Cardiac arrest remains associated with very high morbidity and mortality, and neurologically meaningful recovery (the most important goal of cardiac arrest care) is a regrettably infrequent outcome despite the best efforts of providers. Resuscitative efforts in cardiac arrest patients typically center around ACLS algorithms and attempts to find and correct reversible causes of clinical decompensation, while minimizing damage to vital organs in a low flow state via CPR and medications.

Epinephrine, via alpha-agonism and beta-agonism, is a mainstay of therapy in arrest patients, via peripheral vasoconstriction and cardiac stimulation. Vasopression has been proposed as an adjunctive therapy whose benefit also stems from peripheral vasoconstriction via V1 receptors. Finally, steroid medications have been studied extensively in stress states, and are thought to have a potential benefit in cardiac arrest patients for several reasons: attenuation of the systemic inflammatory response; treatment of adrenal impairment associated with arrest and intracellular potentiation of the effects of vasoconstriction medications. 

On the basis of this reasoning, the authors of this paper first published a small, single-center pilot study with improved outcomes for ROSC and survival to discharge. This study is a followup to that initial pilot study.

Clinical Question

Does a combined vasopressin-epinephrine medication regimen during cardiopulmonary resuscitation, with the addition of corticosteroids during and after arrest, improve neurologically meaningful recovery in in-hospital cardiac arrest?


All consecutive in-hospital cardiac arrest patients requiring vasopressor medications over the age of 18.


Administration of vasopressin (20U/cycle of CPR), plus epinephrine (1mg/cycle of CPR), plus methylprednisolone (40mg) x1 dose; ROSC patients in the intervention arm then received stress-dose hydrocortisone (300mg daily x 7 days followed by taper) if they remained hypotensive after ROSC.


Standard ACLS care; administration of epinephrine (1mg/cycle of CPR) + placebo instead of vasopressin or steroids.


Primary: Return of spontaneous circulation for 20 minutes or longer and survival to hospital discharge with a cerebral performance category (CPC) score of 1 or 2 (i.e. a good neurologic outcome)


Multicenter, randomized, Double-blinding, Placebo-controlled, parallel group, clinical trial.


Patients with cardiac arrest who did not receive epinephrine (immediate VF/VT defibrillation), patients already being treated with open-label hydrocortisone for other indication, age <18, terminal illness (life expectancy < weeks) and cardiac arrest prior to admission (specifically, ED patients not included).

Primary Results

  • 268 patients were included in the study: 138 in the control arm, 130 in the intervention arm.
  • Heterogeneity between centers for ROSC and neurologically favorable survival was low

Critical Findings

  • Primary Outcome (ROSC > 20 minutes)
    • VSE 83.9% (109/130) vs Control 65.9% (91/138)
    • OR = 2.98 (95% CI 1.39 – 6.40 p = 0.005)
    • NNT = 5.6
  • Primary Outcome (Survival to hospital discharge with CPC 1 or 2)
    • VSE 13.9% (18/130) vs. Control 5.1% (7/138)
    • OR = 3.28 (95% CI 1.17 – 9.20 p = 0.02)
    • NNT = 11.4


  • Relatively large, multi-center trial
  • Randomization and blinding appropriately performed
  • Study asked a clear clinical question that was patient centered
  • Follow up was complete
  • Distinguished between ROSC, survival, and survival with favorable neurological status – the most relevant outcome measure in this clinical population


  • The study had co-primary outcomes. This is a problem because the two primary outcomes are not equivalent in importance (one is ROSC for > 20 minutes the other is discharge with good neurologic status). There should only be one primary outcome
  • Despite appropriate randomization, etiologies of cardiac arrest were not equal; a larger proportion of intervention arm patients with cardiac arrest were attributable to primary cardiac etiologies, compared to a larger proportion of control group patients with primary respiratory etiology of arrest. Prior retrospective analyses have suggested better outcomes in the former group, possibly biasing these results.
  • Additionally, there were more patients with “metabolic” causes for cardiac arrest which included sepsis. These patients may respond better to vasopressin and steroids
  • Since the treatment arm received both vasopressin and steroids, we can’t determine the contribution of each individual piece. It is possible that all of the benefit is derived from one of the drugs or the other, that one drug may actually confer harm or that the drugs are synergistic.
  • Post-resuscitative therapeutic hypothermia was not consistently applied, across both groups; this may have changed the difference in neurological outcomes seen in the study.
  • CPR quality was not assessed or followed to compare it across the two groups, another possible source of bias or non-equal treatment.
  • Finally, and most relevant to our ED practice, these patients were all in-hospital cardiac arrest patients. ED patients were excluded. These patients may not represent the same type of cardiac arrest patients seen in the ED; etiologies may be different, overall prognosis for ED cardiac arrest patients may be more favorable, and traumatic causes of arrest are not addressed in the study. Therefore the applicability of these results may be limited in the ED population.

Other Issues

  • The steroid piece of therapy needs to be continued after ROSC for up to 7 days. This means that the intervention requires a hospital protocol and “buy-in” from all the relevant services
  • Since this publication, vasopressin has been removed from standard ACLS algorithms. As a result, the drug may not be readily available on all crash carts.

Author's Conclusions

“Among patients with cardiac arrest requiring vasopressors, combined vasopressin-epinephrine and methylprednisolone during CPR and stress-dose hydrocortisone in postresuscitation shock, compared with epinephrine/saline placebo, resulted in improved survival to hospital discharge with favorable neurological status.”

Our Conclusions

This well performed, multi-center RCT (rare in cardiac arrest care) appears to show a significant benefit for VSE therapy versus standard care in in-hospital cardiac arrest. However, baseline differences, inability of data to show equivalent intra-arrest and post-arrest care and, above all, the exclusion of ED patients makes it difficult to know how this can be applied to our patients.

Potential Impact To Current Practice

The VSE regimen in cardiac arrest has the potential to be practice-altering in the care of cardiac arrest patients, but further study and external validation is needed before current practice should be changed. Without ED based studies, these results can’t be applied to patients presenting to the ED.

Bottom Line

In this trial, the performance of vasopressin, steroids, and epinephrine in the resuscitation of in-hospital cardiac arrest was shown to have significant improvements in neurologically favorable recovery after cardiac arrest, but further study and validation are needed to confirm these findings and alter our practice moving forward.

Read More

The SGEM: SGEM#50: Under Pressure Journal Club

PulmCrit: 2015 ACLS Guidelines: What Happened to VSE

ScanCrit: CPR on Steroids

REBEL EM: Why You Should More Than Consider a Vasopressin, Steroid and Epinephrine (VSE) Cocktail