Customize your JAMA Network experience by selecting one or more topics from the list below.
Out-of-hospital cardiac arrest is a major health problem in the United States. Survival rates vary substantially by geographic region, ranging from 7% to 40% for patients with cardiac arrest due to ventricular fibrillation or ventricular tachycardia (shockable rhythms) and from 2% to 5% for patients with cardiac arrest initially found with pulseless electrical activity or asystole (nonshockable rhythms).1 Overall outcomes have not substantially improved for decades, and improvements in outcomes are made more difficult by an unexplained decline in the incidence of ventricular fibrillation as the first documented rhythm. The medical community’s failure to improve survival and the rate of return to pre-event quality of life has been highlighted by a recent report from the Institute of Medicine (IOM), titled Strategies to Improve Cardiac Arrest Survival: A Time to Act.2 It is indeed time to act.
Out-of-hospital cardiac arrest is a “community” event or catastrophe and logically should include community members in efforts to improve outcomes. Early access to professional rescuers via a 9-1-1 telephone system now exists in nearly all municipalities with an emergency medical services (EMS) system. Accomplishing early access was comparatively easy compared with the tasks ahead and outlined in the IOM report. The report emphasizes the following strategies: “(1) Establish a national registry of cardiac arrest in order to monitor performance in terms of both success and failure, identify problems, and track progress. (2) Enhance performance of EMS systems with emphasis on dispatch-assisted CPR [cardiopulmonary resuscitation] and high-performance CPR. (3) Develop strategies to improve systems of care within hospital settings and special resuscitation circumstances. (4) Expand basic, clinical, translational, and health services research in cardiac arrest resuscitation and promote innovative technologies and treatments. (5) Educate and train the public in CPR, use of automated external defibrillators, and EMS-system activation. (6) Create a national cardiac arrest collaborative to unify the field and identify common goals to improve survival.”2
Community engagement should include early CPR, as captured in the fifth of the IOM strategies. This has been a formidable and challenging task, with training undertaken by a number of entities (eg, the American Heart Association and the American Red Cross) with variable success, at best. At present, only about 30% of those with of out-of-hospital cardiac arrest receive bystander-initiated CPR despite decades of instructional efforts using different platforms, ranging from lectures and mannequin practice to instructional videos. Barriers to success have been identified and include socioeconomic and racial/ethnic factors; reported racial/ethnic disparities in outcomes after cardiac arrest are remarkably large and likely related, in part, to a lower likelihood of early CPR.3,4
The concept and implementation of dispatcher-assisted CPR were first reported in 1984, with the goal of improving the likelihood that a bystander would provide early rescue efforts.5 More than 20 years later, it is still an important idea and is captured in the second of the IOM strategies. Dispatcher- or telephone-assisted CPR is intended to guide the untrained rescuer in the performance of CPR, more specifically chest compression–only resuscitation, after cardiac arrest is confirmed or strongly suspected based on the level of consciousness and the absence of normal breathing. There are now more than 100 publications included in PubMed dealing with this topic. Despite the rapid and widespread adoption of this practice, a recent evidence review,6 undertaken during the course of developing the American Heart Association 2015 guidelines for emergency cardiac care and CPR, suggests that dispatcher instruction in CPR does not lead to more successful resuscitations or improved survival. This evidence review included a meta-analysis, 3 randomized clinical trials, and 11 observational studies.6 Although dispatcher-guided CPR increased the likelihood that chest compressions would be performed, there appeared to be no association with restoration of spontaneous circulation and no benefit when survival with good neurologic outcome was the primary outcome variable. The included meta-analysis reported an absolute survival benefit of only 2.4%.6
In this issue of JAMA Cardiology, Bobrow et al7 report their experience with a structured and guideline-based dispatcher-assisted telephone CPR program in Phoenix, Arizona. Study outcomes were dramatically different from recent studies: survival with a good neurologic outcome was statistically better after implementation of dispatcher-assisted CPR instruction. The investigators used a well-structured before-after study design in an established EMS system and found a 3.1% (95% CI, 1.5-4.9) adjusted absolute increase in survival across all rhythm types and an increase of 9.6% (95% CI, 4.8-14.4) for those with shockable rhythms.7 These improvements translate to a number needed to treat of 32 for all patients and 10 for patients with shockable rhythms. The fraction with good neurological outcome showed an overall absolute adjusted increase of 2.7% (95% CI, 1.3-4.4; number needed to treat, 37). What made this study different?
Bobrow et al7 incorporated guideline-based recommendations for identifying cardiac arrest and a rigorous training and quality-improvement program as key components at the implementation of phase 2 of the study. The quality-improvement program included case reviews and case-level feedback to individual dispatchers with coaching, if needed. Emphasis was placed on chest compression only for cardiac arrests of presumed cardiac etiology, with ventilation instructed only for events not believed to be of primary cardiac origin. Phoenix supports a mature EMS system with a robust data collection platform. Although population data were deidentified, the investigators were able to follow up patients longitudinally from study entry through hospitalization and after discharge. Only 9% of eligible events were excluded because of unknown outcomes or fragmented or incomplete data, minimizing bias in assessing outcomes and the effect of the intervention. Metropolitan Phoenix has adopted a regionalized system for cardiac care, and it is noteworthy that patients with out-of-hospital cardiac arrest transported to any of 23 cardiac receiving centers did better than those who were not. Although specific data are not reported by the investigators, it is possible that patients transported to a cardiac receiving center were more likely to undergo cardiac catheterization and percutaneous coronary intervention if indicated and to receive targeted temperature interventions.
It is difficult to quantify the effect of telephone-assisted CPR on improving outcomes of patients in subgroups defined by the first recorded rhythm because the intervention precedes the assessment of the rhythm. This sequence of events can lead to a bias in the estimation of treatment effects within subgroups, as the provision of CPR undoubtedly slows the degradation of ventricular fibrillation into asystole. Thus, some patients in the shockable rhythm subgroup in the telephone-assisted CPR population would likely have been in the nonshockable rhythm subgroup if they had been treated in the early control phase of the trial and not received early CPR. This movement between subgroups is likely to decrease the apparent benefit of telephone-assisted CPR in patients with shockable rhythms. Although the magnitude of the bias this introduces—bias that only exists when treatment effects are evaluated in the rhythm groups separately—is difficult to predict, it is possible that the true effect of the intervention in patients with shockable rhythms is even greater than reported.
As noted in prior work3,4 and the recent IOM report,2 racial and ethnic disparities in the provision of layperson CPR and outcomes of out-of-hospital cardiac arrest are substantial and disturbing from both ethical and public health perspectives. Bobrow et al7 did not categorize successful instruction and outcomes by racial or ethnic groups nor how or by whom instructions were given to non–English-speaking 9-1-1 callers. Did the regional telephone-assisted CPR program affect outcomes in groups with historically low layperson CPR rates?
Bobrow et al7 acknowledge the complexity of their EMS system and suggest that their findings are generalizable to other settings. The authors imply that if they can do it in Phoenix, then it can be done anywhere. We disagree. In our opinion, the completeness of success in other metropolitan areas will depend on the racial and ethnic diversity of the population and the ability of the system to accommodate the population it serves.
Corresponding Author: Roger J. Lewis, MD, PhD, Department of Emergency Medicine, Harbor–UCLA Medical Center, Bldg D9, 1000 W Carson St, Torrance, CA 90509 (firstname.lastname@example.org).
Published Online: May 4, 2016. doi:10.1001/jamacardio.2016.0278.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Niemann JT, Lewis RJ. The Resuscitative Power of a Telephone Call. JAMA Cardiol. 2016;1(3):302–304. doi:10.1001/jamacardio.2016.0278
Coronavirus Resource Center
Create a personal account or sign in to: