Validation of surgeon-performed emergency abdominal ultrasonography in pediatric trauma patients.

Background/purpose: The focused assessment for the sonographic evaluation of trauma patients (FAST) in adults is effective in detecting intraperitoneal and intrapericardial fluid and can be performed quickly by surgeons in the emergency department (ED). The authors sought to validate the accuracy of FAST performed by surgeons during ED resuscitation of pediatric trauma patients.

Methods: Patients were assigned to one of three groups based on standard clinical criteria: immediate surgery, abdominal computed tomography (CT), or observation alone. FAST was then performed in the ED by a surgery resident (postgraduate year 3 or higher) or an attending trauma surgeon. Four views were used to assess the possible presence of fluid in the pericardial, subphrenic, subhepatic, and pelvic spaces. Time needed to conduct FAST was noted. Presence of peritoneal or pericardial fluid by FAST was compared with that determined by CT or surgery. Sensitivity, specificity, and predictive values were calculated. For those who did not undergo CT or surgery, FAST findings were compared with the clinical course.

Results: Technically adequate studies could be performed on 192 of 196 eligible children. Their ages ranged from 3 months to 14 years (mean, 6.9 years); 119 were boys (62%), and 188 (98%) had sustained a blunt injury. FAST was performed in a mean time of 3.9 minutes (range, 1-17 minutes). All FAST examinations were reviewed by our senior surgeon-sonographer (GSR). Interrater agreement between the performing and reviewing surgeon-sonographer was 100%. Sixty (31%) patients underwent either abdominal CT (n = 56; mean Injury Severity Score (ISS), 9.6) or immediate operation (n = 4; mean ISS, 18.8). Of the 10 patients with verified presence of intraperitoneal fluid, eight had positive and two had false-negative FAST examination results. Of the 50 patients with verified absence of intraperitoneal fluid, none had a positive FAST (ie, no false-positives); sensitivity was 80%; specificity, 100%; predictive value positive, 100%; predictive value negative, 96%. None of the 132 patients followed up clinically without CT or surgery (mean ISS, 4.5) had fluid documented by FAST, and all did well.

Conclusions: The focused assessment for the sonographic evaluation of pediatric blunt trauma patients performed by surgical residents and attendings in the ED rapidly and accurately predicted the presence or absence of intraperitoneal fluid. The FAST is a potentially valuable tool to rapidly prioritize the need for laparotomy in the child with multiple injuries and extraabdominal sources of bleeding.