Transcript
Speaker: Welcome to Critical Matters, a sound podcast covering a broad range of topics related to the practice of intensive care medicine.
Speaker: Sound provides comprehensive critical care programs to hospitals across the country.
Speaker: To learn more about our programs and career opportunities, visit www.soundphysicians.com.
Speaker: And now your host, Dr. Sergio Zanotti.
Speaker: Platelets and fresh frozen plasma are frequently transfused to critically ill patients.
Speaker: Despite their frequent use in the ICU, significant variability in transfusion practices and thresholds persist.
Speaker: In today's episode of the podcast, we will discuss recent clinical guidelines on this topic.
Speaker: Our guest is Dr. Angel Kosiatako, a practicing pulmonary critical care physician.
Speaker: Dr. Kosiatako is on the faculty of the Cleveland Clinic main campus.
Speaker: He was a member of the 2021 Surviving Campaign Guidelines panel and holds multiple leadership positions at CHEST.
Speaker: He is a member of the Board of Regents and the immediate past chair of the Council of Networks of Chest.
Speaker: He has been awarded the Distinguished Chest Educator Award on multiple occasions, is current editor-in-chief of the Chest Physician Newsletter, and the former section editor for the Critical Care Commentary.
Speaker: Dr. Kosciatako is also a member of the editorial board of the Journal of Chest and is the lead author of the recently published American College of Chest Physicians Clinical Practice Guidelines on Transfusion of Fresh Frozen Plasma and Platelets in Critically Ill Adults.
Speaker: Angel, welcome to Critical Matters.
Speaker: Thank you so much for the kind introduction, Sergio.
Speaker: It's really, really, really nice.
Speaker: It's really a pleasure and honor to be here with you.
Speaker: Same here.
Speaker: And as we were talking before we started recording, this is a topic that frequently comes up in our practice.
Speaker: There's still a lot of controversy around what is the right thing to do.
Speaker: So from your perspective, why should intensivists care about this topic?
Speaker: Yeah, so it's really interesting because this is an idea that started, I think about five, six years ago when I was a member, I don't even remember if I was a member or the chair of the Critical Cat Network, and we were working through areas in which there was variability or uncertainty and
Speaker: we got the invitation to submit proposals for guidelines.
Speaker: And this is one that actually was prioritized.
Speaker: It took a little bit of time to get it done because there was some hiccups along the way, not mean something that typically happens.
Speaker: And the reason why it was impetus to work on this was, number one, we saw a lot of variability in clinical practice.
Speaker: And number two is, which we think is the most important is,
Speaker: We, at least based on experience and some cursory review of the literature at the time, there was really no evidence that there was benefit from transfusing either platelets or FFPs.
Speaker: However, we all know about the complications of giving this.
Speaker: And probably the most extreme scenario are patients, for example, who have received multiple transfusions over a period of time for a number of reasons, like call it
Speaker: patients who have multiple transfusions for bone marrow transplant, et cetera.
Speaker: And these patients become allo-immunized.
Speaker: And in those patients, actually the side effects and the complication profile is a lot more severe than for somebody who might receive one transfusion or not.
Speaker: But ultimately, I think that transfusion of multiple blood products over time can have the serious consequences.
Speaker: And we kind of pose this question
Speaker: If we put this on the scale about the risk and the benefit, is there so much upside to this?
Speaker: Are we really preventing complications from happening or are we mostly doing things because we feel they are probably the right thing to do without any evidence to support it?
Speaker: But then also, are we really preventing the dreaded complications that we're trying to prevent?
Speaker: with the concept that transfusing this is going to help.
Speaker: Absolutely.
Speaker: And like you mentioned, it's important for our clinicians to be reminded that blood products have consequences that can be potentially very serious for our patients.
Speaker: But also, in addition to that, there's cost, and these are not unlimited resources.
Speaker: So being more judicious about how we use them and being good stewards of these resources will also allow them to be available for patients who really need them or who would really benefit from them.
Speaker: So I think it's definitely a great area to focus a clinical guideline on.
Speaker: Angel, I would like to ask you about the clinical guideline process.
Speaker: So you mentioned the genesis of how this idea came up within CHESS and the American College of Chess Physicians.
Speaker: Could you give us an overview of the methodology that was used for this particular guideline?
Speaker: Yes.
Speaker: So it is a very strict methodology.
Speaker: CHESS adheres very strictly to great methodology.
Speaker: So what this consists is first is from the beginning,
Speaker: experts are identified and the panel is vetted, then these panelists are reviewed for conflicts of interest or relationships that could preclude them from participating on the guideline or could bias their judgment, meaning playlists and FPs is not too much of a problem because there's not a whole lot of
Speaker: typically conflicts like for example with other guidelines that might involve medications that could be sponsored by pharma etc.
Speaker: Then once the panel is convened then we go in the development of the PICO questions and PICO questions followed the format of patient or population intervention competitor and then the outcome.
Speaker: We
Speaker: select the intervention in this case in most of the cases is the intervention is transfusing versus the outcome or the comparison would be not transfusing then the population will be depending on the question patients generally critically ill patients or for example in the case of procedures patients who are undergoing placement of blah blah central line arterial line etc and then the outcomes we
Speaker: stratify the alcohols based on we rank them.
Speaker: So for example, if we look at
Speaker: the complications of or the outcomes for transfusions we're looking for example at mortality we're looking at bleeding events we're looking at for example in the cases of central line maybe bleeding complications or mechanical complications or infectious complications etc so all those outcomes are ranked and then we select what we call the critical outcomes and those are the ones that
Speaker: are going to be informing the recommendations.
Speaker: So then after that, the methodologists that work with us that are the chest provides that methodologic support, they do the data search.
Speaker: So basically, we screen, I don't remember exactly how many we screen for this
Speaker: because it was two guidelines, the RBCs and the one in FFPs.
Speaker: But it's probably somewhere in the neighborhood of a few thousand articles.
Speaker: And the first step is to screen them by title, meaning, because when search terms are introduced, you would put a lot of studies that might be pediatrics, that might be veterinary, that might have different type of outcomes, or maybe are letters to the editor.
Speaker: The job, all these articles are screened by the panel.
Speaker: They're screened by one of the methodologies and one of the panelist members.
Speaker: And when there are discrepancies, then we hash those out to determine whether the discrepancy was.
Speaker: So from that, that's the first step in which there is a lot of narrowing down of the amount of articles that will be going to the next step.
Speaker: The next step is a full text review, which is all the articles are now
Speaker: that were initially selected, now they're looked at and determine if they are good for inclusion and to which pick a question they will be more applicable.
Speaker: After that, the methodologists do the meta-analysis of all these individual outcomes that we ranked as critical, looks for all these data in all of these studies, and then produces a meta-analysis.
Speaker: Once we have the meta-analysis in all of them, the group convenes again, and we discuss the whole evidence-to-decision table that basically looks at, it's basically how the sausage is made.
Speaker: It basically looks at what does the intervention have any beneficial effects, and then we grade how large, small, trivial, then also for the...
Speaker: Adverse effects.
Speaker: Then we also look at aspects of cost benefit.
Speaker: Then we look at aspects of feasibility.
Speaker: We look at aspects of effects on equity.
Speaker: And based on all of this, the panel issues and the recommendation that we have five choices, strong recommendation for an intervention,
Speaker: a conditional recommendation for an intervention and neutral recommendation basically the panel is saying you can do whichever their equivalent and then conditional recommendation against intervention and then strong recommendation against intervention typically to have a strong recommendation the data needs to come down from randomized controlled trials and one one
Speaker: step that I may have skipped is when we're looking at the great evidence, typically randomized controlled trials start as strong level of evidence and anything that is not a randomized controlled trial starts at a weak level of evidence.
Speaker: from there we look at different variables like um and this is more in the methodology so i'm not going to go into details but typically uh the uh data is upgraded or downgraded i've rarely seen not having worked on several guidelines that any evidence is upgraded is more about the things that make this specific evidence imperfect so they are downgraded so whenever we see that accommodation is
Speaker: low certainty of evidence or very low certainty of evidence, doesn't necessarily mean that the panel does not feel that the recommendation is something that we should be doing.
Speaker: It's just how the GRAE methodology informs the decision making.
Speaker: And anything that comes from randomized controlled trial starts as strong, but then typically gets downgrade to moderate and low.
Speaker: There's another guideline that I'm working on right now that means even despite having
Speaker: All the data on a specific question being randomized trials, it goes on to low rate of evidence because of some of the nuances that go into the great methodology.
Speaker: And everything that comes from observational comes from, um, starts as low.
Speaker: Rarely, I'm yet to see that some evidence gets upgraded.
Speaker: But then based on that, those recommendations, then we draft the manuscript and then sent to review to the guidelines oversight committee, Chess, then to the president.
Speaker: And then after that, it goes to the journal for a separate peer review.
Speaker: So it goes through several layers.
Speaker: So it's a really humongous amount of work.
Speaker: It's daunting.
Speaker: really something that is really gratifying once you see it through after working on those projects for several years.
Speaker: And Angela, as you mentioned, the level of evidence obviously is based on what's available.
Speaker: And by, I guess, not by definition, but by design, you're often looking for PICO questions and clinical guidelines that don't have an overwhelming level of evidence because then obviously you would know the answer what to do, right?
Speaker: So a lot of this is looking in areas where maybe the available evidence is imperfect.
Speaker: But
Speaker: From a perspective of the recommendations itself, the strength of the recommendation, in simple terms for clinicians at the bedside, how do you think of we recommend versus we suggest or a strong recommendation versus a conditional recommendation?
Speaker: How would you apply that to clinicians at the bedside, patients, and maybe even policymakers?
Speaker: Exactly.
Speaker: So I think this is really important because when we...
Speaker: when panels issue a strong recommendation, and a strong recommendation can be done, for example, with low certainty of evidence.
Speaker: And I want to digress a little bit here to just provide some explanation.
Speaker: And surviving sepsis, for example, I think historically the recommendation on antibiotics has been strong,
Speaker: But if I'm not mistaken, it's always been a low grade of evidence.
Speaker: And it's because we're never going to have a randomized controlled trial that is going to evaluate antibiotics versus no antibiotics, right?
Speaker: That's just unethical.
Speaker: But by design, this will make it a low certainty of evidence.
Speaker: Now, going back to your question as far as how should we apply this at the bedside, whenever a clinician who is reviewing guidelines sees that a recommendation is a strong recommendation,
Speaker: recommendation meaning typically that we recommend that should give us pause because it's something that we should consider doing strongly again guidelines are guidelines and as the name says they are meant to provide guidance but they are not meant to be the end all be all because they're nuances right and that is why the clinician at the bedside needs to see if the recommendation fully applies to their patient
Speaker: When we look at the we suggest or the low certainty of evidence, that is basically saying number one,
Speaker: there is not enough or there is not strong enough evidence.
Speaker: However, the panel chose to make a recommendation because at the end of the day, the panel also has the task as typically composed by experts to provide guidance after looking at all the literature available to clinicians at the bedside.
Speaker: It would be a lot of times we debate whether we should just say provide a neutral recommendation.
Speaker: And sometimes that is done.
Speaker: However,
Speaker: that also poses a problem because the clinician at the bedside will have absolutely no guidance because despite being a recommendation, it really does not provide a whole lot of guidance.
Speaker: So,
Speaker: In summary is, if it's a strong recommendation, it's probably that you should strongly consider doing unless there are reasons why you could not or you should not do it based on your assessment of the patient.
Speaker: And then when we suggest with a low certainty of evidence, that's something that we know there's not enough data and nobody should be helped.
Speaker: that they did something wrong by not following this recommendation necessarily.
Speaker: And the same applies to policymakers, right?
Speaker: Because policymakers use their recommendations to determine a lot of things, right?
Speaker: Like core measures and even some, has some downstream effects to the revenue aspect.
Speaker: But again, only strong recommendations should have this implication.
Speaker: The recommendations that are not strong should not have this implication.
Speaker: Excellent.
Speaker: Let's dive into the actual clinical guidelines.
Speaker: And what I really liked is how the team framed the guidelines into kind of general ICU patients, which we'll discuss first.
Speaker: And then second, they talked about procedures, which obviously are very relevant for the practice of critical care.
Speaker: So why don't we talk about general ICU patients first?
Speaker: And what I would like to is just go through some of the questions that you addressed and you can give us a little bit about what was the recommendation, any justification that you think is appropriate and any additional comments.
Speaker: Sure.
Speaker: Sounds good.
Speaker: So the first question is, should critically ill patients with thrombocytopenia receive transfusions of platelets?
Speaker: And I guess this can be discussed in the distinction between low risk of bleeding versus high risk of bleeding.
Speaker: Correct.
Speaker: And that was a million-dollar question because as we framed this, and this stemmed from the viabilities that we saw in clinical practice on the thresholds to transfuse platelets.
Speaker: And I think universally, we know that, and again, this is, for example, data that there's not specific data in critically ill patients.
Speaker: It's more data that is from general or even hematology oncology patients.
Speaker: But the bottom line was what I think the
Speaker: Consensus is for most is 10 is kind of the magic number by which if we keep the playlist lower than that, there is a higher risk of significant bleeding, especially intracranial bleeding.
Speaker: Then we look at the high risk of bleeding.
Speaker: And by that, we dissected that in two different ways.
Speaker: One is that the patient has inherent characteristics that would make them more prone to bleed.
Speaker: And the other one is that the patient has the potential to bleed in an area in which bleeding could be catastrophic.
Speaker: Like, for example, although we did not cover CNS, meaning one good example would be any CNS type of bleed, the ones that we included and thought were, for example, airway lesions in which the patient has a tumor that has a high propensity to bleed, those kind of things.
Speaker: So when we look in the literature as far as how to define high and low risk of bleed,
Speaker: it became very challenging because there was really not a good definition for low or high risk of bleeding outside the ICU, let alone the ICU.
Speaker: And the only data that we found was the IMPROVE score, which is a score that looked at, it basically looked at different patient characteristics and tried to determine if a patient had a high risk of bleeding.
Speaker: Again, this was a
Speaker: Later validated in critically ill patients, but it was not necessarily something that was specifically designed.
Speaker: So based on that, we really couldn't make strong recommendations on that because, uh,
Speaker: Number one, the data was not necessarily completely related or centered on critical ill patients.
Speaker: It was mostly medical patients or patients who were admitted to the ICU.
Speaker: I'm sorry, to the hematologic oncology floor.
Speaker: And then as far as the cut-offs, then the panel had a lot of discussion because at the beginning is the big question is, should we recommend cut-offs?
Speaker: Should we say a number?
Speaker: And
Speaker: It was less complicated when it came to the high risk of bleeding patients, without high risk of bleeding, I'm sorry, because then that's kind of the commonly used threshold.
Speaker: But then for the patients who are at a high risk of bleeding, we knew it had to be a little higher.
Speaker: The evidence, there was really not much.
Speaker: So we took several hours debating this and the panel concluded that it should be over 30,
Speaker: somewhere between 30 and 50 is reasonable.
Speaker: Just because there's not a whole lot of evidence, the panel chose not to say
Speaker: specific number.
Speaker: That's why there is a range there.
Speaker: So that was very tricky.
Speaker: And again, when we look at these recommendations, it was, it was really not a whole lot of evidence of benefit.
Speaker: There was some evidence for potential harm.
Speaker: So again, if we're looking at a patient that does not really have a high risk of bleeding, I mean, I think 10 is fine.
Speaker: And for patients who have a high risk of bleeding, which could be
Speaker: determined by either characteristics that put the patient at a higher risk of bleeding or the patient has a potential area in which if bleeding were to cure, would be catastrophic, those we said 30 to 50.
Speaker: And I think it's important also as a reminder to clinicians that just doing the same thing for every patient eventually can get us into trouble, right?
Speaker: And that's why it's important for us to sometimes pause and think about the individual patient
Speaker: And like you said, this patient might have a higher risk of bleeding.
Speaker: And again, people might differ on how they classify that.
Speaker: But there are certain common areas that we all would agree would put somebody at a higher risk.
Speaker: And maybe those patients, you are a little bit more aggressive with transfusing platelets.
Speaker: But what we also know is that for the vast majority of patients who might have low platelets for multiple reasons and critical illness who are not at really high risk of bleeding,
Speaker: Unless it goes below 10,000, the risk of a severe spontaneous hemorrhage is very, very low.
Speaker: So there's no really point of transfusing at that point, correct?
Speaker: Exactly.
Speaker: I think you summarized it really well is for most patients, then is the most reasonable.
Speaker: And then for the other ones, I think is individualizing the approach using the clinician assessment at the bedside of what specific factors may that patient have that would require.
Speaker: And again, I think this opens the door for, I think one area in which patients
Speaker: guidelines are criticized a lot is because guidelines or clinicians says that they're being told what to do by guidelines.
Speaker: In this case, this guideline, in my opinion, gives the freedom to the clinician to say, hey, in this case, I think my patient has a high risk of bleeding.
Speaker: I can justify using a higher threshold because I think if my patient who has a high risk of bleeding were to bleed, outcomes could be catastrophic.
Speaker: So I think this recommendation gives the opening for that.
Speaker: Perfect.
Speaker: Is there any specific population that you want to address, Angel?
Speaker: In general, this was mostly all critically ill population.
Speaker: I think the population, if I wanted to emphasize specific population, are maybe the populations in which we have to be even more judicious about transfusions.
Speaker: Patients, for example, in which we know that giving platelets can do even more harm than good.
Speaker: Like, for example, patients who have ITP, TTP, who are highly immunized.
Speaker: So, or patients who have leukostasis, for example, those patients are more likely to be harmed by these transfusions than benefited.
Speaker: So, I think those populations are very, very unique.
Speaker: And those are just a few examples of how we should individualize the decision-making for our patients.
Speaker: The second question to this generalized U population pertains to patients who are actively bleeding.
Speaker: So should critically ill patients with thromocyteopenia and active bleeding receive platelets?
Speaker: How did you respond to that PICO question?
Speaker: So we look at all the evidence as far as transfusing platelets on patients who are actively bleeding.
Speaker: And then we came also into a conundrum looking at
Speaker: what active bleeding actually merits transfusion right because it's different if you have a nosebleed versus some sort of a gi bleed that is requiring transfusions right so completely different so
Speaker: We look at all the data and basically notice that the only available data for the most part was in patients who had very sealed bleed from full hypertension and cirrhosis.
Speaker: And what was identified is that reading platelets counts did not affect the results and even the re-bleeding rates were even higher in patients who had received platelets.
Speaker: It was a little bit paradoxical data, but at the end is looking at all the data, what we decided is to, number one, use the existing classification by the WHO on bleeding, kind of degrees of bleeding.
Speaker: Like for example, grade one is epistaxis or purpura, things that are very minor or microscopic hematuria, something for which we would typically not transfuse.
Speaker: Then grade two is epistaxis that lasts more than 30 minutes, maybe some hematemesis, but not producing hemodynamic instability.
Speaker: Then grade three is bleeding that is requiring red blood cell transfusion over routine transfusion needs or bleeding that is causing some degree of hemodynamic instability.
Speaker: And then
Speaker: Number four is basically the ones that we fear in the ICU, the ones that basically start bleeding and you cannot stop.
Speaker: So in those cases, for example, one of the other examples that came, what if somebody is bleeding from the nose or somebody is bleeding from a central line insertion site or an arterial line?
Speaker: In those cases, I think most clinicians would agree that maybe the more sensible approach would be let's apply pressure for a reasonable amount of time, 20, 30 pressure.
Speaker: Let's tamponade that nosebleed.
Speaker: and reassess.
Speaker: And for bleedings that are mild, like grade one or two, I think we should, by all means, prioritize non-transfusion interventions that are going to potentially stop the bleeding, like compression, et cetera.
Speaker: If those don't improve the bleeding, then we should consider the transfusion to 50.
Speaker: And then for
Speaker: I think there is a lot less debate or hesitation that probably 50 is the most commonly accepted threshold to transfuse platelets.
Speaker: So the take-home message here really is to try to, in an objective way, evaluate the severity of bleeding.
Speaker: And there are plenty of patients who might have bleeding that is, like you mentioned, Angel, probably more suitable to some sort of local compression or other treatments to try to stop the bleeding.
Speaker: And when we have more severe bleeding, that requires...
Speaker: higher levels of support and transfusions, those are the ones that we might consider giving platelets.
Speaker: And the target here would be keep them above 50,000.
Speaker: Is that correct?
Speaker: Absolutely.
Speaker: You got it.
Speaker: The second part of the guidelines really focused on procedures, and this has always been not only an issue of tremendous variability within the ICU among intensivists, but also it's not uncommon for our patients to get procedures done by other colleagues, right, interventional radiology or some other colleague, and it's not uncommon for other colleagues to always work
Speaker: want coagulation corrected, they want platelets corrected.
Speaker: And my sense is that over the last several years, that has led probably to an overuse of both platelet transfusions and fresh frozen plasma.
Speaker: So could you tell us how you thought about and how you classified and kind of approached the procedural aspect of this question?
Speaker: Yeah.
Speaker: So we look at different procedures in the ICU that are commonly done at the bedside.
Speaker: As you said, there is a lot of variability.
Speaker: Like, I mean, even among our own colleagues from our same group, in some might feel more comfortable with a specific number.
Speaker: And a lot of that comes from experience.
Speaker: And typically in
Speaker: I think this probably has happened to all of us.
Speaker: We will remember forever a bad complication we had after a procedure, and that tends to inform our practices moving forward.
Speaker: But also it's the other practitioners that are outside our department, right?
Speaker: We need to have a tunnel line to be placed by intervention radiologists.
Speaker: They may have a complete different practice pattern regarding transfusion.
Speaker: So we try to look at all of this, and we...
Speaker: kind of found some trials and there was actually a recent trial in New England that for the most part informed a vast chunk of the recommendation and that's how we came with the recommendation the way it was.
Speaker: Again, unfortunately, because it was only one large randomized trial and there was some discrepancy with the other trials ended up being a low certainty of evidence despite starting as a randomized controlled trial.
Speaker: So, and I think the important message here is that
Speaker: we need to make sure that different factors are considered because the data that has looked at this is two small and one larger on the mild control trial.
Speaker: A lot of these studies excluded patients with INR more than three.
Speaker: So meaning if you have a patient who has thrombocytopenia and an INR more than three, that is when the clinician at the best side judgment is paramount to make those decisions.
Speaker: But if it's something isolated like just thrombocytopenia or just
Speaker: an elevated INR.
Speaker: And that's how we define coagulopathy for these guidelines.
Speaker: There was really, there's so much variability in the definition of coagulopathy that it became difficult.
Speaker: We just took a pragmatic approach, which is the way that most trials in forming these guidelines defined coagulopathy or the endpoint around the needs to transfuse FFP.
Speaker: So that is how we came up to the recommendations.
Speaker: And basically is, uh,
Speaker: You don't need to transfuse specifically for central lines or A lines.
Speaker: I think what you should look at is we should look at the inherent patient transfusion risks and use that as a decision point.
Speaker: Again, acknowledging that the viability practices around the country and around the world, for example, it might be different if there is
Speaker: variability in operator experience, like maybe it's in the middle of the night in a setting in which the person placing the light may not be as experienced.
Speaker: That needs to be factored in.
Speaker: Whether or not there is availability of ultrasound, that also is important, although most ICUs now have ultrasounds.
Speaker: But we're thinking about the parts of the world that are well resourced, but that may not be true in other parts of the world.
Speaker: Then again, the concomitant presence of elevation in INR and also thrombocytopenia.
Speaker: And then the other factor is, for example, if that line is going to be tunneled or if we are going to
Speaker: place the catheter in a site in which compression is not feasible, like the subplavian, for example, then also dose mandate different approaches.
Speaker: But in general, the overall recommendation as the first line should be, we do not need to transfuse platelets or FFPs prior to a central line.
Speaker: What we need to do is look at the patient's individual risk factors, use that as a decision point.
Speaker: But again, consider the nuance of all the different
Speaker: factors like I just mentioned earlier.
Speaker: But like you mentioned, for the majority of patients, we do not need to transfuse FFP or platelets for a central line.
Speaker: And then obviously, like you mentioned, practice has evolved.
Speaker: And in most places, at least here in the U.S., there is ultrasound availability, but we also obviously are creating guidelines and thinking of
Speaker: other places that might not have the same resources.
Speaker: So just to individualize at that point.
Speaker: What exactly?
Speaker: And sorry to interrupt you.
Speaker: One more area that we thought would be important as a future research priority was the use of viscoelastic monitoring.
Speaker: I mean, it has different names depending on what specific test you use and different centers.
Speaker: But there is
Speaker: I mean, growing data of this in the trauma relation, there's growing data on this maybe in the GI bleeding world around liver patients.
Speaker: I think this potentially would have a role in the future of something that we could use as a parameter to determine whether we would consider a transfusion prior to a procedure or not.
Speaker: What was the discussion around thoracentesis and paracentesis?
Speaker: So in this one, it was really interesting because the data, and again, in this case, the data was mostly retrospective.
Speaker: And as you know, there is a publication bias, right?
Speaker: There is the big knowledge that data on bad outcomes tends to be published more than data that does not show that, that doesn't show any benefit from an intervention, right?
Speaker: So the data that we found basically
Speaker: What was an equivocal was that complications from these procedures are extremely rare.
Speaker: To the point that when we discussed the area of research priorities, we came to the conclusion that studies to assess and give a solid answer on this are practically impossible because the complication rates are so low that you will need so large
Speaker: samples, that those studies would not be feasible.
Speaker: So all that being said, knowing that the complications of either Thora or parasynthesis being low-risk procedures are so low and that there is potential harm by giving either of these two interventions, the decision was to
Speaker: suggest against routine transfusion prior to a thorough paracenties.
Speaker: And I want to take a little pause here to preface this, that in the guideline we made very clear that what should inform, so when we go from the section of general critical patients to the section of procedures, we preface this by saying, this should, the overall risk of bleeding and overall decision to transfuse a patient
Speaker: should be assessed first before the individual need to transfuse for a specific procedure.
Speaker: And ultimately, that should, in the patient's individual needs, based on the risk, should inform all the decisions, even before we discuss the procedural need.
Speaker: And I think it's a good exercise, right, when we're making decisions based on risk-benefit to risk stratify our patients individually.
Speaker: Not every patient has the same risk of a transfusion like you mentioned earlier, and including that into our thought process and the equation of our decision-making is extremely important.
Speaker: Absolutely.
Speaker: So now comes my favorite procedure, not necessarily in terms of the procedure itself, but in terms of the discussion about platelets, which is lumbar punctures, which we don't do as frequently, but when they're needed, should be done.
Speaker: And this might be something that a lot of intensivists send to IR, so they might get a lot of pushback.
Speaker: But what is the finding of the discussion by the Clinical Guideline Committee regarding the transfusion of FFP and platelets for patients undergoing lung bar puncture?
Speaker: Yeah.
Speaker: So at the beginning, when we look at the data, we actually identify a lot of studies in patients who were obstetric.
Speaker: and oncologic populations and epidural catheter use.
Speaker: And we evaluated all this data and determined that it was not necessarily something that would be applicable to our patient population or to our population of critically ill patients.
Speaker: So we excluded those patients.
Speaker: Then basically the only two studies that made the criteria were
Speaker: assess the outcomes of traumatic lumbar puncture or spinal hematoma.
Speaker: And again, spinal hematoma was very rare.
Speaker: It was very rare, it's less than 0.2% or something along those lines.
Speaker: So, but in this case, the potential complications, despite not being common, could be potentially catastrophic.
Speaker: We're looking at potentially producing a spinal hematoma and producing a patient to be
Speaker: paraplegic because of a complication like this.
Speaker: So this is a good example of kind of when we look at the high risk of bleeding, low risk of bleeding that we discussed earlier in this case is yes, the complications rate are low, but if they were to happen, it could be catastrophic and something that could potentially have been prevented.
Speaker: Again, the data showing that it prevents this is not strong, is not conclusive.
Speaker: But the way the panel approached this is, this is how practice has been, because there's even a very nice survey that was published that showed that over 90% of clinicians would transfuse for thrombocytopenia or an INR more than two.
Speaker: So that is kind of common practice.
Speaker: So if we were going to suggest switching that, we felt that we needed to have strong evidence to say,
Speaker: This is the reason why we should not be doing it.
Speaker: And in that case, that evidence was not there.
Speaker: So that's why we decided as a panel to say that we should aim for an INR around two or lower and platelet counts around 40 to 50 if we are going to do a lumbar puncture in a patient in the ICU.
Speaker: This is data that, for example, might not be suitable for a randomized controlled trial, and maybe more registry type of data might be something that could be useful to looking at this, or maybe, again, the role of viscoelastic testing to determine if transfusions would change this risk of bleeding.
Speaker: And it's important to reemphasize that lumbar puncture is approached differently than the other procedures that we mentioned, thoracentesis, paracentesis, central venous catheter and arterial line, for all the reasons that you mentioned.
Speaker: And again, studying these questions with randomized controlled trials is not only very hard, but unlikely to happen in the near future for all the reasons that we discussed.
Speaker: Exactly.
Speaker: And that's the difficult part because although we suggested priorities for research, we know that it's going to be very difficult to do them.
Speaker: And because of that is why we chose to make, despite being weak, recommendations or suggestions for or against an intervention because we felt that
Speaker: it was better to issue a recommendation, even if the evidence or the certainty of evidence was not high.
Speaker: It's better to issue this so that clinicians at least have something that is providing them guidance, as opposed to saying we don't have enough evidence, so the evidence is not as strong, but we wanted to make sure that we were providing guidance to clinicians at the pet site.
Speaker: Finally, you did discuss as well bedside endoscopy, specifically bronchoscopy and GI endoscopy, which are both common procedures in critically ill patients.
Speaker: Could you share with us what was the recommendation from the panel?
Speaker: Yeah.
Speaker: So for...
Speaker: Bronchoscopy, first of all, we need to narrow it down to the bronchoscopy that happens in the ICU, right?
Speaker: So we very explicitly said we're not talking about advanced bronchoscopy that will require the use of more sophisticated tools or biopsies, et cetera.
Speaker: So this is the typical airway inspection or a typical airway inspection with BAL that is done in the ICU.
Speaker: And for those, it was basically only
Speaker: two retrospective studies that evaluated platelet transfusions prior to bronchoscopy in the ICU and also in cancer patients with different transfusion thresholds.
Speaker: But ultimately, the conclusion was that transfusions of platelets prior to a bronchoscope did not change the risk of complications.
Speaker: And when you look to FFPs, there were no studies that
Speaker: look at FFPs prior to bronchoscopy.
Speaker: So the panel kind of weighing all the potential benefits of transfusing, which really there's not much because it's a low risk procedure.
Speaker: However, the potential complications of platelets and FFPs are not zero.
Speaker: There are some, especially
Speaker: the more volume we transfuse.
Speaker: So that's why we decided to issue our recommendation saying we suggest against routine transfusions for patients.
Speaker: Again, prefacing that it's important for everything that should be prefaced by the overall assessment of the individual patient's risk of bleeding.
Speaker: Are there any comments that you can share regarding other clinical guidelines on the same topic?
Speaker: It's interesting that a couple of weeks after you published the clinical guidelines, or I saw a different clinical guideline in JAMA from heme societies, they talked about patients that were not certainly specifically critically ill in other populations, but there's obviously other clinical guidelines out there as well.
Speaker: These are unique in that they were specific for critically opatients, which is what we care about on this podcast.
Speaker: But any comments, Angel, on how they kind of align or differ from other big guidelines?
Speaker: Yes, I think we look at different other guidelines.
Speaker: And again, these guidelines are not used to...
Speaker: guide our decision making.
Speaker: We make decision making independently looking at all the data that we identified.
Speaker: This is including the discussion part kind of to compare what we are recommending and why maybe we are agreeing or not so much agreeing with other societies in different areas.
Speaker: So we look at the guidelines from the Association for the Advancement of Blood Therapies, the AABB, the American Society of Clinical Oncology, and also the British Society of Hematology.
Speaker: There's also the European Society of Intensive Care Medicine.
Speaker: And for the most part, there was quite a bit of agreement on that type of recommendations.
Speaker: For example, when we're looking about, we're talking about patients
Speaker: As far as recommendations for general critically ill patients, basically our recommendations align with the AABB and the American Society of Clinical Oncology, and even for the European Society of Intensive Care Medicine when it comes to patients with a high risk of bleeding.
Speaker: I think we're all in agreement that 10 is, if there is a number that we should probably use is the 10.
Speaker: But then for patients with high risk of bleeding,
Speaker: Some of the other guidelines recommending a higher threshold, but did not specify
Speaker: why or what that threshold would be.
Speaker: So we wanted to go a step further there and say, these are the things that you should consider as higher risk of bleeding.
Speaker: And in those patients, we're not going to stop and just say, use a higher threshold.
Speaker: Let's go not to the 50, which is what would be used for active bleeding in most patients, but somewhere in the middle, 30 to 50.
Speaker: Again, judgment of the clinician at the bedside is really important.
Speaker: Then when it comes to the active leading patients, I think there was some data from the International Society of Thrombosis and Hemostasis that basically recommended 50 in patients who were in DIC and bleeding.
Speaker: A little bit different than the population we were looking at, but that's the data that we identified.
Speaker: And then, for example, the ESICM did not make a recommendation in patients who were bleeding patients.
Speaker: That was interesting and we kind of decided to make a recommendation, so in that case.
Speaker: And then when it came to the specific procedures, the recommendations were variable again, because we were not only looking at now recommendations from AAVE and hematology societies, but we were also looking at societies, for example, for interventional radiology.
Speaker: And I think what most people kind of agree is
Speaker: we should not be transfusing at all in patients who have over 50.
Speaker: And I think that makes sense.
Speaker: Now, between 10 and 50 is when the rubber meets the road.
Speaker: And I think the variability in our case, we said,
Speaker: no need to transfuse, assess the patient individual risk factors, right?
Speaker: So if the patient is at low risk of bleeding, we would say 10 is fine.
Speaker: Basically high risk of bleeding, you may want to keep the patient somewhere in the 30 to 50, and then use that as your baseline to determine the needs for transfusion for the procedure.
Speaker: But for example, the interventional radiology side is recommended 20, which might seem as a disagreement with ours, but not necessarily because we're saying
Speaker: we're looking at a patient with high risk of bleeding, we're saying maybe 30 to 50, but again, it's very nuanced.
Speaker: And I think this is where the clinician looking at the patient, looking at the individual factors and all the other parameters that are informing that need to place a line.
Speaker: If it's a tunnel line, if there's concomitant, coagulopathy, et cetera, those are all factors that should be considered.
Speaker: When it came to the fora and para, I think it was unanimous, nobody believes or shouldn't say that.
Speaker: several professional societies classified lower and parsing this as a low risk procedures and there was no recommendation to routinely transfuse.
Speaker: Again, for lumbar puncture, it was very similar in the sense that most societies say, despite a low risk of bleeding from this procedure, the complication is
Speaker: potentially catastrophic.
Speaker: So most of them recommend similar numbers to what we are recommending.
Speaker: And this includes the site interventional radiology as well.
Speaker: And then for bronchoscopy, there was not a whole lot of data from other societies for bronchoscopy.
Speaker: Finally, as we close, putting everything together for the clinicians at the bedside, could you share common pitfalls that clinicians should avoid?
Speaker: Yes, I think number one would be to transfuse to a number.
Speaker: Although we recommended numbers or ranges to provide some guidance, these are not the end all be all.
Speaker: I think that's when we need to look at our patients individually and look at the characteristics to decide if the specific recommendation would apply to them.
Speaker: Number two is we should routinely, or we should get away from the routine
Speaker: use of transfusions prior to the procedure just because a number is low, et cetera.
Speaker: I think in these cases, we need to individualize the care of the patient.
Speaker: And although guidelines are supposed to be meant to, in a way, homogenize the care, actually what we're saying is less homogenizing, but what needs to be homogeneous is the assessment of the individual patient characteristics to determine if there's a need for transfusion or not.
Speaker: And I think the other part that I think is really important, which not necessarily a pitfall, but it's something that I wanted to emphasize is, as you mentioned earlier, playlist and FFPs are scarce resources, they're precious resources.
Speaker: And even talking not necessarily about the cost is if we were to transfuse less using a more judicious approach, we will cut down the number of units
Speaker: that are transfused per year very significantly.
Speaker: I think we did calculations and it would be somewhere in the neighborhood of half a million units per year.
Speaker: And this is important because there are patients who truly need the transfusions, patients who have conditions that will need recurrent transfusions.
Speaker: So in those patients, actually by transfusing patients who may have
Speaker: an indication that it's less solid, and we divert those precious resources to those patients and away from the patients who would truly need it.
Speaker: Ultimately, we are potentially worsening issues of accessibility and also equity because there is data that shows that patients of African-American descent have, African-American ancestry, I'm sorry, have, when it comes to the
Speaker: pool of blood, they are less likely to have units that are more compatible with them than patients from other ethnicities.
Speaker: So it's also important.
Speaker: So it has so many layers in which it can affect patient care just from, number one, having the unit, but also having the correct unit for specific populations.
Speaker: Any pearls of wisdom when we consider transfusion platelets and FFP for our clinicians?
Speaker: I think it's probably to really think about, do I really need to transfuse this patient?
Speaker: Am I really transfusing it because I think it's going to help the patient?
Speaker: Or am I transfusing this unit because that's what I've been doing?
Speaker: But now that I can review the data that shows that maybe there's not much benefit, is that something that I really need to do?
Speaker: And from my perspective, Angel, I think the take-home message really is to pause and be more thoughtful about our individual patients and not just react to a number on a screen and to really think about what are the risks they have for bleeding, what will be potential risk of transfusions in this particular patient, and
Speaker: What are the risks they have based on other comorbidities?
Speaker: And really, like you said, take a little bit more of a thoughtful approach to when we will transfuse.
Speaker: And what it really indicates also is that the available evidence would suggest that we should probably be transfusing a little bit less platelets and less FFPs.
Speaker: But when we do so, do it in a much more deliberate and thoughtful way.
Speaker: Mm-hmm.
Speaker: Totally agree.
Speaker: Thank you for the guidelines.
Speaker: A lot of work goes into this, as you explained, and I know.
Speaker: And these are difficult topics because we don't have all the answers.
Speaker: But one of the things that I've always appreciated about guidelines is that they're wonderful places to read because they review the existing literature and not only give you, like you said, guidance, but allow you to learn more about the topic and what's available and what is missing.
Speaker: So I would encourage all our listeners to take a look at the guidelines.
Speaker: We'll obviously add a reference in the show notes.
Speaker: And as we close, Angel, we like to ask questions unrelated to the clinical topic.
Speaker: To tap in the wisdom of our guest, would that be okay?
Speaker: Absolutely.
Speaker: My first question relates to books.
Speaker: Is there a book or are there any books that have influenced you significantly or a book that you have gifted often to other people?
Speaker: Yeah, there is a book that I actually read.
Speaker: It was a couple, maybe over a year and a half ago, that really made me think again.
Speaker: And it's kind of redundant because the title of the book is Think Again.
Speaker: I don't know if you had the chance to read it by Adam Grant.
Speaker: And actually, it's very interesting because it challenges us to rethink the way we think and maybe kind of think again about our common ways in which we approach things, showing some flexibility in the way we approach clinical problem solving.
Speaker: And I think it's very easy after doing, for example, medicine or just our typical day-to-day things in life,
Speaker: to have a preconceived idea of what we're going to be facing or we're going to be encountering.
Speaker: And sometimes we may tend to anchor.
Speaker: And I think this book's uses kind of shows us, or at least it did to me, how to, by storytelling, really, how it's important to have flexibility in our decision making.
Speaker: Because that leads to, number one, make better decisions.
Speaker: Number two is be able to collaborate with people.
Speaker: And three is to, because we are thinking outside our usual walls, how to innovate and be better.
Speaker: So it's to be curious, to challenge what we know.
Speaker: And I think in medicine, at least we do this routinely, but sometimes because of kind of the day-to-day routine, we may fall into this little habit of
Speaker: the anchor and something because we kind of have seen this many times, but it's kind of looking at everything with a fresh mind.
Speaker: And I think that is, it's really, really, really something that kind of changed how,
Speaker: my way to approach certain things in life.
Speaker: So that was really, really, really a nice book.
Speaker: I agree.
Speaker: I had an opportunity to read Adam Grant's Think Again and agree with everything you said.
Speaker: For me, what was interesting from this book was it made me realize that as much as we like to call ourselves scientists, physicians in general, we do not think like scientists.
Speaker: we definitely hold strong to confirmation bias.
Speaker: And we usually look at the available literature to support those biases.
Speaker: And like you said, we have to be more humble and challenge more what we think is right and have an open mind because it's the only way that we really will do well by our patients, but also continue to advance our practice.
Speaker: So wonderful read.
Speaker: I will add a link in the show notes.
Speaker: Thanks for sharing that.
Speaker: Thank you.
Speaker: The second question I think is a perfect follow-up because it really talks about what we mentioned in the book, which is could you share something you changed your mind about over the last couple of years?
Speaker: Yes.
Speaker: And maybe it's too late now because I've been practicing for what I think 15 years now since finishing training and it's okay to say no.
Speaker: And I think as one finishes training and is...
Speaker: trying to establish a career, trying to establish, I mean, a reputation and trying to kind of expand your clinical acumen, your research profile, your educational profile.
Speaker: Sometimes we may take on, that's happened to me certainly, to take so many opportunities that at the point is you're really, your plate is too full that you,
Speaker: or overwhelm and start sacrificing other parts, other aspects of your life, like family life, personal life, self-care, exercising, meditation, if you practice that, because you have so many obligations.
Speaker: But at the end of the day, what I've learned is, unfortunately, by experiences, it takes you to be in a good state of mind, to care for yourself so that you can do everything.
Speaker: much better in the projects that you are signing up for.
Speaker: But again, that takes the ability to say no to a lot of offers that you might receive
Speaker: that maybe don't align with your long-term goals and kind of being more selective.
Speaker: And again, this might be a little, a lot harder once you, when you're starting, but as you become more experienced or seasoned or old, that's probably the better word.
Speaker: It's, it's a little bit easier, but I think it's kind of working with a mentor, working with somebody who kind of can help you kind of understand what your career goals are and kind of
Speaker: using the opportunities from all the offers that you may receive to kind of further that goal and not just kind of go into, kind of dilute yourself that ultimately you end up sacrificing in the overall, in the wholeness of the person and ultimately affecting what you want to accomplish in the first place.
Speaker: Excellent.
Speaker: And I agree.
Speaker: I think it's something that people usually say no out of fear and the fear is probably unfounded.
Speaker: And ultimately, you are sacrificing what's most important to you and where you can create the most value by adding too many things to your plate.
Speaker: So great, great comment, Angel.
Speaker: Thanks for sharing that.
Speaker: The final question is more like a closing statement.
Speaker: What would you want every listener to know?
Speaker: And this is something that actually was reflecting on this, I was thinking, and it's something that I tell my fellows and sometimes they kind of don't understand at the beginning.
Speaker: And it's the power of aggressive waiting.
Speaker: And I say that in that way, being to kind of raise their eyebrows and it's like, sometimes what we just need to do for a patient for that day is to wait, to see how the interventions have
Speaker: are going to change the course, but we need to give them time.
Speaker: Because if we're going in a zigzag motion,
Speaker: We're less likely to, one, benefit the patient, or two, see what the effects of the interventions we have made.
Speaker: And by saying this, I don't know me by any way that we should not be aggressive, or that we should not take care of our patients aggressively, and I mean, looking at all the information, but sometimes it's just take a pause and wait for the effects of interventions that we have prioritized.
Speaker: Sometimes that's hard, especially early in the career.
Speaker: And that's what I tell my fellows.
Speaker: That is the art of aggressive waiting.
Speaker: Just wait, but when I say aggressive, I mean be vigilant, not passive waiting, vigilant waiting.
Speaker: And if you see there are signs that things are not going the right way, by all means intervene.
Speaker: But sometimes just waiting makes a whole lot of difference.
Speaker: Aggressive waiting.
Speaker: I love that term.
Speaker: I haven't heard it before, but I agree 100%.
Speaker: Angel, thank you so much for sharing your expertise and your time with us.
Speaker: Look forward to having you back on the podcast to discuss other topics relevant to the practice of critical care medicine.
Speaker: It's my pleasure.
Speaker: I had a lot of fun.
Speaker: It has been really great and hope to be able to come again.
Speaker: I think it's really a great resource and I'm really happy that you're doing this to share with our peer critical care ICU practitioners, physicians, APPs.
Speaker: I think it's great and I commend you for that.
Speaker: Thank you.
Speaker: Thank you for listening to Critical Matters, a sound podcast.
Speaker: Make sure to subscribe to Critical Matters on Apple or Google Podcasts and share with your network.
Speaker: Sound's transforming the way critical care is provided in hospitals across the country.
Speaker: To learn more, visit www.soundphysicians.com.
