In summary, for adult patients treated with epinephrine for out-of-hospital cardiac arrest:

Benefits in NNT

    14% achieved ROSC prior to hospital arrival

    No benefit for survival to hospital discharge or survival at one month

    1 in 7 achieved ROSC prior to hospital arrival

    No benefit for survival to hospital discharge or survival at one month

Harms in NNT

    1.2% had a worse long-term neurological outcome
    1 in 83 had a worse long-term neurological outcome

View As:

Source: Loomba RS, Nijhawan K, Aggarwal S, Arora RR. Increased return of spontaneous circulation at the expense of neurologic outcomes: Is prehospital epinephrine for out-of-hospital cardiac arrest really worth it? Journal of Critical Care. 2015;30:1376-1381.

Study Population: Total of 655,853 patients from 13 observational studies and one randomized, controlled trial (RCT) involving patients who experienced out-of-hospital cardiac arrest.

Efficacy Endpoints: Pre-hospital ROSC, survival to hospital discharge, survival at one month

Harm Endpoints: Long-term neurological outcome, defined as Cerebral Performance Category (CPC) score of 1-2 (corresponding to independence in Activities of Daily Living)

Narrative: The original data supporting the use of epinephrine for cardiac arrest is rooted in a poorly controlled canine study1 from the 1960s. The Advanced Cardiac Life Support (ACLS) recommendation that it “may be reasonable” to use epinephrine in cardiac arrest continues in contemporary practice. Yet, when the ACLS guidelines2 are read carefully, they state that “for both survival to discharge and survival to discharge with good neurological outcome, there was no benefit” to receiving epinephrine.

Physiologically, epinephrine is theorized to promote peripheral vasoconstriction, thereby increasing diastolic pressure and coronary perfusion. Epinephrine also increases myocardial work and metabolic demand and may worsen tachydysrhythmia. Despite the purported physiologic benefits, it appears that epinephrine use for out-of-hospital cardiac arrest (OHCA) increases the rate of ROSC but does not increase the chance of survival (Table). Epinephrine may even worsen the neurological outcome in patients who do survive. Epinephrine use for OHCA may therefore extend suffering and increase end-of-life healthcare costs due to intensive care and prolonged hospitalization without clear patient-centered, long-term benefits. Patients who survive to hospital discharge may be more likely to be dependent on others for care. This presents an obvious ethical quandary, as patient and family preferences may differ greatly with regards to life-prolonging therapies.

Caveats: The source meta-analysis3 incorporated 13 observational studies and only one RCT:4 Jacob et al4 studied 601 patients and showed that epinephrine for OHCA was effective at increasing rate of ROSC (OR 3.4, 95% CI 2.0 to 5.6) and survival to hospital admission (OR 2.3, 95% CI 1.4 to 3.6) but lacked a statistically significant effect on survival to hospital discharge (OR 2.2; 95% CI 0.7 to 6.3). The observational trials included in the meta-analysis3 attempted to control for potential confounders (i.e. time to CPR) by propensity-matching individual study subjects, although bias can never be fully accounted for in any observational study. Similarly, the use of random effects methods cannot fully control for heterogeneity when reporting pooled effects in meta-analyses, and 3/4 of reported outcomes had exceedingly high heterogeneity (I2 = 96%). The source of heterogeneity may be patient characteristics, co-interventions, or trial-level, and random effects methods only attempt to adjust for between-trial variability, which can unintentionally inflate the effect of small studies on the pooled results.5 Propensity-matched outcomes in individual studies were generally in agreement with the pooled outcomes in this meta-analysis3 with the exception of survival at one month. The pooled results for one month survival may be confounded by the negative effect of smaller studies that were in favor of withholding epinephrine.

Specific effect modifiers in the included studies such as the timing and dosing of epinephrine administration may have influenced treatment effects on resuscitation outcomes. Earlier epinephrine administration has been associated with more favorable outcomes.6, 7, 8
The ‘standard’ one mg of epinephrine given in cardiac arrest is not weight-based and therefore can have a differential physiologic effect depending on the individual patient.9 The meta-analysis3 included some trials that used ‘high dose’ (0.1 to 0.2 mg/kg) epinephrine, which may have contributed negatively to pooled outcome results. Epinephrine administered at higher doses may be harmful10 and is not recommended by current guidelines.2

Loomba et al3 defines positive neurologic outcome as a cerebral performance category (CPC) of 1 or 2, which represents mild or moderate cerebral dysfunction but ability to independently perform activities of daily living. This is a validated method of measuring neurological outcome; however, trials included in the systematic review3 measured the CPC at different time points (hospital discharge, one month, etc.), resulting in significant heterogeneity in the data. The “recovery time” prior to CPC assessment is an important confounder when determining neurological outcome as are post-ROSC interventions such as targeted temperature management and urgent cardiac catheterization, none of which were adjusted for in the individual trials of this meta-analysis.3 Therefore, the results of the meta-analysis pertaining to this outcome should be interpreted with caution. The data discussed in Loomba et al3 are primarily applicable to cardiac arrests that occur out-of-hospital, and conclusions may not apply to patients that experience cardiac arrest in the setting of an emergency department, hospital floor, intensive care unit, or operating room. In these more controlled hospital settings, epinephrine is more likely to be given earlier along with prompt defibrillation and high-quality CPR, and therefore its use for in-hospital cardiac arrests may result in different outcomes. Finally, this systematic review3 analyzed only adult patients, and its conclusions should not be applied to the pediatric population.

Although in the past, there may have been barriers to performing true RCTs that must withhold potentially life-saving, guideline-recommended, “standard of care” from a control group, a double-blind, RCT (PARAMEDIC-2)11 was recently published and largely agrees with the meta-analysis.3 After enrolling 8,014 patients and following them in pragmatic fashion, PARAMEDIC-2 demonstrated that epinephrine increases survival to hospital admission and 30 days, with an NNT of 112 to prevent one death at 30 days. However, the additional proportion that received epinephrine and survived did so with severe neurological disability, defined as 4 or 5 on the modified Rankin scale. There was no evidence of benefit in the proportion of patients who survived to hospital discharge with a favorable neurologic outcome (OR 1.18, 95% CI 0.86 to 1.61), and the lack of effect persisted at 3 months. This confirms that the use of epinephrine for OHCA is not patient-centered (as defined by patients and the public in preparation for the PARAMEDIC-2 trial) and will cause prolonged suffering in a severely disabled state.

In summary, we chose a color recommendation of “Red” for epinephrine administration in OHCA. There is no patient-centered benefit and probable harm due to increased survival with worse long-term neurological function.

The original manuscript was published in Academic Emergency Medicine as part of the partnership between TheNNT.com and AEM.

Author: Kyle Kelson, MD and Ian S. deSouza, MD

Supervising editors: Kabir Yadav, MD; Shahriar Zehtabchi, MD

Published/Updated: August 13, 2018

  1. The Title Bar

    The title bar is color-coded with our overall recommendation.

    • Green: Benefits outweigh risks.
    • Yellow: Unclear risk/benefit profile.
    • Red: Benefits do not outweigh risks.
    • Black: Obvious harms, no clear benefits.
  2. Tip content...