Therapeutic hypothermia for head injury.

backgroundMild to moderate induced hypothermia has been used in the treatment of head injury for over 50 years, although few randomised controlled trials have been performed. Recent encouraging results from small, single-centre trials and consistent findings of a cerebral protection effect of cooling in laboratory models of global ischaemia has led to a renewed interest in the area.

objectivesTo determine whether the use of mild therapeutic hypothermia in the treatment of moderate and severe head injury improves short-term control of intracranial pressure (ICP) and long-term functional outcome. SEARCH STRATEGY: Electronic searches of the Injuries Group trial registry and EMBASE for any relevant randomised trials, supplemented by hand searching of conference proceedings and reference lists of relevant articles. SELECTION CRITERIA: All randomised controlled trials of mild hypothermia versus control (open or normothermia) in the treatment of patients with any closed head injury requiring hospitalisation. Mild hypothermia was defined as local or systemic cooling to a target temperature of at most 34-35 degrees Celsius for a period of at least 12 hours. Outcome was all-cause mortality and death or severe disability at the end of the scheduled follow-up period. All trials were assessed by two reviewers, and included or excluded on a consensus basis. DATA COLLECTION AND ANALYSIS: Eleven potential trials of therapeutic hypothermia for head injury were found, of which two are ongoing and one is awaiting assessment. The eight remaining trials were included in the systematic review. Data on death, GOS score at final follow-up, complications and ICP were sought and extracted, either from published material or by contact with the investigators. Mantel-Haenzel odds ratios and 95% confidence intervals were calculated for death and death and severe disability for each trial on an intention-to-treat basis. No quantitative synthesis of data on either complications or ICP was attempted. Trials of immediate and deferred hypothermia were analysed separately. MAIN

resultsActive immediate hypothermic treatment was associated with a 33% non-significant (p=0.16) reduction in the odds of death at the end of treatment or final follow-up, (OR 0.67, 95% confidence interval 0.38 to 1.17), and a 61% reduction (p=0.004) in the odds of being dead or severely disabled, (OR 0.39, 95% confidence interval 0.20 to 0.74). Similar effect sizes were found for delayed hypothermia. These results are, however, based on a few small trials each of less than 100 patients. A multi-centre trials of hypothermia versus control in 392 patients will be reporting results in 1999, providing substantially more evidence than is currently available. REVIEWER'S

conclusionsAlthough this review would suggest a strong positive effect of therapeutic hypothermia, the results are based on several small trials carried out in single, specialist centres. The results of a large multi-centre trial are expected in 1999 and will more than treble the available evidence. Until these results have been released, it would be inappropriate to make any short-term recommendations for clinical practice or research.