| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, Glasgow Royal Infirmary, Alexandra Parade, Glasgow,
Scotland G31 2ER.
2 Department of Surgery, Glasgow Royal Infirmary, Glasgow, Scotland.
Received January 27, 2005;
accepted after revision August 15, 2005.
Address correspondence to E. Leen
(gpda01{at}udcf.ac.uk).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. This study included 127 patients with 82 malignant and 52 benign lesions in the liver. Contrast-enhanced sonography was performed using nonlinear imaging modes at low mechanical index (0.1-0.3) to enable real-time visualization of arterial, portal, and late-phase enhancement. Digital recordings of unenhanced sonography and contrast-enhanced sonography were reviewed by on-site investigators and two off-site blinded interpreters. The final diagnosis was based on consensus interpreting of all examinations by another two expert observers with access to CT, MRI, and histologic data; the diagnostic accuracy of contrast-enhanced sonography in identifying the lesion as benign, malignant, or indeterminate and as actual tumor type was compared with baseline sonography.
RESULTS. For on-site investigators, contrast-enhanced sonography reduced the number of indeterminate diagnoses by 67% and improved the sensitivity and specificity to 90.2% and 80.8%, respectively (p < 0.001). For off-site interpreters, contrast-enhanced sonography reduced the number of indeterminate diagnoses by 51-56% (p < 0.001); significantly improved sensitivity and specificity to 90.8-95.4% and 83.7-89.8%, respectively (p < 0.001); eliminated observers' variability (kappa coefficient: 0.66-0.77); and showed no significant difference in all comparisons in the analysis of lesions measuring less than 1.5 cm, 1.5-2.5 cm, and all sizes combined. Contrast-enhanced sonography did not rely on availability of clinical history to enable the diagnoses, and it reduced the need for further imaging investigations 23.7% to 90.4%.
CONCLUSION. Contrast-enhanced sonography improves the characterization of focal liver lesions and may limit the need for further investigations.
Keywords: contrast-enhanced sonography • liver
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
These requirements are fulfilled with second-generation microbubble contrast agents such as Definity (Bristol-Myers Squibb), Sonazoid (Amersham Health), and SonoVue (Bracco) at low mechanical index imaging; SonoVue is widely available in Europe and contains sulfur hexafluoride gas, which optimally lends itself to low mechanical index imaging thereby enabling continuous real-time imaging in the nonlinear imaging modes [6-9]. After IV injection of a small volume (2.4 mL) of SonoVue, the hepatic arterial and portal venous enhancement of the healthy liver and persistent homogeneous parenchymal enhancement for up to 5 min can be observed in real time. This dynamic assessment of the differential temporal enhancement of the focal liver lesions compared with the healthy liver may improve the differentiation between benign and malignant lesions with contrast-enhanced sonography [10].
This study assesses the clinical value and potential impact of real-time SonoVue-enhanced sonography with the nonlinear imaging method in the characterization of focal liver lesions.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SonoVue Administration
SonoVue (2.4-4.8 mL) was administered IV in a peripheral vein at least once
for each of the 152 lesions found in the 127 patients followed by 5 mL of
saline flush. The mean ± SD number of boluses per patient was 2.6
± 0.8, equivalent to a mean ± SD total volume of 7.7 ±
3.2 mL (range, 2.4-16.8 mL), broken down as: 3.1% of the patients received one
bolus, 49.6% received two, 33.1% received three, 11.8% received four, and 2.4%
received five. Multiple bolus administrations were allowed to compensate for
failures/errors (failure to record, unenhanced sonography scanner crash, etc.)
or to enable the use of interval or intermittent imaging techniques in
addition to real-time imaging techniques. A 5-min interval was necessary
between boluses.
Sonography Investigations
Sonography scanners (HDI 5000, ATL, Philips Medical Systems; Elegra and
Sequoia, Siemens Medical Solutions; LOGIQ 7, GE Healthcare; and Technos
MyLab70, Esaote) equipped with nonlinear imaging capabilities were used at
various centers. Technical settings (mechanical index, frame rate, and focal
zone) were optimized to obtain images of the best quality. All patients
underwent unenhanced hepatic sonography using the fundamental, color/power
Doppler techniques (unenhanced sonography) (no baseline harmonic imaging), and
contrast-enhanced sonography in gray-scale was performed with nonlinear
imaging techniques using continuous real-time imaging techniques during the
hepatic arterial (15-25 sec from the time of the injection), portal venous
(25-100 sec), and late parenchymal (100-300 sec) phases. Interval or
intermittent imaging (scanning with 5-8-sec intervals of no scanning [freeze]
at the 0.3 mechanical index in the same plane) was performed at the peak of
the arterial enhancement for some lesions on the second bolus administration
of SonoVue. All sonographic examinations were recorded on super-VHS (SVHS)
videotapes.
The location and size of the lesion and the presence and distribution of color/power Doppler unenhanced sonography flow signals (homogeneous/basket/peripheral/centripetal-flow/centrifugal-flow) within the lesion were assessed on unenhanced sonography. The vascularity and pattern of SonoVue enhancement of the lesion compared with the adjacent liver parenchyma during the hepatic arterial, portal venous, and late phases were evaluated for the contrast-enhanced sonography diagnosis. The criteria for contrast-enhanced sonography for categorizing lesions as benign or malignant are analogous to those of other imaging methods.
For the on-site investigators (investigators from each recruiting center not blinded to their own patients), diagnoses in terms of the nature (malignant, benign, or indeterminate) and histologic type (metastases, hepatocellular carcinomas, hemangiomas, etc.) of the lesions were made based on the unenhanced fundamental color/power Doppler unenhanced sonography and the SonoVue-enhanced sonography. Contrast-enhanced sonography was compared with unenhanced sonography using CT, MRI, biopsy, and any other relevant clinical and biochemical markers as the reference standard.
The reference standard included optimized contrast-enhanced MRI, spiral-CT, and histologic and pathologic examinations scheduled within a week, either before or after the unenhanced sonography investigations. Contrast-enhanced CT or MRI was performed on 81.9% of the patients (62.2% had CT and 19.7% had MRI), and the remaining 18.1% had histologic diagnosis.
Blinded Interpretation
An off-site blinded interpretation study was also performed. After quality
control assessment, recordings from digital videotape or SVHS tapes were
transferred to a Sony PC with a digital capture card using Sony DVgate Motion
software in a standardized videotape format. Patient and center identifying
data were masked, and an identifying code number was inserted such that each
patient had two digital videotape files, one for contrast-enhanced sonography
and one for unenhanced sonography. Although multiple bolus injections were
administered for some of the lesions, only one of the videotape clips (the
optimal) was selected for viewing by the two off-site interpreters.
Two off-site interpreters, radiologist A and gastroenterologist B—who were experienced in SonoVue-enhanced sonography for 4 and 12 months, respectively, and who were unaffiliated with the study centers and blinded to the patient history—reviewed the videotape clips of the unenhanced sonography for each patient separately from that containing the contrast-enhanced sonography videotape clip in computer-generated, randomized fashion. Subsequently, the review was repeated with the benefit of the clinical history separately for contrast-enhanced sonography and unenhanced sonography. Diagnoses in terms of the nature (malignant, benign, or indeterminate) and histologic type of the lesions were made. Both unenhanced sonography and contrast-enhanced sonography diagnoses were then compared against the final diagnosis.
Final Diagnosis
The final diagnosis was made after consensus interpretation of all the
unenhanced sonography and contrast-enhanced sonography videotapes by another
two expert observers (in addition to the two off-site blinded interpreters)
with more than 5 years experience in the field of contrast-enhanced sonography
of the liver who had access to all relevant clinical information including
biochemical markers, CT, MRI, and histologic results. Any disagreement between
the two expert observers led to withdrawal of the case, which occurred in 18
lesions in this study resulting in 134 lesions with a final diagnosis.
Potential Impact on Decision Making
For the two off-site interpreters, potential impact on clinical management
was assessed, with documentation of the interpreter's decision making in
procuring further CT, MRI, biopsy, or no investigation after the diagnoses
were made separately on unenhanced sonography and contrast-enhanced
sonography.
Statistical Analysis
Summarized descriptive statistics were provided for continuous variables,
and absolute and relative frequencies were calculated for categoric data. In
addition, 95% confidence intervals (CIs) were calculated for the main
variables. All statistical comparisons were performed as two-sided tests, and
significance was declared at p £ 0.05 level. The accuracy of
unenhanced sonography and contrast-enhanced sonography in the characterization
of focal liver lesions using the reference standard was estimated in terms of
the nature and tumor type of the lesion.
Accuracy with respect to the nature of the lesion was assessed by means of sensitivity and specificity. A 95% CI, using the Wilson formula, was calculated for sensitivity and specificity. Sensitivity was calculated as the percentage of true malignant lesions of the total number of malignant lesions; in the estimate of sensitivity, indeterminate lesions were classified as being wrongly diagnosed as benign, with the consequence of underestimating true-positives and overestimating false-negatives (benign + indeterminate). Specificity was calculated as the percentage of true benign lesions out of the total number of benign lesions; in the estimate of specificity, indeterminate lesions were considered as wrongly diagnosed as malignant, with the consequence of underestimating true-negatives and overestimating false-positives (malignant + indeterminate).
Changes between unenhanced sonography and contrast-enhanced sonography images in relation to the nature of the lesions were assessed by using cross tabulations and compared by applying the McNemar two-sided test.
Accuracy with respect to tumor type was assessed in terms of the percentage of agreement of the sonographic diagnoses with the final diagnoses.
Agreement between the two blinded interpreters was calculated by Cohen's kappa value; a value more than 0.4 represented fair agreement and more than 0.75 was considered excellent agreement.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Lesion Diameter
The mean ± SD minimum diameter of the lesions was 28.4 ± 20.3
mm on unenhanced sonography, 28.3 ± 20.5 mm on contrast-enhanced
sonography, and 28.2 ± 19.6 mm on the CT and MRI scans. The mean
± SD maximum diameter of the lesions was 35.6 ± 26.0 mm on
unenhanced sonography, 35.2 ± 26.0 mm on contrast-enhanced sonography,
and 35.4 ± 26.8 mm on the CT and MRI scans. There were no significant
differences among any of the three techniques.
Of the 134 lesions, 43 lesions measured less than 1.5 cm, and 34 lesions measured between 1.5 and 2.5 cm in diameter.
Diagnostic Accuracy
For on-site investigators, the percentage of indeterminate diagnoses (i.e.,
uncertain whether lesion is benign or malignant) was markedly reduced with
contrast-enhanced sonography from 67.9% to 10.4%. The reduction in the
percentage of indeterminate diagnoses was higher for malignant lesions (62.2%)
compared with that for benign lesions (50%) (p < 0.01).
For the off-site interpreters A and B, contrast-enhanced sonography reduced the percentage of indeterminate diagnoses from 58.8% and 53.7% to 1.5% and 6.6%, respectively. There was very good agreement between observers as to whether the scan was determinate or indeterminate after contrast-enhanced sonography (91.9% agreement, kappa -0.03, p = 0.7) compared with unenhanced sonography (58.1% agreement, kappa 0.15, p =0.08). In contrast-enhanced sonography, a high level of agreement would be expected by chance because both observers classed nearly all scans as determinate; the low values of kappa showed that the two observers did not agree on which scans were indeterminate, whether there were many (unenhanced sonography) or few (contrast-enhanced sonography).
The sensitivity, specificity, and accuracy of contrast-enhanced sonography and unenhanced sonography for on-site investigators and off-site interpreters A and B are summarized in Table 1. Sensitivity was calculated on the lesions classified as malignant by the final diagnosis (n = 82), specificity was calculated on the lesions classified as benign by the final diagnosis (n = 52), and accuracy was calculated as the percentage of lesions in concordance with the final diagnosis. Contrast-enhanced sonography increased the sensitivity, specificity, and accuracy of on-site investigators and off-site interpreters on the analysis of all lesions, lesions measuring less than 1.5 cm, and lesions measuring 1.5 to 2.5 cm. All differences were statistically significant except for interpreter B's sensitivity analysis of lesions less than 1.5 cm and interpreter A's specificity analysis of lesions less than 1.5 cm.
|
There was very good agreement between off-site interpreters A and B in determining malignant and benign lesions on contrast-enhanced sonography (83.1% agreement, kappa 0.66, p < 0.001) versus unenhanced sonography (57.4% agreement, kappa 0.23, p = 0.001). Although the interpreters frequently did not agree which lesions were indeterminate, they were better at agreeing which were benign and which were malignant, especially after SonoVue administration.
Tumor Type
The percentage of correctly diagnosed histologic types with respect to the
final diagnoses for on-site investigators and off-site interpreters A and B on
contrast-enhanced sonography and unenhanced sonography is summarized in
Table 2. Contrast-enhanced
sonography showed a statistically significant increase in the percentage of
correct diagnosis of tumor for both benign and malignant lesions (p
< 0.001).
|
Lesional Enhancement
Differences in the enhancement patterns observed related to tumor types
(Table 3), based on the mean of
the observations of the two off-site interpreters. For hemangiomas, peripheral
nodular/globular enhancement was seen in 34.8% during the arterial phase and
in 17.4% in the portal venous phase; progressively central filling-in was
observed in 52% over the arterial, portal, and late phases. Also, 95.8% of the
hemangiomas were of higher intensity and 4.2% were lower intensity compared
with the adjacent liver in the late phase.
|
Of the focal nodular hyperplasia (FNH) lesions, 81.8% enhanced brightly and rapidly during the arterial phase, and 72.7% remained predominantly brighter than the adjacent liver during the portal venous phase; in the late phases 54.5% were of higher intensity and 36.4% were isointense to the adjacent liver. A spoke-wheel appearance was present in 63.6% in the arterial phase, and a central scar was more evident in 10% on contrast-enhanced sonography.
Hepatocellular carcinomas (HCCs) showed varying degrees of enhancement during the arterial and portal phases—41.9% were homogeneous, 32.6% had a basketlike pattern, and 11.6% had an irregular combined with peripheral pattern. In the arterial and portal venous phases, 92.9% and 23.3%, respectively, of the HCCs appeared brighter relative to the adjacent liver; however, 4.3% and 12.8%, respectively, had relatively higher intensity and were isointense to the adjacent liver in the late phase.
Metastases were characterized by irregular rimlike enhancement in 61.3% during the arterial and portal venous phases, and 22.6% showed a relatively homogeneous enhancement in the arterial phase; 66.7% and 81.8% had lower intensity relative to the adjacent liver in the arterial and portal venous phases, respectively, and all lesions were of relatively lower intensity in the late phase. Rim enhancement remained present in 12.1% in the late phase.
Clinical History Impact
There was a significant increase in the percentage of diagnoses made on
unenhanced sonography after the availability of patients' clinical history
(6.6% vs 66.7%; p < 0.01). However, no difference was seen in the
percentage of diagnoses made on contrast-enhanced sonography after the
patients' clinical history became available (97.5% vs 97.6%).
Potential Impact on Decision Making
For off-site interpreter A, 65.9% and 34.1% of patients required additional
CT and MRI scans, respectively, after review of the unenhanced sonography
examination with the clinical history. After contrast-enhanced sonography, CT,
MRI, biopsy, and no further investigation were required in 50.4%, 25.9%, 6.8%,
and 16.9% of patients, respectively.
For off-site interpreter B, 47% and 53% of patients required additional CT and MRI scans, respectively, after review of the unenhanced sonography examination with the clinical history. After contrast-enhanced sonography, CT, MRI, biopsy, and no further investigation were required in 4.5%, 5.1%, 13.9%, and 76.5% of patients, respectively.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mechanism underlying the organ-specific phenomenon of the agent remains unknown. There is no evidence of phagocytosis of SonoVue by the reticuloendothelial cells. The persistence of the agent within the healthy liver may be a result of the very slow flow within the sinusoids. Indeed, there is some experimental evidence in rats that microbubble agents containing perfluorocarbon gas enhance the healthy liver by virtue of the slow flow within the sinusoids [11]. Malignant lesions such as metastases and HCCs are devoid of sinusoids and are almost exclusively fed by abnormal arterial channels associated with complex shunts leading to rapid contrast washout; this results in an absence of contrast accumulation relative to healthy liver in the late parenchymal phases (Figs. 1A and 1B). Indeed all metastases and 82.9% of the HCCs were hypointense (Figs. 2A, 2B and 2C). Although FNH lesions have a rich arterial supply, relative accumulation in the late phases occurs because they consist of healthy functioning hepatic tissue similar to adjacent healthy sinusoids with very slow washout; 54.5% and 36.4% of the FNH lesions were of higher intensity and isointensity, respectively, in the late phase (Figs. 3A, 3B, 3C and 3D). The slow blood flow within the hemangiomas, which consists of vascular spaces lined with endothelial cells, is well recognized; persistent and/or progressive accumulation of contrast in the late phases occurred in 95.8% of the cases in this study (Figs. 4A, 4B and 4C). Irrespective of the vascular enhancement pattern, based on the late phase alone, if the lesion is hyperintense, the probability of it being benign is more than 95%.
|
|
|
|
|
|
|
|
|
|
|
|
For the off-site interpreters, characterization of focal liver lesions with contrast-enhanced sonography was improved for sensitivity (90.8-95.4%), specificity (83.7-89.8%), and accuracy (90.4-91.2%) for all lesions. The magnitudes of improvement from unenhanced sonography were similar for lesions smaller than 1.5 cm and those measuring 1.5-2.5 cm in diameter. Recent single center and multicenter studies using Levovist with late-phase pulse inversion harmonic imaging have also shown improvement in the characterization of liver lesions and had similar accuracies ranging from 85% to 90% in the differentiation between malignant and benign liver lesions [4, 12, 13]. A later study using SonoVue shows improved overall diagnostic accuracies of 85-88% compared with unenhanced sonography (49-51%) [10, 14]. Results of triple-phase MRI with gadobenate dimeglumine studies show similar accuracies in improved characterization of focal liver lesions [15, 16]. Contrast-enhanced sonography also significantly improves the determination of the actual tumor type, in particular hemangiomas (from 36% to 84%), FNH lesions (from 30.8% to 76.9%), HCCs (from 32.6% to 87%), and metastases (from 19.3% to 87.1%). This technique can be readily implemented clinically after the identification of a lesion during sonography because it is quick, simple, and practical with a high level of accuracy for common lesions; the substantial reduction in the number of indeterminate diagnoses after contrast-enhanced sonography is also significant in that referrals to other costly diagnostic tests may become unnecessary, with implications on the workload and waiting times for patients.
Pathologic diagnosis was available in only 18.1% of patients and in 45.6% of lesions measuring less than 1.5 cm, which is an important limitation in this study; potential diagnostic errors may be introduced in the determination of actual tumor types of the lesions. Nevertheless, ethical and practical issues exist in carrying out biopsies in all cases, especially in suspected benign cases, and pathologic reports may be inaccurate if sampling errors occur, which may happen with small lesions. To circumvent some limitations, consensus interpreting of all sonograms with the benefit of CT, MRI, histology, and other clinical history results by two additional expert observers were used to provide a final diagnosis.
Differentiation between benign and malignant lesions when the lesions are small is usually difficult even on CT and MRI because they do not display enough characteristic features and biopsy can be very difficult [17]. On contrast-enhanced CT or MRI, the hepatic arterial phase and portal venous phase images of the lesions are obtained at single time frames, and partial voluming effects are limiting when lesions are small. Therefore, the full enhancement characteristics may be difficult to monitor; contrast-enhanced sonography has the advantage in allowing real-time visualization of the enhancement characteristics at frame rates of 10-20 Hz, which is far superior to MRI or CT. In patients without known cancer, most of these lesions are likely benign and are usually evaluated with serial follow-up imaging scans. In patients with known cancer where knowledge of stage and progression is required to determine prognosis and therapeutic management, the presence of a focal lesion carries an 11.6% risk of being malignant; accurate characterization of these small lesions is therefore important [18]. The analysis of small lesions in this study is particularly relevant because contrast-enhanced sonography significantly increased the accuracy in the characterization of these small lesions and may be used as an adjunct noninvasive imaging test for those with coincidental indeterminate lesions.
Evidence of the cost-effectiveness of contrast-enhanced sonography in radiology has not yet been published. Intuitively, the substantial reduction in the number of indeterminate diagnoses after contrast-enhanced sonography by the off-site interpreters might imply that it would reduce the need for further investigations with potential improvement in patient satisfaction as a result of eliminating waiting times for further testing. By improving the diagnostic accuracy of unenhanced sonography, the contrast agent's upfront cost (£39.99 [$70.27 U.S.] per vial in the United Kingdom, which would enable at least two 2.4-mL bolus injections) may be offset by the substantially lower rate of downstream resource use. It may also allow a more selective use of the technology and decrease duplicate testing in most patients because fewer would be referred for further examinations. However, results of the response to further investigations after contrast-enhanced sonography compared with unenhanced sonography separately revealed significant observer variability. Although the sensitivity, specificity, and accuracy of off-site interpreter A were excellent after contrast-enhanced sonography, the response to subsequent management decision making was surprising, with 83.1% of patients still being referred for further investigations. In contrast, off-site interpreter B had a substantial reduction in percentage (76.5%) of patients being referred for further investigations. Although substantial savings would be expected in the case of off-site interpreter B, the cost savings for off-site interpreter A may not be as large.
In conclusion, SonoVue-enhanced sonography improves the characterization of focal liver lesions and may limit the need for further investigations, with potential cost savings dependent on the experience and specialist training of the interpreter.
Acknowledgments
On behalf of the multicenter study group, we would like to acknowledge the
contributions of Dirk Becker, Stefano Gaiani, Emilio Quaia, Paolo Ricci,
Fulvio Stacul, Robert Steinbach, and Hans Peter Weskott.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  P. Ricci, V. Cantisani, M. D'Onofrio, D. Sahani, E. Pagliara, F. Calliada, E. Mehmet, K. Sanjeva, N. Faccioli, R. Pozzi-Mucelli, et al. Behavior of Hepatocellular Adenoma on Real-time Low-Mechanical Index Contrast-Enhanced Ultrasonography With a Second-Generation Contrast Agent J. Ultrasound Med., December 1, 2008; 27(12): 1719 - 1726. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. K. Kim, H.-J. Jang, P. N. Burns, J. Murphy-Lavallee, and S. R. Wilson Focal Nodular Hyperplasia and Hepatic Adenoma: Differentiation with Low-Mechanical-Index Contrast-Enhanced Sonography Am. J. Roentgenol., January 1, 2008; 190(1): 58 - 66. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Lanka, H.-J. Jang, T. K. Kim, P. N. Burns, and S. R. Wilson Impact of Contrast-Enhanced Ultrasonography in a Tertiary Clinical Practice J. Ultrasound Med., December 1, 2007; 26(12): 1703 - 1714. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Quaia, M. D'Onofrio, P. Cabassa, F. Vecchiato, S. Caffarri, F. Pittiani, K. M. Wittkowski, and M. A. Cova Diagnostic Value of Hepatocellular Nodule Vascularity After Microbubble Injection for Characterizing Malignancy in Patients with Cirrhosis Am. J. Roentgenol., December 1, 2007; 189(6): 1474 - 1483. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Murphy-Lavallee, H.-J. Jang, T. K. Kim, P. N. Burns, and S. R. Wilson Are Metastases Really Hypovascular in the Arterial Phase?: The Perspective Based on Contrast-Enhanced Ultrasonography J. Ultrasound Med., November 1, 2007; 26(11): 1545 - 1556. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Catalano, V. Migaleddu, E. Quaia, and G. Caruso Terminology for Contrast-Enhanced Sonography: A Practical Glossary J. Ultrasound Med., June 1, 2007; 26(6): 717 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. R. Wilson, H.-J. Jang, T. K. Kim, and P. N. Burns Diagnosis of Focal Liver Masses on Ultrasonography: Comparison of Unenhanced and Contrast-Enhanced Scans J. Ultrasound Med., June 1, 2007; 26(6): 775 - 787. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
