LAUSR

OBJECTIVES Early pelvic binder placement in the field stabilizes pelvic fractures and tamponades potential hemorrhage within the pelvis. Despite known risk factors for pelvic fracture, it remains challenging to quickly triage and correctly apply a pelvic binder. We aim to develop a prediction model that exclusively uses prehospital criteria to inform the decision to place a pelvic binder. METHODS The trauma registry was used to identify all trauma patients admitted to an urban Level I trauma center between January 2013 and December 2017. Variables collected included patient demographics, mechanism of injury, prehospital vital signs, and the presence of a pelvic fracture. Participants were randomly assigned to a training group (70%) or a validation group (30%). Univariate analyses were used to identify significant predictors for use in multivariate predictive models. RESULTS A total of 8,480 (65% male; median age 49; median ISS 9) and 3,676 (65% male; median age 48; median ISS 9) trauma patients were randomly assigned to the training and validation groups, respectively. Univariate analysis showed significant likelihood of pelvic fracture associated with female sex, hemodynamic instability (initial systolic blood pressure < 90 mmHg), blunt injury type, specific mechanisms of injury (motor vehicle collision, motorcycle collision, pedestrian struck by motor vehicle, crushing injury, and riding an animal), impact location, and position in vehicle. Multivariate models adjusting for blunt type injury, hemodynamic instability, impact location, and position in vehicle showed that presence of two or more of these risk factors is significantly associated with presence of pelvic fracture. CONCLUSION Establishing select prehospital criteria for the empiric application of pelvic binders for patients in the field with blunt injuries, hemodynamic instability, frontal or side motor vehicle collision impact, and non-front seat passenger may improve outcomes among patients with pelvic fractures.