AJR 2005; 184:1085-1090
© American Roentgen Ray Society
Focal Eosinophilic Necrosis Versus Metastasis in the Liver: The Usefulness of Two-Phase Dynamic CT
Jin Hur1,
Mi-Suk Park1,
Jeong-Sik Yu1,
Joon-Suk Lim1,
Soon Won Hong2 and
Ki Whang Kim1
1 Department of Diagnostic Radiology and Research Institute of Radiological
Science, Yonsei University College of Medicine and YongDong Severance
Hospital, 146-92, Dogok-Dong, Kangnam-Ku, Seoul 135-270, South Korea.
2 Department of Pathology, Yonsei University College of Medicine, Seoul, South
Korea.
Received January 31, 2004;
accepted after revision July 1, 2004.
Address correspondence to M-S Park.
Abstract
OBJECTIVE. The purpose of our study was to evaluate the usefulness
of dual-phase dynamic CT in the differentiation of focal eosinophilic necrosis
of the liver and metastasis.
CONCLUSION. Undetected isoattenuating lesions on arterial phase
images that have an indistinct margin, a nonspherical shape, and a homogenous
enhancement pattern on portal venous phase images suggest focal eosinophilic
necrosis rather than metastasis. Two-phase dynamic CT was found to be useful
at differentiating focal eosinophilic necrosis from metastasis.
Introduction
Focal eosinophilic necrosis of the liver is a focal hepatic lesion
caused by eosinophil-related tissue damage and is associated with various
eosinophilia-related conditions such as parasitic infestations, allergic
reactions, hypereosinophilic syndrome, and neoplasms
[1-3].
Several reports have suggested that eosinophils can cause tissue damage by
infiltrating the liver, mainly into the periportal space
[4,
5]. Although the detailed
mechanisms of eosinophil-related tissue damage are not fully understood, the
process might occur as follows: eosinophils infiltrate tissue, causing damage
related to eosinophil function and the products of eosinophils (e.g.,
eosinophil major basic protein and eosinophil cationic protein), leading to
thromboembolic phenomena
[6].
Focal eosinophilic necrosis of the liver was mainly observed as small,
multiple hypoattenuating foci on portal venous phase CT images, simulating
hypovascular metastasis, mainly from adenocarcinoma
[4,
7]. Moreover, focal
eosinophilic necrosis of the liver is often found in patients with underlying
malignant tumors such as gastrointestinal carcinoma, lymphoma, or leukemia
[8,
9]. Such lesions are frequently
encountered as small, multiple, hypoattenuating foci on the liver in patients
with underlying malignancy, and these lesions present diagnostic dilemmas.
The recent widespread use of two-phase helical CT during the hepatic
arterial phase and the portal venous phase has improved the detection and
characterization of focal hepatic lesions. However, to our knowledge, no
comparative studies have been performed of dual-phase helical CT findings in
focal eosinophilic necrosis and metastasis.
The purpose of this study was to evaluate the usefulness of dual-phase CT
in the differentiation of focal eosinophilic necrosis of the liver and
metastasis in patients with underlying malignancy.
Materials and Methods
Patient Selection
By performing a computerized search of medical, radiologic, and pathologic
records, we identified 57 patients with a diagnosis of focal eosinophilic
necrosis of the liver at two institutions from 1999 to 2003. The diagnosis of
focal eosinophilic necrosis was verified by biopsy (n = 18) or by
consistent clinical findings (n = 39) as follows: an association with
initial peripheral eosinophilia (>10%) and spontaneous disappearance on the
follow-up CT with a normalized peripheral blood eosinophil count. Patients who
underwent single-phase CT (n = 19), who had no underlying malignancy
(n = 11), or who underwent chemotherapy during the follow-up period
(n = 9) were excluded. Finally, 18 patients with focal eosinophilic
necrosis (131 lesions) who underwent dual-phase helical CT were included. All
lesions that were included in this study were smaller than 20 mm. These
patients included 13 men and five women, ranging in age from 29 to 71 years
(mean age, 46 years). In these 18 patients, the underlying malignancy was
rectal cancer (n = 5), colon cancer (n = 4), stomach cancer
(n = 8), or pancreatic cancer (n = 1). The diagnosis of
focal eosinophilic necrosis was verified by biopsy in eight of 18 patients and
clinically in the remaining 10. All patients with focal eosinophilic necrosis
were followed up with CT. Follow-up CT scans in eight patients with
biopsy-proven focal eosinophilic necrosis were obtained 3-12 months (mean, 6
months) after the biopsy and in the remaining 10 patients were obtained 1-6
months (mean, 2 months) after the initial CT examinations. The peripheral
eosinophilic count ranged from 10% to 51% (mean, 24%).
For comparison, we searched the pathology records for patients with
pathologically proven hepatic metastasis from adenocarcinoma during the same
periods with focal eosinophilic necroses of similar size that were smaller
than 20 mm in maximum diameter. Patients who had hepatic metastasis, but not
from adenocarcinoma, who underwent single-phase CT, whose lesions were larger
than 20 mm were excluded. We identified 19 patients (56 lesions) with
pathologically proven hepatic metastases who had undergone dual-phase helical
CT and whose lesions were smaller than 20 mm. The 19 patients with hepatic
metastasis included 12 men and seven women ranging in age from 39 to 74 years
(mean age, 52 years). In these 19 patients, the primary malignancy was rectal
cancer (n = 5), colon cancer (n = 5), stomach cancer
(n = 7), and pancreatic cancer (n = 2). The peripheral
eosinophilic count ranged from 3% to 13% (mean, 6%).
CT Technique
Dual-phase CT examinations were performed with a helical CT scanner
(HiSpeed CT/I; GE Healthcare) after the IV administration of 150 mL of
nonionic contrast material (Ultravist 300 [iopromide], Schering) using a power
injector (EnVisionCT, Medrad) at a rate of 3 mL/sec. After the infusion of
contrast material, arterial phase CT was started after a delay of 25-30 sec,
and portal venous phase CT was started after a delay of 65-70 sec. All scans
were acquired in the cephalocaudal direction. The section thickness was 5-7
mm, and the incremental table speed was 5-7 mm/sec.
Image Analysis
Two abdominal radiologists collectively and retrospectively reviewed the
dual-phase helical CT images of 37 patients by consensus, without knowledge of
the final diagnoses. They first evaluated portal venous phase images and then
arterial phase images. The lesions that were detected on portal venous phase
images were included in our study as the reference lesion. Lesion number and
attenuation were evaluated on both arterial phase and portal venous phase
images. Attenuation was assessed as being at one of three levels:
hypoattenuation (less than that of the hepatic parenchyma), isoattenuation
(similar to that of the adjacent hepatic parenchyma), and hyperattenuation
(greater than that of the hepatic parenchyma). For quantitative analysis, the
lesion-to-liver contrast was calculated on arterial phase and portal venous
phase images. Contrast-enhanced mean CT attenuation values (in Hounsfield
units) of lesions and of the liver were obtained by region-of-interest
analysis on both arterial phase and portal venous phase CT images. CT
attenuation values were measured twice by two radiologists respectively and
averaged for each lesion. The difference between the mean attenuation of the
lesion and liver (the lesion- to-liver contrast) was calculated on arterial
phase and portal venous phase CT images. In cases of isoattenuating lesions,
the lesion-to-liver contrast was considered to be 10 H, on the basis of the
belief that the threshold of visual gray-scale distinction is at least 10-15 H
[10].
The size, margin, shape, and enhancement patterns were evaluated on portal
venous phase images. Lesion size of the longest diameter was measured twice by
two radiologists respectively and averaged for each lesion. The margin was
categorized as discrete (a well-demarcated, sharp margin) or indistinct (a
poorly demarcated, blurred margin). Shapes were divided into two groups,
spherical and nonspherical; and enhancement patterns were classified into
three types: homogenous, rim enhancement (hypoattenuation at the center
surrounded circumferentially by a less hypoattenuating border), and mixed
(heterogeneous attenuation that could not be categorized either as homogeneous
or as a targetlike enhanced pattern).
We compared these parameters for focal eosinophilic necrosis and metastasis
using an unpaired one-tailed Student's t test. A p value of
less than 0.01 was considered to indicate statistical significance.
Results
Of the 131 lesions of focal eosinophilic necrosis on portal venous phase CT
images, only 43 lesions (33%) were detected on arterial phase CT images (Figs.
1A, and
1B). Of 56 metastatic lesions
on portal venous phase CT images, 47 lesions (84%) were detected on arterial
phase CT images (Figs. 2A, and
2B). The detection rates of
focal eosinophilic necrosis and of metastasis on arterial phase images were
significantly different (p < 0.01). On arterial phase CT images,
hypoattenuating lesions were far more likely in metastasis than in focal
eosinophilic necrosis (Table
1).
View larger version (105K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —Focal eosinophilic necrosis of liver in 36-year-old man with
early gastric cancer and peripheral eosinophilia (13%). Hepatic artery phase
CT image reveals isoattenuating lesion (arrow) in right hepatic lobe.
Mean lesion-to-liver contrast was 14 ± 7 H.
|
|
View larger version (116K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Focal eosinophilic necrosis of liver in 36-year-old man with
early gastric cancer and peripheral eosinophilia (13%). Corresponding portal
venous phase CT image shows more discrete, nonspherical hypoattenuating lesion
(arrow) without rim enhancement. Mean lesion-to-liver contrast was 36
± 8 H.
|
|
View larger version (113K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A. —Metastasis from gastric cancer in 49-year-old man. Hepatic
artery phase CT scan reveals single focal hepatic lesion (arrow) with
discrete margin, spherical shape, and targetlike hypoattenuation in right lobe
of liver. Mean lesion-to-liver contrast was 28 ± 9 H.
|
|
View larger version (116K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —Metastasis from gastric cancer in 49-year-old man.
Corresponding portal venous phase CT image shows hypoattenuating spherical
lesion (arrow) with discrete margin. Mean lesion-to-liver contrast
was 43 ± 8 H.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 1 Detection Rate and Attenuation of Focal Eosinophilic Necrosis (FEN) and
Metastasis of the Liver in 37 Patients on Dual-Phase Helical CT
|
|
The lesion-to-liver contrast on arterial phase and portal venous phase
images was greater in metastasis (25 ± 7 H and 46 ± 6 H,
respectively) than in focal eosinophilic necrosis (13 ± 8 H and 33
± 7 H, respectively) (Table
2). The lesion-to-liver contrasts of focal eosinophilic necrosis
and metastasis were also significantly different (p < 0.01).
View this table:
[in this window]
[in a new window]
|
TABLE 2 Lesion-to-Liver Contrast on Both Arterial Phase and Portal Venous Phase
CT Images of Focal Eosinophilic Necrosis (FEN) and Metastasis of the Liver in
37 Patients
|
|
The mean lesion sizes of focal eosinophilic necrosis and metastasis were
0.98 cm and 1.17 cm, respectively, which were not significantly different
(p > 0.05). Focal eosinophilic necrosis more frequently showed an
indistinct margin (79% vs. 22%), with a nonspherical shape (82% vs. 29%) than
did metastasis (Figs. 3A,
3B, and
3C). The proportion of rim
enhancement pattern in metastasis was 75% (42/56 lesions) and that in focal
eosinophilic necrosis was 12% (16/131 lesions) (Figs.
4A, and
4B). Focal eosinophilic
necrosis more frequently showed a homogenous enhancement pattern (60%, 79/131
lesions) (Table 3). Lesion
shapes, margins, and enhancement were significantly different for focal
eosinophilic necrosis and metastasis (p < 0.01).
View larger version (106K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A. —Focal eosinophilic necrosis of liver in 55-year-old man with
early gastric cancer. Peripheral eosinophilia was 50%. Hepatic arterial phase
CT image does not show any recognizable lesion.
|
|
View larger version (102K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —Focal eosinophilic necrosis of liver in 55-year-old man with
early gastric cancer. Peripheral eosinophilia was 50%. Corresponding portal
venous phase CT image reveals multiple focal hypoattenuating hepatic lesions
(arrows) with indistinct margins and nonspherical shapes but without
rim enhancement.
|
|
View larger version (164K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C. —Focal eosinophilic necrosis of liver in 55-year-old man with
early gastric cancer. Peripheral eosinophilia was 50%. Photomicrograph of core
needle biopsy specimen shows hepatocellular necrosis and innumerable
inflammatory cell infiltrates predominantly composed of eosinophils and
histiocytes. No tumor cell is identifiable. (H and E, x400)
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 3 Appearance of Focal Eosinophilic Necrosis (FEN) and Metastasis of the
Liver on Portal Venous Phase Images
|
|
Discussion
Localized eosinophilic infiltration into the liver is an uncommon entity
that is characterized by multiple focal lesions. Eosinophilic infiltration can
occur in various conditions such as parasitic infestations, neoplastic
diseases, allergy, drug hypersensitivity, and hypereosinophilic syndrome
[1-3].
"Focal eosinophilic necrosis" is a descriptive term based on
pathologic features and has been referred to in previous reports as
"eosinophil-related hepatic necrosis," "eosinophilic hepatic
necrosis," and "focal eosinophilic abscess"
[4,
11,
12]. The lesion is
characterized by eosinophilic infiltration, mainly in the periportal space,
with necrotic tissue or abscess on pathology.
We frequently encounter small (< 20 mm), multiple hypoattenuating foci
in the liver in patients with underlying malignancy. The lesions may be benign
or malignant and include simple cysts, microhamartomas, cavernous hemangiomas,
focal eosinophilic necrosis, and metastases
[13-15].
Sonography and MRI are recognized problem-solving complementary studies
because they accurately characterize cysts, microhamartomas, and cavernous
hemangiomas.
However, in cases of solid lesions, sometimes they are nonspecific in
imaging findings and give rise to diagnostic dilemmas, especially in patients
with an underlying malignancy. In that situation, we must differentiate
metastasis from other benign conditions, including focal eosinophilic
necrosis. Recently, we have seen many cases of focal eosinophilic necrosis in
the liver in patients with underlying malignancy. Peripheral eosinophilia
would be helpful and give a clue in differentiating focal eosinophilic
necrosis from metastasis. However, sometimes, the peripheral eosinophilic
counts in metastasis and focal eosinophilic necrosis overlap (10-13% in our
study). For those cases, biopsy is needed for definitive diagnosis. However,
biopsy is an invasive procedure and is difficult for small focal lesions,
because some lesions are hard to detect and some lesions are hard to approach
on sonography. Therefore, we think it is worthwhile to differentiate
metastasis from focal eosinophilic necrosis on the basis of CT findings before
biopsy, even though CT cannot replace biopsy.
In our study, all lesions of focal eosinophilic necrosis in patients with
underlying malignancy were smaller than 20 mm in maximum diameter; therefore,
we evaluated metastatic lesions of a similar size. We evaluated several
lesions larger than 20 mm in cases of hypereosinophilic syndrome and parasitic
infections without an underlying malignancy. Those lesions were excluded from
our study. Moreover, because we wanted to assess small hepatic lesions, which
are difficult to prove pathologically, and evaluate any feature that can help
in differentiating between focal eosinophilic necrosis and metastasis in
practice, we excluded lesions larger than 20 mm.
Helical CT can provide a time window for optimal contrast enhancement of
the hepatic parenchyma and focal hepatic lesions, allowing a two- or
three-phase dynamic study. Thus, we undertook a study to differentiate focal
eosinophilic necrosis from metastasis using dual-phase helical CT. Our results
show that most focal eosinophilic necrotic lesions are isoattenuating on
arterial phase CT. On the other hand, most metastases are hypoattenuating on
arterial phase CT. However, both focal eosinophilic necrosis and metastasis
show hypoattenuation on portal venous phase CT. Hypovascular hepatic lesions
are most clearly delineated during the portal venous phase, when the hepatic
parenchyma is maximally enhanced.
Therefore, we suggest that dual-phase helical CT is more useful in the
differentiation of focal eosinophilic necrosis and metastasis than
single-phase (portal) CT. The precise reason that focal eosinophilic necrosis
lesions appear isoattenuating on arterial phase CT and, metastasis,
hypoattenuating is not understood. In the case of metastasis, the metastatic
tumor compresses the hepatic parenchyma, portal vein, and hepatic vein, and
the corresponding area is supplied by arterioportal communication
[16]. So hemodynamic change
around a metastatic lesion, such as an increased blood flow through the
arterioportal communication, may be an explanation. However, in the case of
focal eosinophilic necrosis, eosinophils infiltrate along the periportal space
without deranging arterial blood flow, thus causing no enhancement difference
between the lesion and the surrounding liver parenchyma on arterial phase CT.
We suggest that these different pathophysiologies contribute to their
different enhancement patterns on arterial phase imaging.
Several studies have reported radiologic findings of focal eosinophilic
necrosis of the liver as hypodense nodules with poorly defined margins on
portal venous phase CT [4,
7,
17]. However, findings on
arterial phase CT vary among studies. Lee et al.
[4] reported that all foci of
eosinophil-related necrosis are observed as focal hypoattenuating lesions on
all phases of helical CT, but according to Yoo et al.
[7], all lesions on CT showed
low attenuation in the portal or delayed phase and variable enhancement
patterns (isoattenuation in eight cases, hypoattenuation in six, and
hyperattenuation in one) in the arterial phase. In our series of 131 portal
venous phase-detected focal eosinophilic necrosis lesions, only 43 lesions
(33%) were detected on arterial phase CT as hypoattenuating lesions; the
remaining 88 lesions (67%) were not detected on arterial phase CT and were
isoattenuating. In our study, several hyperattenuating lesions on arterial
phase images were not detected on the corresponding portal venous phase
images. We are not convinced that those hyperattenuating lesions on arterial
phase images are true focal eosinophilic necrosis lesions. We think that they
may be other conditions such as nontumorous arterioportal shunts accompanied
by inflammation or the early stage of focal eosinophilic necrosis. Moreover,
because none of them was detected on sonography, we could not biopsy the
lesions. Therefore, we excluded those lesions from our study to avoid
confusion.
In our study, lesion-to-liver contrast was significantly greater for
metastasis (25 ± 7 H, 46 ± 6 H) than for focal eosinophilic
necrosis (13 ± 8 H, 33 ± 7 H) on both arterial phase and portal
venous phase images. However, because of the lower lesion-to-liver contrast in
focal eosinophilic necrosis, these lesions tended to be fainter on CT images.
On the other hand, metastatic lesions were more distinct than focal
eosinophilic necrosis lesions on CT images, a finding supported by their
higher lesion-to-liver contrast, which explains why the detection rate of
metastatic lesions was higher (84% [47/56] metastatic lesions vs 33% [43/131]
focal eosinophilic necrosis) on arterial phase CT images.
Jang et al. [17] reported
the different imaging features of metastasis and focal eosinophilic necrosis
of the liver based on single-phase CT. They reported that focal hepatic
lesions with an indistinct margin, a nonspherical shape, and subtle
hypoattenuation without rim enhancement are likely to be focal eosinophilic
necrosis. Our series produced similar results. The results of our study show
that lesions of focal eosinophilic necrosis are more likely to have an
indistinct margin and to be nonspherical in shape than those of metastasis.
Pathologically, focal eosinophilic necrosis is a focal area of hepatocellular
necrosis caused by severe eosinophilic infiltration of the perivascular space,
which may explain their frequently noted irregular shapes and indistinct
margins. In our study, targetlike or rim enhancement was seen much more
frequently in metastasis (75%). In focal eosinophilic necrosis, homogeneous
hypoattenuation was seen in 60% (79/131) of lesions, whereas targetlike or rim
enhancement was seen in only 12% (16/131). Thus, enhancement patterns may be a
good means of discriminating metastasis from focal eosinophilic necrosis.
Our study has several limitations. First, not all focal eosinophilic
necrosis lesions were pathologically proven, especially in cases of multiple
lesions, because focal lesions that were detected on helical CT were not
always visible on sonography. So in patients who had multiple lesions, we
biopsied two or three lesions that were detected on sonography and could be
easily approached. Although we confirmed the benign nature of the lesions by
observing their complete resolution during follow-up CT, we still cannot be
absolutely certain whether other transient benign conditions existed. Second,
our study included only patients with adenocarcinoma, and whether results
obtained on this basis can be generally applied to patients with various types
of malignancies is open to debate. Third, because performing dual-phase CT
increases the radiation dose and focal eosinophilic necrosis dose not occur
very often, the problem exists of performing dual-phase CT in all patients
with underlying malignancy. We suggest that dual-phase CT be performed only in
selected cases and not in every patient. A blood test of eosinophilia could
provide a clue, but it is not always possible to perform a blood test before
performing CT, and there could be an overlap.
In conclusion, on dual-phase dynamic CT studies, the higher proportion of
undetected isoattenuating lesions on arterial phase images that were seen as
hypoattenuating lesions on portal venous phase images may suggest focal
eosinophilic necrosis rather than metastasis. Moreover, the characteristic
radiologic features of a poorly defined margin, an irregular shape, and
homogeneous enhancement without rim enhancement may suggest focal eosinophilic
necrosis rather than metastasis. Therefore, we conclude that dual-phase
dynamic CT is useful for differentiating focal eosinophilic necrosis from
metastasis. However, there is still considerable overlap between the two
diseases, and biopsy should be performed in equivocal cases.
References
- Cha SH, Park CM, Cha IH, et al. Hepatic involvement in
hypereosinophilic syndrome: value of portal venous phase imaging.
Abdom Imaging1998; 23:154
-157[Medline]
- Lim JH, Lee WJ, Lee DH, Nam KJ. Hypereosinophilic syndrome: CT
findings in patients with hepatic lobar or segmental involvement.
Korean J Radiol2000; 1:98
-103[Medline]
- Kim GB, Kwon JH, Kang DS. Hypereosinophilic syndrome: imaging
findings in patients with hepatic involvement. AJR1993; 161:577
-580[Abstract/Free Full Text]
- Lee WJ, Lim HK, Lim JH, Kim SH, Choi SH, Lee SJ. Foci of
eosinophil-related necrosis in the liver: imaging findings and correlation
with eosinophilia. AJR1999; 172:1255
-1261[Abstract/Free Full Text]
- Won JH, Kim MJ, Kim BM, et al. Focal eosinophilic infiltration of
the liver: a mimick of hepatic metastasis. Abdom
Imaging 1999;24:362
-372
- Fauci AS, Harley JB, Roberts WC, Ferrans VJ, Gralnick HR, Bjornson
BH. The idiopathic hypereosinophilic syndrome: clinical, pathophysiologic, and
therapeutic considerations. Ann Intern Med1982; 97:78
-92
- Yoo SY, Han JK, Kim YH, Kim TK, Choi BI, Han MC. Focal eosinophilic
infiltration in the liver: radiologic findings and clinical course.
Abdom Imaging2003; 28:326
-332[Medline]
- Iwasaki K, Torisu M, Fujimura T. Malignant tumor and eosinophils.
Cancer 1986;58:1321
-1327[Medline]
- Hong SW, Cho MY, Park C. Expression of eosinophil chemotactic
factors in stomach cancer. Yonsei Med J1999; 40:131
-136[Medline]
- Baron RL. Understanding and optimizing use of contrast material for
CT of the liver. AJR1994; 163:323
-331[Abstract/Free Full Text]
- Hong SW, Kim HG, Park CI, Lee SI. Eosinophilic liver abscess in
patients with gastric cancer. Korean J Pathol1993; 27:27
-33
- Kjell-Arne U, Helen R, Rolf O. Eosinophilic hepatic necrosis in
hypereosinophilic syndrome. J Clin Gastroenterol2000; 31:323
-327[Medline]
- Schwartz LH, Gandras EJ, Colangelo SM, Ercolani MC, Panicek DM.
Prevalence and importance of small hepatic lesions found at CT in patients
with cancer. Radiology1999; 210:71
-74[Abstract/Free Full Text]
- Jones EC, Chezmar JL, Nelson RC, Bernardino ME. The frequency and
significance of small hepatic lesions detected by CT.
AJR 1992;158:535
-539[Abstract/Free Full Text]
- Jang HJ, Lim HK, Lee WJ, Lee SJ, Yun JY, Choi D. Small
hypoattenuating lesions in the liver on single-phase helical CT in
preoperative patients with gastric and colorectal cancer: prevalence,
significance, and differentiating features. J Comput Assist
Tomogr 2002;26:718
-724[Medline]
- Terayama N, Matsui O, Ueda K, et al. Peritumoral rim enhancement of
liver metastasis: hemodynamics observed on single-level dynamic CT during
hepatic arteriography and histopathologic correlation. J Comput
Assist Tomogr 2002;26:975
-980[Medline]
- Jang HJ, Lee WJ, Lee SJ, Kim SH, Lim HK, Lim JH. Focal eosinophilic
necrosis of the liver in patients with underlying gastric or colorectal
cancer: CT differentiation from metastasis. Korean J
Radiol 2002;3:240
-244[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
