New Guidelines Stimulating New Investigation

Trauma is the leading killer of children, and traumatic brain injury (TBI) is involved in approximately half of these cases. It is also a key contributor to long-term morbidity. In 2003, the first “Guidelines for the Medical Management of Severe Traumatic Brain Injury in Infants, Children and Adolescents” were published as a special triple supplement in the journals Pediatric Critical Care Medicine,1 Critical Care Medicine and the Journal of Trauma. In that document, the guidelines committee was faced with the difficult task of writing evidence-based guidelines, despite the existence of very limited pediatric evidence on severe TBI. Notably absent were prospective, randomized clinical trials of any therapy. For example, there was only a single clinical investigation focused on the use of mannitol in pediatric TBI2—this in spite of the fact that mannitol is a staple in the management of pediatric TBI and represents an agent that is specifically a brain-directed therapy. Further complicating the situation, mannitol represents an agent commonly used as part of a standard of care regimen when other therapeutic approaches are being tested.

Although an algorithm for care was generated for pediatric TBI, it was based on a limited number of studies in infants and children and relied importantly on expert opinion and clinical trials in adults. The most important role of the guidelines was to set the research agenda and stimulate new clinical trials in infants and children with severe TBI.3 Remarkably, despite a short two years since the guidelines were written, it is exciting to see that a number of new studies have been published. This article will address some of the highlights.

Descriptive Studies

Several articles addressed new issues in the neuropathology, neurophysiology, cellular and molecular biology, and epidemiology of pediatric TBI. In the area of neuropathology, Tong et al4 took advantage of the well-known association between axonal injury and vascular injury to study diffuse axonal injury in 40 children with severe traumatic brain injury. Using a new technique of susceptibility-weighted magnetic resonance imaging, a significant association between diffuse axonal injury and outcome was demonstrated.

On the topic of neurophysiology, using a prospectively collected dataset of pediatric TBI patients in Seattle, Washington, Vavilala et al5,6 carried out two studies. First, she examined blood pressure in 172 children with severe TBI and demonstrated that an age-appropriate systolic blood pressure < 75th percentile was significantly associated with poor outcome—with an odds ratio of 4.2.5 She also studied 36 children with severe TBI using transcranial Doppler and showed that cerebral autoregulation was impaired 67% of the time.6 Surprisingly, loss of blood pressure autoregulation of cerebral blood flow was associated with poor outcome—a finding that has not been generally seen in adult series.

Cochran et al7 carried out a single center descriptive study of 170 pediatric trauma patients, 64 of whom had isolated TBI. Hyperglycemia (serum glucose concentration > 200 mg/dL) and Glasgow Coma Scale (GCS) score at admission independently predicted death in a logistic regression model. The optimal management of serum glucose after severe TBI in infants and children remains a very controversial area with limited investigation. It is an area that was not addressed in the guidelines and in which careful studies are necessary. It is not clear if the common contemporary practice, in adult neurointensive care,8 of using only 0.9% sodium chloride (without glucose) is either safe or optimal in pediatric TBI.

In the area of molecular biology, Lai et al9 studied 20 infants and children after severe TBI and reported that cerebrospinal fluid (CSF) levels of heat shock protein-70 were markedly elevated after injury—a finding that was strongly associated with child abuse (inflicted childhood neurotrauma) as an injury mechanism. The presence of chronic injury or stress in these victims may be of special importance to this observation.

Finally, in an important article, Keenan et al10 provided the first prospective, population-based estimate of serious or fatal inflicted TBI in children in the United States. Based on data collected from nine hospitals in North Carolina with a pediatric intensive care unit or step-down unit and data on fatal injuries from the chief medical examiner of North Carolina, 152 cases of serious or fatal TBI were observed over a two-year period. Fifty-three percent were cases of inflicted TBI—an incidence of 17 per 100,000 person-years in the first two years of life.

Clinical Trials

The guidelines have stimulated both a number of clinical trials and descriptions of case series in pediatric TBI. Studies in three areas will be discussed, hypothermia, decompressive craniectomy, and CSF drainage. Biswas et al11 in a study of 21 children, reported that 48 hours of mild-moderate hypothermia, initiated in the initial six hours after injury, was safe and reduced the severity of intracranial hypertension, even when applied in conjunction with conventional therapy. We look forward to the results of the prospective, multi-center, randomized, controlled trial of mild hypothermia that was just completed by Adelson and co-investigators.

The largest number of new reports in pediatric TBI has been in the area of decompressive craniectomy. Four new papers have been published on this approach as specifically applied in pediatric cases. Figaji et al12 reported on a series of five children (between five and 12 years of age) in whom decompressive craniectomy was used early after severe TBI with clinical deterioration. In general, these children exhibited favorable outcomes assessed at between 14 and 42 months after injury. Hejazi et al13 reported on the use of unilateral decompressive craniectomy in seven pediatric cases with decorticate posturing and reported favorable outcome in all of the patients. Similarly, Messing-Junger et al14 described 87 cases of decompressive craniectomy after severe TBI. Seven of these cases were in children, and five of the seven were reported to have had survival with good outcome. Finally, Ruf et al15 in a pilot study used decompressive craniectomy (either unilateral or bilateral) to treat intracranial hypertension (ICP > 20 mmHg for longer than 30 minutes) that did not respond to treatment with a single dose of barbiturate and mannitol. Overall, outcomes at six months were favorable. However, none of these studies had a concurrent control group and, as outlined in the study by Ruf et al,15 a randomized controlled trial is needed.

Certainly one could argue that there were many limitations in each of these reports, nevertheless, these studies build on the favorable controlled trial of decompressive craniectomy published by Taylor et al16—which was one of the last studies included in the guidelines. This is an area of considerable interest and promise, but randomized controlled comparisons with assessment of long-term outcome are a must.

Finally, Shore et al17 recently reported a comparison of the effect of continuous vs. intermittent CSF drainage in a twoinstitution series of pediatric patients with severe TBI. Continuous drainage was associated with nearly twice the volume of CSF removal, much reduced levels of a number of biochemical markers—such as interleukin-6 and neuron specific enolase—and lower ICP than the intermittent group. It was interesting that the theoretical complication of clogging of the ventriculostomy catheter was not a significant problem with the use of continuous drainage—even in infants. A randomized controlled trial comparing these two modalities is needed.

In conclusion, it appears that the “Guidelines for the Medical Management of Severe Traumatic Brain Injury in Infants, Children and Adolescents” document has served one of its major purposes and stimulated substantial new investigation in our understanding of severe pediatric TBI and its treatment. Hopefully, by the time that these guidelines are revised, this flurry of research will continue to produce new data upon which stronger recommendations can be made.