Mechanical Ventilation: From Reading Waveforms to AI and Detection Algorithms

Podcast Transcript

Raed Dweik:

Hello and welcome to the Respiratory Exchange Podcast. I'm Raed Dweik, chairman of the Respiratory Institute at Cleveland Clinic. This podcast series of short, digestible episodes is intended for healthcare providers and covers topics related to respiratory health and disease. My colleagues and I will be interviewing experts about timely and timeless topics in the areas of pulmonary, critical care, sleep, infectious disease, and related disciplines. We will share information that will help you take better care of your patients today as well as the patients of tomorrow. I hope you enjoy today's episode.

Hassan Khouli, MD:

Hello, everyone. And welcome to the Respiratory Exchange Podcast. I am your guest host, Dr. Hassan Khouli, chair of the Department of Clinical Care Medicine in the Respiratory Institute at the Cleveland Clinic. My guest today is Dr. Eduardo Mireles. Dr. Mireles is our medical ICU director and the medical director of the Simulation Center at the Cleveland Clinic. He's also the department vice chair. We will be talking today about mechanical ventilation. Eduardo, welcome to the program.

Eduardo Mireles, MD:

Thank you, Hassan. Glad to be here and talk about the topic I really like a lot to talk about.

Hassan Khouli, MD:

I know you're very excited about this and you are an expert on this too. So, I'm actually also looking forward to this conversation there. So, let's start maybe if you can actually describe to us, you know, what is the current state of mechanical ventilation from the point of view of an expert on mechanical ventilation that you are.

Eduardo Mireles, MD:

It's very exciting after, I would say, a long period in which some trials have come out regarding how to use mechanical ventilation. The evolution of both technology and knowledge has been skyrocketing. We have more devices than we ever had with more options than we have ever had. The focus of the research groups has been not only on outcomes of patients and in certain interventions but also in understanding physiology. And that has changed the field tremendously because as we understand more about the physiology of the patients and we create tools that now are at the hands not only of the researchers, but also of the clinicians, we're able to understand better how to tailor and adjust mechanical ventilation to each one of the patients. So right now, it's a hot topic on how to adjust, how to implement mechanical ventilation in its various forms.

Hassan Khouli, MD:

I agree. It's remarkable, you know, the changes that we have really seen. I mean just think about the modes of mechanical ventilation that we have and just to become really more you know, more familiar with them for people who have been in this field for some time or for people who are just really, you know, becoming part of this field in terms of trainees or otherwise.

So that actually is a good segue for, you know, another topic that we can talk about with all this proliferation of mechanical ventilators, the technology itself and also the modes and the physiology, the approach to it too. What are some of the challenges that you see in education in this field of mechanical ventilation?

Eduardo Mireles, MD:

That's a great question, Hassan. Then, the challenge is the following. Mechanical ventilators over the last 50 years have evolved. And we have now several brands. At the last count, we had around 500 modes of mechanical ventilation. It's impossible to know each one of those modes. Each one of those modes has a special name because each one of the manufacturers wants to use a specific name for it and sell and be the best thing out there. However, each one of these modes behaves in a different way. The names don't tell you how they work. There are all these manuals out there that talk about each one of their ventilators. They use nomenclature and words and classifications that don't share across each one of the platforms. So, technology grew rapidly and with an explosion of terms and options.

On the other hand, you have education. And the education of mechanical ventilation has remained at a very slow pace of development, although, because we have been trying to catch up with the technology, one, and number two trying to homogenate all this nomenclature, and understanding what is being developed. That gap that exists between technology, development, and education is a major gap and a threat to our patient care because you now have devices that you think that you know what they do because you read the manual or you saw, you were trained by the representative or you have played with it or used it on other patients. And it turns out that the device has multiple other options that they are getting activated as you're dealing with the patient.

So as a clinician, as a practitioner at the bedside that is putting a patient on a life-saving measure, you need to understand what the machine is doing. And so, that gap between technology and education needs to be filled. And that's our current challenge.

So how do we feel it and how do we get people to understand and our teams to be ready for today and for the future because this technology is gonna continue to evolve. So, we have to prepare them to be able to, in 10 years, understand what's happening with the ventilator.

Hassan Khouli, MD:

Very much so. So, what are we doing about it? What are we doing here at the Cleveland Clinic, you know, about it? And if you can actually share with our audience, you know, a few words about the SEVA Program that we're so excited about that you are leading the efforts here and nationally also a great program there, please, educate us on this Eduardo.

Eduardo Mireles, MD:

Well, I'll tell you where, where it started. Actually, years ago, when I was still a fellow, Rob Chadbourne came to the clinic. And Rob has been a luminary within the respiratory care world. And we started talking. He had been working on how to classify modes of mechanical ventilation. And my interest and passion for mechanical ventilation and as we started talking further and working on this, we realized that there was a major opportunity for education on how to deal with this.

And we created courses in mechanical ventilation. But over the last 15 years, this has evolved to a really methodical way of filling that gap, that, that void that exists. And it starts with understanding what the mode does, what the patient goals or ventilation are, how to read and understand the ventilator graphics and the interactions with the patient, and then how to apply all these physiological techniques and adjust the mode so that it can achieve the best for the patient.

So that's how this started. And over the last six years, we have implemented what we call the SEVA course. The SEVA course stands for Standardized Education on Ventilatory Assistance. And it's a course that whoever takes the course from the very basics of how a mechanical ventilator works all the way to advanced physiology. So, we have currently seven levels that we can move through at different venues initially online at your own pace self-learning.

And then, using simulation and team training to really get the team to work together to set up the ventilator. The overall goal of the SEVA training is to ensure that the practitioners at the bedside have the basic knowledge of how to manage the ventilator and that they have the tools to understand what's coming in the future.

There are several levels that we have created. So, for some of our practitioners that are just gonna be in the intensive care unit and they're gonna be dealing with patients, they need to know the basics. And so, that's SEVA basic level. From there, it starts moving up to a second level which we call advanced in which you understand more in depth how the modes work, how to apply them, how to select them, how to titrate them.

And then, there's a third level which we call the master level in which those that want to advance to that level will learn more about physiological tools, the esophageal manometry, volumetric ethnography, recruitment maneuvers, titration of PEEP, and whatnot. Very comprehensive, without a doubt.

Hassan Khouli, MD:

That's great and very exciting too. You know, you describe the challenges, the exponential growth and how education needs to really, you know, meet these challenges. And this is such a unique innovative program that can certainly provide that ask and expectations there.

I'm just curious to hear, you know, the people who have really gone through this, what has been the feedback about it?

Eduardo Mireles, MD:

We love feedback. Actually, the course has gone through several iterations year after year. And we continue to improve on this. The level to get people to, to understand the current state of mechanical ventilation and abandon some of the prior thoughts, historical thoughts, does come with a challenge. And we have been working on techniques on how to make it easier to digest for our caregivers.

However, in those that have completed the course, the level of understanding of what's happening with the ventilator is eye-opening. Suddenly, they started seeing things that had been in front of their eyes for years and they said, "You know, I, I didn't know that that was what was happening or that's how I had to deal with it."

So, there is power in the knowledge that comes from completing this course of having those basic understandings because then, you can, actually, I'll tell you. This happens pretty regularly. But recently, one of the respiratory therapies actually came to us and said, "You know, Eduardo, I was looking at the, at the waveforms for this ventilator. And suddenly, something happened. I don't know what happened. And there's something abnormal in it."

And they called us. We came and looked at the waveforms and absolutely a strange waveform that had nothing to do with what was happening. But they looked normal but abnormal on the ventilator. What turned out to be that was that the motherboard had a drop of water. And it flipped the image on the ventilator flipped the image on the ventilator which in the normal behavior of people, people may say, "Oh, I would have detected." It had gone unnoticed by several people. But because they are now trained in how to read the waveforms of mechanical ventilation, they were able to detect this thing that probably could have been potentially disastrous, right, having water in the, inside the ventilator. And like this one, I have several stories of the synchrony issues that we had not detected or malfunctions of the ventilator or tricks of the ventilator that we didn't know of that because now, they are trained on it, they can detect and actually help us decide if that's a good thing for the patient or not.

Hassan Khouli, MD:

That's a fantastic example of how education at its best really prevents errors, you know, providing, enhancing patient safety at the bedside. So, you just described to us one of the future elements of mechanical ventilation, the education piece of that. What other future aspect of mechanical ventilation can you foresee for the next decade or longer?

Eduardo Mireles, MD:

Well, as anything that it's a part of our electronic life these days, the evolution is accelerating towards having more automation to have devices that function based on algorithms that may be expert based or maybe based on decision algorithms or it may be based on an artificial intelligence.

As time moves forward, we expect to see more of that integration into our mechanical ventilators. And hence, the importance of understanding how the ventilator is being controlled and what would it do in the presence of certain interactions because, as a practitioner, and currently, we have modes that have certain level of intelligence built into their protocols that when you see the interaction of the patient, you're the only one that knows what the patient has and what the patient needs, and what your goal is. And you apply that tool.

And the tool, the algorithm or the intelligence built into it may not be appropriate for the patient. So, seeing this growing towards more automation in the future, more decision support towards our caregivers on how to manage and adjust the ventilator for certain groups of patients. There's gonna be, and it's already in the pipeline, better detection of patient ventilator interactions that are injurious. So, synchrony issues that may be harmful for the patient, the detection of them and how much, how often, and how those- this develops. So that's where we're going on that arm of mechanical ventilation.

And the next other area that is also developing rapidly is extracorporeal life support. So as the mechanical ventilator will continue to be part of our armamentarium, but how does it relate to extracorporeal life support, how is that relationship and perhaps even how do both of them get regulated under the same environment, right? Right now, we manage extracorporeal life support and the ventilator separately. But perhaps, there is a pathway in which both can achieve minimum impact from the ventilator, minimum impact from the extracorporeal life support. So exciting without a doubt, where we're moving with this.

Hassan Khouli, MD:

Very much so. And maybe, we can stay a little bit on the extracorporeal life support. You know that this has, for sure, expanded during COVID also. Received a lot more attention. It is part right now of some of the mainstream critical care, you know, healthcare delivery that we have applications there. Maybe if you can share with us, you know, your thoughts about where you see this field and how does it really interact with mechanical ventilation support and applications there.

Eduardo Mireles, MD:

Yeah. It has come with a bank during, during COVID, right? So now, it's actually part of the regular conversation. Before, it was a more nuanced conversation. Now, it's more part of our lingo of, well, should this patient go into extracorporeal life support?

There is still a lot of development that needs to occur. And I mean it has accelerated. But a lot of developments in terms of what's the right time for the patient and what are the indications or contraindications for certain groups of patients? We are learning time after time. And especially, I'll say that now during this COVID pandemic, we learned a lot about patients that in the past, we would have said that lung is not gonna recover just based on imaging and function on the physiological parameters that we had. And because now, we had them on extracorporeal life support and we had the benefit of time, those lungs started recovering. And some imaging that you would have deemed end of life turned out not to be end of life in this acute lung injuries.

So, I think that it opened up a huge area of understanding of saying, "Well, the ventilator can take you to, to a certain point." And then, a transition into extracorporeal life support with or without mechanical ventilation is gonna be the next frontier. You're saying, "Well, how much do we need to put them in or not?" We have, as you know, worked on extracorporeal CO2 removal as another maneuver of doing low intensity extracorporeal life support along with mechanical ventilation.

And so, with that trying to decrease the amount of intensity from the ventilator, allowing the patients to have less intense ventilation and perhaps less sedation, less paralysis and, at the same time, not putting them on the full spectrum of extracorporeal life support and just enough so that the lungs can recover during that period of time.

And that also has promises. I mean this is an area in which you're doing trials are coming. And we're trying to learn how to do it appropriately. And sometimes, people read these trials. And it's very easy to try to see black and white and, and say, "No. It doesn't work. Yes, it works." And I think that it's always a gray as we have learned in everything that we do in medicine. You have to start finding what is the actual niche and the group of patients that benefit because we have seen patients that benefit tremendously from others, and you wonder if that was the best decision.

And so, it is because we need to learn how to calibrate. At what point do you put on extracorporeal life support? What time do you continue with mechanical ventilation or at what time do you use both and how to minimize those two so that the patient can go back to their life? That's the key part. So that's where I think that we're going with extracorporeal life support. But certainly, this is not the last conversation that will happen. In the next 10, 20 years, we probably will have more clarity.

Hassan Khouli, MD:

I think so. And we have, you know, we have certainly seen here at the Cleveland Clinic some good examples and meaningful, you know, a way of when you really apply extracorporeal life support with mechanical ventilation being used wisely and with the expert doing that how rewarding that is for the patients as well as for the caregivers who are caring for these patients with many examples of patients surviving with good outcomes, with that approach that you just described here.

Another exciting development in the area of mechanical ventilation is the use of physiological monitoring and all the new applications or really not new, maybe some using them in different ways too like esophageal pressure monitoring or so there. What are your thoughts on this? And where do you really see and how do we use them right now at the Cleveland Clinic from your experience?

Eduardo Mireles, MD:

Yes. This is another area that has now become available to everybody. So before, it was a thing that we would do for research, right? You could do monitoring and measurements. And it was really focused on research for physiology or trying to titrate items in terms of mechanical ventilation.

Now, it's not. Now, the ventilators, actually many of them, have extra ports that can allow you to do esophageal pressure monitoring, volumetric CO2 to better calculate and deliver tidal volumes on this, on our patients. So where do I think this is noble and it's if applied appropriately and protocolized appropriately would be a good thing, is that it allows us to personalize some of the things that we're doing.

When you have knowledge about the intensity of the effort of the patient when they're bringing the ventilator which is really hard to see just with your plane eyes, we're very good at seeing under assistance. But we're not very good at seeing over assistance. And over assistance, we're learning that it may lead to respiratory muscle injury or atrophy which can lead to prolonged patient stays or prolonged mortality.

So having these types of tools like the esophageal pressure monitoring or the electrical activity of the diaphragm may help you recognize and titrate the amount of the intensity of the mechanical ventilation to preserve the function of the diaphragm.

On the other side, the use of volumetric capnography may allow you to not only titrate your levels of PEEP to an up- the least amount of debt space. But also, to make the tidal volume more efficient by adjusting your settings. So, there's all these items that we know that, in physiological studies, work. The question is how do you put them at the bedside? This is also a call for empowerment of some of our interprofessional team. So, our respiratory therapists which, in many places, have the ability to do certain things places for the minority.

But the key is how do we empower them to do this in a protocolized way and guide the rest of the team into better titrating the care for the patient. Whether these will translate into better outcomes well, it's yet to be seen. But I can tell you at least in those groups of patients in which it matters, it helps you titrate and adjust at least with an eye rather than with just a filling. So, you have direct objective observation.

So, it's a very, very exciting field. I look forward for the next years as we are working towards our group of respiratory therapy, obviously, our residents and fellows too, to grow in that area and follow protocols that we can test how they affect patient outcomes.

Hassan Khouli, MD:

That's really a great point. And, you know, you make a good point about training and empowering the respiratory therapists who are, you know, at the bedside with the ventilators, with the patients, you know, a lot more often than, you know, physicians or physicians in training too. So that approach to making them part of the team is, I think, is gonna be quite rewarding not just for them but also for the patients and for the full team to be there.

That's actually a good point to make about, you know, these physiological monitoring applications that we have. I mean we see a significant interaction sometime between the patient and the ventilators. And that can actually affect the outcomes of duration of mechanical ventilation, the injury that can happen there too.

So maybe, you can expand a little bit on that, you know, about the observation, how we manage it, what are your thoughts about what should be really done in this area to address some of these detrimental issues.

Eduardo Mireles, MD:

Yeah. The patient ventilator interaction actually becomes probably the majority of the time a patient is on the ventilator. We are used to being trained on focusing on the first days of mechanical ventilation of patients with acute respiratory distress syndrome or with COPD exacerbation or whatnot. But in reality, now, we are moving towards having them awake and interacting with the ventilator.

And that interaction turns out to have an impact on the amount of sedation that you get. The other co-interventions that come from increasing sedation or paralysis. So, the better that we are at interpreting what's happening on the ventilator at that time, the better it is.

So we work a lot on how to read the waveforms in the ventilator so that there's a basic understanding as you're walking through the bedside how to read that and have a systematic way to understand the context, meaning what's the goal for the patient, what is the, the physiology of the patient, what mode the patient is, and what are you watching. What are you seeing on the screen and, from there, adjust.

I think that in the future there has been developments on how to not only capture but analyze on-site the waveforms that are coming out because, without a doubt, it's a skill that you need to acquire. So artificial intelligence or detection algorithms that can flag the presence of certain types of synchronies are key. And there's a good number of articles coming out the pipeline regarding the presence of detecting through the lifespan of a patient on mechanical ventilation, the presence of synchrony issues. And when there's clusters or high amounts of synchronies, the outcome of those patients seems to be associated with prolonged mechanical ventilation and increased mortality.

So, this is majorly relevant. I mean this is a multi-level announcement. The first one is as providers; we need to learn how to read those waveforms. As technology manufacturers, we have to implement algorithms to read these alarms within the ventilators or within the system where you're living, having automated detection systems that can find those types of patients that are having those issues.

And the last one and the last message is that the ventilators have so many options that can help us deal with those types of patient ventilator interactions. We are past long due from those trying to manage patients with just one mode because that's the one that we know how to manage. The ventilators have many options. And we, as practitioners, need to learn how to use them and then apply them, if that's the right mode for the patient because it may help decrease those dangerous patient ventilator interactions. So exciting times but a lot of responsibilities that we need to work on.

Hassan Khouli, MD:

Very much so. You know, it's hard when, you know, when I hear you talk with all this excitement about technology and the advancement in technology and how we can leverage technology to, you know, to really help us manage patients and improve their outcomes, not to be a little bit nervous about, you know, is technology taking over. What is gonna be the role of physicians and, and real healthcare providers caring for these patients there? So, what can you share, you know, with our professional audience over here, things from your perspective about the role of machines and how we can maintain our role that the ICU at the bedside with mechanical ventilation?

Eduardo Mireles, MD:

I would put it at different levels. The first one is that sensation that they are taking over is a, I share it at times. I wonder if all the work that we're doing on teaching people how to read mechanical ventilation and waveforms is gonna be useless in a couple of years. But at the end of, the picture is what is best for the patient, what is going to be safer and better quality for our patient.

And if you were on that ventilator, the amount of time that me, as a practitioner, spend at the bedside compared to the time that the patient spends on the ventilator and having those interactions, I may not pick it up when I'm there. So there has to be a technological solution to this. It's like if you were driving your car and you were only opening your eyes for 10 minutes or one minute while you're at the bedside, the rest of the time you're driving blind.

So, I think that that's how I think about how to use this technology. At the same time, you are the one that decides where the car is going to go. You were, tell right, left, I mean autonomous driving is around the corner. We keep saying that. But you keep seeing the challenges that come with autonomous driving.

So, in the healthcare environment, it's even more complex. These are not nice, beautiful streets with lines in the middle. These are patients that come with 10,000 other diseases. So, the physician's role on how to apply this technology is gonna remain there for a large amount.

And I don't see it ever separating until there is enough intelligence if you may within those automated systems. So, in all that transition which we need to be aware of how to do it, how to set the goal, what does the machine does and how to get there. So, I am not scared about technology. Actually, I am empowered about it because it will make it safer for our patients. And, I mean, you can imagine how as you're moving through this, this pathway of how to use automated and decision-making systems is how to adjust the ventilator and then, say, "This comes hand in hand with sedation."

Is there a way to actually link those two in which patients are starting to complain of pain like with a PCA and the patient gets that, and the ventilator knows and increases the support because it knows that this is gonna happen? So, there are things that we can do that don't take our autonomy away, but that makes it safer.

Hassan Khouli, MD:

It's really perfectly said. It's reassuring at the same time, you know, saying the word that I think that I certainly resonated with me the most. We drive this as people really who care for the patients. And the drive is, you know, utilizing these, this technology as tools to enable us to help patients and provide them with the best outcomes there.

Eduardo, this has been really a very enjoyable time, learning from you as always. Thank you for being here Dr. Mireles. And thank you, everyone, for listening to our podcast today. I'm your guest host, Dr. Hassan Khouli, MD, chair of the Department of Critical Care Medicine in the Respiratory Institute at the Cleveland Clinic. And my guest today was Dr. Eduardo Mireles, MD. Dr. Morales is the director of the Medical ICU and the medical director of the Simulation Center at Cleveland Clinic and our department vice chair for Critical Care Medicine.

We spent time talking to you today about mechanical ventilation and the future of mechanical ventilation. Thank you.

Eduardo Mireles, MD:

Thank you, Hassan. I appreciate it.

Raed Dweik:

Thank you for listening to this episode of the Respiratory Exchange Podcast. For more stories and information from the Cleveland Clinic Respiratory Institute, you can follow me on Twitter at RaedDweikMD.