| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, University of Ulsan, Asan Medical Center,
388-1 Poongnap-dong, Songpa-ku, Seoul, 138-736, Korea.
2 Department of Surgery, University of Ulsan, Asan Medical Center, Songpa-ku,
Seoul, 138-736, Korea.
Received July 3, 2002;
accepted after revision February 11, 2003.
Address correspondence to T. K. Kim.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Seventy-six patients who underwent living related liver transplantation were evaluated with gadolinium-enhanced 3D MR angiography. All MR angiograms were assessed for patency of the hepatic artery and the portal vein using a four-point scale (grades I–IV). The results were correlated with conventional angiography (n = 23) and clinical follow-up with Doppler sonography (n = 53) for more than 6 months.
RESULTS. Seventy-three of 76 MR angiography procedures were technically adequate. When grades III (focal narrowing [> 50%] at the anastomotic site) and IV (abrupt cutoff at the anastomotic site with nonvisualization of the right [or left] hepatic artery distal to the anastomosis) were regarded as the diagnostic criteria for hepatic artery stenosis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 74%, 29%, 100%, and 77%, respectively. In the portal vein, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 84%, 35%, 100%, and 85%, respectively, when grades III (narrowing [> 50%] without poststenotic dilatation) and IV (narrowing [> 50%] with poststenotic dilatation) were defined as criteria for portal vein stenosis.
CONCLUSION. MR angiography was sensitive but not specific in the detection of significant vascular stenosis after living related liver transplantation. However, normal MR angiography findings reliably exclude the possibility of significant stenosis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The purpose of this study was to evaluate the efficacy of gadolinium-enhanced dual-phase 3D MR angiography for detecting vascular complications in patients who have undergone living related liver transplantation.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The results of conventional angiography and follow-up Doppler sonography and clinical findings (i.e., laboratory findings including liver function tests and physical examinations) were used as the standard of reference. Patients with negative findings on conventional angiography or with no evidence of vascular complications on follow-up clinical and Doppler sonography for at least 6 months were regarded as having no vascular complications.
MR Angiography
MR angiography was performed with a 1.5-T scanner (Magnetom Vision,
Siemens, Erlangen, Germany). A phased array torso coil or body coil was used
in all patients. At first, peak arterial enhancement time was calculated using
a test bolus method. The test bolus injection was performed with 1 mL of
gadopentetate dimeglumine (Magnevist, Berlex Laboratories, Wayne, NJ) followed
by a 30-mL saline flush. Using a test bolus injection, we chose a single-level
axial image that best showed the porta hepatis, and sequential axial images
were obtained at 1-sec intervals until 30 sec from the start of the gadolinium
injection. A region of interest was positioned in the aorta at this level, and
a time–intensity curve was obtained to determine the peak arterial
enhancement time and the optimal injection-to-scanning delay time. A scanning
delay was calculated according to a method described by Prince et al.
[14].
Three-dimensional MR angiography was performed in the coronal plane using fast imaging with a steady-state free precession sequence during end-inspiratory breath-holding. We used two different versions of the fast imaging with a steady-state free precession sequence during the study period. The parameters of contrast-enhanced 3D MR angiography were as follows: TR/TE, 5/2 or 4.6/1.8; section thickness, 3 or 3.5 mm; slab thickness, 96 or 112 mm; number of partitions, 24 or 32; matrix size, 150 x 256 or 200 x 512; field of view, 350 or 400 mm; and acquisition time, 19 or 23 sec. Thirty milliliters of gadopentetate dimeglumine were administered with an automatic power injector (MRS-50, Nemoto, Tokyo, Japan; or Spectris, Medrad, Pittsburgh, PA) at a rate of 4 mL/sec followed by a 10-mL saline flush. The arterial phase was obtained by subtracting the unenhanced image from the enhanced arterial image. The second phases were set with an interscan delay of 20 sec after the arterial phase to fit the portal venous and hepatic venous phases. Zero-filling along the z-axis was performed to obtain 64 reconstructed sections overlapping by 50%. The entire examination required approximately 40 min.
Conventional Angiography
Conventional angiography (n = 23) was performed within 1 week
(mean, 3.5 days). Digital subtraction angiography was performed with a
5-French selective visceral catheter by means of a common femoral arterial
approach. Selective catheterization of the celiac trunk and the superior
mesenteric artery was performed with injection of nonionic contrast medium
([iopamidol] Iopamiro 300, Bracco, Milano, Italy; or [iopromide] Ultravist
300, Schering, Berlin, Germany). In some cases, a vasodilator was administered
to improve visualization of the portal venous system.
Direct portography was performed in six patients. A branch of the right portal vein was punctured in four patients and a branch of the left portal vein in two with a 21-gauge needle under fluoroscopic guidance. The needle was exchanged for a 4-French coaxial dilator and a 6-French sheath combination included in the introducer system. Portal venograms were obtained, and the pressure was measured in the portal vein before and after the portal venous anastomosis.
Conventional angiography was performed only in selected patients who were suspected of having vascular complications on the basis of clinical features and MR angiography and Doppler sonography findings. Clinical features suggestive of vascular complications were as follows: hepatic failure with rapid clinical deterioration (n = 3); development of a delayed bile leak due to ischemic necrosis of bile ducts (n = 3); an unexplained persistent increase in liver enzymes and coagulopathy (n = 9); and signs and symptoms of portal hypertension, persistent ascites, or gastrointestinal hemorrhage from varices (n = 8). Serial Doppler sonography and clinical follow-up were also available for those patients who underwent conventional angiography.
Doppler Sonography and Clinical Follow-Up
The remaining 53 patients underwent serial Doppler sonography and clinical
follow-up for at least 6 months. Doppler sonography was performed by
experienced radiologists with a commercially available unit (HDI 3000, ATL,
Bothell, WA; or Sequoia, Acuson, Mountain View, CA) and a probe with
frequencies of 3–7 MHz. Gray-scale sonography, color Doppler sonography,
and spectral Doppler sonography were performed in all patients. The hepatic
artery and portal vein were examined for patency and characteristics of the
spectral waveforms. Doppler sonography was usually performed every 1–2
months after MR angiography. Doppler sonography criteria for hepatic artery
stenosis were a resistive index less than 0.5 and a systolic acceleration time
greater than 0.08 sec within the intrahepatic arteries or a focal systolic
velocity of more than 2 m/sec at the stenotic site of the hepatic artery
[15]. Doppler sonography
criteria for portal vein stenosis were at least three- or four-fold velocity
gradients across the anastomosis
[16,
17].
Image Analysis
All MR angiography images were retrospectively analyzed by two reviewers in
consensus to assess the patency of the transplanted hepatic artery and portal
vein. During the period of review, no information was provided by conventional
angiography, follow-up data, or Doppler sonography. The MR angiograms were
interpreted at a work-station by viewing raw data, reformatted images,
maximum-intensity-projection images, and the axial T1- and T2-weighted
sequences. Findings of the hepatic artery on MR angiography were classified as
grade I, normal (Fig. 2A,
2B); grade II, visualization of
right (or left) hepatic artery distal to the anastomosis with nonvisualization
of peripheral branches; grade III, focal narrowing (> 50%) at the
anastomotic site with visible right (or left) hepatic artery distal to the
anastomosis; and grade IV, abrupt cutoff at the anastomotic site with
nonvisualization of the right (or left) hepatic artery distal to the
anastomosis. Portal vein findings were classified as grade I, normal (Fig.
2A,
2B); grade II, mild narrowing
(< 50%); grade III, narrowing (> 50%) without poststenotic dilatation;
and grade IV, narrowing (> 50%) with poststenotic dilatation. Poststenotic
dilatation means a fusiform dilatation distal to the anastomotic site of the
portal vein without thrombosis.
|
|
The conventional angiograms were reviewed for the presence or absence of stenosis. Stenoses were interpreted as substantial if narrowing of the diameter of the hepatic artery was greater than 50%. On direct portography, a gradient of more than 5 mm Hg was considered a significant stenosis at the anastomotic site.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We analyzed the MR angiography findings that were technically adequate in 73 patients. Observers classified the MR angiography findings for the hepatic artery in 39 patients (53%) as grade I, 10 (14%) as grade II, 19 (26%) as grade III, and five (7%) as grade IV. Three patients (3/19, 16%) who were classified as grade III (Fig. 3A, 3B) and four patients (4/5, 80%), as grade IV (Fig. 4A, 4B) had hepatic artery stenosis. All patients who were classified as grade IV on MR angiography did not actually have complete occlusion but had a stenosis visualized on conventional angiography. When grade IV alone was regarded as the diagnostic criterion for hepatic artery stenosis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 57%, 99%, 80%, 96%, and 95%, respectively. When grades III and IV were defined as the diagnostic criteria for hepatic artery stenosis, the total number of lesions in the hepatic artery shown on MR angiography was 24, and the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 74%, 29%, 100%, and 77%, respectively (Table 1). Among the seven false-positive cases from the 23 MR angiograms with correlative conventional angiography for the hepatic artery, a direct comparison of findings on MR angiography and conventional angiography confirmed the presence of surgical clips in three (Fig. 5A, 5B).
|
|
|
|
|
|
|
Observers classified the MR angiography findings for the portal vein in 47 patients (64%) as grade I, nine (12%) as grade II, 12 (16%) as grade III, and five (7%) as grade IV. Three patients (3/12, 25%) with findings classified as grade III and three patients (3/5, 60%) with findings classified as grade IV (Fig. 6A, 6B) were found to have portal vein stenosis. When grades III and IV were defined as the diagnostic criteria for portal vein stenosis, the total number of lesions of the portal vein found on MR angiography was 17 (Table 2), and the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 84%, 35%, 100%, and 85%, respectively.
|
|
|
We also analyzed the overall results in 76 patients that included suboptimal examinations due to excessive patient motion. When grades III and IV were defined as the diagnostic criteria for hepatic artery stenosis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 74%, 28%, 100%, and 76%, respectively. When grades III and IV were defined as the diagnostic criteria for portal vein stenosis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of MR angiography were 100%, 83%, 33%, 100%, and 84%, respectively.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gadolinium-enhanced 3D MR angiography has recently become an alternative means of accurate, noninvasive, and rapid evaluation for hepatic vascular disease because image quality can be improved with thinner sections and shorter TRs, resulting in a faster scanning time and fewer motion artifacts from incomplete breath-holding. Improved temporal resolution and repeated sequences after contrast administration allow optimal and separate enhancement of the arteries and veins [21]. A growing body of literature supports the accuracy of 3D MR angiography compared with that of conventional angiography. There has been good agreement between conventional angiography and gadolinium-enhanced 3D MR angiography for the depiction of arterial and venous abnormalities, despite some limitations in small intrahepatic arteries [22]. MR angiography has been reported to have excellent results in the detection of hemodynamically significant stenosis in the renal and aortoiliac arteries (> 50%), although other researchers have found that MR angiography does a poor job of revealing mild stenosis [23]. MR angiography shows a higher sensitivity and accuracy than conventional angiography for the detection of thrombosis or the assessment of vessel patency in any part of the portal venous system [24]. Glockner et al. [13] reported that all cases of hepatic artery and portal vein thrombosis after liver transplantation were detected on MR angiography with no false-positive or false-negative interpretations. The study by Glockner et al. showed that the sensitivity of MR angiography for moderate to severe arterial stenosis after liver transplantation was 86%, with one false-negative and three false-positive interpretations and that there was a discrepancy in estimating the degree of stenosis.
In our study, we evaluated the utility of the criteria for detection of hepatic artery and portal vein stenoses in recipients who underwent living related liver transplantation. Our results indicate that gadolinium-enhanced 3D MR angiography is technically accessible and useful for the evaluation of vascular complications after living related liver transplantation. When an abrupt cutoff at the anastomotic site with nonvisualization of the right (or left) hepatic artery distal to the anastomosis was regarded as the diagnostic criterion for hepatic artery stenosis, the sensitivity and specificity of MR angiography were 57% and 99%, respectively. This finding may have been related to the small amount of contrast material flowing into the intrahepatic artery through the stenosis at the anastomotic site. When the diagnostic criteria for hepatic artery stenosis included focal narrowing at the anastomotic site with visible intrahepatic artery and abrupt cutoff at the anastomotic site with nonvisualization of the intrahepatic artery, we achieved 100% sensitivity and 74% specificity. The results of our study indicate that MR angiography tends to overestimate hepatic artery stenosis. This fact might be explained by the following factors. First, the susceptibility artifact from the surgical clips adjacent to the hepatic arterial anastomotic site creates focal signal loss of the hepatic artery, mimicking stenosis. Second, low-flow states can play an important role in creating false-positive findings of arterial disease. Low-flow states are caused by the inflow of small amounts of contrast material into the intrahepatic artery due to elevated end-organ resistance (i.e., recent transplanted edematous liver, graft rejection) or a decrease of systemic (i.e., systemic hypotension) or localized (i.e., splenic artery steal) blood volume [25]. Third, because of the small size of the lobar hepatic artery in living related liver transplantation grafts, subtle vascular signals could not be distinguished from background signals in a small vessel, thereby producing an overestimation of the amount of stenosis [26]. Therefore, we suggest that this MR angiography finding should be assessed together with the patient's clinical course, the natural history of the hepatic artery stenosis, and Doppler sonography findings before considering conventional angiography.
When narrowing (> 50%) of the portal vein was regarded as the diagnostic criterion for portal vein stenosis, we achieved 100% sensitivity and 84% specificity. MR angiography clearly depicts anastomotic narrowing of the portal vein. In addition, MR angiography may be performed to evaluate the extent and degree of the portosystemic collateral vessels resulting from portal hypertension. Our results show high sensitivity and specificity in the detection of more than 50% of the portal vein stenoses. However, we found a high false-positive rate with the results of portal vein stenosis. For this reason, it is important that a diagnosis of portal vein stenosis correlate with both the clinical findings and Doppler sonography. Most cases of portal vein stenosis are accompanied by symptoms of portal hypertension—that is, varices, upper gastrointestinal hemorrhage, splenomegaly, massive ascites, hepatic failure, and intestinal swelling [11]. It is important that the criteria used in Doppler sonography to identify patients with portal vein stenosis are direct gray-scale sonography depictions of portal vein narrowing in diameter and an acceleration of flow at the stricture (> 3–4 times the peak velocity of the nonstenotic portion) on Doppler sonography. In our study, patients with poststenotic dilatation on MR angiography had fewer false-positive findings than those without poststenotic dilatation. This fact suggests that poststenotic dilatation of the portal vein is caused by turbulent flow immediately distal to the stenosis because of long-standing severe stenosis.
Multidetector CT permits a more rapid acquisition of thinly collimated images than is possible with conventional CT. In addition, the high quality of 3D images can be obtained with volume data. In this regard, the use of 3D images and narrow collimation of section thickness will provide more valuable information about the vascular structure of the transplanted liver and allow detailed examination of the liver parenchyma and extrahepatic tissues. Multiplanar CT has been used to show the dilated pancreatic duct and common bile duct, but the normal or minimally dilated duct may not be visualized [27]. Although nonionic contrast agents have been used to opacify the minimally dilated biliary tree, their use is associated with an increased iodine load in patients [28].
MR imaging combined with MR cholangiography and MR angiography can be an effective diagnostic method in the postoperative workup of a patient who has undergone liver transplantation; provide a more comprehensive evaluation of the transplanted liver; reveal abnormalities of vascular structure; and depict bile ducts, liver parenchyma, and extrahepatic tissues. An important advantage of MR imaging is the low toxicity of its contrast agents; hence, MR angiography can be used, particularly in patients with renal insufficiency. Finally, the amount of radiation exposure associated with an imaging technique must be considered, particularly in young patients, when selecting an imaging modality for a patient who has undergone liver transplantation.
We recognize that our study has specific limitations. First, the definitive proof of patency or stenoses on conventional angiography was not obtained in all patients. Although normal findings on Doppler sonography do not entirely exclude vascular abnormalities because the clinical scenario evolves and the patients return for repeated Doppler sonography of the transplanted liver, it may become apparent that a more definitive diagnostic evaluation of the hepatic arterial system and the portal venous system should be undertaken. Therefore, negative findings on serial Doppler sonography and the absence of clinical symptoms during the follow-up period might prompt us to regard these patients as not having any hemodynamically significant vascular abnormalities. Second, our study was retrospective. We think that further prospective, blinded studies will be needed to determine the applied sensitivity and specificity of the discriminatory criteria selected in the retrospective review. Third, 30 mL of gadopentetate dimeglumine might be insufficient, resulting in the high number of false-positive findings. Using higher doses of IV contrast material may improve vascular signal intensity and diagnostic image quality in low-flow states and deserves further investigation. Fourth, the use of a 4-mL/sec rate with a 10-mL saline flush in our study might be less effective for imaging of the portal venous system. The use of a lower volume rate (2 mL/sec) with a larger volume (> 30 mL) of saline flush might have improved the portal venous phase images [29].
Assessment of the clinical course and Doppler sonography are the first steps for evaluation of vascular complications in patients who have undergone liver transplantation in most hospitals. We suggest a new algorithm using MR angiography that has a high sensitivity and an excellent negative predictive value for the detection of clinically significant vascular stenosis: If a patient has both normal Doppler sonography and clinical findings, a simple follow-up is required. If a patient has both abnormal Doppler sonography and abnormal clinical findings, conventional angiography is recommended. MR angiography is useful and rules out the possibility of vascular complications when Doppler sonography and clinical findings are discordant (Fig. 7).
|
In conclusion, gadolinium-enhanced 3D MR angiography can be a useful, noninvasive technique for evaluating vascular complications in patients who have undergone living related liver transplantation. MR angiography was sensitive but not specific for detection of clinically significant stenosis after living related liver transplantation. However, normal MR angiography findings reliably exclude the possibility of stenosis.
Acknowledgments
We thank Bonnie Hami, Department of Radiology, University Hospitals Health
System, Cleveland, OH, for editorial assistance in preparing this
manuscript.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  S. S. Lee, K. W. Kim, B. J. Park, Y. M. Shin, P. N. Kim, M.-G. Lee, and S. G. Lee Effect of Respiration on the Spectral Doppler Wave of the Right Hepatic Vein in Right Lobe Living Donor Liver Transplant Recipients J. Ultrasound Med., December 1, 2007; 26(12): 1723 - 1733. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. M. Caiado, R. Blasbalg, A. S. Z. Marcelino, M. da Cunha Pinho, M. C. Chammas, C. da Costa Leite, G. G. Cerri, A. C. de Oliveira, T. Bacchella, and M. C. C. Machado Complications of Liver Transplantation: Multimodality Imaging Approach RadioGraphics, September 1, 2007; 27(5): 1401 - 1417. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.-A. Clevert, M. Stickel, H. J. Michaely, F. Loehe, C. Graeb, H. O. Steitz, T. Strautz, P. Flach, S. O. Schoenberg, K. W. Jauch, et al. B-Flow Sonography for Detecting Portal Venous Stenosis After Liver Transplantation Journal of Diagnostic Medical Sonography, July 1, 2006; 22(4): 253 - 257. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
