Welcome again. OK. Good morning, everyone. So if I were to tell you everything I think you need to know about blood, I would be here for much longer than 20 minutes. But I promise I won't do that. So what I wanted to do this morning, this is a little bit of a slight departure from the full version of the talk that you have available to you. I wanted to focus on some practical aspects and including highlighting some of the guidelines documents that have come out over the last couple of years that would be worth knowing about. And I think there was a mention yesterday about some of the guidelines for TBI, et cetera, that you should know. I'd say the transfusion guidelines are other documents that you should know. And we're going to review some of those this morning. I still have no relevant financial disclosures. Anybody want to help me fix that? Let me know. So we're going to start with a question just to think about. So you have a two-year-old child with septic shock who's gotten a fair bit of volume resuscitation, who is still tachycardic and hypotensive. You're titrating up an epinephrine drip. And that CBC that somebody ordered comes back, and the hemoglobin is 8.5. And so my question to you is, would you transfuse this patient with red blood cells? And I want you to just think about that. We don't need to answer now. I'm going to come back to it. So just think about it. The same night, maybe next door, you've got a two-year-old with pneumonia who's on moderate ventilator settings, who is hemodynamically stable and not on vasoactive support, who has the exact same hemoglobin. Would you transfuse this child? Is the answer the same or different? Just think about that. We're going to come back to it. But first, let's talk about why would we transfuse with red blood cells to begin with. Well, in somebody who's acutely bleeding in a life-threatening way, that's easy, right? Blood's coming out. Give blood back. So we just transfuse, right? Anemia in a non-bleeding patient is something a little bit different, right? So I like to tell my trainees that other than ICU being physics, ICU is also all about oxygen delivery. That's what we do. We try to get oxygen into the body, and we try to get the oxygen circulating around the body. And I know mathematically that the higher the hemoglobin concentration, the higher the oxygen content of the blood, and the more oxygen is going to be delivered. In fact, increasing the hemoglobin is the most efficient way to increase the arterial oxygen content. So it makes sense that I should transfuse more hemoglobin if I want to deliver more oxygen, right? So more hemoglobin is better, right? Well, sometimes more isn't more. And so the people who shook their heads when I said hemoglobin of 8.5 were probably thinking of, well, I wouldn't transfuse until hemoglobin of 7. Where did 7 come from? 7 came from this study. So New England Journal of Medicine, getting back there in years some, 2007. So this was an RCT that looked at a red cell transfusion threshold of 9.5 versus 7. It was across different diagnoses in the ICU. Hemodynamically stable patients, that's an important caveat. And the primary outcome was newer progressive mods, saw no difference in outcomes between groups. And the conclusion from this was that a restrictive transfusion strategy is likely safe. So it begs the question, why isn't it better to maintain a higher hemoglobin, right? Why is my favorite question? My trainees know this about me. It's not important to know the what. It's important to know the why. Why is my favorite question? So why isn't maintaining a higher hemoglobin better? So it might be that hemoglobin alone is not a great marker of who will benefit from transfusion. And there's a lot of work happening right now to try to figure out what are those better physiologic markers. But what you'll start to see in some of the guidelines documents is good practice statements and statements to the effect of, really, we probably ought not be using hemoglobin alone, but we should be using other markers of the adequacy of oxygen delivery. An example of this is from a study that looked at a interaction between the severity of shock and whether or not somebody was transfused. The take-home point from this slide was that it seemed that shock severity may have been a better indicator of whether outcomes would be better or worse with transfusion compared to hemoglobin. So maybe what we're using as a marker of whether or not we should transfuse is inadequate. And if you go to the full version, that's Alan Doctor. He will give you many, many more minutes on that concept and the concept of tolerance of anemia. So it also could be that maybe even if a patient would benefit from increased oxygen content and increased oxygen delivery via transfusion, that red cell transfusion is not always efficacious. And this comes to the concept of different blood products are different. This may relate to how long the blood's been sitting on a shelf before it gets transfused, what it's mixed in, et cetera. You'll see some more detail about that in the larger lecture. I also think it could be that for some patients, the benefit of transfusion, even if it's there, is outweighed by the risks of transfusion. Nothing we do in the ICU is risk-free. So it does actually make sense to take a step back and think about, OK, what are the risks of transfusion? And these are probably testable things. You should at least know what different risks of transfusion are and different transfusion reactions, be able to recognize them, and know what to do. So this is a figure from a paper several years ago. But it's kind of a nice figure that illustrates the fact that the things that we think about as risks of transfusion, the things that we put in our consent forms, are actually relatively rare. They're on the order of incidents of things like dying from being struck by lightning or balls, things like that. Relatively rare, but not zero. And it's worth knowing about them. Just for context, like I said, that figure was from 2012, which it hurts me to say that that was a long time ago. But apparently, that was a long time ago. This table is actually an updated version from 2023. So let's talk about different transfusion reactions. So the fundamental aspect of transfusion medicine is ABO compatibility. That's the thing that we learned way back in medical school of why you need to have matched blood products, how you have universal donors and universal recipients. Well, what happens if that goes awry? This is one of those things that you need to know what it is so that you could recognize it if it happens, because it is life-threatening and potentially rapidly fatal if not addressed. Thankfully, rare. And that's where a lot of the safety of blood transfusion these days has come into. And so what happens is you have antibody antigen complexes that get together, and they cause an overwhelming immune activation and complement system activation. And so this could look like sepsis, right? Anytime we say overwhelming immune activation, those signs and symptoms are going to be pretty similar. Things that you'll see specific to this are signs of hemolytic anemia, right? So you're going to have rapid hemolysis. And so if you ever see either increased inflammation all of a sudden while someone's getting a transfusion, dark urine, or any sign of hemolysis, the first thing you need to do, stop the transfusion, contact the blood bank, and initiate the workup for a transfusion reaction. So that's probably the most severe one. Thankfully, also pretty rare. Things that are maybe a little bit less rare, though still not common, but you should know about it. So transfusion-associated circulatory overload, less common in children than in adults. But it's the reason why, if you have somebody with severe anemia, you transfuse slowly and not give them 40 mils per kilo off the bat in an hour, right? Because we're trying to prevent transfusion-associated circulatory overload. So what it is is acute pulmonary edema within six hours of a transfusion in the absence of cardiogenic edema. Like I said, it's most common in adults. They're rare occurrences in children. When it does occur, some of the risk factors are, again, the severe anemia, hemoglobin less than five, or younger age, or if you have pre-existing cardiac or renal impairment, reasons why you would handle extra volume poorly, right? The other one we should know about is transfusion-associated acute lung injury, or TRALI. So TRALI could look similar to TACO. This is also going to have potentially pulmonary edema with it, but the presentation is going to be a little bit more severe, typically, if it's TRALI. So this is within six hours of transfusion. I have a patient who didn't have lung disease before and now has acute lung injury. Now has acute lung injury, including potentially ARDS relatively rapidly. Like I said, it's relatively rare in children, but it does happen. Enough that you should know about it. It used to be that this was related largely to plasma transfusion. That has changed recently with the advent of male-only plasma donation. That has to do with antibodies that are in the plasma of women who've had pregnancies. And now it's actually more commonly reported with red cell transfusion, in fact, as opposed to plasma transfusion. There's also two different mechanisms. So there's the antibody-mediated mechanism of TRALI, but then there's a non-antibody-mediated mechanism of TRALI. And basically, the underlying pathophysiology of that is a two-hit phenomenon. So you have whatever underlying illness, i.e., anything that lands you in the ICU, that will prime the neutrophils. And then either you have HLA or HNA antibodies, so those antibodies that are floating around in the transfused plasma, or other bioactive agents, probably lipid, but we don't quite exactly know what they are, that activate those primed neutrophils. That leads to inflammation, endothelial injury. There's that endothelial injury again. I think ICU medicine is really an endotheliopathy, but that's an editorial. And that's how you get your pulmonary edema. Again, neutrophils, when they're activated, like to go to the lungs. So you get neutrophil infiltration into the lungs, pulmonary edema, and acute lung injury. So one of the caveats here, too, is that the definition of TRALI technically requires that you don't have pre-existing lung disease. How many patients in our ICU don't have pre-existing lung disease? Very few of them. And so it's likely that the incidence of transfusion-associated respiratory dysfunction, or TRALI, is probably higher in our patients than we realize, because it's really hard to detect. Because if your patient is in the ICU and sick enough that they're being transfused, they probably have lots of reasons to have worsening lung injury. It's hard to pinpoint that exactly on the transfusion. But it's something to be aware of and to think about, because that might be a potential mechanism why your patient all of a sudden has worse lung function. OK. So summary from that portion, transfusion is not risk-free. Most of the known risks are rare. There are other risks that are harder to identify, so more occult transfusion-related acute lung injury and potential immunomodulation from blood products. I could talk for an entire hour on that. I will spare you and not do that. We also probably, again, thinking about most of these things are multi-hit phenomenon. Different patients probably have different risks for these adverse reactions. But we don't yet know how to predict that. So I would love to, for the patient in front of me, to say, OK, if I give you this transfusion of this product, you're going to do great. And you're not going to do so great. We're not quite there yet. That's where a lot of the work of the field is going. But I still have to make decisions. So in the meantime, when do we transfuse red cells? So there's a caveat here. I just said hemoglobin might not be the best way to decide this. But our guidelines are all written based on hemoglobin, because that's what the studies are currently. So that's my asterisk here for the next section. The most recent guidelines came out in 2023. This is adult and pediatric. The pediatric version of these guidelines actually are based on this document. This is one of those documents that you should know. These are the TAX-E guidelines for red cell transfusion from the Transfusion Anemia Expertise Initiative, or TAX-E, because everything has to have a great acronym, published in PCCM in 2018. Within that document is a nice algorithm that looks like this. We're going to break this down so that you can actually see it in the next couple of slides. And so first step, if you're in hemorrhagic shock, again, you transfuse. Transfuse with a balanced resuscitation ratio. I think I have another slide that will give you a QR code for a massive transfusion for trauma guideline. If you have a critically ill child or child at risk for critical illness, so when we put these guidelines together, we also considered hospitalized children as children at risk of critical illness. If you're not bleeding to death, there's no hemorrhagic shock, hemoglobin less than 5 probably transfuse. And that's based on the very few studies that exist largely, actually, in African children in settings of malaria and or sickle cell that show a steep increase in mortality as the hemoglobin decreases less than 5. Between 5 and 7, we don't quite know. Probably indication-based, we don't really know the right indication, so use your best judgment. And if the hemoglobin is greater than 7 and your patient is hemodynamically stable, then, oh, sorry, I skipped ahead. If they are not hemodynamically stable, this is the other place, clinical judgment, because we don't have the data yet to say what the right indication or threshold is for patients who are in active resuscitation phases of shock. If your hemoglobin is over 7 and you are hemodynamically stable, for the most part, the recommendation is to not routinely transfuse. So if you don't have another reason to transfuse, hold off on transfusing that child across most populations in the ICU. Where this is a little bit different, acute brain injury, maybe consider 7 to 10. Oncologic diagnoses, maybe consider 7 to 8. The data in the world is a little bit equivocal on that. And then there were a couple of patient populations for whom we really have very few data, and so it was clinical judgment. So severe PARDS, hemolytic anemias, ECMO patients, although there is an ongoing RCT happening right now. You guys met Ravi yesterday and the day before. They're running that out of Boston Children's and actually pretty exciting. So hopefully, we will have some data on that soon. OK, cardiac diseases are a little bit different. We actually do have a couple of RCTs in cardiac disease. And so basically, this has to do with your cardiac lesion that would expected to have poor oxygenation, potentially, or poor oxygen delivery. You might keep a slightly higher threshold. So single ventricle, after stage 1 palliation, considering a threshold of 9. That's based on a small single center RCT. Same thing with after stage 2 or 3 palliation. But for otherwise, congenital heart disease, thinking about transfusion somewhere between hemoglobins of 7 and 9 based on clinical condition. Right. OK, so that was a whirlwind tour of red cell transfusion guidelines. But the bottom line is basically, for most children, probably less is more. I feel like that message has gotten out. Every time I give this and I do that initial question, the number of heads that shake no goes up. But it's important to consider why we're transfusing, right? So it's not just to treat a number, it's to treat the patient. It's to really think about whether or not our oxygen delivery is adequate. The right transfusion strategy is probably not the same for everybody. And like I said, it's unlikely that hemoglobin alone is what's going to be the best marker. So my other genism of ICU is treat the patient, not the number, right? OK, so if I'm answering these questions, let's see if your answers have changed in your head. That two-year-old with septic shock who's still hemodynamically unstable while I'm increasing my vasoactive support, I might actually transfuse them. That hemoglobin of 8.5, oh my gosh, it's above 7. Shh, don't tell anyone. But I might, right? The child who's hemodynamically stable with pneumonia, I probably wouldn't, right? So this bottom line, sort of using other data to help decide those transfusions. And think about what you're doing and why, OK? All right, we're going to switch gears to the yellow stuff. So the point of this slide, it's based on two international point prevalence studies of platelet transfusion and plasma transfusion. And the take-home point here is that the majority of platelet and plasma transfusions that we give in our ICUs are not to treat bleeding, but to try to prevent bleeding. So the majority of these transfusions are prophylactic. And the indications for those prophylactic transfusions are all over the place. They're all over the map. So a huge degree of variation. When you see variation, that's usually a sign of, we don't really know. So one study, it's not in the PICU population. It's in that place with the tiny babies that starts with an N. But it was a really, I would almost say, landmark trial. So this was a trial of platelet transfusion thresholds and neonates that really made a lot of us in the field go, huh, why is that? Because what they did was they randomized infants in the NICU to platelet transfusion thresholds of 50,000 versus 25,000. And they actually found a higher rate of the composite outcome of death or major bleeding in the higher platelet arm. So what we're doing to prevent bleeding could potentially have actually caused worsening bleeding. Whole lot of science behind why we think that might have happened. But it does make us think about, hmm, these platelet transfusion thresholds that we have kind of made up in order to try to prevent bleeding. Maybe that's not the best approach. So when do we transfuse platelets or plasma? There are a couple of different guidelines out there that are not specific to the ICU. But luckily, there is one that is specific to the pediatric ICU. And so we couldn't leave taxi alone because it was so great. So we had to come up with TaxiCab. So this is taxi control and avoidance of bleeding with the handy dandy logo that I think that Oliver Karam is one that came up with that logo. So the TaxiCab came up with evidence-informed consensus guidelines for platelet and plasma transfusion. And basically, this is the diagram from that article. We're going to go through it in detail in the next couple of slides in the next five minutes. So again, massive hemorrhage. You're going to transfuse with a balanced ratio of red cell to platelet to plasma. In the interest of time, I'm not going through the hemorrhagic shock consensus conference recommendations that were published just about a year ago. But the QR code is here, so you guys have the link to it in the slides. I encourage you to check that out as well. When you get to moderate or severe bleeding, so not life-threatening bleeding, but just bleeding, clinically relevant bleeding, what you'll notice there is a lot of gray. Gray means clinical judgment. Again, clinical judgment means we don't really know. And so the places that you will see where there is some data is a suggestion that, well, probably, maybe consider transfusion at a threshold of 50. But it really is a weak recommendation at best. And that was for patients following non-cardiac surgery and for sepsis and DIC. Oop, I went backwards. Let's go forwards. For no bleeding, so still a fair bit of gray, but now you're seeing some red of probably don't transfuse unless you have a really good reason to. So that your patient's not bleeding, there are some scenarios where we consider probably shouldn't transfuse regardless of what the platelet count is. And so that is for simple procedures in the ICU, an IV insertion, right? You probably don't need to give platelets just for that, an NJ, a CVL removal. So the other one is for acute liver failure and liver transplant. There's reasons for that. One of those is in the field of liver transplant. I'm sure that I heard this yesterday. It's actually better to bleed than to clot. Clotting is really, really bad. And so we try to restrict the yellow stuff unless it's absolutely necessary in that population. And then there are a couple of populations where we couldn't really come up with a specific threshold, but considered that at least if the platelet count is over 100, there's probably not efficacy to giving more platelets. OK. So that was for platelets. There's a companion figure for plasma. Again, you see a lot of gray there under the moderate bleeding. But one thing to note on the very bottom that red that says if your INR is less than 1.5, don't transfuse, that's a little bit of a blanket statement. So you heard yesterday about what the INR of FFP is, right? And so if you're giving FFP to try to bring your INR down to less than 1.5, you will continue to give FFP forever. It won't work, right? And in fact, the best data that we have would suggest that if your INR is anything less than 2, giving repeated FFP probably won't do anything to bring that INR down. And we don't know that INR is even the best way to decide when to give plasma. But that's an important caveat, right? The other piece where there is some data is to consider transfusion of plasma for surgical patients if that INR is greater than 2 times normal, so greater than 2. Again, here there's probably even more red of thinking about don't transfuse regardless of what your coags are, right? And so this is the case for sepsis or DIC. If you're not bleeding, this is a non-bleeding patient, so sepsis, liver dysfunction or transplant, your post-surgical patient, and your simple procedures. OK, bottom line here. I have this whirlwind twirl here. Less is probably no. Less is probably no. Less is probably more. But we really need more data to know what the right indications are. But for right now, thinking about, again, nothing is risk-free. So for non-bleeding patients, less is probably more. And a lot of work that's working on trying to actually predict bleeding. And are we even using the right lab tests? If you look at the full companion to those taxicab documents, there's an entire article on different laboratory assays. And there are pluses and minuses if you're interested in that kind of thing. It's kind of a fun read. And my very last note here. This is relatively hot off the presses. Was published just last, it's still August, right? Last month in Pete CCM. This is the transfusion portion of the Pediatric ECMO Anticoagulation Consensus Conference Guidelines. What, I don't know if it's, you may or may not even be able to see that on the screen. But basically, there's not a heck of a lot of evidence within ECMO yet. And so you can see, basically, the transfusion of red cells is use physiologic indicators, as opposed to hemoglobin alone. Your platelet and plasma is probably don't transfuse if the platelets are greater than 100. Below that, think about it. If your INR is less than 1.5, this is sounding familiar, I'm sure, probably don't give plasma. If it's greater than that, think about it. But probably giving repeated plasma, just for the purpose of correcting the INR, not likely to be beneficial. That's for your non-bleeding patient, or even your bleeding patient. For a patient who is about to go through a procedure, there are suggested thresholds there for platelets and plasma that you can read in the slides, because I think I'm about out of time. In my last slide here, think about the components of a patient blood management program. So very similar to what we talked about for the antibiotic stewardship program. Successful programs are multidisciplinary, include bidirectional communication, and things other than just when you're going to decide to transfuse. But how do we prevent bleeding? How do we prevent anemia? And I leave this here for you. And that's it. Thank you.

pediatric ICU

blood transfusion

hemoglobin levels

TAXI guidelines

transfusion risks

multidisciplinary blood management