Ultrasonography by emergency medicine and radiology residents for the diagnosis of small bowel obstruction

Abstract

Our objective was to study the accuracy of emergency medicine [(EM) bedside ultrasonography (BUS)] and radiology residents performed ultrasonography (RUS) in patients with suspected mechanical small bowel obstruction (SBO). After a 6-h training program, from January to June 2009, four EM residents used BUS to prospectively evaluate the patients presenting to the emergency department with suspected SBO. Then, patients underwent RUS. Outcome was determined by surgical findings if they were operated upon or self-reported the condition upon telephone follow-up at 1-month. BUS and RUS results were compared with chi2 testing. Of the 174 enrolled patients, 90 patients were BUS-positive. Of these, surgical findings agreed with the BUS findings in 84 patients. In 78 cases, BUS was negative, and 76 of these patients had benign clinical courses. Six patients were excluded from the study. The sensitivity, specificity, positive predictive value, negative predictive value, and likelihood ratio for BUS were 97.7, 92.7, 93.3, 97.4, and 13.4%, respectively. Sensitivity, specificity, positive predictive value, and negative predictive value for RUS were 88.4, 100, 100, and 89.1%, respectively. The diagnostic accuracy of BUS and RUS were not statistically different from each other (kappa=0.81). The presence of dilated small bowel loops (>25 mm in jejunum or >15 mm in ileum) was the most sensitive (94%) and specific (94%) sonographic finding for SBO. Abdominal sonography for the diagnosis of SBO is a new application of BUS in the emergency department. EM residents can diagnose SBO using BUS with a high-degree of accuracy, comparable with that of radiology residents.