Projects per year
Background: Brain injuries resulting from trauma and stroke are common and costly. Cooling therapy may reduce damage and potentially improve outcome. Head cooling targets the site of injury and may have fewer side effects than systemic cooling, but there has been no systematic review and the evidence base is unclear. Objective: To assess the effect of non-invasive head cooling after traumatic brain injury (TBI) and stroke on intracranial and/or core body temperature, functional outcome and mortality, determine adverse effects and evaluate cost-effectiveness. Review methods: Search strategy Major international databases [including MEDLINE, EMBASE, Cumulative Index to Nursing and Allied Health Literature, Web of Science, the British Library's Electronic Table of Contents (Zetoc)], The Cochrane Library, trial registers, country-specific databases (including China, Japan), Google Scholar, hypothermia conference reports and reference lists of papers were searched with no publication or language restrictions. The searches were conducted from March 2010 to April 2011, with no back date restriction. Selection criteria For formal analysis of effect of head cooling on functional outcome and mortality: randomised controlled trials (RCTs) of non-invasive head cooling in TBI or stroke in adults (aged ≥ 18 years). RCT prespecified in protocol to include adequate randomisation and blinded outcome assessment. For assessment of effect on temperature and adverse effects of cooling methods/devices: studies of any type in TBI, stroke, cardiac arrest and neonatal hypoxic-ischaemic encephalopathy (adverse effects only). Data collection and analysis A study assessment and data collection form was developed and piloted. Data on functional outcome, mortality, temperature change and adverse effects of devices were sought and extracted. Two authors independently assessed RCTs for quality using the Cochrane Renal Group checklist. Results: Out of 46 head-cooling studies in TBI and stroke, there were no RCTs of suitable quality for formal outcome analysis. Twelve studies had useable data on intracranial and core body temperature. These included 99 patients who were cooled after TBI or stroke and 198 patients cooled after cardiac arrest. The data were too heterogeneous for a single summary measure of effect (many studies had no measure of spread) and are therefore presented descriptively. The most effective techniques for which there were adequate data (nasal coolant and liquid cooling helmets) could reduce intracranial temperature by ≥ 1 °C in 1 hour. The main device-related adverse effects were localised skin problems, which were generally mild and self-limiting. There were no suitable data for economic modelling, but an exploratory model of possible treatment effects and cost-effectiveness of head cooling in TBI was created using local patient data. Limitations: We conducted extensive and sensitive searches but found no good-quality RCTs of the effect of head cooling on functional outcome that met the review inclusion criteria. Most trials were small and/or of low methodological quality. However, if the trial reports did not reflect the true quality of the research, there may be some excluded trials that should have been included. Temperature data were often poorly reported which made it difficult to assess the effect of head cooling on temperature. Conclusions: Whether head cooling improves functional outcome or has benefits and fewer side effects compared with systemic cooling or no cooling could not be established. Some methods of head cooling can reduce intracranial temperature, which is an important first step in determining effectiveness, but there is insufficient evidence to recommend its use outside of research trials. The principal recommendations for research are that active cooling devices show the most promise for further investigation and more robust proof of concept of intracranial and core body temperature reduction with head cooling is required, clearly showing whether temperature has changed and by how much.