The fast is positive, now what? Derivation of a clinical decision rule to determine the need for therapeutic laparotomy in adults with blunt torso trauma and a positive trauma ultrasound.

INTRODUCTION: The object of this study was to derive a clinical decision rule for therapeutic laparotomy among adult blunt trauma patients with a positive abdominal ultrasound for trauma (FAST) examination.

METHODS: We retrospectively reviewed the trauma registry and medical records of all critical trauma patients who underwent a FAST examination in the emergency department (ED) in a university Level I trauma center over a 3-year period. Blunt trauma patients aged >16 years who had a positive FAST examination (defined as the presence of intraperitoneal fluid) were eligible. We selected seven clinical and ultrasound variables available during ED resuscitation for analysis: age, presence of an episode of hypotension (systolic blood pressure <90 torr in the ED), presence of abdominal tenderness, chest injury, pelvic fracture, femur fracture, and FAST fluid location (right upper quadrant [RUQ] only; RUQ plus other location; other location only). The primary outcome variable was whether a laparotomy was performed and whether this laparotomy was needed to provide the definitive surgical intervention ("therapeutic laparotomy"). We analyzed the variables using binary recursive partitioning analysis to create a decision rule.

RESULTS: There were 2336 FAST examinations performed during the study period, resulting in 230 (9.8%) positive examinations in patients meeting inclusion criteria. There were 135 patients who had therapeutic laparotomies and 95 who did not need laparotomy. The groups were similar in baseline characteristics. In the recursive partitioning analysis, the first node in the decision tree was the presence of fluid in the RUQ. Of the 144 patients with RUQ fluid, 105 (73%, 95% confidence interval [CI] 64%-80%) required therapeutic laparotomy. Of the 86 patients without RUQ fluid, 30 (35%, 95% CI 25%-46%) nevertheless required therapeutic laparotomies, and the variables blood pressure, femur fracture, abdominal tenderness, and age further divided these patient into high- and low-risk groups. Of the 12 patients without RUQ fluid who had normal blood pressures, no femur fractures, no abdominal tenderness, and were aged 60 years and younger, none (95% CI 0%-22%) required therapeutic laparotomy. In conclusion, given a positive FAST examination, the presence of fluid in the RUQ is an important predictor of the need for therapeutic laparotomy.

CONCLUSION: In the absence of fluid in the RUQ, there are other clinical variables that may allow for the development of a clinical decision rule regarding the need for therapeutic laparotomy.