| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Diagnostic Radiology, Yonsei University College of Medicine,
Seodaemun-ku Shinchondong 134, Seoul 120-752, Republic of Korea.
2 Brain Korea 21 Project for Medical Science, Yonsei University College of
Medicine, Seoul 120-752, Republic of Korea.
3 Department of Surgery, Yonsei University College of Medicine, Seoul 120-752,
Republic of Korea.
Received February 7, 2004; accepted after revision October 19, 2004.
Address correspondence to M.-J. Kim.
OBJECTIVE. This article evaluates the feasibility of single-dose gadobenate dimeglumine–enhanced MRI as both an angiographic and biliary contrast medium for making a preoperative evaluation of the donor candidates for a living-donor-related liver transplantation.
SUBJECTS AND METHODS. Eleven right hepatic lobe donors underwent MRI examinations using T1- and T2-weighted imaging and T2-weighted MR cholangiography (MRC). The MR angiography (MRA) and contrast-enhanced (CE) T1-weighted MRC images then were obtained after injecting a single dose of gadobenate dimeglumine. One radiologist and one surgeon prospectively reviewed all the MRI examinations for hepatic vascular and biliary abnormalities and compared them with the surgical findings and intraoperative cholangiograms. In addition, two blinded reviewers evaluated the two sets of MRC (T2-weighted MRC set and T2-weighted MRC plus CE-T1-weighted MRC set) retrospectively and recorded the anatomic types of the hilar biliary branching pattern along with their confidence in the interpretation.
RESULTS. Prospective analysis detected the following vascular variants: hepatic arterial variation in two patients, portal venous variation in one, and a significantly large accessory hepatic vein (> 5 mm) in one. Biliary variants also were identified in two patients. All the MRI findings on the vascular and biliary anatomy were corroborated intraoperatively. Retrospective analysis showed that the mean diagnostic confidence in the combined set was significantly higher than that of the T2-weighted MRC alone by both reviewers (p < 0.05).
CONCLUSION. Obtaining both MRA and CE-T1-weighted MRC is feasible using a single dose of gadobenate dimeglumine. Therefore, gadobenate dimeglumine–enhanced MRI might play a role as a preoperative imaging technique for the vascular and biliary evaluation of potential living donors.
Adult-to-adult living liver donor transplantation increasingly is being used to overcome the shortage of available cadaveric livers [1]. Preoperative imaging is important for evaluating a potential living donor to exclude a parenchymal, biliary, or vascular abnormality or a variation that might contraindicate or increase the surgical morbidity for a partial hepatectomy [2, 3]. Currently, CT or MRI generally is used for this purpose. However, conventional angiography and cholangiography may also be included in the preoperative evaluation of a potential donor [4, 5].
A comprehensive MRI examination for a preoperative donor includes abdominal MRI for the detection of focal and diffuse liver disease and for a volumetric assessment of the liver; MR angiography (MRA) to determine the patency and anomalies of the hepatic arteries, portal veins, and hepatic veins; and MR cholangiography (MRC) for the detection of biliary anomalies. Therefore, MRI has the potential to simplify the diagnostic technique by providing a preoperative evaluation of the potential donor in a single examination. However, conventional MRC using heavily T2-weighted turbo spin-echo techniques often has diagnostic limitations on the detection and definition of intrahepatic anatomic anomalies, particularly in nondilated systems [6]. The recent development of liver-specific contrast agents that excrete into the biliary tree, such as mangafodipir trisodium, has renewed interest in the contrast-enhanced MRI of the biliary tree, which might counter the limitations of heavily T2-weighted MRC. Recent reports have suggested that mangafodipir trisodium–enhanced T1-weighted MRC may help define the intrahepatic bile duct anatomy in healthy liver transplant donor candidates [2, 7, 8]. However, if preoperative MRI for a donor evaluation includes mangafodipir trisodium–enhanced T1-weighted MRC, both the gadopentetate dimeglumine for MRA and the mangafodipir trisodium contrast material should be administered concurrently. The use of both contrast agents increases the cost of an examination and causes inconvenience to the patient (total increased cost: about $170 in Korea). In addition, mangafodipir trisodium is currently unavailable in the United States.
Gadobenate dimeglumine, which is another intrabiliary contrast agent, is a gadolinium-based contrast agent for MRI. This agent combines the properties of a gadolinium-based extracellular agent with those of a hepatocyte-directed agent [9]. Therefore, a single dose of gadobenate dimeglumine potentially can provide a comprehensive evaluation of the hepatic parenchyma, vasculature, and bile ducts. To the best of our knowledge, gadobenate dimeglumine–enhanced MRI for both vascular and biliary anatomy evaluation has not been reported.
The aim of this study was to evaluate the feasibility and the utility of single-dose gadobenate dimeglumine–enhanced MRI as both an angiographic and biliary contrast medium for making a preoperative evaluation of donor candidates for a living-donor-related liver transplantation.
Subjects and Methods
Patients
Between May 2002 and June 2003, 43 (13 women and 30 men) consecutive
potential living liver donors (age range, 16–52 years; mean age, 31.6
years) underwent an MRI evaluation using a single dose of gadobenate
dimeglumine as both a biliary and angiographic contrast medium. Institutional
review board approval was obtained for this study. Informed consent was
unnecessary according to our institutional review board because the studies
were performed as part of a clinical preoperative evaluation for
living-donor-related liver transplantation. No other imaging tests such as
conventional angiography and endoscopic retrograde cholangiography were
performed in any of the candidates. Nine candidates were excluded as donors
because of the anatomic variants on the MRI findings
(Table 1). Although most
anatomic vascular and biliary variants can be managed safely with the recent
advances in reconstruction techniques
[10], potential donors with a
concomitant vascular and biliary variant were excluded at our institution.
Multiple large accessory hepatic veins in a potential donor can substantially
increase the complexity of the surgical procedure and these candidates also
were excluded. In the case of multiple donor candidates for a recipient, the
donor candidate with the most favorable anatomy was selected for the MRI
examination. Other candidates were excluded based on factors unrelated to an
anatomic variant (insufficient volume, n = 7; large hemangioma in the
liver, n = 1; recipient's issues, n = 5; psychosocial
issues, n = 10). Finally, 11 candidates underwent a successful,
uncomplicated, right hepatectomy and were enrolled in the study.
|
Imaging Techniques
All the MRI examinations were performed using a 1.5-T imaging system
(Gyroscan Intera, Philips Medical Systems) using a surface phased-array coil.
Fasting was recommended before the examinations, but this rule was not
strictly enforced. Antiperistaltic agents or oral contrast agents were not
used. In addition to the MRC and MRA, the donor candidates underwent the
following sequence imaging: a breath-hold axial T1-weighted dual fast
gradient-recalled-echo sequence (in-phase and out-of-phase sequences)
(TR/in-phase TE, 126/4.6 msec; out-of-phase TE = 2.3 msec; flip angle,
90°; field of view, 32–36 x 25–29 cm; matrix, 256
x 192; section thickness, 7 mm; intersection gap, 1 mm; one signal
acquired; number of slices, 24); a breath-hold axial T2-weighted turbo
spin-echo sequence with spectral fat saturation (TR/TE, 3,216/80; echo-train
length, 28; field of view, 32–36 x 25–29 cm; matrix, 256
x 192; section thickness, 7 mm; intersection gap, 1 mm; one signal
acquired; number of slices, 24); and a breath-hold axial- and coronal-2D
balanced turbo field-echo sequence (TR/TE, 3.5/1.4 msec; flip angle, 80°;
field of view, 32–36 x 25–36 cm; matrix, 256 x 256;
section thickness, 5 mm; intersection gap, 0 mm; one signal acquired; number
of slices, 32 in axial imaging and 20 in coronal imaging).
MRA—Dynamic contrast-enhanced MRA was performed in the coronal plane using a breath-hold 3D T1-weighted fast-field-echo sequence (TR/TE, 5.1/1.4; flip angle, 40°; field of view, 36–40 cm; matrix, 512 x 160; slab thickness, 60 mm with 30 partitions interpolated to 60 for a slice thickness of 2 mm at 1-mm interval; one signal acquired; imaging time, 13–14 sec per breath-hold), following an IV bolus of 0.1 mmol gadobenate dimeglumine (Multihance, Bracco Imaging) per kilogram of body weight. This sequence was repeated three times with data acquisition in the hepatic arterial, portal venous, and hepatic venous phases. An automatic infusion system (Spectris MR injector, Medrad Europe) operating at an injection rate of 2 mL/sec was used. To start the sequence after administering the contrast material, a real-time bolus-tracking method was used in the coronal plane with a T1-weighted fast-field-echo sequence (TR/TE, 4.3/1.4; flip angle, 40°; field of view, 53 cm; matrix, 180 x 256; one section; section thickness, 100 mm; one signal acquired). Because of a technical time delay of 5 sec between the bolus-tracking sequence and the beginning of the MRA sequence, the latter was begun when the contrast medium bolus was in the left ventricle. Within these 5 sec, the patient was instructed to hold his or her breath for the hepatic-artery-phase data acquisition. Between the hepatic arterial and portal venous phases, the patient was permitted to breathe once. The patient was allowed to breathe twice between the portal venous and venous phases. Maximum-intensity-projection (MIP) images were reconstructed from the 3D data set for each breath-hold in the hepatic arterial, portal venous, and hepatic venous phases by an experienced technician according to a standardized protocol. Nine standard projections with different projection angles were routinely used in all patients. When necessary, additional segmented MIP images were obtained with a dedicated workstation (Easy-Vision 4.0, Philips Medical Systems). The reconstruction of the images took an average of 15 min per study.
MRC—Breath-hold T2-weighted MRC was performed with a rapid acquisition and relaxation enhancement sequence (TE = 830–1,050 msec; section thickness, 30 mm; field of view, 24 cm) in nine coronal oblique orientations (–30° through 30°) using a single-section thick-slab technique. The acquisition time was 9 sec per section (total, 81 sec). However, the candidates were brought back for contrast-enhanced T1-weighted cholangiography 60 min after the IV administration of gadobenate dimeglumine for MRA. Three-dimensional volumetric breath-hold T1-weighted fast-field-echo acquisitions of the liver and biliary system were performed in the coronal plane. The scanning parameters for the high-resolution sequences that were performed with a limited coverage included TR/TE, 5.1/1.47; flip angle, 40°; matrix, 512 x 160; field of view, 34 cm using a rectangular field of view; and slab thickness, 40 mm with 20 partitions and section thickness of 2 mm, interpolated to 40 slices at 1-mm intervals.
Intraoperative cholangiography—Intraoperative cholangiography was performed by the transplant surgeons in all donors who underwent a partial hepatectomy. After the cystic duct had been cannulated, approximately 5–10 mL of meglumine ioxithalamate (Telebrix, Guerbet) was injected by hand under fluoroscopic control before the right hepatic lobe resection. A minimum of two images were obtained. Although surgical planning was based on the preoperative imaging, the transplantation surgeons made the final decision as to the type of ductal anastomosis to be performed after reviewing the intraoperative cholangiograms in all donors.
Image Analysis
Prospective image evaluation on vascular and biliary
anatomy—All the images, including the source images and 2D or 3D
postprocessed images, were sent to a PACS workstation, which allows
interactive analysis. One radiologist and one surgeon prospectively evaluated
the images with consensus. They defined what would be considered a classic
anatomy for the vascular and biliary abnormalities. For the arteries, they
divided the individual cases into six types (I–VI), according to the
Michels classification [11] of
the branching patterns of the left gastric, common hepatic, and splenic
arteries. The arterial anatomy of type I classification (hepatosplenogastric
trunk) was considered to represent the normal anatomy. In particular, with
regard to the middle hepatic artery, both the nonvisualization case and the
middle hepatic artery from the left hepatic artery were regarded as being a
normal anatomy, whereas the middle hepatic artery from the right hepatic
artery was regarded as being a variation, because vascular supply to segment
IV is essential to ensure the function of this segment in the donor. The
portal vein anatomy was characterized into the patterns described in the
literature, and included the following: normal anatomy (bifurcation into the
normal right and left portal veins)trifurcation (origin of the left portal
vein at the bifurcation of the right portal vein into the anterior and
posterior branches), and origin of the right anterior segmental branch from
the left portal vein [12,
13]. The presence of a normal
right hepatic vein was noted. The locations of any accessory hepatic veins
also were recorded. The biliary anatomy bifurcation patterns were
characterized into the patterns previously described in the literature
[13]. Most notably, an
emphasis was placed on discerning whether there was a right hepatic duct. The
junction of the anterior segmental duct and the posterior segmental duct to
form the right hepatic duct, and the right hepatic duct, in turn, joining the
left hepatic duct in the hilar confluence, were considered to be normal
anatomy. There were three anatomic variations where the anterior and posterior
segmental bile ducts did not form a right hepatic duct: the posterior
segmental duct joined the left hepatic duct, there was a three-branch-type
hilar confluence (trifurcation), and the anterior segmental duct joined the
left hepatic duct. The presence and location of any accessory hepatic ducts
were noted. The MRI definition of the vascular anatomy was compared with the
surgical findings. However, the MRI definition of the biliary anatomy was
compared with the surgical findings and operative cholangiography.
Retrospective image evaluation on hilar biliary branching anatomy—Two experienced radiologists independently reviewed the T2-weighted MRC and a combined set (T2-weighted MRC and gadobenate dimeglumine–enhanced T1-weighted MRC) to verify the utility of gadobenate dimeglumine–enhanced MRC. First, the T2-weighted MRC images were analyzed independently by two radiologists. Second, the same radiologists reviewed the combined sets 2 weeks later. To limit reviewer bias, none of the reviewers was involved in the preoperative donor evaluation. Each reviewer recorded the anatomic types of the hilar bile duct anatomy for the branching pattern and the diagnostic confidence was graded using a 5-point scale (from 1 [very low] to 5 [very high]). A paired Wilcoxon's signed rank test was used to assess the differences in the diagnostic confidence, and a p value < 0.05 was considered significant.
The typing accuracy of the T2-weighted MRC and the combined set was evaluated based on the operative cholangiographic findings and the surgical findings, respectively.
Results
Prospective Image Evaluation on Vascular and Biliary Anatomy
Hepatic arterial anatomy—Most patients (10/11) were found to
have Michels classification type I (hepatosplenogastric trunk). A common
hepatic artery from the superior mesenteric artery (Michels classification
type V) was detected in one donor candidate at MRA
(Fig. 1). All the MRA findings
about the hepatic arterial system were corroborated intraoperatively. The
replaced common hepatic artery from the superior mesenteric artery in the
donor candidate is known to be an important variant because extra steps are
needed to remove the liver tissue. However, this variant was not considered a
contraindication of a liver donation. However, MRA showed the following data
with regard to the middle hepatic artery: an origin from the left hepatic
artery (n = 9), nonvisualization (n = 1), and an origin from
the right hepatic artery (n = 1)
(Fig. 2). The middle hepatic
artery from the right hepatic artery was considered to be a variation because
the hepatectomy line would cross the arterial supply of the medial segment.
However, the donor with the middle hepatic artery from the right hepatic
artery was not excluded because the origin of the middle hepatic artery was
proximal, near the origin of the left hepatic artery, which allowed a
sufficiently long segment of the right hepatic artery distal to the middle
hepatic artery to be resected. In the case of a donor with nonvisualization of
the middle hepatic artery on MRA, the surgeon identified that the middle
hepatic artery did not originate from the right hepatic artery during the
dissection and confirmed the intact left lobe during the transient clamping of
the right hepatic artery and right portal vein.
|
|
|
|
|
Biliary anatomy—MRI identified biliary variants in 2 of 11 patients (trifurcation and right posterior to left duct pattern) (Figs. 5A, 5B and 6A, 6B, 6C). These findings were confirmed by the surgical findings and intraoperative cholangiography. Dual orifices were acquired in both cases.
|
|
|
|
|
Discussion
Before transplantation, potential donors undergo an extensive evaluation with the intent of minimizing the number of complications in the healthy donor and optimizing the graft function in the recipient. Although the donor evaluation protocols vary among institutions, the preoperative assessment of the donor generally is conducted in a series of steps beginning with simple and noninvasive examinations and progressing to more complex and invasive procedures [5]. Recently, several studies reported that an MRI examination using gadolinium-based extracellular agents can be used as the sole imaging technique for making a preoperative evaluation of a potential living donor [3, 14, 15]. However, there is difficulty in evaluating the hilar bile duct anatomy in the nondilated biliary tree. Therefore, an intraoperative cholangiography is necessary for making a precise evaluation of the biliary tree [16]. Some authors used manganese-enhanced MRC in addition to the gadolinium-enhanced MRI for making a better evaluation of the hilar bile duct anatomy [2, 7, 8]. However, this approach requires the combined use of two different contrast agents (manganese agent for MRC and gadolinium-based agent for MRA). Gadobenate dimeglumine can be used for dynamic imaging as an extracellular space agent in the early phase and a biliary contrast agent in the delayed phase. Therefore, this study evaluated the utility of gadobenate dimeglumine–enhanced MRI for both a vascular and biliary anatomy evaluation of donor candidates for living-donor-related liver transplantation.
In this study, the gadobenate dimeglumine–enhanced MRA for the hepatic artery, portal vein, and hepatic vein were corroborated intraoperatively. The preoperative detection of surgically significant vascular variants enabled the surgeons to anticipate this additional component of the operation (acquisition of dual orifices in a donor with a portal venous variant: n = 1, reimplantation of accessory hepatic vein in a recipient: n =1). These results are in agreement with the literature supporting the use of gadolinium-enhanced MRA for defining the hepatic vascular system [17, 18]. In particular, gadobenate dimeglumine–enhanced MRA is known to yield an increased signal-to-noise (SNR) ratio, especially in small vessels, and is superior to MRA using conventional gadolinium-based extracellular agents because of the higher relaxitivity and weak binding with the protein in the body fluid [19, 20].
An evaluation of the biliary anatomy in donor candidates for a living-donor-related liver transplant is essential. Ductal anomalies such as aberrant drainage of the right posterior segmental duct into the left hepatic duct or at the level of the bifurcation (trifurcation) preclude the performance of a single duct-to-duct biliary anastomosis and require an additional biliary anastomosis, which might increase the risk of biliary complications [5, 16]. T2-weighted MRC has been used to assess the intrahepatic biliary anatomy in liver donors [3, 14, 21]. However, previous studies have reported difficulties in depicting the biliary anatomy in potential living donors using T2-weighted MRC [3, 14]. The spatial resolution of T2-weighted single-section single-shot rapid acquisition with relaxation enhancement MRC might be suboptimal owing to the use of thick sections. In addition, the images can be blurred as a result of the use of multiecho techniques with long echo trains. Because of the relatively low spatial resolution, a depiction of the intrahepatic biliary anatomy often is inadequate in individuals without a dilated biliary system. A new approach for MRC is needed for an accurate delineation of the intrahepatic biliary anatomy. Some studies suggest that a new approach will need to rely on contrast agents that are excreted into the bile ducts in order for these approaches to be useful in delineating the intrahepatic biliary anatomy [7, 22]. The spatial resolution potentially can be improved using rapid 3D T1-weighted gradient-echo techniques with contiguous or overlapping thin sections where the bile has been enhanced by the administration of a suitable biliary contrast media. Lee et al. [7] and Kapoor et al. [2] reported that mangafodipir trisodium–enhanced MRI, especially when performed using a high-resolution volumetric T1-weighted gradient sequence, can facilitate the definition of the intrahepatic bile duct anatomy in healthy liver transplant donor candidates. Gadobenate dimeglumine also might be used to depict the biliary anatomy in a delayed phase, because 2–4% of the injected gadobenate dimeglumine dose is excreted into the biliary system [23]. In this study, all the MRC findings about the hilar ductal branching pattern were matched with the operative cholangiography in the prospective evaluation using the combined set. In particular, the acquisition of dual orifices during right lobe harvesting was accurately predicted (trifurcation: n =1 and right posterior to left duct pattern: n =1). In addition, diagnostic confidence in the combined set was significantly improved over that of T2-weighted MRC alone in both reviewers on retrospective analysis, although the accuracy of the combined set was improved only in one donor by one reviewer (Fig. 7A, 7B, 7C). These results suggest that a combination of images obtained from the standard T2-weighted MRC and gadobenate dimeglumine–enhanced T1-weighted MRC may confer improved diagnostic confidence and accuracy in determining the anatomic types of the hilar branching pattern.
|
|
|
This study has some limitations. First, although conventional angiography is considered to be the gold standard for evaluating the vascular anatomy, no conventional angiography was performed in any donor candidates in this study. Therefore, the accuracy in depicting the hepatic arteries and the portal veins could not be assessed. Because the spatial resolution of MRI is inferior in relation to that of digital subtraction angiography, small hepatic arterial branches might not be depicted on the preoperative MRI examination. However, no serious surgical problems were encountered in the patients even though conventional angiography had not been performed before surgery. Gadolinium-enhanced MRA has been reported to be comparable to digital subtraction angiography in evaluating the hepatic arterial system and was superior in showing the portal vein [24]. In addition, MRI and MRA have advantages in depicting the hepatic veins including the anomalous union of the hepatic veins and the presence of accessory hepatic veins. Second, the gadobenate dimeglumine–enhanced T1-weighted cholangiography was obtained approximately 1 hr after the injection. Therefore, the 1-hr delay in the gadobenate dimeglumine–enhanced T1-cholangiography might increase the imaging time and cost. In addition, it has been reported that the optimal time for evaluating the liver parenchyma and the bile duct is between 60 and 120 min after the injection [25]. Because serial images were not obtained at the gadobenate dimeglumine–enhanced MRC, it is unclear whether the delayed images provided the best scan window for acquiring the contrast-enhanced T1-weighted cholangiographic images. Third, the study sample was relatively small; only 11 donors underwent surgery and intraoperative cholangiography. Therefore, the statistical significance of comparing the T2-weighted MRC alone with the combined set of T2- and T1-weighted cholangiography in terms of the accuracy in classifying the bile duct anatomy could not be determined. Although these results showed that the addition of the contrastenhanced T1-weighted cholangiography using a single injection of the gadobenate dimeglumine can improve the radiologists' confidence, the expected bias toward more images (in this case, the combined set) could not be excluded. A larger and more randomized study will be needed to verify the benefits of adding contrast-enhanced T1-weighted cholangiography.
In conclusion, these results showed that it is feasible to obtain both MRA and contrast-enhanced T1-weighted MRC images using a single dose of gadobenate dimeglumine. Therefore, gadobenate dimeglumine–enhanced MRI might play a role as a comprehensive imaging technique for the vascular and biliary anatomy in making a preoperative evaluation of potential living donors for a living-donor-related liver transplant.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. S. Lim, M.-J. Kim, S. Myoung, M.-S. Park, J.-Y. Choi, J.-S. Choi, and S. I. Kim MR Cholangiography for Evaluation of Hilar Branching Anatomy in Transplantation of the Right Hepatic Lobe from a Living Donor Am. J. Roentgenol., August 1, 2008; 191(2): 537 - 545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. A. Catalano, A. H. Singh, R. N. Uppot, P. F. Hahn, C. R. Ferrone, and D. V. Sahani Vascular and Biliary Variants in the Liver: Implications for Liver Surgery RadioGraphics, March 1, 2008; 28(2): 359 - 378. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
/1,088
[effective], 30-mm thickness) (A) was interpreted as ductal
trifurcation with low diagnostic confidence grades by both reviewers, but
serial-enhanced T1-weighted MR cholangiography (coronal 3D T1-weighted
gradient-echo MRI [TR/TE, 5.1/1.47; flip angle, 40°]) (B)
accurately depicts trifurcation pattern (right anterior segmental branch:
long black arrow, left hepatic duct: short black arrow,
right posterior segmental branch: white arrows) (anterior to
posterior).
