Small intestine contrast ultrasonography vs computed tomography enteroclysis for assessing ileal Crohn's disease.

AIM To compare computed tomography enteroclysis (CTE) vs small intestine contrast ultrasonography (SICUS) for assessing small bowel lesions in Crohn's disease (CD), when using surgical pathology as gold standard. METHODS From January 2007 to July 2008, 15 eligible patients undergoing elective resection of the distal ileum and coecum (or right colon) were prospectively enrolled. All patients were under follow-up. The study population included 6 males and 9 females, with a median age of 44 years (range: 18-80 years). INCLUSION CRITERIA (1) certain diagnosis of small bowel requiring elective ileo-colonic resection; (2) age between 18-80 years; (3) elective surgery in our Surgical Unit; and (4) written informed consent. SICUS and CTE were performed ≤ 3 mo before surgery, followed by surgical pathology. The following small bowel lesions were blindly reported by one sonologist, radiologist, surgeon and histolopathologist: disease site, extent, strictures, abscesses, fistulae, small bowel dilation. Comparison between findings at SICUS, CTE, surgical specimens and histological examination was made by assessing the specificity, sensitivity and accuracy of each technique, when using surgical findings as gold standard. RESULTS Among the 15 patients enrolled, CTE was not feasible in 2 patients, due to urgent surgery in one patients and to low compliance in the second patient, refusing to perform CTE due to the discomfort related to the naso-jejunal tube. The analysis for comparing CTE vs SICUS findings was therefore performed in 13 out of the 15 CD patients enrolled. Differently from CTE, SICUS was feasible in all the 15 patients enrolled. No complications were observed when using SICUS or CTE. Surgical pathology findings in the tested population included: small bowel stricture in 13 patients, small bowel dilation above ileal stricture in 10 patients, abdominal abscesses in 2 patients, enteric fistulae in 5 patients, lymphnodes enlargement (> 1 cm) in 7 patients and mesenteric enlargement in 9 patients. In order to compare findings by using SICUS, CTE, histology and surgery, characteristics of the small bowel lesions observed in CD each patient were blindly reported in the same form by one gastroenterologist-sonologist, radiologist, surgeon and anatomopathologist. At surgery, lesions related to CD were detected in the distal ileum in all 13 patients, also visualized by both SICUS and CTE in all 13 patients. Ileal lesions > 10 cm length were detected at surgery in all the 13 CD patients, confirmed by SICUS and CTE in the same 12 out of the 13 patients. When using surgical findings as a gold standard, SICUS and CTE showed the exactly same sensitivity, specificity and accuracy for detecting the presence of small bowel fistulae (accuracy 77% for both) and abscesses (accuracy 85% for both). In the tested CD population, SICUS and CTE were also quite comparable in terms of accuracy for detecting the presence of small bowel strictures (92% vs 100%), small bowel fistulae (77% for both) and small bowel dilation (85% vs 82%). CONCLUSION In our study population, CTE and the non-invasive and radiation-free SICUS showed a comparable high accuracy for assessing small bowel lesions in CD.