The Case for CPR Feedback Devices

When William Kowenhouven, Guy Knickerbocker and Peter Safar first identified an effective method of performing CPR in the 1960s, it promised to revolutionize emergency medicine. In 1994, a study on bystander CPR demonstrated this life-saving technique didn’t have to be limited to professionals–that anyone could step in to help–a concept that could, in theory, increase survival rates significantly. Soon after, the AED promised to continue the revolution: The sooner you defibrillate a patient in v fib, the better the chance for them to survive. Some even predicted AEDs would render CPR obsolete.

And yet, after so many years, after so many innovative leaps forward, cardiac arrest survival rates remain low in most communities.

To be sure, part of the reason is that the window of opportunity to save a patient in cardiac arrest is narrow, and requires a confluence of factors. But John Freese, MD, director of prehospital research for FDNY, believes there’s still room for improvement, and that the key lies in improving our CPR technique. And the key to that lies in CPR feedback devices.

There’s a lot that we still need to understand and study when it comes to delivering a successful resuscitation, but effective CPR forms the basis for our evolving definition. “We’re prescribing therapies we don’t know enough about to a disease process that we know even less about, to human beings that we may not know at all,” Freese says, paraphrasing Voltaire. “There’s a lot of work to be done, but CPR feedback seems to be our first step.”

The Failed Promise of Defibrillation
To understand why CPR feedback devices are needed, we first need to understand why AEDs failed to live up to their promise. After AEDs became available, defibrillation was one of the reasons behind the merger of the FDNY and New York City EMS–the thought was that it would reduce response times and double the number of defibrillators on the streets. Did it work? “Response time was reduced, from 9.3 minutes to 4.7 minutes,” Freese says. “That, combined with essentially doubling the number of defibrillators on the street, should have meant a dramatic increase in cardiac arrest survival. And yet, when you looked at it, the city saw only a small increase in cardiac arrest survival–2.2% to 2.9%.” And when you dig deeper into the statistics, you can see that when defined as return of spontaneous circulation (ROSC), survival actually decreased in the short term, from about 30% to 20%.

Importantly, Freese notes, this trend wasn’t unique to New York City. “Seattle seemed to experience a similar decrease in survival, from around 35% to 20%,” he says. “The problem seems to be the technological imperative–if you only give people a hammer, and endorse the amazing properties of the hammer, the world is going to be a nail. So what probably happened is that we forgot the other key components.”

Fewer Links in the Chain
For patients in v fib, all five links in the American Heart Association (AHA) chain of survival–rapid activation of EMS, early CPR, rapid defibrillation, effective ACLS and integrated post-arrest care–may apply. But Freese points out that for many patients, some of these links are deemphasized. “When you have short EMS response times, bystander CPR doesn’t improve survival, because EMS is there so quickly,” he says. “So you’re down to four links. But there’s a bigger problem, too. We’re seeing less and less v fib. In New York City in the 1990s, one in three patients presented in ventricular fibrillation. In 2012, it was one in eight.”

What does this have to do with CPR feedback? If we’re already providing early CPR, and if defibrillation won’t be effective, we’re now down to three links in the chain of survival where providers can impact survival rates. “For the majority of our patients, the chain of survival is just three links–someone calling us, us getting there and providing effective CPR, and getting them to appropriate post-resuscitation care,” Freese says. “And so for EMS, our link is CPR, regardless of their cause of arrest, regardless of the presenting rhythm.”

That means we better be extremely effective at it.

What Is Quality CPR?
Delivering high-quality CPR lies in understanding what high-quality CPR looks like. By now, most of us are familiar with the six key factors:

  1. Compression depth of at least two inches.
  2. Compression rate of at least 100 compressions per minute.
  3. Allowing the chest wall to recoil fully between every compression.
  4. Spending about 50% of the “duty cycle” of CPR actively compressing.
  5. Limiting interruptions in CPR.
  6. Ventilations of less than 10 per minute.

Unfortunately, knowing we need to perform CPR in this manner and actually doing it are two different things. “A JAMA study showed that the average compression rate is great, but with no feedback, the average patient ends up receiving just 18 effective compressions per minute,” Freese says, due to some not being deep enough and due to pauses in CPR. “That may help to explain the overall 3% survival rate to discharge for these patients.”

Fortunately, using CPR feedback devices has shown promise in improving CPR in real time. In 2009, FDNY purchased new monitors for the entire department, with the requirement that they be able to provide real-time audio and visual feedback.

As with any new technology, there are concerns that must be addressed. “Providers would say, you expect me to use a device that will say out loud, “˜Press deeper, press faster?’ This will put me at risk,” Freese says. “There were legal concerns, concerns about peer perception. It put pressure on our providers that we certainly didn’t intend.” Still, following lengthy discussions, the department’s ultimate decision was that visual and audio feedback was important. After training, the devices were deployed in 2011. Freese believes they are already having an impact.

One note of caution: It’s not enough to simply accept the data CPR feedback devices provide. You must dig into the data for the root causes behind what you’re seeing. If there’s a period where no compressions are being done, is that because the providers were intubating the patient or because the patient had a pulse? If the compressions were consistently too shallow, it could be provider performance, but there could be a patient physical attribute at play.

“If you just look at the data, it’s easy to target the providers who aren’t meeting standards,” Freese says. “But a “˜bad resuscitation’ on paper can actually be a good resuscitation.”

Initial Results
Because CPR feedback in real time is so effective, “most of the time, you’ll see great CPR,” Freese says. “But the question is, does that kind of data make a difference for our patients?”

In 2010, the AHA guidelines designated CPR feedback as effective in training, but noted that there was insufficient evidence to conclude that it had an impact on patient outcomes. In 2011, Freese notes, a study in the British Medical Journal designed to look at ROSC found that as compared to CPR with no feedback, visual and audio feedback provided no change in survival rates. “That study was published in 2011, but it was done prior to the release of the 2010 AHA guidelines, so the compressions were probably too slow and almost certainly too shallow,” Freese says.

So, the FDNY analyzed their data to get a better picture. And the results are promising. “We found that ROSC went from 38.5% in early 2011 when we first introduced the technology to 52.5% in June 2013,” Freese says. His conclusion: CPR feedback does improve at least short-term outcomes. He notes another recent study found that survival to hospital discharge for patients with witnessed ventricular fibrillation arrests was doubled through the use of audiovisual feedback, from 26% to more than 50%.

There’s clearly more research needed before we can truly understand the impact of real-time feedback on CPR quality. In the meantime, Freese argues that it’s our duty to embrace the technology. “Change is difficult to most of us in EMS,” he says. “Think about having to wear florescent vests, think about capnography. Think about when gloves weren’t an essential part of patient care. I think CPR feedback will take the same path as these things. Technology has to be a part of what we do. We have to be willing to use it and accept it because if we don’t, we’ll fall behind the times.” jems

Editor’s note: This article is adapted from the JEMS webcast, “The Weakest Link: Using CPR Feedback to Improve Quality.” View the webcast at

Previous articleFormer Surgeon General Inspires AAA Conference
Next articleThe Argument for BLS CPAP

No posts to display