MR and CT findings of leiomyomatosis peritonealis disseminata with emphasis on assisted reproductive technology as a risk factor.

Leiomyomatosis peritonealis disseminata (LPD) is a rare benign smooth muscle tumour located in the peritoneal cavity. Increased oestrogen exposure appears to be an aetiological factor for LPD. We report two cases of LPD after leiomyomectomy and assisted reproductive technology pregnancy, which can cause a high serum concentration of oestrogen. CT and MR scanning demonstrate many intraperitoneal well-demarcated nodules of varying size that mimic widespread intraperitoneal malignancy.

Clear cell adenocarcinoma of the uterine cervix arising from a background of cervical endometriosis.

The radiological findings of cervical clear cell adenocarcinoma (CCA) have not been described previously. Here, we present MR findings of this neoplasm that included mixed solid and cystic components with eccentric solid components. These are similar to the MR features of ovarian CCA. Endometriosis was also noted in the uterine cervix. Coexistence of CCA and endometriosis at the cervix suggests that the pathogenesis may be similar to that of ovarian CCA.

Silent venous thromboembolism before treatment in endometrial cancer and the risk factors.

Venous thromboembolism (VTE) often occurs after surgery and can even occur before surgery in patients with gynaecological malignancies. We investigated the incidence of VTE before treatment of endometrial cancer and associated risk factors. Plasma D-dimer (DD) levels before initial treatment were examined in 171 consecutive patients with endometrial cancer. Venous ultrasound imaging (VUI) of the lower extremities was performed in patients with DD >or=1.5 microg ml(-1), as the negative predictive value of DD for VTE is extremely high. For patients with deep vein thrombosis (DVT), pulmonary scintigraphy was performed to ascertain the presence of pulmonary thromboembolism (PTE). Risk factors for VTE were analysed using univariate and multivariate analyses for 171 patients. Of these, 37 patients (21.6%) showed DD >or=1.5 microg ml(-1), 17 (9.9%) displayed DVT by VUI and 8 (4.7%) showed PTE on pulmonary scintigraphy. All patients with VTE were asymptomatic. Univariate analysis for various risk factors revealed older age, non-endometrioid histology and several variables of advanced disease as significantly associated with VTE before treatment. Obesity, smoking and diabetes mellitus were not risk factors. Multivariate analysis confirmed extrauterine spread and non-endometrioid histology as independently and significantly associated with risk of VTE. These data suggest that silent or subclinical VTE occurs before treatment in at least around 10% of patients with endometrial cancer. Risk factors for VTE before treatment might not be identical to those after starting treatment.

High incidence of silent venous thromboembolism before treatment in ovarian cancer.

Venous thromboembolism (VTE) such as deep-vein thrombosis (DVT) and pulmonary thromboembolism (PTE) often occurs after surgery and rarely occurs even before surgery in patients with ovarian cancer. It is well known that levels of plasma D-dimer (DD) before treatment in most ovarian cancer patients are increased. This study therefore examined whether increased levels of DD are associated with presence of VTE before treatment of ovarian cancer. Between November 2004 and March 2007, DD levels prior to initial treatment were measured in 72 consecutive patients with presumed epithelial ovarian cancer (final diagnosis: epithelial ovarian cancer, n=60; and epithelial ovarian borderline malignancy, n=12). Venous ultrasound imaging (VUI) of the lower extremity was conducted for all patients except for two patients in whom DVT was detected by pelvic computed tomography (CT). When DVT was found, pulmonary scintigraphy was subsequently performed to ascertain presence of PTE. D-dimer levels were above the cut-off value (0.5 microg ml(-1)) in 65 of 72 patients (90.2%). Venous ultrasound imaging or CT revealed DVT in 18 of 72 patients (25.0%) and pulmonary scintigraphy found PTE in 8 patients (11.1%). All patients with VTE were asymptomatic when VTE was found. D-dimer levels were associated with incidence of VTE (0-1.4 microg ml(-1); 0 of 26 (0%), 1.5-7.4 microg ml(-1); 9 of 30 (30%) and > or =7.5 microg ml(-1); 9 of 16 (56.3%), P for trend=0.0003). However, even if 1.5 microg ml(-1) was used as a cut-off value, this had low specificity and positive predictive value (47.2, 38.3%), though it had high sensitivity and negative predictive value (100, 100%). Therefore, ovarian cancer patients with DD level > or =1.5 microg ml(-1) should be examined using VUI to detect silent DVT. Patients with VTE underwent preventive managements including anticoagulant therapy before initial treatment, chemotherapy or surgery, and after surgery. There was no clinical onset of postoperative VTE in all 72 patients. Measurement of DD levels and subsequent ultrasonography revealed that silent or subclinical VTE frequently occurs before surgery in ovarian cancer. The usefulness of preoperative assessment of VTE needs further confirmation in randomised controlled trials.

Pelvic imaging: multicystic uterine cervical lesions. Can magnetic resonance imaging differentiate benignancy from malignancy?

PURPOSE: To evaluate whether magnetic resonance imaging (MRI) can be used to differentiate between malignant and benign multicystic uterine cervical lesions. MATERIAL AND METHODS: Twenty-two patients with cervical adenocarcinomas, including minimal deviation adenocarcinoma, and five patients with benign multicystic lesions were classified as a percentage of solid components on MRI. Cystic components were further classified by average cyst diameter and signal intensity of the cyst fluid on T1WI. RESULTS: All nine of the entirely solid lesions were malignant. In contrast, two of the entirely cystic lesions were benign. Ten of the ordinary adenocarcinomas had both solid and cystic components. However, three of the 16 solid and cystic lesions were benign. Lesions composed of cysts smaller than 5mm tended to be malignant; however, some lesions composed of larger cysts were also malignant. Three of 5 lesions with low-signal and 7 of 10 with intermediate-signal fluid were malignant. CONCLUSION: The malignancy potential was higher in the lesions with a higher percentage of solid components. However, determining whether multicystic lesions were benign or malignant based on the existence of solid components, the average cyst size, and the signal intensity of cyst fluid was impossible. Although a multicystic lesion with solid components in the deep cervical stroma had been reported as a MR finding of a minimal deviation adenocarcinoma, this does not appear to be pathognomonic.

A decidualized endometrial cyst in a pregnant woman: a case observed with a steady-state free precession imaging sequence.

Decidual changes of the ectopic endometrial stroma during pregnancy are well known among pathologists and obstetricians. However, they appear very similar to endometrial cysts with malignant transformation when imaged. Balanced fast field echo (BFFE) is a steady-state free precession imaging sequence and its contrast is decided by the T1/T2 ratio. The authors report a case of a decidualized endometrial cyst in which mural nodules were isointense with the nomotopic decidualized endometrium on T1- and T2-weighted images and BFFE. Isointensity with the nomotopic endometrium is an MR characteristic that can differentiate a decidualized endometrial cyst from malignant transformation. BFFE is a good alternative sequence during pregnancy because of its shorter acquisition time and lower radiofrequency absorption.

High temporal resolution dynamic contrast MRI in a high risk group for placenta accreta.

Antenatal diagnosis of placenta accreta with MR is not easy even now because T2-weighted images (T2WI) cannot differentiate chorionic villi from decidua basalis. We performed dynamic contrast MRI to study whether trophoblastic villi could be separately demonstrated from the decidua basalis, and whether the contrast resolution between the placenta and myometrium could improve compared to T2WI. Six pregnant women with prior cesarean section were examined at 34-38 gestational weeks. Sagittal T2-weighted images with fast spin echo sequences and dynamic contrast studies with fast field echo sequence every 10-14 s after contrast injection were performed. We analyzed the enhancing pattern of the placenta and compared the contrast between placenta and myometrium. We reviewed medical records to identify complications during the placental delivery and the complications of their newborns. In the early phase after contrast enhancement, multiple foci of the strong lobular enhancement were observed in all cases. Other parts of placenta were slowly but strongly enhanced following them. We speculated that the former corresponded to intervillous space and the latter decidua basalis. The contrast between placenta and myometrium tended to be distinct near the inner cervical os on both T2WI and dynamic contrast study. On the other hand, it was indistinct in the upper part of the uterine body on T2WI despite it was clearly demonstrated on dynamic contrast study. The placentae were delivered without any complication in all cases. Although two neonates showed fetal distress, none of the infant remained any sequelae at the time of the discharge. The other four were well although one of them complicated with meconium staining. As dynamic contrast MRI can differentiate chorionic villi and decidua basalis, and can provide excellent contrast between placenta and myometrium at anywhere within the uterus, it may be a promising technique for antepartum diagnosis of the placenta accreta.

Preliminary estimation of minimum target dose in intracavitary radiotherapy for cervical cancer.

PURPOSE: In intracavitary radiotherapy (ICRT) for cervical cancer, minimum target dose (Dmin) will pertain to local disease control more directly than will reference point A dose (D(A)). However, ICRT has been performed traditionally without specifying Dmin since the target volume was not identified. We have estimated Dmin retrospectively by identifying tumors using magnetic resonance (MR) images. MATERIALS AND METHODS: Pre- and posttreatment MR images of 31 patients treated with high-dose-rate ICRT were used. ICRT was performed once weekly at 6.0 Gy DA, and involved 2-5 insertions for each patient, 119 insertions in total. Dmin was calculated arbitrarily simply at the point A level using the tumor width (W(A)) to compare with D(A). W(A) at each insertion was estimated by regression analysis with pre- and posttreatment W(A). RESULTS: Dmin for each insertion varied from 3.0 to 46.0 Gy, a 16-fold difference. The ratio of total Dmin to total DA for each patient varied from 0.5 to 6.5. Intrapatient Dmin difference between the initial insertion and final insertion varied from 1.1 to 3.4. CONCLUSION: Preliminary estimation revealed that Dmin varies widely under generic dose prescription. Thorough Dmin specification will be realized when ICRT-applicator insertion is performed under MR imaging.

Intrahepatic portosystemic venous shunts: spectrum of CT findings in external and internal subtypes.

PURPOSE: The purpose of this work was to survey the spectrum of CT findings in intrahepatic portosystemic venous shunts (IPSVSs). METHOD: One thousand consecutive liver CT scans examined with an enhanced helical technique were retrospectively reviewed on workstation to find IPSVSs. RESULTS: IPSVSs were noted in 37 patients and divided into two subtypes according to the location of the communicating systemic vein, either outside (external type, n = 34) or within (n = 4) the liver (one patient had both internal and external types). All the external-type shunts were noted in cases with portal hypertension (n = 34). Most shunts (n = 30) ran through segment 4 and/or 3 and came out near the falciform ligament, but five shunts were noted in segment 2 communicating either with the systemic vein (n = 3) or probably with an enlarged left gastric vein (n = 2). Solitary portohepatic venous shunts (internal type) were noted in three cases with (n = 2) or without (n = 1) aneurysmal dilatation and many portohepatic venous shunts in another patient. CONCLUSION: IPSVSs of the external type were far common than those of the internal type, always associated with increased pressure of the portal system and in specific sites probably lacking the hepatic capsule.

MRI of gynaecological solid masses.

Differential diagnosis of gynaecological masses is sometimes difficult, as there are so many histological types. However, magnetic resonance characteristics of some gynaecological tumours have been reported past several years. On the basis of the recent literature, we have made a decision tree for differential diagnosis of solid gynaecological tumours, in which there are some important divergences. Bilateral disease and invasive growth are malignant signs in most cases. Specific findings for different tumour types include: fibrovascular septa in dysgerminomas; preserving ovarian follicles in round cell tumours; pseudolobular patterns in young patients in sclerosing stromal tumours; and extremely hypointense masses on T2WI in Brenner tumours. Distinguishing between sex-cord stromal tumours, Brenner tumours and metastatic tumours may be hard, however, especially in middle age, because they all tend to show well-demarcated, hypointense masses on T2WI. Disproportionately clear zonal anatomy of the uterus, enlarged uterus and thickened endometrium, which are indirect findings of oestrogen-producing tumours, are useful diagnostic findings in children and postmenopausals.

Ovarian carcinoma in patients with endometriosis: MR imaging findings.

OBJECTIVE: Various types of malignancy can develop in patients with endometriosis. Enhancing mural nodules have been reported as an imaging characteristic of malignant transformations. We evaluated contrast-enhanced MR imaging to determine the optimum sequence to reveal mural nodules and other characteristics of malignant transformations. MATERIALS AND METHODS: We examined 10 patients with pathologically proven ovarian adenocarcinoma in endometriosis and 10 patients (the control group) with ovarian endometrioma suggestive of malignant transformation on the basis of sonographic findings. We analyzed the size and nature of the endometriomas in each patient. We compared four types of contrast-enhanced MR imaging to determine which sequence best revealed mural nodules. RESULTS: In the malignant and control groups, 80% of the cysts with findings suggestive of malignant transformation showed unilateral disease or larger endometrial cysts on the suggestive side than on the contralateral side. High signal intensity on T1-weighted images and low signal intensity on T2-weighted images relative to the myometrium were observed only in two of 10 malignant endometrial cysts and in all control cysts. All malignant endometriomas had small mural nodules with low signal intensity on T1-weighted contrast-enhanced images. Only three benign endometriomas had mural nodules and none of them enhanced. The enhancement of mural nodules was easily seen on dynamic subtraction imaging. CONCLUSION: On the basis of our findings, endometrial cysts with malignant transformation rarely show low signal intensity on T2-weighted images and usually have enhancing mural nodules. Because the enhancement of mural nodules is often difficult to evaluate on conventional T1-weighted images, dynamic subtraction imaging can be valuable.

[Pitfalls in diagnosing female genital tract: ovarian disease].

Because the ovaries are composed of surface epithelial cells, germ cells, sex-cord stromal cells and stroma, tumors of various histopathologies arise from this organ. In addition, a number of non-tumorous diseases form adnexal masses such as endometriosis and tubo-ovarian abscess. Furthermore, tumors arising from other pelvic organs are sometimes mis-diagnosed as ovarian tumors. Therefore, it is difficult to predict the histological type of an adnexal mass prior to surgery even though the MR characteristics of various ovarian tumors have been reported. Because of the large number of patients, there are two 'common sense' considerations in diagnosing female intrapelvic masses in daily practice: 1) solid adnexal masses are malignant ovarian tumors, 2) predominantly cystic intrapelvic masses originate from the ovaries. However, certain tumors do not conform to common sense. For example, fibrothecomas and benign teratomas are exceptions to 1), and subserosal pedunculated leiomyomas and cystic adenomyosis are exceptions to 2). The MR characteristics, pitfalls to diagnosis, and clues to correct imaging diagnosis are discussed in this review.

Differential diagnosis of gynaecological "stained glass" tumours on MRI.

Although multilocular cystic gynaecological masses in which the loculi show variable signal intensity on both T1 and T2 weighted images have been considered to be mucinous cystadenoma or adenocarcinoma, other gynaecological tumours can demonstrate this "stained glass" appearance. These include mature cystic teratoma, fibrothecoma, endometrioma, Brenner's tumour of the ovary and degenerated leiomyoma of the uterus, all of which may mimic mucinous tumours of the ovaries.

Angiomyolipoma of the liver: significance of CT and MR dynamic study.

Angiomyolipoma is a benign mesenchymal tumor that has been reported frequently in the kidney but rarely in the liver. In the present study, we present four cases of hepatic angiomyolipoma with different radiologic appearances, discuss differential diagnosis, and review previously reported cases. One of our cases was followed for 8 years. Computed tomography (plain, enhanced, and dynamic study), magnetic resonance imaging (T1-weighted spin echo, T2-weighted spin echo, and dynamic study), ultrasonography, and angiography were performed in all cases. Although different radiologic appearances were observed in the tumors, based on different proportions of fat, blood vessels, and muscle, three cases were diagnosed as angiomyolipoma. In one case, it was quite difficult to make radiologic diagnosis; hepatocellular carcinoma with fatty metamorphosis in part was most likely suspected, but histopathological examination revealed angiomyolipoma with few fat elements. In the present study, early and prolonged enhancement of the lesion with the special pattern of time density/intensity curve was significant for angiomyolipoma, and we suggest that preoperative radiologic diagnosis of the lesion is possible in most of the cases. However, it can be quite difficult to distinguish angiomyolipoma from some hepatocellular carcinomas with fatty metamorphosis.

Testicular feminization: role of MRI in diagnosing this rare male pseudohermaphroditism.

PURPOSE: The purpose of our study was to assess the role of MR in the diagnosis and follow-up of patients with testicular feminization (TF). METHOD: MR scans of seven patients with TF were retrospectively reviewed. RESULTS: MR could diagnose uterus agenesis in all cases. Undescended testes were depicted within or just below the inguinal canal in four cases. They were smaller and more hypointense than those in normal males. Testes that we could not find by MR were found during surgery in two cases. They were just below the inguinal canal where we had not scanned caudally enough in one and they were too small to be detected in the other. CONCLUSION: MR can demonstrate both uterine agenesis and undescended testes in TF. The latter tend to be missed because they are small and can be anywhere in the abdomen and inguinal canal. Radiologists should perform careful MR examination to detect testes in patients suspected having TF, as these organs are more likely to undergo malignant neoplasia.

Malleus bar as a rare cause of congenital malleus fixation: CT demonstration.

Among congenital ossicular anomalies without external ear atresia, malleus fixation is least common. We report a case of congenital malleus fixation by a bony bar connecting the malleus neck to the posterior tympanic wall, which was depicted on thin-section CT scans.

Uterus didelphys associated with obstructed hemivagina and ipsilateral renal agenesis: MR findings in seven cases.

BACKGROUND: Although a specific syndrome of the uterus didelphys, obstructive hemivagina, and ipsilateral renal agenesis had been reported in gynecologic literatures, it is not familiar to radiologists. We introduce this syndrome with radiological, especially MR findings. METHODS: Imaging findings and medical records of seven patients with this syndrome were retrospectively reviewed. RESULTS: MR clearly demonstrated uterus didelphys with obstructed hemivagina in six of the seven patients. A remaining one showed uterus didelphys with left-sided hematometra and hematosalpinx who had undergone vaginal septum incision prior to the MR. We could also recognize ipsilateral renal agenesis in all five patients, in whom renal areas were examined on MR. After vaginal septum resection, symptoms disappeared in six patients. We could observe the disappearance of the obstructive complications with MR in three patients. CONCLUSIONS: MR can do much for the early diagnosis and the prevention against further complications of this condition because it can demonstrate the müllerian duct anomaly complicated with obstructed hemivagina in detail and even ipsilateral renal agenesis.