MR Imaging Evaluation of the Activity of Crohn's Disease
Abstract
OBJECTIVE. The purpose of this study was to evaluate the sensitivity and specificity of MR imaging in assessing the activity of Crohn's disease.
SUBJECTS AND METHODS. Thirty symptomatic patients with Crohn's disease but uncertain disease activity were prospectively examined using MR imaging. Twenty-nine patients were scored using the Crohn's disease activity index. Six hundred milliliters of water orally and 1 mg of glucagon intramuscularly were given before imaging. Breath-hold images were obtained using T2-weighted turbo spin-echo, T1-weighted fast low-angle shot, and fat-suppressed gadolinium-enhanced T1-weighted fast low-angle shot sequences. Images were assessed by two radiologists who were unaware of the patient's symptoms, clinical scoring, and other imaging tests, and who reached a consensus about the imaging findings (bowel wall thickening, bowel wall enhancement, and perienteric changes) and determined the absence or presence of active disease in each patient. MR imaging findings were correlated with endoscopy and surgery.
RESULTS. Twenty-three patients had active disease and seven patients had inactive disease. One hundred twenty-four of a total of 168 bowel segments were examined with both MR imaging and endoscopy or surgery. On a per patient basis, MR imaging had an overall sensitivity of 91% and a specificity of 71% for active disease. The Crohn's disease activity index had a sensitivity of 92% and a specificity of 28%. On a per segment basis, MR imaging had a sensitivity of 59% and a specificity of 93%. Bowel wall thickening of greater than 4 mm, bowel wall enhancement (ratio of signal intensity of abnormal to normal bowel > 1.3:1), and increased mesenteric vascularity were useful in identifying active disease. A layered enhancement pattern after the IV administration of gadolinium was highly specific for active inflammation.
CONCLUSION. MR imaging is useful in assessing the activity of Crohn's disease and may be helpful when clinical scoring is equivocal.
Introduction
Crohn's disease is characterized by remitting and relapsing episodes. Abdominal symptoms may be nonspecific and can result from active inflammation or from fibrotic scarring and stricture formation. Activity assessment is important to identify patients with active inflammation so that optimal therapy may be prescribed. In the light of recent advances in the treatment of active Crohn's disease, the need for accurate activity assessment has become even more important. Antibodies to tumor necrosis factor α have been successfully used to treat patients with active inflammation or fistulas that are refractory to conventional medical therapy [1]. However, treatment with anti—tumor necrosis factor α is expensive and can produce side effects.
Clinical scoring (such as the Crohn's disease activity index [2]), biologic indexes [3], endoscopy, and imaging studies have all been used to monitor activity, but no established gold standard exists. Assessment of activity is usually made using a combination of clinical symptoms, physical findings, laboratory investigations, endoscopy, and imaging tests. The assessment of biologic activity, based on the positivity for three of four acute phase reactants (WBC, erythrocyte sedimentation rate, C-reactive protein, and orosomucoids), has been found to be a sensitive determinant of activity [3], especially when supported by endoscopic or imaging findings [3].
Imaging studies such as leukocyte scintigraphy [4], CT [5, 6], barium examinations [7], sonography [8], MR imaging [3, 9, 10], and positron emission tomography [11] have been used to assess activity. Except for sonography and MR imaging, these modalities use ionizing radiation, which may result in a considerable radiation burden with repeated examinations.
Breath-hold MR imaging can provide assessment of the entire gastrointestinal tract, but its role in Crohn's disease activity assessment is still evolving. The hypothesis is that MR imaging would be useful in detecting actively inflamed segments of bowel, as reflected by an increase in bowel wall thickness and enhancement with IV gadolinium, with or without accompanying perienteric changes.
The purpose of our study was to evaluate the sensitivity and specificity of MR imaging in assessing the activity of Crohn's disease in the gastrointestinal tract.
Subjects and Methods
Patient Population
We undertook a prospective study of 30 clinically symptomatic patients with previously proven Crohn's disease who were referred for MR imaging by gastroenterologists. The symptoms ranged widely from mild, nonspecific abdominal discomfort and dyspepsia to frequent bloody diarrhea and weight loss. The clinical symptoms were withheld from the radiologists. The study population reflected a cross section of patients who are undergoing regular follow-up in the gastroenterology outpatient clinic. Patients were also clinically scored for disease activity using the Crohn's disease activity index [2]. Twenty-nine patients completed the questionnaire required for the scoring (score range, 95-677; mean, 285).
The patients were 14 men and 16 women with a mean age of 37.6 years (range, 18-58 years). The mean duration of illness was 23.8 months (range, 2 months-12 years). Twelve patients had undergone a previous right hemicolectomy and two patients had previous ileal resections. Of the latter two, two patients had an ileostomy and two patients had a colostomy.
Imaging Technique
MR imaging was performed using a 1-T system (Magnetom Impact; Siemens, Erlangen, Germany). All patients drank 600 mL of water 30 min before the examination. Intramuscular glucagon (1 mg) was also administered before imaging. Breath-hold imaging of the abdomen was first performed in the coronal plane using a T2-weighted turbo spin-echo sequence (TR/TE, 2730/13 msec; number of excitations, 1; matrix, 256 × 192; field of view, 350-480 mm; and 5-mm contiguous slices). A large field of view was used to include the entire abdomen between the diaphragm superiorly and the symphysis pubis inferiorly. These sequences were supplemented with T1-weighted fast low-angle shot imaging in the coronal, axial, and sagittal planes (100/6.6; number of excitations, 1; flip angle, 75°; matrix, 256 × 192; and 5-mm sections).
Gadodiamide (Omniscan, 0.1 mmol/kg; Nycomed Amersham, Buckinghamshire, United Kingdom) was administered IV as a hand-injected bolus with a saline flush. Image acquisition began 20 sec after contrast administration using a T1-weighted fast low-angle shot sequence (180/4.8; flip angle, 75°; matrix, 256 × 192; and 5-mm sections) with fat suppression in the axial plane. Images were also acquired in the sagittal and coronal planes.
Image Analysis
MR images were interpreted by two radiologists who reached a consensus without being aware of the clinical scoring and the results of endoscopy, surgery, and other imaging tests.
For the purpose of analysis, the gastrointestinal tract was divided into six segments. These segments consisted of the duodenum and small bowel, the terminal ileum (terminal 10 cm of the native ileum or within 10 cm of an ileocolic anastomosis or ileostomy), the cecum and ascending colon, the transverse colon, the descending colon, and the rectosigmoid. The perianal region was not specifically assessed in this study.
The MR images were reviewed for the number, location, and length (in centimeters) of abnormal bowel segments. A segment of bowel was deemed to be abnormal if it was thickened (wall thickness > 3 mm) or showed increased enhancement after the IV administration of gadolinium. A segment was determined to be actively inflamed if it showed both thickening and abnormal enhancement.
The maximum bowel wall thickness (in millimeters) was obtained on a cross-section T1-weighted image of the thickened bowel using the distance-measuring tool on the display console. In segments in which no wall thickening was perceptible and in which measurement with the tool was not achievable, a measurement of 1 mm for wall thickness was assumed.
The signals returned from abnormal bowel segments on the unenhanced T1- and T2-weighted images were scored visually as to whether they were hypointense, isointense, or hyperintense to the psoas muscle.
The gadodiamide-enhanced fat-suppressed T1-weighted images were surveyed visually to identify loops that were thickened, that showed increased signal compared with adjacent loops, or both. A region of interest was used to obtain the maximum signal intensity from the wall of the abnormal segment. On the same image, another region of interest was placed over normal-appearing bowel. This region of interest was placed remotely from the abnormal segment, usually along adjacent loops of small bowel or segments of collapsed small or large bowel. Care was taken to avoid areas in which the fat suppression was not uniform.
The enhancement of abnormal and normal-appearing bowel was normalized by dividing the signal intensity of these by the signal intensity returned from the psoas muscle on the same fat-suppressed gadolinium-enhanced T1-weighted image. In patients in whom no abnormal segment was identified, the signal intensities from two normal-appearing loops of bowel were obtained and subjected to the same analysis.
For each abnormal segment observed, three normalized values of the signal intensities of abnormal and normal-appearing bowel were obtained, and the average was recorded. In addition, the ratio of the signal intensity of abnormal bowel to the signal intensity of normal-appearing bowel was also computed and recorded.
The mean signal intensities of abnormal and normal-appearing bowel in patients with active and inactive disease, as determined by our gold standard tests, were compared. The median of the ratio of the signal intensity of abnormal to normal-appearing bowel was also compared between patients with active and inactive disease using the nonparametric Mann-Whitney test.
The pattern of enhancement after the IV administration of gadodiamide was also assessed. A diffuse enhancement pattern was recorded when the bowel wall enhanced uniformly. A layered enhancement pattern was characterized by intense enhancement of the mucosa and serosa, with lesser enhancement of the layers between.
We recorded perienteric changes such as an increase in mesenteric vascularity, fatty proliferation, lymphadenopathy, fistula, or abscess. Increase in mesenteric vascularity was recorded if an increase was noted in the prominence and dilatation of the vasa recta supplying the small bowel or colon. Fatty proliferation was deemed to be present if there was an increase in fat resulting in separation or displacement of adjacent bowel loops. Lymphadenopathy was defined as lymph nodes with a short-axis diameter greater than 5 mm in the mesentery, or greater than 1 cm in the retroperitoneum.
After reviewing the images, the radiologists reached agreement as to the presence or absence of active disease in each patient. A patient was deemed to have active disease if one or more inflamed bowel loops were identified that showed abnormal wall thickening (>3 mm) and increased bowel wall enhancement, with or without perienteric changes. The visualization of fistula or abscess was also taken as evidence of active disease.
Comparison with Gold Standard
A total of 124 segments identified on MR imaging were correlated with the findings on endoscopy, surgery, or both. The median time between MR imaging and the comparative test was 21 days.
On a per patient basis, 26 patients had colonoscopy, of which direct visualization of the terminal ileum was achieved in 20 patients. Two patients had both colonoscopy and surgery. Two patients went directly to surgery. Confirmation of disease activity by endoscopy and surgery on a per segment basis is summarized in Table 1.
Segment | Endoscopy | Surgery | Endoscopy and Surgery |
---|---|---|---|
Ileum | 0 | 1 | 0 |
Terminal ileum | 20 | 1 | 0 |
Cecum and ascending colon | 16 | 1 | 1 |
Transverse colon | 27 | 0 | 1 |
Descending colon | 27 | 0 | 1 |
Rectosigmoid | 27 | 0 | 1 |
Visualization of inflamed mucosa in the colon or terminal ileum on colonoscopy, with acute inflammatory infiltrates on biopsy, was taken as evidence of active disease. Among patients who underwent surgery (one right hemicolectomy, one right hemicolectomy and ileal resection, one ileal resection, and one subtotal colectomy), active inflammation in the resected surgical specimen provided proof of activity.
Results
On the basis of the gold standard tests, 23 patients had active disease and seven had inactive disease. Overall, on a per patient basis, MR imaging had a sensitivity of 91% and a specificity of 71% in detecting active disease. The Crohn's disease activity index had a similar sensitivity of 92% but a low specificity of only 28%.
Length, Number, and Location of Abnormal Segments
In the 23 patients with active disease, the mean length of active bowel segments identified on MR imaging was 14.2 cm (range, 0-30 cm). The mean number of active segments identified was 1.2 (range, 0-4).
Of the 124 segments assessed with both MR imaging and the gold standard tests, MR imaging correctly identified active disease in 24 of the 41 actively inflamed segments (Table 2). On a per segment basis, MR imaging had an overall sensitivity of 59% and specificity of 93% (Table 2) but a sensitivity of 89% and a specificity of 67% for disease in the terminal ileum.
Segment | True-Positive | False-Negative | True-Negative | False-Positive |
---|---|---|---|---|
Duodenum and small bowel | 1 | 0 | 0 | 0 |
Terminal ileum | 14 | 1 | 4 | 2 |
Ascending colon | 3 | 6 | 10 | 1 |
Transverse colon | 2 | 4 | 21 | 0 |
Descending colon | 2 | 3 | 23 | 0 |
Rectosigmoid | 2 | 3 | 19 | 3 |
Total | 24 | 17 | 77 | 6 |
Note.—These findings resulted in a sensitivity of 59% and a specificity of 93%. |
The distribution of the 41 actively inflamed segments included one ileum (2%); 15 terminal ilea, ileocolic anastomoses, or ileostomies (37%); nine ceca and ascending colons (22%); six transverse colons (15%); five descending colons (12%); and five rectosigmoids (12%). Of the 17 actively inflamed segments that were missed on MR imaging, 16 were in the colon (Table 2). In addition, MR imaging falsely diagnosed active disease in six segments: two in the terminal ileum, one in the ascending colon, and three in the rectosigmoid.
Bowel Wall Thickening
Overall, the mean maximal wall thickness was 6.7 mm (range, 2-12 mm) in patients with active disease and 3.3 mm (range, 1-5 mm) in patients with inactive disease (Table 3). The mean maximal wall thickness measured on MR imaging was significantly greater in the group of patients with active disease (p<0.01, Student's t test) (Fig. 1A,1B).
Finding | Active Disease (n = 23) | Inactive Disease (n = 7) | p |
---|---|---|---|
Layered enhancement pattern | 7 | 0 | <0.01a |
Increase in mesenteric vascularity | 18 | 3 | <0.03a |
Perienteric fatty proliferation | 7 | 3 | >0.05a |
Mesenteric lymphadenopathy | 2 | 0 | >0.05a |
Abscess | 1 | 0 | >0.05a |
Enteric fistula | 2 | 0 | >0.05a |
Mean bowel wall thickness (mm) | 6.7 | 3.3 | <0.01b |
Median SI (abnormal) / median SI (normal) | 1.6 | 1.0 | <0.01c |
Note.—SI = signal intensity. |
a
Fisher's exact test.
b
Student's t test.
c
Nonparametric Mann-Whitney test.
In addition, the bowel wall thickening was found to be concentric rather than eccentric in most cases (86%). On a per segment basis, a value of 4 mm or greater for abnormal bowel wall thickening has a sensitivity of 54% and a specificity of 98% for active disease (Table 4). On a per patient basis, wall thickening of 4 mm or greater has a sensitivity of 88% and a specificity of 75% for active disease.
Imaging Feature | True-Positive | False-Negative | True-Negative | False-Positive | Sensitivity (%) | Specificity (%) |
---|---|---|---|---|---|---|
Per segment | ||||||
Mean bowel wall thickness > 4 mm | 22 | 19 | 81 | 2 | 54 | 98 |
Median SI (abnormal) / median SI (normal) > 1.3:1 | 28 | 13 | 78 | 5 | 68 | 94 |
Per patient | ||||||
Mean bowel wall thickness > 4 mm | 22 | 3 | 5 | 2 | 88 | 75 |
Median SI (abnormal) / median SI (normal) > 1.3:1 | 18 | 5 | 6 | 1 | 78 | 86 |
Increase in perienteric vascularity | 18 | 5 | 4 | 3 | 78 | 57 |
Note.—SI = signal intensity. |
Bowel Wall Characteristics
In most patients with active disease (73%), abnormal bowel identified on MR imaging was isointense or slightly hypointense to the psoas muscles on T1-weighted imaging. On T2-weighted imaging, the abnormal bowel segments were usually isointense or slightly hyperintense compared with the psoas muscle (98%).
Among the 24 segments in which MR imaging correctly diagnosed active disease, seven had a layer of high T1 and T2 signal in the muscularis and submucosa layers. Another two segments showed areas of low T1 signal and high T2 signal in the muscularis and submucosa. None of the false-positive segments showed either of these appearances.
Bowel Wall Enhancement with IV Gadolinium
In active disease, the signal intensities obtained from the 24 segments of actively inflamed bowel identified on MR imaging were significantly higher than those of the normal-appearing bowel loops after the IV administration of gadolinium (p<0.01, paired t test).
The ratio of the signal intensity of abnormal to normal bowel was also significantly greater in patients with active disease than in those with inactive disease (p<0.05, Mann-Whitney test). In our study, a ratio of 1.3:1 or more (Table 4) had a sensitivity of 68% and a specificity of 94% for active disease on a per segment basis.
Pattern of Enhancement
Of the 24 segments in which MR imaging correctly identified active disease, the enhancement pattern of abnormal bowel was diffuse in 71% (17/24) and layered in 29% (7/24). The layered pattern (Fig. 2A,2B) was seen only in patients with active disease (Table 3). The six segments with false-positive findings showed mild diffuse enhancement.
Perienteric Changes
An increase in the vascularity of the mesentery was seen in 18 (78%) of 23 patients with active disease and three (43%) of seven patients with inactive disease. Although a trend was noted for increased mesenteric vascularity to be more frequently observed in patients with active disease, the trend did not reach statistical significance (Fig. 1B). On a per patient basis, an increase in mesenteric vascularity had a sensitivity of 78% and a specificity of 57% for active disease.
Perienteric fatty proliferation (Fig. 3) was seen in seven (30%) of 23 patients with active disease and three (43%) of seven patients with inactive disease. No statistical difference was seen between the two groups (p>0.05, Fisher's exact test).
Other Findings
A fibrotic stricture with discernable wall thickening or enhancement after the administration of IV gadolinium was seen in one patient with inactive disease (Fig. 7A,7B). On MR imaging, smooth tapering and caliber change were noted at the third part of the duodenum and were associated with proximal dilatation of the duodenum and stomach.
Correlation of MR Imaging Findings with Crohn's Disease Activity Index Score
No statistically significant correlation was found between the Crohn's disease activity index score and the MR imaging findings of wall thickness, the length of bowel involved, and the contrast enhancement ratio of abnormal to normal bowel.
Discussion
Although the role of MR imaging in the evaluation of perianal fistulas is well established [12, 13], the use of MR imaging to study the small bowel and colon has been limited by image acquisition time and suboptimal image quality. However, high-quality images of the gastrointestinal tract can now be obtained using the new ultrafast MR imaging sequences with breath-hold or respiratory triggering. MR imaging can provide multiplanar imaging without the use of ionizing radiation that may identify active disease and thus allow optimal treatment to be instigated. MR imaging may be particularly useful in the evaluation of patients presenting with subacute symptoms.
The MR imaging technique for evaluating Crohn's disease in the gastrointestinal tract is still evolving [3, 8, 9, 14,15,16,17,18,19,20,21,22,23]. Imaging protocols that have been tested include unenhanced and gadolinium-enhanced T1-weighted imaging [3, 8, 9, 21]; T2-weighted imaging with fat saturation [3]; half-Fourier single-shot turbo spin-echo MR enteroclysis with water taken orally [15,16,17]; and MR enteroclysis using methylcellulose, barium, gadolinium, or superparamagnetic iron oxide particles for intraluminal filling [14].
Many recent MR studies have focused on obtaining optimal bowel filling and distention. These techniques require large volumes (>1000 mL) of positive or negative contrast agents to be given orally [17, 19] or via a nasojejunal or rectal tube [14, 15]. However, patients who are acutely ill may not tolerate a large oral fluid load. Other MR studies in the literature have used less [3, 19] or no [8, 9, 17] oral fluids with reportedly good results. Furthermore, despite large oral volumes, distention of the distal small bowel, where disease is most likely to occur, can still be poor.
A recent study compared contrast-enhanced MR imaging (with 1350 mL of oral barium sulfate and rectal water contrast) with single-phase helical CT scanning [18]. In that study [18], MR imaging identified 80-85% of abnormal bowel segments, compared with 60-65% with CT. In our study, MR imaging identified 59% of all segments showing active inflammation, which is comparable to CT for identifying actively inflamed segments. However, our lower detection rate can in part be attributed to suboptimal distention of the colon. Air in the colon can cause substantial susceptibility artifacts. The contents of the colon frequently appear to have high signal. Both factors conspire to further reduce the perception of the colonic wall after the administration of IV contrast material. Of the 17 segments of actively inflamed bowel that were missed in our study, 16 were in the colon. The routine administration of water or saline rectally could significantly improve the detection of colonic abnormalities.
Overall, MR imaging has a high sensitivity (92%) and specificity (75%) for detecting active disease on a per patient basis. Although the pretest probability of active disease was high in our study population, the patients' presenting complaints, clinical findings, laboratory investigations, and other imaging tests were withheld from the radiologists to minimize interpretation bias. We believe the sensitivity and specificity of MR imaging on a per patient basis is useful because the presence of active disease on imaging, regardless of the number of segments involved, can significantly alter the treatment plan, which is tailored toward individual patients rather than individual segments.
The sensitivity and specificity of MR imaging in detecting activity on a per segment basis were 59% and 93%, respectively. In theory, the sensitivity and specificity on a per segment basis should provide a more accurate assessment of the capability of MR imaging in detecting all actively inflamed bowel segments. However, in our study design, an arbitrary segmentation of the gastrointestinal tract was made, in which four of the six segments assessed were located in the colon. Hence, two patients with undiagnosed superficial ulcerations of the entire colon on MR imaging translated to eight missed segments of active disease. This occurrence produced a bias toward a greater false-negative rate and lower sensitivity for MR imaging on per segment analysis.
Of the two patients with false-negative assessment on MR imaging, one had mild inflammation of the ascending and transverse colons on colonoscopy. The other had inflammation of the terminal ileum and the ascending and transverse colons on endoscopy. These changes were not apparent, even on retrospective review of the images. In the two patients with false-positive diagnoses, MR imaging showed mild wall thickening and enhancement of the distal ileum in one patient and of the rectosigmoid in the other patient.
In our study, we found that bowel wall thickening of more than 4 mm and bowel wall enhancement (ratio of signal intensity of abnormal to normal bowel > 1.3:1) were of value in predicting activity. Although a trend was seen toward increased mesenteric vascularity in patients with active disease, the trend did not quite reach statistical significance. In an earlier study, Shoenut et al. [9] found a significant correlation between the severity of the disease on endoscopy with the wall thickness, the length of bowel involvement, and the percentage of contrast enhancement. More recently, a good statistical correlation was shown between the biologic activity and the MR imaging parameters of wall enhancement, wall T2 hyperintensity, and T2 hyperintensity of fibrofatty proliferation on fat-suppressed images [3].
In our series, although fatty proliferation was more frequently observed among patients with active disease than in those with inactive disease, this was not statistically significant. In acute disease, changes in the perienteric fat may result from edema and inflammation in the perienteric fat. Other perienteric changes such as mesenteric lymphadenopathy, fistulas, and abscesses, although relatively uncommon in our series, were observed only in patients with active disease.
The layered pattern of enhancement appears highly specific for active inflammation because it was observed only in segments with proven active disease. To our knowledge, the layered enhancement pattern on MR imaging has not been previously described. In the earlier study of Shoenut et al. [9], transmural enhancement, resembling our description of diffuse enhancement, was seen in all 12 patients with Crohn's disease. The layered enhancement appears analogous to the mural stratification pattern, which has been described on CT [6] and sonography [23, 24]. Visualization of mural stratification on CT or sonography represents active inflammation before the onset of fibrosis, a condition that is potentially reversible with appropriate medical treatment [6, 23, 24].
The pathogenesis and significance of the layered enhancement pattern is uncertain, because none of the segments exhibiting the appearance in our study had surgery or full-thickness biopsy to provide pathologic correlation. However, because the earliest inflammatory changes of Crohn's disease occur in the submucosa, the layered appearance may result from edema in the submucosa, perhaps with associated edema in the muscularis layer. As a result, enhancement of the bowel wall after the IV administration of gadolinium is more apparent over the mucosal and serosal surfaces. The hypothesis is similar to the explanation given for mural stratification observed on CT and sonography [6, 23, 24]. However, of the seven segments that showed the layered enhancement pattern, only two showed low T1 signal and high T2 signal in the submucosa on the unenhanced images, which was consistent with edema. Another two segments exhibited high T1 signal in the submucosa before contrast administration, probably related to fatty proliferation or, less likely, to hemorrhage in the submucosa.
Quantification of what constitutes enhancement is difficult because normal bowel wall shows enhancement after IV gadolinium is administered [12, 25]. In the seven segments with false-positive findings, a perceptible increase in enhancement of the bowel wall was noted. This increased enhancement may reflect local bowel hyperemia that is not related to inflammation [20]. Although it is relatively easy to place a region of interest over a loop of thickened enhancing bowel, an accurate measurement of normal bowel can be challenging, because it is frequently thin. In addition, although recognizing abnormal bowel is easy, the measurement of presumed normal bowel is more subjective. Nevertheless, if we had measured an abnormal segment for the signal intensity of normal-appearing bowel, this would almost certainly have introduced a negative bias on the ratio of the signal intensity of abnormal to normal-appearing bowel. Despite this potential bias, we found that a ratio of the signal intensity of abnormal to normal-appearing bowel of greater than 1.3:1 could be helpful in identifying segments of active disease.
The Crohn's disease activity index, although it has similar high sensitivity (92%), lacks specificity (25%). The Crohn's disease activity index score is not universally accepted because it is derived predominantly from subjective scoring of clinical symptoms, and the perception of the severity of these symptoms can vary considerably among individuals. The assessment of the biologic activity, based on positivity of four laboratory markers, is reportedly a more accurate predictor of activity [3].
The limitations of our study include the relatively small number of patients, who were limited to symptomatic individuals. Furthermore, analysis of the sensitivity and specificity of MR imaging was based almost totally on segments in the colon and the terminal ileum. A patient having activity in the proximal ileum might be missed by our gold standard tests. However, in analyzing the 10 patients in our series who also had barium follow-up examinations, six abnormal segments were identified in the small bowel. MR imaging showed active disease in all six segments. Because these segments were not verified by our gold standard test, they were excluded from the analysis. Nevertheless, it appears that MR imaging can identify active segments in the small bowel. This finding can be further investigated by comparing MR imaging with labeled leukocyte scintigraphy, which has a reported sensitivity of 96% and specificity of 98% [3]. Unfortunately, labeled leukocyte scintigraphy was not available at our institution.
In conclusion, MR imaging is useful in assessing the activity of Crohn's disease. Bowel thickening of more than 4 mm, increase in bowel wall enhancement, and increased mesenteric vascularity may be useful discriminatory signs of active disease. In addition, a layered enhancement pattern after the IV administration of gadolinium appears to be highly specific for active disease. Crohn's disease activity index scoring has a low specificity; and in patients with equivocal clinical findings, MR imaging can be recommended to provide disease activity assessment. Unlike CT, MR imaging does not use ionizing radiation and may be safely repeated.
Acknowledgments
We thank Erica Scurr and Amanda Phillips for their valuable help in performing the MR imaging studies.
Footnote
Address correspondence to D. M. Koh.
References
1.
Bell S, Kamm MA. Antibodies to tumour necrosis factor a as treatment for Crohn's disease. Lancet 2000; 335:858-859
2.
Best WR, Becktel JM, Singleton JW, Kern F. Development of a Crohn's disease activity index: National Cooperative Crohn's Disease Study. Gastroenterology 1976; 70:439-444
3.
Maccioni F, Viscido A, Broglia L, et al. Evaluation of Crohn disease activity with magnetic resonance imaging. Abdom Imaging 2000; 25:219-228
4.
Pullman WE, Sullivan PJ, Barratt PJ, Lising L, Booth JA, Doe WF. Assessment of inflammatory bowel activity by technetium 99m phagocyte scanning. Gastroenterology 1988; 95:989-996
5.
Meyers MA, McGuire PV. Spiral CT demonstration of hypervascularity in Crohn disease: “vascular jejunization of the ileum” or the “comb sign.” Abdom Imaging 1995; 20:327-332
6.
Gore RM, Balthazar EJ, Ghahremani GG, Miller FH. CT features of ulcerative colitis and Crohn's disease. AJR 1996; 167:3-15
7.
Yao T, Okada M, Fuchigami T, et al. The relationship between the radiological and clinical features in patients with Crohn's disease. Clin Radiol 1989; 40:389-392
8.
van Oostayen JA, Wasser MN, van Hogezand RA, Griffioen G, de Roos A. Activity of Crohn disease assessed by measurement of superior mesenteric artery flow with Doppler US. Radiology 1994; 193:551-554
9.
Shoenut JP, Semelka RC, Silverman R, Yaffe CS, Micflikier AB. Magnetic resonance imaging in inflammatory bowel disease. J Clin Gastroenterol 1993; 17:73-78
10.
Shoenut JP, Semelka RC, Magro CM, Silverman RS, Yaffe CS, Micflikier AB. Comparison of magnetic resonance imaging and endoscopy in distinguishing the type and severity of inflammatory bowel disease. J Clin Gastroenterol 1994; 19:31-35
11.
Bicik I, Bauerfeind P, Breitbach T, von Schulthess GK, Fried M. Inflammatory bowel disease activity measured by positron-emission tomography. Lancet 1997; 26:350
12.
O'Donovan AN, Somers S, Farrow R, Mernagh JR, Sridhar S. MR imaging of anorectal Crohn disease: a pictorial assay. RadioGraphics 1997; 17:101-107
13.
Laniado M, Makowiec F, Dammann F, Jehle EG, Glaissen CD. Perianal complications of Crohn disease: MR imaging findings. Eur Radiol 1997; 7:1035-1042
14.
Madsen SM, Thomsen HS, Munkholm P, Schlichting P, Davidsen B. Magnetic resonance imaging of Crohn disease: early recognition of treatment response and relapse. Abdom Imaging 1997; 22:164-166
15.
Hansmann HJ, Kosa R, Dux M, et al. The hydro-MRT of chronic inflammatory bowel diseases [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1997; 167:132-138
16.
Aschoff AJ, Zeitler H, Merkle EM, Reinshagen M, Brambs HJ, Rieber A. MR enteroclysis for nuclear spin tomographic diagnosis of inflammatory bowel diseases with contrast enhancement [in German]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1997; 167:387-391
17.
Lee JKT, Marcos HB, Semelka RC. MR imaging of the small bowel using the HASTE sequence. AJR 1998; 170:1457-1463
18.
Low RN, Francis IR. MR imaging of the gastrointestinal tract with IV gadolinium and diluted barium oral contrast media compared with unenhanced MR imaging and CT. AJR 1997; 169:1051-1059
19.
Ernst O, Asselah T, Cablan X, Sergent G. Breathhold fast spin-echo MR imaging of Crohn's disease. AJR 1998; 170:127-128
20.
Low RN, Francis IR, Politoske D, Benett M. Crohn's disease evaluation: comparison of contrast-enhanced MR imaging and single-phase helical CT scanning. J Magn Reson Imaging 2000; 11:127-135
21.
Marcos HB, Semelka RC. Evaluation of Crohn's disease using half-fourier RARE and gadolinium enhanced SGE sequences: initial results. Magn Reson Imaging 2000; 18:263-268
22.
Debatin JF, Patak MA. MRI of small and large bowel. Eur Radiol 1999; 9:1523-1524
23.
Limberg B, Osswald B. Diagnosis and differential diagnosis of ulcerative colitis and Crohn's disease by hydrocolonic sonography. Am J Gastroenterol 1994; 89:1051-1057
24.
Schwerk WB, Beck K, Raith M, et al. A prospective evaluation of high resolution sonography in the differential diagnosis of inflammatory bowel disease. Eur J Gastroenterol Hepatol 1992; 4:173-182
25.
Mirowitz SA. Contrast enhancement of the gastrointestinal tract on MR images using intravenous gadolinium-DTPA. Abdom Imaging 1993; 18:215-219
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: September 25, 2000
Accepted: July 3, 2001
First published: November 23, 2012
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.