Math & Engineering

A pilot study to evaluate the relative efficacy of endometrial biopsy and full curettage in making a diagnosis of endometriosis by the detection of endometrial nerve fibers

Description
A pilot study to evaluate the relative efficacy of endometrial biopsy and full curettage in making a diagnosis of endometriosis by the detection of endometrial nerve fibers
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  GENERAL GYNECOLOGY A pilot study to evaluate the relative efficacy of endometrialbiopsy and full curettage in making a diagnosis ofendometriosis by the detection of endometrialnerve fibers Moamar Al-Jefout, MD; Natasha Andreadis, MBBS; Natsuko Tokushige, PhD; Robert Markham, PhD; Ian Fraser, MD OBJECTIVE:  The purpose of this study was to evaluate endometrialbiopsy and curettage in detecting small nerve fibers in eutopic endo-metrium for diagnosis of endometriosis. STUDY DESIGN:  Endometrialbiopsieswithprecise,consistenttechniqueandcurettings were taken from 37 women (20 with endometriosis and 17 withoutendometriosis). Sensitivity, specificity, and positive and negative predictivevalue were formally calculated. Endometrial nerve fibers were immunohisto-chemically detected using the pan-neuronal marker PGP9.5. RESULTS:  Small nerve fibers were detected in all endometrial biopsiesand curettings from all 20 women with endometriosis, but were notdetected in endometrium taken from 17 women without endometriosis.Mean (  SD) nerve fiber density in the endometrial biopsies was 26.8per mm 2  55.9 (range, 1.6-125) and for curettings was 21.6 per mm 2  33.1 (range, 0.8-250), with 100% specificity, sensitivity, and posi-tive and negative predictive value. CONCLUSION:  Careful endometrial biopsy combined with immunohis-tochemical staining for nerve fibers may be a reliable means of diag-nosing or excluding endometriosis. Key words:  endometrial biopsy, endometrial curettage,endometriosis, immunohistochemistry, nerve fibers Cite this article as: Al-Jefout M, Andreadis N, Tokushige N, et al. A pilot study to evaluate the relative efficacy of endometrial biopsy and full curettage inmaking a diagnosis of endometriosis by the detection of endometrial nerve fibers. Am J Obstet Gynecol 2007;197:578.e1-578.e4. W e have previously reported thenovel finding of multiple smallunmyelinated sensory C nerve fibers inthe functional layer of eutopic endome-trium in women with endometriosis,while women without endometriosis donot have any nerve fibers in the func-tional layer. 1,2  Women with endometri-osis also have a highly significantly in-creased density of nerve fibers in thebasal layer of endometrium and in themyometrium, compared with womenwithout endometriosis. 1  These nervefibers are also found in ectopic endo-metriotic plaques and express a widerange of neural function markers, 2  al-though nerve fibers in the functionallayer of eutopic endometrium wereonly found to express vasointestinalpeptide (VIP), neuropeptide Y (NPY),substance P (SP), and calcitonin gene-related peptide (CGRP). 3  Throughoutour studies, we have been impressedthat all women with laparoscopically proven endometriosis have had nervefibers detectable in eutopic endome-trium (functional layer), while none of the women without endometriosis(laparoscopically excluded) has hadany detectable nerve fibers. It thereforeappeared that this finding could beused as the basis for a diagnostic testfor endometriosis using endometrialcurettage or even an endometrialbiopsy.Confirmation of diagnosis can be amajor probleminsome womenwith en-dometriosis. At the moment, there is nosimple, reliable, noninvasive way to di-agnose endometriosis, although thereareanumberofstudiescurrentlyunder-way to try and identify “biomarkers” of thisdisease.Certainsymptomsmaysug-gest that a diagnosis is likely, and physi-cal and internal examination may rein-force the possibility of endometriosis.However, laparoscopy is still requiredfor confirmation. The length of timefrom the onset of endometriotic symp-toms to the definite diagnosis is oftenquitelong,withanaverageof6-7yearsinmany centers. 4-7  A simple diagnostictool may greatly help to reduce thisdelay.Wesetupapilotstudytocomparetherelative efficacy of a narrow, disposable“Endosampler” endometrial biopsy suc-tion cannula in comparison with full cu-rettage specimens to detect these smallsensory C nerve fibers in the functional From the Department of Obstetrics andGynaecology, Queen Elizabeth II ResearchInstitute for Mothers and Infants, Faculty of Medicine, College of Health Sciences,University of Sydney, Sydney, Australia. Received Dec. 7, 2006; revised Feb. 11,2007; accepted April 18, 2007.Funding was provided by the Department of Obstetrics and Gynaecology, Faculty of Medicine, College of Health Sciences,University of Sydney, Sydney, Australia.Reprints: M. Al-Jefout, MD, Department of Obstetrics and Gynaecology, Queen ElizabethII Research Institute for Mothers and Infants,University of Sydney, Sydney, NSW, 2006, Australia; aljefoutmr@yahoo.co.uk 0002-9378/$32.00© 2007 Mosby, Inc. All rights reserved.doi: 10.1016/j.ajog.2007.04.032 Research  www. AJOG .org  578.e1  American Journal of Obstetrics &  Gynecology  DECEMBER 2007  layer of endometrium in women withlaparoscopically proven endometriosis.Density of nerve fibers was also calcu-lated both for endometrial biopsy andcurettage specimens. M ATERIALS AND  M ETHODS Participants Inclusion criteria for the study were any women in the reproductive age groupundergoing laparoscopy for suspectedendometriosisorinfertility,andnotcur-rently receiving hormonal treatment forendometriosis. Exclusion criteria for thestudy included pregnancy and unwill-ingness to participate in the study. Writ-teninformedconsentwasobtainedfromall participants.This study was registered with the Aus-tralian Clinical Trials Registry (ACTR)N012606000052538 (registered: June 2,2006).ThisstudywasapprovedbytheEth-ics Review Committee (RPAH Zone) of the Sydney South West Area Health Ser-vice(ProtocolnumberX05-0345).Patients were recruited and treated atThe Royal Prince Alfred Hospital, Syd-ney. Recruitment was based upon pa-tient symptoms of chronic pelvic painand/or infertility. Clinical history wasobtained from the participants beforethe procedure. Tissue processing andstaining were done at the laboratories of theQueenElizabethIIResearchInstitutefor Mothers and Infants, University of Sydney.The reference standard in this study for diagnosing endometriosis was lapa-roscopy and visualization of endometri-otic lesions with surgical staging of thedisease by 3 gynecologists with extensiveexperience in endometriosis. Peritonealbiopsy confirmation was available inmost cases.Endometrial biopsies were obtainedpriortolaparoscopyusingtheEndosam-pler (Medgyn Products, Inc, Lambard,IL), which is a sterile, narrow, and dis-posable plastic cannula for sampling en-dometrium without the need for anes-thesia. The 3-mm cannula has 2 smallsharpslitsatthedistaltipandisattachedto a 10-mm syringe to ensure suctionand efficient sampling.Meticulous attention was paid to thetechnique of endometrial biopsy in or-der to ensure that a narrow but deep en-dometrial strip was obtained (Figure 1). The cannula of the Endosampler has agentlecurvethatisdesignedtofollowthecontour of the uterus. Preliminary study demonstrated that a more efficient bi-opsy was obtained if the cannula was in-serted in line with the contour of theuterusandthenrotatedthrough180°,sothatthecannulaslitswerefirmlypressedagainst the endometrium (Figure 1). Thecannula was then slowly withdrawn in astraight line with full suction. Prelimi-narystudydemonstratedthatrotationof the cannula during withdrawal pro-duced more superficial and fragmentedtissue than straight line withdrawal. Asolidcolumnofendometriumwasmucheasier to assess immunohistochemically.All samples underwent hematoxylin–eosinstainingforconventionalhistolog-ical assessment, and dating was doneblindlyby1experiencedgynecologicpa-thologist at The Royal Prince AlfredHospital, Department of AnatomicalPathology. Immunohistochemistry Specimens were fixed in 10% neutralbuffered formalin for 18-24 hours, pro-cessed, and embedded in paraffin wax according to a standard protocol. Eachsection was cut at 4   m and routinely stained with hematoxylin–eosin. Anti-genretrievaltechniquesforPGP9.5wereused. Serial sections, also cut at 4   m,were immunostained using an antibody for polyclonal rabbit anti-PGP9.5 (dilu-tion 1:1000; Dako Cytomation, Sydney,Australia), a highly specific pan-neuro-nalmarker,for30minutesatroomtem-perature. Sections were washed in Trisbuffer, incubated with Envision-labeledpolymer-AP mouse/rabbit for 30 min-utes (Dako Cytomation), and stained FIGURE 1 Technique of endometrial biopsy Technique of endometrial biopsy. (1) After insertion of Endosampler into the uterine cavity accordingto the curve of uterine cavity, rotate the Endosampler 180° to ensure a firm pressure backwardagainst the endometrium. (2) Pull straight backward; do not rotate during application of suction. Al-Jefout. A pilot study to evaluate the relative efficacy of endometrial biopsy. Am J Obstet Gynecol 2007.  www.AJOG.org   General Gynecology  Research DECEMBER 2007  American Journal of Obstetrics &  Gynecology  578.e2  with permanent fast red chromogen(Dako Cytomation) for 10 minutes. Allimmunostaining was carried out on aDako Autostainer Model S3400 (DakoCytomation, Inc, Carpinteria, CA). Im-ages of the sections were captured usinganOlympusBX51digitalcamera(Olym-pus, Tokyo, Japan). We used normalskin as a positive control as it reliably contains myelinated and unmyelinatednerve fibers expressing PGP9.5. Rabbitimmunoglobulin fraction was used as anegative control, the concentration be-ing matched with the concentration of the PGP9.5 antibodies.An assessment of nerve fiber density was performed using Image Pro PlusDiscovery (MediaCybernetics, SilverSpring, MD). An orthogonal grid mask was sketched above the images capturedfrom the digital camera. The sections of the grid were 50   m a slide. Once thegrid was in position, nerve fibers in theendometrium and the total number of squares covering the sections of endo-metrium were counted. The total num-ber of nerve fibers was divided by the to-tal number of squares covering theendometrium to obtain an average of nerve fibers per square. On average, wecalculated 10 grids for each slide and in-cluded grids with no nerve fibers. Iden-tification and counting of nerve fiberswas undertaken using Image Pro PlusDiscovery (MediaCybernetics) at ap-proximately 20   magnifications by 3different researchers, 2 of them withgood experience in nerve fiber identifi-cation. Blinded counting gave close cor-relations between all the 3 individuals.One single section was assessed fromeach curettage and endometrial biopsy specimenfromindividualpatients.Mul-tiple sections were not assessed in any subject.Theresultswereexpressedasthemean number (  SD) of nerve fibers permm 2 in each endometrial specimen. R  ESULTS During the period from January 1-De-cember 1, 2006, 18 paired samples of eu-topicendometrium(endometrialbiopsy and curettage) and 2 single endometrialbiopsies were obtained from womenwith laparoscopically proven endome-triosis, and another 17 paired samples of endometrium from women with no evi-denceofendometriosis.Allsampleswerecarefully fixed and exhibited clear histo-logic features consistent with normalmenstrualcyclephases.Thewomenwithendometriosis had varying severities of disease according to the Revised Ameri-can Fertility Society Scoring System 8 (rAFS stage 1  10 women; stage 2  5women; stage 3  1 woman; stage 4  4women). Twenty Endosampler biopsieswere obtained, while only 18 full curet-tage samples were achieved because of severe cervical stenosis in 2 cases. Onecaseofmildpostoperativefeveroccurredand was treated by antibiotics anddischarged.The Endosampler provided tissue thatwasofequalqualityforassessmentasthecurettage strips. In women with con-firmed endometriosis, the mean (  SD)nerve fiber density of Endosampler was26.7    55.9 per mm 2 (range, 1.6-125),and for the curettings was 20.4    33.1per mm 2 (range, 0.8-250). All samplesfrom endometriosis subjects from bothEndosampler and curettage were posi-tive for small nerve fibers stained withPGP9.5 (Figure 2). On average, nerve fi-bers were detected in 50-60% of highpower fields; however, these nerve fiberswere not distributed homogeneously throughout the endometrium (Figure 3). The density of nerve fibers was markedly skewed, with few specimens showingcounts above 100 mm 2 and with most be-tween 2-50 per mm 2 (Figure 4). No nerve fibers were found in women withoutendometriosis.We formally calculated specificity,sensitivity, positive predictive value(PPV), and negative predictive value(NPV) for both endometrial biopsy andcurettage, which were all 100%. FIGURE 2 Nerve fiber in endometrialbiopsy Single nerve fiber ( arrow  ) in endometrial biopsy,stained immunohistochemically with PGP9.5and the chromogen fast red (50   m). Al-Jefout. A pilot study to evaluate the relative efficacy of endometrial biopsy. Am J Obstet Gynecol 2007. FIGURE 3 Distribution of endometrialnerve fibers in the eutopicendometrium in a woman whounderwent hysterectomy forendometriosis Identified nerve fibers are circled in red (500  m). Al-Jefout. A pilot study to evaluate the relative efficacy of endometrial biopsy. Am J Obstet Gynecol 2007. FIGURE 4 Box-whisker plots of nervefiber density ec 250200150100500 352772711 This figure demonstrates box-whisker plots ofthe median, 5th, 25th, 75th, and 95th percentilesplus outliers with high values. The number be-side each outlier indicates the identificationnumber of individual subjects. Note the logscale for nerve fiber densities. Al-Jefout. A pilot study to evaluate the relative efficacy of endometrial biopsy. Am J Obstet Gynecol 2007. Research  General Gynecology  www.AJOG.org  578.e3  American Journal of Obstetrics &  Gynecology  DECEMBER 2007  C OMMENT This pilot study suggests that a carefully taken endometrial biopsy may be a reli-able means of making a diagnosis of en-dometriosis in women who are not cur-rently on any hormonal treatment. Wehave preliminary, unpublished evidencethat these nerve fibers may no longer bedetectable in some women on hormonaltherapy. The presence of endometrialnerve fibers in women with other pelvicpathologieswhichmaycausepelvicpain,suchasadenomyosis,uterinefibroids,orendometritis, has not yet beeninvestigated.We believe that meticulous attentionto biopsy technique is of great impor-tanceinprovidinggoodqualitytissueforthis evaluation. With trial and error weimproved our biopsy technique to pro-vide a narrow, solid sample of endome-trialtissue.ItisvaluabletorotatetheEn-dosampler 180° against the curve of theuterussoastoensurefirmpressureback-ward against the endometrium and towithdrawthebiopsycannulainastraightline.Another important issue is the metic-ulous attention to immunohistochemi-cal staining technique. We fine-tunedthe PGP9.5 antibody and assay condi-tions several times until we achievedclear discrimination of these small nervefibers from background staining.The wide range in mean nerve fiberdensities for both techniques (0.8-250per mm 2 ) can be explained by the non-homogeneous distribution of these en-dometrial nerve fibers, which is clearly visible in Figure 3. This variable density  of nerve fibers in the functional layermeans that a narrow endometrial biopsy may sometimes pass between nerve fi-bers, and therefore more than 1 sectionthroughthebiopsymaysometimesneedto be inspected. Data for endometrialnerve fiber density were markedly skewed, with some individuals demon-strating densities greater than 50 nervefibers per mm 2 . However, some individ-uals exhibited densities as low as 2 permm 2 .We find it quite extraordinary thatthere appears to be 100% specificity andsensitivityofthistechniquefordetectionof endometrial nerve fibers in womenwith endometriosis and none in womenwithout endometriosis. 1,3  Undoubtedly,occasional exceptions will be demon-strated in future clinical trials, and thistechnique clearly needs detailed multi-center study.Endometrial biopsy is a less invasiveprocedure than laparoscopy and canusually be achieved without general an-esthesia or strong analgesics. This may encourage clinicians to use endometrialbiopsy in the future for making a diag-nosis of endometriosis at an early stage.The fact that endometrial biopsy can beobtained in an outpatient setting makesthis a very promising technique for early endometriosis diagnosis.Aspecialtargetgroupcouldbethead-olescents with spasmodic dysmenorrheaor atypical pelvic pain and a strong fam-ily history of endometriosis, a group inwhom delay in diagnosis is typically greater than in older women. An endo-metrial biopsy may enable clinicians tomake a more definitive early diagnosisand initiate early therapy. This shorten-ingoftimefromsymptomonsettodiag-nosis could ultimately have a major im-pact on patients’ quality of life, bothintellectual and physical, productivity,and other life aspects, and may contrib-ute to protection of their future fertility.It is recognized that visual assessmentat laparoscopy of apparent peritonealsitesofminimalandmildendometriosis,ordeeperpelvicendometriosis,issubjectto considerable interindividual observervariability. One and the same “lesion”may be assessed differently by differentobservers. Histopathologic confirma-tionmaymakethediagnosisofendome-triosis more reliable. 9  This endometrialbiopsy technique may be useful as a spe-cific tool for initial diagnosis confirma-tion, perhaps with occasional diagnosticlaparoscopyformorepreciseassessmentand staging, or later planned surgicallaparoscopy for effective excision.  f ACKNOWLEDGMENTS  The authors thank Prof Peter Russell, Drs Mi-chael Cooper, Robert Lahoud, Georgina Lus-combe, and Hilmy Salih, and Lawrence Young(DakoAustralia)andFrankManconifortheirex-pert assistance and advice. REFERENCES 1.  Tokushige N, Markham R, Russell P, FraserIS.Highdensityofsmallnervefibresinthefunc-tional layer of the endometrium in women withendometriosis. Hum Reprod 2006;21:782-7. 2.  Tokushige N, Markham R, Russell P, FraserIS. Nerve fibres in peritoneal endometriosis.Hum Reprod 2006;21:3001-7. 3.  Tokushige N, Markham R, Russell P, YoungL,FraserIS.Differenttypesofsmallnervefibersin the eutopic endometrium of women with en-dometriosis. Fertil Steril 2007 (in press). 4.  Endometriosis Association of Victoria (EAV).Members survey on endometriosis symptomsand delay in diagnosis. Available from: http:// www.endometriosis.org.au. Accessed 1989. 5.  Dmowski WP, Lesniewicz R, Rana N, Pep-ping P, Noursalehi M. Changing trends in thediagnosis of endometriosis: a comparativestudy of women with pelvic endometriosis pre-sentingwithchronicpelvicpainorinfertility.Fer-til Steril 1997;67:238-43. 6.  Husby GK, Haugen RS, Moen MH. Diagnos-ticdelayinwomenwithpainandendometriosis. Acta Obstet Gynecol Scand 2003;82:649-53. 7.  Arruda MS, Petta CA, Abrao MS, Benitti-Pinto CL. Time elapsed from onset of symp-toms to diagnosis of endometriosis in a cohortstudy of Brazilian women. Hum Reprod2003;18:756-9. 8.  American Society for Reproductive Medi-cine. Revised American Society for Reproduc-tive Medicine classification of endometriosis.Fertil Steril 1996;67:817-21. 9.  Marchino GL, Gennarelli G, Enria R, Bong-ianniF,LipariG,MassorbioM.Diagnosisofpel-vic endometriosis with use of macroscopic ver-sus histologic findings. Fertil Steril2005;84:12-5.  www.AJOG.org   General Gynecology  Research DECEMBER 2007  American Journal of Obstetrics &  Gynecology  578.e4
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks