Pediatric Critical Care Fellow St. Christopher's Hospital for Children St Christopher's Hospital for Children Philadelphia, Pennsylvania, United States
Background: Pediatric trauma remains a significant source of global mortality. Severity of illness scores are used for outcome prediction and quality of care assessment. Complex calculations needed for current scores limit real time clinical application. The ideal clinical scoring tool would be easy to use, make accurate predictions, and could function as trending parameter throughout the hospital course.
Objective: This study investigated the utility of the VVR score, a robust outcome predictor for pediatric cardiac surgery, as outcome prediction tool for pediatric trauma patients.
Design/Methods: Children aged 0-18 years admitted to level I pediatric trauma center between January 1, 2014 and December 31, 2018 who were intubated upon admission were included. Of 4663 trauma patients, 50 met inclusion criteria. Using mortality as outcome, differences in VVR scores calculated at 6, 12, 24 and 48 hours post admission were compared by Mann-Whitney U test. Area under the receiver-operator curves (AUC) for Pediatric Index of Mortality (PIM2), Pediatric Risk of Mortality (PRISM III) as well as 6-, 12-, 24-, and 48-hour-VVR scores were compared using the non-parametric method of DeLong. In all cases, p-values ≤0.05 were considered significant.
Results: Median age was 3.5 years (IQR 1.75-11.25). Blunt trauma was most common (68%), followed by penetrating trauma (16%), burns (7%) and other mechanisms (1%). 28% of patients did not survive to hospital discharge. VVR was calculable for 35, 43, 40 and 45 patients at the 6-, 12-, 24- and 48-hour time point. PRISM III and PIM2 data was available for 42 and 46 patients, respectively. VVR scores were significantly higher for non-survivors than survivors at all 4 time points (figure 1). There was no significant difference between 24-hour VVR AUC and both PIM2 and PRISM III AUC (p=0.67 and 0.73, respectively). 6-hour VVR AUC, however, was significantly lower (p=0.03). Conclusion(s): VVR score differentiates pediatric survivors and non-survivors of trauma. In its current form, however, it is less robust than established measures during the early hospital course. Future studies will build on VVR score to develop an easy to use prognostic and monitoring tool for pediatric victims of trauma.
Figure 1: Box plot for survivors and non-survivors for different scoring tools and time points (1A-1D: VVR at 6, 12, 24 and 48 hours post admission; 1E: PIM2, 1F: PRISM III)
Authors/Institutions: I. Friederike Strelow, St Christopher's Hospital for Children, Philadelphia, Pennsylvania, United States; Christopher Pennell, St Christopher's Hospital for Children, Philadelphia, Pennsylvania, United States; Autumn D. Nanassy, St Christopher's Hospital for Children, Philadelphia, Pennsylvania, United States; Amit C. Misra, Children's Hospital Los Angeles, Los Angeles, California, United States; Stephen Aronoff, Lewis Katz School of Medicine at Temple University, Philadelphia, Pennsylvania, United States; Vicki Mahan, St Christopher's Hospital for Children, Philadelphia, Pennsylvania, United States