Subscribe to eTOC

Worse Outcomes for Cardiac Arrest in Hospitals, Increased Risk for Brain Damage


  • ✓ An analysis of data in a national CPR registry showed that slightly more than 30 percent of 2,045 patients waited two minutes or more for resuscitation after cardiac arrest, resulting in significantly worse functional and neurological recovery.

Patients who have a cardiac arrest while in the hospital often wait longer for resuscitation and have worse outcomes than people who suffer a cardiac arrest elsewhere, according to a study published in the Jan. 8 New England Journal of Medicine (NEJM).



In an effort to understand the cause of this paradoxical discrepancy, Paul S. Chan, MD, a cardiologist with Saint Luke's Mid-America Heart Institute in Kansas City, MO, and colleagues reviewed survival data on nearly 6,789 patients at 369 hospitals who had cardiac arrests caused by ventricular tachycardia or ventricular fibrillation.

Between 370,000 and 750,000 cardiac arrests occur in U.S. hospitals every year, and there is an approximately 34 percent rate of survival to discharge, according to an analysis of data in the American Heart Association National Registry of Cardiopulmonary Resuscitation. The findings were reported in a 2003 study in Resuscitation, the international journal of the European Resuscitation Council.

About 225,000 cardiac arrests due to ventricular tachycardia or ventricular fibrillation occur in an airport or other public place where an automated external defibrillator is readily available, and the rate of survival there is more than 50 percent, according to a 2002 NEJM study.


In the current study, the hospitals, which all participate in the national CPR registry, achieved an impressive median time to defibrillation of one minute. However, 2,045 patients — slightly more than 30 percent — waited two minutes or more, resulting in significantly worse functional and neurological recovery. Only 22.2 percent of them survived to discharge, compared with 39.3 percent of those who were defibrillated in less than two minutes.

The investigators collected data on cognitive status only for those who survived to hospital discharge, and assessed patients using the Glasgow Coma, a series of tests that measure response — eye, motor, and verbal — after a traumatic brain injury. The scores are combined to indicate levels of disability: a score of 13 or greater indicates no major disability; 9 to 12 is moderate disability; 8 or less, severe disability. And coma is defined as not opening eyes, not obeying commands, and not uttering understandable words.

Of people who survived to discharge, more than 60 percent of those resuscitated in less than two minutes had no apparent neurological disability, 28.2 percent had moderate disability, nearly 10 percent suffered severe disability, and nearly two percent were in a coma or vegetative state.

In contrast, among those who waited two minutes or more for resuscitation, 51.7 percent had no apparent neurological disability, 35.2 percent had moderate disability, 9.4 percent had severe disability, and 3.7 were left in a coma or vegetative state. The authors noted that data on neurological and functional status was missing for 16 percent of patients surviving to hospital discharge.

To compensate for the fact that many in-hospital cardiac arrests occur in people with existing heart problems, the study excluded more than 6,600 patients who had had more than one cardiac arrest, who had an implantable cardioverter-defibrillator, or who were being given intravenous anti-arrhythmia drugs. They also excluded patients who had had a cardiac arrest in an emergency room or an operating room, or were having cardiac catheterization.

Despite this, the remaining people who had a cardiac arrest while in the hospital and waited two minutes or more for defibrillation were nearly half as likely to survive to discharge, and the problem may be even worse than the study indicated, according to Dr. Chan.

“It's actually amazing how well we do,” said Dr. Chan. “I was surprised how few codes had delays. But that may be because the hospitals participating in this registry were high functioning — they were already doing quality improvement and were getting quarterly feedback from the registry about performance compared to other hospitals in the registry. So our finding that 30 percent of patients experienced delays may underestimate the true problem in all hospitals in the United States.”

Although the study issued no recommendations, Dr. Chan believes that rapid response teams may speed resuscitation after cardiac arrest.

“With rapid response teams you have the same people running all the codes and possibly intervening prior to the code,” he said. “It is possible that this would bring more consistent care.”

And since outcomes were poorer on evenings and weekends, especially in hospitals with fewer than 250 beds, the lack of trained medical personnel available for resuscitation may also be important factors, according to Dr. Chan. Making automatic external defibrillators widely available in hospitals also might help, he added.

“We see them used in airports and public spaces,” he said. “The technology is available for rhythms to be automatically read without the need for a whole battalion of people to do that.”


Monitoring all patients would help identify cardiac arrhythmias immediately, thereby speeding resuscitation, but that would be costly with existing technology. “It would be very expensive to have every patient in every hospital in America on a cardiac monitor 24/7,” said Wade Smith, MD, director of the University of California-San Francisco Neurovascular Service and an endowed chair of neurocritical care and stroke. “But this study shows that intermittent monitoring of people outside of the ICU can be problematic.”


Dr. Wade Smith: “It would be very expensive to have every patient in every hospital in America on a cardiac monitor 24/7. But this study shows that intermittent monitoring of people outside of the ICU can be problematic.”

The study reveals that having more personnel is associated with better outcomes, but that too is costly.

“In larger hospitals there are more trained professionals, and when you get to centers of excellence, which tend to have 500 or more beds, you start seeing shorter response times,” said Dr. Smith. “This supports the notion that your chance of having a good outcome is better in larger hospitals. It's a little frightening that you could be in a hospital in America and have a cardiac arrest, and the time it takes to resuscitate you is a function of how many doctors are in the hospital.”

One unexpected finding in the study involved the outcomes of African-American patients, who tended to have poorer outcomes after a cardiac arrest than other patients.

“I think that was one of the most interesting findings,” said William Coplin, MD, associate professor of neurology at Wayne State University School of Medicine in Detroit, where he is director of the Neurological Intensive Care Unit. “I don't think that is a function of doctors or nurses discriminating against black patients. There could be comorbidities such as high blood pressure. Others have shown that patients with lower socioeconomic status are less likely to have follow-ups. Or it may have something to do with the kind of hospital where they may be. However, large inner-city hospitals with large black populations tend to be teaching hospitals that are large and well staffed.”

In an editorial in the same issue of NEJM, Leslie A. Saxon, MD, professor of clinical medicine at the Keck School of Medicine of the University of Southern California, envisions a time when all high-risk hospital patients will wear telemetry electrodes that wirelessly transmit continuous ECG data to a centralized alarm station.

“The centralized computer would continuously analyze these data with the use of automated algorithms,” she writes. “If ventricular tachycardia or ventricular fibrillation were detected, an alarm would be activated outside the patient's room or at the nursing station.”

This would circumvent the problem of staffing shortages, Dr. Saxon said, and speed the process of defibrillation. But that time is not yet here.

“We don't really know how we can shorten times in the foreseeable future,” Dr. Chan said. “We're planning studies that would better describe hospital characteristics such as nurse-to-patient ratios, the presence of rapid response teams, and other factors that could be considered surrogates of quality at the hospital, but we are also looking for interventions that would make a difference.”


• Chan PS, Krumholz HM, Ballamothu BK, et al., for the American Heart Association National Registry. Delayed time to defibrillation after in-hospital cardiac arrest of cardiopulmonary resuscitation investigators. N Engl J Med 2008;358:9–17.
    • Peberdy MA, Kaye W, Lane-Trultt T, et al., for the NRCPR Investigators. Cardiopulmonary resuscitation of adults in the hospital: A report of 14,720 cardiac arrests from the National Registry of Cardiopulmonary Resuscitation. Resuscitation 2003;58:297–308.
      • Caffrey SL, Willoughby PJ, Becker LB, et al. Public use of automated external defibrillators. N Engl J Med 2002;347:1242–1247.