Mortality prediction in pediatric trauma.

Area of Science:

• Pediatric critical care research
• Trauma patient outcomes
• Prognostic scoring systems

Background:

• Accurate mortality prediction is crucial for pediatric trauma patient selection and sample size determination.
• Existing scoring systems like PRISM III, PIM2, and PELOD are used for prognosis in pediatric intensive care units (PICUs).
• The Base Deficit, International normalized ratio and Glasgow Coma scale (BIG) score is a potential alternative for rapid mortality estimation.

Purpose of the Study:

• To compare the accuracy of the BIG score in predicting mortality in pediatric trauma patients.
• To evaluate the BIG score against established scores: PRISM III, PIM2, and PELOD.
• To assess the feasibility of using the BIG score in clinical practice.

Main Methods:

• Analysis of data from 149 PICUs between 2004 and 2015.
• Development of logistic regression models to assess mortality prediction.
• Comparison of Area Under the Curve (AUC) from Receiver Operator Characteristic (ROC) curves for each score.

Main Results:

• The BIG score was calculable for only 0.33% of the 45,377 analyzed pediatric trauma patients.
• Despite limited calculability, the BIG score demonstrated strong predictive accuracy with an AUC of 0.94.
• This accuracy was comparable to PRISM III (0.96), PIM2 (0.97), and PELOD (0.93).

Conclusions:

• The BIG score is a highly accurate tool for predicting mortality in pediatric trauma patients.
• The BIG score's accuracy supports its potential utility, despite challenges in its widespread application.
• Further research may explore methods to improve the calculability of the BIG score in diverse pediatric trauma populations.