Comparison of Cervical Spine Injury Prediction Rule Across Ages

OBJECTIVE: To evaluate the performance of the Pediatric Emergency Care Applied Research Network (PECARN) cervical spine injury (CSI) prediction rule among different pediatric age groups and derive and assess age-specific CSI prediction rules. METHODS: We conducted a planned secondary analysis of the PECARN CSI study of children younger than 18 years with known or suspected blunt trauma evaluated at 1 of 18 children's hospitals' emergency departments. Participants were divided into 4 clinically relevant age groups (<2, 2-7, 8-15, and 16 to <18 years). Age-specific rules were derived from the 4 high-risk predictors of the PECARN CSI Prediction Rule and using a classification and regression tree (CART) analysis to identify additional non-negligible age-specific risk factors. Rule performance was assessed using sensitivity, specificity, and positive and negative predictive values (NPVs) for each age group. RESULTS: There were 22 430 participants included in analysis. Three to 8 risk factors were identified as non-negligible risk in CART analysis for each age-specific rule. The sensitivity of the PECARN CSI Prediction Rule (range 91.7%-94.4% across age groups) was similar to age-specific rules (range 88.4%-97.5%). The specificity of the PECARN CSI Prediction Rule (range 44.6%-72.3% for age groups) was also similar to age-specific rules (range 45.9%-75.9%). NPVs (all ≥99.6%) were similar across age groups for all rules. CONCLUSIONS: Rule performance for the age-specific CSI prediction rules and the PECARN CSI Prediction Rule were similar. These findings support clinicians' use of the PECARN CSI Prediction Rule across the pediatric age spectrum.