This proposed project, the National AcuteBrain Injury Study: Hypothermia II (NABIS:H In will be a multi-center,prospective, randomized Phase III clinical trial in which standard managementat hypothermia (33 degrees C) for 48 hours is tested against standard management atnormothermia in patients with severe brain injury (GCS<8), age 16-45 years, andwith admission temperature less than or equal to 35 degrees C . The primary outcome measure will be the dichotomized Glasgow Outcome Scale at 6 months after injury (GoodRecovery/Moderate Disability vs. Severe Disability/Vegetative/Dead). The samplesize of 220 patients will detect an absolute difference of 17% in thepercentage of poor outcomes in the two groups at a power of 80%. The selectionof this population was based on findings from the original National Acute BrainInjury Study: Hypothermia (NABIS:H I). In that study, with a sample size of392, hypothermia was induced beginning <6 hours after injury, reaching 33 degrees C by8.4 + or - 3 hours after injury, and maintaining hypothermia for 48 hours. Thenull hypothesis was confirmed, with no difference in the percentage of patientsmaking a poor recovery (Severe Disability, Vegetative, Dead) at 6 months afterinjury (Hypothermia 57%, Normothermia 57%, NS). In 52 patients age >45 years,there were more poor outcomes in the hypothermia group (Hypothermia 89%,Normothermia 69%, p=0.08) due to increased medical complications. However, in81 patients (22% of 366 patients with complete data) who were age 16-45 yearswith admission temperature less than or equal to 35 C, maintenance ofhypothermia was associated with a marked decrease in the percentage of pooroutcomes (Hypothermia, 52%, Normothermia 76%, p=0.02) with no difference incomplication rates. The effect was found in all of the 4 high-enrollmentcenters, and there were no confounding variables that could have explained thedifference in outcomes. The entire treatment effect in this subgroup was toshift patients from Severe Disability (poor outcome) to Moderate Disability(good outcome). We believe that the finding is related to a very shorttreatment window for hypothermia induction. The proposed study prospectivelytests whether maintenance of hypothermia present on admission results in betteroutcome than current management.
