Validation of the pediatric surgical risk assessment scoring system

Abstract

Background: Within the adult population, there exist numerous validated risk stratification tools aimed at predicting postoperative outcomes using preoperative and intraoperative parameters. However, similar tools for the pediatric population are scarce. We previously developed and reported on a scoring system to predict postoperative complications in children undergoing surgery at Boston Children's Hospital. The objective of this study was to validate our scoring system to determine its effectiveness in identifying children at risk for major complications or death within 30 days following surgery. Study Design: A computer program randomly selected 2015 out of 11,734 patients who underwent surgery at our institution in 2009. The severity of the complications was graded based on the Clavien classification system, with major complications being a grade III or higher. The Preoperative Complication Score (PCS) and the Overall Complication Score (OCS) were calculated for all patients, and Receiver Operating Characteristic (ROC) curves were generated for each scoring system. Results: The overall incidence of major complications was 3.9% (79 patients). Proportionally, cardiac surgery had the highest percentage of major complications (20% of the performed surgeries). Both PCS and OCS demonstrated excellent correlation with postoperative outcomes with c-statistic values of 0.740 (CI 0.682-0.800, p < 0.001) and 0.767 (CI 0.712-0.822, p < 0.001) respectively. Conclusions: We determined that both the PCS and OCS are effective in identifying children at risk for major complications and death following surgery. Further studies will be needed to determine if these scoring systems are applicable to children undergoing surgery at other institutions and if the use of the scoring systems would result in improved clinical outcomes and reduced costs.