To determine the utility of multidetector computed tomography (MDCT) in patients with clinically evident acute appendicitis and to compare the test characteristics of overall clinical impression, Alvarado scores, and MDCT in suspected appendicitis. A prospective observational cohort study was conducted in two urban emergency departments (ED). Consecutive patients with suspected acute appendicitis were clinically evaluated by an emergency physician who was asked to determine whether appendicitis was clinically evident or not. Elements of the Alvarado scores were collected and all patients then underwent MDCT and a decision to operate, observe, or discharge the patients was made by a surgeon. The final diagnosis was based on surgical pathology or clinical follow-up. The test characteristics of clinical impression, Alvarado scores and MDCT were then calculated and the rates at which acute appendicitis was falsely diagnosed based on clinical impression and MDCT were compared using McNemar's test. Of 157 study patients, 71 were considered to have clinically evident appendicitis before MDCT and 91 had findings of acute appendicitis on MDCT. 19 of the 71 patients with clinically evident appendicitis did not have appendicitis. 14 of 52 patients with an Alvarado score > or = 8 also did not have appendicitis. Three of 91 patients with acute appendicitis based on MDCT did not have appendicitis. The specificities of clinical impression and Alvarado score > or = 8 were 71.6% and 79.1%, respectively, and these were significantly lower than that of MDCT (95.5%, p<0.05). The performance of abdominal MDCT in patients with a high degree of clinical suspicion for acute appendicitis reduces the number of false positives and has the potential to reduce negative appendectomies.