Patients who have a cardiac arrest at the University of Pittsburgh Medical Center don't have to wait long for help to arrive.
“We've put a stopwatch on our response time. From the time of the call to the time that the team is there and deployed is about 90 seconds,” said Michael A. DeVita, FACP, a professor of critical care and internal medicine at Pitt.
Unfortunately, not all hospitalized patients in the U.S. get such speedy responses to their arrests, according to a study published earlier this year in the New England Journal of Medicine. Research by Paul S. Chan, MD, and colleagues found that 30% of 6,789 hospitalized patients with ventricular fibrillation or ventricular tachycardia did not receive defibrillation within two minutes of their arrest, and 11% didn't get a shock until after six minutes.
The study's results have served as a wake-up call for many hospitals. “Up until recently, people have always assumed that if a patient had a cardiac arrest in the hospital, they would be in the best place for that,” said Dr. Chan, a cardiologist at Saint Luke's Health System in Kansas City, Mo.
Now cardiologists and other experts are looking at what changes can be made—in procedure, training and technology—to shorten arrest response times and thereby improve patient outcomes. And they say that hospitalists are in the perfect spot to take charge of the problem and lead improvement efforts.
“Hospitalists are the obvious naturals for that position and they haven't yet come to own the issue of resuscitation, partly because cardiology, critical care, emergency medicine and anesthesia have all sort of been there. The truth is that nobody owns it and that's why we're seeing the poor quality,” said Dana Edelson, MD, a hospitalist at the University of Chicago.
Time means survival
One of the biggest quality issues raised by the Chan study was the clear association between delayed response times and worse outcomes. American Heart Association guidelines had already recommended that patients be defibrillated within two minutes, but evidence was thin on how much of a difference response time really makes, experts said.
“People have talked about the ideal cutoff being somewhere in the two- to three-minute range, but the actual data hasn't been there to have in-hospital standards,” said Dr. Edelson. “Before the recent New England Journal publications, our code committee had always aimed for three minutes, based on previous data.”
That seemingly miniscule time difference could have a dramatic impact on patients' survival and disability, the study found. For patients who were defibrillated in two minutes or less, survival to discharge was 39%, compared with 22% for those who got delayed shocks. The research also found a graded association between longer time to defibrillation and lower survival rates. For example, patients who were defibrillated at four minutes were about half as likely to survive as those who were shocked in a minute or less.
The researchers also uncovered several factors which made patients less likely to get timely defibrillation, including arresting after hours, being at a small hospital and having a noncardiac admitting diagnosis. “The predictors that we cited are tricky because none of them are things that you can change,” Dr. Chan said.
But in some cases, the root causes of the delay are clear and possible to repair, according to Leslie A. Saxon, MD, chief of cardiology at the University of Southern California (USC) University Hospital who wrote an editorial accompanying the Chan study. In particular, she favors more widespread monitoring of patients, including those with noncardiac diagnoses, to reduce defibrillation delays.
“It makes sense that no matter where you are, if you're not monitored and you have a sudden arrhythmia, who's going to know?” she said. At USC, administrators currently are looking into new automated and wireless monitoring technology that could cover more patients, even those in transit within the hospital. “You could create an environment in the hospital where everyone was monitored,” Dr. Saxon said.
Dr. Chan was not so convinced of the merits of universal monitoring. “It's not clear that it's pragmatic, because it's expensive. That's why we don't do it on all patients to begin with,” he said.
There are newer, less expensive monitoring options, but they have yet to be thoroughly studied, according to study co-author Graham Nichol, MD, a professor of medicine at the University of Washington in Seattle. “We need to evaluate how cost-effective monitoring technology is, but I also think we need to assess whether some of the less expensive monitoring technology, which has recently become available, works,” he said.
Responding faster and better
Although Dr. DeVita agrees with universal monitoring, if feasible, it only solves part of the problem behind delayed defibrillation, he said. “Finding the patient is only the first step. There are now a number of studies to show that even if the event is detected, something like 40% of the time people don't do the appropriate next step to action.”
The experts had a variety of ideas for improving that statistic, ranging from the cheap and easy to the more costly and complex. One improvement technique suggested by several experts was simply to give residents and other staff involved in a code more practice with the procedure.
“Most hospitals don't run mock codes. Running them on a routine basis would be beneficial,” Dr. Chan said.
Dr. Edelson agreed, suggesting that hospitals consider adopting the aviation industry's model of more simulation-based training to prepare for high-stress emergencies. “Just because people have taken a class in resuscitation doesn't mean that they can do it in actual practice. It's such a rare event, even in the hospital, that chaos and pandemonium ensue most of the time,” she said.
Training opportunities also arise after, as well as before, arrests. Dr. Edelson and a colleague, Benjamin S. Abella, MD, have studied debriefing as a technique for improving the quality of in-hospital resuscitations. In a study published in the Archives of Internal Medicine in May, they found that having staff routinely meet and discuss the successes and failures of resuscitation efforts after the events improved quality and some outcomes.
The same strategy could be applied to defibrillation delays, said Dr. Abella, who is an assistant professor of emergency medicine at the University of Pennsylvania in Philadelphia. “It would not be unreasonable to have a structure whereby every cardiac arrest that occurs in the hospital would have a debriefing that would include how long did it take the team to respond, was CPR adequate, was resuscitation equipment easily available. Just the act of debriefing, knowing that you're going to be evaluated, often improves care. A lot of system problems often become uncovered,” he said.
One system problem common among hospitals relates to the ease of use of defibrillators, the experts said. “Hospitals really need to focus on the uniformity of their equipment. There's no reason why hospitals should have five different kinds of defibrillators. When you need to figure out a device right away, and you don't use it that often, it's important that everybody be familiar with it,” said Dr. Abella.
A. Maziar Zafari, MD, PhD, a cardiologist at the Atlanta Veterans Administration Medical Center, didn't realize that lack of uniformity was a problem until he began a project to improve cardiac arrest survival at his midsize hospital. “We found eight different types of defibrillators from four or five different companies,” he said.
One solution, which Dr. Edelson's hospital adopted, is to standardize defibrillators. “We're using only one type of defibrillator across the institution,” she said.
Apply outpatient research
Even if all the physicians in the hospital are fully trained on the same defibrillator, there remains the issue of getting the doctor across the hospital, or up the elevator, to the patient in time, the experts noted. “Maybe to get to this two-minute barrier, we can't wait for anybody to respond and we have to allow local responders to defibrillate,” said Dr. Abella.
At the University of Pittsburgh, local response is the key to their speed. “If you want this to happen in under two minutes, basically you have to depend on the unit team to do the first defibrillation before the code team ever gets there,” said Dr. DeVita.
To achieve local response, nurses could be trained to use manual defibrillators and given authority to shock patients without a physician present. But because that isn't standard procedure at most hospitals, there are little data on how it works, Dr. Chan said. “We don't really know whether or not allowing nurses to shock would improve times.”
The other option, which seems to be rapidly gaining popularity around the country, is to make hospitals more like shopping malls and movie theaters. Applying lessons learned from research on automatic external defibrillators in outpatient cardiac arrests, some hospitals are bringing AEDs to the inpatient setting.
“If malls have solved this problem by putting AEDs all over their building, maybe we should learn from them and not think that hospitals are somehow special,” Dr. Abella said.
That's the course that Dr. Zafari and his team chose in his VA quality improvement study in 2000. After all manual monophasic defibrillators were replaced with AEDs and manual defibrillators with AED capability, survival to discharge rates from resuscitations went up from 4.9% to 12.6%. The statistics for patients presenting with ventricular tachycardia or ventricular fibrillation were even more impressive, with a 14-fold increase in survival to discharge. The study was published in the Journal of the American College of Cardiology in 2004.
However, the process was not as simple as just switching out all the defibrillators, Dr. Zafari said. “It was really a culture shock to tell our entire staff and nurses, ‘You are as good as any MD. You can identify the need for CPR.’ That meant re-educating the whole hospital staff in becoming a first provider.” In September 2001, the Veterans Health Administration issued a directive for AEDs to be used in appropriate settings in all its hospitals and clinics.
AEDs may be particularly effective in fixing one of the secondary problems uncovered by the Chan study—that smaller hospitals, those with 250 or fewer beds, had slower response times to cardiac arrest. “A clear potential delay may occur in a smaller hospital in the middle of the night when the only doctor may be in the emergency room,” said Dr. Chan.
Make it work anywhere
Of course, cost is a consideration in the implementation of AEDs, which run from about $1,000 each to several thousand. Budget constraints may already be affecting hospitals' performance in cardiac arrests, the experts noted. The study found that black patients were defibrillated less quickly than whites. Dr. Chan and the other authors are further investigating that finding now, but they suspect that it may be explained by black patients being treated at poorer, lower-performing hospitals.
But any hospital should be able to apply at least some of the improvement strategies, the experts said. “I don't buy that this is only a big hospital capability. I think that hospitals that have worked hard to figure this out have been successful,” said Dr. DeVita.
In other words, hospitals and hospitalists concerned about defibrillation delays have their work cut out for them.
In addition to their general recommendations for improving response times, the experts suggested that hospitalists get involved in code committees and push for synchronization of hospital clocks, a necessary precursor to accurately assessing response time to arrests. (Average defibrillation delays may actually be even longer than those reported in the research because of inconsistencies between clocks, several experts noted.)
With a combination of updated procedures, improved training and new technology, hospitals should be able to get their response times at least within sight of the recommended window and have more patients leave the hospital after a cardiac arrest, the experts concluded.
“I think two minutes is a tall order for most hospitals,” said Dr. Abella. “However, through a multi-pronged approach, we might be able to achieve that goal. It's not going to be easy.”