10336_ftp

8/3/2019 10336_ftp

http://slidepdf.com/reader/full/10336ftp 1/7

Original Research

Hepatocellular Carcinoma of Diffuse Type: MR

Imaging Findings and Clinical Manifestations

Masayuki Kanematsu, MD, Richard C. Semelka, MD,* Polytimi Leonardou, MD,

Maria Mastropasqua, MD, and Joseph K.T. Lee, MD

Purpose: To assess MR imaging findings and clinical man-ifestations of diffuse-type hepatocellular carcinoma (HCC).

Materials and Methods: We retrospectively reviewed our

experience with diffuse HCC from November 1994 to Octo- ber 2001. MR imaging findings and clinical features were

assessed.

Results: Twenty-two consecutive patients with diffuse-typeHCC (19 men and three women, age range 16 –80 years

[mean, 52 years]) were identified in a review of liver MR stud-

ies. This represented 13% of all patients with HCC imagedduring this time period. Diffuse HCC showed a permeative,

infiltrative pattern with ill-defined borders and no evidence of

convex margination in all cases. At least 50% of the liver volume was involved with tumor. Diffuse-type HCC showed

hypointensity in 15 patients, mixed intensity in three, and

isointensity in four on T1-weighted images; heterogeneoushyperintensity in 16 patients; and homogeneous hyperinten-

sity in six on T2-weighted MR images. Diffuse-type HCC

showed patchy enhancement in 12 patients, miliary enhance-

ment in nine, and minimal enhancement in one on postcon-trast early-phase images, and showed heterogeneous wash-

out in all patients on postcontrast late-phase images.Proximal portal venous tumor thrombosis was seen in allpatients. Serum -fetoprotein (AFP) value was elevated (10

ng/mL) in 14 of 18 patients, and 13 showed a value greater

than 500 ng/mL. The four patients who did not have elevated AFP had tumors which were indistinguishable from those in

patientswith elevatedAFP; they also didnot have a distinctive

clinical history.

Conclusion: Diffuse-type HCC was typically seen as an

extensive, heterogeneous permeative hepatic tumor, with

portal venous tumor thrombosis on MR images in all cases.Early enhancement, observed as patchy in 12 and miliary

in nine of 22 patients, was a distinctive imaging feature.

Elevated serum AFP value was a common finding; however,22% had normal values.

Key Words: Magnetic resonance; hepatocellular carcino-ma; diffuse type; cirrhosis; contrast enhancement

J. Magn. Reson. Imaging 2003;18:189–195.

© 2003 Wiley-Liss, Inc.

HEPATOCELLULAR CARCINOMA (HCC) is the most common primary malignant neoplasm arising in pa-tients with chronic liver damage, and is most oftenrelated to hepatitis virus infection (1,2), alcohol abuse(3), or iron overload (4,5). A 1984 pathologic report (6)described that the common major gross patterns wereexpanding, spreading, and multifocal by means of ag-gregating a total of 529 HCC cases from Japan, theUnited States, and South Africa. Further modificationshave divided HCC into infiltrative or expansive, singleor multinodular, and mixed types, using observationson encapsulation and intrahepatic venous spread (7).

Pathologically, diffuse-type HCC has been consideredas a tumor that spreads throughout most of the liver

and is not recognized as a focal tumor, typically accom-panied by extensive portal venous tumor thrombosisand substantial elevation of serum -fetoprotein (AFP) value (8). Although many reports have described theMR appearance of focal forms of HCC, there is a relativepaucity of descriptions of the MR appearance of diffuse-type HCC in the literature. Diffuse-type HCC is oftendifficult to detect on imaging studies because of itspermeative appearance and heterogeneity of back-ground chronic liver disease. The purpose of this study was to describe the MR imaging findings and clinicalmanifestations of diffuse-type HCC.

MATERIALS AND METHODS We retrospectively searched the radiologic records of MR imaging of the liver performed at the Department of Radiology, University of North Carolina, from November 1994 to October 2001, and found records of 171 pa-tients who had HCC and underwent MR imaging of theliver. Among this group, 22 patients were considered tohave diffuse-type HCC based on the criterion that there was no distinct margination on any portion of the tumor on any sequence. Only patients with no prior treatment were included to avoid misinterpreting treatment changes as findings of tumor. Fourteen patients, in-cluding four with normal serum AFP values and four in

Department of Radiology, University of North Carolina, Chapel Hill,North Carolina.

Current address (M.K.): Department of Radiology, Gifu University School of Medicine, Gifu, Japan.

*Address reprint requests to: R.C.S., Department of Radiology, CB7510, University of North Carolina, Chapel Hill, NC 27599-7510.E-mail: [email protected]

Received February 4, 2003; Accepted April 8, 2003.

DOI 10.1002/jmri.10336Published online in Wiley InterScience (www.interscience.wiley.com).

JOURNAL OF MAGNETIC RESONANCE IMAGING 18:189 –195 (2003)

© 2003 Wiley-Liss, Inc. 189

8/3/2019 10336_ftp


whom serum AFP values were not available, had his-

topathologic confirmation (laparotomic wedge biopsy [N 1], percutaneous needle-core aspiration biopsy [N 13]). The remaining eight patients without histo-logic confirmation had MR imaging findings consistent

with extensive, malignant hepatic tumors associated

with markedly high (2,000 ng/mL) serum AFP values.MR imaging was performed with a 1.5-T MR imager

(Vision, Siemens Medical Systems, Iselin, NJ). MR im-aging included a T1-weighted, in-phase, breath-hold,spoiled gradient-echo sequence (repetition time of 140– 175 msec and echo time of 4.1–4.5 msec [140–175/4.1–4.5]; flip angle, 80°; section thickness, 8 mm; in-

tersection gap, 20%; one signal acquired; 21 sections ina 20-second breath hold), and a T2-weighted sequenceperformed on fat-suppressed spin-echo (2400/90; sec-tion thickness, 8 mm; intersection gap, 20%; two sig-nals acquired) (one patient), short tau inversion recov-ery turbo spin-echo (5,110/76/170 [TR/TE/TI]; section

thickness, 8 mm; intersection gap, 20%; two signalsacquired) (four patients), or half-Fourier turbo spin-

echo sequence (/90; section thickness, 8 mm; inter-

section gap, 20%; one signal; 20 sections) (17 patients).

Transverse spoiled gradient-echo images were acquiredprior to and after intravenous bolus injection of 0.1mmol/kg gadolinium chelate (Magnevist, Berlex Labo-ratories, Wayne, NJ or Omniscan, Nycomed, New York,NY). Postcontrast spoiled gradient-echo sequences were

initiated at 18 seconds and one minute, and fat-sup-pressed spoiled gradient-echo sequence was acquiredtwo minutes postcontrast.

Three investigators, Masa Kanematsu, Richard C. Se-melka, and Polytimi Leonardou, who were blinded tothe clinical and histopathologic information and to theoriginal MR imaging reports, retrospectively evaluated,

in consensus, the transverse unenhanced T1- and T2- weighted MR and postcontrast early- and late-phaseMR images in each patient. The MR images were re-

viewed for the following findings: tumor extension in theliver expressed as number of Couinaud’s segments in-

volved by tumors, fibrous capsules or septa, signal in-

tensity characteristics of tumors on unenhanced T1-and T2-weighted MR images, contrast enhancement

characteristics of tumors on early- and late-phase post-

Figure 1. A 51-year-old man with diffuse-type HCC in alcoholic cirrhosis, showing serum alpha-fetoprotein value of 38000ng/mL. A: Unenhanced T1-weighted spoiled gradient-echo (140/4.1) axial MR image shows the tumor (arrows) involving theentire right lobe of the liver as areas of homogeneous, mild hypointensity. Note a small volume of perihepatic and perisplenic

ascites (curved arrow). B: Unenhanced T2-weighted fat-suppressed spin-echo (2400/90) axial MR image shows the tumor asareas of heterogeneous, moderate hyperintensity (arrows). C: Gadolinium-enhanced early-phase spoiled gradient-echo (140/4.1) axial MR image shows the tumor as areas of patchy enhancement (arrow). D: Gadolinium-enhanced fat-suppressedlate-phase spoiled gradient-echo (140/4.1) axial MR image shows the tumor as areas of multiple foci of wash-out (arrow). Portal venous tumor thrombosis in the left main portal vein branch is shown as areas of discrete hypointensity (curved arrow).

190 Kanematsu et al.

8/3/2019 10336_ftp


contrast MR images, bile duct dilatation, portal venoustumor thrombosis, hepatic venous tumor thrombosis,ascites, and upper abdominal lymph-node metastases.

The signal intensity of tumor on T1-weighted images was categorized as moderately hypointense when it wascomparable to that of the spleen, and mildly hypoin-tense when it was intermediate between liver andspleen. The signal intensity of tumor on T2-weightedimages was categorized as moderately hyperintense

when it was comparable to spleen, and mildly hyperin-tense when it was intermediate between liver andspleen. Enhancement characteristics of tumor werecategorized into patchy and miliary: patchy enhance-ment referred to foci of enhancement that were irregu-lar in size and shape and ill-defined, and miliary en-hancement referred to foci that were small (5 mm) and

relatively well defined.Other information on underlying hepatic disease (vi-

ral hepatitis, alcohol abuse, autoimmune hepatitis,

hemochromatosis, etc.), serological laboratory test re-sults (total bilirubin, AFP, alkaline phosphatase, ala-nine aminotransferase, aspartate aminotransferase,-glutamyltranspeptidase), and distant metastases

were obtained using the clinical information system of our institution.

RESULTS

From November 1994 to October 2001, we had 22 pa-tients with diffuse-type HCC (Figs. 1–3), which repre-sented 13% of all patients with HCC. The 22 patientsincluded 19 men and three women, ranging in age from16–80 years (mean age, 52 years). Sixteen patients hadcirrhosis, and the remaining six had chronic hepatitis.

The patient characteristics are summarized in Table 1

and the MR imaging findings in Table 2.Serum total bilirubin value (normal range, 0–1.2 mg/

dL) was elevated in 19 of 20 patients in whom the test

Figure 2. A 44-year-old man with diffuse-type HCC in cirrhosis due to type-C viral hepatitis, showing serum alpha-fetoprotein value of 9000 ng/mL. A: Unenhanced T1-weighted spoiled gradient-echo (170/4.1) axial MR image shows the tumor (arrow)involving the right hepatic lobe and the medial segment of the left hepatic lobe as ill-demarcated areas of homogeneous, mildhypointensity, accompanied by extensive portal venous tumor thrombosis (curved arrow) shown as areas of mixed intensity.Note a moderate amount of perihepatic and perisplenic ascites (small arrows). B: Unenhanced T2-weighted turbo spin-echo(/90) axial MR image shows the tumor (arrow) as areas of heterogeneous, moderate hyperintensity. Portal venous thrombosis(curved arrow) is shown as areas of mixed signal intensity. C: Gadolinium-enhanced early-phase spoiled gradient-echo (170/4.1)axial MR image shows the tumor as areas of miliary enhancement (arrow). Portal venous tumor thrombosis is shown as areasof heterogeneous hypointensity (curved arrow). Note areas of necrosis in the tumor (small arrow). D: Gadolinium-enhancedlate-phase fat-suppressed spoiled gradient-echo (175/4.1) axial MR image shows the tumor as areas of heterogeneous isoin-tensity with some areas of wash-out (arrow). Portal venous tumor thrombosis is shown as areas of discrete, heterogeneous,hypointensity (curved arrow).

MRI of Diffuse HCC 191

8/3/2019 10336_ftp


result was available (0.7–23.5 mg/dL [mean, 6.5 8.1mg/dL]), serum AFP value (0–10 ng/mL) was elevatedin 14 of 18 patients (5–242000 ng/mL, mean; 24822

56380 ng/mL), and 13 showed a value greater than 500ng/mL. Serum alkaline phosphatase value (38–126U/liter) was elevated in 17 of 20 patients (99–969 U/li-ter [356 267 U/liter]), serum alanine aminotransfer-ase (15–48 U/liter) value was elevated in 14 of 20 pa-tients (21–190 U/liter [84 52 U/liter]), serumaspartate aminotransferase value (14–38 U/liter) waselevated in 18 of 20 patients (38–876 U/liter [243

226 U/liter]), serum -glutamyltranspeptidase value(11–48 U/liter) was elevated in all of 18 patients (54– 1042 U/liter [361 270 U/liter]).

Number of Couinaud’s segments involved by diffuse-type HCC was four segments in four patients, five seg-ments in one, six segments in four, seven segments in

one, and eight segments in 12 (mean, 6.7 1.6 seg-ments). Fibrous capsules or septa were seen mildly but partly in seven, minimally in two, and none in 13.

Unenhanced T1-weighted MR images showed the tu-mors as areas of homogeneous hypointensity in fivepatients, heterogeneous hypointensity in ten, mixed in-tensity in three, and isointensity in four. T2-weighted

MR images showed the tumors as areas of heteroge-neous hyperintensity in 16 and homogeneous hyperin-tensity in six. Postcontrast early-phase MR imagesshowed tumors as areas of patchy enhancement in 12patients, miliary enhancement in nine, and minimalenhancement in one. Postcontrast late-phase MR im-ages showed heterogeneous irregular areas of wash-out in all patients.

Portal venous tumor thrombosis was seen in all pa-tients: bilateral main portal vein branches were in-

volved in 12 patients and either of the right or left mainportal vein branch was involved in 10. Hepatic venoustumor thrombosis was not seen in any patient. Intra-hepatic bile ducts were minimally to mildly dilated in

three patients and of normal caliber in 19. Ascites was substantial in one patient, moderate in

seven, minimal in eight, and absent in six. Upper ab-dominal lymphadenopathy was seen in three patients(porta-hepatis nodes in all three and one with addi-tional porto-caval and peripancreatic nodes). Distant metastases were confirmed in three patients (bone inone, adrenal gland in one, and lung in one).

DISCUSSION

A 1981 report by Okuda et al (8) described the clinicaland pathological findings with six autopsy cases of dif-

fuse-type HCC. In the report, the prominent clinicalfeature was the rapid deterioration of the patient’s gen-eral condition terminating in hepatic failure, and theliver size enlarged quickly at a perceptible speed, oftenaccompanied by abdominal pain. They also describedthat the entire liver was studded with minute, uni-formly sized tumor nodules evenly distributed through-out, with some of them grossly indistinguishable fromcirrhotic nodules.

The incidence of diffuse-type HCC in our institution was approximately 13% of all patients with HCC, whichis slightly higher than previously reported (8). This may reflect differences related to geography, as our study involved a North American population, variation relatedto the relative rarity of the disease, or that our study

was imaging-based rather than autopsy-based.Portal venous tumor thrombosis is a common finding

with diffuse-type HCC and may be a clue to the diag-nosis. Accordingly, characterization of portal venousthrombosis is crucial in the diagnosis of diffuse-typeHCC on contrast-enhanced CT or MR imaging. Tublinet al (9) evaluated 58 cirrhotic patients and concludedthat identification of main portal venous thrombosis of 23 mm in diameter or greater or neovascularity of portal

venous thrombosis resulted in a sensitivity and speci-ficity for the CT characterization of malignant portal

venous thrombosis of 86% and 100%, respectively.

No previous reports have described the specific MR imaging findings of diffuse-type HCC cases. We foundthat the unenhanced T1- and T2-weighted MR images

Figure 3. A 64-year-old man with diffuse-type HCC in hemo-

chromatosis, showing negative serum alpha-fetoprotein value(10 ng/mL). A: Unenhanced T2-weighted short tau inversionrecovery turbo spin-echo (5,110/76/170 [TR/TE/TI]) axial MR image shows the tumor involving most of the left hepatic lobeand a part of right hepatic lobe as areas of homogeneous,moderate hyperintensity (arrow). Note extensive portal venoustumor thrombosis (curved arrow). B: Gadolinium-enhancedearly-phase spoiled gradient-echo (160/4.5) axial MR imageshows the tumor as areas of intense, patchy enhancement (arrow). Note the very low signal intensity of the liver paren-chyma (curved arrow), reflecting underlying hemochromato-sis.


8/3/2019 10336_ftp


Table 1

Summary of Patient Characteristics in 22 Patients With Diffuse HCC

Patients Age/sex/raceEtiology of underlying

liver diseaseCirrhosis

TB

(mg/dl)

AFP

(ng/mL)

ALP

(U/liter)

ALT

(U/liter)

AST

(U/liter)

-GTP

(U/liter)

No. of

involved

segments

Capsules

or septa

1 60/M/C Alcohol abuse Yes 3.1 38,000 242 136 395 257 4 segs. –

2 30/M/AA Alcohol abuse Yes 23.5 15,870 114 28 65 350 4 segs.

3 51/M/C Alcohol abuse Yes 19.3 38,000 99 53 118 84 All –4 48/M/AA Alcohol abuse Yes 1.3 2,000 245 66 105 274 All –

5 46/M/C Alcohol abuse Yes 1.5 N.A. 551 28 38 666 All

6 50/M/AA Hep. B Noa 20.8 564 623 155 700 177 All

7 64/M/C Hep. B Yes 3.3 5366 851 147 404 1042 All –

8 49/M/A Hep. B Noa N.A. N.A. N.A. N.A. N.A. N.A. 6 segs. –

9 49/M/AA Hep. C Noa 8.7 42,000 165 47 207 184 All

10 48/M/C Hep. C Yes 2.2 1816 310 21 67 445 5 segs. –

11 49/M/AA Hep. C Yes 1.8 N.A. 173 40 38 139 All

12 51/M/AA Hep. C Yes 2.7 136 195 132 447 135 7 segs.

13 49/M/C Hep. C Noa 2.3 5 122 134 162 254 4 segs. –

14 44/M/C Hep. C Yes 2.0 9000 168 50 268 N.A. All –

15 61/F/C Hep. C Yes 2.0 242,000 784 60 257 N.A. All –

16 43/M/C Alcohol abuse, Hep. C Yes 2.2 7000 174 53 94 249 6 segs.

17 58/M/AA Hep. B, Hep. C Yes 5.2 38,000 280 113 292 288 4 segs. –

18 64/M/C Hemochromatosis Noa

0.7 10 311 55 73 630 6 segs. –19 16/F/C Autoimmune hepatitis Yes N.A. N.A. N.A. N.A. N.A. N.A. All –

20 64/M/C Cryptogenic Yes 23.5 7000 193 190 876 54 All –

21 71/M/C Cryptogenic Noa 2.1 10 969 134 168 822 6 segs.

22 80/F/C Cryptogenic Yes 1.9 10 548 46 86 451 All

aPatients had chronic hepatitis.

TB serum total bilirubin (normal range, 0–1.2 mg/dl), AFP serum alpha-fetoprotein (0–10 ng/mL), ALP serum alkaline phosphatase

(38–126 U/liter), ALT serum alanine aminotransferase (15–48 U/liter), AST serum aspartate aminotransferase (14–38 U/liter), -GTP

serum -glutamyltranspeptidase (11–48 U/liter), C Caucasian, AA African-American, A Asian, Hep. B type-B viral hepatitis, Hep.

C type-C viral hepatitis, N.A. not available, mild, minimal, – none.

Table 2

Summary of MRI Findings in 22 Patients With Diffuse HCC

PatientsT1-weighted imaging

findings

T2-weighted imaging

findings

Enhancement pattern

in early-phase

postcontrast images

Washed-out area %

in late-phase

postcontrast images

Portal venous

tumor thrombosis

in MPV(s)

Bile duct

dilatation

1 Iso Hetero, mild hyper Miliary 50–75% Right –

2 Iso Hetero, mild hyper Patchy 50–75% Right –

3 Homo, mild hypo Hetero, mod. hyper Patchy 75–100% Bilateral –

4 Hetero, mild hypo Hetero, mild hyper Miliary 25–50% Left

5 Hetero, mild hypo Hetero, mod. hyper Patchy 75–100% Bilateral –

6 Hetero, mild hypo Hetero, mild hyper Miliary 75–100% Bilateral

7 Hetero, mod. hypo Hetero, mild hyper Patchy 50–75% Bilateral –8 Hetero, mild hypo Hetero, mild hyper Miliary 75–100% Right –

9 Homo, mod. hypo Homo, mod. hyper Miliary 75–100% Bilateral –

10 Mixed Homo, mild hyper Patchy 25–50% Right

11 Hetero, mild hypo Hetero, mild hyper Patchy 50–75% Right –

12 Hetero, mild hypo Hetero, mild hyper Patchy 25–50% Bilateral –

13 Homo, mild hypo Homo, mild hyper Patchy 25–50% Right –

14 Homo, mild hypo Hetero, mod. hyper Miliary 25–50% Bilateral –

15 Hetero, mild hypo Hetero, mild hyper Miliary 50–75% Bilateral –

16 Hetero, mod. hypo Hetero, mild hyper Patchy 75–100% Right –

17 Iso Homo, mod. hyper Minimal 50–75% Left –

18 Hetero, mod. hypo Homo, mod. hyper Patchy 25–50% Bilateral –

19 Iso Homo, mod. hyper Patchy 50–75% Bilateral –

20 Mixed Hetero, mild hyper Patchy 25–50% Bilateral –

21 Mixed Hetero, mild hyper Miliary 50–75% Right –

22 Homo, mod. hypo Hetero, mod. hyper Miliary 50–75% Bilateral –MPV(s) main portal vein branch(es), mild, minimal, – none, Homo homogeneous, Hetero heterogeneous, mod.

moderate, hypo hypointensity, hyper hyper intensity, iso iso intensity.


8/3/2019 10336_ftp


showed the tumors as ill-demarcated areas of homoge-

neous or heterogeneous, abnormal signal intensities,

typically mildly to moderately hypointense on T1-

weighted images and mildly to moderately hyperintense

on T2-weighted images. We considered the findings on

noncontrast images somewhat nonspecific as these sig-

nal alterations may also be analogous to those in cir-

rhotic liver without tumor. However, multiple small ar-eas of low signal intensity in the liver on T2-weighted

images may suggest the presence of regenerating nod-

ules in cirrhosis, which may help exclude the diagnosis

of diffuse-type HCC (10).

We observed patchy or miliary enhancement corre-

sponding to the tumors on the early-phase postcontrast

images. Miliary enhancement in particular, we believe,

may be relatively specific for diffuse-type HCC, and may

represent the enhancement of extensive micronodules

of tumor as identified at histopathology (8). Heteroge-

neous wash-out of the tumors on the late-phase images

is a feature typical for malignant tumor of all types in

the liver. Increased hepatic parenchymal enhancement on early-phase postcontrast images is known to be

caused by increased hepatic arterial supply in the set-

ting of proximal portal vein obstruction (11), but this

usually is observed as homogeneous, wedge-shaped

enhancement in postcontrast early phase, with fading

to homogeneous isointensity in postcontrast late

phase. We could distinguish such non-neoplastic, tran-

sient early-phase enhancement, which was seen in

some cases, from the enhancement of tumor.

In our clinical experience diffuse HCC may be distin-

guished from other malignant diseases, specifically

cholangiocarcinoma and metastases, in the great ma-

jority of cases. Cholangiocarcinoma, in our experience,has better defined margins and we have not observed

portal vein tumor thrombus with this entity, although

portal vein comparison is common. Massive involve-

ment of the liver in metastatic disease almost invariably

shows a more clearly focal pattern of involvement. In

our experience, breast cancer is the malignancy that

most often may present with extensive infiltration. In

this setting the primary tumor is almost always known.

Coexistence of breast cancer and cirrhosis is rare but

may occur, so uncertainty whether the patient has

breast cancer liver metastases or diffuse HCC is rare.

Finally, while venous tumor thrombus was seen in all

patients with diffuse HCC, it is rare in patients withmetastatic disease.

Lee et al (12) reported that the serum AFP level, which

showed a positive predictive value of 95% for HCC, was

3200 ng/mL in livers with type-B viral hepatitis and

200 ng/mL in livers without type-B viral hepatitis. They

concluded that the presence of underlying type-B viral

hepatitis should be taken into consideration when se-

rum AFP is measured to diagnose HCC. One of our

patients had underlying type-B viral hepatitis and had

a serum AFP value of 5366 ng/mL, which was consid-

ered diagnostic for HCC.

Serum AFP values have been reported to be elevated

in 43%–72% of patients with HCC (13–16). However, asmore small HCCs are currently being detected, due to

the advent of advanced radiologic imaging techniques

and increased imaging screening of patients with hep-

atitis or cirrhosis, this rate is likely much lower, as

small well-differentiated focal HCCs rarely result in el-

evated AFP (17). We observed that 14 (78%) of 18 pa-

tients with diffuse HCC showed elevated serum AFP

values in our study. Radiologists should be apprised of

the fact that although the positive rate of AFP is high

with diffuse HCC, a sizable number (22% in our study)have normal serum AFP value.

There are some limitations to this study. We did not

have a histopathologic diagnosis of HCC in eight pa-

tients. These patients were very ill and the combination

of the MR findings and markedly elevated serum AFP

values (2000 ng/mL) were considered diagnostic for

HCC, as also described in the literature (12).

In conclusion, diffuse-type HCC represented 13% of

the patients with HCC. Diffuse-type HCC was seen as

extensive permeative hepatic tumor with ill-defined

margins, with extensive portal venous tumor thrombo-

sis in all patients, and markedly elevated serum AFP

value in 78% of patients. Serum AFP value was normalin 22% of patients. Unenhanced MR images showed

nonspecific findings of abnormal signal intensities. Ga-

dolinium-enhanced MR images showed patchy or mili-

ary enhancement on immediate postcontrast images,

with the miliary pattern possessing a distinctive ap-

pearance.

REFERENCES

1. Tsukuma H, Hiyama T, Tanaka S, et al. Risk factors for hepatocel-

lular carcinoma among patients with chronic liver disease. N Engl

J Med 1993;328:1797–1801.

2. National Institutes of Health Consensus Development ConferencePanel statement: Management of hepatitis C. Hepatology 1997;26:

2–10.

3. Donato F, Tagger A, Chiesa R, et al. Hepatitis B and C virus infec-

tion, alcohol drinking, and hepatocellular carcinoma: a case-con-

trol study in Italy. Brescia HCC study. Hepatology 1997;26:579–

584.

4. Niederau C, Fischer R, Sonnenberg A, Stremmel W, Trampisch HJ,

Strohmeyer G. Survival and causes of death in cirrhotic and non-

cirrhotic patients with primary hemochromatosis. N Engl J Med

1985;313:1256–1262.

5. Bradbear RA, Bain C, Siskind V, et al. Cohort study of internal

malignancy in genetic hemochromatosis and other non-alcoholic

liver disease. J Natl Cancer Inst 1985;75:81–84.

6. Okuda K, Peters RL, Simson IW. Gross anatomic features of hepa-

tocellular carcinoma from three disparate geographic areas. Pro-

posal of new classification. Cancer 1984;54:2165–2173.7. Anthony PP. Tumours and tumour-like lesions of the liver and

biliary tract: aetiology, epidemiology, and pathology. In: MacSween

RNM, Burt AD, Portmann BC, et al., eds. Pathology of the liver. 4th

ed. London, England: Churchill Livingstone; 2002. p 711–825.

8. Okuda K, Noguchi T, Kubo Y, Shimokawa Y, Kojiro M, Nakashima

T. A clinical and pathological study of diffuse type hepatocellular

carcinoma. Liver 1981;1:280–289.

9. Tublin ME, Dodd 3rd GD, Baron RL. Benign and malignant portal

vein thrombosis: differentiation by CT characteristics. AJR Am J

Roentgenol 1997;168:719–723.

10. Itai Y, Ohnishi S, Ohtomo K, et al. Regenerating nodules of liver

cirrhosis: MR imaging. Radiology 1987;165:419–423.

11. Itai Y, Moss AA, Goldberg HI. Transient hepatic attenuation differ-

ence of lobar or segmental distribution detected by dynamic com-

puted tomography. Radiology 1982;144:835–839.

12. Lee HS, Chung YH, Kim CY. Specificities of serum alpha-fetopro-

tein in HBsAg and HBsAg- patients in the diagnosis of hepatocel-

lular carcinoma. Hepatology 1991;14:68–72.


8/3/2019 10336_ftp


13. Sherman M, Peltekian KM, Lee C. Screening for hepatocellular

carcinoma in chronic carriers of hepatitis B virus: incidence and

prevalence of hepatocellular carcinoma in a North American urban

population. Hepatology 1995;22:432–438.

14. Aoyagi Y, Oguro M, Yanagi M, et al. Clinical significance

of simultaneous determinations of alpha-fetoprotein and des-

gamma-carboxy prothrombin in monitoring recurrence in pa-

tients with hepatocellular carcinoma. Cancer 1996;77:1781–

1786.

15. Okuda H, Nakanishi T, Takatsu K, et al. Measurement of serumlevels of des-gamma-carboxy prothrombin in patients with hepato-

cellular carcinoma by a revised enzyme immunoassay kit with

increased sensitivity. Cancer 1999;85:812–818.

16. Nguyen MH, Garcia RT, Simpson PW, Wright TL, Keeffe EB. Racial

differences in effectiveness of alpha-fetoprotein for diagnosis of

hepatocellular carcinoma in hepatitis C virus cirrhosis. Hepatology

2002;36:410–417.

17. Peterson MS, Baron RL, Marsh Jr JW, Oliver 3rd JH, Confer SR,

Hunt LE. Pretransplantation surveillance for possible hepatocellu-

lar carcinoma in patients with cirrhosis: epidemiology and CT-

based tumor detection rate in 430 cases with surgical pathologiccorrelation. Radiology 2000;217:743–749.


10336_ftp

Documents