Evaluating myometrial and cervical invasion in women with endometrial cancer: comparing subjective assessment with objective measurement techniques

Transcription

1 Ultrasound Obstet Gynecol 2013; 42: Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: /uog Evaluating myometrial and cervical invasion in women with endometrial cancer: comparing subjective assessment with objective measurement techniques F. MASCILINI*, A. C. TESTA*, C. VAN HOLSBEKE, L. AMEYE, D. TIMMERMAN and E. EPSTEIN *Department of Oncology, Catholic University of the Sacred Heart, Rome, Italy; Department of Obstetrics and Gynecology, Catholic University, Leuven, Belgium; Department of Development and Regeneration, KU Leuven - University of Leuven, Leuven, Belgium; Department of Women s and Children s Health, Karolinska University Hospital, Stockholm, Sweden KEYWORDS: cervical invasion; endometrial cancer; measurement; myometrial invasion; ultrasound ABSTRACT Objective To compare the diagnostic accuracy of subjective ultrasound assessment with that of objective measurement techniques in the evaluation of myometrial and cervical invasion in women with endometrial cancer. Methods This was a prospective multicenter study including 144 women with endometrial cancer undergoing transvaginal ultrasound. Myometrial and cervical invasion was evaluated subjectively, as well as objectively measured in different ways: endometrial thickness, tumor/uterine anteroposterior (AP) diameter ratio, minimal tumor-free margin, minimal tumor-free margin/uterine AP diameter ratio, tumor volume (threedimensional (3D)), tumor/uterine volume (3D) ratio, and distance from outer cervical os to lower margin of tumor (Dist-OCO). Histological assessment following hysterectomy was the gold standard. Results The sensitivity (72%) and specificity (76%) of tumor/uterine AP diameter (at cut-off, 0.53) were not significantly different from those of subjective evaluation (sensitivity, 77% (P = 0.44); specificity, 81% (P = 0.32)) for the prediction of deep myometrial invasion; all other objective measurement techniques had either a significantly lower sensitivity or a lower specificity. For all objective measurement techniques, except minimal tumor-free margin/uterine AP diameter ratio, fixing the sensitivity at the same level as that of subjective evaluation (i.e. 77%) gave a significantly lower specificity. Dist-OCO was the only parameter that might have potential to predict cervical invasion; it had a non-significantly higher sensitivity than did subjective evaluation (73% vs 54%, P = 0.06), but a significantly lower specificity (63% vs 93%, P < 0.001). Conclusion Subjective assessment of cervical and myometrial invasion is as good as or better than any objective measurement technique. The tumor/uterine AP diameter ratio and minimal tumor-free margin/uterine AP diameter ratio seem to be the best objective measurement techniques to predict deep myometrial invasion. It remains to be shown if objective measurements are useful to predict cervical invasion. Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. INTRODUCTION Endometrial carcinoma is the most frequent female pelvic malignancy and the seventh most common neoplasm worldwide, with the highest incidence in North America and Europe 1. Prognosis depends on the age of the patient, histological type and grade, tumor size, depth of myometrial invasion, and presence of cervical stromal invasion and lymph node metastases 2. Tumor size, myometrial invasion, and cervical stromal involvement cannot be determined by clinical examination. Preoperative knowledge of these parameters is advantageous in order to plan treatment; the clinical challenge is to select effectively for more radical surgery, with pelvic and paraaortic lymphadenectomy, those patients with a higher risk of relapse, while avoiding overtreatment in low-risk cases. Transvaginal ultrasound examination offers the possibility to evaluate the extent of myometrial infiltration and the presence of cervical stromal invasion 3.Thereis no consensus on how best to assess myometrial invasion by subjective evaluation or by objective measurements. Several studies have been published on the accuracy of subjective assessment in the prediction of deep myometrial invasion, with sensitivities of 68 93% and specificities of Correspondence to: Dr E. Epstein, Karolinska University Hospital, Stockholm, Sweden ( elisabeth.epstein@karolinska.se) Accepted: 11 April 2013 Copyright 2013 ISUOG. Published by John Wiley & Sons Ltd. ORIGINAL PAPER

2 354 Mascilini et al % 4 7. There are fewer publications on objective measurement techniques to assess myometrial invasion. In their study on 30 women with endometrial cancer, Karlsson and co-workers 8 suggested use of the tumor/uterine anteroposterior (AP) diameter ratio, a result of > 0.5 indicating high risk of deep myometrial invasion, with a sensitivity of 79% and a specificity of 100%. Alcazar et al. 6 suggested measuring the tumor-free margin using threedimensional (3D) virtual navigation, with a value < 9mm indicating a high risk of deep myometrial invasion; with this cut-off, they aimed to detect all women with deep myometrial invasion, i.e. with a sensitivity of 100%, and thus achieved a specificity of 61%. There are very few publications on the use of ultrasound to assess cervical invasion, but the results are generally very good for subjective assessment, with sensitivities ranging from 77 to 93% and specificities ranging from 85 to 99% 5,7,9. To the best of our knowledge, no objective measurement technique has been suggested to predict the presence of cervical invasion. The aim of this study was to compare subjective assessment and objective ultrasound measurements of cervical and myometrial invasion, using histology from hysterectomy specimens as the gold standard. A.C.T. in Rome and C.V.H. in Leuven/Genk). An examination protocol was filled out by the investigator directly after the examination. The protocol included assessment of the presence of deep ( 50%) myometrial invasion (yes/no) and cervical invasion (no invasion or stromal invasion suspected) and objective measurements (Figure 1) of: tumor extension/size; endometrial thickness; METHODS This study was a four-center collaboration between the ultrasound unit at the Department of Obstetrics and Gynecology, Lund University Hospital, Lund, Sweden, the Gynecologic Oncology Unit of the Catholic University of Sacred Heart, Rome, Italy, the Department of Obstetrics and Gynecology of Leuven University Hospital, Leuven, Belgium and the Department of Obstetrics and Gynecology of Ziekenhuis Oost-Limburg, Genk, Belgium. Approval was obtained from the ethics committee of Lund University (LU ) and from the local ethics committees of the Catholic University of Sacred Heart, Leuven University and Ziekenhuis Oost-Limburg. All women gave informed consent to participate. From January 2007 until May 2009, we included in the study consecutive women with histologically confirmed (by dilatation and curettage, hysteroscopy or simple endometrial biopsy) endometrial cancer, planned for surgery (which comprised total abdominal or robotic hysterectomy, both with bilateral salpingoophorectomy with systematic pelvic and para-aortic lymphadenectomy if appropriate, according to local protocols). All women underwent transvaginal ultrasound examination within 7 days prior to surgery. During the study period, none of the participating institutions routinely used other imaging techniques, such as magnetic resonance imaging (MRI), to assess the local extent of the tumor. We decided to use the FIGO (International Federation of Gynecology and Obstetrics) 2009 staging criteria 10 post hoc, based on the hysterectomy specimen, since the final analysis was made after all women had been included. All ultrasound examinations were performed by the principal investigators of each center (E.E. in Lund, Figure 1 Objective ultrasound measurements used in this study for the evaluation of myometrial invasion in women with endometrial cancer. (a) A, uterine anteroposterior (AP) diameter; B, tumor AP diameter; C, minimal tumor-free margin. B/A = tumor/uterine AP diameter ratio; C/A = minimal tumor-free margin/uterine AP diameter ratio. (b) D, tumor volume (three-dimensional (3D)). (c) E, uterine volume (3D). D/E = tumor/uterine volume (3D) ratio.

3 Subjective assessment vs objective measurement in endometrial cancer 355 tumor/uterine AP diameter ratio; minimal tumor-free margin; minimal tumor-free margin/uterine AP diameter ratio; tumor volume (3D); tumor/uterine volume (3D) ratio; and distance from outer cervical os to lower margin of tumor (Dist-OCO). In all women we collected still images of all measurements, videoclips and 3D volumes. Women were examined in the lithotomy position and had an empty bladder. The ultrasound equipment used in Lund was a GE Voluson E8 (GE Medical Systems, Zipf, Austria) ultrasound system, equipped with a RIC5-9 transducer, or a Philips IU22 (Philips Healthcare Systems, Bothell, WA, USA) ultrasound system equipped with a 3D9-3v transducer, and in the other two centers the GE Voluson E8, equipped with a RIC5-9 transducer, was employed. The uterus was scanned in the sagittal plane from cornu to cornu and in the (oblique) transverse plane from the cervix to the fundus. Having established an overview of the whole uterus, the image was magnified to contain only the uterine corpus. We defined the tumor mass in the sagittal plane, and measured the endometrial thickness and the tumor/uterine AP diameter ratio at the point at which we found the deepest myometrial invasion. In cases in which we did not suspect myometrial invasion, the tumor/uterine AP diameter ratio was calculated by measuring endometrial thickness and dividing it by the uterine AP diameter. The minimal tumor-free margin was measured in the plane in which we observed the smallest distance from the tumor to the serosa. In cases in which no obvious tumor was seen, Dist-OCO was measured from the lower border of the endometrial cavity to the outer cervical os. Tumor volume (3D) and tumor volume/uterine volume (3D) ratio were calculated from saved volumes using 4D View (GE Medical Systems), Virtual Organ Computer-aided AnaLysis (VOCAL) calculation or Philips QLAB quantification software using manual trace with 30 rotational steps. We included the whole endometrium in the 3D tumor volume measurements, since it is impossible to determine normal or cancer-affected endometrium based on the ultrasound image. The tumor margin was traced in the B-plane. In all centers a dedicated pathologist, with substantial experience in gynecological oncology, assessed the pathological specimens and completed a predetermined protocol. The histopathological variables assessed included histological type, grade and pathological stage (Stage IA, myometrial invasion < 50%; Stage IB, myometrial invasion > 50%; Stage II, cervical invasion; Stage III, spread beyond the uterus/cervix within the pelvis; Stage IV, spread beyond the pelvis. Only epithelial malignant tumors were included. Endometrioid and mucinous adenocarcinoma were classified into three grades (Grade 1, well differentiated; Grade 2, moderately differentiated; Grade 3, poorly differentiated). Serous cancer, clear cell cancer and carcinosarcoma were classified as Grade 3. Receiver operating characteristics (ROC) curves with 95% CIs were calculated for each ultrasound measurement as well as for subjective impression of cervical and myometrial invasion. It is, however, important to understand that a ROC curve based on subjective assessment is constructed from a one-point measurement, which will most probably underestimate the area under the curve (AUC), and this should be taken into consideration when interpreting the results. The optimal cut-off was defined as the point most distant from the diagonal line of discrimination. To compare the objective measurements to subjective evaluation we used McNemar s test and the method of DeLong was used for comparing AUCs. All calculations were performed using SPSS software (Statistical Package for Social Sciences (SPSS), version , SPSS Inc., Chicago IL, USA) and P < 0.05 was considered statistically significant. RESULTS A total of 144 consecutive women with endometrial cancer met the inclusion criteria and were enrolled prospectively. Lund contributed 56 patients, Leuven/Genk 46 and Rome 42. The patient demographics and tumor histological characteristics are presented in Table 1. The mean body mass index was 28.7 (SD ± 6.5), 36% used low-potency estrogens and 28% had a family history of gynecological and/or breast cancer. Approximately half of the women enrolled (49%, 71/144) were Stage IA and 68% (98/144) were Grade 1 or 2 (Table 1). The diagnostic accuracy of subjective assessment and of objective ultrasound measurement techniques in predicting deep myometrial involvement are summarized in Table 2. The AUC was not significantly different Table 1 Patient demographics and tumor histological characteristics in the study group of 144 women with endometrial cancer Characteristic Mean ± SD or n (%) Age (years) 65.8 (± 10.2) Body mass index (kg/m 2 ) 28.7 (± 6.5) Postmenopausal 132 (92) Family history of breast and/or gynecological 40 (28) cancer Current high-/medium-potency* hormone use 15 (10) Current low-potency estrogen use 52 (36) Current tamoxifen use 4 (3) FIGO stage IA 71 (49) IB 34 (24) II 12 (8) IIIA IIIC2 16 (11) IVA IVB 7 (5) Grade Endometrioid adenocarcinoma, Grade 1 56 (39) Endometrioid adenocarcinoma, Grade 2 42 (29) Endometrioid adenocarcinoma, Grade 3 22 (15) Non-endometrioid (serous adenocarcinoma/ 24 (17) clear cell adenocarcinoma/ carcinosarcoma) *Oral/dermal estradiol in combination with sequential or continuous progesterones. Oral/vaginal estriol or vaginal estradiol. In four women there were no remaining tumors at hysterectomy. FIGO, International Federation of Gynecology and Obstetrics (2009 staging criteria) 10.

4 356 Mascilini et al. Table 2 Comparison of diagnostic accuracy of subjective assessment with that of objective ultrasound measurement techniques in predicting deep myometrial involvement in women with endometrial cancer Cut-off AUC (95% CI) P* Sensitivity (% (n)) P Specificity (% (n)) P Subjective assessment 0.79 ( ) 77 (46/60) 81 (68/84) Objective measurement Minimal tumor-free margin/ ( ) (49/60) (61/84) uterine AP diameter ratio Tumor/uterine AP diameter ratio ( ) (43/60) (64/84) 0.32 Minimal tumor-free margin 7.1 mm 0.78 ( ) (51/60) (42/84) < Endometrial thickness 28.5 mm 0.74 ( ) (33/60) (71/84) 0.37 Tumor/uterine volume ratio (3D) ( ) (36/60) (71/84) 0.44 Tumor volume (3D) 9.9 ml 0.72 ( ) (41/60) (56/84) All comparisons were subjective assessment: *vs area under receiver operating characteristics curve (AUC) of objective assessment using DeLong s method; vs sensitivity of objective assessment using McNemar s test; vs specificity of objective assessment using McNemar s test. Table 3 Comparison of specificity of subjective assessment with that of objective ultrasound measurement techniques in predicting deep myometrial involvement in women with endometrial cancer, with sensitivity fixed at same level as that of subjective assessment Cut-off Sensitivity (% (n)) 95% CI (%) Specificity (% (n)) 95% CI (%) P Subjective assessment 77 (46/60) (68/84) Objective measurement Minimal tumor-free margin/ (46/60) (64/84) uterine AP diameter ratio Tumor/uterine AP diameter ratio (46/60) (48/84) < Minimal tumor-free margin 5.5 mm 75*(45/60) (61/84) Endometrial thickness 18.0 mm 75*(45/60) (50/84) < Tumor/uterine volume ratio (3D) *(45/60) (41/84) < Tumor volume (3D) 7.0 ml 77 (46/60) (48/84) < *Not possible to select 77% since too many patients had same value around cut-off point. Specificity of objective ultrasound measurement vs specificity of subjective assessment, at sensitivity fixed at that of subjective evaluation (77%), using McNemar s test. Table 4 Comparison of subjective assessment and objective measurement Dist-OCO (distance from outer cervical os to lower margin of tumor) for prediction of cervical invasion in women with endometrial cancer Cut-off AUC (95% CI) P* Sensitivity (% (n)) P Specificity (% (n)) P Subjective assessment 0.74 ( ) 54 (14/26) 93 (110/118) Dist-OCO 20.5 mm 0.75 ( ) (19/26) (74/118) < All comparisons were subjective assessment: *vs area under receiver operating characteristics curve (AUC) of objective assessment using DeLong s method; vs sensitivity of objective assessment using McNemar s test; vs specificity of objective assessment using McNemar s test. between subjective evaluation and any of the objective measurement techniques, but subjective evaluation showed a significantly higher sensitivity when compared with endometrial thickness and tumor/uterine volume (3D) ratio. Moreover, the subjective evaluation showed a significantly higher specificity compared with minimal tumor-free margin/uterine AP diameter ratio, minimal tumor-free margin and tumor volume (3D). The sensitivity (72%) and specificity (76%) of tumor/uterine AP diameter ratio (at cut-off 0.53) were not significantly different (P = 0.44 and P = 0.32, respectively) from those of subjective evaluation. We also compared the specificity of the objective measurement techniques, fixing sensitivity at the same level as that of subjective assessment (77%), with that of subjective evaluation and found that only minimal tumorfree margin/uterine AP diameter ratio was not significantly different (Table 3). The best objective parameter for the prediction of cervical invasion was Dist-OCO. Other parameters related to endometrial thickness, tumor volume or minimal tumor-free margin were tested but found to be not useful. Table 4 illustrates the subjective assessment and objective measurement of Dist-OCO for the prediction of cervical invasion. Dist-OCO had a nonsignificantly higher sensitivity compared with subjective evaluation (73% vs 54%, P = 0.06), but a significantly lower specificity (63% vs 93%, P < 0.001). The AUC did not differ between these methods. DISCUSSION In this prospective multicenter study on women with endometrial cancer we found that subjective assessment of cervical and myometrial invasion is as good as or better than any objective measurement technique.

5 Subjective assessment vs objective measurement in endometrial cancer 357 The tumor/uterine AP diameter ratio and minimal tumor-free margin/uterine AP diameter ratio seem to be the best objective measurement techniques to predict deep myometrial invasion. 3D volume measurements did not improve the assessment. The best objective parameter to predict cervical invasion was Dist-OCO. Currently, there is no consensus on the use of imaging methods in the preoperative assessment of women with endometrial cancer. Several publications have shown that subjective ultrasound evaluation is a reliable method in the assessment of myometrial 4 7 and cervical 5,7,9 invasion. Nevertheless, ultrasound assessment of myometrial and cervical invasion has not gained widespread use. One reason might be the lack of standardized criteria for the assessment of myometrial and cervical invasion. Two previous studies have suggested the use of objective measurement techniques to assess myometrial invasion. Karlsson and coworkers suggested using the tumor/uterine AP diameter with a cut-off of We also found tumor/uterine AP diameter ratio to be one of the best objective measurement techniques, with a similar cut-off (0.53). Alcazar and coworkers proposed the use of a minimal margin of 9 mm measured from the 3D volume 6, choosing a cut-off point with the aim of detecting all women with deep myometrial invasion, and obtained a specificity of 60%. When we used a 9-mm cut-off for minimal tumor-free margin with our data, we obtained a sensitivity of 90%, but the specificity was only 36%. This discrepancy may have been caused by differences in our study populations, and it highlights the need for prospective validation of cut-off points and measurement techniques. It was reassuring to see that subjective evaluation performed best or as well as any objective measurement technique. Subjective evaluation takes more features into account in addition to size and proportion, including regularity of the endometrial myometrial border and vascular pattern 11,12. For example, in large exophytic tumors, the tumor proportion will be great but the endometrial myometrial border will be regular and the focal vessel pattern will indicate to the examiner that it is most probably a large polypoid lesion bulging into the cavity and stretching the surrounding myometrium, rather than deeply invading. The strength of this study is the multicenter, prospective design and the large size of our series. One limitation might be the lack of generalizability of the results as our data need to be validated in a new series of patients. Reproducibility is also a major issue when dealing with diagnostic tests. An ideal design would be to assess reproducibility for the different measurements as well as for subjective evaluation, for both ultrasound and histological assessment, between several examiners. We feel that such a comparison deserves a separate paper. However, previous studies have shown that the reproducibility of endometrial thickness measurements in the assessment of women with postmenopausal bleeding is very high 13,14, although performance is less reliable in the hands of inexperienced sonographers 14.Inasmall series, Merce and coworkers showed that endometrial tumor volume measurements, using 3D ultrasound, were highly reproducible 15. However, even though the volume calculation has been shown to be reproducible, it is difficult to avoid small inaccuracies in the C-plane if the tumor margins are traced from another plane (as shown in Figure 1b). Another limitation could be the crude analysis of deep myometrial invasion (as yes/no); it would also be interesting to determine if the pathologist and ultrasonographer actually defined the point of deepest infiltration at the same spot. Regarding the statistical analysis comparing the AUC for subjective assessment and objective measurements, one must understand the difficulties in comparing the AUC for subjective evaluation, which is based on only one point and thus has a triangular shape. This might lead to an underestimation as compared to the objective measurement techniques which consisted of several points. There are also limitations related to ultrasound as a technique. Many women with endometrial cancer are obese, which often causes the uterus to take an upright position. It is well-known that endometrial assessment is very difficult if the endometrium is at an angle of 0 relative to the probe. This can, however, sometimes be overcome by using the free hand to manipulate the uterus, by applying slight pressure on the abdomen. Still, in a minority of women the image quality is too poor for appropriate assessment. Another pitfall might be large tumors bulging down into the cervical canal, mimicking cervical invasion. These cases can often be ruled out by the use of a dynamic examination technique; by assessing the presence of a sliding movement of the tumor over the inner cervical os, it is possible to differentiate a tumor that is bulging/sliding down into the cervical canal from one that is truly invading. One can also look for vessels entering the tumor at the inner cervical os, indicating cervical stromal invasion. Assessment of cervical invasion has traditionally been done by cervical curettage. In a study comprising 1297 women undergoing preoperative cervical curettage, the accuracy for prediction of cervical invasion was only 51% 16. In a recent study comprising 338 women, comparing cervical curettage to MRI, the latter yielded a significantly higher specificity (79% vs 91%, P = 0.001) and accuracy (77% vs 84%, P = 0.005) than did cervical curettage, whereas the sensitivity was not significantly different (65% vs 59%) 17. Hysteroscopy has also been used to assess cervical invasion, one study showing that hysteroscopy might have a higher sensitivity compared with both ultrasound and MRI (93%, 53% and 67%, respectively) 18. In our study, the sensitivity of cervical curettage was 56% and the specificity 89% for the assessment of cervical invasion. Moreover, 10% of the specimens could not be assessed due to an insufficient sample. These data suggest that ultrasound is at least as good as and probably better than cervical curettage for the assessment of cervical stromal invasion. Still, one cannot

6 358 Mascilini et al. expect ultrasound to detect minimal stromal invasion that can only be seen using a microscope. In conclusion, we found that subjective assessment of myometrial and cervical invasion performs better than or as well as selected objective measurement techniques in women with endometrial cancer. Further studies are needed to investigate if subjective evaluation or objective measurement techniques have the highest reproducibility, and how well they perform in the hands of examiners from other institutions, in particular in the hands of examiners with more or less experience. The cut-offs suggested for the objective measurements in this paper must also be tested prospectively before they can be used in a clinical setting. ACKNOWLEDGMENTS The study was supported by grants from FoU Region Skane, Sweden and ALF (project ) Stockholms läns landsting, Karolinska University Hospital, Sweden. REFERENCES 1. Parkin DM, Bray FI, Devesa SS. Cancer burden in the year The global picture. EurJCancer2001; 37: S4 S Benedet JL, Bender H, Jones H, 3rd, Ngan HY, Pecorelli S. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. FIGO Committee on Gynecologic Oncology. Int J Gynaecol Obstet 2000; 70: Fischerova D. Ultrasound scanning of the pelvis and abdomen for staging of gynecological tumors: a review. Ultrasound Obstet Gynecol 2011; 38: Weber G, Merz E, Bahlmann F, Mitze M, Weikel W, Knapstein PG. Assessment of myometrial infiltration and preoperative staging by transvaginal ultrasound in patients with endometrial carcinoma. Ultrasound Obstet Gynecol 1995; 6: Akbayir O, Corbacioglu A, Numanoglu C, Guleroglu FY, Ulker V, Akyol A, Guraslan B, Odabasi E. Preoperative assessment of myometrial and cervical invasion in endometrial carcinoma by transvaginal ultrasound. Gynecol Oncol 2011; 122: Alcazar JL, Galvan R, Albela S, Martinez S, Pahisa J, Jurado M, Lopez-Garcia G. Assessing myometrial infiltration by endometrial cancer: uterine virtual navigation with threedimensional US. Radiology 2009; 250: Savelli L, Ceccarini M, Ludovisi M, Fruscella E, De Iaco PA, Salizzoni E, Mabrouk M, Manfredi R, Testa AC, Ferrandina G. Preoperative local staging of endometrial cancer: transvaginal sonography vs. magnetic resonance imaging. Ultrasound Obstet Gynecol 2008; 31: Karlsson B, Norstrom A, Granberg S, Wikland M. The use of endovaginal ultrasound to diagnose invasion of endometrial carcinoma. Ultrasound Obstet Gynecol 1992; 2: Sawicki W, Spiewankiewicz B, Stelmachow J, Cendrowski K. The value of ultrasonography in preoperative assessment of selected prognostic factors in endometrial cancer. Eur J Gynaecol Oncol 2003; 24: Pecorelli S. Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium. Int J Gynaecol Obstet 2009; 105: Leone FP, Timmerman D, Bourne T, Valentin L, Epstein E, Goldstein SR, Marret H, Parsons AK, Gull B, Istre O, Sepulveda W, Ferrazzi E, Van den Bosch T. Terms, definitions and measurements to describe the sonographic features of the endometrium and intrauterine lesions: a consensus opinion from the International Endometrial Tumor Analysis (IETA) group. Ultrasound Obstet Gynecol 2010; 35: Epstein E, Van Holsbeke C, Mascilini F, Masback A, Kannisto P, Ameye L, Fischerova D, Zannoni G, Vellone V, Timmerman D, Testa AC. Gray-scale and color Doppler ultrasound characteristics of endometrial cancer in relation to stage, grade and tumor size. Ultrasound Obstet Gynecol 2011; 38: Epstein E, Valentin L. Intraobserver and interobserver reproducibility of ultrasound measurements of endometrial thickness in postmenopausal women. Ultrasound Obstet Gynecol 2002; 20: Karlsson B, Granberg S, Ridell B, Wikland M. Endometrial thickness as measured by transvaginal sonography: interobserver variation. Ultrasound Obstet Gynecol 1994; 4: Merce LT, Alcazar JL, Engels V, Troyano J, Bau S, Bajo JM. Endometrial volume and vascularity measurements by transvaginal three-dimensional ultrasonography and power Doppler angiography in stimulated and tumoral endometria: intraobserver reproducibility. Gynecol Oncol 2006; 100: Leminen A, Forss M, Lehtovirta P. Endometrial adenocarcinoma with clinical evidence of cervical involvement: accuracy of diagnostic procedures, clinical course, and prognostic factors. Acta Obstet Gynecol Scand 1995; 74: Haldorsen IS, Berg A, Werner HM, Magnussen IJ, Helland H, Salvesen OO, Trovik J, Salvesen HB. Magnetic resonance imaging performs better than endocervical curettage for preoperative prediction of cervical stromal invasion in endometrial carcinomas. Gynecol Oncol 2012; 126: Cicinelli E, Marinaccio M, Barba B, Tinelli R, Colafiglio G, Pedote P, Rossi C, Pinto V. Reliability of diagnostic fluid hysteroscopy in the assessment of cervical invasion by endometrial carcinoma: a comparative study with transvaginal sonography and MRI. Gynecol Oncol 2008; 111: