DOI:10.2214/AJR.07.2810
AJR 2008; 190:81-87
© American Roentgen Ray Society
Radiologic Detectability of Minute Portal Venous Invasion in Hepatocellular Carcinoma
Akihiro Nishie1,
Kengo Yoshimitsu1,
Yoshiki Asayama1,
Hiroyuki Irie1,
Tsuyoshi Tajima1,
Masakazu Hirakawa1,
Kousei Ishigami1,
Tomohiro Nakayama1,
Daisuke Kakihara1,
Yunosuke Nishihara2,
Akinobu Taketomi3 and
Hiroshi Honda1
1 Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu
University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan.
2 Department of Anatomic Pathology, Graduate School of Medical Sciences, Kyushu
University, Higashi-ku, Fukuoka, Japan.
3 Department of Surgery and Science, Graduate School of Medical Sciences, Kyushu
University, Higashi-ku, Fukuoka, Japan.
Received January 15, 2007;
accepted after revision July 16, 2007.
Address correspondence to A. Nishie
(anishie{at}radiol.med.kyushu-u.ac.jp).
Abstract
OBJECTIVE. The objective of this study was to evaluate whether
minute portal venous invasion in hepatocellular carcinoma (HCC) can be
diagnosed radiologically.
MATERIALS AND METHODS. CT hepatic arteriography and CT with
arterioportography (CTAP) of 15 patients with minute portal venous invasion
(group 1) and 30 patients without it (group 0) were evaluated. An area showing
low attenuation on CTAP and high attenuation on CT hepatic arteriography
around the tumor was defined as an area of peritumoral hemodynamic change. The
shape and size of the area were compared between the two groups. The ratio of
the area of peritumoral hemodynamic change volume to tumor volume (area
volume-tumor volume ratio) was used as an indicator of the size of the area of
peritumoral hemodynamic change and was categorized as one of three grades:
grade I, 10% or less; grade II, between 10% and 30%; and grade III, 30% or
more. The detectability of minute portal invasion was assessed when grade III
was considered as an indicator. Each comparison was also made independently
when the tumor diameter either was limited to less than 3 cm or was 3 cm or
more.
RESULTS. Three types of area of peritumoral hemodynamic change were
identified: wedge-shaped, belt-shaped or irregular, and linear. No significant
difference in the frequency of each type of area of peritumoral hemodynamic
change was observed between the two groups. The area volume-tumor volume ratio
in group 1 was larger than that in group 0, with statistical significance when
the tumor diameter was less than 3 cm (p = 0.046). Positive and
negative predictive values were 71.4% and 75.0%, respectively, when the tumor
diameter was less than 3 cm.
CONCLUSION. The area of peritumoral hemodynamic change in HCC
patients with minute portal invasion (group 1) may be larger than in those
without it (group 0), especially when tumors are small.
Keywords: CT hepatic arteriography • CT with arterioportography • hepatocellular carcinoma • liver disease • portal venous invasion • prognosis
Introduction
Hepatocellular carcinoma (HCC) often invades the portal vein
[1]. Involvement of large
vessels, such as the main portal vein or the first- and second-order branches,
can be detected on dynamic CT
[2]. However, detecting portal
venous invasion at the peripheral branches (i.e., minute portal venous
invasion) on dynamic CT and MRI remains difficult
[3]. Radiologic findings
suggestive of peripheral portal venous invasion in cases of HCC have rarely
been examined to date; Imaeda et al.
[4] reported that a punctate or
linear low-attenuation structure adjacent to an HCC nodule on CT indicates a
tumor thrombus in the portal branches.
Portal venous invasion is thought to be one of the important prognostic
factors regulating recurrence and survival after surgical resection in HCC
patients
[5-9].
Microvascular invasion detected histologically, as well as macroscopic
vascular invasion detected by gross examination or on dynamic CT or MRI, also
has prognostic significance [3,
8,
10,
11]. For example, Marsh et al.
[12] reported that
microvascular and major vascular invasions were associated with a three- to
fourfold increased risk of HCC recurrence after liver transplantation for HCC.
In addition, Iwatsuki et al.
[13] stated that microvascular
and major vascular invasions were associated with a 4.4- and 15-fold increased
risk of HCC recurrence in patients with HCC treated with orthotopic liver
transplantation, respectively.
Radiologic detection of minute portal venous invasion may facilitate the
preoperative prediction of a patient's prognosis. We investigated whether
minute portal venous invasion by HCC can be diagnosed using combined CT
hepatic arteriography and CT with arterioportography (CTAP), which are thought
to be the most sensitive imaging techniques for detecting hemodynamic changes
in the liver [14].
Materials and Methods
Patients and Pathologic Classification
From medical data recorded at our hospital between September 1999 and
October 2003, 124 patients who underwent surgery for an untreated solitary,
hypervascular HCC were enrolled in our study. We referred to the initial
pathologic records of these 124 cases and excluded 36 patients with an HCC
nodule showing hepatic venous invasion or portal venous invasion at the level
of the main portal vein or the first- and second-order branches. The reason
HCC nodules with hepatic venous invasion were also excluded was that hepatic
venous invasion could have altered or modified CT hepatic arteriography and
CTAP findings associated with minute portal venous invasion. As a result, the
remaining HCC nodules were classified into one of two groups as follows: group
1, microscopic portal venous invasion was detected in the third-order or in
more peripheral branches and no hepatic venous invasion was identified; and
group 0, neither microscopic portal nor hepatic venous invasion was
identified.
Groups 1 and 0 included 15 and 70 cases, respectively. In this study, 30
consecutive cases were selected in order of date from the oldest for group 0.
The 45 patients included 35 men and 10 women, and their ages ranged from 39 to
82 years, with a mean of 65.8 years. The hepatitis B surface antigen was
present in six patients, and the hepatitis C virus antibody was present in 31
patients. Neither the hepatitis B surface antigen nor the hepatitis C virus
antibody was seen in the remaining eight patients. Liver dysfunction was
preoperatively evaluated using the Child-Pugh classification: 33 and 12 cases
were categorized as grade A or B, respectively. The surgical methods performed
included partial hepatectomy for 32 cases, subsegmentectomy for six cases, and
segmentectomy for seven cases. The histologic features of groups 1 and 0 are
summarized in Table 1.
Imaging Equipment and Techniques
Combined CT hepatic arteriography and CTAP was performed according to a
method reported previously
[15] within 1 month before
surgery as part of the preoperative angiographic examination in all patients.
CT was performed using a 4-MDCT unit (Aquilion, Toshiba). The scanning
parameters were as follows: collimation, 3 mm; pitch, 5.5; and reconstruction,
5 mm. For CTAP, data acquisition was initiated 30 seconds after initiation of
a transcatheter arterial injection of 100 mL of iopamidol solution (150 mg
I/mL [Io-pamiron 150, Nippon Schering]) at a rate of 2.5 mL/s using an
automated power injector. Before the contrast medium was injected into the
patient, a transarterial infusion of prostaglandin E1 (10 mg), used
as a vasodilator, was performed.
CT hepatic arteriography was performed approximately 5 minutes after CTAP.
A single-breath-hold data acquisition began 15 seconds (first phase) and 30
seconds (second phase) after the initiation of a transcatheter hepatic
arterial injection of 20-40 mL of Iopamiron 150 at a rate of 1-2 mL/s using an
automated power injector. The duration of arterial injection was 20 seconds.
The appropriate injection rate for CT hepatic arteriography was determined as
the maximal injection rate that would not lead to backward flow of the
contrast medium on hepatic angiography.
Assessment
The images of CT hepatic arteriography and CTAP were evaluated
independently by two abdominal radiologists and then were evaluated by
consensus review. On the images, a strongly enhanced area on first-phase CT
hepatic arteriography images was defined as a tumor, referring to its actual
macroscopic appearance. An area showing low attenuation on CTAP and high
attenuation on CT hepatic arteriography around the tumor was defined as an
area of peritumoral hemodynamic change. The shape of the area was initially
assessed and recorded for each case. Next, the ratio of the area of
peritumoral hemodynamic change volume relative to the tumor volume (area
volume-tumor volume ratio) was assessed. Qualitatively, the two radiologists
evaluated the CT images and categorized the ratio as one of three grades:
grade I, 10% or less; grade II, between 10% and 30%; or grade III, 30% or
more. Quantitatively, the volume measurement was performed on Windows XP
(Microsoft) with the aid of a software package (Scion Image Beta 4.02) by one
of the two radiologists.
The low-attenuation area on CTAP (A) and the tumor area
(B), as defined earlier, were manually traced by following the edges
of the respective areas in all sections associated with the tumor. Each volume
of A and B was calculated by multiplying the total square
measure obtained with the Scion software by the slice thickness of the images.
The area of peritumoral hemodynamic change volume—namely, (A -
B) (cm3)—was calculated; the area volume-tumor
volume ratio—namely, (A - B) / B x 100
(%)—was also calculated. The quantitative area volume-tumor volume
ratios were also categorized into one of three grades in the same manner as
defined earlier. Sensitivity, specificity, positive predictive value (PPV),
negative predictive value (NPV), and accuracy were calculated when either
grade III or grades II and III were considered to be indicators of minimal
portal venous invasion.
View larger version (10K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —Drawing shows shapes of areas of peritumoral hemodynamic change.
Three types of area of peritumoral hemodynamic change were as follows: a,
wedge-shaped area of peritumoral hemodynamic change with straight boundary
that continues toward peripheral portion from lateral side of tumor; b,
belt-shaped or irregular area of peritumoral hemodynamic change around tumor;
and c, linear area of peritumoral hemodynamic change projecting toward
peripheral portion.
|
|
View larger version (74K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —70-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 2.5 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 73.9%.
First-phase CT hepatic arteriography image reveals strongly enhanced area
suggestive of tumor itself at dome of right lobe. Wedge-shaped minimally
enhanced area (arrow) is also seen. Another wedge-shaped enhanced
area, seen in B, is not observed (arrowhead).
|
|
View larger version (76K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —70-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 2.5 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 73.9%.
Second-phase CT hepatic arteriography image reveals increase in enhancement of
wedge-shaped area (arrow) shown in A. Another wedge-shaped
enhanced area (arrowhead) is also visualized.
|
|
View larger version (68K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —70-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 2.5 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 73.9%. CT
arterioportography image shows focal portal perfusion defects (arrow
and arrowhead) that are equivalent to two wedge-shaped enhanced areas
seen in A and B.
|
|
The size and extent of minute portal venous invasion for each case of group
1 were microscopically reviewed on H and E-stained sections by one
pathologist.
Statistics
The frequency of each shape of area of peritumoral hemodynamic change was
compared between groups 1 and 0 using the Fisher's exact test. The volumes of
the area of peritumoral hemodynamic change and of the tumor itself were
compared between groups 1 and 0 using the Student's t test, whereas
the area volume-tumor volume ratio was similarly analyzed using the Student's
t test and Mann-Whitney U test. Each comparison was also
made when the maximal diameter of the tumor was limited to less than 3 cm or
was 3 cm or more. A p value of less than 0.05 was considered to
indicate a statistically significant difference.
Interobserver agreement with regard to the shape of the area of peritumoral
hemodynamic change for each case was assessed using unweighted kappa
statistics, whereas the qualitative grading for the area volume-tumor volume
ratio between the two abdominal radiologists and the agreement between
qualitative and quantitative grading for the area volume-tumor volume ratio
were analyzed using weighted kappa statistics. A kappa value of less than 0.20
indicated poor agreement; 0.20-0.39, fair agreement; 0.40-0.59, moderate
agreement; 0.60-0.79, substantial agreement; and 0.80 and higher, excellent
agreement.
Results
Qualitative Analysis
No direct finding suggestive of a tumor thrombus (e.g., a defect in the
portal vein) was detected on CTAP in any cases. Three different shapes of
areas of peritumoral hemodynamic change were identified (Figs.
1,
2A,
2B,
2C,
3A,
3B,
3C), and the frequency of each
type of area by consensus assessment is summarized in
Table 2. Although wedge-shaped
areas of peritumoral hemodynamic change tended to be more frequently observed
in group 1 than in group 0 when the diameter of the tumor was less than 3 cm,
no statistically significant difference was obtained. The kappa value,
representing interobserver agreement between the two readers, for wedge-shaped
areas of peritumoral hemodynamic change with a straight boundary continuing
toward the peripheral portion from the lateral side of tumor was 0.766; for
belt-shaped or irregular areas of peritumoral hemodynamic change around the
tumor, 0.821; and for linear areas of peritumoral hemodynamic change,
0.815.
View larger version (127K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —82-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 3.2 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 99.6%.
First-phase CT hepatic arteriography image shows strongly enhanced area
suggestive of tumor itself in right lobe. Belt-shaped (arrow) and
linear (arrowhead) minimally enhanced areas are visualized.
|
|
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —82-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 3.2 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 99.6%.
Second-phase CT hepatic arteriography image reveals increase in enhancement of
belt-shaped (arrow) and linear (arrowhead) enhanced
areas.
|
|
View larger version (118K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —82-year-old man with moderately differentiated hepatocellular
carcinoma (group 1) with maximal diameter of 3.2 cm. Ratio of volume of area
of peritumoral hemodynamic change to tumor volume was quantitatively 99.6%. CT
arterioportography image shows focal portal perfusion defects that are
equivalent to belt-shaped (arrow) and linear (arrowhead)
enhanced areas seen in A and B.
|
|
The numbers of each grade evaluated qualitatively with respect to the area
volumetumor volume ratio, which were obtained by consensus assessment, are
given in Table 3. The area
volume-tumor volume ratio of group 1 was significantly larger than that of
group 0 when tumor size was limited to less than 3 cm in diameter, according
to the Mann-Whitney U test (p = 0.046). The kappa value,
representing interobserver agreement between the two readers, for qualitative
grading of the area volume-tumor volume ratio was 0.892.
View this table:
[in this window]
[in a new window]
|
TABLE 3: Qualitative and Quantitative Between-Group Comparisons in the Ratio of
Volume of Area of Peritumoral Hemodynamic Change to Tumor Volume
|
|
Sensitivity, specificity, PPV, NPV, and accuracy calculated when
qualitative grade III or grades II and III were considered as an indicator of
minute portal venous invasion are summarized in Tables
4 and
5.
View this table:
[in this window]
[in a new window]
|
TABLE 4: Diagnostic Potential in Qualitative and Quantitative Analyses When Grade
III Was Considered as Indicator of Minute Portal Venous Invasion
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 5: Diagnostic Potential in Qualitative and Quantitative Analyses When
Grades II and III Were Considered as Indicators of Minute Portal Venous
Invasion
|
|
Quantitative Analysis
The volume of area of peritumoral hemodynamic change, the tumor volume, and
the area volume-tumor volume ratio, which were obtained using quantitative
analysis, are summarized in Table
6. The average area volume-tumor volume ratio of group 1 was
significantly larger than that of group 0 when tumor size was limited to less
than 3 cm in diameter, according to the Student's t test (p
= 0.02). The numbers of each grade evaluated quantitatively with respect to
the area volume-tumor volume ratio are given in
Table 3. The area volume-tumor
volume ratio of group 1 was significantly larger than that of group 0 when
tumor size was limited to less than 3 cm in diameter, according to the
Mann-Whitney U test (p = 0.049). The kappa value,
representing interobserver agreement between qualitative and quantitative
analyses, regarding the area volume-tumor volume ratio was 0.700.
View this table:
[in this window]
[in a new window]
|
TABLE 6: Volume of Area of Peritumoral Hemodynamic Change, Tumor Volume, and
Ratio of Volume of Area of Peritumoral Hemodynamic Change to Tumor Volume
(Area Volume-Tumor Volume Ratio)
|
|
Sensitivity, specificity, PPV, NPV, and accuracy calculated when either
quantitative grade III or grades II and III were considered as an indicator of
minute portal venous invasion are summarized in Tables
4 and
5. When grade III was used as
an indicator of minute portal venous invasion and tumor size was limited to
less than 3 cm in diameter, a PPV of 71.4% and an NPV of 75% were
achieved.
Pathologic Assessment
Twenty-two sites of minute portal venous invasion were microscopically
identified in the 15 cases in group 1. One site was found in 14 cases and
eight sites in the remaining one case. All sites of minute portal venous
invasion were detected adjacent to the main tumor. Only small clusters of
tumor cells were floating within the lumen of the portal venous branches in
all sites. The proportions of area occupied by tumor cells to that of the
invaded portal venous lumen were less than 5%, 10%, 20%, and 50% in 16 sites,
three sites, two sites, and one site, respectively. The tumor cells in the
portal vein were present away from the main tumor in all cases. There was no
continuity between the main tumor and the tumor cells in the minute portal
vein in any case.
Discussion
The detectability of minute portal venous invasion in HCC was first
evaluated using combined CT hepatic arteriography and CTAP. These techniques
are considered to be the most sensitive for the evaluation of hemodynamic
changes in the liver and were thus used in this study. The HCCs with portal
venous invasion enrolled in this study were limited to those in which
microscopic invasion was pathologically detected in the peripheral portal
veins (i.e., in a third-order or lower-order branch). It is difficult to
detect portal venous invasion at the peripheral branches on sonography,
dynamic CT, or MRI [3]. In
fact, no portal venous invasion was identified in the 15 cases with minute
portal invasion in group 1 on MDCT images obtained using a method reported
previously [16].
Three types of area of peritumoral hemodynamic change were detected in both
groups 1 and 0. A wedged-shaped area of peritumoral hemodynamic change may be
equal to a perilesional portal perfusion defect, with a "straight-line
sign" described in cases involving malignant hepatic tumors
[17]. On the other hand, a
belt-shaped or irregular area of peritumoral hemodynamic change around the
tumor is considered to be an area of blood draining into surrounding liver
parenchyma from an HCC. This effect is referred to as "corona
enhancement," which is identified after tumor enhancement
[18]. Only partially thick
corona enhancement located proximal to the tumor might be identified as a
belt-shaped or irregular area of peritumoral hemodynamic change. To the best
of our knowledge, a linear area of peritumoral hemodynamic change has not yet
been reported in detail.
Given that these types of areas of peritumoral hemodynamic change were
detected in both groups with similar frequency, an area of peritumoral
hemodynamic change may be considered to be almost equivalent to corona
enhancement with or without modification. Although the condition of
noncancerous liver parenchyma may be a possible factor that affects the
formation of an area of peritumoral hemodynamic change, there was no
significant bias in the severity of fibrosis between cases in groups 1 and 0
(Fisher's exact test, p = 0.283).
In this study, the area volume-tumor volume ratio was applied as an
indicator of the size of the areas of peritumoral hemodynamic change. The area
volume-tumor volume ratio in group 1 was larger than that in group 0,
especially when the diameter of the tumor was less than 3 cm. There are two
hypotheses that may explain this phenomenon. One possibility is that corona
enhancement is emphasized by a super-imposed arterioportal shunt caused by
portal venous obstruction due to tumor cells. The somewhat higher frequency of
wedge-shaped areas of peritumoral hemodynamic changes in group 1 may support
this hypothesis. The second possibility is that we are simply observing more
prominent venous drainage from tumors with more arterial blood
supply—namely, more hypervascular tumors. Judging from the microscopic
appearances that only small clusters of tumor cells were floating within the
lumen of portal venous branches and no tumor plugs distending the vessels were
identified in the sites of portal venous invasion in most cases in group 1,
the former hypothesis might be less likely than the latter.
Ueda et al. [18] suggested
that blood flow from an HCC drains to an area of corona enhancement via tumor
sinusoids, tiny vessels in the inner layer and portal veins in the outer layer
of the pseudocapsule, and portal venules in the adjacent liver parenchyma. An
area of corona enhancement is considered to be the first area for tumor spread
via the portal system. Corona enhancement could be amplified by the occurrence
of portal venous invasion, although its precise mechanism is unknown.
Why, then, did we not observe any difference in the area volume-tumor
volume ratios between groups 1 and 0 when tumors with a diameter of 3 cm or
more were considered in the analysis? This finding may be explained as
follows: Arterioportal shunts or hepatic arterial buffer response
[19] due to compression of
portal venous branches may occur around a tumor unrelated to the presence of
portal venous invasion when it gets larger. In fact, the frequency of
wedged-shaped areas of peritumoral hemodynamic change in group 0 was
significantly higher when the tumor diameter was 3 cm or more (seven of 13)
than when it was less than 3 cm (two of 17 cases) (Fisher's exact test,
p = 0.02).
We examined how correctly we can diagnose minute portal venous invasion
using the area volume-tumor volume ratio. We could achieve a relatively high
NPV (71.0%) and a low PPV (42.3%) when grade III was considered as an
indicator. When the tumor size was limited to less than 3 cm, the NPV and PPV
were increased to 75.0%, and 71.4%, respectively, and showed acceptable
values. Minute portal venous invasion can be diagnosed using the area
volume-tumor volume ratio with relatively high probability when a tumor is
small. Thus, combined CT hepatic arteriography and CTAP may be a useful tool
in the preoperative assessment of minute portal venous invasion of HCC.
In this study, we evaluated the area volume-tumor volume ratio both
qualitatively and quantitatively. The qualitative analysis showed results
similar to those obtained in the quantitative analysis, and agreement between
the two analyses was also substantial. The relative size of an area of
peritumoral hemodynamic change can be estimated fairly precisely even in the
qualitative analysis.
One limitation of our study is that it was difficult to determine the tumor
contour in some cases because the area of peritumoral hemodynamic change
itself was also slightly enhanced on first-phase CT hepatic arteriography. The
second limitation is that the time course and transition of enhancement of an
area of peritumoral hemodynamic change could not be evaluated accurately.
These two limitations arose from the constant scan timing of CT hepatic
arteriography used in our protocol. The velocity of hepatic arterial flow is
variable in each patient, and the location of the tumor also may reflect the
enhancement pattern because the entire liver was scanned according to our
protocol. Single-level dynamic CT hepatic arteriography is considered to be
helpful in the evaluation of time course and transition of enhancement.
In conclusion, the area of peritumoral hemodynamic change of an HCC with
minute portal venous invasion may be larger than that of an HCC without such
invasion, especially when a tumor is small. Combined CT hepatic arteriography
and CTAP may thus be useful in the preoperative assessment of minute portal
venous invasion of HCC.
Acknowledgments
We thank Yoshihiko Maehara, Department of Surgery and Science, Kyushu
University, for providing the clinical information for this manuscript. We
also thank Masazumi Tsuneyoshi, Department of Anatomic Pathology, Kyushu
University, for providing the pathologic information for this manuscript.
References
- Albacete RA, Matthews MJ, Saini N. Portal vein thromboses in
malignant hepatoma. Ann Intern Med 1967;67
: 337-348[Abstract/Free Full Text]
- Mathieu D, Grenier P, Larde D, Vasile N. Portal vein involvement in
hepatocellular carcinoma: dynamic CT features.
Radiology 1984;152
: 127-132[Abstract/Free Full Text]
- Esnaola NF, Lauwers GY, Mirza NQ, et al. Predictors of
microvascular invasion in patients with hepatocellular carcinoma who are
candidates for orthotopic liver transplantation. J Gastrointest
Surg 2002; 6:224
-232[CrossRef][Medline]
- Imaeda T, Yamawaki Y, Hirota Y, Suzuki M, Seki M, Doi H. Tumor
thrombus in the branches of the distal portal vein: CT demonstration.
J Comput Assist Tomogr 1989;13
: 262-268[Medline]
- Shimada M, Takenaka K, Taguchi K, et al. Prognostic factors after
repeat hepatectomy for recurrent hepatocellular carcinoma. Ann
Surg 1998; 227:80
-85[CrossRef][Medline]
- Shimada M, Hasegawa H, Gion T, et al. Risk factors of the
recurrence of hepatocellular carcinoma originating from residual cancer cells
after hepatectomy. Hepatogastroenterology1999; 46:2469
-2475[Medline]
- Nagasue N, Ono T, Yamanoi A, et al. Prognostic factors and survival
after hepatic resection for hepatocellular carcinoma without cirrhosis.
Br J Surg 2001;88
: 515-522[CrossRef][Medline]
- Lauwers GY, Terris B, Balis UJ, et al. Prognostic histologic
indicators of curatively resected hepatocellular carcinomas: a
multiinstitutional analysis of 425 patients with definition of a histologic
prognostic index. Am J Surg Pathol 2002;26
: 25-34[CrossRef][Medline]
- Ikai I, Arii S, Kojiro M, et al. Reevaluation of prognostic factors
for survival after liver resection in patients with hepatocellular carcinoma
in a Japanese nationwide survey. Cancer2004; 101:769
-802
- Ochiai T, Sonoyama T, Ichikawa D, et al. Poor prognostic factors of
hepatectomy in patients with resectable small hepatocellular carcinoma and
cirrhosis. J Cancer Res Clin Oncol 2004;130
: 197-202[CrossRef][Medline]
- Yang Y, Nagano H, Ota H, et al. Patterns and clinicopathologic
features of extrahepatic recurrence of hepatocellular carcinoma after curative
resection. Surgery 2007;141
: 196-202[CrossRef][Medline]
- Marsh JW, Dvorchik I, Bonham CA, Iwatsuki S. Is the pathologic TNM
staging system for patients with hepatoma predictive of outcome?
Cancer 2000; 88:538
-543[CrossRef][Medline]
- Iwatsuki S, Dvorchik I, Marsh JW, et al. Liver transplantation for
hepatocellular carcinoma: a proposal of a prognostic scoring system.
J Am Coll Surg 2000;191
: 389-394[CrossRef][Medline]
- Chezmar JL, Barnardino ME, Kaufman SH, Nelson RC. Combined CT
arterial portography and CT hepatic angiography for evaluation of the hepatic
resection candidate. Radiology 1993;189
: 407-410[Abstract/Free Full Text]
- Tajima T, Honda H, Taguchi K, et al. Sequential hemodynamic change
in hepatocellular carcinoma and dysplastic nodules: CT angiography and
pathologic correlation. AJR 2002;178
: 885-897[Abstract/Free Full Text]
- Asayama Y, Yoshimitsu K, Irie H, et al. Poorly versus moderately
differentiated hepatocellular carcinoma: vascularity assessment by CT hepatic
angiography in correlation with histologically counted number of unpaired
arteries. J Comput Assist Tomogr 2007;31
: 188-192[CrossRef][Medline]
- Kanematsu M, Hiroaki H, Tetsuya Y, et al. Overestimating the size
of hepatic malignancy on helical CT during arterial portography: equilibrium
phase CT and pathology. J Comput Assist Tomogr1997; 21:713
-719[CrossRef][Medline]
- Ueda K, Matsui O, Kawamori Y, et al. Hypervascular hepatocellular
carcinoma: evaluation of hemodynamics with dynamic CT during hepatic
arteriography. Radiology 1998;206
: 161-166[Abstract/Free Full Text]
- Lautt WW. Mechanism and role of intrinsic regulation of hepatic
arterial blood flow: hepatic arterial buffer response. Am J
Physiol 1985; 249(5 Pt
1): G549-G556[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
