| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology and Institute of Radiation Medicine, Seoul National
University College of Medicine, 28, Yongon-dong, Chongno-gu, Seoul 110-744,
South Korea.
2 Department of General Surgery, Seoul National University College of Medicine,
Seoul, South Korea.
Received April 3, 2005;
accepted after revision September 18, 2005.
Address correspondence to J. M. Lee
(leejm{at}radcom.snu.ac.kr).
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Forty-four consecutive living donor candidates underwent liver MRI on a 1.5-T MR unit. The MR examination included in- and opposed-phase T1-weighted gradient-echo imaging, T2-weighted MR cholangiography, MR angiography (MRA) and parenchymal phase imaging after the administration of gadobenate dimeglumine, and 60-minute delayed T1-weighted MR cholangiography. Two abdominal radiologists analyzed the images regarding the depiction of the biliary duct anatomy and the hepatic vascular anatomy and for the presence of focal or diffuse liver disease. The findings were compared with intraoperative cholangiographic and surgical findings in 24 patients who underwent partial hepatectomy.
RESULTS. In the 24 patients who underwent liver harvesting, 10 had biliary anatomic variants confirmed by intraoperative cholangiography. T2-weighted MR cholangiography allowed a correct diagnosis in 75% (n = 18/24) and T1-weighted MR cholangiography in 79% (n = 19/24) of these patients. When we evaluated the bile duct anatomy using the combined findings of T2- and T1-weighted MR cholangiographic images, the diagnostic accuracy increased to 92% (n = 22/24), but the difference was not statistically significant (p > 0.05). MRA showed a diagnostic accuracy of 79% (n = 19/24) for the hepatic arterial anatomy, 100% (n = 24/24) for the portal venous anatomy, and 96% (n = 23/24) for the hepatic venous anatomy.
CONCLUSION. Gadobenate dimeglumine-enhanced MRI allows comprehensive assessment of the biliary and hepatic vascular systems and the hepatic parenchyma and can serve as the sole preoperative imaging test for living liver donor candidates.
Keywords: gadobenate dimeglumine • liver disease • liver donor • liver transplantation • MRI
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Complete preoperative imaging of a donor typically includes conventional sonography to evaluate the hepatic steatosis, digital subtraction angiography (DSA) or CT angiography (CTA) to depict the hepatic vascular system, and ERCP or intraoperative cholangiography to assess the biliary anatomy [3, 4]. This type of imaging workup using several different techniques is time consuming and expensive and is also very tiring for the donors.
Recently, there have been many literature reports indicating that contrast-enhanced MRI has the potential to simplify this complex process [1, 5, 6]. However, existing MR angiography (MRA) and cholangiography techniques have several limitations—that is, their limited temporal and spatial resolution compared with direct DSA and intraoperative cholangiography and defining the optimal contrast agent to obtain the best image quality [7, 8].
Based on previous studies, the use of gadobenate dimeglumine for liver MR examinations as a preoperative workup for a living donor transplantation seems to be attractive because of its twofold T1 relaxation effects and biliary excretion property [1, 9, 10]. The motivation for our study was from the possibility for gadobenate dimeglumine to increase the signal-to-noise ratio (SNR) of MRA due to its twofold T1 relaxation effect and its biliary excretion property to be used for 3D cholangiography.
The purpose of this study was to determine the clinical practicability and the diagnostic accuracy of gadobenate dimeglumine-enhanced MRI for the preoperative evaluation of living adult liver donors.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All imaging findings were analyzed prospectively in the 44 candidates before the transplantation surgeons assessed surgical candidacy. Finally, 24 candidates underwent partial hepatectomy, and the surgical and intraoperative cholangiographic results were used for comparison with the MRI findings. This study was performed in accordance with institutional review board guidelines, and written informed consent was obtained from all patients.
MR Examination
We obtained images on a 1.5-T MR scanner (Magnetom Vision, Siemens Medical
Solutions) using a maximum gradient strength of 25 mT/m with a torso
phased-array coil. A 220- to 340-cm field of view was used. Before the start
of the study, a 22-gauge IV catheter was placed in an antecubital or forearm
vein and was attached to an MR-compatible power injector (Spectris, Med-Rad).
Gadobenate dimeglumine (MultiHance, Bracco Imaging) was injected IV at a dose
of 0.1 mmol/kg of body weight. According to the method detailed by Earls et
al. [11], a timing image was
then obtained with a test dose of 1 mL of gadobenate dimeglumine followed by
20 mL of saline solution, both injected at a rate of 2 mL/s. This allowed
estimation of patient circulation time (bolus transit time from arm vein to
abdominal aorta) that was used to ensure acquisition of optimal arterial phase
images.
MR Sequences and Parameters
For evaluation of liver parenchyma, breath-hold axial T1-weighted in-phase
and opposed-phase 2D fast low-angle shot (FLASH) imaging (TR, 128
milliseconds; TE for opposed-phase imaging, 2.7 milliseconds; TE for in-phase
imaging, 5.3 milliseconds; flip angle, 70°; slice thickness, 7 mm; and
matrix, 131 x 256) and T2-weighted HASTE imaging (TR/TE, 4,400/90; flip
angle, 150°; slice thickness, 7 mm; and matrix, 131 x 256) were
performed.
For T2-weighted MR cholangiography, a multislice HASTE imaging sequence (infinite/95.0; flip angle, 150°; slice thickness, 3 mm with no interval gap; and matrix, 240 x 256) and oblique coronal heavily T2-weighted thick-slab turbo spin-echo (TSE) imaging sequence (2,800/1,100; flip angle, 150°; 60-mm-thick slice in coronal oblique planes [-45°, -25°, -15°, 0°, 15°, 25°, 45°] and 40-mm-thick slice in axial plane; and matrix, 240 x 256) were performed.
For MRA, a coronal 3D interpolated fat-suppressed spoiled gradient-echo sequence was performed before and after IV administration of gadobenate dimeglumine. The imaging parameters for the sequence were as follows: 4.6/1.8; flip angle, 25°; matrix, 140 x 256; field of view, 400-450 mm; slab thickness, 80 mm with 40 partitions interpolated to 80 for an effective slice thickness of 1 mm; and an intermittent fat-saturation pulse. The first data set after enhancement, which was timed to show the hepatic arteries using a test dose of gadobenate dimeglumine, was collected an average of 10 seconds after the injection. The second and third data sets for visualization of the portal and hepatic venous systems were collected 60 and 120 seconds after the injection, respectively. For evaluation of incidental focal liver lesions, a second 3D interpolated fat-suppressed spoiled gradient-echo sequence (4.5/1.9; flip angle, 12°) was acquired in the axial plane and included coverage of the entire liver. The acquisition was performed approximately 2-3 minutes after injection. Imaging parameters for the sequence were as follows: matrix, 140 x 256; field of view, 320-350 mm; and slab thickness, 160-200 mm with 32-40 partitions interpolated to 64-80 for an effective slice thickness of 2.5 mm. All acquisition times were kept to less than 25 seconds to facilitate breath-holding at end expiration.
For T1-weighted MR cholangiography, 3D interpolated fat-suppressed spoiled gradient-echo sequences were performed, one in the coronal plane and one in the axial plane, with a time delay of 60 minutes after the administration of gadobenate dimeglumine. This timing was based on our prior experience (unpublished data) because, to our knowledge, there are no documented studies on optimal timing for this agent in this setting. The imaging parameters for the coronal 3D gradient-echo sequence were as follows: 4.5/1.9; flip angle, 12°; matrix, 140 x 256; slab thickness, 81 mm with 27 partitions interpolated to 54 for an effective slice thickness of 1.5 mm; and an intermittent fat-saturation pulse. The imaging parameters for the axial 3D gradient-echo sequence were identical to those used for equilibrium phase MR images, as described earlier.
Image Processing
All images were reviewed on a workstation (Virtuoso, Siemens Medical
Solutions), a satellite console of the MR unit. One experienced technician
performed reformations of the data sets before and after contrast
administration. Coronal multiplanar reconstruction images and
maximum-intensity-projection (MIP) images for MRA and T2- and T1-weighted MR
cholangiography were generated interactively from the source images.
Image Interpretation
All MR images—including source, volume-rendered, and MIP
images—were evaluated by consensus of two MR-fellowship-trained
abdominal radiologists at a PACS monitor regarding anomalies of the biliary
tract; vascular anomalies; focal hepatic masses; diffuse disease of the
hepatic parenchyma, such as steatosis; and those diseases that might
complicate partial hepatectomy
[5]. At the time of this
interpretation, the observers had no information regarding correlative
findings from other imaging studies. The consensus interpretations performed
by the two reviewers of all the MR images were used as the reference standard
for the overall assessment of the accuracy of each MR angiographic and MR
cholangiographic technique.
Initially, the biliary system was analyzed regarding the anatomic variants. To avoid memory-recollection bias, a minimum of 2 weeks separated the reviews of T2- and T1-weighted MR cholangiography examinations with randomization of the order. In addition to separate analyses of T2- and T1-weighted MR cholangiography, a combined analysis of images obtained with the two sequences was performed at a 2-week interval. The observers were unaware of the findings from other sequences.
Normal biliary anatomy was defined when the right posterior duct (from posterior segments VI and VII) drained into the right hepatic duct, and both the right and left hepatic ducts converged into the common hepatic duct. Variants included trifurcation (right posterior duct draining into the junction of the right anterior duct and the left main duct); abnormal right configuration (right posterior duct draining into a left hepatic duct or to the common hepatic duct); accessory hepatic ducts; and other rare variants.
The diagnoses yielded during these three review sessions were later correlated with the intraoperative cholangiographic findings. As the standard reference, intraoperative cholangiography was interpreted by two experienced radiologists who had not participated in the analysis of the MR images. Although we excluded 20 patients who did not undergo liver harvest in analysis of the accuracy due to the lack of a standard of reference, two reviewers determined the findings in accordance between T1- and T2-weighted MR cholangiography in those patients.
The arterial and venous hepatic vascular anatomy was assessed using 3D MRA data sets that included MIP images and source images. Hepatic artery anatomy was classified according to the origins of the right and left hepatic arteries and the presence and the origins of any accessory hepatic arteries [12]. We determined the origin and course of hepatic arteries supplying segment IV. The portal vein anatomy was categorized as normal bifurcation, trifurcation, or abnormal origin of the right anterior portal vein from the left portal vein. Hepatic venous anatomy was evaluated for inferior accessory right hepatic veins and anomalous unions of three hepatic veins [1, 5, 6].
Evaluation of liver parenchymal abnormalities included assessment of fatty
infiltration on the basis of signal loss relative to the spleen on
opposed-phase T1-weighted gradient-echo images versus inphase images
[13,
14]. For quantitative analysis
of hepatic steatosis, an abdominal radiologist, with 7 years of experience in
abdominal MRI who did not participate in qualitative MR analysis, performed
quantitative measurements of signal intensities of the spleen and liver.
Region-of-interest (ROI) measurements (100-400 mm2) were made of an
area as large as possible to contain only hepatic or splenic parenchyma with
no large vessels or biliary trees. At least five ROIs were placed on liver and
spleen on five axial slices, and the ROI values were averaged as a mean liver
signal intensity. Based on our previous study
[15], the percent decrease in
the normalized liver signal on the opposed-phase sequence relative to the
in-phase sequence was calculated using the following formula and defined as
the relative signal decrease:
 |
The detection and characterization of masses relied on T1- and T2-weighted imaging and dynamic contrast-enhanced T1-weighted imaging. The results of MRI regarding fatty infiltration and focal liver lesions were compared with pathologic results of biopsy from the transplanted liver and the results of combined interpretation of sonography and CT by two radiologists who had not participated in the analysis of the MR images.
Intraoperative Correlation
Twenty-four subjects underwent laparotomy and had successful right hepatic
lobectomy for donation to liver transplant recipients. Intraoperative findings
were compared with MRI findings in all 24 patients who underwent
intraoperative cholangiography. As part of the surgical procedure, the hepatic
artery and portal vein were dissected only in the hepatic hilum, and before
the operation, the transplantation surgeons were informed of the results of
CTA regarding vascular variants. Intraoperative cholangiography was used as
the reference standard examination for evaluation of biliary variants. In the
cases in which the intraoperative cholangiographic and MR cholangiographic
findings were discordant, the MR and intraoperative images were compared side
by side to assess the causes of the discrepant findings.
Statistical Analysis
To compare the diagnostic performances, including sensitivity, specificity,
and overall accuracy, of the two individual MR cholangiographic techniques and
a combination of T1- and T2-weighted MR cholangiographic images with that of
intraoperative cholangiography, the McNemar test was used (SPSS software
[version 11.0], Statistical Package for the Social Sciences). A p
value of less than 0.05 was considered to indicate statistical
significance.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
None of the donors who underwent partial hepatectomy experienced biliary ductal or vascular complications related to harvesting. The results of imaging and operative findings are described next according to their anatomic divisions—namely, biliary, vascular, and parenchymal findings. The quality of the MR images was satisfactory to interpret these findings except in one patient with somewhat early venous phase MRA. The results from surgery are summarized in Table 1.
|
Biliary Ductal System
In the 24 patients who underwent liver harvesting, 10 had biliary anatomic
variants that were confirmed by intraoperative cholangiography and surgical
findings (Figs. 1A,
1B,
1C,
2A,
2B, and
2C). MRI findings and
intraoperative cholangiographic findings are summarized in
Table 2.
|
|
|
|
|
|
|
Compared with interpretation of the intraoperative cholangiographic images as the reference standard, T2-weighted MR cholangiography correctly depicted the biliary anatomy in 18 (75%) of the 24 subjects, including 12 of 14 with normal biliary anatomy and six of 10 with biliary anatomy variants. These results represented an overall sensitivity of 60% and an overall specificity of 86% for the detection of biliary anatomy variants.
When considered separately, gadobenate dimeglumine-enhanced T1-weighted MR cholangiography correctly depicted the biliary anatomy in 19 (79%) of the 24 subjects, including all 14 of the subjects with normal anatomy and five of 10 with variants, for a sensitivity of 50% and a specificity of 100% for the detection of variants. In contrast, combined interpretation of both MR cholangiographic image sets resulted in an accurate definition of the biliary anatomy in 22 (92%) of the 24 candidates, including 13 of the 14 donors with normal biliary anatomy and nine of the 10 donors with variants, for a sensitivity of 90% and a specificity of 93% for the detection of variants. Gadobenate dimeglumine-enhanced T1-weighted MR cholangiography was more accurate than T2-weighted MR cholangiography for depicting biliary anatomy, although differences did not reach statistical significance (p = 1.00). However, when we evaluated the bile duct anatomy by combined interpretations of the T2- and T1-weighted MR cholangiographic images, the diagnostic accuracy increased to 92%. However, the McNemar test failed to show a statistically significant increase compared with each T1-weighted (p = 0.248) and T2-weighted (p = 0.134) MR cholangiographic technique.
The differences in sensitivity and specificity among the three interpretation methods for determining the presence of biliary anatomic variants were not statistically significant according to the McNemar test (p = 0.13 for T1-weighted versus combined T1- and T2-weighted MR cholangiography, p = 0.25 for T2-weighted versus combined T1- and T2-weighted MR cholangiography). In the 24 patients who underwent liver harvest, the accordance between T1- and T2-weighted MR cholangiography was 79% (n = 19/24). In addition, in the 20 patients who did not undergo liver harvest and were excluded from analysis of the accuracy of MR cholangiography, T1- and T2-weighted MR cholangiography showed concordant results in 15 patients (75%).
Vascular System
More than half of our patients (n = 14/24, 58%) were found to have
normal hepatic artery anatomy. Thirteen arterial anatomic variants were
observed in 10 patients from surgery: an accessory left hepatic artery
originating from the left gastric artery (n = 4, 31%); a left hepatic
artery originating from the left gastric artery (n = 3, 23%); a right
hepatic artery from the superior mesenteric artery (n = 1, 8%)
(Fig. 3); an accessory right
hepatic artery from the superior mesenteric artery (n = 1, 8%); a
proper hepatic artery from the superior mesenteric artery (n = 1,
8%); and early branching segmental arteries (n = 3, 23%). The artery
to segment IV arose from the left hepatic artery in 21 donors and from the
right hepatic artery in three. In these three donors, the distance from the
hepatic artery bifurcation to the origin of the artery supplying segment IV
was less than 1.5 cm. MRA revealed the origin of the artery to segment IV
correctly in all cases. However, MRA did not reveal two accessory left hepatic
arteries and three early branching segmental arteries. Therefore, the
diagnostic accuracy of MRA was 79% (n = 19/24).
|
Portal venous variants were confirmed in 21% of our patients (n = 5/24) during surgery: trifurcation of the right anterior portal vein, the right posterior portal vein, and the left portal vein in three patients (Fig. 4); common cloaca of right portal vein in one patient; and a right anterior portal vein originating from the left portal vein in one patient. In all patients (n = 24/24), MRA in the portal venous phase produced the correct diagnosis.
|
|
|
|
|
|
|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
However, MRA and cholangiography have limited temporal and spatial resolution compared with direct DSA and intraoperative cholangiography. From this point of view, we speculate that the twofold T1 relaxation afforded by gadobenate dimeglumine provides a higher SNR of the vascular structure on MRA than other gadolinium chelates and its biliary excreting feature contributes to improving the evaluation of the biliary anatomy by allowing delayed 3D T1-weighted MR cholangiography [9, 10].
Indeed, in our study, combined T2- and T1-weighted MR cholangiography increased the diagnostic accuracy to 92% from 75% on T2- and from 79% on T1-weighted MR cholangiography. Many previous studies have shown that MR cholangiography, using either the T2-weighted TSE or HASTE sequence, clearly shows the biliary anatomy, but some biliary anomalies are not easily revealed because of the limited resolution and 2D character of T2-weighted MR cholangiography [16-18]. In addition, given that the complex orthogonal relationships between the right anterior duct, right posterior duct, left hepatic duct, and common hepatic duct are often difficult to define with confidence on conventional 2D MR images, the use of volume-rendering algorithms for reconstruction of gadobenate dimeglumine-enhanced 3D data sets facilitates definition of these relationships for surgical planning [19]. Furthermore, gadobenate dimeglumine-enhanced T1-weighted MR cholangiography can provide information about the differentiation of cystic structures near bile ducts and bile duct lumen. For example, in one patient with numerous peribiliary cysts, no communication of the cysts with a bile duct was confirmed on delayed T1-weighted contrast-enhanced MR cholangiography. Moreover, CT and normal T2-weighted MR cholangiography could not be used to differentiate peribiliary cysts from choledochal cysts.
Recently, several studies using mangafodipir trisodium-enhanced 3D T1-weighted MR cholangiography have reported increased SNR and a greater depiction rate of the ductal anatomy than T2-weighted MR cholangiography and 2D gradient-echo images [7, 19, 20]. In our study, T1-weighted 3D gradientecho images that were obtained 1 hour after injection of gadobenate dimeglumine revealed contrast excretion into the biliary tree and provided cholangiographic images. However, given that only 2-5% of injected gadobenate dimeglumine was excreted into a bile duct but approximately 50% of mangafodipir trisodium was excreted into a bile duct, the SNR of bile on mangafodipir trisodium-enhanced cholangiography could be much higher than that on gadobenate dimeglumine-enhanced cholangiography. Nevertheless, gadobenate dimeglumine-enhanced MRI could have potential advantages over mangafodipir trisodium-enhanced imaging because of its utility for dynamic first-pass imaging. Indeed, given that preoperative evaluation of liver donor candidates includes precise anatomic information about hepatic vasculature, parenchymal abnormalities, and biliary anomalies, the capability of gadobenate dimeglumine to allow simultaneous acquisition of T1-weighted MR cholangiography and parenchymal and vascular imaging has additional value over mangafodipir trisodium. In addition, gadobenate dimeglumine is easier of use than mangafodipir trisodium, which needs to be infused slowly over 10-20 minutes, and is less expensive than mangafodipir trisodium.
Several recent reports show CT cholangiography with iodipamide meglumine enables good visualization of biliary tract anatomy [21, 22]. Wang et al. [21] reported that biliary tract anatomy depicted on CT cholangiography was concordant with findings at surgery in 96% of 24 subjects who underwent right lobe retrieval. However, despite the fact that IV cholangiography with iodipamide meglumine has high accuracy in biliary tract imaging, it also has some risks of contrast material reactions and additional iodinated contrast agent needs to be administered to provide adequate hepatic vascular angiography and parenchymal imaging. In our institute where more than 200 living liver donor transplantations have been performed without a donor death to date, we always give the first priority to the safety of the liver donor. Furthermore, our result in diagnostic accuracy of combined T2- and T1-weighted MR cholangiography (92%) is comparable with the results of Wang et al. Therefore, considering that gadobenate dimeglumine is safer than iodipamide meglumine, use of additional contrast agent for vascular and parenchymal evaluation is not needed, and diagnostic accuracy is comparable, a combined approach of T2-weighted MR cholangiography and gadobenate dimeglumine-enhanced MRA and MR cholangiography could have reasonable standing as a valuable diagnostic imaging tool for liver donor evaluation.
Our results showed an accuracy of 79% for diagnosis of hepatic arterial anatomy, 100% for portal venous anatomy, and 96% for hepatic venous anatomy. Although the spatial resolution of MRI for the evaluation of hepatic arterial anatomy remains inferior to that of DSA, MRA has, in our experience, proven to be an accurate preoperative test for the definition of the portal and hepatic venous anatomy. Several studies have found that CT or MRI can show the anatomy of the hepatic vasculature of a liver donor before transplantation [23-25]. Winter et al. [23] and Chan et al. [24] showed that in liver transplantation candidates, successful 3D CT arteriography was as accurate as catheter angiography for assessing hepatic arterial anatomy. Although the risk of radiation exposure and of a reaction to iodinated contrast material is probably low enough to justify the use of CT for the evaluation of the hepatic vasculature of a living liver donor, the superior safety profile of gadolinium-based agents compared with that of the iodinated contrast agents must be considered one merit of MRI [26, 27]. In addition, when measured using a test dose of contrast material, arterial phase 3D MR images can be obtained reliably and reproducibly [11].
Regarding the assessment of hepatic masses, the unsurpassed soft-tissue contrast inherent to MRI permits assessment of the liver parenchyma using a variety of contrast mechanisms of which the T2- and contrast-enhanced T1-weighted images have been shown to be the most relevant [1]. The 3D data sets of MRA were also valuable for the characterization of focal liver lesions in our study. Based on classic enhancement profiles, in the present study, two hemangiomas were diagnosed with certainty in two patients, thereby obviating biopsy. In addition, chemical shift MRI has been regarded as the best diagnostic approach for detecting fatty liver [13]. For a quantitative MR assessment for fatty liver, researchers have studied using in- and opposed-phase MR sequences [28, 29]. If additional larger studies confirm the accuracy of MRI for the quantification of steatosis, MRI may render the performance of liver biopsy unnecessary for that purpose [5, 15].
Our study was clearly limited. First, our study sample was relatively small. Second, as CT and the surgical findings provided confirmation of our study, DSA was not performed primarily because of the reluctance of the transplant surgeons to subject otherwise healthy individuals to additional testing not deemed necessary. However, two experienced transplantation surgeons interpreted the vascular anatomy in the surgical field, and the possibility that a surgically important vascular variant was missed is very low. Last, the MR examinations were evaluated in consensus and, therefore, observer variability could not be assessed.
We conclude that a comprehensive assessment of the hepatic parenchymal; biliary ductal system; and hepatic arterial, portal, and venous systems can be accomplished with the outlined MR protocol using gadobenate dimeglumine. On the basis of our results, we conclude that MRI can potentially serve as the sole preoperative imaging test for living liver donor candidates.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. H. Choi, J. M. Lee, N. C. Yu, K.-S. Suh, J.-J. Jang, S. H. Kim, and B. I. Choi Hepatocellular Carcinoma in Liver Transplantation Candidates: Detection with Gadobenate Dimeglumine-Enhanced MRI Am. J. Roentgenol., August 1, 2008; 191(2): 529 - 536. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Basaran, A. M. Agildere, F. Y. Donmez, S. Sevmis, I. Budakoglu, H. Karakayali, and M. Haberal MR Cholangiopancreatography with T2-Weighted Prospective Acquisition Correction Turbo Spin-Echo Sequence of the Biliary Anatomy of Potential Living Liver Transplant Donors Am. J. Roentgenol., June 1, 2008; 190(6): 1527 - 1533. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SE |
