| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Developments |
1 From the Departments of Diagnostic Radiology (J.T.H., H.J.W., K.J.K.), Surgery (H.E.), and Nephrology (A.H.), University Hospital, Philipps University, Marburg, Germany; and Department of Radiology, Ohio State University, 645 Means Hall, 1654 Upham Dr, Columbus, OH 43210 (J.T.H., A.L.L.). Received July 26, 2003; revision requested October 9; revision received January 11, 2004; accepted February 4. Address correspondence to J.T.H. (e-mail: heverhagen@gmx.net).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
© RSNA, 2004
Index terms: Magnetic resonance (MR), perfusion study, 770.12144 • Pancreas, function • Pancreas, MR, 770.121411, 770.121412, 770.12143, 770.12144 • Pancreas, transplantation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
In patients who underwent simultaneous pancreas-kidney transplantation, the functional status of the kidney may also serve as an indicator of pancreatic rejection (8). For patients with pancreatic transplantation alone or pancreatic transplantation after kidney transplantation, however, this option does not exist. Moreover, the diagnosis of pancreatic rejection is especially difficult in the absence of kidney rejection. Therefore, pancreatic allograft biopsy is the preferred technique for evaluation of pancreatic allograft rejection or dysfunction. At percutaneous biopsy guided with US, adequate tissue samples are obtained in 89% of patients (9), but this procedure is accompanied by complications that include inflammation and abdominal bleeding (8).
Krebs et al (10) showed that dynamic contrast material–enhanced magnetic resonance (MR) imaging for evaluation of graft enhancement is highly sensitive for the detection of acute pancreatic graft rejection. They calculated the mean percentage of pancreatic enhancement and demonstrated that this value is correlated with the severity of the pancreatic rejection. In some cases, however, they demonstrated an overlap in mean percentage of pancreatic enhancement between functioning and nonfunctioning grafts. Secretin-augmented MR hydrometry has been shown to be a reliable diagnostic tool in the assessment of exocrine organ function in patients with chronic pancreatitis (11,12). Thus, the purpose of our study was to prospectively explore the feasibility of secretin-augmented MR pancreatography in combination with dynamic contrast-enhanced MR measurements for evaluation of the functional status of pancreatic allografts by quantifying the excretion and perfusion of the grafts.
|
Materials and Methods |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
All grafts were transplanted with systemic venous (inferior vena cava) and enteric (jejunal loop) drainage. All patients received standard immunosuppressive therapy. Pancreatic graft rejection was indicated by an increase in circulating blood levels of creatinine, glucose (fasting), lipase, and amylase; Doppler US findings; and/or histologic findings (1). Blood analysis and Doppler US were performed within 1 day prior to or after the MR imaging measurements. Transplant parenchyma and perfusion were evaluated with US (Sonoace SA 9900; Kretz Technik, Zipf, Austria) by one investigator with 14 years of experience. Patients who did not show any signs of graft rejection were followed up clinically by a nephrologist (H.E.) for at least 1 year (mean, 22 months; range, 12–33 months), and this follow-up included quarterly blood analysis of creatinine, lipase, and amylase levels. The need for insulin replacement therapy was determined with blood levels of glucose. Biopsy was not performed in any of the patients without signs of graft rejection or failure. All measured values were tabulated, and mean values with standard deviations were calculated when appropriate.
MR Imaging
Patients were examined by using MR imaging to evaluate exocrine pancreatic function and the perfusion of the transplant between posttransplantation days 14 and 17. All examinations were performed in a 1.0-T clinical magnet (Magnetom Expert; Siemens, Erlangen, Germany) with a quadrature phased-array body coil and commercially available gradients capable of a 1200-µsec rise time and 20 mT/m maximum gradient strength. A negative oral contrast agent was not applied in order to prevent interference between the contrast agent and secreted fluid.
After the acquisition of scout images, a fluid-sensitive fast spin-echo MR imaging sequence was selected to depict all fluid in the examination volume. A dynamic breath-hold T2-weighted single-shot fast spin-echo sequence (repetition time msec/echo time msec, 
/1100 [effective]; flip angle, 150°) with a 7-second acquisition time was used. The echo train consisted of 240 echoes with an echo spacing of 9.2 msec. Single slabs with an in-plane resolution of 1.00 x 0.94 mm and a voxel depth of 65 mm were acquired. The slabs were angled in a coronal oblique plane to cover the graft and adjacent small bowel. This positioning provided a summation of all fluids in the imaging volume.
After the acquisition of the first image, 1 clinical unit per kilogram body weight of secretin (Secrelux; Goldham, Neuss, Germany) was injected intravenously in order to stimulate the transplanted pancreas to produce the exocrine secretion. Then, measurements were repeated every 30 seconds for a total imaging time of 10 minutes. Similar to the protocol proposed by Krebs et al (10), a protocol with a fat-suppressed contrast-enhanced dynamic T1-weighted gradient-echo sequence (153.4/6.0; flip angle, 70°; field of view, 30 cm; matrix, 256 x 256; acquisition time, 24 seconds) was performed in a breath-hold phase. Images were acquired in a coronal oblique plane, similar to the way they were acquired with the T2-weighted sequence. A nonenhanced measurement was conducted. After the injection of 0.1 mmol/kg of gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany), measurements were repeated every minute for a total investigation time of 5 minutes.
Between measurements, the patients were allowed to catch their breath. The contrast agent was administered with a power injector (Spectris; Medrad, Pittsburgh, Pa) at an injection rate of 2 mL/sec. Total examination time, which included patient positioning, was approximately 60 minutes. Patients were questioned by one author (J.T.H.) about secretin or contrast agent side effects and the tolerability of the MR imaging investigation.
Image Evaluation
Evaluation of the MR images and region of interest (ROI) placement were performed by one investigator (J.T.H., with 7 years of experience in MR imaging) who was blinded to all clinical and laboratory results.
MR pancreatographic images were evaluated according to the protocol derived by Heverhagen and colleagues (13). In brief, the mean signal intensity of the ROI was provided by a histogram algorithm. The mean ROI size was 250 cm2± 52. The increase in mean signal intensity over 10 minutes was divided by a calibration factor of 0.043, which was derived from data in a previously performed volunteer study (13). The ROI included only the transplanted pancreas and the associated small bowel; therefore, the native pancreas and the draining duodenum and small bowel were excluded (Fig 1).
|
The enhancement pattern of the transplant in one selected ROI per patient was evaluated. The mean ROI size was 56 mm2± 16. The percentage enhancement in each ROI was calculated as the difference of the signal intensity of each measurement time point and the signal intensity of the first measurement divided by this initial signal intensity. The maximum signal intensity–to-time ratio and the area under the curve were determined. The slope of the curve from the first measurement to peak enhancement represents the maximum signal intensity–to-time ratio.
|
Results |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
|
|
|
|
|
|
US was not able to depict the graft in two cases because of gas overlay. Six functional grafts were demonstrated at US, with homogeneous parenchyma and good perfusion. The chronically rejected graft showed signs of inflammation. One graft with necrotizing pancreatitis was enlarged and showed decreased perfusion at color-coded Doppler US.
Histologic analysis was performed in only those patients with abnormal graft function, and results confirmed the diagnoses of chronic rejection (patient 8, Table) and necrotizing pancreatitis (patients 9 and 10, Table).
MR perfusion measurements (Fig 6) used to plot curves produced area under the curve values that were greater than the threshold value of 0.5 for functional grafts and less than 0.5 for dysfunctional grafts. Maximum signal intensity–to-time ratio values provided a similar threshold value of 0.3 (Fig 7). With secretin-augmented MR pancreatography, a mean excreted volume of 236 mL ± 104 was calculated in functioning grafts, compared with that of 42 mL ± 25 calculated in dysfunctioning grafts. No superposition of fluids secreted from the native pancreas was observed. The graph displayed in Figure 7c summarizes the maximum signal intensity–to-time ratio for MR pancreatographic results.
|
|
|
|
|
|
Discussion |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
Therefore, some noninvasive approaches have been developed to investigate the functional status of the graft, but none of them have been adopted for broad clinical use.
At US, the radiologist uses the variation of the size and the echogenicity of the graft as indicators of disease. US examinations of pancreatic grafts include Doppler imaging of the vascular supply (arteries and veins) and morphologic study of the parenchyma. With US, one can identify arteriovenous fistulas, vascular thrombosis, pseudoaneurysms, and vascular anastomotic stenoses (17). US, however, is investigator dependent, and a sensitivity between 18% and 58% and a specificity between 73% and 100% can be achieved for graft failure (3–5). At percutaneous biopsy guided with US, adequate tissue samples have been obtained in 89% of cases (9).
Laboratory parameters for estimation of graft function, such as enzyme concentrations in the blood, are sensitive but not specific in regard to graft dysfunction (6,7).
Various authors have addressed the problems of pancreatic transplant rejection by using MR imaging. The use of unenhanced T1- and T2-weighted MR images provided varying results. Some groups (18,19) have not been able to show differences in the signal intensity behavior of rejected pancreatic allografts compared with that of grafts without rejection. On the other hand, Yuh et al (20) evaluated signal intensity of the graft in comparison with that of muscle tissue (T1) and of urine (T2), which resulted in a 100% sensitivity and a 76% specificity. In a quantitative approach, Vahey et al (21) measured a higher T2 value for rejected organs than was measured for that of nonrejected grafts. Contrast-enhanced MR imaging revealed differences between the two groups in signal intensity enhancement 1 minute after application of contrast medium (10,19). These results, however, have limited clinical value because of a marked overlap in results between both groups of patients.
Matos et al (22) proposed a semiquantitative visual grading of the duodenal filling after secretin stimulation to diagnose exocrine pancreatic insufficiency. The authors proposed to classify the duodenal filling into four grades, which ranged from 0 when no fluid was secreted to 3 when the duodenum was largely filled beyond the genu inferius. A grade of less than 3 was considered to be indicative of reduced exocrine function. Matos et al showed a significant difference between control subjects and patients with a reduced exocrine secretion. This method has been refined into a fully quantitative analysis of pancreatic output (13).
In a recent study, Krebs et al (10) showed that dynamic contrast-enhanced MR imaging for evaluation of graft enhancement is highly sensitive for the detection of acute pancreatic graft rejection. They calculated the mean percentage of pancreatic enhancement and demonstrated that it was correlated with the severity of the pancreatic rejection. In some cases, however, they demonstrated an overlap in mean percentage of pancreatic enhancement between functioning and nonfunctioning grafts.
In our study, MR imaging of the graft and the anastomosed small bowel was performed with fluid-sensitive T2-weighted sequences after secretin stimulation of the exocrine pancreatic function and with dynamic breath-hold contrast-enhanced T1-weighted sequences. Secretin stimulation was used to demonstrate exocrine function that correlated with pancreatic rejection, and dynamic contrast-enhanced MR imaging was used to provide information about graft perfusion. The results of these investigations were compared with the clinical diagnosis of the patients as assessed with biochemical parameters, biopsy results, and surgical findings.
Normally functioning grafts produced more fluid after secretin stimulation and had stronger enhancement after contrast medium injection. Ten minutes after secretin stimulation, the fluid production in the normal group was always greater than the threshold of 100 mL, whereas the dysfunctional allografts secreted less than 100 mL. Also, there were differences in both of the obtained perfusion parameters, area under the curve and maximum signal intensity–to-time ratio, between groups. In accordance with the results of Krebs et al (10), our area-under-the-curve measurements were always greater than 0.5 and maximum signal intensity–to-time ratio was greater than 0.3 in the functional transplant group, whereas the dysfunctional transplant group consistently had values less than these thresholds. Because of the larger patient population evaluated by Krebs et al, however, an overlap in values between healthy and dysfunctional grafts was found, thus limiting the clinical application of MR imaging perfusion studies in evaluation of graft function.
To overcome this diagnostic problem, we added the measurement of transplant fluid output after secretin stimulation to our study, and this additional measurement enabled us to further discriminate well-functioning from dysfunctioning allografts. Therefore, benefits are expected from the combination of both perfusion and secretion measurements in larger patient groups, especially in providing differentiation between early stages of graft rejection and regular graft function. The method of secretion quantification, however, is a relatively new method that needs to be further evaluated. The suggested thresholds adopted from studies about chronic pancreatitis could be subject to change in future studies.
The use of secretin in patients with acute pancreatitis is controversial. On one hand, the package insert of the manufacturer states that secretin is contraindicated in patients with acute pancreatitis. On the other hand, in case of severe rejection, the gland is already severely damaged, and the additional secretin stimulus seems does not appear to add any more damage. The pathophysiologic features of transplant rejection with graft pancreatitis are different from those of acute pancreatitis. After discussing this issue in our study group that consisted of transplant surgeons, nephrologists, and radiologists, as well as with experienced personnel from other centers, we decided that the use of secretin did not pose a substantial risk to our patients.
Although our study provided results that showed that a combination of perfusion measurements and quantification of exocrine function of the grafts is effective and adds supplementary value to the evaluation of graft function, our study had limitations. First, the patient population was small. This study, however, was planned as a pilot evaluation to demonstrate the feasibility of MR imaging secretion measurements in combination with perfusion imaging results in the same examination. A further obstacle was the limited use of biopsy in the group with normal function. Therefore, we could have missed some minor, but clinically and therapeutically irrelevant, changes in allografts in those patients. An additional limitation was the composition of the study population. Since we could only include normal outcomes and patients with advanced stages of graft dysfunction, no conclusion about the ability to detect early stages of dysfunction could be made.
In summary, MR perfusion measurements proposed by Krebs et al (10) are useful in the differentiation of normal functioning pancreatic allografts from diseased grafts. We have found, however, that by adding the quantification of stimulated exocrine secretion not only is overlap between groups decreased but certainty of the diagnosis may be increased. Therefore, functional MR imaging with secretin-augmented MR pancreatography in combination with dynamic contrast-enhanced MR perfusion measurements appears to be a noninvasive, fast, and reliable diagnostic tool in the evaluation of graft function after pancreatic transplantation. To validate these findings, future studies must be conducted with larger patient populations.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, all authors; study concepts, J.T.H., H.J.W., A.H.; study design, all authors; literature research, J.T.H., A.L.L., K.J.K.; clinical studies, J.T.H., H.E., K.J.K., A.H.; data acquisition, all authors; data analysis/interpretation, J.T.H., H.J.W., A.L.L., K.J.K.; statistical analysis, J.T.H., H.E., A.H.; manuscript preparation, J.T.H., H.J.W., A.L.L., A.H.; manuscript definition of intellectual content, J.T.H., H.J.W., A.H.; manuscript editing, revision/review, and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMaterials and Methods ResultsDiscussionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. T. Heverhagen Noise Measurement and Estimation in MR Imaging Experiments Radiology, December 1, 2007; 245(3): 638 - 639. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
