| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 All authors: Department of Abdominal Imaging and Intervention, Massachusetts General Hospital and Harvard Medical School, 55 Fruit St., White 270, Boston, MA 02114.
Received January 24, 2006;
accepted after revision May 16, 2006.
Address correspondence to D. V. Sahani
(dsahani{at}partners.org).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. In the presence of hepatic tumors, enhancing expansile portal vein thrombus is highly suggestive of hepatocellular carcinoma.
Keywords: abdominal imaging • cancer • liver disease
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The presence of multiple hepatic lesions on initial imaging studies performed for known or unknown hepatic disease can present a diagnostic challenge. Determining whether these lesions are primarily hepatic in origin or due to metastasis is important for treatment. When the imaging findings of cirrhosis coexist with those of hepatic lesions, HCC is the most likely diagnosis. Ethanol, however, which is one of the causative agents of alcoholic cirrhosis, is known also to increase the risk of other gastrointestinal malignant tumors, such as colon, esophageal, pharyngeal, and rectal cancer, all of which can metastasize to the liver [3]. HCC also can be present in patients without imaging evidence of cirrhosis. The objective of this study was to assess whether the presence of enhancing and expansile thrombi in the portal vein is diagnostic of HCC and whether this criterion can be used to differentiate HCC from hepatic metastatic lesions of other tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients
A total of 47 patients satisfied the inclusion criteria. Of these, 36
patients (32 men, four women; age range, 45-81 years; mean age, 62 years) had
HCC, and 11 (six men, five women; age range, 42-77 years; mean age, 60 years)
had other malignant tumors as follows: three, cholangiocarcinoma; three,
ovarian carcinoma; two, pancreatic carcinoma; two, colon carcinoma; and one,
gastrointestinal stromal tumor. Of the 36 patients with a final diagnosis of
HCC, 13 had undergone CT; 17, MRI; and six, both CT and MRI. Five of the 11
patients with malignant tumors other than HCC had undergone CT, and six had
undergone MRI. Images of 15 age-matched control subjects were evaluated for
diameter of the main portal vein and right and left branches of the portal
vein. None of these patients had liver disease.
Imaging Technique
CT—Patients underwent contrast-enhanced arterial and portal
venous phase 4- or 16-MDCT (n = 15) or single-detector CT (n
=9). Arterial phase scans were acquired 25-30 seconds and portal venous phase
scans approximately 65-70 seconds after IV administration of contrast
material. A total of 135-150 mL of nonionic iodinated contrast material was
injected IV at 4 mL/s in all patients. A maximum slice thickness of 5 mm was
used for scanning (1.25-2.5 mm for the arterial phase, 2.5-5 mm for the portal
venous phase).
MRI—MRI was performed on a 1.5-T system. Unenhanced T1- and T2-weighted and gradientecho images were acquired with phased-array coils. Slice thickness varied between 5 and 8 mm. Dynamic contrast-enhanced sequences (TR/TE, 150-200/1.8-4.2) were performed in the arterial (20-second scanning delay), portal venous (70-second scanning delay), and equilibrium phases (180-second scanning delay) after administration of 20 mL of gadolinium contrast material.
Image Analysis
CT scans and MR images were evaluated on a PACS (Impax 4.0, Agfa) by an
experienced radiologist with more than 11 years of radiology experience and 5
years of experience in interpreting CT scans and MR images of the liver. The
reviewer was blinded to the clinical diagnosis. With a predesigned template,
images of each subject were evaluated for the presence of liver tumors. The
characteristics were described as either focal or infiltrative. The main
portal vein and the right and left branches were assessed for thrombus
(present or absent). The maximal diameter of the intrahepatic portion of the
main portal vein and right and left branches was assessed with electronic
calipers. The hepatic veins and inferior vena cava also were evaluated for the
presence of thrombus. The diameters of the main portal vein and right and left
branches were measured in the group of age-matched controls.
In patients with paired unenhanced and contrast-enhanced images, the degree of thrombus enhancement was calculated with a ratio of intensity or density on unenhanced and contrast-enhanced images. Paired unenhanced and contrast-enhanced data were analyzed for 26 patients, 20 who underwent MRI and six who underwent CT. The distribution of images analyzed for enhancement in the HCC and non-HCC groups was four sets of CT scans and 16 sets of MR images for the HCC group and two sets of CT scans and four sets of MR images for the non-HCC group. Utmost care was taken to place the region of interest within the thrombus in the portal vein and to select the slices in which the thrombus was distinctly visible in at least two consecutive images. This step ensured elimination of the possibility of partial volume averaging due to recanalization or partial thrombosis of the portal vein.
Statistical Analysis
The mean diameters of the main portal vein and right and left branches were
compared between veins exhibiting thrombus and control veins by means of an
unpaired Student's t test. The Mann-Whitney U test was used
for analysis of percentage enhancement of the thrombus.
Standard of Reference
The diagnosis of liver tumor was established in 39 patients by biopsy (28
HCC, 11 non-HCC). In the other eight patients, the diagnosis of HCC was
established with an 
-fetoprotein level greater than 280 µg/L (range,
286-169,000 µg/L) and the presence of typical imaging features.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Twenty-one of the 36 patients with the final diagnosis of HCC had thrombi involving the main portal vein and the right branch, left branch, or both. Fifteen of the 36 patients had isolated right or left branch involvement. Seven of the 11 patients with malignant tumors other than HCC had thrombi in the main portal vein and the right branch, left branch, or both, and four of the 11 had involvement of the right or the left branch of the portal vein. The mean maximal diameter of the veins containing thrombus was compared with that of normal veins to assess for expansion of the intrahepatic portion of the main portal vein and its right and left branches. These results are presented in Table 2.
|
In patients with the clinical diagnosis of HCC, the presence of thrombus in the portal vein or its right or left branches resulted in marked expansion of the involved veins (Fig. 2). The mean expansion of the involved veins was not significantly greater than that of the uninvolved veins in patients with malignant tumors other than HCC. Involved right portal veins were markedly contracted in patients with other malignant tumors.
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that the presence of expansile and enhancing intrahepatic portal venous thrombus is strongly suggestive of the diagnosis of HCC in patients with hepatic lesions. On the other hand, intrahepatic portal venous thrombi in patients with other malignant tumors, such as cholangiocarcinoma, ovarian cancer, pancreatic cancer, colon cancer, and gastrointestinal stromal tumor, tend not to expand the portal vein. These portal vein thrombi are likely a reflection of bland thrombi due to a hypercoagulable state or to extrinsic compression on the extrahepatic portal vein, especially in pancreatic cancer. Tublin et al. [4] compared the CT characteristics of benign versus malignant portal vein thrombi in patients with cirrhosis by assessing portal vein diameter and subjective enhancement of the thrombus. In that study, increased diameter and increased subjective enhancement were found in the malignant portal vein thrombus group. In our study, expansion and quantitative enhancement of the portal vein on CT and MRI were found more specific for HCC than for thrombi due to other tumors.
MRI features of portal vein thrombus have been described wherein intraluminal thrombi have variable signal intensity, often appearing isointense on T1-weighted images and having increased signal intensity on T2-weighted images [9]. However, reliable differentiation of tumor thrombus from bland thrombus on the basis of MRI signal intensity alone remains difficult.
The relatively low incidence of portal vein thrombus among patients with malignant tumors other than HCC is likely related to the exclusion of patients with solely or predominantly extrahepatic thrombus. These patients were excluded because of the unpredictability of secondary effects on the intrahepatic veins (e.g., a patient with occlusive extrahepatic portal venous thrombus would be likely to have decreased diameter of the intrahepatic veins as a secondary effect unrelated to the presence or absence of intrahepatic thrombus). Malignancy has been shown to be a major etiologic factor in extrahepatic portal venous thrombosis, accounting for 24% of the cases in one series [10]. Macroscopic portal venous thrombus is rare in cholangiocarcinoma and may reflect mixed HCC and cholangiocarcinoma in a number of cases. Intrahepatic portal venous tumor thrombus in pancreatic cancer also is rare, typically reflecting direct extension from tumor invading the superior mesenteric vein or portal-splenic confluence [11, 12].
The presence of portal vein tumor thrombus has important implications for therapy. Surgical resection and transplantation have been traditionally contraindicated in the presence of portal venous thrombus. Although some reports show a tendency toward aggressive management of advanced HCC such as extending partial hepatic resection and liver transplantation to patients with documented tumor thrombus, macroscopic tumor thrombus is still the most significant predictor of tumor recurrence in patients undergoing orthotopic liver transplantation [7, 13].
Our study had several limitations. Because it was a retrospective review
over 5 years, imaging studies were heterogeneous. Both single-detector and
MDCT studies were included. However, the slice thickness and contrast
injection parameters were comparable in both single-detector and MDCT
protocols. Histopathologic findings on thrombi within the portal vein were not
available for these patients. However, biopsy proof of tumor within the liver
was available for all but eight patients with HCC. For these patients,
elevated -fetoprotein level and imaging criteria were used for
confirmation. The relatively small study population was another
limitation.
In conclusion, the presence of enhancing expansile portal venous thrombus is strongly suggestive of HCC in the presence of hepatic lesions. These imaging characteristics have important implications not only for diagnosis but also for the staging and prognosis of HCC.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
