| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Kurashiki Central Hospital, Kurashiki 710-8602, Japan. From the 1997 RSNA scientific assembly. Received January 6, 1999; revision requested March 5; final revision received June 21; accepted July 21. Address reprint requests to Y.W. (e-mail: yw5904@kchnet.or.jp).
 |
Abstract |
|---|
 TOP Abstract IntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: In 234 patients (102 men, 132 women; age range, 25–80 years), MRCP images obtained by using a single-shot turbo spin-echo sequence were reviewed to assess pseudo-obstruction of the extrahepatic bile duct caused by vascular compression. Dual-phase spiral computed tomography, contrast material–enhanced three-dimensional MR angiography, and/or digital subtraction angiography also were performed to determine the vessel that caused the pseudo-obstruction.
RESULTS: Thirty-six pseudo-obstructions due to vascular compression were found in 33 (14%) patients. The common hepatic duct (27 [75%] sites) was the most common pseudo-obstruction site, followed by the left hepatic duct (four [11%] sites), proximal common bile duct (three [8%] sites), and right hepatic duct (two [6%] sites). The causative vessels were identified as the right hepatic artery at 24 (67%) sites; gastroduodenal artery, two (6%) sites; cystic artery, two (6%) sites; proper hepatic artery, one (3%) site; and an unspecified branch of the common hepatic artery, seven (19%) sites.
CONCLUSION: At MRCP, pseudo-obstruction of the extrahepatic bile duct can be caused by pulsatile vascular compression of the hepatic and gastroduodenal arteries, and it should not be misdiagnosed as a bile duct tumor or biliary stone.
Index terms: Bile ducts, abnormalities, 768.288, 768.291, 768.31, 768.32 • Bile ducts, MR, 768.121411, 768.121416, 768.12142, 768.12143 • Computed tomography (CT), comparative studies, 76.1211 • Hepatic arteries • Magnetic resonance (MR), artifact, 768.121411, 768.121416, 768.12142, 768.12143 • Magnetic resonance (MR), comparative studies, 76.1214
|
Introduction |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In all 234 patients, MR imaging was performed with a 1.5-T MR imaging unit (Gyroscan ACS-NT; Philips Medical Systems, Best, the Netherlands). After localization images were obtained, transverse fat-suppressed turbo spin-echo T1- (repetition time msec/echo time msec, 500/18) and T2-weighted (1,800-2,000/100) images and transverse heavily T2-weighted turbo spin-echo images (6,000/350) were obtained with use of a body coil.
MRCP was performed by using a half-Fourier single-shot turbo spin-echo sequence with a 20-cm circular surface coil to obtain a high signal-to-noise ratio and high spatial resolution. The imaging parameters for MRCP were as follows: 
/400 (effective); echo train length, 128; field of view, 220 mm; section thickness, 4 mm; 18 sections; section overlap, 1 mm; matrix, 205 x 256; and one signal acquired. The imaging time was 18 seconds, which permitted a single breath hold. Two sets of coronal images were obtained for MRCP; one of these sets was obtained with fat suppression by using a spectral presaturation with inversion-recovery pulse, and the other was obtained without using a fat suppression technique. The coronal images obtained with fat suppression were compressed into composite MRCP images by using a MIP algorithm. Oblique MRCP images were reconstructed at 15° intervals from the frontal to lateral view. The coronal images obtained without fat suppression were used as reference images for interpretation of the MIP reconstructed MRCP images, because non–fat-suppressed images were less sensitive to susceptibility artifact from metallic surgical clips and gas in the stomach and duodenum.
Three radiologists (Y.W., M.D., T.I.), who were blinded to the patients' medical history and final diagnosis, reviewed the MRCP images, including the MIP reconstructed MRCP images and two sets of coronal source images obtained with and without fat suppression, with consensus regarding the presence of a pseudo-obstruction of the extrahepatic duct due to artifact from arterial pulsatile compression. The findings at drip-infusion cholangiography and/or ERCP were used as the standards of reference. The diagnosis and site of pseudo-obstruction of the extrahepatic bile duct caused by vascular compression were established by the three radiologists. The extrahepatic bile ducts were defined to include the right, left, and common hepatic ducts and the common bile duct. The criteria used for diagnosis of pseudo-obstruction were (a) a focal stenosis or obstruction of the extrahepatic duct seen on MIP reconstructed MRCP images but not on drip-infusion cholangiographic or ERCP images, (b) minimal or no dilatation of the upstream biliary tree relative to the lower biliary tree, and (c) a vascular structure seen traversing the extrahepatic duct at the site of the focal stenosis or obstruction on the coronal source images obtained without fat suppression.
Other possible causes of pseudo-obstruction, such as duodenal gas, metallic surgical clips, and intravascular metallic coils, were excluded on the transverse fat-suppressed T1- and T2-weighted images (17). A diagnosis of pseudo-obstruction of the extrahepatic duct due to vascular compression was made in 33 patients.
To determine the vessel that caused the pseudo-obstruction of the extrahepatic duct, the dual-phase spiral CT, contrast-enhanced, three-dimensional MR angiographic, and DSA images obtained for clinical reasons were reviewed by four radiologists (Y.W., M.D., T.I., A.O.). The radiologists identified the causative vessel by means of consensus with the MRCP images that were available for correlation. In the 33 patients with a diagnosis of pseudostenosis caused by vascular compression, dual-phase spiral CT, contrast-enhanced, three-dimensional MR angiography, and DSA were performed in 33, nine, and four patients, respectively, within 1 month after MRCP (18,19). Of the 33 patients, three underwent dual- phase spiral CT, contrast-enhanced, three-dimensional MR angiography, and DSA for preoperative evaluation and/or differentiation between benign and malignant lesions; six underwent both spiral CT and MR angiography for preoperative evaluation and/or differentiation between benign and malignant lesions; one underwent both dual-phase spiral CT and DSA for differentiation between benign and malignant lesions; and 23 underwent only dual-phase spiral CT.
Dual-phase spiral CT was performed during a power injection of 100 mL of either iopamidol 300 (Iopamiron 300; Schering, Osaka, Japan) or iohexol 300 (Omnipaque 300; Daiichi Pharmaceutical, Tokyo, Japan) at 2.5–3.0 mL/sec (19). Arterial dominant phase CT scanning (120 kVp, 220–280 mA) began 30 seconds after the start of the contrast material injection, at the dome of the liver and proceeded in a caudal direction to the pancreatic head. A variable pitch of up to 1.5 and 5-mm collimation were used to scan the entire liver and biliary tree during a single breath hold. Transverse images were reconstructed every 5 mm with 5-mm collimation. The equilibrium phase scanning sequence was initiated 120 seconds after the administration of contrast material.
Contrast-enhanced, three-dimensional MR angiography was performed with a fast field-echo sequence (8.0/2.9, 35° flip angle, 370-mm field of view, 5–8-mm section thickness, 2.5–4.0-mm section overlap, 20–30 sections, 198 x 256 matrix, one signal acquired, 14–24-second imaging time) in a coronal plane (18). No cardiac gating or triggering was used. Ten to 15 seconds after the start of an intravenous injection of 0.1 mmol/kg gadopentetate dimeglumine (Magnevist; Schering) (dose range, 7–18 mL) in 5 seconds, the acquisition of five consecutive breath-hold images, obtained in 10-second intervals, was initiated; imaging was performed during the arterial and portal venous phases.
DSA was performed to examine both the celiac trunk and superior mesenteric arteries. Dual-phase spiral CT, contrast-enhanced, three-dimensional MR angiographic, and DSA images were evaluated to determine the artery that caused the pseudo-obstruction of the extrahepatic bile duct.
|
RESULTS |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
Oblique-projection MIP reconstructed MRCP images revealed the aspect of the extrahepatic bile duct that was compressed by the causative artery at the site of pseudo-obstruction. At the 27 sites of the common hepatic duct, the posterior (n = 25 [93%]) and anterior (n = 2 [7%]) aspects were shown to be compressed. At all four (100%) sites of the left hepatic duct, the left posterior aspect was shown to be compressed. At all three (100%) sites of the proximal common bile duct, the pseudo-obstruction was present in the right anterior aspect. At the two sites of the right hepatic duct, compression was found in the posterior aspect at one (50%) site and in the anterior aspect in one (50%) site.
In all 33 patients with a pseudo-obstruction, dual-phase spiral CT, contrast-enhanced, three-dimensional MR angiography, and/or DSA were used to determine the causative vessel of pseudostenosis at the 36 pseudo-obstruction sites. The causative arteries were identified as the right hepatic artery at 24 (67%) of the 36 pseudo-obstruction sites, the gastroduodenal artery at two (6%) sites, the cystic artery at two (6%) sites, the proper hepatic artery at one (3%) site, and an unspecified branch of the common hepatic artery at seven (19%) sites. At the 27 sites of the common hepatic duct, the right hepatic artery at 21 (78%) sites, cystic artery at one (4%) site, and proper hepatic artery at one (4%) site were identified as the causative artery. At four (15%) of the 27 sites, the causative artery was identified as a nonspecified branch of the common hepatic artery. At the four sites of the left hepatic duct, the causative artery was identified as the right hepatic artery at two (50%) sites and as an unspecified branch of the common hepatic artery at two (50%) sites. At the two sites of the right hepatic duct, the causative artery was identified as the right hepatic artery at one (50%) site and as the cystic artery at one (50%) site. At the three sites of the proximal common bile duct, the causative artery was identified as the gastroduodenal artery at two (67%) sites and as an unspecified branch of the common hepatic artery at one (33%) site.
|
DISCUSSION |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
There are criteria to identify pseudo-obstruction due to artifact from arterial pulsatile compression: (a) a focal stenosis or obstruction of the extrahepatic duct seen on MIP reconstructed MRCP images but not on drip-infusion cholangiographic or ERCP images, (b) minimal or no dilatation of the upstream biliary tree relative to the lower biliary tree, and (c) a vascular structure seen traversing the extrahepatic duct at the site of the focal stenosis or obstruction on the coronal source images obtained without fat suppression. In addition, we suggest that the artifact is secondary to pulsatility, because the portal vein rarely causes pseudo-obstruction despite its close anatomic relationship with the extrahepatic bile duct.
In 33 (14%) of the patients in our series, a pseudo-obstruction of the extrahepatic bile duct caused by artifact from arterial pulsatile compression was present; however, the patients were not consecutive. The most common site of pseudo-obstruction was the common hepatic duct, followed by the left hepatic duct, proximal common bile duct, and right hepatic duct. The artery that most commonly caused pseudo-obstruction was the right hepatic artery, followed by the gastroduodenal artery, cystic artery, and proper hepatic artery. The right hepatic artery frequently passes immediately posteriorly to the proximal portion of the common hepatic duct and can create extrinsic compression of the duct. However, the right hepatic artery has variations in its course that cause pseudo-obstruction of not only the common hepatic duct but also the right and left hepatic ducts (20). The accessory right hepatic artery that branches from the superior mesenteric artery may cause a pseudo-obstruction of the common bile duct. When the right hepatic artery crosses dorsally to the portal vein, which occurs rarely, the common hepatic duct may show a mild compression caused by the portal vein, as reported by Holzknecht et al (16).
The length and degree of a pseudo-obstruction of the extrahepatic duct may depend on not only the diameter of the causative artery but also the direction in which the causative artery crosses the extrahepatic bile duct. When the causative arteries, including the right hepatic and cystic arteries, are dilated to supply malignant neoplasms such as hepatocellular carcinoma and gallbladder cancer, the pseudo-obstruction may be severe. In addition, obese patients may be less likely to have pseudo-obstructions: Adipose tissue surrounding the bile duct may work to absorb pulsatile compressive pressure, and this is more likely to occur in obese patients than in thin patients. Other factors may include the intensity of arterial pulsation and the imaging parameters used in MRCP, such as pulse sequence, fat suppression, section thickness, or multiple versus single sections. Chemical-selective fat suppression techniques especially may make MRCP images vulnerable to pulsatile compression (14,17). The caliber of the extrahepatic bile duct does not seem to be associated with pseudo-obstruction: In our study, pseudo-obstruction was seen in both dilated and normal extrahepatic ducts.
The extent of pseudo-obstruction on MIP reconstructed MRCP images can be overestimated compared with the degree of pseudo-obstruction physiologically because of MIP reconstruction, susceptibility, and/or motion artifacts. In contrast, ERCP rarely depicts arterial compression of the extrahepatic bile duct, because the biliary tree can be depicted as more dilated due to the injection pressure of contrast material, compared with its physiologic state.
The differentiation between pseudo-obstruction due to arterial pulsatile compression artifact and true pathologic stenosis or obstruction requires careful interpretation of the coronal source images and transverse T2-weighted images. Coronal source images obtained without using a chemical-selective fat suppression technique can show a vascular structure crossing the extrahepatic bile duct at the site of the pseudo-obstruction. However, there may be some cases in which ERCP or drip-infusion cholangiography is necessary to exclude true stenosis or obstruction. The absence of clinical symptoms caused by the pseudo-obstruction may exclude the misdiagnosis of a biliary pathologic entity. Dual-phase spiral CT or contrast-enhanced, three-dimensional MR angiography also are useful methods for preventing misdiagnoses and identifying the causative artery.
In conclusion, knowledge of the existence and high prevalence of pseudo-obstruction of the extrahepatic bile duct due to artifact from arterial pulsatile compression should preclude the misinterpretation of MRCP images.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Author contributions: Guarantor of integrity of entire study, Y.W.; study concepts and design, Y.W., M.D.; definition of intellectual content, Y.W., M.D., T.I.; literature research, Y.W., M.D., T.I.; clinical studies, all authors; data acquisition and analysis, all authors; manuscript preparation, editing, and review, Y.W., M.D., T.I.
|
References |
|---|
|
TOPAbstractIntroductionMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Sugita, E. Sugimura, M. Itoh, T. Ohisa, S. Takahashi, and N. Fujita Pseudolesion of the Bile Duct Caused by Flow Effect: A Diagnostic Pitfall of MR Cholangiopancreatography Am. J. Roentgenol., February 1, 2003; 180(2): 467 - 471. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
