New AASLD Guidelines Ask Us to Reconsider How We View Bleeding Risk in Cirrhosis

David A. Johnson, MD


March 04, 2021

This transcript has been edited for clarity.

Hello. I'm Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia.

Kudos to the American Association for the Study of Liver Diseases (AASLD) for their latest expert consensus document. Among the several topics covered in these 48-page guidelines are vascular anomalies such as portal vein and hepatic vein thrombosis, and hepatic and splenic aneurysms. However, today I wanted to highlight for you the section dealing with coagulation management in patients with cirrhosis, including how it relates to interventions we commonly deal with as gastroenterologists, such as endoscopic or surgical procedures.

Reconsidering Cirrhosis-Related Bleeding

It would be an understatement to say that patients with cirrhosis have multiple alterations in their hemostatic system. It needs to be better understood, however, that there are hemostatic changes promoting both bleeding and clotting that occur simultaneously in any given individual patient. These changes may in fact counteract each other.

For example, the thrombocytopenia we see in cirrhosis due to hypersplenism is alleviated when the platelet adhesive protein von Willebrand factor is increased. There are decreased levels of the procoagulant proteins that are counteracted by decreased levels in the natural anticoagulant proteins (in particular, protein C; and to a lesser degree, protein S). Decreased levels of the natural antifibrinolytic proteins are counteracted by decreased levels of the profibrinolytics, except in critically ill patients where fibrinogen may be low.

The authors of this guidance document emphasize that, as a result of these changes and clinical observations, cirrhosis should no longer be considered a condition with an overall bleeding tendency, but instead as one with both prohemostatic and antihemostatic properties. As such, rather than globalizing cirrhosis, they suggest that it should be individualized.

This document underscores previous recommendations that found that traditional laboratory tests for measuring coagulation and assessing relative risk — such as prothrombin time, international normalized ratio (INR), or activated partial thromboplastin time — have proven inadequate and often misleading. When assessing risk, we therefore need to stratify by the individual patient.

Intervention-Associated Bleeding Risk

The authors note that clinicians need to consider interventions relative to how they would prevent or treat the bleeding that may occur. If you're performing an intervention where you could visually see and stop the bleeding, that changes the relative risk given your ability to recognize and appropriately treat it. Conversely, if you're performing an intervention in proximity to a major vessel that could potentially be disrupted, that's going to result in a much more significant bleeding risk.

Traditionally, the risk for major bleeding was thought to be < 1.5% in low-risk procedures and > 1.5% in high-risk procedures, yet these figures need to be placed in the context of the present day. This risk dichotomy was developed by expert opinion using periprocedural therapeutic anticoagulation thresholds that we now view as probably inadequate.

This again requires that clinicians identify factors that can impact bleeding risk, including who is performing the procedure and what that procedure is. For example, if a radiologist uses transjugular intrahepatic portosystemic shunt when performing an ultrasound guidance for the portal vein access, this minimizes the risk for capsular perforation, which is a known precipitant of a major intraperitoneal hemorrhage. The risk reduction is dramatically affected.

In my practice, I use a cold snare polypectomy approach, which risk-adjusts for both the operator performing the procedure and the type of technique by avoiding the coagulation-related sloughing of the eschar and late bleed we sometimes observe. It also gives us the ability to intervene acutely and demonstrably if we see active bleeding persist.

Systemic Factors and Periprocedural Considerations

The guidelines' authors also highlighted the role that systemic factors may play in increasing bleeding risk, including acute and chronic kidney disease. They cite a study of nearly 5000 large-volume paracenteses in which 89% of major bleeds were in patients with renal dysfunction. They also looked at a small study with endoscopic variceal ligation in approximately 260 patients, which found that those with a major bleed had significantly greater creatinine (2.2 vs 1.0 mg/dL). These results contributed to their recommendation that acute or chronic renal function needs to be addressed when stratifying for risk.

When considering periprocedural bleeding, these guidelines (like previous ones) indicate that using the specific INR platelet cutoff is not a reliable way to account for elevated risk.

There are also some periprocedural interventions to be considered, including platelet therapy. In vitro studies have suggested that platelet counts > 55,000/μL improve hemostasis. However, these data do not account for the potential compensation by von Willebrand factor or other endothelial-based components, and therefore no longer seem applicable.

Three Available Medications

Three drugs are approved for increasing platelet counts in cirrhosis, all of which are oral thrombopoietin receptor agonists.

One of these is eltrombopag, which doesn't really merit discussion given that it was approved for the management of thrombocytopenia-related, interferon-based hepatitis C, which is by now a bit archaic.

However, the other two — avatrombopag and lusutrombopag — are indicated for treatment of thrombocytopenia in patients with chronic liver disease who are scheduled to undergo a procedure. The guidelines note that these agents require a 2- to 8-day course preceding the scheduled events. Although they've been shown to be superior to placebo in getting the platelets over 50,000, there were no statistical differences in the postprocedural bleeding events. Therefore, the authors do not recommend their routine use.

Another factor to consider for targeting procoagulant factor deficiencies is the use of fresh-frozen plasma. As with INR, fresh-frozen plasma transfusions are not considered reliable. Furthermore, fresh-frozen plasma transfusions have considerable risks. First, there is a potential risk of developing transfusion-related lung injury. Second, they can increase the portal pressure, which is a noteworthy consideration when talking about a variceal bleed. Therefore, the authors concluded that fresh-frozen plasma has no value as an option for routine use in these patients.

The authors also discussed the role of vitamin K. Outside of advanced malnutrition status or chronic cholestasis, vitamin K replacement has no measurable effect on the INR in patients with cirrhosis. That's certainly different from what we were traditionally taught.

We do have data showing that fibrinogen levels < 1000 mg/dL are associated with an increased bleeding risk. Given that fibrin and its precursor fibrinogen are key components of formal clot formation, it makes sense that it might be of value. The correction of plasma fibrinogen to > 100 using cryoprecipitate or fibrinogen concentrate can be considered. However, outside of patients who are systemically and critically ill, it's very unusual to see such low levels of fibrinogen.

Threshold Values for Common Coagulation and Bleeding Parameters

These recommendations on the minimum thresholds for common coagulation and bleeding parameters before performing invasive procedures with high bleeding risk are somewhat different from other societies'.

The AASLD experts recommend no routine preprocedure corrections for platelet count, INR, or fibrinogen level. In comparison, a 2019 guidance from the Society of Interventional Radiology recommended thresholds of > 30,000 for platelet counts, < 2.5 for INR, and > 100 for fibrinogen. Guidelines issued in 2020 from the American College of Gastroenterology recommended thresholds > 50,000 for platelet count, no correction for the INR, and a fibrinogen target > 120-150. Finally, an American Gastroenterological Association panel recommendation from 2019 established thresholds for a platelet count > 50,000, no correction for the INR, and fibrinogen level > 120. Therefore, it's clear that the AASLD has changed the paradigm for how we view coagulation in patients with cirrhosis, so it's very important for us to understand and apply this.

In conclusion, I highly recommend reading this 48-page document in depth. Beyond coagulation effects, it has important information on hepatic and portal vein thrombosis, the management of vascular anomalies in the liver, and other topics. There's a wealth of data here that merits keeping it nearby for easy reference, because this is going to be a meaningful document for years to come. I hope it's helpful in your practice and your management of patients with cirrhosis.

I'm Dr David Johnson. Thanks for listening.

David A. Johnson, MD, a regular contributor to Medscape, is professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia, and a past president of the American College of Gastroenterology. His primary focus is the clinical practice of gastroenterology. He has published extensively in the internal medicine/gastroenterology literature, with principal research interests in esophageal and colon disease, and more recently in sleep and microbiome effects on gastrointestinal health and disease.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.