BACKGROUND: To assess the value of high resolution sonography (HRS) in identifying normal and inflammatory bowel wall in nondiverticular ileitis and colitis by using a segment-by-segment analysis. METHODS: Thirty-five HRS were performed in patients with nondiverticular inflammatory bowel disease, without knowledge of clinical, endoscopic, and radiologic data. HRS evaluated separately five intestinal segments (terminal ileum, cecum/ascending colon, transverse, descending colon, and sigmoid colon) and was considered positive for inflammation when wall thickness during compression exceeded 3 mm. We compared HRS findings with results of endoscopy or enteroclysis performed within 8 days of HRS; endoscopic and radiologic results were classified into two subgroups: mild inflammatory lesions and frank inflammatory lesions. RESULTS: Segment-by-segment analysis resulted in an accuracy of 81%, a sensitivity of 70%, and a specificity of 93%. Sensitivity was significantly lower for mild lesions (52%) than for frank lesions (87%, p < .001). Of the 32 patients having an inflammatory bowel condition, 29 (91%) had at least one segment correctly identified as inflammatory by HRS. CONCLUSION: Even if relatively insensitive for minor lesions, HRS is a promising, minimally invasive method for assessing normal and inflammatory bowel wall in nondiverticular ileitis and colitis.
BACKGROUND: To assess the value of high resolution sonography (HRS) in identifying normal and inflammatory bowel wall in nondiverticular ileitis and colitis by using a segment-by-segment analysis. METHODS: Thirty-five HRS were performed in patients with nondiverticular inflammatory bowel disease, without knowledge of clinical, endoscopic, and radiologic data. HRS evaluated separately five intestinal segments (terminal ileum, cecum/ascending colon, transverse, descending colon, and sigmoid colon) and was considered positive for inflammation when wall thickness during compression exceeded 3 mm. We compared HRS findings with results of endoscopy or enteroclysis performed within 8 days of HRS; endoscopic and radiologic results were classified into two subgroups: mild inflammatory lesions and frank inflammatory lesions. RESULTS: Segment-by-segment analysis resulted in an accuracy of 81%, a sensitivity of 70%, and a specificity of 93%. Sensitivity was significantly lower for mild lesions (52%) than for frank lesions (87%, p < .001). Of the 32 patients having an inflammatory bowel condition, 29 (91%) had at least one segment correctly identified as inflammatory by HRS. CONCLUSION: Even if relatively insensitive for minor lesions, HRS is a promising, minimally invasive method for assessing normal and inflammatory bowel wall in nondiverticular ileitis and colitis.