Seattle, WA and Ann Arbor, MI - A new analysis of out-of-hospital cardiac arrest (OHCA) in 10 areas in North America has found a fivefold difference in survival rates [1]. Dr Graham Nichol (University of Washington, Seattle) and colleagues from the Resuscitation Outcomes Consortium (ROC) report their results in the September 24, 2008 issue of the Journal of the American Medical Association.

Nichol et al say more research is needed to understand why there is such variation but note that their study demonstrates that cardiac arrest is a treatable condition. "If survival after OHCA treated by emergency medical services [EMS] could be increased throughout North America from the study average of 7.9% to the maximum observed rate of 16.3%, an estimated 15 000 premature deaths would be prevented each year," they point out.

In a related paper in the same issue [2], Dr Comilla Sasson (University of Michigan, Ann Arbor) and colleagues say that to try to improve outcomes, attention should be focused on those patients most likely to survive OHCA, and they suggest certain rules should be adopted by EMS to decide when it is futile to continue resuscitation.

If survival after OHCA treated by EMS could be increased . . . from the study average . . . to the maximum observed rate, an estimated 15 000 premature deaths would be prevented each year.

In an accompanying editorial [3], Drs Arthur B Sanders (University of Arizona Health Sciences Center, Tucson) and Karl B Kern (University of Arizona, Tucson) say that OHCA "has a dismal prognosis in many communities, [and] the magnitude of the problem in the US and Canada is such that even small improvements in survival translate into thousands of lives saved."

The study by Nichol et al illustrates that there is much room for improvement in the treatment of OHCA, they say, but they argue that applying rigid termination-of-resuscitation (TOR) criteria—as suggested by Sasson et al—is not necessarily the answer. Nichol agrees, telling heartwire: "Our point is more that there is a wide variation in care, so EMS agencies in [more poorly performing] cities need to try to improve their outcomes, rather than not resuscitating."

In their prospective, observational ROC study, Nichol et al set out to determine whether cardiac-arrest incidence and outcome differed across geographic regions and included data on all OHCAs at seven US and three Canadian sites from May 2006 to April 30, 2007, followed up to hospital discharge and including data available as of June 28, 2008. The 10 sites were: Alabama; Dallas, TX; Iowa; Milwaukee, WI; Ottawa, ON; Pittsburgh, PA; Portland, OR; Seattle, WA; Toronto, ON; and Vancouver, BC.

Among the 10 sites, with a total population of 21.4 million, there were 20 520 cardiac arrests assessed by EMS personnel. Resuscitation was attempted in 11 898 cases (58.0% of the total); 2729 (22.9% of those treated) had initial rhythm of ventricular fibrillation (VF) or ventricular tachycardia or rhythms that were shockable by an automated external defibrillator (AED), but just 954 (4.6%) were discharged alive. Among those in whom resuscitation was attempted, 7.9% of all patients with cardiac arrest and 21% of those with VF survived to hospital discharge.

But there was huge variation in survival rates across the sites, ranging from 3.0% in Alabama to 16.3% in Seattle, the best-performing region for cardiac-arrest survival rates. For VF survival, the figures were 7.7% in Alabama and 39.9% in Seattle. Median survival for cardiac arrest was 8.4% and for VF 22%.

"These findings have implications for prehospital emergency care. The fivefold variation in survival after EMS-treated cardiac arrest and fivefold variation in survival after ventricular fibrillation demonstrate that cardiac arrest is a treatable condition," they write.

Nichol told heartwire: "We are doing further work to understand why these differences occur, but it's probably for several reasons, including differences in patient risk, community response, and EMS response. The key point is for every city to monitor outcomes and try to improve them."

The editorialists agree: "It is time to work to overcome barriers in each community, devote appropriate resources, and optimize survival of all patients so that location by city becomes a minor factor in survival of cardiac arrest."

The key point is for every city to monitor outcomes and try to improve them.

Sasson told heartwire: "The ROC paper does a great job of showing that there are huge variations in the way that we practice out-of-hospital care." What this illustrates, she says, "is that most emergency physicians answering the phone to hear that they have a cardiac arrest coming in have no idea what they are going to get, that's the difficult part."

For example, an ER doctor working in Chicago—where there are 28 EMS agencies—receiving a call about a cardiac arrest would probably say, "To be on the safe side, just bring them in," she explains. But, someone working in Seattle, where the paramedics have a huge amount of training, might have a whole different level of comfort if their paramedic says, "This patient is dead," she says.

Bringing patients into the hospital who have no chance of surviving a cardiac arrest has a number of deleterious consequences, she believes. "It puts EMS providers' lives at risk when they are coming in at high speed, and it takes an ambulance away from someone else who might need it. Also, when that futile resuscitation comes into the ER, all the nurses and doctors there are focusing on it, so care in the ER stops, which may impact other patients for whom time is critical, such as those having a heart attack or stroke. That's the unfortunate part." In addition, patients with OHCA who have little or no chance of surviving to discharge are often admitted to the intensive care unit, wasting further resources, she notes.

"If we are able to accurately predict who will not survive a cardiac arrest, that's a huge step forward in trying to focus our resources on those patients who we know have a chance," she stresses.

In their paper, Sasson and her colleagues describe their retrospective cohort study using surveillance data prospectively submitted by EMS and hospitals in eight US cities to the Cardiac Arrest Registry to Enhance Survival (CARES) between October 2005 and April 2008, including 5505 patients with OHCA.

Their goal was to validate two out-of-hospital rules developed by the Ontario Prehospital Life Support (OPALS) study group for termination of resuscitation without transport to the hospital.

There were three criteria for basic life support (BLS) paramedics: event witnessed by EMS, cardiac arrest with AED/manual shockable rhythm in the field, and cardiac arrest with return of spontaneous circulation in the field. "If you meet even one of these three criteria, then you should be transported to the hospital for further care," says Sasson. But meeting none of them means it would be futile to take the patient to the hospital, she added.

For advanced life support (ALS) paramedics, there are two further criteria. If a bystander has witnessed the cardiac arrest and/or a bystander has delivered cardiopulmonary resuscitation in the field, then again the patient should be transported to the hospital.

"We found when we applied these rules to our patient sample that we could accurately predict, within 0.2%, which patients should be transported," Sasson explained.

The overall rate of survival to hospital discharge was 7.1%.

Using the criteria brought the field pronouncement rate of death up from the 17.2% currently reported by the cities participating in the CARES registry to 47.1% using the BLS rules and to 21.7% using the ALS rules, she said.

"When you extrapolate this to the huge number of OHCAs each year in the US [estimated at 166 000 to 310 000 per year], 60% of which are treated by EMS, this has huge implications," she says.

Sasson says applying the criteria suggested by this research will help to regulate OHCA care by EMS. "If we start to standardize the practice by which patients are transported to the hospital, then I think we can begin to look at why some systems are performing better than others, which is important in improving cardiac-arrest survival overall. At this point, there is so little standardization that we have no idea what the real reasons are as to why certain systems do better."

At this point, there is so little standardization that we have no idea what the real reasons are as to why certain systems do better.

But Nichol disagrees wholeheartedly. "My view is that the glass is half full, while Sasson's is that the glass is half-empty. I don't agree that too many patients are transported to the hospital. Wide implementation of rules for termination of resuscitation would indeed reduce regional variation, but by decreasing survival.

"We observed that regional variation after ventricular fibrillation was as large as variation after treated cardiac arrest," he points out. Because VF is usually regarded as a highly treatable condition, this infers that "differences in outcomes after VF are unlikely to be influenced by differences in decision to initiate or terminate resuscitation," he says. "Instead, the differences are due to other factors.

"Cities need to monitor and improve outcomes after cardiac arrest in their community," he stresses. "Since cardiac arrest is the third leading cause of death in either the US or Canada, we can prevent a lot of premature deaths by improving each community's response to cardiac arrest."

I don't agree that too many patients are transported to the hospital. Rules for termination of resuscitation would indeed reduce regional variation, by decreasing survival.

In their editorial, Sanders and Kern say the key issue is "whether TOR rules are desired and needed. Do such rules best serve patients who have a cardiac arrest?" They argue that the 2005 guidelines from the AHA already "clearly allow for pronouncing as dead those patients who experience OHCA and are unresponsive to advanced cardiac life-support treatment.

"If the problem is that too many patients are being transported to hospitals, education of base station physicians, medics, and EMS directors about what can and cannot be done in the hospital may be preferable to a rule to forgo resuscitation efforts," they say.

Resuscitation science is active, they add, with promising new approaches to improve outcomes. If the cities in the CARE database in the study by Sasson et al improved OHCA outcomes to the level of the best ROC site (the study by Nichol et al), "the TOR rules may no longer be valid."

But Sasson begs to differ. "The one point they missed is that we can't improve anything unless we focus our resources on those patients who actually have a chance. We need to figure out who those patients are and work to improve rates of survival for them. No matter what we do there is going to be a subset of patients who won't survive a cardiac arrest and for whom it will be fatal."

Nichol says: "Both studies show that outcomes differ from city to city. But we disagree on how to reduce these differences."

The editorialists, Sasson, and Nichol all agree on one thing, however: cardiac arrest should be designated a reportable disease.