Abdominal ultrasound is an unreliable modality for the detection of hemoperitoneum in patients with pelvic fracture

Background: Detection of hemoperitoneum in patients with pelvic fracture and hemodynamic instability is important to determine the need for laparotomy versus pelvic angiography. The use of ultrasound (FAST [Focused Assessment with Sonography for Trauma]) for the evaluation of hemoperitoneum after blunt abdominal trauma has become widespread. However, its sensitivity and specificity in patients with pelvic fracture remain poorly defined. The purpose of this study was to determine the sensitivity and specificity of FAST for the detection of hemoperitoneum in patients with pelvic fracture and an increased risk for hemorrhage. Methods: The medical records for all admissions to our Level I trauma center from November 2003 to February 2005 were retrospectively reviewed. Inclusion criteria were presence of pelvic fracture with at least one of the following risk factors for hemorrhage: age <= 55, hemorrhagic shock (systolic blood pressure < 100 mm Hg), or unstable fracture pattern. Emergency department FAST results were recorded. Surgery residents trained and certified in ultrasonography in the acute setting performed all FAST examinations and an in house attending surgeon reviewed them. Presence of hemoperitoneum was confirmed by laparotomy or abdominopelvic computed tomography (CT) scan. Results: There were 146 patients who met entry criteria, 126 of who had a FAST examination performed. A total of 104 patients underwent a confirmatory evaluation of their abdomen with either operative exploration (n = 20) or CT scan (n = 84). Eight patients underwent diagnostic peritoneal lavage before CT confirmation and were excluded. Ninety-six patients constituted the study group. Nineteen patients presented in hemorrhagic shock. There were 11 true-positive, 52 true-negative, 2 false-positive, and 31 false-negative results. Sensitivity and specificity were 26% and 96%, respectively. Positive and negative predictive values were 85% and 63%, respectively. Conclusions: A FAST examination with negative result does not aid in determining the need for laparotomy versus pelvic angiography in patients with pelvic fracture at risk for hemorrhage. These patients should undergo additional confirmatory evaluation to exclude intraperitoneal hemorrhage.