T raumatic brain injury (TBI) is a leading cause of death and disability in children (1). Elevated intracranial pressure (ICP) and reduced cerebral perfusion pressure (CPP) are both associated with worse outcome (2, 3).Even though optimum pediatric ICP and CPP treatment thresholds remain to be determined, an ICP Ͼ20 mm Hg is generally accepted as a threshold for treatment (4), whereas a CPP between 40 and 65 mm Hg is currently advised as an age-related continuum treatment threshold (5).Current treatment for TBI aims to minimize secondary brain injury and increase survival and neurologic outcome by reducing ICP concurrently with max-imizing CPP. For decades, routine treatment of severe pediatric TBI has included maintaining the head in a neutral position with elevation of the head of the bed (HOB) to 30 degrees. The effect of HOB elevation on ICP is attributed to gravity-induced changes in cerebral blood and cerebrospinal fluid volume and to maximizing cerebral venous return (6). A number of clinical trials in adults have quantified the effect of elevating the HOB on ICP, CPP, and, to a lesser extent, cerebral oxygenation (7-17).However, because children differ from adults in physiologic and anatomical characteristics as well as their response to severe TBI, pediatric-specific data are needed (18). Despite years of clinical experience with HOB elevation, little is known of its efficacy in pediatric TBI. One study showed that the optimal (i.e., lowest ICP with highest CPP) HOB elevation in children was actually Ͻ30 degrees in 29% (4 of 14) of the measurements (19), whereas another Objectives: To determine the effect of and dynamic interaction between head elevation on intracranial pressure and cerebral perfusion pressure in severe pediatric traumatic head injury.Design: Prospective, randomized, interventional cohort study. Setting: Two tertiary pediatric critical care referral units. Patients: Ten children admitted with severe traumatic brain injury defined as Glasgow Coma Score <8 necessitating intracranial pressure monitoring (10 yrs ؎ 5 SD; range 2-16 yrs).
Interventions: Head elevation was randomly increased or decreased between 0 and 40 degrees from baseline level (30 degrees) in increments or decrements of 10 degrees.Measurements and Main Results: Intracranial pressure and arterial blood pressure were continuously recorded in combination with time-stamped clinical notations. Data were available for analysis in eight subjects (seven males and one female; mean age, 10 yrs ؎ SD 5; range, 2-16 yrs) during 18 protocol sessions. This resulted in a total of 66 head-of-the-bed challenges. To compare results for a given change in head-of-the-bed elevation across age, we transformed head-of-the-bed angle to change in height (cm) at the level of Monro's foramen. An increase in head elevation of 10 cm resulted in an average change in intracranial pressure of ؊3.9 mm Hg (SD ؎3.2 mm Hg; p < .001), whereas cerebral perfusion pressure remained unchanged (0.1 ؎ 5.6 mm Hg; p ؍ .957). Individual subjects showed marked variabi...