Validation of Clinical Criteria for Obtaining Maxillofacial Computed Tomography in Patients With Trauma

Abstract

More than 180,000 patients present annually with facial trauma to emergency rooms in the United States. Maxillofacial computed tomography is the gold standard in identifying facial fractures. Providers must evaluate patients quickly; therefore, they use decision instruments to determine which patients need imaging. We previously developed a decision instrument that identified patients with trauma at low risk for facial fracture who could avoid imaging. The present study aims to perform an internal validation of that tool. The decision instrument used 5 criteria: bony step-off or instability, periorbital swelling or contusion, Glasgow Coma Scale <14, malocclusion, and tooth absence. The presence of any 1 finding placed the patient at high risk for fracture. In the present study, a retrospective review was conducted on all of the patients with trauma evaluated at a Level I trauma center for >1 year. Inclusion criteria were maxillofacial physical examination, head and maxillofacial computed tomography at presentation. Physical examination findings were collected and imaging reviewed to determine whether the decision tool could accurately detect the presence of a facial fracture in a different patient population from which it was derived. One hundred seventy-nine patients met enrollment criteria. Facial fractures occurred in 81% of patients (n = 145). The decision instrument was 97.4% sensitive (95% confidence interval, 93.8-99.3) for the presence of facial fracture. The negative predictive value was 81.3% (95% confidence interval, 55.0-95.0). Application of the instrument resulted in a missed injury rate of 2.6% (n = 3). All of the missed fractures were nondisplaced and managed nonoperatively. The proposed decision tool identifies patients with trauma at low risk for facial fracture who can avoid maxillofacial imaging. Validation in a prospective study is warranted.