Does therapeutic hypothermia benefit adult cardiac arrest patients presenting with non-shockable initial rhythms?: A systematic review and meta-analysis of randomized and non-randomized studies

Background: The benefit of therapeutic hypothermia (TH) for comatose adult patients with return of spontaneous circulation after cardiac arrest (CA) with non-shockable initial rhythms is uncertain. We evaluated whether TH reduces mortality and improves neurological outcome in comatose adults resuscitated from non-shockable CA.

Methods: We searched PubMed, EMBASE, CENTRAL, and BIOSIS through March 2010, to identify studies using TH after non-shockable CA. Randomized and non-randomized studies (RS and NRS) comparing survival or neurological outcome in TH and standard care or normothermia were selected. We corresponded with authors to clarify data missing from published articles. Individual and pooled statistics were calculated as risk ratios (RRs) with 95% confidence interval (CI). Both fixed- and random-effects models were used for both meta-analyses.

Findings: Two RS and twelve NRS were included in the meta-analysis and separately analyzed. The pooled RR for 6-month mortality of two RS was 0.85 (95% CI 0.65-1.11). The pooled RR for in-hospital mortality for 10 NRS was 0.84 (95% CI 0.78-0.92) and for poor neurological outcome on discharge was 0.95 (95% CI 0.90-1.01) in random-effects model. In subgroup analysis for the NRS with out-of-hospital CA, the pooled RR for in-hospital mortality was 0.86 (95% CI 0.76-0.99) and for the poor neurological outcome on discharge was 0.96 (95% CI 0.90-1.02). For the prospective NRS, the pooled RR for in-hospital mortality was 0.76 (95% CI 0.65-0.89) and for the poor neurological outcome on discharge was 0.96 (95% CI 0.90-1.02). Most of studies had substantial risks of bias and overall quality of evidence was very low.

Interpretation: TH is associated with reduced in-hospital mortality for adults patients resuscitated from non-shockable CA. However, most of the studies had substantial risks of bias and quality of evidence was very low. Further high quality randomized clinical trials would confirm the actual benefit of TH in this population.