- Email: [email protected]
Diagnostic utility of three-dimensional power Doppler ultrasound for postmenopausal bleeding
Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
Contents lists available at ScienceDirect
Taiwanese Journal of Obstetrics & Gynecology journal homepage: www.tjog-online.com
Original Article
Diagnostic utility of three-dimensional power Doppler ultrasound for postmenopausal bleeding Ari Kim a, Ji Young Lee b, Sungwook Chun c, Heung Yeol Kim d, * a
Department of Obstetrics and Gynecology, Institute of Wonkwang Medical Science, College of Medicine, Wonkwang University, Iksan, Republic of Korea Department of Obstetrics and Gynecology, Konkuk University School of Medicine, Seoul, Republic of Korea c Department of Obstetrics and Gynecology, College of Medicine, Inje University, Busan, Republic of Korea d Department of Obstetrics and Gynecology, College of Medicine, Kosin University, Busan, Republic of Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 18 October 2013
Objectives: We evaluated the role of three-dimensional power Doppler ultrasound (3D PD-US) to detect endometrial lesions in women with postmenopausal endometrial bleeding. Materials and methods: In this prospective observational study, from January 2009 to November 2012, we recruited 225 postmenopausal women with postmenopausal uterine bleeding who met the study criteria. Women who had hematologic disease, chronic medical diseases, or nonuterine pelvic diseases were excluded. Prior to endometrial biopsy, the patients underwent a baseline transvaginal ultrasound screening. The vascular indices and endometrial volumes were calculated with 3D PD-US and compared with the endometrial histopathology. Results: Among the endometrial histopathologic findings of 174 women, atrophic endometrium was the most common finding (30.5%). Endometrial malignancy was confirmed in 28 cases (16.1%), and endometrial hyperplasia was diagnosed in 17 cases (9.8%). The prevalence of endometrial cancer was high in patients who had endometrial thickness >9.5 mm (p < 0.001) and volume greater than 4.05 mL (p < 0.001). For the endometrial carcinoma only, the cutoff values of vascular index, flow index, and vascular flow index for predicting malignancy were 13.070, 12.610, and 3.764, respectively. For endometrial hyperplasia, endometrial thickness and vascular flow index were significant findings. Conclusion: Endometrial vasculature and volume can be obtained using 3D PD-US. The diagnostic usefulness of 3D PD-US for endometrial diseases is promising in women with postmenopausal endometrial bleeding. Copyright © 2015, Taiwan Association of Obstetrics & Gynecology. Published by Elsevier Taiwan LLC. All rights reserved.
Keywords: endometrial biopsy postmenopausal bleeding power Doppler three-dimensional ultrasound
Introduction Postmenopausal bleeding is a common symptom in general gynecological practice. The incidence of vaginal bleeding in postmenopausal women is approximately 10% immediately after menopause and 5% of all cases of menopause [1,2]. Various benign genital causes of postmenopausal vaginal bleeding include atrophic vaginitis, endometrial and cervical polyps, endometrial hyperplasia, pyometra, and submucosal fibroids. However, 10% of all women presenting with postmenopausal bleeding may have endometrial
* Corresponding author. Department of Obstetrics and Gynecology, College of Medicine, Kosin University, 31 Amnam-dong, Seo-gu, Busan 607-702, Republic of Korea. E-mail address: [email protected] (H.Y. Kim).
malignancy. Clinical investigations for those patients are mainly directed to exclude malignant and premalignant lesions [3,4]. In the United States, endometrial cancer is the most common type of gynecological cancer, and it is ranked fourth among all types of cancer and seventh among all causes of death by cancer [5]. Cancers in the female reproductive organs account for 15.2% of all types of cancer; however, endometrial cancer comprises only 1.9% of all types of cancer, but the incidence is increasing with the increased average lifespan and popularity of hormone replacement therapy [6]. Endometrial cancer occurs in both the pre- and postmenopausal periods, peaking when patients are in their 50s, and postmenopausal uterine bleeding is the most common symptom of endometrial cancer [3]. To our knowledge, endometrial sampling and histopathologic review can provide a tentative diagnosis [7]. Transvaginal ultrasound as a noninvasive scan is the most commonly used first-line
http://dx.doi.org/10.1016/j.tjog.2013.10.043 1028-4559/Copyright © 2015, Taiwan Association of Obstetrics & Gynecology. Published by Elsevier Taiwan LLC. All rights reserved.
222
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
investigation for women with postmenopausal endometrial bleeding before endometrial sampling. Usually, a thick endometrium is indicative of further invasive evaluations such as endometrial sampling and/or hysteroscopy [8,9]. However, conventional two-dimensional (2D) ultrasound cannot assess all possible endometrial pathologies. The recent developments with ultrasound equipment enable new imaging techniques for volume scanning. Unlike 2D ultrasound, three-dimensional (3D) ultrasound visualizes the whole endometrium on a coronal plane and can integrate Doppler imaging to display the vascularity in the interested areas. In this study, we applied 3D power Doppler ultrasound (3D PDUS) in women with postmenopausal endometrial bleeding and calculated various ultrasonographic vascular markers. Thus, we aimed to determine the useful markers to predict endometrial disease and evaluate the usefulness of 3D PD-US in postmenopausal bleeding.
computer equipped with Virtual Organ Computer-Aided Analysis (VOCAL) software, which can calculate the targeted volume and vasculatures. The results of the adjunctive 3D PD-US assessment did not affect subsequent clinical management procedures because the clinicians did not know the 3D PD-US results. The manual mode of the VOCAL contour editor was used to cover the 3D volume of the endometrium with 15 rotation steps. The A-plane (sagittal view of the uterus) was rotated, and the myometrialeendometrial junction from the fundus to the internal cervical os was outlined in each rotation image (12 slices). Then, the histogram demonstrated only the endometrium and showed the endometrial volume with vascular indices (Figure 1). The three automatically calculated vascular indices of the endometrium included the vascularization index (VI), flow index (FI), and vascularization flow index (VFI). Endometrial sampling
Materials and methods Patients We recruited all 225 women who visited the obstetrics and gynecology departments of three university hospitals from January 2009 to November 2012 with a chief complaint of abnormal postmenopausal endometrial bleeding. The study was conducted prospectively in women who met the criteria. Menopause was defined as amenorrhea for at least 12 months. Patients with bleeding that originated from the cervix, vagina, or vulva were not considered for the observation. Among them, 174 patients were evaluated for prospective correct diagnosis for bleeding after excluding patients who had systemic and hematologic disorders, previous endometrial diagnosis, uterine bleeding due to disorders in the pelvis other than in the uterus, or postmenopausal hormone therapy. All women underwent physical examinations such as weight and body mass index (BMI) evaluation, history, gynecologic evaluation, and basic laboratory tests. Ethical approval for further evaluation and use of data was granted by the Institutional Review Board of Kosin Medical Center.
Within 1 week after ultrasound examination, endometrial biopsy was performed by cervical dilatation and curettage in participating women after signed informed consent was obtained. In the lithotomy position, the patient was prepared with povidoneeiodine topical antiseptic solution and draped. After sedation, the cervix was dilated with a Hegar dilater, and the endometrium was curetted. The endometrial specimens were reviewed by pathologists for final diagnosis. Statistical analysis All results are presented as mean and standard deviation values according to the distribution of data or number with %. The KomologoroveSmirnov test was used to evaluate the normal distribution of the continuous data. Comparisons between two groups were carried out with the Student t test. The receiver operating characteristic curve was applied to calculate the predictive value of the endometrial parameters for endometrial cancer or endometrial hyperplasia. SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) was used for statistical calculations, and p < 0.05 was considered statistically significant.
Ultrasound investigation Results The patients underwent 2D transvaginal ultrasound scanning as the initial investigation to evaluate the endometrium. Routine ultrasound was used for the diagnosis of the anatomical cause of endometrial bleeding, such as uterine fibroid. The endometrial thickness was measured at its thickest point in an anteroposterior dimension from one basal layer to other in a midsagittal plane. Then, adjunctive 3D PD-US (Voluson E8; GE Healthcare, Zipf, Austria) was carried out using uniform ultrasound modes, and the obtained volume data were transferred to one investigator (A. Kim) for volume and Doppler analysis. The settings were as follows: frequency, 5 MHz; power Doppler gain, e5; dynamic range, 20e40 dB; edge, 1; persistence, 2; color map, 5; motion filter, 1; and pulse repetition frequency, 0.8 kHz. When a midsagittal view of the uterus was obtained, the power Doppler mode was turned on. The area of interest was the endometrium. The 3D mode was then activated, and the area of interest was adjusted. Three-dimensional volumetric data were obtained using automatic sweep, with the angle being set to 120 to ensure that a complete endometrial volume was included. The patients were asked to hold their breath during volume acquisition. The multiplanar display was used to ensure that the area of interest had been captured in its entirety. After 3D volume storage, direct ultrasound examination was performed to reduce examination time and patient discomfort. All 3D volume data that were transferred were analyzed with a desktop
Among 225 patients who had undergone 2D and 3D PD-US, 174 completed the endometrial biopsy for diagnostic confirmation (Table 1). In total, 146 patients (83.9%) showed a benign endometrium, which could be of the following types: (1) proliferative endometrium, (2) secretory endometrium, (3) glandular stromal dissociation, (4) endometritis, (5) endometrial atrophy, (6) endometrial polyp, and (7) endometrial hyperplasia. The most common diagnosis among the benign diseases was endometrial atrophy (53 cases, 30.5%). However, endometrial malignancy was confirmed in 28 cases (16.1%). Among these, 25 cases were diagnosed with adenocarcinoma, while two cases were of squamous cell carcinoma and one was of serous carcinoma. Table 1 shows the comparison of clinical characteristics including ultrasonographic data. In the malignant group, the mean age was higher than that of the benign group (59.27 years and 61.04 years, respectively; p ¼ 0.023). Thus, there was a significant difference in the interval since menopause in the two groups (mean, 9.24 years vs. 11.26 years; p ¼ 0.003). Patients with a malignant endometrium tended to have a higher BMI (mean, 27.32 kg/ m2 vs. 29.28 kg/m2; p ¼ 0.16). In the endometrial evaluation with 2D ultrasound, the valuable measurement was endometrial thickness. The endometrial thickness was thicker in malignant cases than in benign cases (mean,
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
223
Figure 1. Histogram of endometrium and endometrial volume with vascular indices using power Doppler and VOCAL software.
8.70 mm and 14.29 mm, respectively; p < 0.001). Additive technical assessment with 3D PD-US can offer more variable parameters, such as endometrial volume, VI, FI, and VFI. The endometrial volume was larger (mean, 2.90 mL vs. 5.56 mL; p < 0.001) and all vascular indices were higher in patients with a malignant endometrium (Table 1). We analyzed the receiver operating characteristic curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for malignancy (Fig. 2). The area under the curve was statistically significant in all parameters, including endometrial
Table 1 Comparison of clinical characteristics between benign and malignant groups. Characteristics
Benign group (n ¼ 146)
Malignant group (n ¼ 28)
Age (y) Body mass index (kg/m2) Years since menopause Endometrial thickness (mm) Endometrial volume (mL) Endometrial vascularization index (%) Endometrial flow index (0e100) Endometrial vascularization flow index (0e100) Histological diagnosis Proliferative Em Secretory Em Glandular stromal dissociation Endometritis Atrophy Polyp Hyperplasia Endometrial cancer
59.27 ± 5.25 27.32 ± 6.31 9.24 ± 2.14 8.70 ± 3.91 2.90 ± 1.26 10.56 ± 6.59
61.04 29.28 11.26 14.29 5.56 17.80
7.28 8.16 4.52 3.60 1.48 3.47
0.023 0.016 0.003 <0.001 <0.001 <0.001
13.96 ± 7.24 2.77 ± 2.23
17.45 ± 6.04 9.17 ± 4.11
0.018 <0.001
29 11 10 11 53 15 17 28
± ± ± ± ± ±
p
(16.67%) (6.32%) (5.75%) (6.32%) (30.46%) (8.62%) (9.78%) (16.09%)
Data are presented as mean ± standard deviation or number (%). Comparisons between two groups were carried out using the Student t test. EM ¼ endometrium.
thickness and volume, for the prediction of endometrial cancer. Table 2 shows that endometrial thickness, volume, VI, FI, and VFI have predictive power for malignancy before the performance of invasive endometrial biopsy. The best predictive cutoff value of endometrial thickness for endometrial cancer was 9.5 mm. An endometrial volume of >4.05 mL was also predictive of endometrial malignancy. The vascular parameters were relative values, but, in this mode setting, the cutoff values were 13.07 for VI, 12.61 for FI, and 3.765 for VFI. We also analyzed the predictive property or ultrasonographic characteristics for endometrial hyperplasia (Figure 3 and Table 3). The area under the curve for endometrial volume was not significant but reflected endometrial hyperplasia. Endometrial VI and VFI were also not significant; there was no significant difference between other benign and hyperplastic endometria. However, endometrial thickness and VFI showed a significant difference between groups. The best predictive values were 7.5 mm for endometrial volume and 2.275 for VFI. Discussion Transvaginal ultrasound is currently used as the first-step technique in women with postmenopausal vaginal bleeding. It is well known that ultrasound may reliably rule out endometrial cancer when endometrial thickness is <5 mm [10]. Furthermore, application of ultrasound to women with postmenopausal bleeding is not invasive and is cost effective [11,12]. However, a thickened endometrium alone is not indicative of endometrial pathology such as malignancy. Other useful additive parameters or sonographic findings should be used to predict endometrial malignancy. Power Doppler angiography and 3D ultrasound have recently been introduced in the gynecologic field. These advanced techniques overcome the limitations of color Doppler ultrasound and conventional 2D ultrasound [13]. Three-dimensional ultrasound can render the whole endometrium even in the coronal plane, and
224
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
Figure 2. Analysis of the ROC curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for endometrial malignancy. FI ¼ flow index; ROC ¼ receiver operating characteristic; VFI ¼ vascular flow index; VI ¼ vascular index.
Figure 3. Analysis of the ROC curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for endometrial hyperplasia. FI ¼ flow index; ROC ¼ receiver operating characteristic; VFI ¼ vascular flow index; VI ¼ vascular index.
the targeted volume is calculated with VOCAL software or another built-in program. It has been accepted that the volume calculated using VOCAL software may represent the true tissue volume [14]. The combination of 3D ultrasound and sensitive power Doppler angiography allow for improved visualization of vessels, even for arterioles or veins with slow velocity. The calculated vascular parameters, even if it they are relative, can provide comparable indices. VI measures the number of color voxels in the volume, which represents the vessels in the tissue and is expressed as a percentage. FI is the mean color value in the color voxels, which indicates the average intensity of blood flow and is expressed as a whole number from 0 to 100. VFI is the mean value of all the color voxels in the volume, which represents both vascularization and blood flow, and is also expressed as a whole number from 0 to 100 [15]. These values are different from color Doppler vascular parameters, such as pulsatility index, resistance index, systolic/diastolic ratio, or velocity. Vascular parameters for a specific diagnostic approach can be calculated for various organs. In the present study, we evaluated the role of 3D PD-US in postmenopausal endometrial bleeding in distinguishing endometrial cancer from benign diseases. Postmenopausal endometrial bleeding is the first and most important symptom of endometrial malignancy. We also found that the studied variables can predict endometrial hyperplasia as compared with other benign diseases.
We included all presenting women with postmenopausal endometrial bleeding and with tentative diagnosis from endometrial curettage and biopsy. The women with a thin endometrium (<5 mm) were not excluded, even though a thin endometrium is not strongly associated with endometrial cancer. We wanted to analyze the natural prevalence of endometrial cancer in postmenopausal bleeding and to reduce the selection bias. Our prevalence of endometrial cancer was 16.1%, which was higher than that of the reported results and therefore does not represent the general population [3,4]. In addition to follow-up loss, exclusion of postmenopausal hormone therapy can explain the higher prevalence. Our results show that endometrial thickness volume may play an important role in discriminating endometrial cancer from benign diseases, which confirm the findings of previous published data regarding the correlation between endometrial cancer and endometrial thickness or volume [11,13,16e18]. The positive cutoff value was offered for the prediction of malignancy with an endometrial thickness of 9.5 mm, not as the passive endometrial thickness of 5 mm. This means that endometrial thickness of >5 mm requires further evaluation with methods such as endometrial biopsy, catheter sampling, whole curettage, or
Table 2 Receiver operating characteristic analysis of endometrial three-dimensional ultrasound and power Doppler angiography parameters for predicting malignancy. Ultrasonographic data
Endometrial thickness (mm) Endometrial volume (mL) Vascularization index (%) Flow index (0e100) Vascularization flow index (0e100) CI ¼ confidence interval.
Table 3 Receiver operating characteristic analysis of endometrial three-dimensional ultrasound and power Doppler angiography parameters for predicting endometrial hyperplasia. Ultrasonographic data
Benign group (n ¼ 146) versus malignant group (n ¼ 28) Area under the curve
95% CI
p
Cutoff value
0.843 0.920 0.827 0.653 0.933
0.770e0.916 0.873e0.966 0.765e0.889 0.550e0.756 0.894e0.972
<0.001 <0.001 <0.001 0.010 <0.001
9.50 4.050 13.070 12.610 3.765
Endometrial thickness (mm) Endometrial volume (mL) Vascularization index (%) Flow index (0e100) Vascularization flow index (0e100) CI ¼ confidence interval.
Other benign group (n ¼ 129) versus endometrial hyperplasia group (n ¼ 17) Area under the curve
95% CI
p
Cutoff value
0.724
0.614e0.834
0.003
0.609 0.636 0.504 0.852
0.480e0.738 0.544e0.728 0.397e0.612 0.753e0.951
0.144 0.069 0.954 <0.001
7.50
2.275
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
hysteroscopic biopsy. However, an endometrium thicker than 9.5 mm is worthy of attention and is evaluated with malignancy in mind. With regard to endometrial volume, a larger volume must be associated with a thick endometrium and, of course, can be related to endometrial cancer. We assessed the cutoff values of endometrial volume for the prediction of cancer as 4.05 mL, which is not small compared with that of the reproductive endometrium [19]. The inter- and intraobserver variations were small in the volume calculation in the 3D VOCAL program, and thus, increased endometrial volume seems to be indicative of endometrial cancer [20]. Endometrial vascular assessment for predicting malignancy has been studied in several reports [13,21e24], all of which concluded that 3D PD-US analysis of vascularization in endometrial cancer correlates with the diagnosis in women with or without endometrial bleeding. In some studies, the subendometrial vasculature calculated by 3D PD-US can also be valuable for the prediction of myometrial invasion of endometrial carcinoma [22]. In the present study, previous published data were confirmed with regard to VI, FI, and VFI. Endometrial cancer tends to have higher vasculature in the endometrium. This study differed from previous studies due to the large number of participants with bleeding, more cases with a diagnosis of cancer, and smaller prospective collection bias of this study. Thus, we can conclude that endometrial cancer increases the vasculature of the endometrium, which can be applied for the prediction of malignancy. The cutoff value in the study cannot be applied to all patients with postmenopausal endometrial bleeding because the parameters vary according to the setting mode and ultrasound model. The group with endometrial cancer showed a higher age distribution, interval from menopause, and BMI, a well-known risk factor for endometrial malignancies [16]. However, we focused on the endometrial characteristics for the prediction of malignancy, so these clinical variables were not considered for correlation with endometrial ultrasonographic characteristics. We focused on the relationship of endometrial characteristics evaluated by ultrasound and endometrial hyperplasia, which has been studied in a previous report [21]. The measurements for endometrial thickness, volume, VI, and FI were significantly higher in both the hyperplasia and the carcinoma groups [21]. Discrimination of hyperplasia was the only meaningful analysis of the study, and the significant variables were endometrial volume and VFI. This finding can be explained by the fact that postmenopausal women have the limitation of volume increase for a decreased uterus and thus a limitation in measuring vascularity. However, typical endometrial thickening >7.5 mm is meaningful for the prediction of endometrial hyperplasia. The cutoff of VFI for endometrial hyperplasia is smaller than that of endometrial cancer, but greater endometrial vasculature can be associated with endometrial hyperplasia. We prospectively analyzed, beyond the previous studies, with a larger group of women with postmenopausal endometrial bleeding. The various analyses were not permitted for the limitation of the recruitment of patients, and a randomized controlled trial is needed to clarify the results of this study. Nevertheless, the present study demonstrated the predictive values for endometrial cancer and hyperplasia. Based on the findings and assessment of the present study, although endometrial thickness appears to be the simplest with 2D ultrasound, endometrial thickness and VFI were more accurate in diagnosing endometrial hyperplasia according to the receiver operating characteristic curves. Furthermore, all parameters of the endometrium, such as thickness, volume, VI, FI, and VFI, have higher scores for endometrial malignancy in women with postmenopausal endometrial bleeding. Thus, physicians must focus on
225
endometrial thickness >9.5 mm for malignancy and >7.5 mm for hyperplasia. Conflicts of interest The authors have no conflicts of interest relevant to this article. Acknowledgments This study was supported by a grant from Kosin University College of Medicine (2012). References [1] Astrup K, Olivarius Nde F. Frequency of spontaneously occurring postmenopausal bleeding in the general population. Acta Obstet Gynecol Scand 2004;83:203e7. [2] Guruwadayarhalli B, Jones SE, Srinivasan V. Hysteroscopy in the diagnosis of postmenopausal bleeding. Menopause Int 2007;13:132e4. [3] Rose PG. Endometrial carcinoma. NEJM 1996;335:640e9. [4] Raouf SA, Gupta P, Papaioannou S, Pradhan P. Endometrial thickness for invasive investigations in women with postmenopausal bleeding. Climacteric 2011;14:117e20. [5] Partridge EE, Shingleton HM, Menck HR. The National Cancer Data Base report on endometrial cancer. J Surg Oncol 1996;61:111e23. [6] Kong TW, Lee KM, Cheong JY, Kim WY, Chang S, Yoo S, et al. Comparison of laparoscopic versus conventional open surgical staging procedure for endometrial cancer. J Gynecol Oncol 2010;21:106e11. [7] Burbos N, Musonda P, Giarenis I, Shiner AM, Giamougiannis P, Morris EP, et al. Predicting the risk of endometrial cancer in postmenopausal women presenting with vaginal bleeding: the Norwich DEFAB risk assessment tool. Br J Cancer 2010;102:1201e6. [8] Dreisler E, Sorensen SS, Ibsen PH, Lose G. Value of endometrial thickness measurement for diagnosing focal intrauterine pathology in women without abnormal uterine bleeding. Ultrasound Obstet Gynecol 2009;33:344e8. [9] Goldstein RB, Bree RL, Benson CB, Benacerraf BR, Bloss JD, Carlos R, et al. Evaluation of the woman with postmenopausal bleeding: Society of Radiologists in Ultrasound-Sponsored Consensus Conference statement. J Ultrasound Med 2001;20:1025e36. [10] Gupta JK, Chien PF, Voit D, Clark TJ, Khan KS. Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis. Acta Obstet Gynecol Scand 2002;81: 799e816. [11] Clark TJ, Barton PM, Coomarasamy A, Gupta JK, Khan KS. Investigating postmenopausal bleeding for endometrial cancer: cost-effectiveness of initial diagnostic strategies. BJOG 2006;113:502e10. €lmann HA, Heintz AP. Cost-effectiveness of the use [12] Dijkhuizen FP, Mol BW, Bro of transvaginal sonography in the evaluation of postmenopausal bleeding. Maturitas 2003;45:275e82. [13] Alcazar JL, Galvan R. Three-dimensional power Doppler ultrasound scanning for the prediction of endometrial cancer in women with postmenopausal bleeding and thickened endometrium. Am J Obstet Gynecol 2009;200: 44.e1e6. [14] Martins WP, Ferriani RA, Barra DA, Dos Reis RM, Bortolieiro MA, Nastri CO, et al. Reliability and validity of tissue volume measurement by threedimensional ultrasound: an experimental model. Ultrasound Obstet Gynecol 2007;29:210e4. [15] Kim A, Han JE, Yoon TK, Lyu SW, Seok HH, Won HJ. Relationship between endometrial and subendometrial blood flow measured by three-dimensional power Doppler ultrasound and pregnancy after intrauterine insemination. Fertil Steril 2010;94:747e52. [16] Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I. Endometrial cancer. Lancet 2005;366:491e505. [17] Burbos N, Musonda P, Crocker SG, Morris EP, Duncan TJ, Nieto J. Outcome of investigations for postmenopausal vaginal bleeding in women under the age of 50 years. Gynecol Oncol 2012;125:120e3. [18] Karlsson B, Granberg S, Wikland M, Ylstalo P, Torvid K, Marsal K, et al. Transvaginal ultrasonography of the endometrium in women with postmenopausal bleedingda Nordic multicenter study. Am J Obstet Gynecol 1995;172:1488e94. [19] Zollner U, Specketer M, Dietl J, Zollner K. 3D-Endometrial volume and outcome of cryopreserved embryo replacement cycles. Arch Gynecol Obstet 2012;286:517e23. [20] Martins WP, Raine Fenning NJ, Leite SP, Ferriani RA, Nastri CO. A standardized measurement technique may improve the reliability of measurements of endometrial thickness and volume. Ultrasound Obstet Gynecol 2011;38: 107e15. [21] Odeh M, Vainerovsky I, Grinin V, Kais M, Ophir E, Bornstein J. Three-dimensional endometrial volume and 3-dimensional power Doppler analysis in
226
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
predicting endometrial carcinoma and hyperplasia. Gynecol Oncol 2007;106: 348e53. ~ G, Alca L, Jurado M, Mínguez JA, Lo pez-GarcAa zar JL. Three[22] Galv an R, Merce dimensional power Doppler angiography in endometrial cancer: correlation with tumor characteristics. Ultrasound Obstet Gynecol 2010;35:723e9. [23] Lieng M, Qvigstad E, Dahl GF, Istre O. Flow differences between endometrial polyps and cancer: a prospective study using intravenous contrast-enhanced
transvaginal color flow Doppler and three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol 2008;32:935e40. €enp€ € JU. Preoperative assessment [24] Saarelainen SK, Vuento MH, Kirkinen P, Ma aa of endometrial carcinoma by three-dimensional power Doppler angiography. Ultrasound Obstet Gynecol 2012;39:466e72.
Contents lists available at ScienceDirect
Taiwanese Journal of Obstetrics & Gynecology journal homepage: www.tjog-online.com
Original Article
Diagnostic utility of three-dimensional power Doppler ultrasound for postmenopausal bleeding Ari Kim a, Ji Young Lee b, Sungwook Chun c, Heung Yeol Kim d, * a
Department of Obstetrics and Gynecology, Institute of Wonkwang Medical Science, College of Medicine, Wonkwang University, Iksan, Republic of Korea Department of Obstetrics and Gynecology, Konkuk University School of Medicine, Seoul, Republic of Korea c Department of Obstetrics and Gynecology, College of Medicine, Inje University, Busan, Republic of Korea d Department of Obstetrics and Gynecology, College of Medicine, Kosin University, Busan, Republic of Korea b
a r t i c l e i n f o
a b s t r a c t
Article history: Accepted 18 October 2013
Objectives: We evaluated the role of three-dimensional power Doppler ultrasound (3D PD-US) to detect endometrial lesions in women with postmenopausal endometrial bleeding. Materials and methods: In this prospective observational study, from January 2009 to November 2012, we recruited 225 postmenopausal women with postmenopausal uterine bleeding who met the study criteria. Women who had hematologic disease, chronic medical diseases, or nonuterine pelvic diseases were excluded. Prior to endometrial biopsy, the patients underwent a baseline transvaginal ultrasound screening. The vascular indices and endometrial volumes were calculated with 3D PD-US and compared with the endometrial histopathology. Results: Among the endometrial histopathologic findings of 174 women, atrophic endometrium was the most common finding (30.5%). Endometrial malignancy was confirmed in 28 cases (16.1%), and endometrial hyperplasia was diagnosed in 17 cases (9.8%). The prevalence of endometrial cancer was high in patients who had endometrial thickness >9.5 mm (p < 0.001) and volume greater than 4.05 mL (p < 0.001). For the endometrial carcinoma only, the cutoff values of vascular index, flow index, and vascular flow index for predicting malignancy were 13.070, 12.610, and 3.764, respectively. For endometrial hyperplasia, endometrial thickness and vascular flow index were significant findings. Conclusion: Endometrial vasculature and volume can be obtained using 3D PD-US. The diagnostic usefulness of 3D PD-US for endometrial diseases is promising in women with postmenopausal endometrial bleeding. Copyright © 2015, Taiwan Association of Obstetrics & Gynecology. Published by Elsevier Taiwan LLC. All rights reserved.
Keywords: endometrial biopsy postmenopausal bleeding power Doppler three-dimensional ultrasound
Introduction Postmenopausal bleeding is a common symptom in general gynecological practice. The incidence of vaginal bleeding in postmenopausal women is approximately 10% immediately after menopause and 5% of all cases of menopause [1,2]. Various benign genital causes of postmenopausal vaginal bleeding include atrophic vaginitis, endometrial and cervical polyps, endometrial hyperplasia, pyometra, and submucosal fibroids. However, 10% of all women presenting with postmenopausal bleeding may have endometrial
* Corresponding author. Department of Obstetrics and Gynecology, College of Medicine, Kosin University, 31 Amnam-dong, Seo-gu, Busan 607-702, Republic of Korea. E-mail address: [email protected] (H.Y. Kim).
malignancy. Clinical investigations for those patients are mainly directed to exclude malignant and premalignant lesions [3,4]. In the United States, endometrial cancer is the most common type of gynecological cancer, and it is ranked fourth among all types of cancer and seventh among all causes of death by cancer [5]. Cancers in the female reproductive organs account for 15.2% of all types of cancer; however, endometrial cancer comprises only 1.9% of all types of cancer, but the incidence is increasing with the increased average lifespan and popularity of hormone replacement therapy [6]. Endometrial cancer occurs in both the pre- and postmenopausal periods, peaking when patients are in their 50s, and postmenopausal uterine bleeding is the most common symptom of endometrial cancer [3]. To our knowledge, endometrial sampling and histopathologic review can provide a tentative diagnosis [7]. Transvaginal ultrasound as a noninvasive scan is the most commonly used first-line
http://dx.doi.org/10.1016/j.tjog.2013.10.043 1028-4559/Copyright © 2015, Taiwan Association of Obstetrics & Gynecology. Published by Elsevier Taiwan LLC. All rights reserved.
222
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
investigation for women with postmenopausal endometrial bleeding before endometrial sampling. Usually, a thick endometrium is indicative of further invasive evaluations such as endometrial sampling and/or hysteroscopy [8,9]. However, conventional two-dimensional (2D) ultrasound cannot assess all possible endometrial pathologies. The recent developments with ultrasound equipment enable new imaging techniques for volume scanning. Unlike 2D ultrasound, three-dimensional (3D) ultrasound visualizes the whole endometrium on a coronal plane and can integrate Doppler imaging to display the vascularity in the interested areas. In this study, we applied 3D power Doppler ultrasound (3D PDUS) in women with postmenopausal endometrial bleeding and calculated various ultrasonographic vascular markers. Thus, we aimed to determine the useful markers to predict endometrial disease and evaluate the usefulness of 3D PD-US in postmenopausal bleeding.
computer equipped with Virtual Organ Computer-Aided Analysis (VOCAL) software, which can calculate the targeted volume and vasculatures. The results of the adjunctive 3D PD-US assessment did not affect subsequent clinical management procedures because the clinicians did not know the 3D PD-US results. The manual mode of the VOCAL contour editor was used to cover the 3D volume of the endometrium with 15 rotation steps. The A-plane (sagittal view of the uterus) was rotated, and the myometrialeendometrial junction from the fundus to the internal cervical os was outlined in each rotation image (12 slices). Then, the histogram demonstrated only the endometrium and showed the endometrial volume with vascular indices (Figure 1). The three automatically calculated vascular indices of the endometrium included the vascularization index (VI), flow index (FI), and vascularization flow index (VFI). Endometrial sampling
Materials and methods Patients We recruited all 225 women who visited the obstetrics and gynecology departments of three university hospitals from January 2009 to November 2012 with a chief complaint of abnormal postmenopausal endometrial bleeding. The study was conducted prospectively in women who met the criteria. Menopause was defined as amenorrhea for at least 12 months. Patients with bleeding that originated from the cervix, vagina, or vulva were not considered for the observation. Among them, 174 patients were evaluated for prospective correct diagnosis for bleeding after excluding patients who had systemic and hematologic disorders, previous endometrial diagnosis, uterine bleeding due to disorders in the pelvis other than in the uterus, or postmenopausal hormone therapy. All women underwent physical examinations such as weight and body mass index (BMI) evaluation, history, gynecologic evaluation, and basic laboratory tests. Ethical approval for further evaluation and use of data was granted by the Institutional Review Board of Kosin Medical Center.
Within 1 week after ultrasound examination, endometrial biopsy was performed by cervical dilatation and curettage in participating women after signed informed consent was obtained. In the lithotomy position, the patient was prepared with povidoneeiodine topical antiseptic solution and draped. After sedation, the cervix was dilated with a Hegar dilater, and the endometrium was curetted. The endometrial specimens were reviewed by pathologists for final diagnosis. Statistical analysis All results are presented as mean and standard deviation values according to the distribution of data or number with %. The KomologoroveSmirnov test was used to evaluate the normal distribution of the continuous data. Comparisons between two groups were carried out with the Student t test. The receiver operating characteristic curve was applied to calculate the predictive value of the endometrial parameters for endometrial cancer or endometrial hyperplasia. SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) was used for statistical calculations, and p < 0.05 was considered statistically significant.
Ultrasound investigation Results The patients underwent 2D transvaginal ultrasound scanning as the initial investigation to evaluate the endometrium. Routine ultrasound was used for the diagnosis of the anatomical cause of endometrial bleeding, such as uterine fibroid. The endometrial thickness was measured at its thickest point in an anteroposterior dimension from one basal layer to other in a midsagittal plane. Then, adjunctive 3D PD-US (Voluson E8; GE Healthcare, Zipf, Austria) was carried out using uniform ultrasound modes, and the obtained volume data were transferred to one investigator (A. Kim) for volume and Doppler analysis. The settings were as follows: frequency, 5 MHz; power Doppler gain, e5; dynamic range, 20e40 dB; edge, 1; persistence, 2; color map, 5; motion filter, 1; and pulse repetition frequency, 0.8 kHz. When a midsagittal view of the uterus was obtained, the power Doppler mode was turned on. The area of interest was the endometrium. The 3D mode was then activated, and the area of interest was adjusted. Three-dimensional volumetric data were obtained using automatic sweep, with the angle being set to 120 to ensure that a complete endometrial volume was included. The patients were asked to hold their breath during volume acquisition. The multiplanar display was used to ensure that the area of interest had been captured in its entirety. After 3D volume storage, direct ultrasound examination was performed to reduce examination time and patient discomfort. All 3D volume data that were transferred were analyzed with a desktop
Among 225 patients who had undergone 2D and 3D PD-US, 174 completed the endometrial biopsy for diagnostic confirmation (Table 1). In total, 146 patients (83.9%) showed a benign endometrium, which could be of the following types: (1) proliferative endometrium, (2) secretory endometrium, (3) glandular stromal dissociation, (4) endometritis, (5) endometrial atrophy, (6) endometrial polyp, and (7) endometrial hyperplasia. The most common diagnosis among the benign diseases was endometrial atrophy (53 cases, 30.5%). However, endometrial malignancy was confirmed in 28 cases (16.1%). Among these, 25 cases were diagnosed with adenocarcinoma, while two cases were of squamous cell carcinoma and one was of serous carcinoma. Table 1 shows the comparison of clinical characteristics including ultrasonographic data. In the malignant group, the mean age was higher than that of the benign group (59.27 years and 61.04 years, respectively; p ¼ 0.023). Thus, there was a significant difference in the interval since menopause in the two groups (mean, 9.24 years vs. 11.26 years; p ¼ 0.003). Patients with a malignant endometrium tended to have a higher BMI (mean, 27.32 kg/ m2 vs. 29.28 kg/m2; p ¼ 0.16). In the endometrial evaluation with 2D ultrasound, the valuable measurement was endometrial thickness. The endometrial thickness was thicker in malignant cases than in benign cases (mean,
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
223
Figure 1. Histogram of endometrium and endometrial volume with vascular indices using power Doppler and VOCAL software.
8.70 mm and 14.29 mm, respectively; p < 0.001). Additive technical assessment with 3D PD-US can offer more variable parameters, such as endometrial volume, VI, FI, and VFI. The endometrial volume was larger (mean, 2.90 mL vs. 5.56 mL; p < 0.001) and all vascular indices were higher in patients with a malignant endometrium (Table 1). We analyzed the receiver operating characteristic curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for malignancy (Fig. 2). The area under the curve was statistically significant in all parameters, including endometrial
Table 1 Comparison of clinical characteristics between benign and malignant groups. Characteristics
Benign group (n ¼ 146)
Malignant group (n ¼ 28)
Age (y) Body mass index (kg/m2) Years since menopause Endometrial thickness (mm) Endometrial volume (mL) Endometrial vascularization index (%) Endometrial flow index (0e100) Endometrial vascularization flow index (0e100) Histological diagnosis Proliferative Em Secretory Em Glandular stromal dissociation Endometritis Atrophy Polyp Hyperplasia Endometrial cancer
59.27 ± 5.25 27.32 ± 6.31 9.24 ± 2.14 8.70 ± 3.91 2.90 ± 1.26 10.56 ± 6.59
61.04 29.28 11.26 14.29 5.56 17.80
7.28 8.16 4.52 3.60 1.48 3.47
0.023 0.016 0.003 <0.001 <0.001 <0.001
13.96 ± 7.24 2.77 ± 2.23
17.45 ± 6.04 9.17 ± 4.11
0.018 <0.001
29 11 10 11 53 15 17 28
± ± ± ± ± ±
p
(16.67%) (6.32%) (5.75%) (6.32%) (30.46%) (8.62%) (9.78%) (16.09%)
Data are presented as mean ± standard deviation or number (%). Comparisons between two groups were carried out using the Student t test. EM ¼ endometrium.
thickness and volume, for the prediction of endometrial cancer. Table 2 shows that endometrial thickness, volume, VI, FI, and VFI have predictive power for malignancy before the performance of invasive endometrial biopsy. The best predictive cutoff value of endometrial thickness for endometrial cancer was 9.5 mm. An endometrial volume of >4.05 mL was also predictive of endometrial malignancy. The vascular parameters were relative values, but, in this mode setting, the cutoff values were 13.07 for VI, 12.61 for FI, and 3.765 for VFI. We also analyzed the predictive property or ultrasonographic characteristics for endometrial hyperplasia (Figure 3 and Table 3). The area under the curve for endometrial volume was not significant but reflected endometrial hyperplasia. Endometrial VI and VFI were also not significant; there was no significant difference between other benign and hyperplastic endometria. However, endometrial thickness and VFI showed a significant difference between groups. The best predictive values were 7.5 mm for endometrial volume and 2.275 for VFI. Discussion Transvaginal ultrasound is currently used as the first-step technique in women with postmenopausal vaginal bleeding. It is well known that ultrasound may reliably rule out endometrial cancer when endometrial thickness is <5 mm [10]. Furthermore, application of ultrasound to women with postmenopausal bleeding is not invasive and is cost effective [11,12]. However, a thickened endometrium alone is not indicative of endometrial pathology such as malignancy. Other useful additive parameters or sonographic findings should be used to predict endometrial malignancy. Power Doppler angiography and 3D ultrasound have recently been introduced in the gynecologic field. These advanced techniques overcome the limitations of color Doppler ultrasound and conventional 2D ultrasound [13]. Three-dimensional ultrasound can render the whole endometrium even in the coronal plane, and
224
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
Figure 2. Analysis of the ROC curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for endometrial malignancy. FI ¼ flow index; ROC ¼ receiver operating characteristic; VFI ¼ vascular flow index; VI ¼ vascular index.
Figure 3. Analysis of the ROC curves of endometrial thickness, volume, VI, FI, and VFI to assess their predictive value for endometrial hyperplasia. FI ¼ flow index; ROC ¼ receiver operating characteristic; VFI ¼ vascular flow index; VI ¼ vascular index.
the targeted volume is calculated with VOCAL software or another built-in program. It has been accepted that the volume calculated using VOCAL software may represent the true tissue volume [14]. The combination of 3D ultrasound and sensitive power Doppler angiography allow for improved visualization of vessels, even for arterioles or veins with slow velocity. The calculated vascular parameters, even if it they are relative, can provide comparable indices. VI measures the number of color voxels in the volume, which represents the vessels in the tissue and is expressed as a percentage. FI is the mean color value in the color voxels, which indicates the average intensity of blood flow and is expressed as a whole number from 0 to 100. VFI is the mean value of all the color voxels in the volume, which represents both vascularization and blood flow, and is also expressed as a whole number from 0 to 100 [15]. These values are different from color Doppler vascular parameters, such as pulsatility index, resistance index, systolic/diastolic ratio, or velocity. Vascular parameters for a specific diagnostic approach can be calculated for various organs. In the present study, we evaluated the role of 3D PD-US in postmenopausal endometrial bleeding in distinguishing endometrial cancer from benign diseases. Postmenopausal endometrial bleeding is the first and most important symptom of endometrial malignancy. We also found that the studied variables can predict endometrial hyperplasia as compared with other benign diseases.
We included all presenting women with postmenopausal endometrial bleeding and with tentative diagnosis from endometrial curettage and biopsy. The women with a thin endometrium (<5 mm) were not excluded, even though a thin endometrium is not strongly associated with endometrial cancer. We wanted to analyze the natural prevalence of endometrial cancer in postmenopausal bleeding and to reduce the selection bias. Our prevalence of endometrial cancer was 16.1%, which was higher than that of the reported results and therefore does not represent the general population [3,4]. In addition to follow-up loss, exclusion of postmenopausal hormone therapy can explain the higher prevalence. Our results show that endometrial thickness volume may play an important role in discriminating endometrial cancer from benign diseases, which confirm the findings of previous published data regarding the correlation between endometrial cancer and endometrial thickness or volume [11,13,16e18]. The positive cutoff value was offered for the prediction of malignancy with an endometrial thickness of 9.5 mm, not as the passive endometrial thickness of 5 mm. This means that endometrial thickness of >5 mm requires further evaluation with methods such as endometrial biopsy, catheter sampling, whole curettage, or
Table 2 Receiver operating characteristic analysis of endometrial three-dimensional ultrasound and power Doppler angiography parameters for predicting malignancy. Ultrasonographic data
Endometrial thickness (mm) Endometrial volume (mL) Vascularization index (%) Flow index (0e100) Vascularization flow index (0e100) CI ¼ confidence interval.
Table 3 Receiver operating characteristic analysis of endometrial three-dimensional ultrasound and power Doppler angiography parameters for predicting endometrial hyperplasia. Ultrasonographic data
Benign group (n ¼ 146) versus malignant group (n ¼ 28) Area under the curve
95% CI
p
Cutoff value
0.843 0.920 0.827 0.653 0.933
0.770e0.916 0.873e0.966 0.765e0.889 0.550e0.756 0.894e0.972
<0.001 <0.001 <0.001 0.010 <0.001
9.50 4.050 13.070 12.610 3.765
Endometrial thickness (mm) Endometrial volume (mL) Vascularization index (%) Flow index (0e100) Vascularization flow index (0e100) CI ¼ confidence interval.
Other benign group (n ¼ 129) versus endometrial hyperplasia group (n ¼ 17) Area under the curve
95% CI
p
Cutoff value
0.724
0.614e0.834
0.003
0.609 0.636 0.504 0.852
0.480e0.738 0.544e0.728 0.397e0.612 0.753e0.951
0.144 0.069 0.954 <0.001
7.50
2.275
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
hysteroscopic biopsy. However, an endometrium thicker than 9.5 mm is worthy of attention and is evaluated with malignancy in mind. With regard to endometrial volume, a larger volume must be associated with a thick endometrium and, of course, can be related to endometrial cancer. We assessed the cutoff values of endometrial volume for the prediction of cancer as 4.05 mL, which is not small compared with that of the reproductive endometrium [19]. The inter- and intraobserver variations were small in the volume calculation in the 3D VOCAL program, and thus, increased endometrial volume seems to be indicative of endometrial cancer [20]. Endometrial vascular assessment for predicting malignancy has been studied in several reports [13,21e24], all of which concluded that 3D PD-US analysis of vascularization in endometrial cancer correlates with the diagnosis in women with or without endometrial bleeding. In some studies, the subendometrial vasculature calculated by 3D PD-US can also be valuable for the prediction of myometrial invasion of endometrial carcinoma [22]. In the present study, previous published data were confirmed with regard to VI, FI, and VFI. Endometrial cancer tends to have higher vasculature in the endometrium. This study differed from previous studies due to the large number of participants with bleeding, more cases with a diagnosis of cancer, and smaller prospective collection bias of this study. Thus, we can conclude that endometrial cancer increases the vasculature of the endometrium, which can be applied for the prediction of malignancy. The cutoff value in the study cannot be applied to all patients with postmenopausal endometrial bleeding because the parameters vary according to the setting mode and ultrasound model. The group with endometrial cancer showed a higher age distribution, interval from menopause, and BMI, a well-known risk factor for endometrial malignancies [16]. However, we focused on the endometrial characteristics for the prediction of malignancy, so these clinical variables were not considered for correlation with endometrial ultrasonographic characteristics. We focused on the relationship of endometrial characteristics evaluated by ultrasound and endometrial hyperplasia, which has been studied in a previous report [21]. The measurements for endometrial thickness, volume, VI, and FI were significantly higher in both the hyperplasia and the carcinoma groups [21]. Discrimination of hyperplasia was the only meaningful analysis of the study, and the significant variables were endometrial volume and VFI. This finding can be explained by the fact that postmenopausal women have the limitation of volume increase for a decreased uterus and thus a limitation in measuring vascularity. However, typical endometrial thickening >7.5 mm is meaningful for the prediction of endometrial hyperplasia. The cutoff of VFI for endometrial hyperplasia is smaller than that of endometrial cancer, but greater endometrial vasculature can be associated with endometrial hyperplasia. We prospectively analyzed, beyond the previous studies, with a larger group of women with postmenopausal endometrial bleeding. The various analyses were not permitted for the limitation of the recruitment of patients, and a randomized controlled trial is needed to clarify the results of this study. Nevertheless, the present study demonstrated the predictive values for endometrial cancer and hyperplasia. Based on the findings and assessment of the present study, although endometrial thickness appears to be the simplest with 2D ultrasound, endometrial thickness and VFI were more accurate in diagnosing endometrial hyperplasia according to the receiver operating characteristic curves. Furthermore, all parameters of the endometrium, such as thickness, volume, VI, FI, and VFI, have higher scores for endometrial malignancy in women with postmenopausal endometrial bleeding. Thus, physicians must focus on
225
endometrial thickness >9.5 mm for malignancy and >7.5 mm for hyperplasia. Conflicts of interest The authors have no conflicts of interest relevant to this article. Acknowledgments This study was supported by a grant from Kosin University College of Medicine (2012). References [1] Astrup K, Olivarius Nde F. Frequency of spontaneously occurring postmenopausal bleeding in the general population. Acta Obstet Gynecol Scand 2004;83:203e7. [2] Guruwadayarhalli B, Jones SE, Srinivasan V. Hysteroscopy in the diagnosis of postmenopausal bleeding. Menopause Int 2007;13:132e4. [3] Rose PG. Endometrial carcinoma. NEJM 1996;335:640e9. [4] Raouf SA, Gupta P, Papaioannou S, Pradhan P. Endometrial thickness for invasive investigations in women with postmenopausal bleeding. Climacteric 2011;14:117e20. [5] Partridge EE, Shingleton HM, Menck HR. The National Cancer Data Base report on endometrial cancer. J Surg Oncol 1996;61:111e23. [6] Kong TW, Lee KM, Cheong JY, Kim WY, Chang S, Yoo S, et al. Comparison of laparoscopic versus conventional open surgical staging procedure for endometrial cancer. J Gynecol Oncol 2010;21:106e11. [7] Burbos N, Musonda P, Giarenis I, Shiner AM, Giamougiannis P, Morris EP, et al. Predicting the risk of endometrial cancer in postmenopausal women presenting with vaginal bleeding: the Norwich DEFAB risk assessment tool. Br J Cancer 2010;102:1201e6. [8] Dreisler E, Sorensen SS, Ibsen PH, Lose G. Value of endometrial thickness measurement for diagnosing focal intrauterine pathology in women without abnormal uterine bleeding. Ultrasound Obstet Gynecol 2009;33:344e8. [9] Goldstein RB, Bree RL, Benson CB, Benacerraf BR, Bloss JD, Carlos R, et al. Evaluation of the woman with postmenopausal bleeding: Society of Radiologists in Ultrasound-Sponsored Consensus Conference statement. J Ultrasound Med 2001;20:1025e36. [10] Gupta JK, Chien PF, Voit D, Clark TJ, Khan KS. Ultrasonographic endometrial thickness for diagnosing endometrial pathology in women with postmenopausal bleeding: a meta-analysis. Acta Obstet Gynecol Scand 2002;81: 799e816. [11] Clark TJ, Barton PM, Coomarasamy A, Gupta JK, Khan KS. Investigating postmenopausal bleeding for endometrial cancer: cost-effectiveness of initial diagnostic strategies. BJOG 2006;113:502e10. €lmann HA, Heintz AP. Cost-effectiveness of the use [12] Dijkhuizen FP, Mol BW, Bro of transvaginal sonography in the evaluation of postmenopausal bleeding. Maturitas 2003;45:275e82. [13] Alcazar JL, Galvan R. Three-dimensional power Doppler ultrasound scanning for the prediction of endometrial cancer in women with postmenopausal bleeding and thickened endometrium. Am J Obstet Gynecol 2009;200: 44.e1e6. [14] Martins WP, Ferriani RA, Barra DA, Dos Reis RM, Bortolieiro MA, Nastri CO, et al. Reliability and validity of tissue volume measurement by threedimensional ultrasound: an experimental model. Ultrasound Obstet Gynecol 2007;29:210e4. [15] Kim A, Han JE, Yoon TK, Lyu SW, Seok HH, Won HJ. Relationship between endometrial and subendometrial blood flow measured by three-dimensional power Doppler ultrasound and pregnancy after intrauterine insemination. Fertil Steril 2010;94:747e52. [16] Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I. Endometrial cancer. Lancet 2005;366:491e505. [17] Burbos N, Musonda P, Crocker SG, Morris EP, Duncan TJ, Nieto J. Outcome of investigations for postmenopausal vaginal bleeding in women under the age of 50 years. Gynecol Oncol 2012;125:120e3. [18] Karlsson B, Granberg S, Wikland M, Ylstalo P, Torvid K, Marsal K, et al. Transvaginal ultrasonography of the endometrium in women with postmenopausal bleedingda Nordic multicenter study. Am J Obstet Gynecol 1995;172:1488e94. [19] Zollner U, Specketer M, Dietl J, Zollner K. 3D-Endometrial volume and outcome of cryopreserved embryo replacement cycles. Arch Gynecol Obstet 2012;286:517e23. [20] Martins WP, Raine Fenning NJ, Leite SP, Ferriani RA, Nastri CO. A standardized measurement technique may improve the reliability of measurements of endometrial thickness and volume. Ultrasound Obstet Gynecol 2011;38: 107e15. [21] Odeh M, Vainerovsky I, Grinin V, Kais M, Ophir E, Bornstein J. Three-dimensional endometrial volume and 3-dimensional power Doppler analysis in
226
A. Kim et al. / Taiwanese Journal of Obstetrics & Gynecology 54 (2015) 221e226
predicting endometrial carcinoma and hyperplasia. Gynecol Oncol 2007;106: 348e53. ~ G, Alca L, Jurado M, Mínguez JA, Lo pez-GarcAa zar JL. Three[22] Galv an R, Merce dimensional power Doppler angiography in endometrial cancer: correlation with tumor characteristics. Ultrasound Obstet Gynecol 2010;35:723e9. [23] Lieng M, Qvigstad E, Dahl GF, Istre O. Flow differences between endometrial polyps and cancer: a prospective study using intravenous contrast-enhanced
transvaginal color flow Doppler and three-dimensional power Doppler ultrasound. Ultrasound Obstet Gynecol 2008;32:935e40. €enp€ € JU. Preoperative assessment [24] Saarelainen SK, Vuento MH, Kirkinen P, Ma aa of endometrial carcinoma by three-dimensional power Doppler angiography. Ultrasound Obstet Gynecol 2012;39:466e72.