| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Radiology, Asan Medical Center, University of Ulsan, 388-1
Poongnap-Dong, Songpa-Ku, Seoul, 138-736, Korea.
2 Present address: Department of Radiology, Cheonan Hospital, Soonchunhyang
University, 23-20 Bongmyungdong, Cheonan, Choongnam, 330-721, Korea.
Received June 14, 2002;
accepted after revision November 15, 2002.
Address correspondence to T. K. Kim.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. This study included 137 patients who underwent combined CT during arterial portography and CT hepatic arteriography for the preoperative evaluation of known or suspected hepatocellular carcinoma. The images were prospectively evaluated to identify focal hepatic lesions and their differential diagnoses (hepatocellular carcinoma versus pseudolesion). We assessed the diagnostic accuracy of our prospective interpretation by comparing the interpretations with the results of histopathology or follow-up imaging. We also retrospectively analyzed imaging features seen on CT during arterial portography and CT hepatic arteriography—the size, shape, and location of the lesion within the liver; attenuation of the lesion; and opacification of the peripheral portal vein branches on CT hepatic arteriography.
RESULTS. One hundred and forty-nine hepatocellular carcinomas (75 lesions confirmed at histopathology and 74 lesions on follow-up imaging) were found in 120 patients, and 104 pseudolesions (15 lesions confirmed at histopathology and 89 lesions on follow-up imaging) were found in 91 patients. The sensitivity of our prospective interpretations was 98.7%, and the specificity of our prospective interpretations was 90.4%. Our positive and negative predictive values were 93.6% and 97.9%, respectively. We found that hepatocellular carcinomas were larger, more frequently nodular, and more likely to be located intraparenchymally than were the pseudolesions (p < 0.01). Opacification of the peripheral portal vein branches on CT hepatic arteriography was detected in 36 pseudolesions (34.6%) but in none of the hepatocellular carcinomas (p < 0.01).
CONCLUSION. Combining CT during arterial portography and CT hepatic arteriography is useful for the preoperative evaluation of patients with known or suspected hepatocellular carcinoma. Familiarity with the imaging features of hepatocellular carcinomas and pseudolesions can help in the accurate differentiation of hepatocellular carcinomas from pseudolesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, many investigators have frequently encountered various types of pseudolesions on CT during arterial portography or CT hepatic arteriography that were depicted as focal perfusion abnormalities, presumably caused by delivery of the blood supply to the liver via the parabiliary venous system or various types of arterioportal shunts [10, 11, 12, 13, 14, 15]. The pseudolesions might mimic true hepatic lesions and thereby evoke diagnostic uncertainty. Moreover, because of the invasiveness, expense, and complexity of CT during arterial portography and CT hepatic arteriography, the practicality of using this combination of modalities for preoperative evaluation has not been determined. Several authors have reported that the effectiveness of MR imaging or triple-phase helical CT equals or exceeds that of combined CT during arterial portography and CT hepatic arteriography for the preoperative detection of malignant hepatic tumors [16, 17, 18].
The imaging features of pseudolesions associated with liver cirrhosis have been described in several previous reports [19, 20, 21]. In our experience, the detailed analysis of images obtained from CT during arterial portography and CT hepatic arteriography is helpful in differentiating pseudolesions from hepatocellular carcinomas if previously described imaging features of pseudolesions are considered at interpretation. To our knowledge, there have been no previous reports on the diagnostic accuracy of the combined use of CT during arterial portography and CT hepatic arteriography for the preoperative evaluation of patients with known or suspected hepatocellular carcinoma that have included detailed analysis of the differentiating features of pseudolesions. Furthermore, to our knowledge, detailed analysis of the imaging features of pseudolesions on CT during arterial portography and CT hepatic arteriography in a large series has not yet been performed.
The purpose of our study was to determine the usefulness of combined CT during arterial portography and CT hepatic arteriography in the preoperative evaluation of patients with known or suspected hepatocellular carcinoma and to describe the imaging findings on CT during arterial portography and CT hepatic arteriography by which hepatocellular carcinomas may be differentiated from pseudolesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
One hundred and twenty-eight patients were excluded from this study because the interval between the combined CT during arterial portography and CT hepatic arteriography examination and follow-up two-phase liver CT was less than 6 months (n = 76) or because the patients had inadequate-quality studies from CT during arterial portography and CT hepatic arteriography due to a replaced hepatic artery (n = 32) or a large area of Lipiodol (iodinated oil, Amersham Health, Seoul, Korea) uptake resulting from a previous transcatheter arterial chemoembolization (n = 20).
The diagnosis of hepatocellular carcinoma was based on the results of
surgical resection, percutaneous biopsy, clinical or laboratory test results
(including elevated 
-fetoprotein levels) typical for hepatocellular
carcinoma combined with a typical angiographic appearance (such as a
hypervascular tumor with neovascularity), and Lipiodol uptake seen on
follow-up CT scans [22]. The
diagnostic criteria for a nontumorous pseudolesion were a hypoattenuated focal
lesion seen on CT during arterial portography or a hyperattenuated focal
lesion seen on CT hepatic arteriography that either showed no interval
increase or could not be detected on a follow-up two-phase liver CT performed
at least 6 months after the combined CT examination, no evidence of compact
nodular uptake of Lipiodol after transcatheter arterial chemoembolization, and
no evidence of the lesion in the surgical resection.
Imaging Techniques
For combined CT during arterial portography and CT hepatic arteriography,
arterial vascular access was obtained with two unilateral femoral artery
punctures using the Seldinger technique. Two 5-French angiographic catheters
(Rosch hepatic, Cook, Bloomington, IN) were selectively placed, one in the
superior mesenteric artery and the other in the common hepatic artery. Before
CT during arterial portography and CT hepatic arteriography, celiac and
superior mesenteric angiography was performed using 50–60 mL of
iopromide (Ultravist 300, Schering, Berlin, Germany) to evaluate tumor
vascularity and the vascular anatomy.
The combination of CT during arterial portography and CT hepatic arteriography was performed on a Somatom Plus-S (Siemens, Erlangen, Germany), HiSpeed CT/i (General Electric Medical Systems, Milwaukee, WI), or a LightSpeed QX/i (General Electric Medical Systems) scanner. CT during arterial portography and CT hepatic arteriography were performed 20–30 min after angiography. We performed CT during arterial portography first. After a pause of at least 5 min, we performed CT hepatic arteriography. For CT during arterial portography, 80 mL of iopromide (Ultravist 370, Schering) was injected with a power injector through the superior mesenteric artery at a rate of 2.5 mL/sec, and CT was performed 35 sec after the start of the injection. For CT hepatic arteriography, 36 mL of contrast material was injected through the common hepatic artery at a rate of 1.8 mL/sec, and CT was performed 6 sec after the start of the injection. The scans were obtained in a craniocaudal plane during a single breath-hold helical acquisition of 20–30 sec depending on the liver size, using a 7-to 8-mm collimation, a 10-mm/sec table speed, and a 7-to 8-mm reconstruction interval. On the LightSpeed QX/i CT scanner, scans were obtained with a 5-mm collimation, an 18.75-mm/sec table speed, and a 5-mm reconstruction interval.
Follow-up two-phase CT of the liver was performed during a 6-to 15-month period (mean, 8.3 months) after the combined CT during arterial portography and CT hepatic arteriography. For the two-phase CT of the liver, unenhanced scans were initially obtained. Scanning for the hepatic arterial phase began 36 sec and scanning for the portal venous phase began 72 sec after injection of 150 mL of nonionic contrast material (iopromide, Ultravist 370) at a rate of 3 mL/sec through the patient's antecubital vein. The scans were obtained in a craniocaudal plane during a single breath-hold helical acquisition of 20–30 sec depending on the liver size, with a 7-to 8-mm collimation, a 10-mm/sec table speed, and a 7-to 8-mm reconstruction interval. On the LightSpeed QX/i CT scanner, scans were obtained with a 5-mm collimation, an 18.75-mm/sec table speed, and a 5-mm reconstruction interval.
Image Analysis
We first evaluated the diagnostic accuracy of our prospective
interpretations of the results of combined CT during arterial portography and
CT hepatic arteriography by comparing the radiologic interpretation with the
results of histopathology or follow-up imaging. We then obtained the
sensitivity, specificity, positive predictive value, and negative predictive
value of our prospective interpretation.
All CT during arterial portograms and CT hepatic arteriograms were then retrospectively reviewed by two radiologists in consensus. During the retrospective analysis, reviewers were unaware of the results of histopathology or follow-up imaging and of the prospective radiologic interpretation. They evaluated the imaging features of each lesion. All images were evaluated on a PACS (picture archiving and communication system) (Radpia, Hyundai Information Technology, Seoul, Korea). With this system, the window width, level, and magnification settings of all images can be freely controlled by the radiologists. The imaging features analyzed included the size, shape, and location of the lesion within the liver and the relative attenuation value of the lesion compared with normal liver parenchyma on CT during arterial portography and CT hepatic arteriography. Visualization of the peripheral portal vein branches within the lesion was also evaluated on CT hepatic arteriography. The lesion size was measured using the measurement tool in the PACS and was defined as the longest diameter of the lesion on the magnified view of the images. The shape of the lesion was classified as either nodular or wedge-shaped. The lesion location was classified as either subcapsular or intraparenchymal. A lesion was defined as subcapsular if it broadly abutted the capsular surface of the liver.
The relative attenuation value was classified as hypoattenuating, isoattenuating, or hyperattenuating compared with the surrounding liver parenchyma. Hypoattenuation was defined as a perfusion defect on CT during arterial portography. Isoattenuation was defined as an attenuation that was similar to that of the surrounding liver parenchyma. Hyperattenuation was defined as a focal enhancement on CT hepatic arteriography. Visualization of the peripheral portal vein branches within the lesion on CT hepatic arteriography was defined as visualization of either linear or branching structures with strong contrast enhancement within the lesion.
We evaluated the significant difference in size between hepatocellular carcinomas and pseudolesions using the Student's t test. To evaluate the statistically significant difference of each imaging feature between hepatocellular carcinomas and pseudolesions on CT during arterial portography and CT hepatic arteriography, we used the chisquare test or the Fisher's exact test, depending on the number of lesions. A p value of less than 0.05 was considered statistically significant.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Among the 149 hepatocellular carcinomas, our prospective interpretation of findings of combined CT during arterial portography and CT hepatic arteriography correctly diagnosed 147 (98.7%). Among the 104 pseudolesions, our prospective interpretation correctly diagnosed 94 (90.4%). We had 10 false-positive results and two falsenegative results for hepatocellular carcinoma. The sensitivity of our prospective interpretation was 98.7%, and our specificity was 90.4%. Our positive and negative predictive values were 93.6% and 97.9%, respectively.
The results of the retrospective review of the imaging features of each lesion detected on CT during arterial portography and CT hepatic arteriography are summarized in Table 1. The hepatocellular carcinomas were significantly larger than the pseudolesions. Most hepatocellular carcinomas had a nodular shape (Figs. 1A, 1B), whereas most pseudolesions were wedge-shaped (Figs. 2A, 2B, 2C). Most of the hepatocellular carcinomas were located intraparenchymally, whereas most of the pseudolesions were subcapsular.
|
|
|
|
|
|
The attenuation of hepatocellular carcinomas on CT during arterial portography and CT hepatic arteriography differed from that of pseudolesions. Hypoattenuation on CT during arterial portography and hyperattenuation on CT hepatic arteriography were more frequently seen in hepatocellular carcinomas than in pseudolesions. Moreover, the frequency with which the combination of hypoattenuation on CT during arterial portography and isoattenuation on CT hepatic arteriography was seen as statistically more significant in pseudolesions (Figs. 3A, 3B) than in hepatocellular carcinomas. We found no statistically significant difference between hepatocellular carcinomas and pseudolesions in other combinations of attenuation on CT during arterial portography and CT hepatic arteriography, possibly because of the small number of patients. The peripheral portal vein branches on CT hepatic arteriography were visualized in 36 pseudolesions (34.6%) (Fig. 2B); however, no hepatocellular carcinomas showed this finding. Among the 10 lesions with a false-positive result, seven were nodular, and eight were subcapsular (Figs. 4A, 4B, 4C). The two lesions with a false-negative result were wedge-shaped and subcapsular (Figs. 5A, 5B).
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Recently, the combination of CT during arterial portography and CT hepatic arteriography has begun to be used again for the preoperative staging of hepatocellular carcinoma. Although these techniques are sensitive for the detection of hepatocellular carcinoma, many pseudolesions are caused by perfusion abnormalities related to the structural changes in the cirrhotic liver. Several investigators have reported that the use of CT during arterial portography and CT hepatic arteriography has resulted in an unacceptably high rate of falsepositive findings; therefore, it has been recommended that CT during arterial portography and CT hepatic arteriography no longer be used for preoperative evaluation of patients with hepatocellular carcinoma [16, 17, 18]. Kondo et al. [16] reported that combining ferumoxides-enhanced MR imaging with unenhanced T1-weighted and T2-weighted imaging and gadolinium-enhanced triphasic dynamic MR imaging was comparable to combining CT during arterial portography and CT hepatic arteriography for the preoperative detection of malignant hepatic tumors. Moreover, a study by Choi et al. [17] found that ferumoxides-enhanced MR imaging produced the same diagnostic accuracy as combined CT during arterial portography and CT hepatic arteriography for the depiction of hepatocellular carcinomas and resulted in a higher specificity than the combined CT techniques. Zacherl et al. [23] found intraoperative sonography to be highly sensitive, even in patients with cirrhosis; the overall accuracy of intraoperative sonography was 95.5%.
Nonetheless, we believe that careful analysis of the imaging features of hepatic lesions detected on CT during arterial portography and CT hepatic arteriography often can lead to the correct diagnosis of hepatocellular carcinomas and pseudolesions. To our knowledge, a prospective analysis of the diagnostic accuracy of combined CT during arterial portography and CT hepatic arteriography for the preoperative evaluation of hepatocellular carcinoma with a detailed analysis of the differentiating features of pseudolesions has not been previously reported.
The possible mechanisms of pseudolesions include small nontumorous arterioportal shunts or an aberrant portal venous supply. Nontumorous arterioportal shunts in patients with cirrhosis is believed to result from the occlusion of the small hepatic venules and the retrograde filling of the small portal vein branches via arterioportal anastomoses [24]. Common locations of pseudolesions resulting from an aberrant portal venous supply include the pericholecystic area, the region anterior to the porta hepatis, and the region adjacent to the falciform ligament [25, 26]. Cross-sectional imaging features that are suggestive of pseudolesions have been described in several previous reports [11, 12, 13, 20]. These features include a lesion that is small and wedge-shaped, a lesion with a subcapsular location in the liver, and early opacification of portal venous branches within the lesion seen during the hepatic arterial phase.
In our study, we prospectively interpreted all hepatic focal lesions detected on CT during arterial portography and CT hepatic arteriography applying knowledge of these imaging features and obtained an acceptable diagnostic performance, with a sensitivity of 98.7%, a specificity of 90.4%, a positive predictive value of 93.6%, and a negative predictive value of 97.9%. Our results for the detection of hepatocellular carcinoma on CT during arterial portography and CT hepatic arteriography images are superior to results reported in previous studies [2, 3, 16, 18], perhaps because we interpreted the CT during arterial portograms and CT hepatic arteriograms while also considering the imaging findings described as typical of nontumorous pseudolesions in patients with cirrhosis.
From the retrospective analysis of our study, we found several imaging findings on CT during arterial portography and CT hepatic arteriography that might be helpful in differentiating pseudolesions from hepatocellular carcinomas. First, the size of a hepatocellular carcinoma was significantly larger than that of a pseudolesion (p < 0.01). Although our analysis showed considerable overlap between the size of hepatocellular carcinomas and the size of the pseudolesions, most pseudolesions (74%) did not exceed 3 cm in diameter. Second, most hepatocellular carcinomas had a nodular shape (91.9%) and intraparenchymal location (75.2%), whereas most pseudolesions had a wedge shape (77.9%) and subcapsular location (82.7%). Third, 36 pseudolesions (34.6%) showed enhancement of the peripheral portal vein branches within the lesion on CT hepatic arteriography, representing early opacification of small portal venous branches in the arterioportal shunt, whereas no hepatocellular carcinoma displayed this finding. We believe that detailed analysis of these imaging features for each lesion may help to differentiate pseudolesions from hepatocellular carcinomas.
In addition, hepatocellular carcinomas and pseudolesions showed a significant difference in attenuation on CT during arterial portography and CT hepatic arteriography. Most hepatocellular carcinomas (90.6%) showed hypoattenuation on CT during arterial portography and hyperattenuation on CT hepatic arteriography. Although most pseudolesions (72.1%) also showed hypoattenuation on CT during arterial portography and hyperattenuation on CT hepatic arteriography, 27 pseudolesions (26%) showed hypoattenuation on CT during arterial portography and isoattenuation on CT hepatic arteriography. Yamagami et al. [15] reported that 78% of the decreased areas of nontumorous portal perfusion were enhanced on CT hepatic arteriography, a finding that was similar to our results. One cause of isoattenuation of pseudolesions on CT hepatic arteriography might be the portal perfusion defect on CT during arterial portography resulting from the aberrant portal venous drainage from the cholecystic, gastric, or subcapsular veins and the normal hepatic arterial supply into the corresponding lesion during CT hepatic arteriography.
Our study has several limitations. First, our patient population included a select group of potential candidates for hepatic resection whose livers were relatively undamaged. This fact might have improved the diagnostic performance of combined CT during arterial portography and CT hepatic arteriography. Second, we excluded many patients who had lesions that were obscured as a result of chemoembolization, insufficient follow-up to determine the presence or absence of interval growth of the lesion, or inadequate quality of CT during arterial portograms and CT hepatic arteriograms resulting from the replaced hepatic artery. If the patients with inadequate CT during arterial portograms or CT hepatic arteriograms had been included in our analysis, the overall utility of our study would be lower. Third, we evaluated the follow-up CT scans obtained at least 6 months after combined CT during arterial portography and CT hepatic arteriography to assess the presence or absence of a pseudolesion. It might be argued that a 6-month interval is too short to confirm the diagnosis of a pseudolesion. However, the diagnostic criteria of the pseudolesion used in a study by Kim et al. [20] was that the lesion showed no increase in size for at least 4 months. Moreover, most patients with pseudolesions in our study had follow-up CT performed more than 6 months after CT during arterial portography and CT hepatic arteriography that showed no increase in size of the lesion.
In conclusion, use of combined CT during arterial portography and CT hepatic arteriography may be feasible in the preoperative evaluation of patients known or thought to have hepatocellular carcinoma. Understanding the typical imaging features of pseudolesions—their small size, wedge shape, subcapsular location, and isoattenuation on CT hepatic arteriography—may help to differentiate pseudolesions from hepatocellular carcinomas.
Acknowledgments
We thank Bonnie Hami, Department of Radiology, University Hospitals Health
System, Cleveland, OH, for editorial assistance in preparing the
manuscript.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  X. Ma, A. E. Samir, N.-S. Holalkere, and D. V. Sahani Optimal Arterial Phase Imaging for Detection of Hypervascular Hepatocellular Carcinoma Determined by Continuous Image Capture on 16-MDCT Am. J. Roentgenol., September 1, 2008; 191(3): 772 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ishigami, K. Yoshimitsu, H. Irie, T. Tajima, Y. Asayama, M. Hirakawa, and H. Honda Accessory Left Gastric Artery from Left Hepatic Artery Shown on MDCT and Conventional Angiography: Correlation with CT Hepatic Arteriography Am. J. Roentgenol., October 1, 2006; 187(4): 1002 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-J. Kim, J. Y. Choi, J. S. Lim, J. Y. Kim, J. H. Kim, Y. T. Oh, E. H. Yoo, J. J. Chung, and K. W. Kim Optimal scan window for detection of hypervascular hepatocellular carcinomas during MDCT examination. Am. J. Roentgenol., July 1, 2006; 187(1): 198 - 206. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
