A randomized controlled trial did not show benefit of vasopressin versus epinephrine in inhospital cardiac arrest. Preceding laboratory data suggest that combined vasopressin and epinephrine ensure long-term survival and neurologic recovery. Also, postresuscitation abnormalities mimic severe sepsis. The investigators hypothesized that combined vasopressin and epinephrine during cardiopulmonary resuscitation (CPR), and steroid supplementation during and after (when required) CPR may improve survival in cardiac arrest.

Inhospital cardiac arrest still constitutes an important clinical problem with survival to discharge ranging within 0-42% (most common range = 15-20%). Survival after witnessed, pulseless ventricular fibrillation/tachycardia(VF/VT) that is responsive to one or two direct current countershock(s) may exceed 30%. However, survival after inhospital asystole, pulseless electrical activity, or refractory VF/VT (defined as not responsive to two countershocks) may be substantially lower (< 5-10%). As in nonsurvivors, both endogenous vasopressin and adrenocorticotrophin are reduced compared to survivors, we hypothesized that the addition of exogenous vasopressin and steroids to the standard CPR protocol may increase the rates of both the return of spontaneous circulation (ROSC) and of post-arrest survival. The mechanistic basis of this hypothesis comprises the simultaneous activation of adrenergic and vasopressin receptors, in conjunction with a potential steroid-mediated enhancement of the vascular reactivity to epinephrine.

Adult in-patients with cardiac arrest not responsive to two direct current countershocks (when applicable) or asystole or pulseless electrical activity are randomized to receive either arginine vasopressin (Pitressin, 20 IU/CPR cycle for the first 5 CPR-cycles in non-VF/VT and from the second to sixth CPR-cycle in VF/VT) plus epinephrine (1 mg/CPR-cycle) plus methylprednisolone (single dose = 40 mg during the first and second CPR-cycle in non-VF/VT and VF/VT, respectively) or normal saline-placebo plus epinephrine (1 mg/CPR-cycle) plus normal saline-placebo during the first 5 or second to sixth CPR-cycles. Further CPR-vasopressor treatment includes epinephrine (1 mg/CPR-cycle) for both groups. Apart from the initial, combined drug administration in the study group, CPR is conducted in full concordance with the 2005 European Resuscitation Council Guidelines. Following ROSC and in the presence of postresuscitation shock (defined as inability to maintain mean arterial pressure > 70 mm Hg without using exogenous catecholamines at infusion rates conferring vasopressor and/or inotropic activity), study group patients receive stress dose hydrocortisone (300 mg/day for a maximum of 7 days and then gradual taper), whereas controls receive saline placebo. Following ROSC, control group patients may receive stress dose steroid treatment if prescribed by the attending physician for indications such as septic shock or known adrenocortical insufficiency. This holds also for study group patients during the follow-up period. Any steroid prescription by attending physicians cancels any concomitant investigational interventions regarding steroid supplementation.

The investigators involved in CPR drug administration are blinded to the use (or no-use) of vasopressin and methylprednisolone, and do not coordinate the CPR procedures. For the study group, steroid treatment is determined by the director of the pharmacy of Evaggelismos hospital, who also performs the computer-based patient randomization and encoding, and supervises the preparation of study drugs for CPR.

Patient follow-up and data recording is being conducted by four associates who are unaware of the CPR interventions. Daily follow-up to day 28 post-arrest includes physiological variables, medication and other treatment interventions, results of laboratory and diagnostic studies (including serum interleukins), and determination of the sequential organ dysfunction assessment (SOFA) score. Physiological variables include hemodynamics (arterial and central venous pressure, and heart rate), gas exchange and respiratory mechanics, body temperature, urinary output and fluid balance. Patient neurological status is being assessed with the Glasgow Coma Score. Following successful weaning from mechanical ventilation, cerebral performance is being assessed with the cerebral performance scale. Additional follow-up data include hospital/intensive care unit (ICU)-related morbidity, length of ICU/hospital stay, and cerebral performance/residual disabilities at hospital discharge.

Primary end-points are ROSC for ≥ 15 min, and survival to discharge either to home or to a rehabilitation facility. Secondary end-points include arterial pressure during CPR and at 15-20 min following ROSC, the intensity of the post-arrest systemic inflammatory response, the number of organ failure-free days during follow-up, and neurological status and cerebral performance during follow up and at discharge from the hospital.

In patients who survived for more than 28 days after the occurrence of the cardiac arrest, it has been ultimately feasible to collect full data on organ failure free days and medication until 60 days following randomization. Consequently, the analysis of the data during April and May 2007, actually enabled the calculation of the organ failure free days, the comparison of the length of the use of various drugs between the two groups, and the construction of survival curves and conduct of Kaplan-Meier analysis and Cox regression analysis until day 60 following randomization.

As in previous cardiac arrest trials, the requirement of informed consent before the administration of the drug combination during CPR has been waived. Informed consent was actually obtained for corticosteroid treatment of postresuscitation shock and for the blood sampling required for determination of plasma cytokine concentration after ROSC.

Inclusion Criteria:

• Adult in-patients with cardiac arrest requiring epinephrine according to current guidelines.

Exclusion Criteria:

• Age < 18 years.
• Documented terminal illness (life expectancy < 6 weeks).
• Do not resuscitate status.
• Cardiac arrest before arrival at hospital.
• Prior enrollment into the study (i.e. second or third inhospital arrest etc.).
• Corticosteroid treatment before the cardiac arrest.
• Any inaccurate documentation of CPR data such as medication, number of countershocks etc.