An international evaluation of ultrasound vs. computed tomography in the diagnosis of appendicitis

BACKGROUND: Abdominal computed tomography scan (CT) is the preferred radiographic study for the diagnosis of appendicitis in the United States, while radiologist-operated ultrasound (US) is often used in Israel. This comparative international study evaluates the performance of CT vs. US in the evaluation of acute appendicitis. METHODS: A retrospective chart analysis was conducted at two tertiary care teaching hospitals, one in each country. Adult patients (age 18-99) with an Emergency Department (ED) working diagnosis of appendicitis between 1 January 2005 and 31 December 2006 were reviewed. Included patients had at least one imaging study, went to the OR, and had documented surgical pathology results. RESULTS: Of 136 patients in the United States with the ED diagnosis of appendicitis, 79 met inclusion criteria for the CT cohort. Based on pathology, CT had a sensitivity of 100% (95% CI 95.4-100%). The negative appendectomy rate in patients with positive CT was 0%. Total median ED length of stay was 533 min [IQR (450-632)] and median time from CT order to completion was 184 min [IQR (147-228)]. Of 520 patients in Israel, 197 were included in the US cohort. Based on final pathology, US had a sensitivity of 68.4% (95% CI 61.2-74.8%). The negative appendectomy rate in patients with positive US was 5.5%. The median ED length of stay for these patients was 387 min [IQR (259-571.5)]. Of the patients, 23.4% had subsequent CT scans. Median time from US order to completion was 20 min [IQR (7-49)]. Both time values were p < 0.001 when compared with CT. We furthermore calculate that a "first pass" approach of using US first, and then performing a confirmatory CT scan in patients with negative US, would have saved an average of 88.0 minutes per patient in the United States and avoided CT in 65% of patients. CONCLUSIONS: Radiologist-operated US had inferior sensitivity and positive predictive value when compared with CT, though was significantly faster to perform, and avoided radiation and contrast in a majority of patients. A "first-pass" approach using US first and then CT if US is not diagnostic may be desirable in some institutions.