DOI: 10.1148/rg.244035158
RadioGraphics 2004;24:921-935
© RSNA, 2004
Blood Flow Patterns in Focal Liver Lesions at Microbubble-enhanced US1
Margot Brannigan, MD,
Peter N. Burns, PhD and
Stephanie R. Wilson, MD
1 From the Department of Medical Imaging, University of Toronto, Toronto General Hospital, University Health Network, 585 University Ave, Toronto, Ontario, Canada M5G 2N2 (M.B., S.R.W.); and the Department of Medical Biophysics, University of Toronto, Imaging Research, Sunnybrook and Women’s College Health Sciences Centre, Toronto, Ontario, Canada (P.N.B.). Presented as an education exhibit at the 2002 RSNA scientific assembly. Received July 1, 2003; revision requested July 30 and received September 24; accepted October 28. Supported by grants from the Canadian Institutes of Health Research and the Terry Fox Program of the National Cancer Institute of Canada. Clinical studies were also supported by Bristol-Myers Squibb Medical Imaging, Berlex Canada, and Mallinckrodt Medical Inc. Address correspondence to S.R.W. (e-mail: stephanie.wilson@uhn.on.ca).
 |
Abstract
|
|---|
Noninvasive diagnosis of liver lesions is usually performed with contrast material–enhanced computed tomography (CT) and magnetic resonance (MR) imaging and is based on enhancement features of the arterial and portal venous phases. Ultrasonography (US) is often limited in characterizing liver lesions because color and spectral Doppler US provide limited vascular information in large patients and in small or deep lesions. However, microbubble contrast agents, together with specialized US techniques, now allow diagnosis of liver lesions based on morphologic evaluation of lesion vascularity and visualization of specific enhancement features. Microbubble contrast agents are purely intravascular, easy to administer, and well tolerated and allow sensitive real-time evaluation of blood flow in hepatic lesions. During the portal venous phase, benign lesions (eg, hemangioma, focal nodular hyperplasia) typically enhance more than the liver, whereas malignant lesions (eg, hepatocellular carcinoma, metastases) enhance less. Microbubble-enhanced US allows characterization of very small lesions that may not be accurately characterized with CT or MR imaging. Findings from initial studies suggest that microbubble-enhanced US of the liver provides enhancement information comparable to that provided by contrast-enhanced CT and MR imaging, along with real-time morphologic evaluation of lesion vascularity.
© RSNA, 2004
Index Terms: Liver, focal nodular hyperplasia, 761.3198 • Liver neoplasms, 761.3194, 761.3198, 761.323, 761.33 • Liver neoplasms, diagnosis, 761.12988 • Liver neoplasms, US, 761.12988 • Ultrasound (US), contrast media • Ultrasound (US), technology
|
Introduction
|
|---|
At present, diagnosis of focal liver lesions with noninvasive imaging relies on well-known vascular enhancement patterns, usually identified at computed tomography (CT) or magnetic resonance (MR) imaging. Although ultrasonography (US) often plays an important role in the detection of masses, this modality offers no such comparable ability to characterize a lesion on the basis of its blood flow. Color and spectral Doppler US are sometimes diagnostic but are more often nonspecific, principally because of their inability to demonstrate blood flow at the parenchymal level.
Now, however, a new generation of microbubble contrast agents for use with US, combined with novel imaging methods optimized for bubble detection, has transformed the ability of US to characterize a liver lesion (Fig 1) (2). The contrast agents are easy to use and are well tolerated by patients. The imaging methods operate in real time, allowing a complete diagnosis to be made more quickly and at a lower cost than with CT or MR imaging.
View larger version (157K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Focal liver mass: characterization with microbubble contrast material. (a) Baseline conventional US image obtained with a high mechanical index (MI) of 1.2 (the MI is the ratio of the peak rarefactional pressure to the square root of the frequency and is related to the tendency of the sound to induce bubble motion [1]) shows a large, highly echogenic focal mass with a hypoechoic rim. The echogenicity of the lesion is related to its tissue properties. (b) Pulse inversion US image (MI = 0.1) shows suppression of the tissue echoes. The entire image now appears black. (c) On an arterial phase US image obtained moments after a small (0.1-mL) bolus injection of a suspension of microbubbles into a peripheral vein, the lesion enhances far more than the adjacent liver, a finding that is consistent with a hypervascular mass. The microbubbles within the vasculature now account for the echogenicity of the lesion. (d) Portal venous phase US image shows enhancement of the liver parenchyma. The lesion is less echogenic than the liver (ie, has "washed out"), a finding that is consistent with a hypervascular malignancy. Hepatocellular carcinoma (HCC) was confirmed at biopsy.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Focal liver mass: characterization with microbubble contrast material. (a) Baseline conventional US image obtained with a high mechanical index (MI) of 1.2 (the MI is the ratio of the peak rarefactional pressure to the square root of the frequency and is related to the tendency of the sound to induce bubble motion [1]) shows a large, highly echogenic focal mass with a hypoechoic rim. The echogenicity of the lesion is related to its tissue properties. (b) Pulse inversion US image (MI = 0.1) shows suppression of the tissue echoes. The entire image now appears black. (c) On an arterial phase US image obtained moments after a small (0.1-mL) bolus injection of a suspension of microbubbles into a peripheral vein, the lesion enhances far more than the adjacent liver, a finding that is consistent with a hypervascular mass. The microbubbles within the vasculature now account for the echogenicity of the lesion. (d) Portal venous phase US image shows enhancement of the liver parenchyma. The lesion is less echogenic than the liver (ie, has "washed out"), a finding that is consistent with a hypervascular malignancy. Hepatocellular carcinoma (HCC) was confirmed at biopsy.
|
|
View larger version (173K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c. Focal liver mass: characterization with microbubble contrast material. (a) Baseline conventional US image obtained with a high mechanical index (MI) of 1.2 (the MI is the ratio of the peak rarefactional pressure to the square root of the frequency and is related to the tendency of the sound to induce bubble motion [1]) shows a large, highly echogenic focal mass with a hypoechoic rim. The echogenicity of the lesion is related to its tissue properties. (b) Pulse inversion US image (MI = 0.1) shows suppression of the tissue echoes. The entire image now appears black. (c) On an arterial phase US image obtained moments after a small (0.1-mL) bolus injection of a suspension of microbubbles into a peripheral vein, the lesion enhances far more than the adjacent liver, a finding that is consistent with a hypervascular mass. The microbubbles within the vasculature now account for the echogenicity of the lesion. (d) Portal venous phase US image shows enhancement of the liver parenchyma. The lesion is less echogenic than the liver (ie, has "washed out"), a finding that is consistent with a hypervascular malignancy. Hepatocellular carcinoma (HCC) was confirmed at biopsy.
|
|
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1d. Focal liver mass: characterization with microbubble contrast material. (a) Baseline conventional US image obtained with a high mechanical index (MI) of 1.2 (the MI is the ratio of the peak rarefactional pressure to the square root of the frequency and is related to the tendency of the sound to induce bubble motion [1]) shows a large, highly echogenic focal mass with a hypoechoic rim. The echogenicity of the lesion is related to its tissue properties. (b) Pulse inversion US image (MI = 0.1) shows suppression of the tissue echoes. The entire image now appears black. (c) On an arterial phase US image obtained moments after a small (0.1-mL) bolus injection of a suspension of microbubbles into a peripheral vein, the lesion enhances far more than the adjacent liver, a finding that is consistent with a hypervascular mass. The microbubbles within the vasculature now account for the echogenicity of the lesion. (d) Portal venous phase US image shows enhancement of the liver parenchyma. The lesion is less echogenic than the liver (ie, has "washed out"), a finding that is consistent with a hypervascular malignancy. Hepatocellular carcinoma (HCC) was confirmed at biopsy.
|
|
In this article, we review the contrast agents and the technology required for their use in liver imaging. We also describe our standard imaging techniques and the enhancement patterns seen at microbubble-enhanced US of commonly encountered focal liver lesions, including HCC, metastasis, hemangioma, focal nodular hyperplasia (FNH), and hepatic adenoma. In addition, we discuss and illustrate some similarities between contrast material–enhanced US and contrast-enhanced CT and MR imaging.
|
US Contrast Agents
|
|---|
US contrast agents consist of microbubbles of air or perfluorocarbon gas stabilized by a protein, lipid, or polymer shell (3). The bubbles are sufficiently small and stable to traverse the pulmonary and cardiac circulations following peripheral venous injection. The bubbles disappear as the gas diffuses through the thin shell, with a typical half-life of a few minutes in blood. A typical contrast material dose for an adult patient consists of 0.2–2 mL of a suspension of bubbles in saline solution. This dose is manually injected into an arm vein, followed by a flush of 5 mL of saline solution. The volume is so small that a bolus can be given without the need for a power injector. In our experience with more than 2,000 injections, patient acceptance has been very high, with no adverse events seen at our institution. The bubbles are approximately the same size as red blood cells and cannot move through the vascular endothelium into the interstitium, even after an extended period of time (1); therefore, they are true blood pool agents. Microbubble contrast agents are approved for radiologic use in many countries, including the European Union, Canada, and Japan. Although US contrast material has been approved for clinical use for cardiac diagnosis in the United States for a number of years, its use for radiologic indications is currently under investigation.
|
Bubble-Specific Imaging: State of the Art
|
|---|
Because parenchymal vessels lie below the resolution limit of radiologic images, contrast-enhanced imaging must allow differentiation between the signal from vessels and that due to solid tissue within the resolution volume. Angiography allows such differentiation by means of subtraction of a precontrast image; US does so by helping distinguish unique characteristics of the echo from the contrast agent itself.
Bubbles respond to the sound emitted by the imaging transducer by changing their size in sympathy with the US pressure wave, which results in their radial oscillation at the US frequency (Fig 2). Like all oscillating systems, bubbles have a natural or resonant frequency, at which their response is greatly enhanced. It happens that bubbles of the size range required for transpulmonary passage following intravenous injection (diameter of 3–5 µm) resonate at diagnostic US frequencies (5). It is this phenomenon that makes contrast-specific imaging possible. Resonant oscillators undergo high-amplitude excursions even when stimulated by a weak driving force, as exemplified by a child on a swing who, as long as he or she is pushed at the correct frequency, does not require a great deal of force to stay in motion. In a similar way, resonant bubbles return detectable echoes even when present in small numbers in the imaged tissue. Furthermore, if the driving force of the incident US is made somewhat stronger, the bubble oscillators become nonlinear in their motion. A bubble’s expansion in the rarefaction phase of the acoustic cycle exceeds its compression in the pressure phase (a consequence of the bubble’s increased stiffness as the gas within it is compressed) (Fig 2). The resulting nonlinear echo contains overtones, or harmonics, of the driving frequency. Harmonics have frequencies that are multiples of the driving frequency and are exploited with a number of imaging methods to distinguish the bubble echo from blood vessels while rejecting the linear echo from the tissue that surrounds the vessels. The simplest of these methods is called harmonic imaging, in which a conventional US imager sends pulses into tissue at one frequency but selectively detects echoes at double that frequency (ie, the second harmonic). Although harmonic imaging was the first bubble-specific imaging method (6), it is now used only rarely for radiologic imaging, principally because of poor image quality (due to reduced bandwidth) and poor solid tissue suppression (due to the propagation nonlinearity of tissue). Instead, a number of methods aid in direct evaluation of the asymmetry of the returning bubble echo by sending a series of pulses into tissue. The most common strategy is based on the principle of pulse inversion (Fig 3) (4,7). Two US pulses are sent sequentially from the transducer into tissue, the second of whose phase is inverted with respect to that of the first. The imager simply adds together the two echoes that return from tissue. Figure 3 shows how echoes from solid tissue cancel each other, whereas those from bubbles combine to yield an echo at double the frequency. In this way, the echo from tissue is suppressed and that from the microbubble contrast material is enhanced. Pulse inversion imaging goes under a number of names with modern imagers, including phase inversion imaging and coherent contrast imaging.
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Graph illustrates microbubble behavior in an acoustic field. Bubbles respond asymmetrically to high-intensity sound waves, stiffening when compressed and yielding when expanded, a nonlinear response that produces harmonics in the scattered wave. (Reprinted, with permission, from reference 4.)
|
|
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Graphs illustrate the principle of pulse inversion. A pulse of sound is transmitted into the body, and echoes are received from the contrast agent and tissue. A second pulse, an inverted version of the first, is then transmitted in the same direction, and the two echoes are summed. Linear echoes from tissue cancel each other. Nonlinear components of the microbubble echoes are reinforced when summed, producing a strong harmonic signal. (Reprinted, with permission, from reference 4.)
|
|
The practical aspects of contrast-enhanced US differ from those of other modalities because the contrast agent interacts with the imaging process itself. The principal determinant of this interaction is the amplitude of the pulses transmitted into tissue, displayed on most US systems as the MI. At a low MI (less than about 0.2, depending on the agent used), bubbles are stimulated into harmonic motion and can be imaged in real time, thereby showing vessel morphology and the evolution of the vascular volume over time. At a higher MI (greater than about 0.3, again depending on the agent), the oscillation of the bubbles becomes so pronounced that the bubbles can be disrupted by the sound beam used to image them. As they are disrupted, they emit a strong, highly nonlinear echo that is relatively easy to detect. The resulting image reveals the distribution of bubbles at the perfusion level of the circulation, creating a single but highly sensitive "vascular volume" image. Because the parenchymal vessels within the scan plane take many seconds to refill with fresh bubbles, this "disruption" method of imaging does not allow real-time frame rates, yet it is the most sensitive means of evaluating the vascular volume in tissue. A combination of these two methods has been used to image capillary flow in myocardium (1) as well as in the liver (8) and other organs (9) and forms the basis for the examinations discussed in this article.
|
Materials and Methods
|
|---|
Over a 5-year period, we performed contrast-enhanced US in 400 patients to characterize focal liver masses. Initial scans in 45 patients were obtained with harmonic imaging. All subsequent scans were obtained with pulse inversion US. In most cases, we used an ATL HDI 5000 scanner (Philips, Bothell, Wash), with an Acuson Sequoia (Siemens, Mountain View, Calif) or Toshiba Aplio (Toshiba, Santa Ana, Calif) scanner used in some cases. The following microbubble contrast agents, approved for clinical use in Canada, were used: Levovist (galactose–palmitic acid; Berlex Canada, Montreal, Quebec, Canada) (n = 80), Optison (albumin microspheres sonicated preparation; Nycomed/Amersham, Little Chalfont, England) (n = 45), and Definity (perflutren lipid microspheres; Bristol-Myers Squibb, Billerica, Mass) (n = 275).
The perfluorocarbon contrast agents (Optison and Definity) are preferable for liver lesion characterization because they allow excellent nonlinear response to insonation at a low MI, thereby permitting continuous real-time evaluation of lesion vascularity and enhancement. On the other hand, galactose–palmitic acid (Levovist) cannot be imaged in real time and relies on a postvascular phase, which makes this contrast agent less valuable for liver lesion characterization (10).
Informed consent was obtained for off-label use of the contrast agents. Methods used to confirm the diagnoses differed as our experience evolved and included contrast-enhanced triphasic CT or MR imaging, sulfur colloid and radiolabeled red blood cell scintigraphy, biopsy, and clinical follow-up.
Unless otherwise noted, the US contrast agent used in the cases illustrated in this article consisted of perflutren lipid microspheres, which were administered in multiple 0.1–0.2-mL bolus aliquots up to a maximum dose of 10 µL/kg via a peripheral venous route. Typically, the contrast agent is injected manually through a three-way stopcock, followed by a 5-mL saline solution flush. A 22-gauge (or larger) "angiocath" is preferred because the bubbles are susceptible to disruption by the pressure drop caused by smaller needles. In all patients, low-MI continuous imaging was performed during the arterial and portal venous phases. In cases in which better visualization of the contrast agent was needed (eg, deep in a large patient), an additional bolus was administered and high-MI "interval delay" or "flash" imaging was performed (3). With this method, after the appearance of contrast agent in the hepatic arteries, the scanner is "frozen" for 5–10 seconds while the agent fills the liver, then is unfrozen for a few seconds. The exposure of the bubbles to US results in their disruption within a few frames, which are then reviewed on the stored cine loop (10). Subsequent delay intervals are then determined based on the rapidity of the filling of the lesion vessels and might be altered as the examination progresses. It should be noted that even high-MI imaging is performed with US exposure settings well below those used for routine US.
Arterial phase imaging is performed continuously from the time when bubbles wash into the hepatic arteries until they are seen in the main portal vein, a period that varies between about 2 and 10 seconds. The presence of lesion vessels as well as their distribution and morphology are documented. The morphology of feeding and intralesion vessels can be useful in diagnosing liver masses (Fig 4). Hypervascular masses such as FNH and HCC show linear lesion vascularity. A stellate vascular pattern is frequently seen in FNH (Fig 4a), whereas dysmorphic vascularity is more often seen in HCC (Fig 4b). Hemangiomas, by comparison, infrequently show linear lesion vascularity, instead demonstrating peripheral puddles or pools of contrast material (Fig 4c).
View larger version (191K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Discriminatory features of vessel morphology. (a) Arterial phase image obtained with galactose-palmitic acid shows FNH with classic stellate vascularity. (Reprinted, with permission, from reference 10.) (b, c) Arterial phase images show HCC with heterogeneous dysmorphic vessels (b) and hemangioma with peripheral puddles and pools of contrast material (c). All three images were obtained with continuous low-MI pulse inversion; none shows linear lesion vessels.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Discriminatory features of vessel morphology. (a) Arterial phase image obtained with galactose-palmitic acid shows FNH with classic stellate vascularity. (Reprinted, with permission, from reference 10.) (b, c) Arterial phase images show HCC with heterogeneous dysmorphic vessels (b) and hemangioma with peripheral puddles and pools of contrast material (c). All three images were obtained with continuous low-MI pulse inversion; none shows linear lesion vessels.
|
|
View larger version (197K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4c. Discriminatory features of vessel morphology. (a) Arterial phase image obtained with galactose-palmitic acid shows FNH with classic stellate vascularity. (Reprinted, with permission, from reference 10.) (b, c) Arterial phase images show HCC with heterogeneous dysmorphic vessels (b) and hemangioma with peripheral puddles and pools of contrast material (c). All three images were obtained with continuous low-MI pulse inversion; none shows linear lesion vessels.
|
|
Arterial phase lesion enhancement is assessed by comparing the echogenicity of the lesion with that of the adjacent liver parenchyma at the peak of the arterial phase. Both the degree and pattern of enhancement are noted (Fig 5). Because low-MI pulse inversion imaging suppresses the normal tissue echoes, the preinjection image appears dark (Fig 1b), or even black, at initiation of the scanning sequence. All echogenicity is then attributed to the presence of the microbubbles. Therefore, lesions with high vascularity, such as FNH and HCC, manifest as regions of increased echogenicity compared with the adjacent liver, indicating that the number of microbubbles within their vascular pool is greater than that in the liver parenchyma during the arterial phase (Figs 1, 5a, 5b). Lesions with low vascularity, including many metastases, manifest as a dark or black area within the enhanced liver during the arterial phase (Fig 5c, 5d).
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5a. Arterial phase enhancement. Assessments are made by comparing the echogenicity of the lesion with that of the adjacent liver during the arterial phase. (a, b) Baseline conventional US image (a) and corresponding arterial phase US image (b) demonstrate FNH as a hypervascular mass. (c, d) Baseline conventional US image (c) and corresponding arterial phase US image (d) obtained in a different patient show hypovascular masses from metastatic colon cancer. In both cases, the lesions are very subtle on the baseline images, whereas the contrast-enhanced images show the lesions clearly.
|
|
View larger version (126K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5b. Arterial phase enhancement. Assessments are made by comparing the echogenicity of the lesion with that of the adjacent liver during the arterial phase. (a, b) Baseline conventional US image (a) and corresponding arterial phase US image (b) demonstrate FNH as a hypervascular mass. (c, d) Baseline conventional US image (c) and corresponding arterial phase US image (d) obtained in a different patient show hypovascular masses from metastatic colon cancer. In both cases, the lesions are very subtle on the baseline images, whereas the contrast-enhanced images show the lesions clearly.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5c. Arterial phase enhancement. Assessments are made by comparing the echogenicity of the lesion with that of the adjacent liver during the arterial phase. (a, b) Baseline conventional US image (a) and corresponding arterial phase US image (b) demonstrate FNH as a hypervascular mass. (c, d) Baseline conventional US image (c) and corresponding arterial phase US image (d) obtained in a different patient show hypovascular masses from metastatic colon cancer. In both cases, the lesions are very subtle on the baseline images, whereas the contrast-enhanced images show the lesions clearly.
|
|
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5d. Arterial phase enhancement. Assessments are made by comparing the echogenicity of the lesion with that of the adjacent liver during the arterial phase. (a, b) Baseline conventional US image (a) and corresponding arterial phase US image (b) demonstrate FNH as a hypervascular mass. (c, d) Baseline conventional US image (c) and corresponding arterial phase US image (d) obtained in a different patient show hypovascular masses from metastatic colon cancer. In both cases, the lesions are very subtle on the baseline images, whereas the contrast-enhanced images show the lesions clearly.
|
|
Portal venous phase lesion enhancement is assessed by comparing the echogenicity of the lesion with that of the adjacent liver parenchyma between 45 and 75 seconds following the appearance of the contrast agent in the field of view (Fig 6). We define positive enhancement as echogenicity equal to or greater than that of the enhanced liver and negative enhancement as echogenicity less than that of the liver. Because the liver is supplied predominantly by the portal vein, the liver parenchyma shows progressively increased enhancement through the arterial phase and into the portal venous phase. Malignant liver tumors, whether primary or secondary, derive the majority of their blood supply from the hepatic artery. Therefore, during the portal venous phase they generally appear hypoechoic or black, in contrast with the enhanced adjacent liver parenchyma (Figs 1, 6a–6c).
View larger version (149K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (169K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6c. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6d. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6e. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6f. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (167K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6g. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6h. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6i. Portal venous phase enhancement. (a-c) Baseline conventional US image (a), arterial phase image (b), and portal venous phase image (c) show HCC as a hypervascular mass with washout during the portal venous phase. (d-f) Baseline conventional US image (d), arterial phase image (e), and portal venous phase image (f) obtained in a different patient show FNH as a hypervascular mass, with "sustained enhancement" during the portal venous phase. (g-i) Baseline conventional US image (g), arterial phase image (h), and portal venous phase image (i) obtained in a third patient show hemangioma with peripheral nodular enhancement during the arterial phase and sustained enhancement during the portal venous phase.
|
|
Observations regarding the progressive changes in enhancement that occur from the arterial phase into the portal venous phase and beyond contribute to liver lesion diagnoses (Fig 6). Washout refers to a situation in which a lesion shows arterial phase enhancement greater than that of the liver and portal venous phase enhancement less than that of the liver. This phenomenon is typical of malignant lesions, including HCC and hypervascular metastases (Fig 6a–6c). Sustained enhancement refers to a situation in which a lesion demonstrates greater enhancement than the liver during the arterial phase and continues to show enhancement equal to or greater than that of the liver during the portal venous phase. This phenomenon is typical of benign lesions (Fig 6d–6i).
|
Focal Liver Lesions
|
|---|
In the following sections, we present our perspective on the impact of microbubble-enhanced US in the evaluation of the four most commonly encountered focal liver lesions: HCC, metastasis, hemangioma, and FNH. Specific enhancement features of each liver lesion are illustrated in the remaining figures and summarized in the Table.
Hepatocellular Carcinoma
HCC is the most frequently encountered primary malignant tumor of the liver. It frequently complicates chronic liver disease, most often secondary to viral hepatitis B or C. HCC is of worldwide significance, and its early detection provides the only hope for successful treatment, especially in populations where viral infections are endemic. Imaging is often complicated by the underlying chronic liver disease.
Noninvasive imaging diagnosis of HCC is most often based on the CT detection of a hypervascular mass in the liver of a patient at high risk for this diagnosis. Washout of the contrast agent during portal venous phase CT has also been described, although a well-recognized interstitial phase, which occurs when contrast agent passes through the vascular endothelium, may alter this washout.
Although many malignant liver tumors are readily seen at US, surveillance US can be difficult in patients with chronically diseased livers who are at risk for HCC. In addition, nonenhanced US has low specificity because a variety of nonmalignant lesions such as hemangioma, FNH, and regenerative nodules may appear similar to HCC. In many cases, CT or MR imaging is required for further characterization.
The addition of microbubble contrast agents has greatly improved the ability to diagnose HCC with US. Hypervascularity with frequent vessel dysmorphology and washout during the portal venous phase are characteristic (Fig 7). Tumor necrosis is common and manifests as nonenhancing regions within the enhanced tumor during the arterial phase. Our experience with US diagnosis of HCC includes more than 100 lesions as small as 2 cm in diameter (Fig 8). Consistent results have been achieved in patients who present with an obvious liver mass; the observations shown in the Table are comparable to those made at CT and MR imaging.
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. HCC. (a) Arterial phase US image shows a lesion whose echogenicity exceeds that of the liver, a finding that is consistent with a hypervascular mass. There is a central nonenhancing area that suggests scar or necrosis. (b) On a portal venous phase US image, the lesion is washed out, appearing less echogenic than the adjacent liver. (c, d) Arterial phase (c) and portal venous phase (d) CT scans show the hypervascular mass with necrosis and with washout on the portal venous phase image.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. HCC. (a) Arterial phase US image shows a lesion whose echogenicity exceeds that of the liver, a finding that is consistent with a hypervascular mass. There is a central nonenhancing area that suggests scar or necrosis. (b) On a portal venous phase US image, the lesion is washed out, appearing less echogenic than the adjacent liver. (c, d) Arterial phase (c) and portal venous phase (d) CT scans show the hypervascular mass with necrosis and with washout on the portal venous phase image.
|
|
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7c. HCC. (a) Arterial phase US image shows a lesion whose echogenicity exceeds that of the liver, a finding that is consistent with a hypervascular mass. There is a central nonenhancing area that suggests scar or necrosis. (b) On a portal venous phase US image, the lesion is washed out, appearing less echogenic than the adjacent liver. (c, d) Arterial phase (c) and portal venous phase (d) CT scans show the hypervascular mass with necrosis and with washout on the portal venous phase image.
|
|
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7d. HCC. (a) Arterial phase US image shows a lesion whose echogenicity exceeds that of the liver, a finding that is consistent with a hypervascular mass. There is a central nonenhancing area that suggests scar or necrosis. (b) On a portal venous phase US image, the lesion is washed out, appearing less echogenic than the adjacent liver. (c, d) Arterial phase (c) and portal venous phase (d) CT scans show the hypervascular mass with necrosis and with washout on the portal venous phase image.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Small HCC. A baseline conventional US image (not shown) revealed a small, subtle, hypoechoic mass. (a) Arterial phase US image shows a hypervascular lesion with avid enhancement. (b) On a portal venous phase US image, the lesion is hypoechoic relative to the enhanced background liver. This washout is strongly suggestive of the malignant nature of the mass.
|
|
View larger version (163K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Small HCC. A baseline conventional US image (not shown) revealed a small, subtle, hypoechoic mass. (a) Arterial phase US image shows a hypervascular lesion with avid enhancement. (b) On a portal venous phase US image, the lesion is hypoechoic relative to the enhanced background liver. This washout is strongly suggestive of the malignant nature of the mass.
|
|
To our knowledge, however, no prospective studies showing the ability to screen populations at risk for this malignant tumor with contrast-enhanced US have been published. Anticipated difficulties include problems with adequately scanning the entire cirrhotic liver because of its small size, difficulty in differentiating regenerative nodularity from malignant tumors, and problems related to the alteration of liver blood flow in cirrhosis, including increased relative hepatic arterial supply.
Metastasis
Metastatic disease is a common indication for imaging the liver, and lesion detection and characterization are both important considerations. Multiple solid liver masses in a patient with a known primary tumor pose little problem in diagnosis. Nonetheless, characterization may be an issue, especially in a patient who presents with liver masses without a known primary tumor or in a patient with a solitary tumor of uncertain origin.
Contrast-enhanced US shows lesion blood flow in metastases as a reflection of the vascularity of the primary tumor. Therefore, considerable variation is seen during the arterial phase of enhancement. Findings include hypovascularity (Fig 5c, 5d) and rim enhancement (Fig 9), both typical of metastatic lesions, including those from the breast, lung, and colon. Tumor enhancement may be inhomogeneous because of necrosis. Hypervascular primary tumors, including those from the kidney and thyroid gland and those of neuroendocrine origin, show hypervascular metastases with enhancement features that overlap with those of HCC (Fig 10). Regardless of their appearance during the arterial phase, however, metastases have consistently shown less enhancement than the liver during the portal venous phase of contrast-enhanced US.
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9c. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (147K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9d. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (120K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9e. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (148K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9f. Metastasis from primary carcinoma of the anal canal with rim enhancement. (a, b) Baseline US (a) and CT (b) images show an expansive mass in the left lobe of the liver. (c, d) Arterial phase US (c) and CT (d) images demonstrate the mass with rim enhancement. (e) On a portal venous phase US image, the lesion is completely washed out and appears black. (f) Portal venous phase CT scan shows continued rim enhancement of the tumor, a finding that probably represents interstitial contrast material.
|
|
View larger version (137K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Hypervascular metastases from carcinoid tumor. (a) Baseline US image shows the liver with high fat content and hypoechoic lesions. (b) On an arterial phase US image, the lesions enhance avidly, appearing more echogenic than the adjacent liver. (c) On a portal venous phase US image, the liver remains brightly enhanced, and the lesions have washed out and appear hypoechoic. The patient was allergic to iodinated contrast material. (Fig 10 reprinted, with permission, from reference 8.)
|
|
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Hypervascular metastases from carcinoid tumor. (a) Baseline US image shows the liver with high fat content and hypoechoic lesions. (b) On an arterial phase US image, the lesions enhance avidly, appearing more echogenic than the adjacent liver. (c) On a portal venous phase US image, the liver remains brightly enhanced, and the lesions have washed out and appear hypoechoic. The patient was allergic to iodinated contrast material. (Fig 10 reprinted, with permission, from reference 8.)
|
|
View larger version (146K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10c. Hypervascular metastases from carcinoid tumor. (a) Baseline US image shows the liver with high fat content and hypoechoic lesions. (b) On an arterial phase US image, the lesions enhance avidly, appearing more echogenic than the adjacent liver. (c) On a portal venous phase US image, the liver remains brightly enhanced, and the lesions have washed out and appear hypoechoic. The patient was allergic to iodinated contrast material. (Fig 10 reprinted, with permission, from reference 8.)
|
|
Hemangioma
Hemangioma is the most frequently encountered benign liver tumor and is reported to occur in 5%–20% of the population. The US features of both typical and atypical hemangiomas are well recognized, and in an asymptomatic patient without risk factors for malignancy, no further imaging is recommended. However, large lesions, those with uncertain imaging features, and those detected in patients at risk for any type of malignancy require noninvasive imaging, which most often consists of either radiolabeled red blood cell scintigraphy or of CT or MR imaging.
Hemangiomas demonstrate a reproducible and apparently specific pattern of enhancement at contrast-enhanced US (11). At our institution, we now frequently perform contrast-enhanced US to confirm the nature of a hemangioma that is incidentally detected at nonenhanced US. Lesions rarely show linear vessels; instead, they show peripheral puddles and pools of enhancement that expand in a centripetal pattern during the portal venous phase and beyond, often progressing to complete fill-in of the lesion (Fig 11). The real-time nature of US allows this phenomenon to be appreciated even in small, rapidly perfusing lesions (Fig 12). A "bridge" of enhancement with similar puddling and pooling of contrast agent may be seen traversing the lesion (Fig 13). Although the speed of enhancement may differ, the degree of enhancement of the peripheral nodules seen during the arterial phase exceeds that of the adjacent liver (Figs 11–13). Sustained enhancement, in which the lesion has an echogenicity equal to or greater than that of the liver through the portal venous phase and beyond (Fig 11), is requisite to confident diagnosis. Complete enhancement does not always occur, especially in large lesions, which often undergo central thrombosis with scarring.
View larger version (144K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Hemangioma. (a) Baseline conventional US image shows a heterogeneous, bulbous, slightly exophytic lesion that extends from the posterior right lobe. (b) Early arterial phase vascular US image shows no linear vessels. There is peripheral nodular enhancement. (c) Late arterial phase US image shows centripetal progression of the enhancement. (d) On a portal venous phase US image, the lesion appears uniformly enhanced and brighter than the background liver. This sustained enhancement is consistent with a benign condition.
|
|
View larger version (140K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Hemangioma. (a) Baseline conventional US image shows a heterogeneous, bulbous, slightly exophytic lesion that extends from the posterior right lobe. (b) Early arterial phase vascular US image shows no linear vessels. There is peripheral nodular enhancement. (c) Late arterial phase US image shows centripetal progression of the enhancement. (d) On a portal venous phase US image, the lesion appears uniformly enhanced and brighter than the background liver. This sustained enhancement is consistent with a benign condition.
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11c. Hemangioma. (a) Baseline conventional US image shows a heterogeneous, bulbous, slightly exophytic lesion that extends from the posterior right lobe. (b) Early arterial phase vascular US image shows no linear vessels. There is peripheral nodular enhancement. (c) Late arterial phase US image shows centripetal progression of the enhancement. (d) On a portal venous phase US image, the lesion appears uniformly enhanced and brighter than the background liver. This sustained enhancement is consistent with a benign condition.
|
|
Top
Abstract
Introduction
