(Published by P Andrews and colleagues on behalf of the Eurotherm3235 Trial collaborators, in New England Journal of Medicine October 2015)
Written by Ewan Ross (Research Assistant, Imperial College London)
Edited by Lucia M Li
A major new trial has found that patients with Traumatic Brain Injury (TBI) treated with hypothermia on the Intensive Care Unit seemed to have worse overall outcomes.
A clinical trial testing whether or not patients with Traumatic Brain Injury (TBI) should be put in to a state of hypothermia was recently published. Hypothermia is when the body is cooled to below normal body temperature (which is around 37°C). In the Eurotherm3235 trial, patients were cooled to between 32°C and 35°C. Intensive Care Units across the world have used this practice for many years with patients in the very early stages after injuries. However, the published evidence hasn’t been very clear about the benefits, with some studies showing benefits, while others have shown that the practice leads to worse outcomes.
The theory is that cooling patients reduces the swelling and inflammation that can occur as a result of TBI. One result of these processes is raised pressure inside the skull, which has been associated with problems with blood flow and structural damage. Hypothermia is just one of a number of treatments that can be used to try to combat this raised pressure and it does indeed control the pressure quite well, in this study and in others before. However, beyond controlling this pressure, it is much more important to find out if people who receive hypothermia treatment have a better recovery than those that don’t.
This was a large trial led by a team at the University of Edinburgh but included patients in many centres across the world (47 Hospitals in 18 countries), since 2009. The team intended to follow the progress of 600 patients, but the trial was stopped early by an independent committee because they were concerned that the cooling treatment was leading to worse outcomes. In the end, only 376 patients who had suffered a head injury within 10 days and had a raised pressure in the skull requiring treatment. Exactly half of these patients, 188, were randomly allocated to be given hypothermia for at least 48 hours, while the other half of the patients were allocated to receive other treatments commonly used to control the pressure as part of the best practice guidelines, but they did not receive therapeutic hypothermia.
Patients from both groups were contacted again in 6 months and their outcomes were scored according to an 8-point scale called the ‘Glasgow Outcome Scale – Extended’ or GOS-E. Favourable outcome is classed as scoring 5 or above. Scoring was done by investigators who were ‘blind’ - they didn’t know which patients had had hypothermia and which did not. The results show that both treatment groups were able to control the pressure in the skull, though the non-hypothermia group patients more often had to use additional measures to reduce the pressure - 54% of them compared to 43.8% in the hypothermia group. With hypothermia, 25.7% patients had a favourable outcome at 6 months, but without hypothermia this was 36.5%. This difference was statistically significant, meaning that this result was unlikely to have happened just by chance. Patients in the hypothermia group were also more likely to die, and they also experienced more than 3 times the number of ‘Serious Adverse Events’, for example, serious bleeding.
It isn’t clear that the worse outcomes after receiving hypothermia were purely a result of the hypothermia, or perhaps because of the benefits of the alternative treatments given instead. There is too little evidence on other potential treatments to be able to say and the researchers note that this will be important to understand in future. It is important to remember that this is only one trial out of several that have tackled this question. Currently, there is no single study which can conclusively answer whether patients with TBI should be given hypothermia or not. This result does not add encouragement to using therapeutic hypothermia for patients with TBI.