Liver Mass Evaluation with Ultrasound: The Impact of Microbubble Contrast Agents and Pulse Inversion Imaging

• ABSTRACT
• MICROBUBBLE CONTRAST AGENTS: WHERE ARE WE TODAY?
• THE BEHAVIOR OF MICROBUBBLE CONTRAST AGENTS IN AN ULTRASOUND FIELD
• Specialized Imaging Techniques
• THE SONOGRAPHIC EXAMINATION
• Lesional Vascularity Assessment
• Vascular Volume Assessments
• LESION-SPECIFIC ANALYSIS
• Hepatocellular Carcinoma
• Metastastatic Disease
• Hemangiomas
• Focal Nodular Hyperplasia
• Summary Regarding Liver Lesion Characterization
• POSTVASCULAR LIVER-SPECIFIC IMAGING WITH LEVOVIST
• LIVER MASS DETECTION WITH POSTVASCULAR LEVOVIST SCANNING
• LIVER MASS CHARACTERIZATION WITH POSTVASCULAR LEVOVIST SCANNING
• SUMMARY
• ABBREVIATIONS
• REFERENCES

ABSTRACT

Liver mass evaluation includes two essential elements-lesion detection and lesion characterization. Both of these are greatly improved on sonography with the addition of contrast agents and the use of specialized imaging techniques, particularly pulse inversion imaging. Ultrasound contrast agents are comprised of tiny microbubbles of gas that interact with the ultrasound beam producing an enhancement of the Doppler signal from blood. Pulse inversion imaging allows preferential detection of the signal from the microbubble agents with suppresion of the signal from background tissue. Two imaging techniques include a low mechanical index (MI) nondestructive method to show lesional vascularity and a high MI destructive mode that produces disruption of the bubbles in a single frame. The latter allows for quantitative assessment of the relative enhancement of a lesion as compared with the adjacent liver parenchyma, which is a reflection of the relative vascular volumes. Vascular imaging has shown characteristic and reproducible features of common liver masses, including hemangioma, focal nodular hyperplasia, hepatocellular carcinoma, and liver metastases. Delayed postvascular enhancement of the normal liver, a phenomenon that is unique to certain classes of microbubble contrast agents, allows detection of more and smaller malignant lesions than on baseline.

Evaluation of a focal liver mass is a complex issue that is often the major focus of a cross-sectional imaging study. There are two basic questions that may be posed. The first deals with the characterization of a known liver lesion and answers the question-What is it? The second issue is that of detection and answers the question-Is it there? The resolution of either problem requires a focused examination that is often adjusted according to the clinical situation or on the basis of information already known from previous imaging tests.

Characterization of a liver mass on sonography is based on the appearance of the mass on gray scale imaging as well as on vascular information that may be obtained on a conventional Doppler examination. Conventional Doppler, however, often fails in the evaluation of a focal liver mass, particularly in a large patient or on a small or deep liver lesion, or on one with weak Doppler signals. Motion artifact is also highly problematic for abdominal Doppler studies, and a left lobe liver mass, close to the pulsation of the cardiac apex, is virtually always a failure for conventional Doppler. To remedy the problem of failed Doppler examination of a focal liver lesion, there are two basic approaches to improve the study. The first remedy is to inject a microbubble contrast agent that enhances the Doppler signal from blood. The second remedy is to utilize specialized imaging techniques that allow preferential detection of the signal from the ultrasound contrast agent with suppression of the signal from background tissue.

MICROBUBBLE CONTRAST AGENTS: WHERE ARE WE TODAY?

The last decade has been an exciting time for ultrasound with the introduction of microbubble contrast agents into clinical practice. Paralleling the development and investigation of these agents, there have been equally exciting advancements on the technical side as the major manufacturers have focused research and development efforts on specialized imaging techniques to replace conventional Doppler for contrast agent studies.

Contrast agents for ultrasound are blood pool agents that do not diffuse through the vascular endothelium. They comprise tiny bubbles of gas contained within a stabilizing shell. There are two types of gas commonly used in contrast agents available today: air and perfluorocarbons. Our experience is based on the use of Levovist℗(Schering Ag, Berlin, Germany), an air microbubble agent with a lipid shell, and two perfluorocarbon agents, Definity (DuPont Inc., Billerica, MA) and Optison (Mallinckrodt, St Louis, MO). Levovist is the most widely used contrast agent worldwide. It is currently available for use in more than 60 countries, though not in the United States. Several perfluorocarbon agents are in various stages of clinical trials. These include Optison, Definity, Sonovue (Bracco spa, Milan, Italy), and Sonazoid (Nycomed Amersham, Oslo, Norway). It is anticipated that these perfluorocarbon agents will be available for clinical use in the near future.

Microbubble contrast agents are extremely well tolerated by patients. To date more than 100,000 injections have been given, with no significant adverse events recorded. The doses administered are much smaller than for the iodinated agents used in the typical computed tomography (CT) exam, with whole body adult doses varying between 0.5 and 10 mL, depending on the agent. Despite their evident lack of toxicity, much work remains to be done in understanding the biological action of an intravascular microbubble when excited by an external acoustic field such as that used in ultrasound imaging.

Cardiac applications for microbubble contrast agents were the first to receive widespread acceptance, and contrast agents are now available in many countries for ventricular opacification, defining the endocardial outline, and documenting myocardial perfusion.[1] Levovist was first approved for radiology applications as a Doppler enhancing agent for salvaging otherwise unsuccessful Doppler examinations. Current investigations suggest that the potential uses for these contrast agents in medical imaging extend far beyond these initial indications. Liver mass evaluation, the focus of this publication, is now well described, and recent publications suggest great potential for this technique.

THE BEHAVIOR OF MICROBUBBLE CONTRAST AGENTS IN AN ULTRASOUND FIELD

Microbubble contrast agents for ultrasound are unique in that they interact with the imaging process. The major determinant of this interaction is the peak negative pressure of the transmitted ultrasound pulse, reflected by the mechanical index (MI), a number that is displayed on the ultrasound machine. The bubbles show stable nonlinear oscillation when exposed to an ultrasound field with a low MI. As the ultrasound wave, which comprises alternate compressions and rarefactions, propagates over the bubbles, they experience a periodic change in their radius (Fig. [1]). As a bubble is compressed by the ultrasound pressure wave, it becomes stiffer and hence resists further reduction in its radius. Conversely, in the rarefaction phase of the ultrasound pulse, the bubble becomes less stiff and therefore enlarges much more. The resulting nonlinear oscillation is responsible for the production of harmonics of the transmitted frequency, including the frequency double that of the sound emitted by the transducer, the second harmonic. When the MI is raised sufficiently, the bubbles undergo irreversible disruption with the production of a bright, but brief, high-intensity ultrasound signal.[2]

Specialized Imaging Techniques

The intravenous injection of microbubble contrast agents produces enhancement of the Doppler signal from blood. Early clinical investigations of contrast agents for liver imaging used conventional color and power Doppler, however, and discriminatory features of specific liver masses were not evident. Furthermore, the anticipated advantages of microbubble contrast agents on conventional Doppler are often offset by the disadvantages of color blooming and motion artifact (Fig. [2]). To overcome the problems of imaging contrast agents with traditional color Doppler, specialized techniques have been developed that provide preferential detection of the signal from the microbubble agents with suppression of the signal from the background tissue. The first developed of these techniques was harmonic imaging.[3] [4] In comparison to conventional imaging, which sends an ultrasound pulse and receives the returning echo at the same frequency, harmonic ultrasound sends at a standard frequency and receives at double that frequency, or at the second harmonic. As the microbubble contrast agents oscillate in response to the ultrasound field, higher frequency sound waves or harmonic signals are produced. Background tissue, by comparison, shows a linear response and sends a returning echo at the same frequency as the insonating pulse. Therefore, the harmonic imaging technique detects the signal from the microbubble agents in preference to the signal from the background tissue. This provides the advantage of improved sensitivity to the contrast agent echo and a reduction in both motion and blooming artifact. However, these improvements are achieved at a price. A narrow bandwidth is required for the transmitted and the receiver frequency, so that the range of frequencies transmitted does not overlap with the range of frequencies around the second harmonic that is received. This narrow bandwidth results in a trade-off between contrast and resolution that is inherent to the harmonic imaging method. A second disadvantage of harmonic imaging is related to the filtering approach. The contrast in the harmonic image depends on the strength of the echoes from the bubbles, which is dependent on the concentration of bubbles and the intensity of the incident utrasound pulse. In practice, therefore, scanning in harmonic mode must be performed with a high MI. Because this causes disruption of the bubbles as they enter the scan plane, the bubbles will show a brief bright signal and then disappear. Vessels are, therefore, not seen in their entirety but appear rather as punctate areas of increased echogenicity.

Pulse inversion imaging is a more recent technical advance that provides an even greater sensitivity to the detection of the signal from a microbubble contrast agent without the compromises encountered with harmonic imaging. Pulse inversion imaging involves the rapid transmission of two ultrasound pulses, the second pulse a phase inverted or mirror image copy of the first (Fig. [3]). Normal tissue responds in a linear way, so that the echo from the second pulse is a mirror image of the echo from the first. The scanner adds these received signals, causing the echoes from linear structures to cancel. The successive insonation of the microbubbles, however, produces two nonlinear echoes that do not cancel when summed. Indeed, it can be shown that the sum contains the even harmonics of the bubble response.[5] Pulse inversion imaging overcomes the trade-off of contrast and resolution in harmonic imaging and provides greater sensitivity (Fig. [4]),[6] thus allowing low incident power and nondestructive, continuous imaging of microbubbles in an organ such as the liver.[7] Long lengths of vessels in both the liver and liver lesions can be seen and assessed for number, distribution, and morphology.

THE SONOGRAPHIC EXAMINATION

Liver lesion characterization with microbubble contrast agents is dependent on imaging of the agents while they are within the vascular pool. These imaging sequences are performed early, and have two objectives: to show the lesional vascularity and to predict the relative vascular volume of the lesion as compared with the adjacent liver.

Lesional Vascularity Assessment

The perfluorocarbon agents are best suited for vascularity assessment because they produce stable harmonic signals when insonated at low MI. A low MI is preferred because this will preserve the contrast agent population without destruction of the bubbles in the imaging field. If a high MI is selected, bubble destruction will not allow for visualization of long lengths of lesional vessels because the bubbles are destroyed as they enter the imaging plane, showing as small pools of contrast accumulation rather than as long lengths of lesional vessels.

Vascularity assessments are performed immediately following the injection of the contrast agent by an intravenous access and last for about 2 min following completion of the injection. Our preference is for injection of small boluses of contrast agent, each followed by a saline flush. Low MI continuous imaging in the vascular phase allows for assessment of the presence of lesional vascularlity, vessel number, and distribution, as well as vessel morphology (Fig. [5]). These characteristics have allowed differentiating features to be attributed to specific liver lesions.

Vascular Volume Assessments

Vascular volume assessments are most easily performed using some variation of an interval delay technique. Interruption of the imaging process for several seconds allows the entire vascular volume to fill with the contrast agent. By freezing the mechanism on the ultrasound machine for an interval of time followed by a brief reinsonation at high MI, bubble destruction will result in a brief and bright enhancement of the image as the bubbles accumulated during the delay are burst by the ultrasound beam. The intensity of the brightness will be in proportion to the number of bubbles that have accumulated during the interval delay. Therefore, comparison of the relative change in the echo level of the liver lesion to the uninvolved liver will give a relative measurement of their vascular volumes.[8] The timing of the interval delay relative to the time of injection will allow for assessments in the different phases of liver enhancement. We routinely perform our first interval delay sequence at the peak of arterial enhancement as assessed by the appearance of the contrast agent in the liver. The first frame following a cessation of insonation for 8 to 10 sec will produce an arterial phase image. An interval delay of 50 to 70 sec following injection of the saline flush is usually optimal for the portal venous phase. Longer interval delays, up to several minutes, may be appropriate for the evaluation of lesions with slow internal flow such as hemangiomas.[9]

Optimally performed with a large contrast agent bolus and a high MI, the interval delay technique also works with smaller injections over a range of MI selections. It is dependent on bubble destruction, which is most efficient at high MI but continues to occur at MI levels that are in the midrange. It is only with very low MI techniques that bubble destruction does not occur. Current technology developments offer the possibility that in the future these relative vascular volume assessments will be made in real time with very low MI techniques, which allow for continuous imaging without any bubble destruction.[10] Currently, however, bubble destruction following interval delay is the most sensitive reproducible method to determine a relative vascular volume assessment of a liver lesion. These interval delay techniques are more sensitive than low MI continuous imaging techniques for vascularity assessment. They are well performed with both air and perfluorocarbon agents, as all agents will disrupt with a high MI.

Insonation of the liver following an interval delay may show one of two appearances. A single-frame bright flash is the most common, although a multiple-frame event, referred to as the gray scale veil, is seen if a large bolus of a perfluorocarbon agent is used with a maximal MI. In this situation, the bubbles burst first in the near, then in the mid, and finally in the far field. In both appearances the enhancement of the liver lesion may be compared with the enhancement of the adjacent liver.

The gray scale appearance of a lesion in the interval delay sequence is a reflection of its relative vascular volume. The liver routinely appears enhanced or whiter than at baseline related to the disruption of the accumulated microbubbles of the liver vascular bed. A hyperarterialized mass will therefore appear white against a less white liver on an arterial phase interval delay sequence (Fig. [6]). Conversely, a hypoperfused lesion will appear as a dark hole in the enhanced liver on an arterial phase interval delay (Fig. [7]). To appreciate these differences optimally, storage of a brief cine loop of 10 to 30 frames allows for reevaluation of the events and close inspection of the first frame, which is invariably the most valuable for evaluation of this phenomenon.

LESION-SPECIFIC ANALYSIS

Using the above descriptions for performing the vascular analysis of a liver lesion, discriminatory features have been found in association with specific liver lesions.[7] [8] [11]

Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is a malignant tumor of the liver that is most commonly associated with previous liver injury from chronic viral liver infection (hepatitis B or C) or cirrhosis of any cause. The identification of a focal liver abnormality on the background of chronic liver disease, therefore, should raise the possibility of HCC.

Using microbubble contrast agents HCC shows as a highly vascular lesion in the overwhelming majority of patients studied. This lesional vascularity is usually diffuse and appears to be in excess of the adjacent liver parenchyma (Figs. 5 and 8). Aberrant vascular morphology is frequent and consists of highly tortuous blood vessels with corkscrew appearances (Fig. [5]). In real time, during the wash-in of the contrast, or following interval delay in the arterial phase, HCC shows as a hypervascular lesion appearing as a white ball against a less white but also enhanced liver parenchyma. Areas of tumor necrosis show as hypoechoic zones within the enhanced mass (Figs. 5 and 8). A nonenhancing scar is also an inconsistent and infrequent observation. Longer interval delays of 50 to 70 sec, which are in the portal venous phase of enhancement, continue to show liver enhancement, with the lesion showing as an area of lesser enhancement, similar to the appearance of HCC in the portal venous phase on enhanced CT scan. In fact, the CT features of HCC are highly consistent with the findings on contrast-enhanced ultrasound: a mass that is hypervascular in the arterial phase, with rapid washout in the portal venous phase of enhancement.

Metastastatic Disease

Metastases in the liver are a heterogeneous group of tumors whose vascularity reflects that of the primary lesion. The commonly encountered tumors from primaries located within the gastrointestinal tract, breast, and lung most often show as relatively hypovascular tumors. The visualized lesional vascularity often consists of fairly sparse and straight vessels concentrated in the periphery of the lesion. Following interval delay, lesions are often weakly enhanced or show marginal enhancement with a nonenhancing center that remains hypoechoic (Fig. [9]). As the lesion margin enhances, the lesion may appear both smaller and rounder on the interval delay image than on the baseline scan. In the portal venous phase of liver enhancement, lesions may become more conspicuous as the liver enhances around them (Fig. [10]).

Hypervascular metastases are seen with renal cell, neuroendocrine, and carcinoid primaries and show a hypervascular pattern that is frequently similar to that seen with HCC (Fig. [11]). Interval delay in the arterial phase, therefore, shows an enhancing tumor that is frequently brighter than the adjacent enhancing liver parenchyhma.

Hemangiomas

Hemangiomas are the most frequently encountered benign tumor of the liver and are seen in 5 to 20% of the population. Invariably detected as an incidental observation on a cross-sectional imaging examination of the abdomen performed for other and unrelated reasons, hemangiomas are insignificant observations. Their importance lies in their potential for confusion with more significant liver pathology.

The gray scale appearance of a hemangioma on sonography is often suggestive of the correct diagnosis, and two varieties of lesion are recognized. The typical hemangioma is a uniformly echogenic mass with infrequent increased through transmission. These lesions are often small and frequently multiple. An atypical hemangioma is a frequent variant of the above, showing as a lesion with a mixed echogenicity, a hypoechoic center, and a scalloped echogenic rim that may be either a thin rim or a thick rind. Recognition of the hemangioma in an otherwise asymptomatic patient, who is not at risk for either HCC or metastatic colon cancer, is usually not confirmed with other imaging. Conversely, frequent lesions, ultimately proven to be hemangiomas, are seen on sonography, CT scan, or magnetic resonance imaging (MRI), where the original assessment does not allow for a confident diagnosis of the nature of the mass. In these patients, there is a variety of confirmatory examinations. Contrast-enhanced CT scan and MRI both show hemangiomas as having peripheral nodular enhancement in the arterial phase with evidence of centripetal progression of the enhancement on delayed images. Recent studies suggest that microbubble contrast agents also have the ability to characterize hemangiomas with ultrasound.[9] [12] [13]

In the arterial phase of enhancement with microbubble contrast agents, hemangiomas often show sparse or marginal vascularity (Fig. [12]). This vascularity is distinctive, occurring in pools or puddles without definition of long lengths of lesional vessels. Following an interval delay in the arterial phase, a pattern of peripheral nodular enhancement, analagous to that described on CT scan, is evident (Figs. 12 and 13). These enhancing nodules typically are brightly enhanced and appear whiter than the adjacent enhanced liver. Interval delay sequences performed with longer intervals of delay, up to several minutes, characteristically will show centripetal progression of the pattern of peripheral nodular enhancement. Our routine is to perform these delays at the peak of arterial enhancement, and then at 30, 60, 90, and 120 sec following the saline flush at the end of the contrast injection. Unlike HCC, which is invariably white or enhancing on the arterial phase interval delay and then gray or nonenhancing on the portal venous phase and beyond, hemangiomas will tend to get whiter as the enhancement proceeds from the margin of the mass, in the arterial phase, toward its center by the portal venous phase and beyond.

The identification of an appearance that raises the possibility of peripheral nodular enhancement in the arterial phase interval delay is the first clue to the presence of a hemangioma. However, occasional metastatic lesions may also show this pattern of enhancement on arterial phase interval delay. Therefore, performance of further interval delay sequences with progressive elongation of the delay interval is essential to show the centripetal progression of the enhancement of the hemangiomas in distinction to the lack of progression that is seen with a metastatic lesion.

Focal Nodular Hyperplasia

Focal nodular hyperplasia (FNH) is the second most commonly encountered benign liver lesion and is seen most often as an incidental and insignificant observation in a young adult female. FNH is not an actual neoplastic tumor but represents instead a hyperplastic response to an underlyling vascular malformation in the liver. Histologically, it is made up of normal liver components, including hepatocytes and Kupffer cells. There are no normal portal veins, and the central area is characterized by both a hyalinized scar and hyperplastic arteries. Recognized as a hypervascular tumor in the arterial phase of enhancement on CT scans and MRI, the lesions also show characeristic scar and enhancement patterns. As a vascular scar neoplasm, FNH is rarely confused with other benign tumors, such as hemangiomas, but is frequently in question in the differential diagnosis of HCC, particularly of the fibrolamellar variety.

Vascular imaging with microbubble contrast agents shows FNH as a hypervascular lesion, often showing a central stellate vascularity and a large and tortuous feeding artery((Fig. [14]).[14] Following an interval delay at the peak of arterial enhancement, FNH invariably shows greater enhancement than the adjacent liver and appears most often as a bright white ball against the enhanced parenchyma. A nonenhancing scar is frequently observed.

Summary Regarding Liver Lesion Characterization

The enhancement of the Doppler signal from blood associated with the injection of a microbble contrast agent and the use of specialized imaging techniques, particularly pulse inversion imaging, have allowed for excellent depiction of lesional vascularity within focal liver masses with ultrasound. Pulse inversion imaging is currently the imaging method of choice, allowing for superb vessel clarity with the resolution associated with regular gray scale ultrasound imaging. Vascular morphology features have been highly contributory to the recognition of discriminatory features of specific lesions. Vascular volume assessments, not previously possible with ultrasound, show comparable vascular information to that provided on contrast-enhanced CT scans and MRI for liver lesion assessment. Future developments, focused on the ability to use very low MI nondestructive imaging, will undoubtedly afford the opportunity to further improve visualization of the vascularity of hypoperfused and slowly perfused lesions.

A potential negative aspect of characterizing a liver lesion with ultrasound is the difficulty of applying the contrast agent technique to the entire liver. A CT scan will show the entire liver on each and every scan, whereas a single lesion is optimal for an ultrasound characterization study. Although two or even three lesions could also be studied with ultrasound, certainly screening of the entire liver would be difficult.

POSTVASCULAR LIVER-SPECIFIC IMAGING WITH LEVOVIST

Levovist℗has a liver-specific phase that follows the clearance of the contrast agent from the vascular pool. The microbubbles will clear from the vascular pool from 2.5 to 3.0 min after an intravenous injection of the contrast agent.. The microbubbles, however, persist in the normal liver parenchyma possibly within the sinusoids, which are areas of low shear stress, or possibly within the reticuloendothelial cells following phagocytosis. Although the exact location of the persistent microbubbles remains controversial, the benefit of these persistent microbubbles is established. A high MI sweep through the liver will produce a bright enhancement of the liver parenchyma, in proportion to the bubble distribution.

Our technique for performing postvascular scans with Levovist includes separate intravenous injections followed by a delay of 3.0 to 4.5 min, during which time no liver scanning is performed. A high MI sweep is then performed to include all of the liver, if possible, in a single attempt. Rescanning generally is not successful because the bubbles burst on the first insonation. The timing of the delayed sweeps can be varied but seems to be optimal between about 2 and 5 min. Each circulation of the agent through the liver vasculature will result in some persistent contrast agent. We chose a time at 4 min postinjection to minimize the overlap in the contrast effects that could occur if the scans were performed earlier while there is still contrast agent within the vascular pool.

LIVER MASS DETECTION WITH POSTVASCULAR LEVOVIST SCANNING

The vascular imaging described above is highly useful in the evaluation of a known focal liver lesion and greatly increases the ability to diagnose a liver mass with ultrasound alone. Liver lesion detection, however, is a different problem and not surprisingly requires a different approach than that used for lesion characterization. In fact, the choice of contrast agent and imaging technique is also different. Delayed postvascular scanning with Levovist has shown consistently good results in the role of liver lesion detection. Uniform enhancement of the liver parenchyma on postvascular Levovist scans shows as an increase in the echo level as compared with baseline. Well recognized is the lack of enhancement of malignant lesions in the liver, including metastases and HCC, on delayed postvascular scans. Therefore, metastases typically show increased conspicuity on postvascular delayed scans of the liver, allowing for the detection of both smaller and more lesions than on baseline scans (Fig. [15]).[15] [16] A multicenter trial conducted in Europe and Canada showed competitive results from postvascular delayed Levovist scans with both CT scan and MRI in terms of the number of metastatic lesions detected in 150 patients. Our own experience has shown that HCC, which has a predilection for multifocal growth, is also more easily and reliably detected with postvascular scanning with Levovist.

The actual use of Levovist in clinical practice for the detection of liver lesions is controversial. Suffice it to say that postvascular delayed scans with Levovist unquestionably improve the ability to detect liver metastases with ultrasound. In many countries, CT scan is firmly established as the imaging test of choice in the patient at risk for metastatic disease. CT scan and MRI are routinely performed with contrast agent injection. Although this has not been the standard for an ultrasound scan, the detection of metastatic lesions with ultrasound is clearly superior with the use of contrast agents. The extent to which contrast agents will be used in the future is yet to be resolved. Successful application for the general use of contrast agents in clinical practice, reimbursement issues, and the impact of doing an intravascular injection as part of an ultrasound scan are all issues that will influence this decision in the future.

Other perfluorocarbon contrast agents undergoing clinical trials suggest they also will show persistence in the liver similar to that seen with Levovist. This may occur earlier in the enhancement phase, with the potential that high MI sweeps in the portal venous phase may also offer the ability of improved detection of malignant lesions with ultrasound. These agents, however, are not yet available for clinical use.

LIVER MASS CHARACTERIZATION WITH POSTVASCULAR LEVOVIST SCANNING

Very little material is published regarding the characterization of liver masses with postvascular delayed scans with Levovist.We have found, however, that FNH has positive enhancement on these scans and invariably shows an unenhancing scar. This enhancement pattern, in association with a hypervascular mass on vascular imaging, lends strong support to the suspicion that a hypervascular lesion is FNH.

SUMMARY

Liver mass evaluation with ultrasound contrast agents is still in its early stages. Nonetheless, current and recent investigations suggest great promise for microbubble-enhanced ultrasound in terms of both liver mass characterization and liver mass detection. Contrast-enhanced CT scan and MRI are considered the standards of practice for liver mass evaluation, and one would not consider interpreting a scan performed without contrast enhancement. It is our belief that the future for ultrasound should also include widespread and frequent use of contrast enhancement.

ACKNOWLEDGMENT

Supported by the Medical Research Council, the Terry Fox Program of the National Cancer Institute of Canada; Mallinckrodt Medical Inc., St Louis, MO; DuPont Inc, Billerica, MA; Berlex Canada; Lachine, Quebec.

ABBREVIATIONS

CT computed tomography

FNH focal nodular hyperplasia

HCC hepatocellular carcinoma

MI mechanical index

MRI magnetic resonance imaging

REFERENCES

• 1 Becher H, Burns P N. Handbook of contrast echocardiography. Berlin: Springer; 2000
• 2 Burns P N, Wilson S R, Muradali D. Microbubble destruction is the origin of harmonic signals from FS069.  Radiology . 1996;  201 158
  PubMedSearch in Google Scholar
• 3 Burns P N, Powers J E, Fritzsch T. Harmonic imaging: a new imaging and Doppler method for contrast enhanced ultrasound.  Radiology (Abst). 1992;  185(P) 142
  PubMedSearch in Google Scholar
• 4 Burns P N, Powers J E, Hope Simpson D. Harmonic imaging: principles and preliminary results.  Angiology . 1996;  47 63-73
  PubMedSearch in Google Scholar
• 5 Hope Simpson D, Chin C T, Burns P N. Pulse inversion Doppler: a new method for detecting nonlinear echoes from microbubble contrast agents.  IEEE Transactions UFFC . 1999;  46 372-382
  PubMedSearch in Google Scholar
• 6 Burns P N, Hope Simpson D, Averkiou M. Nonlinear imaging.  Ultrasound Med Biol . 2000;  26 19-22
  CrossrefPubMedSearch in Google Scholar
• 7 Burns P N, Wilson S R, Hope Simpson D. Pulse inversion imaging of liver blood flow: an improved method for characterization of focal masses with microbubble contrast.  Invest Radiol . 2000;  35 58-71
  CrossrefPubMedSearch in Google Scholar
• 8 Wilson S R, Burns P N, Muradali D. Harmonic hepatic ultrasound with microbubble contrast agent: initial experience showing improved characterization of hemangioma, hepatocellular carcinoma, and metastasis.  Radiology . 2000;  215 153-161
  CrossrefPubMedSearch in Google Scholar
• 9 Kim T K, Choi B I, Han J K. Hepatic tumors: contrast agent-enhancement patterns with pulse-inversion harmonic US.  Radiology . 2000;  216 411-417
  PubMedSearch in Google Scholar
• 10 Tiemann K, Lohmeier S, Kuntz S. Real-time contrast echo assessment of myocardial perfusion at low emission power: first experimental and clinical results using power pulse inversion imaging.  Echocardiography . 1999;  16 799-809
  CrossrefPubMedSearch in Google Scholar
• 11 Ophir J, Parker K J. Contrast agents in diagnostic ultrasound [published erratum appears in Ultrasound Med Biol 1990; 16(2):209].  Ultrasound Med Biol . 1989;  15 319-333
  CrossrefPubMedSearch in Google Scholar
• 12 Bertolotto M, dalla Parma L, Quaia E, Locatelli M. Characterization of unifocal liver lesions with pulse inversion harmonic imaging after Levovist injection: preliminary results.  Eur Radiol . 2000;  10 1369-1376
  CrossrefPubMedSearch in Google Scholar
• 13 Dill-Macky M J, Burns P N, Khalili K, Wilson S R. Focal liver masses: interpreting patterns of enhancement with Levovist and pulse inversion imaging.  Radiology (in press).
  PubMed
• 14 Uggowitzer M, Kugler C, Groll R. Sonographic evaluation of focal nodular hyperplasia (FNH) of the liver with a transpulmonary contrast agent (Levovist).  Br J Radiol . 1998;  71 1026-1032
  CrossrefPubMedSearch in Google Scholar
• 15 Blomley M JK, Albrecht T, Cosgrove D O. Improved imaging of liver metastases with stimulated acoustic emission in the late phase of enhancement with the US contrast agent SH U 508A: early experience.  Radiology . 1999;  210 409-416
  CrossrefPubMedSearch in Google Scholar
• 16 Blomley M J, Albrecht T, Wilson S R. Improved detection of metastic liver lesions using pulse inversion harmonic imaging with Levovist: a multicenter study.  Radiology . 1999;  213(P) 491
  PubMedSearch in Google Scholar

REFERENCES

• 1 Becher H, Burns P N. Handbook of contrast echocardiography. Berlin: Springer; 2000
• 2 Burns P N, Wilson S R, Muradali D. Microbubble destruction is the origin of harmonic signals from FS069.  Radiology . 1996;  201 158
  PubMedSearch in Google Scholar
• 3 Burns P N, Powers J E, Fritzsch T. Harmonic imaging: a new imaging and Doppler method for contrast enhanced ultrasound.  Radiology (Abst). 1992;  185(P) 142
  PubMedSearch in Google Scholar
• 4 Burns P N, Powers J E, Hope Simpson D. Harmonic imaging: principles and preliminary results.  Angiology . 1996;  47 63-73
  PubMedSearch in Google Scholar
• 5 Hope Simpson D, Chin C T, Burns P N. Pulse inversion Doppler: a new method for detecting nonlinear echoes from microbubble contrast agents.  IEEE Transactions UFFC . 1999;  46 372-382
  PubMedSearch in Google Scholar
• 6 Burns P N, Hope Simpson D, Averkiou M. Nonlinear imaging.  Ultrasound Med Biol . 2000;  26 19-22
  CrossrefPubMedSearch in Google Scholar
• 7 Burns P N, Wilson S R, Hope Simpson D. Pulse inversion imaging of liver blood flow: an improved method for characterization of focal masses with microbubble contrast.  Invest Radiol . 2000;  35 58-71
  CrossrefPubMedSearch in Google Scholar
• 8 Wilson S R, Burns P N, Muradali D. Harmonic hepatic ultrasound with microbubble contrast agent: initial experience showing improved characterization of hemangioma, hepatocellular carcinoma, and metastasis.  Radiology . 2000;  215 153-161
  CrossrefPubMedSearch in Google Scholar
• 9 Kim T K, Choi B I, Han J K. Hepatic tumors: contrast agent-enhancement patterns with pulse-inversion harmonic US.  Radiology . 2000;  216 411-417
  PubMedSearch in Google Scholar
• 10 Tiemann K, Lohmeier S, Kuntz S. Real-time contrast echo assessment of myocardial perfusion at low emission power: first experimental and clinical results using power pulse inversion imaging.  Echocardiography . 1999;  16 799-809
  CrossrefPubMedSearch in Google Scholar
• 11 Ophir J, Parker K J. Contrast agents in diagnostic ultrasound [published erratum appears in Ultrasound Med Biol 1990; 16(2):209].  Ultrasound Med Biol . 1989;  15 319-333
  CrossrefPubMedSearch in Google Scholar
• 12 Bertolotto M, dalla Parma L, Quaia E, Locatelli M. Characterization of unifocal liver lesions with pulse inversion harmonic imaging after Levovist injection: preliminary results.  Eur Radiol . 2000;  10 1369-1376
  CrossrefPubMedSearch in Google Scholar
• 13 Dill-Macky M J, Burns P N, Khalili K, Wilson S R. Focal liver masses: interpreting patterns of enhancement with Levovist and pulse inversion imaging.  Radiology (in press).
  PubMed
• 14 Uggowitzer M, Kugler C, Groll R. Sonographic evaluation of focal nodular hyperplasia (FNH) of the liver with a transpulmonary contrast agent (Levovist).  Br J Radiol . 1998;  71 1026-1032
  CrossrefPubMedSearch in Google Scholar
• 15 Blomley M JK, Albrecht T, Cosgrove D O. Improved imaging of liver metastases with stimulated acoustic emission in the late phase of enhancement with the US contrast agent SH U 508A: early experience.  Radiology . 1999;  210 409-416
  CrossrefPubMedSearch in Google Scholar
• 16 Blomley M J, Albrecht T, Wilson S R. Improved detection of metastic liver lesions using pulse inversion harmonic imaging with Levovist: a multicenter study.  Radiology . 1999;  213(P) 491
  PubMedSearch in Google Scholar