Ultrasound in Crohn’s Disease - bowel wall characteristics and perfusion estimates using microbubbles

Abstract

Crohn’s disease (CD) is a chronic inflammatory bowel disease which often presents early in life and sometimes can be debilitating. The patients need frequent diagnostic work-up to assess disease activity, location, extent and if complications have occurred. This warrants diagnostic tools which are of little nuisance to the patient, available, affordable and objective. Diagnostic ultrasound imaging could potentially fulfil these criteria. A specific problem in patients with CD is the differentiation of inflammation and fibrosis in the thickened bowel wall when an obstruction develops. Improved differentiation could lead to better tailoring of treatment. The main aim of this thesis was to examine if there were ultrasound criteria which could separate fibrosis and inflammation. In the first study 14 bowel specimens from patients operated for CD was examined with ultrasound in vitro and compared directly to histology. We found that some histological findings typical for CD which cannot be investigated in mucosal biopsies can be identified with ultrasound. A thickened muscular mucosa, echo changes in the submucosa and proper muscles are features of fibrotic disease while lymphocyte aggregates along the outer border of muscularis propria is a feature of chronic inflammation. Since the main measurement used for detecting bowel disease is bowel wall thickness and proper reference values were wanting, study two was performed. The bowel wall thickness was measured in several locations in the gastrointestinal tract in 122 healthy volunteers. The results indicated that the normal bowel wall should not exceed 2 millimetres in most of the GI tract except for the stomach, duodenum and rectum. Furthermore, the reference values can be used for ultrasound probes with a frequency >8 megahertz regardless of fasting state, age, weight sex and height. In paper three two patient groups allocated for medical treatment (n=19) or surgical treatment (n=20) through a regular clinical work up were compared to identify features separating the two groups. Contrast enhanced ultrasound was used in combination with a perfusion model to estimate blood flow in the bowel wall. We found that the surgical group had decreased blood volume and flow when compared to the medical group and markedly thicker bowel wall including the muscularis propria and mucosa. From our results we conclude that a thickened muscular mucosa and echo changes in the submucosa as well as a reduced blood volume and flow could be indications of fibrostenotic disease and thus might support a decision for surgical treatment. Work remains on verifying the perfusion results with histology and to see if these findings can be reproduced in a larger group of patients. The perfusion model also needs further validation before implementation in regular clinical work.