Late-Onset Neurologic Dysfunction in Pediatric Sepsis—What Brains Might Learn From Kidneys and Persistent Acute Kidney Injury*

Abstract

T his issue of Pediatric Critical Care Medicine (PCCM) includes an important article describing the relationship between the timing of development of encephalopathy relative to the identification of pediatric sepsis (1). The authors used the existing data set collected for a large, multicenter point prevalence study of pediatric sepsis, the Sepsis Prevalence, Outcomes and Therapies Study (2). Leveraging this large data set, the authors set out to determine if there was an association between the timing of development of encephalopathy and outcome. They identified two distinct cohorts of encephalopathy in pediatric sepsis: encephalopathy identified at sepsis recognition and encephalopathy, which developed or was identified 1-5 days after sepsis recognition. The criteria for encephalopathy are quite strict, and while both cohorts had increased risk of mortality or poor outcome, there was a greater odds of mortality associated with delayed onset of neurologic dysfunction in pediatric sepsis.The association between CNS dysfunction and risk of mortality in the setting of sepsis is well described in the literature. Although we are yet to develop an evidence or consensus-based definition of so-called "septic encephalopathy, " any identified symptom referable to CNS dysfunction is associated with increased risk of mortality. In 1990, Sprung et al (3) described increased mortality in sepsis for patients whose symptoms included acute altered mental status when compared with patients with normal mental status (49% vs 26%; p < 0.000001). A prospective case series of 50 adults with severe sepsis identified a stepwise increase in mortality associated with decreasing Glasgow Coma Scale score (4). Radiological evidence also demonstrates that the dysfunction in some of the sickest patients is associated with brain injury, including infarction, blood-brain barrier breakdown and edema, and focal and global ischemic injury (5). Sanchez-Pinto et al (6) derived several multiple organ system failure (MOSF) phenotypes from a large, retrospective data set. This group reported greatest odds of mortality in the multiple organ dysfunction syndrome phenotype with severe, persistent neurologic dysfunction. All these investigations, including the current report in PCCM (1), used different specific definitions for CNS dysfunction in sepsis, and all described meaningfully increased association with mortality (6). At this point, it is reasonable to infer that the development of any CNS dysfunction in a septic patient, at any point in their illness, should be considered an extremely concerning. Further, the relationship between CNS dysfunction and injury and mortality in the PICU has been well described, that is, previously well children who die in a PICU almost always die with a presenting or acquired CNS insult (7).

Editor’s Choice Articles for August

“…Start with the 2022 consensus criteria for acute neurologic dysfunction in critically ill children developed by the Pediatric Organ Dysfunction Information Update Mandate (PODIUM) collaborative (24). Also, consider the Sepsis Prevalence, Outcomes and Therapies (SPROUT) study investigators report on outcomes associated with the late onset of ABD relative to recognition of pediatric sepsis (25, 26). Like the PODIUM criteria, SPROUT used the Glasgow Coma Scale score and pupil examination findings.…”

Editor’s Choice Articles for October

“…As a reader of PCCM you may also want to review our recent material about timing of acute neurologic dysfunction in relation to sepsis recognition (13, 14) and the choice of clinical assessment (15). Finally, also consider the computational phenotype of “acute brain dysfunction” regarding database research—based on using neuroimaging or electroencephalography as part of evaluating neurologic change—which had better diagnostic performance than the GCS in sepsis (16).…”

Editor’s Choice Articles for August

“…In a cohort of over 550 pediatric patients, 15% had new neurologic dysfunction at some time during severe sepsis, and this finding was associated with greater odds of death or new moderate disability in survivors. The editorial is worth reading for its in-depth commentary on "septic encephalopathy", and it draws on parallels between acute kidney injury and brain pathophysiology in sepsis (6). My 'third' Editor's Choice covers the practice of using peripheral venous infusions of vasoactive agents-a topic that was identified as important in the recent pediatric surviving sepsis campaign international guidelines (10,11).…”

“…In a cohort of over 550 pediatric patients, 15% had new neurologic dysfunction at some time during severe sepsis, and this finding was associated with greater odds of death or new moderate disability in survivors. The editorial is worth reading for its in-depth commentary on “septic encephalopathy”, and it draws on parallels between acute kidney injury and brain pathophysiology in sepsis (6).…”