The Oncologist, Vol. 9, No. 4, 385–397,
July 2004
© 2004 AlphaMed Press
Imaging the Liver
Dushyant V. Sahani,
Sanjeeva P. Kalva
Division of Abdominal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Boston, Massachusetts, USA
Correspondence: Dushyant V. Sahani, M.D., Division of Abdominal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, White 270, 55 Fruit Street, Boston, Massachusetts 02114, USA. Telephone: 617-726-8396; Fax: 617-726-4891; e-mail: dsahani{at}partners.org
 |
LEARNING OBJECTIVES
|
|---|
After completing this course, the reader will be able to:
- Select an appropriate imaging modality to diagnose a suspected liver tumor or stage an existing liver tumor.
- Describe the relative advantages and disadvantages of various imaging modalities in characterizing liver tumors.
- Explain the common radiological findings in the diagnosis of focal and diffuse liver disease.
Access and take the CME test online and receive 1 hour of AMA PRA category 1 credit at CME.TheOncologist.com
|
ABSTRACT
|
|---|
Imaging of the liver is undertaken for the detection and characterization of suspected primary or secondary neoplasms, prior to planning a surgery or chemotherapy pump placement, for assessing treatment response, for evaluating biliary pathology, and for screening for liver neoplasms in high-risk groups. In this article, we review the advantages and disadvantages of various imaging modalities in the evaluation of the liver and formulate guidelines for the imaging of common clinical indications. A brief review of imaging findings in focal and diffuse liver disease is also presented.
Key Words. CT • MRI • PET • Liver tumors
|
INTRODUCTION
|
|---|
The liver is an important organ from an oncologic perspective. Primary hepatic neoplasms are common, especially in the presence of diffuse liver disease such as cirrhosis, hemochromatosis, and steatohepatitis. The liver is the most common site of metastasis from gastrointestinal tumors. High blood flow (about 25% of cardiac output), a favorable microscopic anatomy (liver sinusoids and gaps in subendothelial basement membrane), and a rich biochemical environment favor the rapid growth of metastatic deposits in the liver [1]. The objectives of liver imaging in oncology are the detection of the liver disease, the characterization of liver lesions, the staging of neoplasms, the evaluation of biliary ductal status, the evaluation of treatment response, and the assessment of vascular anatomy for surgical planning and chemotherapy pump placement [2]. It is important to understand the utility of various imaging modalities to optimally address the clinical question at hand.
|
ULTRASONOGRAPHY
|
|---|
Ultrasonography (US) is inexpensive and easily available. It is an excellent test to screen the liver for biliary obstruction or gall bladder disease and to assess vascular patency. It is highly sensitive at differentiating a cyst from a solid liver lesion. However, it is not as sensitive as computerized tomography (CT) or magnetic resonance imaging (MRI) at detecting focal, solid liver lesions [3]. Though a few experienced operators have quoted high detection rates for colorectal liver metastases [4] and hepatocellular carcinoma (HCC) [5] with ultrasound, similar results could not be reproduced in the United States, which may be due to the patient body habitus and subspecialty practice patterns. The reported sensitivity of ultrasound for the detection of liver metastases varies from 40%–70% [6]. The main limitations of US are high operator dependency, inability to detect lesions <1 cm in size, and low specificity. The presence of diffuse liver disease also lowers the sensitivity of US for the detection of focal lesions. Similarly, pseudolesions, such as focal fatty infiltrations or focal fatty sparings, are sometimes difficult to differentiate from other pathologic liver lesions. On the other hand, intraoperative US (IOUS) and the recently introduced laparoscopic US are highly sensitive for detecting liver lesions not seen on routine preoperative imaging, for assessing the relationship between tumors and hepatic vessels, and for assessing vascular patency [7, 8]. Likewise, endoscopic US (EUS) is useful for assessing the left lobe of the liver and the lymph nodes in the gastrohepatic ligament, and fine-needle aspiration of liver lesions can be performed under EUS guidance [9]. The recent addition of US contrast agents (not yet approved in the United States) for imaging the liver has shown promise in the characterization of various hepatic tumors [10].
|
COMPUTERIZED TOMOGRAPHY
|
|---|
CT offers the best spatial resolution and the ability to study the entire liver in a single breath-hold. It serves as an ideal screening examination for the entire abdomen and pelvis. Recent technological advances in CT technology, such as helical CT and multidetector row helical CT, have further improved the performance of CT scanners in terms of speed of acquisition, resolution, and the ability to image the liver during various phases of contrast enhancement more precisely than was possible previously [11]. Advances in image postprocessing and reconstruction methods have enabled the acquisition of three-dimensional (3D) images of the liver vasculature (CT angiography) to map the liver vascular anatomy and to define the liver and tumor volume.
Intravenous iodinated contrast media are routinely used in the imaging of the liver. They improve the contrast-to-noise ratio between focal liver lesions and normal liver and thus aid in the detection of focal liver lesions. They also help to characterize liver lesions, based on the enhancement patterns of liver lesions during various phases of contrast circulation in the liver [12]. When performed properly, CT suffices for most clinical indications. Its limitations include the need for a high radiation dose and a low sensitivity for the detection and characterization of lesions smaller than 1 cm. Contrast-enhanced CT is contraindicated in patients with a history of anaphylaxis from contrast agents and renal failure. CT fluoroscopy is a new tool that assists in performing biopsies of liver lesions. Current multislice CT fluoroscopy systems allow real-time monitoring of the needle during biopsies and may increase the yield of biopsies and decrease the time required for performing a biopsy, with an acceptable radiation dose [13].
|
MAGNETIC RESONANCE IMAGING
|
|---|
MRI has emerged as the best imaging test for liver lesion detection and characterization, because this modality provides high lesion-to-liver contrast and does not use ionizing radiation. Recent advances in MRI, including breath-hold 3D imaging and rapid half-Fourier acquisition, help image the liver in a single breath-hold with a high spatial resolution. In addition, chemical shift imaging is very useful to differentiate pseudolesions, such as focal fatty infiltrations and focal fatty sparings, from pathologic liver lesions (Fig. 1
). Various contrast agents are available to image the liver [14]. Gadolinium diethylenetriaminepentaacetic acid (DTPA), the most commonly used MRI contrast agent, has an extracellular distribution and behaves similarly to the iodinated contrast agents used in CT. Its main applications include the characterization of liver tumors and MR angiography. Liver-specific contrast media, such as mangafodipir trisodium (taken up by hepatocytes) and ferrumoxides (taken up by Kupffer cells), demonstrate selective uptake in the liver and are primarily used for lesion detection [15]. These two contrast agents are also useful in characterizing specific liver tumors, such as fibronodular hyperplasia, hepatic adenoma, and HCC.
View larger version (85K):
[in this window]
[in a new window]
|
Figure 1. Imaging study of a 54-year-old male with a round cell tumor of the nasal cavity. A) Unenhanced CT demonstrates a large irregular lesion in the right lobe of the liver. Chemical shift imaging with MRI obtained B) in phase and C) out of phase confirms that the lesion is focal fatty infiltration.
|
|
The main advantages of contrast-enhanced MRI include a high spatial resolution, better contrast sensitivity, better lesion detection and characterization than with CT, and lack of ionizing radiation. MRI contrast agents can be safely used in patients with renal failure. In addition, in patients with allergies to contrast media, unenhanced MRI is superior to unenhanced CT for lesion detection and characterization. MRI contrast agents can safely be used in patients with allergies to iodinated contrast agents. The main drawbacks of MRI include its high cost, a long procedure time, and the need for the patient to hold his breath for longer periods.
|
ANGIOGRAPHY
|
|---|
Catheter angiography is the gold standard for vascular evaluation of the liver prior to tumor resection. However, with the present advances in CT and MR technology, the noninvasive evaluation of hepatic vessels can be performed reliably by contrast-enhanced CT or MRI. At present, catheter angiography is performed for therapeutic liver tumor embolization and to assess complex vascular anatomy demonstrated on CT or MRI.
|
POSITRON EMISSION TOMOGRAPHY
|
|---|
Positron emission tomography (PET) has emerged as an important diagnostic tool in the evaluation of metastatic liver disease. A greater metabolic activity in malignant tissue is accompanied by a greater glucose uptake relative to that of surrounding normal tissue. This greater focal glucose uptake can be identified with 18F-fluoro-2-deoxy-D-glucose (FDG)-PET, which allows for the identification of malignant tumor foci [16]. This procedure is highly sensitive; however, any focal area of hypermetabolism can give false-positive results. The advantages are its high sensitivity and the ability to survey the entire body at a single sitting. The main disadvantages include its high cost, poor availability, poor lesion localization, and limited sensitivity for lesions smaller than 1 cm [17]. PET-CT combines the advantages of CT with the functional ability of PET, by the fusion of PET images with CT images acquired at the same time, and helps to accurately localize the greater metabolic activity [18]. However, its use in the evaluation of liver tumors needs further evaluation [19]. PET was not found to be useful in detecting hepatomas [5], and its ability to detect liver metastases is comparable with that of MRI [20]. However, it is extremely useful for detecting extrahepatic metastatic disease.
|
CHOICE OF IMAGING TEST
|
|---|
The choice of imaging test largely depends on the clinical question, availability, the clinician’s familiarity with the test, and the patient’s clinical condition. In general, US and CT remain the first imaging tests for screening and characterizing most patients with suspected liver tumors.
Suspected Liver Metastases
CT is the imaging modality of choice for evaluating suspected liver metastases. Unenhanced CT is useful in a few cases, especially for detecting calcifying or hemorrhagic metastases from colon or breast cancer or melanoma and for detecting metastases from neuroendocrine tumors. A technically adequate contrast-enhanced CT of the abdomen can detect metastatic liver disease, the presence of diffuse liver disease, and biliary ductal abnormalities and helps to survey the entire abdomen for potential lymph nodal and peritoneal metastases. It also aids in mapping the vessels for surgery or chemotherapy pump placement (Fig. 2). Indeterminate lesions on CT can be evaluated with MRI. Diffuse liver disease and fatty infiltration limit the sensitivity of CT in lesion detection. If the clinical suspicion of a lesion is high and the CT scan is negative, MRI or PET can be considered as alternative tests for further assessment. MRI with liver-specific contrast agents has a sensitivity equal to or better than CT arterioportography [21, 22]. In a meta-analysis comparing US, CT, MRI, and PET in the detection of hepatic metastases from gastrointestinal tract cancers, Kinkel et al. reported the highest sensitivity for FDG-PET [23]. Recently, Yang et al. reported equivalent sensitivities and specificities for MRI and PET in detecting liver metastases [20].
View larger version (152K):
[in this window]
[in a new window]
|
Figure 2. CT angiography oblique maximum intensity projection in a patient scheduled for liver resection. The left hepatic artery is seen arising from left gastric artery.
|
|
Preoperative Evaluation of Liver Tumors
Patient selection and surgical planning are the two broad objectives of preoperative evaluation of patients for liver resection for primary or secondary neoplasms. Contrast-enhanced MRI is superior to CT in view of its high sensitivity for detecting small lesions (Fig. 3) and its ability to characterize small lesions, which may be indeterminate on CT. Tissue characterization with T1- and T2-weighted images, a high spatial resolution, and a high sensitivity for contrast enhancement contribute to the greater accuracy of MRI versus CT. MRI findings may alter planned procedures. Schultz et al. reported altered therapeutic management in 67% of patients assessed by ferrumoxide-enhanced MRI of the liver, with the added benefit of a significant cost reduction [24]. IOUS is highly sensitive and can detect more lesions than preoperative MRI. However, in our experience, it has rarely altered the planned procedure following contrast-enhanced MRI. CT and MR angiography provide excellent opportunities for vascular mapping and delineation of tumor spread into vessels and assist in surgical planning. FDG-PET has not yet been established for preoperative evaluation. It can be used to exclude extrahepatic disease not evident on CT that may alter the planned surgery. Lai et al. demonstrated previously unsuspected extrahepatic disease in 32% of patients scheduled for hepatic metastatectomy, predominantly involving the celiac lymph nodes [25]. Recently, Desai et al. reported that PET had altered therapy (from that planned after CT imaging) in 40% of patients with recurrent or metastatic colorectal cancer and showed extrahepatic disease in 72% of patients that was not found on CT [26]. PET-CT, as mentioned above, may assist in the accurate localization of extrahepatic disease.
View larger version (69K):
[in this window]
[in a new window]
|
Figure 3. Imaging study of a 55-year-old male patient with pancreatic adenocarcinoma. A) CT scan shows single focal liver metastasis in the medial segment of left lobe of the liver (arrow). B) Mangafodipir trisodium-enhanced MRI shows multiple lesions in both liver lobes (arrows).
|
|
Screening for Liver Tumors in Diffuse Hepatic Disease
There is no formal recommendation of any imaging tests for the screening of liver tumors in patients with diffuse liver diseases such as cirrhosis, hemochromatosis, and hepatosteatosis. Based on a Markov model, Arguedas et al. reported that screening for HCC with CT was a cost-effective strategy in transplant-eligible patients with cirrhosis secondary to chronic hepatitis C viral infection [27]. Markov models are useful when a decision problem involves risk that is continuous over time, when the timing of events is important, and when important events may happen more than once. Markov models assume that a patient is always in one of a finite number of discrete health states, called Markov states. All events are represented as transitions from one state to another. A Markov model may be evaluated by matrix algebra, as a cohort simulation, or as a Monte Carlo simulation [28]. Screening with alpha-fetoprotein (AFP) for HCC was found to be more useful in patients with cirrhosis with nonviral etiologies [29]. Though US can be used for initial screening, CT or MRI should be used once cirrhosis sets in. In patients with equivocal findings on CT, MRI is useful for its ability to accurately identify fatty infiltration and small HCCs (Fig. 4).
View larger version (120K):
[in this window]
[in a new window]
|
Figure 4. Imaging study of a 45-year-old man with cirrhosis and elevated AFP level. A) CT scan demonstrates no focal liver lesion. B) Gadolinium DTPA-enhanced MRI reveals multiple focal enhancing liver lesions consistent with multifocal HCC (arrows).
|
|
Incidentally Detected Lesions with Other Imaging Modalities
Liver lesions incidentally detected on imaging should be characterized. Although CT can characterize most liver lesions, MRI is considered superior for characterizing liver tumors. Lesions that are indeterminate or considered too small to be characterized on CT may benefit from MRI. Occasionally, MRI may fail to completely characterize liver lesions and no formal recommendation currently exists for such lesions. Follow-up imaging to demonstrate stability (suggesting benign lesions) or progression (suggesting malignant lesions) or biopsy is recommended for these lesions.
Assessing Tumor Response to Surgery or Chemotherapy
CT is the imaging modality of choice to assess tumor response or recurrence following resection, intravenous chemotherapy, intra-arterial chemoembolization, ethanol ablation, or radiofrequency ablation. MRI and PET may be used for problem solving in cases where CT results are indeterminate. The presence of enhancement along surgical margins usually represents postoperative changes, but nodular enhancement or a discrete soft-tissue mass at the surgical margins indicates recurrence. PET is not useful within 2 months of resection due to a high false-positive rate, which may be the result of inflammation or granulation tissue. Rarely, biopsy may be required. Following chemotherapy or chemoembolization, changes in tumor measurements help assess response. Some drugs may result in changes to the liver architecture and may induce changes in the size and shape of the liver as a result of chemotoxicity. The classic example is methotrexate-induced fibrosis in liver parenchyma. Contrast-enhanced CT is useful in the assessment of complete tumor necrosis following radiofrequency ablation. In a small group of patients, Donckier et al. showed a high sensitivity of PET for detecting local recurrence following radiofrequency ablation of liver tumors [30].
|
IMAGING OF FOCAL LIVER DISEASE
|
|---|
Benign Liver Tumors
Hepatic Cysts
Hepatic cysts, either single or multiple, are common and occur in 2%–7% of the general population. Cysts vary in size from a few millimeters to several centimeters. On US, they appear as well-defined, anechoic lesions with distal acoustic enhancement and thin imperceptible walls. The presence of nodules or septae may indicate infection or neoplasm. On CT, these lesions appear similar to how they appear on US, with low attenuation (0–10 HU), and do not enhance on contrast study. On MRI, these lesions follow the signal intensity of cerebrospinal fluid on all pulse sequences. Any enhancement on CT or MRI indicates infection, inflammation, or neoplasm.
Hemangiomas
Hemangiomas are the most common benign tumors of the liver. They may be isolated or multiple and are more common in women [31]. Their size may vary from <1 cm to 10 cm. Large hemangiomas show varied imaging features due to internal hemorrhage and fibrosis. Both CT and MRI can characterize hemangiomas based on their distinctive enhancement pattern (Fig. 5). The enhancement pattern is classically described as "peripheral globular enhancement with progressive fill-in" [32]. MRI offers the additional benefit of T2-weighted imaging, in which heavy T2-weighted imaging with an echo time >112 ms is generally used to differentiate hepatic hemangiomas from malignant lesions, as the former retain their higher signal on this sequence [33]. The sensitivity and specificity of contrast-enhanced MRI exceed 95% in diagnosing hemangiomas, and technitium-labeled red blood cell blood-pool scintigraphy is rarely needed to characterize hemangioma [34]. However, atypical hemangiomas are sometimes difficult to differentiate from metastases and require biopsy for confirmation.
View larger version (138K):
[in this window]
[in a new window]
|
Figure 5. Imaging study of a 40-year-old male with incidentally detected lesion on US who was further studied for lesion characterization on MRI. A) Gadolinium DTPA-enhanced MRI demonstrates a hypointense lesion with peripheral globular enhancement on arterial phase images, which progressively fills in during delayed images (B). These features are diagnostic of hemangioma.
|
|
Focal Nodular Hyperplasia
Focal nodular hyperplasia occurs in young women and is more common in women taking oral contraceptives [35]. These are benign lesions and do not require any treatment unless causing mass effects or pain. They are common under the liver surface and usually measure <5 cm. On US, these lesions appear hypoechoic. On unenhanced CT, they are hypodense to liver and show significant enhancement on arterial phase images and become isodense to liver on portal venous phase images and delayed phase images. The central scar may enhance on delayed images. As the lesion is isodense to liver on unenhanced and portal venous phase CT, even relatively large lesions may be missed if an arterial phase study is not obtained. On MRI, these lesions are isointense on T1-weighted images and isointense to mildly hyperintense on T2-weighted images. The central scar is bright on T2-weighted images. Gadolinium-enhanced MRI demonstrates features similar to those observed on contrast-enhanced CT (Fig. 6). Small lesions may not demonstrate the central scar and may be difficult to differentiate from other malignant lesions. Due to the presence of Kupffer cells, these lesions lose signal on ferrumoxide-enhanced MRI [36]. The sensitivity and specificity of MRI exceed 95% for diagnosing focal nodular hyperplasia. Focal nodular hyperplasia shows increased uptake on sulfur colloid scintigraphy. This uptake depends on the percentage of Kupffer cells in the tumor.
View larger version (92K):
[in this window]
[in a new window]
|
Figure 6. Imaging study of a 35-year-old female patient with an incidentally detected lesion on US who was further studied for characterization on MRI. A) T2-weighted image shows mildly hyperintense lesion with central bright scar (arrow). B) On arterial phase images, there is significant enhancement of the tumor. C) The scar (arrow) enhances on the delayed images. These features are diagnostic of fibronodular hyperplasia.
|
|
Hepatic Adenomas
Hepatic adenomas are less common than focal nodular hyperplasia and occur in young women taking oral contraceptives. They also occur in men who use anabolic steroids. Patients present with acute abdominal pain or hepatomegaly, due to their large size, or intratumoral hemorrhage. Surgery is usually indicated for bleeding adenomas. Hepatic adenomas are large, measuring 5–10 cm, and have well-defined capsules. These lesions often show intratumoral hemorrhage and infarction on pathology. This results in varied imaging appearances on US, CT, and MRI. On unenhanced CT, they are hypodense, but lesions with hemorrhage appear hyperdense or heterogeneous [37]. On contrast study, these lesions show significant enhancement on arterial phase images as they are supplied by the hepatic artery. As their characteristics overlap with those of focal nodular hyperplasia, a heterogeneous appearance or the presence of hemorrhage on an unenhanced CT scan is highly diagnostic. The presence of a capsule and demonstration of intralesional fat on out-of-phase T1-weighted images are helpful in diagnosing hepatic adenomas on MRI. The enhancement pattern on gadolinium-enhanced MRI is similar to that observed on CT. The imaging characteristics of hepatic adenoma overlap with those of hepatomas, including the presence of intracellular fat, and sometimes it may be difficult to differentiate these two entities on imaging.
Hepatic Adenomatosis
Hepatic adenomatosis is a distinct entity and is unrelated to oral contraceptive or anabolic steroid use. The number of adenomas varies between 10 and 50, and these tumors demonstrate similar imaging features (Fig. 7) as described above for hepatic adenomas. However, they tend to be progressive, symptomatic, and are more likely to lead to impaired liver function, hemorrhage, and malignant degeneration [38]. Follow-up with CT or MRI may be required in these patients due to the high incidence of hemorrhage and malignancy.
View larger version (65K):
[in this window]
[in a new window]
|
Figure 7. Imaging study of a 35-year-old female with abdominal pain. A) Ultrasound shows three focal hypoechoic lesions. B, C) Contrast-enhanced CT demonstrates multiple small enhancing lesions. Biopsy of two lesions demonstrated hepatic adenomatosis.
|
|
Benign Lipomatous Tumors
Benign lipomatous tumors include lipoma, myelolipoma, and angiomyelolipoma. Their imaging features depend on the amount of fat in the lesion. CT and MRI demonstrate fat in simple lipoma, which does not enhance on contrast study. Angiomyelolipoma shows enhancing soft tissue in addition to fat [39]. This should be interpreted with caution if significant soft-tissue components are present, because HCC can contain fat.
Other Benign Tumors
Mesenchymal hamartoma is a rare benign lesion that occurs in children less than 2 years of age. Its appearance on imaging studies depends on the relative contribution of stromal elements and cysts. The appearance of mesenchymal hamartoma on CT and MRI varies from that of a solid mass containing multiple small cysts to a multilocular cystic mass with septations and enhancing solid portions. Hemangioendotheliomas occur in infants less than 6 months of age. These lesions are large and well defined and show an enhancement pattern similar to that of hemangioma. Bile duct hamartomas are incidental lesions and may be confused with metastases on CT. They measure <1.5 cm and are irregular and hypodense on unenhanced CT, with no enhancement on contrast study. MRI appearances resemble cysts, but contrast study may reveal homogeneous or rim enhancement or no enhancement. Biliary cystadenoma and cystadenocarcinoma are rare and present with a large solitary cystic mass with a well-defined thick fibrous capsule, mural nodules, and internal septae [40].
Malignant Hepatic Lesions
Hepatocellular Carcinoma
HCC is the most common primary malignancy of the liver. HCC can be focal, multifocal, or diffuse. On sonography, these lesions appear hypoechoic. Color Doppler study may show intratumoral vessels and arteriovenous shunting. The presence of arterial waveforms on color Doppler sonography distinguishes tumor thrombi from bland thrombi in the portal vein. Contrast-enhanced US has shown promise in differentiating malignant tumors from nonmalignant tumors [41] based on their enhancement patterns, the presence of intratumoral vessels, and the detection of early hepatoportal shunting, associated with hepatomas. Sonography is also useful to guide percutaneous biopsy.
On unenhanced CT, HCCs are well defined and appear hypodense to liver. In diffuse liver disease, the lesions may not be seen on unenhanced CT. These lesions are actively enhancing on arterial study and appear hyperdense to liver. On portal venous phase images, they appear hypodense to liver. When a capsule is present, it is usually hypodense on hepatic arterial phase images, is of mixed density on portal venous phase images, and shows enhancement on delayed phase images. CT helps detect tumor extension into the portal vein or hepatic veins and the presence of biliary obstruction, regional nodes, and peritoneal implants [42]. It is difficult to differentiate tumor thrombi from bland thrombi; however, enhancing thrombi favor tumor thrombi. CT is also useful in guiding percutaneous biopsy and follow-up after surgery or radiofrequency ablation for detecting recurrence.
On T1-weighted MRI, these lesions usually appear hypointense to liver and are moderately hyperintense on T2-weighted images; however, their appearances may be variable. The critical sequence in the detection of HCC is dynamic gadolinium-enhanced imaging [43]. HCC, being supplied by the hepatic artery, shows significant enhancement on arterial study and rapid washout on portal venous phase images. The capsule is hypointense on T1- and T2-weighted images and may show delayed enhancement. Arterial phase MRI was found to be superior to arterial phase CT in detecting hypervascular HCC [44]. MRI is equally sensitive as CT in detecting vascular invasion (Fig. 8). On ferrumoxide-enhanced MRI, these lesions appear hyperintense to normal liver. Rarely, well-differentiated HCCs may accumulate ferrumoxides. In the presence of cirrhosis, ferrumoxide-enhanced MRI was found to be superior to dual-phase helical CT for the depiction of small HCCs [45].
View larger version (75K):
[in this window]
[in a new window]
|
Figure 8. Imaging study of a 52-year-old male patient with cirrhosis and HCC. A) Axial T2-weighted image shows an isointense mass within the inferior vena cava (IVC) (arrow). B) Gadolinium-enhanced T1-weighted image shows a mass within the hepatic vein, with minimal enhancement within the mass (arrow), suggesting tumor extension within the hepatic vein and IVC.
|
|
Fibrolamellar HCC is distinct from other forms of HCC and is more common in patients with no existing diffuse liver disease, and in women. The tumor is usually large at presentation. This lesion is hypointense on T1-weighted images and hyperintense on T2-weighted images with a central nonenhancing hypointense scar.
Cholangiocarcinomas
Cholangiocarcinomas arise from the epithelium of small intrahepatic bile ducts. These tumors demonstrate variable echotextures due to the presence of varying amounts of fibrosis and necrosis. Approximately 60%–70% of cholangiocarcinomas occur at the hepatic duct bifurcation, and the remainder occur in the distal common bile duct (20%–30%) or within the liver (5%–15%) [46]. The associated biliary ductal dilatation is well demonstrated on US. The tumor itself may not be visible on US or on CT. On unenhanced CT, the lesion is iso- to hypodense and shows delayed enhancement. On MRI, the lesion is hypointense on T1-weighted images and hyperintense on T2-weighted images, with delayed enhancement on contrast study (Fig. 9). Capsular retraction is common. Diffuse infiltrating types are best evaluated with MRI. Multicentric cholangiocarcinoma can occasionally occur where multiple focal cholangiocarcinomas arise de novo in the liver. Mangafodipir trisodium-enhanced MRI and ferrumoxide-enhanced MRI are useful in detecting small lesions. Magnetic resonance cholangiopan-creatography is extremely useful in planning surgery and palliative therapy.
View larger version (150K):
[in this window]
[in a new window]
|
Figure 9. Imaging study of a 47-year-old female with jaundice and an abdominal mass due to cholangiocarcinoma. A) Gadolinium-enhanced MRI shows a large minimally enhancing mass in the right lobe of the liver. B) There is a central hypointense scar (arrow) that enhances on the delayed image.
|
|
Metastases
Metastases comprise the most common malignant liver neoplasms. On US, metastases appear as round or oval hypoechoic lesions; central areas of necrosis may appear hypo- or anechoic. Metastases from mucinous adenocarcinomas and calcifying metastases appear hyperechoic. Unenhanced CT is rarely necessary for the evaluation of liver metastases except for hemorrhagic or calcifying metastases. Metastases from neuroendocrine tumors appear hyperdense on unenhanced CT. On arterial-phase helical CT, metastases appear as well-defined or ill-defined hypodense lesions and may show minimal peripheral rim enhancement. Portal venous phase study is important in screening for metastases. On portal venous phase images, metastases appear hypodense to adjacent enhancing liver. Some lesions may show the peripheral washout sign. This refers to peripheral enhancement during the arterial and portal venous phases, which disappears in the delayed phase. On CT arterioportography, metastases appear hypodense. Multidetector CT is superior to conventional CT, and its sensitivity reaches that of CT arterioportography in evaluating liver metastases. On MRI, metastases appear hypointense on T1-weighted images and hyperintense on T2-weighted images, with enhancement patterns similar to those observed with CT (Fig. 10). MRI with liver-specific contrast agents was found to be more sensitive than CT in detecting liver metastases [47, 48]. In a meta-analysis comparing US, CT, MRI, and PET for the detection of liver metastases from colorectal, gastric, and esophageal cancers, Kinkel et al. found that, in studies with a specificity higher than 85%, the mean weighted sensitivity was 55% (95% confidence interval [CI] = 41–68) for US, 72% (95% CI = 63–80) for CT, 76% (95% CI = 57–91) for MRI, and 90% (95% CI = 80–97) for FDG-PET [23].
View larger version (74K):
[in this window]
[in a new window]
|
Figure 10. Imaging study of a 52-year-old female with colon cancer with metastases to the liver. A) Gadolinium-enhanced MRI during the arterial phase reveals peripheral rim enhancement of the lesions (arrows). B) On the delayed image, these exhibit peripheral washout (arrows).
|
|
|
IMAGING OF DIFFUSE LIVER DISEASE
|
|---|
Diffuse liver disease comprises many different pathological conditions, including abnormalities of metabolism, infections, chronic injury due to toxins, and malignant diseases. Many of the diffuse liver diseases predispose to the development of primary hepatic neoplasms. The aim of imaging in diffuse liver disease is to characterize some of these conditions, detect early hepatic neoplasms, and differentiate benign lesions from their malignant counterparts in the presence of diffuse liver disease.
Because of its intrinsic soft tissue resolution, that is, the ability to characterize tissue based on T1- and T2-weighted image characteristics, MRI is suitable for imaging diffuse liver disease. Diffuse liver disease, based on its distribution and the presence of abnormal signals, can be divided in four imaging patterns [49]. A diffuse homogenous distribution is observed in hemochromatosis, steatohepatitis, and glycogen storage diseases. Segmental distribution is seen with focal fatty infiltration and subacute hepatitis. A diffuse nodular distribution is seen with postviral cirrhosis, Wilson’s disease, and sarcoidosis. Congested liver and Schistosomiasis japonica infection result in perivascular abnormalities.
Hemochromatosis
Primary hemochromatosis is due to genetic mutation in the HFE gene; however, the cause is not known in juvenile and neonatal hemochromatosis. Secondary hemochromatosis may result from excessive transfusions, as in thalassemia patients. Iron overload results in increased attenuation on CT, and the liver appears uniformly hyperdense. The sensitivity of CT at diagnosing hemochromatosis is high if iron overload exceeds five times the normal value, and it falls dramatically if the overload is less than 2.5 times normal [50]. Additionally, the presence of diffuse fatty infiltration concomitantly results in low attenuation. MRI is highly sensitive at diagnosing iron overload, as the liver turns dark on T2-weighted images (Fig. 11). It is better appreciated on the gradient echo T2* (T2 star) weighted pulse sequence due to its inherent high magnetic susceptibility. The presence of HCC can be detected with high confidence, as it does not contain iron and, as such, appears bright against the dark background of normal liver on T2-weighted images.
View larger version (68K):
[in this window]
[in a new window]
|
Figure 11. Hemochromatosis of liver. A) Axial T1-weighted and B) T2-weighted images reveal a low signal intensity from the liver versus the spleen, due to iron deposition in the liver.
|
|
Steatohepatitis and Fatty Liver
Pathologically, steatohepatitis demonstrates varying degrees of steatosis (fatty liver), mixed cellular inflammatory infiltrate across the lobule, the presence of hepatocyte injury, and fibrosis [51]. This condition is unrelated to alcohol abuse. Radiologically, steatohepatitis cannot be reliably distinguished from other causes of fatty liver. The presence of >33% fat on liver biopsy was shown to be optimal for detecting steatosis on radiological imaging [52]. CT demonstrates uniform low density compared with spleen or muscle on noncontrast and contrast-enhanced CT. MRI is more reliable, as it detects microscopic fat better than CT. Chemical shift imaging techniques with MRI, such as in-phase and out-of-phase T1 gradient-recalled echo techniques, are useful, as the liver loses signal on out-of-phase imaging in the presence of intracellular fat. Fatty liver occurs in patients with diabetes, hyperalimentation, alcohol abuse, pregnancy, chemical toxicity, and transplanted liver. The segmental distribution results from regional differences in perfusion. The imaging features are similar to those of steatohepatitis except for localized segmental distribution.
Cirrhosis
Cirrhosis is an end result of chronic liver injury secondary to viral infections, alcohol abuse, and toxins. Pathologically, the liver demonstrates varying degrees of fibrosis and regenerative nodules. These nodules may be small or large. Cirrhosis secondary to Wilson’s disease demonstrates similar features to those secondary to viral or alcoholic cirrhosis. Other secondary changes follow, including portal hypertension and ascites. Imaging is not reliable in the diagnosis of cirrhosis. The role of imaging in cirrhosis is the early detection of HCC and the differentiation of regenerative nodules from dysplastic nodules and HCC. Regenerating nodules appear hypointense on T2-weighted MRI, as opposed to the hyperintense HCC. On both CT and MRI, dysplastic nodules and HCCs demonstrate arterial enhancement versus nonenhancing regenerative nodules [53]. However, in a recent report by Freeny et al., hyperattenuating nodules on arterial phase CT were either regenerating nodules or dysplastic nodules, and it was not possible to differentiate these nodules from HCC [54]. Ferrumoxide-enhanced MRI is highly useful in demonstrating HCCs in cirrhosis, as they do not contain functioning Kupffer cells and appear bright on T2-weighted postcontrast images [55]. It has a better sensitivity and accuracy for detecting malignant lesions in cirrhosis, but its specificity is less due to high false-positive rates [56]. Ward et al. reported superior sensitivity and accuracy for detecting HCC with double-contrast MRI with ferrumoxides and gadolinium DTPA [57]. However, well-differentiated HCCs have been reported to contain Kupffer cells and may take up ferrumoxides. The sensitivity of gadolinium-enhanced MRI for detecting HCC in cirrhosis increases with an increase in the size of the tumor. In one study, MRI had a 100% sensitivity for lesions larger than 2 cm, a 50% sensitivity for 1- to 2-cm lesions, and only a 4% sensitivity for lesions smaller than 1 cm [58].
Congested Liver
Passive venous congestion of the liver occurs with any cause resulting in right heart failure, constrictive pericarditis, and hepatic venous outflow obstruction. Inhomogeneous mottled reticular mosaic parenchyma and periportal low attenuation may appear on contrast enhanced CT [59]. MRI may demonstrate periportal hyperintensity secondary to perivascular lymphedema on T2-weighted images [49].
|
CONCLUSION
|
|---|
Hepatic imaging is usually undertaken to search for primary or metastatic liver disease. CT is the initial diagnostic test for most indications due to its versatility, availability, high sensitivity and specificity, and the fact that it surveys the entire abdomen for potential metastatic disease in the lymph nodes and peritoneum. MRI is superior to CT in lesion detection and characterization and should be used in cases in which CT is equivocal or noncontributory. PET is useful in evaluating the entire body for potential metastatic spread; however, its high cost and lack of availability make it a poor choice.
|
REFERENCES
|
|---|
- Robinson PJ. Imaging liver metastases: current limitations and future prospects. Br J Radiol 2000;73:234–241.[Abstract]
- Sica GT, Ji H, Ros PR. CT and MR imaging of hepatic metastases. AJR Am J Roentgenol 2000;174:691–698.[Free Full Text]
- Glover C, Douse P, Kane P et al. Accuracy of investigations for asymptomatic colorectal liver metastases. Dis Colon Rectum 2002;45:476–484.[CrossRef][Medline]
- Lamb G, Taylor I. An assessment of ultrasound scanning in the recognition of colorectal liver metastases. Ann R Coll Surg Engl 1982;64:391–393.[Medline]
- Teefey SA, Hildeboldt CC, Dehdashti F et al. Detection of primary hepatic malignancy in liver transplant candidates: prospective comparison of CT, MR imaging, US, and PET. Radiology 2003;226:533–542.[Abstract/Free Full Text]
- Paulson EK. Evaluation of the liver for metastatic disease. Semin Liver Dis 2001;21:225–236.[CrossRef][Medline]
- Schmidt J, Strotzer M, Fraunhofer S et al. Intraoperative ultrasonography versus helical computed tomography and computed tomography with arterioportography in diagnosing colorectal liver metastases: lesion-by-lesion analysis. World J Surg 2000;24:43–47; discussion 48.[CrossRef][Medline]
- Catheline JM, Turner R, Champault G. Laparoscopic ultrasound of the liver. Eur J Ultrasound 2000;12:169–177.[CrossRef][Medline]
- DeWitt J, LeBlanc J, McHenry L et al. Endoscopic ultrasound-guided fine needle aspiration cytology of solid liver lesions: a large single-center experience. Am J Gastroenterol 2003;98:1976–1981.[Medline]
- Furuse J, Nagase M, Ishii H et al. Contrast enhancement patterns of hepatic tumours during the vascular phase using coded harmonic imaging and Levovist to differentiate hepatocellular carcinoma from other focal lesions. Br J Radiol 2003;76:385–392.[Abstract/Free Full Text]
- Francis IR, Cohan RH, McNulty NJ et al. Multidetector CT of the liver and hepatic neoplasms: effect of multiphasic imaging on tumor conspicuity and vascular enhancement. AJR Am J Roentgenol 2003;180:1217–1224.[Abstract/Free Full Text]
- Saini S. Imaging of hepatobiliary tract. N Engl J Med 1997;336:1889–1894.[Free Full Text]
- Bissoli E, Bison L, Gioulis E et al. [Multislice CT fluoroscopy: technical principles, clinical applications and dosimetry.] Radiol Med (Torino) 2003;106:201–212. English, Italian.
- Hahn PF, Saini S. Liver-specific MR imaging contrast agents. Radiol Clin North Am 1998;36:287–297.[CrossRef][Medline]
- Kamel IR, Bluemke DA. MR imaging of liver tumors. Radiol Clin North Am 2003;41:51–65.[CrossRef][Medline]
- Zealley IA, Skehan SJ, Rawlinson J et al. Selection of patients for resection of hepatic metastases: improved detection of extrahepatic disease with FDG PET. Radiographics 2001;21:S55–S69.[Abstract/Free Full Text]
- Fong Y, Saldinger PF, Akhurst T et al. Utility of 18F-FDG positron emission tomography scanning on selection of patients for resection of hepatic colorectal metastases. Am J Surg 1999;178:282–287.[CrossRef][Medline]
- Israel O, Mor M, Gaitini D et al. Combined functional and structural evaluation of cancer patients with a hybrid camera-based PET/CT system using (18)F-FDG. J Nucl Med 2002;43:1129–1136.[Abstract/Free Full Text]
- Yao SZ, Zhang CQ, Chen J et al. [Clinical evaluation of 18F-FDE PET-CT in detecting malignant liver tumors.] Di Yi Jun Yi Da Xue Xue Bao 2003;23:1214–1216. Chinese.[Medline]
- Yang M, Martin DR, Karabulut N et al. Comparison of MR and PET imaging for the evaluation of liver metastases. J Magn Reson Imaging 2003;17:343–349.[CrossRef][Medline]
- Semelka RC, Cance WG, Marcos HB et al. Liver metastases: comparison of current MR techniques and spiral CT during arterial portography for detection in 20 surgically staged cases. Radiology 1999;213:86–91.[Abstract/Free Full Text]
- Vogl TJ, Schwarz W, Blume S et al. Preoperative evaluation of malignant liver tumors: comparison of unenhanced and SPIO (Resovist)-enhanced MR imaging with biphasic CTAP and intraoperative US. Eur Radiol 2003;13:262–272.[Medline]
- Kinkel K, Lu Y, Both M et al. Detection of hepatic metastases from cancers of the gastrointestinal tract by using noninvasive imaging methods (US, CT, MR imaging, PET): a meta-analysis. Radiology 2002;224:748–756.[Abstract/Free Full Text]
- Schultz JF, Bell JD, Goldstein RM et al. Hepatic tumor imaging using iron oxide MRI: comparison with computed tomography, clinical impact, and cost analysis. Ann Surg Oncol 1999;6:691–698.[Abstract]
- Lai DT, Fulham M, Stephen MS et al. The role of whole-body positron emission tomography with [18F]fluorodeoxyglucose in identifying operable colorectal cancer metastases to the liver. Arch Surg 1996;13:703–707.
- Desai DC, Zervos EE, Arnold MW et al. Positron emission tomography affects surgical management in recurrent colorectal cancer patients. Ann Surg Oncol 2003;10:59–64.[Abstract/Free Full Text]
- Arguedas MR, Chen VK, Eloubeidi MA et al. Screening for hepatocellular carcinoma in patients with hepatitis C cirrhosis: a cost-utility analysis. Am J Gastroenterol 2003;98:679–690.[CrossRef][Medline]
- Sonnenberg FA, Beck JR. Markov models in medical decision making: a practical guide. Med Decis Making 1993;13:322–338.
- Soresi M, Magliarisi C, Campagna P et al. Usefulness of alpha-fetoprotein in the diagnosis of hepatocellular carcinoma. Anticancer Res 2003;23:1747–1753.[Medline]
- Donckier V, Van Laethem JL, Goldman S et al. [F-18] fluorodeoxyglucose positron emission tomography as a tool for early recognition of incomplete tumor destruction after radiofrequency ablation for liver metastases. J Surg Oncol 2003;84:215–223.[CrossRef][Medline]
- Ohtomo K, Itai Y, Yoshikawa K et al. Hepatocellular carcinoma and cavernous hemangioma: differentiation with MR imaging. Efficacy of T2 values at 0.35 and 1.5 T. Radiology 1988;168:621–623.[Abstract/Free Full Text]
- Quinn SF, Benjamin GG. Hepatic cavernous hemangiomas: simple diagnostic sign with dynamic bolus CT. Radiology 1992;182:545–548.[Abstract/Free Full Text]
- McFarland EG, Mayo-Smith WW, Saini S et al. Hepatic hemangiomas and malignant tumors: improved differentiation with heavily T2-weighted conventional spin-echo MR imaging. Radiology 1994;193:43–47.[Abstract/Free Full Text]
- Middleton ML. Scintigraphic evaluation of hepatic mass lesions: emphasis on hemangioma detection. Semin Nucl Med 1996;26:4–15.[CrossRef][Medline]
- Nagorney DM. Benign hepatic tumors: focal nodular hyperplasia and hepatocellular adenoma. World J Surg 1995;19:13–18.[CrossRef][Medline]
- Zheng WW, Zhou KR, Chen ZW et al. Characterization of focal hepatic lesions with SPIO-enhanced MRI. World J Gastroenterol 2002;8:82–86.[Medline]
- Mathieu D, Bruneton JN, Drouillard J et al. Hepatic adenomas and focal nodular hyperplasia: dynamic CT study. Radiology 1986;160:53–58.[Abstract/Free Full Text]
- Grazioli L, Federle MP, Ichikawa T et al. Liver adenomatosis: clinical, histopathologic, and imaging findings in 15 patients. Radiology 2000;216:395–402.[Abstract/Free Full Text]
- Hooper LD, Mergo PJ, Ros PR. Multiple hepatorenal angiomyelolipomas: diagnosis with fat suppression, gadolinium-enhanced MRI. Abdom Imaging 1994;19:549–551.[CrossRef][Medline]
- Mortele KJ, Ros PR. Cystic focal liver lesions in the adult: differential CT and MR imaging features. Radiographics 2001;21:895–910.[Abstract/Free Full Text]
- Fracanzani AL, Burdick L, Borzio M et al. Contrast-enhanced Doppler ultrasonography in the diagnosis of hepatocellular carcinoma and premalignant lesions in patients with cirrhosis. Hepatology 2001;34:1109–1112.[CrossRef][Medline]
- Szklaruk J, Silverman PM, Charnsangavej C. Imaging in the diagnosis, staging, treatment, and surveillance of hepatocellular carcinoma. AJR Am J Roentgenol 2003;180:441–454.[Free Full Text]
- Oi H, Murakami T, Kim T et al. Dynamic MR imaging and early-phase helical CT for detecting small intrahepatic metastases of hepatocellular carcinoma. AJR Am J Roentgenol 1996;166:369–374.[Abstract/Free Full Text]
- Tomemori T, Yamakado K, Nakatsuka A et al. Fast 3D dynamic MR imaging of the liver with MR SmartPrep: comparison with helical CT in detecting hypervascular hepatocellular carcinoma. Clin Imaging 2001;25:355–361.[CrossRef][Medline]
- Lee JM, Kim IH, Kwak HS et al. Detection of small hypervascular hepatocellular carcinomas in cirrhotic patients: comparison of superparamagnetic iron oxide-enhanced MR imaging with dual-phase spiral CT. Korean J Radiol 2003;4:1–8.[Medline]
- Ahrendt SA, Pitt HA. Biliary tract. In: Townsend C, ed. Sabiston Textbook of Surgery. Philadelphia: W.B. Saunders Company, 2001:1076–1111.
- Bartolozzi C, Donati F, Cioni D et al. Detection of colorectal liver metastases: a prospective multicenter trial comparing unenhanced MRI, MnDPDP-enhanced MRI, and spiral CT. Eur Radiol 2004;14:14–20.[CrossRef][Medline]
- Furuhata T, Okita K, Tsuruma T et al. Efficacy of SPIO-MR imaging in the diagnosis of liver metastases from colorectal carcinomas. Dig Surg 2003;20:321–325.[CrossRef][Medline]
- Tani I, Kurihara Y, Kawaguchi A et al. MR imaging of diffuse liver disease. AJR Am J Roentgenol 2000;174:965–971.[Free Full Text]
- Jager HJ, Mehring U, Gotz GF et al. Radiological features of the visceral and skeletal involvement of hemochromatosis. Eur Radiol 1997;7:1199–1206.[CrossRef][Medline]
- Neuschwander-Tetri BA. Nonalcoholic steatohepatitis: an evolving diagnosis. Can J Gastroenterol 2000;14:321–326.[Medline]
- Saadeh S, Younossi ZM, Remer EM et al. The utility of radiological imaging in nonalcoholic fatty liver disease. Gastroenterology 2002;123:745–750.[CrossRef][Medline]
- Hussain SM, Zondervan PE, Ijzermans JN et al. Benign versus malignant hepatic nodules: MR imaging findings with pathologic correlation. Radiographics 2002;22:1023–1036; discussion 1037–1039.[Abstract/Free Full Text]
- Freeny PC, Grossholz M, Kaakaji K et al. Significance of hyperattenuating and contrast-enhancing hepatic nodules detected in the cirrhotic liver during arterial phase helical CT in pre-liver transplant patients: radiologic-histopathologic correlation of explanted livers. Abdom Imaging 2003;28:333–346.[CrossRef][Medline]
- Ward J, Robinson PJ. How to detect hepatocellular carcinoma in cirrhosis. Eur Radiol 2002;12:2258–2272.[Medline]
- Mori K, Scheidler J, Helmberger T et al. Detection of malignant hepatic lesions before orthotopic liver transplantation: accuracy of ferumoxides-enhanced MR imaging. AJR Am J Roentgenol 2002;179:1045–1051.[Abstract/Free Full Text]
- Ward J, Guthrie JA, Scott DJ et al. Hepatocellular carcinoma in the cirrhotic liver: double-contrast MR imaging for diagnosis. Radiology 2000;216:154–162.[Abstract/Free Full Text]
- Krinsky GA, Lee VS, Theise ND et al. Transplantation for hepatocellular carcinoma and cirrhosis: sensitivity of magnetic resonance imaging. Liver Transpl 2002;8:1156–1164.[CrossRef][Medline]
- Gore RM, Mathieu DG, White EM et al. Passive hepatic congestion: cross-sectional imaging features. AJR Am J Roentgenol 1994;162:71–75.[Abstract/Free Full Text]
Received October 1, 2003;
accepted for publication April 22, 2004.
Top
Learning Objectives
Abstract
