Ultrasound Elastography of Ethanol-induced Hepatic Lesions: In Vitro Study

Chin Med Sci jVoL 24 No. 2June 2009P.81-85CHINESEMEDICAL SCIENCESJOURNALORIGINAL ARTICLEUltrasound Elastography of Ethanol-inducedHepatic Lesions: In Vitro StudyLi-gang Cui, Jin-hua Shao, Jin-rui Wang, Jing Bai, and Yi-zhuo ZhangDepartment of Diagnostic Ultrasound, Peking University Third Hospital, Beijing 100083, ChinaDepartment of Biomedical Engineering, School of Medicine, Tsinghua University,Beijing 100084, ChinaDepartment of Diagnostic Ultrasound, Beijing Shijitan Hospital, Beijing 100038, ChinaKey words: elastography; ultrasound hepatic cirrhosis; ethanolObjective To study the value of ultrasound elastography in evaluation of ethanol-induced IMethods Alcohol with a dose of 2 ml was injected into a fresh porcine liver under ultrasounduidance to create stiff necrosis. Then freehand elastography of the lesion from the identical scan plance wasobtained with SONOLINE Antares system using VF10-5 probe at about every 30 seconds till 6 minutes laterThe original high quality radiofrequency data were acquired through an ultrasound research interface whichas provided by the ultrasound system. Then, corresponding elastograms were produced offline usingcross-correlation technique and compared with gross pathology findingResults Gray-scale sonogram showed a hyperechoic area with acoustic shadow below appearedimmediately after alcohol injection. The hyperechoic area tended to be diffuse and its boundary to be illegiblewith time. On the contrary, the ethanol-induced lesion in elastogram appeared as a low strain hard regionsurrounded by high strain soft hepatic tissues, with clear but irregular boundaries. Sequential elastogramswith the sketched lesion boundaries showed that the lesion area increased in the first 3 minutes after ethanolinjection, and then reached a plateau which corresponding to gross specimenConclusion Ultrasound elastography is capable of detecting and evaluating the diffusion ofethanol-induced hepatic lesion, and more sensitive and accurate than routine sonographyISTOLOGIC evaluation of liver biopsy remains chronic liver disease. Ultrasonography has been the mainthe gold standard for assessing fibrosis and imaging modality for chronic liver disease follow-up, for itscirrhosis. As an invasive method, liver biopsy convenient and multiple-application advantages. Howevercan not be a perfect tool for follow-up of the sensitivity of ultrasonography is poor for diagnosinghepatic cirrhonn nhvintie change of acousticimpedance diff中国煤化工 a coefficient ofReceived for publication October 21, 20Corresponding author Tel: 86-10-62017691-2582, E-mail: jinruiCNMHGroSis. Sowang@sina.com.cnultrasonic researches did not get substantial results onA Supported by National Natural Science Foundation of Chinaearly diagnosis of hepatic fibrosis. Actually, liver elasticity(3047046)CHINESE MEDICAL SCIENCES JOURNALJune 2009has changed at this time, because the elastic properties of plate which was fixed on a micrometer caliper motionsoft tissues depend on their molecular building blocks and control systemon the microscopic and macroscopic structural organizationThe compression was imposed about every 30 secondsof these blocks. If we can observe these changes of me- until 6 minutes. During each compression, the radiofrechanical properties of the liver, we will get the information quency data were stored through URI. During the studof liver fibrosis. Yeh et al measured the elastic modulus of all parameters of the system were kept the same.After19 fresh human liver samples and 1 hepatic tumor sample, the data acquisition, the liver was dissected along the scanand compared with histologic grade of liver fibrosis. They plane, and a gross specimen photograph was taken as thefound the correlation between the fibrosis score and the reference picture. The stored radiofrequency data wereelastic modulus was significant (P<0.01). So, if we can then transferred to a laptop computer for offline processingaccurately measure the elastic modulus of liver, we will get using MATLAB(The Mathworks, Natick, MA). The MATLABinformation about liver fibrosis order. This study was based program was developed on the basis of an open-sourceonourpreviousultrasoundelastographywork.inthistoolboxdesignedforUridatafiles(http://ultrasonicsstudy, we applied ultrasound elastography to detect and bioen uiuc. edu/URI/index. htm). According to Luo et afsevaluate liver lesions induced by ethanol injection, and method, we rebuilt the sequential elastograms by aaimed to assess the value of our ultrasound elastographic cross-correlation method. Finally, the temporal formationsystem in diagnosing hepatic cirrhosis in vitroof the lesion was plotted as the elastographic area versusMATERIALS AND METHODSStatistical analysisInstrumentThe lesion area in elastogram was shown as x tsA SONOLINE Antares system(Siemens Medical SolutionsUSA, Inc, Mountain View, CA)with a VF10-5 transducerRESULTSwhich acted as a sonographic guidance system and anelastographic system, was used in this study. The systemDuring the injection, the gray-scaleprovides an ultrasound research interface(URI, which is showed a hyperechoic area with an obviously posteriorable to store high-quality radiofrequency data for offline acoustic shadow. The hyperechoic area was diffused andthe echogenicity was fading. Four minutes after injectionthe injection area was interfaced with liver tissue withoutMethodsclear boundary(Fig. 1)Eight fresh porcine livers were used in this study, and 15Dissected along the scan plane after ethanol injection,lesions were created by injection of 2 ml of 99.9% sterile the necrotic lesion was a white area in gross specimenethanol(Bei Hua Fine Chemicals Co, Ltd, Beijing, China) photogram. Although lesion shape was similar to someusing a 23-gauge needle. The liver was obtained from a extent in gray-scale sonogram, elastogram, and grossslaughterhouse about 4 hours after its excision, and was specimen photogram, the elastogram showed the lesionkept in a refrigerator until the experimentboundary more clearly than gray-scale sonogram, asInitially, the liver was placed on a flat metal board. The low-strain area which was easy discriminated from surinjection site was carefully chosen to make sure that it was rounding normal soft liver tissue. Also, the shape of lesionkept away from the vasculature and duct systems of the area as depicted by elastogram had high consistency withliver. Then the ethanol (a dose of 2 ml) was slowly injected gross pathology photograph finding(Fig. 2)into the specimen at the depth of 1.5 cm far from the liverSequential elastograms demonstrated local liver tissuesurface under ultrasound guidance. A successful injection hardness increasing after ethanol injection. When com-would encounter great resistance from hepatic tissue. pressed with probe, the injection area shape changed lessOtherwise, the ethanol was thought to have flowed alonthan surrounding liver tissue, since it had a larger elasticthe vasculature, and another scan plane was chosen for a modulus. The injection area which appeared as a blue areanew try.in elastogram中国煤化工 rmal liver tissueAfter the injection, the injection plane was marked on which appearedhe liver surface to indicate the position of this plane. Then the hard areaH出tograms showedNMHinutes after inthe compression was performed at a step of 1% of liver jection and then reached a plateau. The hard areas in thedepth with ultrasound transducer mounted to a plexiglass sequential elastograms 3 minutes after injection wereCHINESE MEDICAL SCIENCES JOURNALhighly similar(Fig. 4), and the average area at 4 minutes was 1.173+0.045 cm01015-10-50510151015Figure 1. Gray-scale sonograms of porcine liver after ethanol injectionA. Hyperechoic area with acoustic shadow appears immediately after ethanol injection. Small arrow indicates needle, andtriangle indicates the hyperechoic areaB. The hyperechoic area is diffused and its boundary is illegible 4 minutes after injectionThe number unit is mm. Y-axis indicates depth, and O means liver surface. The strap hyperechoic lines of the bottom of imagingrepresent the floor of container. X-axis indicates width and 0 means the position of needle100.002051015sIIr OIQ8AFigure 2. Lesion induced by ethanol injection shows a hyperechoic area with unclear boundary on sonogram(A); the lesion as a lowstrain, boundary clear area on elastogram(B); and gross specimen photographic view of necrotic area in white color(c)In Figure A and B, the number unit is mm. Y-axis indicates depth, and 0 means liver surface. The strap hyperechoic lines ofthe bottom of imaging represent the floor of container X-axis indicates width, and o means the position of needle tip the colorbar represents hardness scale0.00615-1015-10-50510-5深物■0000015-10-50515-10-50515-10-50515-10-505Figure 3. Elastograms of the lesion obtained at 0.5(A),1.0(B),1.5(C), 2.0(D),2.5(E)1中国煤化工s() after ethanolCNMHGhe number unit is mm. Y-axis indicates depth, and 2 means the location is 2ave. X-axis indicateswidth, ando means the position of needle tip. The color bar represents hardness scale, in which red represents high strain area(soft) and blue represents low strain area(hard)CHINESE MEDICAL SCIENCES JOURNALJune 2009were performed with FibroScan(EchoSens, Paris, France)a one-dimensional transient elastometry system", whichuses mild amplitude and low frequency vibrations to getstiffness information FirbroScan was easy to operate andcould be helpful for monitoring and evaluation of dilevelopment and assess therapeutic efficacy. HoweverFibroScan works only in one direction and detects theelasticity of organs by measuring the velocity of wavegoing through a 1 cmx2 cm circular cylinder area in theliver. It can not give the distribution of elastic modulus intwo-dimension. Further, during measurement, this systemcan not show gray-scale sonogram meantime. HepaticFigure 4. Induced necrotic area-time curve on elastogram aftervascular structures will also influence the accuracy ofmeasurementhe solid line and error bar denote the mean valueIn our study, we used the elastographic system builtand standard deviation of lesion area in 15 elastobefore. A small external compression was imposed on thetissue using probe or probe-crushing plate system alongthe longitudinal direction. Radiofrequency data were colDISCUSSIONlected pre-and post-compression and then strain distri-bution in two-dimension was estimated by a cross-corIn China, there are many chronic hepatitis B patients of relation method. The strain distribution was indirectlywhom quite a lot develop chronic liver disease to liver fi- described as elastic modulus in the tissue At the meantimebrosis and then liver cirrhosis at last. Effective therapeutic we can select proper plane for elastograstrategies for chronic liver disease to postpone and even ultrasound guidance. The previous study showed thireverse fibrosis have led to the compelling need of how to elastography system was workable, and it could describeearly detection of liver fibrosis. Liver palpation is a basic different elastic modulus distribution in the tissue but withskill of clinician to judge liver hardness, and it is based ora low signal to noise ratio. Besides controllable factorsassessment of tissue elastic modulus that is a mechanical such as keeping in the same plane and evenly compressioninformation. However, palpation is subjective and influwith probe, high quality raw radiofrequency data, is a keyenced by clinical experience. This method only has qualipoint for improvement of elastogram outcometative information with low sensitivity. There is no quanDifferent from previous study, this study usedtitative standard for comparison between different physi- SONOLINE Antares ultrasound system. The system pro-cians or same physician at different time. As a universal vides an URI, which is able to store high-quality radiofield with lots of research but with unsatisfied results 46 uimaging modality, ultrasonography has brequency data of pre- and post-compression. Reconstructed tissue elastogram has a higher signal to noise ratioIn 1991, Ophir et al introduced ultrasound elasto- We found compared with normal gray-scale sonogram,graphy, whicha new ultrasonic elasticity imaging elastogram was easier to read and showed the lestechnique that produced images of the strain distribution in boundary more clearly. It could evaluate the temporcompliant tissue. After that, ultrasound elastography has formation of the lesions. And further, findings of elasto-been the hot spot area in the medical imaging field. It had grams was highly similar to ones of pathological photobeen the subject of many studies, such as differentiation of graphs. All these results demonstrated that ultrasoundbreast lesions, evaluation of arterial sclerotic plaqueselastography could find stiffness changes of liver tissue anddetection of lesions induced by high intensity focuseddelineated the scope of stiffness. It could depict thetrasound and measurement of the lesions resulting from process of gradual stiffness with diffusion of ethanolradiofrequency ablation or ethanol injection. Although temporallymany studies are still in the laboratory or preclinical phaseThe mainthere were some gratifying applications on differential based which中国煤化工 at it is in vitrodiagnosis between benign and malignant lesions in breast circumstanceCNMHG vivo study withand thyroidanimal liver cirrhosis model which will be the basic step forMany studies about liver fibrosis and liver cirrhosis clinical applicationVol 24. No. 2CHINESE MEDICAL SCIENCES JOURNALREFERENCESCharacterization of plaque components with intravascularultrasound elastography in human femoral and coronary1. Chen Y, Wang BE, Bao-En, Jia JD, et al. Nonirarteries in vitro. Circulation 2000: 102: 617-23evaluation of liver flbrosis in chronic hepatitis b patients10. Luo JW, Ding CHX, Bai J, et al. Ultrasound elastographyChin J Hepato 2003; 11: 354-7applied to the detection of HIFU-induced lesions. Beijing2. Yeh wct al elasticBiomed Eng 2006; 25: 235-9urements of human liver and correlation with patholog11. Lyshchik A, Higashi T, Asato R, et al. Thyroid gland tumorUltrasound med Biol 2002: 28: 467-74diagnosis at US elastography. Radiology 2005; 237: 202Luo JW, Bai J, Wang w, et al. An experimental system forultrasound elastography. Chin J Sci Instrum 2006; 2Luo Jw, Shao JH, Bai J, et al. Using non-invasive transientelastography for the assessment of hepatic fibrosis. Chin J4. Meng FK, Zhent Y, Ge HY, et al. Correlation between theHepatol2006;14:395-7waveform changes of hepatic veins in patients with13i M. Handra-Luca A. Kettanehchronic hepatitis tested by color Doppler ultrasonographyassessment of liver fibrosis by measurement of stiffness inand the staging of liver fibrosis. Chin J Ultrasonographypatients with chronic hepatitis. Hepatology 2005; 41: 482006;15:2975. Meng FK, Ding L, Qu M, et al. Morphological changes of14. Castera L, Vergniol J, Foucher J, et al. Prospective com-hepatic parenchyma in staging of liver fibrosis in patientsparison of transient elastography, fibrotest, APRl, andith chronic liver diseases by high frequency ultrasoundliver biopsy for the assessment of fibrosis in chronicChin J Med Imaging Technol 2006; 22: 916-8hepatitis C. Gastroenterology 2005: 128: 343-506. Meng FK, Ding L, Zheng L, et al. Deviation of fibrosis15. Colletta c. Smirne C, fabris c, et al. value of two non-staging in patients with chronic liver diseases observedinvasive methods to detect progression of fibrosis amongwith ultrasonography from the pathological results: aHCV carriers with normal aminotransferases. Hepatologydiscussion J Ultrasound clin Med 2007:9: 394-72005:42:838-457. Ophir J, Cespedes I, Ponnekanti H, et al. Elastography: a16. Obara N. ueno y. Fukushima K. et al. transient elaquantitative method for imaging the elasticity of biologicalgraphy for measurement of liver stiffness measurementtissues. Ultrason Imaging 1991: 13: 111-3ly significant hepatic fibrosis in Japanese8. Ou B, Luo BM, Feng x, et al. The value of ultrasonicpatients with viral and nonviral liver disease. J Gastroelastography in differential diagnosing breast smallentero|2008;43:720-28masses. Chin J Ultrasonography 2007: 16: 506-89. de Korte CL, Pasterkamp G, van der Steen AFw, et alERRATUMIn the March 2009 article by Shih-chen Wang(Chin Med SciJ2009: 24: 3-11),"Recommendations on Strengthening theDevelopment of Nuclear Medicine in China, "the following corrections should be notedThe first sentence of the fourth paragraph on page 5 should read"At present, there are about 850 nuclear medicinedepartments2. In line 15 on page 6, the word"pathological"should be used instead of"physiological3. In line 28 on page 8, the year should be"instead of"1995中国煤化工CNMHG