AJR 2004; 183:1573-1576
© American Roentgen Ray Society
CT of Benign Hypervascular Liver Nodules in Autoimmune Hepatitis
Aliya Qayyum1,
Anno Graser1,
Antonio Westphalen1,
Raphael B. Merriman2,
Linda D. Ferrell3,
Benjamin M. Yeh1 and
Fergus V. Coakley1
1 Department of Radiology, University of California–San Francisco, 505
Parnassus Ave., San Francisco, CA 94143-0628.
2 Department of Medicine, Division of Gastroenterology, University of
California–San Francisco, San Francisco, CA 94143-0628.
3 Department of Pathology, University of California–San Francisco, San
Francisco, CA 94143-0628.
Received November 11, 2003;
accepted after revision March 8, 2004.
Address correspondence to A. Qayyum.
Abstract
OBJECTIVE. The purpose of this report is to describe the frequency
and histopathologic basis of benign hypervascular liver nodules seen on CT in
patients with autoimmune hepatitis.
CONCLUSION. Benign hypervascular liver nodules may be seen on CT in
patients with cirrhosis due to autoimmune hepatitis and may represent large
regenerative nodules. This phenomenon is important to recognize because of the
potential for confusion with hepatocellular carcinoma.
Introduction
Autoimmune hepatitis has an annual incidence of 1.9 per 100,000
[1] and is the indication for
2.6% of liver transplants in Europe
[2] and 3.9% in the United
States [3]. The diagnosis of
autoimmune hepatitis is established by characteristic liver histopathology,
clinical and immunologic features, and the exclusion of other conditions that
may resemble autoimmune hepatitis, including viral hepatitis, drug-induced
hepatitis, Wilson's disease, hemochromatosis, and
1-antitrypsin deficiency. Autoimmune hepatitis is
characterized histologically by interface hepatitis (piecemeal necrosis) with
a predominant lymphoplasmacytic cell infiltrate, usually with fibrosis.
Autoimmune hepatitis progresses to cirrhosis in at least 40% of patients
within 5 years [4,
5]. Solid nodules, other than
hemangiomas, in a cirrhotic or precirrhotic liver are generally considered
suggestive of hepatocellular carcinoma, especially if the nodules are
hypervascular in the arterial phase of enhancement
[6–8].
However, we recently have encountered several patients with autoimmune
hepatitis in whom CT showed benign hypervascular nodules. This previously
unreported phenomenon is especially important to distinguish from potentially
complicating hepatocellular carcinoma. Therefore, we undertook this study to
describe the frequency and histopathologic basis of benign hypervascular liver
nodules seen on CT in patients with autoimmune hepatitis.
Materials and Methods
This was a retrospective single institutional study and was approved by the
local committee on human research. Patient consent was not required. We
performed a computerized search of our radiology and pathology information
systems to identify patients who met the inclusion criteria of histopathologic
diagnosis consistent with autoimmune hepatitis and multiphase
contrast-enhanced abdominal CT reported at our institution between 1998 and
2002.
Seven patients, out of 38,239 abdominal CT examinations reported at our
institution over the same period, fulfilled these criteria. One patient was
imaged at an outside institution, but the study was submitted to our
institution for reinterpretation. No exclusion criteria were present. The
seven patients forming the study group consisted of five women and two men
with a mean age of 21 years (age range, 17–29 years). None of the
patients had positive markers for viral hepatitis, and all had liver biopsies
consistent with autoimmune hepatitis. Indications for abdominal CT were
cirrhosis (n = 5), left upper quadrant pain (n = 1), and
evaluation of a renal mass seen on sonography (n = 1).
CT Technique
CT examinations at our institution were performed using helical CT scanners
(HiSpeed [n = 3] or LightSpeed [n = 3], GE Healthcare). All
patients received 150 mL IV iohexol (Omnipaque 350, Nycomed Amersham) at a
rate of 3–5 mL/sec and 800 mL oral meglumine diatrizoate (Hypaque,
Sanofi Winthrop). Arterial and portal venous phase images were acquired after
a scanning delay of 45 and 70 sec, respectively, using a slice collimation of
5 mm, which was the standard protocol at the time studies were performed. All
images were contiguous. The CT scan from an outside institution was obtained
using oral and IV contrast material, with images acquired in the arterial,
portal venous, and delayed phases of enhancement.
Image Interpretation and Analysis
Two attending radiologists reviewed the images and reached a consensus
regarding the presence and location of focal hypervascular liver lesions.
Lesions were considered hypervascular if they were of greater attenuation than
the surrounding hepatic parenchyma during the arterial phase of enhancement.
The enhancement of all detected hypervascular lesions was further classified
as hypervascular, isovascular, or hypovascular on portal venous phase images
on the basis of whether the lesion was of greater, equal, or lesser
attenuation, respectively, in relation to the surrounding hepatic parenchyma
during the portal venous phase. Lesions with the typical enhancement pattern
of hemangiomas were excluded. CT signs of cirrhosis, including liver surface
nodularity, lobar atrophy, ascites, varices, and splenomegaly also were
recorded. In patients with hypervascular lesions, all histopathologic records
were examined for correlation with imaging. In patients without
histopathologic correlation, serial CT examinations, serum -fetoprotein
levels, and clinical evolution were used to determine lesion benignity or
malignancy.
Results
Five of the seven patients had hypervascular liver lesions (enhancement on
arterial phase imaging), and two of the five patients had lesions that
enhanced on both arterial and portal venous phase imaging.
(Table 1) (Figs.
1A,
1B and
2A,
2B). The portal venous phase
washout pattern observed in three of the five patients with hypervascular
lesions was uniform without the presence of a persistent hyperdense rim. For
each patient, all hypervascular lesions showed a similar enhancement pattern
on arterial and portal venous phase imaging. The mean size of the
hypervascular lesions was 13 mm (range, 7–32 mm). No other suspicious
hyperor hypovascular lesions were seen. In the five patients with
hypervascular lesions, the mean number of hypervascular lesions per patient
was three (range, 2–4). Histopathologic correlation was available in
four of the five patients with hypervascular liver lesions, consisting of
sonographically guided fine-needle aspiration (n = 1), core biopsy
(n = 2), and liver explantation (n = 1). The fine-needle
aspiration biopsy showed benign liver tissue. The two core biopsies and the
liver explant showed regenerative nodules. All patients with hypervascular
lesions were found to have stable findings on follow-up CT at a mean interval
of 18 months (range, 8–40). In one of the five patients with
hypervascular liver lesions, confirmation of a benign cause was based on
stable CT findings and -fetoprotein levels over 23 months of follow-up.
In all seven patients, the liver showed CT features suggestive of cirrhosis
and features of portal hypertension also were detected
(Table 1). Ascites was seen in
four of the seven patients.
View larger version (136K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A. —20-year-old woman with cirrhosis secondary to autoimmune
hepatitis. Contrast-enhanced CT scan shows hypervascular lesion
(arrow) in segment V of liver. Fine-needle aspiration showed benign
histopathology.
|
|
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —20-year-old woman with cirrhosis secondary to autoimmune
hepatitis. Contrast-enhanced CT scan shows second similar hypervascular lesion
(arrow) in liver dome on arterial phase of enhancement.
|
|
View larger version (84K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A. —29-year-old woman with cirrhosis secondary to autoimmune
hepatitis. Histopathology of explant showed benign regenerative nodules.
Contrast-enhanced CT scan shows hypervascular lesion (arrow) in liver
dome on arterial phase of enhancement.
|
|
View larger version (95K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B. —29-year-old woman with cirrhosis secondary to autoimmune
hepatitis. Histopathology of explant showed benign regenerative nodules.
Contrast-enhanced CT scan shows lesion (arrow) is isoattenuating on
venous phase of enhancement.
|
|
Discussion
Our study suggests that benign hypervascular nodules may be seen on CT in
patients with cirrhosis due to autoimmune hepatitis, with such lesions seen in
five of seven patients. In three patients, these nodules were proven to be
regenerative nodules on the basis of core biopsy and liver explantation. The
similar appearance and setting of the lesions in the other two patients
suggests the same diagnosis, as does the finding of benign hepatic tissue on
fine-needle aspiration biopsy, though the distinction of a regenerative nodule
from healthy liver cannot be made reliably on a fine-needle biopsy specimen
[9]. It should be noted that
the CT findings in these patients are not typical for regenerative nodules,
which typically measure 3–10 mm in diameter and show enhancement
characteristics similar to the background liver. Generally, regenerative
nodules have a predominant portal venous supply and enhance on CT arterial
portography but not on CT hepatic arteriography. However, large hypervascular
regenerative nodules with an increased hepatic arterial blood supply have been
described in patients with long-standing Budd-Chiari syndrome
[10]. These regenerative
nodules are considered a consequence of elevated hepatocellular growth factors
resulting from a disturbance in the hepatic microcirculation associated with
obstruction of hepatic veins
[10]. A few studies, including
a recent radiologic–pathologic correlation, suggest that growth of
benign hypervascular nodules in Budd-Chiari syndrome may be related to a
reduction in portal venous flow
[8], although it should be
noted that decreased portal venous flow occurs in cirrhosis without the
development of hypervascular regenerative nodules. An alteration in the liver
microcirculation in patients with autoimmune hepatitis could be responsible
for the development of similar benign hypervascular regenerative nodules,
although this is speculative. Variation in the portal venous enhancement
pattern of the hypervascular lesions might indicate differences in relative
arterial and portal venous flow contribution to the regenerative nodules,
although an alternative explanation might be differences in patient
circulation as a result of fixed bolus timing.
The existence of benign hypervascular nodules in autoimmune hepatitis is
important because these lesions could mimic hepatocellular carcinoma;
hypervascular nodules that are not hemangiomas in a cirrhotic liver generally
are considered hepatocellular carcinomas until proven otherwise
[7]. Hypervascular nodules in
cirrhosis due to autoimmune hepatitis and long-standing Budd-Chiari syndrome
may be exceptions to this rule. In fact, hepatocellular carcinoma is unusual
in both these specific conditions, despite the markedly increased risk of
hepatocellular carcinoma in patients with cirrhosis
[4,
8,
11]. Only one in 200 patients
with cirrhosis due to autoimmune hepatitis develops hepatocellular carcinoma
[4] although patients with
Budd-Chiari syndrome have a hepatocellular carcinoma prevalence of 6.4%
[12]. Core biopsy was
performed on two patients and is considered to be superior to fine-needle
biopsy for diagnosing a particular type of malignancy, but fine-needle biopsy
reliably can distinguish benign liver tissue from hepatocellular carcinoma,
even though problems exist in determining the difference between healthy
liver, cirrhosis, or other benign hepatocytic tumors
[9]. In the setting of chronic
liver disease, hepatocellular carcinoma (and not metastatic tumors) is the
main concern, and fine-needle biopsy should be a good screening tool in these
patients.
Our study has a number of limitations. A major limitation is the small
number of patients with autoimmune hepatitis, which may reflect the rarity of
this condition. A second major limitation is the potential selection bias
inherent in the methodology. Five patients were referred for evaluation of
cirrhosis. It is possible that the more specific reason for referral in these
patients was the detection of focal hepatic lesions on imaging performed at
outside institutions. These two factors restrict the ability to extrapolate
our results to the wider population with autoimmune hepatitis. A final
limitation is the nature of the histopathologic correlations, with one patient
undergoing only fine-needle aspiration biopsy, which precludes evaluation of
nodule architecture. It is important to mention that one could consider in the
differential diagnosis of these hypervascular nodules the possibilities of
hemangiomas and perfusional pseudolesions, and in spite of limited
histopathologic analysis the former diagnosis was excluded. Perfusional
pseudolesions in cirrhosis are reported to measure less than 2 cm in diameter
and are not apparent or regress on serial imaging. All five patients with
hypervascular lesions underwent follow-up CT, which showed persistent, stable
lesions. Three of our patients underwent core biopsy or liver explantation,
and histopathologic evaluation confirmed regenerative nodules, excluding the
possibility of pseudolesions. Although biopsies only were performed on a
single lesion in each patient, the similar appearance and stability on
follow-up CT (10–40 months) favors the same diagnosis.
In conclusion, benign hypervascular liver nodules may be seen on CT in
patients with cirrhosis due to autoimmune hepatitis and may represent large
regenerative nodules. This phenomenon is important to recognize because of the
potential for confusion with hepatocellular carcinoma.
References
- Boberg KM, Aadland E, Jahnsen J, Raknerud N, Stiris M, Bell H.
Incidence and prevalence of primary biliary cirrhosis, primary sclerosing
cholangitis, and autoimmune hepatitis in a Norwegian population.
Scand J Gastroenterol1998; 33:99
–103[Medline]
- Milkiewicz P, Hubscher SG, Skiba G, Hathaway M, Elias E. Recurrence
of autoimmune hepatitis after liver transplantation.
Transplantation1999; 68:253
–256[Medline]
- U.S. Department of Health and Human Services. 2000
Annual report of the U.S. Scientific Registry for Transplantation Recipients
and the Organ Procurement and Transplantation Network: transplant data:
1989–1999. Rockville, MD: Department of Health and Human
Services, 2000
- Park SZ, Nagorney DM, Czaja AJ. Hepatocellular carcinoma in
autoimmune hepatitis. Dig Dis Sci2000; 45:1944
–1948[Medline]
- Czaja AJ, Freese DK, American Association for the Study of Liver
Disease. Diagnosis and treatment of autoimmune hepatitis.
Hepatology2002; 36:479
–497[Medline]
- de Lédinghen V, Laharie D, Lecesne R, et al. Detection of
nodules in liver cirrhosis: spiral computed tomography or magnetic resonance
imaging? A prospective study of 88 nodules in 34 patients. Eur J
Gastroenterol Hepatol2002; 14:159
–165[Medline]
- Coakley FV, Schwartz LH. Imaging of hepatocellular carcinoma: a
practical approach. Semin Oncol2001; 28:460
–473[Medline]
- Maetani Y, Itoh K, Egawa H, et al. Benign hepatic nodules in
Budd-Chiari syndrome: radiologic–pathologic correlation with emphasis on
the central scar. AJR2002; 178:869
–875[Abstract/Free Full Text]
- Ljung BM, Ferrell LD. Fine needle aspiration biopsy of the liver:
diagnostic problems. Pathology (Phila)1994; 3:161
–84
- Brancatelli G, Federle MP, Grazioli L, Golfieri R, Lencioni R.
Large regenerative nodules in Budd-Chiari syndrome and other vascular
disorders of the liver: CT and MR imaging findings with clinicopathologic
correlation. AJR 2002;178
: 877–883[Abstract/Free Full Text]
- Hayashi M, Matsui O, Ueda K, Kawamori Y, Gabata T, Kadoya M.
Progression to hypervascular hepatocellular carcinoma: correlation with
intranodular blood supply evaluated with CT during intraarterial injection of
contrast material. Radiology2002; 225:143
–149[Abstract/Free Full Text]
- Okuda H, Yamagata H, Obata H, et al. Epidemiological and clinical
features of Budd-Chiari syndrome in Japan. J Hepatol1995; 22:1
–9[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
R. F. Hanna, D. A. Aguirre, N. Kased, S. C. Emery, M. R. Peterson, and C. B. Sirlin
Cirrhosis-associated Hepatocellular Nodules: Correlation of Histopathologic and MR Imaging Features
RadioGraphics,
May 1, 2008;
28(3):
747 - 769.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
