IMR Press / EJGO / Volume 40 / Issue 1 / DOI: 10.12892/ejgo4256.2019
Open Access Original Research
Role of transvaginal ultrasound in finding endometrial cancer in postmenopausal women with or without bleeding
Show Less
1 University Medical Center, Ljubljana
2 General Hospital, Trbovlje, Slovenia
Eur. J. Gynaecol. Oncol. 2019 , 40(1), 60–64;
Published: 10 February 2019

Purpose of Investigation: To determine the endometrial thickness (ET) threshold to predict high risk of endometrial cancer (EC) in symptomatic and asymptomatic postmenopausal patients. It is also assessed whether additional ultrasonic findings (vascularisation, myometrial invasion, and inhomogeneity) improves diagnostic accuracy. Materials and Methods: In a prospective study the authors included 447 postmenopausal patients, treated for endometrial pathology. Ultrasound assessment was followed by endometrial sampling and histopathological examination. Results were compared according to presence of symptoms and histopathological findings (benign and malignant). Results: Eighty-six percent (385) patients were symptomatic and 14% (62) were asymptomatic. In the symptomatic patients with ET ≥ 5 mm, 17.5% had EC (CI: 13.5 -22); there was no patient with EC and ET less than 5 mm (CI: 0-15). All asymptomatic patients with ET ≥ 11 mm had benign findings. Three asymptomatic patients with malignancy were discovered because of other suspicious findings than ultrasound. Additional ultrasonic assessment showed statistically relevant results in symptomatic group (p < 0.05). Conclusions: In symptomatic women, less than 5 mm ET threshold rules out patients with EC with good certainty. In asymptomatic patients, ET measurement played no role in discovering EC.

Endometrial cancer
Postmenopausal bleeding
Endometrial thickness

More than one in 20 female cancers in Europe affect the endometrium, and in many countries the incidence is increasing [1]. According to Slovene Cancer Registry, crude rate of endometrial cancer was 30.2 per 100,000 women in 2016. In the majority of patients (75%), the disease is still confined to uterus at time of diagnosis. Approximately 80% of endometrial carcinomas arise in post-menopause, and 90% of cases present with vaginal bleeding [2].

Postmenopausal bleeding is a common clinical problem; the leading cause is atrophic endometrium (59%), followed by endometrial polyps (12%), endometrial carcinoma (10%), hyperplasia (9.8%), other causes are less common [3]. Transvaginal ultrasound is a non-invasive method that can help select patients for further, more invasive diagnostic and therapeutic procedures [4]. Thin endometrium is considered as normal while thick endometrium can represent a polyp, hyperplasia, or carcinoma [4]. Postmenopausal patients with bleeding and thick endometrium require biopsy and histopathological verification [5], while patients with thin endometrium require biopsy after second bleeding.

In asymptomatic postmenopausal patients, management of endometrial changes, presented at ultrasound imaging, remains controversial. Although routine ultrasound scanning for endometrial thickness as screening for endometrial cancer is not recommended [6,7], a cut-off value of endometrial thickness of 11 mm is suggested as the threshold to instigate further investigations [8].

Besides endometrial thickness, other ultrasonic characteristics (vascularisation, tissue texture, presence of fluid in the cavity, and estimation of myometrial invasion) are used to describe endometrial changes and researched in possible ultrasonic prediction models for endometrial cancer [9,10].

The present research focused on estimating risk of endometrial malignancy, considering presence of symptoms (bleeding) and endometrial thickness, and established if additional ultrasonic assessment can improve diagnostic value of ultrasound.

Materials and Methods

In a prospective study from January to December 2014, the authors included 447 postmenopausal patients treated for endometrial pathology at Gynaecological department of University medical centre Ljubljana. Patients with vaginal bleeding were assigned to symptomatic and all others to asymptomatic group. Prior to planned diagnostic procedure, 352 patients had transvaginal ultrasound with endometrial thickness measurement. Additional to endometrial thickness, protocol included estimation of endometrial vascularisation (any presence of vascularisation of endometrial changes visible with power Doppler or none), myometrial invasion (subjective impression of disturbed endometrial- myometrial border), and subjective impression (suspicious/unsuspicious, based on findings on homogeneity, and inhomogeneous changes were considered as malignant). Measurements were performed by several doctors with various range of experience in sonography (mostly trainees) on various ultrasound machines. Diagnostic procedures included dilation and curettage, office hysteroscopy, operative hysteroscopy, and aspirational biopsy and tissues were sent for histopathological evaluation. Symptomatic and asymptomatic patients were further divided in group with benign (atrophy, polyp, hyperplasia without atypia) and malignant changes of endometrium (hyperplasia with atypia, low grade, high grade carcinomas). Results were statistically analysed.

A chi-square test was used to identify risk factors for malignancy. Odds ratios (OR) and 95% confidence intervals (95% CI) were calculated with two-sided probability (p) values, and p-value of < 0.05 was considered significant. Statistical analysis was performed using SPSS Statistics, version 21.


The authors analysed data of 447 postmenopausal patients; 385 (86%) patients were symptomatic and 62 (14%) asymptomatic. Average age was 62.7 years (45 to 92), average time from menopause to intervention was 11.2 years (SD 9.8) in symptomatic and 14.3 years (SD 8.1) in asymptomatic patients; 27 patients after hormonal replacement therapy (25 symptomatic, 2 asymptomatic) and 20 during therapy with tamoxifen (14 symptomatic, 6 symptomatic).

Histopathology revealed malignant changes in 54 (14%) symptomatic and three (6%) asymptomatic women. Results are shown in Table 1. Most common histopathological finding was endometrial polyp (198 patients, 49.3%), followed by endometrial atrophy (119 patients, 29.6%). Fifty-one patients had endometrial carcinoma and six hyperplasia with atypia. In patients with carcinoma, 36 had endometrioid adenocarcinoma, six serous, one mucinous adenocarcinoma, six mixed carcinomas, one primary squamous cell carcinoma, and one malignant tumour with neuroendocrine differentiation.

Table 1Incidence of malignant and benign changes of endometrium in symptomatic and asymptomatic postmenopausal women.
Histopathological finding Symptomatic patients Asymptomatic patients Total number
Endometrial atrophy 113 (32.3%) 6 (11.5%) 119 (29.6%)
Polyp 157 (44.9%) 41 (78.8%) 198 (49.3%)
Hyperplasia without atypia 26 (7.4%) 2 (3.8%) 28 (7.0%)
Hyperplasia with atypia 6 (1.7%) 0 (0.0%) 6 (1.5%)
Carcinoma 48 (13.7%) 3 (5.8%) 51 (12.7%)

Data on endometrial thickness on transvaginal ultrasound was available for 352 patients. In symptomatic patients, there was no malignant finding on histopathology if edometrial thickness was less than 5 mm (CI: 0-15%), while in patients with endometrial thickness 5 mm or more histopathology showed malignancy in 17.5% (49/280) patients (CI: 13.5-22%), as shown in Table 2.

Table 2Endometrial thickness in symptomatic patients according to histopathological findings.
Endometrial thickness Benign HP Malign HP p value
Less than 5 mm 21 (100.0%) 0 (0.0%) <0.05
5 mm or more 231 (82.5%) 49 (17.5%)

HP = histopathological finding.

In asymptomatic patients, histopathology showed malignancy in 5.9% (3/51) cases; however all patients with malignancy had endometrial thickness less than 11 mm, as shown in Table 3. In asymptomatic patients with endometrial thickness less than 11 mm, chances for malignancy were 21.4% (CI: 7-47%); there was no asymptomatic patient with malignancy and endometrial thickness 11 mm or more (CI: 0-9%).

Table 3Endometrial thickness in asymptomatic patients according to histopathological findings.
Endometrial thickness Benign HP Malign HP p value
Less than 11 mm 11 (78.6%) 3 (21.4%) NS
11 mm or more 37 (100.0%) 0 (0.0%) NS

HP= histopathological finding, NS = not significant.

The authors analysed the correlation between histopathological finding (benign/malign), endometrial thickness, and occurrence of symptoms (bleeding). In the symptomatic group, women with malignant histopathological findings had thicker endometrium (11 mm and more) in higher percentage than women with benign findings (84.0% vs. 44.1%). Symptomatic women with endometrial thickening 11 mm and more had 6.66 (95% CI: 3.02, 14.68) higher risk for malignant histopathological finding than women with endometrial thickness less than 11 mm. In asymptomatic group, all women with malign histopathology had endometrium less than 11 mm thick. Besides endometrial thickness, ultrasonic assessment with transvaginal ultrasound comprised of additional ultrasonic features: suspicion of myometrial invasion (disturbed endometrial - myometrial border), presence of vascularisation in endometrial changes with Doppler ultrasound and subjective impression (suspicious, unsuspicious, according to tissue texture). Presence of vascularisation with Doppler US was estimated in 152 patients, suspicion of invasion into myometrium in 198 patients, and subjective impression of malignancy according to inhomogeneity of tissue texture was estimated in 418 patients. In symptomatic group, patients were more likely to have malignant changes of endometrium if subjective impression was suspicious (50.8% vs. 6.2%), invasion into myometrium was suspected (45.7% vs. 12.7%) and vascularisation of tissue was present (54.5% vs. 6.1%). In asymptomatic group, additional ultrasonic assessment showed no significant results. Statistical analysis of additional ultrasonic assessment is shown as comparison between patients with benign and malign histopathology for symptomatic and asymptomatic group in Table 4 and 5.

Table 4Additional ultrasonic assessment on TVUS in symptomatic patients
Ultrasonic feature Benign HP Malign HP p value
Subjective impression Suspicious 30 (49.2%) 31 (50.8%) < 0.05
Unsuspicious 285 (93.8%) 19 (6.2%) (< 0.001)
Suspicion of myometrial invasion Yes 25 (54.3%) 21 (45.7%) < 0.05
No 110 (87.3%) 16 (12.7%) (< 0.001)
Presence of vascularisation in endometrial changes Yes 5 (45.5%) 6 (54.5%) < 0.05
No 108 (93.9%) 7 (6.1%) (< 0.001)

HP= histopathological finding.

Table 5Additional ultrasonic assessment on TVUS in asymptomatic patients.
Ultrasonic feature Benign HP Malign HP p value
Subjective impression Unsuspicious 44 (95.7%) 2 (4.3%) NS
Suspicious 7 (100.0%) 0 (0.0%)
Suspicion of myometrial invasion Yes 4 (100.0%) 0 (0.0%) NS
No 20 (90.9%) 2 (9.1%)
Presence of vascularisation in endometrial changes Yes 2 (100.0%) 0 (0.0%) NS
No 24 (100.0%) 0 (0.0%)

HP= histopathological finding, NS= not significant.


In this study group, endometrial carcinoma was discovered in 14% (54/385) of symptomatic and 5% (3/62) of asymptomatic patients. Patients with malignancy presented with vaginal bleeding in 94.6% cases.

In symptomatic patients, chances of malignancy were very low (CI: 0-15%) if endometrial thickness was less than 5 mm and 17.5% (CI: 13.5-22%) if endometrial thickness was 5 mm or more. Gupta et al. in their meta-analysis of 57 studies suggested that ≤ 5 mm threshold should rule out endometrial pathology with good certainty [11]. In the present study one patient with malignancy and endometrial thickness of 5 mm would have been missed. However Fer-razi et al. suggested a ≤ 4 mm cut-off to predict endometrial atrophy as the cause for bleeding, with 99% negative predictive value and 98% sensitivity in 930 patients. Endometrial cancer prevalence in his group was 11.5% and 49.2% atrophy (29.6% in the present group) [12]. Timmermans et al. in their meta-analysis suggest 3 mm cut-off value for exclusion of endometrial carcinoma in women with postmenopausal bleeding with 98% sensitivity [13]. Billingsley et al. found that 27.5% of women with endometrial cancer type II had endometrial thickness ≤ 5 mm or indistinct, so they suggested that all patients with postmenopausal bleeding should be investigated, regardless of endometrial thickness [14]. Eighty to 90 percent of endometrial cancers are type I (low grade endometrial carcinomas, that are estrogen dependant). Mostly they are more hyperechoic and have tendency for exofitic growth into the cavity. Other 10-20% of endometrial cancers are type II (high grade endometrial cancers that are estrogen independent and usually presented in older patients) [15]. Type II cancers are more aggressive and have more tendency for early invasion in the myometrium, sometimes leaving endometrium thin in appearance, which further compromises ultrasonic assessment. Ultrasonic features of the two types of carcinomas are different and cannot be simplified.

In the present authors’ experience, < 5 mm threshold confirms endometrial cancer with good certainty. To avoid unnecessary procedures, invasive diagnostic procedures in patients with thin endometrium should be reserved to those with recurrent bleeding.

Numerous theoretical and retrospective data showed higher risk of malignancy in asymptomatic patients with endometrial thickness 11 mm and more. In their metaanalysis in 2004, Smith-Bindman et al. found that in asymptomatic patients, chances of endometrial cancer were 6.7% if patients had endometrial thickness 11 mm or more and 0.002% if endometrial thickness was less than 11 mm [8]. Based on this meta-analysis, asymptomatic patients with endometrial thickening 11 mm or more should undergo further investigations. The present results do not confirm that as all asymptomatic women with malign changes had endometrial thickness less than 11 mm; there was no asymptomatic patient with malignancy and endometrial thickness 11 mm or more (CI 0-9%). Lowest endometrial thickness in patients with malignancy was 6 mm in asymptomatic and 5 mm in symptomatic group. In the three patients with malignancy in asymptomatic group, causes for intervention were atypical PAP smear, suggestive of endometrial pathology in two patients, and enlarged lymph nodes with positive cytology, suggestive of endometrial carcinoma in one patient.

Breijer et al. found in their meta-analysis that there is no role for endometrial ultrasound screening in asymptomatic patients [6]. Goldstein et al. found that prevalence of asymptomatic endometrial thickening in post-menopause is 10-17% and should not trigger automatic intervention [7]. Saatli et al. evaluated retrospectively investigations of 530 women with asymptomatic endometrial thickening in post-menopause and found 0.9% incidence of adenocarcinoma and 12.2% of simple or complex atypical hyperplasia; mean endometrial thickness was 8.7 (ranging from 6-26) mm [16]. Giannella et al. in their prospective study in 268 asymptomatic women found that endometrial cut-off value ≥ 10 mm did not miss any endometrial cancer patients [17]. Gambacciani et al. found in their group of 148 asymptomatic patients only one patient with carcinoma based on ultrasonic endometrial thickening measurement alone [18]. Famuyide et al. found two malignant cases in 154 asymptomatic patients, however both patients had endometrial thickness 17 mm or more [19]. Menzies et al. reported in their study two cases of malignancy from 142 asymptomatic patients; their average thickness was 17.5 mm [20], and Ferrazi et al. suggested 15 mm as threshold to exclude endometrial cancer in asymptomatic women with polyps in post-menopause [21].

A possible explanation for the present different results is small number of asymptomatic patients (62 asymptomatic vs. 385 symptomatic patients). Based on the present study results, endometrial thickness measurement in asymptomatic patients plays no role in discovering endometrial cancer, as all patients with malignant changes had other causes for intervention. All other patients had benign changes, regardless of endometrial thickness. Although endometrial cancer can occur in asymptomatic women, the role of ET measurement and threshold value still needs to be determined in this group to avoid unnecessary procedures and possible complications.

Statistical analysis of additional ultrasound assessment (suspicion of myometrial invasion, presence of vascularisation, subjective impression of malignancy) showed statistically relevant results (p < 0.05) in symptomatic patients. However results were not accurate enough to be of value in clinical practice, which the present authors attribute to large number of investigators, their lack of special expertise in that field, and to lower quality of ultrasound machines used. Other authors [9,10] have shown that various models including endometrial thickness and power Doppler with or without other ultrasonic and clinical features (percentage of vascularisation on power Doppler, estimation of colour content, interrupted endo-myometrial junction, irregular surface at gel infusion sonography, age, and body mass index) perform well in predicting endometrial cancer, however models were used only in symptomatic patients and their results cannot be extrapolated to asymptomatic group.

Endometrial thickness < 5 mm in symptomatic patients ruled out endometrial cancer with good certainty in the present clinical setting. Number of asymptomatic patients included was low and endometrial thickness measurement played no role in discovering endometrial cancer; however symptomatic women with malignancy had an endometrium thicker than 11 mm in a higher percentage than asymptomatic women. Ultrasonic endometrial thickness measurement does play a role in diagnosing endometrial cancer, but more data should be collected to establish its role in asymptomatic women. The present data analysis of additional ultrasonic assessment showed statistically significant results in the symptomatic group. More training and education of gynaecologists is needed to further improve diagnostic value of ultrasound. Using endometrial thickness measurement with additional ultrasonic features and other risk factors were proven to be a successful prediction model for endometrial cancer in symptomatic women [9,10].


In post-menopause, leading cause to induce further investigations in diagnosing endometrial cancer remains vaginal bleeding. In symptomatic women with one episode of bleeding and endometrial thickness less than 5 mm, chances of endometrial cancer are low according to the present results. Symptomatic women with endometrial thickness 11 mm and more have higher risk for endometrial cancer than women with thinner endometrium. In symptomatic patients, presence of vascularization of tissue, inhomogeneity of tissue, and disturbed endometrial-myometrial borders are strongly suggestive of malignant findings. In asymptomatic women, risk for endometrial cancer is low; in the present study, endometrial thickness measurement played no role in discovering endometrial cancer in asymptomatic group.

Bray F., Dos Santos Silva I., Moller H., Weiderpass E.: “Endometrial cancer incidence trends in Europe: underlying determinants and prospects for prevention”. Cancer Epidemiol. Biomarkers Prev., 2005, 14, 1132.
Wolfman W., Leyland N., Heywood M., Singh S.S., Rittenberg D.A., Soucy R., et al.: “Asymptomatic endometrial thickening”. J. Obstet. Gynaecol. Can., 2010, 32, 990.
Goodman A.: “ Postmenopausal uterine bleeding”. In: UpToDate. Waltham (MA): UpToDate, 2016 [cited 5 May 2016]. Available from:
Smith-Bindman R., Kerlikowske K., Feldstein V.A., Subak L., Scheidler J., Segal M., et al.: “Endovaginal ultrasound to exclude endometrial cancer and other endometrial abnormalities”. JAMA, 1998, 280, 1510.
Amant F., Mirza M.R., Koskas M., Creutzberg C.L.: “Cancer of the corpus uteri”. Int. J. Gynaecol. Obstet., 2012, 119, S110.
Breijer M.C., Peeters J.A., Opmeer B.C., Clark T.J., Verheijen R.H., Mol B.W., et al.: “Capacity of endometrial thickness measurement to diagnose endometrial carcinoma in asymptomatic postmenopausal women: a systematic review and meta-analysis”. Ultrasound. Obstet. Gynecol., 2012, 40, 621. 10.1002/uog.12306
Goldstein S.R.: “Significance of incidentally thick endometrial echo on transvaginal ultrasound in postmenopausal women”. Menopause, 2011, 18, 434. 10.1097/gme.0b013e31820ad00b5aa9af4a-925a-4087-81a3-441782aed5fe
Smith-Bindman R., Weiss E., Feldstein V.: “How thick is too thick? When endometrial thickness should prompt biopsy in postmenopausal women without vaginal bleeding”. Ultrasound Obstet. Gynecol., 2004, 24, 558. 10.1002/uog.170415386607
Opolskiene G., Sladkevicius P., Valentin L.: “Prediction of endometrial malignancy in women with postmenopausal bleeding and sonographic endometrial thickness ≥ 4.5 mm”. Ultrasound Obstet. Gynecol., 2011, 37, 232. 21061264
Dueholm M., Moller C., Rydbjerg S., Hansen E.S, Ortoft G.: “An ultrasound algorithm for identification of endometrial cancer”. Ultrasound Obstet. Gynecol., 2014, 43, 557. 10.1002/uog.1320524009152
Gupta J.K., Chien P.F., Voit D., Clark T.J., Khan K.S.: “Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis”. Acta Obstet. Gynecol. Scand., 2002, 81, 799. 10.1034/j.1600-0412.2001.810902.x12225294
Ferrazzi E., Torri V., Trio D., Zannoni E., Filiberto S., Dordoni D.: “Sonographic endometrial thickness: a useful test to predict atrophy in patients with postmenopausal bleeding. An Italian multicenter study”. Ultrasound Obstet. Gynecol., 1996, 7, 315. 10.1046/j.1469-0705.1996.07050315.x8774095
Timmermans A., Opmeer B.C., Khan K.S., Bachmann L.M., Epstein E., Clark T.J., et al. “Endometrial thickness measurement for detecting endometrial cancer in women with postmenopausal bleeding: a systematic review and meta-analysis”. Obstet. Gynecol., 2010, 116, 160. 10.1097/AOG.0b013e3181e3e7e820567183
Billingsley C.C., Kenne K.A., Cansino C.D., Backes F.J., Cohn D.E., O’Malley D.M., et al.: “The use of transvaginal ultrasound in type II endometrial cancer”. Int. J. Gynecol. Cancer, 2015, 25, 858. 10.1097/IGC.000000000000042325768078
Bebar S.: “Kirurško zdravljenje endometrijskega karcinoma - nerešena vprašanja”. Onkologija, 2012, XVI, 23.
Saatli B., Yildirim N., Olgan S., Koyuncuoglu M., Emekci O., Saygılı U.: “The role of endometrial thickness for detecting endometrial pathologies in asymptomatic postmenopausal women”. Aust. N. Z. J. Obstet. Gynaecol., 2014, 54, 36. 10.1111/ajo.1217424471845
Giannella L., Mfuta K., Setti T., Boselli F., Bergamini E., Cerami L.B.: “Diagnostic accuracy of endometrial thickness for the detection of intra-uterine pathologies and appropriateness of performed hysteroscopies among asymptomatic postmenopausal women”. Eur. J. Obstet. Gynecol. Reprod. Biol., 2014, 177, 29. 10.1016/j.ejogrb.2014.03.02524766900
Gambacciani M., Monteleone P., Ciaponi M., Sacco A., Genazzani A.R.: “Clinical usefulness of endometrial screening by ultrasound in asymptomatic postmenopausal women”. Maturitas, 2004, 48, 421. 10.1016/j.maturitas.2003.10.00615283934
Famuyide AO, Breitkopf DM, Hopkins MR, Laughlin-Tommaso SK.: “Asymptomatic thickened endometrium in postmenopausal women: malignancy risk”. J. Minim. Invasive Gynecol., 2014, 21, 782. 10.1016/j.jmig.2014.03.00424632398
Menzies R., Wallace S., Ennis M., Bennett A., Jacobson M., Yip G., et al.: “Significance of abnormal sonographic findings in postmenopausal women with and without bleeding”. J. Obstet. Gynaecol. Can., 2011, 33, 944. 10.1016/s1701-2163(16)35020-421923992
Ferrazzi E., Zupi E., Leone FP., Savelli L., Omodei U., Moscarini M., et al.: “How often are endometrial polyps malignant in asymptomatic postmenopausal women? A multicenter study”. Am. J. Obstet. Gynecol., 2009, 200, 235. 10.1016/j.ajog.2008.09.87619027096
Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Back to top