Identifying, Managing and Treating Sepsis

Podcast Transcript

Hassan Khouli, MD:

Hello everyone and welcome to the Respiratory Exchange Podcast. I am your guest host, Dr. Hassan Khouli, director of the Critical Care Center and Chair of Critical Care medicine in the Integrated Hospital Care Institute at the Cleveland Clinic. My guests today are Dr. James Morrison, staff physician in critical care medicine and emergency medicine, and the chair of the Enterprise Sepsis Committee at the Cleveland Clinic. Dr. Matthew Dettmer, staff physician in critical care medicine and emergency medicine, and co-medical director of the Sepsis Emergency Response Team, SERT, at the Cleveland Clinic main campus. Today, we will be talking about sepsis. James and Matt, welcome to the program. We have a lot to talk about today in the area of significant interest to many of us in healthcare, as well as the public.

James and Matt, welcome to the program.

James Morrison, MD:

Thanks, Hassan. Thanks very much.

Hassan Khouli, MD:

We have a lot to talk about today in the area of sepsis. As you can imagine, all of us know the significant interest among the public, among many of us in healthcare in this, including really the government. I'm looking forward to our discussion today and to inform our audience of how things are being done in the area of sepsis and where we want to go there.

James, I'm going to start with you with the first question. Can you give us an overview of what is a definition of sepsis and how do you really qualitatively in a way define it?

James Morrison, MD:

Thanks for the question, Hassan. I think it's a good place to start because sepsis is so broad that it can almost be hard to tackle at points. When we think of it in the broadest sense, we think of it as a life-threatening medical emergency. It's caused by the body's response to an infection. We know that sepsis can lead to tissue damage, organ failure, and death, particularly if not treated quickly. That's how we define it in a broad sense.

We know that not every infection can lead to sepsis, but when it does, it's as dangerous or more dangerous than other common conditions that we think of, like heart attack or stroke. That leads us into one of the key parts of it is that sepsis is important to us and to our mission because it's a leading cause of death in hospitals worldwide. That's the broad strokes of what sepsis means to us.

Hassan Khouli, MD:

Yeah, I thank you, James, for really framing it this way. I think when people, even among us physicians think about stroke, think about acute MI or so, we become quite alert, and our responses are quite different. I appreciate how you're framing it, that this is really as life-threatening and as much of a concern for all of us in healthcare, too.

Maybe actually we can dig a little bit deeper into this and give us a little bit of maybe some numbers about when you say it is a leading cause of death, what is that like?

James Morrison, MD:

Yeah, we want to recognize mortality because sepsis is one of the leading causes of death in hospitals. In our own system, we know that sepsis causes two out of five hospital deaths. That's something that's seen pretty much across the board in the country and globally, so a massive, massive impact in terms of mortality. But also huge in terms of for as many patients as are dying, there's many more that are injured and have long-term consequences. The impact really can't be understated.

Hassan Khouli, MD:

I agree. You both as physicians who really work at the point of entry when patients come through the emergency room and then take care of the more seriously ill patients, critically ill patients in the ICUs. You really identify and understand the emergency of being able to care for these patients the best way we can and identifying them the best we can, too.

As you frame the definition the way you did and then the importance of us really focusing on this, what are some of the clinical criteria that allow us to further define sepsis?

Matthew Dettmer, MD:

Yeah, thanks, Hassan. I can take that one. I think there's actually what makes it a little bit challenging is that there's probably a couple sets of clinical criteria that exist that are used in different applications in the present day. I think to understand why that happens, just having a little bit of historical perspective is helpful because there's really been a lot of development about our understanding of sepsis, how we manage sepsis over the last 25 years or so.

In the '90s, sepsis carried a really high mortality. That was the prompting for a lot of research into why that was happening and how we should manage sepsis in a better way. A real pivotal study that came out in 2001 is referred to as the Early Goal-Directed Therapy for Severe Sepsis and Septic Shock study, which showed that prompt management and resuscitation for septic patients was associated with a significantly improved mortality. That study really is important in its own right, but I think it's very important because it's opened the floodgates with respect to even further research into understanding how sepsis works, how to improve mortality in it.

In correlation with some of those scientific developments over the last 25 years, there's been an increasing focus on sepsis from a government regulatory standpoint. There was a case in New York called the Rory Staunton case, which was somewhat unfortunate. In 2013, a pediatric death due to sepsis that really raised awareness that sepsis is a cause of mortality and that because it's such a significant contributor that it warrants some regulation from a governmental standpoint. That prompted the development of sepsis guidelines and metrics from the Center for Medicare and Medicaid Services.

Through those parallel veins of sepsis evolution over the last 25 years, we have definitions that are pertinent that have developed from the scientific community and then definitions that are more pertinent for regulatory bodies as they define sepsis and when sepsis occurred. In summary, we're left with, and there's probably a few other definitions as sepsis, but I think two that are really important to emphasize right now are the sepsis three definition criteria. That's a consensus set of guidelines from the scientific community, and then a SIRS-based criteria that's used by the Center for Medicare Medicaid Services for their SEP-1 national benchmark.

Hassan Khouli, MD:

Thank you. Thank you, Matt, for sharing this historical perspective. Certainly over the years sepsis definition has really evolved, and you talked about that tragic case that increased the public awareness and triggered some of the follow-up in terms of [inaudible] focus, although two, and we are where we are right now in terms of a more defined criteria for sepsis there.

Maybe, Matt, we can follow up on that question. There are two important sets of these diagnostic criteria that you summarized there. Are these currently fully implemented throughout our hospitals? I'm now talking about the clinic and beyond that, too. For us in healthcare, this is where we're really going to focus our area of how we define sepsis here.

Matthew Dettmer, MD:

Yeah, I'd say both are at play in different venues. I'd say the sepsis three diagnosis is a key one in the research sphere in terms of understanding patients who not only have infection but are at higher risk for death due to sepsis. And then the SIRS-based sepsis definition is very important from the CMS standpoint. That's how they assign time of onset for sepsis so they can grade how well or not well hospitals provide care for sepsis.

Both are at play. I'd say one more and in the scientific research side and one in the regulatory venue.

Hassan Khouli, MD:

Thank you. Matt. James, would you like to add more to that in terms of that life-threatening sepsis?

James Morrison, MD:

Yeah, so I think Matt has successfully laid out the paradigm where the sepsis three definition is based on defining groups that are at higher risk for mortality rather than having a really rule-based criteria. The application is through SOFA score changes at the bedside, but that can be a little challenging. I think it'd be helpful for us to just talk about what's literally in the definitions and then maybe we can hit some of the nuances between the interplay of how they're used and how we use them.

Sepsis three defines sepsis as a life-threatening organ dysfunction caused by a regulated host response to infection, which is clinically meaningful, defined as roughly a 10% mortality we expect to see in that group. Operationally it's deployed using a change in SOFA score, so a SOFA score greater than or equal to two, or a difference of two from a baseline known organ injury. Again, that's key because it defines a 10% mortality.

There's a second subgroup of patients with septic shock, and those are really just defined as particularly profound circulatory salivary metabolic abnormalities with a greater risk of mortality. In this case, we're talking a mortality of up to 40%. Clinically this is defined by the need for vasopressors to maintain an adequate blood pressure and more specifically if the lactate is over two without hypovolemia.

Hassan Khouli, MD:

Thank you, James. These are really more clear definitions and have evolved over time. There's still obviously some of these patients who are the nuances about how we manage this at the bedside, too.

I'm curious actually to hear from you, James, about where does SIRS fit into this definition framework?

James Morrison, MD:

Yeah, I think we touched on how they're used differently in a scientific sense, in a clinical bedside sense. I think you really can't give an up-down vote on the definitions, but you have to understand their strengths and weaknesses.

SIRS still has a role to play because it's a rule-based criteria. We know another strength of it is its sensitivity. It's good in so far that it helps us catch sepsis cases and sensitivity is high, and that's at an exchange of sometimes of specificity. SIRS still has a role to play in that sense in having concrete rules that people can use and assess either in the medical record or at the bedside. That contributes to its use in regulatory affairs, too.

Hassan Khouli, MD:

Thank you, James.

Matt, I'm going to actually switch to you and ask you a question about the patient populations and what patient populations seem to be at a higher risk of developing sepsis or a severe form of sepsis.

Matthew Dettmer, MD:

Sure. I think that's a really important question. I think, again, thinking historically in the last quarter decade or so, we've thought about gradations of sepsis, whether it's sepsis, septic shock, et cetera.

I think another way to think about the disease process is whether there's different subpopulations that develop sepsis that should be considered maybe a little bit differently in terms of how they manifest the syndrome and therapies that can improve them. One way to subdivide septic subtypes or phenotypes might be patients who have significant underlying diseases. Patients with malignancy who might be on chemotherapy obviously impacting their immune system might present with sepsis and have worse outcomes than other patients with other underlying comorbidities. I think that's one way to think about different patient populations as sort of the primary underlying comorbidities.

Another way that I think is valuable to think about sepsis has to do with where the patient is when they develop sepsis and where they are when they're receiving care for sepsis. The vast majority of patients in this country, for sure, who develop sepsis do it in the community or outside of the hospital. And then based on the symptoms that they're having, they come to the emergency department. That's where a lot of their care is initially conducted. We call that patient population community onset, or maybe present on admission sepsis. It's important patient population because, again, it's the majority of patients that have sepsis. It's also because it's the majority of patients, it's where a lot of the clinical research that's been conducted is in that patient population.

The opposite of that being patients who are already hospitalized for another reason, whether it's for cancer care or heart failure exacerbation, were already in the hospital and then develop sepsis in the hospital, a population that you might call hospital onset sepsis or not present on admission. That's an important population. It represents a significant minority of patients globally who develop sepsis but historically has consistently had a higher mortality than that community onset population. And so, I think is an important subpopulation because of that worsened mortality, a population that's an important target for research.

Hassan Khouli, MD:

Thank you, Matt. It is interesting how you divided these patient populations either by comorbidities, underlying diseases, or where they develop sepsis. Almost two different phenotypes of patient population: the ones that arrive through the emergency room and they already have signs of infection and sepsis versus the ones who develop it during their hospital stay there. We'll come back and maybe ask you a couple of questions about that later, too.

James, you have referred both of you and Matt to how the government and the regulatory agencies in the government are focusing on this and are interested in how to maybe regulate sepsis in the hospitals or outside the hospitals. Can you give us some examples of this, and how does that really play out in the healthcare systems?

James Morrison, MD:

Yeah, I think the obvious one that's been a huge topic of discussion for years really now and is still a persistent and active issue for the government is CMS and SEP-1 measure, which really was derived out of some of the seminal early research and understanding that protocolized early care treating sepsis and emergency led to better outcomes. It's been brought together into a measure that uses SIRS criteria as an on-ramp and then recommends or asks providers to provide protocolized care in bundles.

This is where we hear this nomenclature a lot, the three- and six-hour bundles. Yes, these are meant to be evidence-based approaches to help providers give standardized care, protocolized care, and then to assess the ground game or the performance of a hospital based on that measure.

Hassan Khouli, MD:

Thank you, James. You mentioned the word performance as the government wants to measure the performance of how hospitals are doing in the management of patients with sepsis there. How do we measure our performance in sepsis care here at the Cleveland Clinic?

James Morrison, MD:

Well, I think I alluded a little bit to controversy, and I'm minimizing that, but there are strong opinions about SEP-1. We recognize the challenge of trying sepsis as we are highlighting. It's very broad, it's syndromic, and it's hard to get a clear thing that applies to everybody.

In that sense it's been a challenge and we've really had success by appealing to our caregivers to our core mission, which is to take care of our patient and save lives. When we talk about success at Cleveland Clinic, we want what our key outcome is to reduce death from sepsis, to prevent sepsis deaths, to decrease harm from sepsis. That's how we talk about it, that's how we talk about our mission, and that's what keeps our providers interested.

To quickly answer that, we look at mortality and we do some risk adjustment to understand our performance based on a benchmark, based on what our patients come in with, what our risk model suggests would be mortality, and we make sure that we're beating, that we're preventing sepsis deaths and exceeding the benchmark.

Hassan Khouli, MD:

Yeah, certainly. When you talk about outcome mortality ultimately is the most important outcome, especially in life-threatening disease like sepsis there.

What are some of the best practices that we tend to think about that will lead processes that will lead to improving outcome in the area of sepsis?

James Morrison, MD:

Great question, Hassan. Because you take this broad sense of a provider walks in the room and they want to reduce sepsis mortality, how do we make sure that if we have a standard of care and that it's delivered across the board and to everyone? We need key processes to measure, to give ourselves feedback, and we've based our ours on things that we know lead to differences in patient outcomes.

The key one that I think is pretty broadly recognized and the best supported by evidence is antibiotic timing. We have looked at the literature. We're all aware of some of the early studies that show an hour to hour deterioration and increased sepsis mortality in delaying antibiotics and septic shock, and we have a sense of how it does with other sepsis conditions as well. That's our key metric, and we know that in our own system that our mortality decreases from around 10% to 7% for all types of sepsis when we get our antibiotics in on time.

I'll cover just a couple of other quick ones is I've highlighted the importance to us of standardized care so that we make sure that the patients are getting good evidence-based and guideline-supported care. We use an order set which helps us achieve that. Really takes a lot of the nitty-gritty pieces and makes it easier for the provider to deliver that.

And then we have some standardization in how the care is delivered. We use responses that we call, we have specific names for a sepsis alert and a code sepsis, but the key is we activate the team, we come to the bedside, and we work together to meet those key metrics on time, which are getting antibiotics in time and doing the other standardized care contained in our order set.

Hassan Khouli, MD:

Thank you, James. It is really interesting what it seems to be to many of us a simple intervention, like antibiotics administration within a defined period of time, is how challenging it is to achieve that outcome, that target there, too. But a lot of good work is being done. It's good that people can focus on a specific intervention, specific process, and now drill into it and be able to deliver on time consistently I think as you mentioned, too.

James Morrison, MD:

I agree, Hassan.

One point that I'll bring up is we push a specific metric, but it bumps up the quality of care across the board. When we call an emergency response and really push on antibiotic timing, it elevates the concern about sepsis and the response across the board, there's better. And so this communication within providers and the activations have made a difference for us to meet those things. The core is treating this like an emergency and working together. Good communication, right?

Hassan Khouli, MD:

Thank you, James.

On that word of emergency, I'm going to actually come back to Matt, and then we talked about sepsis emergency response team. Matt, I maybe already answered that question, but I would like to hear more from you about SERT and what that program is all about here at the Cleveland Clinic.

Matthew Dettmer, MD:

I can certainly elaborate on that point.

The sepsis emergency response team conceptually is a team of providers that tries to identify and assist with management patients with early signs of sepsis on the regular nursing floor. Speaking to that population that we already discussed about hospital onset sepsis, trying to help identify those patients and then provide some of the evidence-based interventions to improve their outcomes.

Functionally we have a team of, we run 24 hours a day, seven days a week with a team of APPs, physician assistants and nurse practitioners as well as dedicated nursing support. Our teams identify patients through a couple different mechanisms. The primary mechanism is using an electronic screening system that identifies patients with SERT's criteria and objective evidence of organ dysfunction. Our teams also hear about patients via either our nurses soliciting a concern from bedside providers or from our other rapid response team, our AMET team, who may be called outpatients. If they're concerned about sepsis, they can forward that to us as well. We try to identify those patients, and then it's based on either a combination of chart review as well as in-person evaluation. See if those patients really do have sepsis and if there's interventions that are indicated.

Obviously, all those patients are under care at the time that we identify them of an existing primary team. And so our teams do a lot of very active collaboration with those teams, recognizing that especially at main campus we have a lot of patients with very super specialized diagnoses that don't present the same way all the time when they do have infections or sepsis.

It's really a unique team in the sense that not only our model of electronic screening but really our focus on patients on the regular nursing floor. A lot of teams that exist in this capacity focus primarily in the intensive care unit or the emergency department, which are obviously important places to intervene early on septic patients. But our team is really taking the idea that early interventions on sepsis is really valuable to try to capture that hospital onset sepsis, whether on the floor before they decompensate and go to the ICU is really the focus of our team.

Hassan Khouli, MD:

That's such a really comprehensive program that has been put in place here at the Cleveland Clinic with, as you mentioned, a lot of resources and data collection to really help us move forward in this area there.

I have a couple maybe follow-up questions on this, Matt, if you can elaborate more on the SERT there. We talked a little bit about that phenotype of patients coming through the emergency room and then patients possibly developing sepsis within the hospital there. How are these two different phenotypes different?

Matthew Dettmer, MD:

Yeah, and this is, again, I think an area of ongoing study that those patients coming through the ED versus that develops sepsis in the hospital. While I frame those as two different kinds of phenotypes, I think there there's probably a little bit of room in between the patient that spikes a fever on maybe day two of the hospital stay. Did they really have an infection before they came in? Maybe it was subtle and didn't present until later, or truly the patient developed the infection and the subsequent sepsis totally in the hospital. Realizing that there's probably a little bit of a spectrum in between those two broad categories.

But I think really diving into what explains that NPOA mortality is a really interesting question. I think obviously there's other elements of acute illness that are present if somebody's already in a hospital, so their risk of death from an infection and subsequent sepsis is probably to some extent higher because they're sick enough to be in the hospital already.

Some of the organisms that grow in the hospital tend to be a little bit more resistant. I think that plays a role. I think those are some potential explanations, but really it's an area that's ripe, I think, for research to understand what's going on there and if there's preexisting things that contribute to that mortality. Or if some of the interventions that we do, some of the CMS three and six-hour bundles that are important for the broad population, whether that is specifically applicable to this population or maybe deserve to be tweaked a little bit, I think are active areas of inquiry.

Hassan Khouli, MD:

Thank you, Matt. I agree. I think with all the data that is being collected and the experience that we have in a comprehensive program like this in the future, we're looking forward to hear more about how can we really be more informed about the value of a program like this. Also, how we can really predict sepsis in a different patient population within the Cleveland Clinic and [inaudible]

Matthew Dettmer, MD:

Yeah, I think that's a really good point. It's as valuable obviously as our clinical apparatus is something that I think provides a lot of value is we have some dedicated data abstractors as well as folks that pull pieces of data out of the electronic medical record.

I think one of the challenges to studying hospital onset sepsis is patients are already in the hospital for such a prolonged period of time. It can be challenging to sort out when they developed infection, when they developed sepsis, when they got antibiotics, et cetera, with our very dedicated data. That's something that we're able to provide and so provides the groundwork for, again, I think a lot of interesting research endeavor.

Hassan Khouli, MD:

Thank you, Matt.

This is such a strong program here at the Cleveland Clinic. How else can providers really contribute in general to the process, to the outcome, of patients who have sepsis?

Matthew Dettmer, MD:

That's great. I'll speak maybe to SERT, and then James can probably talk about the enterprise in general.

I think from our standpoint, we recognize this is a new idea to have a team that's really on the lookout for sepsis. And so as a primary team with a lot of patients and patients that you know very well, it's I think it's a new evolution to collaborate with this outside team that sees everything through a sepsis-colored glasses, so to speak. And so I think that for our providers, the bedside, just being aware of who we are and why we're there and that we're really trying to hone in on this very, very fragile patient population.

I think it's important for everybody to know who we are and so that when there are opportunities to collaborate. Because we learn a lot obviously about some of these subspecialized patient populations as we're trying to intervene on the patients, but also provide education about sepsis. I think that's one mechanism.

Hassan Khouli, MD:

Thank you, Matt.

Maybe, James, you can expand more on this and share with us some upcoming sepsis initiatives that are already on the horizon here.

James Morrison, MD:

Well, I wanted to bring up a point which is talking about sepsis and sepsis regulation can really just disengage providers. I think we've done a good job speaking to what's important and motivating people about mortality in that and engaging them with the OKR. That engagement, we've seen it across the board, and it's fostered a lot of our own improvements. While we want to bake in some of these core practices, we've also seen a lot of innovation that's started at a local level with a provider's idea or a team's idea and is deployed across our entire system, which means the globe, really.

I can give an example of in our cancer institute. They really engaged with antibiotic timing and made a dashboard about antibiotic turnaround time that has been deployed across the enterprise. In one of our specific hospitals, we had ED nurses competing for antibiotic times in between the specific teams. That kind of competition was spread across the way as a way to just engage people and keep them motivated and celebrate teams that are having success. We've really found that that's an important part. Not always a slap on the wrist, but these teams are doing a good job of working on this hard. When we celebrate them, they dig in. They do it even more.

And then we have wonderful providers who go out and they're in the community talking to people at community events at their homes drumming up, talking about the importance of substances and connecting with people on a daily basis. And so, if you're a Cleveland Clinic provider and you see all this regulation, we want you to know that we listen and these things come to us and have really impacted our platform and made it better.

I think also you did ask me about upcoming initiatives, and it's fascinating to listen to Matt. I like to hear about the SERT team. We obviously collaborate a lot on that. We've done a good job of we use the same terms. Whereas the resources at the hospitals are different, there is a unifying process. I'm excited to hear about the next frontier is if there's heterogeneity to treatment. If you're going to have an impact on patient subpopulations, they're going to respond differently to treatment and maybe we should find those patients.

I think that's exciting. We're deploying technology to help us one, find sepsis patients and then two, hone in on some subgroups. We have work in sepsis AI and teams that are working on that. I think that probably hands it right back to Matt, so he could tell us maybe you want to tell us some about specific populations or any unique metrics you're working on.

Matthew Dettmer, MD:

Yeah, I think as we've been living since the end of February 2022, which is we've really learned a lot about workflow and how to roll the program out across the hospital and engage with different service lines.

I think one of the things we're really excited about moving forward is thinking about some of the distinct patient populations that may be having hospital onset sepsis but fall into the subcategorization of having different underlying disease states and thinking a little bit about how we're identifying those patients, if there's unique criteria that might be helpful for the patients with hepatic cirrhosis or the patient with heart failure. And then also thinking about how we're managing those patients and whether there's things that improve their outcomes that, again, might be subtle tweaks from the CMS bundle.

I think that's an area we're really excited about moving forward with in the future.

Hassan Khouli, MD:

That's great.

James, I like how you creating these incentives in a way that is a friendly competition that people really can think through that these different initiatives and then be able to follow it with that kind of engagement strategies that you have put in place there.

Closing remarks? Matt, brief closing remarks with all this good information you shared with us, please.

Matthew Dettmer, MD:

Sure. I mean, I would be remiss, and I'm sure James feels the same way, that it's very exciting to be chatting about this in this platform. But there's obviously we're part of huge teams, and there's a lot of folks that are involved in every level, as well as just the providers that are taking care for the septic patients. Too many to name, but just so many folks that I think were effective representatives of here that make sepsis care in our environment very successful.

I'd say globally take-home points I think from my standpoint is we've learned a lot about sepsis, again, in the last 25 or so years. We've, I think, done a lot of work more recently to sort of translate those lessons into good care. I think we're in a stage now where we can learn even more about how to take good care about maybe some of these unique patient populations to try to improve outcomes even further.

Hassan Khouli, MD:

Thank you, Matt.

James?

James Morrison, MD:

I think I would start at the same place: gratefulness to our providers. We've always cared about sepsis, but in particular it's been one of our top focuses for two years now. Providers' engagement I really think is what is driving our improved mortality, which has been significant. We've seen our risk adjusted mortality go down by 30 or 40% across the board, and I think it's providers recognizing this as an emergency, taking it seriously, and working together to address it.

My big picture is having a sepsis program, looking at your performance, recognizing the emergency, and then giving key things that the providers are engaged in and believe impact patient care. That they believe helps the patient live, and that being the base of your sepsis program and the rest falls into place.

I would say the same thing. I'm grateful to all our teams and the dedication to make sure that if the patient's in the ED in one state, in one city, in one country, that they're committed to a standard of care. I think I like working here because we saw the challenge and we've risen to it, and that's the take home message to me is that teams can be effective and then they can reduce sepsis mortality.

Hassan Khouli, MD:

Thank you, James. That is the last word here, so it brings us really to closure on this podcast. Thank you, Dr. Morrison, and Dr. Dettmer. Thank you everyone for listening to our podcast today. I am your host, Hassan Khouli, director of the Critical Care Center and chair of critical care medicine in the Integrated Hospital Care Institute at the Cleveland Clinic.

My guests today were Dr. James Morrison, staff physician in critical care medicine and emergency medicine and chair of the Enterprise Sepsis Committee at the Cleveland Clinic. Joined by Dr. Matt Dettmer, staff physician in critical care medicine and emergency medicine, and a co-Medical, director of the sepsis emergency response team, SERT, at the Cleveland Clinic main campus. Today we had a good discussion about sepsis. Thank you both.

James Morrison, MD:

Thanks, Hassan.

Matthew Dettmer, MD:

Thank you.