Continuous chest compressions during CPR failed to improve survival or neurologic function compared with standard chest compressions that are briefly interrupted for ventilation, based on findings in the first large randomized trial to compare the two strategies for out-of-hospital, nontraumatic cardiac arrest.
In a presentation at the American Heart Association scientific sessions, simultaneously published online Nov. 9 in the New England Journal of Medicine, Dr. Graham Nichol and his associates analyzed data from the Resuscitation Outcomes Consortium, a network of clinical centers and EMS agencies that have expertise in conducting research on out-of-hospital cardiac arrest.
Data were analyzed for 23,711 adults treated by 114 EMS agencies affiliated with eight clinical centers across the United States and Canada. These agencies were grouped into 47 clusters that were randomly assigned to perform CPR using either continuous chest compressions (100 per minute) with asynchronous positive-pressure ventilations (10 per minute) or standard chest compressions interrupted for ventilations (at a rate of 30 compressions per two ventilations) at every response to an out-of-hospital cardiac arrest. Twice per year, each cluster crossed over to the other resuscitation strategy, said Dr. Nichol of the University of Washington–Harborview Center for Prehospital Emergency Care and Clinical Trial Center in Seattle.
A total of 12,653 patients were assigned to continuous chest compressions (the intervention group) and 11,058 to interrupted chest compressions (the control group). The primary outcome – the rate of survival to hospital discharge – was 9.0% in the intervention group and 9.7% in the control group, a nonsignificant difference. Similarly, the rate of survival with favorable neurologic function did not differ significantly, at 7.0% and 7.7%, respectively, the investigators said (N Engl J Med. 2015 Nov 9. doi:10.1056/NEJMoa1509139).
The reason for these unexpected findings is not yet known. It is plausible that continuous chest compressions really don’t improve outcomes and that the previous positive results were actually due to improvements in the CPR process, such as more consistent rate and depth of compressions; concurrent improvements in the system of care; or Hawthorne effects, in which CPR providers altered their behavior during the studies because they were aware they were being observed.
However, it is also possible that three important limitations of this trial unduly influenced the results.
First, the per-protocol analysis, which used an automated algorithm to assess adherence to the compression assignments, could not classify many patients as having received either continuous or interrupted chest compressions. Second, the quality of postresuscitation care, which certainly influences outcomes, was not monitored. And third, actual oxygenation levels were not measured, nor were minutes of ventilation delivered. Thus, “we do not know whether there were important differences in oxygenation or ventilation between the two treatment strategies,” he said.