Part 3: Adult Basic and Advanced Life Support: 2020 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care

Agonal breathing is characterized by slow, irregular gasping respirations that are ineffective for ventilation. Agonal breathing is described by lay rescuers with a variety of terms including, abnormal breathing, snoring respirations, and gasping.⁴ Agonal breathing is common, reported as being present in up to 40% to 60% of victims of OHCA.⁵ The presence of agonal breathing is cited as a common reason for lay rescuers to misdiagnose a patient as not being in cardiac arrest.⁶ In patients who are unresponsive, with absent or abnormal breathing, lay rescuers should assume the patient is in cardiac arrest, call for help, and promptly initiate CPR. These 2 criteria (patient responsiveness and assessment of breathing) have been shown to rapidly identify a significant proportion of patients who are in cardiac arrest, allowing for immediate initiation of lay rescuer CPR. Further, initiation of chest compressions in patients who are unconscious but not in cardiac arrest is associated with low rates of significant adverse events.² The adverse events noted included pain in the area of chest compressions (8.7%), bone fracture (ribs and clavicle) (1.7%), and rhabdomyolysis (0.3%), with no visceral injuries described.²

Protracted delays in CPR can occur when checking for a pulse at the outset of resuscitation efforts as well as between successive cycles of CPR. Healthcare providers often take too long to check for a pulse^7,8 and have difficulty determining if a pulse is present or absent.^7–9 There is no evidence, however, that checking for breathing, coughing, or movement is superior to a pulse check for detection of circulation.¹⁰ Thus, healthcare providers are directed to quickly check for a pulse and to promptly start compressions when a pulse is not definitively palpated.^9,11

CPR is the single-most important intervention for a patient in cardiac arrest, and chest compressions should be provided promptly. Chest compressions are the most critical component of CPR, and a chest compression–only approach is appropriate if lay rescuers are untrained or unwilling to provide respirations. Beginning the CPR sequence with compressions minimized time to first chest compression.^2–4 Nationwide dissemination of chest compression–only CPR for lay rescuers was associated with an increase in the incidence of survival with favorable neurological outcome after OHCAs in Japan, likely due to an increase in lay rescuers providing CPR.⁵ Chest compressions should be provided as soon as possible, without the need to remove the victim’s clothing first.

The optimal timing of CPR initiation and emergency response system activation was evaluated by an ILCOR systematic review in 2020.¹ An observational study of over 17 000 OHCA events reported similar results from either a “call-first” strategy or a “CPR-first” strategy.⁶ In the current era of ubiquitous mobile devices, ideally both the call to activate EMS and the initiation of CPR can occur simultaneously.

Four observational studies^7–10 reported outcomes from patients who were not in cardiac arrest and received CPR by lay rescuers. No serious harm from CPR was found in patients when they were later determined not to have been in cardiac arrest.¹ This is in contrast to the significant risk of withholding CPR when a patient is in cardiac arrest, making the risk:benefit ratio strongly in favor of providing CPR for presumed cardiac arrest.

In some observational studies, improved outcomes have been noted in victims of cardiac arrest who received conventional CPR (compressions and ventilation) compared with those who received chest compressions only.^5,11,12 Other studies have reported no difference in outcomes for patients receiving conventional versus compression-only CPR.^11,13–21 Given the potential benefit of conventional CPR, if lay rescuers are appropriately trained, they should be encouraged to concurrently deliver ventilation with compressions. A thorough review of the data concerning the ratio of compressions to ventilation when performing conventional CPR is discussed in Ventilation and Compression-to-Ventilation Ratio.

The 2010 Guidelines for CPR and Emergency Cardiovascular Care included a major change for trained rescuers, who were instructed to begin the CPR sequence with chest compressions rather than with breaths (circulation, airway, and breathing versus airway, breathing, and circulation) to minimize the time to initiation of chest compressions. This approach is resupported by new literature, summarized in a 2020 ILCOR systematic review (Table 2).^1–4 In the recommended sequence, once chest compressions have been started, a single trained rescuer delivers rescue breaths by mouth to mask or by bag-mask device to provide oxygenation and ventilation. Manikin studies demonstrate that starting with chest compressions rather than with ventilation is associated with faster times to chest compressions,^3,23 rescue breaths,⁴ and completion of the first CPR cycle.⁴

Healthcare providers are trained to deliver both compressions and ventilation. Delivery of chest compressions without assisted ventilation for prolonged periods could be less effective than conventional CPR (compressions plus ventilation) because arterial oxygen content decreases as CPR duration increases. This concern is especially pertinent in the setting of asphyxial cardiac arrest.¹¹ Healthcare providers, with their training and understanding, can realistically tailor the sequence of subsequent rescue actions to the most likely cause of arrest.

1 and 2. The head tilt–chin lift has been shown to be effective in establishing an airway in noncardiac arrest and radiological studies.^2–5 No studies have compared head tilt–chin lift with other airway maneuvers to establish an airway during cardiac arrest.

3. Although there is no evidence examining the effectiveness of their use during cardiac arrest, oropharyngeal and nasopharyngeal airways can be used to maintain a patent airway and facilitate appropriate ventilation by preventing the tongue from occluding the airway. Incorrect placement, however, can cause an airway obstruction by displacing the tongue to the back of the oropharynx.^6,7

4. The benefit of an oropharyngeal compared with a nasopharyngeal airway in the presence of a known or suspected basilar skull fracture or severe coagulopathy has not been assessed in clinical trials. However, an oral airway is preferred because of the risk of trauma with a nasopharyngeal airway. Multiple case reports have observed intracranial placement of nasopharyngeal airways in patients with basilar skull fractures.^8,9

5. There is no evidence that cricoid pressure facilitates ventilation or reduces the risk of aspiration in cardiac arrest patients. There is some evidence that in non–cardiac arrest patients, cricoid pressure may protect against aspiration and gastric insufflation during bag-mask ventilation.^10–13 However, cricoid pressure may also impede ventilation and the placement of a supraglottic airway (SGA) or intubation,^14–20 and increase the risk of airway trauma during intubation.²¹

Healthcare providers should consider the possibility of a spinal injury before opening the airway. If a spinal injury is suspected or cannot be ruled out, providers should open the airway by using a jaw thrust instead of head tilt–chin lift.²

Maintaining a patent airway and providing adequate ventilation and oxygenation are priorities during CPR. If a jaw thrust and/or insertion of an airway adjunct are ineffective in opening the airway and allowing ventilation to occur, a head tilt–chin lift may be the only way to open the airway. In these cases, this maneuver should be used even in cases of potential spinal injury because the need to open the airway outweighs the risk of further spinal damage in the cardiac arrest patient.

When spinal injury is suspected or cannot be ruled out, rescuers should maintain manual spinal motion restriction and not use immobilization devices. Manual stabilization can decrease movement of the cervical spine during patient care while allowing for proper ventilation and airway control.^23,24 Spinal immobilization devices may make it more difficult to maintain airway patency^25,26 and provide adequate ventilation.

A 2020 ILCOR systematic review identified 3 studies involving 57 total patients that investigated the effect of hand positioning on resuscitation process and outcomes.⁴ Although no difference in resuscitation outcomes was noted, 2 studies found better physiological parameters (peak arterial pressure, mean arterial pressure [MAP], end-tidal carbon dioxide [ETCO₂]) when compression was performed over the lower third of the sternum compared with the middle of the sternum.^5,6 A third study found no difference.⁷ Radiographic studies show the left ventricle is typically located inferior to the internipple line, corresponding with the lower half of the sternum.⁸ However, hand placement inferior to the internipple line may result in compression over the xiphoid.⁹ Although data from manikin studies conflict, it does not appear to matter whether the dominant or nondominant hand is placed in contact with the sternum.^10,11

The primary considerations when determining if a victim needs to be moved before starting resuscitation are feasibility and safety of providing high-quality CPR in the location and position in which the victim is found. This is a separate question from the decision of if or when to transport a patient to the hospital with resuscitation ongoing.

The effectiveness of CPR appears to be maximized with the victim in a supine position and the rescuer kneeling beside the victim’s chest (eg, out-of-hospital) or standing beside the bed (eg, in-hospital).¹² It is thought that optimal chest compressions are best delivered with the victim on a firm surface.^13,14 Manikin studies show generally acceptable thoracic compression with CPR performed on a hospital mattress.

An older systematic review identified 22 case reports of CPR being performed in the prone position (21 in the operating room, 1 in the intensive care unit [ICU]), with 10/22 patients surviving.¹⁵ In a small case series of 6 patients with refractory IHCA, prone positioning with the use of a board with sandbag to compress the sternum improved hemodynamics during CPR but did not result in ROSC.¹⁶ The efficacy of CPR in the prone position is not established, but the very limited evidence suggests it may be better than providing no CPR, when a patient cannot be placed in supine position, or until this can be done safely.

Observational evidence suggests improved outcomes with increased chest compression fraction in patients with shockable rhythms.^18,19 Specifically, studies have also reported increased ROSC with shorter perishock pauses.^20–22

This recommendation is based on the overall principle of minimizing interruptions to CPR and maintaining a chest compression fraction of at least 60%, which studies have reported to be associated with better outcome.^18,19,23

Chest compression depth begins to decrease after 90 to 120 seconds of CPR, although compression rates do not decrease significantly over that time window.²⁴ A randomized trial using manikins found no difference in the percentage of high-quality compressions when rotating every 1 minute compared with every 2 minutes.²⁵ Rotating the designated chest compressor every 2 minutes is sensible because this approach maintains chest compression quality and takes advantage of when CPR would ordinarily be paused for rhythm analysis.

Two RCTs enrolling more than 1000 patients did not find any increase in survival when pausing CPR to analyze rhythm after defibrillation.^26,27 Observational studies show decreased ROSC when chest compressions are not resumed immediately after shock.^28,29

Because chest compression fraction of at least 60% is associated with better resuscitation outcomes, compression pauses for ventilation should be as short as possible.^18,19,23

A 2015 systematic review reported significant heterogeneity among studies, with some studies, but not all, reporting better rates of survival to hospital discharge associated with higher chest compression fractions.^18,19,23 In 2 studies, higher chest compression fraction was associated with lower odds of survival.^2,30 Compression rate and depth and cointerventions such as defibrillation, airway management, and medications, are also important and may interact with chest compression fraction. High-performing EMS systems target at least 60%, with 80% or higher being a frequent goal.

A 2020 ILCOR scoping review³² identified 12 studies, including over 12 500 patients, looking at chest compression components. Several studies found better outcomes, including survival to hospital discharge and defibrillation success, when compression depth was at least 5 cm compared with less than 4 cm.^3,20,33,34

The same review³² identified 13 studies, involving 15 000 patients, that looked at compression rate. Results were somewhat inconsistent across studies, with only 3 observational studies in adults showing an association between higher compression rate and outcomes.^1,35,36 The only RCT identified included 292 patients and compared a rate of 100 to a rate of 120, finding no difference in outcomes.³⁷ There is no evidence to suggest altering the suggested compression rate of 100 to 120/min in adults. Three studies have reported that depth decreases as rate increases, highlighting the pitfalls of evaluating a single CPR quality metric in isolation.^1–3

The ILCOR review³² identified 2 observational studies that provided inconsistent results on the association between chest compression release velocity and survival, with 1 study finding no association and the other finding that faster release velocity was associated with increased survival.^38,39 Not allowing complete chest wall recoil has been associated with increased intrathoracic pressure and decreased coronary perfusion.^40,41

CPR duty cycle refers to the proportion of time spent in compression relative to the total time of the compression plus decompression cycle. The 2010 Guidelines recommended a 50% duty cycle, in which the time spent in compression and decompression was equal, mainly on the basis of its perceived ease of being achieved in practice. Notably, in a clinical study in adults with out-of-hospital VF arrest (of whom 43% survived to hospital discharge), the mean duty cycle observed during resuscitation was 39%.⁴² A study in children also found the mean duty cycle was 40%, suggesting that shorter duty cycles may be the norm in clinical practice.⁴³ Although many animal studies have observed higher blood flows and better outcomes when the duty cycle was less than 50%, the optimal duty cycle is not known. Currently, there is insufficient evidence to warrant a change from the existing recommendation, which remains a knowledge gap that requires further investigation.

A 2020 ILCOR systematic review found that most studies did not find a significant association between real-time feedback and improved patient outcomes.⁴ However, no studies identified significant harm, and some demonstrated clinically important improvement in survival. One recent RCT reported a 25.6% increase in survival to hospital discharge from IHCA with audio feedback on compression depth and recoil (54% versus 28.4%; P<0.001).⁴⁵

An analysis of data from the AHA’s Get With The Guidelines-Resuscitation registry showed higher likelihood of ROSC (odds ratio, 1.22; 95% CI, 1.04–1.34; P=0.017) when CPR quality was monitored using either ETCO₂ or diastolic blood pressure.⁴⁶ An observational study in adult patients (IHCA and OHCA) reported that for every 10 mm compression depth increase, ETCO₂ increased 1.4 mm Hg.⁴⁷ A 2018 systematic review of ETCO₂ as a prognostic indicator for ROSC⁴⁸ found variability in cutoff values, but less than 10 mm Hg was generally associated with poor outcome and greater than 20 mm Hg had a stronger association with ROSC than a value of greater than 10 mm Hg. The combination of the association of higher ETCO₂ with ROSC and the finding that increased chest compression depth can increase ETCO₂ suggests that targeting compressions to a value of at least 10 mm Hg, and ideally 20 mm Hg or greater, may be useful. The validity and reliability of ETCO₂ in nonintubated patients is not well established. When available, invasive arterial blood pressure monitoring may also help assess and guide CPR efforts. The use of diastolic blood pressure monitoring during cardiac arrest was associated with higher ROSC,⁴⁶ but there are inadequate human data to suggest any specific pressure.

Studies have reported that enough tidal volume to cause visible chest rise, or approximately 500 to 600 mL, provides adequate ventilation while minimizing the risk of overdistension or gastric insufflation.^6–9

Both mouth-to-mouth rescue breathing and bag-mask ventilation provide oxygen and ventilation to the victim.¹⁰ To provide mouth-to-mouth rescue breaths, open the victim’s airway, pinch the victim’s nose, create an airtight mouth-to-mouth seal, and provide a breath.

Taking a regular rather than a deep breath prevents the rescuer from getting dizzy or light-headed and prevents overinflation of the victim’s lungs. The most common cause of ventilation difficulty is an improperly opened airway,¹¹ so if the victim’s chest does not rise with the first rescue breath, reposition the head by performing the head tilt–chin lift again and then give the second rescue breath. The recommendation for 1 second is to keep the pauses in CPR as brief as possible.

Excessive ventilation is unnecessary and can cause gastric inflation, regurgitation, and aspiration.^12,14 Excessive ventilation can also be harmful by increasing intrathoracic pressure, decreasing venous return to the heart, and diminishing cardiac output and survival.¹⁴

Mouth-to-nose ventilation may be necessary if ventilation through the victim’s mouth is impossible because of trauma, positioning, or difficulty obtaining a seal. A case series suggests that mouth-to-nose ventilation in adults is feasible, safe, and effective.¹⁶

Effective ventilation of the patient with a tracheal stoma may require ventilation through the stoma, either by using mouth-to-stoma rescue breaths or by use of a bag-mask technique that creates a tight seal over the stoma with a round, pediatric face mask. There is no published evidence on the safety, effectiveness, or feasibility of mouth-to-stoma ventilation. One study of patients with laryngectomies showed that a pediatric face mask created a better peristomal seal than a standard ventilation mask.¹⁷

Since the last review in 2010 of rescue breathing in adult patients, there has been no evidence to support a change in previous recommendations. A study in critically ill patients who required ventilatory support found that bag-mask ventilation at a rate of 10 breaths per minute decreased hypoxic events before intubation.¹⁸

A 2017 ILCOR systematic review found that a ratio of 30 compressions to 2 breaths was associated with better survival than alternate ratios, a recommendation that was reaffirmed by the AHA in 2018.^19,20 Most of these studies examined “bundles” of cardiac arrest care, making it impossible to know if the improvement was due to the compression-to-ventilation ratio itself. This ratio is supported by a large OHCA RCT in which the use of 30:2 (with a pause in compressions of less than 5 seconds) was at least as good as continuous chest compressions.²¹

In a large trial, survival and survival with favorable neurological outcome were similar in a group of patients with OHCA treated with ventilations at a rate of 10/min without pausing compressions, compared with a 30:2 ratio before intubation.²¹

A 2017 systematic review identified 1 observational human study and 10 animal studies comparing different ventilation rates after advanced airway placement.²² No clear benefit from a rate of 10 was identified, but no other rate was found to be superior. A 2017 ILCOR systematic review did not identify any new evidence to alter this recommendation, which was reiterated in the “2017 AHA Focused Update on Adult BLS and CPR Quality: An Update to the AHA Guidelines for CPR and Emergency Cardiovascular Care.”^19,20

A 2017 ILCOR systematic review concluded that although the evidence from observational studies supporting the use of bundles of care including minimally interrupted chest compressions was of very low certainty (primarily unadjusted results), systems already using such an approach may continue to do so.¹⁹

Emergent electric cardioversion and defibrillation are highly effective at terminating VF/VT and other tachyarrhythmias. No shock waveform has distinguished itself as achieving a consistently higher rate of ROSC or survival. Biphasic and monophasic shock waveforms are likely equivalent in their clinical outcome efficacy.³

No shock waveform has proved to be superior in improving the rate of ROSC or survival. However, biphasic waveform defibrillators (which deliver pulses of opposite polarity) expose patients to a much lower peak electric current with equivalent or greater efficacy for terminating atrial⁴ and ventricular tachyarrhythmias than monophasic (single polarity) defibrillators do.^5–10,13 These potential differences in safety and efficacy favor preferential use of a biphasic defibrillator, when available. Biphasic defibrillators have largely replaced monophasic shock defibrillators, which are no longer manufactured.

The rationale for a single shock strategy, in which CPR is immediately resumed after the first shock rather than after serial “stacked” shocks (if required) is based on a number of considerations. These include the high success rate of the first shock with biphasic waveforms (lessening the need for successive shocks), the declining success of immediate second and third serial shocks when the first shock has failed,¹⁴ and the protracted interruption in CPR required for a series of stacked shocks. A single shock strategy results in shorter interruptions in CPR and a significantly improved survival to hospital admission and discharge (although not 1-year survival) compared with serial “stacked” shocks.^15–17 It is unknown whether stacked shocks or single shocks are more effective in settings of a monitored witnessed arrest (for example, see the section on Cardiac Arrest After Cardiac Surgery).

Regardless of waveform, successful defibrillation requires that a shock be of sufficient energy to terminate VF/VT. In cases where the initial shock fails to terminate VF/VT, subsequent shocks may be effective when repeated at the same or an escalating energy setting.^18,19 An optimal energy setting for first or subsequent biphasic defibrillation, whether fixed or escalating, has not been identified, and its selection can be based on the defibrillator’s manufacturer specification.

There is no conclusive evidence of superiority of one biphasic shock waveform over another for defibrillation.²⁰ Given the variability in electric characteristics between proprietary biphasic waveforms, it is reasonable to use the energy settings specified by the manufacturer for that specific device. If a manufacturer’s specified energy setting for defibrillation is not known at the time of intended use, the maximum dose setting for that device may be considered.

Commercially available defibrillators either provide fixed energy settings or allow for escalating energy settings; both approaches are highly effective in terminating VF/VT.¹⁸ An optimal energy setting for first or subsequent biphasic defibrillation, whether fixed or escalating, has not been identified and is best deferred to the defibrillator’s manufacturer. A randomized trial comparing fixed 150 J biphasic defibrillation with escalating higher shock energies (200–300–360 J) observed similar rates of successful defibrillation and conversion to an organized rhythm after the first shock. However, among patients who required multiple shocks, escalating shock energy resulted in a significantly higher rate of conversion to an organized rhythm, although overall survival did not differ between the 2 treatment groups.¹⁹ When VF/VT is refractory to the first shock, an equivalent or higher energy setting than the first shock may be considered. As yet, there is no conclusive evidence of superiority of one biphasic shock waveform over another for defibrillation.²⁰ It is reasonable to use the energy settings specified by the manufacturer for that specific device. If a manufacturer’s specified energy setting for defibrillation is not known at the time of intended use, the maximum dose setting for that device may be considered.

Anterolateral, anteroposterior, anterior-left infrascapular, and anterior-right infrascapular electrode placements are comparably effective for treating supraventricular and ventricular arrhythmias.^24–28 A larger pad/paddle size (within the limits of 8–12 cm in diameter) lowers transthoracic impedance.^29,30 Self-adhesive pads have largely replaced defibrillation paddles in clinical practice. Before pad placement, remove all clothing and jewelry from the chest.

AEDs are highly accurate in their detection of shockable arrhythmias but require a pause in CPR for automated rhythm analysis.^32,33 Manual defibrillation can result in a shorter hands-off period for rhythm confirmation in operators with a sufficient skill for rapid and reliable rhythm interpretation.^34,35

CPR is the single-most important intervention for a patient in cardiac arrest and should be provided until a defibrillator is applied to minimize interruptions in compressions.

When VF/VT has been present for more than a few minutes, myocardial reserves of oxygen and other energy substrates are rapidly depleted. If replenished by a period of CPR before shock, defibrillation success improves significantly.^1,2,36,37 Because no differences in outcome were seen in studies comparing short (typically approximately about 30 seconds) with prolonged (up to 3 minutes) periods of CPR preceding the initial rhythm analysis, a brief period of CPR while the defibrillator is readied for use may be sufficient in unmonitored cardiac arrest.^38–40 Even in monitored arrests, it can take time to attach pads, power on a defibrillator, and charge the capacitor before shock delivery, during which there is good reason to administer CPR.

Early defibrillation improves outcome from cardiac arrest.^41–43 When VF is of short duration, myocardial reserves of oxygen and other energy substrates are likely to remain intact. During this early electric phase, the rhythm is most responsive to defibrillation.^44,45 Thus, if the onset of VF is monitored or witnessed with a defibrillator that is already applied, or to which there is immediate access, it is reasonable to administer a shock as soon as possible. Interim CPR should be provided if there is any delay in obtaining or readying the defibrillator for use.

There are differing approaches to charging a manual defibrillator during resuscitation. It is not uncommon for chest compressions to be paused for rhythm detection and continue to be withheld while the defibrillator is charged and prepared for shock delivery. This approach results in a protracted hands-off period before shock. Precharging the defibrillator during ongoing chest compressions shortens the hands-off chest time surrounding defibrillation, without evidence of harm.⁴⁷ Although no study has directly evaluated the effect of precharging itself on cardiac arrest outcome, shorter perishock pauses (which could result from such a strategy) are associated with improved survival from VF arrest.⁴⁸ Two approaches are reasonable: either charging the defibrillator before a rhythm check or resuming compressions briefly after a rhythm check while the defibrillator charges. Either approach may reduce no-flow time.^49,50

Immediate resumption of chest compressions after shock results in a shorter perishock pause and improves the overall hands-on time (chest compression fraction) during resuscitation, which is associated with improved survival from VF arrest.^16,48 Even when successful, defibrillation is often followed by a variable (and sometimes protracted) period of asystole or pulseless electrical activity, during which providing CPR while awaiting a return of rhythm and pulse is advisable. Whether resumption of CPR immediately after shock might reinduce VF/VT is controversial.^52–54 This potential concern has not been borne out by any evidence of worsened survival from such a strategy. Should there be physiological evidence of return of circulation such as an arterial waveform or abrupt rise in ETCO₂ after shock, a pause of chest compressions briefly for confirmatory rhythm analysis may be warranted.

CPR obscures interpretation of the underlying rhythm because of the artifact created by chest compressions on the ECG. This makes it difficult to plan the next step of care and can potentially delay or even misdirect drug therapies if given empirically (blindly) based on the patient’s presumed, but not actual, underlying rhythm. Time taken for rhythm analysis also disrupts CPR. Artifact-filtering and other innovative techniques to disclose the underlying rhythm beneath ongoing CPR can surmount these challenges and minimize interruptions in chest compressions while offering a diagnostic advantage to better direct therapies.^55–60 Despite the theoretical advantages, no study has evaluated these technologies in a real-time clinical setting or validated their clinical effectiveness compared to current resuscitation strategies. At present, filtering algorithms are strictly used for visual (manual) rhythm interpretation and not for automated VF/VT rhythm detection in AEDs during ongoing CPR. This added potential application remains untested. Recognizing the need for further clinical research, a 2020 ILCOR systematic review recommended against adopting artifact-filtering algorithms for rhythm analysis during CPR at the present time.⁵¹ The writing group also endorses the need for further investigation and clinical validation before these technologies are adopted into clinical practice.

The electric characteristics of the VF waveform are known to change over time.⁶¹ VF waveform analysis may be of value in predicting the success of defibrillation or other therapies during the course of resuscitation.^62–64 The prospect of basing therapies on a prognostic analysis of the VF waveform in real-time is an exciting and developing avenue of new research. However, the validity, reliability, and clinical effectiveness of an approach that prompts or withholds shock or other therapies on the basis of predictive analyses is currently uncertain. The only prospective clinical trial comparing a standard shock-first protocol with a waveform analysis-guided shock algorithm observed no differences in outcome.⁶⁵ The consensus of the writing group is that there is currently insufficient evidence to support the routine use of waveform analysis to guide resuscitation care, but it is an area in which further research with clinical validation is needed and encouraged.

There is limited evidence examining double sequential defibrillation in clinical practice. A number of case reports have shown good outcomes in patients who received double sequential defibrillation. However, these case reports are subject to publication bias and should not be used to support its effectiveness.⁶⁷ A handful of observational studies demonstrated no difference in outcomes (ROSC, survival, neurological outcome) with the use of double sequential defibrillation compared with standard defibrillation.^68–71 These studies should also be interpreted with caution, because the use of double sequential defibrillation was not protocolized and was often used late in the resuscitation after standard resuscitation was unsuccessful. Published reports also do not distinguish the application of double sequential defibrillation for truly shock-refractory (incessant) VF versus VF that recurs during the period of CPR after a successful shock, which is the more common clinical scenario.^3,7 A 2020 ILCOR systematic review found no evidence to support double sequential defibrillation and recommended against its routine use compared with standard defibrillation.⁵¹ A recent pilot RCT (not included in the systematic review) of 152 patients who remained in VF after at least 3 shocks found higher rates of VF termination and ROSC with double sequential defibrillation or alternative defibrillator pad placement compared with standard defibrillation but was not powered for these outcomes and did not report patient survival.⁷² A number of unanswered questions remain about double sequential defibrillation, including intershock timing, pad positioning, technique, and the possibility of harm with increased energy and defibrillator damage.^73,74 It is premature for double sequential defibrillation to be incorporated into routine clinical practice given the lack of evidence. Its usefulness should be explored in the context of clinical trials. An ongoing RCT (NCT04080986) may provide answers to some of these questions.

Existing evidence, including observational and quasi-RCT data, suggests that pacing by a transcutaneous, transvenous, or transmyocardial approach in cardiac arrest does not improve the likelihood of ROSC or survival, regardless of the timing of pacing administration in established asystole, location of arrest (in-hospital or out-of-hospital), or primary cardiac rhythm (asystole, pulseless electrical activity).^2–6 Protracted interruptions in chest compressions while the success of pacing is assessed can also be detrimental to survival. It is not known whether the timing of pacing initiation may influence pacing success such that pacing may be useful in the initial seconds of select cases of witnessed, monitored cardiac arrest (see the section on Cardiac Arrest After Cardiac Surgery). If pacing is attempted during cardiac arrest related to the special circumstances described above, providers are cautioned against its performance at the expense of high-quality CPR, particularly when assessing electric and mechanical capture.

1 and 2. The intent of precordial thump is to transmit the mechanical force of the “thump” to the heart as electric energy analogous to a pacing stimulus or very low-energy shock (depending on its force) and is referred to as electromechanical transduction.¹ There is no evidence that the use of precordial thump during routine cardiac arrest care in the out-of-hospital or in-hospital settings improves rates of ROSC or survival to hospital discharge.^8–12 It may be beneficial only at the very early onset of VT when the arrhythmia is most vulnerable to lower-energy termination such as in responder-witnessed, monitored events, or in a controlled laboratory environment, but even then it is rarely effective.¹³ Although there are case reports of success without evidence of harm from a precordial thump,^9,14,15 if fortuitously administered on the electrically vulnerable portion of an organized rhythm (T wave), the thump (like an unsynchronized shock) risks acceleration or conversion of the rhythm to VF,^16–19 analogous to commotio cordis.²⁰ Thus, although the thump may be useful as a single brief intervention under specific circumstances (ie, when a cardiac arrest is witnessed by the responder and monitor-confirmed to be due to VF/VT and a defibrillator is not readily available for use), it should not delay CPR or deployment of a defibrillator.

Fist, or percussion, pacing is administered with the goal of stimulating an electric impulse sufficient to cause depolarization and contraction of the myocardium, resulting in a pulse. There are a number of case reports and case series that examined the use of fist pacing during asystolic or “life-threatening bradycardic” events^1,22–25 showing favorable outcomes of survival²² and ROSC.²³ None of these studies, however, were controlled or comparative, and it is not known if the use of fist pacing itself improves rates of ROSC or survival compared with standard therapy. There is no role for fist pacing in patients in cardiac arrest.

It is important to underscore that while cough CPR by definition cannot be used for an unconscious patient, it can be harmful in any setting if diverting time, effort, and attention from performing high-quality CPR. Cough CPR is described as a repetitive deep inspiration followed by a cough every few seconds before the loss of consciousness. It is feasible only at the onset of a hemodynamically significant arrhythmia in a cooperative, conscious patient who has ideally been previously instructed on its performance, and as a bridge to definitive care. There are no studies comparing cough CPR to standard resuscitation care. Limited evidence from case reports and case series demonstrates transient increases in aortic and intracardiac pressure with the use of cough CPR at the onset of tachyarrhythmias or bradyarrhythmias in conscious patients.^10,26–28 These studies suffer from considerable selection bias and lack of comparison groups, and do not control for the confounding effect of other treatments, making them hard to interpret.

The peripheral IV route has been the traditional approach to vascular access for emergency drug and fluid administration during resuscitation. The pharmacokinetic properties, acute effects, and clinical efficacy of emergency drugs have primarily been described when given intravenously.^3–6 The IV route has precedence, is usually accessible, and affords a potentially more predictable drug response, making it a reasonable initial approach for vascular access.

The paucity of information on the efficacy of IO drug administration during CPR was acknowledged in 2010, but since then the IO route has grown in popularity. IO access is increasingly implemented as a first-line approach for emergent vascular access. A 2020 ILCOR systematic review⁷ comparing IV versus IO (principally pretibial placement) drug administration during cardiac arrest found the IV route was associated with better clinical outcomes compared with IO in 5 retrospective studies.^8–12 There were significant concerns for bias, particularly due to the fact that need for IO placement may indicate patient or arrest characteristics that are also risk factors for poor outcome. Subgroup analyses of IV versus IO route from 2 RCTs were also included in this systematic review. In these, no statistically significant effect modification by route of administration was identified. Point estimates favored IV access except for the outcome of ROSC in the PARAMEDIC2 trial, where the effect of epinephrine was similar regardless of route.^13,14 Site specificity may also be an issue with IO administration, because IO access was nearly always pretibial in these studies. On the basis of these results, the writing group concluded that establishing a peripheral IV remains a reasonable initial approach, but IO access may be considered when an IV is not successful or feasible. Further research is needed to assess the efficacy of drugs delivered intravenously as compared with intraosseously (tibial and humeral).

Drug administration by central venous access (by internal jugular or subclavian vein) achieves higher peak concentrations and more rapid circulation times than drugs administered by peripheral IV do,^15–17 but there are currently no data comparing clinical outcomes between these access routes. Central access is associated with higher morbidity, takes time to perform, and may also require interruption of CPR. Current use of this approach is largely in the hospital and may be considered by skilled providers when IV and IO access are not successful or feasible.

Endotracheal drug administration is regarded as the least-preferred route of drug administration because it is associated with unpredictable (but generally low) drug concentrations^18–20 and lower rates of ROSC and survival.²¹

The suggestion to administer epinephrine was strengthened to a recommendation based on a systematic review and meta-analysis,³ which included results of 2 randomized trials of epinephrine for OHCA, 1 of which included over 8000 patients,^1,2 showing that epinephrine increased ROSC and survival. At 3 months, the time point felt to be most meaningful for neurological recovery, there was a nonsignificant increase in survivors with both favorable and unfavorable neurological outcome in the epinephrine group.² Any drug that increases the rate of ROSC and survival, but is given after several minutes of downtime, will likely increase both favorable and unfavorable neurological outcome. Determining the likelihood of favorable or unfavorable neurological outcome at the time of arrest is currently not feasible. Therefore, continuing to use a drug that has been shown to increase survival, while focusing our broader efforts on shortening time to drug for all patients so that more survivors will have a favorable neurological outcome, seems the most beneficial approach.

The existing trials have used a protocol of 1 mg every 3 to 5 minutes. Operationally, administering epinephrine every second cycle of CPR, after the initial dose, may also be reasonable.

Of 16 observational studies on timing in the recent systematic review, all found an association between earlier epinephrine and ROSC for patients with nonshockable rhythms, although improvements in survival were not universally seen.³

For shockable rhythms, trial protocols have directed that epinephrine be given after the third shock. The literature supports prioritizing defibrillation and CPR initially and giving epinephrine if initial attempts with CPR and defibrillation are not successful.³

The recent systematic review³ found no difference in outcomes in trials comparing vasopressin alone or vasopressin combined with epinephrine to epinephrine alone for cardiac arrest, although these studies were underpowered.

Multiple RCTs have compared high-dose with standard-dose epinephrine, and although some have shown higher rates of ROSC with high-dose epinephrine, none have shown improvement in survival to discharge or any longer-term outcomes.^4–11

Administration of amiodarone or lidocaine to patients with OHCA was last formally reviewed in 2018¹ and demonstrated improved survival to hospital admission but did not improve overall survival to hospital discharge or survival with good neurological outcome.^1,2 However, amiodarone and lidocaine each significantly improved survival to hospital discharge in a prespecified subgroup of patients with bystander-witnessed arrest, potentially arguing for a time-dependent benefit and a group for whom these drugs may be more useful. Other antiarrhythmic agents were not specifically addressed in the most recent evidence review and merit further evaluation. These include bretylium tosylate, which was recently reintroduced in the United States for treatment of immediately life-threatening ventricular arrhythmias but without any new information on its effectiveness or safety.³ Sotalol requires administration as a slow infusion, rendering it impractical to use in cardiac arrest.⁴ Similar limitations also apply to procainamide, although it has been given by rapid infusion as a second-line agent in cardiac arrest, with uncertain benefit.⁵ The efficacy of antiarrhythmic drugs when given in combination for cardiac arrest has not been systematically addressed and remains a knowledge gap. The role of prophylactic antiarrhythmic medications on ROSC after successful defibrillation is also uncertain. Though not associated with improved survival to hospital discharge, lidocaine decreased the recurrence of VF/pVT when administered prophylactically after successful defibrillation and ROSC.⁶ The “2018 AHA Focused Update on Advanced Cardiovascular Life Support Use of Antiarrhythmic Drugs During and Immediately After Cardiac Arrest: An Update to the AHA Guidelines for CPR and Emergency Cardiovascular Care”¹ concluded that lidocaine use could be considered in specific circumstances (such as during EMS transport) when treatment of recurrent VF/pVT might be compromised. There is no evidence addressing the use of other antiarrhythmic drugs for this specific indication.

Two randomized trials from the same center reported improved survival and neurological outcome when steroids were bundled in combination with vasopressin and epinephrine during cardiac arrest and also administered after successful resuscitation from cardiac arrest.^7,8 However, nonrandomized studies of strictly intra-arrest corticosteroid administration, in addition to standard resuscitation, show mixed outcomes.^9,10 Due to the only studies suggesting benefit being from a single center with a bundled intervention, and observational data having conflicting results, whether steroids are beneficial during cardiac arrest remains unclear. At least 1 trial attempting to validate the findings of Mentzelopoulos et al is ongoing (NCT03640949).

Since last addressed by the 2010 Guidelines, a 2013 systematic review found little evidence to support the routine use of calcium in undifferentiated cardiac arrest, though the evidence is very weak due to lack of clinical trials and the tendency to use calcium as a “last resort” medication in refractory cardiac arrest.¹¹ Administration of calcium in special circumstances such as hyperkalemia and calcium blocker overdose is addressed in Electrolyte Abnormalities and in Toxicity: β-Adrenergic Blockers and Calcium Channel Blockers.

Clinical trials and observational studies since the 2010 Guidelines have yielded no new evidence that routine administration of sodium bicarbonate improves outcomes from undifferentiated cardiac arrest and evidence suggests that it may worsen survival and neurological recovery.^12–14 Use of sodium bicarbonate in special circumstances such as hyperkalemia and drug overdose is addressed in Electrolyte Abnormalities and in Toxicity: Sodium Channel Blockers, Including Tricyclic Antidepressants.

Magnesium’s role as an antiarrhythmic agent was last addressed by the 2018 focused update on advanced cardiovascular life support (ACLS) guidelines.¹ RCTs have not found it to improve ROSC, survival, or neurological outcome regardless of the presenting cardiac arrest rhythm,^15–18 nor useful for monomorphic VT.¹⁹ There are anecdotal reports and small case series attesting to magnesium’s efficacy in the treatment of torsades de pointes (See Torsades de Pointes).

Point-of-care cardiac ultrasound can identify cardiac tamponade or other potentially reversible causes of cardiac arrest and identify cardiac motion in pulseless electrical activity.^1,2 However, cardiac ultrasound is also associated with longer interruptions in chest compressions.³ A single small RCT found no improvement in outcomes with the use of cardiac ultrasound during CPR.⁴

No adult human studies directly compare levels of inspired oxygen concentration during CPR. A small number of studies has shown that higher Pao₂ during CPR is associated with ROSC, but this is likely due to differences in patients or resuscitation quality.^5–7

Observational studies have found that increases in ETCO₂ of more than 10 mm Hg may indicate ROSC, although no specific cutoff value indicative of ROSC has been identified.⁸

Arterial Po₂ and Pco₂ values are dependent on cardiac output and ventilation and therefore will depend on both patient characteristics and CPR quality. One small study found wide discrepancies in blood gases between mixed venous and arterial samples during CPR and concluded that arterial samples are not accurate during resuscitation.⁹

If an arterial line is in place, an abrupt increase in diastolic pressure or the presence of an arterial waveform during a rhythm check showing an organized rhythm may indicate ROSC.

The BLS TOR rule recommends TOR when all of the following criteria apply before moving to the ambulance for transport: (1) arrest was not witnessed by EMS providers or first responder; (2) no ROSC obtained; and (3) no shocks were delivered. In a recent meta-analysis of 7 published studies (33 795 patients), only 0.13% (95% CI, 0.03%–0.58%) of patients who fulfilled the BLS termination criteria survived to hospital discharge.³

The ALS TOR rule recommends TOR when all of the following criteria apply before moving to the ambulance for transport: (1) arrest was not witnessed; (2) no bystander CPR was provided; (3) no ROSC after full ALS care in the field; and (4) no AED shocks were delivered. In a recent meta-analysis of 2 published studies (10 178 patients), only 0.01% (95% CI, 0.00%–0.07%) of patients who fulfilled the ALS termination criteria survived to hospital discharge.³

The BLS TOR rule, otherwise known as the universal TOR rule (arrest not witnessed by EMS providers; no shock delivered; no ROSC), has been prospectively validated in combined BLS and ALS systems.⁴ Although the rule did not have adequate specificity after 6 minutes of resuscitation (false-positive rate: 2.1%) it did achieve better than 99% specificity after approximately 15 minutes of attempted resuscitation, while still reducing transportation by half. A retrospective analysis found that application of the universal TOR at 20 minutes of resuscitation was able to predict futility, identifying over 99% of survivors and patients with good neurological outcome.⁵

In intubated patients, an ETCO₂ measurement less than 10 mm Hg indicates low to no blood flow. Several small studies provide evidence showing that an ETCO₂ less than 10 mm Hg after 20 minutes of ALS resuscitation is strongly but not perfectly predictive of futility.^6–9 These small observational studies suffer from high risk of bias. Alternative ETCO₂ thresholds and timepoints have been proposed. The use of ETCO₂ alone to predict patient outcome needs to be validated in a large prospective study.

A recent systematic review found that no sonographic finding had consistently high sensitivity for clinical outcomes to be used as the sole criterion to terminate cardiac arrest resuscitation.¹⁰ Although some findings demonstrated higher ranges of sensitivity and/or specificity, studies examining the use of point-of-care ultrasound during cardiac arrest demonstrate varying results and are hindered by significant bias. There is considerable heterogeneity between studies in terms of timing and application of point-of-care ultrasound as well as inconsistent definitions and terminology in terms of cardiac motion. Further there is little research examining the interrater reliability of ultrasound findings during cardiac arrest.^11,12 In addition, see Adjuncts to CPR for ultrasound as an adjunct to CPR.

No studies were found that specifically examined the use of ETCO₂ in cardiac arrest patients without an advanced airway. It is not known whether ETCO₂ values during bag-mask ventilation are as reliable as those with an advanced airway in place. Because of the lack of evidence, there is nothing to support using any cutoff value of ETCO₂ for decisions about TOR in a nonintubated patient.

One large RCT in OHCA comparing bag-mask ventilation with endotracheal intubation (ETI) in a physician-based EMS system showed no significant benefit for either technique for 28-day survival or survival with favorable neurological outcome.² The success rate of ETI in this study was 98%, suggesting a relatively optimal setting for the potential success of ETI as an intervention. Further research is required to determine equivalence or superiority between the 2 approaches for acute airway management.

1, 2, and 3. One RCT in OHCA comparing SGA (with iGel) to ETI in a non–physician-based EMS system (ETI success, 69%) found no difference in survival or survival with favorable neurological outcome at hospital discharge.³ A second RCT in OHCA comparing SGA (with laryngeal tube) with ETI in a non–physician-based EMS system (ETI success, 52%) found both better survival to hospital discharge and better survival to hospital discharge with good neurological outcome in the patients managed with SGA.⁴ These results are challenging to contextualize because they both allowed for provider deviation from protocol based on clinical judgment. Additionally, precise thresholds for high or low tracheal intubation success rates have not been identified, though guidance can be taken from the existing clinical trials. Thus, it is difficult to understand the potential benefit (or harm), per individual, that drove the decision to place the specific advanced airway device. The decision on placement of an advanced airway requires an understanding of patient and provider characteristics that are not easily defined in a global recommendation. Because of a paucity of studies on advanced airway management for IHCA, the IHCA recommendations are extrapolated from OHCA data. Based on these issues, there is a need for further research specifically on the interface between patient factors and the experience, training, tools, and skills of the provider. Given these reasons, a recommendation for SGA in preference to ETI would be premature.

To maintain provider skills from initial training, frequent retraining is important.^5,6 However, future research will need to address the specific type, amount, and duration between training experiences.

Although an advanced airway can be placed without interrupting chest compressions,⁷ unfortunately, such interruptions still occur. Therefore, providers should weigh the potential benefits of an advanced airway with the benefits of maintaining a high chest compression fraction.^8–10

In a small clinical trial and several observational studies, waveform capnography was 100% specific for confirming endotracheal tube position during cardiac arrest.^11–13 The sensitivity of waveform capnography decreases after a prolonged cardiac arrest.^11–13 The use of waveform capnography to assess the placement of other advanced airways (eg, Combitube, laryngeal mask airway) has not been studied.

The rationale for tracking the overall success rate for systems performing ETI is to make informed decisions as to whether practice should allow for ETI, move toward SGA, or simply use bag-mask ventilation for patients in cardiac arrest; recommendations will vary depending on the overall success rate in a given system.

1 and 2. Studies of mechanical CPR devices have not demonstrated a benefit when compared with manual CPR, with a suggestion of worse neurological outcome in some studies. In the ASPIRE trial (1071 patients), use of the load-distributing band device was associated with similar odds of survival to hospital discharge (adjusted odds ratio [aOR], 0.56; CI, 0.31–1.00; P=0.06), and worse survival with good neurological outcome (3.1% versus 7.5%; P=0.006), compared with manual CPR.³ In the CIRC trial (n=4231), use of load-distributing band–CPR resulted in statistically equivalent rates of survival to hospital discharge (aOR, 1.06; CI, 0.83–1.37) and survival with good neurological outcome (aOR, 0.80; CI, 0.47–1.37).⁴ In the PARAMEDIC trial (n=4470), use of a mechanical piston device produced similar rates of 30-day survival (aOR, 0.86; CI, 0.64–1.15), and worse survival with good neurological outcome (aOR, 0.72; CI, 0.52–0.99), compared with manual CPR.⁵ In the LINC trial (n=2589), survival with good neurological outcome was similar in both groups (8.3% versus 7.8%; risk difference, 0.55%; 95% CI, –1.5% to 2.6%).⁶

A 2013 Cochrane review of 10 trials comparing ACD-CPR with standard CPR found no differences in mortality and neurological function in adults with OHCA or IHCA.⁸ An important added consideration with this modality is that of increased rescuer fatigue, which could impair the overall quality of CPR.

ACD-CPR and ITD may act synergistically to enhance venous return during chest decompression and improve blood flow to vital organs during CPR. The ResQTrial demonstrated that ACD plus ITD was associated with improved survival to hospital discharge with favorable neurological function for OHCA compared with standard CPR, though this study was limited by a lack of blinding, different CPR feedback elements between the study arms (ie, cointervention), lack of CPR quality assessment, and early TOR.^9,10 The 2015 AHA Guidelines Update for CPR and Emergency Cardiovascular Care⁷ evaluated this topic and noted that though a large RCT of low-quality demonstrated benefit of its use, additional trials were needed to confirm the results because of study limitations noted. Thus, ACD-CPR plus ITD was not recommended in previous versions of the AHA Guidelines. However, in settings where the equipment and trained personnel are available, ACD-CPR plus ITD could be an alternative to standard CPR.

In the PRIMED study (n=8178), the use of the ITD (compared with a sham device) did not significantly improve survival to hospital discharge or survival with good neurological function in patients with OHCA.¹¹ Despite the addition of a post hoc analysis of the PRIMED trial for ITD,¹² the routine use of the ITD as an adjunct during conventional CPR is not recommended.

Interposed abdominal compression CPR is a 3-rescuer technique that includes conventional chest compressions combined with alternating abdominal compressions. The dedicated rescuer who provides manual abdominal compressions will compress the abdomen midway between the xiphoid and the umbilicus during the relaxation phase of chest compression. This topic was last reviewed in 2010 and identified 2 randomized trials, interposed abdominal compression CPR performed by trained rescuers improved short-term survival¹³ and survival to hospital discharge,¹⁴ compared with conventional CPR for adult IHCA. One RCT of adult OHCA¹⁵ did not show any survival advantage to interposed abdominal compression CPR. More evaluation is needed to further define the routine use of this technique.

There are no RCTs on the use of ECPR for OHCA or IHCA. Fifteen observational studies were identified for OHCA that varied in inclusion criteria, ECPR settings, and study design, with the majority of studies reporting improved neurological outcome associated with ECPR.² For ECPR use in the in-hospital setting, all studies were assessed as having very serious risk of bias (primarily due to confounding) and the overall certainty of evidence was rated as very low for all outcomes.² In 3 studies, ECPR was not associated with beneficial effects for short- or long-term neurological outcomes,^3–5 while 1 study⁶ did report associated short- and long-term neurological outcome benefit. Despite many studies reporting favorable outcomes with the use of ECPR, the vast majority of the studies are from single centers with varying inclusion criteria and settings, with decisions to perform ECPR made on a case-by-case basis. While there is currently no evidence to clearly define what should constitute “selected patients,” most of the studies analyzed included younger patients with fewer comorbidities. More data are clearly needed from studies of higher methodologic quality, including randomized trials.

Before embarking on empirical drug therapy, obtaining a 12-lead ECG and/or seeking expert consultation for diagnosis is encouraged, if available. If a regular wide-complex tachycardia is suspected to be paroxysmal SVT, vagal maneuvers can be considered before initiating pharmacological therapies (see Regular Narrow-Complex Tachycardia). Adenosine is an ultra–short-acting drug that is effective in terminating regular tachycardias when caused by AV reentry. Adenosine will not typically terminate atrial arrhythmias (such as atrial flutter or atrial tachycardia) but will transiently slow the ventricular rate by blocking conduction of P waves through the AV node, afford their recognition, and help establish the rhythm diagnosis. While ineffective in terminating ventricular arrhythmias, adenosine’s relatively short-lived effect on blood pressure makes it less likely to destabilize monomorphic VT in an otherwise hemodynamically stable patient. These features make adenosine relatively safe for treating a hemodynamically stable, regular, monomorphic wide-complex tachycardia of unknown type⁴ and as an aid in rhythm diagnosis, although its use is not completely without risk.^5,6

IV antiarrhythmic medications may be considered in stable patients with wide-complex tachycardia, particularly if suspected to be VT or having failed adenosine. Because of their longer duration of action, antiarrhythmic agents may also be useful to prevent recurrences of wide-complex tachycardia. Lidocaine is not included as a treatment option for undifferentiated wide-complex tachycardia because it is a relatively “narrow-spectrum” drug that is ineffective for SVT, probably because its kinetic properties are less effective for VT at hemodynamically tolerated rates than amiodarone, procainamide, or sotalol are.^7–10 In contrast, amiodarone, procainamide, and sotalol are “broader-spectrum” antiarrhythmics than lidocaine and can treat both SVT and VT, but they can cause hypotension. Since the 2010 Guidelines, a new branded bioequivalent formulation of amiodarone has become available for IV infusion with less hypotensive effects than the older generic formulation.¹¹ There are few direct comparisons of efficacy between amiodarone, procainamide, and sotalol themselves,¹² which the writing group felt were insufficient to favor one of these drugs over another, apart from cautioning about their use in patients with long QT, amiodarone in suspected pre-excited arrhythmias, or giving these drugs in combination without prior expert consultation. Any of these drugs can also worsen wide-complex tachycardia, converting it to an arrhythmia that is more rapid, less hemodynamically stable, or more malignant, such that availability of a defibrillator is encouraged when these drugs are administered.¹³

Verapamil is a calcium channel blocking agent that slows AV node conduction, shortens the refractory period of accessory pathways, and acts as a negative inotrope and vasodilator. Its effects are mediated by a different mechanism and are longer lasting than adenosine. Though effective for treating a wide-complex tachycardia known to be of supraventricular origin and not involving accessory pathway conduction, verapamil’s negative inotropic and hypotensive effects can destabilize VT¹⁴ and accelerate pre-excited atrial fibrillation and flutter.¹⁵ Similar concerns may also apply to other drugs commonly used to treat SVTs, such as diltiazem and β-adrenergic blockers, which are not addressed in this recommendation and require evidence review.

The combination of adenosine’s short-lived slowing of AV node conduction, shortening of refractoriness in the myocardium and accessory pathways, and hypotensive effects make it unsuitable in hemodynamically unstable patients and for treating irregularly irregular and polymorphic wide-complex tachycardias. Adenosine only transiently slows irregularly irregular rhythms, such as atrial fibrillation, rendering it unsuitable for their management. The drug’s hypotensive and tissue refractoriness–shortening effects can accelerate ventricular rates in polymorphic VT and, when atrial fibrillation or flutter are conducted by an accessory pathway, risk degeneration to VF.¹⁶ Thus, the drug is not recommended in hemodynamically unstable patients or for treating irregularly irregular or polymorphic wide-complex tachycardias.

When available, expert consultation can be helpful to assist in the diagnosis and management of treatment-refractory wide-complex tachycardia. Electric cardioversion can be useful either as first-line treatment or for drug-refractory wide-complex tachycardia due to reentry rhythms (such as atrial fibrillation, atrial flutter, AV reentry, and VT). However, electric cardioversion may not be effective for automatic tachycardias (such as ectopic atrial tachycardias), entails risks associated with sedation, and does not prevent recurrences of the wide-complex tachycardia. Notably, when the QRS complex is of uniform morphology, shock synchronized to the QRS is encouraged because this minimizes the risk of provoking VF by a mistimed shock during the vulnerable period of the cardiac cycle (T wave).¹⁸ In contrast, polymorphic wide-complex tachycardias cannot be synchronized reliably because of the differing characteristics of each QRS complex, and require high-energy defibrillation.¹⁹

Regardless of the underlying QT interval, all forms of polymorphic VT tend to be hemodynamically and electrically unstable. They may repeatedly recur and remit spontaneously, become sustained, or degenerate to VF, for which electric shock may be required. When the QRS complex of a VT is of uniform morphology, electric cardioversion with the shock synchronized to the QRS minimizes the risk of provoking VF by a mistimed shock during the vulnerable period of the cardiac cycle (T wave).¹⁰ In contrast, polymorphic VT cannot be synchronized reliably because of the differing characteristics of each QRS complex and requires high-energy unsynchronized defibrillation.¹¹ While effective in terminating polymorphic VT, electric shock may not prevent its recurrence, for which pharmacological therapies are often required and the primary focus of the ensuing recommendations

Torsades de pointes typically presents in a recurring pattern of self-terminating, hemodynamically unstable polymorphic VT in context of a known or suspected long QT abnormality, often with an associated bradycardia. Immediate defibrillation is the treatment of choice when torsades is sustained or degenerates to VF. However, termination of torsades by shock does not prevent its recurrence, which requires additional measures. In small case series, IV magnesium has been effective in suppressing and preventing recurrences of torsades.^13–16 Magnesium is believed to suppress early afterdepolarizations, which are fluctuations in the myocardial action potential that can trigger the salvos of VT seen in torsades.¹⁷ Correcting any electrolyte abnormalities, particularly hypokalemia, is also advisable. Torsades is not treatable with antiarrhythmic medications, which can themselves prolong the QT interval and promote the arrhythmia. When given acutely, β-adrenergic blockers can also precipitate torsades by causing or worsening bradycardia. In patients with bradycardia or pause-precipitated torsades, expert consultation is best sought for additional measures such as overdrive pacing or isoproterenol,^18–20 if needed. The use of magnesium in torsades de pointes was addressed by the 2010 Guidelines and updated in a 2018 focused update on ACLS guidelines,²¹ with an interim evidence review that identified no new information that would modify previous recommendations.

Polymorphic VT that is not associated with QT prolongation is often triggered by acute myocardial ischemia and infarction,^4,5 often rapidly degenerates into VF, and is treated similarly to other ventricular arrhythmias (VT and VF). However, termination of polymorphic VT with defibrillation may not prevent its recurrence, which often requires additional measures. No RCTs have been performed to determine the best practice for pharmacological management of polymorphic VT. However measures to treat myocardial ischemia (eg, β-adrenergic blockers or emergent coronary intervention) as well as lidocaine and amiodarone may be effective^22–29 in concert with defibrillation when the arrhythmia is sustained. β-Adrenergic blockers have also been shown to reduce the incidence of ventricular arrhythmias in acute coronary syndromes.^30,31 Expert consultation is advisable when other causes of polymorphic VT are suspected, for which β-adrenergic blockers and antiarrhythmics may also have efficacy.^6,32 This topic was last addressed by the 2010 Guidelines, with an interim evidence update that identified no new information that would modify previous recommendations. Newer defined diagnostic entities causing polymorphic VT merit future evidence evaluation.

In the absence of long QT, magnesium has not been shown to be effective in the treatment of polymorphic VT ¹³ or to afford benefit in the acute management of other ventricular tachyarrhythmias.¹⁶

1 and 2. Management of hemodynamically unstable patients with SVT must start with prompt restoration of sinus rhythm through the use of cardioversion. Cardioversion has been shown to be both safe and effective in the prehospital setting for hemodynamically unstable patients with SVT who had failed to respond to vagal maneuvers and IV pharmacological therapies.³ Cardioversion is advised in patients who present with hypotension, acutely altered mental status, signs of shock, chest pain, or acute heart failure. Though rare, cardioversion may also be necessary in stable patients with SVT. Most stable patients with SVT have high conversion success rates of 80% to 98% with pharmacological management (eg, adenosine, diltiazem).^4,5 However, if drugs fail to restore sinus rhythm, cardioversion is safe and effective for stable patients after adequate sedation and anesthesia.

Success rates for the Valsalva maneuver in terminating SVT range from 19% to 54%.⁷ Augmenting the Valsalva maneuver with passive leg raise is more effective.⁸ Caution is advised when deploying carotid massage in older patients given the potential thromboembolic risk.

The 2015 American College of Cardiology, AHA, and Heart Rhythm Society Guidelines evaluated and recommended adenosine as a first-line treatment for regular SVT because of its effectiveness, extremely short half-life, and favorable side-effect profile.⁶ A Cochrane systematic review of 7 RCTs (622 patients) found similar rates of conversion to sinus rhythm with adenosine or calcium channel blockers (90% versus 93%) and no significant difference in hypotension.² Adenosine may have profound effects in post–heart transplant patients and can cause severe bronchospasm in asthma patients.

Treatment of hemodynamically stable patients with IV diltiazem or verapamil have been shown to convert SVT to normal sinus rhythm in 64% to 98% of patients.^4,9–11 These agents are particularly useful in patients who cannot tolerate β-adrenergic blockers or who have recurrent SVT after treatment with adenosine. Caution should be taken to administer these medications slowly to decrease the potential for hypotension.¹¹ Diltiazem and verapamil are not appropriate in the setting of suspected systolic heart failure.⁶

Evidence for the effectiveness of β-adrenergic blockers in terminating SVT is limited. In a trial that compared esmolol with diltiazem, diltiazem was more effective in terminating SVT.⁵ Nonetheless, β-adrenergic blockers are generally safe, and it is reasonable to use them to terminate SVT in hemodynamically stable patients.⁶

1 and 2. Uncontrolled tachycardia may impair ventricular filling, cardiac output, and coronary perfusion while increasing myocardial oxygen demand. While an expeditious trial of medications and/or fluids may be appropriate in some cases, unstable patients or patients with ongoing cardiac ischemia with atrial fibrillation or atrial flutter need to be cardioverted promptly.^1–3 When making the decision for cardioversion, one should also consider whether the arrhythmia is the cause of the tachycardia. Potential exacerbation of rapid ventricular response by secondary causes (eg, sepsis) should be considered and may inform initial attempts at hemodynamic stabilization with pharmacotherapy. There are few data addressing these strategies in hemodynamically unstable patients. However, studies demonstrating hemodynamic benefits of successful cardioversion have been published.^4,5 In addition, risks of hypotension and hypoperfusion with use of negative inotropes have been demonstrated even in normotensive patients.^6–8 Hemodynamically unstable patients and those with ongoing cardiac ischemia are likely to benefit from the improved hemodynamic status associated with restoration of sinus rhythm and avoidance of hypotension caused by the alternative pharmacological therapies. Depending on the clinical scenario, patients cardioverted from atrial fibrillation or atrial flutter of 48 hours’ duration or longer are candidates for anticoagulation. Details about anticoagulation selection can be found elsewhere.²

3 and 4. The electric energy required to successfully cardiovert a patient from atrial fibrillation or atrial flutter to sinus rhythm varies and is generally less in patients with new-onset arrhythmia, thin body habitus, and when biphasic waveform shocks are delivered.^9–15 Obese patients may require greater energy.¹⁶ If initial cardioversion is unsuccessful, energy is increased in subsequent attempts. Less energy is generally required for atrial flutter than for atrial fibrillation.¹¹ Higher energies of 200 J or more are associated with improved first shock success and decreased total energy delivery. In addition, a retrospective analysis found that lower energy shocks were associated with higher risk of cardioversion-induced VF.¹⁷ Previous guidelines included a comparison of monophasic and biphasic waveforms. This recommendation now focuses primarily on biphasic waveforms. Recommended energy levels vary with different devices, reducing the validity of generalized recommendations. This topic requires further study with a comprehensive systematic review to better understand the optimal electric doses with current devices. The writing group assessment of the LOE as C-LD is consistent with the limited evidence using modern devices and energy waveforms.

1 and 2. Clinical trial evidence shows that nondihydropyridine calcium channel antagonists (eg, diltiazem, verapamil), β-adrenergic blockers (eg, esmolol, propranolol), amiodarone, and digoxin are all effective for rate control in patients with atrial fibrillation/flutter.^{6–8,19–23} Calcium channel blockers may be more effective than amiodarone, and cause more hypotension.⁶ Digoxin is rarely used in the acute setting because of slow onset of effect.^1,2

3. Based on limited case reports and small case series, there is concern that patients with concomitant preexcitation and atrial fibrillation or atrial flutter may develop VF in response to accelerated ventricular response after the administration of AV nodal blocking agents such as digoxin, nondihydropyridine calcium channel antagonists, β-adrenergic blockers, or IV amiodarone.^24–27 In this setting, cardioversion is recommended as the most appropriate management.

4. Because of their negative inotropic effect, nondihydropyridine calcium channel antagonists (eg, diltiazem, verapamil) may further decompensate patients with left ventricular systolic dysfunction and symptomatic heart failure. They may be used in patients with heart failure with preserved ejection fraction. β-Adrenergic blockers may be used in compensated patients with cardiomyopathy; however, they should be used with caution or avoided altogether in patients with decompensated heart failure. This recommendation is based on expert consensus and pathophysiologic rationale.^2,18,28 β-Adrenergic blockers may be used in patients with chronic obstructive pulmonary disease because multiple studies have shown no negative effects.²⁹