Antiarrhythmic drugs in cardiac arrest

[toc]

The role of antiarrhythmic drugs in cardiac arrest has gradually shifted over the decades. Only amiodarone remains, among drugs which are used as a part of the cardiac arrest algorithm. Even amiodarone's position in the ALS algorithm is not perpetually assured. 

Question 26 from the second paper of 2005 pre-dated the changes in ARC guidelines which have done away with lignocaine and atropine.

In brief, the objective of giving amiodarone after the third cycle of CPR for a shockable rhythm is to convert a defibrillation-refractory VF into one which is defibrillation-sensitive. The evidence for its use is supported by two trials (Dorian et al 2002, and Somberg et al 2002) which found some benefit of amiodarone over lignocaine in the context of shock refractory or recurrent VT and VF. There was no benefit in survival to hospital discharge. However, Kudenchuk et al (1999) found some benefit in survival to hospital admission. A meta-analysis in 2013 (Huang et al) re-analysed the data and came to the same depressing conclusion. This better than the evidence for any other anitarrhytmic drug, and thus amiodarone remains in the guidelines ...for now.

If one only had time to read only one source for this, the LITFL page would be enough to write an SAQ answer. There, one may also find a link to the 2011 review article by Ong et al, which is an excellent detailed overview.

Medications recommended for use in cardiac arrest

All of this information is available in the ARC Guideline 11.5: Medications in Adult Cardiac Arrest. In brief summary, other drugs which are covered by this guidelines statement are as follows:

• Adrenaline: Favoured because retrospective studies have found an improvement in the rates of ROSC with adrenaline; however there has never been any confirmed improvement in survival associated with it.
• Calcium: Not recommended routinely; no benefit in terms of survival (6.8mmol calcium chloride as a bolus)
• Lignocaine: Not as good as amiodarone, and thus recommended only for those situations when amiodarone cannot be used (1mg/kg bolus)
• Magnesium: Recommended for torsades des pointes, but not recommended for any other situation, as there is no survival benefit. (5mmol bolus)
• Potassium: Recommended for hypokalemic arrests only (5mmol bolus)
• Sodium bicarbonate: Not recommended, as it is associated with poor short-term and long-term outcomes.
• Vasopressin: Not recommended as an alternative to adrenaline, as there is insufficient data to support its use.
• Aminophylline: There is no evidence of harm, but there is insufficient evidence to recommend its routine use.
• Thrombolytics: Recommended only in confirmed or strongly suspected massive PE as a cause of cardiac arrest, in which case one is committed to performing CPR for 60-90 minutes.

Pros and cons of putting antiarrhythmic drugs into the ALS algorithm

Rationale for the use of antiarrhythmic drugs in cardiac arrest

• Cardiac arrest is often the consequence of a non-perfusing arrhythmia. Ergo, an antiarrhythmic drug is the correct treatment.
• The energy required to defibrillate is decreased by acute administration of amiodarone (in dogs - Fain et al, 1987). Thus, antiarrhythmic drugs may improve the effectiveness of defibrillation.
• The use of anti-arrhythmics may not be guided by any scientific principles, but it appears so deeply ingrained that it has become accepted as standard practice. Therefore to abjure the use of antiarrhythmics would be viewed as a substantial departure from standard practice. What would the coroner say?
• The presence or absence of these drugs in the ALS algorithm is purely the "official stance", and in every situation the use of these drugs will be guided by the judgement and experience of the clinican. It is possible to conjure in the imagination a series of scenarios where their use is well justified, eg. where an arrhythmia degenerates from perfusing to pulseless.
• The use of antiarrhythmics has been demonstrated to improve pre-hospital survival, but not overall survival; but one might argue that the effect of cardiac arrest medications is mainly aimed at improving the outcome of the immediate cardiac arrest episode, and that overall survival will be more heavily influenced by unmodifiable patient factors and post-arrest intensive care management.

Arguments against the use of antiarrhythmic drugs in cardiac arrest

• Pro-arrhythmic properties of antiarrhythmics must be taken into account; even amiodarone can produce QTc prolongation and torsade.
• Apart from the drugs themselves, there are harmful excipients which are safest in slow infusion, but could have cardiotoxic effects as a rapid bolus. For instance, amiodarone is rendered more soluble by the addition of benzyl alcohol and polysorbate 80, a surfactant and emulsifier. These have been implicated in causing negative haemodynamic effects when the drug is given rapidly (Masi et al, 2009). One must amend this objection by mentioning that the more recent formulations of amiodarone are excipient-free, instead using an acid buffer (100mmol/L of acetic acid, pH ~ 3.8) to increase the water solubility of amiodarone. This supposedly safer formulation was tested in the 2002 Somberg paper which is discussed below. 
• The addition of extra drugs or steps to the algorithm complicates it, and makes it more difficult to teach (and to follow).
• The improvement of survival to hospital admission may not translate into any improvement in survival (in fact, none of the studies have found any improvement in survival)
• Most of the trials involving athiarrhythmics have compared one drug to another; there has been little in the way of placebo-controlled trials. We do not know what would happen without these drugs. Would survival rates drop sharply? According to ROC-ALPS, they would not (placebo was only 3.2% less effective that the active drugs).

Evidence in support of antiarrhythmic drugs for shockable rhythm arrest 

Numerous papers have been published, and only a selection of a few interesting ones is offered here.

ARREST Trial 

Kudenchuk et al (1999) - randomized, double-blind, placebo-controlled study in the prehospital setting;  amiodarone (246 patients) or placebo (258 patients). Methodology resembles current practice (maybe because current practice is based mainly on this study) - 300mg amiodarone was given after the third shock. Survival to admission was improved (44% vs 34%). There was no significant improvement in survival until hospital discharge, nor was the study powered to detect such an improvement anyway. 

ALIVE trial 

Dorian et al (2002) performed a randomized, double-blind, placebo-controlled study in the prehospital setting; 347 patients in total. Methodology near-identical to Kudenchuk et al, except the authors compared amiodarone (2.5mg/kg) to lignocaine (1.5mg/kg). Unlike Kudenchuk et al, the protocol allowed repeated doses of either drug if the arrest continued. Amiodarone was the clear winner again: survival to admission was improved (22.8% vs 12.0%). Again there was no statistically significant difference in survival until discharge (5% vs 3%) in this grim pre-hypothermia dark age.

Somberg et al (2002)

A double-blind, parallel-design cross-over trial; 29 patients with shock-resistant VT received either two boluses 150mg amiodarone or two boluses of 100mg lignocaine, and crossed over to the other drug if the first one didn't work. The first line treatment was always cardioversion: patients were only eligible for enrolment if the first shock had failed.  Amiodarone was markedly better at lignocaine, both at terminating the arrhythmia (78% vs 27%) and in terms of 24-hour survival (39% vs 9%).

Of the lignocaine group, only one patient survived the first hour after VT termination. The rest ended up failing two boluses of lignocaine, then got shocked and crossed over to amiodarone. Then, the majority failed two boluses of that, developed asystole or brady-PEA and died. This brings up some questions. What the hell kind of resuscitation were these patients receiving? Was there even CPR? The average heart rate was 120; what about overdrive pacing?

ROC-ALPS 

In 2014 Kudenchuk et al published the rationale and methodology of their large scale of amiodarone vs. lignocaine vs. nothing. They expected n=3000, providing 90% power to detect a 6.3% survival benefit (from 23.4% to 29.7%). By the end of enrolment in 2015, they had 3026 patients. The detected survival benefit for amiodarone (vs placebo) was unfortunately only 3.2%, and 2.6% for lignocaine (i.e. a statistically insiginificant  0.7% difference). The treatment effect was somewhat better in the group of patients with witnessed arrest, suggesting that bystander CPR maintains some myocardial oxygenation so that the antiarrhythmics can have a chance to work. Ultimately, the authors were forced to conclude that on the basis of their data it was impossible to tell which drug was better, and in fact neither performed much better than placebo. 

Results of meta-analysis

Huang et al (2013) a meta-analysis of the above trials, plus a whole bunch of others. In total ten RCTs and seven observational studies were scraped together. The authors were able to firmly conclude in favour of amiodarone (vs. lignocaine and nifekalant) and against sotalol or bretylium. However, all the improvement was focused on surviving the arrest, rather than the hospital admission. Similarly, the 2016 meta-analysis by Laina et al redemonstrated the trend towards better prehospital survival and unchanged in-hospital mortality. On the basis of this, one may continue to make an argument for the ongoing use of amiodarone in cardiac arrest: the first step to overall survival is survival to hospital admission.