| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gastrointestinal Imaging |
1 From the Department of Radiology, Hôpital Universitaire E. Herriot, 3 Place d'Arsonval, 69003 Lyon, France. From the 2005 RSNA Annual Meeting. Received August 25, 2005; revision requested October 25; revision received December 14; final version accepted March 20, 2006. Address correspondence to F.P. (e-mail: frank.pilleul{at}chu-lyon.fr).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
Materials and Methods: The study protocol was approved by the Human Research Committee of the institution, and all patients gave written informed consent. Two hundred nineteen patients (108 male, 111 female; age range, 17–98 years; mean, 53.1 years) with clinical suspicion of small-bowel neoplasm underwent contrast- and water-enhanced multidetector CT enteroclysis after normal findings of upper and lower gastrointestinal endoscopy. The prospective interpretations of CT enteroclysis results include evaluation of focal bowel wall thickening, small-bowel masses, small-bowel stenosis, mesenteric stranding, enlarged mesenteric lymph nodes, and visceral metastasis. Positive enteroclysis findings were compared with results of pathologic examination after surgical (n = 35) or endoscopic (n = 20) procedures. Negative results were compared with results of surgery (n = 8), enteroscopy (n = 15), capsule endoscopy (n = 14), and clinical follow-up (n = 127). Sensitivity, specificity, accuracy, and positive and negative predictive values were calculated on a per-patient basis with 95% confidence intervals.
Results: Findings of CT enteroclysis were positive in 55 cases and negative in 164. The overall sensitivity and specificity in identifying patients with small-bowel lesions were 84.7% and 96.9%, respectively. The negative and positive predictive values were 94.5% and 90.9%, respectively. Findings of pathologic examination confirmed small-bowel tumor in 50 patients with carcinoid tumor (n = 19), adenocarcinoma (n = 7), lymphoma (n = 5), jejunal adenoma (n = 9), stromal tumor (n = 5), ectopic pancreas (n = 2), angiomatous mass (n = 2), or metastasis (n = 1). Five examinations resulted in false-positive findings.
Conclusion: Contrast- and water-enhanced multidetector CT enteroclysis had an overall accuracy of 84.7% for depiction of small-bowel neoplasms.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
In contrast to conventional barium enteroclysis, cross-sectional imaging modalities have the potential to simultaneously demonstrate intraluminal, mural, and extraintestinal abnormalities (9). During the past 10 years, spiral computed tomographic (CT) enteroclysis has evolved into an established modality for the investigation of various small-bowel disorders (10,11). The thinner collimation possible with multidetector CT along with oral administration of water and an intravenous bolus of contrast material may improve the sensitivity of CT for depicting small-bowel tumors. Thus, the purpose of our study was to prospectively evaluate the sensitivity and specificity of contrast material–enhanced and water-enhanced multidetector CT enteroclysis (hereafter, CT enteroclysis) in depicting small-bowel neoplasms in symptomatic patients, with endoscopic, tissue, and follow-up findings as reference standards.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
CT Enteroclysis
CT enteroclysis was performed with the same protocol by one of three radiologists (P.J.V., F.P., and M.P., with 20, 12, and 5 years of experience, respectively, in interpretation of abdominal CT images). The patients fasted for 8 hours prior to the investigation without other digestive preparation. An 8-F nasojejunal tube (815 NF; Biosphere Medical, Louvres, France) was positioned into the proximal jejunum with fluoroscopic guidance. Room temperature water was infused with a pressure-controlled pump (NMP V01; Guerbet Group, Roissy, France) at a rate of 180–200 mL/min. The pressure was controlled at less than 1000 mm Hg in all patients. The quantity of water infused via the nasojejunal tube into the small bowel immediately before CT acquisition was 2500 mL in all patients. Sixty seconds before image acquisition, the patients received an intravenous administration of 15 mL of an antispasmodic drug (tiemonium methylsulphate, Viscéralgine; Laboratoires Organon, Puteaux, France) to reduce motion artifacts caused by small-bowel peristalsis.
CT enteroclysis was performed with a commercially available multidetector CT unit (Somatom Plus 4 Volume Zoom; Siemens, Erlangen, Germany) by using the following scanning parameters: 4 x 2.5 mm collimation (four detector rows and 2.5-mm section thickness), 120 kVp, 165 mAs, and 12.5-mm table speed per gantry rotation. Images were reconstructed at 3-mm intervals with the B30 soft-tissue algorithm and 1.5-mm image reconstruction. At the start of the procedure, 120 mL of nonionic iodinated contrast material (Iomeron 400; Altana, Milan, Italy) was injected intravenously through a 20-gauge cannula at a rate of 4 mL/sec with an automated power injector (Medex, Carlsbad, Calif). The delay between the start of contrast material administration and the start of helical scanning was 60 seconds. Images were obtained from the dome of the liver to the lower margin of the symphysis pubis during a single breath hold. The infusion of water was maintained during image acquisition. Multiplanar reformatted images were created interactively by using an attached workstation (Somatom Volume Wizard; Siemens). Reformatted images were routinely obtained in the coronal plane. For better analysis of abnormal bowel loops, the radiologist who performed the examination created a curved multiplanar reformation of individual loops. Processing of the reformatted images required approximately 15 minutes. Both transverse and multiplanar reformatted images were originally acquired with the standard window level (50 HU) and width setting (350 HU) that are used in our department for viewing the abdomen and pelvis.
CT enteroclysis was successfully performed in all patients. It was well tolerated in 211 patients; vomiting at the end of the procedure occurred in eight patients. The patients experienced no side effects from the intravenous administration of tiemonium methylsulphate or iodinated contrast material. The nasojejunal intubation was also well tolerated by all patients, with minor pain during the nasal introduction in 20 patients.
Image Analysis
CT enteroclysis images were prospectively interpreted by one of three experienced gastrointestinal radiologists (P.J.V., F.P., and M.P.) blinded to other imaging results. During the reading session, transverse and multiplanar reformatted enteroclysis images were specifically analyzed for the presence of focal bowel wall thickening, small-bowel masses, small-bowel stenosis, mesenteric stranding, enlarged mesenteric lymph nodes, and visceral metastasis. Small-bowel thickening was considered to be present if the thickness of the bowel wall was greater than 3 mm. Small-bowel stenosis was considered to be present if segmental narrowing of the bowel lumen could be seen on at least two planes. A small-bowel mass was considered to be present if a flat, sessile, or pedunculated structure could be identified on at least two planes and could be differentiated from small-bowel folds. Visceral metastasis was considered to be present if nodules were identified within the peritoneal cavity or hypointense lesions were identified within the liver. Mesenteric stranding was considered to be present if an infiltration of the mesenteric fat was observed. An enlarged mesenteric lymph node was considered to be present when a lymph node was greater than 10 mm in short-axis diameter.
Standard of Reference
As a standard of reference, we used intraoperative enteroscopy (n = 15), surgery without resection (n = 12), histopathologic findings from surgical (n = 35) or enteroscopic biopsy specimens (n = 13), videocapsule endoscopy (n = 14), and clinical follow-up (n = 127). In two other patients, CT enteroclysis helped identify pathognomonic sign of small-bowel lipoma. Intraoperative enteroscopy and nonsurgical push-type enteroscopy (performed with a videoenteroscope during general anesthesia; jejunal exploration was performed after introducing a gastric overtube, and the progression was ensured by means of a push-pull motion and was helped if necessary by means of manual compression of the abdomen and changing the patient's position; ileal exploration was performed after introducing a colonoscope up to the ileoceacal valve) were performed by the attending gastroenterologist (J.C.S., with 15 years of experience). Videocapsule endoscopy was performed with an imaging system (Pillcam type, Given Diagnostic System; Given Imaging, Yoqneam, Israel) that comprised three main components: an M2A capsule, a data recorder, and an image and data reporting and processing workstation. Of the 219 patients who underwent enteroclysis, 89 patients underwent at least one invasive procedure (ie, intraoperative enteroscopy, push enteroscopy, histopathologic study, and/or videocapsule endoscopy). One hundred thirty patients had clinical follow-up at least 2 years after enteroclysis that was used a standard of reference.
Statistical Analysis
The numbers of missed small-bowel lesions (thickening, stenosis, or masses) and correctly detected lesions at the prospective study of enteroclysis images were counted and further analyzed according to their histopathologic diagnosis and/or endoscopic identification. The sensitivity in lesion detection was defined as the true-positive rate, that is, the number of patients with small-bowel disease who were correctly identified as having the disease at CT enteroclysis divided by the number of patients with small-bowel disease who were actually present in our study population (as defined by the standard of reference), with 95% confidence intervals (CIs). Specificity (with 95% CIs) was calculated on a per-patient basis. Accuracy was defined as the number of cases of small-bowel disease that were correctly diagnosed in our study population. Positive predictive value was defined as the probability that the patient had small-bowel disease when restricted to all patients with small-bowel disease at CT enteroclysis. Negative predictive value was defined as the probability that the patient had no small-bowel disease when restricted to all patients without small-bowel disease at CT enteroclysis.
The Youden index is defined as sensitivity plus specificity minus 1 and represents the proportion of patients correctly classified at CT enteroclysis. The Yule Q is based on the odds ratio and a symmetric measure that takes values between –1 and + 1; 1 implies perfect negative or positive association and 0 implies no association.
Statistical analysis was performed with commercially available software (SPSS 10.0 for Windows; SPSS, Chicago, Ill).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
CT enteroclysis helped characterize at the same time extradigestive disease in 15 patients: carcinoid tumors (n = 12), lymphoma (n = 2), and jejunal adenocarcinoma (n = 1). Extradigestive disease was consistent with liver metastases in eight patients; with peritoneal metastases, in seven patients; and with enlarged mesenteric lymph node, in six patients. In these cases, reformatted images did not provide additional information when compared with transverse CT enteroclysis images.
Carcinoid tumors visible at CT enteroclysis were 5–30 mm in axial diameter. These tumors were depicted as focal nodular lesions that were located in the small-bowel wall and showed marked enhancement after intravenous administration of iodinated contrast material (Fig 2). A stromal tumor was detected with CT enteroclysis in five patients. Three of these patients had a benign stromal tumor, and two had a malignant stromal tumor, as documented with histopathologic analysis after surgical resection in all patients (Fig 3). Seven small-bowel adenocarcinomas were detected with CT enteroclysis (Fig 4) and were confirmed with histopathologic findings after surgical removal. These tumors were detected as a mass in five patients and as a stenosis in two patients. Lower and upper endoscopic procedures were performed before CT enteroclysis in all patients.
|
|
|
|
Negative findings.—Negative findings from CT enteroclysis were obtained in 164 patients. In this group, 128 patients underwent only a clinical follow-up and 36 patients underwent other investigations, including enteroscopy, videocapsule endoscopy, or histopathologic examination. Interpretation of CT enteroclysis images resulted in false-negative findings in nine patients. In four patients, the cause of gastrointestinal bleeding was found at enteroscopy: jejunal polyps smaller than 10 mm in context of familial disease (n = 3) and ileal angiodysplasia (n = 1). In the remaining five patients, the cause of substantial abdominal pain was found at surgery (n = 2), videocapsule endoscopy (n = 1), and push enteroscopy (n = 2). Invasive examination demonstrated polyps (n = 3) and 3–5-mm ectopic pancreas (n = 2). The clinical follow-up was at least 24 months (up to 43 months). We believe that no patient developed any disease that might have been missed at CT enteroclysis.
Overall results.—Among 219 patients who underwent CT enteroclysis for possible small-bowel tumors after negative findings of lower and upper endoscopy, false-positive findings were obtained in five (2.3%), false-negative findings were obtained in nine (4.1%), true-positive findings were obtained in 50 (22.8%), and true-negative findings were obtained in 155 (70.8%) patients. Sensitivity of CT enteroclysis for the depiction of small-bowel disease was 84.7% (95% CI: 75.5%, 93.9%), specificity was 96.9% (95% CI: 94.2%, 99.6%), positive predictive value was 90.9% (95% CI: 83.3%, 98.5%), negative predictive value was 94.5% (95% CI: 91.1%, 97.9%), and accuracy was 84.7%. The Youden index was 0.82, and the Yule Q was 0.99.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
In addition, CT enteroclysis allowed the detection of liver metastases and/or peritoneal carcinomatosis in the same session in 15 patients. These results suggest that CT enteroclysis facilitates an adequate evaluation of patients with small-bowel tumors. Our results suggest that CT enteroclysis could replace the combination of abdominopelvic CT and barium follow-through examination that is commonly performed for the evaluation of patients with small-bowel tumors, thus decreasing the patient's exposure to ionizing radiation. In a preliminary study (11), CT enteroclysis was evaluated in a series of 48 patients with small-bowel tumors, and the authors concluded that it was more sensitive than small-bowel follow-through and less invasive than enteroscopy. Although direct comparison between the results obtained in our preliminary study is difficult because the technique and patient population vary, our results confirmed the usefulness of CT enteroclysis.
Five patients had false-positive findings that were a result of a pseudotumor (which proved be a simple fold thickening or normal mucosa). Although these false-positive findings had a negative effect on the specificity of CT enteroclysis, we found the same specificity of push enteroscopy for the detection of small-bowel lesions (13). In addition, nine patients had false-negative findings.
In our study, we found that infusion of water with the automatic pump was well tolerated in 211 patients, with vomiting observed in eight patients. We routinely used transverse CT enteroclysis images in association with multiplanar reformatted images. We found that the multiplanar reformatted images provided additional information in the group of patients suspected of having small-bowel polyps. In this particular group of patients, multiplanar reformatted images were helpful in determining the site and presence of a polyp compared to a simple fold thickening.
Furthermore, wireless capsule endoscopy was performed in 14 patients and was more informative than CT enteroclysis for one patient. In some series, however, wireless capsule endoscopy demonstrated more lesions than did CT enteroclysis (14). Methods used to determine the standard of reference and lack of pathologic proof probably explained the discrepancy. Although capsule endoscopy seems to be an excellent diagnostic tool for small-bowel disease, tumors can be missed with this technique (15) and a barium study is required to exclude small-bowel stricture. Furthermore, only one study (16) compared findings of capsule endoscopy with those of CT enteroclysis in eight patients without significant results.
The main limitations of the present study were that use of clinical follow-up as a standard of reference allows smaller and slow-growing tumors to be missed. Although we did not directly address this in our study, small-bowel loop distention can have a strong influence on tumoral detection. To minimize this bias, the intubation technique provides a gastric bypass for rapid enteral administration, a prerequisite of homogeneous small-bowel distension (16). As in the study of Boudiaf et al (16), we overestimated the sensitivity of CT enteroclysis because clinical follow-up was our main standard of reference for negative findings (164 of 219, 75%). Furthermore, the design of our study provides a limitation of verification bias, because for ethical reasons we decided to perform invasive examinations only in patients with a positive finding at CT enteroclysis.
In conclusion, CT enteroclysis allows depiction of small-bowel neoplasms with an accuracy of 84.7%. We believe that CT enteroclysis should be used in patients who are suspected of having small-bowel neoplasms with negative findings of upper and lower endoscopic procedures.
|
ADVANCE IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: CI = confidence interval
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, F.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, F.P., M.P., L.M.; clinical studies, F.P., M.P., J.C.S., J.A.C., P.J.V.; experimental studies, F.P., M.P., J.C.S., J.A.C., P.J.V.; statistical analysis, F.P., M.P., L.M.; and manuscript editing, F.P., M.P., L.M., P.J.V.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Masselli, E. Polettini, E. Casciani, L. Bertini, A. Vecchioli, and G. Gualdi Small-Bowel Neoplasms: Prospective Evaluation of MR Enteroclysis Radiology, June 1, 2009; 251(3): 743 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Soyer, M. Boudiaf, Y. Fargeaudou, X. Dray, L. Hamzi, K. Vahedi, A. Lavergne-Slove, and R. Rymer Celiac Disease in Adults: Evaluation with MDCT Enteroclysis Am. J. Roentgenol., November 1, 2008; 191(5): 1483 - 1492. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Matsumoto, M. Esaki, S. Yada, Y. Jo, T. Moriyama, and M. Iida Is Small-Bowel Radiography Necessary Before Double-Balloon Endoscopy? Am. J. Roentgenol., July 1, 2008; 191(1): 175 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Ehman, W. R. Hendee, M. J. Welch, N. R. Dunnick, L. B. Bresolin, R. L. Arenson, S. Baum, H. Hricak, and J. H. Thrall Blueprint for Imaging in Biomedical Research Radiology, July 1, 2007; 244(1): 12 - 27. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
