| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology (B.M.Y., B.T., R.S.B., A.Q., F.V.C.) and Surgery (J.P.R.), University of California San Francisco, Box 0628, C-324C, 505 Parnassus Ave, San Francisco, CA 94143-0628. From the 2002 RSNA scientific assembly. Received December 26, 2002; revision requested March 4, 2003; revision received May 30; accepted July 23. Address correspondence to B.M.Y. (e-mail: benyeh@itsa.ucsf.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Eight living potential liver donors underwent iodipamide meglumine–enhanced CT cholangiography. Eight different potential liver donors then underwent conventional MR cholangiography and mangafodipir trisodium–enhanced excretory MR cholangiography. Two readers independently scored all first-, second-, and third-order biliary branches with a four-point scale from 0 (not seen) to 3 (excellent visualization). Interobserver agreement was calculated by using the weighted 
statistic. Scores were compared between imaging modalities by using generalized estimating equations. Imaging findings of second-order biliary tract anatomy were compared with intraoperative findings for nine patients.
RESULTS: Interobserver agreement for overall biliary tract visualization was good for CT, conventional MR, and excretory MR cholangiography (with weighted values of 0.76, 0.66, and 0.79, respectively). The mean second-order biliary branch visualization scores for readers 1 and 2, respectively, were significantly higher at CT cholangiography (2.81 and 2.75) than at conventional MR (1.84 and 1.75, P < .001), excretory MR (2.00 and 2.06, P < .001), and combined conventional and excretory MR cholangiography (2.31 and 2.25, P < .01). At CT, conventional MR, and excretory MR cholangiography, respectively, second-order biliary branching anatomy was discernible in eight, five, and seven patients, with second-order biliary branch variants seen in three, two, and two patients. Surgical findings confirmed the pattern of second-order biliary branching seen at CT in five patients, that seen at conventional MR imaging in one patient, and that seen at excretory MR cholangiography in three patients. At surgery, one case of variant biliary anatomy was found to have been missed at CT cholangiography.
CONCLUSION: In living potential liver donors, CT cholangiography enables significantly better biliary tract visualization than conventional or excretory MR cholangiography either alone or in combination.
© RSNA, 2004
Index terms: Bile ducts, anatomy, 763.13, 764.13, 765.13, 766.13 • Bile ducts, CT, 763.12116, 764.12116, 765.12116, 766.12116 • Bile ducts, MR, 763.12142, 764.12142, 765.12142, 766.12142 • Liver, transplantation, 761.89 • Magnetic resonance (MR), cholangiopancreatography, 763.12142, 764.12142, 765.12142, 766.12142
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The standard examination for defining biliary anatomy—diagnostic endoscopic retrograde cholangiography—has a major complication rate of 1.4%–3.2% (9,10). Development of a safer method of evaluating biliary anatomy would be beneficial. Noninvasive evaluation of the biliary tract in living potential liver donors is challenging owing to the small caliber of normal-sized bile ducts, although promising results have been obtained with conventional MR cholangiography (11), mangafodipir trisodium–enhanced excretory MR cholangiography (12,13), and CT cholangiography (4,14). The advantages of conventional and excretory MR cholangiography include a high contrast-to-noise ratio and absence of ionizing radiation exposure. The advantages of multi–detector row CT cholangiography include higher spatial resolution and lower cost.
To our knowledge, comparative data on the accuracy of these noninvasive modalities have not been reported. We undertook this study to compare biliary tract visualization in living potential liver donors at conventional MR, mangafodipir trisodium–enhanced excretory MR, and multi–detector row CT cholangiography.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
All patients underwent multi–detector row CT for evaluation of hepatic vasculature, liver volume, and liver parenchyma before undergoing CT or MR cholangiography. Intravenous morphine sulfate (Abbott Laboratories, Chicago, Ill) (0.04 mg per kilogram of body weight) was administered to all patients before cholangiography to contract the sphincter of Oddi (15–17), retard transit of contrast material, and improve biliary imaging.
Of the total of 16 patients, nine proceeded to liver donation. Of these nine patients, six had undergone CT cholangiography and three had undergone conventional and excretory MR cholangiography. Two patients who underwent CT cholangiography did not donate portions of their livers owing to the presence of nonalcoholic steatohepatitis in the donor (n = 1) or to the recipient receiving a cadaveric liver transplant (n = 1). Five patients who underwent MR cholangiography did not donate portions of their livers due to improvement in the status of the recipient (n = 2), personal reasons (n = 2), or the presence of numerous hepatic cysts (n = 1).
CT Cholangiography
Before infusion of cholangiographic contrast material, all patients were first given 25 mg of intravenous diphenhydramine (Benadryl; Pfizer, New York, NY). Contrast material was then administered as a 30-minute infusion of 20 mL of iodipamide meglumine 52% (Cholografin; Bracco Diagnostics, Princeton, NJ) diluted in 80 mL of normal saline. The liver was imaged in high-speed mode with a four-section multi–detector row CT scanner (LightSpeed LX/i; GE Medical Systems, Milwaukee, Wis) and 2.5-mm collimation 15 minutes after completion of the infusion and during a single breath hold. All patients were observed for reactions to the contrast agent, and none occurred. Images were reconstructed at 1.25-mm intervals with a reduced field of view.
It should be noted that CT cholangiography has been evaluated in studies in the United States (18–20) and has been used extensively in Asia (21–26) and Europe (4,27–29). In addition, intravenous iodipamide meglumine is approved by the United States Food and Drug Administration for imaging the biliary tract. For these reasons, institutional review board approval for use of this contrast agent in clinical examinations of the biliary tract was not required.
MR Cholangiography
With a 1.5-T MR imaging unit (Gyroscan Intera, software version 8.37; Philips, Best, the Netherlands) and a phased-array torso surface coil, the following MR sequences were performed through the biliary tract: a coronal breath-hold single-shot rapid acquisition with relaxation enhancement (RARE) sequence (repetition time msec/echo time msec, 
/80; matrix, 256 x 144–192; echo train length, 91; section thickness, 4 mm; gap, 0 mm; field of view, 380 mm), a coronal respiratory-triggered volumetric RARE sequence (1,800/650 [effective]; matrix, 256 x 191; echo train length, 122; section thickness, 0.9 mm; gap, 0 mm; field of view, 280 mm), and five oblique coronal thick-slab RARE sequences (/1,200; matrix, 512 x 256; echo train length, 256; section thickness, 4–7 cm; gap, 0 mm; field of view, 250 mm) in a radial array.
Then, after slow (1-minute) intravenous administration of 5 µmol/kg of mangafodipir trisodium (Teslascan; Nycomed, Princeton, NJ), the following T1-weighted MR sequences were performed through the biliary tract at delays of 10, 25, and 40 minutes: coronal and transverse gradient-echo sequences (6.8/2.1; flip angle, 25°–40°; matrix, 512 x 94–128; section thickness, 1.0–1.5 mm; gap, 0 mm; field of view, 250–300 mm) and a coronal gradient-echo sequence (5.3/1.5; flip angle, 40°; matrix, 512 x 204–256; section thickness, 1.0 mm; gap, 0 mm; field of view, 250–300 mm). Parallel acquisition imaging with a dual-coil array was used in all transverse sequences (30). All breath-hold sequences had imaging times of 35 seconds or less. All living potential liver donors in our series were healthy and not known to have cardiac or pulmonary disease, and none had difficulty with breath holding.
Image Interpretation
Maximum-intensity-projection and volume-rendered reformations of the CT and MR cholangiograms were obtained by using a dedicated workstation (Advantage Windows 4.0 or 3.1; GE Medical Systems). Two radiologists with subspecialty training in abdominal imaging and 5 and 7 years of experience in interpreting these types of images (B.M.Y. and B.T., respectively) independently reviewed all source and reconstructed CT and MR images at a picture archiving and communication system, or PACS, workstation (Agfa, Mortsel, Belgium).
The CT cholangiograms, conventional MR cholangiograms, and excretory MR cholangiograms were first evaluated separately at different sessions to reduce recall bias (with a 2-week time difference between the reading session for the conventional MR cholangiograms and that for the excretory MR cholangiograms), without reference to results of other imaging studies. The conventional and excretory MR cholangiograms were then evaluated together at a separate session (with a 3-month time interval between the reading session for the excretory MR cholangiograms and that for the combined conventional and excretory MR cholangiograms).
A total of 12 biliary segments were scored for each examination as follows: The common (one segment), cystic (one segment), first-order branch (two segments), and second-order branch (four segments) bile duct segments were scored by using a four-point scale in which a score of 0 indicated that the segment was not seen; a score of 1, that the segment was faintly seen; a score of 2, that the segment was well seen but the confluence or a portion of the duct was not seen; and a score of 3, that there was excellent visualization of the segment from its proximal commencement to its distal confluence. The visualization of the third-order branch (four segments) for each second-order branch segment was also scored by using a four-point scale in which a score of 0 indicated that the tertiary branches were not seen; a score of 1, that tertiary branches were faintly seen; a score of 2, that one or two tertiary branches were clearly seen; and a score of 3, that tertiary branches were well seen.
The biliary tract anatomic variants initially not evaluated in the independent review sessions by B.M.Y. and B.T. were subsequently evaluated by them in consensus 5 months later. The CT cholangiograms, conventional MR cholangiograms, and mangafodipir trisodium–enhanced excretory MR cholangiograms were evaluated separately at different sessions without reference to results of other imaging studies; conventional and excretory MR cholangiograms were then evaluated together in a different session.
Second-order biliary tract anatomy was classified as conventional or variant. Conventional biliary tract anatomy was defined as an arrangement in which the right posterior duct (which drains liver segments VI and VII) drained into the right anterior duct (which drains liver segments V and VIII) to form a right hepatic duct that then joined the left hepatic duct (formed by ducts draining liver segments II, III, and IV) (6). All other anatomic configurations were considered variant.
For the nine patients (six of whom had undergone CT and three of whom had undergone conventional and excretory MR cholangiography) who proceeded to liver donation, imaging findings of biliary anatomy were compared with findings at intraoperative cholangiography (n = 3) or, when intraoperative cholangiography was not performed, with surgical findings at both the liver retrieval (donor) and the liver transplantation (recipient) procedures (n = 6).
Statistical Analysis
Statistical analyses were performed by using the Stata software package, version 7.0 (Stata, College Station, Tex). Interobserver agreement for biliary tract visualization scores was determined with the weighted statistic. Interobserver agreement was classified as follows: A value of 0.00–0.20 was considered to indicate poor agreement; a value of 0.21–0.40, fair agreement; a value of 0.41–0.60, moderate agreement; a value of 0.61–0.80, good agreement; and a value of 0.81–1.00, excellent agreement.
The visualization scores for the biliary tract branches (first-, second-, and third-order) were compared between CT and MR imaging and between MR techniques by using generalized estimating equation models to account both for clustering of multiple biliary branch visualization scores within the same patient and for multiple readers (31). A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
, 0.76), conventional MR cholangiography (weighted , 0.66), and mangafodipir trisodium–enhanced excretory MR cholangiography (weighted , 0.79) (Table 1). Examples of images from CT, conventional MR, and excretory MR cholangiography are shown in Figures 1–3. Visualization scores for the common, cystic, and first-order branch (ie, right and left main) ducts were excellent for both readers with all three modalities, without significant differences between modalities. The scores for visualization of first-, second-, and third-order biliary tract branches are summarized in Figure 4.
|
|
|
|
|
Second-order biliary tract branching anatomy findings at CT, conventional MR, and excretory MR cholangiography are summarized in Table 2. In cases in which biliary tract anatomy was discernible at both conventional MR cholangiography alone and excretory MR cholangiography alone (n = 5), determinations of biliary tract branching anatomy were in agreement for these modalities without discrepancy. Combined viewing of conventional and excretory MR cholangiograms enabled determination of biliary anatomy in seven of eight patients.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Our findings are similar to those of Carlos et al (32), who also achieved significantly improved biliary tract visualization with use of combined conventional and excretory MR cholangiography as compared with use of either modality alone. Carlos et al (32) rated bile duct visualization in 10 patients with a five-point scale, with a score of 0 representing no visualization and a score of 4 representing excellent visualization. In the study of Carlos et al (32), the mean second-order duct visualization score was 0.9 at conventional MR cholangiography and 1.4 at gadolinium ethoxybenzyl diethylenetriaminepentaacetic acid–enhanced excretory MR cholangiography alone but was 2.3 at combined conventional and excretory MR cholangiography.
Our results also support the results of several authors who express caution with regard to the consistency of conventional MR cholangiography in enabling adequate visualization of biliary anatomy in living potential liver donors (3,11,33). Lee et al (12) recommended mangafodipir trisodium–enhanced excretory MR cholangiography as a useful supplement to conventional T2-weighted MR cholangiography to improve biliary visualization after observing promising results in 10 living potential liver donors, one of whom had intraoperative confirmation of biliary anatomic findings at mangafodipir trisodium–enhanced MR cholangiography.
A subsequent report by a different research group (13) described concordance in findings of conventional and variant biliary tract anatomy in eight living potential liver donors between mangafodipir trisodium–enhanced excretory MR cholangiography and intraoperative cholangiography. In that study, all patients also underwent conventional MR cholangiography, but distinction between the use of conventional MR cholangiography and the use of mangafodipir trisodium–enhanced excretory MR cholangiography was not made (13).
CT cholangiography has been described for the evaluation of living potential liver donors. Cheng et al (14) reported results in 16 living potential liver donors, of whom three underwent endoscopic retrograde cholangiography and seven underwent intraoperative cholangiography; the results confirmed CT cholangiographic findings of biliary tract anatomy (ie, that there were eight cases of conventional and two cases of anomalous biliary tract anatomy) in all cases. Schroeder et al (4) also observed excellent results in 16 living potential liver donors evaluated with same-day multisection CT cholangiography and CT angiography. In their study, intraoperative confirmation of CT cholangiographic biliary tract anatomic findings was available for four patients (4).
Our finding that biliary tract anatomy was fully visualized in eight living potential liver donors at CT cholangiography, with concordance between CT cholangiographic and surgical findings in five of six patients, conservatively supports the results of these previous studies. However, our finding of one case of discrepancy between biliary tract anatomy as determined at CT cholangiography and biliary tract anatomy as determined with intraoperative findings raises the concern that CT may have limitations in revealing biliary branching variants when bile ducts course close together. Improvements in multi–detector row CT technology, with the use of a section thickness thinner than 2.5 mm, may improve the performance of CT cholangiography; further work is needed to define the role of CT in identifying biliary variants.
Intravenous cholangiography is rarely performed in North America partly because of the perceived high risk of contrast agent reactions (34,35). However, in several studies of CT cholangiography, minor contrast agent reactions were observed in 1%–3% of patients, a rate similar to that at conventional intravenous contrast material–enhanced CT (36–39). The low frequency and mild severity of reactions in these studies may reflect the use of slow contrast agent infusion rates, as well as the premedication of patients with intravenous diphenhydramine (19,36). Further experience will help us to assess this potential risk of CT cholangiography. The use of oral contrast agents at CT cholangiography does not appear to be a sufficiently robust alternative to the use of intravenous contrast material: In a study involving five healthy volunteers examined with CT cholangiography after administration of an oral contrast agent, first-order biliary branches could not be visualized in one volunteer, and third-order branches could not be visualized in five volunteers (20).
The administration of intravenous morphine to contract the sphincter of Oddi (15,40,41) is widely performed before scintigraphic studies of the gallbladder (16,17). The use of morphine may have improved the visualization of bile ducts in our study, but this would not be a confounding factor in the comparison of modalities because morphine was used in both the CT and MR imaging patient groups.
Our study was limited by its small sample size and by the examination of different patient groups with CT and MR imaging. Despite these limitations, our findings suggest that normal-caliber bile ducts of living potential liver donors are significantly better visualized at excretory CT cholangiography than at conventional or excretory MR cholangiography either alone or in combination. This is likely because of the higher spatial resolution of CT as compared with MR imaging. Additional advantages of CT over MR imaging include the ease, speed, and lower cost of CT examinations. However, the use of iodinated contrast material and ionizing radiation are drawbacks of CT. Our study was also limited by the small number of patients (n = 9) for whom intraoperative confirmation of CT or MR cholangiographic findings was avaliable. Evaluation of a larger series of patients is necessary to determine if CT or MR cholangiography can accurately and consistently depict variant biliary tract anatomy depicted at conventional cholangiography.
In conclusion, in the evaluation of living potential liver donors, multi–detector row CT cholangiography enables significantly better biliary tract visualization than conventional or mangafodipir trisodium–enhanced excretory MR cholangiography either alone or in combination. In patients evaluated with MR imaging, use of mangafodipir trisodium–enhanced excretory MR cholangiography in conjunction with conventional MR cholangiography, as compared with the use of conventional MR cholangiography alone, significantly improves visualization of second-order biliary tract branches. Further work to determine the accuracy of these examinations for delineating second-order biliary tract anatomy is needed.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, B.M.Y.; study concepts, B.M.Y., R.S.B., F.V.C., B.T., J.P.R.; study design, B.M.Y., R.S.B., F.V.C., B.T.; literature research, B.M.Y., R.S.B.; clinical studies, B.M.Y., R.S.B., F.V.C., A.Q., J.P.R.; data acquisition and analysis/interpretation, B.M.Y., B.T.; statistical analysis, B.M.Y.; manuscript preparation, definition of intellectual content, and editing, B.M.Y., R.S.B., F.V.C., J.P.R.; manuscript revision/review, all authors; manuscript final version approval, B.M.Y., R.S.B., F.V.C.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. I. Sainani, O. A. Catalano, N.-S. Holalkere, A. X. Zhu, P. F. Hahn, and D. V. Sahani Cholangiocarcinoma: Current and Novel Imaging Techniques RadioGraphics, September 1, 2008; 28(5): 1263 - 1287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Breiman, F. V. Coakley, E. M. Webb, J. J. Ellingson, J. P. Roberts, J. Kohr, J. Lutz, N. Knoess, and B. M. Yeh CT Cholangiography in Potential Liver Donors: Effect of Premedication with Intravenous Morphine on Biliary Caliber and Visualization Radiology, June 1, 2008; 247(3): 733 - 737. [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]
|
|
|
|
|
|
B. M. Yeh, F. V. Coakley, A. C. Westphalen, B. N. Joe, C. E. Freise, A. Qayyum, R. A. McTaggart, and J. P. Roberts Predicting Biliary Complications in Right Lobe Liver Transplant Recipients according to Distance between Donor's Bile Duct and Corresponding Hepatic Artery Radiology, January 1, 2007; 242(1): 144 - 151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. K. An, J. M. Lee, K.-S. Suh, N. J. Lee, S. H. Kim, Y. J. Kim, J. K. Han, and B. I. Choi Gadobenate dimeglumine-enhanced liver MRI as the sole preoperative imaging technique: a prospective study of living liver donors. Am. J. Roentgenol., November 1, 2006; 187(5): 1223 - 1233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Schroeder, A. Radtke, H. Kuehl, J. F. Debatin, M. Malago, and S. G. Ruehm Evaluation of Living Liver Donors with an All-inclusive 3D Multi-Detector Row CT Protocol Radiology, March 1, 2006; 238(3): 900 - 910. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. M. Merkle, P. A. Haugan, J. Thomas, T. A. Jaffe, and C. Gullotto 3.0- Versus 1.5-T MR Cholangiography: A Pilot Study Am. J. Roentgenol., February 1, 2006; 186(2): 516 - 521. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kitami, K. Takase, G. Murakami, S. Ko, M. Tsuboi, H. Saito, S. Higano, Y. Nakajima, and S. Takahashi Types and Frequencies of Biliary Tract Variations Associated with a Major Portal Venous Anomaly: Analysis with Multi-Detector Row CT Cholangiography Radiology, January 1, 2006; 238(1): 156 - 166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Kim, A. Y. Kim, S. S. Hong, M.-H. Kim, J. H. Byun, H. J. Won, Y. M. Shin, P. N. Kim, H. K. Ha, and M.-G. Lee Biliary Ductal Evaluation of Hilar Cholangiocarcinoma: Three-dimensional Direct Multi-Detector Row CT Cholangiographic Findings versus Surgical and Pathologic Results--Feasibility Study Radiology, December 1, 2005; 238(1): 300 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. B. Macdonald, M. A. Haider, K. Khalili, T. K. Kim, M. O'Malley, P. D. Greig, D. R. Grant, G. Lockwood, and M. S. Cattral Relationship Between Vascular and Biliary Anatomy in Living Liver Donors Am. J. Roentgenol., July 1, 2005; 185(1): 247 - 252. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. J. Wang, B. M. Yeh, J. P. Roberts, R. S. Breiman, A. Qayyum, and F. V. Coakley Living Donor Candidates for Right Hepatic Lobe Transplantation: Evaluation at CT Cholangiography--Initial Experience Radiology, June 1, 2005; 235(3): 899 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Chen, B. M. Yeh, Z. J. Wang, J. P. Roberts, R. S. Breiman, A. Qayyum, and F. V. Coakley Concordance of Second-Order Portal Venous and Biliary Tract Anatomies on MDCT Angiography and MDCT Cholangiography Am. J. Roentgenol., January 1, 2005; 184(1): 70 - 74. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
