Adrenaline has been used in the treatment of cardiac arrest for many years. It increases the likelihood of return of spontaneous circulation (ROSC), but some studies have shown that it impairs cerebral microcirculatory flow. It is possible that better short-term survival comes at the cost of worse long-term outcomes. This narrative review summarises the rationale for using adrenaline, significant studies to date, and ongoing research. Keywords: Cardiac arrest, Cardiopulmonary resuscitation, Adrenaline, Epinephrine, Outcome Background The CPP is strongly associated with return of spontan- Adrenaline has been included in resuscitation guidelines eous circulation (ROSC) . worldwide since the 1960s and, through its action of Although global cerebral and coronary blood flow is increasing coronary and cerebral perfusion pressure, is increased by adrenaline, microcirculatory flow may be thought to increase the chance of restoring a heartbeat reduced. Once ROSC has been achieved, excessive (return of spontaneous circulation (ROSC)) and of plasma concentrations of adrenaline will cause tachycar- improving long-term neurological outcome. However, dia (which increases oxygen demand) and arrhythmias, there are no human data to show that long-term neuro- including ventricular tachycardia and ventricular fibrilla- logical outcome is improved with injection of adrenaline tion (VF). during cardiac arrest. Several observational studies docu- ment an association between the injection of adrenaline Animal studies and worse neurological outcome, but all of these are A study of 36 adult pigs, which were randomised to one confounded because of indication bias (those with more of two adrenaline doses (20 or 30 μg/kg) or to placebo, prolonged cardiac arrests are more likely to be given bolused every 3 minutes, documented increased arterial adrenaline and are more likely to have a poor outcome). blood pressure and increased CePP in the adrenaline This narrative review summarises the rationale for using groups . These two groups, however, had lower SpO adrenaline, significant studies to date, and ongoing values and lower cerebral tissue oximetry values than research. the placebo group, consistent with reduced organ and brain perfusion. A six-pig study measuring cerebral, Why is adrenaline used in cardiac arrest and why coronary, and aortic pressures and blood flow identified might it be harmful? that injection of 40 μg/kg of intravenous (IV) adrenaline Adrenaline has been a key component of advanced life significantly increased mean aortic pressure (29 ± 5 vs 42 support algorithms for many years. Its mechanism of ± 12 mmHg, p = 0.01), cerebral perfusion pressure (12 ± action—stimulation of α receptors in vascular smooth 5 vs 22 ± 10 mmHg, p = 0.01) and coronary perfusion muscle—causes vasoconstriction. This increases the aor- pressure (8 ± 7 vs 17 ± 4 mmHg, p = 0.02), but mean cor- tic diastolic pressure, which increases coronary perfusion onary blood flow decreased (29 ± 15 vs 14 ± 7.0 mL/min, pressure (CPP) and cerebral perfusion pressure (CePP). p =0.03) . * Correspondence: firstname.lastname@example.org Microcirculatory blood flow was evaluated with orthog- Anaesthesia and Intensive Care Medicine, Royal United Hospital, Bath BA1 onal polarization spectral imaging in ten pigs that were 3NG, UK 2 randomised to receive either adrenaline 30 μg/kg or vaso- Resuscitation Medicine, Bristol Medical School, University of Bristol, Bristol, UK pressin 0.4 units/kg during CPR . Post-resuscitation © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Gough and Nolan Critical Care (2018) 22:139 Page 2 of 8 microvascular flows and cerebral oxygen tension (PbO ) (10 μg/kg/min). CBF was monitored continuously. The were higher and cerebral carbon dioxide tension (PbCO ) adrenaline bolus groups had transient increases in CBF lower after vasopressin compared with adrenaline. In an- after each bolus, but the infusion group had higher CBF other study by the same group, cerebral blood flow (CBF; overall (p < 0.01) . assessed with microcirculatory imaging), cerebral oxygen In summary, adrenaline increases the mean aortic tension (PbO ), and carbon dioxide tension (PbCO )were pressure, but the effect on coronary and cerebral blood 2 2 measured in four groups of five pigs. The pigs receiving flow is inconsistent. In many cases, adrenaline reduces bolus adrenaline (30 μg/kg) achieved a higher mean aortic microcirculatory flow, even if global organ blood flow is pressure than those given placebo during and after CPR either increased or unchanged. The different techniques (p < 0.05), but had lower PbO values (p < 0.01) and higher used to monitor cerebral blood flow may contribute to PbCO values (p < 0.01) after resuscitation . Microcir- differences in results. culatory blood flow was lower in the adrenaline groups than the placebo group after resuscitation (p < 0.01). This Human physiological studies was also observed in a separate study where 15 pigs were In an early study of 100 patients, to enable continuous subjected to 5 min of VF, and 5 minutes of precordial pressure monitoring, during cardiac arrest invasive lines compression before electrical defibrillation was attempted were placed into the right atrium via the subclavian vein . Microcirculatory blood flow was assessed in the sub- and into the aortic arch via the femoral artery . lingual mucosa at regular intervals, and CPP was also re- Twenty-four patients had ROSC. The maximal CPP was corded. Six of the pigs received 1 mg of adrenaline after much higher in the patients who had ROSC, and no 1 min of precordial compression. Injection of adrenaline patient with a maximal CPP less than 15 mmHg had reduced microcirculatory blood flow (p < 0.05), which per- ROSC. sisted for several minutes. An observational study of regional cerebral oxygen- In another study, piglets were randomised to vasopres- ation (rSO ) measured by near-infrared spectroscopy sin, or vasopressin and adrenaline, with the adrenaline (NIRS) in 36 patients with in-hospital cardiac arrest doc- given by bolus (20 μg/kg) followed by infusion (10 μg/ umented rSO for 5 minutes before and after 89 doses kg/min) . Although the adrenaline with vasopressin of adrenaline . Of note, 66.7% of patients received group had higher mean blood pressure (p = 0.03) and only a single dose of adrenaline. Excluding 33 adrenaline CBF (p < 0.05) during CPR, after resuscitation the CBF events that were preceded by a previous dose of adren- was numerically 36% lower, although this was not statis- aline given in the 5-minute window, the effect on rSO tically significant (p = 0.06). Neuronal injury and signs of of 56 doses was assessed. The mean rSO increased by disruption to the blood–brain barrier were both greater 1.4% in the 5 minutes after adrenaline dosing compared in the adrenaline group. to the 5 minutes before (95% confidence interval (CI) In a study of 20 adult dogs, coronary, cerebral and 0.41–2.40%, p = 0.006). However, the rSO values were renal blood flow were measured, and cardiac tissue sam- already increasing by 0.88%/minute before injection of ples were taken for lactate concentration and myocardial adrenaline and this trend was not significantly altered by adenosine 5′-triphosphate (ATP) . The dogs were the adrenaline (p = 0.583). Whether or not NIRS is suffi- allocated randomly into two groups—those that received ciently sensitive and reliable for detecting changes in re- CPR alone and those that also received adrenaline (1 mg gional cerebral oxygenation associated with adrenaline bolus then 0.2 mg/min). The adrenaline group had remains to be established . higher myocardial blood flow (48 ± 11 vs 21 ± 4 ml/min/ Patients in cardiac arrest may transition from one 100 g, p < 0.05), but lower renal blood flow (1 ± 0 vs 74 rhythm to another, for example from PEA to VF, which ± 23 ml/min/100 g, p < 0.01). There was no significant may in turn give them a higher chance of achieving difference between the groups in ATP values but the ROSC. In an Oslo study of 174 patients with adrenaline group had a higher lactate concentration in out-of-hospital cardiac arrest (OHCA) with an initial the epicardium (6.3 ± 0.6 vs 4.2 ± 0.6 nmol/mg, p < 0.05, rhythm of PEA, patients given adrenaline were signifi- with a rise from baseline of 5.6 ± 0.5 vs 3.8 ± 0.5 nmol/ cantly more likely to transition into a different rhythm mg, p < 0.05). The higher lactate values associated with (rate ratio = 1.6, p < 0.001) . Although the rate of administration of adrenaline could reflect either in- transition from PEA to ROSC increased markedly in the creased myocardial oxygen demand and/or stimulation adrenaline group, the rate of transition from ROSC to of glycolysis. VT/VF also increased (regression parameter = 0.3, p < The effect on CBF of bolus adrenaline compared with 0.01), as well as from ROSC to PEA (regression param- an infusion was evaluated in 24 pigs that were rando- eter = 1.07, p < 0.01). mised to receive either boluses of adrenaline every 3 min In summary, adrenaline increases CPP and this is asso- (20 μg/kg) or a bolus (20 μg/kg) followed by an infusion ciated with a higher rate of ROSC. However, adrenaline Gough and Nolan Critical Care (2018) 22:139 Page 3 of 8 also increases instability and although it increases the on a sufficient sample size can provide a useful approxi- likelihood of transition to ROSC, it also makes the pa- mation of the effect of an intervention. tient more prone to develop arrhythmias, including VF. Clinical observational studies, including Propensity analysis systematic reviews and meta-analyses Several Japanese studies have documented associations Among 417,188 OHCAs in the Japanese nationwide between adrenaline use and short- and long-term out- registry between 2005 and 2008, ROSC before hospital comes. These observational studies are prone to consid- arrival was achieved in 18.5% of 15,030 patients who re- erable bias (e.g. patients successfully resuscitated early ceived adrenaline, and in 5.7% of 402,158 patients who are much less likely to have received adrenaline) and a did not receive adrenaline (p < 0.001; unadjusted odds variety of statistical techniques are used to adjust for ratio (OR) 3.75; 95% CI 3.59–3.91; Table 1). After confounders. One such technique is propensity analysis; propensity matching the adjusted odds ratio (aOR) for this is used when two groups of patients have dissimilar ROSC was 2.51 (95% CI 2.24–2.80). Although the raw characteristics that could account for any observed dif- outcome data indicate a higher rate of one-month sur- ference in outcome. A score is calculated that is the vival in those receiving adrenaline, after propensity probability that a patient would receive the treatment of matching the aOR for one-month survival was 0.54 (95% interest, based on characteristics of the patient, treating CI 0.43–0.68) and for CPC 1–2 the aOR was 0.21 (95% clinician, and environment [13, 14]. Many observational CI 0.10–0.44). These data suggest that more patients studies of the management of OHCA, such as some of who received adrenaline survived to hospital admission, those from the Japanese nationwide OHCA registry, use but that longer-term outcomes were better in the propensity score matching, which creates two groups of no-adrenaline group. study participants—one group that received the treat- Another analysis of the same Japanese nationwide ment of interest and the other that did not—while OHCA registry, but using a different period (2007 and matching individuals with similar propensity scores. This 2010), showed that among patients receiving adrenaline, approach has several limitations. Firstly, only the mea- the unadjusted rate of ROSC was higher in those with sured characteristics can be adjusted for, so any unmeas- an initial non-shockable rhythm (18.5 vs 5.7%) but lower ured confounders that affect treatment selection or in those with an initial shockable rhythm (21.6 vs 28.1%) outcome will not be corrected for. Secondly, the quality  (Table 1). The unadjusted survival rates (at one of the propensity model used will affect its outcome, as month or to discharge) and rates of survival CPC 1–2in will the size and quality of the included data. Observa- all patients were lower in those receiving adrenaline. tional data cannot establish causal relationships or treat- The authors identified 1990 propensity-matched pairs of ment effects, but appropriately used propensity analysis patients with and without adrenaline with an initial Table 1 Summary of outcomes from analyses of the All-Japan out-of-hospital cardiac arrest registry Author Hagihara Nakahara Nakahara Period 2005–2008 2007–2010 2007–2010 Subset NA Shockable Non-shockable Total number of cases 417,188 14,943 81,136 ROSC ROSC with adrenaline (unadjusted) 18.5% 21.6% 18.5% ROSC without adrenaline (unadjusted) 5.7% 28.1% 5.7% Adjusted OR (95% CI) 3.75 (3.59–3.91) NA NA One-month survival One-month survival with adrenaline (unadjusted) 5.4% 16.5% 3.9% One-month survival without adrenaline (unadjusted) 4.7% 28.8% 4.2% a b b Adjusted OR (95% CI) 0.54 (0.43–0.68) 1.34 (1.12–1.60) 1.72 (1.45–2.04) CPC 1–2 CPC 1–2 with adrenaline (unadjusted) 1.4% 6.9% 0.6% CPC 1–2 without adrenaline (unadjusted) 2.2% 19.8% 1.5% a b b Adjusted OR (95% CI) 0.21 (0.10–0.44) 1.01 (0.78–1.30) 1.57 (1.04–2.37) Data adjusted for propensity and all covariates Time-dependent propensity score-matched data Gough and Nolan Critical Care (2018) 22:139 Page 4 of 8 shockable rhythm, and 9058 propensity-matched pairs of with a worse neurological outcome (aOR for favourable patients with an initial non-shockable rhythm. In con- neurological outcome 0.32, 95% CI 0.22 to 0.47), even trast to the Hagihara study , after propensity match- after adjusting for in-hospital interventions. Although ing, the aOR for survival favoured adrenaline for both the authors made considerable effort to adjust for shockable (aOR 1.36, 95% CI 1.13–1.63) and confounders, the observational nature of this study pre- non-shockable rhythms (aOR 1.78, CI 1.49–2.13). The cludes any firm conclusion on causality. aORs for survival with CPC 1–2 in those with The effect of adrenaline can be inferred from a non-shockable rhythms (OR 1.55, CI 0.99–2.41) and before-after trial in Ontario, Canada, which studied the those with shockable rhythms (aOR 1.02, CI 0.78–1.33) impact of introducing prehospital advanced life support indicate no significant difference with and without (ALS) to an optimised basic life support automated adrenaline. Nakahara and colleagues  used a external defibrillation (BLS-AED) system . The ALS time-dependent propensity analysis which may account phase included tracheal intubation and intravenous for the contradictory findings between their study and drugs. Of the 4247 patients enrolled in the ALS phase, that of Hagihara and co-investigators. Time-dependent 95.8% received adrenaline. Patients in the ALS phase propensity analysis better adjusts for what has recently had higher rates of ROSC (18.0 vs 12.9%, p < 0.001) and been described as ‘resuscitation time bias’ where inter- survival to hospital admission (14.6 vs 10.9%, p < 0.001) ventions such as injection of adrenaline are more likely but no difference in survival to hospital discharge (5.1 vs to be implemented the longer the duration of cardiac ar- 5.0%, p = 0.83) . The limitation of this study is that it rest, and longer durations of cardiac arrest are associated is difficult to separate the impact on outcome of tracheal with worse outcome . intubation and injection of adrenaline. For example, any A third analysis of the Japanese nationwide registry, beneficial effect of adrenaline could be offset by harm this time covering the period 2009–2010, identified caused by tracheal intubation, and vice versa. Determin- 209,577 OHCA . Among the 15,492 patients who ing the impact of single interventions when they are had an initial shockable rhythm, the rate of ROSC, delivered as components of a package of care is one-month survival and one-month CPC 1–2 was 27.7, challenging. 27.0, and 18.6% in those who did not receive adrenaline A recent systematic review and meta-analysis includ- and 22.8, 15.4, and 7.0% in those who did receive adren- ing 13 observational studies and one randomised con- aline (all p < 0.001). In the 194,085 patients who initially trolled trial, with 655,653 OHCA patients, found that had a non-shockable rhythm, the rate of ROSC and the administration of adrenaline before hospital arrival one-month survival was 3.0%, and 2.2% in those who did was associated with an increase in ROSC (OR 2.84, 95% not receive adrenaline, and 18.7 and 3.9% in those who CI 2.28–3.54, p < 0.001), but was also associated with an did receive adrenaline (both, p < 0.001). There was no increase in the risk of poor neurological outcome at hos- significant difference in one-month CPC 1–2 between pital discharge (OR 0.51, 95% CI 0.31–0.84, p < 0.01), the two groups. Injection of adrenaline within 20 min of without affecting survival at one month (Figs. 1 and 2) onset of CPR was associated with better survival. For . non-shockable rhythms, injection of adrenaline within In summary, these observational data suggest that 10 min and 10–19 min of the onset of CPR was associ- adrenaline increases the rate of ROSC, but may have ated with increased one-month survival (aOR 1.78, 95% detrimental effects on overall survival, particularly CI 1.50–2.10 and aOR 1.29, CI 1.17–1.43, respectively). neurologically intact survival. It appears to have greatest Delayed injection of adrenaline was associated with benefit—or least harm—in patients with cardiac arrest worse neurological outcomes at one month (aOR 0.63, with an initial non-shockable rhythm. 95% CI 0.48–0.80 and aOR 0.49, CI 0.32–0.71) for adrenaline injected at 10–19 min and greater than Adrenaline timing 19 min, respectively. Several studies have shown that An analysis of 25,095 adult in-hospital cardiac arrest early adrenaline administration is associated with better (IHCAs) with an initial non-shockable rhythm in the Ameri- outcomes compared with later adrenaline (see ‘Adren- can Heart Association Get with the Guidelines-Resuscitation aline timing’ below). (AHA GWTG-R) registry between 2000 and 2009 identi- A Paris study, including all patients with OHCA who fied an association between survival and time to injection achieved ROSC and were admitted to a single centre be- of adrenaline . Time to adrenaline administration was tween 2000 and 2012, found that 17% of patients who analysed by 3-minute intervals, with an aOR of survival to received adrenaline had a favourable neurological out- hospital discharge of 1.0 for 1–3 min as the reference come (CPC 1–2) while 63% of patients who did not re- group. The aOR for survival to hospital discharge was ceive adrenaline had a CPC 1–2. After adjusting for 0.91 (95% CI 0.82–1.00, p =0.055) for 4–6 min, 0.74 (95% known confounders, use of adrenaline was associated CI 0.63–0.88, p <0.001) for 7–9 min, and 0.63 (95% CI Gough and Nolan Critical Care (2018) 22:139 Page 5 of 8 Fig. 1 Forrest plot comparing ROSC for those who did, and did not, receive adrenaline (epinephrine) 0.52–0.76, p < 0.001) for over 9 min. The results were A further analysis of the AHA GWTG-R registry similar for good neurological survival. evaluated the impact on outcome of time to adrenaline Another analysis of the AHA GWTG-R registry included administration among children (age < 18 years) with patients with an initial shockable rhythm who were defibril- IHCA and an initial non-shockable rhythm . Among lated within 2 minutes of the cardiac arrest and who 1558 children, 31.3% survived to hospital discharge. Al- remained in a shockable rhythm after defibrillation . though the median time to first adrenaline dose was 1 The authors focused on the patients who were given adren- minute (interquartile range 0–4), multivariate analysis aline within 2 minutes after the first defibrillation, which is identified that longer time to adrenaline administration counter to the guidelines of the AHA and European Resus- was associated with lower risk of survival to discharge, citation (these organisations recommend adrenaline deliv- with a risk ratio (RR) of 0.95 per minute delay (95% CI ery only after the second or third shocks, respectively). Of 0.93–0.99), as well as a lower risk of survival with 2978 propensity-matched patients, 1510 received adren- favourable neurological outcome, RR 0.95 per minute aline within 2 minutes of defibrillation and this intervention delay (95% CI 0.91–0.99). Children in whom the time to was associated with decreased odds of survival (OR 0.70, adrenaline administration was longer than 5 minutes 95% CI 0.59–0.82, p < 0.001). Early injection of adrenaline had a lower risk of survival to discharge compared with was also associated with a decreased rate of ROSC (OR those given adrenaline within 5 minutes (21.0 vs 33.1%, 0.71, 95% CI 0.60–0.83, p < 0.001) and good functional out- aRR 0.75, 95% CI 0.60–0.93, p = 0.01). come (OR 0.69, 95% CI 0.58–0.83, p < 0.001). As well as Another analysis of the Japanese nationwide registry be- the potential decrease in cerebral and coronary microcircu- tween 2008 and 2012 included 119,639 patients with a latory flow, it is possible that the increase in myocardial witnessed OHCA . The 20,420 patients who received oxygen demand associated with adrenaline may be particu- adrenaline were divided into four groups based on timing larly harmful in the first few minutes of a VF cardiac arrest. of adrenaline administration: early adrenaline (5–18 min), Fig. 2 Forrest plot comparing favourable neurological outcome (CPC 1–2) for those who did, and did not, receive adrenaline (epinephrine) Gough and Nolan Critical Care (2018) 22:139 Page 6 of 8 intermediate adrenaline (19–23 min), late adrenaline A recent before–after study of 2255 patients with (24–29 min), and very late adrenaline (30–62 min). Mul- non-traumatic OHCA compared different doses of tiple logistic regression analyses and aORs were deter- adrenaline. Initially, 1 mg adrenaline was given at mined for CPC 1–2 at one month, and for ROSC. Overall, 4 min, followed by additional 1 mg doses every 2 min the adrenaline group had a higher rate of ROSC (18 vs in those with non-shockable rhythms, and every 8 min 9.4%) but a lower rate of CPC 1–2(2.9vs5.2%). Incom- in those with shockable rhythms . During the parison with the late group, CPC 1–2 was highest in the intervention period, 0.5 mg of adrenaline was given at 4 early adrenaline group (aOR 2.49, 95% CI 1.90–3.27), and 8 min, followed by every 2 min in those with followed by the intermediate group (aOR 1.53, 95% CI non-shockable rhythms, and every 8 min in those with 1.14–2.05); the very late adrenaline group had the worst shockable rhythms. Although the dose of adrenaline neurological outcomes (in comparison with the late group: per patient reduced during the intervention period, aOR 0.71, 95% CI 0.47–1.08). there was no difference in survival to hospital discharge Other observational studies have shown that adren- or favourable neurological outcome in either the shock- aline is rarely given very early in a cardiac arrest. In a able or non-shockable groups. literature review where drug delivery time was reported in 7617 patients, the mean time to first drug delivery by Adrenaline dosing intervals any route was 17.7 min . Another US retrospective A review of 20,909 IHCAs from the AHA GWTG-R de- study of 686 patients reported similar findings—the fined the adrenaline average dosing interval as the time mean time to adrenaline administration was 14.3 min, between the first adrenaline dose and the resuscitation while those who received early adrenaline (within endpoint, divided by the total number of adrenaline 10 min) were more likely to have ROSC (32.9 vs 23.4%, doses received after the first dose . Compared with OR 1.59, 95% CI 1.07–2.38), although there was no sig- an average dosing interval of 4 to < 5 min per dose, sur- nificant difference in survival to discharge . vival to hospital discharge was higher in patients with In summary, these observational data indicate that longer dosing intervals: aOR 1.41 (95% CI 1.12–1.78) for 6 earlier use of adrenaline is associated with better out- to < 7 min/dose; aOR 1.30 (95% CI 1.02–1.65) for 7 to < comes than later use of adrenaline, but in patients with 8 min/dose; aOR 1.79 (95% CI 1.38–2.32) for 8 to < 9 min/ an initial shockable rhythm, administration of adrenaline dose; aOR 2.17 (95% CI 1.62–2.92) for 9 to < 10 min/dose. within 2 minutes of the first defibrillatory shock may be A much smaller single-centre study of 896 IHCAs in detrimental. Taiwan also found an association between shorter adren- aline dosing intervals and worse outcome . Adrenaline dose An analysis of 1630 IHCAs among children in the The optimal dose of adrenaline remains unknown. A same registry categorised average dosing intervals as meta-analysis of six randomised controlled trials (RCTs) 1–5 min, > 5 to < 8 min, and 8 to < 10 min/dose . comparing standard dose adrenaline (1 mg; SDA) with Compared with a reference of 1–5 min/dose, the aOR high-dose adrenaline (> 1 mg; HDA) found that SDA had for survival to hospital discharge was 1.81 (95% CI 1.26– a lower rate of ROSC (RR 0.85, 95% CI 0.75–0.97, p = 2.59) for > 5 to < 8 min/dose, and 2.64 (95% CI 1.53– 0.02) (Fig. 3), and survival to admission (RR 0.87, 95% CI 4.55) for 8 to < 10 min/dose. 0.76–1.00, p = 0.049). However, there was no difference in In summary, although high-dose adrenaline had no ap- survival to discharge (RR 1.04, 95% CI 0.76–1.42; Fig. 4) parent benefit over standard-dose adrenaline, a higher or neurologically favourable survival (RR 1.20, 95% CI rate of survival to hospital discharge was associated with 0.74–1.96) . longer adrenaline dosing intervals. Fig. 3 Forrest plot comparing ROSC for those who received high-dose adrenaline (HDA) compared with standard dose adrenaline (SDA) Gough and Nolan Critical Care (2018) 22:139 Page 7 of 8 Fig. 4 Forrest plot comparing survival to hospital discharge for those who received high-dose adrenaline (HDA) compared with standard dose adrenaline (SDA) Clinical randomised controlled trials Ongoing studies In a study from Norway, 851 OHCA patients were ran- The PARAMEDIC-2 trial (Pre-hospital Assessment of the domised to receive either ALS with IV access and drugs Role of Adrenaline: Measuring the Effectiveness of Drug as indicated (IV group) or ALS with IV access delayed administration In Cardiac arrest) has recently finished until 5 min after ROSC (no IV group) . Eighty per- recruiting more than 8000 patients. This individually cent of the patients in the IV group received adrenaline randomised, double-blind, placebo-controlled trial in- during resuscitation. In the 286 patients whose initial cluded OHCA patients in whom ALS was initiated, while rhythm was shockable (VF/pVT), there were no differ- excluding patients in cardiac arrest from anaphylaxis or ences between the groups in the rates of ROSC, survival life-threatening asthma, under-16 year olds, and those to ITU admission, or survival to hospital discharge. In who were pregnant. Adrenaline and placebo were pre- the 565 patients with an initial non-shockable rhythm pared in identical syringes and placed in pre-randomised (asystole or pulseless electrical activity (PEA)), those in packs of ten syringes. Outcomes will be survival to 30 days, the IV group had higher rates of ROSC (29 vs 11%, p < hospital discharge, 3, 6, and 12 months, health-related 0.001) and survival to ITU admission (19 vs 10%, p = quality of life, and neurological outcomes at hospital dis- 0.003), but survival to hospital discharge was similar (2 charge and 3 and 6 months . The results of this study vs 3%, p = 0.65). will be reported in 2018. A post hoc analysis of this study compared out- comes for patients actually receiving adrenaline with Conclusions those not receiving adrenaline . Patients receiving Although the administration of adrenaline remains one adrenaline had a higher rate of hospital admission of the most common ALS interventions, and likely in- (OR 2.5, 95% CI 1.9–3.4) but lower rate of survival to creases rate of ROSC after cardiac arrest, its effect on hospital discharge (OR 0.5, 95% CI 0.3–0.8) and lower long-term outcomes is far less certain. Several animal rate of neurologically intact survival (OR 0.4, 95% CI studies indicate that whilst global blood flow to vital or- 0.2–0.7). gans is generally increased, microcirculatory flow may be A double-blind placebo-controlled RCT from West- made worse by adrenaline. Many clinical observational ern Australia randomised 534 patients to ALS with and studies document an association between the injection without adrenaline. The adrenaline group had a higher of adrenaline and worse long-term outcomes, yet others rate of hospital admission (25.4 vs 13.0%, OR 2.3, 95% show an association between early injection of adren- CI 1.4–3.6) but survival to hospital discharge was not aline and better long-term outcome. Ultimately, it is statistically different between the groups (4 vs 1.9%, p = hoped that the recently completed large RCT comparing 0.15). The effect of adrenaline on pre-hospital ROSC adrenaline with placebo will provide some clarity on the was particularly marked in non-shockable rhythms (OR role of adrenaline, if any, in the treatment of cardiac 6.9, 95% CI 2.6–18.4) than in shockable rhythms (OR arrest. 2.4, 95% CI 1.2–4.5). With the exception of two pa- Authors’ contributions tients in the adrenaline group, all survivors had good CJRG and JPN drafted the manuscript and amended this in response to neurological outcomes (CPC 1–2) . reviewer comments. Both authors read and approved the final manuscript. In summary, these data from prospective clinical trials Competing interests suggest that adrenaline increases the rate of ROSC, but CJRG declares no competing interests. JPN is a co-investigator for the not long-term survival or neurologically favourable National Institute of Health Research (NIHR) funded PARAMEDIC-2 trial and is survival. Editor-in-Chief of the journal Resuscitation. Gough and Nolan Critical Care (2018) 22:139 Page 8 of 8 Publisher’sNote 23. Andersen LW, Kurth T, Chase M, et al. Early administration of epinephrine Springer Nature remains neutral with regard to jurisdictional claims in published (adrenaline) in patients with cardiac arrest with initial shockable rhythm in maps and institutional affiliations. hospital: propensity score matched analysis. BMJ. 2016;353:i1577. 24. Andersen LW, Berg KM, Saindon BZ, et al. Time to epinephrine and survival Received: 7 February 2018 Accepted: 10 May 2018 after pediatric in-hospital cardiac arrest. JAMA. 2015;314:802–10. 25. Tanaka H, Takyu H, Sagisaka R, et al. Favorable neurological outcomes by early epinephrine administration within 19 minutes after EMS call for out-of- hospital cardiac arrest patients. Am J Emerg Med. 2016;34:2284–90. References 26. Rittenberger JC, Bost JE, Menegazzi JJ. Time to give the first medication 1. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure and the during resuscitation in out-of-hospital cardiac arrest. Resuscitation. 2006; return of spontaneous circulation in human cardiopulmonary resuscitation. 70:201–6. JAMA. 1990;263:1106–13. 27. Koscik C, Pinawin A, McGovern H, et al. Rapid epinephrine administration 2. Hardig BM, Gotberg M, Rundgren M, et al. Physiologic effect of repeated improves early outcomes in out-of-hospital cardiac arrest. Resuscitation. adrenaline (epinephrine) doses during cardiopulmonary resuscitation in the 2013;84:915–20. cath lab setting: A randomised porcine study. Resuscitation. 2016;101:77–83. 28. Lin S, Callaway CW, Shah PS, et al. Adrenaline for out-of-hospital cardiac 3. Burnett AM, Segal N, Salzman JG, McKnite MS, Frascone RJ. Potential arrest resuscitation: A systematic review and meta-analysis of randomized negative effects of epinephrine on carotid blood flow and ETCO2 during controlled trials. Resuscitation. 2014;85:732–40. active compression-decompression CPR utilizing an impedance threshold 29. Fisk CA, Olsufka M, Yin L, et al. Lower-dose epinephrine administration and device. Resuscitation. 2012;83:1021–4. out-of-hospital cardiac arrest outcomes. Resuscitation. 2018;124:43–8. 4. Ristagno G, Sun S, Tang W, Castillo C, Weil MH. Effects of epinephrine and 30. Warren SA, Huszti E, Bradley SM, et al. Adrenaline (epinephrine) dosing vasopressin on cerebral microcirculatory flows during and after period and survival after in-hospital cardiac arrest: a retrospective review of cardiopulmonary resuscitation. Crit Care Med. 2007;35:2145–9. prospectively collected data. Resuscitation. 2014;85:350–8. 5. Ristagno G, Tang W, Huang L, et al. Epinephrine reduces cerebral perfusion 31. Wang CH, Huang CH, Chang WT, et al. The influences of adrenaline dosing during cardiopulmonary resuscitation. Crit Care Med. 2009;37:1408–15. frequency and dosage on outcomes of adult in-hospital cardiac arrest: A 6. Fries M, Weil MH, Chang YT, Castillo C, Tang W. Microcirculation during retrospective cohort study. Resuscitation. 2016;103:125–30. cardiac arrest and resuscitation. Crit Care Med. 2006;34:S454–7. 32. Hoyme DB, Patel SS, Samson RA, et al. Epinephrine dosing interval and 7. Halvorsen P, Sharma HS, Basu S, Wiklund L. Neural injury after use of survival outcomes during pediatric in-hospital cardiac arrest. Resuscitation. vasopressin and adrenaline during porcine cardiopulmonary resuscitation. 2017;117:18–23. Ups J Med Sci. 2015;120:11–9. 33. Olasveengen TM, Sunde K, Brunborg C, Thowsen J, Steen PA, Wik L. 8. Ditchey RV, Lindenfeld J. Failure of epinephrine to improve the balance Intravenous drug administration during out-of-hospital cardiac arrest: a between myocardial oxygen supply and demand during closed-chest randomized trial. JAMA. 2009;302:2222–9. resuscitation in dogs. Circulation. 1988;78:382–9. 34. Olasveengen TM, Wik L, Sunde K, Steen PA. Outcome when adrenaline 9. Johansson J, Gedeborg R, Basu S, Rubertsson S. Increased cortical cerebral (epinephrine) was actually given vs. not given - post hoc analysis of a blood flow by continuous infusion of adrenaline (epinephrine) during randomized clinical trial. Resuscitation. 2012;83:327–32. experimental cardiopulmonary resuscitation. Resuscitation. 2003;57:299–307. 35. Jacobs IG, Finn JC, Jelinek GA, Oxer HF, Thompson PL. Effect of adrenaline 10. Deakin CD, Yang J, Nguyen R, et al. Effects of epinephrine on cerebral on survival in out-of-hospital cardiac arrest: a randomised double-blind oxygenation during cardiopulmonary resuscitation: a prospective cohort placebo-controlled trial. Resuscitation. 2011;82:1138–43. study. Resuscitation. 2016;109:138–44. 36. Perkins GD, Quinn T, Deakin CD, et al. Pre-hospital Assessment of the Role 11. Putzer G, Braun P, Strapazzon G, et al. Monitoring of brain oxygenation during of Adrenaline: Measuring the Effectiveness of Drug administration In Cardiac hypothermic CPR - a prospective porcine study. Resuscitation. 2016;104:1–5. arrest (PARAMEDIC-2): trial protocol. Resuscitation. 2016;108:75–81. 12. Nordseth T, Olasveengen TM, Kvaloy JT, Wik L, Steen PA, Skogvoll E. Dynamic effects of adrenaline (epinephrine) in out-of-hospital cardiac arrest with initial pulseless electrical activity (PEA). Resuscitation. 2012;83:946–52. 13. Haukoos JS, Lewis RJ. The propensity score. JAMA. 2015;314:1637–8. 14. Andersen LW, Kurth T. Propensity scores - a brief introduction for resuscitation researchers. Resuscitation. 2018;125:66–9. 15. Hagihara A, Hasegawa M, Abe T, Nagata T, Wakata Y, Miyazaki S. Prehospital epinephrine use and survival among patients with out-of-hospital cardiac arrest. JAMA. 2012;307:1161–8. 16. Nakahara S, Tomio J, Takahashi H, et al. Evaluation of pre-hospital administration of adrenaline (epinephrine) by emergency medical services for patients with out of hospital cardiac arrest in Japan: controlled propensity matched retrospective cohort study. BMJ. 2013;347:f6829. 17. Andersen LW, Grossestreuer AV, Donnino MW. “Resuscitation time bias”–a unique challenge for observational cardiac arrest research. Resuscitation. 2018;125:79–82. 18. Goto Y, Maeda T, Goto Y. Effects of prehospital epinephrine during out-of- hospital cardiac arrest with initial non-shockable rhythm: an observational cohort study. Crit Care. 2013;17:R188. 19. Dumas F, Bougouin W, Geri G, et al. Is epinephrine during cardiac arrest associated with worse outcomes in resuscitated patients? J Am Coll Cardiol. 2014;64:2360–7. 20. Stiell IG, Wells GA, Field B, et al. Advanced cardiac life support in out-of- hospital cardiac arrest. N Engl J Med. 2004;351:647–56. 21. 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? J Crit Care. 2015;30:1376–81. 22. Donnino MW, Salciccioli JD, Howell MD, et al. Time to administration of epinephrine and outcome after in-hospital cardiac arrest with non- shockable rhythms: retrospective analysis of large in-hospital data registry. BMJ. 2014;348:g3028.
– Springer Journals
Published: May 29, 2018