Cardiac arrest is a human problem, not a training problem

Listen and followApple PodcastsSpotifyYouTubeMore apps · Browse all episodes

When a patient’s heart stops, the person leading the code is running on memory alone, and memory fails under that kind of pressure. Michael Peck is an anesthesiologist and retired faculty member at George Washington University who spent more than 30 years watching skilled clinicians get cognitively overwhelmed during a cardiac arrest. In this episode he explains why the problem is not a lack of training but the hard limits of human attention, memory, and decision making during a crisis. This conversation is based on his article “Cognitive overload in cardiac arrest is a human problem,” published on KevinMD. You will hear what it actually feels like to run a code with no reference to lean on, why aviation and nuclear power lean on checklists while medicine still does not, and why hospital reviews built on memory may miss what went right as well as what went wrong. Press play to hear why he believes patient safety starts with admitting our limits.

Partner with me on the KevinMD platform. With over three million monthly readers and half a million social media followers, I give you direct access to the doctors and patients who matter most. Whether you need a sponsored article, email campaign, video interview, or a spot right here on the podcast, I offer the trusted space your brand deserves to be heard. Let’s work together to tell your story.

PARTNER WITH KEVINMD → https://kevinmd.com/influencer

SUBSCRIBE TO THE PODCAST → https://www.kevinmd.com/podcast

RECOMMENDED BY KEVINMD → https://www.kevinmd.com/recommended

Transcript

Kevin Pho: Hi, and welcome to the show. Subscribe at KevinMD.com/podcast. Today, we welcome Michael Peck. He’s an anesthesiologist, and today’s KevinMD article is “Cognitive overload in cardiac arrest is a human problem.” Michael, welcome to the show.

Michael Peck: Thank you so much, Kevin, I appreciate it. This is really quite exciting for me.

I’ll give you a little background. As you said, I’m an anesthesiologist. I have recently retired from hands-on work with patients, and I have gotten a little more into my creative side, so to speak. Let me tell you how I got there, from my years in residency all the way to becoming a faculty member, and then moving forward.

I actually became an anesthesiologist almost by accident. What I mean is that I started out in a general surgery residency. It was a fantastic experience. It was also ridiculously grueling. By my third year, my wife and I had our first child. Between residency and a newborn at home, there was not much time to sleep or to think clearly about the future.

Then fate intervened. I was in a pyramid program, which always made me a little nervous, and I was let go. Suddenly I had a newborn, no residency, and a very real question staring me in the face: what was I going to do now? Someone suggested anesthesiology to me. My first reaction was, “No way. I’m a surgeon.”

But eventually I thought, “What do I have to lose?” I started over again in a whole new residency, and suddenly I was immersed in a completely different way of thinking about patients. Instead of operating on the anatomy, I was now managing physiology in real time. That is a very big difference. There was a steep learning curve, but once I caught on, I found that I actually loved it.

It was fantastic. What surprised me was that I had not left problem-solving behind. I had simply found a different way to solve a problem. I did a fellowship in neuroanesthesiology that led me to George Washington University, where I was an assistant professor, and teaching became one of the most rewarding parts of my entire career.

So looking back, it is kind of remarkable. I became a father and was forced out of surgery, and that shaped everything that followed, including eventually filing my first patent and developing a perioperative anesthesia software program. I did not know it at the time, but that was the beginning of the things I am doing now.

Kevin Pho: Excellent, and I love hearing stories like that, because one of the things I like to share on this podcast and on my site is that the paths of a lot of physicians are non-linear. Sometimes they do not go as expected, and of course, talking to you, we are going to be talking about those tangents away from what most physicians do clinically.

Your KevinMD article is titled “Cognitive overload in cardiac arrest is a human problem.” For those who get a chance to read it, tell us what it is about.

Michael Peck: Well, the article came from years of observing cardiac arrests and what they actually look and feel like in real life. Even if you are highly trained, you can become cognitively overloaded during a code.

You have multiple medications, timers, rhythm changes, communication demands, and documentation, all happening at the same time. We often talk about errors as individual failings, but many are really the predictable consequences of human cognitive limits under stress.

Part of my interest came from personal experience years ago, when I was leading a cardiac arrest. I was relying on algorithms I had learned during a recertification course that might have been a week before, or it could have been a year before. When the code was over, I remember feeling completely mentally exhausted and asking myself a simple question. There had to be a better way.

Not long afterward, while I was still contemplating all this, I walked into an operating room, stared at the physiologic monitor, and said, “Oh, I have a great idea. Why can’t the cardiac arrest algorithms be built directly into the monitor?” The AHA owned them, so it seemed to me that there was no liability on behalf of the monitor’s manufacturer. All they had to do was license it. I thought it was a great marketing ploy.

But anyway, getting back to checklists, we use checklists and decision support tools in aviation and nuclear power. Why should cardiac arrest management deviate from that and rely completely on our memories? So I approached that manufacturer thinking, “Oh, I have a great idea for these guys. They are going to love it,” and they said no.

Then you fast-forward. This was back in 1991 or 1992, when the technology was just not ready for what I wanted to do.

What surprised me while researching the article was that most hospitals do review cardiac arrests. The challenge is that those reviews rely on documentation and memory again, and memory is not always reliable after an extremely stressful event. So we may be missing opportunities to learn, not just from the mistakes, but also from the successes.

What was missed even when a patient did fine? What worked? What were the delays? Could the process have been better? So in many ways, the article reflects a question that followed me for more than 30 years: why do we expect clinicians to rely on memory? It just seemed ridiculous to me. So it was never about replacing clinical judgment, and the whole idea of the article was not even about building a product to fix that.

Judgment becomes even more important during a crisis. We have all been there. What do I have to do? Everybody is waiting for you to make a decision and tell them what to do, and you are trying to recall those algorithms you once memorized, or that you hope are still stuck in your head somewhere.

But we all have cognitive limits. So once we accept the reality that we have cognitive limits, and that we are not superhuman, we can begin designing systems that help people perform at their best when they are the most stressed.

Kevin Pho: Now you obviously have run many codes, and even I, in internal medicine when I was training, ran a few, although it has been a while. Certainly in primary care, we do not run as many ACLS codes as you have to. For those who are outside of medicine and may not know what it is like to run a code, take us into the middle of one and give us a sense of why it is so overwhelming. Even though we have these defined algorithms for what to do, in the midst of a code, in the midst of a crisis, why is it so overwhelming for physicians?

Michael Peck: Well, let me put it this way. For me, as an anesthesiologist, if you are on a regular floor, in the ER, or in the ICU, they have these code teams that all rush in. But in the operating room, this one code I was running might have been after midnight, and there is just a dearth of people. There is nobody around.

And here you are, the sole person in the room who has to take charge. So let’s say you ask for people to show up and no one is coming into the operating room. They do not know what to do. And by the time anybody shows up, you are already in the throes of it with the people who are in the operating room.

So you have the nursing staff, you have, let’s say, the attending surgeon, the resident surgeon, maybe an intern, maybe a medical student, and there is me and one of the anesthesia residents, whatever number of people that adds up to. And then you have to say, “OK, everybody, somebody has to start the chest compressions.”

So somebody has to take that on and manage it, and you have to make sure they are doing it properly, and that has also changed over time. We used to do maybe 80 to 100. Now we are doing multiples of that, rapidly, and that has changed. So then you have to look at the monitor.

What kind of rhythm are you dealing with? And then, based on that rhythm, you have to say, “OK, for this particular rhythm, of course we are going to do our chest compressions, and then we are going to start figuring out what is next. What medication might we need?” And there is the patient. Are they in v-fib and do they need to be shocked?

And you have to tell somebody, “Please get the defibrillator ready, let’s shock them,” and then you shock them and wait to see what happens. If they continue in the same rhythm, you keep continuing your CPR, and you shock them again. And then what happens? Either it gets better and they go into a normal rhythm, and you say, “Oh, thank God,” or you sit there and they go into a different rhythm.

Then you have to try to remember what that different rhythm calls for. All the while, people are doing things. You may need somebody to put in a better intravenous line if it is necessary. That could be anybody in the room who is skilled, as long as they are not already doing something else. And you actually have to tell somebody, “Go over to the corner.” Usually it is one of the nursing staff, or maybe even the medical student, and you say, “I need you to document everything that is happening here.”

“I need the time it is happening, when the code started, the drugs we used, and when we used them, as best as you can.” So we need to keep track of all of this. And then, if you need somebody to put in an IV, the patient is obviously in the operating room, whether it is during their surgery, at the end of it, or even when they first arrive at the OR. If they need a breathing tube, somebody has to put that in.

If they need a central line, somebody has to do that. So you are basically the person in control. You are telling everybody what to do, and you are following them and making sure the timing gets done, because these algorithms are separated by two minutes, five minutes, and it all depends on what drug they need to give next.

One thing I forgot to mention is that whenever there is a code, every operating room, every part of the hospital, has these things called code carts, and on those carts is the equipment. If you need to place an airway, or you need medications, or you need the pad to shock somebody, all of that is on the cart, so everything is in one place, which makes it really convenient, and we wheel one around all over the place.

So that might give you some idea of it. I sometimes call it controlled chaos, but as long as somebody is in control of that chaos, it runs better.

Kevin Pho: Now, for the codes that unfortunately fail, what are the most common reasons you have seen in your experience running codes?

Michael Peck: Well, if you think about it, in the ER and the ICU, people are coding for one reason or another. Did they come in with an initial cardiac issue and then arrest? Was it a trauma event? Have they lost a lot of blood, and now they are having a cardiac arrest? Did they have an issue in the field, because they came in by ambulance and there was trouble getting a breathing tube in, so suddenly their oxygen levels are really low and they are arresting from that? You have to pull all these different possibilities together and figure out what happened and why.

So part of what you do during a code is sit there and ask, “What happened? Where are they?” Are they acidotic? Are they hypovolemic, meaning is their blood level low, are their fluid levels low? And then you say, “OK, if that is what is happening, or what we think is happening,” and you run some labs, including what are called arterial blood gases. That gives you an idea of where they are, and you try to correct all of those things while, remember, somebody is still doing the CPR, all the CPR is going on, and you are trying to diagnose the cause of the event.

That is probably one of the most stressful parts of it, especially when the patient keeps cycling through this fatal rhythm and you are left asking why, why is this happening, and what can you do about it?

Kevin Pho: So as technology has evolved, like you said, it was the early 1990s when you suggested putting the algorithms on a monitor. Where are we now in terms of technology to reduce some of that cognitive burden?

Michael Peck: Well, it is interesting you ask that, because within the hospitals, not much really exists. You are still relying on whoever is in charge of the code to remember the algorithms, whether they are experienced or not. There are protocols to follow, but there is no crutch, which is what drove me crazy from 1991 or 1992 all the way up to when I decided to put together a small team to build the product.

I did not like not having a crutch. I do not want to say I was arrogant, though we can say that, but I mean, why not have something I can look at and rely on to make sure I do not miss a step? So for those running codes, it is pretty much the same technology. They have to rely on memory.

Kevin Pho: Are there any cards that come up for the codes? Is there really no reference they can turn to in case they have a momentary lapse of memory?

Michael Peck: No. Right now it is pretty much just somebody taking charge and telling everybody what to do. I do not recall anybody pulling up an algorithm, because to do that you have to stop. You say, “Let me pull up an algorithm,” and then there is a lull in the care. A cardiac arrest is very dynamic, and you need to keep things moving. Once you stop something, the prognosis gets worse.

Kevin Pho: And there is no evolution in terms of digital tools or AI? Right now, when you talk about health care, it is AI everything, but when it comes to ACLS and code care, nothing you see can help supplement judgment?

Michael Peck: I think right now, as you know, in medicine, yes, there is AI involved in a lot of things, but I have yet to see, and maybe this is part of my ignorance, AI involved in a practical way during a cardiac arrest. Not much has really changed in all these years, no.

Kevin Pho: Now, if you were in charge of things, what kind of suggestions would you make? Would you still push for the algorithms on monitors? What would you like to see?

Michael Peck: Well, what I would like to see, for example, is this. The monitors get a little wonky, but the operating room would be fine, because you put it all on a monitor. Somebody like me could sit there and follow it. For the person running the code, it is a practicality thing: it could be a monitor, or it could be a laptop, or anything like that, as long as they can refer to it easily while watching the team do what you are telling them to do, making sure everything is OK.

But right now that is not all that practical. A program can be written, and programs can be downloaded onto a particular device, so either the person doing the documenting can say, “Oh, this is what we need to do next, we just missed that step, because it is here in the algorithm,” or the person running the code can look at a monitor, or a laptop, and watch it scroll through and say, “Oh, what is due? We need to do this next.”

Now that you bring it up, it is very interesting. I almost wonder where VR is going to come into all this, where somebody puts on a helmet or a pair of glasses, and as they are telling people what to do, the algorithm runs through their field of vision. That may be the most practical way of not missing anything.

Kevin Pho: We’re talking to Michael Peck. He’s an anesthesiologist, and today’s KevinMD article is “Cognitive overload in cardiac arrest is a human problem.” Michael, let’s end with some take-home messages you want to leave with the KevinMD audience.

Michael Peck: Yes. So I have always been the kind of guy who likes to build things. As a kid, I was obsessed with building things, and I used to drive my mother crazy. But I think it all led me to always wanting to solve a problem.

So the central message is pretty simple. Medicine works best when we stop pretending that we are superhuman as clinicians, and that is sometimes hard for us to do. Every one of us has limits, and we have to know them. Certainly our memory has limits, our attention has limits, and decision-making under stress has limits. Patient safety improves not when we demand perfection from individuals, but when we build systems that support people during the moments that matter most.

So the goal is not to replace clinicians. The goal is to help good clinicians perform at their best when the stakes are highest and most stressful.

Kevin Pho: Michael, thank you so much for sharing your perspective and insight. Thanks again for coming on the show.

Michael Peck: Thank you, Kevin. A pleasure.

Listen and followApple PodcastsSpotifyYouTubeMore apps · Browse all episodes