Hi everybody, my name is Shekhar Gamande, I'm from Baylor, Scott and White in Temple. Welcome to the session on advanced topics in non-pulmonary critical care. We want to thank the ACCP and the CHEST and the Critical Care Network for inviting us for this session. This is going to be an exciting session where we will talk about high-yield topics and literature review in a very brief time. We have a distinguished faculty here, Dr. Tim Crozier from Australia from Monash University, Dr. Anshul Kapoor from the Cleveland Clinic, and Dr. Mark Warner from University of Texas in Houston. I just want to clarify that the app indicated the last one was going to be on neurocritical care, and that's not true, it's going to be on endocrine emergencies. The neurocritical care truck is going to be on Wednesday at 7.15. So I think with that, we'll get started. We're going to have Dr. Crozier talk on obstetrics emergencies in ICU. Okay, thank you. All right, so there's lots to get through, so I'll just go quickly. I don't have any disclosures. And objectives, I'm just going to use a case study today, and it's going to go fairly quickly, because as I said, there's lots to get through. So challenges in obstetrics, especially for general critical care practitioners who don't see a lot of maternity patients. There's the physiology, which proves challenging for diagnostic and management purposes. Pregnancy-specific conditions, which are, let's face it, pretty rare, and even rarer to come to intensive care. Chronic illness and pregnancy, which we're not going to touch on. That could be anything, and in the modern world, it's becoming more and more common. And then there's the fact that you have more than one patient, often, in the antepartum setting. So today, we're going to talk really about physiology and the fact that there's two people in the bed. So, quick case study. Young for intensive care purposes, but rather older from a maternity point of view. High BMI, first pregnancy with ART, 32 weeks, presents with pleuritic chest pain, cough, and breathlessness. 60% oxygen, sat in the low 90s, and relatively stable hemodynamics. So, is it similar? Well, yes, it is, and that's sort of stately obvious, but there are some other things that you need to think about. So, pregnancy's a procoagulant state, so you must think about thromboembolic disease. Over the last 15 years, we've had a couple of pandemics, swine flu, then COVID, and you always think about sort of viral pneumonitis. Cardiac dysfunction, both in the hypertensive disorders of pregnancy, peripartum cardiomyopathy, and in women with no antenatal care, especially you might see women with undiagnosed heart disease, especially valvular heart disease. You've got to think about your radiology, think about your anti-infective choice. And you've got to remember that the utero-placental circulation doesn't auto-regulate, and so you need to be conscious of cardiac output and blood pressure. Fetus tolerates all the stuff that we don't tolerate well, even worse than we don't. If you're seeking signs of end-organ dysfunction, so if you look at mentation, urine output, et cetera, you have to include a utero-placental or fetal assessment, as well as the usual stuff. And I've come full circle on this, but I'm now absolutely firmly of the view that you must have an obstetrician directly involved from moment dot. That's MFM, obstetrics, whatever you want to choose. Some people used to say you could get away without it. I do not believe that's true, and you need someone in your hospital all the time. And if you don't have them, I think the patient should probably go as soon as possible to a place with them. So, bearing all of that in mind, can I do what I normally would do, is probably the big question. So, should I perform diagnostic radiology? The answer is yes, you should. If you have a question that needs answering for the safety of the mother, and remember the safety of the mother as well as the safety of both patients, you should answer it. And there's some stuff there, just talking about radiation effects and stuff, but the short answer is yes. So, in Our Lady, we do a chest X-ray, we do a CT pulmonary angiogram. There's just multi-lobe anemonia. Fortunately, there's no PE, and the echo, for the sake of simplicity, we just made everything normal, okay? Blood gas, and so, it's not a complete gas, obviously. Forgive me, because I'm not used to working in KPA, but sort of CO2, oxygen, pretty hypoxic, bicarb's a bit low, you'd expect that. The CO2 is normal, but probably a bit higher than you'd expect, and you'll remember the physiological changes of pregnancy later in pregnancy, so that's interesting. Pretty hypoxic, you know, PF ratio of 100, or PAF ratio of 100, and metabolic state, there's missing too many values to make a big judgment. So, is this safe, and is this any safer or any less safe than an equivalent non-pregnant 38-year-old woman with multi-lobe anemonia? The CO2 level's high, but it's normal, but it's high. Now, is that a sign of ventilatory failure, or is it a sign that you did the gas at the wrong time? What are you worried about, is someone getting tired? You want much more information on the metabolic state, and then the big question is, should I intubate? Now, obviously, there's not enough information there, but do you wanna get stuck on non-invasive in these women? No. Do you wanna leave it too late? No, you don't. And so, if you're considering intubating, you must respect the obstetric airway. If anyone only remembers one thing from this, even when there's no pointers of difficulty, and you think, oh, this looks fine, it's a young, slim woman, it's a high incidence of not only difficult airway, but unexpected difficult airway, okay? The hormonal changes that were talked about yesterday, you might need a smaller tube, there's lots of edema. We had some interesting stuff yesterday on malampate grade actually changing intrapartum, so changing from one to three while women are laboring, which is not the case here, but it's interesting anyway. High BMI makes it harder, and when you render these women apneic, when you induce them, they desaturate very quickly, okay? Now, there are guidelines around. This is from the Obstetric and Genesis Association in the UK. There's three different ones here that master algorithm, but going into it, you just need to think hard. A, should I be doing it? B, am I the right person to be doing it? And C, can I get some help here to do it? Okay, because if it goes bad, it goes bad really quickly, and it's not very pleasant for anyone. All right, then you ventilate in. There's lots and lots of potential things that you think about. All the stuff that you do automatically, normally, that you think, oh, can I do this if this woman's 30-odd weeks pregnant? And we're not gonna talk about all of these. There's not enough time, but generally, how should I ventilate? There's minimal evidence regarding modes and pressures. In practice, it's essentially the same as the non-pregnant patient. I'll show you some data in a second. Sedation and paralysis worries everyone. It worries all of us, because there's very little data. Most people follow their normal practice around the world, as far as we can tell, and there's no data that things are particularly harmful outside of what you do normally. Okay, and oxygenation, targets of 70 have been suggested. That can be generous and hard to get to, so you'd have to really adapt to where you are. Okay, now this is Indian data from the pre-COVID period, looking at pregnant ARDS and non-pregnant ARDS. There's some suggestion here that the plateau pressure's higher and the driving pressure's higher. Compliance is unchanged, and if you look at the pregnant later, their PF ratio was worse. But this is difficult to really judge. This is an isolated set of data. There's also no data on weight, which I think can affect both of these, too. And this is very recently, looking at COVID. This is the study that Daniela did in Argentina, and they thought the lung mechanics was very similar in pregnant and un-pregnant patients, at least with COVID. Okay, so you've all seen this before. This is from 2018. There's a sort of nice little algorithm on moderate, severe ARDS, and you think, can I do the same things I do there? Well, yes, and yes, but. So there's not a lot of evidence. There's lots and lots of strong, strong recommendations out there about mimicking the respiratory alkalosis of pregnancy. It says as soon as you put the tube in, bang, you've gotta ventilate down to this. That's not what we really do, and at least observationally, we've shown it's not aggressively pursued. Wrecking the mother's lungs to get that is not justifiable, is the best thing I can say about that, and you should not do it. There doesn't seem to be any need to do it, and having someone getting chronic, severe lung disease after you've ruined their lungs is not good, okay? If you need permissive hypercapnia, there's some suggestion that if you keep it under 60, it's probably okay, and that's quite an old quote there from nearly 10 years ago, but really should try, and if you need to get it up to protect the maternal lungs, it's probably okay. Now the blood pressure's sagging. Now what do you do? Now optimize volume state and cardiac output. Optimize is not a helpful word. I'm aware of that, so what does that mean? Look, you could do a whole talk on this itself, except to say I think it's important that given the utero-placental circulation in the fetus, it's very reliant on the cardiac output and the volume state, okay, so you need to try and get it right. There's been a lot of angst over the years about using pressors and inotropes in these women. You know, you're gonna cut off all the blood to the fetus, et cetera, et cetera, so getting the volume state is right, but pouring in fluid after fluid just blindly is not a good idea either, so I'll leave it at optimize. Remember, all of these women should be in a lateral tilt. Keep the uterus off the IVC and the Aorta, okay? So it's an easy and simple thing to do, so you avoid the supine hypertensive syndrome, and if you need a pressor, noradrenaline or norepinephrine is the first line. There's lots and lots of suggested blood pressures. I mean, often pregnant women have lower blood pressures than they might otherwise do due to the changes in pregnancy, but if you generally go for a MAP of 65, you're probably gonna be okay. I've written the FDA's discussion about noradrenaline up there because a lot of people say they don't wanna use it. Once again, if you don't resuscitate the mother properly and treat the mother properly, the baby's not gonna do very well, so please use it, and there is a sort of masterful understatement there saying life-sustaining therapy for pregnant women should not be withheld due to the fetus, so please just give what you normally do. So speaking of the baby, what should we do about monitoring the baby? Now, this woman and her baby are in RICU, so there's no consensus guidelines that I know about. There's different modalities of doing it. Should it be intermittent? Should it be continuous? This is where you need MFM and obstetric people involved. Many units use continuous fetal monitoring. If you choose to do, that's not what we do in my unit, but if that's what you do, it needs to be appropriate, so it needs to be appropriate gestation, et cetera, and like any monitoring, you need to actually have someone who understands what it means, how to interpret it, and then if you discover a problem, what are you gonna do about it? There's not a lot of point having it if there's nothing you can do after that, and this is quite a nice article that came from New York talking about building an obstetric ICU during COVID, and they used continuous electronic fetal monitoring if fetal resuscitation was appropriate and desired, and obviously, at an appropriate gestation, and as I said before, they found that it was maybe a subtle early warning sign to talk about adequate maternal resuscitation if the fetus was unhappy, so this is some examples in that of blood pressure dropping or the stats dropping. The fetus is the first thing that shows that there might be a problem, so it's quite an elegant thing. Then next, should we deliver the baby? Perhaps that'll fix everything, or at least make my stress level less, okay? So it's not a simple decision, okay, and it needs, if you remember the second thing, apart from the airway thing, it needs to be individualised, okay, it's multidisciplinary, there's lots and lots of factors to take in, might be for maternal or obstetric indications. Unfortunately, there's a high rate of spontaneous labour, especially in early gestation, and some conditions that we haven't talked about today, such as preeclampsia, you know, delivery is high on the treatment list there, depending on a variety of things we won't go into now. So there is some evidence out there, now this is from quite a while ago now, this is 25 years old, this is talking about maternal indications, and so delivery did result in a sort of hefty reduction in the FIO2 in the first 24 hours, but there wasn't uniform, dramatic, sustained improvement. This was a case series we did a few years ago now, and in some women, there was a big change in the OI, big change in compliance, and people who you'd say were the sickest, i.e. with the lowest PF ratio, there was a big improvement post-delivery. But that wasn't the same across the board, and delivery was for a variety of indications, because this was a sort of retrospective observational study, and the data was insufficient to identify those patients who might benefit the best from that. Okay, this is a European COVID study, looking at the same thing, so delivery they found improved the driving pressure, improved the PF ratio, they were both statistically significant, improved compliance by up to a quarter, but that wasn't significant, and the plateau pressure came down as well, that wasn't significant either. And the study I showed you before, this was a prospective study, delivery didn't change the plateau pressure or the driving pressure, and there was some improvement in PF ratio, but the conclusion of that was that we shouldn't be doing this just because someone comes to our unit who's very ill and pregnant, we must deliver them, no, please don't. Okay, it's a decision that you need to take, individualized for each mother and each child, depending on the circumstance. All right, so in conclusion, sort of please remember the physiological and anatomical changes that occur in pregnancy, and think about them when you're interpreting things like blood gases, other stuff, thinking about intubating, thinking you should be fine, that you might not always be fine, and get an expert. Involve a multidisciplinary team early, many and most things are actually very similar, and there's state literature out there doing, do what you'd normally do, because saving the mother is the best way of saving the baby, but please think about everything you do. Okay, it doesn't cost anything to check things quickly, prescribing, et cetera, et cetera, all right. And delivery might be from maternal or obstetric indications but it always requires careful decision making and planning. Thank you. All right, so switching gears to nutrition in the ICU. I don't have any disclosures for this talk, so you can take a moment to take a picture of this, so you can participate in the audience response question. I do have one. All right, so I want to emphasize that as we talk about nutrition, it should not be viewed as its own silo, it should be part of the evidence-based medicine that is delivered to these patients who are critically ill, and nutrition by itself should not be expected to change outcomes. With that caveat in mind, the objectives of today's talk are to briefly review the existing data on early nutrition in the ICU, which seems to be the most controversial, and then we'll discuss some new data that was published on that. We'll explore the possibility of preventing ICU acquired weakness with that, and then we'll revisit the decade-long debate about bolus versus continuous feeds. So the question is a 63-year-old healthy woman with BMI of 32 and type 2 diabetes and hypertension. She was admitted with septic shock with severe chlamydia-acquired pneumonia and acute respiratory failure to the ICU. You picked up the ICU service and are seeing her 16 hours since her admission. She's sedated and her exam is unremarkable except for predilectema. She's in volume-assist control ventilation with norepinephrine at 0.2, antibiotics, dexperitamidine, and fentanyl infusions. As you're reviewing all of her management, the nurse asks you about nutrition at this time. Which of the following statements would be most appropriate? A, wait for the shock to resolve and then start enter nutrition at 20 to 25. B, start enter tube feeds now at six to 10 kilocalories and protein at 0.2 to 0.4. C, start total parental nutrition aimed at 20 to 25 and protein at 1.2 to two. And D, delay any nutritional therapy until day four to five. I guess we'll wait for the polls to close. All right, everybody got it right. That's awesome. Okay. So the recently released European Guidelines on Nutrition basically take you through this algorithm where basically everybody has to be considered to go on some sort of nutrition if they're gonna be staying in the ICU for 48 hours or more. Oral route is obviously preferred. If they can eat, that's the first choice. If not, you start with enter nutrition. But it's important to recognize that they have said that it should not be a full dose, but 70% of the estimated resting energy expenditure. And there's some controversy over how you go about doing that. The ideal way to do it is by using indirect calorimetry because the predictive equations which I displayed in the question do not correlate very well with indirect calorimetry. However, this tool is not used very frequently in a lot of the ICUs. And then if you look at most of the research studies, they do talk about predictive equations in their randomized arms. So if they're not able to tolerate internal nutrition, you do an assessment about malnutrition. And if it's present, you would start early parenteral nutrition. And if it's not present, you could afford to wait for a little longer. So in terms of what we know so far, there have been a paradigm shift in the last decade or so where we are moving away from full dose feeding in these patients earlier on. And there is a change in how we deliver the nutrition early versus in the subacute phases and in the late phases. What we know from prior studies is that assimilation of carbohydrates and proteins at that time is kind of blunted because the anabolic response is blunted. And a provision of higher amino acids doesn't translate to higher muscle protein synthesis. So the original study that came out was the EDEN trial in 2012. And after that, two more large randomized control trials indicated that there was no merit really to do a full dose internal feeds except that there was more GI intolerance when you use the full dose. In terms of parenteral nutrition, the large EPANIC study was worrisome because using full dose of the parenteral nutrition, which was about 25 to 30 or more kilocalories per day, led to worse outcomes, including organ failures, infections, and ICU-acquired weakness. However, when this was controlled in the subsequent randomized control trials, parenteral nutrition was not found to be as detrimental. And it sounded like it may have had to do with dose and perhaps control of the hyperglycemia. So moving forward, the big study released this year was the NUTRIARIA-3 from France by Jean Renier and his group which specifically looked at a very low-calorie and low-nitrogenous load earlier on in patients who were on mechanical ventilation and in septic shock. So this is a pragmatic randomized control trial across 61 French ICUs. They were on mechanical ventilation and vasopressors. The average dose of norepinephrine was about 0.5 microns per kilo per minute. And nutrition was started within 24 hours. In the study arm, the dose was six kilocalories per day, and the protein was limited at .2. In the standard arm, it was about 25, and the protein of one to 1.3. I would also let you know that in most randomized control trials, as well as in real life, it is very hard to actually deliver the dose you have intended them to get, and that happened in this trial as well. After the first seven days, they were liberalized to 30 kilocalories and 1.2 to two grams of protein a day. The primary endpoint was time to readiness for ICU discharge, which was basically defined as being free of mechanical ventilation, free of vasopressors, as well as not having sedation or metabolic issues. And then they also looked at 90-day mortality. So in terms of the dose delivered, they were able to start feeding within 17 hours of randomization in both arms, and most of the patients were on enteral nutrition. It was a pragmatic trial, so the providers had the liberty of choosing parental nutrition, which occurred in about a third of the patients. But look at the dose, it was about 7.4 in the study arm, and about 22 in the control arm. The protein was limited nicely at .2 in the study arm, and it was .9 in the control arm. There was a more fluid intake in the standard group. In terms of the outcomes, the 90-day mortality was similar. However, there was a significant decrease by about one day of time to readiness for ICU discharge, and this came mostly from decreased time on mechanical ventilation. The other benefits observed included less hyperglycemia in the low protein and calorie group, and there was also less hypophosphatemia, although they did not detect any overfeeding syndrome. So essentially, it said that there was no harm done when you go low, and there was a possible benefit. There was some controversy around the outcome of time to readiness for ICU discharge, because it was not really blinded, and so there's a possibility of bias playing a role in there. The second trial was also interesting, was an effort protein trial, which specifically now controlled just the protein intake. This was on enrolled patients in 85 ICUs across 16 countries. These were mechanically ventilated and at nutritional risk. That was defined as low frailty scores or the low sarcopenia scores, evidence of malnutrition, or if the BMI was 25 or less or 35 or more, or if they were expected to be on mechanical ventilation for four days or more. The study arm used a high protein of 2.2, versus less than 1 or 1.2 in the control arm, and this was continued up to 28 days. This study was carried out during the COVID pandemic, and the recruitment fell short, so they revised the target from 4,000 patients down to 1,200, and they revised the primary outcome from 60-day mortality to time to discharge alive from 60 days, and the 60-day mortality becomes secondary. So once again, the doses fell short in the study arm. It came to about 1.6 grams per kilo per day, and in the control arm, it was 0.9. The internal caloric dose delivered was also similar between the two, about 14 approximately in both arms. There was no difference in the primary outcome, which was time to discharge alive at 60 days. Similarly, for the secondary outcome, the mortality was not significantly different. However, there was a difference noticed in patients who had AKI on admission, who had a higher mortality with high protein feeds, and similarly, those who were sicker with a SOFA score of nine or more also had a higher mortality, and this has also been shown in two prior studies and post-hoc analysis that the higher nitrogenous load in patients who have AKI seems to lead to increased mortality. So taken together, these studies indicate that it's important to go low initially in the initial part of the nutritional and critical illness. As you move into the subacute phase and the anabolic resistance is blunted, at that point, you can probably start going up. Switching gears to ICU-acquired weakness, this was a registry study from 69 ICUs in Spain who were on mechanical ventilation for at least 48 hours and received at least a week of internal nutrition. The primary outcome was ICU weakness, and they managed to enroll 319 patients. What they found is that the dose of the energy and protein delivered was similar. In logistic regression, there was no difference on those who had and did not have ICU-acquired weakness, so it did not make an impact. The only impact really came from those who were mobilized early, according to protocol, and these are the ones who had less ICU-acquired weakness. They also noticed that there was overfeeding in patients who had obesity, but did not explore this any further. The last paper is, again, looking at, is there any merit to doing bolus feeds, because it seems more physiological and you induce at least a partial fasting state, which perhaps may be important because the fasting state does help improve autophagy, which removes the dead cellular debris. So this battalion is of 13 randomized control trials, also included a trial sequential analysis to make the results more robust. It was a heterogeneous population, and the primary outcome was hospital mortality. They found overall that in the studies assessed in hospital mortality, which is about seven studies, there was a decrease in hospital mortality with continuous feeding instead of bolus feeds. However, the trial sequential analysis suggested that the number of patients for this analysis seemed to be falling short. Nevertheless, there were other complications also seen. There was more constipation with continuous feeds, which we know, but interestingly, there was no difference in diarrhea, gastric residual volumes, aspiration or pneumonia rates. So in conclusion, going low with low calorie and protein intake early in critical illness does not lead to poor outcomes. High protein intake earlier on may be detrimental in patients who have acute kidney injury. And internal nutrition strategies really have not been shown to prevent ICU-acquired weakness, and I think we should still stick to continuous feeds at this point. Thank you for your attention. So my name is Anshul Kapoor. I'm from Cleveland Clinic. We are gonna talk about updating acute liver failure and referral to the transplant center. I want to state that for the safety of our presenters and attendees like this time, I urge the ACCP to commit to ensuring future meetings are held in locations where elected leaders understand the importance of freedom and sanctity of patient-physician relationships. I'll start off with a great note. Thank you. So this was a study done 10 year ago where somebody like the pioneers of acute liver failure predicted that in 2024 it will be a curable disease. So we are approaching 24 and see where we are. So what we are gonna talk today is appraise etiology and eligibility for transplant as soon as possible in acute liver failure, recognize the need of early multidisciplinary care and organ-specific therapy for acute liver failure. How do we early prognosticate and refer patients to the transplant center? We all know the definition, but just to remind ourselves, the acute liver failure definition is combination of acute liver injury, which is defined by no incidence of or no confirmatory disease of chronic liver disease, sedum transaminitis more than two to three times normal, impaired liver function defined by coagulopathy and jaundice plus hepatic encephalopathy by Westhaven grade criteria up to 12 weeks post-jaundice. So within 12 weeks when you see combination of ALI and encephalopathy, the automatic definition becomes acute liver failure. Multiple studies have been done to subclassify acute liver failure into hyperacute, acute, and subacute. As the name states, it all depends on severity of illness, severity of coagulopathy, jaundice, intracranial hypertension, chances of spontaneous recovery, and in certain cases, the type of disease or type of injury to the liver. For example, acetaminophen and hepatitis A virus are more commonly associated with hyperacute liver failure. Multiple etiologies, again, we are not gonna talk about it, but in general, the classification is viral toxins, pregnancy, vascular diseases, drug-induced, and other causes which are separate than the acute liver failure. These patients might have chronic liver disease and now have an acute flare, which includes Wilson's disease, autoimmune disease, and in addition to that, lymphoma, malignancy, and HLH. Depends what the most common disease etiology is depends on where you are in the globe. In the United States, the most common cause of acute liver failure still remains the Tylenol overdose, but what I want to point out is over the last one decade, there has been increasing rise in hepatitis B infection thanks to opioid overuse and chemotherapy medications, so hepatitis B incidence is increasing, more so the drug-induced liver injury has been increasing over the last one decade. Surprisingly, the drug-induced injury is not the medications we use, but is the herbal medication, so make sure you ask those medications in the history of patient's disease. We are now focusing on how to manage by knowing these are the common factors, this is how the acute liver injury happens. What we as intensivists should know about when we see these patients, what are the immediate concerns for these patients? The first and foremost is exclude chronic liver disease because if it is a component of acute or chronic liver disease, then your trajectory and clinical course and your management is different. So exclude cirrhosis, exclude alcohol-associated liver injury and alcohol hepatitis, screen these patients intensively for hepatic encephalopathy, remember the definition includes acute liver injury and hepatic encephalopathy, so sometimes that encephalopathy could be that subtle that we might miss it, so you do not want to miss that window. So three components of it, determine cause to guide your treatment and also give you a determination of prognosis, assess if this patient is suitable for transplant, so you see social workers, transplant coordinators coming into the unit on your alert, saying let's assess while you are managing this patient, this is happening in the background. Transfer, if you have any clue or if you have a hinge that this is acute liver failure, not doing well, transfer to a specialized unit or a transplant center as soon as possible. The signs would be patients rising INR and rising or change in hepatic encephalopathy, so when you see those two things, transfer to the referral center. Determine cause, why it is important, because if it is primary liver disease, that patient could be a transplant candidate, so that could be drug-related, viral, autoimmune, acetaminophen versus extra-hepatic, it could be ischemic hepatitis, bad RV failure, so you have to differentiate those two types of etiologies because the transplant indications would be different, there is no indication for transplant and right heart failure and liver disease or liver congestion. Another thing to note here is that there are multiple mechanisms of liver injury, the progression of acute liver failure would be different in viral-induced or hepatitis B-induced acute liver failure versus drug-induced liver failure. Viral diseases mainly act by PAMPs, and the drug-induced mainly caused by damage-associated molecular proteins. Depending on what the initial cause or the initial insult to the liver disease is, your progression of clinical course would be different. In this scenario, there are pro-inflammatory markers and anti-inflammatory markers, they are trying to balance each other, once that balance is disbalanced, you do see multi-organ dysfunction. So you want to catch these patients before they go into multi-organ dysfunction. So again, as expected, how do you find the diagnosis? You look for all the liver markers as well as assess disease severity based on your PT-INR liver function test, renal function test, and also looking at other organ dysfunctions, like for example, ABTs, arterial pneumonia. So this is how you will send the plethora of labs as soon as you see these patients to find the cause as well as severity of illness. Just to note on this slide, sorry, I forgot to put the animation on this, but look at the ALT in three thousandths and two thousandths in acetaminophen and hepatitis A-induced viral acute liver failure. So there are phenotypes of acute liver failure which are more inclined towards increased transaminitis, and some are more towards where you see the increase in bilirubin. The difference is, when you see severe transaminitis, that means hepatic necrosis is happening very, very fast. You have to act very, very fast on those patients, so those are mainly Tylenol, hepatitis A-induced acute liver failure. In bilirubin, yes, you have to be acting fast, but those patients still give you some time to salvage their liver. The next is assess suitability of transplant. So after years and years, still the criteria remains as King's College criteria to define severity of illness and define whether this patient is a transplant candidate. Same criteria as you have read years ago in your medical school. Acetaminophen-induced etiologies cause arterial pH less than 7.3, INR, renal failure, increase in creatinine, altered level of consciousness, hypoglycemia. The more recent than the original criteria was add of lactate. So once they added lactate to the King's College criteria, the specificity went up. So once you define this patient as a transplant candidate and you have the ball rolling on transplant candidate selection, you are in the ICU doing organ-specific management. So a few pointers on what is the updated organ-specific management in acute liver failure. Don't have to read the slide, but we will go through multiple organs, just showing that this is a multi-system organ dysfunction. Neurological. So as I said before, pay attention to hepatic encephalopathy. You have to probably watch those patients every hour to do neurological exam to assess and not miss the award signs of encephalopathy. The course of neurological dysfunction depends on the phenotype of hepatic disease, like for example, viral versus drug-induced liver injury, and the outcome of the disease, like how this disease is progressing, how fast. Every six hours, transaminitis is going up. Those patients would be more likely to have neurological injury. It is associated with worse outcome if there's a coexistence of other organ dysfunction, for example, infection, inflammation without sepsis, like severe sores, other organ dysfunctions, use of vasopressors, and arterial ammonia level persistently elevated more than 100. If it is more than 100, patients are at risk of high-grade encephalopathy. More than 200 is associated with intracranial hypertension. So what can you do when you see these criteria? Knowing one thing, that the incidence of intracranial hypertension has been decreasing over decades. What you saw in 1973 is not true in 2023. The incidence has decreased significantly, thanks to us, thanks to a better care in ICU, thanks to all the therapies we implement early on to prevent intracranial hypertension. But still understand the risk of intracranial hypertension is high in patients who have either hyperacute or acute phenotype. Patients are young, they have small sulci, tight brains, increases risk of intracranial hypertension. Any fluid imbalances in the body, including renal impairment, concomitant renal failure, AKI, need of anatropic support, and persistent elevation of arterial ammonia causing astrocyte swelling. So any of these criteria, you should be taking more attention to, that this patient has high risk for intracranial hypertension. In the United States, acute liver failure registry showed that ICH has decreased from 52% to 30% in last decade. And death from cerebrodema has decreased from 12% to 4.5%. So yes, we have to be vigilant, but we have to be also taking this data into account that this is not that common anymore. So what you should do when the patient is at risk, or any patient with acute liver failure, take these precautions. Head off bed elevated to 20 to 30 degrees to increase the venous drainage from the brain. Brain imaging to be done so that you can exclude the other causes, and also look for the bleeding in these patients that can cause intracranial hypertension in the setting of severe cragulopathy. Grade of encephalopathy should be noted. Multiple times I said, it can be done multiple ways. The neurological exam has evolved over time. Pupillary reflex, now we have pupillometer so that you take the subjectivity out of it. Glasgow Coma Scale, EEG. There is a good correlation between ICP monitoring and optic nerve sheath diameter. So these days, many of the, like you have a talk on neurointensivist neurocritical care. So these days, optic nerve sheath diameter is taking precedence over all other non-invasive measures to look for the risk of intracranial hypertension. Intubation with sedation and mechanical ventilation. As soon as you see the grade of encephalopathy is worsening, intubate these patients early on. Medications, mannitol, or you start CRRT early on these patients. Target sodium level, 145 to 155. Moderate hypothermia. So there are multiple trials in the past. Equivocal results, but this is the recent meta-analysis, systematic review done on what should be the targeted temperature management in acute liver failure. Nine studies were done. Nine studies were included in this systematic analysis, which did show that there was a benefit, the older studies did show a benefit in intracranial hypertension in the treatment. Like if the patients do have intracranial pressure, high pressure, by doing targeted temperature therapy, you can treat that intracranial hypertension. But the recent studies, more so randomized control trials did not show that effect. Rather, their intention was to look for prevention of intracranial hypertension. They did not show any potential benefit in preventing intracranial hypertension. Continuous renal replacement therapy, same study. There were six adult liver transplant ICU across Australia and New Zealand looked at ALF, like the CRRT effect on hyperammonemia. But they did show that there is a correlation between early start of CRRT and decreased ammonia level. They did not look at further what happened post-transplant, whether these patients were optimized for transplant or what happened to post-transplant survival by decreasing ammonia before. The whole intent of study was to decrease the ammonia level. More studies needs to be done, but this is not in literature in nephrology yet. Other systems we all know, the cardiovascular, these patients are hyperdynamic in the beginning. But as the liver necrosis progresses, there's a severe vasodilatation and hypotension. Use of vasopressors, cautious fluid resuscitation is needed in these patients. Pulmonary, so they have hyperventilation and alkalosis, which by literature is a good idea, but do not let it happen for too long, because hyperventilation will cause cerebral ischemia because of cerebral vasoconstriction. So when we say prophylactic or do, there is no prophylactic hyperventilation in these patients. You do hyperventilate only for little procedures. You are getting a CT, putting the patient down, you do hyperventilation only for small bits and bits, but not permanently. Metabolic, they are at risk for hypoglycemia, hyponatremia, and elevated lactate. Correct their metabolites. Recent meta-analysis showed patients with more than 200 milligram of glucose do better. Not that you aim for 200, but again saying, do not do tight glycemic control as in sepsis. Nephrology, again, early CRRT for control of hyponatremia. Also, there are randomized control trials looking at early CRRT, despite of ammonia level, is beneficial because it balances fluid and causes less intracranial hypertension in short of ammonia too. Coagulopathy, don't start transfusing these patients. There's a balanced coagulopathy dysfunction. As much as there are procoagulant decrease, you have anticoagulant decrease and procoagulant decrease. So these patients normally do not bleed. Do not correct INR for this reason, and also because you do not want to give too much FFP and raise intracranial pressure. Thirdly, if you give too much of blood transfusions, the INR is the marker of severity of disease. You do not want to correct INR because that is telling you how the patient is progressing in course. Infections, no use of prophylactic antibiotics, but whenever you see clinical signs of infection, criteria for SIRS, or worsening hepatic encephalopathy in these patients, think about infection. That's the time to start empiric antibiotics. Patients staying in the hospital for too long can think of antifungal. Therapeutic plasma exchange, great data that these patients, once you start TPE, significant improvement in transplant survival rate. The paper is so good that it made a guidelines to, it's 1A guideline by Aphoresis Literature to do therapeutic plasma exchange in acute liver failure. Just want a word on extracorporeal support. We all talk about MARS, we forget about MARS, but this is a recent and the only meta-analysis done on extracorporeal liver support. For the sake of time, just take this graph as, the higher the bar value, that means this higher is the probability of intervention to be seen. So this study was done, this meta-analysis was done to see what these treatments do in terms of in-hospital mortality and encephalopathy. They could not look at encephalopathy because these patients were all intubated, sedated. So they could not look at that primary outcome, but they did look at in-hospital mortality. Biological detoxification, which is not in market anymore, did show, based on these studies, the maximum effect on mortality. The next is MARS. So at least by this meta-analysis, we know that MARS has a tendency to decrease mortality when used in acute liver failure patients. But at the same time, this study, or this meta-analysis has a lot of drawbacks. Small sample size, heterogeneity of the patients, done over 40 years, and inconsistent definition of acute liver failure in these diseases. Standard medical therapy has changed over time, so we can't correlate these results. Last word, status A in the OPTN score is the acute liver failure. So this is the disease priority, that's why diagnosis is very, very important, and early diagnosis is important to us. What we know is criteria for early liver transplant and emergent liver transplant have shown improved survival over time. We have seen survival change from 16.7% to 62.2. Multiple prognostic markers, including King's College criteria, MELD, and SOFA. So King's College criteria predicts really well, but it lacks sensitivity and fails to identify patients that will not survive without transplant. MELD predicts more in non-acetaminophen-induced liver failure. SOFA score is equally good as King's College criteria. So what I would say, look at the clinical trajectory, still be the intensivist, and look at SOFA score on these patients, and correlate with King's College criteria to make a diagnosis whether this patient is getting sicker, whether this patient even will survive the transplant, even if it's a transplant candidate. Multiple scoring systems have been in place, but again, clinical judgment is the key. We believe in scores, but at the same time, a clinical course of the disease will tell us more than the scores. Lastly, liver transplantation has been very, very successful over 40 years in acute liver failure. One year survival after emergent liver transplant, regardless of the etiology, is now around 80%. So these patients do survive transplant. Survival rate in acetaminophen-induced liver injury is becoming closer to the post-transplant survival rate. So by means that medical therapy alone can make these patients skip the transplant. So survival is as good as post-transplant. So I think we are living the dream of our pioneers in 2014 that we have found a cure for acute liver failure, at least in acetaminophen-induced liver injury. 60% of the patient weight listed do not go under liver transplant because they get better. Emergent liver transplant is associated with 80 to 90% survival. So what are key take-home points? Our need for early diagnosis and referral to transplant center is must. Organ-specific therapeutic approach should be taken for these patients early on. Multidisciplinary team approach is necessary, including hepatology, nephrology, neurology. Prognostication is important for us so that we can bridge these patients to transplant early on because their post-transplant survival is great. Thank you. Good morning, how y'all doing out there? All right, thank you very much for coming to our ICU speed dating session today. So my name is Mark Warner, I'm from UT Health Houston, I have no disclosures. I was tempted to say that I'm not an endocrinologist, but I didn't want to lose you for the last 15 minutes. We're going to attempt to cover four different things here. We're going to talk about hyperglycemic syndromes, we're going to talk about hyperthyroid syndromes, we're going to manage decompensated hypothyroidism, and then we're going to talk about adrenal crisis. So just a show of hands, who is taking care of a patient with DKA? Okay, so I'm not going to reinvent the wheel here, I think what we want to do here is give you a public service announcement to say that, contrary to some of the talks that you'll attend, I don't think the management is the hard part. I think the diagnosis and the recognition is the hard part. So I'm going to show you some data to say that delay in diagnosis is increasing mortality in some of these conditions. So when we talk about glycemic disorders, a lot of endocrine disorders go, is it too much or is it too little? So I don't think you're going to miss hypoglycemia. I think this is something that we're all attuned to, and so all of the allied health providers, nursing principally, are going to be looking for this. They're going to be sticking fingers, they're going to be drawing blood, and you're not going to miss this. So let's focus with hyperglycemia. So when we talk about hyperglycemic syndromes, we're going to talk about a few things here. So traditionally we're taught there's two. It turns out there's probably a Venn diagram, and so I'm going to show you a little bit of data about hyperosmolar hyperglycemic syndrome. We can't say honk or hyperosmolar non-ketotic coma and things like that, right? We've got to take the coma out because we've got to think about hyperosmolar hyperglycemic syndromes. DKA, I don't think we're going to miss DKA, so we're looking for acidosis. And so there's a Venn diagram of overlap here where you're going to have a hyperosmolar patient that does have a little bit of acidosis, but they're not frankly in DKA. And then I'm going to briefly talk about euglycemic DKA. So DKA, it's common. We know it. Mortality is decreasing, so that means we're doing a really good job. So similar to all three of my predecessors, it looks like comprehensive critical care is winning, right? We're winning on some fronts. When it comes to hyperosmolar hyperglycemia, we've got some work to do. So it's not very common. So we're talking about 13% or less of admissions, and this could be up to 20% of the time the initial presentation of diabetes now. And so this is working its way into not only adult care, but also pediatric care. So we know that childhood obesity is increasing, and so if you have a patient that presents with a sugar over 600, you have to think about HHS. If you look at U.S. and U.K. data, this is staggering to me that 10 times more people are dying from hyperosmolar hyperglycemia than DKA. And this is international. So when you compare these side by side, we're looking at sugar. If your sugar's less than 250, it's not hyperosmolar hyperglycemia. If it's over 600, it probably is, but we have to look for overlap. And so we're looking for acidosis markers here, and we're looking for hyperosmolarity. One of the key features here as we're talking about hyperosmolar hyperglycemia is the profound dehydration. So when we think about this, most of the causes are going to be infection. The majority of the infections are going to be pneumonias. I don't think this surprises you. 21% of the time, or about a fifth of the time, patients are noncompliant with their antidiabetic medications, principally insulin. But when we're looking for this, we're thinking about older patients, we're thinking about patients that have had time in life to acquire medical comorbidities, and we're also thinking about kids with undiagnosed medical comorbidities. So when we talk about this profound amount of dehydration, really this is size dependent, but classically the teaching was that the hyperosmolar hyperglycemic patient is going to be twice as volume depleted as the DKA patient. And so the principal thing that we need to do is give volume repletion. When we talk about pathophysiology here, there's an underlying trigger that's really small in the middle, but we're thinking about the classic three things that we always teach, infection, infarction, medication noncompliance. But we think about the different reasons that adaptive hormones are increasing and you're going to get hyperglycemic. That hyperglycemia is going to drive an osmotic diuresis, and that osmotic diuresis is going to be the principal thing that over days to weeks causes a patient to end up in an emergency department and end up in your ICU. They do have continued presence of insulin, right? And so this is going to be the thing that prevents them from being, frankly, ketotic. So I thought this was very nice from the British guidelines. And so this provides a two by two table here. And so what you're looking at on the y-axis here is serum ketones, principally beta-hydroxybutyrate. And then on the x-axis, you're looking at plasma osmolarity. So you can see over here, we've got DKA. And again, I don't think anybody's going to miss that. And when you come down the ketosis here, you have a patient with low osms that still has ketosis. This is where we're going to talk about euglycemic DKA in a minute. Up here, though, is where it gets interesting in your hyperosmolar syndromes. So if I have a patient who's very volume depleted, very hyperglycemic, and they have a little bit of ketosis, then I may have an overlap here. So it doesn't have to be all one or all the other. When we talk about management, this is where this is going to get nuanced a little bit. When and how you start the insulin is going to depend upon whether or not the patient's, frankly, ketotic or not. So if they're not ketotic, then we're going to replete the volume, and then we're going to watch the sugar naturally come down with volume depletion, and then we'll start insulin. So we'll get there. Euglycemic DKA. So I brought it up. I teased you with it a little bit. It's not incredibly common. But when we see SGLT2, I left out a T there, SGLT2 inhibitors, then there's approximately five of those on the market that I'm aware of. These patients that present with euglycemic DKA, it doesn't fit neatly into a box, which is why 50% of them have a delayed diagnosis. And so you can see the criteria here. And then here's a differential diagnosis. So I'm not going to touch pregnancy. I think that was done very well, and I'm not going to attempt to touch endocrine disorders in pregnancy, for sure, at the end of the hour here. But we have to think about pregnancy. We have to think about our advanced medications for diabetes. And then we think about glycogen storage disease, liver disease is on the list. And so this does a good job of kind of encapsulating several different things, including caloric deficiencies. Here's a schematic of why this might be the case. And so we know, sorry, as soon as I touch it, it wants to move forward. So we know SGLT2 inhibitors prevent glucose reabsorption. And it turns out they can also prevent reabsorption and processing of ketones, which is what allows a patient to get ketotic. But because the glucose is not being reabsorbed, they don't get hyperglycemic. And so this creates a little bit of a conundrum in terms of diagnosis. Down the left side of the graph here, we have the management. It's fluids, right? So when you talk about hyperglycemic patients, give fluids for the hyperglycemia, right? So the classic teaching was if the patient's in DKA and you have a lot of ketosis, you give the insulin to force the liver to process ketones, right? If the patient's hyperglycemic, you give fluids to replete the volume. And as you replete volume, the sugar will come down. So there's a couple of nuanced things here, and I'm going to try not to touch it. So on the top pane here, this is also from the British guidelines, you'll see as you're following the osmolarity, we want to be careful that as we replete volume that we're using isotonic fluid, but we're also watching and remeasuring constantly osmoles. Because if you're three to eight milliosmoles per kilo per hour is what we're shooting for. If you're dropping the serum osmolarity too fast, you have to worry about cerebral edema and osmotic demyelination. If you're not dropping it fast enough, then we need to think about giving hypotonic fluid. I also mentioned about blood glucose here. So in a HHS patient, we want to give volume to replete this dehydrated state, and we want to watch the sugar coming down. As the sugar comes down, when you get a plateau of the sugar decrease, then you can start insulin. So in the background here in the British guidelines, the FRIII, that's insulin infusion, fixed rate IV insulin infusion. And so you're going to use about half as much in HHS as you're going to use in DKA. So in DKA, remember, the insulin is going to be the driver of the treatment, and HHS fluids are the driver of the treatment. Okay, rapidly switching gears, thyroid disease. Everybody's with me. We're managing diabetes. We're managing DKA. Everybody's good. Okay, thyroid disease. So too much or too little. We're going to talk about both here for just a second. So first of all, hyperthyroidism. So when you think about thyrotoxicosis or hyperthyroidism, it's elevated thyroid hormone plus symptoms, right? And I'll show you the symptom cluster here, but principally the patient has tachycardia, the patient has AFib, things like that. The prevalence is relatively low but noticeable. Thyroid storm obviously is much less common than that, but it's thyrotoxicosis, and I think these are the old versions of my slides. So the hallmark of thyroid storm is elevated thyroid hormone plus altered mentation, and then you either have fever, tachycardia, or heart failure. The mortality can be up to 40% in thyroid storm, and I'm going to show you why. So this is a very interesting depiction here. So there's two large studies, and I'm going to show graphs from each. So in this study, you're comparing patients that had CT, which is clinical thyrotoxicosis versus TS, which is thyroid storm. So when you look at this, what this is telling you is that in the thyroid storm patients, 100% of them were tachycardic, right? And each one of these rings is 20% moving out from the center. So most of them, thyroid storm had a precipitating event, they had altered mentation more than half the time, and some of them had fever. When you look at inside of that in the thyrotoxicosis, the symptoms were less frequent, but principally you were seeing tachycardia in there. When you talk about another longitudinal study, this was an 18-year French study, when they looked at 100 patients that had thyroid storm, the in-ICU mortality was 17%, right? So that's worse than sepsis. And 22% in six months, but you look at the survival curve of the patients that presented with cardiogenic shock versus those that didn't, the cardiogenic shock is a principal driver of mortality in thyroid storm, right? So we're going to need to figure out how to support their cardiac symptoms system here. So you talk about management, if the patient has thyrotoxicosis, you give beta blocker. If the patient's over 65, you think about giving a beta blocker when their heart rate's 90 instead of waiting until it gets to 100, and then you have time to think about your antithyroid therapies. When the patient's in thyroid storm, they give you a laundry list of the normal things, right? The patient's in thyroid storm, so they have altered mentation and they probably have shock. So the beta blocker may not be a great idea. I know it's first on the list, let's think about not doing that if the patient's in cardiogenic shock. Let's replete steroids, let's replete volume, and then they may need vasopressors, inotropes, they may need mechanical ventilation, which is probably likely, they may need mechanical circulatory support. So this is not a talk about ECMO, but ECMO utility. In this cohort, 50% of them got ECMO, and almost 95% of them were mechanically ventilated on pressors. And so these are very sick patients, and so I think the sooner we can recognize this and figure out that there's a precipitating event, the sooner that we can comprehensively manage this. And I think my public service announcement with this would also be, don't be afraid to call an endocrinologist into the ICU to help out, right? So once we realize we're in trouble, let's get as many people on the bus as possible. Okay, so switching gears again to decompensated hypothyroidism. So we're moving away from mixed edema coma, because we recognize most of the patients don't have coma, right? So we want you to recognize this as decompensated hypothyroidism. So your 60-year-old woman with a history of hypothyroidism, I realize we're going over time so we'll make this fast, is likely to present with some of these symptoms. So if your patient has hypotension, bradycardia, hypoglycemia, and altered mentation, if they're not improving with volume repletion, think about decompensated thyroid disease. When you look for that, look for an underlying cause, think about supporting their hemodynamic system. We want to give IV steroids first before we give the IV levothyroxine, and then you want to think about passively rewarming them. So don't get crazy, just put a warming blanket on them and let them slowly rewarm. Okay, last but not least, adrenal disease. So I'm not going to talk about Cushing's today, I'll talk about adrenal crisis and then two slides on this. So adrenal crisis, there's a lot of different criteria out there that I'm going to show you. Most of these patients have a preexisting diagnosis and something happens with their steroid repletion or they get an infection or some reason for decompensation. So think about it, take a good history, and then replete the steroids, give them aggressive volume resuscitation, and support them. So key points, coming in for a landing. I've highlighted and read delay in diagnosis, so I think this is a good one slide to kind of think about the hypoglycemic conditions are more deadly than hyperglycemic ones. But hyperglycemic ones are nuanced here. And delay in diagnosis in all these entity are things that are going to hurt the patient. So thank you very much for your attention.

non-pulmonary critical care

obstetrics emergencies

ICU challenges

nutrition in the ICU

acute liver failure

transplant center referral

hyperglycemic syndromes

mortality rates

euglycemic DKA

multidisciplinary care