Vasopressors During Cardiopulmonary Resuscitation. A Network Meta-Analysis of Randomized Trials.
Review
Overview
abstract
OBJECTIVES: Several randomized controlled trials have compared adrenaline (epinephrine) with alternative therapies in patients with cardiac arrest with conflicting results. Recent observational studies suggest that adrenaline might increase return of spontaneous circulation but worsen neurologic outcome. We systematically compared all the vasopressors tested in randomized controlled trials in adult cardiac arrest patients in order to identify the treatment associated with the highest rate of return of spontaneous circulation, survival, and good neurologic outcome. DESIGN: Network meta-analysis. PATIENTS: Adult patients undergoing cardiopulmonary resuscitation. INTERVENTIONS: PubMed, Embase, BioMed Central, and the Cochrane Central register were searched (up to April 1, 2017). We included all the randomized controlled trials comparing a vasopressor with any other therapy. A network meta-analysis with a frequentist approach was performed to identify the treatment associated with the highest likelihood of survival. MEASUREMENTS AND MAIN RESULTS: Twenty-eight studies randomizing 14,848 patients in 12 treatment groups were included. Only a combined treatment with adrenaline, vasopressin, and methylprednisolone was associated with increased likelihood of return of spontaneous circulation and survival with a good neurologic outcome compared with several other comparators, including adrenaline. Adrenaline alone was not associated with any significant difference in mortality and good neurologic outcome compared with any other comparator. CONCLUSIONS: In randomized controlled trials assessing vasopressors in adults with cardiac arrest, only a combination of adrenaline, vasopressin, and methylprednisolone was associated with improved survival with a good neurologic outcome compared with any other drug or placebo, particularly in in-hospital cardiac arrest. There was no significant randomized evidence to support neither discourage the use of adrenaline during cardiac arrest.
