Arch Gynecol Obstet (2012) 285:839–843 DOI 10.1007/s00404-011-2068-6 GYNECOLOGIC ONCOLOGY Role of hysteroscopy in the detection of endometrial pathologies in women presenting with postmenopausal bleeding and thickened endometrium Amr K. Elfayomy ã Fawzia A. Habib Mohamed A. Alkabalawy ã Received: 14 May 2011 / Accepted: 11 August 2011 / Published online: 26 August 2011 Ó Springer-Verlag 2011 Abstract Objectives The goal of this study was to define the diagnostic value of hysteroscopy in eva
of 6
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
  GYNECOLOGIC ONCOLOGY Role of hysteroscopy in the detection of endometrial pathologiesin women presenting with postmenopausal bleeding and thickenedendometrium Amr K. Elfayomy  ã Fawzia A. Habib  ã Mohamed A. Alkabalawy Received: 14 May 2011/Accepted: 11 August 2011/Published online: 26 August 2011   Springer-Verlag 2011 Abstract Objectives  The goal of this study was to define thediagnostic value of hysteroscopy in evaluating uterinecavity compared to endometrial biopsy in women pre-senting with postmenopausal bleeding and thick endome-trial mucosa with particular attention to endometrialhyperplasia and carcinoma.  Methods  Eighty-three consecutive women presentingwith postmenopausal bleeding and endometrial thicknessof 5 mm or more measured by transvaginal ultrasound(TVU) were enrolled in a prospective study between May2008 and July 2010. They underwent diagnostic hysteros-copy and endometrial biopsy. Hysteroscopic data wascompared with the final diagnosis established by histo-logical examination.  Results  The women’s mean age was 61.2  ±  5.2 years(range 44–80). The most frequent endometrial lesion wasendometrial polyps (31.1%). Hyperplastic endometriumwas confirmed in 23 (27.8%), only 13 cases were suspectedby the hysteroscope. Out of the 14 (16.9%) proven cases of endometrial cancer, only half of the cases were suspected.In benign endometrial lesions, the sensitivity of thehysteroscopic view was 94.7%, specificity was 97.8%,positive (PPV) and negative (NPV) predictive values were97.3 and 95.7%, respectively. On the other hand, hyster-oscopy demonstrated an overall sensitivity, specificity,PPV, and NPV of 56.5, 91.6, 72.2, and 84.6%, respectively,in endometrial hyperplasia, whereas the same parametersfor endometrial cancer were 50, 94.2, 63.6, and 90.2%. Conclusion  Hysteroscopy can be used as the first linediagnostic tool for evaluating the benign endometriallesions, such as endometrial polyp and submucosal myoma,nonetheless hysteroscopy has poor validity for excludingendometrial hyperplasia and cancer in women presentingwith the postmenopausal bleeding and thick endometrium. Keywords  Postmenopausal bleeding    Diagnostichysteroscopy    Endometrial biopsy    Endometrial neoplasia Introduction The main objective in the diagnostic work up of post-menopausal women presenting with abnormal uterinebleeding was to detect or rule out endometrial cancer.Approximately 90% of women with endometrial carci-noma report vaginal bleeding as their only complaint, sothis symptom should always be carefully investigated.However, just 10–15% of women with postmenopausalbleeding have endometrial carcinoma [1–3]. Hysteroscopy is the most appropriate method to assessuterine cavity in postmenopausal women with sonog-raghically thickened endometrium with or without symp-toms. It is a dynamic exam and allows a direct view of theendometrium. Its main advantage is that biopsies arepossible, which improves the diagnostic accuracy [4, 5]. The hysteroscopic macroscopic diagnosis in the case of  A. K. ElfayomyDepartment of Obstetrics and Gynecology,Faculty of Medicine, Zagazig University,Zagazig, EgyptA. K. Elfayomy ( & )    F. A. HabibDepartment of Obstetrics and Gynecology,Faculty of Medicine, Taibah University,Al-Madinah Al-Munawarah, Saudi Arabiae-mail: amr.fyomy@yahoo.comM. A. AlkabalawyDepartment of Pathology, Faculty of Medicine,Taibah University, Al-Madinah Al-Munawarah, Saudi Arabia  1 3 Arch Gynecol Obstet (2012) 285:839–843DOI 10.1007/s00404-011-2068-6  (pre)-malignant uterine cavity abnormalities must beaccurate enough and histological sampling must be reli-able especially in case of endometrial polyps [6]. Themost important benefit of hysteroscopy is its ‘‘see andtreat’’ potential, which not only avoids multiple hospitalvisits, but also provides higher patient satisfaction [7]. Theobjective of this prospective study was to evaluate thediagnostic value of hysteroscopy in the detection of endometrial lesions compared with the histologic resultsof endometrial biopsy in postmenopausal women withuterine bleeding and endometrial thickness of 5 mm ormore with special attention to endometrial hyperplasia andcarcinoma. Patients and methods Women with postmenopausal uterine bleeding attendingthe Obstetrics and Gynecology Department, Ohud Hospi-tal, one of the Taibah University Hospitals, Al-MadinahAl-Munawarah Province, Saudi Arabia, between May 2008and July 2010 were enrolled in a prospective study. Thesepatients were consecutively recruited from gynecologicoutpatient clinics, emergency departments or directly fromgeneral practice for investigation of postmenopausalbleeding. Menopause was defined as spontaneous cessationof menses for 12 months or more. The exclusion criteriawere use of hormonal replacement therapy, obvious causeof bleeding from cervix and vagina, surgical menopause orTVU showing adnexal pathology. After complete medical,surgical history and a thorough physical examination eachpatient gave informed consent to participate in the study.The study protocol was approved ‘‘Medical and HealthSciences Research Committee Involving Human Subjectsof Ohud hospital’’ which conforms to the provisions of theDeclaration of Helsinki.A double-layer measurement of the endometrium at itsthickest part in the longitudinal plane was performed byTVU using Toshiba SSA 270A/HG Tokyo Japan, vaginalprobe 7.5 MHz. Endometrial thickness was measured asthe maximal distance between the two myometrial inter-faces in a longitudinal scan. When ET \ 5 mm, the patientwas reassured but instructed to contact her physician whenbleeding recurs. A cut-off level of the endometrial thick-ness in the postmenopausal women in this study waschosen at  C 5 mm. The adnexal region was also coveredduring TVU to exclude extrauterine pelvic masses. Out of the 147 women with abnormal postmenopausal bleedingevaluated for eligibility in this study, 52 were excludedowing to endometrial thickness \ 5 mm, hysteroscopy wasimpossible to perform in 9 women due to cervical stenosis,and 3 patients abandoned the study; only 83 women hadcompleted the protocol.Hysteroscopy was performed in our gynaecologicalendoscopy unit in outpatient clinic with a Storz Hamou IImicro-hysteroscope (4 mm diameter, 30   angle and 5.0 mmsheath). No local anaesthesia or systemic drugs were givento perform hysteroscopies. Sodium chloride was used as adistension medium. Illumination was provided by a Storzcold light source via a fibre-optic cable. The hysteroscopewas guided through the endocervical canal into the uterinecavity under visual control. The cavity and endometrialsurface were inspected systematically, and the tubal ostiaidentified by rotation of the hysteroscope. Hysteroscopywas defined as completed when the entire uterine cavity wasvisualized. If hysteroscopic features showed uniform non-vascular endometrium then there no abnormality consid-ered. Hyperplastic endometrium looks vascular andthickened in polypoid appearance, when pressed with hys-teroscope, endometrial groove is seen. Abnormal vascu-larization and irregular friable polypoid formations withnecrosis or bleeding were considered suspicious endome-trial carcinoma [8, 9]. Guided biopsies were performed in all the patients with suspected alterations. Four sampleswere taken (one for each uterine wall) when the endome-trium had homogeneous characteristics using a 3 mmstainless steel Novak curette after removal of the scope. Thebiopsied material was placed immediately in 10% formal-dehyde and sent to the pathology laboratory. Oral nonste-roidal anti-inflammatory drugs (NSAIDs) were prescribedwhenever analgesia was required following it.Histological diagnosis is taken as a gold standard todetermine the efficacy of hysteroscopy in diagnosingendometrial pathologies. Evaluation of predictive power of hysteroscopy in diagnosing various endometrial lesionswas based on sensitivity, specificity, positive (PPV), andnegative (NPV) predictive values according to Bayes’theorem [10, 11]. Results The mean age of 83 women participating in the study was61.2  ±  5.2 years (range 44–80 years) and they were post-menopausal for an average of 11.6  ±  8.4 years (range1–31 years). The mean endometrial thickness 10.51  ± 4.92 mm ranged from 5 to 29 mm.Endometrial polyps was the most common hystero-scopic pathologies observed in 25 cases (30.1%), whileendometrial hyperplasia and carcinoma diagnosed visuallyby the hysteroscopist in 18 (21.7%) and 11 cases (13.3%),respectively. The hysteroscopic findings and the histolog-ical diagnoses were compared in 83 cases (Table 1). Of note, hyperplastic endometrium was confirmed by histol-ogy in 23 (27.8%), 13 cases had been suspected by thehysteroscope, the other 10 cases were not suspected during 840 Arch Gynecol Obstet (2012) 285:839–843  1 3  visual diagnosis of the uterine cavity. On the other hand,the hysteroscopic findings were not suspicious for malig-nancy in seven cases proved by histological examination.All six myomas at hysteroscopy were confirmed by biopsy.Because the endometrial polyps was the most commonhysteroscopic pathologies in our series, we follow thedefinitive histological examination after total removal of the polyp, proved the presence of an endometrial carci-noma hidden in either stalk or centre of 5 (20%) out of 25polyps. Outpatient hysteroscopy was a feasible and well-tolerated technique. All women for whom cervical stenosisand hysteroscopy could not be performed were excluded.We did not report adverse experiences, such as uterineperforation or failure to visualize the uterine cavity duringthe performance of all the hysteroscopies.Using pathologic diagnoses as gold standard, the cal-culated sensitivity, specificity, PPV, and NPV showinggood hysteroscopic performance in visual diagnosis of theoverall benign uterine avity abnormality. However, thesame parameters showing its limited value for the detectionof endometrial hyperplasia and carcinoma (Table 2). Discussion Abnormal postmenopausal bleeding remains an importantreason for referral in general gynaecological practice. Incurrent clinical practice, measurement of double-layerendometrial thickness by transvaginal sonography (TVS) isused to determine whether further endometrial sampling isrequired [12, 13]. The cut-off value for endometrial thickness beyond which further investigation can be rec-ommended is beyond 5 mm [14–16]. Hysteroscopy has emerged as a useful diagnostic pro-cedure for investigating postmenopausal bleeding that issafe with a low incidence of clinically significant compli-cations. In the presence of intracavitary lesion, hysteros-copy allows for the visualization of the probable uterinesource of bleeding, improving the chance that the tissueobtained by directed biopsy will yield an accurate histo-logic diagnosis [16].In this study the most frequent hysteroscopic patholo-gies in patients with postmenopausal bleeding and thick-ened endometrium were most often represented byendometrial polyps (30.1%). Our results agree with thosereported in the literature [17, 18]. Nevertheless, histologi- cal examination, after total removal of these polyps, provedthe presence of an endometrial carcinoma hidden in eitherstalk or centre of 5 (20%) out of 25 polyps despite negativebiopsy. Endometrial cancer may be overlooked in the casesof focal neoplasia within endometrial polyps or in settingswhen endouterine visualization is suboptimal [19], there-fore it is not safe to consider the identification of endo-metrial polyps by hysteroscopic visualization or even afterbiopsy to be conclusive in postmenopausal women withabnormal uterine bleeding; this leads to the discussion Table 1  Comparison between hysteroscopic and histologic findingsHysteroscopy Histology Total  N  (%)Atrophy Polyp Myoma Hyperplasia Carcinoma OtherNo abnormality 2 – 6 3 6 b 17 (20.5)Atrophy 6 – – – – 1 a 7 (8.4)Polyp – 25 – – – – 25 (30.1)Myoma – – 5 – – – 5 (6)Hyperplasia – – – 13 4 1 a 18 (21.7)Carcinoma – – – 4 7 – 11 (13.3)Total 8 (9.6) 25 (30.1) 5 (6) 23 (27.8) 14 (16.9) 8 (9.6) 83 (100) a Insufficient endometrium b Endometritis Table 2  Predictive power of hysteroscopic view in diagnosing various endometrial lesionsHysteroscopic findings Sensitivity Specificity PPV NPVOverall benign abnormality (%) 94.7 (36/38) 97.8 (44/45) 97.3 (36/37) 95.7 (44/46)Endometrial hyperplasia (%) 56.5 (13/23) 91.6 (55/60) 72.2 (13/18) 84.6 (55/65)Carcinoma (%) 50 (7/14) 94.2 (65/69) 63.6 (7/11) 90.2 (65/72)Values in parenthesis are in ( n  /  n ) PPV   positive predictive value,  NPV   negative predictive valueArch Gynecol Obstet (2012) 285:839–843 841  1 3  whether an endometrial polyp should always be resected.In this regard, the removal of polyps under local anesthesiais an option following diagnostic hysteroscopy. In ouropinion, complete removal of such a polyp using D & Ccan be technically difficult even after the previous hyster-oscopy. Hysteroscopy-controlled extraction was superiorespecially in those patients with an endometrial thickness of  [ 10 mm [20].Endometrial hyperplasia may produce obvious spaceoccupying lesions in which diagnosis is easy with hyster-oscopy, but it may be not obvious especially in early stagesof the disease [21].The diagnosis of endometrial hyperplasia proved not tobe easy during diagnostic hysteroscopy. In our series onlyabout two-third of the cases, in which the hyperplasia wassuspected could be confirmed histologically. On the otherhand histology showed the presence of hyperplasia tentimes when this was not suspected during hysteroscopy.Similar study observed that hysteroscopy did not improvesensitivity compared to D & C in case of hyperplasia. Toovercome the risk of missing hyperplasia, the authorssuggest always to take biopsies of the endometrium whendiagnostic hysteroscopy is performed [22].On the other hand, endometrial carcinoma cannot bereliably detected by hysteroscopy in our series. Out of the14 (16.9%) patients with endometrial cancer, 7 had beensuspected by hysteroscopy and 7 cases with no abnormalitydetected in the cavity or endometrial hyperplasia. Thespecificity of visual diagnosis in detecting endometrialcancer seems to be low and not all the endometrialmalignancies will be detected. This is according to theopinion in literature reported the limited role of diagnostichysteroscopy in establishing the diagnosis of endometrialcarcinoma [22]. Indeed, hysteroscopy without endometrialbiopsy is unreliable in the diagnosis of pre-malignant andmalignant disease in the uterine cavity [23, 24]. In this current study, direct visualization of uterinecavity by the hysteroscope was superior in detectingendometrial polyps, submucous fibroids, and endometrialatrophies. These entities represent unique conditions wherea hysteroscopic diagnosis can be clearly established. In ourstudy, the calculated sensitivity, specificity, PPV, and NPVvalues regarding visual diagnostic hysteroscopy for theselesions were quite compatible with others [25, 26]. On the other hand, there was inadequate performance of hyster-oscopy as a diagnostic tool in the case of hyperplasia andcarcinoma compared with the other endometrial lesions.In endometrial hyperplasia, hysteroscopy showed anoverall sensitivity, specificity, PPV, and NPV of 56.5, 91.6,72.2, and 84.6%, respectively. There are some statisticallimitations due to the small number of observationsregarding the diagnostic usefulness of hysteroscopy for thepatients with postmenopausal bleeding and thickenedendometrium. A previous published study [27] evaluatedthe diagnostic accuracy of hysteroscopic view in endome-trial hyperplasia among postmenopausal patients. The sen-sitivity, specificity, PPV, and NPV of hysteroscopic view toanticipate hyperplasia were 61.6, 95.2, 79.4, and 89.3%,respectively, and reported that the current hysteroscopiccriteria suggesting endometrial hyperplasia are inaccurate;in order to exclude hyperplasia, a pathologic assessment iswarranted in all hysteroscopies showing an irregularly linedendometrium, while another study under similar conditionsdemonstrated that the sensitivity of the hysteroscopic viewfor hyperplasia was 58.33% and the specificity was 98.45%,the PPV was 63.64% and the NPV was 98.07% [16]; theseresults are consistent with our findings.Considering endometrial carcinoma in our series, hys-teroscopic view parameters revealed values of sensitivity50%, specificity 94.2%, PPV 63.6%, and NPV 90.2%.Other similar study showed a sensitivity and positive pre-dictive values of hysteroscopy in diagnosing endometrialcarcinoma were only 58.8 and 20.8%, respectively [28].Moreover, a recent study reported that, the sensitivity of the hysteroscopy in endometrial cancer recognition was61%, whereas the specificity of this method was 90% andconcluded that the diagnostic criteria of hysteroscopicimages in endometrial cancer are not perfect [29]. There-fore, our study as well as other researchers showed a lowdiagnostic ability of hysteroscopy in the detection of endometrial cancer.In conclusion, the findings of this study showed thathysteroscopy is a safe and reliable diagnostic procedure forevaluating benign endometrial lesions, such as endometrialpolyp and submucosal myoma, in view of poor validity toexclude endometrial hyperplasia and cancer in our series,we recommend always to perform diagnostic hysteroscopycomplemented with endometrial microprobe biopsy(Micro-curettage) in women presenting with postmeno-pausal bleeding and endometrial double-layer 5 mm ormore measured by transvaginal ultrasound even if noabnormalities were seen by hysteroscopy. In addition, thistrial demonstrated that 20% of resected polyps had endo-metrial carcinoma after histological examination despitenegative biopsy. With regard to these data, it seems rationalto remove the whole macroscopic polyp for a histologicalexamination in all the postmenopausal patients with uterinebleeding. Conflicts of interest  We declare that we have no conflict of interest. References 1. Bakour SH, Dwarakanath LS, Kahn KS, Newton JR, Gupta JK (1999) The diagnostic accuracy of ultrasound scan in predicting842 Arch Gynecol Obstet (2012) 285:839–843  1 3
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