Brain oxygen tension in severe head injury.

OBJECTIVE Ensuring adequate cerebral oxygenation and perfusion is of fundamental importance in the treatment of patients with acute cerebral disorders. Online continuous monitoring of brain oxygenation is possible with a parenchymal microelectrode that measures local brain oxygen tension. The ultimate question is whether therapeutic approaches can be targeted on the basis of such monitoring. Before this question can be addressed, the technique requires validation in the clinical setting. The frequency of occurrence of low values and its relation to outcome need to be established. METHODS One hundred one comatose head-injured patients (Glasgow Coma Scale score < or = 8) were studied. Local brain oxygen tension probes were inserted in an undamaged part of the frontal region. Patients were treated in conformance with the European Brain Injury Consortium guidelines. Outcome at 6 months was determined by Glasgow Outcome Scale score. RESULTS Early brain tissue hypoxia was frequently observed, despite aggressive treatment for intracranial pressure and cerebral perfusion pressure. Values lower than 15 mm Hg, for a duration longer than 30 minutes, were observed in 57 patients. Values lower than 10 mm Hg in 42 patients, and lower than 5 mm Hg in 22 patients, were observed during the first 24 hours. Depth and duration of tissue hypoxia were related to outcome and proved to be an independent predictor of unfavorable outcome and death. CONCLUSION Monitoring the partial oxygen pressure of local brain tissue is a safe and reliable method for regulating cerebral oxygenation. Because brain tissue hypoxia occurs frequently and is significantly related to poor outcome, future efforts should be aimed at the treatment of brain tissue hypoxia. The effects of such brain hypoxia-targeted treatment need to be established in a multicenter study.