Sonographic scoring for operating room triage in trauma

Abstract

Study objectives: Focused assessment with sonography for trauma (FAST) provides a timely noninvasive method for patient assessment in blunt trauma. We examined which physiologic criteria combined with the FAST examination can be used to more accurately predict which patients require laparotomy than using the FAST examination alone. Methods: This was an observational, prospective study of emergency department (ED) patients who underwent FAST examination during their initial trauma evaluation. The study was performed at an urban, regional Level I trauma center. FAST examinations were performed by an emergency medicine resident or attending physician certified in bedside ultrasonography. FAST examinations were scored as described by Huang, with values ranging from 0 to 8, depending on the amount and location of intraperitoneal blood. Physiologic criteria recorded include out-of-hospital vital signs, vital signs at FAST examination, arterial blood gas, hemoglobin, age, Glasgow Coma Scale score, and whether a pelvis fracture was present. Records of all included patients who underwent laparotomy in the first 72 hours after admission were reviewed by a trauma surgeon to determine the necessity of the laparotomy according to the degree of injury to the spleen, liver, intraperitoneal free blood, hollow viscous organs, great vessel, diaphragm, pancreas, or bladder rupture. A statistician performed area under the curve analysis using 1-, 2-, and 3-variable models for all possible combinations of physiologic data and ultrasonographic score. Results: There were 1,113 patients enrolled in the study. Nine hundred seventy-three patients had blunt trauma, and 140 had penetrating injuries. Of the 973 blunt trauma patients, 48 patients went to the operating room for laparotomy. The average ultrasonographic score for laparotomy cases was 2.5. For nonlaparotomy cases, the average score was 0.1. A 3-variable model consisting of ED systolic blood pressure, ED respiratory rate, and ultrasonographic score had a value of 0.88970 for area under the curve. Conclusion: A model combining ultrasonographic score and immediately available physiologic criteria can predict the need for laparotomy in blunt trauma. If validated prospectively, the rule may allow emergency and trauma surgeons to expedite laparotomy for those who need it.