Trends and variation in cervical spine imaging utilization across children's hospitals for pediatric trauma

Abstract

Background Cervical spine evaluation is a critical component in trauma evaluation, and though several pediatric cervical spine evaluation algorithms have been developed, none has been widely implemented. Here, we assess rates of cervical spine imaging use across children’s hospitals, specifically temporal trends in imaging use, variation across hospitals in imaging used, and timing of magnetic resonance imaging in admitted patients. Methods Data from the Children’s Hospital Associations Pediatric Health Information System was abstracted from 2015 to 2020. Patients less than 18 years of age seen in the emergency department with an International Classification of Diseases (ICD)-10 code indicative of trauma and cervical spine plain radiograph or computed tomography in the emergency department were included. Data visualization and descriptive statistics were used to assess rates of imaging use by age, year, hospital, injury severity, and day of service. Changes in rates of imaging use over time were evaluated via Chi-square test. Results Across 25,238 patient encounters at 35 children’s hospitals, there was an increase in use of cervical spine computed tomography from 2015 to 2020 (28.5 to 36.5%). There was substantial inter-institutional variation in rates of use of plain radiographs versus computed tomography for initial evaluation of the cervical spine across all age groups and regardless of rates of severe injury across institutions. Magnetic resonance imaging was obtained more than three days after admission in 31.5% of intensive care patients who received this imaging. Conclusions Increasing use of computed tomography, substantial inter-institutional variation in rates of use of plain radiographs versus computed tomography, and heterogenous timing of magnetic resonance imaging for evaluation of the pediatric cervical spine demonstrate the growing need for development and implementation of an age-specific cervical spine evaluation algorithm to guide judicious use of diagnostic resources. Level of Evidence Level III, Epidemiologic