This study assessed the ability of a six-point trauma ultrasound (US) evaluation (a) to identify the presence of free fluid in the abdomen or pelvis, with computed tomography (CT) and laparotomy used as diagnostic standards and (b) to predict the presence of abdominal or pelvic injury, particularly injury requiring surgical intervention. Of 156 patients who underwent US evaluation for free fluid after sustaining blunt and penetrating trauma, 147 were entered into the prospective study and underwent follow-up CT and/or laparotomy (n = 79), in-hospital observation, or outpatient examination. The sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of US for identifying abdominal or pelvic free fluid were 69%, 100%, 100%, 95%, and 95%, respectively. The corresponding values for predicting abdominal and pelvic injury on the basis of free fluid status alone were 57%, 99%, 80%, 96%, and 95%, respectively. Performing repeated US examinations in patients with deteriorating clinical status decreased the false-negative rate by 50%, increasing the sensitivity for free fluid detection to 85% and the negative predictive value to 97%. Similarly, the sensitivity and negative predictive value for detection of injury increased to 71% and 97%, respectively. A learning curve was also observed, with 67% of the false-negative findings occurring in the first 3 months of the 19-month study. A six-point trauma US evaluation can reliably identify abdominal and pelvic free fluid, which can be a reliable indicator of abdominal or pelvic injury. Scanning conditions must be optimized, and the approach to clinical management must be cautious.