Vasopressin for cardiac arrest: meta-analysis of randomized controlled trials.

BACKGROUND Prior meta-analyses-reported results of randomised controlled trials (RCTs) published between 1997 and 2004 failed to show any vasopressin-related benefit in cardiac arrest. Based on new RCT-data and a hypothesis of a potentially increased vasoconstricting efficacy of vasopressin, we sought to determine whether the cumulative, current evidence supports or refutes an overall and/or selective benefit for vasopressin regarding sustained restoration of spontaneous circulation (ROSC), long-term survival, and neurological outcome. METHODS Two reviewers independently searched PubMed, EMBASE, and Cochrane Database for RCTs assigning adults with cardiac arrest to treatment with a vasopressin-containing regimen (vasopressin-group) vs adrenaline (epinephrine) alone (control-group) and reporting on long-term outcomes. Data from 4475 patients in 6 high-methodological quality RCTs were analyzed. Subgroup analyses were conducted according to initial cardiac rhythm and time from collapse to drug administration (T(DRUG))<20 min. RESULTS Vasopressin vs. control did not improve overall rates of sustained ROSC, long-term survival, or favourable neurological outcome. However, in asystole, vasopressin vs. control was associated with higher long-term survival {odds ratio (OR)=1.80, 95% confidence interval (CI)=1.04-3.12, P=0.04}. In asystolic patients of RCTs with average T(DRUG)<20 min, vasopressin vs. control increased the rates of sustained ROSC (data available from 2 RCTs; OR=1.70, 95% CI=1.17-2.47, P=0.005) and long-term survival (data available from 3 RCTs; OR=2.84, 95% CI=1.19-6.79, P=0.02). CONCLUSIONS Vasopressin use in the resuscitation of cardiac arrest patients is not associated with any overall benefit or harm. However, vasopressin may improve the long-term survival of asystolic patients, especially when average T(DRUG) is <20 min.