| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Ohio State University Medical Center, 450 W. 10th
Ave., S-255 Rhodes Hall, Columbus, OH 43210.
2
Department of Gastroenterology, St. Paul's Hospital, University of British
Columbia, Vancouver, British Columbia, Canada.
3
Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline
Ave., Boston, MA 02215.
4
Department of Radiology, Lehigh Valley Hospital, Cedar Crest and 1-78, P. O.
Box 689, Allentown, PA 18105-1556.
5
Department of Radiology, Duke University Medical Center, Box 3189, Durham, NC
27710.
6
Department of Gastroenterology, Duke University Medical Center, Durham, NC
27710.
Received June 1, 2001;
accepted after revision August 9, 2001.
Presented at the annual meeting of the American Roentgen Ray Society, New
Orleans, May 1999.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Twenty patients with sclerosing cholangitis were evaluated on MR cholangiopancreatography using the single-shot fast spin-echo technique at 1.5 T. A group of 19 healthy volunteers underwent MR cholangiopancreatography as controls. Thick-slab (2-cm sections) coronal oblique and thin-slab (5-mm sections) interleaved straight coronal MR images were obtained. All patients with sclerosing cholangitis had an MR cholangiopancreatogram within 12 months of a contrast-enhanced cholangiogram (mean, 3.8 months). Seventy-five percent of patients had an MR cholangiopancreatogram within 3 months of the contrast-enhanced cholangiogram. The MR cholangiopancreatograms and contrast-enhanced cholangiograms were reviewed independently in a random fashion by two radiologists who were unaware of clinical history for the degree of ductal visualization and for the presence and location of strictures of the intrahepatic and extrahepatic bile ducts. All discrepancies were resolved by a consensus, and the contrast-enhanced cholangiograms were regarded as the gold standard. Statistically significant data were calculated using the signed rank test (p < 0.01), and agreement analysis was calculated using Cohen's kappa.
RESULTS. All findings on MR cholangiopancreatograms in healthy
subjects were interpreted as normal, and all findings on MR
cholangiopancreatograms in patients with sclerosing cholangitis were
interpreted as abnormal. When compared with the control group, scans of
patients with sclerosing cholangitis usually showed good visualization
(>50%) of the intrasegmental (86% vs 9%) and peripheral (67% vs 0%)
intrahepatic ducts on thick-slab MR cholangiopancreatography. Thick-slab MR
cholangiopancreatography showed good visualization in more ducts than contrast
cholangiography (84% vs 70%; p = 0.10) and showed more strictured
ducts than contrast cholangiography (47% vs 36%; p = 0.22). When
comparing those ducts with good visualization on both MR
cholangiopancreatography and contrast cholangiography, we found that
disagreement occurred regarding 32% of ducts. Most of the discrepancies (60%)
resulted when a stricture was noted on MR cholangiopancreatography but not on
contrast-enhanced cholangiography. Good interobserver agreement (
>
0.4) was noted for detecting strictures of the extrahepatic, left hepatic,
left medial, and right posterior ducts, with the greatest agreement for
extrahepatic ductal strictures ( = 0.8).
CONCLUSION. Thick-slab MR cholangiopancreatography is the best technique for depicting normal and strictured bile ducts and allows the differentiation of healthy patients from patients with sclerosing cholangitis. Although endoscopic retrograde cholangiopancreatography was considered the standard, MR cholangiopancreatography was superior for intrahepatic biliary ductal visualization. Therefore, this technique is of value in the diagnosis and follow-up of patients with sclerosing cholangitis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MR cholangiopancreatography is a relatively new imaging technique shown to be comparable to endoscopic retrograde cholangiopancreatography for the evaluation of extrahepatic ductal abnormalities [1,2,3,4,5]. However, many radiologists and gastroenterologists argue that its role in the evaluation of intrahepatic bile ducts is limited mainly because of decreased spatial resolution and nondistention of the ducts. Although some recent publications indicate that MR cholangiopancreatography may show intrahepatic ductal abnormalities in a comparable fashion to that of endoscopic retrograde cholangiopancreatography [1, 2, 6,7,8], further comparison study between MR cholangiopancreatography and direct cholangiography is needed before the utility of MR cholangiopancreatography for patient treatment can be assessed. The purpose of this study was to compare MR cholangiopancreatography and direct contrast cholangiography in patients with sclerosing cholangitis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the study group of 20 patients with sclerosing cholangitis, 13 were men and seven were women, ranging in age from 22 to 79 years (mean age, 48.5 years). The etiology of sclerosing cholangitis was primary in 12 and secondary in eight patients. Secondary causes of sclerosing cholangitis included bacterial cholangitis (n = 4; three postoperative strictures and one Oriental cholangiohepatitis), hepatic arterial stenosis after liver transplantation (n = 1), sclerosing cholangitis after liver transplantation not related to hepatic arterial stenosis (harvesting or recurrence) (n = 2), and cholangitis caused by fluroxidine after chemotherapy (n = 1). The duration of sclerosing cholangitis ranged from 1 day to 60 months with a mean of 13.5 months.
Of the 20 patients, five were inpatients who were being treated for acute exacerbation of sclerosing cholangitis. These five patients underwent contrast-enhanced cholangiography, serum analysis, and MR cholangiopancreatography during their hospitalization. Symptoms included jaundice (n = 4) and fever (n = 1).
In the 20 patients, bilirubin and alkaline phosphatase ranged from 0.4 to 22.8 mg/dL (mean, 3.64 mg/dL; normal, 2-1.2 mg/dL) and from 96 to 1387 U/L (mean, 518 U/L; normal, 30-135 U/L), respectively. In the 15 asymptomatic outpatients, bilirubin and alkaline phosphatase ranged from 0.4-7.6 mg/dL (mean, 1.6 mg/dL) and 87-1068 U/L (mean, 383.6 U/L), respectively. In the five inpatients who were evaluated during hospitalization for acute disease, bilirubin and alkaline phosphatase ranged from 1.4 to 22.8 mg/dL (mean, 9.2 mg/dL) and from 177 to 1029 U/L (mean, 584.7 U/L), respectively. In no patient was there worsening of the laboratory values or symptoms between the contrast-enhanced cholangiography and MR cholangiopancreatography.
A group of 19 healthy subjects with no history of abdominal symptomatology, pancreaticobiliary disease, or inflammatory bowel disease underwent MR cholangiopancreatography as controls. All subjects were men whose ages ranged from 25 to 60 years with a mean of 34 years. Bilirubin and liver enzymes were not obtained in these patients. In addition, they did not undergo endoscopic retrograde cholangiopancreatography. These volunteers underwent the same MR imaging protocol as the patients with sclerosing cholangitis.
MR Imaging Technique
All patients had MR cholangiopancreatography within 12 months of their most
recent contrast-enhanced cholangiography. The MR cholangiopancreatograms were
obtained in the supine position with a 1.5-T Signa MR imaging unit (General
Electric Medical Systems, Milwaukee, WI), using a torso phased array coil. No
oral or IV contrast agents were administered. A single-shot fast spinecho
technique was used for MR cholangiopancreatography. Multiple thick-slab (2-cm
sections) coronal oblique MR images were obtained during a breath-hold (TR/TE,
250/infinite; bandwidth, 32 MHz; field of view, 32 cm; echo-train length, 128;
echo spacing, 10.8 msec; matrix size, 256 x 256). In addition, thin-slab
(5-mm sections) interleaved straight coronal MR cholangiopancreatograms were
also obtained with similar parameters during one or two breath-holds. There
were no major software upgrades to the MR imaging scanner during the course of
the study.
The number of slices ranged from seven to eight for the thick-slice coronal oblique MR images and from 19 to 28 for the thin-slice straight coronal MR images. Because each image took 2 sec to acquire, the time of imaging for thick-slice MR cholangiopancreatography was 10-14 sec and for the thin-slice MR cholangiopancreatography was 38-56 sec (more than one breath-hold was necessary in several patients).
The time interval between contrast-enhanced cholangiography and MR cholangiopancreatography ranged from 1 day to 12 months, with a mean of 3.8 months. Fifteen (75%) of the 20 patients underwent MR cholangiopancreatography within 3 months of the contrast-enhanced cholangiography. The most recent contrast-enhanced cholangiograms in these patients, including endoscopic retrograde cholangiopancreatograms (n = 12), percutaneous transhepatic cholangiograms (n = 6), and T-tube cholangiograms (n = 2), were used as the gold standard.
Image Interpretation
Endoscopic retrograde cholangiopancreatograms were obtained in a standard
fashion by one of three gastroenterologists.
The MR cholangiopancreatograms and contrastenhanced cholangiograms of the 20 patients with sclerosing cholangitis and the MR cholangiopancreatograms of the 19 healthy subjects were reviewed using hard-copy film independently in a random fashion by two radiologists experienced in abdominal imaging who were unaware of clinical history. Although the radiologists were unaware of clinical history at the time of each image interpretation, they knew that the study involved the interpretation of images from both patients with sclerosing cholangitis and healthy subjects. The thick-slab and the thin-slab MR cholangiopancreatograms were interpreted separately to determine normal or abnormal findings; the degree of extrahepatic, central, and peripheral ductal visualization (not seen, <50%, >50% and 100%); and for the presence of strictures in each of nine ducts. The nine ducts individually assessed in each patient included the extrahepatic bile duct, right main hepatic duct, anterior right hepatic duct, posterior right hepatic duct, left main hepatic duct, medial left hepatic duct, lateral left hepatic duct, right peripheral ducts, and left peripheral ducts. The ductal systems were categorized into central (extrahepatic, main right hepatic, main left hepatic), intrasegmental (anterior and posterior right, medial and lateral left hepatic ducts), and peripheral (right peripheral, left peripheral) ducts. Each of the nine ductal segments was assessed for its extent of visualization on each MR cholangiopancreatogram. If the extent of visualization was greater than 50%, then the degree of visualization was considered good. If the extent of visualization was less than 50%, then the degree of visualization was considered poor. In addition, if the findings of the examination were interpreted as abnormal, each individual duct was assessed for the location of strictures. On MR cholangiopancreatography, a stricture was defined as a short or long segment of ductal narrowing with proximal ductal dilatation or nonvisualization of a central duct with good visualization of the peripheral ducts (Fig. 1). All discrepancies were resolved by consensus. The contrast-enhanced cholangiograms were regarded as the gold standard.
|
To determine if there was overlap between the MR cholangiopancreatographic appearance of healthy subjects and patients with sclerosing cholangitis, the cholangiograms of healthy subjects were intermixed with the cholangiograms of patients with sclerosing cholangitis and reviewed independently. Thus, each observer interpreted three cholangiograms for the 20 sclerosing cholangitis patients (n = 60) and two cholangiograms for the 19 healthy subjects (n = 38) for a total of 98 cholangiograms.
Statistically significant data were calculated, using the signed rank test (p < 0.01). Agreement analysis was calculated using Cohen's kappa. Agreement was considered good if the kappa value was greater than 0.4.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Visualization of Ducts
In patients with sclerosing cholangitis, the consensus interpretation of
thick-slab coronal oblique MR cholangiopancreatograms showed the greatest
degree of ductal visualization, but the difference was not statistically
significant when compared with the gold standard (p = 0.10). The
thick-slab MR images showed greater than 50% ductal visualization in 56 (93%)
of 60 of the central, 69 (86%) of 80 of the intrasegmental, and 27 (68%) of 40
peripheral ducts. In all, 152 (84%) of 180 ducts were noted with greater than
50% ductal visualization (Fig.
2).
|
The consensus interpretation thin-slab MR cholangiopancreatograms showed greater than 50% ductal visualization in 51 (85%) of 60 central, 53 (66%) of 80 intrasegmental, and 21 (52%) of 40 peripheral ducts. In all, 125 (69%) of 180 ducts were visualized with greater than 50% ductal visualization (Fig. 2).
The consensus contrast-enhanced cholangiographic interpretation, which was considered the gold standard, was inferior to the thick-slab MR cholangiopancreatographic interpretation for the degree of ductal visualization. Greater than 50% ductal visualization was noted in 53 (88%) of 60 central, 54 (67%) of 80 intrasegmental, and 20 (50%) of 40 peripheral ducts. In all, 127 (70%) of 180 ducts were seen with greater than 50% ductal visualization (Fig. 2).
The degree of ductal visualization in the healthy subjects was different from that in patients with sclerosing cholangitis. Findings in patients with sclerosing cholangitis usually revealed good visualization (>50% visualization) of the intrasegmental and peripheral ducts on MR cholangiopancreatography, whereas these ducts were rarely well visualized in the control group. On the thick-slab MR cholangiopancreatograms, 56 (98%) of 57 central, seven (9%) of 76 intrasegmental and 0 (0%) of 38 peripheral ducts were well visualized. On the thin-slab MR cholangiopancreatograms, 49 (86%) of 57 central, 7 (9%) of 76 intrasegmental, and 0 (0%) of 38 peripheral ducts were well visualized (Fig. 3). In total, good visualization was noted in 119 (35%) of 342 ducts.
|
Visualization of Strictures
The thick-slab MR cholangiopancreatograms revealed the most strictures, but
the difference was not statistically significant when compared with the gold
standard (p = 0.22). Forty-eight percent of central and 37% of
intrasegmental ducts were strictured. As we would expect, 62% of the
peripheral ducts were strictured on this sequence. In all, 84 (47%) of 180
ducts were strictured on the thick-slab MR cholangiopancreatograms
(Fig. 4).
|
The thin-slab MR cholangiopancreatograms showed central duct strictures in 27 (45%) of 60 ducts, intrasegmental strictures in 20 (25%) of 80 ducts, and peripheral strictures in 22 (55%) of 40 ducts. In all, 69 (38%) of 180 ducts were strictured on the thin-slice MR cholangiopancreatograms (Fig. 4).
The consensus direct contrast cholangiograms showed central strictures in 26 (43%) of 60 ducts, intrasegmental strictures in 18 (22%) of 80 ducts, and peripheral strictures in 20 (50%) of 40 ducts. In all, 64 (36%) of 180 ducts were strictured (Fig. 4).
A total of 111 (62%) of 180 ducts were well visualized on both thick-slab MR cholangiopancreatography and contrast-enhanced cholangiography. Of these 111 ducts, disagreement between the findings of MR cholangiopancreatography and contrast-enhanced cholangiography occurred in 35 ducts (32%). Most of the discrepancies resulted when a stricture was noted on MR cholangiopancreatography but not on contrast-enhanced cholangiography (21 [60%] of 35).
Forty-one (23%) of 180 ducts were well visualized on thick-slice MR cholangiopancreatography but poorly visualized on contrast-enhanced cholangiography. Of these 41 ducts, disagreement between the findings on MR cholangiopancreatography and contrast-enhanced cholangiography occurred in 16 (39%) of 41. Of these 16 discrepancies, 14 (87%) resulted when a stricture was identified on the consensus thick-slice MR cholangiopancreatogram but not on the consensus contrast-enhanced cholangiogram.
Good interobserver agreement was noted for detecting strictures of the
extrahepatic, left hepatic, left medial, and right posterior ducts, with the
greatest agreement for extrahepatic ductal strictures ( = 0.8) on
thickslab MR cholangiopancreatography.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
= 0.79). In addition, the sensitivity and specificity of MR
cholangiopancreatography for localizing extrahepatic primary sclerosing
cholangitis were 83-89% and 83%, respectively, with good interobserver
agreement ( = 0.62). The sensitivity of MR cholangiopancreatography for
localizing intrahepatic primary sclerosing cholangitis was 83%, also with good
interobserver agreement ( = 0.71). Although that study makes a strong
case for MR cholangiopancreatography in the diagnosis and localization of
changes of primary sclerosing cholangitis, the authors limited the evaluation
to two broad bile duct locations, intrahepatic and extrahepatic. To our
knowledge, there has been no study evaluating the correlation of MR
cholangiopancreatography and contrast cholangiography for individual ductal
segments. By dividing the biliary tree into nine ductal segments and comparing
each segment on MR cholangiopancreatography with contrast-enhanced
cholangiography, which is the current gold standard, our goal was to determine
if MR cholangiopancreatography could rival contrastenhanced cholangiography as
an alternative noninvasive imaging test in these patients. In our study, the degree of ductal visualization in healthy male subjects was greatest for the central ducts on thick-slab MR cholangiopancreatography. In addition, the central ducts were well visualized, whereas the intrasegmental and peripheral ducts were poorly visualized (Fig. 5). In fact, good visualization was noted in only 35% of ducts. The central ducts were better visualized because of their larger diameter. The intrasegmental and peripheral ducts were poorly visualized because of their small diameter and because imaging was performed when the ducts were in a physiologic nondistended state. In addition, thick-slab MR cholangiopancreatography was superior to thin-slab MR cholangiopancreatography because of increased spatial resolution and signal-to-noise ratio. Furthermore, all findings of MR cholangiopancreatograms in healthy male subjects were interpreted as normal.
|
In patients with sclerosing cholangitis, we have shown that the consensus interpretation thick-slab MR images showed greater overall ductal visualization when compared with both thin-slab MR cholangiopancreatograms and contrast-enhanced cholangiograms (Fig. 6A,6B). In addition, thick-slab MR cholangiopancreatography showed superior visualization of all the ductal regions when compared with thin-slab MR cholangiopancreatography and gold standard contrast-enhanced cholangiography. The greater spatial resolution and signal-to-noise ratio obtained from thick-slab MR cholangiopancreatography probably account for the greater ductal visualization when compared with the thin-slab MR cholangiopancreatography as described previously. The lesser degree of ductal visualization on contrast-enhanced cholangiography, especially of the intrasegmental ducts, probably reflects a combination of several factors, including the impedance of contrast material due to strictures, low infusion pressure, suboptimal positioning of the catheter, preferential filling of the gallbladder, and the experience of the endoscopist. All these factors may limit the opacification of peripheral ducts on contrastenhanced cholangiography and may result in underestimation of the degree of biliary disease in patients with sclerosing cholangitis [7].
|
|
Although the degree of central ductal visualization was similar between the healthy male volunteers and sclerosing cholangitis group (98% vs 93%), the degree of visualization of the intrasegmental and peripheral ducts was strikingly different between the two groups on thick-slab MR cholangiopancreatography. The intrasegmental ducts were well visualized in 13% and 86%, and the peripheral ducts were well visualized in 0% and 67% in healthy subjects and sclerosing cholangitis groups, respectively. Multiple small focal strictures are usually present in both the intrasegmental and peripheral ducts in patients with sclerosing cholangitis. On MR cholangiopancreatography, a focal stricture will cause upstream dilatation of the ducts, which produces better visualization of these ducts. Unlike the central bile ducts, which because of their larger caliber contain enough bile to produce a strong enough signal to be well visualized on MR cholangiopancreatography, the smaller intrasegmental and peripheral ducts contain little bile and produce little-to-no signal and, thus, are poorly visualized on MR cholangiopan-creatography in healthy subjects. In patients with sclerosing cholangitis, we postulate that the intrasegmental and peripheral ducts are well visualized because of secondary dilatation from strictures of the more central ducts. Thus, the stricture that is a disadvantage on endoscopic retrograde cholangiopancreatography because of impedance of contrast is an advantage on MR cholangiopancreatography. This difference is illustrated in our study. Of the 180 ducts that were well-visualized on thick-slab MR cholangiopancreatography, only 31% were well visualized on contrast-enhanced cholangiography. This dilatation of the ducts may make strictures easier to detect on MR cholangiopancreatography.
Thick-slab MR cholangiopancreatography showed the most strictures when compared with the other modalities. In all sites, strictures were identified on thick-slab MR cholangiopancreatography, thin-slab MR cholangiopancreatography, and contrast-enhanced cholangiography in 47%, 38%, and 36% of ducts, respectively. As we expected, peripheral strictures were more common than intrasegmental and central strictures. Many more intrasegmental (37% vs 22%) and peripheral (62% vs 50%) strictures were detected on MR cholangiopancreatography. As we discussed previously, stricture detection should be closely related to the degree of ductal visualization. Because the peripheral ducts were better visualized on thick-slab MR cholangiopancreatography, strictures were more often detected. However, because of the limited literature regarding MR cholangiopancreatography in the evaluation of sclerosing cholangitis, we do not yet know the accuracy of MR cholangiopancreatography for detecting peripheral strictures in patients with sclerosing cholangitis. Thus, these results should be viewed cautiously. For example, in our study, strictures detected on MR cholangiopancreatography may be false-positive results. Because of our small study population, more work in this area is required.
From the results of our study, it may be argued that MR cholangiopancreatography may be more sensitive than contrast-enhanced cholangiography in detecting strictures in sclerosing cholangitis. First, we have shown that the degree of ductal visualization in patients with sclerosing cholangitis is greatest on thick-slab MR cholangiopancreatography when compared with contrast-enhanced cholangiography. Second, we have shown that although the peripheral ducts are never well visualized (>50% ductal visualization) in healthy subjects, they are well visualized in 67% of patients with sclerosing cholangitis on thick-slab MR cholangiopancreatography (compared with 50% on contrast-enhanced cholangiography). Therefore, we postulate that an increase in ductal caliber due to central strictures in patients with sclerosing cholangitis causes the ducts to become better visualized on MR cholangiopancreatography and translates to increased detection of strictures. Thus, MR cholangiopancreatography may be more sensitive to early changes of sclerosing cholangitis than contrast-enhanced cholangiography.
In the ducts with suboptimal visualization on contrast-enhanced cholangiography and good visualization on thick-slab MR cholangiopancreatography, discrepancies occurred in 39% of cases. Of these discrepancies, 87% included a strictured duct on MR cholangiopancreatography with a nonvisualized or nonstrictured duct on contrast-enhanced cholangiography (Figs. 6A,6B and 7A,7B). In those ducts with good visualization on contrast-enhanced cholangiography and MR cholangiopancreatography, there were fewer discrepancies (32%), as expected. However, most of the discrepancies (60%) showed a strictured duct on MR cholangiopancreatography and a normal duct on contrast-enhanced cholangiography. Without further studies using a larger population of patients, these results can imply that either MR cholangiopancreatography is more sensitive than the gold standard endoscopic retrograde cholangiopancreatography for detecting strictures or that the evaluation of strictures on MR cholangiopancreatography tends to produce false-positive results.
|
|
Good interobserver agreement (k > 0.4) between thick-slab MR
cholangiopancreatography and gold standard contrast-enhanced cholangiography
in sclerosing cholangitis was noted in four of the nine ductal units (Figs.
8A,8B
and
9A,9B).
In addition, the detection of extrahepatic bile duct strictures showed the
greatest interobserver agreement ( = 0.8), probably because of greater
size and visibility of ducts. However, a more meaningful result is that all
patients with sclerosing cholangitis and all healthy subjects were interpreted
as having sclerosing cholangitis or as being healthy, respectively. Thus, MR
cholangiopancreatography may be sensitive in diagnosing sclerosing
cholangitis.
|
|
|
|
In this study, we have separated the bile ducts into nine units. Because we were comparing MR cholangiopancreatography with the gold standard contrast-enhanced cholangiography, the separation of the ducts into small units should provide a more accurate means to determine agreement between the modalities. However, because of the increased number of variables, there is a greater likelihood of obtaining discordant interpretations. The clinical impact of the discordant observations in our study is expected to be low, because therapy is based not on the number of bile duct strictures per patient but on the presence of a dominant stricture that may be amendable to percutaneous or endoscopic therapy. We hypothesize that although the number of strictures detected may vary from reviewer to reviewer, the diagnosis of sclerosing cholangitis should be made on MR cholangiopancreatography, and there should be no negative clinical impact, particularly as all patients with sclerosing cholangitis and all healthy volunteers were correctly identified as having sclerosing cholangitis and as being healthy, respectively. In our experience, MR cholangiopancreatography should identify clinically important dominant strictures in patients with sclerosing cholangitis.
The greatest potential advantage of MR cholangiopancreatography in depicting sclerosing cholangitis is the potential for noninvasive follow-up of these patients. The evolution of sclerosing cholangitis varies from patient to patient, with some patients living many years and others succumbing quickly to liver failure or death after initial diagnosis. Cholangiocarcinoma can develop in up to 15% of patients with sclerosing cholangitis [9, 10]. In addition, it is difficult to predict which patients will develop mild disease with long survival, rapidly progressive disease with short survival, or cholangiocarcinoma. Furthermore, some patients may have progression of disease that is not manifest at clinical examination or serology [11,12,13]. Finally, it is not practical or acceptable to subject asymptomatic patients to routine follow-up endoscopic retrograde cholangiopancreatography, because of the cost, morbidity, and mortality associated with the procedure. Thus, MR cholangiopancreatography may have a potential role in the noninvasive serial follow-up of patients with sclerosing cholangitis. Using this procedure may ultimately result in a decrease in morbidity and mortality rates. An advantage of MR cholangiopancreatography is that the study may be supplemented as needed with conventional T1- (including contrast-enhanced) and T2-weighted images for a comprehensive study when complications such as cholangiocarcinoma are being investigated.
Several potential pitfalls in using MR cholangiopancreatography to image patients with sclerosing cholangitis include the following: first, pneumobilia caused by a biliary-enteric anastomosis, stent, or recent endoscopic retrograde cholangiopancreatogram can produce signal voids and obscure ductal visualization in those ducts containing air. Second, diffuse stricturing can produce visualization of the ducts. For example, nonvisualization of a central duct that should normally be visualized, associated with upstream dilatation of the peripheral ducts, is diagnostic of a stricture. However, diffuse strictures of the central and peripheral ducts preclude their visualization and, thus, their assessment for disease on MR cholangiopancreatography. Third, thick bile will cause decreased signal on single-shot fast spin-echo sequences, thereby decreasing the visualization of the duct. Fourth, the presence of a stent in a duct may produce a large-enough filling defect to preclude accurate evaluation of that duct. Fifth, imaging the ducts in their physiologic nondistended state may decrease the sensitivity of MR cholangiopancreatography to subtle ductal abnormalities. Sixth, imaging of the bile ducts in patients with cirrhotic livers may preclude visualization of the ducts because of extrinsic compression by regenerative nodules.
In conclusion, our study shows that thick-slice MR cholangiopancreatography is the best technique for the visualization of normal and strictured bile ducts and allows differentiation of healthy patients from patients with sclerosing cholangitis. Therefore, we suggest that this technique may be of value in the diagnosis and follow-up of patients with sclerosing cholangitis.
Acknowledgments
We thank David DeLong for his assistance with the statistical analysis.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  R Caprilli, M A Gassull, J C Escher, G Moser, P Munkholm, A Forbes, D W Hommes, H Lochs, E Angelucci, A Cocco, et al. European evidence based consensus on the diagnosis and management of Crohn's disease: special situations Gut, March 1, 2006; 55(suppl_1): i36 - i58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. M. Vitellas, A. El-Dieb, K. K. Vaswani, W. F. Bennett, M. Tzalonikou, C. Mabee, R. Kirkpatrick, and J. G. Bova MR Cholangiopancreatography in Patients with Primary Sclerosing Cholangitis: Interobserver Variability and Comparison with Endoscopic Retrograde Cholangiopancreatography Am. J. Roentgenol., August 1, 2002; 179(2): 399 - 407. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. M. Vitellas, A. El-Dieb, K. K. Vaswani, W. F. Bennett, J. Fromkes, C. Ellison, and J. G. Bova Using Contrast-Enhanced MR Cholangiography with IV Mangafodipir Trisodium (Teslascan) to Evaluate Bile Duct Leaks After Cholecystectomy: A Prospective Study of 11 Patients Am. J. Roentgenol., August 1, 2002; 179(2): 409 - 416. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
