Validation of a pediatric early warning system for hospitalized pediatric oncology patients in a resource-limited setting

Background Pediatric oncology patients are at high risk of clinical deterioration, particularly in hospitals with resource limitations. The performance of pediatric early warning systems (PEWS) to identify deterioration has not been assessed in these settings. This study evaluates the validity of PEWS to predict the need for unplanned transfer to the pediatric intensive care unit (PICU) among pediatric oncology patients in a resource-limited hospital. Methods A retrospective case-control study comparing the highest documented and corrected PEWS score before unplanned PICU transfer in pediatric oncology patients (129 cases) with matched controls (those not requiring PICU care) was performed. Results Documented and corrected PEWS scores were found to be highly correlated with the need for PICU transfer (area under the receiver operating characteristic, 0.940 and 0.930, respectively). PEWS scores increased 24 hours prior to unplanned transfer (P = .0006). In cases, organ dysfunction at the time of PICU admission correlated with maximum PEWS score (correlation coefficient, 0.26; P = .003), patients with PEWS results ≥4 had a higher Pediatric Index of Mortality 2 (PIM2) (P = .028), and PEWS results were higher in patients with septic shock (P = .01). The PICU mortality rate was 17.1%; nonsurvivors had higher mean PEWS scores before PICU transfer (P = .0009). A single-point increase in the PEWS score increased the odds of mechanical ventilation or vasopressors within the first 24 hours and during PICU admission (odds ratio 1.3-1.4). Conclusions PEWS accurately predicted the need for unplanned PICU transfer in pediatric oncology patients in this resource-limited setting, with abnormal results beginning 24 hours before PICU admission and higher scores predicting the severity of illness at the time of PICU admission, need for PICU interventions, and mortality. These results demonstrate that PEWS aid in the identification of clinical deterioration in this high-risk population, regardless of a hospital's resource-level. Cancer 2017;123:4903-13. © 2017 American Cancer Society.