Commentary:

Continuous or Interrupted Chest Compressions for Cardiac Arrest

Rudolph W. Koster, M.D., Ph.D.

November 9, 2015DOI: 10.1056/NEJMe1513415

Article

References

High-quality cardiopulmonary resuscitation (CPR) is identified as a critical but often poorly performed component of the rescue efforts for patients in cardiac arrest. Chest compressions have often been too shallow, and compression rates too low or too high. Prolonged interruptions of chest compressions have been observed during resuscitation both in the hospital and outside the hospital.1,2 All prolonged pauses (not only those for defibrillation) are associated with worse survival.3 Interruptions of chest compressions cause a rapid decline in coronary perfusion pressure, reducing myocardial blood flow, which has previously been shown with shorter interruptions for rescue breathing.4 Experiments in animals have suggested that the rate of survival may increase if CPR is performed with continuous chest compressions, not interrupted for ventilations. Retrospective cohort studies have seemed to confirm this concept. A prospective statewide observational study in Arizona showed that training the population in continuous chest compressions until the arrival of emergency medical services (EMS) increased the rate of bystander-initiated CPR and increased the rate of survival to discharge from the hospital.5

Randomized studies involving patients with cardiac arrest are difficult and require considerable resources that are often not available. In the EMS setting, the concept of continuous chest compressions has been introduced and its potential benefit has been studied in observational studies with historical controls. In the largest of these studies, several measures were introduced simultaneously as a “bundle of care.” In addition to three periods of 200 chest compressions each, which were interrupted only for rhythm analysis and defibrillation, this bundle of care included a single-shock scenario (three stacked shocks were allowed previously), no delay of chest compressions for rhythm or pulse checks, deferred insertion of an advanced-airway device, and passive oxygen insufflation replacing positive-pressure ventilation until 6 minutes had passed during which the three periods of 200 chest compressions were delivered. The introduction of this bundle of care resulted in a significant increase in the rate of survival to discharge, from 1.8% to 5.4%; among patients with witnessed arrest and ventricular fibrillation, the rate increased from 4.7% to 17.6%.6 On the basis of this study and similar studies, the 2015 American Heart Association (AHA) guidelines for resuscitation included a new class IIb recommendation that it may be reasonable for EMS to initiate resuscitation with three initial periods of 200 continuous chest compressions with passive oxygen insufflation.7 This recommendation was not made in the concurrent 2015 guidelines for resuscitation from the European Resuscitation Council (ERC).8

Bundles of care are a pragmatic way to introduce and study new treatments. But if studies show higher rates of survival with the new techniques, it is not clear which components of the bundle contributed to the improved survival. The results of a new randomized clinical trial from the Resuscitation Outcomes Consortium (ROC) have now been published in the Journal.9 This trial was designed as a cluster-randomized study of non–trauma-related cardiac arrest treated by EMS providers. Patients received either continuous chest compressions or the standard approach of chest compressions that were interrupted for positive-pressure ventilation in a ratio of 30 compressions to two ventilations (termed “interrupted chest compressions”). In the group that received continuous chest compressions, asynchronous positive-pressure ventilations were given with a recommended rate of 10 ventilations per minute. The primary outcome of the study was the rate of survival to hospital discharge.

A total of 12,653 patients were included in the group that received continuous chest compressions (intervention group) and 11,058 in the group that received interrupted chest compressions (control group). The overall rate of survival to hospital discharge was 9.0% in the intervention group and 9.7% in the control group — a nonsignificant difference. Survival with favorable neurologic function at discharge, defined as a score of 3 or less on the modified Rankin scale (on which scores range from 0, indicating no symptoms, to 6, indicating death), did not differ significantly between the two groups. A prespecified per-protocol analysis that was based on strict adherence to the treatment algorithm showed significantly lower rates of survival among patients in the intervention group than among those in the control group (7.6% vs. 9.6%).

Why did this new large, randomized study show no benefit from continuous chest compressions, whereas previous observational studies showed a clear survival benefit among patients treated with this approach? First, in the bundle-of-care studies, measures other than the continuous chest compressions could be the changes that improved the rate of survival. Second, in this study, the mean chest-compression fraction (the proportion of each minute during which compressions were given), which is an important marker of interruptions of chest compressions, was already high (0.77) in the control group and was not much lower than the mean chest-compression fraction of 0.83 in the intervention group. Both values were much higher than the target for chest-compression fraction of more than 0.60 that is recommended in the new AHA and ERC resuscitation guidelines.7,8 Third, pauses for ventilation may be less critical, and less detrimental for survival, than is currently believed.10 And of course, the randomized trial is the best tool to investigate causality.

The new 2015 AHA resuscitation guidelines were published only recently.7 If the results of the current ROC study had been available, the guidelines committee might have decided to retain the previous recommendation to give chest compressions interrupted for ventilations and perhaps even to upgrade that recommendation to a class IIa recommendation for EMS providers. Should the AHA reconsider their recommendation?