| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology and Nuclear Medicine, Kyoto University School of
Medicine, Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan.
2
Department of Pathology, Kyoto University Hospital, Kyoto, 606-8507,
Japan.
Received July 12, 2000;
accepted after revision November 13, 2000.
Address correspondence to Y. Maetani.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. MR images of 50 patients with pathologically proven intrahepatic cholangiocarcinoma were reviewed retrospectively. T1- and T2-weighted spin-echo images were obtained in all patients. Contrast-enhanced T1-weighted imaging was performed in 25 patients. Signal intensity and enhancement pattern of the tumors were correlated with pathology findings. The frequency of central hypointense regions on T2-weighted images and the intrahepatic bile duct dilatation of several other hepatic tumor types were investigated. Results were compared with imaging results of cholangiocarcinoma.
RESULTS. On T2-weighted images, central hypo- and hyperintense regions were detected in tumors in 27 and 17 patients, respectively. Contrast-enhanced T1-weighted imaging revealed central hypointense areas exhibiting homogeneous, heterogeneous, and no enhancement in six, three, and five, respectively, of 14 patients. Regions of fibrosis displayed enhancement, whereas those of coagulative necrosis showed no enhancement. The signal intensity difference on T2-weighted images between the center and the edge of the tumor correlated well with the fibrotic ratio difference between those two areas corresponding to the MR image (Spearman's rank correlation test, r = 0.72, 95% confidence interval = 0.48-0.86). T2-weighted images revealed central hypointense regions in 16 of 34 instances of hepatic colorectal metastases. However, hypointensity was observed in only 26 of 234 other hepatic tumors. Intrahepatic bile duct dilatation was evident in 27 of 50 cases of cholangiocarcinoma but occurred in only a single case of 34 instances of hepatic colorectal metastases.
CONCLUSION. The combination of the signal intensity on T2-weighted images and the enhancement pattern on contrast-enhanced T1-weighted images showed good correlation with the pathologic findings of cholangiocarcinoma. The occurrence of a central hypointense area on T2-weighted images is not pathognomonic; however, this finding, which reflects severe fibrosis, appears to be a characteristic marker of intrahepatic cholangiocarcinoma. The presence of intrahepatic bile duct dilatation may indicate cholangiocarcinoma, although it is difficult to differentiate cholangiocarcinoma from hepatic colorectal metastasis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Pathologic Review
The maximum diameter of the primary tumor ranged from 2.6 to 21 cm (mean,
7.1 cm). Partial hepatectomy was performed in 43 patients, and wedge resection
or intraoperative biopsy was performed in seven patients in whom unresectable
tumor was indicated at surgery. A pathologist with no knowledge of the MR
imaging findings reviewed the histologic specimens retrospectively. The degree
of fibrosis, coagulative necrosis, cell debris, and mucin accumulation in the
lesion were semiquantitatively graded as none (0% of tumor area), little
(<10% of tumor area), moderate (
10 to <50% of tumor area), or severe
(50% of tumor area).
MR Imaging
Because this was a retrospective investigation dating back to 1988, a
variety of pulse sequences were used. In brief, all examinations were
performed with a 1.5-T superconducting unit, and T1-weighted spin-echo images
and T2-weighted conventional (n = 41) or fast spin-echo images
(n = 9) were obtained in all patients. Dynamic MR imaging was
performed in 13 patients, and contrast-enhanced T1-weighted imaging was
performed in 25 patients.
From January 1988 through November 1996, MR imaging was performed on a 1.5-T superconducting unit (Signa; General Electric Medical Systems, Milwaukee, WI) in 41 patients. Axial T1-weighted spin-echo MR images were produced with a TR/TE range of 600/20-25 msec. T2-weighted spin-echo images were obtained with a TR/TE range of 2000/60-80. Multiple sections were simultaneously obtained in all patients. Section thickness was 5-8 mm with a gap of 2-5 mm. The number of signals averaged was 4 for the T1-weighted images and 2 for the T2-weighted images. The image matrix was 256 x 192-256, and the field of view was 26-32 cm. Respiratory compensation and flow compensation were used in 27 patients. Dynamic MR images were obtained in five patients using a gradient-echo sequence with 100/13, a 90° flip angle, a 256 x 128 matrix, and 1 excitation. The single-slice technique was used, and the appropriate axial slice level across the largest diameter of the main tumor was determined on the basis of T1- or T2-weighted spin-echo images. After the injection of a bolus of 0.1 mmoL/kg of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) at a rate of 1-2 mL/sec, dynamic MR images were obtained immediately and then every 1 min for 5 min. Contrast-enhanced T1-weighted images were obtained 6-20 min after the administration of gadopentetate dimeglumine in 16 patients, including five patients in whom dynamic studies had been conducted. Planes and parameters were identical to those used for the unenhanced imaging.
From December 1996 to December 1998, MR imaging was performed on a 1.5-T superconducting unit (Signa Horizon; General Electric Medical Systems) in nine patients. All MR images were obtained in the axial plane with a phased array multicoil for the body. T1-weighted spin-echo images were obtained with the following parameters: 500/9, a 512 x 224 matrix, 2 acquisitions, and a field of view of 32 x 32 cm. T2-weighted images were obtained with a fat-suppressed respiratory-triggered fast spin-echo sequence with a TR range/effective TE range of 3529-7500/104-130, a 512 x 256 matrix, 2 acquisitions, a field of view of 32 x 24 cm, an echo train length of 10, and gradient moment nulling in the frequency-encoding direction. The chemical shift—selective fat-suppression technique was used for T2-weighted images. Section thickness was 7-8 mm with a 2-3 mm intersection gap. Contrast-enhanced T1-weighted images, including dynamic MR images, were obtained in all nine patients. In four patients, the enhanced three-dimensional Fourier transform fast gradient-echo sequence was performed with the following imaging parameters; 6.2/1.4, a flip angle of 30°, a 256 x 160 matrix, 1 acquisition, a field of view of 32 x 24 cm, a slab thickness of 96-128 mm, and 12-16 partitions. The spectrally selected inversion recovery pulse was used as a fat-suppression technique, and the inversion time was 50 msec. In five patients, fast multiplanar spoiled gradient-recalled imaging was performed with the following imaging parameters: TR range/TE, 150-180/1.6 or 4.2, a flip angle of 60°, a 256 x 160 matrix, 1 acquisition, a field of view of 32 x 24 cm, and a section thickness of 7-8 mm with an intersection gap of 2-3 mm. In three patients examined with a TE of 1.6 msec, the fat-suppression technique was used. Saturation pulses superior and inferior to the imaging volume were used for all images. After a set of unenhanced MR images was obtained, a bolus of 0.1 mmoL/kg of gadopentetate dimeglumine was injected at a rate of 1-2 mL/sec followed by a 20-mL flush of normal saline by hand or using a power injector. The scan delays for triphasic dynamic imaging were 20, 80, and 180 sec after the initiation of contrast injection. Contrast-enhanced T1-weighted spin-echo imaging was conducted 5-10 min after injection.
Image Analysis
The MR images were retrospectively evaluated in conference by two
radiologists with knowledge of the diagnosis of cholangiocarcinoma. However,
the histologic findings were not disclosed. In five patients showing
intrahepatic metastases, the primary tumor was chosen exclusively for
analysis. The radiologists reached a consensus regarding the following
features: lesion signal intensity compared with that of normal liver, signal
homogeneity, the presence of a central area or band, the lesion enhancement
pattern after contrast material injection, and dilatation of the intrahepatic
bile duct. Because of the retrospective nature of this study, tumor signal
intensity was assessed in relation to normal liver and was scored on a
seven-point scale (1 = very high signal intensity, 2 = hyperintense relative
to normal liver, 3 = slightly hyperintense relative to normal liver, 4 =
isointense relative to normal liver or not seen, 5 = slightly hypointense
relative to normal liver, 6 = hypointense relative to normal liver, and 7 =
very low signal intensity). Homogeneity was defined as uniform signal
intensity on T1-, T2-, and contrast-enhanced T1-weighted images. In the event
of a central region of low or high signal intensity relative to the tumoral
edge on a T2-weighted image, the enhancement pattern of the area on
contrast-enhanced T1-weighted images was investigated. In the 13 patients who
underwent dynamic MR imaging, the pattern of enhancement was also
analyzed.
Correlation of MR Imaging and Histologic Findings
MR imaging results and histologic appearance were compared by the
radiologists and pathologist in concert. Histologic specimens taken from the
tumoral edge were available for 43 patients who had undergone partial
hepatectomy; however, precise localization of the tissue block to the imaged
tumor was impossible in 10 cases. Furthermore, the tissue block corresponding
to the central portion of the tumor was not available in these 10 cases and in
three additional cases because of the retrospective nature of this study.
Consequently, it was possible to strictly correlate MR imaging and histologic
findings in 30 of 50 patients.
The degree of fibrosis was evaluated as follows. Initially, histologic specimens stained by the Masson trichrome method were downloaded from an electron microscope (high-vision camera [HQ-130Ci]; Nikon, Tokyo, Japan) onto a computer (Power Macintosh 8500/150; Apple Computer, Cupertino, CA). Staining appeared in blue. The stained areas, which reflected fibrosis, were chosen using Photoshop 5.0 software (Adobe Systems, San Jose, CA). Subsequently, the area was calculated by means of National Institutes of Health Image 1.61 software [20] (written by Wayne Rasband at the National Institutes of Health, Bethesda, MD). The fibrotic ratio of the lesion was defined as follows: Fibrotic ratio (%) = (blue area / entire area input) x 100.
Two separate regions were chosen for analysis: the tumoral edge and the central hypo- or hyperintense area revealed on T2-weighted images. In cases involving homogeneous intensity on T2-weighted images, the extreme central portion was chosen. Subsequently, the signal intensity difference on T2-weighted images between the center and the edge of the tumor was correlated with the fibrotic ratio difference between the two areas corresponding to the MR image (Fig. 1A,1B,1C,1D,1E,1F,1G). Statistical analysis was performed using Spearman's rank correlation test. The correlation of the enhancement pattern and the histologic finding of the area was investigated when a central hypo- or hyperintense region occurred on a T2-weighted image.
|
|
|
|
|
|
|
MR Imaging Features of Several Hepatic Tumor Types
In addition, we attempted to investigate the frequency of occurrence of
central hypointense areas on T2-weighted images of several hepatic tumor
types. From January 1999 to December 1999, 423 patients presenting with
several hepatic tumor types underwent MR imaging examination in our hospital.
T2-weighted fast spin-echo images (parameters described previously) were
obtained for 274 proven, untreated hepatic tumors (144 instances of
hepatocellular carcinoma, 88 instances of metastasis, 32 instances of
hemangioma, six cases of intrahepatic cholangiocarcinoma, three cases of focal
nodular hyperplasia, and one case of angiomyolipoma) in 149 patients (102
males and 47 females; age range, 16-84 years). The frequencies of central
hypointense regions on T2-weighted images and of intrahepatic bile duct
dilatation were evaluated retrospectively for all pertinent tumor types except
cholangiocarcinoma. Data were examined in conference by two radiologists with
no knowledge of the diagnoses.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR Imaging
The signal intensity of the tumoral edge was predominantly low on
T1-weighted images in most cases. Signal intensity was high on T2-weighted
images in all but two cases (Table
1). Three lesions exhibited high signal intensity on T1-weighted
images, and the same areas were also of high signal intensity on T2-weighted
images. Signal intensity was homogeneous in 33 of 50 lesions on T1-weighted
images. In contrast, signal intensity was homogeneous in only 12 lesions on
T2-weighted images. Of the 50 patients, 27 malignancies showed central areas
that were of lower signal intensity than the higher intensity of the tumoral
edge on T2-weighted images. These regions varied from small central foci to
large areas bounded by a thin hyperintense rim (Figs.
1A,1B,1C,1D,1E,1F,1G,2A,2B,2C,2D,2E,2F,3A,3B,3C,3D).
Some tumors were sharply defined; others were not. No region was linear or
stellate in shape. In contrast, focal areas of high signal intensity on
T2-weighted images occurred in 17 of the 50 lesions (Figs.
2A,2B,2C,2D,2E,2F
and
4A,4B,4C,4D).
Eight tumors displayed areas of both high and low signal intensity on
T2-weighted images. Additionally, high-signal-intensity regions within the
tumor were surrounded by low-signal-intensity areas (Fig.
2A,2B,2C,2D,2E,2F)
in three cases.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
On dynamic MR images, nine tumors showed initial rim enhancement characterized by progressive and concentric filling with contrast material (Fig. 1A,1B,1C,1D,1E,1F,1G). However, complete filling of the tumor with the contrast agent was observed in only a single case. Furthermore, the central regions of the tumor displayed heterogeneous enhancement in the delayed phase of contrast-enhanced T1-weighted images in eight patients. In three patients, rim enhancement was observed without concentric filling. In one patient, whole, albeit gradual, heterogeneous enhancement of the tumor was evident. Evaluation of enhancement in the remaining patient proved impossible. Segmental enhancement of liver parenchyma in the arterial phase resulting from the portal vein occlusion caused by tumor invasion prevented evaluation in this instance. No early phase hypervascularity was observed. However, of 17 patients who underwent dynamic CT in lieu of dynamic MR imaging, one malignancy displayed strong enhancement in the early phase (Fig. 2A,2B,2C,2D,2E,2F).
On contrast-enhanced T1-weighted images, all but one tumor exhibited heterogeneous enhancement. The central hypointense areas on T2-weighted images typically revealed heterogeneous enhancement; however, the primary enhancement pattern was homogeneous (Fig. 1A,1B,1C,1D,1E,1F,1G), heterogeneous, and slightly enhanced (Fig. 3A,3B,3C,3D) in six, three, and five, respectively, of 14 cases. On the other hand, central areas of high signal intensity on T2-weighted images showed no enhancement (Fig. 4A,4B,4C,4D) in 10 of 11 cases (Table 2).
|
Dilatation of intrahepatic bile ducts was observed in 27 of 50 cases.
Correlation of MR Imaging and Histologic Findings
We found that the signal intensity difference on T2-weighted images between
the edge and the center of the tumor correlated well with the fibrotic ratio
difference between the two areas corresponding to the MR image
(Fig. 5). This finding
indicated that, with respect to the central portion of the tumor, the stronger
the fibrotic change, the lower the signal intensity on T2-weighted images
(Fig.
1A,1B,1C,1D,1E,1F,1G).
However, this relationship did not hold true in all cases. Some tumors
displaying little fibrotic change showed central hypointense areas, whereas
other malignancies showed central hyperintense areas despite fibrotic change.
Histologic specimens corresponding to the central low-signal-intensity areas
were available for review in 11 of the 14 patients in whom contrast-enhanced
T1-weighted imaging was performed. In regions showing homogeneous enhancement,
fibrosis was predominant histologically (Figs.
1A,1B,1C,1D,1E,1F,1G
and
2A,2B,2C,2D,2E,2F).
In regions revealing little enhancement, coagulative necrosis was prominent
(Fig.
3A,3B,3C,3D).
Thus, areas of coagulative necrosis showed no enhancement, whereas areas of
fibrosis typically displayed enhancement. However, enhancement was not
entirely uniform. Histologic specimens corresponding to the central
high-signal-intensity areas were also available for review in nine of the 11
patients in whom contrast-enhanced T1-weighted imaging was performed.
Pathologically, these areas corresponded to coagulative necrosis and cell
debris (Figs.
2A,2B,2C,2D,2E,2F
and
4A,4B,4C,4D).
|
In a single case of mucinous carcinoma, signal intensity was very high and homogeneous on T2-weighted images. In contrast, signal intensity was low on T1-weighted images. On contrast-enhanced T1-weighted images, no enhancement was evident except within the peripheral portion of the tumor.
The correlation between MR imaging and histologic findings is summarized in Table 3.
|
MR Imaging Features of Several Hepatic Tumor Types
The frequency of central hypointense areas occurring on T2-weighted images
of several hepatic tumor types was reviewed retrospectively. Furthermore, the
likelihood ratio for cholangiocarcinoma versus several other hepatic tumor
types was investigated (Table
4). The central hypointense areas on T2-weighted images were not
uncommon in cases of liver tumors derived from colorectal cancer; however, it
was observed in only 26 of the 234 cases exhibiting other hepatic tumors.
|
Intrahepatic bile duct dilatation occurred in a single instance among 34 cases of hepatic colorectal metastasis. Additionally, dilatation was evident in only two instances among 144 cases of hepatocellular carcinoma. Dilatation was not observed in other hepatic tumors.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Internal desmoplastic change was usually intermingled with various degrees of coagulative necrosis. In addition, lesion signal intensity was influenced by the coexisting necrosis. Outwater et al. [23] conducted a study involving 16 cases of hepatic colorectal metastatic tumors exhibiting prevalent coagulative necrosis. Of the 16 cases, nine lesions were of low signal intensity, whereas seven tumors displayed high signal intensity on T2-weighted images. This observation indicated that coagulative necrosis shows both low and high signal intensity on T2-weighted images. Consequently, despite a large amount of fibrosis, no central hypointense area is observed because of the display of high signal intensity by the coexisting coagulative necrosis. In contrast, a tumor showing a low-signal-intensity central region on T2-weighted images does not necessarily consist of dense fibrosis. However, it may well show coagulative necrosis.
Therefore, to differentiate fibrosis from coagulative necrosis, it is necessary to obtain contrast-enhanced T1-weighted images. Central aspects of hepatic tumors consisting of massive coagulative necrosis have been known to show no enhancement in the delayed phase of contrast-enhanced T1-weighted imaging [11, 24]. However, some hepatic tumors exhibiting intratumoral fibrous stroma display marked delayed or prolonged enhancement on the delayed phase of contrast-enhanced T1-weighted imaging [7, 8] as well as on CT [25,26,27]. We found, however, that regions of moderate to severe fibrosis did not show obvious enhancement on contrast-enhanced T1-weighted images in some cases. We surmise that scanning delay times may influence the enhancement pattern. In our study, contrast-enhanced T1-weighted imaging was performed 5-20 min after the administration of gadopentetate dimeglumine. Gabata et al. [24] emphasized the importance of ultradelayed imaging performed 1-4 hr after the administration of gadopentetate dimeglumine. Those researchers determined that, as a result of the higher contrast resolution inherent to MR imaging, ultradelayed imaging, rather than delayed (5-10 min) imaging, was useful in the delineation of fibrous stroma and in the detection of coagulative necrosis. In addition, some hypervascular cholangiocarcinomas, characterized by very little fibrosis, showed prolonged enhancement in the late phase (2-3 min after initiating contrast injection) on dynamic imaging [15]. Consequently, it may prove beneficial to evaluate the pattern of delayed enhancement of hepatic tumors during the ultradelayed phase in order to investigate the different tissue components present in these malignancies.
A unique subtype of intrahepatic cholangiocarcinoma has been described by Nakajima et al. [21], who performed histopathologic studies involving 102 cases of intrahepatic cholangiocarcinoma. Nine histologic types emerged. A single case was classified as mucinous carcinoma because of the presence of small nests of tumor cell suspensions occurring in a large mucinous lake. On MR imaging, signal intensity is influenced by the large amount of mucin, which was extremely hypo- and hyperintense on T1- and T2-weighted images, respectively. Hayashi et al. [12] reported two cases of mucinous-type cholangiocarcinoma with MR imaging findings similar to those in our patient.
Consequently, as shown in Table 3, when the signal intensity on T2-weighted images and the enhancement pattern on contrast-enhanced T1-weighted images are combined, the pathologic findings of intrahepatic cholangiocarcinoma may be predicted.
We found central hypointense regions on T2-weighted images in 27 of 50 intrahepatic cholangiocarcinomas. We submit that the finding, which reflects severe fibrosis, appears to be a characteristic marker of cholangiocarcinoma. It is true, however, that central hypointense regions on T2-weighted images are evident in several other hepatic tumors [16]. We believe that dynamic MR imaging may provide additional information. That is, intrahepatic cholangiocarcinoma may be distinguished from other hepatic tumors on the basis of different enhancement patterns [28,29,30].
Dynamic MR imaging findings of intrahepatic cholangiocarcinoma have been well documented [7,8,9, 11, 13]. In our study and in others, cholangiocarcinoma typically revealed minimal or moderate initial rim enhancement, followed by progressive and concentric filling with contrast material. Several cases of cholangiocarcinoma exhibiting little fibrosis have shown early enhancement on dynamic studies [11, 14, 15]. We also found a single tumor displaying abundant tumor cells and a small degree of fibrosis that revealed early enhancement on dynamic CT.
Differentiating intrahepatic cholangiocarcinoma from metastatic adenocarcinoma is difficult despite histopathologic evidence. Consequently, reasonable exclusion of an extrahepatic primary tumor is necessary [5]. Central hypointense regions on T2-weighted images were observed in 27 of 50 patients with intrahepatic cholangiocarcinoma. In contrast, Outwater et al. [23] described central low-signal-intensity changes on T2-weighted images in 77 lesions among a sample of 154 patients with colorectal metastases. This observation is not surprising considering the histologic similarity of these two entities. Therefore, this finding may be presumed to be an MR imaging feature characteristic of cholangiocarcinoma and hepatic colorectal metastasis.
Hamrick-Turner et al. [7] and Fan et al. [8] reported that the presence of a central scar on MR images may indicate intrahepatic cholangiocarcinoma because a central scar is usually not observed in metastatic liver tumors [16]. However, those researchers did not clarify the pathologic findings of what they considered a central scar based on MR findings. In contrast, Vilgrain et al. [13] reported that central hypointense bands or areas on T2-weighted images of cholangiocarcinomas did not correspond to a central scar at gross pathologic examination. Therefore, this finding does not necessarily indicate the presence of a central scar. We believe that central low-signal-intensity regions may not be reliable markers for distinguishing these two malignancies on MR imaging.
Ascertaining the presence of intrahepatic bile duct dilatation may help to differentiate intrahepatic cholangiocarcinoma from hepatic colorectal metastasis. Intrahepatic bile duct dilatation was observed in 27 of 50 patients with cholangiocarcinoma. However, in our study, it appeared in one of 34 cases of hepatic colorectal metastases. Furthermore, Jinzaki et al. [31] reported that among 90 cases of surgically resected liver metastases, only five cases of biliary dilatation were seen on imaging, although metastatic liver tumors are reported to invade bile ducts frequently at pathology examination [32].
In conclusion, we believe that our study, which consisted of a relatively large sample size, revealed several patterns and combinations of MR imaging features that correlate well with the pathologic characteristics of intrahepatic cholangiocarcinoma. Although a central hypointense area on T2-weighted images is not pathognomonic, this finding that reflects severe fibrosis appears to be a characteristic marker of intrahepatic cholangiocarcinoma. Additionally, when a distinctive dynamic enhancement pattern is observed—that is, one exhibiting minimal or moderate initial rim enhancement followed by progressive and concentric filling with the contrast agent—MR imaging may permit the radiologist to considerably narrow the scope of the diagnosis. The differentiation of intrahepatic cholangiocarcinoma from hepatic colorectal metastasis is challenging. However, the presence of intrahepatic bile duct dilatation may indicate intrahepatic cholangiocarcinoma.
Acknowledgments
We thank Tomoko Okuno for her advice and help in the computer analysis of
the histopathologic findings.
|
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 Correlation1 RadioGraphics, May 1, 2009; 29(3): 683 - 700. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Colagrande, M. L. Paolucci, L. Messerini, W. Schima, A. Stadler, T. V. Bartolotta, A. Vanzulli, and G. Brancatelli Solitary Necrotic Nodules of the Liver: Cross-Sectional Imaging Findings and Follow-Up in Nine Patients Am. J. Roentgenol., October 1, 2008; 191(4): 1122 - 1128. [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]
|
|
|
|
|
|
Y. Asayama, K. Yoshimitsu, H. Irie, T. Tajima, A. Nishie, M. Hirakawa, T. Nakayama, D. Kakihara, A. Taketomi, S.-i. Aishima, et al. Delayed-Phase Dynamic CT Enhancement as a Prognostic Factor for Mass-forming Intrahepatic Cholangiocarcinoma Radiology, December 1, 2005; 238(1): 150 - 155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Elsayes, V. R. Narra, Y. Yin, G. Mukundan, M. Lammle, and J. J. Brown Focal Hepatic Lesions: Diagnostic Value of Enhancement Pattern Approach with Contrast-enhanced 3D Gradient-Echo MR Imaging RadioGraphics, September 1, 2005; 25(5): 1299 - 1320. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
