British Journal of Radiology (2005) 78, 355-357
© 2005 British Institute of Radiology
doi: 10.1259/bjr/36150185
The reverse target sign in liver disease: a potential ultrasound feature in cirrhotic liver nodules characterization
G J Kraus, MD1,
P Schedlbauer, MD1,
S Lax, MD2,
D Zebedin, MD3 and
F Flueckiger, MD1
Institutes of 1 Radiology and 2 Pathology, General Hospital Graz-West, Goestingerstrasse 22, 8020 Graz and 3 Department of Radiology, University Hospital Graz, Auenbruggerplatz 1, 8036 Graz, Austria
 |
Abstract
|
|---|
In patients with suspected liver disease, ultrasound is the most commonly performed initial imaging modality. We report a patient who had previously undiagnosed liver cirrhosis with target-shaped lesions interspersed throughout the liver parenchyma on ultrasound seen as multiple uniform round shaped lesions with varying isoechoic to hyperechoic centres surrounded by a hyperechoic rim. We have termed this the "reverse" target sign as there is inversion of the typical echoic pattern that is normally seen in metastatic liver disease. We suggest this ultrasound sign may represent a method for differentiating cirrhotic liver nodules from other nodular liver lesions.
|
Introduction
|
|---|
Ultrasound is often the preliminary and most widely available imaging modality in evaluation of liver disease. The ultrasound and the CT appearance of metastatic liver disease and cirrhotic nodules can be quite similar. If there are no concomitant morphological imaging findings, the differential diagnosis is based on histology and clinical history.
|
Case report
|
|---|
A 55-year-old patient was referred to our institution for the liver ultrasound because of a weight loss of 17 kg over 3 months. His alanine aminotransferase (56 U l–1; normal range 5–22 U l–1), aspartate aminotransferase (33 U l–1; normal range 5–18 U l–1), alkaline phosphatase (214 U l–1, normal range 60–170 U l–1) and gamma glutamyle transferase (35 U l–1; normal range 5–28 U l–1) were elevated, and his cholinesterase (1444 U l–1; normal range 5300–12 920 U l–1) was below the reference value. All other laboratory values concerning the liver were within the normal range.
The prior medical history revealed that the patient suffered from pulmonary tuberculosis, as well as non insulin dependent diabetes mellitus. Serology for viral hepatitis A, B and C proved to be negative.
Ultrasound (Combison 401; General Electric Medical Systems, Milwaukee, WI) showed multiple lesions dispersed throughout the whole liver parenchyma with an isoechoic to hyperechoic centre and a hyperechoic rim (Figure 1
). On colour Doppler ultrasound, there was regular hepatopetal blood flow in the portal vein.
View larger version (86K):
[in this window]
[in a new window]
|
Figure 1. Ultrasound (B-mode) shows disseminated isoechoic to hyperechoic macronodular lesions with a hyperechoic rim (reverse target sign) throughout the liver parenchyma.
|
|
On a triple phase CT-scan the lesions were best visible in the unenhanced phase (Figure 2), faintly visible as slightly hypodense lesions in the portalvenous phase at 70 s delay (not shown) and isodense to liver parenchyma in a delayed phase at 180 s. In the unenhanced phase the large centre was slightly hypodense to liver parenchyma imbedded in a more hypodense rim (Figure 2). There was delayed uptake of contrast media in the centre and subtle enhancement of the rim in the portalvenous phase (not shown). There was no nodularity of the liver contour and the ratio of the caudate lobe to the right lobe was within the normal range (i.e.<0.6). Secondary changes of cirrhosis such as ascites, varicose venous collaterals and splenomegaly were absent.
View larger version (126K):
[in this window]
[in a new window]
|
Figure 2. On an unenhanced spiral CT-scan, the lesions are isodense to liver parenchyma with a hypoattenuating rim.
|
|
The most likely diagnosis was of metastatic liver disease. We did not perform contrast-enhanced ultrasound or MRI.
For further diagnostic work-up biopsy under ultrasound guidance was performed, and a 16 G SurecutTM biopsy needle (Boston Scientific, Natick, MA) was used. The histological specimens of two representative lesions showed features of large regenerating nodules with changes such as architectural distortion, cirrhosis and steatosis of hepatocytes (Figure 3).
View larger version (153K):
[in this window]
[in a new window]
|
Figure 3. Histology: regenerative nodule with fibrocirrhotic changes (white arrows) surrounding partly steatotic hepatocytes (black arrows) (Chronotrop-Aniline-Blue stain, magnification 40 x ).
|
|
|
Discussion
|
|---|
Cirrhotic nodules and dysplastic nodules (DN), with different grades of histological differentiation, are viewed as a spectrum of changes towards possible hepatocellular carcinoma [1] with up to 21.5% of nodules in cirrhotic livers proving to be dysplastic [2]. Cirrhotic or regenerative nodules (RN) and DN can show a variety of ultrasound and CT patterns [3]. Many malignant liver lesions may typically present a sonolucent halo, the so called target sign or bulls-eye lesion [4]. The histomorphological aetiology of this sonolucent halo is controversial and has been correlated with an aggressive growth pattern, although it is non-specific and can also be seen in hepatocellular carcinoma, adenoma, focal nodular hyperplasia, haemangioma, lymphoma and fungal microabscesses [4].
However, in our patient, the rim was hyperechoic, which is believed to correspond with a histological pattern of septal fibrosis and increased vascularity (Figure 3
). Owing to this sonomorphological pattern, we have termed it the "reverse" target sign. The ultrasound patterns of metastatic liver disease are variable, and correlated with the histology of the primary tumour can be echogenic, hypoechoic, cystic, infiltrative, and may even calcify [4]. Haemangiomas larger than 3 cm can also have the sonomorphological pattern of the reverse target sign and should be considered in the differential diagnosis [5]. On CT the typical nodular dynamic contrast enhancement with "filling-in" centripetally is missing, the lesions are more radio-opaque than haemangiomas, and haemangiomas do not occur in the disseminated fashion seen in our patient.
The multiplicity of the lesions and missing signs of cirrhosis, such as contour irregularity, led us to the hypothesis of metastatic liver disease. Surface nodularity of the liver has the highest diagnostic accuracy and specificity for cirrhosis compared with caudate lobe hypertrophy and hepatic venous blood flow patterns in ultrasound [6], and it might have been useful to assess the liver surface, using a high frequency linear array probe.
The role of MRI in the diagnosis of nodular liver lesions is still uncertain [3]. Ferumoxide-particles administered intravenously are cleared by phagocytosis by Kupffer cells, and this shortens the T2* signal [7]. DN and RN have an almost identical or even increased number of Kupffer cells compared with the surrounding parenchyma. Metastatic tumours are devoid of Kupffer cells; therefore ferumoxide-enhanced MRI can help to differentiate metastases from RN. But moderately and poorly differentiated HCC development in a cirrhotic liver are also found to have a significantly decreased number of Kupffer cells [8]. MRI is more accurate in the detection of nodular liver lesion in the cirrhotic liver than CT but tumour size is a restricting factor and small lesions can easily be missed [9].
The characterization of focal liver lesions with Gd-BOPTA and dynamic contrast enhanced MRI may be of greater importance in the future [10].
In summary, the diagnostic work-up of nodular liver lesions in some patients, even using all radiodiagnostic modalities, can be misleading and lead to misdiagnosis. Clinical and laboratory findings, as well as patient history may help in achieving the correct diagnosis. Although liver biopsy can yield false-negative results in about one-third of patients, for accurate characterization of liver nodules biopsy and histology remains the gold standard [11].
Received for publication June 21, 2004.
Revision received November 12, 2004.
Accepted for publication January 5, 2005.
|
References
|
|---|
- International Working Party. Terminology of nodular hepatocellular lesions. Hepatology 1995;22:983–93.[CrossRef][Medline]
- Terada T, Terasaki S, Nakanuma Y. A clinicopathologic study of adenomatous hyperplasia of the liver in 209 consecutive cirrhotic livers examined by autopsy. Cancer 1993;72:1551–6.[CrossRef][Medline]
- Lim JH, Choi BJ. Dysplastic nodules in liver cirrhosis: imaging. Abdom Imaging 2002;27:117–8.[CrossRef][Medline]
- Harvey CJ, Albrecht T. Ultrasound of focal liver lesions. Eur Radiol 2001;11:1578–93.[CrossRef][Medline]
- Leifer DM, Middleton WD, Teefey SA, Menias CO, Leahy JR. Follow-up of patients at low risk for hepatic malignancy with a characteristic hemangioma at US. Radiology 2000;214:167–72.[Abstract/Free Full Text]
- Colli A, Fraquelli M, Andreoletti M, Marino B, Zuccoli E, Conte D. Severe liver fibrosis or cirrhosis: accuracy of US for detection-analysis of 300 cases. Radiology 2003;227:89–94.[Abstract/Free Full Text]
- Stark DD, Weissleder R, Elizondo G, Hahn PF, Saini S, Todd LE, et al. Superparamagnetic iron oxide: clinical application as a contrast agent for MR imaging of the liver. Radiology 1988;168:297–301.[Abstract/Free Full Text]
- Tanaka M, Nakashima O, Wada Y, Kage M, Kojiro M. Pathomorphological study of Kupffer cells in hepatocellular carcinoma and hyperplastic nodular lesions in the liver. Hepatology 1996;24:807–12.[CrossRef][Medline]
- de Ledinghen V, Laharie D, Lecesne R, Le Bail B, Winnock M, Bernard PH, 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 Hepatol 2002;14:159–65.[CrossRef][Medline]
- Schneider G, Grazioli L, Morana G, Kirchin M, Seidel R, Altmeyer K. Characterization of focal liver lesions with Gd-BOPTA: a review of imaging findings on unenhanced, dynamic, and hepatobiliary phase MR imaging. (Abstract) RSNA 2003 URL: http://rsna2003.rsna.org/rsna2003/VBK/conference/event_display.cfm?em_id=3106019 (cited 10 December 2003).
- Poniachik J, Bernstein DE, Reddy KR, Jeffers LJ, Coelho-Little ME, Civantos F, et al. The role of laparoscopy in the diagnosis of cirrhosis. Gastrointest Endosc 1996;43:568–71.[CrossRef][Medline]
Top
Abstract
Introduction
