First published online December 10, 2007
British Journal of Radiology (2008) 81, 113-119
© 2008 British Institute of Radiology
doi: 10.1259/bjr/25294912
Haemodynamic parameters of the hepatic artery and vein can detect liver metastases: assessment using contrast-enhanced ultrasound
J H ZHOU, MD
A H LI, MD
1
L H CAO, MD
2
H H JIANG, MD
1
L Z LIU, MD
1
X Q PEI, MD
1 and
F HAN, MD
1
Departments of 1 Ultrasound and 2 Anaesthesiology, Cancer Center, Sun Yat-Sen University, Guangzhou, People's Republic of China
Correspondence: Dr A H LI, Department of Ultrasound, Cancer Center, Sun Yat-Sen University, 651 Dongfeng Road East, Guangzhou, PR China 510060. E-mail: anhuali{at}hotmail.com
 |
Abstract
|
|---|
This study was to assess the changes in the haemodynamic parameters of the hepatic artery and vein in the diagnosis of liver metastases by contrast-enhanced ultrasound (CEUS). 52 patients with proven liver metastases (patient group) and 23 normal volunteers (control group) were recruited in this study. Each subject was administered an intravenous bolus injection of SonoVue (0.6 ml). The arrival time in the hepatic artery (AT HA), time to peak in the hepatic artery (TTP HA), peak intensity of the hepatic artery (PI HA), arrival time in the hepatic vein (AT HV), time to peak in the hepatic vein (TTP HV) and peak intensity of the hepatic vein (PI HV) were measured with the use of time–intensity curve software. The hepatic artery to vein transit time (HAVTT) was calculated as the difference between the arrival times in the hepatic artery and the hepatic vein. AT HA, TTP HA, AT HV, TTP HV and HAVTT in the patient group were significantly shorter than those of the control group (P<0.01). PI HA and PI HV in the patient group were significantly higher than those of the control group (P<0.01).These results suggest that CEUS assessment of changes in the haemodynamic parameters of the hepatic artery and vein help to diagnose liver metastases. This functional imaging technique may contribute to the early detection of micrometastases in the liver.
|
Introduction
|
|---|
The liver is the second most frequent site after the regional lymph nodes [1] for the spread of metastases of primary tumours to other organs because of two hepatic features, the dual blood supply via the portal vein and the hepatic artery, and because it is the biggest viscus of the human body. The presence of overt hepatic metastases is associated with poor prognosis and survival rates that do not usually exceed 2 years after initial diagnosis [2]. Therefore, early and accurate detection of liver metastases is important to determine the stages of malignancies and guide the treatment as well as to estimate the prognosis of patients.
Unfortunately, currently available imaging techniques such as ultrasound (US), CT and MRI cannot detect hepatic metastases early. The sensitivity of current imaging techniques for the detection of liver lesions smaller than 1 cm is in the region of 50% when surgery and intraoperative ultrasound are used as gold standards [3]. However, it is these microscopic liver metastases that account for early disease recurrence within the liver and failure of surgical cure. Functional imaging techniques, which detect occult liver metastases based on the changes in hepatic haemodynamics with metastatic seeding in the liver, were proposed to solve this problem and showed promising value in detecting liver metastases [4–6]. Leen et al [7, 8] developed a Doppler perfusion index (the ratio between hepatic arterial and total hepatic blood flow) that was the most sensitive method for detecting colorectal metastases and had a strong predictive value in identifying patients who were likely to relapse, but failed to gain wide acceptance because of the poor reproducibility [9]. Recently, measurement of hepatic transit time using an ultrasound contrast agent as a tracer has shown promising preliminary results [10, 11].
The purpose of our study was to investigate the value of measuring the changes in different haemodynamic parameters of the hepatic artery and vein in the diagnosis of hepatic metastases using an ultrasound contrast agent as a tracer and to analyse some of the relative changes.
|
Patients and methods
|
|---|
The control group consisted of 23 normal volunteers (19 men and 4 women; median age, 37 years; age range, 25–76 years), with no history of liver disease and malignancies. The volunteers were recruited from within the Ultrasound Department. The patient group consisted of 52 patients (32 men and 20 women; median age, 52 years; age range, 31–80 years) with confirmed and untreated hepatic metastases. The metastatic nature of the liver lesions was confirmed by liver biopsy or liver surgery in 24 patients. In the other 28 patients, hepatic metastases were diagnosed on the basis of characteristic features on images obtained with at least two modalities together with objective evidence of an increase in the size and number of the lesions in the 3–6 months prior to enrolment in the study. All the recruited patients had a known primary malignancy: nasopharyngeal in 13, colorectal in 11, oesophageal in 7, pulmonary in 7, mammary in 6, gastric in 2, renal in 2, uterine in 1, ovarian in 1, lymphatic in 1 and lingual in 1. All patients and volunteers were required to starve for at least 4 h prior to being examined. Patients and normal volunteers who were suspected of having alcoholic, viral, schistosomal or other diffuse hepatic disease were excluded from this study. After obtaining approval from the institutional research ethics committee, we received written informed consent from all subjects.
According to the arterial contrast ultrasound enhancement pattern of hepatic metastases, patients were divided into two groups: a hypovascular group and a hypervascular group. According to the size and number of hepatic metastases, patients were divided into four groups: group A, group B, group C and group D (Table 1
). According to the type of primary malignancies, patients were divided into five groups: group I (nasopharyngeal carcinoma), group II (colorectal carcinoma), group III (pulmonary carcinoma), group IV (mammary carcinoma) and group V (oesophageal carcinoma) (Table 2).
Sonography examinations were performed on an Acuson Sequoia 512 ultrasound system (Siemens, Mountain View, CA) using standard probes (4C1 convex transducer with frequencies 2.0–5.0 MHz). The ultrasound machine was equipped with cadence contrast pulse sequencing imaging (CPS) and time–intensity curve software (autotracking contrast quantification, ACQ). The CPS was used in "contrast only" mode. The ACQ was used in the "log" preset. A second-generation ultrasound contrast agent, SonoVue (Bracco, Italy), containing sulphur hexafluoride phospholipid-stabilized microbubbles, was used for this study. Two experienced sonologists (J.H.Z., A.H.L.) who were blinded to any other imaging data and clinical information performed all examinations.
The transducer was placed in the right intercostal space with the region of interest (ROI) over the porta hepatis and one of the hepatic veins (usually the right hepatic vein was studied and if impossible the middle hepatic vein) to simultaneously visualize the hepatic artery and the hepatic vein in an intercostal US section. The patients and volunteers were instructed to try their best to hold their breath or breathe shallowly during data acquisition. After entering the CPS mode (mechanical index, 0.66), SonoVue (0.6 ml) was administered as a bolus into the antecubital vein using a 21-gauge intravenous catheter, immediately followed by a 5 ml flush of 0.9% saline. Data acquisition was started after the injection of contrast agent. A continuous video clip of 40 s was stored in the machine. After activation of the ACQ software, two ROI, with size adapted to the vessel diameter, were placed on the hepatic artery and hepatic vein. Gentle breathing was compensated by the automatic movement correction, and if the automatic motion correction could not compensate completely for the interference of the subjects' breathing, manual correction was applied in these subjects to erase the interference of breathing. If a sudden change in breathing occurred during data acquisition and these data could not provide optimal traces even when manual correction was used, these studies were repeated with a bolus injection of 0.6 ml of SonoVue. The arrival time in the hepatic artery (AT HA), time to peak in the hepatic artery (TTP HA), peak intensity of the hepatic artery (PI HA), arrival time in the hepatic vein (AT HV), time to peak in the hepatic vein (TTP HV) and peak intensity of the hepatic vein (PI HV) were acquired automatically by means of time–intensity curve software. The arrival time was defined as the time when the signal was greater than the threshold of 110% of the baseline signal. The peak intensity was defined as the maximum increase in the signal produced by the injection of the contrast agent. The time to peak was defined as the time interval from the beginning of injection to the peak of the smoothed curve. The hepatic artery to vein transit time (HAVTT) was calculated as the time difference between arrival of microbubbles in the hepatic artery and the hepatic vein (Figure 1).
View larger version (53K):
[in this window]
[in a new window]
|
Figure 1. Illustration of the measurement of the haemodynamic parameters of the hepatic artery and vein(AT, arrival time; TTP, time to peak; PI, peak intensity). (a) A hepatic artery time–intensity curve. (b) Haemodynamic parameters of the hepatic artery obtained from the hepatic artery time–intensity curve. (c) A hepatic vein time–intensity curve. (d) Haemodynamic parameters of the hepatic vein obtained from the hepatic vein time–intensity curve.
|
|
In 42 patients, the second injection of SonoVue (2.4 ml) was given at least 10 min after the first (0.6 ml of SonoVue) to assess the blood supply of the hepatic metastases. The mechanical index was set at 0.16. Thorough examination of the whole liver was performed in conventional B mode and a colour Doppler energy image was used to evaluate the vascularity of each lesion before injection of SonoVue (2.4 ml). Solid lesions without apparent necrosis in conventional B mode and/or with rich vascularity in colour Doppler energy image were selected as marker lesions to characterize further in contrast-enhanced ultrasound. According to the classification of the arterial enhancement pattern of the lesions, metastases were divided into two classes: hypervascular metastases (outstandingly homogeneous enhancement) and hypovascular metastases (randomly distributed enhanced or non-enhanced regions within the lesion or peripheral enhancement).
All data were analysed using the SPSS program (version 11.0). Summary data are presented as means±standard deviation. The Kolmogorov–Smirnova test was applied to evaluate normal distribution. The Levene test was applied to evaluate the homogeneity of variance. A t-test on independent samples was used to determine the significant differences of measurement data between the two groups. One-way analysis of variance (ANOVA) was used to determine the significant differences among the groups. Statistical significance was defined as p<0.05.
|
Results
|
|---|
There was an age difference between the patient group and the control group (p<0.01). However, there was no age difference between the hypovascular group and the hypervascular group (p>0.05), in groups A, B, C and D (p>0.05), and among groups I, II, III, IV and V (p>0.05). In 50 patients, lesions were present in the right lobe of the liver, whereas lesions were present only in the left lobe in the remaining two patients. Diagnostic curves with a clearly identifiable start and peak were obtained in all subjects after manual correction (n = 30) and repeated injection of 0.6 ml of SonoVue (n = 5) was used. Mean values and standard deviations for the seven indices measured in the patient and control groups are shown in Table 3. Significant differences were shown in all seven indices between the patient group and the control group (p<0.01). To diagnose liver metastases, the area under the ROC (receiver operating characteristic) curve was 0.742 for AT HA, 0.815 for TTP HA, 0.691 for PI HA, 0.915 for AT HV, 0.749 for TTP HV, 0.741 for PI HV and 0.957 for HAVTT. Sensitivity, specificity, positive and negative predictive values of using an AT HV value of <20 s to denote liver metastases were 85%, 83%, 92% and 70%, respectively. Sensitivity, specificity, positive and negative predictive values of using an HAVTT value of <9 s to denote liver metastases were 89%, 78%, 90% and 75%, respectively.
Mean values and standard deviations for the seven indices measured in the hypovascular group and hypervascular group are shown in Table 4. There were significant differences in AT HA and TTP HA between the hypovascular group and the hypervascular group (p<0.01). There were no significant differences in the mean AT HV, TTP HV, HAVTT, PI HA and PI HV between the two groups (p>0.05).
ANOVA did not show significant differences in AT HA, TTP HA, AT HV, TTP HV, HAVTT, PI HA and PI HV among groups A, B, C and D (p>0.05) (Table 5).
View this table:
[in this window]
[in a new window]
|
Table 5. Haemodynamic parameters of the hepatic artery and vein in the four study groups(by size and number of hepatic metastases)
|
|
There were no significant differences in AT HA, TTP HA, AT HV, TTP HV, HAVTT, PI HA and PI HV among groups I, II, III, IV and V (p>0.05) (Table 6).
View this table:
[in this window]
[in a new window]
|
Table 6. Haemodynamic parameters of the hepatic artery and vein in the five study groups(by type of primary malignancies)
|
|
|
Discussion
|
|---|
It is known from experimental animal studies [1, 12, 13] and clinical patient studies [4–8] that the presence of even small hepatic metastases may lead to subtle changes in hepatic haemodynamics that are similar to the haemodynamic changes in hepatic cirrhosis. Although the mechanism is not clear, these changes seem to originate from the early phase of liver metastases [1, 13]. In order to detect metastases early, researchers have been trying to monitor the changes in hepatic haemodynamics by using currently available imaging techniques. Recent developments in the US technique have introduced the possibility of measuring the changes in the hepatic transit times of US contrast agents caused by hepatic metastases, thus allowing to discriminate between patients with and without hepatic metastases. A short transit time of an ultrasound contrast agent, observed in patients with hepatic metastases [10, 11, 14] and patients with hepatic cirrhosis [15–17] provides a non-invasive index for hepatic metastases and hepatic cirrhosis. Besides hepatic transit time, several haemodynamic parameters of the hepatic artery and vein were analysed in this study. Moreover, the blood supply, the size and number of liver metastases and primary malignancies were considered for the first time.
This clinical study showed that AT HA and TTP HA, AT HV and TTP HV, and the HAVTT were significantly shorter in patients with liver metastases than in normal volunteers, as well as higher values of PI HA and of the PI HV. Among these seven indices, the differences of AT HV and HAVTT between the patient group and the control group are the most significant. In this clinical study, an AT HV value of <20 s to denote liver metastases is 85% sensitive and 83% specific, whereas an HAVTT value of <9 s to denote liver metastases is 89% sensitive and 78% specific. The thresholds of 20 s (AT HV) and 9 s (HAVTT) were chosen because they could provide the best sensitivity, specificity, positive predictive values and negative predictive values to discriminate between patients with and without metastases.
Although the haemodynamic changes associated with liver metastases have not been fully explained yet, the following mechanisms are likely to contribute to these differences. It is known that angiogenesis is essential to the growth of all tumours, whether primary or secondary, since growth beyond 1–2 mm3 requires adequate blood supply [18]. Lin et al [19] found that the arterial supply was dominant in most human liver metastases. Portal connections may exit veins but are generally draining veins [20]. Therefore, angiogenic changes due to hepatic metastases will lead to increased arterial perfusion. It is well known that a normal liver receives about 30% of its blood supply via the hepatic artery and the remainder through the portal vein, but, when metastatic seeding in the liver occurs, the normal percentage of blood supply changes. Previous studies have shown a significant increase in the ratio of hepatic arterial to total liver blood flow in a group of patients with liver metastases, which would lead to the arterialization of the liver blood supply. The hepatic perfusion index [4, 5] and the Doppler perfusion index [6–8] have been proposed as a measurement of this arterialization of liver blood supply to predict the presence of liver metastases. Moreover, arteriovenous shunts [21] in the liver and some humoral factors that may lead to the changes in extrahepatic haemodynamics are involved.
Arterialization of the liver blood supply obviously increases the amount of microbubbles that reach the liver during the early arterial phase rather than during the later portal venous phase of liver first pass. Arteriovenous shunts can decrease the time of microbubbles passing through the liver, since a significant amount of microbubbles bypass the hepatic sinusoids. A significantly shorter AT HA, TTP HA, AT HV and TTP HV in patients with liver metastases reflects these hepatic haemodynamic changes associated with liver metastases. The peak intensity is linearly proportional to the concentration of the contrast agent present in the vessel [22]. The increase in the peak intensity showed that the maximum microbubble concentration in the hepatic artery and vein was significantly higher in patients with metastases than in normal subjects. The arterialization of the liver blood supply would lead to the increase in the maximum microbubble concentration in the hepatic artery. SonoVue is generally regarded as a pure vascular agent with no tissue tropism, ensuring a certain duration of blood pool enhancement [23, 24]. Normally, the microbubbles have to traverse three microvascular beds (the capillaries of the lungs, the gut and the hepatic sinusoids) before reaching the hepatic vein [25]. The microbubbles will gradually dissolve in plasma and each passage of capillary beds will accelerate this process. Therefore, bypassing the capillaries of the gut via the hepatic artery and the hepatic sinusoids via the arteriovenous shunts as well as the shorter TTP HV will significantly decrease microbubble destruction, and thus lead to the increase in the maximum microbubble concentration in the hepatic vein in patients with metastases. However, patients (n = 2) with lesions remote from the vein sampled also showed pathological AT HA, TTP HA, AT HV, TTP HV, HAVTT, PI HA and PI HV. This phenomenon had been noted in Bernatik's study [11]. It is possible that there were occult metastases in the drainage segment of the vein sampled or some humoral factors may be involved.
Because patients' haemodynamic conditions will affect AT HV, Bernatik et al [11] introduced HAVTT to describe the real hepatic transit time that is far less dependent on patients' haemodynamic conditions. In people with liver metastases, the HAVTT found in our study (5.8±2.2 s) was similar to that obtained by Bernatik et al [11] (6.6±1.8 s), whereas in people without liver metastases the HAVTT was shorter in our study (11.4±2.6 s) than in the study by Bernatik et al [11] (15.0±2.0 s). The differences may result from the use of a different contrast agent (Optison was used in the Bernatik study). Different contrast agents have different chemical properties and different pharmacokinetics in the liver. However, using the same contrast agent (SonoVue), AT HV in volunteers in the study by Lim et al [26] was 29.4±6.9 s compared with 23.7±4.14 s in our study. The differences can probably be attributed to the different methods to record the arrival of contrast agent in the hepatic vein. The brightness used in our study may be more sensitive than the Doppler signal intensity used in the study by Lim et al [26] in the assessment of the change of signal intensity caused by the contrast agent.
In this study, we could observe that AT HA and TTP HA were significantly shorter in the hypervascular group than in the hypovascular group, but there were no significant differences in AT HV, TTP HV, HAVTT, PI HA and PI HV between these two groups. The shorter AT HA and TTP HA in the hypervascular group could be explained as a higher degree of arterialization of the liver blood supply. Furthermore, ANOVA did not show significant differences in these seven variables among the four groups divided by size and number of hepatic metastases. The size and number of the hepatic metastases did not correlate with these seven variables. This result could not be explained by the arterialization of the liver blood supply and arteriovenous shunts, which suggests that some unknown humoral factors may be involved. Likewise, there were no significant differences in these seven variables among the five groups divided by type of primary tumour, which suggests that measuring the changes in the haemodynamic parameters of the hepatic artery and vein can be extensively applied to discriminate between patients with and without hepatic metastases.
This study had several limitations. There was an age difference between the patient and the control groups due to the difficulty of finding older volunteers who had no history of liver disease and malignancies for comparison with patients. This study lacked data on patients with occult metastases because it is difficult to detect occult metastases with currently available techniques. Further longitudinal studies are needed to evaluate this method on patients who appeared normal on imaging but subsequently showed metastases. The patient group included patients with several types of primary malignancies, which could confound our results since different tumour types may differ in many aspects, such as the degree of arterialization and the amount of arteriovenous shunts. It would have been more helpful to have larger numbers of patients in the various subgroups.
A high MI (mechanical index) (0.66) was used in this study, which may result in bubble destruction. However, a low MI (0.1) would make it difficult to clearly discern the lumens of the hepatic vein from the surrounding hepatic tissue, and the bubble in the surrounding hepatic sinusoids would interfere with the measurement of the haemodynamic parameters of the hepatic vein. We did not study the relationship between the segmental position of the metastases and the indices because the hepatic artery and the hepatic vein must be visualized simultaneously in an intercostal US section, so we could not choose the hepatic vein according to the segmental position of the metastases and could not study all three hepatic veins. Finally, we did not perform a specialized study to assess reproducibility. Future studies are under way to address all these limitations.
|
Conclusions
|
|---|
Measuring the changes in the haemodynamic parameters of the hepatic artery and vein by contrast-enhanced ultrasound provides a non-invasive parameter for hepatic metastases. Yarmenitis et al [1] found that significant haemodynamic changes had happened in the liver without the presence of fully developed and well-vascularized metastatic infiltrates, so this functional imaging technique may have the potential to provide a valuable method for early detection of micrometastases in the liver. This needs further longitudinal studies to be confirmed in the future.
Received for publication September 23, 2006.
Revision received March 23, 2007.
Accepted for publication April 5, 2007.
|
References
|
|---|
- Yarmenitis SD, Kalogeropoulou CP, Hatjikondi O, Ravazoula P, Petsas T, Siamblis D, et al. An experimental approach of the Doppler perfusion index of the liver in detecting occult hepatic metastases: histological findings related to the haemodynamic measurements in Wistar rats. Eur Radiol 2000;10:417–24.[CrossRef][Medline]
- Oxley EM, Eills H. Prognosis of carcinoma of large bowel in the presence of liver metastases. Br J Surg 1969;56:149–52.[Medline]
- Robinson PJA. Imaging liver metastases: current limitations and future prospects. Br J Radiol 2000;73:234–41.[Abstract]
- Leveson SH, Wiggins PA, Giles GR, Parkin A, Robinson PJ. Deranged liver blood flow patterns in the detection of liver metastases. Br J Surg 1985;72:128–30.[Medline]
- Totman JJ, O'Gorman RL, Kane PA, Karani JB. Comparison of the hepatic perfusion index measured with gadolinium-enhanced volumetric MRI in controls and in patients with colorectal cancer. Br J Radiol 2005;78:105–9.[Abstract/Free Full Text]
- Leen E, Goldberg JA, Robertson J, Angerson WJ, Sutherland GR, Cooke TG, et al. Early detection of occult colorectal hepatic metastases using duplex colour Doppler sonography. Br J Surg 1993;80:1249–51.[Medline]
- Leen E, Angerson WJ, Wotherspoon H, Moule B, Cook TG, McArdle CS. Detection of colorectal liver metastases: comparison of laparotomy, CT, US, and Doppler perfusion index and evaluation of postoperative follow-up results. Radiology 1995;195:113–16.[Abstract/Free Full Text]
- Leen E, Angerson WG, Cooke TG, McArdle CS. Prognostic power of doppler perfusion index in colorectal cancer correlation with survival. Ann Surg 1996;223:199–203.[CrossRef][Medline]
- Roumen RM, Scheltinga MR, Slooter GD, van der Linden AW. Doppler perfusion index fails to predict the presence of occult hepatic colorectal metastases. Eur J Surg Oncol 2005;31:521–7.[CrossRef][Medline]
- Blomley MJK, Albrecht T, Cosgrove DO, Jayaram V, Eckersley RJ, Patel N, et al. Liver vascular transit time analyzed with dynamic hepatic venography with bolus injections of an US contrast agent: early experience in 7 patients with metastases. Radiology 1998;209:862–6.[Abstract/Free Full Text]
- Bernatik T, Becker D, Neureiter D, Hansler J, Frieser M, Schaber S, et al. Hepatic transit time of an echo enhancer: an indicator of metastatic spread to the liver. Eur J Gastroenterol Hepatol 2004;16:313–17.[CrossRef][Medline]
- Nott DM, Grime SJ, Yates J, Day DW, Baxter JN, Jenkins SA, et al. Changes in hepatic perfusion index during the development of experimental hepatic tumours. Br J Surg 1989;76:259–63.[Medline]
- Kruskal JB, Thomas P, Kane RA, Goldberg SN. Hepatic perfusion changes in mice livers with developing colorectal cancer metastases. Radiology 2004;231:482–90.[Abstract/Free Full Text]
- Bang N, Nielsen MB, Rasmussen AN, Osterhammel PA, Pedersen JF. Hepatic vein transit time of an ultrasound contrast agent: simplified procedure using pulse inversion imaging. Br J Radiol 2001;74:752–5.[Abstract/Free Full Text]
- Albrecht T, Blomley MJK, Cosgrove DO, Taylor-Robinson SD, Jayaram V, Eckersley R, et al. Non-invasive diagnosis of hepatic cirrhosis by transit time analysis of an ultrasound contrast agent. Lancet 1999;353:1579–83.[CrossRef][Medline]
- Blomley MJK, Lim AKP, Harvey CJ, Patel N, Eckersley RJ, Basilico R, et al. Liver microbubble transit time compared with histology and Child–Pugh score in diffuse liver disease: A cross-sectional study. Gut 2003;52:1188–93.[Abstract/Free Full Text]
- Lim AKP, Taylor-Robinson SD, Patel N, Eckersley RJ, Goldin RD, Hamilton G, et al. Hepatic vein transit times using a microbubble agent can predict disease severity non-invasively in patients with hepatitis C. Gut 2005;54:128–33.[Abstract/Free Full Text]
- Ellis LM, Fidler IJ. Angiogenesis and metastasis. Eur J Cancer 1996;32A:2451–60.[CrossRef]
- Lin G, Lunderquist A, Hagerstrand I, Boijsen E. Postmortem examination of the blood supply and vascular pattern of small liver metastases in man. Surgery 1984;96:517–26.[Medline]
- Ridge JA, Bading JR, Gelbard AS, Benua RS, Daly JM. Perfusion of colorectal hepatic metastases: relative distribution of flow from the hepatic artery and portal vein. Cancer 1987;59:1547–53.[CrossRef][Medline]
- Goldberg JA, Thomson JAK, Mccurrach G, Anderson JH, Willmott N, Bessent RG, et al. Arteriovenous shunting in patients with colorectal liver metastases. Br J Cancer 1991;63:466–8.[Medline]
- Yamada S, Komuro K, Mikami T, Kudo N, Onozuka H, Goto K, et al. Novel quantitative assessment of myocardial perfusion by harmonic power Doppler imaging during myocardial contrast echocardiography. Heart 2005;91:183–8.[Abstract/Free Full Text]
- Harvey CJ, Blomley MJK, Eckersley RJ, Cosgrove DO. Developments in ultrasound contrast media. Eur Radiol 2001;11:675–89.[CrossRef][Medline]
- Correas JM, Bridal L, Lesavre A, Mejean A, Claudon M, Helenon O. Ultrasound contrast agents: properties, principles of action, tolerance, and artifacts. Eur Radiol 2001;11:1316–28.[CrossRef][Medline]
- Cosgrove D, Eckersley R, Blomley M, Harvey C. Quantification of blood flow. Eur Radiol 2001;11:1338–44.[CrossRef][Medline]
- Lim AKP, Patel N, Eckersley RJ, Goldin RD, Thomas HC, Cosgrove DO, et al. Hepatic vein transit time of SonoVue: a comparative study with Levovist. Radiology 2006;240:130–5.[Abstract/Free Full Text]
Top
Abstract
Introduction
