SCCM Resource Library-Anticoagulation Controversies in Extracorporeal Life Support

Video Transcription

Hello, thank you for taking the time to listen to this talk today. My name is Ryan Ravisecki, and I'm the clinical pharmacist in the Cardiothoracic Intensive Care Unit at UPMC  Presbyterian Hospital. My session today will be discussing anticoagulation controversies for extracorporeal life support. I have no relevant disclosures in relation to this  program or presentation. However, I will be discussing off-label indications of certain anticoagulation medications and how they relate to extracorporeal support.  Just to start off with some basics of the ECLS circuit, moving from left to right, you have the most traditional circuits with VV and VA ECMO. In the middle, you have hybrid  circuits looking at some VAV, and then lastly, on the right, parallel circuits. For the sake  of the discussions today, it's really just going to be the basics, thinking of the cannulas, the oxygenator, the pump, heater-coolers, and various connections between the circuits.  It's a very delicate balance of ECLS between bleeding and thrombotic complications. The picture on the left is of an oxygenator that clearly has a lot of visible thrombus present.  The picture on the right, however, is the flip side of that coin, which is a very large intracranial hemorrhage that was seen after a bleed. Up to 40% to 60% of patients during  ECLS support will experience a major bleeding complication. However, thrombosis is also relatively common, with up to 40% of patients experiencing a major bleeding complication.  Thrombosis is also relatively common, with upwards of 20% venous, 20% arterial clot, and perhaps one of the worst complications occurring in 4% to 8% of patients being in  ischemic stroke. Knowing that there's a very delicate balance between bleeding and thrombosis  and the extracorporeal support patient, I'd like to focus this talk on two main controversies  that I have identified. Controversy one, what is the optimal anticoagulant for a patient requiring extracorporeal life support? And two, is anticoagulation actually necessary  requiring extracorporeal support? So going into the first controversy, I'm going to discuss and  walk through a little bit of literature on the different types of anticoagulants that have been  used to support these patients. So why do I find that this is very controversial? I really think  it goes down to four key points here. One, there is a lack of a standardized protocol or guideline  from hospital to hospital, system to system, or even country to country. There's various uses of APTTs, anti-10As, ACTs, et cetera. Also, studies are typically limited to very small  sample sizes, heterogeneous patient populations, retrospective designs, and really convenience of  sample size and convenience studies. And lastly, there's a lot of different equipment and techniques  that's used from center to center. Is the likelihood of a bleed or thrombus the same  based on these pumps, based off the oxygenators? Is the cannulation strategy matter? A lot of this  stuff is very difficult to compare and really truly have an apples to apples comparison across  different levels of evidence. So I don't want to spend too much time on the table, but really,  I pulled this out of a systematic review by Cato and colleagues. And of these roughly eight or nine  papers, what I want to highlight is that the largest study that was sampled was 153 patients.  The patient populations that are seen here are a mix of VA, VV, some just VV altogether.  The anticoagulants that were used were mixed between heparin, bivalorudin, and some used both. And lastly, the anticoagulation targets throughout.  As you can see on the far right column, some are looking at ACTs, APTTs, TEGs. And lastly,  even if you were looking at the same APTT or the same ACT, the goals are different from study to  study. And lastly, of these studies identified, only one was randomized in fashion. So I thought  this is a really interesting group. The same group worked on this in 2013, and then 2021,  they repeated the same survey. In 2013, 100% of surveyed centers utilized unfractionated heparin  as their primary anticoagulant, with ACT doing the lion's share of the monitoring, with only  8% of centers saying that they did ever use an alternative agent. In 2021, what you can see is  that 94% of centers still say that heparin is their workhorse and do the majority of the work.  The ACT monitoring, however, has switched to anti-10A monitoring for the majority of those centers. And now 6% of centers are still using bivalorudin as a primary anticoagulant.  So over an eight-year stretch of time, there really hasn't been a dramatic shift in which anticoagulant is preferred in centers across the nation. So what is the main pharmacologic  differences between unfractionated heparin and bivalorudin? In terms of a mechanism of action,  the biggest difference is that bivalorudin is a direct thrombin inhibitor. It does not rely on the actions of antithrombin 3 to exhibit its anticoagulant effect. One of the common  criticisms of bivalorudin is that there is no reversal agent. There is no protamine like you  have for unfractionated heparin. The benefit, there is no heparin-induced thrombocytopenia  with bivalorudin. Some folks may be concerned with the idea of the smoke effect in bivalorudin  due to its metabolism of plasma esterases that could increase the risk of a thrombus in the  left ventricle. And perhaps the biggest critique of bivalorudin, it is the dramatically more expensive  than unfractionated heparin. So I like to lump these three studies together as the, quote, early studies comparing bivalorudin to unfractionated heparin from 2011 until 2013.  So you can see in Ranucci, Pieri, and Nagel, looking at 15 to 20 patients for most of them.  So really more proof of concept than really trying to drive change with these three papers. But these  were the papers that I think started to get things going. And when I say that, I like to say these are  the next waves of papers. So these are the three papers that I'm going to highlight for this section  of bivalorudin. And going right ahead in no particular order, start with the work that we've  done in our center, published in Critical Care Medicine. And this paper was looking at all VV  ECMO patients. As you can see in the plot on the left, this is the probability of no clot over time.  And bivalorudin was dramatically superior to heparin in that paper with a p-value of 0.01.  And going back to the beginning of the talk, looking at both bleeding and thrombotic outcomes, we were also able to see a decrease in major bleeding events from 41% to 12%  after the switch to bivalorudin. Again, this was a retrospective in nature.  And perhaps the biggest of the next wave of papers is by Dr. Seelheimer and his colleagues  in his group up there. And this is a really, really nice paper that looked at over 400 patients  altogether, one of the largest bivalorudin cohorts. What I wanted to pull out of this is,  again, retrospective in nature, lumping pediatrics and adults together, VA and VV patients together as well. But what's really awesome about this paper and those that  like to believe in bivalorudin is that Dr. Seelheimer was able to see reduced mortality  in the adult extracorporeal support patient that received bivalorudin, which is the first of all  of them that has been able to demonstrate a mortality benefit. And lastly, Erica Sheridan  and her group out in Kentucky looked at the same paper that a lot of us had looked at. They had a  N of 50 heparin patients to 100 bivalorudin patients. And despite that they were able to find that there was more time in therapeutic range, they got to therapeutic levels faster  in the bivalorudin group, they were unable to identify any difference in bleeding or thrombosis.  So looking at back all three of these papers together, you can still see retrospectives in  nature, different cohorts looking at VA, mixed VV and VA, and different results found out throughout.  So one thing I'd like to say here is other than the Sheridan paper, both our paper here and then Dr. Seelheimer's paper did show at least some benefit with the use of bivalorudin.  So as I said, one of the most common drawbacks of bivalorudin is the increase in cost. So  is that increased drug spend of bivalorudin really worth it? So a couple of papers have looked at this. Again, not a ton of sample size here. So in the first one with Dr. Hamza and  colleagues, you can see that bivalorudin was actually decreased in daily spend due to some  of the other benefits. There was no difference in Bieri and colleagues. Then Renucci and colleagues  did find that there is no difference in the adult patient, but there is a dramatic benefit in pediatric patients. I think looking at all of this, it's a little challenging to say,  is this definitive? And really need to hopefully look at some of these bigger studies like our own or Dr. Seelheimer's and say, is there a cost benefit in a true cost analysis?  So this is a great time to move into the next controversy that I wanted to discuss, which is  we know that we had a little bit of benefit from bivalorudin. It's possibly a little bit better  from an overall cost to the patient, but do patients need any degree of anticoagulation at  all? The first paper looking at this was a systematic review by Olson and colleagues in 2021  with the idea of do ECMO patients need to be anticoagulated? So they reviewed 34 publications, got a total of 200 patients out of it, relatively even mix between VV and VA patients.  Of the group that was not anticoagulated, 129 of that group were not due to bleeding-related  complications, which I think is a very important thing to see. I think the planned non-anticoagulated  ECMO run is very different than the ECMO run that is anticoagulation-free due to major bleeding  events. Looking at a little less than five days as a median of anticoagulation-free support,  and notably about 50% of these patients did require some type of blood product administration  during the run, even being anticoagulation-free. Outcomes-wise, the thrombosis was about 23%,  sercoelated making up a little bit more of that at 13%, and there was still a very high rate of  bleed with 33% of patients experiencing some sort of bleed and 28 of those being major bleeding  events. So this is the breakdown of the table. It's a little bit small, but I wanted to kind of say that this is a great paper and that they pulled all of these patients together,  but we kind of need a comparator. So just looking at all patients that didn't receive anticoagulation doesn't necessarily help answer that question. So the right sides of each of  these tables is looking at systematic reviews of patients that were anticoagulated. One of the nice  things is a couple of these papers have sample sizes up over a thousand patients, which might  give a little better idea. So looking at the thrombosis side, we're able to see is that the thrombosis occurred in the 23% and the anticoagulation-free event, and there's across  the board 24%, 5%, and 10% of thrombosis in the groups that were anticoagulated. So it does look like there is a little bit of an increase in thrombotic complications when it's an  anticoagulation-free circuit. Now the flip side to that being bleeding events, which I think is  interesting, that the Olson systematic review found that 33% of patients had a bleeding event,  and in the anticoagulated cohort systematic reviews, those numbers were 29% and 33%.  So not a dramatic difference in bleeding across the two cohorts either. So this is just to  highlight those two main endpoints there. So a study missing out of that review, just out of  the nature of timing, was Kuhari and colleagues in 2020. And here they looked at 36 patients that  were anticoagulation-free and 38 patients that were anticoagulated, a decent number of patient  days with 225 versus 780 patient days. The only two things that they found to be statistically  significant were gastrointestinal bleeding and the rates of oxygenator exchanges. So a little  bit interesting here though, that the anticoagulation group actually had a near threefold  increase in the need for oxygenator exchange. I think the GI bleeding difference is reasonable  to think through it as 6% versus 29%, but a little challenging to evaluate the difference  in oxygenator exchange here. And on the right, you can see is their plot there that there is  actually no difference when looking out through the total of the numbers at risk. So in this paper  though, they did utilize three risk mitigation strategies that I think really need to be  discussed, and that is that all of their circuits were anticoagulation-bonded. They kept flow rates  over three and a half liters, so three and a half to four liters, knowing that the faster the blood  is flowing through the corporeal oxygenator, that it's decreasing the likelihood of circuit-related  thrombus. And lastly, they utilized early and aggressive VTE prophylaxis to try to mitigate  some of the thrombotic risks that these patients may be experiencing. So I think the final piece I  wanted to discuss is, is there a happy medium? So I said, started the talk with bivalorudin versus  heparin, and it looked like bivalorudin may have had some benefit. Presented a little bit of  evidence that says, maybe you don't need it at all. The bleeding and thrombotic rates don't really  differ, so why do we need it? But is there that place that you can go that gives you a little bit more protection against those thrombotic events and doesn't actually increase your  bleeding events? So these are three papers that looked at that happy medium. So the first by  Selinger and colleagues looked at an APTT goal of 35 to 40 compared to an ACT goal of 140 to 180.  So to me, thinking of a PTT of 35 to 40, you're kind of seeing something along the lines of a  aggressive VTE prophylaxis, more so than a therapeutic anticoagulation.  And the results of this paper did find that the higher doses of anticoagulation did lead to a reduction in need for oxygenator exchanges. Secondly, Carter and colleagues looked at a  comparison of an ACT goal of 140 to 160 compared to 180 versus 200. So here, a little bit more of  comparing the same anticoagulation measurement with ACTs. In this manuscript, they were able to  they were able to find that there was not a difference in outcomes between the two and  recommended that ECMO could be safely done with that ACT of 140 to 160. Then lastly is the trial  that looked at an APTT goal of less than 45 seconds compared to 50 to 70. I think this is really  interesting to me personally because that 50 to 70 a little bit closer mimics the protocols that  we utilize here at our institution. They did structure this primarily as a pilot study, not  necessarily looking for big outcomes, but what they were able to see is that it did appear that there  was no difference and said that this could be the foundation to potentially go forward with a larger  randomized control trial. So thank you for all of your attention during this this discussion today.  So of the two controversies that I that I mentioned, I think there's a couple of main  takeaway points to me. One is that for controversy number one, bivalirudin does appear to be at least  as safe and efficacious as unfractionated heparin. I say at least as because none of the manuscripts  that I discussed today were randomized in nature. There was no great homogenous cohorts, so it is  difficult for me to truly discuss superiority of one anticoagulation over the other. So I do think  that some better designed randomized studies will really helpfully answer this question.  But two, I think what's really needed in the meantime is an in-depth cost analysis that really  may say, yes, you're going to spend a little bit more money on bivalirudin from a pharmacy  perspective, but if you're helping patients and you're helping the overall cost of care, that it  may be worth it. For controversy number two, I think that at this point in time, anticoagulation-free  ECLS may have a role in VVECMO. I don't think from the available literature in until some bigger studies are completed that it's an acceptable approach right now for veno-arterial  ECMO due to the increased risk of ischemic stroke if a patient were to be not anticoagulated enough.  I do think there's a role to continue to expand that work, but right there, I would not be willing  in my practice to go there. I think that focusing on limiting that anticoagulation in the VVECMO  population through either the use of aggressive VTE prophylaxis or lower anticoagulation goals  whenever the risk mitigation strategies have been employed, such as anticoagulation bonded circuits,  maintaining and watching flows, and careful examination of oxygenator connectors, etc., I think there's a role for that. Hopefully, some of the papers moving forward will really  be able to answer that question for the ECLS community as well. With that, I do want to acknowledge a couple of groups here. One, the picture on the left is a majority of our ECLS  multidisciplinary team. You have a couple of our intensivists, our lung transplant surgeon, Dr.  Pablo Sanchez, a couple of our fellows, and our nurse leadership there. Then on the right is Dr.  Penny Sappington, who was one of my personal mentors in life, as well as helped to devise our  bivalveirudin protocol for ECLS here at UPMC. I do want to acknowledge them and thank them for all  of their help throughout the years. Thank you, and please feel free to reach out with any questions.

Video Summary

In this talk, Ryan Ravisecki, a clinical pharmacist, discusses controversies around anticoagulation for extracorporeal life support (ECLS). He highlights the delicate balance between bleeding and thrombotic complications in ECLS patients. He mentions that up to 40-60% of patients experience major bleeding complications during ECLS, while thrombosis is also common, occurring in around 20% of patients. The talk focuses on two controversies: the optimal anticoagulant for ECLS patients and whether anticoagulation is necessary for ECLS support. Ravisecki discusses studies comparing unfractionated heparin and bivalirudin as anticoagulants, noting mixed results. He also mentions studies that explore the possibility of anticoagulation-free ECLS but emphasizes that more research is needed, particularly for veno-arterial ECMO. Ravisecki concludes that bivalirudin appears to be safe and effective, but cost analysis and further studies are necessary. He also suggests that anticoagulation may not be necessary for certain ECLS patients, but caution is required.

Asset Subtitle

Pharmacology, Procedures, 2022

Asset Caption

Anticoagulation strategies and monitoring vary by institution. Laboratory monitoring protocols and applications in a variety of situations, including extracorporeal membrane oxygenation, will be reviewed. Quality assurance for this high-risk medication with continuous evaluation of protocols and areas of practice improvement will be highlighted though clinical cases.

Meta Tag

Content Type Presentation

Knowledge Area Pharmacology

Knowledge Area Procedures

Knowledge Level Advanced

Membership Level Select

Tag Anticoagulation

Tag Extracorporeal Membrane Oxygenation ECMO

Year 2022

Keywords

anticoagulation

extracorporeal life support

bleeding complications

thrombotic complications

unfractionated heparin

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