Therapeutic Hypothermia Compared with Normothermia in Adults with Traumatic Brain Injury; Functional Outcome, Mortality, and Adverse Effects: A Systematic Review and Meta-Analysis

Background The main focus of traumatic brain injury (TBI) management is prevention of secondary injury. Therapeutic hypothermia (TH), the induction of a targeted low core body temperature, has been explored as a potential neuroprotectant in TBI. The aim of this article is to synthesize the available clinical data comparing the use of TH with the use of normothermia in TBI. Methods A systematic search was conducted through MEDLINE, EMBASE, and Cochrane Central Register of Controlled Trials for randomized clinical trials including one or more outcome of interest associated with TH use in TBI. Independent reviewers evaluated quality of the studies and extracted data on patients with TBI undergoing TH treatment compared with those undergoing normothermia treatment. Pooled estimates, confidence intervals (CIs), and risk ratios (RRs) or odds ratios were calculated for all outcomes. Results A total of 3,909 patients from 32 studies were eligible for analysis. Pooled analysis revealed a significant benefit of TH on mortality and functional outcome (RR 0.81, 95% CI 0.68–0.96, I 2 = 41%; and RR 0.77; 95% CI 0.67–0.88, I 2 = 68%, respectively). However, subgroup analysis based on risk of bias showed that only studies with a high risk of bias maintained this benefit. When divided by cooling method, reduced poor functional outcome was seen in the systemic surface cooling and cranial cooling groups (RR 0.68, 95% CI 0.59–0.79, I 2 = 35%; and RR 0.44, 95% CI 0.29–0.67, I 2 = 0%), and no difference was seen for the systemic intravenous or gastric cooling group. Reduced mortality was only seen in the systemic surface cooling group (RR 0.63, 95% CI 0.53–0.75, I 2 = 0%,); however, this group had mostly high risk of bias studies. TH had an increased rate of pneumonia (RR 1.24, 95% CI 1.10–1.40, I 2 = 32%), coagulation abnormalities (RR 1.63, 95% CI 1.09–2.44, I 2 = 55%), and cardiac arrhythmias (RR 1.78, 95% CI 1.05–3.01, I 2 = 21%). Once separated by low and high risk of bias, we saw no difference in these complications in the groups with low risk of bias. Overall quality of the evidence was moderate for mortality, functional outcome, and pneumonia and was low for coagulation abnormalities and cardiac arrhythmias. Conclusions With the addition of several recent randomized clinical trials and a thorough quality assessment, we have provided an updated systematic review and meta-analysis that concludes that TH does not show any benefit over normothermia in terms of morta