Approaches and Decisions for Acute Pediatric TBI Trial

Traumatic brain injury (TBI) is the leading cause of death in children in the US. According to the CDC, 7440 children died of TBI in 2005, but this likely underestimates the full burden of the disease. Based on the current best estimates for severe pediatric TBI (20% mortality, 50.6% unfavorable 6 mo outcome, mean age 9 y), each year 37,200 children suffer a severe TBI with up to 1.3 million life-years potentially adversely affected. Incremental improvement in outcomes could make enormous differences for the health of children, but such advances have remained elusive. Dozens of injury mechanisms have been identified after experimental TBI, yet no mitigating treatments have been translated into clinical practice. Randomized controlled trials (RCTs) of therapies, from steroids to novel pharmaceutical agents to hypothermia, have failed for adult and pediatric TBI victims. Single-center experiences have contributed to understanding, yet these largely remain insufficiently powerful to change practice. Recently, evidenced-based guidelines for 15 aspects of pediatric TBI were published that provide no level I and only 4 Level II recommendations - with such recommendations indicating therapies that "must be done" or "should be considered" based on the literature, respectively. Disappointingly, 3 of these recommendations advised against specific interventions (hypothermia, steroids and immune-enhanced diets) - emphasizing the uncertainty of the effectiveness of many commonly used therapies that leads to wide variations in clinical practice. Unsurprisingly, significant variations of clinical outcomes and basic treatment strategies for TBI have been observed. The IMPACT study merged data from over 9000 adults with TBI from 11 trials and demonstrated significant variations in outcomes from clinical sites. Similar variations in outcomes in children with TBI can be found using various administrative databases, with mortality rates varying between 12.2% - 34.4% in 11 US states. There are also variations in strategies within an international consortium and a recently completed RCT - with marked variations in strategies for first-line intracranial hypertension treatments, prevention of common secondary insults and metabolic support after pediatric TBI. The paucity of data to create robust guidelines, the failure of RCTs that tested a wide-variety of putative mechanisms and variations in outcomes and in clinical practices argues that the current understanding of contemporary therapies is inadequate. Because neither retrospective analyses from available databases nor self-reported variations in practices can determine optimal therapeutic strategies for these contemporaneous strategies, a new approach is urgently needed. ADAPT is a large, prospective, observational cohort study using an international consortium including sites from the US, EU and UK. Children with severe TBI \[Glasgow coma scale (GCS) score ≤ 8 with intracranial pressure (ICP) monitoring, n = 1000, \>32 sites\] will be studied. The local standard of care at each site will be used and extensive data collection over the first 7 days after TBI will be performed to interrogate the effectiveness of strategies for intracranial hypertension, mitigation of specific secondary insults and metabolism. Several statistical approaches, often used in comparative effectiveness research (CER) to control for measured confounding effects, will test the following aims: Specific Aim 1: Compare the effectiveness of first-line intracranial hypertension strategies on outcome. Intracranial hypertension management is a mainstay of TBI care yet evidence for utilization the first-line therapies of cerebrospinal fluid (CSF) diversion and use of hyperosmolar solutions, is incomplete. Aim 1a: Determine the effect of CSF diversion strategies (continuous drainage, intermittent drainage and none) on outcome. Aim 1b: Determine the effect of hyperosmolar therapies (hypertonic saline, mannitol) on outcome. Specific Aim 2: Compare the effectiveness of strategies that mitigate specific secondary insults on outcome. Prophylactic hyperventilation (HV) and hypoxia may worsen outcome after TBI but have been inadequately studied. Aim 2a: Determine the effect of prophylactic HV (CO2 \< 30 mm Hg) on outcome. Aim 2b: Determine the effect of hypoxia detection with brain tissue oxygen monitoring (PbO2) on outcome.