| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology (L.A., R.M., E.D., B.E.V.B.) and the Laboratory of Gastroenterology (Y.H.), Université Catholique de Louvain, St-Luc University Hospital, Ave Hippocrate 10, B-1200 Brussels, Belgium; and the Center of Biostatistics and Medical Documentation, Université Catholique de Louvain, Mont-Godinne University Hospital, Yvoir, Belgium (J.J.). Received September 3, 2002; revision requested November 6; final revision received March 13, 2003; accepted April 16. Supported by grant 3.4578.00 from Fonds National de la Recherche Scientifique, Belgium. L.A. is a research fellow of the Fonds National de la Recherche Scientifique. Address correspondence to L.A. (e-mail: laurence.annet@clin.ucl.ac.be).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Forty-six patients referred for measurements of portal venous pressure (three with normal liver, 12 with chronic hepatitis, and 31 with cirrhosis [10 with Child-Pugh class A cirrhosis; 13 with class B cirrhosis; and eight with class C cirrhosis]) were included in the study. Apparent liver perfusion, apparent arterial and portal perfusion, portal fraction, distribution volume, and mean transit time were measured with dynamic contrast material–enhanced MR imaging. Portal velocity, portal flow, congestion index, right hepatic artery resistance index, and modified hepatic index were measured with Doppler US. Results in patients with cirrhosis and those without cirrhosis were compared with the Wilcoxon rank sum test. Correlations were assessed with Spearman rank correlation coefficients.
RESULTS: With MR imaging, all flow parameters except distribution volume were significantly different between patients with and those without cirrhosis (P < .05). There was a significant correlation between all flow parameters measured with MR imaging and portal pressure (P < .02). Apparent arterial (P = .024) and portal (P < .001) perfusion, portal fraction (P < .001), and mean transit time (P = .004) were correlated with Child-Pugh class. Flow parameters measured with Doppler US did not differ significantly between patients with and those without cirrhosis. Only right hepatic arterial resistance (P < .007) and portal flow (P < .043) were weakly (r < 0.7) correlated with portal pressure. No Doppler US parameter was correlated with Child-Pugh class.
CONCLUSION: Hepatic flow parameters measured with MR imaging correlate with the severity of cirrhosis and portal hypertension. Doppler US parameters are only weakly correlated with portal pressure.
© RSNA, 2003
Index terms: Liver, blood supply, 761.12144 • Liver, cirrhosis, 761.288 • Liver, MR, 761.121412, 761.12143, 761.12144 • Liver, US, 761.12984 • Magnetic resonance (MR), perfusion study, 761.12144
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Doppler ultrasonography (US) has been proposed to assess the hemodynamic changes of liver cirrhosis (3,4). Several Doppler US parameters related to the flow in the portal vein and the hepatic artery have been used for this purpose. However, the value of Doppler US in the diagnosis and grading of portal hypertension and liver dysfunction remains limited. This is at least partially explained by the relatively high intra- and interobserver variability and observer dependency of Doppler US (1,5,6).
Compartmental analysis of intensity versus time curves for magnetic resonance (MR) images of the liver after injection of a gadolinium chelate is an alternative method to measure hemodynamic parameters. This MR imaging method has been described and validated recently (7). The objective of the present study was to determine the correlations between hemodynamic parameters of hepatic flow measured with Doppler US and MR imaging and the severity of cirrhosis and portal hypertension.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Five patients were excluded: Three had hepatocellular carcinomas, one had a heart valve prosthesis, and one had cerebral vascular aneurysmal clips. Ten patients refused to participate in the study. Forty-six patients (31 men, with a mean age of 55 years ± 12 and age range of 18–75 years; and 15 women, with a mean age of 57 years ± 12 and age range of 40–83 years; P > .5) formed the final study group. The subjects were studied after an overnight fast. The three parts of the study were performed consecutively. The portal pressure measurements and liver biopsy were performed first, followed by MR imaging and Doppler US, respectively. The average duration of portal pressure measurements and liver biopsy, MR imaging, and Doppler US was 30, 20, and 30 minutes, respectively. Because the patients fasted the whole night before the study, we think that the short additional time of fasting between MR imaging and Doppler US did not substantially affect the measurements of the hemodynamic parameters. In addition, the patients were asked not to stand up until the end of the study to avoid postural variations in portal flow.
Portal pressure was measured by an experienced hepatologist (Y.H.). After administration of local anesthesia to the neck, a 7-F balloon-tipped catheter (MediTech; Boston Scientific, Galway, Ireland) was introduced into the right jugular vein and guided with fluoroscopy into the right main hepatic vein. No additional sedation was administered. Portal pressure was measured as the difference between the occluded hepatic venous pressure and the free hepatic venous pressure (8). Liver biopsy was performed through the same right jugular vein approach. Hematoma at the puncture site occurred in three patients. Bleeding from the liver biopsy site was not observed at MR imaging or Doppler US. On the basis of the results of these biopsies, three patients had normal liver, 12 had chronic hepatitis without cirrhosis, and 31 had cirrhosis. The severity of cirrhosis was graded according to Child-Pugh classification (9). Ten patients had Child-Pugh class A cirrhosis, 13 had class B cirrhosis, and eight had class C cirrhosis. The cause of chronic liver disease was viral (associated with chronic viral hepatitis B or C) in 22 patients, alcoholic in 15 patients, autoimmune in two patients, drug-related in two patients, and mixed viral and alcoholic in two patients.
MR Imaging and Analysis
After measurements of portal pressure were obtained, MR perfusion studies were performed with a 1.5-T MR imager (Gyroscan Intera; Philips Medical Systems, Best, the Netherlands). A transverse section passing through the portal vein, aorta, and right hepatic lobe was obtained continuously with a T1-weighted fast spoiled gradient-echo sequence, preceded by a non–section-selective 90° pulse and a spoiler gradient (7). The body coil was used for signal reception. The following parameters were used: 6-mm section thickness, 5.4/1.3 (repetition time msec/echo time msec), one signal acquired, flip angle of 15°, 40-cm field of view with 65% rectangular field of view, 256 x 256 matrix, linear phase-encoding order, cardiac triggering, and respiratory tracking with a navigator prepulse positioned on the right hemidiaphragm.
One image was acquired per heartbeat, and a total of 120 images were acquired in each patient. A low-molecular-weight gadolinium chelate (gadoterate dimeglumine, Dotarem; Guerbet, Roissy, France) was injected at the beginning of the MR data acquisition. A low dose of 0.05 mmol per kilogram of body weight was used, and the contrast agent was injected with a mechanical MR injector (Spectris; Medrad, Indianola, Pa) at a rate of 3 mL/sec. Image analysis was performed on a Silicon Graphics O2 workstation (Silicon Graphics, Mountain View, Calif), as explained previously (7).
Briefly, signal intensity versus time curves were obtained by one investigator (L.A.). The reproducibility of the analysis has been reported previously (13). The largest possible regions of interest in the aorta, portal vein, and right hepatic lobe were drawn. Care was taken to exclude large vessels from the regions placed in the liver. The mean sizes of the regions of interest in the aorta, portal vein, and liver parenchyma were 179 mm2 ± 50 (SD), 102 mm2 ± 19, and 2,606 mm2 ± 252, respectively. The signal intensity versus time curves were converted into longitudinal relaxation rate (R1) versus time curves according to an in vitro calibration curve (7). Next, a linear relationship was assumed between R1 and contrast agent concentration, according to the following classical equation:
|
|
|
The apparent liver perfusion k1 was defined as k1a + k1p.
The distribution volume (percentage) of the contrast agent was calculated as
|
The mean transit time(s) of the contrast agent in the liver was calculated as 1/k2.
Doppler US
Examination of the liver with Doppler US was performed after MR imaging. Three radiologists (L.A., R.M., E.D.) performed this part of the study with a Sequoia 512 unit (Acuson, Mountain View, Calif) and a 4-MHz transducer. The three operators had 4, 7, and 14 years of experience in Doppler US, respectively. To reproduce real clinical practice and shorten the examination time, Doppler US was performed by only one operator for each patient. The operators agreed on common rules to perform the US measurements before the start of the study. The measurements in the portal vein were performed approximately halfway between the venous confluence and the portal bifurcation. The sampling size was equal to the vessel diameter, and the angle between the Doppler ultrasound beam and the longitudinal axis of the portal vein was 30°–60°. The diameter of the portal vein was the average of two perpendicular measurements of a cross-sectional sample of the portal vein. Each result was the mean of three measurements performed during suspended respiration. The mean portal velocity was calculated as the time-averaged maximum velocity multiplied by a factor of 0.5 (14).
Arterial measurements were performed in the right branch of the hepatic artery near the hepatic artery bifurcation, as recommended by Piscaglia et al (15). According to previous reports (14,16–18), the portal blood flow (in milliliters per minute) was calculated as the product of the mean portal velocity and the cross-sectional area of the portal vein, assuming a circular conformation of the portal vein; the congestion index of the portal vein (in centimeters times seconds) was calculated as the ratio of the sectional area and the mean flow velocity of the portal vein; the hepatic artery resistance index (percentage) was the ratio of 100 times the difference of peak systolic minus end diastolic velocity to peak systolic velocity; and the modified hepatic index (in centimeters per second) was the portal flow velocity divided by the hepatic artery resistance index. Dilatation of a paraumbilical vein greater than 3 mm was also noted at color Doppler US, but paraumbilical flow was not assessed in the present study, since it may be difficult to measure in small paraumbilical veins (19).
Statistical Analysis
Data are expressed as mean ± SD. The MR imaging and Doppler US results in patients with and those without cirrhosis were compared with the Wilcoxon rank sum test. The same test was used to compare the patients with and those without dilatation of the paraumbilical vein. Spearman rank correlation coefficients were used to assess the correlation between each of the MR imaging and US parameters and the portal pressure, as well as the correlation between the MR imaging and US parameters and the Child-Pugh class in patients with cirrhosis. Statistically significant correlations (P < .05) were considered substantial when the correlation coefficient (r value) was at least 0.700 and weak when it was less than 0.700 (20).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Another MR imaging parameter than correlates with the severity of cirrhosis and portal hypertension is mean transit time. This increase of mean transit time may be explained by the capillarization of fenestrated sinusoids and the deposition of collagen in the extravascular spaces that occur in cirrhosis (25,26). Small molecules, such as gadolinium chelates, are still able to pass through the modified sinusoids. However, their diffusion in the extravascular space is slowed (27,28).
In contrast to the observations made with MR imaging, the flow parameters calculated with Doppler US did not differ significantly between patients with and those without cirrhosis. Only weak correlations were observed between right hepatic artery resistance index and portal flow on one hand and portal pressure on the other. No correlations were observed between Doppler parameters and Child-Pugh classification. Some investigators have claimed in previous studies that Doppler US has a high accuracy in the diagnosis and grading of cirrhosis and portal hypertension (17,29,30). However, several problems remain when Doppler US is used for this purpose.
First, it is important to look at the population being studied. In most of the studies realized to date, investigators compared the flow parameters between healthy subjects and patients with cirrhosis (3,14,17,29,30). Few studies included patients with chronic liver disease without cirrhosis, as we did. Disturbances of hepatic flow parameters are already observed in hepatic fibrosis before cirrhosis develops (4,18). For example, Piscaglia et al were unable to show a significant difference of hepatic artery resistance index between patients with cirrhosis and age-matched patients with chronic hepatitis (15).
Second, assessment of the portal venous flow and flow velocity can provide misleading results if a paraumbilical vein is dilated, which allows underestimation of the degree of portal hypertension (19). This is also shown in our study: Patients with a dilated paraumbilical vein had higher portal pressure than did patients without dilatation, but they did not have decreased portal flow or flow velocity at Doppler US. It has been recommended to exclude the measurements of portal venous flow velocity in patients with a dilated paraumbilical vein or to calculate the effective portal flow as the portal flow minus the paraumbilical flow (17). We, like others (14), did not exclude the velocity measurements in patients with dilated paraumbilical veins because they represent a large fraction of the patients with cirrhosis (32% in our study). We also did not attempt to measure the paraumbilical flow because it is difficult to perform these measurements in patients with a small paraumbilical vein.
The third problem with Doppler US measurements of the portal vein is interobserver variability, even if cooperative training programs are able to reduce this variability (5). These limitations explain that the use of Doppler US in patients with cirrhosis remains controversial. Only weak (r < 0.700) correlations have been demonstrated between Doppler US parameters and the severity of cirrhosis and portal hypertension (14,16,29). Because of the limitations of simple Doppler US parameters, new composite indexes, such as the modified hepatic index, have been evaluated (18,31). However, the reliability of these composite indexes is not yet proved.
Our results show that measurement of hepatic flow parameters with MR imaging may overcome some of the limitations of Doppler US measurement. We think that the better correlations observed between MR imaging parameters and disease severity may be explained by several factors. Portal perfusion (in milliliters per minute per 100 mL) is measured with MR imaging in the right hepatic lobe. This measurement quantifies the amount of venous blood arriving in 100 mL of liver parenchyma. This MR imaging measurement is thus not influenced by the flow in the paraumbilical vein and the volume of the liver in contrast to portal flow (in milliliters per minute) measured with Doppler US in the portal vein. Similarly, arterial perfusion can be measured with MR imaging, but flow in the hepatic artery is difficult to measure with Doppler US because of the small arterial caliber (32). The reproducibility of the MR imaging perfusion and the Doppler US measurements was not assessed directly in the present clinical study. However, these results of reproducibility have been published previously. The interobserver variations in flow measurements obtained with the model that we use for MR imaging are small (13). Interobserver intraclass correlation coefficients above 0.9 have been reported in a previous study. This contrasts with the known large interobserver variability in Doppler US parameters. The interobserver intraclass correlation coefficient for portal venous flow was only 0.49 in the study of Iwao et al (33). The intraobserver variability of Doppler US is less important. Therefore, the use of Doppler US has mainly been recommended for the follow-up of portal hypertension and for the assessment of therapeutic interventions (34–36).
Our study is limited by the fact that the flow parameters that were measured with MR imaging and Doppler US were not exactly the same. Measurements of portal blood flow with a phase-contrast sequence at MR imaging have been performed previously (36,37). This method has the same limitations as Doppler US —namely, the absence of measurements of arterial flow and errors in measurements of portal flow and velocity when the paraumbilical vein is dilated. Another limitation of our study is that only simple correlations were performed. Multiple regression was not performed because of the interactions between the parameters. For example, portal velocity, portal flow, congestion index, and modified hepatic index derive from the same measurements. Similarly, all MR imaging parameters derive from k1a, k1p, and k2.
In conclusion, hepatic flow parameters measured with MR imaging differed between patients with and those without cirrhosis. Substantial correlations were observed between MR imaging flow parameters and the severity of cirrhosis and portal hypertension. Only weak correlations were seen between Doppler US flow parameters and portal hypertension. Therefore, the measurement of flow parameters with MR imaging in patients with cirrhosis and portal hypertension may be potentially useful to determine the severity of disease and its prognosis.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Congestion index
of the portal vein. AJR Am J Roentgenol 1986; 146:735-739.This article has been cited by other articles:
|
|
 |
|
  T. S. Koh, C. H. Thng, P. S. Lee, S. Hartono, H. Rumpel, B. C. Goh, and S. Bisdas Hepatic Metastases: In Vivo Assessment of Perfusion Parameters at Dynamic Contrast-enhanced MR Imaging with Dual-Input Two-Compartment Tracer Kinetics Model Radiology, August 11, 2008; (2008) 2483071958. [Abstract] [Full Text]
|
|
|
|
|
|
H. Kim, C. J. Booth, A. B. Pinus, P. Chen, A. Lee, M. Qiu, M. Whitlock, P. S. Murphy, and R. T. Constable Induced Hepatic Fibrosis in Rats: Hepatic Steatosis, Macromolecule Content, Perfusion Parameters, and Their Correlations--Preliminary MR Imaging in Rats Radiology, June 1, 2008; 247(3): 696 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Bali, T. Metens, V. Denolin, V. De Maertelaer, J. Deviere, and C. Matos Pancreatic Perfusion: Noninvasive Quantitative Assessment with Dynamic Contrast-enhanced MR Imaging without and with Secretin Stimulation in Healthy Volunteers--Initial Results Radiology, April 1, 2008; 247(1): 115 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Hagiwara, H. Rusinek, V. S. Lee, M. Losada, M. A. Bannan, G. A. Krinsky, and B. Taouli Advanced Liver Fibrosis: Diagnosis with 3D Whole-Liver Perfusion MR Imaging--Initial Experience Radiology, March 1, 2008; 246(3): 926 - 934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zhang, Y.-Y. Duan, J.-M. Li, and J.-K. Yin Hemodynamic Features of Doppler Ultrasonography in Patients With Portal Hypertension: Intraoperative Direct Measurement of Portal Pressure in the Portal Venous System J. Ultrasound Med., December 1, 2007; 26(12): 1689 - 1696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Huwart, C. Sempoux, N. Salameh, J. Jamart, L. Annet, R. Sinkus, F. Peeters, L. C. ter Beek, Y. Horsmans, and B. E. Van Beers Liver Fibrosis: Noninvasive Assessment with MR Elastography versus Aspartate Aminotransferase to-Platelet Ratio Index Radiology, November 1, 2007; 245(2): 458 - 466. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Taouli, A. J. Tolia, M. Losada, J. S. Babb, E. S. Chan, M. A. Bannan, and H. Tobias Diffusion-Weighted MRI for Quantification of Liver Fibrosis: Preliminary Experience Am. J. Roentgenol., October 1, 2007; 189(4): 799 - 806. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. A. Aguirre, C. A. Behling, E. Alpert, T. I. Hassanein, and C. B. Sirlin Liver Fibrosis: Noninvasive Diagnosis with Double Contrast Material-enhanced MR Imaging. Radiology, May 1, 2006; 239(2): 425 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M Pinzani Non-invasive evaluation of hepatic fibrosis: don't count your chickens before they're hatched Gut, March 1, 2006; 55(3): 310 - 312. [Full Text] [PDF]
|
|
|
|
|
|
A. Haktanir, B. S. Cihan, C. Celenk, and S. Cihan Value of Doppler Sonography in Assessing the Progression of Chronic Viral Hepatitis and in the Diagnosis and Grading of Cirrhosis J. Ultrasound Med., March 1, 2005; 24(3): 311 - 321. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. V. Pandharipande, G. A. Krinsky, H. Rusinek, and V. S. Lee Perfusion Imaging of the Liver: Current Challenges and Future Goals Radiology, March 1, 2005; 234(3): 661 - 673. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
