failure of large, multicenter phase II clinical trial to produce an improvement in outcome in TBI patients (Marklund et al., 2006) and raised concern about the ability to translate from bench to bedside. (For detailed discussion of why therapies shown to be efficacious in animals failed to translate to human TBI, see recent reviews: Statler et al., 2001; Cernak, 2005; Morales et al., 2005; Thompson et al., 2005; Marklund et al., 2006; Kokiko and Hamm, 2007.)
In phase 3, secondary events—such as hypoxia, hypotension, ischemia, increased intracranial pressure (ICP) and brain swelling, and metabolic failure—perturb brain function further and augment cell injury (Statler et al., 2001; Graham et al., 2002; Marklund et al., 2006). As in phase 2, those secondary insults occur over an extended period of time postinjury and so are suitable targets for therapeutic intervention.
In human TBI, hypoxia and hypotension occur in one-third of patients with TBI (Statler et al., 2001; Morganti-Kossmann et al., 2007). Hypotension is a principal predictor of poor outcome after TBI: a single episode of systolic blood pressure of less than 90 mm Hg is associated with a 150% increase in mortality (Statler et al., 2001). In a recent prospective multicenter study, about 40% of patients with TBI sustained a secondary insult before reaching the hospital (Chi et al., 2006). Of those patients, 65% had episodes of hypoxia, 11% were hypotensive, and 24% showed a combination of those two secondary insults. Episodes of hypoxia were particularly notable in that they were independently predictive of death. Moreover, the episodes occurred despite aggressive medical management, including endotracheal intubation.
Posttraumatic ischemia has been demonstrated in both animal models and humans after TBI and is associated with poor neurologic outcome. The factors mediating posttraumatic ischemia include mechanical damage to blood vessels, hypotension in concert with autoregulatory failure, and lack of available endogenous vasomodulators, such as nitric oxide and prostaglandins (Werner and Engelhard, 2007).
Injury severity is a determinant of the magnitude of increase in ICP (Thompson et al., 2005). Increased ICP is usually a consequence of brain swelling that occurs in response to disruption of the blood–brain barrier and the later evolution of cerebral edema (Pasternak and Lanier, 2007). Edema has a profound effect on the brain because of the restrictions imposed by the bony calvarium, which limit further expansion of the edematous brain and lead to an increase in ICP.
Metabolic failure arises from mitochondrial dysfunction, reduced availability of the nicotinic coenzyme pool, and intramitochondrial overload (Werner and Engelhard, 2007). Outcome is worsened in accordance with the degree of metabolic failure (Werner and Engelhard, 2007). An alternative finding has been hypermetabolism of glucose, which reflects an uncoupling of cerebral blood flow and metabolism. It is a scenario whereby massive transmembrane ionic fluxes and neuroexcitation are not adequately met by increases in cerebral blood flow (Werner and Engelhard, 2007).
Phase 4, representing recovery and functional outcome, is influenced by primary and secondary injury responses and by wound-healing events, including phagocytic removal of cellular debris, glial scar formation, and plastic changes in neural networks. Recovery of function after brain injury is described in three phases (Wieloch and Nikolich, 2006). Phase 1 involves reversal of inhibition of function and initiation of cell repair, phase 2 entails a change in the properties of existing pathways, and phase 3 involves the formation of new connections