SCCM Resource Library-Pro/Con Debate: Does Postoperative Critical Care Improve Outcomes for Patients Following Major Surgery?

Video Transcription

Very good morning and warm welcome to our PROCON debate about postoperative critical care, do we need it for every patient or not?  The PRO side will be delivered by Dr. Nix from Kansas City. I'm not, my pronunciation is obviously not the best. I'm Thomas Zagmani, I work in the UK and in the UK you need  to be very sick and recoverable to get onto a critical care unit even after surgery. So I will be looking forward to hear both the pros and the cons of the debate.  What we would like to do is that once everybody is settled down we will start the talks. We will have both of the talks and after that we will have a panel discussion  and I will invite you to come to the mics and ask your questions. This is one of the sessions which will be on demand later on so please do use the mics  so we can capture the questions and the answers. And with that, I'd like to invite Dr. Nix. How much time do we have again? You've got about 25 minutes. So 25 minutes, all right.  Can you hear me? Is this mic on? I don't know which mic I'm using, I'm using some mic. Well, good morning and glad to be here.  I'm from St. Luke's Hospital in Kansas City and that is on the Missouri side, not the whole state of Kansas.  Everybody knows a little bit of humor there. Sorry, I'm putting my watch on. I'll lose track time and that's, can I move that? Is that okay? Good, so you can still hear me.  So anyway, I apologize. So I'm super happy to do this, apologize in advance. I gave my talk to my wife who is a brilliant clinical pharmacist of the day  and she kind of gave me the look for the mentally infirm who don't know they're mentally infirm. But which I get through it and hopefully maybe I'll learn something  while we're doing it. Maybe we will. That's me, St. Louis. I've been out there for a couple of years now and hopefully we're building a pretty good trauma program.  Nothing to disclose, no one's paying me.  And yes, I was kind of voluntold to do this. Apparently a speaker had some issues and couldn't do it kind of last minute and so here I am. And yes, I was up till midnight,  I think last night, look at some stuff but we'll try to get through it. Learning objectives, I feel like I should put some up there just because I've always been taught, right?  We've always been taught to do it. And of course we try to use Bloom's taxonomy and I've never really learned anything from Bloom's taxonomy, but we're gonna kind of look  at this ICU overuse problem, it's significant.  Think about some risk factors. And there's a plethora of literature out there and I apologize I couldn't get through them all but I wanted to pick some stuff  that might show big picture why we're in the ICUs or why we're being admitted in ICUs and maybe a little bit smaller and then down to the surgical factors.  So who gets admitted to the ICU? So we all know, because we're intensivists, only the priority one and two, oh, there's a typo there, isn't it? Did I fix that?  From the, I did, from the SACM guidelines, right? So priority one, you have the physiology, you're sick, you're critically ill, right?  Priority two are the folks who are potential to be that way and need close monitoring. And then the threes are the people who aren't expected  really to survive, aren't gonna do well, they're really sick and the fours shouldn't be admitted, they're not gonna make it at all. So, but show of hands.  So who only gets in their ICUs ones and twos? Right, none of us, right?  All right, it would be nice. I was just, I was asked to do this talk when I was admitting like the third or fourth patient, I just want them watched overnight kind of thing,  nothing's wrong with them. So I'm like, well, how do I give this talk? I don't want to.  And so we all get the, we need a flap check, right? At least in my house, we don't have a step down, so I'm constantly getting these flap checks.  And then I love the phrase, you know, I go, hey, Dr. Smith, that's a generic term, I'm protecting the innocent there, right? Or the guilty.  Hey, Dr. Smith, you know, this patient looks pretty good. You know, it's been three days since their flap, you think they can move out? And he goes, you know, it's the weekend.  If I keep them on Q1 hour, can I stay in the ICU, right? I mean, show of hands, anybody seen that? Yeah, a couple of us, right? Drives you crazy.  And then the other thing I think about when I'm thinking about ICU use, is I've been through where we've expanded ICUs. Most of us probably expanded ICUs along the way.  Show of hands, after you expanded your ICU, you had more bed capacity, how many actually ended up with a surplus of beds?  Okay, we got one, right? And I always equate it to kind of the field of dreams thing. You build it and they will come, right? In my last institution, we doubled our ICU capacity  and we filled them up. You know, I still didn't have it. So should they have been there? And now we're kind of, which side is he arguing for, right? I don't know.  No, I think there's a definite overuse and we got to figure that out.  So yes, there's overuse problem, over 50% of beds not needed. This is Dr. Dong, Dr. Shapiro did a big database analysis. Looked at really medical, I know this is surgical,  but it gives us some perspective on this. And then applying criteria and SSM guidelines for who should be in the ICU, 50% of the patients admitted for upper GI bleeds,  CKs, PE, things of that nature didn't need to be in the ICU, right? And these numbers are from the SSM statistics, critical care statistics page.  And $2,056 billion were spent on ICU in the United States.  And then in 2010, 108 billion, I don't have more recent and 0.72% of the GDP.  That's a lot.  And so again, you go, well, okay, which side are you arguing for? I promise I'll eventually get to my side.  And I'm going to go a little quicker. There's more information I wanna tell you about that, but we'll just keep going. But think about that. Think about if you walked in your unit  and you had 50% more beds available, right? I mean, what could you do with that? How many times do you have patients sitting in outside hospitals who need to be in the ICU  who are waiting on one of your beds? How many times do you have patients being housed in the PACU for several hours while you're trying to figure out a way to get a bed?  All right, it happens a lot. So there are patients who need to be in the ICU. And then there are patients who their surgeons just kind of want them to be in the ICU.  So definitely something worth looking at. And not to usurp her or her own article, and I actually didn't know it was my co-speaker's article before I'd referenced this,  but I thought it's worth referencing this. It's a well done study. It's also been referenced a lot in the literature.  So this is a study looked at Medicare database. She could probably go into more about database, but about 7,800 patients looked at endo AAAs, cystectomy and whipples.  What'd they find? No real benefit, no difference, no protective effect of the ICU.  But when you kind of read this study a little more, when you look at the different hospitals, there's a wide variation on who was admitted to the ICU, like endo AAA, zero to 100%.  Whipple went from 36 to 100%, or an open AAA was 39 to 100%. So you have a really heterogeneous data set. So it's hard to apply or get clinical equipoise of how I treat my patients  just based on this large data set. I thought it was interesting because I'm going to talk about it later too, was the Charleston Comorbidity Index  really didn't differ that significantly. A few parameters, whipples differed, but overall didn't differ that much in the different hospitals.  So no benefit, maybe we should just stop. We'll go have a beer or whatever your beverage is of choice.  But we can't, we gotta keep going, right? So this was an international study, seven-day prospective cohort study. They analyzed at all these recruited hospitals  from different countries around the world.  27 countries, 44,000 patients, right?  So they kind of divided low, middle, high income  countries as to how they came out. And you look at this, P-value wasn't significant. They said it was kind of a trend. I always loved that phrase is,  I first heard it, I think Craig Coopersmith told me about the corticosteroid. There was a trend towards improvement, right? And on this, there was a trend that low income  had a higher mortality rate than ICU admission. But when they looked at the total data set, overall, there was a higher mortality rate with ICU admission patients.  But what they didn't find was a protective effect of ICU. So when they drew down the data, looked at stuff, they didn't find that the ICUs had any protective effect.  And we look at high-risk patients in different countries, they didn't find an effect. When you look at different hospitals, tertiary hospitals,  community hospitals, high income countries, low income countries, high income countries always had greater ICU availability, right?  And they didn't see a difference, but then you go back and look at the data again, there is a huge heterogeneity of the data. So it's hard to say, giving that,  I mean, it's one hand and go, oh, it's done, there you go. But you really have, you can't control. Maybe some hospitals, as the authors pointed out,  maybe some hospitals really just do really complex patients, right? And so maybe if you look at that individual hospital, it's some cardiovascular tertiary center  that they do the worst of the worst, and that's what I feel like my hospital is. So if you looked at that center and you really compared apples to apples,  then you could figure out that, wait a minute, these guys are having these routine admissions and they're saving lives. But then you have this really heterogeneous data set,  it's hard to say. And even the authors said that you really had to  use caution when looking at this.  Another interesting thing is complications in high income countries versus low income countries was about 9% higher, 20 to 11%.  They had more ICU capacity and more direct ICU transfers. So again, there is some data, I think this adds to the data that patients going to ICU may have a higher mortality rate,  but does it protect? That's the question we have to answer, right?  That's a little snippet I got from a group that was just looking at where patients die, but I thought it was worth putting up there. 22% of all deaths occur in the ICU.  So we think about ICU overuse problem, we have unnecessary admissions from surgery and from medicine. And then we also have some folks that maybe they're at the end of life  and they just shouldn't be there to start with. Where do patients die in the US? Well, most folks die there.  But for our surgery patients, you got to decide, was this the best care? Kind of didn't seem like it's raising the cost. Maybe it's not. Can it free up some beds if we alter things?  Maybe, maybe. And do post-op patients need those beds? And that's the question we got to think about. So eventually, I told you, eventually,  I promise you, I eventually think about it.  So Weissman et al, they published a couple of papers at Tel Aviv, some well done statistics and they really looked at comparing elective to emergent surgeries,  which is not our role here. Our role here is really to look at elective. But if you drill down in the data, it's about 1000 patients elective, 174 emergent.  It's academic center greater than 14 years of age.  Varied things.  But then if we look at the data, they actually did have in their tables where they divided out the elective patients that went to the floor or the general ICU of some sort.  And the ICU again had higher mortality rate.  The interesting thing too is the ASA score was higher and length of service was higher for these folks. And the age was kind of in range a little bit higher, but it's hard to say.  So now we're starting to maybe develop a picture of maybe there is a group of patients that might benefit from ICU admission. We just know that these folks  who ended up in the ICU direct transfers did still have a higher mortality rate. Would the ICU have been elective? Protective, it's hard to say. We're gonna keep going.  He's telling me I have five minutes, so I'm gonna talk really fast.  Jareth et al did this study out of Ontario, looked at days alive and out of the hospital, 30 and 60 days.  Wait a minute, who am I arguing for? It still says no difference when you look at hospitals, but there's a huge variation. Some hospitals had 0.9% admission for certain procedures.  Some hospitals had over 90% admission for those same procedures. So a huge variation. But when you look at patient-centric factors, what they did find was days alive in the hospital  was decreased by increasing age, comorbidity, COPD, and surgery. And we're gonna just kind of flip through. Apologize for running late.  So, and then we looked at patients who were admitted, transferred from the floor to the ICU, no early ICU versus an early ICU admission.  So their comorbidities were lower. But look at this. The early ICU patients had more comorbidities.  They had higher rates of stroke, MI, and other things. They tend to be older. So the surgeons or the intensivist, somebody who was recognized, these patients could get in trouble.  And then could we help them? The 30-day mortality rate for early was 2.4 versus the late ICU. So if we let a patient languish on the floor, then the mortality rate was higher  than patients who initially admitted to ICU. So I still think you could equate from this that there was this group of patients who were sick, had more comorbidities, right,  that were identified and transferred to ICU, and they had a lower mortality rate than patients who maybe were on the floor. You know, it depends on where you're from.  You know, some of our floor nurses have 10 patients. Maybe they don't get to that patient for three, four hours just because they're working really hard.  All right, and we're just gonna keep going.  This is some disease-specific. This is a little study done.  400-some-odd patients got normal care. Unplanned ICU transfers, seven direct transfers, like 22. And this is a cystectomy with early recovery. And they felt it was overall safe.  But I did wanna show you this. The unplanned ICU patients had a higher rate of comorbidities. They tended to be older.  But really, the direct, again, were also had a much higher rate, 90-day re-admit. Those are patients that knew they had a tough time in surgery, did bad,  and they got re-admitted for other reasons. But still, these ICU patients tended to be older and sicker. We're also at all ovarian cancers. Look at that real quick.  The unplanned ICUs were 66 compared to 61 who stayed on the floor. BMIs were higher in the unplanned.  And then the risk factors they found, once they did a multivariate analysis, optimal debulking. If you didn't have that, there's a risk factor. Blood loss.  I know all surgeons in here think that we don't lose blood, and it's not a risk factor, but it is, and they showed it. I like that. But unplanned ICU was predicted by,  sorry, would increase your length of service, complications, re-admit, and death. Odds ratio was 2.16.  Craniotomies, I think this was done in England. Said, can you admit them right to the floor?  Sure. Only 2% went to the ICU. 10% had MET calls or rapid response. But again, I just wanna show here, and they had no mortalities. It was a smaller study,  but they showed that they were all older. The patients who ended up having to emergently go into the ICU were older. Bariatric patients, big, large, infirmal surgery,  ICU versus no ICU. ICU was older. Death rate was higher. And if you had an emergent transfer ICU, you had an increased risk. This is a systematic review,  and I'll share this with anybody. I got some references at the end. And so they look at factors for unplanned ICU admission.  They looked at 37,000 journals. They whittled it down just to 44 studies due to size or conference abstracts or things of that nature. Once they whittled all this thing down,  they had 44 studies, which is a lot to exclude. And some of the things they found were age,  predicted transfer, unplanned transfer ICU, ASA status, and the risk of surgery and anemia. They didn't find a difference in mortality, which kind of plays into my opponent's hand,  but they still identified some risk factors. And in my book, it's always better to have a planned ICU transfer than that unplanned ICU transfer.  As a medical director, I'm always looking at those cases. Why did they end up in the ICU accidentally, so to speak? There's more information than that,  but I'm gonna keep moving, just get this. This is the National Association, National Anesthetic Clinical Outcomes Registry. This was published, shoot, it was just published, right?  Yeah, I'm gonna go back. Whoop, all right. So, what they identified, just looking at this, 23 million records and data points, and there were two million records they looked at.  What they identified as risk factors for unplanned ICU admission was age, ASA status, as you see there,  and then mortality rate. So if you had an unplanned ICU transfer, your odds ratio for mortality was 17.63,  and two biggest factors were age, or the biggest factors were age, ASA, and surgery duration for that. So, increased age, difficulty of surgery, surgery duration, blood loss,  incomplete tumor, dual blocking, things that lead, or BMI, things that lead to more difficult surgery, the comorbid conditions,  and it'll lead to ICU transfers, and unplanned is associated with a higher mortality rate. But what I really wanna do is, why are we here, what do we do,  we're intensivists, and so who cares? If you go to the ICU, if we can't help you, it doesn't matter. And there's a couple little studies that he's done out of Korea,  and this one really showed that the closed versus open unit decreased mortality rate once they started a closed unit with the intensivists. This one was a standard of care,  but they put an intensivist in the ICU, and just having the intensivist there did have a mortality benefit. And I guess we're taking questions at the end, correct?  All right, thank you very much. You gave me 18 minutes.  You want your notes?  Thanks very much, and now I would like to invite Hannah Wunsch from Sunnybrook, Toronto, to give the con version. All right, thank you. My mic's on, great. So, thank you very much.  I feel like some of my work has been done for me, which is always fantastic. I was given the con side about, specifically around routine ICU admission for patients after surgery,  and it's a topic I've thought a lot about. As you heard, I've published a bit about it. I'm gonna take you a little bit more into the weeds of that study that was mentioned.  I should just mention, I don't have any conflicts of interest. I am an associate editor at the Blue Journal, and I have some funding for unrelated research  from a number of organizations. So, the crux of it. I don't know about you, but this is me rounding on a Whipple patient, right? So, all those routine post-op patients  who come to the ICU, my rounding is basically smiling and waving at the end of the bed and saying, good to see ya, before moving on to the next patient.  So, I feel like right there is the entire talk in terms of what we routinely have in terms of value added for our patients. But I wanna go over three questions.  First, just to kind of quickly define what we're talking about when we talk about things like major surgery. What is critical care when we're talking about that?  And then finally, the kind of the main issue, does it help to send everyone to critical care after major surgery?  This is just an example of a paper. There are many out there that look at populations that are dubbed major surgery, and I'm highlighting it because it's a common way  of describing major surgery patients. It's patients who were admitted to a SICU for at least 24 hours. And there's a chicken and egg issue here, right?  That we are defining our high-risk patients by someone's decision that it's a high-risk patient and they should go to the ICU. And so, I would urge you to be very wary  of scores that predict admission to ICU or sort of studies that use this as their measure because fundamentally, you're predicting what some physician or nurse or someone out there  has decided is the right thing to do for that patient, not that intensive care is actually going to add value or prevent any complications or do anything for them.  So that's one way that it gets defined. The other way is expected risk of death, right? And we actually know this. We've got pretty good data now for most surgeries  about what the short-term risk of death looks like. And probably one of the most important ones to come out that really tried to define  for different surgical procedures was this one, development of a list of high-risk operations for patients 65 years and older. And you think, great, this will tell me  which surgeries I should really focus on. Their definition for high-risk was 1% mortality. So that's lower than the average mortality for patients in the hospital, right?  It's pretty low. And they also ended up with 264 procedures, an enormous number of procedures, an enormous number of patients. So if you take that as your cutoff,  you wouldn't need to turn your entire hospital into an ICU at all times, right, to just care for those patients. And so there's also this fundamental tension  around where's our cutoff in terms of deciding that based on our baseline information about people that they are actually, quote, high-risk or having major surgery.  Okay, quickly, critical care. I was giving you a very quick definition of what I think of when I think of critical care because this comes up a lot in talks I give.  How are you defining it? So care in an ICU bed staffed appropriately care for the care of the sickest patients in the hospital. Pretty vague, right?  But I think we all kind of know what we mean by it. And I've just provided a list of things that we traditionally have when we think about ICU.  Ability plus minus to do invasive mechanical ventilation, to provide invasive monitoring, continuous infusions, high nurse-to-patient ratio,  and I'm gonna come back to that at the end. And then I like to think presence of an intensivist, and I'm going to ignore the issue of open ICUs.  What are the downsides of ICU? Obviously, it's very simple. Limited resource and unbelievably expensive relative to other options for care of patients. And this is a universal.  Any country you go to is grappling this problem of the fact that ICU beds are expensive. So how do we decide to triage to the ICU? Well, you can break this down into three things.  Availability, need, and desire of a patient to be there. And I would argue that for surgical patients, we're down to availability and need because we've identified the desire  before they go in for surgery that this is someone who's willing to go to the ICU if they have a problem or they may be anticipating going.  I have an entire talk on availability of resources, ICU resources, and how much that drives the perception of the need for intensive care and the appropriateness of intensive care.  But I wanna get to the crux of this issue, which is does more intensive care help?  This is a study that Rupert Pierce did now a number of years ago that was published in The Lancet in 2012. It looked at variation across countries in Europe  who were across a whole range of surgical procedures and found a fair amount in variability, although I wanna caution you that most of these cross one  in terms of their odds ratios across different countries. But one of the key findings that was noted was that the overall mortality for this population  across all these countries in Europe was about twice what they expected. It was 4% and they had anticipated 2%. And one of his observations was that 73% of the patients  who died were never admitted to an ICU. And so this really raised that question of if some of these countries or hospitals more routinely admitted patients to the ICU,  maybe that was the explanation for this high mortality. And this is this concept of rescue or failure to rescue. And what this is showing you across hospitals in the US  divided by the across quartile, sorry, quintiles of hospital mortality that these complication rates were the same across these hospitals,  but the death after major complications was different. And the argument again here is if you stick intensive care into the middle of that,  that that's the explanation for this differing mortality, this ability to rescue patients faster and better.  However, Rupert Pierce went on to try to answer this question, as was mentioned, looking at the data and really looking by hospital at their use of intensive care for their patients  to see if there was any relationship with the mortality. And what this kind of icky looking scatter plot shows you is on the X axis, the percentage of patients admitted to the ICU  and on the Y axis, the risk adjusted mortality for all the hospitals that participated in that study. And the basic answer was this,  at hospital level, there was no association between mortality and critical care admission directly after surgery, critical care admission to treat complications,  or provision of critical care beds. Could not find a signal for benefit with more intensive care. As was mentioned, this was work we did using Medicare data,  so more relevant to the US in terms of population UC. And what we did differently here was that one of the challenges with that study done in Europe  was the heterogeneity of the population. So we said, okay, let's get more specific here because we think in procedures, right? You don't sort of say, I've got a patient  having major surgery who needs to come to the ICU. You say, I have a patient who's having an open triple A who needs to come to the ICU, right?  You're very generally specific in your language around that. So we chose five procedures that have a range of mortality associated with them.  From esophagectomy, PD is Whipple, pancreatic otoadenectomy, cystectomy, open triple A, and endovascular triple A. And endovascular triple A has a mortality  about 1% open triple A, and then you wear for sort of 10 to 30% depending on where you are. And so what this is showing you is just the variability in use of intensive care.  And that bottom gray is the hospitals that admitted less than 50% of the patients to the ICU. So what I call really discretionary admission.  And that top yellow is the number of hospitals that admitted more than 90%, basically 100% of their patients to the ICU. So what I consider routine or mandatory admission.  Someone's having a Whipple, it's on the schedule that they're going to the ICU afterwards, whatever happens, whoever they are, right? And you can see there's a lot of hospitals  where that is the case across every one of these surgical procedures.  So it's important to note, if you look at the patient level,  the going to the ICU and outcomes, what you see is an incredible signal for harm associated with ICU, which of course is not the reality.  What this is telling you is across the board, risk adjusted for comorbidities, age, all of that, is that we're pretty good at identifying the patients  who are truly high risk of death, because we tend to send them to the ICU no matter what. We also try to do this with propensity matching,  same thing, big signal for increased mortality going to the ICU. But then when we flipped it around and said, okay, let's look at just all the patients in that hospital,  whether they went to the ICU or not, and compare across hospitals, those that use that discretionary admission versus those that use that mandatory admission,  and there was absolutely no signal for benefit, for reducing mortality in those hospitals that were really aggressive in sending everybody to the ICU.  Now, in an ideal world, we would randomize people to go to the ICU or not, that would answer our question. It hasn't been done in a surgical population,  I would have told you it can't be done at all ever, but then the French came along and they did this. They actually randomized hospitals  to a more aggressive approach to sending patients to the ICU from the emergency room versus not. So the caveat here, this is not surgical patients,  but it's a really interesting example of seeing what do you get in terms of value added of sending more patients to the ICU. So they had their normal approach,  patients would be assessed for need for intensive care, and then they would go to the ICU or not, and then in the hospitals that were randomized to a more aggressive approach,  they really said, you know, really try to get them there, really kind of, if you have any concern, they need to go to the ICU, send them to the ICU, and they got a great split.  They had hospitals that had their routine approach, sent about 30% of the patients they evaluated to the ICU, and the hospitals that use this more aggressive approach  sent 60% of the patients to the ICU. So a really nice difference in terms of saying, okay, if we're gonna see a signal for benefit, we should see it here.  And here's the mortality curves, right? Absolutely no benefit out to six months, and in fact, in the short term, higher mortality in the more aggressive arm,  which I wouldn't say is a good thing or a bad thing, but it certainly does not give you evidence that that more aggressive intensive care is providing benefit.  And I think it gets back to the question of what are you able to do for people in an ICU? And for many people, it's not a lot if they already have some problem,  like a stroke or something, that's going to kind of play out, however it's gonna play out. You're not going to change their trajectory through that.  So I sort of think about it in terms, and this was alluded to by my colleague before this, kind of the tails on the curve, right? You've got those very sick ones  who you kind of know aren't gonna do well, but I think the bigger and more important one that's really that major surgery population is this not-so-sick tail.  And I like to think about it in terms of number needed to treat for ICU benefit, right? We think about this in terms of taking aspirin or giving a medication.  We don't think about it in terms of admission to intensive care. This is an old study done by Jack Zimmerman and Andrew Kramer years ago, looking at patients who may or may not need  to be in the ICU. And what they had in these data was information on admission to ICU, were they getting active treatment or were they being there for monitoring, right?  And then they also had information about who was there for monitoring went on to require some sort of intervention that we associate with intensive care,  intubation, vasopressors, whatever it was. So initially on admission, 60% were in the active treatment group and 40% were in the monitoring group.  Again, caveat here, this is not just surgical patients, this is everybody.  And what they did is they took those patients who were in the monitoring group and they split them into what they called high-risk versus low-risk monitoring.  And then looked at the percent who actively received an intervention. So in their high-risk monitoring group, about 20% received an intervention. And you can think about that  as a number needed to treat of five. You have to admit five of those monitoring patients to the ICU for you to then have to do something other than wave from the end of the bed.  In their low-risk monitoring group, only about 5% received active intervention. That's a number needed to treat of 20 admissions to your ICU before you're doing something  that we consider sort of benefit of intensive care. And I put benefit sort of with an asterisk because we don't even know if it ultimately  really makes a difference of doing those things. So I think that this is a really helpful way to think about it, right? So we could put everybody post-op in the ICU  who needs that monitoring, but we really need to think about it in terms of where is our cutoff? Is it a number needed to treat of 10, of 50,  or that 1% mortality I showed you in the beginning constituting major surgery? That would be a number needed to treat of 100 to potentially prevent.  And there's that caveat that it's potentially able to prevent that one death. And then finally, the sort of issue of, if you build it, they will come.  Just to highlight this, this is not a parking lot. This is a highway in China before coronavirus because it's probably empty now. But it's, anyone who does traffic theory  can tell you that there's this fallacy, the idea that if you've got a jammed up highway, just build more lanes, right? And that will solve the problem. In fact, it doesn't.  More people start to drive, more people take that route, and you end up just as jammed up as you were before, looking like a parking lot now. And there's some really neat studies  that have teased this out in intensive care. So this is work by Joan Tenno looking at patients admitted to hospitals with advanced dementia.  And she tracked hospitals in terms of their building of intensive care beds over a time period to ask the question, in hospitals that start out at some certain baseline rate  of providing mechanical ventilation for patients with advanced dementia, which I think most of us try to avoid if we can, and then looking to see whether that rate changed  as they added more ICU beds, so that there was greater availability of beds. And sure enough, there was an increased odds of mechanical ventilation per 10 ICU beds.  And interesting as sort of a negative comparison for those places that increased step-down beds that obviously can't provide invasive mechanical ventilation,  there was no change in the rates. So just that caveat that sort of, we can provide this service, but it's not clear what value added we have,  and we risk just needing more and more and more and getting into that vicious cycle.  Want to highlight, of course, you've got to have other options for care. If you're saying don't send them to the ICU, PACU, recovery room, observation unit, ward,  I think the key and the thing that makes it very hard in some places for us to do this is this, the nursing. Of all the things I highlighted about what intensive care provides,  I would argue that 99% of the time, it's the quality nursing and the ability to call for help and intervene if there is an issue  rather than everything else that's there in the unit. And so there was this editorial number of years ago now, we need more nurses. And I would argue that, as mentioned,  sometimes nurses are with 10 patients on the ward. That is obviously not conducive to good quality care of your post-operative patients.  But I think that that's really where we should be focusing in terms of trying to minimize the de-skilling of our nurses and teams on the wards and in other locations,  rather than just building more beds that are equipped to provide invasive mechanical ventilation and ECMO for everybody who ultimately isn't gonna need it  sitting there in those beds. So in conclusion, post-operative monitoring in an ICU has not been proven to help. We now have a number of studies that really show this.  Particularly for low-risk surgeries, it's clearly expensive and inefficient. I don't know about you, but my patients get stuck in the ICU, unable to get to the ward,  and I end up discharging them home from the ICU, feeling very nervous because I have no ability to really know whether they're gonna be safe when they get home  because that's not what we're trained to do. And I think alternative care pathways with good nursing are key, and the alternative care pathways really is key  in terms of making sure we have safe, good care of our post-operative patients. So I will stop there, and I think we have time for questions from both of us. Thank you. Thank you.  Thank you very much for the two fascinating talks,  and they are open for questions. Mike Russell from WVU, and you sort of alluded to part of this in your closing remarks. We had a problem. It was a legitimate problem.  We were having unplanned admissions for our geriatric orthopedic patients at an alarming rate, and we solved that problem.  But the definition of the benefit of intensive care is not,  we lowered our readmission rates by admitting everybody to the ICU on the front end. So, of course, the number went down, and the fundamental problem, in fact, was floor-based care.  Now, that's a pretty tough nut to crack and requires a lot of system changes to improve that, but you talked about definitions of sick patients.  Anybody admitted to the ICU in the first 24 hours is by definition sick. Is there any sort of consensus building about those definitions  and what we actually should be trying to target? Readmissions are expensive, and they're lethal, so something needs to be done to prevent that problem,  but I suspect it has to be done outside the ICU. So, if I understand correctly, the question is that how can we identify the high-risk patients?  So, in the one, and I apologize for going fast, and the slides are gonna be available, but the one meta-analysis they did, they looked at risk factors.  The NACOR, they looked at risk factors. There's news, there's modified news, there's, what else, modified frailty. There's a number of them.  I don't think there's anyone that's perfect, as you point out, but, yeah, I think a good clinical judgment, you can identify some patients that probably need some extra care,  and it depends on the capability of your hospital. If you've got a good step-down unit, if you have a good nursing ratio,  it's great, but if you've got a nurse that's seeing one, 10, 12 patients taking care of them in the ward all night, maybe not, I mean, what do you think?  Yeah, I think one of the challenges of talking about surgical patients is this issue that it gets very specific in terms of care needs and different risks,  depending on which surgery we're talking about, right? And that's always, for me, one of the, not frustrations, but challenges of trying to think about this topic,  because if you talk about carotid endarterectomies versus hepatectomies, right, we're already talking very, very different care needs, populations, and we do have a tendency  to lump them all together in trying to think about this. So it often feels like you need to almost sort of chip away at it, one surgical procedure at a time,  figuring out who are the high-risk patients within that surgical procedure, right, because the surgery itself at that point isn't giving you that answer.  The frailer patients, the patients who have a huge blood loss, all of those things, to then create your care pathways that are siphoning off the truly high-risk into the ICU.  So, you know, I certainly would never say we shouldn't ever admit a carotid endarterectomy to the ICU after surgery. Certainly, there are those patients,  for one reason or another, who absolutely belong there, and I appreciate your point that sort of trying to avoid readmissions is the goal.  We just have to be careful we don't end up with the, just keep them in the ICU forever, and then we avoid the readmission problem, right?  You know, the kind of, the usual gaming of the system that can go on that doesn't really solve the problem. So yeah, bit of hand-waving, I'm afraid.  No real answers there, but I think your point about the ward and changing things on the wards and empowering nurses there, empowering teams there,  having rapid response teams or others who can help when there's a problem is important. I always sort of say that buried in those data about sort of the French data showing you  that the randomized control trial didn't make a difference,  I am sure there is an anecdote or a patient and a family and a physician who can point to someone and say the fact that they went to the ICU saved their life, right?  You know, it's always there in any data whether or not you can see it. So I don't wanna minimize that because for an individual, for a family, for a patient,  and the team taking care of them, that is extremely important, of course, but we're not gonna see that necessarily in the data.  Please, sir. Hi, my name's Greg Halgen. Actually, I'm the department head in Surrey, British Columbia, and one of the struggles as we're developing a research program  around PICS specifically, when I look at that data, we know there's a benefit to monitoring. We know there's a benefit to intervention.  We know there's a benefit to a rapid response. In other words, failure to rescue, particularly in surgical patients, leads to bad outcomes.  When I look at that data, what I see is a harm matching the potential for benefit, and reflecting on the harm of critical care, it's a lack of mobility, disruption of day and night  routines, central line infections. In your views of the literature, is there any assessment of specifically those harms? Because it may not be that we're actually, you know,  just, you know, increasing our mortality from being in the ICU, but rather harming our patients with our interventions, and we could maybe  see a mortality benefit from mobilizing patients like our surgical colleagues, maybe? Yeah, no, it's a great question. I've never been able to really get at that,  but I agree, your list of things, you know, I always say, would you rather, do you really want your grandmother next to the person with C. diff on three vasopressors, and, you know,  on the other side with influenza pneumonia, or would you rather have her, you know, in a quiet ward somewhere where she's, no, you're not gonna have people coming in  and out of the room all the time. Yeah, no, I agree, I think that we underestimate the negatives of intensive care  in terms of the potential impact, and, you know, as you've said, I think people have gotten better at things like mobilizing, right, recognizing that that has been a downside  of intensive care, and we can change something like that, but the ability that, at least in our ICU, like, they can't go to the bathroom anywhere, right,  they're stuck using a bedpan or something by the side of the bed, it's dehumanizing in that way, they're hooked up to all kinds of things, they're at risk for infection,  the urge to sort of put more lines in people because we can, and I do think that we're changing that culture slowly with the less is more approach,  but there are fundamental aspects of intensive care that don't change, and the sort of, the physical plant is one of them. And the Dong and Shapiro data show that,  that harm, but there wasn't an increase in mortality with it, but they did show more procedures and higher costs, and that was there,  I think, in your paper, you looked at the cost too, was it yours? We did, yes. Yeah, looked at cost, but only on the endotriple A did you actually see an increased cost.  Right, not much difference. And there's plenty of data, you know, in any of the papers there wasn't any increase in mortality because you went to the ICU,  but you certainly have hospital-associated badness that can happen, certainly true. Yeah, I'm just looking at the fact that there's no benefit in mortality,  and yet we know that there is. Well, there's no benefit in mortality in total populations, guaranteed. But, you know, I think it's pretty clear  that unplanned ICU admissions has an increase in mortality, and so there would be a benefit if we can find those patients, and that's the hard thing,  find those patients and avoid unplanned ICU admissions I think is definitely a benefit. I can't tell you how to do that, so I'm kind of arguing both sides of the line, all right?  We just submitted a patient paper where we looked at, we built an HAU that is two-to-one nursing and private rooms, private bathrooms,  and that is, since we've instituted that, we're actually, we've reduced our ICU mortality as a result, and we can show that by flow in and flow out of the ICU,  but that doesn't translate to the ward, and it doesn't really answer the question that you're asking. Yeah, there was a paper out of New Zealand 10 years ago,  I think it was, where they said the best optimal care was two nurses, 24 hours a day, and you had the best outcomes, but that's just not the real world, right?  There's another question? Thank you for the presentation, I enjoyed it. My question is, when you think about admitting post-surgical patients to the ICU,  most of those operations like Whipple and AAA were anticipating physiologic derangement, so that's the impetus to get an ICU bed in advance.  However, as you know, one vascular surgeon may do an open AAA in 90 minutes with 200 cc's of blood loss, and another vascular surgeon does a 10-hour open AAA  with four liters of blood loss, and both of those patients have profound, different physiologic status at the end of the case. So when you group the patients by operation,  you get a wide variability. So maybe one of the criteria for admission to the ICU is not the operation, but it's the physiologic status  at the end of the case. So admission criteria to the ICU would be a pH less than 7.30, or a CO2 greater than 55. Instead of the operation, it becomes their status.  You know, in terms of the physiology, that would give us a more solid framework to decide if the patients coming to the ICU are really sick and need resuscitation,  or if the patient is stable enough to go to the floor. I'm just interested in your comments on that.  Well, in most of the papers I read, those direct ICU transfers from the OR were kind of lumped in together.  But, you know, they did, even in the ovarian cancer, you know, increased blood loss, you know, was the indication, or actually most of the surgeries,  they actually, you know, analyzed that as increased risk for unplanned and increased risk for mortality, like increased blood loss, more difficult surgeries.  And so, no, that's, and that's a thing. He's not in the room, I don't think.  But Pierre Cobb, you know, he told me that one night, I was asleep, and he woke me up, and he said, could you join me by this bedside, which is kind of weird, I thought,  and I knew I was in trouble, but the thing was this, I was resident, and the thing was this, he said, you know, the patients deserve  the physician or the practitioner bedside to evaluate them, and that's what it is, you gotta do the right thing at the right time, like Armando was a couple years ago.  You're right, you gotta evaluate the patient with the clinical parameters and good clinical judgment and make that decision. That's why everybody going to ICU after surgery  is a bad idea, evaluating them, making them go to the ICU is a good idea, I think. I don't know, what do you think?  I think the first step is to get away from this, you've had this one surgery, you're going to the ICU, which, as you can see, is still very common  in terms of our approach, and with limited ICU beds, we really kind of need to stop. How we then determine who's at risk? Well, it turns out we're pretty good,  even without fancy machine learning to tell us who's high risk. You know, just from the data we have, we can see that we're, just showing you  how high the risk of death was amongst the patients who ended up in the ICU, actually tells you, as humans were good, and the head-to-head comparisons  of humans versus sort of scores and things generally comes out in favor of clinicians making good decisions at the end of case, knowing your pH is low, knowing you may not be able  to extubate, whatever it is. However, can that ultimately be improved in the long run by additional information, you know, all the pre-op information added  to everything that's gone on inter-op? You know, 10, 20 years from now, we're probably going to have those scores to assist us to sort of flag and say, high risk for death.  I think the problem is a lot of people try to create those models using intensive care as the outcome, and then we're back at square one because that doesn't tell you whether or not  they're actually high risk, that again, just tells you that somebody decided they're high risk. So I think we just have to be careful  as we start to go that route, and we are going that route with machine learning to, there was just a paper in JAMA, right, a day or two ago, looking at trying to change  amount of hypotension, whether or not ultimately that makes any difference, time will tell. But I agree, I think that that's where we're headed.  And I think we've got time for one last question. Thank you, appreciate it. Bill Hartford from the University of Cincinnati.  We talk about dichotomous interventions, ICU versus not, and you generally have been talking about dichotomous outcomes, dead or not.  Both are fairly easy to define, but as we all know,  life is more than a dichotomy, and our patients are very different.  But we've also talked a little bit about design, administrative structure.  How do we make our hospitals more fluid and our care more fluid? So what's wrong with going by and waving to a Whipple? It's fun, they'll talk back to you.  So they don't need that care. But in the past, we sort of thought about universal bed concepts, changes in status, all that. How are you addressing making your care  and your administration and your physical space more fluid to provide just the amount of care that patients need?  So it's a great question.  I think the problem with the universal bed model really is things like the amount of nursing and the geographic location issues that you get into. And so it sounds great in theory  when you just take a single bed and say, great, we can scale up and we can scale down and they can get whatever they need in any bed in the hospital,  whether they need invasive mechanical ventilation or just oxygen. But it's just not a model that I think works in reality when you say, well, okay, then we're scaling  from someone who only needs a nurse to check on them once every four hours to someone who needs a nurse bedside 24 seven. And the reality is of the nurses saying,  we don't work, we can't work that way. We can't always be on call to kind of come in and add in as needed. So those are the challenges. Also, I work in a hospital in Toronto  that is very constrained in terms of resources. So trying to make any changes to any beds, we still have magic curtains between our beds in the ICU.  We don't even have separate rooms for most of our patients. So we're starting from the basics, but it is trying to create more in between levels of care.  That I think is the big one is to say, we know not everybody needs the highest intensive care. We don't need to use those beds for those patients.  We don't need to have a one-to-one nursing for those patients. And so let's kind of downgrade those patients to a level that does seem like a better match,  maybe not ward in terms of the nurse patient ratios, but we now have what we call level one for things like carotid endarterectomies, flap checks,  all the things that need good nursing, but don't need me swinging by every 30 seconds to check on them. And so that's been the little piece that we've added in.  I don't know if you've been with more resources in this hospital to be able to do more. Yeah, I'm building a new, it's been there,  but I'm revamping, remodeling the trauma ACS program and critical care. And essentially, we're trying to tackle each one at a time with a lot of physician education.  But the problem I have, the biggest problem we have, one of our other intensive spec room, is that the administration is loathe to tell some of these private docs,  hey, I know you always put these patients in ICU for observation for two days after surgery, but they don't need to do it. And so that's the battle I'm fighting right now,  just getting the authority to say, hey, even though it's a closed unit, they still just admit with impunity,  they can't tell me this was coming up, they don't ask, hey, what do you think? So me having a new trauma ACS program there is one I've been able to have intensivists there  more hours of the day, so we can help evaluate patients and pack you.  And then I'm kind of doing it physician by physician, trying to get that education to the physicians and docs. I hate to, I apologize, but I'm late for my next thing. All right.  All right, all right. Thank you very much for both our speakers. All right, bye. Thank you.

Video Summary

In this debate on postoperative critical care, the two speakers present contrasting views on the benefits of routine ICU admission for every patient after surgery. Dr. Nix argues in favor of routine ICU admission, highlighting the overuse of ICU beds and the potential to identify high-risk patients who can benefit from intensive care. He asserts that there is evidence of a higher mortality rate for patients who are not admitted to the ICU when they should be. Dr. Wrench, on the other hand, argues against routine ICU admission, pointing out that postoperative monitoring and care in an ICU has not been proven to improve outcomes. She emphasizes the limited resources and high costs associated with ICU care, and suggests that alternative care pathways with adequate nursing support may be more effective and efficient for postoperative patients. The speakers also discuss the challenge of identifying high-risk patients and the need for more specific criteria beyond the type of surgery performed. They highlight the importance of good clinical judgment and evaluation of the patient's physiological status at the end of surgery. Overall, this debate raises important questions about the appropriate use of critical care resources for postoperative patients and the potential for alternative care pathways.

Asset Subtitle

Quality and Patient Safety, Procedures, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Discuss the effect of critical care admission on patient outcomes in the perioperative period
Analyze the different organizational structures to support optimal care
Evaluate the financial benefit of routine critical care admissions following major surgical procedures
Examine the possibilities of value-driven protocolized care in the perioperative setting in and outside of the critical care unit

Pro: Routine Admission to Critical Care Following Major Surgery Improves Patient Outcomes

Sean A. Nix

Con: Routine Admission to Critical Care Following Major Surgery Does Not Improve Patient Outcomes

Hannah Wunsch

Meta Tag

Content Type Presentation

Knowledge Area Quality and Patient Safety

Knowledge Area Procedures

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Surgery

Tag Mortality

Year 2020

Keywords

postoperative critical care

routine ICU admission

high-risk patients

intensive care

mortality rate

postoperative monitoring

care pathways

nursing support

alternative care

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.