Endovascular treatment in postpartum haemorrhage.

PURPOSE: Postpartum haemorrhage (PPH) is one of the main causes of maternal mortality and occurs in 5% of total deliveries. In this study we consider the indications for and technique and results of endovascular treatment for this serious event. MATERIALS AND METHODS: Between January 2004 and December 2010, we conducted a nonrandomised prospective study on ten women with severe PPH who were treated endovascularly in an emergency setting. The procedure was considered to be clinically successful when the PPH resolved completely without the need for further surgical intervention. Laboratory values and the number of transfused blood packs were assessed for each patient. RESULTS: The endovascular procedure completely stopped the bleeding in 8/10 women. After embolisation, the remaining two patients underwent a second laparotomy, which completely arrested the bleeding. No patient died as a result of PPH, and no patient with PPH who avoided hysterectomy before endovascular treatment underwent it after the procedure. CONCLUSIONS: In keeping with the literature, our study indicates that endovascular therapy can significantly help reduce the rates of hysterectomy and mortality due to PPH. This treatment, when performed in the angiography room, is safe and effective and is probably relatively uncommon and underused.

Multifocal, multicentric and contralateral breast cancers: breast MR imaging in the preoperative evaluation of patients with newly diagnosed breast cancer.

PURPOSE: This study was done to verify the usefulness of preoperative breast magnetic resonance (MR) imaging in patients with newly diagnosed breast cancer. MATERIALS AND METHODS: A retrospective analysis of 291 patients with invasive breast cancer newly diagnosed with conventional breast imaging (mammography and ultrasound) was performed. All patients underwent MR imaging prior to surgery. The MR imaging detection rate of additional malignant cancers occult to mammography and ultrasound was calculated. Data were analysed with Fisher's exact test (p<0.05) according to the following parameters: histopathological features of the index tumour (histological type and size) and mammographic density [according to the Breast Imaging Reporting and Data System (BI-RADS) classification from 1 fatty to 4 dense). The gold standard was the histological examination on the surgical specimen. RESULTS: MR imaging identified 40 mammographically and sonographically occult malignant lesions other than the index cancer in 27/291 patients (9%). These additional cancers were located in the same quadrant as the index cancer in 13 women (4%), in a different quadrant in 12 (4%) and in the contralateral breast in the remaining two (1%). The cancer detection rate in the subgroup of index cancers with lobular histological type was 25%, significantly higher (p=0.03) than the detection rate of 11% recorded in the subgroup of ductal cancers. The cancer detection rate in the subgroup of index cancers >2 cm was 27%, significantly higher (p=0.001) than the rate of 8% found in the subgroup of index cancers <2 cm. Mammographic density was not correlated (p=0.48) with MR detection of additional cancer, with 14% of additional malignancies being detected in both dense and fatty breasts. CONCLUSIONS: In patients with newly diagnosed invasive breast cancer, preoperative MR imaging is useful for detecting additional synchronous malignancies that are not detected on conventional breast imaging. The cancer detection rate is 9%. The use of preoperative MR imaging as an adjunct to conventional breast imaging in women with an infiltrating lobular index cancer and an index cancer >2 cm is especially beneficial.

MR-guided vacuum-assisted breast biopsy in the management of incidental enhancing lesions detected by breast MR imaging.

PURPOSE: The objective of this study was to determine the clinical value and accuracy of magnetic resonance (MR)-guided vacuum-assisted breast biopsy (VAB). MATERIALS AND METHODS: We retrospectively analysed 106 incidental breast lesions detected on MR imaging in 98 patients. Patients with nonpalpable suspicious lesions that were only MR visible were referred for MR-VAB performed with a 10-gauge needle. All patients with a VAB diagnosis of infiltrating carcinoma, carcinoma in situ or atypical epithelial hyperplasia were referred for surgery. Histopathology of the surgical specimen was considered the reference standard. RESULTS: MR-guided VAB was attempted in 29/106 lesions (27%); in 2/29 patients, the procedure could not be performed owing to failure to visualise the lesion. Lesions with clearly malignant features and borderline lesions (atypical ductal hyperplasias) were identified in 12 cases (44%) and benign entities in 15 (56%). Seven of 12 (58%) malignant lesions were <10 mm. Among the 27 successful MR-VAB procedures, VAB yielded one false-negative diagnosis (4%) and underestimation (4%). MR-guided VAB sensitivity and specificity were 92% and 100%, respectively, with a positive predictive value of 100% and a negative predictive value of 93%. CONCLUSIONS: The results of this study indicate that MR-guided VAB offers good accuracy in characterising nonpalpable breast lesions visible on MR imaging alone. Small lesion size (<1 cm) did not prove to be a limitation for the success of the procedure.

Inflammatory breast carcinoma and locally advanced breast carcinoma: characterisation with MR imaging.

PURPOSE: This study was done to identify the typical magnetic resonance (MR) imaging findings of inflammatory breast carcinoma (IBC) in comparison with noninflammatory locally advanced breast carcinoma (LABC). MATERIALS AND METHODS: MR images of 30 patients with IBC (T4d) were compared with those of a cohort of 30 patients with LABC (T3/T4a-c). The age distribution was approximately equal in the two groups. MR images were assessed for the following features: skin thickening (>4 mm), skin oedema, architectural distortion, enhancement pattern (mass-like/non-mass-like), time-signal intensity curve (continuous-persistent type/wash-out type), skin enhancement. Fisher's exact text was used to compare MR imaging appearances of IBC and LABC (significant p value <0.05). RESULTS: Skin involvement and enhancement pattern differed between groups: skin thickening was present in 16/30 IBC (53%) vs 8/30 LABC cases (27%, p=0.06), skin oedema was present in 26/30 IBC (87%) vs 8/30 LABC (27%, p < 0.0001), and skin enhancement in 10/30 IBC (33%) vs 2/30 LABC (7%, p=0.02); non-mass-like enhancement was present in 22/30 IBC (73%) vs 12/30 LABC (40%, p=0.02). CONCLUSIONS: IBC is a distinct clinical and pathological entity resulting in typical MR imaging features. Skin changes (thickening, oedema, enhancement) related to neoplastic involvement of the dermal lymphatics are suggestive of IBC and should prompt a skin biopsy to confirm or rule out the diagnosis.

Utility of second-look ultrasound in the management of incidental enhancing lesions detected by breast MR imaging.

PURPOSE: To assess the utility of second-look ultrasound (US) for identifying and characterising incidental enhancing lesions detected by breast magnetic resonance imaging (MRI). MATERIALS AND METHODS: From among 655 consecutive breast MRI studies, 62 lesions (MRI visible, nonpalpable, occult at first-look US and mammography) were recommended for second-look US. MRI enhancement of lesions was mass-like in 59 cases (95%) and non-mass-like in three (5%). Forty-two lesions (68%) were ≤10 mm; only three lesions (5%) were >20 mm. Of all lesions, the Breast Imaging Reporting and Data System (BI-RADS) MRI category was highly suggestive of malignancy in six cases (10%), suspicious abnormality in 33 (53%) and probably benign in 23 (37%). The correlation between MRI lesion appearance, lesion size, histopathology findings and detection rate at second-look US were analysed. The reference standard was histopathology and/or follow-up (range 18-24 months). Statistical analysis was performed with the Fisher exact test. RESULTS: Second-look US identified 44 out of 62 (71%) lesions depicted at MRI. The detection rate at second-look US was higher for mass-like MRI lesions (75%) than non-mass-like lesions (0%), for lesion size >10mm (90%) and for BI-RADS 4 lesions (88%). Second-look US-guided biopsy detected 12 out of 17 (71%) malignant lesions. There was no correlation between the likelihood of carcinoma and the presence of a sonographic correlate. CONCLUSIONS: Second-look US is a reliable problem-solving tool in identifying and characterising most incidental MRI findings. It contributes to accurately selecting the cases in which MRI-guided biopsy is required.

Combined radiofrequency and kyphoplasty in painful osteolytic metastases to vertebral bodies.

PURPOSE: The aim of this study was to assess retrospectively the safety and efficacy of combined radiofrequency (RF) and kyphoplasty (KP) in managing painful osteolytic metastases to vertebral bodies resistant to conservative treatments. MATERIALS AND METHODS: Eleven patients (9 women and 2 men; mean age 68 years; age range 58-82) with painful osteolytic vertebral body metastases unresponsive to conservative treatments underwent RF combined with KP under general anaesthesia. Primary neoplasms were kidney carcinoma (n=1), breast carcinoma (n=1), thyroid carcinoma (n=2) and multiple myeloma (n=7). Lesion levels were cervical (n=1), thoracic (n=9) and lumbar (n=1). Combined RF and KP was well-tolerated by all patients. The procedures were performed using fluoroscopic guidance and intraoperative neurophysiology monitoring. Pain relief with the visual analogue scale (VAS) pain score and analgesic consumption were evaluated before and after treatment. RESULTS: No complication occurred. In one case, we observed an asymptomatic cement leakage. Pain significantly decreased after treatment: the mean VAS pain score before treatment was 8 (range 7-10) vs. 1.8 (range 0-3) and 1.9 (range 1-3), respectively, 72 h and 6 weeks after the treatment. Analgesic reduction was achieved in all patients. CONCLUSIONS: RF combined with KP represents a potential alternative method for palliation of painful spinal osteolytic metastases in selected patients. The procedures are safe and provide pain relief with bone augmentation and improvement in quality of life.

Fischer's score criteria correlating with histopathological prognostic factors in invasive breast cancer.

PURPOSE: This study aimed to evaluate whether the Fischer score criteria on contrast-enhanced magnetic resonance (CE-MR) imaging could correlate with histopathological prognostic factors in invasive breast cancer. MATERIALS AND METHODS: Seventy-two women with histologically proven invasive breast cancer underwent preoperative CE-MR imaging. Images were assessed for the following parameters, according to the scoring system described by Fischer in 1999: tumour shape, margins, internal enhancement, signal intensity increase, signal intensity course and overall Fischer score. Evaluated histopathological prognostic factors included histological type, associated extensive intraductal component, diameter, lymph node metastasis, tumour grade, and oestrogen receptor (ER), progesterone receptor (PgR), Ki67 proliferation, oncogene c-erbB-2 (HER2/neu) expression. Fisher's exact test was used to correlate the CE-MR imaging parameters and histopathological findings (with significance set a p < 0.05). RESULTS: Fischer's score was 0-4 in 14/72 (19%) cases, >4 in 58/72 (81%) and 3 in 5/72 (7%; false negative), with a sensitivity of 93%. A significant correlation (p=0.02) was found between stellate-dendritic shape and the presence of an associated extensive intraductal component (EIC), which was found in 78% of stellate tumours vs. 49% of round-oval tumours. A significant correlation (p=0.039) was found between Ki67 expression and signal intensity course (Ki67 overexpression was present in 81% of tumours with washout course vs. 21% with plateau course). CONCLUSIONS: The CE-MR imaging findings that correlate with prognostic factors are shape and signal intensity curve. Fischer's multifactorial analysis was helpful in the interpretation of CE-MR images, showing a sensitivity of 93% for invasive breast cancer.

Inflammatory breast cancer: MR imaging findings.

PURPOSE: This paper describes the magnetic resonance (MR) imaging features of primary inflammatory breast cancer (IBC). MATERIALS AND METHODS: Two radiologists reviewed the MR examinations of 14 women with a pathological diagnosis of IBC. Images were assessed for skin thickening, oedema, nipple retraction, architectural distortion, type and extent of parenchymal and cutaneous enhancement and enhancement kinetics over time, axillary and internal mammary lymphadenopathy, pectoral muscle enhancement and additional findings. RESULTS: Skin thickening was identified in eight patients (58%), oedema in nine (64%), nipple retraction in two (14%), architectural distortion in eight (58%), mass-like enhancement in five (36%), non-mass-like enhancement in nine (64%) with washout enhancement curve in 12 (86%) and plateau curve in two (14%), axillary lymphadenopathy in 12 (86%) and internal mammary artery lymphadenopathy in two (14%), and pectoral muscle enhancement in one (7%). Additional findings included increased breast volume in two patients (14%), prepectoral fluid in four (28%) and hypertrophic internal mammary artery in three (21%). CONCLUSIONS: The most characteristic MR findings of IBC are skin thickening, oedema, architectural distortion, mass-like enhancement with washout curve and axillary lymphadenopathy; less frequent ones are nipple retraction, mass-like enhancement and internal mammary lymphadenopathy. Prepectoral fluid is frequent but is not a sign of infiltration.

Autoimmune pancreatitis: imaging findings on contrast-enhanced MR, MRCP and dynamic secretin-enhanced MRCP.

PURPOSE: This study retrospectively determined magnetic resonance (MR), MR cholangiopancreatography (MRCP) and secretin-MRCP findings in patients with autoimmune pancreatitis (AIP). MATERIALS AND METHODS: The MR examinations of 28 patients with histopathologically proven AIP were reviewed. In 14 cases, secretin-enhanced MRCP was performed. The observers evaluated pancreatic parenchymal enlargement, signal intensity abnormalities, enhancement, vascular involvement, bile-duct diameter and main pancreatic duct (MPD) narrowing (diffuse/focal/segmental). After secretin administration, the presence of the "duct-penetrating" sign was evaluated. RESULTS: MR imaging showed diffuse pancreatic enlargement in 8/28(29%) cases, focal pancreatic enlargement in 16/28 (57%) cases and no enlargement in 4/28 (14%) cases. The alteration of pancreatic signal intensity was diffuse in 8/28 (29%) cases (eight diffuse AIP) and focal in 20/28 (71%) cases (20 focal AIP). Delayed pancreatic enhancement was present in all AIP, with peripheral rim of enhancement in 8/28 (29%) AIP (1/8 diffuse, 7/20 focal); vascular encasement was present in 7/28 (25%) AIP (1/8 diffuse, 6/20 focal); distal common bile duct narrowing was present in 12/28(43%) AIP (5/8 diffuse, 7/20 focal). MRCP showed MPD narrowing in 17/28 (61%) AIP (4/8 diffuse, 15/20 focal), MPD dilation in 8/28(29%) AIP (3/8 diffuse, 5/20 focal) and normal MPD in 1/8 diffuse AIP. Secretin-MRCP showed the duct-penetrating sign in 6/14(43%) AIP (one diffuse AIP with MPD segmental narrowing, five focal AIP with MPD focal narrowing), demonstrating integrity of the MPD. CONCLUSIONS: Delayed enhancement and MPD stenosis are suggestive for AIP on MR and MRCP imaging. Secretin-enhanced MRCP is a problem-solving tool in the differential diagnosis between focal AIP and ductal adenocarcinoma.

Treatment of type-A3 amyelic thoracolumbar fractures (burst fractures) with kyphoplasty: initial experience.

The authors propose a minimally invasive procedure for treating type-A3 amyelic thoracolumbar fractures according to Magerl classification (compression burst fractures). The procedure, percutaneous kyphoplasty, allows the fracture to be reduced and vertebral height to be restored by injecting bone cement into a cavity created in the vertebral body by an inflatable balloon introduced with the percutaneous approach. Four patients were successfully treated, with clinical and functional benefit in all cases. There were no complications.

Assessment of pelvic endometriosis: correlation of US and MRI with laparoscopic findings.

PURPOSE: The purpose of this study was to evaluate the contribution of ultrasound (US) and magnetic resonance (MR) imaging in the diagnosis and local staging of endometriosis by comparing results with laparoscopic findings. MATERIALS AND METHODS: We evaluated 36 consecutive women with suspected or clinically diagnosed endometriosis. Thirty-two out of 36 patients met the following inclusion criteria: transabdominal and endocavitary (US) examination and MR imaging, followed by laparoscopy performed within 2 weeks. US and MR findings were classified based on location, number and morphology (small nodules, large nodules, laminar lesions, cystic lesions, complex lesions, adhesions, cul-de-sac obliteration). RESULTS: Laparoscopy, considered the gold standard, identified 143 lesions in 32 patients. US detected 101 lesions, and MR detected 92 lesions, which were subsequently divided by morphologic appearance. Sensitivity and specificity of the two imaging techniques in the recognition of the different locations were 58% and 25%, respectively, for US and 56% and 50%, respectively, for MR imaging. Results of the two techniques in the different locations examined were similar, with the exception of lesions in the rectovaginal septum, which were better detected by US, and for adhesions and cul-de-sac obliteration, which were more easily detected by MR. CONCLUSIONS: Both US and MR are accurate in the diagnosis of endometriosis. There are no significant differences in staging of pelvic endometriosis between US and MR. US examination is the primary evaluation in cases of suspected disease and for the rectovaginal septum. MR examination is recommended for correct classification in doubtful cases and in cases of suspected extrapelvic lesions and adhesions.

Angiosonography in suspicious breast lesions with non-diagnostic FNAC: comparison with power Doppler US.

PURPOSE: The purpose of this study was to evaluate the accuracy of angiosonography in comparison with colour Doppler ultrasound (US) in the discrimination of suspicious breast lesions with nondiagnostic fine-needle aspiration cytology (FNAC). MATERIALS AND METHODS: Pre-operative Power Doppler US and angiosonography were prospectively performed in 20 suspicious breast lesions with non-diagnostic FNAC. A second-generation US contrast agent was utilised with a high-frequency transducer and a contrast-specific algorithm (low acoustic pressure CnTI). The enhancement characteristics of all lesions were analysed using qualitative and quantitative parameters obtained from time-intensity curves with the different imaging modalities. The final diagnosis was confirmed at pathology in all cases. Microvessel density (MVD) was assessed in the surgical specimen using CD34. RESULTS: The correct assessment of biological behaviour was achieved in all cases by angiosonography (sensitivity of 100%; specificity of 91%) and colour Doppler US (45% sensitivity; 78% specificity). MVD correlated with the biological behaviour. CONCLUSIONS: Angiosonography is more accurate than colour Doppler US in the correct assessment of biological behaviour of suspicious breast lesions.

Branch duct IPMTs: value of cross-sectional imaging in the assessment of biological behavior and follow-up.

Branch duct intraductal papillary mucinous tumors (IPMTs) are increasingly being described and represent a challenge for the physician. Recent imaging modalities, especially computed tomography and magnetic resonance, allow for a correct diagnosis, but IPMTs can manifest with different degrees of cellular atypia so it is very important to be able to establish the biological behavior of the lesion. Sixty-five patients were included in this study: 29 of them underwent surgery, and the other 36 were followed with cross-sectional imaging. Among patients who underwent surgery, 11 had benign lesions and 18 had malignant lesions. Overall sensitivity, specificity, and accuracy in the diagnosis of malignancy for all the cross-sectional imaging techniques were 61%, 81%, and 69%, respectively. Among patients who were followed and had a magnetic resonance cholangiopancreatographic diagnosis of branch duct IPMT with imaging features indicative of benignity, only two showed modifications over the follow-up period. Even considering some important biases (small number of patients and relatively short follow-up), interesting conclusions can be drawn: the imaging diagnosis of malignancy can be trusted, whereas that of benignity cannot be relied upon, but if the signs of malignancy are absent, the slow growth of the lesions justifies watchful follow-up in the correct clinical setting.

Pelvic endometriosis: US and MRI features.

Endometriosis represents a common and important clinical problem of women of childbearing age. It is a disabling disorder manifesting with pain and infertility. The exact pathogenesis of the disease remains unclear, despite the different theories that have been formulated. The literature on endometriosis is extensive, but often in regard to classic endometrioma. It is surprising that, to the best of our knowledge, the many radiologic features of extraovarian endometriosis have not been well documented thus far. Although ultrasound (US) remains the imaging modality of choice in the radiologic evaluation of female patients with pelvic pain, the role of magnetic resonance imaging (MRI) in the evaluation of abdominal pain is expanding. In the young patient, MRI may be performed if a gynecologic disorder is not suspected at first, especially if US findings are equivocal or the abnormality extends beyond the field of view of the sonographic probe. Moreover, MRI is useful whenever further characterization of pelvic disorder is required. In fact, many causes of pelvic disorders and of endometriosis in particular demonstrate characteristic MRI findings. For these reasons, in this work we describe the protean US and MRI appearances of endometrial foci as encountered in daily experience.

Spiral computed tomography assessment of resectability of pancreatic ductal adenocarcinoma: analysis of results.

BACKGROUND: Despite more recent technical advances, single detector spiral computed tomography is still the most widely used imaging technique for the detection and staging of pancreatic adenocarcinoma. Many reports have recently focused on single detector spiral computed tomography imaging findings indicative of unresectability AIM: To evaluate the ability of single detector spiral computed tomography in the pre-operative staging of ductal adenocarcinoma of the pancreas in selecting surgically resectable versus unresectable cases, considering different parameters (vascular and local infiltration, liver metastases, lymphadenopathy, and peritoneal carcinomatosis). PATIENTS AND METHODS: A total of 100 patients diagnosed with ductal adenocarcinoma of the pancreas, pathologically verified, and examined with single detector spiral computed tomography were considered. Of these, 63 underwent surgery; 37 were excluded (imaging evidence of unresectability in biopsy-proven adenocarcinomas). Single detector spiral computed tomography studies, performed at the time of surgery, were retrospectively reviewed by two radiologists. The final results were reached by consensus, without being aware of the operation. All prospective clinical readings were also considered, and tabulated. RESULTS: In the patients undergoing surgery, 44 proved to be unresectable and 19 resectable. Considering the above-mentioned parameters, the prospective assessment of resectability demonstrated a rather low sensitivity (68%), with high specificity (95%). Accordingly, a high positive predictive value (97%) was achieved, while the negative predictive value was 56%. After the retrospective evaluation, the value of sensitivity increased (84%) to the expense of the specificity (84%). Moreover, a positive predictive value of 92.5%, and a negative predictive value of 69.5% were obtained. CONCLUSIONS: The demonstration of many signs of unresectability of pancreatic ductal adenocarcinoma reported in the literature yields a better sensitivity in the diagnosis, but unfortunately, is associated with an inevitable reduction in specificity.

Cystic tumours of the pancreas: a review.

The detection of a cystic tumour of the pancreas is a challenge which puts not only the surgeon's knowledge and expertise to the test, but also those of the team of radiologists and pathologists with whom he works. The diagnosis of a suspected pancreatic cystic tumour is morphological and is based on modern imaging techniques and, in the case of intraductal papillary mucinous tumours, on endoscopic findings. In the search for the correct preoperative diagnosis, however, it is of fundamental importance to bear in mind the limitations of the various instrumental investigations, and particularly those of fine-needle aspiration cytology. In this light the main morphological and clinicopathological features of serous cystadenomas, mucinous adenomas and adenocarcinomas, intraductal papillary mucinous tumours and papillary cystic and solid tumours are analysed as well as the surgical indications. In fact only the surgeon, on the basis of his knowledge of the patient's medical history and symptoms, will be in a position to determine to which nosological "cystic" entity the morphological findings described belong. A deeper knowledge of the natural history of each of these cystic tumours will help the surgeon formulate the most appropriate treatment indication. Providing the patient's condition fulfills the necessary operability criteria, resection will be mandatory whenever there exists a doubt that the tumour may be malignant or whenever its natural history suggests a malignant potential.

Intraductal papillary mucinous tumors of the pancreas: spectrum of CT and MR findings with pathologic correlation.

The reports of intraductal papillary mucinous tumors (IPMT) of the pancreas are increasingly more frequent in the literature. The diagnosis by means of cross-sectional imaging of these tumors is not easy, especially in the early stages, when they can mimic an inflammatory disease of the pancreas. Prompt identification of the disease is nevertheless extremely important, especially in the case of tumors originating from the collateral branches, since its recognition can modify the management of the patient, in some cases obviating recourse to surgery.

[Serous cystadenoma of the pancreas: imaging findings]. / Cistoadenoma sieroso del pancreas: i rilievi del'imaging.

PURPOSE: To present the imaging findings of serous cystadenoma (SCA) and discuss the main problems of differential diagnosis with other cystic pancreatic lesions. MATERIALS AND METHODS: 55 cases of SCA were reviewed; all the lesions proved to be benign at pathology. 35 tumors were evaluated with Ultrasonography (US), 49 with Computed Tomography (CT), and 15 with Magnetic Resonance Imaging (MRI). RESULTS: Among the patients who underwent either demolitive (47) or derivative (3) interventions, two different morpho-structural patterns were identified: microcystic (37), and oligocystic (13). The diagnosis of SCA, possible in the presence of microcystic pattern, was achieved in 63% of cases by US (22/35), in 63% of cases by CT (31/49), and in 73% of cases by MRI (11/15). In 12 patients evaluated with all the imaging modalities, the combined information allowed a correct diagnosis in 10 cases (83%). The 13 oligocystic tumors were almost always undistinguishable from other cystic masses of the pancreas. CONCLUSIONS: The diagnosis of SCA can be considered certain if a microcystic pattern is detected. A correct diagnosis is not achievable in the presence of oligocystic architecture. MRI is the best imaging modality in characterizing this tumor.

CT features of malignant mucinous cystic tumors of the pancreas.

This study is aimed at evaluating which pre-operative findings at CT are prevailingly associated with histologically malignant appearance of mucinous cystic tumor (MCT) of the pancreas. The CT examinations of 52 female patients affected with pathologically proved MCT were retrospectively evaluated by two radiologists, blinded to the histopathological assessment of the biologic behavior (either benign or malignant). A multivariate logistic regression analysis was performed in order to identify the most important features associated with the malignant nature of MCT. Calcifications in the wall and/or in the septa, thick wall, and septations resulted in the most important features associated with malignancy. The simultaneous presence of all these radiological signs constitutes an almost "certain" marker of malignancy, being the risk equal to 0.95, whereas the simultaneous presence of at least two of them entails a risk of malignancy ranging from 0.56 to 0.74, according to the type of morphological features. On the other hand, the absence of the mentioned radiological features entails a negligible risk of malignancy (0.02). Mucinous cystic tumors characterized by multilocular macrocystic architecture, with thick wall and calcifications in the wall and/or the septa, present the highest risk of malignancy, and advocate prompt surgical intervention. When those signs are absent, the probability of malignancy is lower. In this case a non-operative management is possible.

Nonfunctioning endocrine tumors of the pancreas: possibilities of spiral CT characterization.

The aim of this study was to assess the ability of spiral CT to adequately characterize the nonfunctioning endocrine tumors (NFETs) of the pancreas, distinguishing this lesion from the other pancreatic tumors. The spiral CT examinations of 21 cases of histologically proven NFETs, along with those of 29 cases of other pancreatic tumors and tumor-like lesions, were retrospectively reviewed in a blinded fashion by two radiologists, in order to correctly classify the lesions, highlighting the typical signs reported in the literature. Discordant cases were further analyzed in the presence of a third radiologist. The final diagnosis was acquired by means of a majority or overall consensus. The histopathologic examination was considered the gold standard. The sensitivity, specificity, and positive and negative predictive values of CT were calculated. After the consensus evaluation, the correct diagnosis was reached in 72% of cases, with 10% of nonspecific diagnoses of solid pancreatic tumor and 18% of wrong diagnoses. The sensitivity and specificity of spiral CT in identifying NFETs were 66.6 and 82.7%, respectively. The positive and negative predictive values were 73.7 and 77.4%, respectively. In up to 70% of cases the NFET demonstrates a typical aspect of a mass hyperdense in the arterial contrastographic phase eventually associated with hyperdense hepatic metastases in more than half of the patients. This finding does allow the diagnosis of NFET but without certainty indeed, since other tumors can show a similar densitometric behavior and among them particularly the ductal adenocarcinoma. On the other hand, both the solid, hypovascularized NFETs, and the cystic form, cannot be differentiated from the other solid and cystic tumors of the pancreas.