectin$.tw. (67300)\n69 Cadherin$.tw. (27150)\n70 Aromatase P450.tw. (202)\n71 estrogen receptor$.tw. (46656)\n72 progesterone receptor$.tw. (19861)\n73 MTMMP$.tw. (15)\n74 cyr61.tw. (755)\n75 exp cysteine rich protein 61/ (753)\n76 cysteine-rich heparin-binding protein$.tw. (12)\n77 (ANXA 1 or ANXA1).tw. (452)\n78 (Annexin 1 or Annexin1).tw. (425)\n79 (PGP 9?5 or PGP9?5 or protein gene product$).tw. (2620)\n80 serum marker$.tw. (7720)\n81 neural marker$.tw. (1119)\n82 cell surface marker$.tw. (5851)\n83 inflammatory marker$.tw. (17339)\n84 microarray$.tw. (101846)\n85 microRNA$.tw. (40082)\n86 proteomic$.tw. (55191)\n87 genomic$.tw. (217184)\n88 (endometri$ adj2 biops$).tw. (4369)\n89 Follistatin$.tw. (1945)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n100\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n90 exp vasculotropin/ (69810)\n91 Vascular Endothelial Growth Factor A.tw. (2275)\n92 exp vitamin D binding protein/ (2064)\n93 exp cytokine/ (1034772)\n94 exp interleukin derivative/ (2790)\n95 exp interleukin 1/ (48499)\n96 exp interleukin 6/ (136328)\n97 exp interleukin 8/ (48884)\n98 exp interleukin 12/ (31842)\n99 exp interleukin 13/ (13584)\n100 exp epidermal growth factor/ (32130)\n101 exp fibroblast growth factor/ (13858)\n102 cytokeratin 19/ (3601)\n103 platelet derived growth factor/ (18930)\n104 cytokeratin-19.tw. (1918)\n105 (VDBP or vitamin D-binding protein$).tw. (1413)\n106 urinary peptide$.tw. (174)\n107 VDBP-Cr.tw. (1)\n108 urinary VDBP corrected for creatinine expression.tw. (1)\n109 urinary marker$.tw. (830)\n110 exp blood analysis/ (118854)\n111 exp endometrium biopsy/ (4988)\n112 exp urinalysis/ or exp biological marker/ (210153)\n113 (biomarker or biomarkers).tw. (159748)\n114 or/1-113 (2734501)\n115 endometriosis/di [Diagnosis] (4979)\n116 114 or 115 (2738583)\n117 exp endometriosis/ (25923)\n118 Endometriosis.tw. (22110)\n119 117 or 118 (27911)\n120 116 and 119 (10326)\n121 Animal/ not Human/ (1204497)\n122 120 not 121 (10279)\nAdditional search February 2015 - May 2015\nEmbase <1980 to 2015 Week 35 (3.09.2015)>\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n101\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n1 Laboratory Test$.tw. (44290)\n2 growth factor$.tw. (335543)\n3 scatter factor$.tw. (1407)\n4 cytokine$.tw. (343623)\n5 hepatocyte growth factor.tw. (10104)\n6 (FGF or fibroblast growth factor$).tw. (39159)\n7 (PDGF or platelet derived growth factor$).tw. (24591)\n8 (EGF or epidermal growth factor$).tw. (73599)\n9 (IGF-I or insulin-like growth factor$ or IGF1).tw. (51838)\n10 (TGF-a or transforming growth factor alfa or TGFa).tw. (583)\n11 (TGF-b or transforming growth factor beta or TGFb).tw. (32580)\n12 (EGFR or epidermal growth factor receptor$).tw. (71526)\n13 (VEGF or vascular endothelial growth factor$).tw. (79087)\n14 exp luteinizing hormone/ec [Endogenous Compound] (22767)\n15 leptin$.tw. (34921)\n16 exp progesterone blood level/ or exp progesterone urine level/ (6534)\n17 Proteolytic enzyme$.tw. (9903)\n18 exp matrix metalloproteinase/ (20462)\n19 matrix metalloproteinase$.tw. (44380)\n20 MMP$.tw. (63208)\n21 TIMP$.tw. (15146)\n22 exp \"tissue inhibitor of metalloproteinase 2\"/ (5136)\n23 exp \"tissue inhibitor of metalloproteinase 1\"/ (9381)\n24 exp glycoprotein/ec [Endogenous Compound] (260024)\n25 (Ca-125 or Ca125 or cancer antigen 125).tw. (10051)\n26 (Ca-19-9 or Ca19-9 or cancer antigen 19-9).tw. (6446)\n27 (PP 14 or PP14).tw. (243)\n28 serum placental protein$.tw. (44)\n29 exp follistatin/ (2283)\n30 Osteopontin$.tw. (9173)\n31 exp intercellular adhesion molecule 1/ (33492)\n32 exp selectin/ (3217)\n33 soluble intercellular adhesion.tw. (1865)\n34 Soluble adhesion molecule$.tw. (944)\n35 sICAM.tw. (3049)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n102\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n36 sVCAM$.tw. (1924)\n37 (sEcadherin or soluble E-cadherin).tw. (125)\n38 (sEselectin or soluble E-selectin).tw. (861)\n39 exp T lymphocyte/ (394405)\n40 exp natural killer T cell/ (6310)\n41 Immune cells alteration$.tw. (6)\n42 (T helper$ or T supressor$ or T helper$ T supressor$ ratio).tw. (26082)\n43 Total complement level$.tw. (20)\n44 Autoantibodies.tw. (44153)\n45 exp phospholipid antibody/ (10362)\n46 Anti-endometrial.tw. (25)\n47 Antiphospholipid$.tw. (14399)\n48 exp HLA antigen/ (83748)\n49 exp HLA A1 antigen/ (622)\n50 exp HLA A2 antigen/ (3409)\n51 (HLA or human leucocyte antigen$).tw. (109332)\n52 Anti-laminin-1.tw. (43)\n53 Anti-thyroid.tw. (2059)\n54 Anti-Thomsen Friedenreich antigen$.tw. (7)\n55 Anti-transferrin.tw. (297)\n56 Anti-LDL.tw. (191)\n57 (Anti-2HSG or Heremans-Schmidt glycoprotein).tw. (4)\n58 interleukin$.tw. (210083)\n59 (MCP-I or monocyte chemoattractant protein-I).tw. (114)\n60 (MIF or migration inhibitory factor$).tw. (5342)\n61 (TNF-a or tumour necrosis factor$ alfa).tw. (6488)\n62 Fas ligand$.tw. (6895)\n63 Endometrial marker$.tw. (18)\n64 CAMs.tw. (2198)\n65 cell adhesion molecule$.tw. (25207)\n66 exp integrin/ (30330)\n67 Integrin$.tw. (50938)\n68 Selectin$.tw. (71624)\n69 Cadherin$.tw. (29496)\n70 Aromatase P450.tw. (207)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n103\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n71 estrogen receptor$.tw. (49530)\n72 progesterone receptor$.tw. (21068)\n73 MTMMP$.tw. (16)\n74 cyr61.tw. (822)\n75 exp cysteine rich protein 61/ (829)\n76 cysteine-rich heparin-binding protein$.tw. (12)\n77 (ANXA 1 or ANXA1).tw. (500)\n78 (Annexin 1 or Annexin1).tw. (440)\n79 (PGP 9?5 or PGP9?5 or protein gene product$).tw. (2760)\n80 serum marker$.tw. (8158)\n81 neural marker$.tw. (1234)\n82 cell surface marker$.tw. (6222)\n83 inflammatory marker$.tw. (19492)\n84 microarray$.tw. (110181)\n85 microRNA$.tw. (47554)\n86 proteomic$.tw. (60599)\n87 genomic$.tw. (233444)\n88 (endometri$ adj2 biops$).tw. (4589)\n89 Follistatin$.tw. (2081)\n90 exp vasculotropin/ (74115)\n91 Vascular Endothelial Growth Factor A.tw. (2526)\n92 exp vitamin D binding protein/ (2196)\n93 exp cytokine/ (1094317)\n94 exp interleukin derivative/ (3281)\n95 exp interleukin 1/ (50850)\n96 exp interleukin 6/ (147379)\n97 exp interleukin 8/ (52281)\n98 exp interleukin 12/ (33479)\n99 exp interleukin 13/ (14685)\n100 exp epidermal growth factor/ (33057)\n101 exp fibroblast growth factor/ (14499)\n102 cytokeratin 19/ (3886)\n103 platelet derived growth factor/ (19655)\n104 cytokeratin-19.tw. (2030)\n105 (VDBP or vitamin D-binding protein$).tw. (1520)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n104\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n106 urinary peptide$.tw. (189)\n107 VDBP-Cr.tw. (1)\n108 urinary VDBP corrected for creatinine expression.tw. (1)\n109 urinary marker$.tw. (883)\n110 exp blood analysis/ (124468)\n111 exp endometrium biopsy/ (5197)\n112 exp urinalysis/ or exp biological marker/ (232619)\n113 (biomarker or biomarkers).tw. (182609)\n114 or/1-113 (2911073)\n115 endometriosis/di [Diagnosis] (5173)\n116 114 or 115 (2915302)\n117 exp endometriosis/ (27433)\n118 Endometriosis.tw. (23449)\n119 117 or 118 (29532)\n120 116 and 119 (10922)\n121 Animal/ not Human/ (1261620)\n122 120 not 121 (10862)\n123 (201501$ or 201502$ or 201503$ or 201504$).em. (49200)\n124 122 and 123 (34)\nAppendix 4. Biomarkers search strategy for CINAHL (EBSCO platform)\nDatabase: CINAHL Plus with Full Text (EBSCOhost) <1980 to 20.04.2015>\n/uni00A0\n# Query Results\nS97 S3 AND S96 1131\nS96 S4 OR S5 OR S6 OR S7 OR S8 OR S9 OR S10 OR S11 OR S12 OR S13 OR S14 OR\nS15 OR S16 OR S17 OR S18 OR S19 OR S20 OR S21 OR S22 OR S23 OR S24 OR\nS25 OR S26 OR S27 OR S28 OR S29 OR S30 OR S31 OR S32 OR S33 OR S34 OR\nS35 OR S36 OR S37 OR S38 OR S39 OR S40 OR S41 OR S42 OR S43 OR S44 OR\nS45 OR S46 OR S47 OR S48 OR S49 OR S50 OR S51 OR S52 OR S53 OR S54 OR\nS55 OR S56 OR S57 OR S58 OR S59 OR S60 OR S61 OR S62 OR S63 OR S64 OR\nS65 OR S66 OR S67 OR S68 OR S69 OR S70 OR S71 OR S72 OR S73 OR S74 OR\nS75 OR S76 OR S77 OR S78 OR S79 OR S80 OR S81 OR S82 OR S83 OR S84 OR\nS85 OR S86 OR S87 OR S88 OR S89 OR S90 OR S91 OR S92 OR S93 OR S94 OR\nS95\n341775\nS95 TX urinary peptide* 1598\nS94 TX (VDBP or vitamin D-binding protein*) 134\nS93 TX cytokeratin-19 109\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n105\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nS92 TX (Luteinizing Hormone* or LH) 18041\nS91 (MH \"Diagnosis, Laboratory+\") 101773\nS90 \"Keratin-19\" 2\nS89 (MH \"Platelet-Derived Growth Factor\") 394\nS88 (MH \"Epidermal Growth Factors\") 1264\nS87 (MH \"Interleukins\") 6584\nS86 (MH \"Cytokines\") 6860\nS85 TX Vitamin D-Binding Protein 131\nS84 (MH \"Vascular Endothelial Growth Factor A\") 194\nS83 TX (endometri* N2 biops*) 432\nS82 TX (endometri* adj2 biops*) 0\nS81 TX genomic$ 7487\nS80 TX proteomic* 2434\nS79 TX microRNA 824\nS78 TX microarray 3123\nS77 TX (PGP 95 or PGP95 or protein gene product*) 9925\nS76 TX (Annexin 1 or Annexin1) 472\nS75 TX (ANXA 1 or ANXA1) 41\nS74 TX cysteine-rich heparin-binding protein* 12\nS73 (MH \"Protein Array Analysis\") 73\nS72 TX cyr61 34\nS71 TX MTMMP* 0\nS70 TX progesterone receptor* 1927\nS69 TX estrogen receptor* 5193\nS68 TX Aromatase P450 38\nS67 TX Cadherin* 900\nS66 TX Selectin* 28411\nS65 TX Integrin* 1587\n/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n106\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nS64 TX cell adhesion molecule* 1578\nS63 TX CAMs 550\nS62 TX Endometrial marker* 54\nS61 TX Fas ligand 338\nS60 TX (TNF-a or tumour necrosis factor* alfa) 1489\nS59 TX (MIF or migration inhibitory factor*) 399\nS58 TX (MCP-I or monocyte chemoattractant protein-I) 13\nS57 TX interleukin 13809\nS56 TX (Anti-2HSG or Heremans-Schmidt glycoprotein) 7\nS55 TX Anti-LDL 9\nS54 TX Anti-transferrin 3\nS53 TX Anti-Thomsen Friedenreich antigen* 1\nS52 TX Anti-thyroid 109\nS51 TX Anti-laminin-1 15\nS50 TX (HLA or human leucocyte antigen*) 4202\nS49 (MM \"HLA Antigens\") 638\nS48 TX Antiphospholipid* 1249\nS47 TX Anti-endometrial 34\nS46 (MH \"Antibodies/BL/DU\") 1294\nS45 TX Autoantibodies 4385\nS43 TX Total complement level 3\nS42 TX (T helper* or T supressor*) 2341\nS41 TX Immune cells alteration* 24\nS40 TX natural killer t-cells 669\nS39 (MM \"T Lymphocytes\") 2404\nS38 TX (sEselectin or soluble E-selectin) 91\nS37 TX (sEcadherin or soluble E-cadherin) 8\nS36 TX sVCAM 100\n/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n107\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nS35 TX sICAM 173\nS34 TX Soluble adhesion molecule 368\nS33 TX soluble intercellular adhesion 237\nS32 (MM \"Cell Adhesion Molecules\") 52\nS31 TX Osteopontin* 416\nS30 TX Follistatin 74\nS29 TX serum placental protein* 11\nS28 TX (Ca-19-9 or Ca19-9 or cancer antigen 19-9) 262\nS27 TX (Ca-125 or Ca125 or cancer antigen 125) 831\nS26 (MM \"Glycoproteins/BL/DU\") 224\nS25 TX tissue inhibitor of metalloproteinase 423\nS24 TX TIMP* 1845\nS23 TX MMP* 4244\nS22 TX matrix metalloproteinase* 3325\nS21 TX Proteolytic enzyme* 1461\nS20 (MM \"Progesterone/BL/DU\") 51\nS19 TX leptin* 3258\nS18 (MM \"Luteinizing Hormone/BL/DU\") 38\nS17 TX (VEGF or vascular endothelial growth factor*) 7166\nS16 TX (EGFR or epidermal growth factor receptor*) 6188\nS15 TX (TGF-b or transforming growth factor beta or TGFb) 2972\nS14 TX (TGF-a or transforming growth factor alfa or TGFa) 464\nS13 TX (IGF-I or insulin-like growth factor* or IGF1) 3588\nS12 TX (EGF or epidermal growth factor*) 6250\nS11 TX (PDGF or platelet derived growth factor*) 3195\nS10 TX (FGF or fibroblast growth factor*) 3395\nS9 TX hepatocyte growth factor* 880\nS8 TX cytokine* 20821\n/uni00A0/uni00A0(Continued)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n108\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nS7 TX scatter factor* 1864\nS6 TX growth factor* 76163\nS5 TX Laboratory Test* 82732\nS4 TX (biomarker* or marker*) 84857\nS3 S1 OR S2 2841\nS2 TX Endometrio* 2841\nS1 (MM \"Endometriosis\") 889\nS4 TX (biomarker* or marker*) 61,794\nS3 S1 OR S2 2,174\nS2 TX Endometrio* 2,174\nS1 (MM \"Endometriosis\") 1,306\n/uni00A0/uni00A0(Continued)\n/uni00A0\nAppendix 5. Biomarkers search strategy for other databases\nSearches for clinical studies\nDatabase: PsycINFO (Ovid) <1806 to April Week 2 2015 (20.04.2015)>\nSearch strategy:\n1. endometriosis.tw. (174)\nDatabase: Web of Science Core Collection (Thomson Reuters) <1900 to Present (20.04.2015)>\nSearch strategy:\n1. Topic=(endometrio*) AND Topic=(diagnos* OR test* OR imag*); Timespan=All Years (7425)\nDatabase: LILACS <20.04.2015>\nSearch strategy:\n1. (tw:(endometriosis)) AND (tw:(diagnos*)) (420)\nDatabase: OAIster (WorldCat.org) <20.04.2015>\nSearch strategy:\n1. endometriosis and (marker* or biomarker*) (11)\n2. endometriosis and diagnos* (446)\nDatabase: TRIP <20.04.2015>\nSearch strategy:\n1. (endometriosis and diagnos*) (1648)\nSearches of trial registers for ongoing and registered trials\nDatabase: ClinicalTrials.gov (US NIH) <20.04.2015>\nSearch strategy:\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n109\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n1. endometriosis (220)\n2. endometriosis AND diagnosis (22)\nDatabase: WHO International Clinical Trials Registry Platform (ICTRP) <20.04.2015>\nSearch strategy:\n1. endometriosis (523)\nSearches for the reviews as source of references to potentially relevant studies\nDatabase: MEDION <10.01.2014>\nSearch strategy:\nICP Code female genital system (including breast), Signssymp medical imaging, laboratory tests, histology and cytology, endoscopy and\nlaparoscopy. Filter: systematic reviews of diagnostic studies. (2)\nDatabase: DARE (CRD) <20.04.2015>\nSearch strategy:\n1. endometriosis (99)\nPubMed, a ‘Systematic Review’ search under the ‘Clinical Queries’ link <20.04.2015>\nSearch strategy:\n1. (endometriosis) AND systematic[sb] (418)\nCategory: Diagnosis; Scope: Broad\nSearches for papers recently published and not yet indexed in the major databases\nSearch engine: PubMed <20.10.2014 to 20.04.2015>\nSearch strategy:\n/uni00A0\n1. marker (14979)\n2. test (61151)\n3. diagnos* (69743)\n4. biomarker (10806)\n5. or/1-4 (7943)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nIndex test(s) set\n6. Endometriosis (584)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nTarget condition set\n7. 5 and 6 (267)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nCombined sets\n/uni00A0\n/uni00A0\nAppendix 6. Imaging search strategy for CENTRAL\nDatabase: EBM Reviews - Cochrane Central Register of Controlled Trials <April 2015 (20.4.2015)>\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n110\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n1. exp magnetic resonance imaging or exp ultrasonography or exp Imaging, Three-Dimensional or\nexp radiography (772)\n2. (ultraso* or magnetic resonance imaging or MRI or imag*).tw. (36)\n3. diagnos* (106503)\n4. [mh diagnosis] (257329)\n5. or/1-4 (310878)\nIndex test(s) set\n6. exp endometriosis (142)\n7. endometrio*.tw. (22)\n8. [mh endometriosis] (553)\n9. or/6-8 (681)\nTarget condition set\n10. 5 and 9 (465)\n11. (animals not (humans and animals)).sh. (36)\n12. 10 not 11 (445)\nCombined sets\n/uni00A0\n/uni00A0\nAppendix 7. Imaging search strategy for MEDLINE\n/uni00A0\n1. exp magnetic resonance imaging/ or exp ultrasonography/ or exp Imaging, Three-Dimensional/\nor exp radiography/ (1114639)\n2. ultraso$.tw. or magnetic resonance imaging.tw. or MRI.tw. or imag$.tw. (1020000)\n3. diagnos$.tw. (1750239)\n4. or/1-3 (3048652)\nIndex test(s) set\n5. exp Endometriosis/ (17415)\n6. Endometrio$.tw. (21775)\n7. or/5-6 (25236)\nTarget condition set\n8. 4 and 7 (8107)\n9. (animals not (humans and animals)).sh. (3931867)\n10. 8 not 9 (7391)\nCombined sets\n/uni00A0\n/uni00A0\nAppendix 8. Imaging search strategy for EMBASE\n/uni00A0\n1. Ecography/exp or radiodiagnosis/exp (1988601)\n2. ‘magnetic resonance imaging’:ab,ti or MRI:ab,ti or imag*:ab,ti or ultraso*:de,ab,ti (1370683)\n3. diagnos*:ab,ti (2373625)\nIndex test(s) set\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n111\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n4. ‘diagnostic accuracy':de or‘diagnostic test accuracy study’:de or 'diagnostic value':de (298281)\n5. or/1-4 (4437871)\n/uni00A0\n/uni00A0\n/uni00A0\n6. Endometrio*:de,ab,ti (37439)\n7. 'endometriosis'/exp/dm_di (4976)\n8. or/6-7 (37439)\nTarget condition set\n/uni00A0\n/uni00A0\n/uni00A0\n9. #5 and #8 (13500)\n10. animal:de not (animal:de and human:de) (3861389)\n11. #9 not #10 (12161)\nCombined sets\n/uni00A0\n/uni00A0\nAppendix 9. Imaging search strategy for CINAHL\n/uni00A0\n# Query Results /uni00A0\n/uni00A0\n/uni00A0\n/uni00A0\nS9 S3 AND S8\nSearch modes - Boolean/Phrase\nSearch Screen - Advanced Search\n668 Combined sets\n/uni00A0\n/uni00A0\n/uni00A0\nS8 S4 OR S5 OR S6 OR S7 258011\n/uni00A0\n/uni00A0\n/uni00A0\nS7 TX imag* 258011\n/uni00A0\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n112\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nS6 TX ultraso* 58570\n/uni00A0\n/uni00A0\n/uni00A0\nS5 TX (magnetic resonance imaging or MRI) 58387\n/uni00A0\n/uni00A0\n/uni00A0\nS4 TX (biomarker* or marker*) 84857\n/uni00A0\n/uni00A0\n/uni00A0\nS3 S1 or S2 2841\n/uni00A0\n/uni00A0\n/uni00A0\nS2 TX Endometrio* 2841\n/uni00A0\n/uni00A0\n/uni00A0\nS1 (MM \"Endometriosis\") 889\n/uni00A0\n/uni00A0\nAppendix 10. Imaging search strategy for other databases\nDatabase: Web of Science Core Collection (Thomson Reuters) <1900 to Present (20.04.2015)>\nSearch strategy:\n1. Topic=(endometrio*) AND Topic=(diagnos* OR test* OR imag*); Timespan=All Years (7425)\nDatabase: LILACS <20.04.2015>\nSearch strategy:\n1. (tw:(endometriosis)) AND (tw:(diagnos*)) (420)\nDatabase: OAIster (WorldCat.org) <20.04.2015>\nSearch strategy:\n1. endometriosis and (marker* or biomarker*) (11)\n2. endometriosis and diagnos* (446)\nDatabase: TRIP <20.04.2015>\nSearch strategy:\n1. (endometriosis and diagnos*) (1648)\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n113\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nSearches of trial registers for ongoing and registered trials\nDatabase: ClinicalTrials.gov (US NIH) <20.04.2015>\nSearch strategy:\n1. endometriosis (220)\n2. endometriosis AND diagnosis (22)\nDatabase: WHO International Clinical Trials Registry Platform (ICTRP) <20.04.2015>\nSearch strategy:\n1. endometriosis (523)\nSearches for the reviews as source of references to potentially relevant studies\nDatabase: MEDION <10.01.2014>\nSearch strategy:\nICP Code – female genital system (including breast), Signssymp – medical imaging, endoscopy and laparoscopy. Filter: systematic reviews\nof diagnostic studies (190)\nDatabase: DARE (CRD) <20.04.2015>\nSearch strategy:\n1. endometriosis (99)\nPubMed, a ‘Systematic Review’ search under the ‘Clinical Queries’ link <20.04.2015>\nSearch strategy:\n1. (endometriosis) AND systematic[sb] (418)\nCategory: Diagnosis; Scope: Broad\nSearches for papers recently published and not yet indexed in the major databases\nSearch engine: PubMed <20.10.2014 to 20.04.2015>\nSearch strategy:\n/uni00A0\n1. marker (14979)\n2. test (61151)\n3. diagnos* (69743)\n4. biomarker (10806)\n5. or/1-4 (7943)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nIndex test(s) set\n6. Endometriosis (584)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nTarget condition set\n7. 5 and 6 (267)\nFilters: Publication date from 2014/10/20 to 2015/04/20\nCombined sets\n/uni00A0\n/uni00A0\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n114\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\nC O N T R I B U T I O N S /uni00A0 O F /uni00A0 A U T H O R S\nVicki Nisenblat co-ordinated the production of the protocol and the review series and was involved in literature search, quality appraisal\nand data extraction for the included studies and took a primary role in writing the review. Louise Hull participated in co-ordination of the\nprotocol and the review series. Lucy Prentice participated in literature search, quality appraisal and data extraction for the included studies\nand contributed to the first dra/f_t of the review. Neil Johnson contributed to the conception and the design of the review and was involved\nin quality appraisal, data extraction for the included studies and critical revision of the manuscript. Patrick Bossuyt provided advice on\nstatistical methods for the review. Cindy Farquhar critically reviewed the methodological aspects and participated in the study design. All\nthe authors contributed to the revision and dra/f_ting of the review.\nD E C L A R A T I O N S /uni00A0 O F /uni00A0 I N T E R E S T\nCindy Farquhar is a director/shareholder of a fertility/gynaecology clinic and undertakes private practice within those premises.\nNeil Johnson is involved in research funded by Abb-Vie. He has received support to attend conferences from MSD, Merck-Serono and Bayer.\nHe has been on an advisory board for Vifor Pharma.\nNo other authors have any conflict of interest to declare.\nS O U R C E S /uni00A0 O F /uni00A0 S U P P O R T\nInternal sources\n• Cochrane Gynaecology and Fertility Group, University of Auckland, New Zealand.\nTechnical support\n• The Robinson Institute, University of Adelaide, Australia.\nAccess to academic resources\nExternal sources\n• None, Other.\nD I F F E R E N C E S /uni00A0 B E T W E E N /uni00A0 P R O T O C O L /uni00A0 A N D /uni00A0 R E V I E W\nGeneral scope: this review is a part of the review series arising from the same generic protocol. The following sections were adjusted to\nthe main topic of the review as following.\nBackground: the section on the index test was modified and all the irrelevant information to combination of different testing modalities\nwas removed. The \"Rationale\" section was updated and now includes a clearer definition of triage diagnostic tests.\nObjectives\n1. During revision of the literature on the subject, we identified a substantial body of studies looking at the biomarkers, expression of which\ndid not change by presence of endometriosis (no statistically significant difference was found in women with and without the disease).\nWe believe that presenting this type of data, obtained from the adequately designed studies is important for both the clinicians and the\nresearchers in the field, which has been explained in the Background section under \"Rationale\", in the Methods section under \"Criteria\nfor considering studies for this review - Index test\" and added to \"Objectives\" as a secondary objective: '2.To assess the biomarkers\nwhich were not affected by endometriosis and hence were unlikely to discriminate between women with and without the disease'.\n2. The list of the sources of heterogeneity has been updated.\nMethods\n1. Criteria for considering studies for this review were updated as following.\na. Types of studies: we removed the 'cohort' and 'case control' classifications and introduced the concepts of 'single-gate design'\nand 'two-gate' design'. This was defined as the presence of a single or multiple sets of inclusion criteria by clinical condition or by\nreference standard. We found this classification more informative in the description of diagnostic studies, all of which are cross-\nsectional in nature. We limited the inclusion criteria to the studies with a single set of inclusion criteria by reference standard (i.e.\nall women who underwent abdominal surgery), but included single or multiple sets of inclusion criteria by clinical presentation (i.e.\nwomen with suspected endometriosis or other indications for abdominal surgery), referring to these as 'single-gate design' and 'two-\ngate' design', respectively.\nb. Likewise, we removed the terminology 'prospective studies' and introduced 'studies performed on prospectively collected samples'.\nThis decision was guided by the fact that most diagnostic studies are retrospective in nature, as they aim to compare the result\nof an index test with the result of a reference standard in the same group of participants, where the groups are classified by the\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n115\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\noutcome of the reference standard. Also, the analysis of the index test could have been performed retrospectively in a single batch\non stored samples a/f_ter the prospective collection of samples. The timing of sample collection (before or a/f_ter surgical treatment\nof the disease) from a preoperatively recruited population has more impact on the test result than the timing of the laboratory\nassay. Therefore, we included only studies where the biological sample was collected before the reference surgical procedure, i.e.\n'prospectively collected', irrespective of the actual timing of test performance and abandoned labelling studies as 'prospective'\nor 'retrospective' to avoid confusion. This allowed us to include the studies from well-established high-quality tissue banks using\nwell characterised archived samples as omission of these studies would have resulted in the loss of potentially valuable data. This\nis presented in the Methods under \"Criteria for considering studies for this review\". For the combined tests that include imaging\nmodality, only prospectively recruited women and prospectively performed tests were considered eligible.\nc. Index tests were modified to pertain only to the combined test of different testing modalities and the table listing the tests of interest\n(Table 2) was updated accordingly. As a summary review of the series, this review also presents the full list of the tests evaluated\nin each sister review.\nd. Target condition now also includes deep pelvic endometriosis in view of the growing body of literature on this condition as a separate\nentity and its diagnostic importance to optimise the surgical approach.\ne. Spectrum of disease: following an ad hoc observation, we included the studies that involved only selected populations of women\nwith endometriosis (i.e. specific rASRM stages) in view of the emerging evidence on poor correlation of this classification with\ninfertility and pain symptoms. Exclusion of such studies could result in the loss of potentially important diagnostic information from\nthe otherwise eligible publications. Where possible, we aimed to address the impact of the inclusion of these studies in investigations\nof heterogeneity.\n2. Search methods for identification of studies\na. In the protocol we stated that the grey literature (unpublished studies including conference proceedings and reports) would be\nidentified and defined specific search strategies. In practice, the paucity of relevant data that was available from abstracts made\nit impossible to apply the selection criteria and methodological quality judgement to these studies. Identification of these types\nof studies and attempts to obtain the necessary information directly from the study investigators was anticipated to increase the\nalready labour intense work involved in preparation of this review. Therefore, by consensus between the key authors, we removed\nalready identified unpublished studies and did not complete an intended search for unpublished material.\nb. The search strings were updated for all biomarkers excluding imaging (searched separately), applying the same principles as\npresented in the protocol.\n3. Assessment of methodological quality: the QUADAS-2 tool was tailored for the topic of the review. The differences between the original\nQUADAS-2 tool and the designed for this review are outlined in the relevant section in the Methods.\nAnalysis\n1. The section on statistical methods was amended and tailored to the types of tests included in the review.\n2. We performed no sensitivity analyses and no assessment of heterogeneity due to insufficient data.\n3. When a test performance was judged against the predetermined diagnostic criteria, only the point estimates of sensitivity and specificity\nwere considered as we believe that presenting these metrics of test performance is the most helpful and informative way to summarise\nthe diagnostic data. We acknowledge that the choice of the most helpful summary is subjective. There are tests where the point\nestimates did not reach the predetermined criteria, but the confidence intervals (CIs) contain the values above the thresholds for\nreplacement or triage tests. These tests could have diagnostic value if the point values underestimated their diagnostic potential. For the\ntests where the point estimates reached the criteria for a replacement or triage test but the CIs contained values below the thresholds,\npoint values could have overestimated the diagnostic performance of the test. If the range of the CIs rather than the point estimates of\nthe data are used, the predetermined cut-off becomes meaningless. We did not consider CIs in qualifying the test performance, however,\nwe utilised the CIs in interpreting the reliability of the obtained data.\nThe authors list and order changed to accurately reflect their contribution to the review.\nN O T E S\nThe initially planned single review on the non-invasive tests for diagnosis of endometriosis was split into five smaller reviews in order to\nfacilitate data handling and interpretation, due to the abundance and diversity of the suggested tests. The review was generated from\na generic protocol, which was designed for all the reviews in these series. The other reviews from the series include 1) Endometrial\nbiomarkers for the non-invasive diagnosis of endometriosis; 2) Urinary biomarkers for the non-invasive diagnosis of endometriosis; 3)\nImaging modalities for the non-invasive diagnosis of endometriosis; 4) Blood biomarkers for the non-invasive diagnosis of endometriosis.\nI N D E X /uni00A0 T E R M S\nMedical Subject Headings (MeSH)\nAromatase /uni00A0[analysis];/uni00A0 Biomarkers /uni00A0[*analysis];/uni00A0 CA-125 Antigen /uni00A0[blood];/uni00A0 CA-19-9 Antigen /uni00A0[blood];/uni00A0 Endometriosis /uni00A0[*diagnosis]\n/uni00A0[diagnostic imaging];/uni00A0 Interleukin-6 /uni00A0[blood];/uni00A0 Leukocytes /uni00A0[cytology];/uni00A0 Ovarian Diseases /uni00A0[*diagnosis] /uni00A0[diagnostic imaging];/uni00A0 Pelvis\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n116\n\nCochrane\nLibrary\nTrusted evidence.\nInformed decisions.\nBetter health.\n/uni00A0\n/uni00A0\nCochrane Database of Systematic Reviews\n/uni00A0[diagnostic imaging];/uni00A0 Peritoneal Diseases /uni00A0[*diagnosis] /uni00A0[diagnostic imaging];/uni00A0 Phosphopyruvate Hydratase /uni00A0[urine];/uni00A0 Sensitivity and\nSpecificity;/uni00A0 Ubiquitin Thiolesterase /uni00A0[analysis];/uni00A0 Ultrasonography;/uni00A0 Vitamin D-Binding Protein /uni00A0[urine]\nMeSH check words\nFemale; Humans\nCombination of the non-invasive tests for the diagnosis of endometriosis (Review)\nCopyright © 2016 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.\n117","source_license":"CC0","license_restricted":false}