CPR Survival Rates Can Differ Greatly by City

If you have a stroke, your odds of survival are similar whether you are in Boston or Boise. But not so if you fall victim to cardiac arrest.

Dr. Sam Parnia, the director of resuscitation research at Stony Brook Hospital, compared CPR to driving an outdated automobile.Credit...Uli Seit for The New York Times

Since the 1960s, cardiopulmonary resuscitation — chest compressions, ventilation, defibrillation and epinephrine — has been the go-to approach to reanimate a stopped heart. But in recent years, experts in critical care have developed an arsenal of modern treatment refinements.

These improvements remain underused in practice, however. As a result, survival rates after cardiac arrest vary drastically across the country.

In the United States, there are roughly 350,000 resuscitation attempts outside hospitals each year, with average survival rates of 5 to 10 percent, and 750,000 attempts in hospitals, with about a 20 percent survival rate.

But exceptions exist in certain areas. In Seattle and King County, survival rates for cardiac arrests treated by emergency medical services providers improved by 22 percent over the last seven years to 19.9 percent, according to Dr. Graham Nichol, professor of medicine at the University of Washington. In places like Detroit, the survival rate is about 3 percent.

These statistics show “a lot more variation in survival than we see for stroke and heart attack,” Dr. Nichol said. “Cardiac arrest is indeed a treatable condition. Providing care, and providing better care, is important.”

Seattle and King County’s improvements come from training E.M.S. providers better, continually measuring the care they provide, and spreading awareness that cardiac arrest is a treatable condition, so citizens are quick to perform bystander CPR, Dr. Nichol said.

If Seattle’s innovations could be implemented nationwide for out-of-hospital cardiac arrests, he added, as many as 30,000 lives annually could be saved.

The problem with cardiac arrest treatment begins with the administration of CPR. It requires endurance and training to perform 100 to 120 chest compressions a minute, each at a depth of about five centimeters. Resuscitators often interrupt compressions for too long in order to check for a patient’s pulse, starving internal organs of oxygen, studies have found.

“You should only be feeling for a pulse for 10 seconds, while people fumble around trying to feel a pulse for one minute or more,” said Dr. Stephan Mayer, the director of neurocritical care at the Mount Sinai Health System.

Doctors and nurses also tend to give up too soon. CPR is typically performed for 15 to 20 minutes, but research shows that longer attempts at CPR, up to one hour, can lead to survival. These patients ultimately may fare as well those who are resuscitated more quickly. In patients with a chance of recovery, experts now advise attempting CPR for at least 45 minutes.

If no pulse returns after 20 minutes, however, experts say more powerful interventions should be considered. Often they are not.

“Doing CPR is like today driving a Model T Ford that itself isn’t even being operated properly much of the time,” said Dr. Sam Parnia, the director of resuscitation research at Stony Brook Hospital. “When it struggles to go uphill, we should switch to a more modern car — say a Ferrari with a powerful engine.”

One alternative to CPR is extra corporeal membrane oxygenation (ECMO), a procedure in which blood is drawn from a patient in cardiac arrest through a catheter placed in a central vein, circulated through an oxygenated filter, and then returned to the body carrying oxygen.

ECMO is more widely used in countries like Japan and South Korea than in the United States. “They routinely bring people back to life who would remain dead here,” Dr. Parnia said.

Once circulation is restored, a chain of interventions must occur to prevent further injury to the body and brain. But there’s no guarantee that patients will receive these treatments, which include avoiding toxic amounts of oxygen, maintaining normal carbon dioxide levels and high blood pressure, and sometimes a cardiac catheterization procedure.

“It’s a lottery of what you will get in the hospital,” Dr. Parnia said. “It may depend on which doctor happens to receive you, since none of these treatments are regulated.”

Among the most crucial procedures is therapeutic hypothermia. Patients who remain comatose after being in cardiac arrest should be cooled for at least 24 hours to a temperature from 89.6 to 96.8 degrees Fahrenheit, to slow down the metabolic processes that cause cells to die.

But this treatment is far more common in Europe than in the United States, said Dr. Romergryko Geocadin, professor of neurology at Johns Hopkins University School of Medicine. And even when hypothermia is used in conjunction with other life-support measures, physicians and relatives sometimes decide to give up on the patient in the first three days, before the prognosis is truly clear.

“A lot of progress has been made in the ICU, with hypothermia, more ventilators, more ways of supporting blood pressure and glucose,” Dr. Geocadin said. “So we should give these people more time to survive.”

A major cause of the variation in outcomes, experts say, is the lack of systematic benchmarking of cardiac arrest data, which would allow hospitals to see where they rank against others and motivate better performance.

“Right now, even at Mount Sinai, where I am the chairman of the CPR committee, we know what our success rates are for resuscitation but we have no idea if it is above average, average or below average,” Dr. Mayer said. “Everyone’s performance is looked at in isolation.”

This lack of accountability and transparency prompted the Institute of Medicine to call in June for the creation of a national registry to track the incidence and outcome of cardiac arrests.