Sage Journals: Discover world-class research

Abstract

Real-time elastography is an emerging sonographic imaging technique that provides a noninvasive method of evaluating biomechanical properties of a lesion: specifically, the elasticity, or stiffness, of the cellular composition relative to surrounding tissue. Clinical studies have demonstrated the diagnostic capability of elastography to detect specific cancers with accuracy that may substantially affect patient care and improve outcome. This literature review examines the basics of elastography, the advantages and limitations of this technique, and the results of research centered on its role in cancer detection and evaluation of tissue response to cancer therapies.

Keywords

elastography tumor response sonographic elastography cancer detection ultrasound

Cancer detection maintains a central role in diagnostic imaging; continued research is performed and technological improvements are made to increase the sensitivity and specificity of virtually all imaging modalities. The objective in such trials is to improve patient outcome through early and accurate characterization of malignancies as well as prevent costly and unnecessary follow-up studies and/or surgical biopsies. Similarly, imaging methods used to evaluate tumor response to treatment, such as adjuvant chemotherapy and ablative interventions, have been the focus of many clinical trials. Researchers are making increasing efforts to determine the efficacy of imaging techniques at characterizing these changes earlier in the plan of care, so treatment can be tailored to each individual in an attempt to decrease mortality rates and improve patient outcomes.

Real-time elastography (RTE) is an FDA-approved method of assessing soft tissue structures via a transdermal approach as well as intraoperatively.¹ This diagnostic imaging method assesses the biomechanical properties of tissue for elasticity, or the ability to resume its original shape following compression.² RTE methods consist of several compression-generating techniques including quasistatic, shear wave, and acoustic radiation force impulse (ARFI) imaging. Shear-wave elasticity imaging generates shear waves through acoustic radiation force. The velocity of these waves is measured to determine the stiffness of the tissue.^3,4 ARFI uses a focused ultrasound beam aimed at the targeted tissue for a short period of time (~0.1 milliseconds) to create and measure resulting tissue displacements. Quasistatic elastography can be implemented on many existing ultrasound systems with the addition of software. This method uses operator-induced mechanical compression to measure the degree of deformity, or tissue strain. This is calculated by the displacement of adjacent tissues producing an elastogram that maps the relative stiffness of the targeted lesion.⁵ Fibrotic, inflammatory, and malignant tissues are characteristically less elastic than other tissues, demonstrate a higher degree of stiffness, and will compress less than healthy tissue.^2,5,6 Tumors with initial elastographic interrogation can be monitored at intervals following treatment to assess for tissue changes such as a decrease in stiffness, or an increase in elasticity, resulting from tumor necrosis relating to a positive response to therapy.⁷ Assessment of tissue elasticity for malignant characterization has been performed in multiple organs including the pancreas, liver, prostate, and breast.

Discussion

Advantages

Ultrasound is a noninvasive, widely available, cost-effective method of diagnostic imaging with no risk of radiation exposure, adverse reaction to contrast, or other contraindications such as previous surgeries or impaired renal function common to other methods such as magnetic resonance imaging (MRI) and computed tomography (CT). In cancer detection, RTE has been reported to differentiate benign from malignant breast lesions with sensitivities of 78% to 100% and specificities of 21% to 98%.⁶ In conjunction with other imaging techniques, RTE can potentially improve the radiologist’s ability to accurately characterize malignant lesions and distinguish fibrotic tissue from cancerous growths. This capability has the potential to reduce unnecessary biopsies of characteristically benign masses, lowering costs and improving throughput and overall patient management.^5,6

Limitations

A particular limitation of RTE in cancer assessment is specific to the histologic cell type of the cancer being evaluated. RTE measures the stiffness of tumors based on the assumption that malignancies will possess a greater cell density; however, atypical cancers such as ductal cancers, medullary cancers, mucinous cancers, and papillary cancers do not follow the principles of this assumption and may therefore be underreported by RTE assessment alone.² Similarly, adjacent inflammatory tissues can reduce the sensitivity of RTE in cancer detection, as is seen in the presence of pancreatitis.^8,9 Other factors that increase interobserver variance, and thus reduce reported accuracy, include the type of cancer and the size of the lesion. These two factors affect the elasticity of the lesion and can influence interpretation.^2,10 These limitations have the potential to increase false-negative results.

Ultrasound is an inherently operator-dependent modality, and variance in interobserver agreement can result from inconsistent technique, level of experience, the available technology, and subjective interpretation of disease.^2,11,12 RTE is still in the development stages, and there is little consistency in the scoring system relative to tissue characteristics indicative of malignancy. Independent scoring systems have been used in many trials assessing RTE accuracy, but this lack of standardization reduces the reliability of published results. A great deal of literature centering on body mass index (BMI) as a substantial limitation suggests that increased body habitus perpetuates variance in RTE ultrasound. However, technological advancement in probe development is anticipated to significantly help overcome this persistent issue in RTE application.¹³

Literature Review

Variance

Variance in RTE results largely from a lack of protocol and standards for the application, measurement, scoring, and interpretation in this developing technique. The resulting disagreement between operators, observers, and interpreters can be misconstrued as evidence against the reliability of elastography, and this continues to be a challenging limitation of RTE. A clinical trial conducted to measure the variance in elasticity images was conducted to determine what factors have the greatest influence on quality. The results reported that image quality was inadequate in 21 cases (6.7%), low in 134 cases (42.9%), and high in 157 cases (50.3%).¹⁴ According to this study, higher image quality was reported in conjunction with smaller lesion size, shallower lesion depth, decreased breast thickness at the location of the tumor, and benign pathologic findings. The greatest impact on quality was inversely proportional to the thickness of the breast at the location of the target lesion where increasing thickness resulted in decreased quality. Other variables measured included age, BMI, mammographic density, and distance from the nipple; none of these factors had any appreciable impact on image quality. The reported sensitivity in differentiating benign from malignant masses between higher quality and lower quality images was 87.0% and 56.8%, respectively.

A study that reviewed previous cases in which RTE was used to measure liver tissue stiffness as it relates to fibrosis found a rate of measurement failures, or nondiagnostic quality, of 3.1%, and an additional 15.8% of reported results were determined to be unreliable.¹⁵ The authors highlighted the strong correlation of failure/unreliability with variables such as increasing age (>52 years), increased BMI (>30 kg/m²), coexisting type 2 diabetes, and operator inexperience. A similar study reflected comparable measurements of failure and unreliability at 5.3% and 16%, respectively, as did studies in France and China, reporting failure rates of approximately 5%. The implication of these studies points to obesity as a primary factor in unreliable or failed results in RTE. Despite the limitations noted in such studies, the variance depicted threatens reliability of RTE in clinical applications. Contrary to these findings, a study by Săftoiu et al¹⁶ of interobserver variability in the efficacy of elastography in differentiating focal masses in patients with chronic pancreatitis reported correlations between 0.86 and 0.94, with good reliability in reproduction of images between observers. The sensitivity was 93.4%, specificity 66.0%, positive predictive value 92.5%, negative predictive value 68.9%, and overall accuracy 85.4%.¹⁶

Pancreas

Chronic pancreatitis and pancreatic cancer are often coexistent, and the detection of focal abnormalities in the presence of inflammation is challenging. The diagnosis and plan of care for both pancreatic inflammation and malignancy, however, are largely dictated by imaging results.⁸ RTE is a safe and effective technique that has been reported to be instrumental in accurately diagnosing chronic pancreatitis and pancreatic cancer^8,17 (Figure 1). When compared with results of other imaging modalities, results of RTE assessment and biopsy of pancreatic masses have achieved a sensitivity of 85% to 90% and a specificity of virtually 100% in the absence of chronic or pseudo-tumoral pancreatitis. Considering that 20% to 35% of patients with pancreatic lesions have coexistent pancreatitis and that in this condition RTE typically has a lower sensitivity (approximately 75%), caution must be used when using this technique for diagnosis. A trial to determine the accuracy of RTE in differentiating between normal pancreas, chronic pancreatitis, and pancreatic cancer reported a sensitivity of 91.4%, specificity 88.9%, and accuracy of 90.6%.^8,9 A subgroup analysis within this study differentiating pancreatic cancer from pseudo-tumoral pancreatitis reflected good sensitivity at 93.8% and overall accuracy of 86%, but with low specificity of only 63.6%.

Figure 1.

Elastography image side by side with conventional B-mode image of a pancreatic carcinoma, demonstrating the increased stiffness (blue) of the tumor.

Another trial measuring RTE sensitivity and specificity in differentiating benign from malignant pancreatic lesions compared with conventional sonography showed a sensitivity and specificity for elastography of 92.3% and 80.0%, respectively, compared with 92.3% and 68.9%, respectively, for conventional B-mode images.¹⁸ A trial conducted by Larino-Noia et al¹⁹ evaluating RTE accuracy in characterizing solid pancreatic masses included RTE assessment of the mass compared with adjacent tissue as reference areas. The results were confirmed by histopathologic examination of the gross specimen. Endoscopic ultrasound (EUS) elastography had a sensitivity and specificity of strain ratio for detecting pancreatic malignancies of 100% and 92.9%, respectively.¹⁹

Liver

Hepatocellular carcinoma (HCC) is the third most common cause of cancer-related mortality worldwide, with the majority (80%) developing in patients with advanced liver cirrhosis or fibrosis, making it the greatest risk factor for HCC development.²⁰ Fibrotic changes in the liver have a strong correlation with later development of HCC, which may be treated by ablative therapies²⁰ (Figure 2). Elastography has been used to localize hepatic masses to improve the accuracy of biopsies and to determine the response of malignancies to therapy²¹ (Figure 3). Tissue response to ablation therapy has been researched to determine whether RTE can detect changes in the biomechanical properties of the tumor compared with surrounding tissues. In one such study, elastography demonstrated the ablated region as a well-circumscribed area of increased stiffness compared with nonablated surrounding tissue. These findings correlated well with contrast-enhanced CT images as well as with the gross specimen following resection.²²

Figure 2.

Elastography image side by side with conventional B-mode image of a liver with diffuse fibrotic changes, showing the diffuse nature of the areas of increased stiffness (blue).

Figure 3.

Real-time elastography image side by side with conventional B-mode image in a patient with cholangiocarcinoma acquired during an endoscopic, ultrasound-guided, fine-needle aspiration. The increased stiffness of the tumor (blue) can be seen clearly in the elastography image.

Prostate

Results of RTE evaluation of the prostate gland for cancer have been equivocal regarding its diagnostic value (Figure 4). Despite a clinical trial reporting 76% diagnostic accuracy of endorectal elastography for prostate cancer detection,²³ other studies found significantly lower reliability of this modality in prostate cancer evaluation. In a study by Magnoni et al²⁴ examining the sensitivity of RTE in characterizing malignant prostate masses when compared with histological samples obtained via transrectal biopsies, only 1 of 102 patients was determined to be true positive for prostate cancer, and 6 cases demonstrated false negatives. A clinical trial to evaluate malignant prostate tissue response to high-intensity focused ultrasound by elastographic imaging demonstrated a marked underestimation of residual tumor volume when compared with MRI.²⁵ The trial did note that technical limitations such as bandwidth and frame rate affected the diagnostic quality of elastographic ultrasound images. Both studies concluded that the limited accuracy, sensitivity, and specificity do not justify the routine application of real-time elastography in prostate cancer detection.

Figure 4.

Real-time elastography image side by side with conventional B-mode image showing a small lesion with increased stiffness (blue) on the right side of a prostate. The lesion was later confirmed to be prostate cancer.

Breast

Breast cancer tissue is less elastic than normal breast tissue; this increased hardness, or stiffness, is the property that allows some breast cancers to be palpated as well as characterized by comparative elasticity through RTE assessment²⁶ (Figure 5). The principle of elastography is that tissue compression produces strain (displacement) within the tissue and that the strain is smaller in harder tissue than in softer tissue. Therefore, by measuring the tissue strain induced by compression, we can estimate tissue hardness, which may be useful in diagnosing breast cancer. A study conducted by Ueno et al²⁶ evaluated the diagnostic value of RTE by examining 111 nodules and applied varying scoring system standards for characterization in determining its diagnostic accuracy. Elastography achieved a sensitivity, specificity, and accuracy of 86.5%, 89.8%, and 88.3%, respectively. Applying a different set of threshold values yielded a sensitivity, specificity, and accuracy of 71.2%, 96.6%, and 84.7%, respectively. A separate study using the same scoring system as Ueno et al demonstrated RTE sensitivity and specificity of 79% and 89%, respectively.^2,10 A study using a scoring system different from the preceding studies that included 874 breast lesions found a high specificity in benign lesions with a negative predictive value of 98% related to the entire group of lesions and 100% in lesions less than 5 mm.²⁷ An imaging comparison trial conducted by Ou et al²⁸ centered on differentiating benign from malignant breast lesions in dense breasts. Imaging modalities included B-mode ultrasound, RTE, and mammography, and the study concluded that RTE demonstrated the highest specificity (95.7%) and the lowest false-positive rate (4.3%). When compared with B-mode ultrasound, RTE diagnostic accuracy was higher at 88.2% vs 72.6%. Positive predictive values (PPVs) also exceeded B-mode at 87.1% vs 52.5%, respectively. Despite these results, sensitivity, negative predictive value, and false-negative rate were comparable to the other two methods. Increased false-negative rates in RTE were seen with invasive ductal carcinomas and those malignancies with a large area of central necrosis²⁸ (Figure 6). A combination of RTE and B-mode ultrasound had an improved sensitivity (89.7%), accuracy (93.9%), false-negative rate (9.2%), specificity (95.7%), and positive predictive value (89.7%).

Figure 5.

Elastography image side by side with conventional B-mode image of a fibroadenoma of the breast. The difference in stiffness between the lesion and the surrounding breast tissue is clearly contrasted in the elastography image.

Figure 6.

Elastography image side by side with conventional B-mode image of an invasive ductal carcinoma of the breast. Note the difference in stiffness of this lesion (blue) compared with the fibroadenoma of Figure 5.

Destounis et al¹¹ published results of a multicenter study evaluating the sensitivity and specificity of RTE in characterizing and differentiating breast lesions. Sensitivity and specificity obtained by the various centers participating in the study ranged between 96.7% and 100% and between 66.7% and 95.4%, respectively. The marked variance in specificity was attributed by the authors to differences in the examination technique. This concern about interoperator variance was also raised by Moon et al¹² as a potential limitation that undermines the reliability of published data and overall utility.

Tumor Response

Ensuring accurate characterization, staging, and monitoring of tumors and their response to therapy is a challenging but critical role of diagnostic imaging modalities (Figure 7).

Figure 7.

Representative elastography and B-mode images in patients with locally advanced breast cancer from a nonresponder (A) and a responder (B) taken at baseline prior to treatment, at week 1, at week 4, at week 8, and preoperatively.⁷ (The color bar on the right indicates relative stiffness; the scale bar equals 1 cm.)

Second to malignancies of the skin, breast cancer is the most frequent type of cancer diagnosed in women; more than 200,000 new cases of invasive breast cancer were diagnosed in the United States during 2012.⁷ Approximately 5% to 20% of these patients will present with locally advanced breast cancer (LABC), which is defined as stage III or inoperable disease, characterized by tumors that are larger than 5 cm and/or involving the skin or chest wall, with or without lymphatic involvement. When compared with early stage breast cancer, LABC has a much poorer prognosis and higher rate of recurrence (10%-20%). Only 55% of LABC patients survive to 5 years because of the high risk for metastatic spread. Approximately 75% of LABCs show marked response to initial chemotherapy, improving surgical outcome. In more than 50% of cases there is only microscopic tumor, or no residual tumor at all, following surgical intervention.²⁷

Imaging to assess for early functional changes that indicate the extent of therapy response is critical in determining the plan of care for cancer patients. The earlier a response can be detected, the more tailored a patient’s treatment can be to improve outcome. In LABC, administration of neoadjuvant therapy is a standard protocol prior to surgical resection to ensure disease-free margins and lower the chance of in situ reoccurrence. Such neoadjuvant therapy has been linked to increased survival rates up to 70%.^7,27 A recent study by Falou et al⁷ centered on elastographic assessment of tumor response to neoadjuvant therapy. Nine patients demonstrated positive response to neoadjuvant therapy by elastography evaluation that was confirmed surgically, and five patients demonstrated poor response to therapy by RTE. One patient demonstrated a false-positive response to therapy due to the invasive, mucinous nature of her specific LABC, a pattern that presents with biomechanical properties of decreased stiffness, atypical of LABC cancers.

Studies have measured tumor response to therapy in order to determine criteria for treatment efficacy. One such treatment that has been under development for the past two decades is percutaneous ethanol injection (PEI), studied for its effect on small HCCs. Ethanol has a pattern of diffusion in tissue that creates a cytotoxic environment resulting from protein denaturation, cellular dehydration, and microvessel thrombosis contributing to coagulation necrosis in local HCC cells. Studies have shown that up to 70% of treated HCC tumors smaller than 3 cm result in complete coagulation necrosis, and the 5-year survival rate is between 40% and 65% for PEI-treated patients who have concomitant hepatic cirrhosis.²⁹ To evaluate the potential of RTE to measure tumor response to treatment, Bai et al²⁹ conducted RTE following PEI, using the area of a lesion created in vivo to depict temporal formation of the ethanol-induced response. The results demonstrated the formation of a focal area of lower strain with well-defined borders within 2 minutes of PEI, the maximum area being reached at 2 minutes. The authors concluded that RTE is a valuable tool for monitoring tumor response to PEI. Their study also indicated some value in using RTE for real-time assessment of PEI response by necrotic formation. This will allow physicians to adjust the dose of PEI based on RTE findings, thus improving patient outcome and treatment efficacy and reducing recurrence rate of inadequately treated tumors.

Conclusion

RTE is an emerging imaging modality that provides data related to the biomechanical properties of tissue for characterization of malignant and benign masses. Limitations of RTE include operator dependence, increased BMI, tissue thickness anterior to breast masses, histologic composition of atypical cancers, and lack of standard scoring methods and protocols, which hamper reliability. However, RTE remains a cost-effective, noninvasive, and widely available technique that poses less risk to patients compared with other imaging modalities, making it ideal for screening and monitoring disease processes.² RTE has established a developing role in distinguishing benign and malignant masses in the pancreas, and the high degree of sensitivity in breast imaging suggests that this modality may reduce unnecessary biopsies. In addition to screening, published reports have reflected a strong correlation between RTE and pathologic response of breast tumors following neoadjuvant chemotherapy.⁷ This correlation has been documented in RTE determination of tumor response to ablative therapy as well.^22,29 These findings facilitate the establishment of protocols for techniques that monitor the response of cancer to specific therapies. RTE can be instrumental in tailoring treatment to patients exhibiting a negative tumor response. This ability of response monitoring has the potential to improve patient outcome, efficacy, and cost of care, reducing recurrence rates and overall mortality in some cancers. Overall, while RTE is a relatively new technique, research has supported the value of this modality in multiple cancer-related applications that promise to aid in the screening, detection, and monitoring of malignancies and enhancement of cancer therapies through measured response.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author received no financial support for the research,authorship,and/or publication of this article.

References

United States Food and Drug Administration. 2013. Available from: http://www.accessdata.fda.gov/cdrh_docs/pdf13/k131527.pdf. Accessed on June 5, 2013.

Bonardi

Alessi

Goddi

: Breast elastography: a literature review. J Ultrasound 2012;15:192–198.

Lee

Karadayi

Luo

YKS

: Real-time ultrasound elastography on a multi-core DSP. University of Washington Engineering Department, Seattle, 2011. http://www.ti.com/lit/wp/sprabn9/sprabn9.pdf. Accessed on June 5, 2013.

Orenstein

: Hard decisions—ultrasound elastography seeks to help characterize breast lesions and, more recently, throughout the body. Radiology Today 2011;12(4):26. http://www.radiologytoday.net/archive/rt0411p26.shtml#sthash.sgBcFdZ4.dpuf. Accessed on June 5, 2013.

Vinnicombe

Planche

: Breast imaging in the new era. Cancer Imaging 2004;4(2):39–50.

Brennan

Dershaw

Ginsberg

: Advances in oncologic imaging update on 5 common cancers. Ca Cancer J Clin 2012;62:364–393.

Falou

Sadeghi-Naini

Prematilake

: Evaluation of neoadjuvant chemotherapy response in women with locally advanced breast cancer using ultrasound elastography. Transl Oncol 2013;6:17–24.

Wallace

Gill

: EUS elastography for pancreatic mass lesions: between image and FNA? Gastrointest Endosc 2008;68(6):1095–1097.

Vilmann

Gorunescu

Suaftoiu

: Neural network analysis of dynamic sequences of EUS elastography used for the differential diagnosis of chronic pancreatitis and pancreatic cancer. Gastrointest Endosc 2008;68:1086–1094.

10.

Martegani

Di Cioccio

Baldassarre

Giuseppetti

: Elastosonography in the diagnosis of the nodular breast lesions: preliminary report. Radiol Med 2005;110:69–76.

11.

Destounis

Lackey

Svensson

: Evaluation of breast lesions using sonographic elasticity imaging: a multicenter trial. J Ultrasound Med 2012;31:281–287.

12.

Moon

Choi

Cho

Jang

Kim

Chung

: Differentiation of benign from malignant nonpalpable breast masses: a comparison of computer-assisted quantification and visual assessment of lesion stiffness with the use of sonographic elastography. Acta Radiol 2010;51(1):9–14.

13.

Afdhal

Cohen

: Ultrasound-based hepatic elastography: origins, limitations, and applications. J Clin Gastroenterol 2010;44(9):637–645.

14.

Moon

Cho

Kim

Chang

: Breast mass evaluation: factors influencing the quality of US elastography. Radiology 2011;259(1):59–64.

15.

Kemp

Hodge

: Transient elastography: the bigger we are, the harder to scan. J Gastroenterol Hepatol 2011;26:300–305.

16.

Săftoiu

Vilmann

Gorunescu

: Accuracy of endoscopic ultrasound elastography used for differential diagnosis of focal pancreatic masses: a multicenter study. Endoscopy 2011;43(7):596–603.

17.

Hitachi Medical Systems Europe. Real time elastography. http://www.hitachi-medical-systems.eu/products-and-services/ultrasound/hitachi-real-time-tissue-elastography-hi-rte/clinical-applications.html#Further-examples-5. Accessed on June 5, 2013.

18.

Botelberge

Borie

Pesenti

: Endoscopic ultrasound elastography for evaluation of lymph nodes and pancreatic masses: a multicenter study. World J Gastroenterol 2009;15(13):1587–1593.

19.

Larino-Noia

Abdulkader

Forteza

Dominguez-Munoz

Iglesias-Garcia

: Quantitative endoscopic ultrasound elastography: an accurate method for the differentiation of solid pancreatic masses. Gastroenterology 2010;139(4):1172–1180.

20.

Gordon-Walker

Aucott

van Deemter

: Matrix stiffness modulates proliferation, chemotherapeutic response, and dormancy in hepatocellular carcinoma cells. Hepatology 2011;53(4):1192–1205.

21.

Sandulescu

Padureanu

Dumitrescu

: A pilot study of real time elastography in the differentiation of focal liver lesions. Current Health Sciences J 2012;38:1–4.

22.

Gauthier

Fernandez

Xie

: Ultrasound-based elastography: a novel approach to assess radio frequency ablation of liver masses performed with expandable ablation probes: a feasibility study. J Ultrasound Med 2008;27(6):935–946.

23.

Stamatia

Barr

Castaneda

Strang

Rubens

Ginat

: US elastography of breast and prostate lesions. Radiographics 2009;29:2007–2016.

24.

Magnoni

Giusti

Seveso

: Impact of real-time elastography versus systematic prostate biopsy method on cancer detection rate in men with a serum prostate-specific antigen between 2.5 and 10 mg/mL. ISRN Oncol 2013;2013:584672.

25.

Souchon

Rouvière

Gelet

Chapelon

Curiel

: Elastography for the follow-up of high-intensity focused ultrasound prostate cancer treatment: initial comparison with MRI. Ultrasound Med Biol 2005;31(11):1461–1468.

26.

Ueno

Tohno

Kamma

Takahashi

Shiina

Itoh

: Breast disease: clinical application of US elastography for diagnosis. Radiology 2006;239:341–350.

27.

Aiani

Baldassarre

Bulzacki

: Characterization of breast lesions with real-time sonoelastography: results from the Italian multicenter clinical trial. SKK01-09 RSNA Nov 29, 2006. Available from: http://rsna2006.rsna.org/rsna2006/V2006/conference/event_display.cfm?em_id=4430351.

28.

Luo

Feng

: Comparison of ultrasound elastography, mammography, and sonography in the diagnosis of solid breast lesions. J Ultrasound Med 2007;26:807–815.

29.

Bai

Cui

Wang

: Elastographic evaluation of the temporal formation of ethanol-induced hepatic lesions: preliminary in vitro results. J Ultrasound Med 2007;26:1191–1199.