Identifying the critically ill pediatric oncology patient

Hospitalized pediatric oncology patients face potential life-threatening complications from both the disease and its treatment, which may require a transfer to the pediatric intensive care unit (PICU). Timely recognition of a patient’s decline is crucial to start early interventions, aiming to prevent further progress to irreversible organ damage and death. This thesis focuses on the timely identification of clinically deteriorating patient, both at the inpatient ward and the PICU. Ultimately, we aim to improve the outcomes of critically ill pediatric oncology patients. Prior to the research presented in this thesis, there were gaps of knowledge in the optimal standard of care for critically ill pediatric oncology patients, as well as how we can best recognize a deteriorating patient.

As a first step towards improving the outcome of this specific patient group, we have formulated a top-5 list of research priorities for the next decade in Chapter 2, based on international consensus across Europe. One of the priorities is the improvement of detection of clinically deteriorating patients at the inpatient ward.

Despite the widespread use of pediatric early warning system (PEWS) scores, few studies have validated a PEWS in pediatric oncology patients. Therefore, we summarized and critically appraised the existing evidence of the ability of a PEWS to predict inpatient deterioration, and the effect of implementation of PEWS on patient outcomes (Chapter 3). We found limited evidence on both research questions, and although high predictive ability of a PEWS score were frequently reported, most studies were at high risk of bias. Therefore, in this thesis, we assessed the performance of a modified Bedside Pediatric Early Warning System (PEWS) score in predicting patient deterioration (Chapter 4 and 5). Our study is the first to include all modified Bedside PEWS scores of all hospitalized pediatric oncology patients, while accounting for the longitudinal, time-dependent nature of a PEWS score. The main finding of this study was that this score can be used to assist clinical decision-making to increase the level of care. In addition, we found nuances to the use of this score as a clinical prediction model for the timely detection of clinical deterioration, that may be valuable for future studies to further optimize the risk prediction of clinical deterioration and the efficiency of escalation of care.

In Chapter 6, in a study at the PICU, we identified hemato-oncological diagnosis, unplanned admission, and the number of failing organs at PICU admission as significant risk factors for new or progressive multi-organ failure (MOF). Remarkably, in a large part of PICU admissions, dysfunction of one or more organs was already present at the start of the admission. These findings may help to identify patients at high risk for organ failure who may benefit from closer monitoring and early interventions at the ward, prior to PICU admission, aiming to halt progression to irreversible organ damage.

Overall, the research in this thesis represents a significant step towards the overarching goal of improving survival rates and enhancing the quality of life for critically ill pediatric oncology patients.