Twelve RCTs with a total of 1,069 participants were included.
Two studies were graded as high, three as moderate and five as low methodological quality.
Therapeutic hypothermia was associated with a significant reduction in the risk of death compared with normothermia (RR 0.81, 95% CI: 0.69, 0.98). The results were statistically homogeneous across the included studies (Q=8.12, P value not provided). Subgroup analyses found that hypothermia longer than 48 hours was associated with a statistically significant reduction in the risk of death (RR 0.70, 95% CI: 0.56, 0.87). However, none of the other subgroup analyses were statistically significant. Analyses exploring methodological quality did not detect any evidence of bias in the estimation of treatment effect for mortality. The reduction in the risk of death remained significant after the largest clinical trial was removed (RR 0.75, 95% CI: 0.62, 0.91). A funnel plot did not find any evidence of publication bias.
Therapeutic hypothermia was associated with a significant reduction in the risk of poor neurological outcome (RR 0.78, 95% CI: 0.63, 0.98), but this effect was statistically heterogeneous (Q=16.05, P value not given). There remained a reduction in risk when hypothermia was induced for more than 48 hours (RR 0.65, 95% CI: 0.48, 0.89) or for 24 hours (RR 0.61, 95% CI: 0.39, 0.97), but not for 48 hours. Studies that included both induced hypothermia at 32 to 33 degrees C and rewarming within 24 hours after discontinuation of hypothermia were associated with a significant reduction in the risk of poor neurological outcome (RR 0.61, 95% CI: 0.45, 0.83 and RR 0.79, 95% CI: 0.63, 0.98, respectively). No other subgroup analyses had a statistically significant effect on neurological outcome. Analyses according to methodological quality did not show any evidence of bias in the treatment effects.