SCCM Resource Library-Is Blood Thicker Than Water? Management of Coagulopathy in Liver Failure

Video Transcription

Hello, my name is Stephanie Bass, and I'm a medical ICU pharmacist at Cleveland Clinic, and I'm happy to speak to you today about the management of coagulopathy of liver failure.  These are my disclosures, but nothing is relevant to the content of this presentation. So, the goal for this presentation is to review the 2020 SCCM guideline recommendations and  the literature supporting the use of viscoelastic testing and reversal agents in the coagulopathy of liver failure.  And we'll begin with evaluating the role of viscoelastic testing in the coagulopathy of liver failure.  So, the 2020 SCCM guidelines for the management of acute and acute-on-chronic liver failure in the ICU have really two recommendations as it concerns viscoelastic testing.  So, for overall evaluation of bleeding and thrombosis risk in these patients, viscoelastic testing is suggested over traditional measures such as INR, platelet, and fibrinogen.  However, in the setting of bleeding risk prior to procedures, viscoelastic testing is actually recommended over traditional measures, and this is a much stronger recommendation due  to a moderate quality of evidence. We'll dive into the data behind where the level of evidence for these recommendations  comes from in a minute, but first let's take a step back and remember why the traditional measures of coagulation might not be helpful in patients with liver disease.  So, these traditional measures of coagulation include platelets, INR, and fibrinogen. These though are often inadequate because they can only measure hemostasis and neglect  counterbalance factors that are often present in patients with liver disease. So, let's take, for instance, platelet count.  Though a platelet count might be reduced in liver failure, it isn't a great marker for overall platelet function because some factors, such as von Wildebaran factor, are increased  which can compensate for the thrombocytopenia. And now, let's take a look at INR. We all likely know about the inadequacies of INR as a marker of coagulopathy in this  population, and in fact, due to decreased levels of protein C as an antithrombin and elevated levels of factor VIII, there's actually often normal to enhanced thrombin-generating  capacity in this population, but the INR number itself does not reflect that. And finally, the other traditional measure of coagulation is fibrinogen.  Although lower levels are more common in cirrhosis, it isn't clear what level is associated with  hemostasis as most of this data for correcting fibrinogen comes from trauma and cardiac surgery literature.  So because of the concerns with traditional measures of coagulation, viscoelastic testing is a promising technique for the evaluation of hemostasis in patients with liver disease.  There are three types of viscoelastic tests, thromboelastography or TEG, rotational thromboelastography or ROTEM, and sonoraiometry.  The major advantage of viscoelastic testing is that it can provide real-time measurement of clot formation and breakdown, and that it's performed on whole blood.  The major disadvantage, though, is that there are not well-defined thresholds for transfusion based on the results of viscoelastic testing.  So at this point in time, applicability of the test is somewhat limited. The results of viscoelastic testing for TEG and ROTEM provide five values which correspond  to really the time to start forming a clot, the time until maximum clot formation, the speed of clot formation, the maximal clot strength, and finally, the time to clot breakdown  respectively. On this slide, really, I just wanted to remind you of the terminology of the output for the  test, since if you're like me, you might only be familiar with one of those terminologies because your site probably only has one of these types of tests, as well as the normal  values. Notably, R-time and clotting time, as well as maximum lysis and LY30, have fairly different normal ranges between the two tests.  There are five studies that have evaluated viscoelastic testing for transfusion guidance in patients with liver disease, particularly looking at replacement of FFP, platelets,  and cryoprecipitate based on the output of the TEG or ROTEM. Although the definition of coagulopathy were similar between the studies, the risk of bleeding  was actually markedly different, with the first three studies being paired procedural studies and actually two of those studies including quite low risk bleeding procedures  such as central line insertion, while the last two studies included patients with active GI bleeds.  The primary outcome, though, in all of the studies was similar at looking at the difference in transfusion requirements from viscoelastic guided strategies compared to standard of  care. And in almost all the studies, there was a significant reduction in transfusions with the use of viscoelastic testing compared to standard of care.  And additionally, the studies found no difference in either post-procedural bleeding or control of bleeding in the active bleeding studies between the two transfusion protocols.  So based on these studies, then, it appears that transfusions based on viscoelastic testing results might be safe and can likely reduce transfusion requirements.  However, as I mentioned earlier, one concern with the use of viscoelastic testing is that there are no well-established transfusion thresholds.  And as you can see with these five studies, the transfusion thresholds varied widely between studies.  Looking only at the R-time threshold or clotting time threshold shown in the second column, the studies varied anywhere between greater than 10 minutes to greater than 40 minutes  when the normal value for R-time is 5 to 10 minutes. One important piece to keep in mind is that the most conservative thresholds were utilized in the setting of active bleeding.  For example, the Kumar study, the last study shown on the slide, whereas the more liberal  transfusion thresholds, for example, the Dipetri study, the top study, were utilized paraprocedurally.  Another thing to note is what test output measures were utilized as transfusion thresholds. For all the periprocedural studies, only R-time or clotting time and maximum amplitude or  maximum clotting time were used as transfusion thresholds, whereas in the setting of active  bleeding, essentially any and all test output measures were considered as a potential transfusion threshold. And particularly, again, I would like to point out the Kumar study.  So to summarize viscoelastic testing in liver patients, there's honestly still a lot we do not know.  But although the transfusion goals are not validated at this time, the actual transfusion goal and which results even to transfuse based on likely depend on your clinical scenario.  Our discussion to this point has focused on the evaluation of coagulopathy in liver patients.  But now we'll turn our attention to the treatment of coagulopathy in liver patients and we'll discuss reversal agents.  And we are going to talk about a lot of these in a short amount of time, so please bear with me a little bit.  We'll begin by reviewing agents that have been investigated for the reversal of your elevated INR and cirrhosis, the first being recombinant activated factor VII.  So factor VII is thought to be helpful because it might limit the volume associated with correcting your elevated INR using FFP.  And it has been evaluated in two settings for patients with liver disease, both for, again, acute bleeding as well as prior to procedures.  For acute variceal bleeding, there have actually been two randomized controlled trials using factor VII that have evaluated nearly 500 patients.  And they found that using a high dose factor VII strategy resulted in no overall difference in failure to control bleeding compared to standard of care.  Although they did see a potential benefit in patients who, the subgroup of patients with active bleeding found on endoscopy.  The very short duration of action of factor VII though limits its use for active bleeding in some ways.  So again, in this active bleeding study, factor VII had to be given as 100 mics per kilo every  few hours until that bleeding was controlled, which led to sometimes very high overall doses. Patients received up to 800 mics per kilo.  And because of that, increased thromboembolic events. In those studies, 16 total overall, including five arterial thromboembolic events.  However, peri-procedurally, factor VII having this short duration of action is somewhat beneficial.  For peri-procedural use, the doses that have been studied for liver patients have been much lower, 20 to 40 mics per kilo, which has been shown to provide full reversal of  INR for about two hours, which is usually enough time to do most procedures. Similar to factor VII, prothrombin complex concentrates, or PCCs, can reduce fluid administration  required to correct the INR. However, compared to factor VII, PCC may theoretically provide a longer duration of reversal, usually a few hours, like six to eight hours.  A retrospective study of, again, pretty small, 45 patients with cirrhosis and coagulopathy considered an INR greater than 1.5 compared the use of FFP, PCCs, and factor VII prior  to invasive procedures. And their primary outcome was looking at the proportion of patients who were able to achieve full reversal defined as an INR of 1.5.  And they found that significantly more patients were able to achieve that full reversal with both PCCs and factor VII compared to FFP.  And additionally, the median INR reduction was similar with PCC and factor VII at 1.6 and 1.8, respectively, compared to only 0.5 with FFP.  There was no difference in this study in major bleeding post-procedure between any group, although the majority of these procedures were low-risk procedures.  And although there were no significant differences in thromboembolic events between groups, they were numerically higher with both PCC and factor VII.  So although this is a very small study and quite limited data, this potentially implies that PCC and factor VII can most effectively temporarily reduce an INR in this patient  population.  However, the biggest question when using PCC in patients with liver disease is how to dose it effectively, since the established dosing for PCC comes from anticoagulant reversal  indication and is because of that base on your INR. So if we use this label dosing based on INR in patients with liver failure, that would result in significant overdosing of PCC.  So a couple of strategies, then, have been investigated for dosing PCC in this population.  The first strategy listed in this table, suggested by Huang and colleagues, is to use a fixed dose of 500 units.  This strategy was evaluated in a small population of 10 patients with cirrhosis. However, it seems to lead to inadequate coagulopathy reversal because it's not enough.  As you can see in this table, no patients achieved INR reversal.  The strategy that seems to be probably more effective for PCC dosing is actually using an INR-based strategy, but then using a maximum dose of around 25 to 30 units per kilo.  So this strategy was used in two small retrospective studies in which the median PCC doses were 23 and 22 units per kilo, respectively.  And this strategy resulted in some, but not full, INR reversal. However, low rates of thromboembolic and bleeding complications.  So based on this data, plus additional reports of significantly increased thromboembolic events with doses above 30 units per kilo in this population, this strategy of limiting  doses to less than 25 to 30 units per kilo seems to be the most effective until we have further data.  So besides reversing the coagulopathy of liver disease due to INR, medications to reverse the thrombocytopenia of cirrhosis have also been evaluated.  And actually, thrombopoietin receptor agonists are FDA-approved to increase platelet count in this population.  I've included the major randomized trials that led to each drug's FDA approval for  L-thrombopag, avatrombopag, and lucitrombopag, respectively. Notably, all of these studies were done in cirrhotic patients with platelet counts less  than 50 and were performed prior to known procedure. None were done in a setting of active bleeding. The reason that I want to point this out is due to the timing here.  So depending on the drug and their various timing, these agents are started several days  prior to procedure in order for the platelet count to increase enough to perform the procedure.  The primary outcome, though, in all of the studies were the proportion of patients not requiring a platelet transfusion.  And the TPO agonists all significantly increased this number. So essentially, platelet transfusions were avoided.  The studies also looked at the difference in post-procedural bleeding, and there was no difference between placebo and patients that received these drugs.  But actually, the SCCM acute liver failure guidelines, as well as the AASLD procedural bleeding guidelines, point to this lack of difference as a reason to avoid these agents,  implying essentially these agents are no different than just giving platelets. Another reason to avoid these agents is actually the increased risk of thrombotic events, including  portal vein thromboses. And in these studies, the thrombotic events usually occurred when the platelet count increased above 200.  However, because of how the platelet count increases, it usually peaks after drug discontinuation.  So this increased thrombotic event cannot be easily mitigated by early drug discontinuation. So in this example here on the chart shown, this was from the Ava Trombopag study.  Although the duration of the study drug was only five days, you can see that the platelet count didn't even start to increase until around day four, and then it peaked around  day 10 to 13, and then it eventually returned to baseline. Finally, the last consideration for reversal of coagulopathy is hyperfibrinolysis.  If hyperfibrinolysis is present, two agents are available, tranexamic acid or aminocoproic acid.  Neither agent has really been well studied in acute bleeding secondary specifically to hyperfibrinolysis.  In fact, I could really only find one small study of 37 patients with cirrhosis and clinical evidence of hyperfibrinolysis.  But in that study, these patients were given oral aminocoproic acid throughout their hospitalization,  which did result in 92% of them having improvement or resolution of bleeding and no thromboembolic complication.  However, of the 16 patients that had aminocoproic acid discontinued during their hospitalization, nearly half had recurrent bleeding after drug discontinuation.  So we don't have a lot of data for bleeding secondary specifically to hyperfibrinolysis,  but whether antifibrinolytics could be considered for patients with acute bleeding in general without known hyperfibrinolysis was recently studied for patients with GI bleeding, of  which only a small percentage of patients with acute bleeding in general had acute bleeding in general.  However, antifibrinolysis was recently studied for patients with GI bleeding, of which 40% actually had cirrhosis in the HALT-IT study.  In this study, tranexamic acid or placebo was given for 24 hours to over 12,000 patients with acute GI bleeds.  Their primary outcome was death due to bleeding, and they found no overall difference in their primary outcome between TXA and placebo.  There was also no difference in death due to bleeding in the subgroup of patients with variceal bleeds or concomitant liver disease, so no benefit of TXA, but potentially a signal  for harm.  Overall, VTE risk was higher with TXA in this study at 0.8% compared to 0.4% with placebo,  but this increased risk was actually confined to the subgroup of patients with liver disease, and this was actually confirmed by the interaction test for heterogeneity.  So, this implies potentially an increased risk of harm if antifibrinolytics are given to patients with liver disease without known hyperfibrinolysis.  Of note, a similar study is being completed in only cirrhosis patients with results expected shortly, so hopefully we may be able to confirm the results of this subgroup finding.  So, in conclusion, there is still much we don't know about the utility of viscoelastic testing to guide transfusions for patients with liver disease, but transfusion goals  likely reflect the indication. And in terms of reversal agents, there are many agents that have been evaluated for the  coagulopathy of liver disease, but it's important to remember that all have pretty limited data and are associated with increased thromboembolic events.  For Factor VII, PCC, and antifibrinolytics, consider using lower doses and short-term durations to limit the increased thromboembolic events.  And finally, thrombopoietin receptor agonists aren't recommended because of that delay in time to effect.  So, thank you again for your attention, and again, thank you to SCCM.

Video Summary

Stephanie Bass, a medical ICU pharmacist at Cleveland Clinic, discusses the management of coagulopathy in liver failure. She reviews the 2020 SCCM guideline recommendations and the use of viscoelastic testing and reversal agents. Viscoelastic testing is suggested over traditional measures such as INR, platelet, and fibrinogen for evaluating bleeding and thrombosis risk in patients with liver failure. Additionally, viscoelastic testing is recommended over traditional measures for bleeding risk prior to procedures. Viscoelastic testing provides real-time measurement of clot formation and breakdown. However, the lack of well-defined transfusion thresholds limits its applicability. Several studies have evaluated viscoelastic testing for transfusion guidance and found a significant reduction in transfusions when guided by viscoelastic testing compared to standard of care. For reversing coagulopathy, recombinant activated factor VII and prothrombin complex concentrates (PCCs) have been investigated. PCCs have shown promise when dosed at less than 25 to 30 units per kilo. Thrombopoietin receptor agonists are FDA-approved to increase platelet count but come with an increased risk of thrombotic events. Antifibrinolytics have limited data for use in liver disease, and a recent study showed potential harm and increased risk of venous thromboembolism.

Asset Subtitle

Hematology, GI and Nutrition, 2022

Asset Caption

SCCM's Guidelines for the Management of Adult Acute and Acute-on-Chronic Liver Failure in the ICU were published in March 2020. This session will cover these recommendations and continued controversies surrounding managing these complex patients. The guidelines cover recommendations for cardiovascular, endocrine, hematologic, pulmonary, and renal considerations. This session will cover selected topics, including hemodynamics and resuscitation, coagulopathy, and renal complications.

Meta Tag

Content Type Presentation

Knowledge Area Hematology

Knowledge Area GI and Nutrition

Knowledge Level Intermediate

Knowledge Level Advanced

Membership Level Select

Tag Coagulation

Tag Liver

Year 2022

Keywords

coagulopathy

liver failure

viscoelastic testing

reversal agents

transfusion guidance

	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.