Lewis · The Cochrane database of systematic reviews 2017 · systematic review · n=37 studies (3110 participants)

Hypothermia for traumatic brain injury.

Cited 94 times in the scientific literature.

Level 1 - systematic review of randomized trials

Systematic review of randomized controlled trials

PubMed 28933514 · doi:10.1002/14651858.CD001048.pub5 · record verified 2026-08-26

What was done

This Cochrane systematic review searched multiple databases through March 2016 (updated June 2017) for randomized controlled trials comparing mild hypothermia (≤35 °C for ≥12 hours, maintained for ≥24 hours) with normothermia (36.5–38 °C) in hospitalized adults and children with closed traumatic brain injury. The authors evaluated mortality, unfavorable functional outcome (Glasgow Outcome Scale: death, vegetative state, or severe disability), and pneumonia, grading evidence certainty using GRADE.

What was found

Across 37 trials (3,110 participants), 33 reported mortality, 31 reported unfavorable outcomes, and 14 reported pneumonia. Results showed severe inconsistency with conflicting directions of effect across studies, which persisted after subgrouping by hypothermia duration and follow-up length. Consequently, no pooled numerical meta-analysis was performed. Evidence quality was rated very low for mortality and unfavorable outcomes, and low for pneumonia.

Why it matters

Despite numerous trials, there is no high-quality evidence that therapeutic hypothermia improves survival or functional outcomes in traumatic brain injury.

Limits

Included studies were generally poorly reported, preventing adequate risk-of-bias assessment. Pronounced clinical and methodological heterogeneity precluded quantitative pooling, and potential sources of bias or inconsistency were not evaluated through detailed subgroup or sensitivity analyses.

Cited by