Christian W. Hoffmann

Improved detection of liver metastases with phase inversion imaging during the liver-specific phase of the ultrasound contrast agent levovist.

The complete text will be published in Radiology.

Improved detection of liver metastases with phase inversion ultrasound during the late phase of levovist

The sensitivity of ultrasound (US) in the detection of liver metastases ranges between 53% and 84% (1,2); thus, US is inferior to computed tomography (CT) and magnetic resonance imaging (MRI) in this application. Most lesions missed on US are either small—the sensitivity for detecting lesions 1 cm is only 20% (1)—or near-isoechoic and, thus, mainly reflect the relatively poor contrast between lesion and normal liver parenchyma. It was recently shown that the US contrast agent Levovist (Schering AG, Berlin, Germany), which was developed to enhance Doppler signals in technically difficult vascular studies, accumulates in liver parenchyma during a late liver-specific phase (3). This follows after the agent’s blood pool phase and starts approximately 2.5 min after an intravenous (IV) bolus injection. It lasts approximately 30 min provided no bubble destruction has occurred due to earlier scanning (4,5). The late-phase contrast effect is specific to normal liver (and spleen) parenchyma and spares focal lesions such as metastases (3,5,6). The increase in backscatter provided by Levovist is not sufficient to obtain late-phase liver enhancement on conventional (fundamental) B-mode US. This requires highly sensitive microbubble-specific imaging techniques that use the nonlinear (ie, distorted) signals that are returned from microbubbles when scanning at high transducer output (still within the range of diagnostic US). In the case of Levovist, stimulated acoustic emission (SAE) is the most important nonlinear phenomenon in this context. SAE is a process of bubble destruction caused by the insonating US beam, which produces a characteristic, strong, transient, nonlinear signal of a wide frequency range (7,8). So far, SAE enhancement of liver parenchyma has been mainly imaged using velocity-encoded color Doppler US, in which the wide frequency range of the returned SAE signal is interpreted by the scanner as a “pseudo Doppler shift” and, thus, displayed as a characteristic dense color mosaic with a range of velocities randomly distributed in the liver parenchyma. This color mosaic spares focal lesions such as metastases and markedly improves lesion-to-liver contrast. In a preliminary series of 19 patients with liver metastases, Blomley and associates compared color SAE imaging during the late phase of Levovist with conventional B-mode US. The authors found not only that SAE imaging improved lesion conspicuity in all cases but it also detected three additional metastases (6). However, this color technique suffers from several limitations: The enhancement effect is limited to the color box, and temporal as well as spatial resolution is inferior to B-mode US. Furthermore, the enhancement is encoded as “yes or no” information (color present or absent), which can make image interpretation difficult especially with regards to artifacts. We recently presented the first human results of a new, highly sensitive, microbubble-specific US technique called Acad Radiol 2002; 9(suppl 1):S236–S239

Characterization of Focal Liver Lesions with Phase Inversion Ultrasound During the Late Liver-Specific Phase of Levovist

On baseline scans mean C-values were: metastasis: 2.8 ( 5.5) dB, HHC: 3.2 ( 4.0) dB, haemangioma: 3.9 ( 5.1) dB, FNH 0.9 ( 4.8) dB, FF: 1.0 ( 7.6) dB, all malignant lesions: 2.9 ( 5.1) dB, all benign lesions: 1.6 ( 6.1) dB. On contrast enhanced PIUS mean C-values were: metastasis: 17.9 ( 7.3) dB, HHC: 14.5 ( 9.4) dB, haemangioma: 1.4 ( 5.6) dB, FNH 4.4 ( 5.15) dB, FF: 0.4 ( 2.1) dB, all malignant lesions: 17.0 ( 7.9) dB, all benign lesions: 1.8 ( 5.1) dB. The differences between benign and malignant lesions were significant both pre and post contrast (p 0.02 and p 0.0001). However on baseline there was a big overlap between the 2 groups and no discrimination was possible. Conversely, post Levovist all malignant lesions showed a C 5 dB while all but two benign lesions (one haemangioma and one FF) had a C 5dB.

Sonographie von Lebermetastasen mit leberspezifischem Kontrastmittel

ZusammenfassungSeit wenigen Jahren stehen auch für die Sonographie Kontrastmittel zur Verfügung, die für die Diagnostik fokaler Leberläsionen eingesetzt werden können. Einige dieser Kontrastmittel (z. B. Levovist, Schering AG, Berlin) verfügen über eine leberspezifische Spätphase. Um diese Spätphase klinisch nutzbar darzustellen, sind kontrastmittelspezifische Bildgebungstechniken wie die Phasen- oder Pulsinversion erforderlich. Unter Verwendung der Phaseninversion zeigt das normale Leberparenchym in der Spätphase ein ausgeprägtes Enhancement. Metastasen bleiben von diesem Enhancement ausgespart und zeigen sich als scharf begrenzte echoarme oder nahezu echofreie Läsionen. Dadurch wird die Abgrenzbarkeit der Metastasen erheblich verbessert. So können mit dieser Technik bei bis zu 45% der Patienten im Vergleich zur nativen Sonographie zusätzliche Metastasen nachgewiesen werden. In einer multizentrischen Studie an 128 Patienten konnte dadurch die Sensitivität im Nachweis individueller Läsionen von 71 auf 88% gesteigert werden. Gleichzeitig verbesserte sich die Spezifität von 59 auf 88%. Somit stellt die kontrastmittelgestützte Sonographie eine vielversprechende Alternative zu anderen Schnittbildverfahren in der Diagnostik von Lebermetastasen dar.AbstractIn recent years sonographic contrast agents which can be used for liver imaging have become available. Some of these agents (e. g. Levovist, Schering AG, Berlin) display a liver-specific late phase. Visualisation of this late phase requires contrast-specific imaging techniques such as phase or pulse inversion. When scanned in phase inversion during the late phase, normal liver parenchyma shows strong enhancement. This enhancement spares metastases which stand out as echo-poor or almost echo-free enhancement defects. This improves the conpicuity of metastases markedly. The technique increases the number of detectable metastases in up 45% of patients in comparison to unenhanced sonography. In a multi-centre study on 128 patients the sensitivity in the detection of individual metastases was increased from 71% to 88% and specificity improved from 59% to 88%. Contrast-enhanced sonography thus represents a promising alternative to other cross-sectional imaging modalities in the diagnosis of hepatic metastases.