Beyond Staying Alive…

With the advent of centralized pediatric intensive care units, specialized training in pediatric critical care, and advances in medicine and technology, children have experienced improved survival during critical illness.[1, 2] The reduction in mortality has propelled an interest in characterizing and improving PICU survivorship.[3, 4] In this issue of Pediatric Critical Care Medicine, Slater et al. report on mortality outcomes of a cohort of >96,000 children discharged from Australian PICUs between 1997–2018. In doing so, they characterize long-term survival patterns and factors associated with mortality after pediatric critical illness. (5)

Strengths of the study include its use of linked registry data to cultivate a large cohort with available longitudinal data, which is particularly notable given the duration of follow-up over a 20-year period. The study focuses on mortality outcomes, characterizing mortality trends over time as well as patient-level risk factors for mortality. The authors found that risk of death after discharge decreased over time. The hazard ratio for death was highest in children admitted in 1997–2002 while mortality risk was 30–40% lower for those admitted after 2008. The highest risk of mortality was seen in the period immediately following discharge, with risk decreasing over the subsequent four years. Overall mortality risk was higher in children with specific reasons for PICU admission (e.g., neurologic and gastrointestinal), highest predicted risk of PICU mortality, and recurrent critical illness as well as those who lived in very remote areas. Notably, the authors did not find a relationship between risk of death and higher social disadvantage.

The decreasing risk of long-term mortality perhaps represents a change in the case mix of children receiving critical care services, availability of new pharmaceutical and/or medical technology in both in inpatient and outpatient settings, shifts in clinical management (e.g. sedation practices, early mobility, etc.) or other factors. In this cohort, risk of injury as a primary reason for PICU admission decreased over the course of the study period, while respiratory diagnoses increased. While the authors do not present the data, it is likely that patients in more recent epochs have more complex medical needs and increased medical technology use, as has been described in U.S. PICUs.[6] It is plausible that newer technology may improve risk of mortality. For example, in a subset of U.S. children’s hospitals, PICU admissions for bronchiolitis have doubled while the proportion of children requiring invasive mechanical ventilation has remained stable with shifts in practice towards increased use of non-invasive ventilation.[7] As a result, these children may require less sedative exposure or experience increased early mobility. It is unknown if and how this changing case mix and PICU therapeutic strategies affect longer term outcomes.

The findings reported by Slater et al. are an important call to action. Consistent with trends in PICU mortality during hospitalization, this work shows that reductions in PICU mortality after hospitalization also have occurred over time. At first glance, we can and should celebrate that the field of pediatric critical care has achieved exactly what we set out to do with the establishment of the specialty not only at hospital discharge but extending for years after discharge. However, PICU survivorship research demonstrates that although these children are staying alive, they increasingly experience morbidities that challenge their ability to thrive.[8] For example, prior studies have identified PICU survivors at-risk of long-term morbidity including functional limitations, changes in health-related quality of life, emotional challenges, financial strain, and difficulties returning to school and leisure routines.[9–14] The trajectory of mortality trends seen in this report is akin to what is thought to be understood about trajectories of recovery for post-intensive care syndrome in pediatrics (PICS-p): peak morbidity typically exhibited in the immediate post-discharge period and variable recovery over the subsequent year.[15–18] While the authors are to be commended for their comprehensive evaluation of mortality, this report is also remarkable for what it does not describe: any sense of the long-term physical health, cognitive or academic performance, emotional well-being, or social effects of critical illness on surviving children and their families.

Finally, this report provides food for thought regarding potential interventions to reduce long-term mortality and morbidity among survivors. This manuscript suggests that we likely need to focus on targeted interventions for high-risk survivors in the early post-discharge period to yield the highest impact. However, we will need comprehensive measures of their wellbeing to effectively study the interventions’ effects. In addition to highlighting potentially more vulnerable patients and suggesting a timeline for intervention, this study examined other factors which may be associated with adverse long-term outcomes. The authors found no association between mortality risk and increased social disadvantage, while residing in very remote settings was associated with increased risk of death. These factors are likely to impact the feasibility of any supportive interventions and may still be linked with morbidity outcomes. These nuanced associations are likely to vary across societies and across health care delivery systems, and disparities may be particularly marked in jurisdictions where we lack the ability to study them at all.

This study, then, should encourage PICU clinicians and researchers to consider ways of meaningfully capturing longitudinal post-critical illness morbidity. Published longitudinal studies beyond the first year after discharge tend to be limited regarding sample size, specific patient populations, and single outcomes and demonstrate challenges with cohort retention over time.[3] The authors’ ability to provide longitudinal mortality data over a twenty-year period is encouraging for future efforts to capture extensive longitudinal morbidity data. As an often-quoted saying goes, “the best time to plant a tree was twenty years ago, but the second best is now”—the same is true for a focus on PICU-related morbidity and our ability to monitor long-term outcomes over time. Improvements in overall PICU survival at discharge have rendered the study of mortality as the singular long-term outcome of interest obsolete; we know (most of) our patients are staying alive. While administrative linkages certainly should be used, when possible, PICU researchers also should move toward intentionally including long-term, patient- and family-focused outcomes in all future pediatric critical care trials.