The importance of diffuse brain swelling in the pathophysiology of pediatric severe and moderate traumatic brain injury (TBI) has long been touted as a distinct feature compared with adults. Seminal studies carried out between 1979 and 1981 by Bruce and collagues reported on the importance of this phenomenon in pediatric TBI, calling it "the syndrome of malignant brain edema." That early work suggested that the etiology of diffuse brain swelling after severe and moderate pediatric TBI was largely due to hyperemia, based on groundbreaking assessments of cerebral blood flow (CBF) quantified by xenon in a small sample of children. Hyperemia was purported to increase cerebral blood volume and thus produce diffuse swelling and intracranial hypertension. Those findings served as the basis for the use of aggressive hyperventilation as a first-line therapy in severe pediatric TBI for 2 decades. However, subsequent studies in the late 1990s in larger series of patients suggested that hyperemia was uncommon in pediatric TBI, with low rather than high CBF associated with unfavorable outcome, and with the concern that hyperventilation could exacerbate ischemia. Subsequently, a study by Marmarou et al, in 76 adults with severe TBI during their intensive care unit (ICU) admission and 30 healthy volunteers using a combination of stable xenon computed tomography to measure cerebral blood volume, magnetic resonance imaging to quantify brain water, and gadolinium to assess blood-brain barrier (BBB) permeability revealed that edema was the overwhelmingly predominant mechanism of brain swelling, and surprisingly, based on limited BBB permeability, that cytotoxic rather than vasogenic edema was largely driving brain swelling and intracranial hypertension in adults. A similar study has never been carried out in pediatric TBI, to our knowledge, and despite its potential importance, the pathophysiological underpinnings of diffuse cerebral swelling in pediatric TBI have remained relatively unexplored at the physiological, pathological, and molecular level.

In this study, Fullerton et al provide a simple and elegant neuropathological report on cerebral swelling and BBB injury in 81 pediatric and 62 adult patients with moderate to severe TBI who died within 2 weeks of injury. Specimens were randomly selected from the Glasgow TBI archive. The cause of death, survival time, and postmortem intervals were comparable between groups; however, road traffic collisions predominated (73%) in pediatric cases, while falls predominated (44%) in adults, and ischemic brain injury was also more common in the pediatric patients. Brain swelling was more common in pediatric vs adult TBI, 83% vs 65%, respectively, and was bilateral in 95% of pediatric patients, vs only 54% of adults. Despite similar proportions of BBB disruption (assessed by fibrinogen and immunoglobin G immunohistochemistry) in pediatric (80%) and adult (91%) patients, the patten of BBB injury differed greatly between groups. Pediatric patients had BBB disruption localized largely (81%) in capillaries, while adults had BBB damage largely in medium to larger diameter vessels, with only 31% seen in capillaries. Preferential capillary BBB disruption was seen in only 1 adult aged 20 years. This study by Fullerton et al provides important possible clues about the long-known phenomenon of diffuse cerebral swelling after TBI in infants and children, and suggests that it is produced by vasogenic edema, rather than cytotoxic edema or hyperemia.

Many potential mechanisms contributing to diffuse cerebral swelling in infants and children have been suggested in preclinical and clinical studies and comprehensively reviewed. These include biomechanical differences related to unique mechanical properties of the skull in infants and young children. Children, with a proportionally larger head-to-body ratios and less neck support may also be more susceptible to acceleration-deceleration forces, like those experienced in road traffic collisions, with subsequent development of diffuse axonal injury, which is characterized by scattered hemorrhages in the microvasculature on neuroimaging and neuropathology. Also, the important contribution of hypoxic-ischemia brain injury in pediatric TBI (confirmed in the study by Fullerton et al), which can superimpose a global insult; the high prevalence of seizures in pediatric TBI producing CBF increases that can fail to meet metabolic demands; greater diffusion of excitotoxic amino acids through the developing vs adult brain; enhanced BBB permeability in the developing brain; and putative greater risk for spreading depolarization could all play a role. Regarding age-related differences in BBB permeability, seminal studies performed in pediatric models of cardiopulmonary arrest, not TBI, suggested enhanced BBB disruption in the developing vs adult brain.

The underpinnings of the observed age-associated difference in the topography of BBB injury shown by Fullerton et al remain to be defined. It is possible that biomechanical differences in injury mechanism contributed to the findings. The high prevalence of motor vehicle collisions in the pediatric patients vs falls in adults could differentially impact the shear forces imparted on the cerebral vasculature. Sensitivity analysis of patients matched for injury mechanism and severity could be helpful. Similarly, given that these patients received fatal injuries with as long as 2 weeks of ICU care, a role of treatment effects cannot be excluded. Gonda et al reported that aggressive hyperosmolar therapy in pediatric severe TBI is associated with microvascular injury from neutrophil degranulation and platelet aggregation, and high-dose intra-arterial mannitol administration has long been used to open the BBB via osmotic effects on the cerebrovascular endothelium to deliver therapies. A contribution to endothelial damage cannot thus be ruled out in some patients even with intravenous administration. Hyperosmolar therapy is the most common therapy in pediatric severe TBI, used in more than 81% of patients in the 2022 ADAPT comparative effectiveness trial of 1000 pediatric TBI patients treated in 44 sites. Details of the intensity of hyperosmolar therapy use here could be helpful. Nevertheless, the fact that a substantial number of patients with early mortality showed this pathology might argue against an iatrogenic etiology, since the mean (SD) survival time was 57.8 (56.9) hours in the pediatric group. Although a role of BBB injury in repetitive concussion is a topic of emerging interest, caution is also in order in extrapolating these findings to explain the underpinnings of the second impact syndrome. Catastrophic brain swelling and death in young athletes with multiple concussions could be linked to these age-related differences in BBB disruption, but it must be recognized that the findings presented by Fullerton et al are all from patients with moderate to severe TBI, not concussions.

Based on this provocative work, future studies should explore potential age-related differences in the cellular and molecular response to TBI in the cerebral microcirculation. Recent preclinical studies using single-cell RNA sequencing have revealed important differences in the endothelial, neuroinflammatory, and ependymal responses across multiple TBI models in adult mice, including different effects across brain regions, and important sex differences. Similar approaches should be applied to preclinical models of developmental vs adult TBI, and in pediatric vs adult clinical samples, to explore possible molecular mechanisms underlying the differences revealed by Fullerton et al. Some new insight is also emerging into the mechanisms underlying the development of cerebral edema, such as the role of the sulfonylurea receptor 1-transient receptor potential melastatin 4 cation channel and the interactions between distinct forms of cerebral edema, including vasogenic, cytotoxic, and contusional swelling. New therapeutic approaches, such as glibenclamide and agents to reseal an injured BBB, are also being evaluated in preclinical and clinical studies to prevent the development of cerebral edema rather than respond to it. Thus, this long-overdue work by Fullerton et al is timely, given the ultimate goal of precision therapy for this aspect of pathophysiology of special importance to pediatric TBI.

Corresponding Author: Patrick M. Kochanek, MD, Department of Critical Care Medicine, Safar Center for Resuscitation Research, UPMC Children's Hospital of Pittsburgh, University of Pittsburgh School of Medicine, 4401 Penn Ave, Pittsburgh, PA 15224 ([email protected]).

Conflict of Interest Disclosures: Dr Kochanek reported receiving grants from Chuck Noll Foundation during the conduct of the study. Dr Simon reported receiving grants from the National Institutes of Health during the conduct of the study. No other disclosures were reported.