| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Radiology and Institute of Radiation Medicine, Seoul National University College of Medicine, 28, Yongon-dong, Chongno-gu, Seoul 110-744, Korea.
Received February 12, 2005;
accepted after revision June 7, 2005.
Address correspondence to J. M. Lee
(leejm{at}radcom.snu.ac.kr).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. CT images of 30 patients with hepatolithiasis and pathologically proven cholangiocarcinoma (n = 14) or periductal fibrosis (n = 16) were retrospectively reviewed. Helical CT scans were obtained before, 30 seconds after, and 65 seconds after the start of contrast material injection. Analysis of CT findings included evaluation for the presence of periductal soft-tissue density, bile duct wall thickening at the stricture site, ascites, portal vein obliteration, lymph node enlargement, and a duct stone; assessment of the degree of ductal dilatation; and evaluation of the enhancement pattern of periductal lesions, thickened ductal wall, and hepatic parenchyma. The CT attenuation coefficients of the thickened ductal wall and adjacent normal-looking bile duct were measured on images obtained during each phase. Among these findings, statistically significant variables were determined using the Fisher's exact test and Student's t test. Sensitivity and specificity values of the CT criteria were also calculated.
RESULTS. The presence of periductal soft-tissue density (p = 0.002), higher enhancement of the duct than adjacent bile duct on portal venous phase images (p = 0.008), ductal wall thickening (p = 0.026), portal vein obliteration (p = 0.031), and lymph node enlargement (p = 0.031) were found to be the significant findings for differentiating cholangiocarcinoma from fibrosis in patients with hepatolithiasis. When any two or more of these five criteria were used in combination, we could identify 100% of the patients with cholangiocarcinoma but only 12.5% of the patients with fibrosis.
CONCLUSION. Cholangiocarcinoma in patients with hepatolithiasis can be diagnosed using specific CT criteria.
Keywords: biliary system • cholangiocarcinoma • CT • fibrosis • hepatolithiasis • liver
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The natural history of the disease is marked by recurrent attacks of cholangitis, usually once or twice a year [2, 3]. Because of the recalcitrant nature of the strictures and stones, the complications of the condition—that is, cholangitis, abscess, and jaundice—are relentless. Repeated surgery for recurrent cholangitis and abscess formation due to re-formed or residual stones results in substantial morbidity and mortality [3, 5]. The incidence of hepatolithiasis is known to be high throughout East Asia; hepatolithiasis is rare in Europe and the United States, but the recent pattern of increased population migration to the West has led to the increased prevalence and recognition of hepatolithiasis in Western countries [5].
According to previous studies, hepatolithiasis is known to be associated with cholangiocarcinoma, and the incidence rates of cholangiocarcinoma in patients with hepatolithiasis have been reported in the range of 5-6% [6-9]. Alternatively, the percentage of patients with cholangiocarcinoma with concomitant hepatolithiasis has been reported to be in the range of 17-27% [10-12]. Chronic irritation by intrahepatic calculi, bile stasis, and bacterial infection have been proposed as possible factors responsible for its development [6]. A careful search for the presence of cholangiocarcinoma is crucial in the treatment of patients with hepatolithiasis. However, because the clinical features of hepatolithiasis have been reported to be similar in patients with or without the coexistence of cholangiocarcinoma, preoperative imaging workup, especially with CT, is important for the selection of appropriate surgical therapy and treatment planning [6].
Typical CT findings of cholangiocarcinoma are parenchymal low-attenuated mass with rim enhancement, capsular retraction and delayed enhancement, biliary duct wall thickening, or intraductal polypoid mass with dilatation of up-stream ducts [13-16]. However, only a few reports describe the radiologic findings of cholangiocarcinoma occurring in patients with hepatolithiasis [17-19]. In patients with hepatolithiasis, fibrotic masses in the bile duct walls and periductal hepatic parenchyma are often seen as a form of periductal tumorlike lesions and may appear similar to cholangiocarcinoma on CT [19]. Furthermore, in patients with hepatolithiasis, the stricture or stenosis caused by cholangiocarcinoma may not be easy to differentiate from that caused by stones [6, 10, 20].
Despite the fact that differentiation between periductal fibrosis and cholangiocarcinoma in patients with hepatolithiasis is important because of their different prognoses, no report, to our knowledge, has compared the CT findings of cholangiocarcinoma occurring in the background of hepatolithiasis with those of fibrosis. The purpose of this study was to identify the useful CT findings for differentiating cholangiocarcinoma manifesting as periductal lesions from periductal fibrosis in patients with hepatolithiasis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Finally, 30 patients were included in our retrospective study—that is, 14 patients with cholangiocarcinoma (male-to-female ratio, 7:7; mean age, 59.6 years; age range, 45-74 years) and 16 patients with periductal fibrosis (male-to-female ratio, 9:7; mean age, 52.9 years; age range, 41-71 years). Cholangiocarcinoma was proven by percutaneous sonographically guided liver biopsy (n = 4), choledochoscopic biopsy (n = 5), or surgery (n = 3 [left lobectomy in two patients and left lateral segmentectomy in one]). Peritoneoscopic and open omental biopsy were performed on each of the remaining two patients, and they were proven to have metastatic adenocarcinoma clinically from the liver. All 16 patients with fibrosis underwent hepatic surgery: left lateral segmentectomy (n = 8), left lobectomy (n = 7), or right lobectomy (n = 1). For the limited and anonymous review of these patients' data for this study, we were not required to have formal approval or informed patient consent according to the institutional review board of our hospital.
CT Examination
The CT examinations evaluated for this study were performed on different
helical CT scanners (Somatom Plus 4, Siemens Medical Solutions; or HighSpeed,
GE Healthcare) using the following parameters: 5-mm collimation, 5-mm
reconstruction interval, and a 1:1 table pitch. Transverse images were
reconstructed with a soft-tissue algorithm. The X-ray tube voltage used was
120-140 kV, and the current varied between 240 and 260 mA. The CT images were
routinely obtained with the patient in a supine position during full
inspiration. Both unenhanced and contrast-enhanced CT scans were obtained.
With IV injection of 120 mL of nonionic contrast material (iopromide
[Ultravist 370, Schering]), both hepatic artery phase (HAP) and portal venous
phase (PVP) images were obtained with a scanning delay of 30 and 65 seconds,
respectively. Contrast material was administered at a rate of 3 mL/s using a
mechanical power injector (CT 9000 ADV Digital Injection System,
Liebel-Flarsheim) through an 18-gauge angiographic catheter inserted into a
forearm vein.
CT Analysis
CT scan data were available on a PACS, and all images were reviewed at a
PACS monitor. CT scans were reviewed retrospectively by two experienced
abdominal radiologists (15 and 7 years of experience) in consensus who were
blinded to the final pathology results. For the analysis of the CT features,
cases of cholangiocarcinomas and benign fibrotic masses were randomly
intermixed. For subjective analysis, the following CT findings were included:
periductal soft-tissue-density lesion and its enhancement pattern; ductal wall
thickening at the stricture site and, if present, its extent and its
enhancement pattern; degree of ductal dilatation proximal to the stricture
site; presence of intrahepatic duct or common bile duct stone; portal vein
patency; ascites; and lymph node enlargement. A periductal soft-tissue-density
lesion was defined as a low-attenuated lesion compared with adjacent liver
parenchyma along the involved bile duct on contrast-enhanced CT. Relative
enhancement was obtained by calculating the absolute value of attenuation of
the periductal soft-tissue density subtracted by that of adjacent hepatic
parenchyma. Ductal dilatation proximal to the stricture was defined as mild
(
15 mm in diameter) or marked (> 15 mm in diameter)
[21]. Portal vein status was
analyzed as one of the following three categories: intact; narrowed, but
patent; or obliterated. A lymph node was considered to be significantly
enlarged when the short diameter exceeded 1 cm or it had a nonenhancing,
low-attenuated portion on PVP images, thereby suggesting necrosis even if the
diameter was less than 1 cm.
To assess the degree of enhancement of the ductal wall at the stricture site, the attenuation of the ductal lesion and the normal bile duct wall were measured (in Hounsfield units [H]) by one of the authors using circular regions of interest (ROIs) on unenhanced, HAP, and PVP images. The ROI cursors were carefully placed to encompass as much of the ductal wall as possible and to avoid adjacent structures; the mean size of the ROI cursors was 10 mm2 (range, 6-15 mm2). At least three ROIs were placed on the ductal wall of the stricture site and on normal bile duct, and the ROI values were averaged as a mean ductal wall attenuation. Relative enhancement of the ductal wall was obtained by calculating the ROI value of the attenuation of the lesion subtracted from that of adjacent normal bile ductal wall. Two of the authors who were not involved in the review process measured the thickness of the ductal wall at the stricture portion on a maximally magnified CT image on the PACS monitor at our institute with electronic calipers.
Statistical Analysis
The Kolmogorov-Smirnov test was used to determine whether continuous
variables followed a normal distribution. Sensitivity and specificity analyses
were performed for a range of ductal wall thicknesses and relative ductal wall
enhancement values to generate a receiver operating characteristic (ROC) curve
[22] and to determine the
optimal cutoff of ductal wall thickness and relative ductal enhancement for
the detection of cholangiocarcinoma in patients with hepatolithiasis. The
optimal cutoff point was defined as the value at which the sum of the
sensitivity and the specificity was maximized. From the subjective analysis
regarding the imaging findings of the cholangiocarcinoma and fibrosis groups,
statistically significant variables were determined using the Fisher's exact
test and Student's t test. A p value of less than 0.05 was
required for rejection of the null hypothesis. Also, the sensitivity and
specificity values of each of the CT criteria were calculated. Numbers used in
this statistical analysis were the numbers of lesions. For statistical
analysis, we used SPSS for Microsoft Windows (version 10.0, SPSS) and MedCalc
for Windows (version 8.0.0.1, MedCalc).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
Table 2 shows the different imaging features observed in cholangiocarcinoma and fibrosis in patients with hepatolithiasis. On CT scans, periductal soft-tissue-density lesions were identified in 10 patients with cholangiocarcinoma (71.4%) and in two patients with periductal fibrosis (12.5%) (p = 0.002, Fisher's exact test) (Figs. 3A and 3B). As for the enhancement pattern of periductal soft-tissue-density lesions of cholangiocarcinoma, all 10 cases showed low attenuation in both the HAP and the PVP and they showed progressive enhancement from the HAP to the PVP. The average value of relative enhancement on the HAP and PVP images was 24.6 ± 14.9 H and 45.9 ± 19.2 H, respectively, which indicates that the lesions are seen more conspicuously on the PVP images than on the HAP images. Two cases of periductal soft-tissue-density lesions in patients with periductal fibrosis also showed low attenuation in both the HAP and PVP and progressive enhancement from the HAP to the PVP. The mean relative enhancement was 13.5 H in the HAP and 44.5 H in the PVP.
|
|
|
When we used 10 H as a cutoff value indicating high ductal enhancement and 2.8-mm thickness indicating ductal wall thickening, these parameters were significant in differentiating cholangiocarcinoma from benign fibrosis (Figs. 4A, 4B, 4C, 5A, 5B, 5C, 5D, 6A, 6B, and 6C). Portal vein obliteration (p = 0.031) (Figs. 3A and 3B) and lymph node enlargement (p = 0.031) (Figs. 4A, 4B, and 4C) were also significant parameters, and there was no necrotic lymph node in our study population. On the contrary, the degree of ductal dilatation, the presence of intrahepatic or common duct stones, and the presence of ascites were statistically insignificant in differentiating cholangiocarcinoma from benign fibrosis (p > 0.05).
|
|
|
|
|
|
|
|
|
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our hospital, which is an academic center caring for a large number of patients with cancer and receives many tertiary referrals, multiphasic helical CT, including unenhanced and contrast-enhanced HAP and PVP images, is used as a main imaging study for the evaluation of patients with suspected hepatolithiasis and intrahepatic stones. Based on our experience, the challenge for radiologists interpreting these CT images is to differentiate cholangiocarcinoma from benign fibrotic lesions. Indeed, there have been a limited number of reports regarding the radiologic findings of cholangiocarcinoma complicating hepatolithiasis [7, 18], but there are as yet no reports published in the English-language literature. To our knowledge, no reports describing the differential points of cholangiocarcinoma that distinguish it from fibrosis in patients with hepatolithiasis have been published.
In our study, which determined the CT findings that are useful for differentiating benign stricture caused by stones from concomitant malignant strictures in patients with hepatolithiasis, several CT findings suggesting concomitant cholangiocarcinoma in patients with hepatolithiasis were found by univariate analysis including periductal soft-tissue density, ductal wall thickening, portal vein obliteration, lymph node enlargement, and a high degree enhancement of ductal wall in the PVP (p < 0.05). When at least two of these five signs were present, correct diagnosis of cholangiocarcinoma could be made with a specificity of 87.5%, and when at least four signs were present, diagnosis of cholangiocarcinoma could be made with a specificity of 100% (Table 4). We believe that the results of our study are encouraging, although further research with a larger series of patients is required to test whether the findings we have noted can be successfully applied to similar patient populations.
In this study, concomitant cholangiocarcinoma in patients with hepatolithiasis showed progressive enhancement from the HAP to PVP and appeared more conspicuous in the PVP than the HAP. This finding agrees with a previous study regarding cholangiocarcinoma in which periductal infiltrating intrahepatic cholangiocarcinoma appeared as a periductal soft-tissue density or a focal ductal wall thickening on enhanced CT, usually showing early or late enhancement or both [13, 16]. There were two cases of benign biliary stricture (12.5%, 2/16), which appeared as periductal low-attenuated soft-tissue-density lesions similar to malignant strictures in their enhancement pattern and were pathologically proven to be periductal fibrosis. Given that differentiating a periductal low-attenuated lesion of cholangiocarcinoma from that of benign fibrosis was not easy and that the incidence of periductal lesions of benign fibrosis is considerably lower than that of malignant stricture (71.4%, 10/14), the differential diagnosis of cholangiocarcinoma should be included when there is a periductal soft-tissue-density lesion on helical CT in patients with hepatolithiasis.
In an analysis of the ductal wall thickness, the cholangiocarcinoma group had a significantly thicker ductal wall than the fibrosis group (p = 0.026) and cholangiocarcinoma also showed higher enhancement than normal duct wall on the PVP (p = 0.008). There are a few reports regarding the usefulness of delayed phase imaging in the evaluation of cholangiocarcinoma and fibrosis [16, 26, 27]. Some authors have suggested that delayed postequilibrium phase tumor contrast enhancement is a typical feature of intrahepatic cholangiocarcinoma [16, 26] that may be attributed to the inherent dense fibrous stroma of the tumor. In addition, Keogan et al. [27] reported that delayed CT images were helpful for tumor characterization and improved tumor detection in the evaluation of hilar or intrahepatic cholangiocarcinoma and that the optimal time for acquisition of delayed images is 10-20 minutes after contrast medium injection. However, according to a study by Min et al. [7], a typical enhancement pattern of intrahepatic cholangiocarcinoma associated with hepatolithiasis on CT was progressive ductal wall enhancement in the arterial through the portal phases with decreased enhancement in the delayed phase.
Because in this study we did not acquire delayed CT images 10-20 minutes after contrast medium injection, we do not know the exact value of the delayed phase images for the differentiation of benign strictures from concomitant cholangiocarcinoma with hepatolithiasis. Considering the histologic feature of the abundant fibrotic component in both benign and malignant biliary strictures, cholangiocarcinoma containing a substantial fibrous component may be difficult to differentiate from fibrosis even on the delayed phase images [28, 29]. Further study will be necessary to determine the value of delayed phase imaging for differentiating benign from malignant biliary strictures.
Our study results indicate that the presence of portal vein obliteration was significant in differentiating cholangiocarcinoma from benign fibrotic stricture (p = 0.031). According to previous studies regarding hepatolithiasis [30, 31], the degree of portal vein narrowing correlated with the severity of liver atrophy and the pruned-tree appearance of portal veins reflected slight to moderate liver atrophy. Combined malignancy must be suspected, therefore, when there is complete portal vein obstruction associated with or without segmental lobar atrophy of the liver in hepatolithiasis.
Our study had several limitations. First, because this study was based on retrospective design, the precise correlation of the CT features with the histopathologic findings was not possible. Second, because our routine CT protocol for biliary diseases did not include delayed images—that is, longer than a 15-minute delay—that could be helpful for the detection of cholangiocarcinoma [23], we were not able to assess the value of delayed enhancement of the lesions for differentiation between cholangiocarcinoma and fibrosis. Third, as to the analysis of lesion enhancement, we did not take account of the severity of the cholangitis when the CT scans were obtained and this may well have influenced the results of this study. Fourth, although we obtained multiphasic CT images using 5-mm collimation and 5-mm reconstructions, the scanners that were used are out of date; thus, the CT technique was not state-of-the-art. Last, only 3 of the 14 patients with cholangxiocarcinoma underwent hepatic surgery; the others did not undergo surgery because of the advanced disease stage, expected inadequate hepatic reserve after resection, or medical comorbidities. This indicates that the study population included many advanced cases, and the results may therefore be optimistically biased. However, this indirectly indicates that preoperative detection of cholangiocarcinoma in underlying hepatolithiasis is not easy and explains why our study is necessary. Despite the aforementioned problem of selection bias, we believe that our study results may help radiologists to differentiate the two diseases in patients with hepatolithiasis.
In summary, certain CT findings are helpful in detecting cholangiocarcinoma in patients with hepatolithiasis, and the combination of these CT findings may further improve accuracy in making the diagnosis.
Acknowledgments
We thank Bonnie Hami for her editorial assistance in the preparation of
this manuscript.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  Y. E. Chung, M.-J. Kim, Y. N. Park, J.-Y. Choi, J. Y. Pyo, Y. C. Kim, H. J. Cho, K. A. Kim, and S. Y. Choi Varying Appearances of Cholangiocarcinoma: Radiologic-Pathologic Correlation RadioGraphics, May 1, 2009; 29(3): 683 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W. Anderson, E. Rho, and J. A. Soto Detection of Biliary Duct Narrowing and Choledocholithiasis: Accuracy of Portal Venous Phase Multidetector CT Radiology, May 1, 2008; 247(2): 418 - 427. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
