Traumatic brain injury (TBI) is an important critical care problem with wide-ranging impact both for the individual patient and for society as a whole. Over the past decade, research has focused on identifying the cellular mechanisms involved in excitotoxic cell death. This information was translated into a number of pharmacologic trials each designed to attack a single mechanism of action. Unfortunately, this extensive experimental work did not translate into a new clinical treatment and trials have by-and-large been negative. This has resulted in a deficit of level-one evidence to support many, if not all, clinical treatment guidelines. At the same time, basic and clinical pathophysiologic studies have shed new light into the alteration in brain metabolism after brain injury. This emerging information is fundamental to understanding brain injury and may provide useful principles for critical care.

Does Brain Ischemia After TBI Occur?

Most important among the main questions surrounding the critical care of brain trauma is the question of whether brain ischemia occurs after TBI. The answer to this question is yes and no. Numerous clinical studies have documented ischemia early after traumatic brain injury.2,6,10,13 However, the overall prevalence of ischemia after the initial 24 hours has been difficult to document despite persistent markers of disturbed brain metabolism seen in microdialysis monitoring.25 Positron emission tomography (PET) studies in TBI patients have recently been performed to elucidate whether ischemia is occurring.7 In the Diringer study, neither regional nor global brain ischemia was found under typical intensive care unit (ICU) conditions of normal ventilation or upon provocative hyperventilation. Similarly, Coles used oxidative PET studies in TBI patients and found that at baseline CO2 of 30-34 mm Hg, the mean ischemic brain volume was 67 cc.5 This translates into approximately ischemia being present in 6% of the brain.

At the same time, significant changes in glucose metabolism have been noted on PET imaging,24 with a prominent finding being a reduction in glucose metabolism in the grey matter28 and an abnormal ratio of glucose metabolism to oxidative metabolism throughout the brain.3 Reduction in brainstem glucose metabolism correlates with the depth of coma.11 In contrast to these studies, global ischemia as evidenced by jugular venous desaturation and reduction in brain tissue oxygen tension has been found.9 Ultra-early monitoring of jugular venous saturation indicates that ischemia can occur in the emergency room or early hours of injury.27 Desaturation was usually not coupled with measures of oxidative capacity but nonetheless indicate a relative shortage of oxygen supply compared with demand. Comparatively, using state-of-the-art PET imaging, it appears that in the ICU, ischemia can occur but is not common beyond the initial 12 hours after injury, but brain metabolism of glucose and oxygen is altered in a complex fashion that is yet to be fully explained.

Is Monitoring Brain Metabolism Important?

Experimental models demonstrate that brain oxygen metabolism is reduced due to dysfunction of mitochrondria,14 and human studies have confirmed that reductions in oxidative metabolism are an important prognostic factor.8 While PET imaging in TBI has been criticized for its inability to serial measure brain metabolism and hence brain monitors such as brain tissue oxygen probes, jugular venous oximetry and cerebral microdialysis are now starting to be used to determine moment-to-moment changes in oxygen metabolism after TBI. Brain tissue oxygen tension has been demonstrated to vary with changes in cerebral perfusion pressure,16 increased inspired oxygen,18 temporary vessel occlusion during surgery,29 excessive hyperventilation,12 cerebral vasospasm,4 and terminal increases in intracranial pressure with herniation.24

Brain tissue oxygen tension measures the balance between oxygen supply and demand rather than the oxygen content. As such, it is not clearly a measure of oxidative metabolism. Using the combination of PET and brain tissue oxygen tension monitoring,17 has demonstrated a gradient between the partial pressure of oxygen delivered to the cerebral circulation and that found in the brain, thus concluding that a diffusion barrier to oxygen entry into the damaged brain exists. This is a novel concept that suggests that supraphysiological levels of brain tissue oxygen may be needed to supply adequate oxygen delivery to the brain. Indeed, clinical studies have demonstrated lower than expected brain tissue oxygen tension and subnormal response of the brain to supplemental inspired oxygen.

Does Altering Brain Metabolism Improve Outcome?

Coupled with evidence of disturbed brain metabolism found on PET, fundamental clinical observational studies indicate that brain metabolism is altered after TBI and may be modifiable. Modification of brain metabolism may result in new therapeutic possibilities. Strategies to increase oxygen delivery to the brain, such as normobaric hyperoxia and hyperbaric hyperoxia have not been extensively studied.15,18,21,23 While the Rockwold study23 is showing that hyperbaric hyperoxia increased cerebral blood flow after TBI, there was no difference in mortality or overall outcome. In addition, short-term physiological studies have demonstrated that hyperoxia leads to inconsistent changes in cerebral hemodynamics and metabolism when delivered under basal conditions. Experimental administration of cyclosporin A increases the ability of the brain mitochondria to utilize oxygen,1,14 decreases the amount of axonal injury,19 and may improve outcome.22

Increasing oxidative metabolism is a potential new treatment for brain injury with clinical trials currently being planned. In addition to oxygen, brain metabolism may be altered by the delivery of non-traditional or alternative fuels to the brain. For example, ketones may be used as a source of fuel in hypoglycemic states. Recently, Prins reported that supplemental administration of three hydroxy-butyrate (beta-HB) after experimental brain injury resulted in increased uptake of beta-HB and preservation of ATP levels.20 Taken together, these studies suggest that upregulation of brain metabolism is possible but may require alternative fuels or supratherapeutic levels of conventional fuels to do so.

In summary, alterations in brain metabolism are dynamic after brain injury. Early ischemia does occur, and monitoring for ischemia may be useful. However, the more long lasting alterations in brain metabolism may be based on factors other than ischemia. An emerging concept is that alternative metabolic treatments may prove to be useful in the treatment of brain injury and preserve vulnerable brain cells from dying after injury.