Clinical Imaging xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: http://www.clinicalimaging.org
Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings☆ Guo-Fu Zhang a, He Zhang a,⁎, Xiao-Mei Tian a, Hao Zhang b a b
Department of Radiology, Obstetrics and Gynecology Hospital, Fudan University Department of Pathology, Obstetrics and Gynecology Hospital, Fudan University
a r t i c l e
i n f o
Article history: Received 9 March 2014 Received in revised form 19 May 2014 Accepted 2 June 2014 Available online xxxx Keywords: Uterine sarcoma Magnetic resonance imaging Diffusion-weighted imaging Apparent diffusion coefficient
a b s t r a c t Objectives: We investigated the utility of magnetic resonance imaging (MRI) and diffusion-weighted imaging (DWI) in the categorization of uterine sarcoma (US) and compared them with pathological findings. Methods: The baseline and MRI characteristics were recorded and compared across the subtypes of USs. Results: There were no differences in the conventional or DWI signals among the four subtypes of US, except in the heterogeneity of T2-weighted imaging. A difference in the mean apparent diffusion coefficient value for USs and uterine fibroids differed significantly (P=.019). Conclusions: MRI characteristics showed no specific differences between any subtypes of US.
© 2014 Elsevier Inc. All rights reserved.
1. Introduction Uterine sarcoma (US) is rare malignant tumors of mesenchymal origin, accounting for about 1%–3% of uterine malignancies [1]. They are histologically categorized into four common variants: leiomyosarcoma (LS), endometrial stromal sarcoma (ESS), adenosarcoma (AS), and carcinosarcoma (CS) [2]. USs display aggressive biological behavior, and the prognosis is very poor, with a reported 5-year survival rate of only 22%–39%, even when they are detected at an early stage [3]. Because the prevalence of US is low, a preoperative diagnosis is often difficult, although it is critical for stratifying patients to case-effective treatments. Magnetic resonance imaging (MRI) has been widely used in obstetrics and gynecology in recent decades [4–10]. Because it offers the advantages of superb soft-tissue resolution, no radiation, and multiplanar imaging, MRI plays a key role in determining the etiology of any suspected malignancy in the female genital tract detected with ultrasound or computed tomography [11–13]. Diffusion-weighted imaging (DWI) is a functional imaging technique currently available in the majority of clinical MRI units. By displaying water molecule mobility (Brownian motion), DWI permits the quantitative evaluation of tumor tissues and has shown promising results in distinguishing malignancies from benign gynecological lesions in several recently reported studies [6,12,14–17]. ☆ We declare that we have no conflicts of interest. ⁎ Corresponding author. No. 419 Fang xie Road, Shanghai, 200011, P.R. China. Tel.: + 86- 21 63770161; fax: + 86 21 63770768. E-mail address: [email protected] (H. Zhang).
However, research that uses MRI or DWI to image USs is still limited. Therefore, the purpose of the study was twofold: first, to clarify the MRI and DWI-MRI findings for the four subtypes of US in a relatively large sample and to compare them with the pathological findings at our single institution; and second, to determine whether DWI-MRI is useful in discriminating USs from other benign uterine conditions. 2. Materials and methods 2.1. Study subjects From June 2010 to December 2012, 437 consecutive patients with clinically suspected gynecological disease underwent prospective MRI examinations before pelvic or laparoscopic surgery at our institution. Twenty-two patients (19–73 years of age; average age, 52.7±12.3 years) with pathologically proven US were included in this study. To compare the apparent diffusion coefficients (ADCs) values of benign and malignant uterine lesions, we also selected a limited numbers of leiomyomas and adenomyomas for the comparative group. The criteria were as follows: first, to minimize the selection bias caused by multiple lesions in one patient, we chose only patients with a solitary lesion for either the leiomyoma or adenomyoma group; second, patients with any previous pelvic surgery or radiation history were arbitrarily excluded because the inherent structure of the uterus may have been altered. Thus, 21 leiomyomas (24–72 years of age; average age, 47.7±13.9years) and 20 adenomyomas (34–51 years of age; average age, 41.4±5.2years) were included. Our institutional review board approved the study, and the requirement for informed consent was waived for all participants.
http://dx.doi.org/10.1016/j.clinimag.2014.06.004 0899-7071/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
2
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Table 1 Details of parameters for MRI protocols Parameters
T1WI
T2WI
FS-T2WI
DWI
Contrast-enhanced MRI
Repetition/Echo time (msec) Echo trains per slice Sequence Bandwidth (Hz) Thickness (mm) Field of view (mm) Voxel size (mm) Flip angle (degrees)
550/10 44 TSE 178 4 350 (APa) 1.5×1.1×4.0 150
4000/83 19 TSE 260 4 350 (LRb) 1.1×1.1×4.0 144
8000/83 19 TSE 260 4 350 (LR, AP) 1.4×1.4×4.0 150
2800/81
4.89/2.38
EP2D 1250 5 300 (AP) 2.7×1.9×5.0
VIBEc 400 3 380 (LR, AP) 1.7×1.2×3.0 10
a b c
AP indicated that the scan plane was from anterior to posterior. LR indicated that the scan plane was from left to right. Volumetric Interpolated Breath-hold Examination.
2.2. Image acquisition MRI was performed using a 1.5-T MR system (Magnetom Avanto, Siemens, Erlangen, Germany) with a phased-array coil. The routine MRI protocols used for the assessment of pelvic masses included the axial turbo spin-echo (TSE) T1-weighted imaging (T1WI), sagittal TSE T2-weighted imaging (T2WI), and axial/sagittal TSE fat-suppressed T2WI (FS T2WI). For axial images, the transverse plane was perpendicular to the long axis of uterine body; for sagittal images, the longitudinal plane was parallel to the main body of uterus. DWI using an echo-planar imaging two-dimensional (EP2D) sequence performed in the axial plane with parallel acquisition technique by using b value=0, 100, and 800 s/mm 2. Contrast-enhanced pelvic imaging was acquired at multiple phases of contrast medium enhancement in both sagittal and axial planes. The detailed MRI acquisition parameters were listed in Table 1. 2.3. Image analysis The location, size (largest dimension in two orthogonal planes), shape (round, oval, and lobulated), and margin (clear or ambiguous); visibility of endometrium and hemorrhage component (high signal on T1WI) within the lesion; and presence of pelvic-free fluid and lymph nodes were noted. Further, the homogeneity of T1WI/T2WI signals, accompanying lesions and extrauterine extension were separately evaluated and recorded. On T1WI, hypo-, iso-, and hyperintensity were similar for the pelvic fluid, pelvic wall muscle, and fat signal; on T2WI, hypo-, iso-, and hyperintensity were similar for the pelvic bone, pelvic wall muscle, and fat signal; on b=800 mm−2/s DWI images, the low, intermediate, and high-signal intensities were similar for the pelvic bone, myometrium, and endometrium. After the intravenous injection of the contrast medium, the degree of lesion enhancement was graded as follows: 1, weak enhancement (less than the myometrium); 2, mild enhancement (equal to the myometrium); 3, avid enhancement (superior to the myometrium). ADCs were measured manually on postprocessing workstation (Leonardo, Siemens, Germany) by one reviewer (H.Z.). Two observers (G. F. Z. and H.Z., with 10 and 6 years of experience in gynecological imaging, respectively), who were blinded to the histological results independently, analyzed MRI datasets of each participant. At the end of the study, two observers were also required to give the tumor stage according to new FIGO system for US [18]. For interobserver discrepancies in the evaluation of uterine lesions, consensus was achieved. 2.4. Statistical analyses Continuous variables were expressed as the means±standard deviation (S.D.) and compared with the unpaired t test if normally distributed or the Mann–Whitney test if not normally distributed. A nonparametric test (Mann–Whitney) was used to test other nonparametric variables within each group. Kappa statistics were used to
Table 2 The details of baseline and MRI characteristics of four subtypes of the histologically proven 22 USs Baseline characteristics
Age (years) Menstruation Premenopause Perimenopausal Menopause Symptomsa Pelvic mass Virginal bleeding (discharge) Menstruation disorder MRI characteristicsb Maximum diameter (mm) Component Solid Cyst Mixed Hemorrhage Present/Absent Margin Clear/Unclear Extrauterine extension Yes/No Accompanying lesion Present/Absent Normal endometrium Present/Absent T1WI signal character Low Intermediate High Mixed Homogeneous/ Inhomogeneous T2WI signal character Low Intermediate High Mixed Homogeneous/ Inhomogeneous DWI signal character Low Intermediate High Mixed ADC value (10−3/mm2/s)c Enhancement Avid enhancement Moderate enhancement Weak enhancement
Pathology group
Total
LS
ESS+AS
CS
53±10.6
51±16.8
55±6.7
3 2 2
2 1 6
2 2 2
7 5 10
4 3
1 7
5
5 15
1
1
2
55.6±49.7
49.4±31.9
36.3±19.5
6
6
5
1
2
5/2
2/6
1/4
8/12
3/4
6/2
3/2
12/8
4/3
0/8
0/5
4/16
4/3
1/8
1/5
6/16
3/4
3/5
3/2
9/11
3 1
8
4
3 2/5
5/3
1 4/1
3 13 3 4 11/9
6 0/7
4 2 2 4/4
1 2 2 1/4
6 4 10 5/15
1 4 2 0.93±0.30
1 5 2 1.19±0.33
5
2 14 4
6
3 5
P value
.034
1
17 3
.021
1.05±0.22 .05
1
5
9 10 1
a If both symptoms were observed, then the virginal bleeding (discharge) was seemed as the primary symptom. b One CS and one AS could not be evaluated at MRI for the undetected lesion. c ADC values could not been calculated in one case in AS and two cases in CS group because the small-sized (b10 mm) lesions were not detected on DWI-MRI images.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
3
Fig. 1. A 53-year-old female patient with histologically proven LS (Stage Ib). (A) Axial T1WI revealed an oval soft mass occupying the left part of cavity, in which patchy high signals (arrowhead) were displayed, indicating the hemorrhage components. (B) Axial FS-T2WI revealed the solid tumor exhibited inhomogeneous signals. On DWI-MRI (b=800s/mm2) (C) the lesion appeared as the inhomogeneous high signal with foci of low signal and on ADC map (D), it displayed the homogenous intermediate signal. On early (E) and delay (F) contrast-enhanced images, the solid parts of the tumor displayed the obvious enhancement, while necrosis components did not enhance. Enhancement type of tumor (red line) was well demonstrated compared with normal myometrium (yellow line) and muscle tissue (green line) (G, H).
calculate the interobserver reliability as percentage of agreement between the observer evaluations before the consensus reading was made. The area under the receiver operating characteristic curve (AUC) was calculated for ADCs to discriminate US from leiomyoma. SPSS (version 13.0, SPSS, Chicago, USA) was used to perform statistical analyses. P values b .05 were considered statistically significant.
3. Results 3.1. Baseline characteristics The histopathological specimens revealed 22 primary USs, including seven LSs (39–73 years of age; average age, 53.0±10.6 years),
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
4
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Fig. 2. A 28-year-old female patient with histologically proven ESS (Stage IIIc). (A) Axial T1WI showed a homogeneous solid mass with isointensity, in the right uterine body. (B) On sagittal T2WI, the tumor occupied the whole anterior wall of the uterus, and the inhomogeneous signal and normal endometrium were displaced posteriorly (arrowheads). On DWIMRI (b=800 s/mm2) (C), the lesion appeared as a solid mass with inhomogeneous high and intermediate signal on the ADC map (D). The tumor displayed inhomogeneous enhancement on arterial-phase contrast-enhanced images (E), and later, the signal intensity was similar to that of the myometrium on venous-phase contrast-enhanced images (F). Note that the lymphoadenopathy near bilateral parailiac arteries (arrow) was clearly evident, which was confirmed in subsequently evaluated histopathological specimens. The tumor was clearly enhanced (red line) compared with the normal myometrium (green line) and muscle tissue (yellow line) (G, H).
three ESSs (19–66 years of age; average age, 37.7±24.9 years), six ASs (49–65 years of age; average age, 58.0±6.0 years), and six CSs (43–62 years of age; average age, 54.7±6.7 years). Three tumors were in Stage I, one in Stage II, and three in Stage III in the LS group; one tumor
was in Stage I, one in Stage II, and one in Stage III in the ESS group; six tumors were in Stage I in the AS group; and five tumors were in Stage I and one in Stage III in the CS group. The details of baseline characteristics for each subtype of US are summarized in Table 2.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
5
Fig. 3. A 19-year-old female patient with histologically proven ESS (Stage Ic). (A) Axial T1WI outlined a homogeneous solid mass (arrowheads) with isointensity, in the right uterine body. Note the high signals indicative of fluid in the cavity (*). (B) On T2WI, the well-defined tumor margin was located in the myometrium with isohyperintensity. On DWIMRI (b=800 s/mm2) (C) and the ADC map (D), the lesion was well displayed (arrowheads) and appeared as a homogeneous intermediate signal. The tumor showed homogeneously avid enhancement on contrast-enhanced images (E), and on a time–signal intensity curve (F), the degree of enhancement of the lesion (red line) was greater than that of the normal myometrium (yellow line).
3.2. MRI characteristics Two lesions (one AS and one CS) could not be detected on MRI by either observer because the lesions were too small. Eleven USs (11/20, 55%; Fig. 1) were confined to the uterine cavity, four (4/20, 20%) to the
Table 3 The details of locations of four pathological types in the histologically proven 22 USs on MRI Pathology
Cavitya
LocationCervical canal
Myometrium
Total
LS ESS ASb CSb Total
3
1 1 1 1 4
3 2
7 3 5 5 20
a
4 4 11c
5
If the lesion involved both cavity and cervical canal and the main body is located in the cavity, then the lesion was regarded in the cavity. b There was one case in both AS and CS group wherein the lesion was too small and could not be detected on MRI. c There was a significant difference in the location of US occurrence between cavity and myometrium (P=.032).
cervical canal, and five (5/20, 25%) to the myometrium. Two ESSs were located in the anterior part of the uterine body (Figs. 2, 3, Table 3). The uterine cavity was more likely to be involved than the other two sites (P= .032). The maximum diameter of the lesions in the LS group was largest (55.6±49.7 mm) among the four subtypes. On MRI, most of the USs (17/20, 85%) appeared as a solitary, solid tumor. In this study, 12 lesions were clearly demarcated, but in another eight patients, the margin was ambiguous. On T1WI, most patients (13/20, 65%) showed intermediate signals. On T2WI, the tumor signals appeared more diverse, and relatively, the LS lesions (6/22, 27.3%) showed more mixed signals than the other groups. In terms of T1WI/T2WI signal heterogeneity, the T1WI signals of the lesions in each group had no specific character. All tumors (7/7, 100%) in the LS group had an inhomogeneous signal on T2WI, which was clearly different from those of the ESS/AS group (P=.021) and CS group (P=.003). On MRI, normal endometrium could be seen in 11 patients. Extrauterine extension was only observed in the LS group (4/20, 20%). Hemorrhage components were usually observed in the LS group (5/7, 71.4%), which appeared as patchy high signals on T1WI images. However, hemorrhage signals were less well detected in the ESS, AS, and CS groups.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
6
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Fig. 4. A 59-year-old patient with histologically proven CS (Stage Ia). An oval mass centrally located in the uterine cavity (*) with homogeneous isointensity on axial T1WI (A) and T2WI (B) was vaguely displayed. On DWI-MRI (b=800 s/mm2) (E), the lesion displayed homogeneous high signal and intermediate signal on the ADC map (F). On early (C) and delayed (D) enhanced images, the tumor displayed moderate enhancement, whereas the subendometrial enhancement line (arrowheads) was preserved. The enhancement of the tumor (red line) was similar to that of the myometrium, although the delay enhancement was observed compared with that of the myometrium (G, H).
On contrast-enhanced MRI images, most tumors (6/7, 85.7%) in the LS group showed rapid and avid enhancement. In the ESS/AS group, most tumors (5/8, 63%) appeared mild enhancement, and in the CS group, all lesions (5/5, 100%) showed moderate and delayed enhancement (Fig. 4). There was a little statistically significant difference in the level of enhancement between any groups (P=.05).
3.3. DWI-MRI characteristics On DWI-MRI images, 14 tumors (14/20, 70%) showed a homogeneously high signal. The mean ADC value was 0.93±0.30 in the LS group, 1.19±0.33 in the ESS/AS group, and 1.05±0.22 in the CS group, with a little overlap (Fig. 5). A significant difference in ADCs was only observed
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
7
Fig. 5. Box-and-whisker plots of the calculated ADC values (10−3/mm2/s) for the US subtypes. There were no significant differences in mean ADC value of the US subtypes. A significant difference in ADCs was only observed between LS (0.93±0.30) and leiomyoma (1.34±0.42) (P=.018).
between LS and leiomyoma (P=.018). When the leiomyomas and adenomyomas were compared, the mean ADC value for US (1.05±0.30) was lower than that for leiomyoma (1.34±0.42) and adenomyoma (1.12±0.14) (Fig. 6). There was a significant difference between the mean ADC value for US and fibroids (P=.019). If the cutoff value for ADC was set at 1.06×10−3/mm2/s, its sensitivity and specificity were 81% and 62%, respectively. The AUC was 0.72 (95% confidence interval, 0.556–0.883) for ADC in discriminating LS from leiomyoma.
were confirmed by the histopathological findings. When compared with the final histological staging, both two reviewers did correct tumor staging based on MRI images (Figs. 1–2, 3–4). There was excellent agreement between the two observers in both MRI findings (kappa=0.90) and MRI tumor staging (kappa=0.93).
3.4. Interobserver agreement and radiological–pathological correlations
Until now, most of the literature related to the MRI features of US have been published as case reports [9,15,19,20]. Based on this limited literature, several reviews of US imaging are especially noted [1,13,21].
Tumor extension, necrosis, hemorrhage, and lymphoadenopathy were clearly observed on both pre- and postcontrast MRI images and
4. Discussion
Fig. 6. Box-and-whisker plots of the calculated ADC values (10−3/mm2/s) of three groups. The mean ADC value for leiomyoma (1.34±0.42) was higher than that for US (1.05±0.30), with a little overlap. The mean ADC value for US and leiomyoma differed significantly (P=.019).
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
8
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
LS is one of the smooth muscle tumors and comprises almost half of all USs [22]. It is sometimes difficult to differentiate histologically or radiologically between LS and leimyoma in which the process of malignant transformation has occurred [21,23]. It has been suggested that the presence of high-signal intensity on T2WI/T1WI and irregular tumor margins are indicators of LS that distinguish it from leiomyoma [24]. Our findings support this view because necrosis or hemorrhage components were always observed on the T1WI (5/7) and T2WI (7/7) images of the rapidly growing mass (Fig. 1), which displayed highsignal intensity on T1WI and an inhomogeneous signal on T2WI. In the present study, most LS tumors (6/7) showed marked enhancement on contrast-enhanced images [1]. ESS is an indolent neoplasm with a favorable prognosis involving long-term survival but with a propensity for late recurrence and metastasis [13]. In our study, one tumor was located in the vestigial part of the cervix after a hysterectomy, and two tumors were located in the myometrium. Three ESSs, in patients in whom the normal endometrium was well-preserved, invariably appeared as solid masses with homogeneous signals and avid enhancement on MRI, which mimicked LS or adenomyoma. Similar findings were also described in another study [19]. In this context, the MRI features are nonspecific for the diagnosis of ESS, and a histological diagnosis is mandatory to confirm the etiology of the tumor. AS of the uterus is a rare low-grade mixed epithelial mesenchymal tumor containing benign glandular epithelial elements and malignant mesenchymal elements and constitutes only 8% of all USs [25]. The imaging features of AS are similar to those of CS; it is typically seen as a large polyploidal mass within the endometrial cavity, which may protrude into the endocervical canal. In one study [9], the authors described foci of hyperintense signal within the mass on T2WI, reflecting glandular epithelial components, which may be a characteristic finding. In the newly revised histopathological classification of US, CS is reclassified as a subtype of endometrial carcinoma [18]. Because CS behaves more aggressively than ordinary endometrial carcinomas, it is still included in most studies of US, as in this study. Upon imaging, CSs are predominantly endometrial-based heterogeneous masses causing distension of the endometrial cavity and associated with varying degrees of myometrial invasion. The craniocaudal dimension of the CS is greater than that of endometrial carcinomas [26]. They also usually show more marked contrast enhancement than endometrial carcinomas, which are usually hypovascular [27]. In our study, all cases of AS and CS displayed heterogeneous enhancement on MRI, which was similar to the appearance of the myometrium. The degree of enhancement of the ESS/AS and CS lesions was less than that of the LS lesions. In this study, tumor staging by MRI correlated perfectly with the histopathological staging. Interobserver agreement was excellent, also indicating that MRI is a reliable imaging modality for accurate preoperative staging. As the parametric index in DWI imaging, ADC may provide a quantitative measurement for any suspected tumor tissues [16,28] with encouraging results in distinguishing malignancies from benign diseases of the female genital tract [12,17,29,30]. Tamai et al. reported in their study that for b= 500, 1000 s/mm 2 DWI images, the mean ADC value (10 −3mm 2/s) of LS was 1.17±0.15, which was lower than that of the normal myometrium (1.62±0.11) or degenerated leiomyomas (1.70±0.11), without any overlap [6]. In another study, the authors showed that on b= 1000 s/mm 2 DWI images, endometrial carcinomas, and CSs had lower ADCs (less than 1.15×10 −3mm 2/s) than endometrial polyps or submucosal leiomyomas [14]. In a recent study, the authors also reported that ADCs could be used as a useful reference tool with which to distinguish benign uterine lesions from malignancies [12]. Our results are consistent with the results reported above. In the present study, the mean ADC value for b= 800 s/mm 2 DWI images of US was lower than those for leiomyoma or adenomyoma, whereas a significant difference in the mean ADC value for LS and leiomyoma was also observed. However, there was no significant difference observed in the ADCs for US and adenomyoma.
To the best of our knowledge, this result has not been described in previously published studies. There were several limitations of our study. First, although the number of USs studied was the largest sample so far reported from one institution, the numbers of histopathological subtypes of US (especially ESS) were still limited and unevenly distributed. To investigate prospective MRI findings in larger samples will be needed to clarify the true differences between the subtypes. Second, the regions of interest, in addition to the ADCs, were manually drawn and individually calculated on a case-by-case basis, and this lack of standardization may have influenced the final results. In summary, MRI characteristics show no specific sign between any subtype of US, except for the heterogeneity of T2WI signal. LS often shows more intense enhancement than the other subtypes, and ADCs could be useful in discriminating USs from uterine fibroids. MRI can be used as a reliable imaging modality for staging preoperative US.
References [1] Sohaib SA, Verma H, Attygalle AD, Ind Thomas EJ. Imaging of uterine malignancies. Semin Ultrasound CT MR 2010;31:377–87. [2] D'Angelo E, Prat J. Uterine sarcomas: a review. Gynecol Oncol 2010;116:131–9. [3] Uterine sarcoma. In: Schorge JO, Schaffer JI, Halvorson LM, et al, editors. Williams gynecology. Columbus, OH: The Mc-Graw-Hill Companies; 2008. p. 706–16. [4] Sala EWS, Senior E, Lomas D. Mri of malignant neoplasms of the uterine corpus and cervix. Am J Roentgenol 2007;188:1577–87. [5] Teo SYBK, Peters HE, Mortele KJ. Primary malignant mixed müllerian tumor of the uterus: findings on sonography, ct, and gadolinium-enhanced mri. Am J Roentgenol 2008;191:278–83. [6] Tamai K, Koyama T, Saga T, Morisawa N, Fujimoto K, Mikami Y, Togashi K. The utility of diffusion-weighted mr imaging for differentiating uterine sarcomas from benign leiomyomas. Eur Radiol 2008;18:723–30. [7] Tanaka YO, Tsunoda H, Minami R, Yoshikawa H, Minami M. Carcinosarcoma of the uterus: Mr findings. J Magn Reson Imaging 2008;28:434–9. [8] Namimoto T, Yamashita Y, Awai K, Nakaura T, Yanaga Y, Hirai T, Saito T, Katabuchi H. Combined use of t2-weighted and diffusion-weighted 3-t mr imaging for differentiating uterine sarcomas from benign leiomyomas. Eur Radiol 2009;19:2756–64. [9] Takeuchi M, Matsuzaki K, Yoshida S, Kudo E, Bando Y, Hasebe H, Kamada M, Nishitani H. Adenosarcoma of the uterus: magnetic resonance imaging characteristics. Clin Imaging 2009;33:244–7. [10] Cornfeld D, Israel G, Martel M, Weinreb J, Schwartz P, McCarthy S. Mri appearance of mesenchymal tumors of the uterus. Eur J Radiol 2010;74:241–9. [11] Stamatopoulos CP, Mikos T, Grimbizis GF, Dimitriadis AS, Efstratiou I, Stamatopoulos P, Tarlatzis BC. Value of magnetic resonance imaging in diagnosis of adenomyosis and myomas of the uterus. J Minim Invasive Gynecol 2012;19:620–6. [12] Thomassin-Naggara I, Dechoux S, Bonneau C, Morel A, Rouzier R, Carette MF, Daraï E, Bazot M. How to differentiate benign from malignant myometrial tumours using mr imaging. Eur Radiol 2013;23:2306–14. [13] Tirumani S, Ojili V, Shanbhogue A, Fasih N, Ryan J, Reinhold C. Current concepts in the imaging of uterine sarcoma. Abdom Imaging 2013;38:397–411. [14] Fujii S, Matsusue E, Kigawa J, Sato S, Kanasaki Y, Nakanishi J, Sugihara S, Kaminou T, Terakawa N, Ogawa T. Diagnostic accuracy of the apparent diffusion coefficient in differentiating benign from malignant uterine endometrial cavity lesions: Initial results. Eur Radiol 2008;18:384–9. [15] Fujii S, Kaneda S, Tsukamoto K, Kakite S, Kanasaki Y, Matsusue E, Kaminou T, Ogawa T. Diffusion-weighted imaging of uterine endometrial stromal sarcoma: a report of 2 cases. J Comput Assist Tomogr 2010;34:377–9 [310.1097/ RCT.1090b1013e3181cfc1676]. [16] Sala E, Rockall A, Rangarajan D, Kubik-Huch RA. The role of dynamic contrastenhanced and diffusion weighted magnetic resonance imaging in the female pelvis. Eur J Radiol 2010;76:367–85. [17] Zhang H, Zhang G-F, He Z-Y, Li Z-Y, Zhu M, Zhang G-X. Evaluation of primary adnexal masses by 3 t mri: categorization with conventional mr imaging and diffusion-weighted imaging. J Ovarian Res 2012;5:33. [18] Prat J. Figo staging for uterine sarcomas. Int J Gynecol Obstet 2009;104:177–8. [19] Furukawa R, Akahane M, Yamada H, Kiryu S, Sato J, Komatsu S, Inoh S, Yoshioka N, Maeda E, Takazawa Y, Ohtomo K. Endometrial stromal sarcoma located in the myometrium with a low-intensity rim on t2-weighted images: report of three cases and literature review. J Magn Reson Imaging 2010;31:975–9. [20] Kato H, Kanematsu M, Furui T, Imai A, Hirose Y, Kondo H, Goshima S, Tsuge Y. Carcinosarcoma of the uterus: radiologic–pathologic correlations with magnetic resonance imaging including diffusion-weighted imaging. Magn Reson Imaging 2008;26:1446–50. [21] Qiu L-L, Yu R-S, Chen Y, Zhang Q. Sarcomas of abdominal organs: computed tomography and magnetic resonance imaging findings. Semin Ultrasound CT MR 2011;32:405–21. [22] Chiang S, Oliva E. Recent developments in uterine mesenchymal neoplasms. Histopathology 2013;62:124–37.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx [23] Lim D, Alvarez T, Nucci MR, Gilks B, Longacre T, Soslow RA, Oliva E. Interobserver variability in the interpretation of tumor cell necrosis in uterine leiomyosarcoma. Am J Surg Pathol 2013;37:650–8 [610.1097/PAS.1090b1013e3182851162]. [24] Tanaka YO, Nishida M, Tsunoda H, Okamoto Y, Yoshikawa H. Smooth muscle tumors of uncertain malignant potential and leiomyosarcomas of the uterus: Mr findings. J Magn Reson Imaging 2004;20:998–1007. [25] Chourmouzi DBG, Zarampoukas T, Drevelengas A. Sonography and mri of tamoxifen-associated müllerian adenosarcoma of the uterus. Am J Roentgenol 2003;181:1673–5. [26] Genever AV, Abdi S. Can mri predict the diagnosis of endometrial carcinosarcoma? Clin Radiol 2011;66:621–4.
9
[27] Bharwani NNA, Tunariu N, Babar S, Sahdev A, Rockall AG, Reznek RH. Mri appearances of uterine malignant mixed müllerian tumors. Am J Roentgenol 2010;195:1268–75. [28] Punwani S. Diffusion weighted imaging of female pelvic cancers: concepts and clinical applications. Eur J Radiol 2011;78:21–9. [29] Zhang P, Cui Y, Li W, Ren G, Chu C, Wu X. Diagnostic accuracy of diffusionweighted imaging with conventional mr imaging for differentiating complex solid and cystic ovarian tumors at 1.5t. World J Surg Oncol 2012;10:237. [30] Li W, Chu C, Cui Y, Zhang P, Zhu M. Diffusion-weighted mri: a useful technique to discriminate benign versus malignant ovarian surface epithelial tumors with solid and cystic components. Abdom Imaging 2012;37:897–903.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
Contents lists available at ScienceDirect
Clinical Imaging journal homepage: http://www.clinicalimaging.org
Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings☆ Guo-Fu Zhang a, He Zhang a,⁎, Xiao-Mei Tian a, Hao Zhang b a b
Department of Radiology, Obstetrics and Gynecology Hospital, Fudan University Department of Pathology, Obstetrics and Gynecology Hospital, Fudan University
a r t i c l e
i n f o
Article history: Received 9 March 2014 Received in revised form 19 May 2014 Accepted 2 June 2014 Available online xxxx Keywords: Uterine sarcoma Magnetic resonance imaging Diffusion-weighted imaging Apparent diffusion coefficient
a b s t r a c t Objectives: We investigated the utility of magnetic resonance imaging (MRI) and diffusion-weighted imaging (DWI) in the categorization of uterine sarcoma (US) and compared them with pathological findings. Methods: The baseline and MRI characteristics were recorded and compared across the subtypes of USs. Results: There were no differences in the conventional or DWI signals among the four subtypes of US, except in the heterogeneity of T2-weighted imaging. A difference in the mean apparent diffusion coefficient value for USs and uterine fibroids differed significantly (P=.019). Conclusions: MRI characteristics showed no specific differences between any subtypes of US.
© 2014 Elsevier Inc. All rights reserved.
1. Introduction Uterine sarcoma (US) is rare malignant tumors of mesenchymal origin, accounting for about 1%–3% of uterine malignancies [1]. They are histologically categorized into four common variants: leiomyosarcoma (LS), endometrial stromal sarcoma (ESS), adenosarcoma (AS), and carcinosarcoma (CS) [2]. USs display aggressive biological behavior, and the prognosis is very poor, with a reported 5-year survival rate of only 22%–39%, even when they are detected at an early stage [3]. Because the prevalence of US is low, a preoperative diagnosis is often difficult, although it is critical for stratifying patients to case-effective treatments. Magnetic resonance imaging (MRI) has been widely used in obstetrics and gynecology in recent decades [4–10]. Because it offers the advantages of superb soft-tissue resolution, no radiation, and multiplanar imaging, MRI plays a key role in determining the etiology of any suspected malignancy in the female genital tract detected with ultrasound or computed tomography [11–13]. Diffusion-weighted imaging (DWI) is a functional imaging technique currently available in the majority of clinical MRI units. By displaying water molecule mobility (Brownian motion), DWI permits the quantitative evaluation of tumor tissues and has shown promising results in distinguishing malignancies from benign gynecological lesions in several recently reported studies [6,12,14–17]. ☆ We declare that we have no conflicts of interest. ⁎ Corresponding author. No. 419 Fang xie Road, Shanghai, 200011, P.R. China. Tel.: + 86- 21 63770161; fax: + 86 21 63770768. E-mail address: [email protected] (H. Zhang).
However, research that uses MRI or DWI to image USs is still limited. Therefore, the purpose of the study was twofold: first, to clarify the MRI and DWI-MRI findings for the four subtypes of US in a relatively large sample and to compare them with the pathological findings at our single institution; and second, to determine whether DWI-MRI is useful in discriminating USs from other benign uterine conditions. 2. Materials and methods 2.1. Study subjects From June 2010 to December 2012, 437 consecutive patients with clinically suspected gynecological disease underwent prospective MRI examinations before pelvic or laparoscopic surgery at our institution. Twenty-two patients (19–73 years of age; average age, 52.7±12.3 years) with pathologically proven US were included in this study. To compare the apparent diffusion coefficients (ADCs) values of benign and malignant uterine lesions, we also selected a limited numbers of leiomyomas and adenomyomas for the comparative group. The criteria were as follows: first, to minimize the selection bias caused by multiple lesions in one patient, we chose only patients with a solitary lesion for either the leiomyoma or adenomyoma group; second, patients with any previous pelvic surgery or radiation history were arbitrarily excluded because the inherent structure of the uterus may have been altered. Thus, 21 leiomyomas (24–72 years of age; average age, 47.7±13.9years) and 20 adenomyomas (34–51 years of age; average age, 41.4±5.2years) were included. Our institutional review board approved the study, and the requirement for informed consent was waived for all participants.
http://dx.doi.org/10.1016/j.clinimag.2014.06.004 0899-7071/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
2
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Table 1 Details of parameters for MRI protocols Parameters
T1WI
T2WI
FS-T2WI
DWI
Contrast-enhanced MRI
Repetition/Echo time (msec) Echo trains per slice Sequence Bandwidth (Hz) Thickness (mm) Field of view (mm) Voxel size (mm) Flip angle (degrees)
550/10 44 TSE 178 4 350 (APa) 1.5×1.1×4.0 150
4000/83 19 TSE 260 4 350 (LRb) 1.1×1.1×4.0 144
8000/83 19 TSE 260 4 350 (LR, AP) 1.4×1.4×4.0 150
2800/81
4.89/2.38
EP2D 1250 5 300 (AP) 2.7×1.9×5.0
VIBEc 400 3 380 (LR, AP) 1.7×1.2×3.0 10
a b c
AP indicated that the scan plane was from anterior to posterior. LR indicated that the scan plane was from left to right. Volumetric Interpolated Breath-hold Examination.
2.2. Image acquisition MRI was performed using a 1.5-T MR system (Magnetom Avanto, Siemens, Erlangen, Germany) with a phased-array coil. The routine MRI protocols used for the assessment of pelvic masses included the axial turbo spin-echo (TSE) T1-weighted imaging (T1WI), sagittal TSE T2-weighted imaging (T2WI), and axial/sagittal TSE fat-suppressed T2WI (FS T2WI). For axial images, the transverse plane was perpendicular to the long axis of uterine body; for sagittal images, the longitudinal plane was parallel to the main body of uterus. DWI using an echo-planar imaging two-dimensional (EP2D) sequence performed in the axial plane with parallel acquisition technique by using b value=0, 100, and 800 s/mm 2. Contrast-enhanced pelvic imaging was acquired at multiple phases of contrast medium enhancement in both sagittal and axial planes. The detailed MRI acquisition parameters were listed in Table 1. 2.3. Image analysis The location, size (largest dimension in two orthogonal planes), shape (round, oval, and lobulated), and margin (clear or ambiguous); visibility of endometrium and hemorrhage component (high signal on T1WI) within the lesion; and presence of pelvic-free fluid and lymph nodes were noted. Further, the homogeneity of T1WI/T2WI signals, accompanying lesions and extrauterine extension were separately evaluated and recorded. On T1WI, hypo-, iso-, and hyperintensity were similar for the pelvic fluid, pelvic wall muscle, and fat signal; on T2WI, hypo-, iso-, and hyperintensity were similar for the pelvic bone, pelvic wall muscle, and fat signal; on b=800 mm−2/s DWI images, the low, intermediate, and high-signal intensities were similar for the pelvic bone, myometrium, and endometrium. After the intravenous injection of the contrast medium, the degree of lesion enhancement was graded as follows: 1, weak enhancement (less than the myometrium); 2, mild enhancement (equal to the myometrium); 3, avid enhancement (superior to the myometrium). ADCs were measured manually on postprocessing workstation (Leonardo, Siemens, Germany) by one reviewer (H.Z.). Two observers (G. F. Z. and H.Z., with 10 and 6 years of experience in gynecological imaging, respectively), who were blinded to the histological results independently, analyzed MRI datasets of each participant. At the end of the study, two observers were also required to give the tumor stage according to new FIGO system for US [18]. For interobserver discrepancies in the evaluation of uterine lesions, consensus was achieved. 2.4. Statistical analyses Continuous variables were expressed as the means±standard deviation (S.D.) and compared with the unpaired t test if normally distributed or the Mann–Whitney test if not normally distributed. A nonparametric test (Mann–Whitney) was used to test other nonparametric variables within each group. Kappa statistics were used to
Table 2 The details of baseline and MRI characteristics of four subtypes of the histologically proven 22 USs Baseline characteristics
Age (years) Menstruation Premenopause Perimenopausal Menopause Symptomsa Pelvic mass Virginal bleeding (discharge) Menstruation disorder MRI characteristicsb Maximum diameter (mm) Component Solid Cyst Mixed Hemorrhage Present/Absent Margin Clear/Unclear Extrauterine extension Yes/No Accompanying lesion Present/Absent Normal endometrium Present/Absent T1WI signal character Low Intermediate High Mixed Homogeneous/ Inhomogeneous T2WI signal character Low Intermediate High Mixed Homogeneous/ Inhomogeneous DWI signal character Low Intermediate High Mixed ADC value (10−3/mm2/s)c Enhancement Avid enhancement Moderate enhancement Weak enhancement
Pathology group
Total
LS
ESS+AS
CS
53±10.6
51±16.8
55±6.7
3 2 2
2 1 6
2 2 2
7 5 10
4 3
1 7
5
5 15
1
1
2
55.6±49.7
49.4±31.9
36.3±19.5
6
6
5
1
2
5/2
2/6
1/4
8/12
3/4
6/2
3/2
12/8
4/3
0/8
0/5
4/16
4/3
1/8
1/5
6/16
3/4
3/5
3/2
9/11
3 1
8
4
3 2/5
5/3
1 4/1
3 13 3 4 11/9
6 0/7
4 2 2 4/4
1 2 2 1/4
6 4 10 5/15
1 4 2 0.93±0.30
1 5 2 1.19±0.33
5
2 14 4
6
3 5
P value
.034
1
17 3
.021
1.05±0.22 .05
1
5
9 10 1
a If both symptoms were observed, then the virginal bleeding (discharge) was seemed as the primary symptom. b One CS and one AS could not be evaluated at MRI for the undetected lesion. c ADC values could not been calculated in one case in AS and two cases in CS group because the small-sized (b10 mm) lesions were not detected on DWI-MRI images.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
3
Fig. 1. A 53-year-old female patient with histologically proven LS (Stage Ib). (A) Axial T1WI revealed an oval soft mass occupying the left part of cavity, in which patchy high signals (arrowhead) were displayed, indicating the hemorrhage components. (B) Axial FS-T2WI revealed the solid tumor exhibited inhomogeneous signals. On DWI-MRI (b=800s/mm2) (C) the lesion appeared as the inhomogeneous high signal with foci of low signal and on ADC map (D), it displayed the homogenous intermediate signal. On early (E) and delay (F) contrast-enhanced images, the solid parts of the tumor displayed the obvious enhancement, while necrosis components did not enhance. Enhancement type of tumor (red line) was well demonstrated compared with normal myometrium (yellow line) and muscle tissue (green line) (G, H).
calculate the interobserver reliability as percentage of agreement between the observer evaluations before the consensus reading was made. The area under the receiver operating characteristic curve (AUC) was calculated for ADCs to discriminate US from leiomyoma. SPSS (version 13.0, SPSS, Chicago, USA) was used to perform statistical analyses. P values b .05 were considered statistically significant.
3. Results 3.1. Baseline characteristics The histopathological specimens revealed 22 primary USs, including seven LSs (39–73 years of age; average age, 53.0±10.6 years),
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
4
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Fig. 2. A 28-year-old female patient with histologically proven ESS (Stage IIIc). (A) Axial T1WI showed a homogeneous solid mass with isointensity, in the right uterine body. (B) On sagittal T2WI, the tumor occupied the whole anterior wall of the uterus, and the inhomogeneous signal and normal endometrium were displaced posteriorly (arrowheads). On DWIMRI (b=800 s/mm2) (C), the lesion appeared as a solid mass with inhomogeneous high and intermediate signal on the ADC map (D). The tumor displayed inhomogeneous enhancement on arterial-phase contrast-enhanced images (E), and later, the signal intensity was similar to that of the myometrium on venous-phase contrast-enhanced images (F). Note that the lymphoadenopathy near bilateral parailiac arteries (arrow) was clearly evident, which was confirmed in subsequently evaluated histopathological specimens. The tumor was clearly enhanced (red line) compared with the normal myometrium (green line) and muscle tissue (yellow line) (G, H).
three ESSs (19–66 years of age; average age, 37.7±24.9 years), six ASs (49–65 years of age; average age, 58.0±6.0 years), and six CSs (43–62 years of age; average age, 54.7±6.7 years). Three tumors were in Stage I, one in Stage II, and three in Stage III in the LS group; one tumor
was in Stage I, one in Stage II, and one in Stage III in the ESS group; six tumors were in Stage I in the AS group; and five tumors were in Stage I and one in Stage III in the CS group. The details of baseline characteristics for each subtype of US are summarized in Table 2.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
5
Fig. 3. A 19-year-old female patient with histologically proven ESS (Stage Ic). (A) Axial T1WI outlined a homogeneous solid mass (arrowheads) with isointensity, in the right uterine body. Note the high signals indicative of fluid in the cavity (*). (B) On T2WI, the well-defined tumor margin was located in the myometrium with isohyperintensity. On DWIMRI (b=800 s/mm2) (C) and the ADC map (D), the lesion was well displayed (arrowheads) and appeared as a homogeneous intermediate signal. The tumor showed homogeneously avid enhancement on contrast-enhanced images (E), and on a time–signal intensity curve (F), the degree of enhancement of the lesion (red line) was greater than that of the normal myometrium (yellow line).
3.2. MRI characteristics Two lesions (one AS and one CS) could not be detected on MRI by either observer because the lesions were too small. Eleven USs (11/20, 55%; Fig. 1) were confined to the uterine cavity, four (4/20, 20%) to the
Table 3 The details of locations of four pathological types in the histologically proven 22 USs on MRI Pathology
Cavitya
LocationCervical canal
Myometrium
Total
LS ESS ASb CSb Total
3
1 1 1 1 4
3 2
7 3 5 5 20
a
4 4 11c
5
If the lesion involved both cavity and cervical canal and the main body is located in the cavity, then the lesion was regarded in the cavity. b There was one case in both AS and CS group wherein the lesion was too small and could not be detected on MRI. c There was a significant difference in the location of US occurrence between cavity and myometrium (P=.032).
cervical canal, and five (5/20, 25%) to the myometrium. Two ESSs were located in the anterior part of the uterine body (Figs. 2, 3, Table 3). The uterine cavity was more likely to be involved than the other two sites (P= .032). The maximum diameter of the lesions in the LS group was largest (55.6±49.7 mm) among the four subtypes. On MRI, most of the USs (17/20, 85%) appeared as a solitary, solid tumor. In this study, 12 lesions were clearly demarcated, but in another eight patients, the margin was ambiguous. On T1WI, most patients (13/20, 65%) showed intermediate signals. On T2WI, the tumor signals appeared more diverse, and relatively, the LS lesions (6/22, 27.3%) showed more mixed signals than the other groups. In terms of T1WI/T2WI signal heterogeneity, the T1WI signals of the lesions in each group had no specific character. All tumors (7/7, 100%) in the LS group had an inhomogeneous signal on T2WI, which was clearly different from those of the ESS/AS group (P=.021) and CS group (P=.003). On MRI, normal endometrium could be seen in 11 patients. Extrauterine extension was only observed in the LS group (4/20, 20%). Hemorrhage components were usually observed in the LS group (5/7, 71.4%), which appeared as patchy high signals on T1WI images. However, hemorrhage signals were less well detected in the ESS, AS, and CS groups.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
6
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
Fig. 4. A 59-year-old patient with histologically proven CS (Stage Ia). An oval mass centrally located in the uterine cavity (*) with homogeneous isointensity on axial T1WI (A) and T2WI (B) was vaguely displayed. On DWI-MRI (b=800 s/mm2) (E), the lesion displayed homogeneous high signal and intermediate signal on the ADC map (F). On early (C) and delayed (D) enhanced images, the tumor displayed moderate enhancement, whereas the subendometrial enhancement line (arrowheads) was preserved. The enhancement of the tumor (red line) was similar to that of the myometrium, although the delay enhancement was observed compared with that of the myometrium (G, H).
On contrast-enhanced MRI images, most tumors (6/7, 85.7%) in the LS group showed rapid and avid enhancement. In the ESS/AS group, most tumors (5/8, 63%) appeared mild enhancement, and in the CS group, all lesions (5/5, 100%) showed moderate and delayed enhancement (Fig. 4). There was a little statistically significant difference in the level of enhancement between any groups (P=.05).
3.3. DWI-MRI characteristics On DWI-MRI images, 14 tumors (14/20, 70%) showed a homogeneously high signal. The mean ADC value was 0.93±0.30 in the LS group, 1.19±0.33 in the ESS/AS group, and 1.05±0.22 in the CS group, with a little overlap (Fig. 5). A significant difference in ADCs was only observed
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
7
Fig. 5. Box-and-whisker plots of the calculated ADC values (10−3/mm2/s) for the US subtypes. There were no significant differences in mean ADC value of the US subtypes. A significant difference in ADCs was only observed between LS (0.93±0.30) and leiomyoma (1.34±0.42) (P=.018).
between LS and leiomyoma (P=.018). When the leiomyomas and adenomyomas were compared, the mean ADC value for US (1.05±0.30) was lower than that for leiomyoma (1.34±0.42) and adenomyoma (1.12±0.14) (Fig. 6). There was a significant difference between the mean ADC value for US and fibroids (P=.019). If the cutoff value for ADC was set at 1.06×10−3/mm2/s, its sensitivity and specificity were 81% and 62%, respectively. The AUC was 0.72 (95% confidence interval, 0.556–0.883) for ADC in discriminating LS from leiomyoma.
were confirmed by the histopathological findings. When compared with the final histological staging, both two reviewers did correct tumor staging based on MRI images (Figs. 1–2, 3–4). There was excellent agreement between the two observers in both MRI findings (kappa=0.90) and MRI tumor staging (kappa=0.93).
3.4. Interobserver agreement and radiological–pathological correlations
Until now, most of the literature related to the MRI features of US have been published as case reports [9,15,19,20]. Based on this limited literature, several reviews of US imaging are especially noted [1,13,21].
Tumor extension, necrosis, hemorrhage, and lymphoadenopathy were clearly observed on both pre- and postcontrast MRI images and
4. Discussion
Fig. 6. Box-and-whisker plots of the calculated ADC values (10−3/mm2/s) of three groups. The mean ADC value for leiomyoma (1.34±0.42) was higher than that for US (1.05±0.30), with a little overlap. The mean ADC value for US and leiomyoma differed significantly (P=.019).
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
8
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx
LS is one of the smooth muscle tumors and comprises almost half of all USs [22]. It is sometimes difficult to differentiate histologically or radiologically between LS and leimyoma in which the process of malignant transformation has occurred [21,23]. It has been suggested that the presence of high-signal intensity on T2WI/T1WI and irregular tumor margins are indicators of LS that distinguish it from leiomyoma [24]. Our findings support this view because necrosis or hemorrhage components were always observed on the T1WI (5/7) and T2WI (7/7) images of the rapidly growing mass (Fig. 1), which displayed highsignal intensity on T1WI and an inhomogeneous signal on T2WI. In the present study, most LS tumors (6/7) showed marked enhancement on contrast-enhanced images [1]. ESS is an indolent neoplasm with a favorable prognosis involving long-term survival but with a propensity for late recurrence and metastasis [13]. In our study, one tumor was located in the vestigial part of the cervix after a hysterectomy, and two tumors were located in the myometrium. Three ESSs, in patients in whom the normal endometrium was well-preserved, invariably appeared as solid masses with homogeneous signals and avid enhancement on MRI, which mimicked LS or adenomyoma. Similar findings were also described in another study [19]. In this context, the MRI features are nonspecific for the diagnosis of ESS, and a histological diagnosis is mandatory to confirm the etiology of the tumor. AS of the uterus is a rare low-grade mixed epithelial mesenchymal tumor containing benign glandular epithelial elements and malignant mesenchymal elements and constitutes only 8% of all USs [25]. The imaging features of AS are similar to those of CS; it is typically seen as a large polyploidal mass within the endometrial cavity, which may protrude into the endocervical canal. In one study [9], the authors described foci of hyperintense signal within the mass on T2WI, reflecting glandular epithelial components, which may be a characteristic finding. In the newly revised histopathological classification of US, CS is reclassified as a subtype of endometrial carcinoma [18]. Because CS behaves more aggressively than ordinary endometrial carcinomas, it is still included in most studies of US, as in this study. Upon imaging, CSs are predominantly endometrial-based heterogeneous masses causing distension of the endometrial cavity and associated with varying degrees of myometrial invasion. The craniocaudal dimension of the CS is greater than that of endometrial carcinomas [26]. They also usually show more marked contrast enhancement than endometrial carcinomas, which are usually hypovascular [27]. In our study, all cases of AS and CS displayed heterogeneous enhancement on MRI, which was similar to the appearance of the myometrium. The degree of enhancement of the ESS/AS and CS lesions was less than that of the LS lesions. In this study, tumor staging by MRI correlated perfectly with the histopathological staging. Interobserver agreement was excellent, also indicating that MRI is a reliable imaging modality for accurate preoperative staging. As the parametric index in DWI imaging, ADC may provide a quantitative measurement for any suspected tumor tissues [16,28] with encouraging results in distinguishing malignancies from benign diseases of the female genital tract [12,17,29,30]. Tamai et al. reported in their study that for b= 500, 1000 s/mm 2 DWI images, the mean ADC value (10 −3mm 2/s) of LS was 1.17±0.15, which was lower than that of the normal myometrium (1.62±0.11) or degenerated leiomyomas (1.70±0.11), without any overlap [6]. In another study, the authors showed that on b= 1000 s/mm 2 DWI images, endometrial carcinomas, and CSs had lower ADCs (less than 1.15×10 −3mm 2/s) than endometrial polyps or submucosal leiomyomas [14]. In a recent study, the authors also reported that ADCs could be used as a useful reference tool with which to distinguish benign uterine lesions from malignancies [12]. Our results are consistent with the results reported above. In the present study, the mean ADC value for b= 800 s/mm 2 DWI images of US was lower than those for leiomyoma or adenomyoma, whereas a significant difference in the mean ADC value for LS and leiomyoma was also observed. However, there was no significant difference observed in the ADCs for US and adenomyoma.
To the best of our knowledge, this result has not been described in previously published studies. There were several limitations of our study. First, although the number of USs studied was the largest sample so far reported from one institution, the numbers of histopathological subtypes of US (especially ESS) were still limited and unevenly distributed. To investigate prospective MRI findings in larger samples will be needed to clarify the true differences between the subtypes. Second, the regions of interest, in addition to the ADCs, were manually drawn and individually calculated on a case-by-case basis, and this lack of standardization may have influenced the final results. In summary, MRI characteristics show no specific sign between any subtype of US, except for the heterogeneity of T2WI signal. LS often shows more intense enhancement than the other subtypes, and ADCs could be useful in discriminating USs from uterine fibroids. MRI can be used as a reliable imaging modality for staging preoperative US.
References [1] Sohaib SA, Verma H, Attygalle AD, Ind Thomas EJ. Imaging of uterine malignancies. Semin Ultrasound CT MR 2010;31:377–87. [2] D'Angelo E, Prat J. Uterine sarcomas: a review. Gynecol Oncol 2010;116:131–9. [3] Uterine sarcoma. In: Schorge JO, Schaffer JI, Halvorson LM, et al, editors. Williams gynecology. Columbus, OH: The Mc-Graw-Hill Companies; 2008. p. 706–16. [4] Sala EWS, Senior E, Lomas D. Mri of malignant neoplasms of the uterine corpus and cervix. Am J Roentgenol 2007;188:1577–87. [5] Teo SYBK, Peters HE, Mortele KJ. Primary malignant mixed müllerian tumor of the uterus: findings on sonography, ct, and gadolinium-enhanced mri. Am J Roentgenol 2008;191:278–83. [6] Tamai K, Koyama T, Saga T, Morisawa N, Fujimoto K, Mikami Y, Togashi K. The utility of diffusion-weighted mr imaging for differentiating uterine sarcomas from benign leiomyomas. Eur Radiol 2008;18:723–30. [7] Tanaka YO, Tsunoda H, Minami R, Yoshikawa H, Minami M. Carcinosarcoma of the uterus: Mr findings. J Magn Reson Imaging 2008;28:434–9. [8] Namimoto T, Yamashita Y, Awai K, Nakaura T, Yanaga Y, Hirai T, Saito T, Katabuchi H. Combined use of t2-weighted and diffusion-weighted 3-t mr imaging for differentiating uterine sarcomas from benign leiomyomas. Eur Radiol 2009;19:2756–64. [9] Takeuchi M, Matsuzaki K, Yoshida S, Kudo E, Bando Y, Hasebe H, Kamada M, Nishitani H. Adenosarcoma of the uterus: magnetic resonance imaging characteristics. Clin Imaging 2009;33:244–7. [10] Cornfeld D, Israel G, Martel M, Weinreb J, Schwartz P, McCarthy S. Mri appearance of mesenchymal tumors of the uterus. Eur J Radiol 2010;74:241–9. [11] Stamatopoulos CP, Mikos T, Grimbizis GF, Dimitriadis AS, Efstratiou I, Stamatopoulos P, Tarlatzis BC. Value of magnetic resonance imaging in diagnosis of adenomyosis and myomas of the uterus. J Minim Invasive Gynecol 2012;19:620–6. [12] Thomassin-Naggara I, Dechoux S, Bonneau C, Morel A, Rouzier R, Carette MF, Daraï E, Bazot M. How to differentiate benign from malignant myometrial tumours using mr imaging. Eur Radiol 2013;23:2306–14. [13] Tirumani S, Ojili V, Shanbhogue A, Fasih N, Ryan J, Reinhold C. Current concepts in the imaging of uterine sarcoma. Abdom Imaging 2013;38:397–411. [14] Fujii S, Matsusue E, Kigawa J, Sato S, Kanasaki Y, Nakanishi J, Sugihara S, Kaminou T, Terakawa N, Ogawa T. Diagnostic accuracy of the apparent diffusion coefficient in differentiating benign from malignant uterine endometrial cavity lesions: Initial results. Eur Radiol 2008;18:384–9. [15] Fujii S, Kaneda S, Tsukamoto K, Kakite S, Kanasaki Y, Matsusue E, Kaminou T, Ogawa T. Diffusion-weighted imaging of uterine endometrial stromal sarcoma: a report of 2 cases. J Comput Assist Tomogr 2010;34:377–9 [310.1097/ RCT.1090b1013e3181cfc1676]. [16] Sala E, Rockall A, Rangarajan D, Kubik-Huch RA. The role of dynamic contrastenhanced and diffusion weighted magnetic resonance imaging in the female pelvis. Eur J Radiol 2010;76:367–85. [17] Zhang H, Zhang G-F, He Z-Y, Li Z-Y, Zhu M, Zhang G-X. Evaluation of primary adnexal masses by 3 t mri: categorization with conventional mr imaging and diffusion-weighted imaging. J Ovarian Res 2012;5:33. [18] Prat J. Figo staging for uterine sarcomas. Int J Gynecol Obstet 2009;104:177–8. [19] Furukawa R, Akahane M, Yamada H, Kiryu S, Sato J, Komatsu S, Inoh S, Yoshioka N, Maeda E, Takazawa Y, Ohtomo K. Endometrial stromal sarcoma located in the myometrium with a low-intensity rim on t2-weighted images: report of three cases and literature review. J Magn Reson Imaging 2010;31:975–9. [20] Kato H, Kanematsu M, Furui T, Imai A, Hirose Y, Kondo H, Goshima S, Tsuge Y. Carcinosarcoma of the uterus: radiologic–pathologic correlations with magnetic resonance imaging including diffusion-weighted imaging. Magn Reson Imaging 2008;26:1446–50. [21] Qiu L-L, Yu R-S, Chen Y, Zhang Q. Sarcomas of abdominal organs: computed tomography and magnetic resonance imaging findings. Semin Ultrasound CT MR 2011;32:405–21. [22] Chiang S, Oliva E. Recent developments in uterine mesenchymal neoplasms. Histopathology 2013;62:124–37.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
G-F. Zhang et al. / Clinical Imaging xxx (2014) xxx–xxx [23] Lim D, Alvarez T, Nucci MR, Gilks B, Longacre T, Soslow RA, Oliva E. Interobserver variability in the interpretation of tumor cell necrosis in uterine leiomyosarcoma. Am J Surg Pathol 2013;37:650–8 [610.1097/PAS.1090b1013e3182851162]. [24] Tanaka YO, Nishida M, Tsunoda H, Okamoto Y, Yoshikawa H. Smooth muscle tumors of uncertain malignant potential and leiomyosarcomas of the uterus: Mr findings. J Magn Reson Imaging 2004;20:998–1007. [25] Chourmouzi DBG, Zarampoukas T, Drevelengas A. Sonography and mri of tamoxifen-associated müllerian adenosarcoma of the uterus. Am J Roentgenol 2003;181:1673–5. [26] Genever AV, Abdi S. Can mri predict the diagnosis of endometrial carcinosarcoma? Clin Radiol 2011;66:621–4.
9
[27] Bharwani NNA, Tunariu N, Babar S, Sahdev A, Rockall AG, Reznek RH. Mri appearances of uterine malignant mixed müllerian tumors. Am J Roentgenol 2010;195:1268–75. [28] Punwani S. Diffusion weighted imaging of female pelvic cancers: concepts and clinical applications. Eur J Radiol 2011;78:21–9. [29] Zhang P, Cui Y, Li W, Ren G, Chu C, Wu X. Diagnostic accuracy of diffusionweighted imaging with conventional mr imaging for differentiating complex solid and cystic ovarian tumors at 1.5t. World J Surg Oncol 2012;10:237. [30] Li W, Chu C, Cui Y, Zhang P, Zhu M. Diffusion-weighted mri: a useful technique to discriminate benign versus malignant ovarian surface epithelial tumors with solid and cystic components. Abdom Imaging 2012;37:897–903.
Please cite this article as: Zhang G-F, et al, Magnetic resonance and diffusion-weighted imaging in categorization of uterine sarcomas: correlation with pathological findings, Clin Imaging (2014), http://dx.doi.org/10.1016/j.clinimag.2014.06.004
