The sensitivity and negative predictive value of a pediatric cervical spine clearance algorithm that minimizes computerized tomography

Abstract

BackgroundIt is crucial to identify cervical spine injuries while minimizing ionizing radiation. This study analyzes the sensitivity and negative predictive value of a pediatric cervical spine clearance algorithm. MethodsWe performed a retrospective review of all children <21years old who were admitted following blunt trauma and underwent cervical spine clearance utilizing our institution's cervical spine clearance algorithm over a 10-year period. Age, gender, International Classification of Diseases 9th Edition diagnosis codes, presence or absence of cervical collar on arrival, Injury Severity Score, and type of cervical spine imaging obtained were extracted from the trauma registry and electronic medical record. Descriptive statistics were used and the sensitivity and negative predictive value of the algorithm were calculated. ResultsApproximately 125,000 children were evaluated in the Emergency Department and 11,331 were admitted. Of the admitted children, 1023 patients arrived in a cervical collar without advanced cervical spine imaging and were evaluated using the cervical spine clearance algorithm. Algorithm sensitivity was 94.4% and the negative predictive value was 99.9%. There was one missed injury, a spinous process tip fracture in a teenager maintained in a collar. ConclusionsOur algorithm was associated with a low missed injury rate and low CT utilization rate, even in children <3years old.Level of Evidence: IV.