Diagnostic Imaging Europe
March 1997

Gastrointestinal Imaging:
CT and MRI assume role in bowel disease

Imaging simplifies differentiation of Crohn’s disease from ulcerative colitis

By Hans-Martin Klein, M.D.

The diagnosis of Crohn’s disease (CD) and ulcerative colitis (UC) remains problematic.1 The differential diagnosis of inflammatory bowel disease (IBD) cannot always be achieved by endoscopy and conventional imaging techniques, and the chronic course of the disease necessitates an efficient and careful diagnostic rationale that causes as little discomfort to the patient as possible. In this respect, the role of CT and MRI is developing, leading to significant changes in gastrointestinal imaging.

At present, the primary diagnostic tool for the assessment of stomach and colon pathology is endoscopy. It enables not only investigation of the mucosa but also bioptic studies. The latter are mandatory for the definite diagnosis of CD by demonstration of granulomatous inflammation and for the detection of malignant metaplasia in UC. Furthermore, minimally invasive procedures, such as dilatation of stenoses, therapy of intestinal bleeding, or resection of small polyps, can be carried out under endoscopic guidance.

The mainstay of conventional radiology/fluoroscopy comprises a double-contrast examination of the stomach and a colon enema. The results of fluoroscopy and endoscopy are equivalent in 95% of cases.2 Indications for fluoroscopy are seen more rarely today, however, and are restricted to regions that are difficult to assess endoscopically.

The situation is different for the small bowel because the procedure is impossible with endoscopy. Small-bowel imaging will thus continue to be the domain of fluoroscopy for enteroclysis and follow-through. Nevertheless, fluoroscopy is not the final solution for small-bowel imaging. A small-bowel follow-through does not provide adequate distention of the lumen, which is essential for detailed investigation of the bowel wall. Enteroclysis has a better diagnostic performance, but it is invasive, uncomfortable for the patient and the doctor, and involves considerable x-ray exposure for the patient, particularly if the probe positioning is difficult.3

CT and MRI provide only limited visualization of the mucosa, but they can demonstrate the thickness and texture of the bowel wall, as well as extramural structures like vessels, mesentery, and parenchymal abdominal organs.4-6

Usefulness of CT

In recent years a marked improvement in image quality has been achieved with the implementation of new CT scanners, particularly those using powerful computer techniques. The maximum spatial resolution of contemporary scanners is now about 10 line pairs/cm for low-contrast objects. The acquisition time of one second or less has resulted in reduced movement artifacts.

In 1989 spiral CT (also known as helical or volumetric CT) was developed and established by Kalender et al.7 This new acquisition method has transformed many areas of CT diagnostics. The advantage of the method is evident: A complete abdominal status can be obtained in a few seconds. An abdominal examination during a single breath-hold is possible, which avoids artifacts caused by respiratory movement. An intravenous contrast injection can be timed exactly, so that a dedicated enhancement of vessels, gut, and parenchymal abdominal organs is possible (Figure 1).8 A reduction in x-ray exposure at the same spatial resolution is possible, as well as more efficient evaluation of the raw data.

The most recent development is the integration of powerful postprocessing options; two- and three-dimensional reformatting programs can compensate for the axial slice orientation that is a drawback of CT (Figure 2). Until now, experience with image processing in gastrointestinal imaging has been limited. The postprocessing procedure, known as virtual endoscopy,9 involves using a ray-tracing algorithm to render the inner view of the bowel wall. Previous studies in virtual colonoscopy were restricted to visualization of a “pseudo-surface” (Figure 3).

Hara et al proposed a new variant of 2-D reformatting, which they called CT colography. They obtained a spatial trajectory from a virtual endoscopy procedure and reformatted the axial images accordingly. The resulting images “enroll” the bowel, produce a useful anatomical orientation, and give a detailed demonstration of the bowel.10

In addition, the technique of intraluminal contrast application has been improved. Hyperdense media, containing barium or iodine, were designed to simply tag the bowel. This made sense because CT scanners were not able to provide detailed visualization of the gastrointestinal tract because of their limited spatial, contrast, and time resolution.

Distinguishing bowel, on the one hand, and lymph nodes, vessels, and parenchymal organs or masses on the other hand, was assisted by hyperdense contrast agents. Assessment of the bowel wall structures was complicated, however, by partial volume effects caused by the high intraluminal density. New methods for hypodense filling of the bowel lumen were therefore developed.

To provide a hypodense gastrointestinal contrast in the stomach and colon, water or tea are economical and easy to use.11 The patient should drink one liter immediately before a CT examination of the stomach and upper small bowel. The application of a hypoperistaltic agent like glucagon is important to avoid gastric emptying. To make the medium more potable, it is acceptable to add a flavoring.

To achieve isotonicity, 10 teaspoons of sugar and a teaspoon of salt can be added to a liter of water. This also helps to delay fluid resorption. In 1992 Zwaan et al proposed adding paraffin oil, methyl cellulosis, and an emulsifier to the contrast solution. The results were encouraging, but not optimal.12 To examine the large bowel, rectal administration of 1.5 liters of hand-warm isotonic water solution results in a well-filled colon frame.

Positioning of the patient depends on the intestinal region of interest. To obtain adequate distention and filling of the gut, consideration must be given to the part of the body being examined. For example, the gastric antrum has to be examined in the left anterior oblique position.

Distention is an important factor for evaluation of the intestines. While good distention is relatively easy to achieve in the stomach and colon, optimal distention of the small bowel can be produced only after duodenal intubation.

There are some reports in the literature on the application of duodenal intubation for gastrointestinal CT. Thiele uses a barium dispersion, together with methyl-cellulosis, as in conventional enteroclysis.13 Rollandi et al performed a small- bowel enema with methyl-cellulose through a nasojejunal probe, which resulted in an excellent mucosal delineation after additional administration of an intravenous contrast agent.14 The insertion of a duodenal probe is uncomfortable for the patient, however, and the most obvious advantage of cross-sectional imaging over fluoroscopy—the lack of invasiveness—is lost.

Impact On Patient Care

What do these changes in technique mean in terms of improvements in clinical imaging? The main implication is that the correct use of CT enables differentiation of CD from UC in 94% of cases.15 Except in very early cases of CD, wall thickening is seen (Figure 4).16-18 It is mostly homogeneous in CD, but a hypodense inner ring may be seen in the early phase, due to inflammatory mucosal edema. Wall thickness may be extensive in CD; it measured up to 25 mm in our study (Figure 5).19 The wall contours are an important criterion for the differential diagnosis. In CD, the outer contour is mostly irregular, particularly in the later phase of the disease, due to scars and fibrous retraction (Figure 6).

In UC, the bowel wall is thickened, but the thickening is less extreme than in CD (Figure 7). The wall may show a layering with a hyperdense inner ring (mucosa), a hypodense intermediate ring (submucosal fatty deposits), and a hyperdense outer ring (muscular and adventitial layer). The morphology can vary because there may be a strong thickening of the muscularis mucosae instead of a lipomatous submucosal ring. In UC, the inner contour shows ulcerative changes and is coarsely irregular, while the outer contour is smooth, even in advanced stages of the disease (Figure 8).

Bowel wall enhancement after intravenous contrast administration is an important parameter because it indicates local changes in vessel wall permeability, caused by inflammatory, neoplastic, or ischemic damage. Furthermore, it improves the delineation from extraintestinal structures. Use of a hypodense luminal contrast agent is advantageous for evaluation of bowel wall enhancement characteristics. The detection and classification of fistulas and abscesses depend on sufficient bowel dilation and luminal filling with a hyperdense contrast agent. The decision to use conventional iodine hyperdense agents or hypodense substances like oil or water depends on the individual clinical situation.20

Finally, the course of IBD can be controlled unobtrusively and reliably with CT. The detection of abscesses or fistulas, which are present in about 40% of patients, is possible with CT. CT should be the method of second choice, after ultrasound, for diagnosis of IBD in children, according to Jabra et al.21

MRI’s Scope

MRI has some principal advantages over CT because of its multiplanar imaging capabilities and the higher soft-tissue contrast.20,22 As with CT, new generations of scanners offer increased imaging speed, and thus fewer motion artifacts, but the chances of improving spatial and contrast resolution are limited. With MRI, it is possible to obtain a series of images in about 16 seconds and to repeat this sequence up to five times with an optional delay. This facilitates a dynamic MRI of the ileocecal region and large parts of the ileum.

There are some disadvantages of gastrointestinal MRI: Peristaltic and respiratory movements produce motion artifacts, the pulsation of the aorta causes ghosting artifacts, and chemical shift artifacts lead to a deterioration of the borderline between mesenteric fat and the bowel wall. Furthermore, the spatial resolution of MRI is low compared with CT, so the normal bowel wall, which is thinner than 3 mm, can be difficult to delineate from surrounding structures.

The considerations of hyper- or hypointense intestinal contrast are analogous to the considerations with CT. The first attempts to obtain an intraluminal contrast used a gadolinium-containing solution to produce a hyperintense intraluminal signal.23 Koelbel et al analyzed MRI for detection of fistulas caused by CD and found a sensitivity of more than 90% among 17 patients.24 Hypodense agents containing oral magnetic particles (OMP) have become available (Abdoscan from Nycomed and Lumirem from Guerbet). Other agents include kaopectin,25 or even blueberry juice, which is hypodense due to its manganese content.

In our own prospective study on 20 patients with CD, we found a good tolerability of OMP given orally and rectally (Figure 9). The imaging results for determining the location of the disease were comparable to enteroclysis. The clinical assessment was improved in most cases.

To conclude, endoscopy is the diagnostic method of first choice for the upper gastrointestinal tract (esophagus and stomach) and the large intestine, while CT, MRI, and fluoroscopy are complementary methods. For the small bowel, enteroclysis is the best way to image the mucosal surface, and for early inflammatory disease, it is still important, but is rivaled by CT and MRI.

CT and MRI are becoming more important and enteroclysis should be properly indicated for follow-up examinations in IBD, particularly when there are complications. In the development of an optimal regimen for gastrointestinal cross-sectional imaging—including patient preparation, intraluminal and intravenous contrast administration, and imaging parameters—CT and MRI are worth more investigative work.

Dr. Klein is an associate professor of diagnostic radiology at RWTH Aachen, Aachen, Germany. Assisting in the preparation of this manuscript were Andreas Schmitz, Gerhard Adam, Christian Hohl, Bernd Klosterhalfen, and Rolf W. Gunther.

References

  1. Crohn BB, Ginzburg L, Oppenheimer GD. Regional ileitis: a pathological and clinical entity. JAMA 1932;99:1323-1329.
  2. Reeders JWAJ, Rosenbusch G. Clinical radiology and endoscopy of the colon. Stuttgart: Thieme, 1994;112.
  3. Klein HM, Gunther RW. Double contrast small bowel follow-through with a new acid resistant effervescent agent. Invest Radiol 1993;28:581-585.
  4. Goldberg HI, Gore RM, Margulis AR. Computed tomography in the evaluation of Crohn’s disease. Am J Roentgenol 1983;140:277-283.
  5. Goldberg HI, Thoeni RF. MRI of the gastrointestinal tract. Radiol Clin N Am 1989;27:805-812.
  6. Gore RM, Marn CS, Kirby DF. CT findings in ulcerative, granulomatous and indeterminate colitis. Am J Roentgenol 1984;27:279-284.
  7. Kalender WA, Vock P, Polacin A, Soucek M. Spiral-CT: a new technique for volumetric scans. 1. Basic principles and methodology. Rõntgenpraxis 1990;43:323-330.
  8. Meyers MA, McGuire PV. Spiral CT demonstration of hypervascularity in Crohn’s disease: “vascular jejunization of the ileum” or the “comb sign.” Abdom Imaging 1995;20:327-332.
  9. Vining DJ, Gelfand DW, Bechtold RE, et al. Technical feasibility of colon imaging with helical CT and virtual reality. Am J Roentgenol 1994;162:104(abstract).
  10. Hara AK, Johnson CD, Reed JE, et al. Colorectal polyp detection with CT colography: two- versus three-dimensional techniques. Radiology 1996;250:49-54.
  11. Gazelle GS, Gaa J, Saini S, Shellito P. Staging of colon carcinoma using water enema CT. JCAT 1995;19:87-91.
  12. Zwaan M, Gmelin E, Borgis KJ, Neubauer B. Normal wall thickness and tumorous changes in the gastrointestinal tract demonstrated by CT. Fortschr Rõntgenstr 1991;155:423-427.
  13. Thiele J, Kloeppel R, Schulz HG. CT-Sellink—a new method for evaluating the intestinal wall. Fortschr Rõntgenstr 1993;159:213-217.
  14. Rollandi GA, Curome PF, Bertolotto M, et al. Crohn’s disease: evaluation with spiral CT and transparent bowel enema. Radiology 1995;197(P):314.
  15. Fasse J, Klose KC, Raguse T. An evaluation of preoperative computed tomography for chronic inflammatory bowel disease. Colo-proctology 1987;4:241-246.
  16. Düx M, Richter GM, Roeren T, et al. Gastrointestinal imaging with hydrosonography and hydro-CT. Colorectal carcinoma. Fortschr Rõntgenstr 1996;164:359-367.
  17. Frager DH, Goldman M, Beneventano TC. Computed tomography in Crohn’s disease. JCAT 1983;7:819-824.
  18. Philpotts LE, Heiken JP, Westcott MA, Gore RM. Colitis: use of CT findings in differential diagnosis. Radiology 1994;190:445-449.
  19. Klein HM, Wein B, Adam G, et al. Computed tomography of Crohn’s disease and ulcerative colitis. Fortschr Rõntgenstr 1995;163:9-15.
  20. Li KCP, Ho-Tai PCK. The search for the ideal enteric MR contrast agent. Diag Imaging 1990;3:110-114.
  21. Jabra AA, Fishman EK, Taylor GA. Crohn’s disease in the pediatric patient: CT evaluation. Radiology 1991;179:495-498.
  22. Werthmüller WC, Margulis AR. Magnetic resonance imaging of the alimentary tube. Invest Radiol 1991;26:195-200.
  23. Laniado M, Kornmesser W, Hamm B, et al. MR imaging of the gastrointestinal tract: value of Gd-DTPA. Am J Roentgenol 1987;150:817-821.
  24. Koelbel G, Schmiedl U, Majer MC, et al. Diagnosis of fistulae and sinus tracts in patients with Crohn’s disease: value of MR imaging. Am J Roentgenol 1989;152:999-1003.
  25. Neuerburg J, Klose KC, Bohndorf K, et al. MR tomography of colorectal disease with a negative rectal contrast medium. Fortschr Rõntgenstr 1992;156:258-263.