FLORIAN VAN BOMMEL, M.D.1THOMAS WUNSCHE, M.D.2DIRK SCHURMANN, M.D.2THOMAS BERG, M.D.11Medizinische Klinik mit Schwerpunkt Hepatologie und

Gastroenterologie2Medizinische Klinik mit Schwerpunkt InfektiologieChariteCampus Virchow-KlinikumHumboldt-Universitat zu BerlinBerlin, Germany

References1. Lok AS, Heathcote EJ and Hoofnagel JH. Management of hepatitis B

2000—summary of a workshop. Gastroenterology 2001;120:1828-1853.2. Stuyver LJ, Locarnini SA, Lok A, Richman DD, Carman WF, Dienstag JL,

Schinazi RF. Nomenclature for antiviral-resistant human hepatitis B virusmutations in the polymerase region. HEPATOLOGY 2001;33:751-757.

3. Perrillo R, Schiff E, Yoshida E, Statler A, Hirsch K, Wright T, Gutfreund K,et al. Adefovir Dipivoxil for the treatment of lamivudine-resistant hepatitisB mutants. HEPATOLOGY 2000;32:129-134.

4. James JS. Tenofovir approved: broad indication. AIDS Treat News 2001;26:2-3.

5. Cihlar T, Birkus G, Greenwalt DE, Hitchcock MJM. Tenofovir exhibitslow cytotoxicity in various human cell types: comparison with other nucle-oside reverse transcriptase inhibitors. Antiviral Res 2002;54:37-45.

6. Schwetz BA. New drug for HIV-1 infection. From the Food and DrugAdministration. JAMA 2001;286:2660.

7. Barditch-Crovo P, Deeks SG, Collier A, Safrin S, Coakley DF, Miller M,Kearney BP, et al. Phase I/II trial of the pharmacokinetics, safety, andantiretroviral activity of tenofovir disoproxil fumarate in human immuno-deficiency virus-infected adults. Antimicrob Agents Chemother 2001;45:2733-2739.

8. Ying C, De Clercq E, Nicholson W, Furman P, Neyts J. Inhibition of thereplication of the DNA polymerase M550V mutation variant of humanhepatitis B virus by adefovir, tenofovir, L-FMAU, DADP, penciclovir andlobucavir. J Viral Hepat 2000;7:161-165.

Copyright © 2002 by the American Association for the Study of Liver Diseases.doi:10.1053/jhep.2002.35328

Liver Nodule and US Medium Contrast Agents: Past and Future

To the Editor:

We read with interest the article by Fracanzani et al. inthe December issue of HEPATOLOGY1 dealing withnoninvasive tissue characterization of small nodules incirrhosis by means of contrast-enhanced Doppler ultra-sonography (US).

We would like to comment on the methodologic as-pects of the report concerning the US technique.

First of all, it is quite surprising that no case has beenexcluded from the study owing to technical difficulties.Liver lesions under the diaphragm, especially if near theheart, are poorly explored with color Doppler because ofboth US beam attenuation and motion artifacts.2 In arecent study using contrast-enhanced US, 7% of 60 hep-atocellular carcinomas displayed no vascular signs becauseof their deep location.3

Such technical limitations are overwhelming whenthin, slow blood flow vessels inside nodules less than 2 cmhave to be studied (54% of the focal lesions included inthe report by Fracanzani et al. had a diameter less than 2cm). Injecting the US medium contrast, such as Levovist(Schering, Berlin, Germany), does not improve the visu-alization of the microvasculature because images fromsuch examinations show extensive color blooming andsaturation of the color box, artifacts that are exceedinglyprominent if the fundamental (conventional) harmonic isused, as Fracanzani et al. did. In addition, coupling con-ventional color Doppler with a US contrast medium al-lows only the macrovasculature of the lesions to be

displayed because with continuous insonation the bub-bles are continuously destroyed in larger vessels and thusnever reach the smaller downstream vessels.4 If the aim ofthe study by Fracanzani et al. was especially to differenti-ate hepatocellular carcinomas from either atypical orbenign nodules by showing subtle differences in micro-vasculature, then the method used by Fracanzani et al. isclearly too rough and inaccurate.

To overcome the limitations of methodologic ap-proaches such as that used by Fracanzani et al., over thepast few years US equipment has been refined and newercontrast agents have been extensively studied.3-5 The newstrategies for tissue characterization of liver nodules (ineither a normal or cirrhotic liver) entails the combinationof wide-band harmonic imaging, US contrast agents athigh and low mechanical index, and different scanningtechniques (e.g., intermittent imaging, pulse inversion).

Harmonic imaging preferentially is used to detecthigher-frequency echoes, specifically those at the secondharmonic: characteristically moving tissue does not giveharmonic echoes so that the motion artifacts are sup-pressed.3-5

On the other hand, intermittent transmission of theultrasound beam (intermittent harmonic imaging) allowsa high mechanical index contrast agent (such as Levovist)to be visualized in small vessels,3-5 whereas to image thecontrast agent in blood vessels without destroying thebubbles a low mechanical index contrast agent (the sec-ond generation of contrast agents made of microbubblescontaining gas different from air) must be used, which

508 CORRESPONDENCE HEPATOLOGY, August 2002

reduces the sensitivity of the image but allows a real-timeevaluation of the different vascular phases of the contrastmedium through the lesion.4,5

The final aim of these efforts is to display, possibly inreal-time and at acceptable costs, the microcirculatory ki-netic characteristics of space-occupying liver lesions thatare more informative about their nature and competewith the current available contrast-enhanced images ofcomputed tomography and magnetic resonance.

In light of these considerations, the high diagnosticperformance of the contrast-enhanced US method byFracanzani et al. seems exceedingly optimistic and re-flects, at least in part, both the high prevalence of thedisease in the population studied (hepatocellular carcino-mas represented half of the cases included in the study)and the clinical setting of the study (i.e., a scientific workperformed by highly motivated operators).

The clinical audience of HEPATOLOGY should be awarethat the methods used by Fracanzani et al. represent theimmediate past and not an updated work in progress. Thenew US equipment, scanning techniques, and contrastagents are on the way and the results of their clinicalapplications will supply a new body of information onnoninvasive tissue characterization of liver lesions, espe-cially when arising in a cirrhotic liver.

GIAMPIERO FRANCICA, M.D.Servizio di Ecografia ed EcointerventisticaPresidio Ospedaliero S. Maria della PietaCasoria, Italy

ANTONIO GIORGIO, M.D.Servizio di Ecografia ed EcointerventisticaAzienda Ospedaliera D. CotugnoNapoli, Italy

References1. Fracanzani AL, Burdick L, Borzio M, Roncalli M, Bonelli N, Borzio F, Mara-

schi A, et al. Contrast-enhanced Doppler ultrasonography in the diagnosis ofhepatocellular carcinoma and premalignant lesions in patients with cirrhosis.HEPATOLOGY 2001;34:1109-1112.

2. Koito K, Namieno T, Morita K. Differential diagnosis of small hepatocel-lular carcinoma and adenomatous hyperplasia with power Doppler sonog-raphy. AJR Am J Roentgenol 1998;170:157-161.

3. Tanaka S, Ioka T, Oshikawa O, Hamada Y, Yoshioka F. Dynamic sonog-raphy of hepatic tumors. AJR Am J Roentgenol 2001;177:799-805.

4. Wilson SR, Burns PN, Muradali D, Wilson JA, Lai X. Harmonic hepaticUS with microbubble contrast agent: initial experience showing improvedcharacterization of hemangioma, hepatocellular carcinoma, and metastasis.Radiology 2000;215:153-161.

5. Solbiati L, Cova L, Ierace T, Marelli P, Dellanoce M. Liver cancer imaging;the need for accurate detection of intrahepatic disease spread. J ComputAssist Tomogr 1999;23(Suppl):S29-S37.

Copyright © 2002 by the American Association for the Study of Liver Diseases.doi:10.1053/jhep.2002.33633

Reply:

We thank Drs. Francica and Giorgio who have givenus the opportunity to better explain a few points appar-ently not too clear in our manuscript. Contrary to whatDrs. Francica and Giorgio imply, we did not analyze thetumor vascularity pattern after contrast enhanced powerDoppler, but we performed Doppler spectral analysis onthe intratumoral arterial vessel detected. Hepatocellularcarcinoma is a highly vascular tumor and contrast-en-hanced power Doppler is a sensitive method of detectingarterial flow signals in tumor nodules.1,2 Further charac-terization of the nodules is possible using spectral analysisof the Doppler shift, which allows the calculation of thepulsatility and resistive indices of the arterial signals asindicated in our report. As for artifacts such as color“blooming” and saturation of the color box, they are of noclinical relevance if the appropriate concentration of con-trast medium and correct gain settings are used as re-ported in the literature.1 We stress the fact that our reportis not a morphologic study of the tumor vascularity suchas those using pulse inversion imaging with high mechan-ical index agents. The newer low mechanical index agentsseem to be more promising in this respect. However it isimportant to realize that these new techniques only re-cently available to the clinician are still, in part, experi-mental and require dedicated equipment with the relativeeconomical investments. Our study began 4 years agowhen these new techniques were not available. Themethod we described is useful in the characterization ofsmall lesions that occur during follow-up in cirrhotic pa-tients. Finally we pointed out that this approach does notrequire specialized equipment but can be performed withcurrently available machines with power-Doppler capa-bility. We obtained a sensitivity of contrast enhanced ul-trasound of 95%, which is similar to that obtained byTanaka et al.3 with pulse inversion harmonic imaging andcontrast agent (Levovist), and we found a similar percent-age of HCC with no arterial vascularization (5% vs. 7%).

Regarding the point risen by Drs. Francica and Gior-gio on the prevalence of HCC, it is worthwhile notingthat the prevalence of HCC in our study does not seem tobe higher than that reported in Italy (3.5% per year). Infact, from 500 cirrhotic patients in follow-up for 4 yearswe have found 20 monofocal HCC and 21 dysplastic orregenerative nodules. The remaining 12 lesions not in-cluded in the study were all HCC.

Thus, although we appreciate the interest of Drs. Fran-cica and Giorgio in our work, we conclude that the settingthey mention (equipment and series of patients) is not theone described in our report.

HEPATOLOGY, Vol. 36, No. 2, 2002 CORRESPONDENCE 509