| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Radiology and Nuclear Medicine, Universitätsklinikum Benjamin Franklin, Freie Universität Berlin, Hindenburgdamm 30, D-12200 Berlin, Germany, e-mail: t.albrecht@medizin.fu-berlin.de; Department of Imaging, Hammersmith Hospital, Imperial College, London, England
Editor:
We read with great interest the article by Dr Wilson and colleagues in the April 2000 issue of Radiology (1) that reported the use of a microbubble contrast agent and harmonic gray-scale imaging for characterizing focal liver lesions. The authors are to be congratulated on their study in which they demonstrate the advantage of contrast material–enhanced harmonic gray-scale imaging compared with conventional ultrasonography (US) for characterizing focal liver lesions.
The authors used interval-delay imaging: They interrupted scanning for 5–8 seconds at peak enhancement in the systemic circulation so that no bubble destruction would take place during the interval. The liver, including the lesion, could then fill with fresh bubbles. After the delay, imaging was commenced with a high mechanical index, which destroyed the bubbles and thus produced a strong signal from the bubbles. Using this technique, the authors found marked enhancement of hepatocellular carcinoma, a mixed pattern with mainly peripheral enhancement in metastases, and little or no enhancement in most hemangiomas. The authors concluded that the enhancement at interval-delay imaging reflects the vascular volume of a lesion and that hemangiomas have a vascular volume lower than that of normal liver.
This conclusion is at odds with the appearance of hemangiomas with other imaging modalities, such as computed tomography (CT), magnetic resonance (MR) imaging, and blood-pool scintigraphy. Hemangiomas are highly vascular tumors consisting of interconnected endothelial-lined vascular channels enclosed in a loose fibroelastic stroma (2). They typically show peripheral enhancement with gradual centripetal filling and are regarded as having a large slowly enhancing blood volume. The authors claim that this discrepancy is due to the fact that MR and CT contrast agents diffuse into the interstitial spaces of these lesions, while microbubbles do not.
We disagree with this statement, which is also at odds with the known data about hemangioma enhancement at scintigraphy with labeled red blood cells, which is, of course, a blood-pool technique. We would argue that the discrepancy is in fact explicable by means of the scanning method used by Dr Wilson and colleagues.
In their study, interval-delay imaging was performed after the interruption of scanning for 5–8 seconds. Thus, the time allowed for the undestroyed bubbles to enter the imaging plane and to refill the lesion was only a few seconds. It is well established that the majority of hemangiomas require several minutes to centripetally fill with contrast material at CT and MR imaging. Therefore, it is likely that the lack of enhancement of hemangiomas seen by Dr Wilson and colleagues was due to the short delay used, which allowed only the periphery of the lesions to enhance so that the majority of it appeared nonenhancing.
We suspect that a longer delay of 2 or 3 minutes would have produced different results and that such an approach would have better mimicked the use of delayed imaging at CT or MR imaging. Indeed, in two recently published articles (3,4), 37 hemangiomas with contrast-enhanced pulse-inversion harmonic US were examined by using the microbubble agent SH U 508A (Levovist; Schering, Berlin, Germany) and intermittent imaging. Findings of both studies showed complete or, in cases of large lesions, partial centripetal filling in almost all hemangiomas at 48 seconds to 4 minutes after the injection of the contrast agent. The enhancement was either of the same degree as or exceeded that of normal liver parenchyma. These results are in keeping with our own experience of using both an air-based microbubble agent (SHU 508A) and perfluoro-gas agents, as used by Dr Wilson and colleagues.
Without wishing to detract from a stimulating and important article, we make these comments in the hope of maintaining a distinction between this article and others in the literature.
REFERENCES
Department of Medical Imaging, Toronto General Hospital, 200 Elizabeth Street, Toronto, Ontario, Canada M5G 2C4* e-mail: stephanie.wilson@uhn.on.ca; Department of Medical Biophysics, Sunnybrook and Women’s Health Science Centre, Toronto, Ontario, Canada
We are grateful to Drs Albrecht and Blomley for their comments on our recent article (1) and for highlighting an important limitation of our interpretation of the behavior of the hepatic hemangiomas that we studied. Our studies were performed only in the arterial phase of enhancement. The fact that the hemangiomas showed an enhancement lower than that of the adjacent liver and other tumors, such as hepatocellular carcinoma, indeed does not mean that they have a lower vascular volume.
We agree that longer interval delays are required to show the slow accumulation of contrast agent with the centripetal filling that is characteristic of hemangioma. Investigations with microbubble contrast agents are still in the early stages, and the technology has rapidly evolved since these initial studies were performed more than 2 years ago. At that time, the sensitivity of harmonic imaging was such that we needed a large bolus of microbubble contrast agent, and we could detect the agent only shortly after the peak of the arterial phase. Now we (2), as well as Kim et al (3), Bertolotto et al (4), and others, have more sensitive methods, such as pulse inversion imaging, at our disposal that routinely reveal the slower filling of hemangiomas.
If we suspect a hemangioma, it is now our practice to add to the 8–10-second delay described in this article additional delays of 45, 60, and 90 seconds or longer to show centripetal progression of the enhancement, which we now regard as a crucial element for the diagnosis of hemangioma (5). Occasional metastases and, even more rarely, hepatocellular carcinomas may show a pattern reminiscent of peripheral nodular enhancement with arterial-phase interval-delay sequences. These lesions, however, do not show centripetal progression of the enhancement with longer interval delays.
The newest developments in US imaging technology heighten the sensitivity to contrast agents, eliminating the need to destroy the bubbles to detect them (6). This allows us to perform these studies in real time and dispense with the interval delay and to use doses smaller than those used in our first study.
We believe that the ability to image the dynamics of the arterial and portal phases and beyond in real time will bring new utility to US in the diagnosis of liver lesions. We thank Drs Albrecht and Blomley for their clarification of what we can expect to see with hemangiomas. We are happy to have any misconception related to our publication clarified.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
