Part 4: Advanced Life Support

Introduction

All newly manufactured defibrillators currently deliver shocks using biphasic waveforms. Although it has not been shown conclusively in randomized clinical studies that biphasic defibrillators save more lives than monophasic defibrillators, biphasic defibrillators achieve higher first-shock success rates at lower energy levels and appear to cause less postshock myocardial dysfunction.^12,13

Consensus on Science

No new randomized trials of biphasic waveforms since 2010 were identified.

Treatment Recommendation

We recommend that a biphasic waveform (biphasic truncated exponential [BTE] or rectilinear-biphasic [RLB]) is used for both atrial and ventricular arrhythmias in preference to a monophasic waveform (strong recommendation, very-low-quality evidence). In the absence of biphasic defibrillators, monophasic defibrillators are acceptable.

Values, Preferences, and Task Force Insights

In making this strong recommendation, we place a high value on the reported higher first-shock success rate for termination of fibrillation with a biphasic waveform, the potential for less postshock myocardial dysfunction, and the existing 2010 CoSTR.^12,13 The task force acknowledges that many emergency medical services (EMS) systems and hospitals around the world continue to use older monophasic devices.

Introduction

The pulsed biphasic waveform that is used in clinical practice had not previously been reviewed in 2010. The single published study¹⁴ of this waveform used a non–impedance compensated waveform (ie, the current delivered is not adjusted for the impedance of the chest), whereas the waveform in clinical use is an impedance-compensated waveform (ie, the current delivered is adjusted for the impedance of the chest).

Consensus on Science

For the critical outcome of survival to hospital discharge, very-low-quality evidence (downgraded for very serious risk of bias and serious indirectness) from 1 cohort study (ie, no control group)¹⁴ with a total of 104 patients that used a 130 J-130 J-180 J pulsed biphasic waveform protocol documented a survival rate of 9.8%. This compares with a weighted average BTE survival rate of 33.1% at 150 to 200 J.¹⁴

For the important outcome of termination of fibrillation, the same very-low-quality evidence (downgraded for very serious risk of bias and serious indirectness) from 1 cohort study¹⁴ with a total of 104 patients documented first-shock termination rates at 130 J of 90.4% with a pulsed biphasic waveform, comparable with BTE waveforms (weighted average 91.8%) at 150 to 200 J.¹⁴

Treatment Recommendation

We recommend following the manufacturer’s instructions for first and subsequent shock energy levels for the pulsed biphasic waveform (strong recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this strong recommendation, we have placed a high value on following the manufacturer’s guidance in the absence of high-quality data to suggest otherwise. The available very-low-quality data showing the efficacy of a non–impedance compensated pulsed biphasic waveform do not enable direct comparison with other biphasic waveforms. In addition, no clinical studies attest to the efficacy of this waveform in its current impedance-compensated form.

Introduction

In 2010, it was concluded that it was reasonable to start at a selected energy level of 150 to 200 J for a BTE waveform, and no lower than 120 J for an RLB waveform for defibrillation of VF/pVT cardiac arrest, acknowledging that the evidence was limited.^12,13

Consensus on Science

For the important outcome of termination of VF/pVT, low-quality evidence (downgraded for imprecision and risk of bias, respectively) from a post hoc report from an RCT and a cohort study showed a first-shock success rate of 73 of 86 (85%) and 79 of 90 (87.8%), respectively, when using a 120 J initial shock with an RLB waveform.^15,16

Treatment Recommendations

We recommend an initial biphasic shock energy of 150 J or greater for BTE waveforms, and 120 J or greater for RLB waveforms (strong recommendation, very-low-quality evidence). If a monophasic defibrillator is used, we recommend an initial monophasic shock energy of 360 J (strong recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making these strong recommendations, the working group was keen to acknowledge manufacturer’s instructions and recognize that evidence for the optimal first-shock energy level was lacking. We also considered that although monophasic defibrillators are no longer manufactured, they are still used in many countries.

Introduction

In 2010, it was recommended that when defibrillation was required, a single shock should be provided with immediate resumption of chest compressions after the shock.^12,13 This recommendation was made for 2 reasons: (1) in an attempt to minimize perishock interruptions to chest compressions, and (2) because it was thought that with the greater efficacy of biphasic shocks, if a biphasic shock failed to defibrillate, a further period of chest compressions could be beneficial. It was acknowledged that there was no clinical evidence to support improved outcomes from this strategy.

Consensus on Science

For the critical outcome of survival to 1 year, we have identified low-quality evidence (downgraded for serious risk of bias and serious indirectness) from 1 RCT enrolling 845 OHCA patients showing no difference in single versus 3 stacked shocks (odds ratio [OR], 1.64; 95% confidence interval [CI], 0.53–5.06).¹⁷

For the critical outcome of survival to hospital discharge, we have identified low-quality evidence (downgraded for serious risk of bias and serious indirectness) from 1 RCT enrolling 845 OHCA patients showing no difference in single versus 3 stacked shocks (OR, 1.29; 95% CI, 0.85–1.96).¹⁷

For the critical outcome of survival to hospital admission, we have identified very-low-quality evidence (downgraded for serious risk of bias and serious indirectness) from 1 RCT enrolling 845 OHCA patients showing no difference in single versus 3 stacked shocks (OR, 1.02; 95% CI, 0.78–1.34).¹⁷

For the critical outcome of ROSC, we have identified low-quality evidence (downgraded for serious risk of bias and serious indirectness) from 1 RCT enrolling 845 OHCA patients showing no difference in single versus 3 stacked shocks (OR, 0.94; 95% CI, 0.72–1.23).¹⁷

For the important outcome of recurrence of VF (refibrillation), we have identified low-quality evidence (downgraded for serious risk of bias, serious indirectness, and serious imprecision) from 1 RCT enrolling 136 OHCA patients showing no difference in single versus 3 stacked shocks (OR, 1.00; 95% CI, 0.47–2.13).¹⁸

Treatment Recommendation

We recommend a single-shock strategy when defibrillation is required (strong recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making this strong recommendation, the task force has placed a greater value on not changing current practice and minimizing interruptions in chest compressions whilst acknowledging that studies since 2010 have not shown that any specific shock strategy is of benefit for any survival end point. There is no conclusive evidence that a single-shock strategy is of benefit for ROSC or recurrence of VF compared with 3 stacked shocks, but in view of the evidence suggesting that outcome is improved by minimizing interruptions to chest compressions, we continue to recommend single shocks. The task force is aware that there are some circumstances (eg, witnessed, monitored VF cardiac arrest with defibrillator immediately available) when 3 rapid stacked shocks could be considered.

Introduction

In 2010, we recommended that for second and subsequent biphasic shocks, the same initial energy level was acceptable, but that it was reasonable to increase the energy level when possible (ie, with manual defibrillators).^12,13

Consensus on Science

For the critical outcome of survival with favorable neurologic outcome at hospital discharge, we identified very-low-quality evidence (downgraded for serious risk of bias, serious imprecision, and serious indirectness) from 1 RCT enrolling 221 OHCA patients showing no benefit of one strategy over the other (OR, 0.78; 95% CI, 0.34–1.78).¹⁹

For the critical outcome of survival to hospital discharge, we have identified very-low-quality evidence (downgraded for serious risk of bias, serious imprecision, and serious indirectness) from 1 RCT enrolling 221 OHCA patients showing no benefit of one strategy over the other (OR, 1.06; 95% CI, 0.52–2.16).¹⁹

For the critical outcome of ROSC, we have identified very-low-quality evidence (downgraded for serious risk of bias, serious imprecision, and serious indirectness) from 1 RCT enrolling 221 OHCA patients showing no benefit of one strategy over the other (OR, 1.095; 95% CI, 0.65–1.86).¹⁹

Treatment Recommendation

We suggest if the first shock is not successful and the defibrillator is capable of delivering shocks of higher energy, it is reasonable to increase the energy for subsequent shocks (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we have considered that an escalating shock energy may prevent the risk of refibrillation (see ALS 470). We also consider this to be in line with current practices where rescuers will escalate shock energy if initial defibrillation attempts fail and the defibrillator is capable of delivering a higher shock energy.

Introduction

Refibrillation is common and occurs in the majority of patients after initial first-shock termination of VF.²⁰ Refibrillation was not specifically addressed in 2010 guidelines. Distinct from refractory VF, defined as fibrillation that persists after 1 or more shocks, recurrence of fibrillation is usually defined as recurrence of VF during a documented cardiac arrest, occurring after initial termination of VF while the patient remains under the care of the same providers (usually out-of-hospital).

Consensus on Science

For the important outcome of termination of refibrillation, low-quality evidence (downgraded for serious risk of bias) from 2 observational studies^16,21 with a total of 191 cases of initial fibrillation showed termination rates of subsequent refibrillation were unchanged when using fixed 120 or 150 J shocks, respectively, and another observational study²⁰ (downgraded for confounding factors) with a total of 467 cases of initial fibrillation showed termination rates of refibrillation declined when using repeated 200 J shocks, unless an increased energy level (360 J) was selected.

Treatment Recommendation

We suggest an escalating defibrillation energy protocol to prevent refibrillation (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making this weak recommendation, we considered the lack of studies showing myocardial injury from biphasic waveforms, making it reasonable to consider increasing defibrillation energy levels when delivering shocks for refibrillation if the energy dose delivered by the defibrillator can be increased. It is unclear from current studies whether repeated episodes of VF are more resistant to defibrillation and require a higher energy level or whether a fixed energy level is adequate.

Introduction

It has generally been considered appropriate to administer 100% oxygen, whenever available, during cardiac arrest; however, in some other medical emergencies, the use of 100% is now being challenged.

Consensus on Science

There are no adult human studies that directly compare maximal inspired oxygen with any other inspired oxygen concentration.

For the critical outcome of survival to hospital discharge with favorable neurologic outcome (Cerebral Performance Category [CPC] 1 or 2), we identified very-low-quality evidence (downgraded for very serious risk of bias, very serious indirectness, and serious imprecision) from 1 observational study²² enrolling 145 OHCA patients who had a Pao₂ measured during CPR that showed no difference between an intermediate Pao₂ and low Pao₂ (11/83 [13.3%] versus 1/32 [3.1%]; relative risk [RR], 4.2; 95% CI, 0.57–31.52; P=0.16), or between a high Pao₂ and low Pao₂ (7/30 [23.3%] versus 1/32 [3.1%]; RR, 7.45; 95% CI, 0.98–57.15; P=0.053).

For the important outcome of ROSC, we identified very-low-quality evidence (downgraded for very serious risk of bias, very serious indirectness, and serious imprecision) from 1 observational study²² enrolling 145 OHCA patients who had a Pao₂ measured during CPR that showed improved ROSC in those with a higher Pao₂: intermediate Pao₂ versus low Pao₂ (47/83 [56.6%] versus 7/32 [21.9%]; RR, 2.59; 95% CI, 1.31–5.12; P=0.006); high Pao₂ versus low Pao₂ (25/30 [83.3%] versus 7/32 [21.9%]; RR, 3.81; 95% CI, 1.94–7.48; P=0.0001); high Pao₂ versus intermediate Pao₂ (25/30 [83.3%] versus 47/83 [56.6%]; RR, 1.47; 95% CI, 1.15–1.88; P=0.002).

In the single identified study,²² all patients had tracheal intubation and received 100% inspired oxygen during CPR. The worse outcomes associated with a low Pao₂ during CPR could be an indication of illness severity.

Treatment Recommendation

We suggest the use of the highest possible inspired oxygen concentration during CPR (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we have considered the limited available evidence and the need to correct tissue hypoxia during CPR, and see no reason to change the current treatment recommendation.

Knowledge Gaps

• The optimal arterial or tissue oxygen targets during CPR are unknown.

• A method of reliably monitoring oxygen targets during CPR has not been established.

• The feasibility of controlling inspired oxygen concentration during CPR remains unclear.

• Prospective clinical trials may be warranted to explore different inspired oxygen concentrations during CPR.

• The role and feasibility of alternatives to oxygen/air mixtures during CPR are unknown.

Consensus on Science

Treatment Recommendation

We suggest using either an advanced airway or a bag-mask device for airway management during CPR (weak recommendation, very-low-quality evidence) for cardiac arrest in any setting.

Values, Preferences, and Task Force Insights

In the absence of sufficient data obtained from studies of IHCA, it is necessary to extrapolate from data derived from OHCA.

The type of airway used may depend on the skills and training of the healthcare provider. Tracheal intubation may result in unrecognized esophageal intubation and increased hands-off time in comparison with insertion of an SGA or a bag-mask device. Both a bag-mask device and an advanced airway are frequently used in the same patient as part of a stepwise approach to airway management, but this has not been formally assessed.

Knowledge Gaps

• There are no RCTs of initial airway management during cardiac arrest.

• The type and duration of training required for each device is unknown.

• During cardiac arrest, is a stepwise approach to airway management commonly used? It is not clear how this can be studied rigorously.

Introduction

SGAs are generally considered easier to insert than tracheal tubes are, and their use in cardiac arrest has been increasing.

Consensus on Science

Treatment Recommendation

We suggest using either an SGA or tracheal tube as the initial advanced airway during CPR (weak recommendation, very-low-quality evidence) for cardiac arrest in any setting.

Values, Preferences, and Task Force Insights

In the absence of sufficient data obtained from studies of IHCA, it is necessary to extrapolate from data derived from OHCA.

The type of airway used may depend on the skills and training of the healthcare provider. Tracheal intubation requires considerably more training and practice. Tracheal intubation may result in unrecognized esophageal intubation and increased hands-off time in comparison with insertion of an SGA. Both an SGA and tracheal tube are frequently used in the same patients as part of a stepwise approach to airway management, but this has not been formally assessed.

Knowledge Gaps

• There are no RCTs of initial airway management during cardiac arrest.

• The type and duration of training required for each device is unknown.

• During the management of cardiac arrest, is a stepwise approach to airway management commonly used? It is not clear how this can be studied rigorously.

Introduction

Unrecognized esophageal intubation is a serious complication of attempted tracheal intubation during CPR. There are several potential methods for confirming correct placement of a tracheal tube: capnography and detection of CO₂, use of an esophageal detection device, and tracheal ultrasound.

Consensus on Science

Treatment Recommendations

We recommend using waveform capnography to confirm and continuously monitor the position of a tracheal tube during CPR in addition to clinical assessment (strong recommendation, low-quality evidence).

We recommend that if waveform capnography is not available, a nonwaveform CO₂ detector, esophageal detector device, or ultrasound in addition to clinical assessment is an alternative (strong recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making these strong recommendations, and despite the low-quality evidence, we place a high value on avoiding unrecognized esophageal intubation. The mean incidence of unrecognized esophageal intubation in cardiac arrest was 4.3% (range, 0%–14%) in the 11 studies we assessed. Unrecognized esophageal placement of an advanced airway is associated with a very high mortality. We, therefore, place value on recommending devices with a low FPR (ie, the device indicates tracheal placement but the tube is in the esophagus).

In addition, waveform capnography is given a strong recommendation, because it may have other potential uses during CPR (eg, monitoring ventilation rate, assessing quality of CPR).

Knowledge Gaps

• The evidence is limited on the value of CO₂ devices after prolonged cardiac arrest.

• There are very few studies comparing the practical implications (cost, timeliness) of these devices.

• The use of ultrasound requires further studies.

Introduction

Hyperventilation during CPR has been shown to be harmful, but once an advanced airway has been placed, the optimal ventilation rate remains uncertain.

Consensus on Science

We did not identify any evidence to address the critical outcomes of survival with favorable neurologic/functional outcome at discharge, 30 days, 60 days, 180 days, and/or 1 year; survival at discharge, 30 days, 60 days, 180 days, and/or 1 year.

We identified very-low-quality evidence (downgraded for very serious risk of bias and indirectness, and serious inconsistency and imprecision) from 10 animal studies^54–63 and 1 human observational study⁶⁴ that does not enable us to estimate with confidence the effect of a ventilation rate of 10/min compared with any other rate for the important outcome of ROSC.

Treatment Recommendation

We suggest a ventilation rate of 10 breaths/min in adults with cardiac arrest with a secure airway receiving continuous chest compressions (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we have valued the need to suggest a ventilation rate that is already in use. We note that the Australian and New Zealand Committee on Resuscitation (ANZCOR) currently recommends a ventilation rate of 6 to 10 breaths/min and would see no reason for this to change. We did not assess effect of tidal volume and any other ventilation variables during CPR and have therefore not addressed these in the treatment recommendation.

Knowledge Gaps

• Ventilation rates lower than 10/min need to be assessed during ALS.

• We do not know the ideal tidal volume and any other ventilation variables during CPR.

Introduction

The ITD is designed to reduce the intrathoracic pressure during the decompression phase of chest compression. There is some evidence the ITD increases blood flow during CPR. The ITD has been studied during conventional CPR and during ACD CPR.

Consensus on Science

Treatment Recommendation

We recommend against the routine use of the ITD in addition to conventional CPR (strong recommendation, high-quality evidence).

A consensus recommendation could not be reached for the use of the ITD when used together with ACD CPR.

Values, Preferences, and Task Force Insights

In making a recommendation against the routine use of the ITD alone, we place a higher value on not allocating resources to an ineffective intervention over any yet-to-be-proven benefit for critical or important outcomes.

Because of the concern about allocating resources to an intervention with equivocal benefit for critical or important outcomes, a consensus recommendation could not be reached for ITD combined with ACD CPR. The task force thought that the decision on use of the ITD plus ACD combination should be left to individual Council guidelines.

Public comments posted online were reviewed and considered by the task force, specifically regarding the use of the ITD and ACD CPR combination and the task force’s interpretation of the data from 2 publications from the same study^68,69 using the GRADE process, and how the data from these studies had been analyzed and interpreted. The task force received feedback from the investigator(s) of this study in the public commenting period and in an open session. In addition, it considered an editorial on the analysis of this study⁷¹ and discussed the publications^68,69 and their clinical significance in its closed sessions. The NNTs were discussed and the use of the CI closest to unity as a measure of study precision. It was also noted that the critical and important endpoints for this and the other ALS PICO questions were agreed a priori and posted for public commenting before searches took place, hence the difference in our hierarchy of outcomes compared with the actual primary and secondary outcomes reported in the study that made up the 2 publications. The task force appreciated the challenges of studying a combined intervention and conducting a large cardiac arrest study. There was also discussion of the involvement of the manufacturer in the design and reporting of the study and that sponsorship of drug and device studies by manufacturers can lead to more favorable results and conclusions.⁷² There was considerable debate on this topic in both closed and open task force sessions such that a consensus could not be achieved by the task force on a treatment recommendation for the use of the ITD when used together with ACD CPR.

Knowledge Gaps

• Optimal compression and ventilation rates for ITD CPR and ACD plus ITD CPR may or may not be different from those for conventional CPR.

• The independent effects of ITD and ACD CPR are uncertain.

• Effectiveness studies should examine other geographical settings and populations.

Introduction

Providing high-quality manual CPR is tiring, and there is evidence that CPR quality deteriorates with time. Mechanical CPR devices may enable the delivery of high-quality CPR for a sustained period, but at the time of writing the 2010 CoSTR, their impact on outcome was unclear.

Consensus on Science

For the critical outcome of survival to 1 year, we identified moderate-quality evidence (downgraded for serious risk of bias) from 1 cluster RCT⁷³ using the Lund University Cardiac Arrest System (LUCAS) device showing no benefit or harm when compared with manual chest compressions (5.4% versus 6.2%; RR, 0.87; 95% CI, 0.68–1.11).

For the critical outcomes of survival at 180 days with good neurologic outcome and survival at 30 days with favorable neurologic outcome, we identified moderate-quality evidence (downgraded for serious risk of bias) from 1 RCT⁷⁴ using a LUCAS device and enrolling 2589 OHCA patients that did not show benefit or harm when compared with manual chest compressions at 180 days (8.5% versus 7.6%; RR, 1.11; 95% CI, 0.86–1.45) or 30 days (7.3% versus 8.1%; RR, 1.11; 95% CI, 0.84–1.45).

For the critical outcome of survival to 180 days, we identified moderate-quality evidence (downgraded for serious risk of bias) from 1 RCT⁷⁴ using a LUCAS device enrolling 2589 OHCA patients showing no benefit or harm when compared with manual chest compressions where quality of chest compressions in the manual arm was not measured (8.5% versus 8.1%; RR, 1.06; 95% CI, 0.81–1.41).

For the critical outcome of survival to hospital discharge with favorable neurologic outcome (defined as CPC 1–2 or mRS 0–3), we have identified moderate-quality evidence (downgraded for serious risk of bias) from 3 RCTs enrolling 7582 OHCA patients showing variable results.^74–76 One study⁷⁵ (n=767) showed harm with the use of a load-distributing band mechanical chest compression device compared with manual chest compressions (7.5% of patients in the control group versus 3.1% in the intervention group; P=0.006; RR, 0.41; 95% CI, 0.21–0.79). Two other RCTs^74,76 (n=6820), one using a load-distributing band and the other using a LUCAS, did not show benefit or harm when compared with manual chest compressions: load-distributing band study: 4.14% survival in the intervention group versus 5.25% for manual compressions (RR, 0.79; 95% CI, 0.60–1.03); LUCAS: 8.31% intervention versus 7.76% manual compressions (RR, 1.07; 95% CI, 0.83–1.39).

For the critical outcome of survival to hospital discharge, we identified moderate-quality evidence (downgraded for serious risk of bias) from 5 RCTs^74–78 enrolling 7734 OHCA patients and 150 IHCA patients showing heterogeneous results. One study of patients with IHCAs⁷⁷ (n=150) showed benefit with use of a piston device compared with manual chest compressions (32.9% versus 14.7%; P=0.02; RR, 2.21; 95% CI, 1.17–4.17). Two other RCTs^74,78 of LUCAS did not show benefit or harm (9.0% versus 9.15%; RR, 0.98; 95% CI, 0.77–1.25 and 8.0% versus 9.72%; RR, 0.82; 95% CI, 0.29–2.33, respectively, for LUCAS versus manual compressions). One large RCT⁷⁶ (n=4231) using a load-distributing band device showed equivalence when compared with high-quality manual chest compressions (9.34% versus 10.93%; RR, 0.85; 95% CI, 0.71–1.02).

For the critical outcome of survival to 30 days, we identified moderate-quality evidence (downgraded for serious risk of bias) from 2 RCTs^73,74 (n=7060) using the LUCAS device showing no benefit or harm when compared with manual chest compressions and where quality of compressions in the manual arm was not measured (6.3% versus 6.85%; RR, 0.92; 95% CI, 0.73–1.16 and 8.82% versus 8.07%; RR, 1.02; 95% CI, 0.97–1.31, respectively).

For the important outcome of ROSC, we identified low-quality evidence (downgraded for serious risk of bias and serious inconsistency) from 7 RCTs enrolling 11 638 cardiac arrest patients (IHCA and OHCA).^{73,74,76–80} Two studies^77,79 (n=167) showed benefit with mechanical chest compression devices compared with manual compressions: 14.29% versus 0% (RR, not applicable) and 55.26% versus 37.84% (RR, 1.46; 95% CI, 1.02–2.08), respectively. One study⁷⁶ (n=4231) showed harm with mechanical devices; however, there was no adjustment for interim analyses: 28.59% versus 32.32% (RR, 0.88; 95% CI, 0.81–0.97). Four studies^73,74,78,80 (n=7240) did not show benefit or harm when compared with manual chest compressions: 47.06% versus 17.75% (RR, 2.67; 95% CI, 0.85–8.37), 31.60% versus 31.39% (RR, 1.01; 95% CI, 0.92–1.10), 35.38% versus 34.60% (RR, 1.02; 95% CI, 0.92–1.14), and 40.54% versus 31.94% (RR, 1.27; 95% CI, 0.82–1.96), respectively.

Treatment Recommendations

We suggest against the routine use of automated mechanical chest compression devices to replace manual chest compressions (weak recommendation, moderate-quality evidence).

We suggest that automated mechanical chest compression devices are a reasonable alternative to high-quality manual chest compressions in situations where sustained high-quality manual chest compressions are impractical or compromise provider safety (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

The task force placed value on ensuring high-quality chest compressions with adequate depth, rate, and minimal interruptions, regardless of whether they are delivered by machine or human. The task force also considered that application of a mechanical chest compression device without a focus on minimizing interruptions in compressions and delay to defibrillation could cause harm.

In making a recommendation for mechanical compression devices for use in some settings, we place value on the results from a large, high-quality RCT⁷⁶ showing equivalence between very-high-quality manual chest compressions and mechanical chest compressions delivered with a load-distributing band in a setting with rigorous training and CPR quality monitoring. Also, the task force acknowledges the existence of situations where sustained high-quality manual chest compressions may not be practical. Examples include CPR in a moving ambulance where provider safety is at risk, the need for prolonged CPR where provider fatigue may impair high-quality manual compressions (eg, hypothermic arrest), and CPR during certain procedures (eg, coronary angiography or preparation for ECPR).

Our task force agreed that there was an adequate amount of data generated from RCTs for the systematic review to exclude observational studies. We agreed that despite the availability of several observational studies comparing manual and mechanical chest compressions, the inherent risk of bias related to patient selection, group allocation, and uncontrolled confounders supports a decision to exclude them from the process of developing this CoSTR statement.

We conducted a universal literature search for RCTs studying any type of automated mechanical chest compression device. Prior to initiating the review, we planned to parse the data by device type if an effect specific to device was observed in the analysis. Although we did not undertake a formal analysis by device, there were no obvious device-specific effects observed.

The task force did consider some data that are not included in the evidence profile tables or CoSTR statement. Specifically, the PARAMEDIC (prehospital randomized assessment of a mechanical compression device in cardiac arrest) study⁷³ showed an association between mechanical chest compressions and worse survival with good neurologic outcome (CPC 1–2) at 3 months (adjusted OR, 0.72; 95% CI, 0.52–0.99). This was not included in our consensus on science, because survival with good neurologic outcome at 90 days was not an a priori outcome identified by the group.

After assessing the evidence, there was much debate over the ultimate wording of our recommendation. Some members thought a weak recommendation supporting mechanical chest compression devices as a reasonable alternative to manual chest compressions was most appropriate, whereas others thought a recommendation against the routine use of mechanical chest compression devices was more appropriate. There was general agreement that the bulk of evidence reviewed suggests no significant difference or equivalence between mechanical and manual chest compressions related to critical and important clinical outcomes. The task force weighed this with the data from a few studies suggesting a negative association between mechanical chest compression and outcomes as well as the potential resource implications associated with implementation of mechanical devices in any setting. With these factors in mind, the task force concluded that available clinical evidence did not support a recommendation for broad and universal implementation of mechanical chest compression devices across all clinical settings in favor of high-quality manual chest compressions.

Public comments provided online were reviewed by the task force. Comments suggested that we consider special circumstances where mechanical chest compressions may be more practical than the continued provision of high-quality chest compression and circumstances where provider safety might be improved with the use of mechanical versus manual chest compressions. Delivery of manual compressions in a moving ambulance by an unrestrained provider was seen as a particularly unsafe situation. Mechanical devices may allow providers to remain seated and restrained in this situation while chest compressions continue. Accordingly, we have included a treatment recommendation to address these situations not directly addressed in the literature reviewed but deemed to represent reasonable situations for the use of this technology.

Knowledge Gaps

• Are mechanical chest compression devices superior to manual chest compressions in special situations such as the moving ambulance, prolonged CPR, or during procedures such as coronary angiography?

• Are there certain subgroups of patients who may benefit differentially from mechanical or manual chest compressions (eg, shockable versus nonshockable initial rhythm)?

• Is one type of mechanical chest compression device superior to another with respect to important clinical outcomes?

Introduction

Extracorporeal techniques require vascular access and a circuit with a pump and oxygenator and can provide a circulation of oxygenated blood to restore tissue perfusion. This has the potential to buy time for restoration of an adequate spontaneous circulation and treatment of reversible underlying conditions. This is commonly called extracorporeal life support (ECLS), and more specifically ECPR when done during cardiac arrest. These techniques are increasingly being used for OHCA. We considered ECLS for IHCA and OHCA separately.

Consensus on Science

Treatment Recommendation

We suggest ECPR is a reasonable rescue therapy for selected patients with cardiac arrest when initial conventional CPR is failing in settings where this can be implemented (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this weak recommendation, we note that the published series used selected patients for ECPR and that guidelines for clinical practice should apply to similar populations. Published comparative studies are limited by the bias created when experienced clinicians select the best candidates to receive ECPR, perhaps using unmeasured variables. We acknowledge that ECPR is a complex intervention that requires considerable resource and training that is not universally available, but put value on an intervention that may be successful in individuals where usual CPR techniques have failed. In addition, ECPR can buy time for another treatment such as coronary angiography and percutaneous coronary intervention (PCI).

Knowledge Gaps

• Controlled clinical trials are needed to assess the effect of ECPR versus traditional CPR on clinical outcomes in patients with cardiac arrest.

• What is the optimal flow rate for ECPR in the treatment for cardiac arrest?

• Which subgroups of patients can benefit most from a strategy of ECPR?

• What type of patients should be considered for ECPR?

• What role, if any, should prehospital ECPR play in resuscitating patients from OHCA?

• What is the optimal target temperature for patients on ECPR after cardiac arrest?

• What are reliable prognostic factors for patients treated with ECPR after cardiac arrest?

Introduction

ETCO₂ is the partial pressure of CO₂ at the end of an exhaled breath. It reflects cardiac output (CO) and pulmonary blood flow, as CO₂ is transported by the venous system to the right side of the heart and then pumped to the lungs by the right ventricle. During CPR, ETCO₂ values are low, reflecting the low CO generated by chest compression. Although ETCO₂ values higher than 10 mm Hg have been correlated to ROSC,^85–89 there is uncertainty if any ETCO₂ value measured during CPR can reliably predict survival or survival with good neurologic outcome.

Consensus on Science

We did not identify any evidence to address the critical outcome of neurologically intact survival.

For the critical outcome of survival at discharge, we have identified low-quality evidence (downgraded for serious risk of bias and serious imprecision) from 1 observational study enrolling 127 patients⁹⁰ showing a correlation with initial ETCO₂ 10 mm Hg (1.33 kPa) or greater when compared with less than 10 mm Hg (OR, 11.4; 95% CI, 1.4–90.2).

For the critical outcome of survival at discharge, we have identified low-quality evidence (downgraded for serious risk of bias and serious imprecision) from 1 observational study enrolling 127 patients⁹⁰ showing a correlation with 20 minutes of ETCO₂ 20 mm Hg (2.67 kPa) or greater when compared with less than 20 mm Hg (OR, 20.0; 95% CI, 2.0–203.3).

For the important outcome of ROSC, we have identified moderate-quality evidence (downgraded for serious risk of bias) from 3 observational studies enrolling 302 patients^90–92 showing a correlation with initial ETCO₂ 10 mm Hg or greater when compared with less than 10 mm Hg (OR, 10.7; 95% CI, 5.6–20.3).

For the important outcome of ROSC, we have identified very-low-quality evidence (downgraded for very serious risk of bias, serious inconsistency, and serious imprecision) from 3 observational studies enrolling 367 patients^90,93,94 showing correlation with 20 minutes ETCO₂ 10 mm Hg or greater when compared with less than 10 mm Hg (OR, 181.6; 95% CI, 40.1–822.6).

Treatment Recommendations

We recommend against using ETCO₂ cutoff values alone as a mortality predictor or for the decision to stop a resuscitation attempt (strong recommendation, low-quality evidence).

We suggest that an ETCO₂ 10 mm Hg or greater measured after tracheal intubation or after 20 minutes of resuscitation may be a predictor of ROSC (weak recommendation, low-quality evidence).

We suggest that an ETCO₂ 10 mm Hg or greater measured after tracheal intubation or an ETCO₂ 20 mm Hg or greater measured after 20 minutes of resuscitation may be a predictor of survival to discharge (weak recommendation, moderate-quality evidence).

Values, Preferences, and Task Force Insights

In making the strong recommendations against using a specific ETCO₂ cutoff value alone as a mortality predictor or for the decision to stop a resuscitation attempt, we have put a higher value on not relying on a single variable (ETCO₂) and cutoff value when their usefulness in actual clinical practice, and variability according to the underlying cause of cardiac arrest, has not been established and there are considerable knowledge gaps.

The task force was concerned that the etiology (eg, asphyxia, PE) of cardiac arrest could affect ETCO₂ values, and that there was a risk of self-fulfilling prophecy if specific threshold values were followed. There was concern about the accuracy of ETCO₂ values measured during CPR. During open discussions there were requests that the ILCOR recommendation be far more prescriptive to prevent futile and prolonged resuscitation attempts.

Knowledge Gaps

• The effects on ETCO₂ of timing, etiology of arrest, ventilation rate, and chest compression quality are not fully understood.

• The role of ETCO₂ with a bag-mask device or SGA requires further study.

• Are ETCO₂ values measured during CPR accurate?

• The ETCO₂ cutoff values to reliably predict short- and long-term outcomes is not known.

Introduction

Several physiological variables such as ETCO₂, coronary perfusion pressure, aortic diastolic pressure, and cerebral oximetry measurements have been used to assess and guide the quality of CPR.

Consensus on Science

We found no studies that addressed the critical and important outcomes.

For the outcome of change in physiologic values by modifications in CPR, we identified 13 observational studies that provided very-low-quality evidence (downgraded for serious risk of bias, serious inconsistency, serious indirectness, and serious imprecision) comparing different CPR techniques (standard, lower sternal, active compression-decompression, intra-abdominal compression, mechanical thumper, ITD, band chest compression, load-distributing band, vest CPR) with the use of physiologic monitoring (arterial line, ETCO₂, oxygen saturation as measured by pulse oximetry (Spo₂), coronary perfusion pressure, cerebral oximetry, near-infrared spectroscopy) in 469 subjects.^{66,80,95–105} Differences were detected between different CPR techniques, although this was not consistent across different modalities. Given the heterogeneity of CPR techniques used across studies, data could not be pooled. There were no studies that were found that used physiologic feedback to evaluate CPR quality.

Treatment Recommendation

We make no treatment recommendation for any particular physiological measure to guide CPR, because the available evidence would make any estimate of effect speculative.

Values, Preferences, and Task Force Insights

In making no recommendation, we have placed high value on the lack of evidence and the need for further studies in this area.

Knowledge Gaps

• Studies of the effect of using physiologic feedback to evaluate CPR quality and modifications in CPR technique are required.

• Studies that measure the effect of physiological monitoring to guide resuscitation on ROSC and survival with good neurologic outcome are required.

Introduction

Ultrasound has been increasingly used as a diagnostic and prognostic tool for critically ill patients, particularly in intensive care units (ICUs).¹⁰⁶ Specific protocols for evaluation during CPR enable assessment of myocardial contractility and may help identify potentially treatable causes, such as hypovolemia, pneumothorax, pulmonary thromboembolism, or restrictive pericardial effusion, without interfering in patient care.¹⁰⁷

Consensus on Science

For the critical outcome of survival, we identified 1 observational study.¹⁰⁸ The evidence was downgraded for very high risk of bias (significant confounding, selection bias) and imprecision (small sample size). Therefore, we concluded that the data do not provide enough evidence to address the PICO question.

For the important outcome of ROSC, we identified very-low-quality evidence (downgraded for imprecision [small sample size] and very high risk of bias [no information about randomization allocation, lack of blinding, lack of blinding in outcome assessors]) from 1 RCT investigating the use of cardiac ultrasound during ACLS, compared with no use of cardiac ultrasound during ACLS in adult patients with pulseless electrical activity arrest.¹⁰⁹ This study enrolled 100 patients in a convenience sample and reported ROSC for at least 10 seconds in 34% of patients in the ultrasound group versus 28% in the group with no ultrasound (P=0.52).

Treatment Recommendations

We suggest that if cardiac ultrasound can be performed without interfering with standard ACLS protocol, it may be considered as an additional diagnostic tool to identify potentially reversible causes (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation we have placed a higher value on the potential harm from interruptions in chest compressions. There is currently inadequate evidence to evaluate whether there is any benefit of cardiac ultrasound during ACLS. Although this was not specifically part of the question, the task force discussed the importance of the need for an individual trained in ultrasound during resuscitation to minimize interruption in chest compression. The task force agreed there will be circumstances where ultrasound identification of a potentially reversible cause of cardiac arrest or ‘pseudo’ pulseless electrical activity may be useful.

Knowledge Gaps

It remains unclear if the addition of ultrasound during ACLS improves outcomes:

• All data are from OHCA. All data are from non-VF patients, primarily assessing pulseless electrical activity.

• A systematic review of the diagnostic utility of ultrasound should be done. There are some articles investigating whether ultrasound findings predict probability of survival.

• Pretest probability (suspicion of an ultrasound-detectable etiology) is important for choosing to do ultrasound, because ultrasound will interfere to some extent with CPR.

• It is unknown if the findings of ultrasound during CPR are correctly interpreted, because images are compared with findings from patients with pulse (eg, right ventricular dilation occurs in all cardiac arrest, separate from PE).

• It remains unclear if the addition of ultrasound during CPR improves outcomes. The vast majority of literature on ultrasound during cardiac arrest has focused on the prognostic value of cardiac ultrasound findings. Randomized trials investigating whether use of ultrasound during CPR has an effect on patient outcomes are needed.

Introduction

Since 2010, there have been 2 randomized drug trials in cardiac arrest: one compared drugs with no drugs¹¹⁰ and another compared epinephrine with placebo.¹¹¹ The Olasveengen trial compared a bundle of drugs given intravenously to a control group of patients randomized to no intravenous access and, therefore, no drugs. A post hoc subgroup analysis of the trial comparing those patients who did or did not receive epinephrine¹¹² revealed an advantage with epinephrine for admission to hospital but suggested an association with harm for the outcomes of survival to discharge and functional survival as measured by CPC. The Olasveengen original trial¹¹⁰ was excluded from our review; however, the post hoc subgroup analysis was included in the systematic review of observational and randomized trials, which we have used to comment on the body of work defined by adjusted and unadjusted observational studies.¹¹³

Consensus on Science

For all 4 long-term and short-term outcomes, we found 1 underpowered RCT that provided low-quality evidence (downgraded for selection and ascertainment bias) comparing SDE with placebo¹¹¹ in 534 subjects.

For the critical outcome of survival to discharge, there was uncertain benefit or harm of SDE over placebo (RR, 2.12; 95% CI, 0.75–6.02; P=0.16; absolute risk reduction [ARR], 2.14%; 95% CI, −0.91% to 5.38%, or 21 more patients/1000 survived with epinephrine [95% CI, 9 fewer patients/1000 to 54 more patients/1000 survived with epinephrine]).

For the critical outcome of survival to discharge with good neurologic outcome (defined as CPC of 1–2), there was uncertain benefit or harm of SDE over placebo (RR, 1.73; 95% CI, 0.59–5.11; P=0.32; ARR, 1.4%; 95% CI, −1.5% to 4.5%, which translates to 14 more patients/1000 survived with a CPC score of 1 or 2 with epinephrine [95% CI, 15 fewer patients/1000 to 45 more patients/1000 survived with a CPC score of 1 or 2 when given epinephrine]).

For the important outcome of survival to admission, patients who received SDE had higher rates of survival to admission (RR, 1.95; 95% CI, 1.34–2.84; P=0.0004; ARR, 12%; 95% CI, 5.7%–18.9%, which translates to 124 more patients/1000 survived to admission with epinephrine [95% CI, 57–189 more patients/1000 survived to admission]).

For the important outcome of ROSC in the prehospital setting, 151 more patients/1000 achieved ROSC with epinephrine (95% CI, 90–212 more patients/1000 achieved ROSC with epinephrine) when compared with those who received placebo (RR, 2.80; 95% CI, 1.78–4.41; P<0.000 01; ARR, 15%; 95% CI, 9%–21%).

While observational studies were excluded from the primary evidence evaluation, the task force did make some comparison of the randomized trial data with prior conclusions drawn from large observational data sets. Using the analysis published by Patanwala¹¹³ in 2014 when the Jacobs trial¹¹¹ is compared with adjusted observational trials^114,115 for the critical outcome of survival to discharge and functional survival with a CPC of 1 or 2, in settings with very low survival rates after cardiac arrest, 4.7% OHCA¹¹⁴ and 14% IHCA,¹¹⁵ in the OHCA setting, the use of epinephrine was associated with worse outcomes for survival to discharge (5.4% with epinephrine versus 4.7% without epinephrine; unadjusted OR, 1.15; 95% CI, 1.07–1.53; adjusted OR, 0.46; 95% CI, 0.42–0.51) and for functional survival (1.4% with epinephrine versus 2.2% without epinephrine; unadjusted OR, 0.61; 95% CI, 0.53%–0.71%; adjusted OR, 0.31; 95% CI, 0.26–0.36).¹¹⁴ In the in-hospital setting, the use of epinephrine was not significantly associated with either survival to discharge (OR, 1.16; 95% CI, 0.52–2.58) or functional survival (CPC, 1–2; OR, 0.43; 95% CI, 0.08–2.29).

Treatment Recommendation

We suggest SDE be administered to patients in cardiac arrest (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

We make this statement after considering the observed benefit in short-term outcomes (ROSC and admission to hospital) and our uncertainty about the benefit or harm on survival to discharge and neurologic outcome given the limitations of the observational studies. Our statement is not intended to change current practice until there are high-quality data on long-term outcomes. We have considered 1 mg to be the standard dose of epinephrine.

Knowledge Gaps

• Dose response and placebo-controlled efficacy trials are needed to evaluate the use of epinephrine in cardiac arrest. We are aware of an ongoing randomized study of epinephrine (adrenaline) versus placebo for OHCA in the United Kingdom (PARAMEDIC 2: The Adrenaline Trial, ISRCTN73485024).

Consensus on Science

A single RCT¹¹⁶ (n=336) of low quality (downgraded for high risk of bias) compared multiple doses of SDE with multiple doses of standard-dose vasopressin in the ED after OHCA. Much of the methodology is unclear, and there was 37% postrandomization exclusion. The primary outcome measure was a CPC score of 1 or 2; however, neither the sample size estimate nor power calculation were included in the article.

For the critical outcome of survival to discharge with favorable neurologic outcome (CPC 1 or 2), there was no advantage with vasopressin (RR, 0.68; 95% CI, 0.25–1.82; P=0.44 or ARR, −1.6; 95% CI, −6 to 2.4, which translates to 16 fewer patients/1000 surviving with CPC 1 or 2 with vasopressin [95% CI, 60 fewer patients/1000 to 24 more patients/1000 survive with CPC 1 or 2]).

For the critical outcome of survival to discharge, RR was 0.68 (95% CI, 0.25–1.82; P=0.44 or ARR, 1.8%; 95% CI, −3.1 to 6.7, which translates to 18 more patients/1000 surviving to discharge with vasopressin [95% CI, 31 fewer patients/1000 surviving to discharge with vasopressin to 67 more patients/1000 surviving to discharge]).

For the important outcome of ROSC, there was no observed advantage with vasopressin (RR, 0.93; 95% CI, 0.66–1.31; P=0.67).

Treatment Recommendations

We suggest vasopressin should not be used instead of epinephrine in cardiac arrest (weak recommendation, low-quality evidence).

We suggest that those settings already using vasopressin instead of epinephrine can continue to do so (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

The recommendation considers the fact that vasopressin is already used in some settings, and the available data do not indicate any reason to stop using vasopressin if current treatment protocols already include vasopressin instead of epinephrine. Conversely, there is also no evidence to indicate that settings that use epinephrine should switch to using vasopressin.

Knowledge Gaps

• Until high-quality, adequately powered trials are completed comparing epinephrine with placebo, trials involving vasopressin are not required unless as a third arm against epinephrine and placebo.

Consensus on Science

For the critical outcome of survival to hospital discharge with CPC of 1 or 2, we found very-low-quality evidence (downgraded for very serious bias and serious imprecision) from 3 RCTs^117–119 (n=2402) comparing SDE with vasopressin and epinephrine combination therapy that showed no superiority with vasopressin and epinephrine combination (RR, 1.32; 95% CI, 0.88–1.98 and ARR, 0.5%; 95% CI, −0.2% to 1.3%, which translates to 5 more patients/1000 [95% CI, 2 fewer patients/1000 to 13 more/1000] surviving to hospital discharge with a CPC of 1 or 2 with vasopressin in combination with epinephrine).

For the critical outcome of survival to hospital discharge, we found very-low-quality evidence (downgraded for very serious bias and serious imprecision) from 5 RCTs^117–121 (n=2438) comparing SDE to vasopressin and epinephrine combination therapy that did not show superiority with vasopressin and epinephrine combination therapy in survival to discharge (RR, 1.12; 95% CI, 0.84–1.49; P=0.45 and ARR, −0.17%; 95% CI, −1.3 to 1, which translates to 2 fewer patients/1000 [95% CI, 13 fewer patients/1000 to 10 more/1000] surviving to hospital discharge with vasopressin in combination with epinephrine).

For the important outcome of survival to admission, we found moderate-quality evidence (downgraded for serious bias) from 5 RCTs^117–121 (n=2438) showing no significant differences in survival to hospital admission with vasopressin and epinephrine combination therapy (RR, 0.88; 95% CI, 0.73–1.06; P=0.17).

For the important outcome of ROSC, we found moderate-quality evidence (downgraded for serious bias) from 6 RCTs^117–122 showing no ROSC advantage with vasopressin and epinephrine combination therapy (RR, 0.96; 95% CI, 0.89–1.04; P=0.31).

Treatment Recommendation

We suggest against adding vasopressin to SDE during cardiac arrest (weak recommendation, moderate-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we preferred to avoid the additional expense and implementation issues required to add a drug (vasopressin) that has no evidence of additional benefit for patients.

Knowledge Gaps

• Until high-quality, adequately powered trials are completed comparing epinephrine with placebo, trials involving vasopressin in combination with epinephrine are not required unless as a third arm against epinephrine and placebo.

Consensus on Science

For the critical outcome of survival to hospital discharge with CPC 1 or 2, we found very-low-quality evidence (downgraded for very serious indirectness and serious imprecision) from 2 RCTs comparing SDE with HDE^123,124 (n=1920) and cumulative RR that did not show any CPC 1 or 2 survival to discharge advantage with HDE (RR, 1.2; 95% CI, 0.74–1.96; ARR, −0.4%, 95% CI, −1.2 to 0.5, which translates to 3 fewer patients/1000 surviving to discharge with a CPC score of 1 or 2 [95% CI, 12 fewer to 5 more patients/1000 surviving to discharge with a CPC score of 1–2]).

For the critical outcome of survival to hospital discharge, we found very-low-quality evidence (downgraded for very serious indirectness and serious imprecision) from 5 RCTs comparing SDE with HDE^123–127 (n=2859) that did not show any survival to discharge advantage with HDE (RR, 0.97; 95% CI, 0.71–1.32; ARR, −0.1%; 95% CI, −0.1 to 0.7, which translated to 1 fewer patient/1000 surviving to discharge with HDE [95% CI, 10 fewer patients/1000 to 7 more patients/1000]).

For the important outcome of survival to hospital admission, we found low-quality evidence (downgraded for very serious indirectness) from 4 RCTs comparing SDE with HDE^{123–125,128} (n=2882) showing a survival to hospital admission advantage with HDE (RR, 1.15; 95% CI, 1.0–1.32).

For the important outcome of ROSC, we found low-quality evidence (downgraded for very serious indirectness) from 6 RCTs comparing SDE with HDE^123–128 (n=3130) showing a ROSC advantage with HDE (RR, 1.17; 95% CI, 1.03–1.34).

Treatment Recommendation

We suggest against the routine use of HDE in cardiac arrest (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making this statement, we acknowledge that HDE improves short-term outcomes but note that the low-quality evidence failed to show an improvement in the critical outcomes of survival and neurologic outcome. The absolute magnitude of effects of HDE versus SDE on ROSC (RR, 1.17; 95% CI, 1.03–1.34) and admission to hospital (RR, 1.15; 95% CI, 1.0–1.32) are modest. These HDE studies were published in the 1990s, and since then care and outcomes for cardiac arrest have changed dramatically, making it hard to interpret the relevance of these results for current care.

Knowledge Gaps

• Until high-quality, well-powered trials are completed comparing epinephrine with placebo, trials addressing dose response of epinephrine are not required except as a third arm embedded in an epinephrine-versus-placebo trial.

Consensus on Science

Treatment Recommendation

For cardiac arrest with an initial nonshockable rhythm, we suggest that if epinephrine is to be administered, it is given as soon as feasible after the onset of the arrest (weak recommendation, low-quality evidence).

For cardiac arrest with an initial shockable rhythm, we found insufficient evidence to make a treatment suggestion regarding the timing of administration of epinephrine, particularly in relation to defibrillation, and the optimal timing may vary for different groups of patients and different circumstances.

Values, Preferences, and Task Force Insights

In making the recommendation for nonshockable rhythms, we place a higher value on being able to modify a current (standard) treatment at minimal cost.

For shockable rhythms, we place a higher value on early defibrillation than on administration of epinephrine but did not think there is sufficient evidence to make a treatment recommendation. Although we acknowledge that the pathophysiology of IHCA and OHCA is likely to be different, we were confident that the same recommendations could apply to both settings.

Knowledge Gaps

• Until high-quality, well-powered trials are completed comparing epinephrine with placebo, trials addressing the timing of epinephrine doses are not required except as a third arm embedded in an epinephrine-versus-placebo trial.

Introduction

We identified studies that assessed the use of methylprednisolone, hydrocortisone, or dexamethasone during CPR. Studies usually bundled the steroid with other vasoactive drugs. All studies examined either IHCA or OHCA. Because the pathophysiology and epidemiology of IHCA and OHCA are so different, we considered these situations separately.

Consensus on Science

Treatment Recommendation

For IHCA, the task force was unable to reach a consensus recommendation for or against the use of steroids in cardiac arrest.

We suggest against the routine use of steroids during CPR for OHCA (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation for IHCA, it was noted that there were no studies assessing the effect of the addition of steroids alone to standard treatment for IHCA. Also, although the triple-agent drug regimen appears to suggest an association with improved outcome, the population studied had very rapid ALS, a high incidence of asystolic cardiac arrest, and low baseline survival compared with other IHCA studies, so some of the observed effects might be peculiar to the population studied.

In making this recommendation for OHCA, we considered the cost and distraction from the addition of treatments for which there is very low confidence in any effect. The different recommendation for OHCA and IHCA was influenced by the physiological differences between these conditions, such as the incidence of sepsis, adrenal insufficiency from critical illness, and cardiovascular etiologies.

Knowledge Gaps

• It is unclear which aspect of bundled treatments such as epinephrine, vasopressin, and steroids are related to any observed treatment effect. The alternative possibility is that bundled treatments require synergistic action, because other studies with each agent (vasopressin and steroids) have failed to find the same effect.

• Confidence in the treatment effects from bundled treatments will increase if confirmed in further studies.

Introduction

Antiarrhythmic drugs can be used during cardiac arrest for refractory ventricular dysrhythmias. Refractory VF/pVT is defined differently in many trials but generally refers to failure to terminate VF/pVT with 3 stacked shocks, or with the first shock. In an ongoing clinical trial from which results are not yet available, refractory VF refers to “persistent or recurrent VF/pVT after 1 or more shocks.”¹⁴⁰

Consensus on Science

Comparative data on the use of antiarrhythmic drugs were identified for amiodarone, lidocaine, magnesium, and nifekalant. The data reviewed for magnesium only addressed the use of this drug for undifferentiated VF/pVT and not the treatment of torsades de pointes or known hypomagnesemic patients. Nifekalant is only available in certain regions.

Treatment Recommendations

We suggest the use of amiodarone in adult patients with refractory VF/pVT to improve rates of ROSC (weak recommendation, moderate-quality evidence).

We suggest the use of lidocaine or nifekalant as an alternative to amiodarone in adult patients with refractory VF/pVT (weak recommendation, very-low-quality evidence).

We recommend against the routine use of magnesium in adult patients (strong recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making these recommendations, we considered the reported beneficial effects of amiodarone on the important outcome of survival to hospital admission. We acknowledged that there was the uncertainty about any beneficial or harmful effects of these drugs on the critical outcomes of survival or favorable neurologic survival. Although the evidence supporting their use is weaker, in making a recommendation for lidocaine and nifekalant as alternatives to amiodarone, the task force recognized that amiodarone is not available or currently used in some countries. The small quantity of new data made the task force place value on not changing current clinical practice.

Knowledge Gaps

• There is a need for sufficiently powered RCTs to detect a difference in survival to hospital discharge or favorable neurologic outcomes.

• A potential source of bias reducing confidence in prior trials of amiodarone is use of the polysorbate solvent for the drug. This solvent is known to reduce blood pressure, and its use as placebo may have created a bias for worse outcomes in placebo groups. Future studies should account for this effect or use other solvents.

• There is an ongoing trial comparing amiodarone to lidocaine and to placebo designed and powered to evaluate for functional survival.¹⁴⁰

• No data address how to select a second-line agent when VF/pVT is refractory to the first drug.

Introduction

The aim of this PICO review was to assess whether commonly applied additions to the standard practice of resuscitation led to improved outcomes in pregnant women. Specific emphasis was placed on uterine displacement for the purpose of decreasing aortocaval compression, and perimortem cesarean delivery as interventions to improve outcome in the mother and newborn.

Consensus on Science

There were no comparative studies of uterine displacement for women in cardiac arrest before delivery. No studies compared different maneuvers (eg, manual displacement versus left pelvic tilt) to achieve optimal uterine displacement for women in cardiac arrest before delivery.

Physiologic reviews and studies of uterine displacement maneuvers in nonarrest pregnant women support that uterine displacement might be physiologically beneficial for women in cardiac arrest.¹⁴⁹ Any benefit would have to be weighed against the potential interference or delay with usual resuscitation care.

For the critical outcomes of survival with favorable neurologic/functional outcome at discharge, 30 days, 60 days, 180 days, and/or 1 year, and survival only at discharge, 30 days, 60 days, 180 days, and/or 1 year, and the important outcomes of ROSC, we found 3 observational studies of 154 subjects collectively^150–152 that provided very-low-quality evidence (downgraded for very serious risk of bias and imprecision, and serious inconsistency) comparing cardiac arrest resuscitation with or without perimortem cesarean delivery. The procedures to ascertain cases and controls in these studies were significantly different so that the pooled comparison of any of the assigned outcomes is considered inappropriate.

Treatment Recommendations

We suggest delivery of the fetus by perimortem cesarean delivery for women in cardiac arrest in the second half of pregnancy (weak recommendation, very-low-quality evidence).

There is insufficient evidence to define a specific time interval by which delivery should begin. High-quality usual resuscitation care and therapeutic interventions that target the most likely cause(s) of cardiac arrest remain important in this population.

There is insufficient evidence to make a recommendation regarding the use of left lateral tilt and/or uterine displacement during CPR in the pregnant patient.

Values, Preferences, and Task Force Insights

In making this statement, we place value on maternal and neonatal survival, on the absence of data on left lateral tilt and uterine displacement in women with cardiac arrest, and on our uncertainty about the absolute effect of either uterine displacement or perimortem delivery during CPR on any of the assigned outcomes. The task force thought not making a recommendation for or against the use of left lateral tilt or uterine tilt is unlikely to change current practice or guidelines.

Knowledge Gaps

• Research in the area of maternal resuscitation is lacking because cardiac arrest in pregnancy is rare. Most evidence is extrapolated from nonpregnant people, manikin or simulation studies, and case reports.

• The heterogeneous nature of the etiologies of maternal cardiac arrest, variations in gestational age and body mass index of the cases, variations in the location (eg, out-of-hospital, ED, obstetric unit), and context of arrest and personnel available to immediately respond, and absence of information about the quality of usual resuscitation care all further hamper interpretation of the limited available data.

• Systematic data collection in pregnant women who have experienced cardiac arrest will require a national or international registry and/or coordinated prospective population-level surveillance to compile a sufficiently large and robust data set to evaluate the effect of either uterine displacement or perimortem delivery on maternal ROSC, maternal survival, functionally intact maternal survival, neonatal survival, and functionally intact neonatal survival.

• A particular emphasis on cardiovascular etiologies of arrest is warranted given increasing numbers of women with congenital heart conditions having children, the increasing prevalence of cardiomyopathy among pregnant and postpartum women, and the preponderance of cardiovascular disease evident in maternal mortality surveillance reports.¹⁵³

Introduction

Lipid therapy for cardiac arrest associated with drug toxicity, and in particular local anesthetic toxicity, is becoming increasingly common. Based on laboratory and preclinical data showing that IV administration of lipid solutions can absorb lipid-soluble drugs, studies examined whether this therapy would be useful for cardiac arrest related to drug overdose. We set out to identify studies comparing outcomes with IV lipids to no IV lipids.

Consensus on Science

We identified no human comparative studies in cardiac arrest and periarrest states relevant to the PICO question. Many case reports and case series described resuscitation that included administration of lipid.

Treatment Recommendation

We are unable to make any evidence-based treatment recommendation about the use of IV lipid emulsion to treat toxin-induced cardiac arrest.

Values, Preferences, and Task Force Insights

Although there are many case reports and case series of patients who were resuscitated after administration of IV lipid, the absence of any comparative data made it impossible to determine anything besides temporal association of the therapy with outcome. Despite the paucity of data, we do not wish to discourage the use of an antidote with some theoretical basis in a dire clinical situation.

Knowledge Gaps

• Comparisons are needed of patients with similar clinical characteristics who were treated and who were not treated with IV lipids after suspected drug toxicity.

Introduction

Opioid toxicity is associated with respiratory depression that can lead to cardiorespiratory arrest. This is becoming an increasingly common cause of death in many countries.¹⁵⁴ The specific role of education and availability of naloxone for those with a high risk of opioid overdose is addressed in “Part 3: Adult Basic Life Support and Automated External Defibrillation.” Here we address whether any specific modifications to ALS are required when cardiac arrest is precipitated by opioid toxicity.

Cardiac arrest and respiratory arrest were considered separately. We sought evidence that compared results with any changes in usual resuscitation sequences or interventions in the setting of opioid overdose. Administration of the opioid antagonist naloxone was the only intervention for which literature was identified.

Consensus on Science

For the important outcome of survival with favorable neurologic outcome at discharge, 30 days, 60 days, 180 days, and/or 1 year, survival only at discharge, 30 days, 60 days, 180 days, and/or 1 year, ROSC, after opioid-induced cardiac arrest, we found no study with comparative data beyond standard ALS care.

For the important outcome of survival with favorable neurologic outcome at discharge, 30 days, 60 days, 180 days, and/or 1 year, survival only at discharge, 30 days, 60 days, 180 days, and/or 1 year, ROSC, after opioid-induced respiratory arrest, we found no comparative studies. There were 12 studies of which 5 compared intramuscular and intranasal routes of naloxone administration (2 RCT,^156,157 3 non-RCT,^158–160 and 7 assessed the safety of naloxone use or were observational studies of naloxone use).^161–169, These studies report that naloxone is safe and effective in treatment of opioid-induced respiratory depression, that complications are rare and dose related, and that mortality is rare when patients refuse transfer after initial naloxone administration.

Treatment Recommendation

We recommend the use of naloxone by IV, intramuscular, subcutaneous, IO, or intranasal routes in respiratory arrest associated with opioid toxicity (strong recommendation, very-low-quality evidence). The dose of naloxone required will depend on the route.

We can make no recommendation regarding the modification of standard ALS in opioid-induced cardiac arrest.

Values, Preferences, and Task Force Insights

In making these recommendations, we place a high value on the potential of the opioid antagonist naloxone to reverse opioid-induced respiratory depression.

Knowledge Gaps

• There are no data on the use of any additional ALS therapies in opioid-induced cardiac arrest. In respiratory arrest, there is only evidence for the use of naloxone—no other adjuncts or changes in sequence of interventions. Studies of naloxone use in respiratory arrest were observational, looked at safety, or compared routes of administration.

Introduction

The possible treatments for massive PE include fibrinolytic therapy, surgical embolectomy, and percutaneous mechanical thrombectomy. Most retrospective studies do not make subgroup analysis of patients with suspected or confirmed PE. These treatments were assessed separately as therapies during cardiac arrest as a consequence of PE. The reported outcomes and follow-up of patients is very heterogeneous between studies.

Consensus on Science

Treatment Recommendations

We suggest administering fibrinolytic drugs for cardiac arrest when PE is the suspected cause of cardiac arrest (weak recommendation, very-low-quality evidence).

We suggest the use of fibrinolytic drugs or surgical embolectomy or percutaneous mechanical thrombectomy for cardiac arrest when PE is the known cause of cardiac arrest (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making these recommendations, we acknowledge the use of thrombolytic drugs, surgical embolectomy or percutaneous mechanical thrombectomy, or a combination for known PE in non–cardiac arrest patients. We acknowledge the potential risk of bleeding after fibrinolysis and place value in the choice of intervention taking into account location, availability of interventions, and contraindications to fibrinolysis.

Knowledge Gaps

• There is a paucity of data on the topic of pulmonary embolus and its diagnosis and management during cardiac arrest. Further high-quality studies are required.

Introduction

We examined the literature for any studies comparing novel treatments during cardiac arrest that occurs during cardiac catheterization in addition to standard ALS approaches (eg, defibrillation) to cardiac arrest. The search was intended to find studies about any changes in sequence of interventions or about routine use of advanced circulatory support techniques.

Consensus on Science

There were no comparative studies evaluating the survival benefit of mechanical CPR; however, individual noncomparative case series reported variable survival rates.

For the critical outcomes of survival with favorable neurologic/functional outcome at discharge, 30 days, 60 days, 90 days, 180 days, and 1 year, and the outcomes of survival at 30 days, 60 days, 90 days, and 180 days, and 1 year, no studies were identified.

For the critical outcomes of survival to discharge and survival to 6 months, and the important outcome of ROSC, very-low-quality evidence (downgraded for very serious imprecision and risk of bias) from 1 observational study¹⁷⁶ compared ECLS with intra-aortic balloon pump and medical therapy for cardiogenic shock during PCI for ST-segment elevation myocardial infarction. There were 21 subjects with cardiac arrest during PCI, and all survivors were in the ECLS group.

Treatment Recommendation

We suggest the use of ECLS as a rescue treatment when initial therapy is failing for cardiac arrest that occurs during coronary catheterization (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this weak recommendation, the task force puts a higher value on usual ALS measures such as defibrillation.

We have not made a specific recommendation here regarding the use of automated mechanical chest compressions, because we found no studies that addressed this question. We have suggested previously that automated mechanical chest compression devices are a reasonable alternative to high-quality manual chest compressions in situations where sustained high-quality manual chest compressions are impractical or compromise provider safety. This earlier weak recommendation could, therefore, apply to cardiac arrest during coronary catheterization. ECLS encompasses ECPR. We have already suggested ECPR is a reasonable rescue therapy for selected patients with cardiac arrest when initial conventional CPR is failing in settings where this can be implemented.

Knowledge Gaps

• There is a lack of data about specific interventions to treat cardiac arrest during coronary catheterization.

Introduction

Previous preclinical work suggests that hyperoxia may be injurious in the post–cardiac arrest period. However, whether these findings apply to humans remains unclear. This PICO question evaluated whether titration of oxygen in post–cardiac arrest patients alters outcome.

Consensus on Science

Treatment Recommendations

We recommend avoiding hypoxia in adults with ROSC after cardiac arrest in any setting (strong recommendation, very-low-quality evidence).

We suggest avoiding hyperoxia in adults with ROSC after cardiac arrest in any setting (weak recommendation, very-low-quality evidence).

We suggest the use of 100% inspired oxygen until the arterial oxygen saturation or the partial pressure of arterial oxygen can be measured reliably in adults with ROSC after cardiac arrest in any setting (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making these recommendations, we think, despite the very-low-quality evidence, there is likely to be far greater actual harm from hypoxia and, therefore, make a strong recommendation that hypoxia should be avoided. The evidence for harm associated with hyperoxia is of very low quality and inconsistent, hence the weak recommendation.

Knowledge Gaps

• There is a lack of clinical trials evaluating titration of oxygen after ROSC.

• Observational data vary considerably on definitions of hyperoxia and the optimal timing and mechanisms for measurement (arterial oxygenation versus oxygen saturation).

• Future studies are necessary to define the optimal approach to titration of oxygen in post–cardiac arrest patients taking into account measurement as well as timing/duration.

Introduction

Post–cardiac arrest patients often have pulmonary injury and/or aspiration, and also have the added concern of ischemia-reperfusion injury to the brain. Thus, the post–cardiac arrest ventilator management may need to consider both brain and lung injury when determining specific strategies for mechanical ventilation. This PICO question addressed whether mechanical ventilation after cardiac arrest to achieve any specific Paco₂ goal was superior to any other Paco₂ goal.

Consensus on Science

No studies have specifically randomized patients to ventilation to a specific Paco₂ goal.

Treatment Recommendation

We suggest maintaining Paco₂ within a normal physiological range as part of a post-ROSC bundle of care (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, the task force did not find good evidence to suggest or recommend either hypercarbia or hypocarbia. In the absence of evidence to that end, combined with a potential suggestion of harm, we suggest maintaining normocarbia. Many physiological considerations may influence selection of Paco₂ goals for individual patients.

Knowledge Gaps

• There are no randomized prospective controlled trials evaluating different Paco₂ goals in post–cardiac arrest patients.

• Evaluation of optimal Paco₂ goals may need to be determined in populations both with and without lung injury.

Introduction

In the post–cardiac arrest period, patients often have persistent tissue hypoperfusion/hemodynamic instability. The optimal approach to resuscitation of patients after cardiac arrest remains unknown.

Consensus on Science

There are no RCTs addressing hemodynamic goals after resuscitation.

Treatment Recommendations

We suggest hemodynamic goals (eg, MAP, SBP) be considered during postresuscitation care and as part of any bundle of postresuscitation interventions (weak recommendation, low-quality evidence).

There is insufficient evidence to recommend specific hemodynamic goals; such goals should be considered on an individual patient basis and are likely to be influenced by post–cardiac arrest status and pre-existing comorbidities (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making these recommendations, we place a higher value on the recognition that while hemodynamic goals are likely important to optimize outcome, specific targets remain unknown and likely vary depending on individual physiology and comorbid status.

Knowledge Gaps

• There are no prospective, randomized trials on specific hemodynamic targets or goals with respect to outcome.

• Comorbidities and the complexities of individual-based physiology should ideally be taken into account in future investigations into hemodynamic targets/goals.

• Future studies of measurement of actual blood flow and tissue perfusion, particularly cerebral perfusion, and the role of noninvasive technology are desirable.

Introduction

After ROSC from cardiac arrest, the decision to initiate or continue therapy with antiarrhythmic medications remains uncertain. Literature was found for both β-blocking medications and lidocaine. We identified no studies of magnesium, amiodarone, procainamide, bretylium, or nifekalant.

Consensus on Science

Treatment Recommendation

We make no recommendation about the routine prophylactic use of antiarrhythmic drugs post after ROSC (GRADE used for evidence evaluation and synthesis only, very low confidence in effect estimate).

Values, Preferences, and Task Force Insights

The available data were too limited to have any confidence in any effect, and, therefore, no recommendation is made. We also place value on avoiding known side effects of medications when the treatment effect was unproven or unknown. The studies evaluated were all observational, and no causal relation could be determined. Moreover, they were performed before changes in current practice (ie, currently amiodarone is used during cardiac arrest more than lidocaine).

Knowledge Gaps

• There are no randomized trials for any antiarrhythmic drug in the post–cardiac arrest period.

• Patients resuscitated from VF/pVT who have received an antiarrhythmic medication during the cardiac arrest period are a specific population of interest.

Introduction

This PICO review was divided into 2 questions. The first question evaluated whether induced hypothermia should be initiated for postarrest patients. Evidence was evaluated separately for OHCA with shockable rhythms, OHCA with nonshockable rhythms, and IHCA (all rhythms). The second question evaluated the optimal target temperature for postarrest patients.

Consensus on Science

Treatment Recommendations

We recommend selecting and maintaining a constant target temperature between 32°C and 36°C for those patients in whom temperature control is used (strong recommendation, moderate-quality evidence). Whether certain subpopulations of cardiac arrest patients may benefit from lower (32°C–34°C) or higher (36°C) temperatures remains unknown, and further research may help elucidate this.

We recommend TTM as opposed to no TTM for adults with OHCA with an initial shockable rhythm who remain unresponsive after ROSC (strong recommendation, low-quality evidence).

We suggest TTM as opposed to no TTM for adults with OHCA with an initial nonshockable rhythm who remain unresponsive after ROSC (weak recommendation, very-low-quality evidence).

We suggest TTM as opposed to no TTM for adults with IHCA with any initial rhythm who remain unresponsive after ROSC (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making these recommendations, we place a higher value on the potential for increased survival with good neurologic outcome as compared with the possible risks (which appear to be minimal) and the cost of TTM. We emphasize that the mortality after cardiac arrest is high and the treatment options are limited. Although the evidence for TTM compared with no temperature management is of low quality, it is the only post-ROSC intervention that has been found to improve survival with good neurologic outcome. We have, therefore, made our recommendation strong in spite of the low-quality evidence.

Knowledge Gaps

• There is no high-quality evidence to support or refute the use of TTM in adults with OHCA and nonshockable initial rhythm, or adults with IHCA and any initial rhythm.

• There is no evidence to support or refute specific temperature targets tailored to individual patients based on post–cardiac arrest injury severity.

• Studies including more detailed neurocognitive evaluations to determine outcome in a more granular fashion are also needed.

Introduction

The hypothermia trials published in 2002 used 12 hours and 24 hours of cooling,^201,202 which was adopted in subsequent guidelines.²⁰⁹ The optimal duration for TTM remains unknown.

Consensus on Science

We found no human trials comparing different durations of TTM after cardiac arrest.

For the critical outcome of favorable neurologic outcome, very-low-quality evidence (downgraded for risk of bias and imprecision) from 2 observational trials found no difference in duration of hypothermia²¹⁰ and no difference in mortality or poor neurologic outcome with 24 hours compared with 72 hours of hypothermia.²¹¹ Previous trials treated patients with 12 to 28 hours of TTM.^201,202,208 One trial provided strict normothermia (less than 37.5°C) after hypothermia until 72 hours after ROSC.²⁰⁸

Treatment Recommendations

We suggest that if TTM is used, duration should be at least 24 hours, as done in the 2 largest previous RCTs^201,208 (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we place a high value on not changing current clinical practice, which most commonly is a TTM duration of 24 hours. We further note that the 2 largest trials related to TTM both used at least 24 hours, one of which found an outcome benefit when compared with not using TTM.

Knowledge Gaps

• There is no direct evidence to support or refute any specific duration of TTM.

• Controlled, randomized, human trials to evaluate duration of TTM are needed.

Introduction

Prior recommendations suggest that cooling should be initiated as soon as possible after ROSC, but this recommendation was based only on preclinical data and rational conjecture.²⁰⁹ This question addressed whether early cooling was superior to delayed cooling. Early cooling was defined as prehospital cooling before hospital arrival. Because multiple trials were available, only RCTs were included.

Consensus on Science

Five RCTs^212–216 used cold IV fluids after ROSC to induce hypothermia, 1 trial used cold IV fluid during resuscitation,²¹⁷ and 1 trial used intra-arrest intranasal cooling.²¹⁸ The volume of cold fluid ranged from 20 to 30 mL/kg and up to 2 L, though some patients did not receive the full amount before hospital arrival. One small feasibility trial was not included.²¹⁹ All 7 trials suffered from the unavoidable lack of blinding of the clinical team, and 3 also failed to blind the outcomes assessors.

For the critical outcome of favorable neurologic outcome, 5 trials with a total of 1867 subjects with OHCA^{213–⇓216,218⇓} provided overall moderate-quality evidence (downgraded for risk of bias), showing that neurologic outcomes did not differ after initiation of induced hypothermia in the prehospital environment compared with later initiation (RR, 1.00; 95% CI, 0.95–1.06).

For the critical outcome of mortality, 7 trials with a total of 2237 subjects provided moderate-quality evidence (downgraded for risk of bias), showing no overall difference in mortality for patients treated with prehospital cooling (RR, 0.98; 95% CI, 0.92–1.04) compared with those who did not receive prehospital cooling. No individual trial found an effect on either poor neurologic outcome or mortality.

For the outcome of rearrest, 4 RCTs provided low-quality evidence (downgraded for risk of bias and inconsistency) for an increased risk among subjects who received prehospital induced hypothermia (RR, 1.22; 95% CI, 1.01–1.46).^{212,213,215,216} This result was driven by data from the largest trial.²¹⁶

For the outcome of pulmonary edema, 3 trials reported no pulmonary edema in any group. Two small pilot trials^212,217 found no difference between groups, and 1 trial showed an increase in pulmonary edema in patients who received prehospital cooling (RR, 1.34; 95% CI, 1.15–1.57).²¹⁶ These trials provided overall low-quality evidence (downgraded for risk of bias and inconsistency).

Treatment Recommendations

We recommend against routine use of prehospital cooling with rapid infusion of large volumes of cold IV fluid immediately after ROSC (strong recommendation, moderate-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we place high value on not recommending an intervention with no proven benefit despite a large number of patients studied. We further note that the meta-analysis driven by the results from the largest study found also noted an increased risk of rearrest with prehospital induction of mild hypothermia using rapid infusion of cold IV fluid.²¹⁶ This recommendation is specific to the prehospital setting after ROSC, and we acknowledge that cold IV fluid might still be used in patients who have been further evaluated or in other settings.

Knowledge Gaps

• Early cooling strategies, other than rapid infusion of large volumes of cold IV fluid, and cooling during CPR in the prehospital setting have not been studied adequately.

• Whether certain patient populations (eg, patients for whom transport time to a hospital is longer than average) might benefit from early cooling strategies remains unknown.

Introduction

Fever is associated with poor outcome in many critical illnesses with neurologic injury. Increased temperature may aggravate ischemia-reperfusion injury and neuronal damage through increased metabolic activity. We examined whether fever prevention improves outcomes in patients not receiving TTM and in patients after the use of TTM.

No interventional or observational studies were identified addressing whether fever prevention (or treatment) after cardiac arrest improves outcome. We, therefore, included studies that examined the association between fever and outcomes.

Consensus on Science

Treatment Recommendation

We suggest prevention and treatment of fever in persistently comatose adults after completion of TTM between 32°C and 36°C (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we recognize that TTM always should be used in comatose patients after cardiac arrest, and that fever will not occur during this time. Thus, fever management is primarily a concern after TTM has been completed. Despite substantial limitations of the included studies, expert opinion within the task force combined with the fact that fever prevention is common practice for other neurologic injuries in the ICU and the relative low risk of harm associated with fever prevention prompted us to recommend in favor of fever prevention.

Knowledge Gaps

• In the absence of RCTs, whether the prevention or treatment of fever after cardiac arrest is beneficial remains unclear.

• It is unclear how long fever prevention is necessary, and what technique (eg, external, internal, pharmacologic) is best.

• Data to date cannot distinguish whether fever causes increased neurologic injury or severe neurologic injury causes temperature dysregulation.

Introduction

Seizures, and particularly status epilepticus (SE), are associated with poor outcomes in comatose post–cardiac arrest patients. While seizure and SE can be the result of severe brain injury caused by cardiac arrest, these disorders also have the potential to exacerbate brain injury caused by cardiac arrest. We examined whether seizure prophylaxis or effective seizure management improve outcomes of post–cardiac arrest patients.

Consensus on Science

For the critical outcome of survival with favorable neurologic/functional outcome, moderate-quality evidence (downgraded for indirectness) from 2 prospective double-blinded randomized clinical trials involving a total of 312 subjects^233,234 and 1 nonrandomized prospective clinical trial that used historic controls with 107 subjects²³⁵ detected no benefit of seizure prophylaxis.

In 1 block randomized trial,²³⁴ OHCA patients with ROSC received either placebo, diazepam, magnesium sulfate, or diazepam plus magnesium sulfate. The percentage of patients independent at 3 months was 25.3% (19/75) in the placebo group, 34.7% (26/75) in the magnesium group, 17.3% (13/75) in the diazepam group, and 17.3% (13/75) in the diazepam plus magnesium group (for magnesium: RR, 1.22; 95% CI, 0.81–1.83). After adjusting for baseline imbalances, outcomes did not differ between groups. In a trial of thiopental versus placebo within 1 hour of ROSC,²³³ 1-year survival with good cerebral function was 15% (20/131) in the placebo group and 18% (24/131) in the thiopental group (RR, 1.20; 95% CI, 0.70–2.06). After multivariate adjustment, groups did not differ (OR, 1.18; 95% CI, 0.76–1.84). A nonrandomized clinical trial²³⁵ showed no benefit of barbiturate therapy in comatose post–cardiac arrest patients using a combination of thiopental and phenobarbital when compared with historic controls. In this study, survival to hospital discharge with favorable neurologic outcome was 38% (20/53) in the barbiturate group and 26% (14/54) in the historic control group (ARR, 11.8%; 95% CI, −5.8 to 28.5; or 118 more patients/1000; 95% CI, 58 fewer to 285 more patients/1000). One case series showed that 9 of 10 patients with anesthesia-associated cardiac arrest survived with good neurologic outcome when single-dose phenytoin was administered early after ROSC.²³⁶

For the important outcome of seizure prevention, we identified low-quality evidence downgraded for indirectness from 2 prospective double-blinded RCTs^233,234 showing no benefit of seizure prophylaxis. In 1 trial of thiopental treatment,²³³ 21% (28/131) of control subjects and 13% (17/131) of thiopental-treated subjects had seizures (ARR, −8.4%; 95% CI, −17.5 to 0.8; 84 fewer patients/1000; 95% CI, 175 fewer to 8 more patients/1000). The incidence of seizures in a second trial²³⁴ was 11.9% in all treatment groups (double placebo, magnesium plus placebo, diazepam plus placebo, and diazepam plus magnesium).

Treatment Recommendation

We suggest against routine seizure prophylaxis in post–cardiac arrest patients (weak recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, the task force acknowledged the lack of confidence in a treatment effect on the critical outcome of survival with good neurologic function treatment. The task force also considered that seizure prophylaxis in other forms of acute brain injuries is not associated with improved outcomes, and that most drugs have significant side effects.

Knowledge Gaps

• Standardized definitions for diagnosing seizures in comatose post–cardiac arrest patients have not been used.

• The utility of continuous electroencephalogram (EEG) versus intermittent EEG monitoring versus no EEG in the diagnosis and treatment of seizures in comatose post–cardiac arrest patients remains controversial due to resource utilization and lack of evidence for improved outcomes.

• There are no RCTs designed to evaluate the impact of seizure prophylaxis after ROSC on incidence of seizures and neurologic outcome.

• There are inadequate data regarding the timing, duration, dosing, and choice of antiepileptic drugs for seizure prophylaxis in comatose post–cardiac arrest patients.

Consensus on Science

There are no RCTs addressing this question.

For the critical outcome of survival with favorable neurologic outcome at discharge, 30 days, 60 days, 180 days, and/or 1 year, very-low-quality evidence (downgraded for lack of concurrent comparative data) from 3 case series^237–239 showed only 1/47 post–cardiac arrest patient treated for seizures or SE survived with good neurologic function. The antiepileptic drugs used were widely variable (phenytoin, levetiracetam, sodium valproate, clonazepam, propofol, and midazolam); included general anesthetics; and the drug, dose, and timing of therapy were not consistently reported. In these reports, no post–cardiac arrest patients with seizures were left untreated, providing no insight into the impact of antiepileptic drug therapy on survival or neurologic outcome. In 1 study, effective seizure control was achieved in 4/5 patients treated for seizures, and 0/5 survived with good neurologic function.²³⁹ In 1 study, effective seizure control was achieved in 0/24 patients with SE, and 1/24 patients survived with good neurologic function.²³⁷

Treatment Recommendation

We recommend the treatment of seizures in post–cardiac arrest patients (strong recommendation, very-low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we acknowledge very low confidence in the estimated treatment effect. However, ongoing seizures have the potential to worsen brain injury, and treatment of recurrent seizures and SE is the standard of care in other patient populations.

Knowledge Gaps

• Standardized definitions for diagnosing seizures in comatose post–cardiac arrest patients have not been used.

• The utility of continuous EEG versus intermittent EEG monitoring versus no EEG in the diagnosis and treatment of seizures in comatose post–cardiac arrest patients remains controversial due to resource utilization and lack of evidence for improved outcomes.

• There are no RCTs designed to evaluate the impact of seizure prophylaxis after ROSC on incidence of seizures and neurologic outcome.

• There are inadequate data regarding the timing, duration, dosing, and choice of antiepileptic drugs for seizure prophylaxis in comatose post–cardiac arrest patients.

• The threshold for treating epileptiform activity other than convulsive seizures (eg, generalized epileptiform discharges) is poorly defined.

Introduction

Glycemic control with insulin is now common for critically ill patients, and hyperglycemia after cardiac arrest is associated with worse neurologic outcomes. We examined whether specific glucose values other than those selected for other critically ill patients should be targeted in patients after resuscitation from cardiac arrest.

Consensus on Science

For the critical outcome of survival to hospital discharge, there was moderate-quality evidence (downgraded for risk of bias due to lack of blinding) from 1 RCT of 90 subjects showing reduced 30-day mortality (RR, 0.94; 95% CI, 0.53–1.68) when subjects were assigned to strict (4–6 mmol/L) versus moderate (6–8 mmol/L) glucose control.²⁴⁰ One before-and-after observational study of 119 subjects provided very-low-quality evidence (downgraded for multiple potential confounding variables) of reduced in-hospital mortality (RR, 0.46; 95% CI, 0.28–0.76) after implementation of a bundle of care that included defined glucose management (5–8 mmol/L).¹⁹³ The effect of glucose management cannot be separated from the effects of other parts of the bundle.

Treatment Recommendation

We suggest no modification of standard glucose management protocols for adults with ROSC after cardiac arrest (weak recommendation, moderate-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we considered the lack of evidence that the approach to glucose management chosen for other critical care populations should be modified for the post–cardiac arrest patients. Moreover, we noted that strict glycemic control is labor intensive, and in other populations, implementation of strict glycemic control is associated with increased episodes of hypoglycemia, which might be detrimental. Avoiding hypoglycemia was considered more important than the unproven benefits of treating moderate hyperglycemia.

Knowledge Gaps

• It remains unknown whether maintaining serum glucose within a specific range or minimizing variability in post–cardiac arrest patients will improve survival and/or neurologic outcome.

Consensus on Science

Treatment Recommendations

We suggest against the use of clinical criteria alone before 72 hours after ROSC to estimate prognosis (weak recommendation, low-quality evidence).

We suggest that multiple modalities of testing (clinical exam, neurophysiological measures, imaging, or blood markers) be used to estimate prognosis instead of relying on single tests or findings (weak recommendation, low-quality evidence).

Knowledge Gaps

Consensus on Science

No study on clinical examination reported blinding of the treating team to the results of the index test. Blinding of the treating team is very difficult to achieve for predictors based on clinical examination, which implies a risk of self-fulfilling prophecy.

For the critical outcome of survival with unfavorable neurologic status or death at discharge, we identified 2 studies on pupillary reflex and motor response or oculocephalic reflex (151 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).^298,299

For the critical outcome of survival with unfavorable neurologic status or death at 30 days, we identified 1 study on GCS (97 patients; very-low-quality evidence downgraded for very serious bias and very serious imprecision).³⁰⁰

For the critical outcome of survival with unfavorable neurologic status or death at 90 days, we identified 2 studies on corneal reflex, pupillary reflex, motor response, oculovestibular reflex, GCS, or myoclonus (97 patients; very-low-quality evidence downgraded for serious or very serious bias and serious or very serious imprecision).^301,302

For the critical outcome of survival with unfavorable neurologic status or death at 180 days, we identified 4 studies on brainstem reflexes, motor response, or myoclonus (650 patients; very-low-quality evidence downgraded for serious or very serious bias and very serious imprecision).^303–306

For the critical outcome of survival with unfavorable neurologic status or death at 1 year, we identified 3 studies on brainstem reflexes, motor response, GCS, or myoclonus (172 patients; very-low-quality evidence downgraded for serious or very serious bias and very serious imprecision).^307–309

Treatment Recommendations

Knowledge Gaps

Introduction

Resuscitation from cardiac arrest is not always successful, and many patients who are initially resuscitated from cardiac arrest will subsequently die in the hospital. Whether these nonsurviving patients can become organ donors has been debated because of the potential injury to organs during the initial cardiac arrest.

The committee reviewed experience about donation from this population that has accumulated in recent years. Two situations were separately considered. In the first, an individual who dies after being resuscitated by successful CPR may become an organ donor after brain death or having withdrawal of life-sustaining treatment. In the second situation, an individual may die because of unsuccessful CPR in a center with a rapid response system that allows procurement of organs after unsuccessful CPR. For kidney transplants, the primary outcomes were graft function, because recipients can survive with renal replacement therapy even with graft failure. For other organs, recipient death was considered equivalent to graft failure. Only studies that allowed comparison of organs procured in these situations with other organs from non-CPR donors were selected for review.

Consensus on Science

Treatment Recommendation

We recommend that all patients who have restoration of circulation after CPR and who subsequently progress to death be evaluated for organ donation (strong recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we consider the absence of any evidence of worse graft function from donors with antecedent CPR, the desirability of providing more organs to waiting recipients, and the absence of any risk to the donor. As in all organ donations, the function of the donated organ determines whether procurement and transplantation proceed. Therefore, there is also precaution to ensure the safety of the recipient.

Treatment Recommendation

We suggest that patients who fail to have restoration of circulation after CPR and who would otherwise have termination of CPR efforts be considered candidates for kidney or liver donation in settings where programs exist (weak recommendation, low-quality evidence).

Values, Preferences, and Task Force Insights

In making this recommendation, we consider the evidence that kidney grafts obtained from donors in whom CPR failed can function at rates comparable to kidneys obtained from other donors, and that recipients can safely tolerate delayed graft function that is common with kidneys obtained in this manner. We also consider the immediate lifesaving potential of liver grafts, which offsets the potentially greater rate of long-term graft failure in livers obtained from donors with ongoing CPR.

Knowledge Gaps

• The optimal methods for organ procurement after failed CPR are unknown.

• Barriers to consenting to this organ donation and the acceptability of these practices in different settings are unknown.