The battery-operated device -- known as the Zoll AutoPulse Non-invasive Cardiac Support Pump -- consists of a board that patients lie on and a band that tightens across their chest to give compressions with the push of a button.
  
"When you are giving chest compressions to someone in cardiac arrest, the most important thing is that you do it fast and hard, and it needs to be consistent each and every time," said Nicole Seraphine, a nurse manager who was instrumental in bringing the device to Baptist Hospital East. AutoPulse "gives continuous compressions until you need to stop it for whatever reason."

Conflicting research on the efficacy of the federally approved device appeared in the June 14 issue of the Journal of the American Medical Association. The two studies involved the use of AutoPulse by emergency medical personnel outside of hospitals.

A Virginia study, involving about 700 patients, showed AutoPulse outperforming manual CPR. But a higher-quality study known as the ASPIRE trial that involved more than 1,000 patients was halted early because it was associated with lower survival than manual CPR.

The results of the ASPIRE trial were a surprise to Dr. Michael Sayre, an investigator at one of the trial's five sites.

"Certainly, when we started on this study, we didn't dream that it might actually come out that the device was worse" than manual CPR, Sayre said. But he stopped short of saying that people should discontinue use of the device, noting that the issue isn't that clear-cut.

Since some research findings have been positive, that "raises the possibility, at least, that the AutoPulse is better than not-very-good CPR, but maybe it's not as good as really good manual CPR, so perhaps it's somewhere in the middle, or there could be other effects that we just don't really understand about how the human body responds to CPR," said Sayre, an emergency-medicine physician at Ohio State University Medical Center in Columbus.

Zoll Medical Corp., which produces the device, considers the ASPIRE trial's results to be inconclusive and is planning another trial, which starts soon, said Ward Hamilton, vice president of marketing.

"The researchers at Zoll would tell you that the data, although it was troubling to the researchers, is not the final data on the AutoPulse," Hamilton said. "The predominance of information from various research on this device is all very favorable to the device," which Zoll maintains is safe and "life-saving."

Baptist Hospital East has been using the AutoPulse for more than a year, and "we certainly have found it very valuable," said Cheryl Stout, the hospital's director of critical care. "We have no documented problems at all or no issues with it."

Regarding the trial that had to be halted, she said, "I can't say it's not concerning; I'm just saying that as we do with all codes, we'll continue to monitor it and look at outcomes."

In training
Clark Memorial Hospital in Jeffersonville, Ind., was in the midst of training selected personnel how to use AutoPulse last week. It will be at least a few weeks before the hospital officially begins using the device, said Clark spokeswoman Mary Jennings, but a specific date was unavailable.

"I think it will be very beneficial" in cardiac-arrest situations, said Tiffany Sierota, a cardiopulmonary administrator at Clark.

Without AutoPulse, giving compressions deep enough and fast enough "is something that doesn't typically happen a lot in practice," Sierota said. "This device allows you to be able to give compressions at the appropriate rate, the appropriate depth, and the goal would be then to improve your outcomes to survival from the hospital."

Sierota said Clark would be monitoring how the device performs at the hospital during a three-month critical-care trial before deciding whether to expand its use.

She also said Clark would look into the ASPIRE trial and compare it to "what we see here at the hospital."

"Every code that's attended, we monitor it, and we go back and sort of evaluate it and see if there was anything that should have been done differently," she said.

Sierota also noted that in an actual emergency, if AutoPulse didn't appear to be helping a patient, hospital personnel "would just immediately start manual compressions."

Limited numbers
Fewer than 10 entities in Kentucky and Indiana have acquired the device, which costs about $14,000, according to Zoll. They include hospitals and fire departments.

Louisville Metro Emergency Medical Services doesn't use AutoPulse for a couple of reasons, said Dr. Neal Richmond, chief executive officer of EMS.

The device would be a "pretty pricey" investment since "we're running up to 32 response units at a time in the city," he said, and "the evidence is not 100 percent convincing about what's right."

Baptist has two AutoPulse devices -- one in the intensive-care unit that can be taken throughout the hospital and one in the emergency department.

Seraphine said she has seen the difference the device can make in cardiac-arrest patients.

"We've seen them go from completely cyanotic, which is blue, to pinking up within seconds," she said.

The device also can cut down on the number of people needed to respond to a cardiac arrest or "code," Seraphine said.

"We've run a code where it's just been two nurses and a doctor, and you're able to do everything you need to do because compressions are going, you're administering medications out of the crash cart, the doctor is intubating," she said. "It's just a very smooth process."

The goal is to bring back people who are essentially dead, Hamilton said.

Eliminates fatigue factor
Because AutoPulse uses a band instead of a single pressure point on the chest, there is "much more blood flowing when the AutoPulse is used versus the kind of blood flow that can occur when you're using your hands," Hamilton said. "So the two main advantages are it moves a lot of blood -- almost as much as normal circulation -- and it can do it over a long period of time, very consistently."

Without AutoPulse, "rescuers get fatigued," he said. "They also have some tendency, perhaps, to interrupt CPR too much, and when you have a device, which is mechanical, as the AutoPulse is, it overcomes some of those things quite well."

Seraphine said tiredness sets in "very easily and quickly" while someone is doing manual CPR, necessitating that someone else take over chest compressions.

"Every time you pause and you don't give compressions, that's when you're losing time because the blood flow is actually stopped," she said. "If you keep pausing, you're less likely to bring this individual back to a sustainable rhythm."

Even with Zoll's device, Sayre said "every system has to look at what they're doing carefully, and be certain that if they are choosing to use the AutoPulse that they're using it in a way that is really minimizing the amount of time that the patient goes without any compressions."

He added, "The way we used it in our study apparently did not accomplish that goal, and so we are not using it anymore in any of the study sites."

When it's not used
Stout said AutoPulse isn't used at Baptist if personnel think that it might cause a delay in someone getting chest compressions.

AutoPulse also isn't used on certain types of patients, such as those who weigh more than 300 pounds, those who are pregnant, or those who've gone into cardiac arrest because of blood loss from trauma, such as a shooting or car accident, Hamilton said.

It's unclear exactly how many of the devices are in use, but Zoll said it has more than 450 AutoPulse customers worldwide, mostly in the United States.

In the ASPIRE trial, the device was tested in an ambulance setting in five cities. In the AutoPulse group, 6 percent survived to discharge from the hospital, compared with 10 percent for the group of patients who got manual CPR.

But a study involving emergency medical personnel in Virginia showed "a very substantial improvement in long-term survival with the device being used by the Richmond EMS," Hamilton said

The company hopes its next study will answer questions raised by the ASPIRE trial.

Learning curve
After reading an article on the trial last week, Stout said the study might have had some limitations. She wondered whether AutoPulse might have delayed the start of CPR in some cases because some users might not have been that familiar with the device.

Sayre acknowledged that there are differences between using the device inside a hospital and outside the hospital.

For instance, "EMTs (emergency medical technicians) only use the device once or twice a year, and naturally that means they don't have a whole lot of chance to practice in using it with real patients," he said. "In the hospital, it's possible that the team using it might use it much more often and therefore get a whole lot better."

Authorities in Riverside County, Calif., suspended use of AutoPulse last year after an elderly man treated with the device received cracked ribs and later died.

Zoll said last week that it doesn't believe there is any stronger likelihood of injury with its device than with manual CPR, which can lead to rib fractures and other injuries.

The company is not "aware of any instances to date where the AutoPulse alone was the cause of CPR-related complications; however, it is not out of the realm of possibility to expect that this might occur at some time," communications manager Robert Minicucci said in a written reply.