Prospective evaluation of prehospital trauma ultrasound during aeromedical transport.

BACKGROUND Ultrasound is widely considered the initial diagnostic imaging modality for trauma. Preliminary studies have explored the use of trauma ultrasound in the prehospital setting, but the accuracy and potential utility is not well understood. OBJECTIVE We sought to determine the accuracy of trauma ultrasound performed by helicopter emergency medical service (HEMS) providers. METHODS Trauma ultrasound was performed in flight on adult patients during a 7-month period. Accuracy of the abdominal, cardiac, and lung components was determined by comparison to the presence of injury, primarily determined by computed tomography, and to required interventions. RESULTS HEMS providers performed ultrasound on 293 patients during a 7-month period, completing 211 full extended Focused Assessment with Sonography for Trauma (EFAST) studies. HEMS providers interpreted 11% of studies as indeterminate. Sensitivity and specificity for hemoperitoneum was 46% (95% confidence interval [CI] 27.1%-94.1%) and 94.1% (95% CI 89.2%-97%), and for laparotomy 64.7% (95% CI 38.6%-84.7%) and 94% (95% CI 89.2%-96.8%), respectively. Sensitivity and specificity for pneumothorax were 18.7% (95% CI 8.9%-33.9%) and 99.5% (95% CI 98.2%-99.9%), and for thoracostomy were 50% (95% CI 22.3%-58.7%) and 99.8% (98.6%-100%), respectively. The positive likelihood ratio for laparotomy was 10.7 (95% CI 5.5-21) and for thoracostomy 235 (95% CI 31-1758), and the negative likelihood ratios were 0.4 (95% CI 0.2-0.7) and 0.5 (95% CI 0.3-0.8), respectively. Of 240 cardiac studies, there was one false-positive and three false-negative interpretations (none requiring intervention). CONCLUSIONS HEMS providers performed EFAST with moderate accuracy. Specificity was high and positive interpretations raised the probability of injury requiring intervention. Negative interpretations were predictive, but sensitivity was not sufficient for ruling out injury.