CRD summary: This review assessed the effectiveness of vasopressin compared with epinephrine in the treatment of cardiac arrest. The authors concluded that vasopressin has no clear advantage over epinephrine in this context. This appears to be a well designed and conducted systematic review, so the authors' conclusions are likely to be reliable.

CRD commentary: The review question was explicit in terms of the population, intervention, outcomes and study design of interest. A thorough search was performed, without language restrictions and unpublished literature was also sought. The authors assessed publication bias. The validity assessment, data extraction and screening of studies for inclusion were all performed in duplicate, thus minimising the possibility of reviewer bias and error. Sufficient details of the primary studies were reported, allowing the reader to draw conclusions about the appropriateness of statistically synthesising the results of the studies. The techniques used for the data synthesis were appropriate and were well defined a priori, with the authors investigating possible sources of heterogeneity. This appears to be a well designed and conducted systematic review, so the authors' conclusions are likely to be reliable.