BACKGROUND
Enteroclysis, intestinal wall ultrasound (IWU) and abdominal magnetic resonance imaging (MRI) are three established methods in the diagnosis of Crohn's disease (CD). To date, however, the three methods have not been compared in one patient collective.
AIMS
The present prospective study compared the relative performance of IWU, MRI and enteroclysis in determining the extent of disease involvement and intestinal complications in patients with CD both at initial diagnosis and during follow-up.
PATIENTS AND METHODS
Included in the present study were 48 patients with confirmed CD (age: 19-66 years) examined with all three methods between August 1999 and December 2000. IWU was performed in B-mode with a 4-7 MHz convex transducer head and a 5-12 MHz linear transducer head by an experienced examiner. At MRI, T1 and T2 weighted sequenced (Flash 2D before and after intravenous application of gadolinium DTPA or TSE) were acquired in coronal and transverse planes. Enteroclysis was performed using conventional biphasic technique. Interpretation was conducted on the basis of a standardized catalogue of findings.
RESULTS
Changes in bowel segments consistent with inflammation were identified in 41 of 48 patients. All three methods returned equivalent findings with regard to the length of inflamed bowel segments (IWU, range: 3-25 cm, mean: 12 cm; MRI, range: 3-25 cm, mean: 10 cm; enteroclysis, range: 3-30 cm, mean: 11 cm) and wall thickness (IWU, range: 4-10 mm, mean: 7 mm; MRI, range: 5-10 mm, mean: 7 mm; of nine patients with stenotic change, five were correctly diagnosed with IWU (sensitivity, 55.6%; specificity, 97.4%), four with MRI (sensitivity, 44.4%; specificity, 100%) and six with enteroclysis (sensitivity, 66.7%; specificity, 100%). Fistulae were correctly identified in five patients with IWU (sensitivity, 55.6%; specificity, 97.4%), in four with MRI (sensitivity, 44.4%; specificity, 100%) and in six with enteroclysis (sensitivity, 66.7%; specificity, 100 %) of a total of nine patients with confirmed fistula formation. Abscesses were correctly identified in five patients with IWU (specificity, 66.7%; specificity, 100%), in five with MRI (sensitivity, 83.3%; specificity, 100%) and in no patients with enteroclysis (sensitivity, 0%; specificity, 100%) in six patients with abscesses.
CONCLUSION
Both IWU and MRI identify extent, severity and intestinal complications with adequate diagnostic accuracy in patients with CD. Both techniques possess the potential for replacing enteroclysis in the work-up of CD. Enteroclysis should be reserved for the work-up of complex fistula systems.