Efficacy of anatomic and physiologic indicators versus mechanism of injury criteria for trauma activation in pediatric emergencies.

BACKGROUND: In pediatric trauma patients, adult triage criteria that use mechanism of injury (MOI) have been shown to result in overactivation of trauma teams. Anatomy- and physiology-based (APB) triage criteria have been recommended to improve the accuracy of trauma activations. At our Level 1 academic tertiary pediatric trauma referral center, we recently changed our triage criteria by emphasizing APB criteria and de-emphasizing MOI. This study was conducted to analyze the resulting change in accuracy of activations.

METHODS: This was a criterion standard, cohort-controlled retrospective study comparing patients triaged by MOI criteria (January 2006 to March 2009) to those triaged by APB criteria (April 2009 to June 2010). Patients were subdivided according to trauma activation level as major (TMaj), minor (TMin), or consult (TC). Demographic, vital sign, injury pattern, trauma activation level, and emergency department disposition data were collected. Triage criteria were retrospectively applied to the patients according to the criteria that were in effect when they arrived. Patients were assigned to either high-risk (HR) or low-risk (LR) groups based on the need for urgent intervention (emergency department procedure, emergent operation, or blood transfusion), admission to intensive care unit, Injury Severity Score [ISS] of greater than 12, or death. Sensitivity and specificity of major activations were calculated using the following groups: true positive, trauma activation and HR; false positive, trauma activation and LR, false negative, no trauma activation and HR; true negative, no trauma activation and LR. Comparisons were then made between the MOI to the APB patients.

RESULTS: The MOI and APB patients were similar in race (p = 0.201), sex (p = 0.639), and age (p = 0.643). The APB criteria resulted in 14% TMaj, 35% TMin, and 51% TC, compared with 41%, 23%, and 36%, respectively, for MOI. Median ISS in the APB group was 16 for TMaj, 5 for TMin, and 4 for TC compared with 8, 4, and 4, respectively, for MOI. Sensitivity for trauma activation of HR patients was 89.2% versus 89.1% (equivalent), while specificity increased from 45.8% to 65.8% for MOI versus APB, respectively.

CONCLUSION: For pediatric trauma patients, the emphasis on APB triage criteria and de-emphasis on MOI results in selection of higher-acuity patients for major activation while maintaining acceptable undertriage and overtriage rates overall. This improved accuracy of major activation results in a more cost-efficient resource use and fewer unnecessary disruptions for the surgeon, operating room, and other staff while maintaining appropriate capture and evaluation of trauma patients. The low sensitivity noted in both the MOI and APB groups is largely caused by the broad definition of HR patients used in this study. We recommend the use of APB criteria for pediatric trauma triage.

LEVEL OF EVIDENCE: Therapeutic study, level IV.