(Radiology. 1999;212:866-875.)
© RSNA, 1999
Hepatocellular Carcinoma and Intrahepatic Peripheral Cholangiocarcinoma: Enhancement Patterns with Quadruple Phase Helical CT-A Comparative Study1
Evelyne M. Loyer, MD,
Hsiao Chin, MD 2,
Ronelle A. DuBrow, MD,
Cynthia L. David, MD,
Farzin Eftekhari, MD and
Chusilp Charnsangavej, MD
1 From the Department of Diagnostic Radiology, the University of Texas M. D. Anderson Cancer Center, Box 57, 1515 Holcombe Blvd, Houston, TX 77030. Received January 7, 1998; revision requested March 6; final revision received November 23; accepted March 8, 1999. Address reprint requests to E.M.L. (e-mail: eloyer@di.mdacc.tmc.edu).
 |
Abstract
|
|---|
PURPOSE: To define the hemodynamic features of hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma by using quadruple phase helical computed tomography (CT) and determine the value of this information in characterizing tumors.
MATERIALS AND METHODS: Helical CT of the liver was performed in 45 patients with newly diagnosed HCC or peripheral cholangiocarcinoma. Scans were obtained before and 25 seconds, 70 seconds, and 2–6 minutes after the start of the contrast material injection. The intensity and spatial distribution of contrast material uptake were evaluated during all phases. Time-attenuation curves were established for each lesion. Relative attenuation and lesion conspicuity were assessed. A diagnostic confidence level was assigned to each lesion.
RESULTS: In the majority of HCC lesions, a single, early peak of enhancement followed by a continuous decrease in tumor attenuation over time was seen. The greatest tumor conspicuity occurred during the delayed phase. In cholangiocarcinoma, tumor attenuation increased during the delayed phase. In the majority of lesions, the greatest tumor conspicuity was seen during the portal venous phase. In both tumor types, the diagnostic confidence level improved when the delayed phase was used.
CONCLUSION: The variation over time in the intensity of contrast enhancement in HCC and cholangiocarcinoma differs sufficiently to make this a useful diagnostic criterion. The delayed phase is particularly important because it amplifies this difference.
Index terms: Bile ducts, neoplasms, 76.321 • Computed tomography (CT), contrast enhancement, 76.12113, 76.12114 • Computed tomography (CT), helical, 76.12115 • Liver, CT, 76.12111, 76.12113, 76.12114, 76.12115 • Liver neoplasms, 76.321, 76.323
|
Introduction
|
|---|
The enhancement characteristics of hepatic tumors at dynamic computed tomography (CT) are influenced by their histologic type. The capability of helical CT to enable scanning of the entire liver sequentially during the hepatic arterial and portal venous phases (HAP and PVP, respectively) of enhancement affords the opportunity to study the temporal evolution of the contrast enhancement of hepatic tumors and to use this information to predict, to some extent, the pathologic diagnosis. Nodular hepatocellular carcinoma (HCC) and peripheral cholangiocarcinoma exhibit very different pathologic features and have been shown to have different patterns of enhancement (1–6). In this study, we compared the enhancement patterns of HCC and peripheral cholangiocarcinoma by using quadruple phase helical CT. Our purpose was to define the evolution of enhancement of each tumor type over time by using helical CT and determine the usefulness of this information for tumor characterization.
|
MATERIALS AND METHODS
|
|---|
Forty-five consecutive patients (26 men, 19 women; age range, 22–75 years; mean age, 66 years) with newly diagnosed nodular HCC or peripheral cholangiocarcinoma underwent imaging between April 1995 and February 1996 by means of a uniform quadruple phase helical CT technique. All of these patients had biopsy-proved HCC or cholangiocarcinoma. Twenty-five patients had HCC, 19 patients had cholangiocarcinoma, and one patient had mixed HCC and cholangiocarcinoma. The dominant lesion, and at least one smaller lesion when the disease was multifocal, was characterized. Lesions were classified by size into one of the following three groups: 5 cm or larger, at least 2 cm but smaller than 5 cm, or smaller than 2 cm (mean, 5.7 cm).
All scans were obtained by using a CT HiSpeed Advantage scanner (GE Medical Systems, Milwaukee, Wis). First, nonenhanced axial scans with 10-mm collimation were acquired. Then, contrast material–enhanced helical CT of the entire liver was performed during a single breath hold by using 7–10-mm collimation with a 1.0:1.0 or 1.5:1.0 pitch in a craniocaudal direction. The scanning sequences were initiated 25 seconds, 70 seconds, and 2-6 minutes (mean, 3 minutes 10 seconds) after the start of the intravenous injection of 150 mL of 60% nonionic contrast material (iohexol [Omnipaque 300]; Nycomed, Princeton, NJ) at a rate of 2.5 mL/sec. This rate was chosen because it can be achieved consistently. This scanning protocol provided three sets of scans that corresponded to the HAP (25 seconds), PVP (70 seconds), and delayed phase (2-6 minutes) of contrast enhancement. The time required to acquire each set of scans was 20–30 seconds. Because this technique did not add any risk compared with conventional abdominal CT, and because at our institution patients sign a general consent form that covers all diagnostic studies, neither institutional review board approval nor informed consent was considered to be necessary.
The lesions were assessed objectively and subjectively. In the objective analyses, the CT attenuation coefficients of tumors were measured during four different phases. The regions of interest were chosen by one radiologist (H.C.). The enhancement of the heterogeneously enhancing lesions was evaluated in the areas where there was more homogeneous enhancement. Film hard-copy images were reviewed by at least three of the radiologists (E.M.L., R.A.D., C.L.D., F.E., C.C.) who participated in the study. If more than one region of interest was selected in large heterogeneously enhancing tumors, then the region of interest used for analysis was chosen by consensus. Measurements of the solid component of the tumor and of the normal liver parenchyma were obtained from the same areas on the four scans (obtained during the four different phases). By using these data, time-attenuation curves and curves of differences in attenuation between the tumor and liver parenchyma were established.
The attenuation values of uninvolved liver parenchyma measured in both tumor types were compared by using the Mann-Whitney test. P values were computed by using Stat Exact software (Cytel, Cambridge, Mass). A P value of less than .05 denoted statistical significance.
In the subjective analyses, tumor attenuation was graded with respect to the surrounding liver parenchyma. Five radiologists (E.M.L., H.C., R.A.D., C.L.D., F.E., C.C.) interpreted the scans and established a consensus grade without knowledge of the tissue diagnosis. The scans were evaluated with both narrow (window width, 100 HU; window level, 60–90 HU) and soft-tissue (width, 400 HU; level, 70–80 HU) window settings. On all four sets of scans, the lesions were classified as hyperattenuating, isoattenuating, or hypoattenuating relative to the adjacent parenchyma. Heterogeneously enhancing lesions were classified as hyperattenuating when most of the solid tumor was enhancing.
The lesions that were hyperattenuating or hypoattenuating were further subdivided into the following three categories according to their conspicuity: 1, barely visible against the surrounding liver parenchyma; 2, intermediately obvious; or 3, extremely obvious. Consequently, the following seven categories were defined: hyperattenuating 3, hyperattenuating 2, hyperattenuating 1, isoattenuating, hypoattenuating 1, hypoattenuating 2, hypoattenuating 3. Lesions with the highest relative attenuation were described as hyperattenuating 3, and those with the lowest relative attenuation were described as hypoattenuating 3. Each lesion was assigned to four categories, one for each set of scans, which together expressed a qualitative assessment of the tumor attenuation and conspicuity over time. By using these data, a curve of the relative attenuation of the tumor over time was established for each lesion.
Ancillary findings also were documented and included the presence of a capsule, tumor thrombus, arterioportal shunting, cirrhosis, ascites, adenopathy, biliary obstruction, calcification, and/or necrosis. A diagnosis of each case was established by consensus on the basis of the subjective appreciation of the enhancement pattern, morphologic characteristics, and ancillary findings. A confidence level for each diagnosis was then assigned.
To determine the importance of the delayed phase, the studies were reviewed a second time, after a delay of 2 months, by using the scans obtained before contrast material administration and those obtained during the HAP and PVP. Diagnoses were established a second time, and the diagnostic confidence levels were assigned and compared with the confidence levels reached when all four phases were used initially.
|
RESULTS
|
|---|
Hepatocellular Carcinoma
Morphologic features.—We analyzed 40 HCC lesions in 25 patients. Twenty-two lesions were 5 cm or larger, 10 were 2 cm or larger but smaller than 5 cm, and eight were smaller than 2 cm. The tumor was unifocal in 13 (52%), multifocal with a dominant mass in 10 (40%), and multifocal without a dominant mass in two (8%) patients. A mosaic pattern was present in seven (28%), a capsule was seen in 16 (64%), and necrosis characterized by an area of attenuation close to that of water was seen in 10 (40%) patients. Portal or hepatic vein thrombosis was seen in 10 patients (40%), and arterioportal shunting was seen in two (8%). Radiologic evidence of cirrhosis was present in five (20%) patients. Other features included the following: ascites in 10 (40%) patients, lymphadenopathy in eight (32%) patients, focal biliary dilatation in two (8%) patients, and calcification in two (8%) patients.
Enhancement characteristics.—The CT attenuation measurements revealed a pattern of enhancement characterized by a single peak of enhancement during the HAP and/or PVP followed by a decrease in attenuation over time in 38 (95%) of the 40 lesions (Fig 1). The mean attenuation values during the precontrast phase, HAP, PVP, and delayed phase were 33, 61, 90, and 69 HU, respectively. The highest attenuation coefficients were measured on the PVP scans in 30 lesions, on the PVP and HAP scans in eight lesions, and on the HAP scans alone in none. Attenuation measurements demonstrated no enhancement in two lesions. There was no difference in the delay time of enhancement when the size of the lesions was considered. On average, the CT attenuation of tumors during the PVP increased 180% compared with those during the precontrast phase. A decrease in contrast enhancement over time was seen in all but one lesion.
Although attenuation measurements on the delayed phase scans were, in general, higher than those on the HAP scans, they were lower than those on the PVP scans by 20 HU or more in 21 lesions and by less than 20 HU in 15 lesions. The attenuation during the delayed phase was equal to that during the PVP in three lesions. In one lesion, the attenuation was higher during the delayed phase, but only by 13 HU. The difference in attenuation coefficients between the tumor and the surrounding liver parenchyma was highest during the PVP and/or delayed phase in 29 lesions, during the HAP and PVP in four lesions, and during the HAP only in one lesion. In six lesions, the differences in attenuation between the tumor and the normal liver parenchyma were similar during all phases.
In the subjective analysis, the lesions were classified into two groups based on the presence (group 1) (Fig 2) or absence (group 2) of hyperattenuation relative to the liver parenchyma. The lesions in both groups, however, were similarly characterized by a decrease over time in tumor attenuation relative to the liver parenchyma (Fig 2). There were 30 (75%) lesions in group 1. Sixteen lesions were 5 cm or larger, 10 lesions were at least 2 cm but smaller than 5 cm, and four lesions were smaller than 2 cm.
View larger version (128K):
[in this window]
[in a new window]
|
Figure 2a. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP, (c) PVP, and (d) delayed phase demonstrate the HCC enhancement pattern. (a) Precontrast scan shows enlargement of the liver. There is loss of the normal vascular anatomy but no definite focal change in attenuation. During the subjective evaluation, this tumor was judged to be isoattenuating. One small tumor nodule (arrow) is of lower attenuation than the surrounding liver parenchyma. The high-attenuating material (arrowhead) in the left lobe is a result of prior chemoembolization. (b) HAP scan shows marked increase in relative tumor attenuation and conspicuity. The dominant mass in the left lobe and the tumor nodules were judged to be hyperattenuating 3. (c) PVP scan shows decrease in relative tumor attenuation and conspicuity. The tumor was judged to be isoattenuating. (d) Delayed phase scan shows further decrease in relative tumor attenuation and increase in conspicuity. The tumor was judged to be hypoattenuating 3. The areas of very low attenuation (arrows) correspond to tumor nodules demonstrating a more pronounced loss of contrast.
|
|
View larger version (130K):
[in this window]
[in a new window]
|
Figure 2b. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP, (c) PVP, and (d) delayed phase demonstrate the HCC enhancement pattern. (a) Precontrast scan shows enlargement of the liver. There is loss of the normal vascular anatomy but no definite focal change in attenuation. During the subjective evaluation, this tumor was judged to be isoattenuating. One small tumor nodule (arrow) is of lower attenuation than the surrounding liver parenchyma. The high-attenuating material (arrowhead) in the left lobe is a result of prior chemoembolization. (b) HAP scan shows marked increase in relative tumor attenuation and conspicuity. The dominant mass in the left lobe and the tumor nodules were judged to be hyperattenuating 3. (c) PVP scan shows decrease in relative tumor attenuation and conspicuity. The tumor was judged to be isoattenuating. (d) Delayed phase scan shows further decrease in relative tumor attenuation and increase in conspicuity. The tumor was judged to be hypoattenuating 3. The areas of very low attenuation (arrows) correspond to tumor nodules demonstrating a more pronounced loss of contrast.
|
|
View larger version (134K):
[in this window]
[in a new window]
|
Figure 2c. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP, (c) PVP, and (d) delayed phase demonstrate the HCC enhancement pattern. (a) Precontrast scan shows enlargement of the liver. There is loss of the normal vascular anatomy but no definite focal change in attenuation. During the subjective evaluation, this tumor was judged to be isoattenuating. One small tumor nodule (arrow) is of lower attenuation than the surrounding liver parenchyma. The high-attenuating material (arrowhead) in the left lobe is a result of prior chemoembolization. (b) HAP scan shows marked increase in relative tumor attenuation and conspicuity. The dominant mass in the left lobe and the tumor nodules were judged to be hyperattenuating 3. (c) PVP scan shows decrease in relative tumor attenuation and conspicuity. The tumor was judged to be isoattenuating. (d) Delayed phase scan shows further decrease in relative tumor attenuation and increase in conspicuity. The tumor was judged to be hypoattenuating 3. The areas of very low attenuation (arrows) correspond to tumor nodules demonstrating a more pronounced loss of contrast.
|
|
View larger version (134K):
[in this window]
[in a new window]
|
Figure 2d. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP, (c) PVP, and (d) delayed phase demonstrate the HCC enhancement pattern. (a) Precontrast scan shows enlargement of the liver. There is loss of the normal vascular anatomy but no definite focal change in attenuation. During the subjective evaluation, this tumor was judged to be isoattenuating. One small tumor nodule (arrow) is of lower attenuation than the surrounding liver parenchyma. The high-attenuating material (arrowhead) in the left lobe is a result of prior chemoembolization. (b) HAP scan shows marked increase in relative tumor attenuation and conspicuity. The dominant mass in the left lobe and the tumor nodules were judged to be hyperattenuating 3. (c) PVP scan shows decrease in relative tumor attenuation and conspicuity. The tumor was judged to be isoattenuating. (d) Delayed phase scan shows further decrease in relative tumor attenuation and increase in conspicuity. The tumor was judged to be hypoattenuating 3. The areas of very low attenuation (arrows) correspond to tumor nodules demonstrating a more pronounced loss of contrast.
|
|
On the precontrast scans, all lesions were either isoattenuating (n = 13) or hypoattenuating (n = 17) relative to the surrounding liver parenchyma. After intravenous contrast material enhancement, all lesions became hyperattenuating. Enhancement relative to the surrounding liver appeared to be at its maximum during the HAP in 18 lesions (60%), during the PVP in eight (27%) lesions, and during both phases in four (13%) lesions (Fig 3a). In the four latter cases, the degree of enhancement achieved during the HAP persisted to be unchanged during the PVP. This peak in relative attenuation was followed in all cases by a continuous decrease in relative attenuation (Fig 3a). There was no difference in the evolution of contrast material uptake when the size of the lesions was considered.
There were 10 (25%) lesions in group 2. Six lesions were 5 cm or larger, and four were smaller than 2 cm. On the precontrast scans, all lesions were either isoattenuating (n = 4) or hypoattenuating (n = 6). Following intravenous contrast material enhancement, none of these lesions became hyperattenuating; they remained isoattenuating or hypoattenuating relative to the liver parenchyma. The attenuation scores (Fig 3b), however, indicated a change in the relative tumor attenuation over time. Three lesions had a relative increase in attenuation during the HAP, and one lesion had an increase in attenuation during the PVP; the peaks of enhancement in both these groups of lesions were followed by a decrease in attenuation thereafter. In six lesions, only a decrease in the relative tumor attenuation was observed over time; this occurred during the PVP and delayed phase (Fig 3b). A comparison with the objective attenuation measurements in these six lesions, however, showed that a peak of enhancement did occur during the PVP in three lesions and during the HAP and PVP in one lesion. Attenuation measurements confirmed the absence of enhancement in two of these six tumors.
Overall, in the subjective analysis, a single peak of enhancement was seen in 34 (85%) of the 40 lesions. The lowest relative tumor attenuation was observed on the scans obtained during the delayed phase in 31 (78%) lesions. The attenuation was lowest during the delayed phase alone in 23 (58%) lesions. In 17 lesions (42%), the lowest relative tumor attenuation was observed during more than one phase; the delayed phase was one of these phases in eight of these cases. Tumor conspicuity varied with the phase of enhancement and was judged to be maximal during one phase or frequently during two phases, but rarely during three of the four phases. The delayed phase was among the phases chosen for its highest conspicuity in 30 lesions (75%), whereas the HAP, PVP, or precontrast phase was among the phases chosen respectively, in six, 16, and three lesions.
In 16 (40%) lesions, a capsule was identified and the time of enhancement of the capsule differed from that of the tumor. The capsule was seen on the precontrast study as a hypoattenuating, often incomplete ring in six lesions. It was also seen as a hypoattenuating ring during the HAP in five of these six lesions. The capsule was seen as an enhancing rim during the PVP in seven lesions and during the delayed phase in all 16 lesions.
The diagnosis of HCC was established in 24 (96%) of the 25 patients on the basis of morphologic and subjective analysis of the enhancement characteristics. When the cases were reviewed a second time by using the same scoring system but without the aid of the delayed phase scans, the correct diagnosis was made again in the same 24 patients. Although there was no difference in the diagnostic accuracy, the diagnostic confidence level was greater in 14 (56%) cases when the delayed phase scans were available.
Intrahepatic Peripheral Cholangiocarcinoma
Morphologic features.—A total of 42 cholangiocarcinoma lesions in 19 patients were assessed. Sixteen lesions were 5 cm or larger, 13 lesions were at least 2 cm but smaller than 5 cm, and 13 lesions were smaller than 2 cm. The disease was unifocal in six patients (32%) and multifocal in 13 (68%) patients. In 12 of the 13 patients with multifocal lesions, the dominant mass was larger than 5 cm. Tumor necrosis characterized by an area of attenuation similar to the attenuation of water was present in four patients. Biliary dilatation was seen in seven (37%), underlying cirrhosis was seen in four (21%), and ascites was seen in four (21%) patients. Lymphadenopathy was seen in 12 (63%) patients. Neither calcification, tumorous capsule, tumor thrombus, nor arterioportal shunting was seen in any patient.
Enhancement characteristics.—CT attenuation measurements indicated tumoral enhancement in all lesions, with maximum enhancement during the PVP and/or delayed phase (Fig 4). On average, CT attenuation increased by 132% during the PVP compared with the attenuation during the precontrast phase. The mean CT attenuations during the precontrast phase, HAP, PVP, and delayed phase were 23, 31, 56, and 61 HU, respectively. Measurement of CT attenuation on the delayed phase scans indicated a persistence of or further increase in tumoral enhancement. Attenuation measurements during the delayed phase were higher than those during the PVP in 31 (74%) lesions (by more than 20 HU in five lesions, by 10–20 HU in 11 lesions, and by less than 10 HU in 15 lesions). A decrease in the objective measurement was seen in 11 lesions (26%), but it was a difference of only 2–4 HU. The difference in attenuation between the tumor and surrounding liver parenchyma was maximal during the PVP.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 4. Graph illustrating objective evaluation of cholangiocarcinoma.  ,  ,  ,  , open asterisk,  ,  ,  , *,  , •,  , and  represent the attenuation values of the 13 tumors. Time-attenuation curves for 13 tumors measuring at least 2 cm but less than 5 cm show a tendency toward increased attenuation over time. Note that the y axis differs from that in Figure 1. The time-attenuation curves for tumors equal to or larger than 5 cm and for tumors smaller than 2 cm were identical to these.
|
|
The results of subjective analysis showed that all lesions were either hypoattenuating (35 lesions) or isoattenuating (seven lesions) on the precontrast scans. After intravenous contrast material enhancement, three lesions became hyperattenuating and three lesions became isoattenuating. The other 36 lesions remained predominantly hypoattenuating during the HAP and PVP (Fig 5). In 18 lesions, however, the attenuation scores indicated that despite persistent hypoattenuation, the relative tumor attenuation actually increased during the HAP (Fig 6). The attenuation scores remained similar to those during the precontrast study in 21 lesions and decreased in three lesions (Fig 6). The lowest relative tumor attenuation was seen during the PVP in 36 (86%) of the 42 lesions. In contrast, the delayed phase compared with the PVP was characterized by an increase in relative tumor attenuation in 27 lesions (64%) (Fig 6a), no change in 12 lesions (Fig 6b), and a decrease in relative attenuation in three lesions. In addition, in 14 lesions, the distribution of contrast material in the lesion migrated centrally during the delayed phase; this bull's eye pattern is also seen with adenocarcinoma but not with HCC. Lesion conspicuity varied according to the enhancement phase and was judged to be maximal during the PVP in 36 (86%) lesions.
View larger version (126K):
[in this window]
[in a new window]
|
Figure 5a. (a-d) Axial abdominal CT scans obtained during the (a) precontrast, (b) HAP, (c) PVP, and delayed phase demonstrate the enhancement pattern of cholangiocarcinoma. (a) Precontrast scan shows an area of ill-defined hypoattenuation (arrows) in the right lobe. The mass was judged to the hypoattenuating 2 in the subjective assessment. (b) HAP scan shows peripheral enhancement (arrows). The attenuation of most of the tumor, however, similar to that on the precontrast study (a), is hypoattenuating 2. (c) PVP scan shows the mass (arrows) to have decreased relative attenuation and increased conspicuity. This tumor was judged to be hypoattenuating 3 in the subjective evaluation. (d) Delayed phase scan shows increased relative tumor attenuation with central migration of contrast material. This tumor was judged to be hyperattenuating 3 in the subjective evaluation. An enhancing capsular vein (arrow) also is seen.
|
|
View larger version (123K):
[in this window]
[in a new window]
|
Figure 5b. (a-d) Axial abdominal CT scans obtained during the (a) precontrast, (b) HAP, (c) PVP, and delayed phase demonstrate the enhancement pattern of cholangiocarcinoma. (a) Precontrast scan shows an area of ill-defined hypoattenuation (arrows) in the right lobe. The mass was judged to the hypoattenuating 2 in the subjective assessment. (b) HAP scan shows peripheral enhancement (arrows). The attenuation of most of the tumor, however, similar to that on the precontrast study (a), is hypoattenuating 2. (c) PVP scan shows the mass (arrows) to have decreased relative attenuation and increased conspicuity. This tumor was judged to be hypoattenuating 3 in the subjective evaluation. (d) Delayed phase scan shows increased relative tumor attenuation with central migration of contrast material. This tumor was judged to be hyperattenuating 3 in the subjective evaluation. An enhancing capsular vein (arrow) also is seen.
|
|
View larger version (105K):
[in this window]
[in a new window]
|
Figure 5c. (a-d) Axial abdominal CT scans obtained during the (a) precontrast, (b) HAP, (c) PVP, and delayed phase demonstrate the enhancement pattern of cholangiocarcinoma. (a) Precontrast scan shows an area of ill-defined hypoattenuation (arrows) in the right lobe. The mass was judged to the hypoattenuating 2 in the subjective assessment. (b) HAP scan shows peripheral enhancement (arrows). The attenuation of most of the tumor, however, similar to that on the precontrast study (a), is hypoattenuating 2. (c) PVP scan shows the mass (arrows) to have decreased relative attenuation and increased conspicuity. This tumor was judged to be hypoattenuating 3 in the subjective evaluation. (d) Delayed phase scan shows increased relative tumor attenuation with central migration of contrast material. This tumor was judged to be hyperattenuating 3 in the subjective evaluation. An enhancing capsular vein (arrow) also is seen.
|
|
View larger version (126K):
[in this window]
[in a new window]
|
Figure 5d. (a-d) Axial abdominal CT scans obtained during the (a) precontrast, (b) HAP, (c) PVP, and delayed phase demonstrate the enhancement pattern of cholangiocarcinoma. (a) Precontrast scan shows an area of ill-defined hypoattenuation (arrows) in the right lobe. The mass was judged to the hypoattenuating 2 in the subjective assessment. (b) HAP scan shows peripheral enhancement (arrows). The attenuation of most of the tumor, however, similar to that on the precontrast study (a), is hypoattenuating 2. (c) PVP scan shows the mass (arrows) to have decreased relative attenuation and increased conspicuity. This tumor was judged to be hypoattenuating 3 in the subjective evaluation. (d) Delayed phase scan shows increased relative tumor attenuation with central migration of contrast material. This tumor was judged to be hyperattenuating 3 in the subjective evaluation. An enhancing capsular vein (arrow) also is seen.
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Figure 6a. Graph illustrating subjective analysis of cholangiocarcinoma. (a, b) The y axis reflects the conspicuity of the lesions. Hypodense 3 and Hyperdense 3 indicate maximal conspicuity for hypoattenuating and hyperattenuating lesions, respectively; Hypodense 1 and Hyperdense 1 indicate the lowest conspicuity. Schematic curves for the representative lesions are shown. Peak enhancement occurred during the HAP in 18 tumors ( and ). The remaining 24 tumors did not appear to enhance during the HAP ( and  ). (b) All cholangiocarcinoma lesions were characterized by a decrease in relative attenuation during the PVP. Unlike in HCC lesions, a further decrease in relative attenuation was not seen during the delayed phase, but rather an increase in relative attenuation occurred in 27 lesions (, , , and in a). The relative tumor attenuation remained stable in 12 lesions (, , , and in b). A decrease in relative tumor attenuation during the delayed phase (not shown) was seen in three of 42 tumors.
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Figure 6b. Graph illustrating subjective analysis of cholangiocarcinoma. (a, b) The y axis reflects the conspicuity of the lesions. Hypodense 3 and Hyperdense 3 indicate maximal conspicuity for hypoattenuating and hyperattenuating lesions, respectively; Hypodense 1 and Hyperdense 1 indicate the lowest conspicuity. Schematic curves for the representative lesions are shown. Peak enhancement occurred during the HAP in 18 tumors ( and ). The remaining 24 tumors did not appear to enhance during the HAP ( and ). (b) All cholangiocarcinoma lesions were characterized by a decrease in relative attenuation during the PVP. Unlike in HCC lesions, a further decrease in relative attenuation was not seen during the delayed phase, but rather an increase in relative attenuation occurred in 27 lesions (, , , and in a). The relative tumor attenuation remained stable in 12 lesions (, , , and in b). A decrease in relative tumor attenuation during the delayed phase (not shown) was seen in three of 42 tumors.
|
|
In all 19 patients, the diagnosis of cholangiocarcinoma was predicted accurately on the basis of morphologic characteristics and subjective analysis of enhancement. When the scans were reviewed a second time by using the same scoring system but without the benefit of the delayed phase scans, the diagnosis was made correctly in 18 patients. Confidence in the diagnosis was greater in nine patients (47%) when the delayed phase scans were available.
Mixed HCC and Cholangiocarcinoma
One lesion had mixed characteristics at CT. According to subjective analysis, the relative attenuation increased continuously over time. By using only the precontrast phase, HAP, and PVP scans, this lesion was judged to be HCC. The delayed phase scans showed a further increase in attenuation, which suggested the diagnosis of cholangiocarcinoma. The pathologic diagnosis was mixed HCC and cholangiocarcinoma.
Attenuation of Nontumorous Liver Parenchyma
The attenuation values of uninvolved liver were higher in the HCC group than in the cholangiocarcinoma group (Table). These differences were found to be significant during the precontrast (Mann-Whitney test, P = .015) and delayed (Mann-Whitney test, P = .028) phases and showed a trend during the HAP (Mann-Whitney test, P = .088).
View this table:
[in this window]
[in a new window]
|
Mean Attenuation Coefficients of the Normal Liver Parenchyma in Patients with HCC and in Patients with Cholangiocarcinoma
|
|
|
DISCUSSION
|
|---|
Tumor enhancement may be characterized by intensity, spatial distribution of contrast material, and most important, the evolution over time of both of these parameters. In this study, we focused on the temporal evolution of contrast material uptake by HCCs and peripheral cholangiocarcinomas. We assessed the changes over time in relative tumor attenuation (subjective evaluation), in the attenuation coefficient values of the tumors (objective evaluation), and in the spatial distribution of contrast material.
The results obtained with subjective grading and objective measurement of attenuation coefficients differed in some aspects. With HCCs, on the basis of subjective measurements, peak enhancement occurred during the HAP in 21 (52%) lesions, whereas with objective measurements, it occurred during the PVP in 38 (95%) lesions. Both sets of data, however, showed a single, early peak of enhancement followed by a continuous decrease in attenuation. This was observed by using subjective measurements in 34 (85%) tumors and by using attenuation measurements in 38 (95%) tumors.
With cholangiocarcinomas, enhancement could be defined by using attenuation measurements, but was more difficult to appreciate by using subjective analysis because the majority of the lesions remained hypoattenuating during the HAP and PVP. Both the objective and subjective data, however, demonstrated a similar tendency for delayed accumulation of contrast material as opposed to the washout seen with HCC. This accumulation was seen by using subjective measurements in 27 (64%) lesions and by using attenuation measurements in 31 (74%) lesions.
The apparent discordance between the subjective and objective evaluations can be explained. Subjective evaluation involves the visual appreciation of the difference in attenuation between the liver parenchyma and tumor and is dependent on the enhancement of the normal liver parenchyma as well as that of the tumor. With HCC, the attenuation of the tumor during the PVP is maximal, but it is lower than that of the surrounding liver parenchyma; therefore, the tumor appears to be hypoattenuating. In contrast, during the HAP, the attenuation measurement of the tumor is equal to or higher than that of the surrounding liver parenchyma; therefore, the tumor appears to be isoattenuating or hyperattenuating even though the absolute tumor attenuation measurement is lower than that during the PVP.
With cholangiocarcinoma, the tumor attenuation is lower than that of the surrounding liver parenchyma during all phases; therefore, the tumor remains hypoattenuating according to subjective analysis. Although enhancement during the PVP is not appreciated because of the high attenuation of the surrounding liver parenchyma, it may be appreciated during the delayed phase as the attenuation of the surrounding liver parenchyma decreases.
Our findings with respect to the enhancement of HCC concur with those of studies by Araki et al (3) and Hosoki et al (7,8), who also used measurements of tumor attenuation coefficients after contrast material injection to show that tumoral enhancement follows a time-attenuation curve that ascends and descends rapidly. In these studies, maximal enhancement occurred during the HAP. In our study, the highest attenuation coefficients were measured during the PVP. This difference is possibly related to a more rapid administration of contrast material in these studies than that in our protocol.
A similar temporal evolution of enhancement in HCCs has also been described by using subjective evaluations. Honda et al (9) assessed tumoral enhancement at 45 seconds and 6 minutes after contrast material injection. Thirty-eight (56%) of 68 lesions showed diffuse, peripheral, or mixed enhancement during the early phase and apparent decreased attenuation during the delayed phase. Twenty-eight (41%) of the 68 lesions did not show enhancement, and two (3%) lesions increased in attenuation on the delayed phase scans. The long interval between the two scanning phases, however, raises the possibility that enhancement, which is transient, might have been missed in some of these tumors. Cho et al (10) evaluated the scans obtained in 84 HCC lesions during the HAP (20–45 seconds), PVP (55–90 seconds), and delayed (2–4 minutes) phases of contrast enhancement. Seventy-three (87%) of these lesions showed moderate to marked hyperattenuation during the HAP, and 67 (80%) were hypoattenuating during the delayed phase. Similarly, Takayasu et al (11) evaluated 18 HCCs smaller than 3 cm and found a single peak of enhancement. They described this pattern as low attenuating on nonenhanced scans, high attenuating on early enhanced scans, and low attenuating on late enhanced scans. This pattern was seen in 16 (89%) of the lesions.
Most frequently in HCC, relative hyperattenuation is maximal during the HAP, but peak enhancement during the PVP or during the overlapping HAP and PVP can occur, as it did in some of our cases (22% and 10% of lesions, respectively, when they were evaluated subjectively). Variations in the time to peak enhancement among lesions have been reported by other investigators (10,12) as well. Improved rates of detecting HCCs by using biphasic contrast-enhanced helical CT were reported by Baron et al (12); this suggests that the use of both the HAP and PVP optimizes the examination of patients suspected of having HCC. Although we did not study the sensitivity of our technique, the results of our study also suggest that the use of both phases is important in this patient population because the time to maximal enhancement is not constant.
The results of our study confirm the value of the delayed phase in the detection and characterization of HCCs. Capsular or septal enhancement, when present, occurs during this phase (4,10,13). In our study, capsular enhancement was seen during the delayed phase in 16 lesions; it was seen during the PVP in seven lesions. Hypoattenuation of tumors during the delayed phase has been reported in 88% and 80% of lesions in two different studies (10,14). This phase assisted in the detection of 13% of lesions in the study by Honda et al (14). Hwang et al (15) showed that a small number of lesions may be seen during the delayed phase only. In our study, loss of contrast during the delayed phase was seen subjectively in 85% and objectively in 95% of lesions. When delayed phase scans were available, the diagnostic confidence increased in 56% of the patients.
Although lesion conspicuity was not the focus of this study, we assessed this parameter with both subjective grading and attenuation measurements. With both methods, we found that the conspicuity of HCCs was highest during the delayed phase and poor during the HAP. The reason for the poor enhancement during the HAP compared with that in prior reports may be twofold. We used an injection rate of 2.5 mL/sec, which is slower than the rates used by some of the other investigators. More important, the average tumor size in our population was 5.7 cm; only eight of 40 lesions were smaller than 2 cm, and four were 2 cm. The value of the HAP in the detection of small as well as large HCC lesions has been shown (12), but in general, this phase is more useful in the detection of small tumors (10,16,17). In our study, of the six lesions that were best seen on HAP scans, four were smaller than 2 cm.
Our findings on the enhancement of peripheral cholangiocarcinomas also concur with those of other researchers (5,18–20). Cholangiocarcinomas are most frequently hypoattenuating during the early phases of contrast material administration but show relatively increased attenuation on delayed phase scans (19). Cholangiocarcinomas may be hyperattenuating during the early phase of contrast enhancement, as they were in three of our cases, but this is uncommon (18,19). In our study, a delayed increase in attenuation was found subjectively in 27 (64%) lesions and objectively in 31 (74%) lesions. Three lesions showed a decrease in relative tumor attenuation during the delayed phase. Lacomis et al (5) showed that delayed tumor enhancement on scans obtained 6–36 minutes after contrast material injection may aid in the detection and characterization of cholangiocarcinomas. The tumor was seen on the delayed phase scans only in three of the 47 patients in their study (5). This pattern of delayed enhancement is indistinguishable from that observed in metastatic adenocarcinoma and may be due in part to the fibrous stroma of these tumors.
The spatial distribution of contrast material within tumors during the delayed phase may prove to be useful in evaluating cholangiocarcinomas. Peripheral enhancement on early phase scans and central enhancement surrounded by a peripheral low-attenuating area (ie, bull's eye) on delayed phase scans are known patterns of enhancement with adenocarcinomas (6,21) and have been reported with cholangiocarcinoma (21,22). In 14 of the cholangiocarcinomas that we evaluated, enhancement was initially seen at the periphery of the lesion and later in the center of the lesion with a hypoattenuating rim. Overall, we found that the use of the delayed phase increased diagnostic confidence in 47% of the patients with cholangiocarcinoma.
In conclusion, the evolutions over time of the relative CT attenuation of HCCs and of peripheral cholangiocarcinomas differ sufficiently to be a differential diagnostic criterion. Adenocarcinoma and cholangiocarcinoma cannot, however, be differentiated on the basis of their enhancement patterns. A quadruple phase spiral CT technique appears to be optimal for the characterization of primary hepatic tumors. This technique ensures the demonstration of peak HCC lesion enhancement despite individual variation in the time to maximal enhancement. With cholangiocarcinoma, the technique allows demonstration of the delayed increase in tumor attenuation. The delayed phase is particularly helpful because it exposes the opposite enhancement trends of the two tumor types (Fig 7). Enhancement of the capsule or septa, when present, in HCC and of the central migration of contrast material in cholangiocarcinoma are also demonstrated during this phase. The delays do not need to be longer than 6 minutes (delayed time was <2 minutes 30 seconds in 20 patients), because the focus is not on an individual phase but rather on the overall pattern of enhancement over time. This short delay time allows the patient to remain on the table until completion of the examination.
View larger version (127K):
[in this window]
[in a new window]
|
Figure 7a. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP phase, (c) PVP phase, and (d) delayed phase demonstrate the value of the delayed phase. (a) Precontrast scan shows ill-defined area of low attenuation (arrowheads) occupying the lateral segment of the left liver lobe, with changes in the contour. Several hypoattenuating nodules (arrows) are seen in the right lobe. The tumor was judged to be hypoattenuating 2 in the subjective assessment. (b) HAP scan shows increased relative tumor attenuation and decreased conspicuity. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. (c) PVP scan shows decreased relative attenuation of tumor nodules (arrows) and increased conspicuity. The tumor was judged to be hypoattenuating 2 in the subjective evaluation. (d) Delayed phase scan shows an increase in relative tumor attenuation. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. Because this usually was not seen with HCC, the diagnosis of cholangiocarcinoma or adenocarcinoma is more likely.
|
|
View larger version (135K):
[in this window]
[in a new window]
|
Figure 7b. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP phase, (c) PVP phase, and (d) delayed phase demonstrate the value of the delayed phase. (a) Precontrast scan shows ill-defined area of low attenuation (arrowheads) occupying the lateral segment of the left liver lobe, with changes in the contour. Several hypoattenuating nodules (arrows) are seen in the right lobe. The tumor was judged to be hypoattenuating 2 in the subjective assessment. (b) HAP scan shows increased relative tumor attenuation and decreased conspicuity. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. (c) PVP scan shows decreased relative attenuation of tumor nodules (arrows) and increased conspicuity. The tumor was judged to be hypoattenuating 2 in the subjective evaluation. (d) Delayed phase scan shows an increase in relative tumor attenuation. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. Because this usually was not seen with HCC, the diagnosis of cholangiocarcinoma or adenocarcinoma is more likely.
|
|
View larger version (121K):
[in this window]
[in a new window]
|
Figure 7c. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP phase, (c) PVP phase, and (d) delayed phase demonstrate the value of the delayed phase. (a) Precontrast scan shows ill-defined area of low attenuation (arrowheads) occupying the lateral segment of the left liver lobe, with changes in the contour. Several hypoattenuating nodules (arrows) are seen in the right lobe. The tumor was judged to be hypoattenuating 2 in the subjective assessment. (b) HAP scan shows increased relative tumor attenuation and decreased conspicuity. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. (c) PVP scan shows decreased relative attenuation of tumor nodules (arrows) and increased conspicuity. The tumor was judged to be hypoattenuating 2 in the subjective evaluation. (d) Delayed phase scan shows an increase in relative tumor attenuation. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. Because this usually was not seen with HCC, the diagnosis of cholangiocarcinoma or adenocarcinoma is more likely.
|
|
View larger version (130K):
[in this window]
[in a new window]
|
Figure 7d. (a-d) Axial abdominal CT scans obtained during the (a) precontrast phase, (b) HAP phase, (c) PVP phase, and (d) delayed phase demonstrate the value of the delayed phase. (a) Precontrast scan shows ill-defined area of low attenuation (arrowheads) occupying the lateral segment of the left liver lobe, with changes in the contour. Several hypoattenuating nodules (arrows) are seen in the right lobe. The tumor was judged to be hypoattenuating 2 in the subjective assessment. (b) HAP scan shows increased relative tumor attenuation and decreased conspicuity. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. (c) PVP scan shows decreased relative attenuation of tumor nodules (arrows) and increased conspicuity. The tumor was judged to be hypoattenuating 2 in the subjective evaluation. (d) Delayed phase scan shows an increase in relative tumor attenuation. The tumor was judged to be hypoattenuating 1 in the subjective evaluation. Because this usually was not seen with HCC, the diagnosis of cholangiocarcinoma or adenocarcinoma is more likely.
|
|
Knowledge of the described characteristic patterns of enhancement will help in the evaluation of lesions that are atypical during one or more individual phases, because the overall pattern is often preserved. The results of our study also indicate that the technique of contrast material administration and the scanning delay are very important parameters to consider in the interpretation of single phase CT scans, because the expected enhancement of a particular type of lesion will vary substantially with the phase of enhancement. Finally, the observation that the uninvolved liver parenchyma has higher attenuation in the HCC group may be related to underlying cirrhosis and iron deposits in the liver of these patients.
|
Acknowledgments
|
|---|
The authors thank Elihu Estey, MD, for performing statistical analysis and Debbie Smith for secretarial assistance.
|
Footnotes
|
|---|
2 Current address: Department of Radiology, Memorial City Hospital, Houston, Tex. 
Abbreviations: HAP = hepatic arterial phase
HCC = hepatocellular carcinoma
PVP = portal venous phase
Author contributions: Guarantor of integrity of entire study, E.M.L.; study concepts, C.C.; study design, E.M.L., C.C.; definition of intellectual content, C.C.; literature research, H.C.; clinical studies, all authors; data acquisition, H.C., R.A.D., E.M.L., C.L.D., F.E.; data analysis, all authors; manuscript preparation and editing, H.C., E.M.L.; manuscript review, C.C., E.M.L., R.A.D.
|
References
|
|---|
-
Shugihara S, Kojiro M. Pathology of cholangiocarcinoma. In: Okuda K, Ishak KG, eds. Neoplasms of the liver. Tokyo, Japan: Springer-Verlag, 1987; 143-158.
-
Kojiro M, Nakashima T. Pathology of hepatocellular carcinoma. In: Okuda K, Ishak KG, eds. Neoplasms of the liver. Tokyo, Japan: Springer-Verlag, 1987; 81-103.
-
Araki T, Itai Y, Furi S, Tasaka A. Dynamic CT densitometry of hepatic tumors. AJR 1980; 135:1037-1043.[Abstract]
-
Itai Y, Ohtomo K, Kokubo T, et al. CT of hepatic masses: significance of prolonged and delayed enhancement. AJR 1986; 146:729-733.[Abstract/Free Full Text]
-
Lacomis JM, Baron RL, Oliver JH, III, et al. Cholangiocarcinoma: delayed CT contrast enhancement patterns. Radiology 1997; 203:98-104.[Abstract/Free Full Text]
-
Muramatsu Y, Takayasu K, Moriyama N, et al. Peripheral low-density area of hepatic tumors: CT-pathologic correlation. Radiology 1986; 160:49-52.[Abstract/Free Full Text]
-
Hosoki T, Chatani M, Mori S. Dynamic computed tomography of hepatocellular carcinoma. AJR 1982; 139:1099-1106.[Abstract/Free Full Text]
-
Hosoki T, Toyonaga Y, Araki Y, Mori S. Dynamic computed tomography of isodense hepatocellular carcinoma. J Comput Assist Tomogr 1984; 8:263-268.[Medline]
-
Honda H, Ochiai K, Adachi E, et al. Hepatocellular carcinoma: correlation of CT, angiographic, and histopathologic findings. Radiology 1993; 189:857-862.[Abstract/Free Full Text]
-
Cho JS, Kwag JG, Oh YR, et al. Detection and characterization of hepatocellular carcinoma: value of dynamic CT during the arterial dominant phase with uniphasic contrast medium injection. J Comput Assist Tomogr 1996; 20:128-134.[Medline]
-
Takayasu K, Furukawa H, Wakao F, et al. CT diagnosis of early hepatocellular carcinoma: sensitivity, findings and CT-pathologic correlation. AJR 1995; 164:885-890.[Abstract/Free Full Text]
-
Baron RL, Oliver JH, III, Dodd GD, III, et al. Hepatocellular carcinoma: evaluation with biphasic, contrast-enhanced, helical CT. Radiology 1996; 199:505-511.[Abstract/Free Full Text]
-
Lalonde L, van Beers B, Jamart J, et al. Capsule and mosaic pattern of hepatocellular carcinoma: correlation between CT and MR imaging. Gastrointest Radiol 1992; 17:241-244.[Medline]
-
Honda H, Matsuura Y, Onitsuka H, et al. Differential diagnosis of hepatic tumors (hepatoma, hemangioma, and metastasis) with CT: value of two-phase incremental imaging. AJR 1992; 159:735-740.[Abstract/Free Full Text]
-
Hwang GJ, Kim MJ, Yoo SK, et al. Nodular hepatocellular carcinomas: detection with arterial-, portal-, and delayed-phase images at spiral CT. Radiology 1997; 202:383-388.[Abstract/Free Full Text]
-
Hollet MD, Jeffrey RB, Jr, Nino-Murcia M, et al. Dual-phase helical CT of the liver: value of arterial phase scans in the detection of small (<1.5 cm) malignant hepatic neoplasms. AJR 1995; 164:879-884.[Abstract/Free Full Text]
-
Choi BI, Cho JM, Han JK, et al. Spiral CT for the detection of hepatocellular carcinomas: relative value of arterial- and late-phase scanning. Abdom Imaging 1996; 21:440-444.[Medline]
-
Choi BI, Park JH, Kim YI, et al. Peripheral cholangiocarcinoma and clonorchiasis: CT findings. Radiology 1988; 169:149-153.[Abstract/Free Full Text]
-
Honda H, Onitsuka H, Tasumori K, et al. Intrahepatic peripheral cholangiocarcinoma: two-phased dynamic incremental CT and pathologic correlation. J Comput Assist Tomogr 1993; 17:397-402.[Medline]
-
Thorsen MK, Quiroz F, Lawson TL, et al. Primary biliary carcinoma: CT evaluation. Radiology 1984; 152:479-483.[Abstract/Free Full Text]
-
Itai Y, Araki T, Furui S, et al. Differential diagnosis of hepatic masses on computed tomography, with particular reference to hepatocellular carcinoma. J Comput Assist Tomogr 1981; 5:834-842.[Medline]
-
Ros PR, Buck JL, Goodman ZD, et al. Intrahepatic cholangiocarcinoma: radiologic-pathologic correlation. Radiology 1988; 167:689-693.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
T. Uenishi, O. Yamazaki, H. Tanaka, S. Takemura, T. Yamamoto, S. Tanaka, S. Nishiguchi, and S. Kubo
Serum Cytokeratin 19 Fragment (CYFRA21-1) as a Prognostic Factor in Intrahepatic Cholangiocarcinoma
Ann. Surg. Oncol.,
February 1, 2008;
15(2):
583 - 589.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. J. Kim, J. M. Lee, J. K. Han, K. H. Kim, J. Y. Lee, and B. I. Choi
Peripheral Mass Forming Cholangiocarcinoma in Cirrhotic Liver
Am. J. Roentgenol.,
December 1, 2007;
189(6):
1428 - 1434.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H.-X. Xu, M.-D. Lu, G.-J. Liu, X.-Y. Xie, Z.-F. Xu, Y.-L. Zheng, and J.-Y. Liang
Imaging of Peripheral Cholangiocarcinoma With Low-Mechanical Index Contrast-Enhanced Sonography and SonoVue: Initial Experience
J. Ultrasound Med.,
January 1, 2006;
25(1):
23 - 33.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
Z.-C. Zeng, Z.-Y. Tang, J. Fan, J. Zhou, L.-X. Qin, S.-L. Ye, H.-C. Sun, B.-L. Wang, J.-Y. Zhang, Y. Yu, et al.
Radiation Therapy for Adrenal Gland Metastases from Hepatocellular Carcinoma
Jpn. J. Clin. Oncol.,
February 1, 2005;
35(2):
61 - 67.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. I. Choi, J. K. Han, S. T. Hong, and K. H. Lee
Clonorchiasis and Cholangiocarcinoma: Etiologic Relationship and Imaging Diagnosis
Clin. Microbiol. Rev.,
July 1, 2004;
17(3):
540 - 552.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. H. Y. Lee, M. E. O'Malley, M. A. Haider, and A. Hanbidge
Triple-Phase MDCT of Hepatocellular Carcinoma
Am. J. Roentgenol.,
March 1, 2004;
182(3):
643 - 649.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Introduction
) was seen.