Identification of severely injured children at very low risk of emergent surgery: A national trauma registry study

• Published in: Injury 2024-09, Vol.55 (9), p.111678, Article 111678
• Main Authors: Gimelraikh, Yulia, Epstein, Danny, Radomislensky, Irina, Lipsky, Ari M., Givon, Adi, Berant, Ron, Berzon, Baruch, Raz, Aeyal, Ben-Eli, Danny, Bodas, Moran, Samuel, Nir
• Format: Article
• Language: English
• Subjects: Adolescent; Child; Child, Preschool; Children; Emergency Service, Hospital; Female; Humans; Infant; Infant, Newborn; Injury Severity Score; Israel - epidemiology; Male; Prediction; Registries; Retrospective Studies; Risk Assessment; Rule; Surgery; Trauma; Triage; Wounds and Injuries - surgery
• ISSN: 0020-1383; 1879-0267; 1879-0267
• DOI: 10.1016/j.injury.2024.111678

•The decrease in pediatric trauma cases necessitating immediate surgery facilitates the optimization of surgical resource allocation.•The study aimed to create a decision support tool to rapidly identify children unlikely to need emergency surgery upon hospital arrival.•Among children with blunt trauma mechanism, normal GCS, and low-risk heart rate, only 0.02 % required emergent surgery.•A limited set of physiological parameters available at hospital admission can effectively identify children at a low risk for emergent surgery. The number of pediatric trauma patients requiring surgical interventions has been steadily decreasing allowing for a judicious approach to immediately available resources. This study aimed to derive and validate a prediction rule that reliably identifies injured children who are at very low risk for requiring emergency surgery upon emergency department (ED) arrival. A retrospective cohort study of data included in the Israeli National Trauma Registry from January 1, 2011, through December 31, 2020, was conducted. We included children aged 0–14 years who presented to EDs from the scene of injury and were hospitalized. We excluded patients transferred between facilities or with isolated burns. The primary outcome was emergency operative intervention (EOI) performed within one hour of ED arrival. We tested mechanism, GCS, heart rate, and blood pressure as candidate predictors. We then randomized patients to two cohorts, derived and internally validated a prediction rule. During the study period, 83,859 children met enrollment criteria. The median age was 6 years (IQR 2–10) and 56,867 (67.8 %) were male; 75,450 (90.0 %) sustained blunt trauma. One hundred sixty-nine (0.20 %) children underwent EOI. In the derivation and validation cohorts, 34,138 (81.4 %) and 34,271 (81.7 %) patients, were classified as low risk based on blunt trauma mechanism, normal GCS (15), and low-risk heart rate (according to age). Of those, 8 (0.02 %) and 13 (0.04 %) required an EOI, respectively. In the validation cohort, the prediction rule for EOI had a sensitivity of 84 % (95 % CI 75–91), a specificity of 82 % (95 % CI 81–82), and a negative predictive value of 99.96 % (95 % CI 99.94–99.98). Among children with an Injury Severity Score>15, the sensitivity was 87 % (95 % CI 77–94), the specificity of 57 % (95 % CI 54–59), and the negative predictive value was 98.97 % (95 % CI 98.13–99.44). A limited set of physiologic parameters, readily available at hospital