| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Faculty of Medicine, Kagoshima University, 8-35-1 Sakuragaoka, Kagoshima-shi 890-8520, Japan. Received March 27, 2001; revision requested April 30; revision received July 9; accepted August 9. Address correspondence to Y.F. (e-mail: fukukura@m.kufm.kagoshima-u .ac.jp).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Single-shot turbo spin-echo T2-weighted dynamic magnetic resonance cholangiopancreatography (MRCP) was performed in 85 patients who did not have pancreatic diseases (group 1) and in 50 patients who had focal severe stenosis of the main pancreatic duct (group 2). The visualization and diameter of the pancreatic duct before and after secretin administration were assessed.
RESULTS: In group 1, after secretin administration, the best visualization of the main pancreatic duct in the head, body, accessory pancreatic duct, and branch ducts was achieved in 4.7 minutes ± 1.6 (SD), 4.8 minutes ± 1.6, 4.6 minutes ± 1.6, and 4.7 minutes ± 1.3, respectively. Improvement in the delineation of the main pancreatic duct in the head (78 [92%] patients), body (80 [94%] patients), accessory pancreatic duct (35 [41%] patients), and branch ducts (14 [16%] patients) was achieved. Overlap of fluid in the organ and the pancreatic duct was observed in 20 (24%) of 85 patients. Overlap was especially observed after 5 minutes of secretin injection. In group 2, the best visualization of the distal main pancreatic duct was achieved 4.9 minutes ± 1.4 after secretin administration. Improvement in the delineation of the distal main pancreatic duct was achieved in 17 (85%) of 20 patients.
CONCLUSION: MRCP is best performed during the first 5 minutes after secretin administration.
© RSNA, 2002
Index terms: Magnetic resonance (MR), cholangiopancreatography, 770.121411 • Pancreas, MR, 770.121411 • Pancreatic ducts, MR, 774.121411 • Secretin
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
To our knowledge, published articles (13–16) that include results with MRCP after secretin stimulation have been limited. Moreover, the effectiveness of MRCP with secretin stimulation in the evaluation of severe stenosis of the main pancreatic duct and the influence of duodenal filling have not been elucidated.
The goals of this study were (a) to evaluate the pancreatic duct after secretin stimulation in patients with normal ducts and in patients with focal severe stenosis of the main pancreatic duct and (b) to determine the best time to perform imaging in these patients after secretin administration.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The first group included 85 patients with a normal pancreas, the second group included 50 patients with pancreaticobiliary diseases that caused severe stenosis of the head or body of the main pancreatic duct, and the third group included 68 patients with other pancreaticobiliary diseases (cystic lesion of the pancreas [44 patients], anomalous pancreaticobiliary junction [11 patients], tumor of the pancreatic tail [nine patients], pancreas divisum [three patients], or autoimmune pancreatitis [one patient]). Thus, our study included only the 135 patients (79 men, 56 women; age range, 19–82 years; mean age, 62.7 years) who were assigned to groups 1 and 2.
Group 1 included 85 patients (44 men, 41 women; mean age, 61.9 years; age range, 19–82 years). Laboratory data and US, CT, and MR imaging findings revealed a normal pancreas in all 85 patients. All 43 patients who underwent ERCP had normal pancreatic ducts. Final diagnoses for the 85 patients were confirmed with surgical findings in 38 patients, with biopsy findings in three patients, and with clinical findings—medical history, laboratory data, and imaging findings—in 44 patients. In 31 of 85 patients, US, CT, and MR imaging revealed normal pancreaticobiliary findings. Of the remaining 54 patients, 21 patients had cholelithiasis, 15 had extrahepatic bile duct cancers without pancreatic invasion, nine had adenomyomatosis, four had gallbladder cancers, four had liver tumors, and one had a hepatic arterial aneurysm.
Group 2 included 50 patients (35 men, 15 women; mean age, 63.9 years; age range, 36–82 years). All 38 patients who underwent ERCP had severe stenosis of the head or body of the main pancreatic duct. In 33 of 38 patients, ERCP could not depict the proximal main pancreatic duct to the stenosis. Final diagnoses were verified with surgical findings in 40 patients and with clinical findings—medical history, laboratory data, and imaging findings—in 10 patients. Twenty patients had pancreatic cancer, 17 had pancreatolithiasis, 10 had papilla of Vater cancer, two had pancreatic metastases, and one had extrahepatic bile duct cancer with pancreatic invasion.
MR Examinations
All examinations were performed with a 1.5-T MR system (Magnetom Vision; Siemens, Erlangen, Germany) with a phased-array body coil. Coronal single-shot turbo spin-echo T2-weighted dynamic MRCP was performed before and at every minute for 10 minutes after an intravenous injection of secretin (Secrepan; Eisai, Tokyo, Japan) at a dose of 1 unit per kilogram of body weight.
The imaging parameters for the single-shot rapid acquisition with relaxation enhancement (RARE) MR sequence in which one signal was acquired were as follows: repetition time msec/effective echo time msec, 
/1,100; echo spacing, 11.5 msec; echo train length, 240; slab thickness, 30–50 mm; field of view, 200–300 mm; matrix, 240 x 256; and acquisition time, 7.12 seconds. The patients were asked to fast for at least 9 hours before examination. To suppress the signal intensity of overlapping fluid-containing organs, the contrast-agent-and-milk mixture, prepared by dissolving 600 mg of ferric ammonium citrate (FerriSeltz; Otsuka Pharmaceutical, Tokushima, Japan) in 50 mL of water with 150 mL of milk added, was administered before dynamic MRCP imaging.
Image Analysis
The visualization of the pancreatic duct and the occurrence of overlapping fluid-containing organs were assessed by two radiologists (F.F., M.S.) who together reached a consensus. The maximum diameter of the pancreatic duct and the length of the stenosis were measured by one researcher (Y.F.). To minimize learning bias, patient data and imaging parameters were removed from images, and each sequence was reviewed in random order. The radiologists were aware that a duct stenosis may have been documented with use of other imaging methods, but they had no information regarding the length of the stenosis.
The observers were asked to record the degree of visualization and the diameter of the pancreatic duct before and at every minute for 10 minutes after secretin stimulation. The time to achieve the best visualization of the pancreatic duct after secretin stimulation was also evaluated in each patient. MR cholangiopancreatograms (those in which visualization of the pancreatic duct was the best) obtained before and after secretin administration were compared on the basis of the degree of visualization and the diameter of the duct. The grades of visualization were as follows: poor (the anatomic part was difficult to detect or partially visible), fair (the anatomic part was mostly visible), good (the entire anatomic part was visible), or excellent (the entire anatomic part was clearly visible). Improvement in visualization after secretin administration was assessed by means of a change in the grades of visualization.
The diameters of the pancreatic ducts were measured at the same location before and after secretin administration. When the pancreatic duct was not visualized, the diameter of the duct was measured as 0 mm. In group 1, the main pancreatic duct in the head, body, accessory pancreatic duct, and branch ducts was evaluated. In group 2, the proximal (n = 50 of 50) and distal (n = 20 of 50) portions of the main pancreatic duct to the stenosis were evaluated. In 20 patients, we also measured the length of the stenosis before and after secretin stimulation.
We evaluated the degree of overlap of fluid in the gastrointestinal tract and the pancreatic duct or biliary tract in both groups. The grades of the degree of overlap were as follows: no overlap (overlap not present), minimum overlap (overlap present but not degrading the image), medium overlap (overlap present and degrading the image), or great overlap (overlap present and the image was nondiagnostic).
Statistical Analysis
The visualization of the pancreatic duct before and after secretin administration was statistically evaluated by using the McNemar test. A Student t test was subsequently applied to analyze the difference in the diameter of each pancreatic duct before and after secretin administration. A P value of less than .05 was considered statistically significant. The Bonferroni adjustment for multiple tests was also applied.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The best visualization of the main pancreatic duct in the head, body, accessory pancreatic duct, and branch ducts in each patient was achieved after a mean delay of 4.7 minutes ± 1.6, 4.8 minutes ± 1.6, 4.6 minutes ± 1.6, and 4.7 minutes ± 1.3, respectively. Improvement in the grades of visualization of the main pancreatic duct in the head (78 [92%] patients), body (80 [94%] patients), accessory pancreatic duct (35 [41%] patients), and branch ducts (14 [16%] patients) was achieved (Fig 1). The difference in the mean diameter of the main pancreatic duct in the head, body, accessory pancreatic duct, and branch ducts before (1.4 mm ± 0.7, 1.1 mm ± 0.6, 0.1 mm ± 0.3, and 0.01 mm ± 0.11, respectively) and after (2.4 mm ± 0.9, 2.2 mm ± 0.8, 0.5 mm ± 0.6, and 0.2 mm ± 0.5, respectively) secretin administration was significant (P < .0001 for the head, body, and accessory pancreatic duct and P < .0005 for the branch ducts). Even after the Bonferroni adjustment for multiple comparisons, with a P value of .0125 considered to indicate a statistically significant difference, the mean diameter of each segment of the pancreatic duct was significantly larger after secretin administration.
|
|
|
|
|
|
|
Before and after secretin administration, the distal main pancreatic duct was visualized in 18 (90%) and 19 (95%) of 20 patients, respectively. The best visualization of the distal main pancreatic duct in each patient was obtained after a mean delay of 4.9 minutes ± 1.4. Improvement in the grades of visualization of the distal main pancreatic duct was obtained in 17 (85%) of 20 patients (Fig 8). Before and after secretin administration, the mean diameter was 1.4 mm ± 0.7 and 2.5 mm ± 0.8 (P < .0001), respectively. The difference in the mean diameter of the distal main pancreatic duct before and after secretin administration was significant, and this difference was assessed by using the Bonferroni adjustment for multiple comparisons, with a P value of .0166 considered to indicate a statistically significant difference. The length of severe stenosis before and after secretin administration was 22.8 mm ± 16.4 and 18.4 mm ± 12.1 (P = .0088), respectively. The difference in the length of severe stenosis before and after secretin administration was also significant after the Bonferroni adjustment for multiple comparisons, with a P value of .0166 considered to indicate a statistically significant difference. In five (25%) of 20 patients, the length before secretin administration was greater than that after secretin administration, with a difference of more than 10 mm.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In the patients with a normal pancreas in this study, the main pancreatic duct in the head and body was visualized in 80 (94%) and 71 (84%) patients, respectively. These results are similar to other published data in investigations with half Fourier RARE MRCP (5–7,11,12), and they are superior to those in earlier investigations (visualization rates, 53%–86%) with fast spin-echo MRCP (1–3,10,20). The normal main pancreatic duct is, therefore, not always visualized even with the single-shot RARE technique, which provides higher quality MRCP images than other techniques (6,19). Moreover, Masui et al (11) reported that without dilation of the pancreatic ducts, stenotic lesions were difficult to evaluate, sensitivity decreased to 62.5%, and false-positive diagnoses occurred.
The evaluation of branch ducts is important and is essential for diagnosing some pancreatic abnormalities, such as chronic pancreatitis or pancreatic cancer originating from the ducts. In this study, the normal accessory pancreatic duct and branch ducts were visualized in only eight (9%) and one (1%) patient, respectively. Irie et al (6) reported that undilated branch ducts were rarely demonstrated, even with half Fourier RARE MRCP. This is caused by the limited spatial resolution of the technique, since the normal accessory pancreatic duct and branch ducts are very thin. In contrast to the situation with MRCP, at ERCP slight overdistention of the ducts may enable better characterization of subtle structures and filling of ductal side branches, both of which have important diagnostic implications. A limitation of MRCP is, therefore, an occasional difficulty in evaluation of the thin main pancreatic duct and branch ducts.
The hormone secretin stimulates the duct cells of the pancreas and biliary tract to secrete fluid and bicarbonate, with a consequent increase in the volume of fluid in the pancreatic duct system. For the diagnosis of pancreatic disease, it may be useful to perform MRCP during dilation of the pancreatic duct after secretin administration. Takehara et al (21) reported that the frequency of false-positive stenosis significantly decreased at MRCP after secretin administration. Manfredi et al (22) reported that MRCP with secretin stimulation helped in identifying pancreas divisum and cystic dilatation of the distal Santorini duct, which is called "santorinicele." However, to our knowledge, the role of MRCP with secretin stimulation in the behavior of the pancreatic duct has not been explored extensively.
In the patients with a normal pancreas in this study, improvement in the delineation of the main pancreatic duct in the head (92%), body (94%), accessory pancreatic duct (41%), and branch ducts (16%) was achieved after the intravenous administration of secretin. These results suggest that MRCP with secretin stimulation may not be adequate for the assessment of the undilated accessory pancreatic duct and branch ducts, although it may be useful in the assessment of the main pancreatic duct. The degree of duct dilatation depends on the rate of secretion of pancreatic juice, the volume of the ductal system, the compliance of the duct, and intraduodenal pressure, in addition to the resistance of the pancreatic duct sphincter (23). In this study, a significant increase in the diameter of the normal pancreatic duct and improvement in its visualization at MRCP, especially 4–5 minutes after the administration of secretin, were observed.
These results are similar to those in previous articles (15,23) about US and MRCP. This is explained by the fact that intravenous administration of a bolus of secretin induces increased tonus of the sphincter of the hepatopancreatic ampulla (Oddi sphincter) during the first 5–6 minutes, and the tonus of the sphincter decreases after that time (24). Cappeliez et al (16), however, reported that the mean delay to reach the maximum diameter of the pancreatic duct was 175 seconds in patients with a normal pancreas and 289 seconds in those with chronic pancreatitis. Our results in patients with a normal pancreas were similar to their results in patients with chronic pancreatitis. The limitation of our study was that results from only 43 of 85 patients with a normal pancreas could be correlated with ERCP images, so the group of patients with a normal pancreas in this study might include some patients with mild chronic pancreatitis. The more tardy dilatation after secretin administration seen in our patients with a normal pancreas may be due to their older ages, because periductal fibrosis seen in advanced age reduces the flexibility of the pancreatic duct (23).
In cases of severe stenosis of segments of the pancreatic duct, MRCP can depict both the proximal and distal main pancreatic ducts to the stenosis, whereas depiction of the proximal duct with ERCP is difficult. Because stenotic portions can be reconstructed or resected surgically, the evaluation of stenosis is an important role of MRCP. In this study, the length of severe stenotic portions at MRCP before secretin administration was significantly longer than the length of severe stenotic portions at MRCP with secretin stimulation. The relatively high frequency of overestimation of severe stenotic portions with MRCP before secretin administration was thought to be attributable to insufficient visualization of the distal main pancreatic duct to the stenosis. After secretin administration, visualization of the distal main pancreatic duct to the stenosis was improved, as was visualization of the main pancreatic duct in patients with a normal pancreas, because normal ductal segments are considered to be soft and flexible and are, therefore, dilated with secretin.
On the other hand, segments that are affected with chronic desmoplastic pancreatitis or that show involvement of a malignant tumor are not likely to be dilated with secretin. The use of secretin was thought to increase the accuracy of MRCP in evaluating pathologic severe stenosis of the main pancreatic duct, although we did not correlate results with ERCP findings because ERCP could not depict the proximal duct to the severe stenosis in many patients.
The duodenal filling grade at MRCP after secretin stimulation allows specific estimation of the pancreatic exocrine function (13–15,25), but high signal intensity in the bowel is also an obstacle to evaluation of the pancreaticobiliary system on images obtained with the single-shot RARE sequence (26). In this study, duodenal filling impaired evaluation of the pancreaticobiliary system after secretin stimulation, although we used peroral negative contrast media prior to MR imaging to eliminate intensified signals from the gastrointestinal tract. In particular, overlap of fluid in the gastrointestinal tract and the head or tail of the normal pancreas was observed after 5 minutes of secretin injection. Bolondi et al (27) reported that a progressive filling of the duodenal lumen was visualized at US at 6–9 minutes after secretin injection. In this study, the most prominent appearance of the overlap was obtained 7–8 minutes after secretin injection.
Secretin is also associated with a risk of pancreatitis owing to dilatation and increased pressure within side branches proximal to a pancreatic duct stenosis. However, none of our patients experienced complications associated with secretin administration. We believe that secretin is safe except in some cases of contraindication, such as acute pancreatitis.
In conclusion, despite the additional time and cost of secretin administration in MRCP examinations, results of this study suggest that MRCP, performed during the first 5 minutes after the administration of secretin, is useful in assessing the pancreatic duct.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, Y.F.; study concepts, Y.F.; study design, Y.F., F.F.; literature research, Y.F.; clinical studies, Y.F.; data acquisition, Y.F., F.F., M.S.; data analysis/interpretation, Y.F.; statistical analysis, Y.F.; manuscript preparation, Y.F.; manuscript definition of intellectual content, Y.F., F.F.; manuscript editing, Y.F., M.N.; manuscript revision/review, Y.F., F.F., M.S.; manuscript final version approval, Y.F.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Tamura, T. Ishibashi, and S. Takahashi Chronic Pancreatitis: MRCP versus ERCP for Quantitative Caliber Measurement and Qualitative Evaluation Radiology, March 1, 2006; 238(3): 920 - 928. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Monill, J. Pernas, J. Clavero, A. Farre, A. Morales, M. Gonzalez, V. Artigas, and A. Martinez-Noguera Pancreatic Duct After Pancreatoduodenectomy: Morphologic and Functional Evaluation with Secretin-Stimulated MR Pancreatography Am. J. Roentgenol., November 1, 2004; 183(5): 1267 - 1274. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
), body (
), accessory pancreatic duct (
), and branch ducts (+) in a normal pancreas before and at every minute for 10 minutes after secretin administration.
