Pediatric Delirium: We’ve come a long way, and we’ve only just begun

In this month’s issue of Critical Care Medicine, Semple et al present a systematic review of delirium prevalence in critically ill children. The authors show an overall pooled prevalence rate of 34%, with variability based on the subgroup studied. Thirty-one papers -- including single- and multi-center studies, with 9,756 children from 14 countries -- met inclusion criteria [1].

There was striking consistency between studies with respect to findings amongst specific subgroups. As an example, nine out of the ten studies that took place in general pediatric intensive care units (ICUs) demonstrated delirium rates below the pooled prevalence, ranging from 17-28%. By contrast, delirium rates in all four studies focusing on patients in cardiothoracic ICUs were above the pooled prevalence, ranging from 40-66% [1]. Regardless of subgroup studied, this systematic review confirms what individual studies have suggested: delirium affects a substantial proportion of critically ill children.

Overall, these findings are consistent with the robust body of literature describing delirium in critically ill adults. In a 2018 meta-analysis published in this journal, the overall pooled prevalence of delirium in adult ICUs was 31% [2]. Similar to the findings in this pediatric systematic review, the majority of delirium was of the hypoactive subtype. This is noteworthy, as without routine screening, most cases of hypoactive delirium remain undiagnosed [3]. This represents a missed opportunity for early detection and intervention, which can shorten delirium duration and improve patient outcomes [4]. In contrast to adult ICUs, where delirium screening is now considered standard-of-care, a recent study revealed that only 44% of pediatric ICUs engage in routine delirium monitoring [5]. Implementation of standardized delirium screening poses an important opportunity for quality improvement in all pediatric ICUs. This systematic review argues that, in good conscience, pediatric intensivists can no longer afford to ignore this widespread problem.

Strengths of this study include rigorous search methodology, following the guidelines for Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [6]. In addition, the authors undertook careful quality assessment of the included studies. A weakness of the study is the cursory review of risk factors associated with pediatric delirium (PD). Despite stating this as a secondary study aim, the search criteria used for the primary review question (“what is the prevalence of pediatric delirium reported in critically ill children?”) led to the exclusion of several important studies focusing on potential risk factors for PD. As such, the risk factors and associations described in the results are incomplete. There is an unmet need for a high-quality systematic review specifically focused on risk factors for delirium in critically ill children.

To quote the groundbreaking singer-songwriter Loretta Lynn, “We’ve Come A Long Way, Baby” [7]. A pooled prevalence study of delirium in children would have been impossible only 10 years ago. Although no date barriers were set in the search criteria for this systematic review, every single one of the included studies was published within the last decade. The recent availability of well-accepted validated tools for bedside delirium screening in children of all ages has allowed for establishment of a consistent operational definition for delirium amongst studies [8,9]. This was essential for advancement of the field.

In addition, establishment of uniform terminology has enabled tremendous progress. Until recently, a multiplicity of terms was used to describe delirium in children: ICU psychosis, encephalopathy of critical illness, acute brain dysfunction, apathy syndrome, acute confusional state, and others. This lack of common language severely limited communication across disciplines (for example: between pediatric neurologists, psychiatrists, and intensivists) and hindered collaborative research [10]. More recently, “delirium” has been widely accepted as the term describing the acute and fluctuating change in awareness and cognition that occurs in the setting of pediatric critical illness [11]. This common terminology has been vital in the advancement of pediatric delirium research.

Despite the progress that has been made, the authors note that the bulk of pediatric delirium studies have been observational. This was entirely necessary, as until recently, we did not have a clear picture of the epidemiology of delirium in children. We first needed to understand the scope of this problem: Who gets delirious? When do they get delirious? How long does the delirium last? What are the modifiable risk factors for PD? What are the associations between delirium and outcomes?

We did not know the answers to these questions a decade ago; we now do. This systematic review has added an important piece to our understanding of the epidemiology of pediatric delirium. With the knowledge we have gained, it is time to begin the important work of designing interventional studies. There is a golden opportunity for pediatric delirium researchers to engage in large-scale multisite longitudinal investigations exploring treatment and prevention approaches for this frequent complication of pediatric critical illness. In conclusion, although we’ve come a long way, we have also only just begun.