| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Division of Abdominal Imaging and Intervention, Department of Radiology,
Brigham and Women's Hospital and Harvard Medical School, 75 Francis St.,
Boston, MA 02115.
2 Division of Ultrasound, Department of Radiology, Brigham and Women's Hospital
and Harvard Medical School, Boston, MA.
Received October 10, 2007;
accepted after revision January 18, 2008.
Address correspondence to K. J. Mortele
(kmortele{at}partners.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Uterine MMMT most commonly presents as an intracavitary mass with coexistent dilatation of the endometrial canal. Tumors tend to appear hyperechoic on sonography, heterogeneously hypodense and ill defined on contrast-enhanced CT, and heterogeneously hyperintense on T2-weighted MR images with signal abnormalities indicating subacute hemorrhage on T1-weighted MR images. Myometrial invasion is common and has a predilection for the uterine fundus.
Keywords: CT • mixed müllerian tumor • MRI • sonography • uterus
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Prior reports defining the imaging appearances of uterine MMMT are sparse. To the best of our knowledge, only two case reports illustrating the sonographic findings of uterine MMMT are available in the literature [4, 5]. A study on the CT features of MMMT reported 27 patients with locally advanced and metastatic disease; 11 patients were evaluated before hysterectomy [6]. A series describing the MRI findings of primary and recurrent uterine sarcomas included four patients with primary MMMT [7]; MRI findings of uterine MMMT were also previously touched on in a study of seven patients using unenhanced MRI only [8] and in a pictorial essay illustrating all subtypes of uterine sarcomas [9].
Therefore, the objective of this study is to present the radiologic findings of primary uterine MMMT using various imaging techniques, including sonography, contrast-enhanced CT, and gadolinium-enhanced MRI. To our knowledge, our study is the largest series of primary uterine MMMTs with a comprehensive analysis of the radiologic findings.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All patients with available preoperative imaging (sonography, contrast-enhanced CT, gadolinium-enhanced MRI) of pathologically proven primary uterine MMMT from January 2000 to May 2006 were included in our study. Patients were identified through the surgical pathology and radiology information electronic databases in our institution. We retrospectively reviewed the radiologic images of 15 women who were eligible for the study. Of them, 14 patients underwent total abdominal hyster ectomy and bilateral salpingo-oophorectomy, omental and pelvic lymph node biopsy, and peritoneal washing. The other patient was diagnosed with uterine MMMT on the basis of endometrial curettage; she did not undergo surgery because widespread thoracic disease was present at pre sentation. A total of 17 imaging studies, in cluding five sono graphic studies, 10 contrast-enhanced CT studies, and two MR studies, were performed in the 15 women. One patient underwent a combination of sonography, CT, and MRI; the other 14 patients were evaluated with a single imaging technique before surgery.
The mean age of our study population was 60.8 years (age range, 19–82 years). Thirteen women were of white descent and two were of African American descent. Eleven patients presented with postmenopausal bleeding; our youngest patient presented with abnormal vaginal bleeding. Two patients presented with abdominal pain, and one patient suffered from vaginal discharge. Four (26.6%) patients received previous tamoxifen therapy, and two (13.3%) patients underwent previous exogenous hormonal therapy. None of our patients underwent previous pelvic irradiation.
Imaging Techniques
Five patients underwent both transabdominal and transvaginal pelvic
sonography. A sono grapher and an attending radiologist performed the
sonography examinations in accordance with our department protocol using an
Acuson Sequoia C512 sonography unit (Siemens Medical Solutions).
Contrast-enhanced CT was performed in 10 patients using a 4- or 16-MDCT scanner (Somatom VolumeZoom or Sensation, Siemens Medical Solutions). All patients received 100 mL of iopromide IV (Ultravist 300, GE Healthcare) and 900 mL of oral contrast material (Readi-Cat barium sulfate suspension, 2.1% w/v, E-Z-EM). Contrast-enhanced CT was performed 60–70 seconds after IV contrast administration. The CT slice thickness ranged from 3 to 5 mm. In one patient, multiplanar reformatted images were available for analysis.
MRI examinations were performed in two patients using a 1.5-T magnet (EchoSpeed, GE Healthcare). The MRI protocol included T2-weighted fast spin-echo images in axial, coronal, and sagittal planes (TR range/TE range, 3,417–6,067/88–126.5), fat-suppressed T1-weighted spin-echo images in axial, coronal, and sagittal planes (310–800/4.2–20), and fat-suppressed spoiled gradient-recalled echo (SPGR) T1-weighted images (260–450/1.6–6.4; field of view, 20 cm; matrix, 256 x 128; flip angle, 75°; number of excitations, 2; and approximate scanning time, 1 minute 30 seconds) after IV administration of 2 mmol/kg of gadopentetate dimeglumine (Magnevist, Bayer HealthCare). After gadolinium administration, images were obtained in the axial, coronal, and sagittal planes, respectively. No dynamic imaging was performed. The slice thickness used for all sequences varied between 4 and 5 mm.
Image Analysis
Two radiologists evaluated all images retrospectively in consensus. For
sonography, they evaluated images for the presence of dilatation of the
endometrial cavity; the maximal dimension of the endometrial cavity in the
sagittal plane if dilatation was present; the presence of a mass within the
endometrial cavity; the size (largest dimension) and echogenicity (defined as
echo genicity of > 50% of the mass compared with the myometrium) of the
mass, if present; and the presence and degree (< 50%, superficial; 
50%, deep) of myometrial invasion (defined as loss of the normal
endometrial–myometrial interface) and its location. When possible, tumor
staging, using International Federation of Gynecology and Obstetrics (FIGO)
uterine cancer staging, was performed
[10].
For CT, the radiologists evaluated the scans for dilatation of the
endometrial cavity and for the presence of a mass within the endometrial
cavity. When dilatation was present, the maximal dimensions of the endometrial
cavity were measured; this was performed in the transverse and anteroposterior
axes if the uterus was seen in an axial plane and in the transverse axis only
if the uterus was seen in a coronal plane. When a mass was present, the
margins (defined as > 75% of the circumference of the mass), the size
(largest dimension), appearance (homogeneous or hetero geneous), and
attenuation (region of interest [ROI] measurement) were assessed. Attenuation
of the mass was compared with normal myometrium and was designated as hypo-,
iso-, or hyperdense (on the basis of attenuation > 50% of the mass compared
with myometrium at the level of greatest endo metrial cavity dilatation).
Attenuation values of the uterine mass as well as normal myometrium were
measured and expressed as a ratio. CT images were also reviewed for the
presence and depth (< 50%, superficial; 50%, deep) of myometrial
invasion (defined as loss of the normal enhancing
endometrial–subendometrial layer) and its location. When possible, tumor
staging, using FIGO uterine cancer staging, was performed
[10].
Both MRI series were evaluated for dilatation of the endometrial cavity;
the presence of a mass within the endometrial cavity; the size (largest
dimension) of the mass, if present; the signal characteristics of the mass on
T1-weighted and T2-weighted sequences (compared with normal myometrium); the
amount (mild, moderate, or avid) and type (homogeneous or heterogeneous) of
enhancement; and the presence and depth of myo metrial invasion (< 50%,
superficial; 50%, deep) and its location. Maximal dilatation of the
endometrial cavity was measured in the sagittal (anteroposterior dimension)
and axial (transverse dimension) planes. Myometrial invasion was defined as
disruption or discontinuation of the junctional zone or visualization of tumor
extension into the myometrium. When possible, tumor staging, using FIGO
uterine cancer staging, was performed
[10].
Radiologic–Pathologic Surgical Correlation
Results from histopathologic analysis of the surgical specimens (presence
and depth [< 50%, superficial; 50%, deep] of myometrial invasion;
presence and location of metastatic disease) were reviewed and compared with
radiologic findings in each patient. In addition, disease stage, as defined by
imaging, was compared with the disease stage as derived from surgical
exploration.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
MMMT mostly appeared heterogeneous (n = 9) except for the smallest mass, which appeared homogeneous. All masses appeared hypodense to normal myometrium (Figs. 2A, 2B and 3A, 3B, 3C, 3D, 3E, 3F, 3G). Of these, two had areas of hyper density that did not constitute 50% of the mass but affected the measurement of its attenuation; one showed a 0.8-cm calcified nodule and the other showed a tiny focus of calcification, both within the right lateral aspect of the endometrial cavity.
|
|
|
|
|
|
|
Eight of 10 patients were deemed to have myometrial involvement on CT scans (Figs. 2A, 2B and 3A, 3B, 3C, 3D, 3E, 3F, 3G). This involvement was present at the fundus (n = 4), fundus and left wall (n = 1), left wall (n = 2), and right wall (n = 1). Using CT, the local tumor extension in the 10 patients was staged as IA (n = 2), IB (n = 2), IC (n = 2), IIIA (n = 1), IIIC (n = 2), and IVB (n = 1). Metastatic pelvic and retroperitoneal lymphadenopathy was identified in three patients: omental carcinomatosis and right adnexal involvement, in one patient each.
MRI Findings
Dilatation of the endometrial cavity and an intracavitary mass were present
in both patients. The dilation of the endometrial cavity measured 5.2 x
8 cm and 7.2 x 8.1 cm, respectively; the masses themselves measured 10
x 8.6 cm and 18 x 9 cm, respectively. Both intracavitary masses
were heterogeneously hyperintense on T2-weighted images and slightly
hypointense with scattered areas of high signal intensity on T1-weighted
images (Fig. 3A,
3B,
3C,
3D,
3E,
3F,
3G). Both masses showed
heterogeneous but avid enhancement on gadolinium-en hanced images. Deep
myometrial invasion at the fundus and metastatic pelvic and retroperitoneal
lymphadenopathy were identified in both patients; therefore, both patients
were thought to have stage IIIC disease.
Radiologic–Pathologic Surgical Correlation
All five patients imaged with sonography underwent total abdominal
hysterectomy and bilateral salpingo-oophorectomy, omental biopsy, peritoneal
washing, and pelvic lymph node sampling. Compared with histo pathologic
analysis of the specimen, sonography was accurate in predicting deep
myometrial invasion in two patients, underestimated the presence of
superficial (n = 1) and deep (n = 1) myometrial invasion in
two patients, and overestimated the presence of deep myometrial invasion in
one patient who had superficial (20%) invasion on pathology. Surgical staging
in the five patients revealed stage IB (n = 2), IC (n = 1),
IIIC (n = 1), and IVB (n = 1). Using sonography, one patient
was staged correctly, three patients were understaged, and one patient was
overstaged.
Of 10 patients imaged with contrast-enhanced CT, nine underwent total abdominal hysterectomy and bilateral salpingo-oophorectomy, omental biopsy, peritoneal washing, and pelvic lymph node sampling. Compared with histopathologic analysis of the specimen, contrast-enhanced CT was accurate in predicting the extent of myometrial invasion in eight of nine patients and underestimated the presence of superficial (10%) myometrial invasion in one patient. Surgical staging in the nine patients revealed stage IA (n = 1), IB (n = 3), IC (n = 2), IIIA (n = 1), IIIC (n = 1), and IVB (n = 1). Using contrast-enhanced CT, eight patients were staged correctly, and one patient was understaged.
Both patients imaged with MRI underwent total abdominal hysterectomy and bilateral salpingo-oophorectomy, omental biopsy, peritoneal washing, and pelvic lymph node sampling. Compared with histopathologic analysis of the specimen, MRI was accurate in predicting deep myometrial invasion in both patients. Surgical staging in both patients revealed stage IIIC; using MRI, both patients were staged correctly. The areas of hemorrhage seen on T1-weighted images in both patients were confirmed pathologically.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Overall, in our retrospective study, 14 of 15 patients imaged with sonography, CT, or MRI showed abnormal dilation of the endometrial canal (n = 12) or the presence of an intracavitary mass (n = 14) or both. In one patient, the MMMT appeared as minimal diffuse thickening of the endometrium only; in another patient, sonography showed a large hetero geneous mass replacing the entire uterus, with no dilatation of the endometrial canal; and in one patient, a 9-mm intracavitary mass was identified on contrast-enhanced CT without associated dilation of the endometrial canal. These data confirm that dilatation of the endo metrial cavity should alert the radiologist to the possibility of an endometrial-based abnormality.
Two prior case reports have illustrated the sonographic appearance of uterine MMMT [4, 5]. Chourmouzi et al. [4] reported a 9-cm tamoxifen-induced MMMT that expanded the endometrial canal and appeared heterogeneous and hyperechoic, with numerous small cystic spaces; no myometrial invasion was present. Mastrantonio et al. [5] similarly reported a case of MMMT, which was of high acoustic impedance, richly and inhomogeneously echogenic, with anechoic areas. In our study, all tumors also appeared hyperechoic com pared with the myometrium, and coexistent expansion of the endometrial canal was seen in three of five cases. None of our cases, however, displayed the presence of small cystic spaces. Not surprisingly, sonography failed to accurately predict the surgical tumor stage in most of our patients. Indeed, compared with CT and MRI, sonography does not have the capability to accurately evaluate the retro peritoneum and deep pelvic lymph node chains. Therefore, although sonography is moderately accurate in predicting the presence of myometrial invasion (three of five patients), remote tumor staging warrants further imaging with CT or MRI.
Smith et al. [6] evaluated the contrast-enhanced CT features of primary uterine MMMT in 11 patients; uterine cavity dilatation was present in 73% of patients and irregular low-attenuation intracavitary masses in 64% of patients. Data on the presence, depth, or location of myometrial invasion and size of the lesions were not provided. In our study, most MMMTs evaluated with contrast-enhanced CT also pre sented as heterogeneous, ill-defined, hypodense masses with coexistent dilation of the endo metrial canal. Myometrial invasion was present in 80% and predominantly involved the uterine fundus (62.5%); myometrial invasion was accurately detected with contrast-enhanced CT, presumably because of the different attenuation of the tumor and the enhancing myometrium. In our study, contrast-enhanced CT showed pro mise for predicting the surgical tumor stage of MMMT (89%). This is likely because of the obser vations made by Smith et al. that, although MMMT may metastasize hemato genously, local and lymphatic spread and intraperitoneal seeding are the most common metastatic patterns [6] and may be detected by CT as in our series.
In a study by Sahdev et al. [7] on the MRI features of uterine sarcomas, two patterns of disease were seen including a large heterogeneous mass and an endometrial mass indistinguishable from endometrial carci noma [5]. Leiomyosarcomas tended to present with the former pattern, whereas four of five uterine MMMTs included in the afore mentioned study showed the latter pattern. The tumors displayed intermediate T1 and T2 signal intensity, in vaded the myometrium, and showed avid enhancement after gadolinium administration. The remaining case of MMMT presented as a large uterine mass. Another study of seven uterine MMMTs evaluated with unenhanced MRI showed that MMMTs commonly present as large endometrial-based masses that are heterogeneously hyperintense on T2-weighted sequences and homogeneously hypointense on T1-weighted images; areas of high-signal-intensity hemorrhage were seen in three (43%) patients [8]. Our study showed similar findings, with both tumors evaluated with MRI showing heterogeneously hyperintense signal on T2-weighted sequences and on T1-weighted im ages, slight hypointensity com pared with normal myometrium mixed with areas of high signal in tensity, caused by subacute hemorrhage. Although both tumors enhanced heterogene ously, areas of avid enhancement were identified in both.
In our series, myometrial invasion was assumed on imaging in 10 patients; pathologically, 12 (80%) patients had myometrial invasion. On imaging, the uterine fundus was the most common site for myometrial invasion, occurring in 60% of patients (n = 6) on sonography, CT, or MRI. To the best of our knowledge, this predilection for fundal myometrial invasion has not been described with other endometrial neoplasms.
Imaging characteristics of uterine MMMT are not unique, and the radiologic differential diagnoses of uterine MMMT include endometrial carcinoma; endometrial stromal sarcoma; and, uncommonly, leiomyosarcoma [8, 9]. A recent study suggested that dynamic gadolinium-enhanced MRI might be useful in differentiating MMMT from endometrial carcinoma. Ohguri et al. [16] studied four uterine MMMTs and 11 endometrial carcinomas and found that the sarcomatous components of uterine MMMT showed early and persistent avid enhancement compared with endometrial carcinoma. Endometrial stromal sarcomas, particularly the high-grade type, tend to present as a large endometrial mass with irregular margins, peripheral nodular lesions, and myometrial nodules on MRI [17]. Leiomyosarcomas are mesenchymal tumors arising from the myometrium, which may arise either from a known leiomyoma or de novo, and thus are more likely to present as a myometrial-based mass rather than an intracavitary endometrial mass [5].
A limitation of our study is the retrospective study design; a blinded review or prospective analysis may provide more information on the value of imaging techniques in the evaluation of patients with postmenopausal bleeding, specifically in the diagnosis of MMMT. Another limitation, due to the lack of multitechnique imaging in the same patient, is the inability to compare the accuracy of imaging techniques with regard to the detection of myometrial invasion and local tumor staging. Finally, although we reported the enhancement patterns of MMMT on CT compared with normal myometrium during the portal venous phase, a dynamic contrast-enhanced time–intensity analysis is required to accurately assess lesion enhancement. However, obtaining a time–density ana lysis would have required serial dynamic contrast-enhanced CT of the lesions, including com parison with time–density curves for normal uterine myometrium.
In conclusion, uterine MMMT most commonly presents on imaging as an intracavitary abnormality with associated expansion of the endometrial canal; uterine MMMT presenting as a heterogeneous mass replacing the uterus is an uncommon radiologic manifestation. Tumors tend to appear echogenic on sonography and heterogeneously hypodense on contrast-enhanced CT. On MRI, MMMT appears heterogeneously hyperintense on T2-weighted MR images with signal ab normalities indicating subacute hemorr hage on T1-weighted MR images. Myometrial invasion is common and has a predilection for the uterine fundus.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
