Antiarrhythmic drugs for cardiac arrest – Adults
Citation
Soar J, Donnino MW, Andersen LW, Berg KM, Böttiger BW, Callaway CW, Deakin CD, Drennan I, Neumar RW, Nicholson TC, O’Neil BJ, Paiva EF, Parr MJ, Reynolds JC, Ristagno G, Sandroni C, Wang TL, Welsford M, Nolan JP, Morley PT.
Antiarrhythmic Drugs for Cardiac Arrest in Adults and Children Consensus on Science and Treatment Recommendations [Internet] Brussels, Belgium: International Liaison Committee on Resuscitation (ILCOR) Advanced Life Support Task Force, 2018 May 30. Available from: http://ilcor.org
Anti-arrhythmic drugs for cardiac arrest PICOST
The PICOST (Population, Intervention, Comparator, Outcome, Study Designs and Timeframe)
Population: Adults and children in any setting (in-hospital or out-of-hospital) with cardiac arrest and a shockable rhythm at any time during cardiopulmonary resuscitation (CPR) or immediately after return of spontaneous circulation (ROSC).
Intervention: Administration (intravenous or intra-osseous) of an antiarrhythmic drug during CPR and immediately (within 1 hour) after ROSC.
Comparators: Another anti-arrhythmic drug or placebo or no drug during CPR or immediately after ROSC.
Outcomes: Survival to hospital discharge with good neurological outcome and survival to hospital discharge were ranked as critical outcomes. Return of spontaneous circulation (ROSC) was ranked as an important outcome. For antiarrhythmic drugs after ROSC – re-arrest was included as an important outcome.
Study Designs: Randomized controlled trials (RCTs) and non-randomized studies (non-randomized controlled trials, interrupted time series, controlled before-and-after studies, cohort studies) are eligible for inclusion.
Timeframe: All years and all languages were included as long as there was an English abstract; unpublished studies (e.g., conference abstracts, trial protocols) were excluded. Literature search updated to August 15, 2017.
Comparative data on the use of antiarrhythmic drugs during cardiac arrest were identified for amiodarone versus placebo, lidocaine versus placebo, amiodarone versus lidocaine, magnesium versus placebo, bretylium versus placebo, lidocaine versus bretylium, amiodarone versus nifekalant, lidocaine versus nifekalant, and lidocaine versus sotalol. There were no RCTs identified that addressed the use of antiarrhythmic drugs immediately after ROSC (defined as within 1 hour post ROSC).
Amiodarone versus placebo
For the critical outcome of survival to hospital discharge with good neurological outcome, the combined evidence with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 2 RCTs involving 2,526 out-of-hospital cardiac arrest (OHCA) patients showed no difference in effect for amiodarone compared with placebo (16.5% versus 14.6%; P= 0.18; RR, 1.13; 95% CI, 0.95–1.36, 19 more per 1,000 treated, from 7 fewer to 53 more) (Kudenchuk 1999 871, Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence for the Cordarone (amiodarone in polysorbate 80) preparation of amiodarone with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 1 RCT involving 504 patients showed no difference in effect for amiodarone compared with an active polysorbate 80 ‘placebo’ (7.3% versus 6.6%; P=0.75; RR, 1.11; 95% CI, 0.59–2.10, 7 more per 1,000, from 27 fewer to 72 more) (Kudenchuk 1999 871).
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence for the Nexterone preparation of amiodarone with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 2,022 patients showed no difference in effect for amiodarone compared with saline placebo (18.8% versus 16.6%; P=0.19; RR, 1.13; 95% CI, 0.94–1.37, 22 more per 1,000, from 10 fewer to 61 more) (Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge, the combined evidence with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 2 RCTs involving 2,530 patients showed no difference in effect for amiodarone compared with placebo (22.2% versus 19.5%; P=0.08; RR 1.14; 95% CI 0.98-1.33, 27 more per 1,000, from 4 fewer to 64 more) (Kudenchuk 1999 871, Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge, evidence for the Cordarone preparation of amiodarone with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 1 RCT involving 504 patients showed no difference in effect for amiodarone compared with an active polysorbate 80 ‘placebo’ (13.4% versus 13.2%; P=0.94; RR, 1.02; 95% CI, 0.65–1.59, 3 more per 1,000, from 46 fewer to 78 more) (Kudenchuk 1999 871).
For the critical outcome of survival to hospital discharge, evidence for the Nexterone preparation of amiodarone with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 2,026 patients showed no difference in effect for amiodarone compared with saline placebo (24.4% versus 21.0%; P=0.07; RR, 1.16; 95% CI, 0.99–1.37, 34 more per 1,000, from 2 fewer to 78 more) (Kudenchuk 2016 1711).
For the important outcome of ROSC, the combined evidence with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 2 RCTs involving 2,537 patients showed no difference in effect for amiodarone compared with placebo (37.5% versus 34.5%; P=0.11; RR, 1.13; 95% CI, 0.93–1.37, 45 more per 1,000, from 24 fewer to 128 more) (Kudenchuk 1999 871, Kudenchuk 2016 1711) .
For the important outcome of ROSC, evidence for the Cordarone preparation of amiodarone with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 1 RCT involving 504 patients showed a benefit favoring amiodarone compared with an active polysorbate 80 ‘placebo’ (43.9% versus 34.5%; P=0.03; RR, 1.27; 95% CI, 1.02–1.59, 93 more per 1,000, from 7 more to 204 more) (Kudenchuk 1999 871).
For the important outcome of ROSC, evidence for the Nexterone preparation of amiodarone with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 2,033 patients showed no difference in effect for amiodarone compared with saline placebo (35.9% versus 34.6%; P=0.52; RR, 1.04; 95% CI, 0.92–1.17, 14 more per 1,000, from 28 fewer to 59 more) (Kudenchuk 2016 1711).
Lidocaine versus placebo
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 2,039 patients showed no difference in effect for lidocaine compared with placebo (17.5% versus 16.6%; P=0.59; RR 1.05; 95% CI 0.87-1.28, 8 more per 1,000, from 22 fewer to 46 more) (Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge, evidence with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 2,041 patients showed no difference in effect for lidocaine compared with placebo (23.7% versus 21.0%; P=0.15; RR 1.13; 95% CI 0.96-1.32, 27 more per 1,000, from 8 fewer to 67 more) (Kudenchuk 2016 1711).
For the important outcome of ROSC, evidence with high certainty from 1 RCT involving 2,051 patients showed a benefit favoring lidocaine compared with placebo (39.9% versus 34.6%; P=0.01; RR 1.16; 95% CI 1.03-1.29, 55 more per 1,000, from 10 more to 100 more) (Kudenchuk 2016 1711).
Amiodarone versus lidocaine
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 1,951 patients showed no difference in effect for amiodarone compared with lidocaine (18.8% versus 17.5%; P=0.44; RR 1.08; 95% CI 0.89-1.30, 14 more per 1,000, from 19 fewer to 52 more), (Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge, the combined evidence with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 2 RCTs involving 2,302 patients showed no difference in effect for amiodarone compared with lidocaine (21.4% versus 20.7%; P=0.59; RR 1.04; 95% CI 0.89-1.22, 8 more per 1,000, from 23 fewer to 45 more) (Dorian 2002 884; Kudenchuk 2016 1711).
For the critical outcome of survival to hospital discharge, evidence for lidocaine with polysorbate 80 with very low certainty (downgraded for serious concerns for risk of bias, indirectness and imprecision) from 1 RCT involving 347 patients showed no difference in effect for amiodarone compared with lidocaine with polysorbate 80 (5.0% versus 3.0%; P=0.35; RR, 1.67; 95% CI, 0.57–4.88, 20 more per 1,000, from 13 fewer to 116 more) (Dorian 2002 884).
For the critical outcome of survival to hospital discharge, evidence for lidocaine with moderate certainty (downgraded for serious concerns for imprecision) from 1 RCT involving 1,955 patients showed no difference in effect for amiodarone compared with lidocaine (24.4% versus 23.7%; P=0.69; RR, 1.03; 95% CI, 0.88–1.21, 7 more per 1,000, from 28 fewer to 50 more) (Kudenchuk 2016 1711).
For the important outcome of ROSC, evidence with high certainty from 1 RCT involving 1,966 patients showed no difference in effect for amiodarone compared with lidocaine (35.9% versus 39.9%; P=0.07; RR 0.90; 95% CI 0.80-1.01, 40 fewer per 1,000, from 80 fewer to 4 more) (Kudenchuk 2016 1711).
Magnesium versus placebo
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 3 RCTs involving 332 patients showed no difference in effect for magnesium compared with placebo (8.0% versus 3.5%; P=0.10; RR 2.08; 95% CI 0.87-4.97, 38 more per 1,000, from 5 fewer to 140 more) (Fatovich 1997 237, Thel 1997 1272, Allegra 2000 245).
For the critical outcome of survival to hospital discharge, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 4 RCTs involving 437 patients showed no difference in effect for magnesium compared with placebo (9.8% versus 9.0%, P=0.81; RR 1.07; 95% CI 0.62-1.86, 6 more per 1,000, from 34 fewer to 77 more) (Fatovich 1997 237, Thel 1997 1272, Allegra 2000 245, Hassan 2002 57).
For the important outcome of ROSC, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 4 RCTs involving 437 patients showed no difference in effect for magnesium compared with placebo (33.2% versus 32.7%, P=0.83; RR 0.97; 95% CI 0.77-1.24, 10 fewer per 1,000, from 75 fewer to 79 more) (Fatovich 1997 237, Thel 1997 1272, Allegra 2000 245, Hassan 2002 57).
Bretylium versus placebo
For the critical outcome of survival to hospital discharge, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT involving 29 patients showed no difference in effect for bretylium compared with placebo (38.9% versus 9.1%, P=0.15; RR 4.28; 95% CI 0.60-30.26, 298 more per 1,000, from 43 fewer to 535 more) (Nowak 1981 404).
Lidocaine versus bretylium
For the critical outcome of survival to hospital discharge, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 2 RCTs involving 237 patients showed no difference in effect for lidocaine compared with bretylium (19.7% versus 23.5%, P=0.99; RR 0.84; 95% CI 0.51-1.36, 38 fewer per 1,000, from 143 fewer to 66 more) (Haynes 1981 353, Olson 1984 807).
For the important outcome of ROSC, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 2 RCTs involving 237 patients showed no difference in effect for lidocaine compared with bretylium (58.2% versus 49.6%, P=0.37; RR 1.23; 95% CI 0.78-1.92, 114 more per 1,000, from 109 fewer to 456 more) (Haynes 1981 353, Olson 1984 807).
Amiodarone versus nifekalant
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT with 30 patients showed no difference in effect for amiodarone compared with nifekalant (26.7% versus 26.7%; P=1.0; RR 1.00; 95% CI 0.31-3.28, 0 more per 1,000, from 184 fewer to 608 more) (Amino 2010 391).
For the critical outcome of survival to hospital discharge, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT involving 30 patients showed no difference in effect for amiodarone compared with nifekalant (53.3% versus 26.7%, P=0.16; RR 2.00; 95% CI 0.76-5.24, 267 more per 1,000, from 77 fewer to 536 more) (Amino 2010 391).
For the important outcome of ROSC, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT involving 30 patients showed no difference in effect for amiodarone compared with nifekalant (66.7% versus 46.7%, P=0.28; RR 1.43; 95% CI 0.75-2.73) 201 more per 1,000, from 117 fewer to 807 more) (Amino 2010 391).
Lidocaine versus nifekalant
For the critical outcome of survival to hospital discharge, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT involving 28 patients showed no difference in effect for lidocaine compared with nifekalant (survival rate = 0% in both groups) (Igarashi 2005 S155).
For the important outcome of ROSC, evidence with very low certainty (downgraded for serious concerns for risk of bias and very serious concerns for imprecision) from 1 RCT involving 22 patients showed a lower rate of ROSC with lidocaine compared with nifekalant (14.3% versus 62.5%, P=0.04; RR 0.23; 95% CI 0.06-0.92, 481 fewer per 1,000, from 587 fewer to 50 fewer) (Igarashi 2005 S155).
Lidocaine versus sotalol
For the critical outcome of survival to hospital discharge with good neurological outcome, evidence with low certainty (downgraded for very serious concerns for imprecision) from 1 RCT with 129 patients showed no difference in effect for lidocaine compared with sotalol (4.35% versus 0.0%; P=0.23; RR 6.10; 95% CI 0.32-115.76, 43 more per 1,000, from 23 fewer to 120 more) (Kovoor 2005 518).
For the critical outcome of survival to hospital discharge, evidence with low certainty (downgraded for very serious concerns for imprecision) from 1 RCT involving 119 patients showed no difference in effect for lidocaine compared with sotalol (8.5% versus 3.3%, P=0.34; RR 2.17; 95% CI 0.44-10.80, 39 more per 1,000, from 19 fewer to 327 more) (Kovoor 2005 518).
For the important outcome of ROSC, evidence with low certainty (downgraded for very serious concerns for imprecision) from 1 RCT involving 129 patients showed no difference in effect for lidocaine compared with sotalol (37.7% versus 26.7%, P=0.19; RR 1.41; 95% CI 0.84-2.37, 109 more per 1,000, from 43 fewer to 365 more) (Kovoor 2005 518).
Treatment Recommendations
We suggest the use of amiodarone or lidocaine in adults with shock refractory ventricular fibrillation/pulseless ventricular tachycardia (VF/pVT) (weak recommendation, low quality evidence).
We suggest against the routine use of magnesium in adults with shock refractory VF/pVT (weak recommendation, very low-quality evidence).
For the use of bretylium, nifekalant or sotalol in adults with shock refractory VF/pVT the confidence in effect estimates is currently too low to support an ALS Task Force recommendation.
For the use of prophylactic antiarrhythmic drugs immediately after ROSC in adults with VF/pVT cardiac arrest the confidence in effect estimates is currently too low to support an ALS Task Force recommendation.
Values and Preferences
- This topic was prioritized by the ALS Task Force based on a large RCT comparing amiodarone, lidocaine and placebo (‘ROC ALPS’) (Kudenchuk 2016 1711) that was published after the previous CoSTR in 2015 (Callaway 2015 s84, Soar 2015 e71).
- In considering the importance of this topic we noted that in a large RCT (n= 23,711) of continuous or interrupted chest compressions during cardiopulmonary resuscitation (CPR) for out-of-hospital cardiac arrest (OHCA) (Nichol 2015 2203), 22.5% of patients had an initial rhythm of VF/pVT and about 6.7% of all patients received an antiarrhythmic drug (amiodarone 4.7%, lidocaine 2.0%) during CPR. A large observational study (n= 108,079) on airway management using data from the American Heart Association Get With The Guidelines Registry of in-hospital cardiac arrest (IHCA) reported that about 18% of all patients had an initial rhythm of VF/pVT, and 25% of all patients received an antiarrhythmic drug (amiodarone 17%, lidocaine 8%) during CPR (Andersen 2017 494).
- Given the availability of comparative data from large RCTs, we did not include non-RCTs in establishing our confidence in the estimated effect size of these drugs.
In making these recommendations, the ALS Task Force considered the following:
Amiodarone or lidocaine
- We considered the predefined and reported subgroup analysis of the ROC ALPS study (Kudenchuk 2016 1711) that showed an improvement in the critical outcome of survival to hospital discharge with amiodarone or lidocaine compared with placebo in those patients who had a bystander witnessed cardiac arrest. In addition, survival rate was also higher among amiodarone recipients than placebo recipients with EMS-witnessed arrest – this was associated with earlier drug use: the time from cardiac arrest to the first dose of trial drug was 11.7±5.8 min for EMS-witnessed arrest versus a time from 911-call to the first study drug of 19.3±7.1 for non-EMS-witnessed cardiac arrest.
- We did not identify any RCTs for in-hospital cardiac arrest (IHCA). The EMS-witnessed subgroup analysis data from a large OHCA RCT does suggest the use of antiarrhythmic drugs in the hospital setting could be useful as drugs tend to be given much earlier after IHCA. We acknowledge the lack of RCT data for IHCA in our knowledge gaps.
- In making a weak recommendation, we considered the reported increase in the important but short-term outcome of ROSC of both amiodarone (Kudenchuk 1999 871) or lidocaine (Kudenchuk 2016 1711) with no evidence of improved or worse longer-term outcomes ranked as critical: survival or good neurological survival to hospital discharge.
- We considered that in the ROC ALPS study there was no difference between amiodarone and lidocaine in ROSC, survival or good neurological survival to hospital discharge.
- We considered the differences between the two amiodarone versus ‘placebo’ RCTs (Kudenchuk 1999 871, Kudenchuk 2016 1711), and also the two amiodarone versus lidocaine RCTs (Dorian 2002 884, Kudenchuk 2016 1711). We discussed the benefits of pooling or keeping the studies separate in the systematic review and meta-analyses. The benefits of increasing precision of an estimate of effect were weighed against the detrimental effects of combining distinctly different studies. We have provided both pooled estimates based on combining studies and also just those from the individual studies.
- The earlier RCTs (Kudenchuk 1999 871, Dorian 2002 884) used polysorbate 80 as placebo in the amiodarone v placebo study (Kudenchuk 1999 871), and mixed polysorbate 80 with lidocaine (Dorian 2002 884) in the amiodarone v lidocaine study. The effects of polysorbate 80 on the outcome of these studies is uncertain.
Magnesium
- We did not identify any new RCTs published since the 2015 CoSTR.
- In making a suggestion against the routine use of magnesium for refractory VF/pVT cardiac arrest, we recognize that there are specific circumstances where magnesium could be useful during refractory VF/pVT (e.g. hypomagnesemia, torsade de pointes).
Bretylium, nifekalant, sotalol
- In making no recommendation, we considered guidance from the GRADE handbook: [Section 6.1.4 No recommendation, accessed 12 April 2018]: http://gdt.guidelinedevelopmen...
- We did not identify any RCTs that compared nifekalant with a placebo.
- We identified only one very small RCT with 30 patients that compared amiodarone with nifekalant (Amino 2010 391) and another very small RCT with 28 patients that compared lidocaine with nifekalant (Igarashi 2005 S155).
- We recognize bretylium is not available in most settings for clinical use and is not part of current guidelines.
- Sotalol is not part of current guidelines. The role of beta-blocker drugs during and after cardiac arrest remains a knowledge gap.
Prophylactic antiarrhythmic drugs immediately after ROSC
- We did not identify any RCTs for the use of prophylactic antiarrhythmic drugs in patients immediately (within 1 hour) after ROSC following a VF/pVT cardiac arrest and have identified this as a knowledge gap.
Knowledge Gaps
Current knowledge gaps include but are not limited to:
- What is the role of antiarrhythmic drugs for in-hospital cardiac arrest?
- What is the optimal bundle of care for shock refractory VF/pVT (defibrillation attempts versus drugs versus mechanical CPR/extracorporeal CPR/percutaneous coronary intervention(PCI))?
- Does the etiology of cardiac arrest (e.g. coronary artery disease, cardiomyopathy, inherited heart rhythm disorder, congenital heart disease, drug-induced arrhythmia, long-QT syndromes and pulmonary embolism) have an impact on the effectiveness of antiarrhythmic drugs during CPR?
- Do patients and families value short term outcomes (e.g. ROSC, intensive care unit admission) after cardiac arrest for those patients who subsequently die prior to hospital discharge?
- What is the cost effectiveness of antiarrhythmic drug treatment during CPR?
- What is the effect of antiarrhythmic drugs during CPR on long term outcomes and health related quality of life?
- Does adrenaline (epinephrine) alter effectiveness of antiarrhythmic drugs? We have no data on the effectiveness of antiarrhythmic drugs used prior to or without adrenaline.
- What is the optimal timing of antiarrhythmic drugs during CPR (how early, after how may defibrillation attempts?
- Is multiple antiarrhythmic drug use (e.g. amiodarone followed by lidocaine) more effective than single drug use?
- What is the impact of bystander CPR on the effectiveness of antiarrhythmic drugs?
- Are there differences in the effectiveness of different amiodarone preparations during CPR?
- What are the effects of polysorbate 80 during CPR for VF/pVT cardiac arrest?
- Is there a difference in effectiveness between intravenous (IV) and intraosseous (IO) antiarrhythmic drug use during VF/pVT cardiac arrest?
- How effective is nifekalant compared with placebo or alternative antiarrhythmic drugs in adult VF/pVT cardiac arrest?
- Does CPR quality impact antiarrhythmic drug effectiveness during CPR?
- Does magnesium have a role in cardiac arrest (e.g. neuroprotection) other than its effects on heart rhythm?
- What is the role of prophylactic antiarrhythmic drugs or beta blockers immediately after ROSC?
Attachments
Should LIDOCAINE vs no lidocaine be used for adults with shock refractory VF/pVT
Should AMIODARONE vs LIDOCAINE be used for adults with shock refractory VF/pVT
Should AMIODARONE vs. no amiodarone be used for adults with shock refractory VF/pVT
Should MAGNESIUM vs No Magnesium be used for adults with shock refractory VF/pVT