Introductory Chapter: Anticoagulation in Liver Cirrhosis – A Contemporary Point of View from Hepatologists

1. Introduction

In the past, anticoagulation had rarely been explored in patients with liver diseases, especially liver cirrhosis. This is primarily because they were often considered at a high risk of bleeding due to reduced synthesis of coagulation factors secondary to liver injury, decreased platelets count caused by hypersplenism, and rupture of gastroesophageal varices related to portal hypertension. However, in the contemporary era, it has been well recognized that patients with liver cirrhosis are also more prone to develop venous thromboembolism, especially portal vein thrombosis (PVT). Accordingly, there is increased evidence regarding the efficacy and safety of prophylactic and therapeutic anticoagulation in such patients. In this introductory chapter, we would like to briefly discuss two important issues regarding the use of anticoagulation in liver diseases based on the experiences of our group.

2. PVT in liver cirrhosis: anticoagulation therapy or not?

PVT is common in liver cirrhosis with a prevalence of 13.92% and an incidence of 10.42% [1]. Risk stratification of PVT by evaluating its stage, grade, and extension is very important to judge the necessity of anticoagulation therapy [2].

According to the interval from onset of symptoms to diagnosis of PVT, it was divided into acute and chronic stages [3]. However, PVT is often asymptomatic in patients with liver cirrhosis, the date of onset of symptoms may be obscure, and the threshold of this interval greatly varies among studies. Recently, the Chinese expert consensus has supported the term “acute symptomatic PVT” characterized as the presence of abdominal pain secondary to extensive mesenteric vein thrombosis and its related bowel ischemia [4]. Generally, this term is more convenient and accurate to guide immediate initiation of antithrombotic therapy for acute symptomatic PVT. Indeed, this group of population is the absolute indication of anticoagulation therapy [5, 6], because they can develop lethal complications, such as intestinal infarction, in a very short time, if untreated. Notably, in some cases where PVT is occlusive and extensive, anticoagulation alone may be ineffective [7], and thus, thrombolysis should be very necessary [8].

In cirrhotic patients with asymptomatic PVT, which refers to the absence of abdominal symptoms related to bowel ischemia, the initiation of anticoagulation therapy is dependent upon the grade and extension of thrombosis as well as the risk of its potential complications. It has been shown that PVT with and without thrombosis extension to the superior mesenteric vein, especially when the thrombus occupies more than 50% of the main portal vein, can increase the risk of hepatic decompensation and death in patients with liver cirrhosis [9]. Thus, it should be very necessary to maximize the portal vein recanalization. Meta-analyses supported the benefits of anticoagulation for dissolving PVT and preventing PVT progression in liver cirrhosis [10, 11]. Besides, it seemed that anticoagulation therapy did not increase the risk of bleeding [11]. Of course, this finding should be cautiously interpreted, because the investigators delicately selected the patients who received anticoagulation therapy [12], and few cirrhotic patients with severe thrombocytopenia or liver failure were given anticoagulation for treatment of asymptomatic PVT among these published studies. On the other hand, the fact that PVT can be spontaneously recanalized in the absence of antithrombotic therapy has been increasingly recognized [13], which is named as “transient PVT in cirrhosis” by a group of hepatologists [14]. Accordingly, how to precisely distinguish between the candidates who do not require anticoagulation but achieve portal vein recanalization spontaneously and those who should receive anticoagulation for recanalization of PVT becomes an important issue to be addressed in the future.

Collectively, in the case where the recanalization of PVT is considered as the primary goal of anticoagulation therapy in patients with liver cirrhosis, several conditions where anticoagulation must or should be employed and where anticoagulation may be ineffective or unnecessary have been summarized in Table 1.

Absolute indication of its use	Acute symptomatic PVT.
	i.e., acute and progressive abdominal pain related to bowel ischemia.
Conditions where it should be considered	Asymptomatic PVT which will influence the outcomes of cirrhotic patients.
	i.e., clinically significant PVT; or thrombosis within the main portal vein occupied more than 50% of the lumen.
Conditions where it may be ineffective	Asymptomatic PVT which is occlusive and extensive.
	i.e., occlusive PVT or fibrotic cord with cavernous collaterals. The rate of portal vein recanalization is often very low by anticoagulation alone.
Conditions where it may be unnecessary	Asymptomatic PVT which will be spontaneously recanalized. i.e., transient PVT.; More data should be necessary to identify the candidates who have a high probability of transient PVT.

Table 1.

Indications of anticoagulation for recanalization of PVT in cirrhosis.

3. Type of anticoagulation in liver cirrhosis: direct oral anticoagulants or conventional anticoagulants?

At present, conventional anticoagulants, including warfarin, unfractionated heparin, and low molecular weight heparin, have been employed for the management of thromboembolic disorders for decades in clinical practice. Nowadays, they are being replaced by direct oral anticoagulants (DOACs), because DOACs are more convenient in anticoagulation monitoring, more compliable via an oral route, and equally effective for the prevention and treatment of thromboembolic disorders. However, DOACs are not recommended in patients with advanced liver diseases, such as those with Child-Pugh class B and C. Additionally, its safety profile in liver cirrhosis is still a matter of debate. Based on the recent evidence, a meta-analysis demonstrated a significantly lower risk of all bleeding and major bleeding in cirrhotic patients who received DOACs than those who received conventional anticoagulants, but statistically similar risk of fatal bleeding, gastrointestinal bleeding, and intracranial hemorrhage between them [15]. Notably, as expected, elder age, a longer duration of DOACs, and Child-Pugh class C are associated with a higher risk of bleeding in cirrhosis. Thus, DOACs should be a preferred choice of anticoagulation in selected cases of liver cirrhosis.

In conclusion, the clinical decision making should be individualized when the initiation of anticoagulation aims to recanalize PVT in patients with liver cirrhosis. Besides, DOACs seem to be a promising choice of anticoagulation therapy in such patients.