Cirrhosis Paradox: Is the Risk for Clotting or for Bleeding?

David A. Johnson, MD


April 10, 2017

This feature requires the newest version of Flash. You can download it here.

Coagulopathy in Cirrhosis

Hello. I am Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School.

We were all taught at some point, unless perhaps you are still in training, that most people with cirrhosis are "anticoagulated." We see these patients routinely for preoperative evaluation, or perhaps in a catastrophic presentation with acute gastrointestinal (GI) bleeding. What should we do?

Here are two patient scenarios.

Scenario 1. You get a call from the emergency room (ER). Patient 1 has cirrhosis and vomited blood. The platelet count is 65,000 cells/µL, and the international normalized ratio (INR) is 1.8 to 2.0.

What do you tell the patient? What do you tell the ER physician?

Most of us would say, "Give the patient some fresh frozen plasma (FFP) to reverse the anticoagulation, and we will be in and do endoscopy." You may add sandostatin. The idea is that you would target the abnormal INR in patients with cirrhosis.

Scenario 2. You get a call from your hospital regarding patient 2. The patient, who is cirrhotic, is in for acute pneumonia. The patient is thrombocytopenic, with a platelet count around 60,000-65,000 cells/µL, and the INR is 2. The patient meets criteria for deep venous thrombosis (DVT) prophylaxis with low-molecular-weight heparin or subcutaneous heparin.

What would you tell the hospitalist? Would you say to give it? Or would you say no, because they are "pseudo" anticoagulated?

If you said to give FFP or not to give them anticoagulation, you are wrong.

Are Patients With Cirrhosis 'Anticoagulated'?

This is a profound understanding, advanced by a brilliant article that was just published in the American Journal of Gastroenterology by Dr Steve Caldwell,[1] working with the group from the Thrombophilia and Hemophilia Center in Milano. They are really doing a tremendous service to our understanding of the coagulopathy of patients with cirrhosis.

Dr Caldwell and his colleague, Dr Pat Northup, have taught us about the coagulation defects in cirrhosis over the past several years. An article published in Clinical Gastroenterology and Hepatology[2] is another brilliant review. I have looked at it for the past several years.

We need to start to understand that the cirrhotic patient may be hypercoagulable, not just hypocoagulable, and we need to start to understand how we make disease prevention interventions.

Cirrhosis Alters Coagulation

Let us talk about what happens with cirrhosis.

We know that the vitamin K-dependent factors II, VII, IX, and X decrease. Other factors—factor V and particularly factor XI—also may be decreased. There is a tendency for a reduction in coagulation.

There is also a reduction in procoagulation. We have all seen evaluations for protein S and protein C levels. In cirrhotics, protein C in particular is decreased. You have a balance between procoagulant and reduction in anticoagulant, and this balance is a fine tightrope.

Patients may actually be hypercoagulable. What a paradox that seems to be. If you look at what happens in patients with cirrhosis, we know from studies[3,4,5,6] that a patient with cirrhosis may have an increased tendency for DVT; pulmonary embolus; and, particularly, portal vein thrombosis (PVT). We see PVT in about 11% of our cirrhotic patients.[7] If you look at autopsy studies, this number goes up, maybe in the range of 35%-40%.[8]

We know that PVT accelerates the decompensation of liver disease. In fact, a randomized study[9] (needing to be corroborated) found that giving low-dose heparin in cirrhotic patients may prevent or decrease some of the deceleration effect of PVT, promote longevity, and decrease the risk for decompensation and the need for liver transplant. We are coming full circle from considering them to be autoanticoagulated to giving them anticoagulation.

Evaluating the Patient

The Patient With Cirrhosis

Let us look at this from the standpoint of the patient with a bleeding episode.

The new paradigm is to measure fibrinogen levels, which are where we can make a meaningful intervention. The level of 100-120 mg/dL is really a branchpoint for whether or not to give patients something supplemental, such as cryoprecipitate, which may be helpful. Remember that fibrinogen is an acute phase reactant. A low-normal result may still warrant an intervention with cryoprecipitate.

In the setting of acute major GI bleeding, you may want to think about FFP. But remember from this discussion that you do not want to give FFP to reequilibrate the vitamin K-dependent factors. To do so would require approximately 20-40 mL/kg, which is a tremendous amount of volume in a patient with liver disease. You may actually increase the portal pressure and precipitate ongoing bleeding or portal hypertensive complications, anasarca, pulmonary edema, volume overload, et cetera. Again, the new paradigm is that FFP is not the answer.

If you think the patient needs prothrombin complex concentrate, it can be ordered and would be best guided by a hematologist. This typically includes the vitamin K-dependent factors. Some include protein S and C and antithrombin III. These are lower-volume. Beware, because you may actually precipitate more of a procoagulant effect.

The Patient Undergoing an Elective Procedure

When a patient comes in for an elective procedure, ask yourself, "Does the patient need the procedure now?"

As much as I said these coagulation abnormalities are apparent, they may also be transient and interchanging on a very variable basis, particularly driven by such things as infection, alcoholic hepatitis, or renal insufficiency. Correct those things, before you really have to do something that may be forestalled as an elective procedure, because these may rebalance over a short period.

If a patient comes in for an elective surgery or procedure, the typical recommendation now would be to measure fibrinogen in patients with platelet counts that are deemed adequate—greater than 50,000- 60,000 cells/µL. In the patients with a low fibrinogen level of < 120 to 150 mg/dL, consider replacing fibrinogen with cryoprecipitate. There is no need to give prophylaxis with FFP to these patients. Recognize that that was in recommendations from a lot of societies and that it has no merit. In fact, the INR was really developed around vitamin K-dependent deficient factors induced by drugs, such as Coumadin (warfarin). It had nothing to do with liver disease in factoring in all these other things.

The Patient Who Needs Urgent Attention

If a patient is undergoing a surgical procedure or has an imminent need for some type of intervention (eg, endoscopic procedure) and the platelets are low, consider measuring the fibrinogen level and consider platelet transfusion. These things should be done only in the context of what the relative risk for volume is.

If you are called to the ER or called by a surgeon about a patient with active bleeding, preoperatively (or as quickly as you can) determine a fibrinogen level. Remember, this is something that is a phase reactant, and may be somewhat elevated. Consider using prothrombin complex concentrate. Again, this is best done in concert with a hematologist.

Fresh frozen plasma is not the answer. You may increase the portal pressures. You will never get there with the volumes that you need to replete. If the fibrinogen level is low—under 120 to 150 mg/dL—consider repletion with the cryoprecipitate.

Address remedial factors, such as infection and uremia. We do give DDAVP (desmopressin) to uremic patients, which helps the platelet aggregation factors.

Where We've Been, and Where We Are

There is a paradox of where we have been and where we are. Consider looking at your "pseudo" anticoagulated patients as being in a hypercoagulable state. Not all of them are, but if we miss the opportunity to recognize that, we may actually accelerate their liver disease and potentially increase their liver decompensation.

Certainly, do not hold off on standard things that you would do for other patients to prevent DVT risk. But do put on your thinking caps, and take off what you have been taught in the past.

Kudos to Dr Steve Caldwell and the group in Milano, and also his colleague Dr Pat Northup at the University of Virginia, for teaching us new insights into the coagulation abnormalities of cirrhosis.

I am Dr David Johnson. Thanks again for listening.


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.
Post as: