OBJECTIVES:
The objective of this systematic review was to evaluate the impact of intraarrest corticosteroids on neurologic outcomes and mortality in patients with cardiac arrest.
DATA SOURCES:
We conducted a systematic search using the Cochrane Central Register of Controlled Trials, EMBASE, and MEDLINE databases.
STUDY SELECTION:
We included all randomized controlled trials and comparative observational studies. We excluded single arm studies, case reports/series, narrative reviews, and studies irrelevant to the focus of this article.
DATA EXTRACTION:
Two reviewers independently assessed trial eligibility. Data were collected for the following outcomes: primary outcomes included good neurologic outcome, survival to hospital discharge, and survival at greater than or equal to 1 year. Secondary outcomes included incidence of return of spontaneous circulation, ICU and hospital length of stay, duration of vasopressor and inotropic treatment, and blood pressure during cardiopulmonary resuscitation and after return of spontaneous circulation.
DATA SYNTHESIS:
The pooled estimates from randomized controlled trials for the following subgroups were analyzed using random-effects models: 1) patients with in-hospital cardiac arrest who received vasopressin, steroids, and epinephrine; 2) patients with in-hospital cardiac arrest who used corticosteroids only (i.e., no vasopressin); and 3) patients with out-of-hospital cardiac arrest who used corticosteroids only. Results included an increase in good neurologic outcomes (relative risk, 2.84; 95% CI, 1.36–5.94) and survival to hospital discharge (relative risk, 2.58; 95% CI, 1.36–4.91) in in-hospital cardiac arrest patients receiving vasopressin, steroids, and epinephrine followed by corticosteroids for postresuscitation shock. This was further supported by an increase in return of spontaneous circulation (relative risk, 1.35; 95% CI, 1.12–1.64) and hemodynamics in this population. There was no benefit observed in in-hospital cardiac arrest or out-of-hospital cardiac arrest patients receiving corticosteroids alone.
CONCLUSIONS:
Our study found that there are limited high-quality data to analyze the association between corticosteroids and reducing mortality in cardiac arrest, but the available data do support future randomized controlled trials. We did find that corticosteroids given as part of a vasopressin, steroids, and epinephrine regimen in in-hospital cardiac arrest patients and for postresuscitation shock did improve neurologic outcomes, survival to hospital discharge, and surrogate outcomes that include return of spontaneous circulation and hemodynamics. We found no benefit in in-hospital cardiac arrest or out-of-hospital cardiac arrest patients receiving corticosteroids only; however, a difference cannot be ruled out due to imprecision and lack of available data.