Hepatology

Review

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Anticoagulation in cirrhosis
Lara Roberts ‍ ‍1,2

► Additional supplemental ABSTRACT

material is published online KEY POINTS
Patients with cirrhosis are at increased risk of

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
only. To view, please visit the
atrial fibrillation and venous thromboembolism, ⇒ Patients with cirrhosis are at increased
journal online (https://​doi.​
org/​10.​1136/​flgastro-​2024-​ for which anticoagulation is standard of
risk of atrial fibrillation (AF) and venous
102994). thromboembolism (VTE).
care. Historical misunderstanding of the
⇒ Anticoagulation remains the mainstay of
1
Institute of Pharmaceutical complex haemostatic changes which occur
therapy to reduce the risk of stroke in AF
Sciences, King’s College London, in cirrhosis, along with the increased baseline
London, UK and thrombosis extension/embolisation
2 risk of bleeding, has contributed to concerns in deep vein thrombosis/pulmonary
King’s Thrombosis Centre,
Department of Haematological regarding anticoagulation in this patient embolism.
Medicine, King’s College Hospital group. It is now accepted that haemostasis ⇒ Prolonged coagulation times and
NHS Foundation Trust, London,
in patients with cirrhosis is rebalanced and thrombocytopenia do not reflect
UK
that the risk of bleeding is predominantly underlying haemostatic capability and are
Correspondence to mediated by portal hypertension. However, not a contraindication to anticoagulation.
Dr Lara Roberts; ​lara.​roberts@​ patients with cirrhosis were excluded from ⇒ Varices should be evaluated and managed
nhs.​net
seminal randomised controlled trials of direct
as per existing guidance and are not a
contraindication to anticoagulation.
Received 28 March 2025 oral anticoagulants, resulting in low-­quality
⇒ Direct oral anticoagulants can be
Accepted 5 May 2025 evidence to inform best practice in this patient
considered for patients with Child-­Pugh A
group. Cohort studies predominantly in atrial and B cirrhosis.
fibrillation suggest anticoagulation reduces ⇒ Low molecular weight heparin may be
stroke risk and improves survival in patients preferred initially for patients with VTE,
with cirrhosis. Similarly, anticoagulation for particularly in the context of acute illness
portal vein thrombosis reduces thrombosis or recent bleeding.
progression, promotes recanalisation and
may improve survival. Additionally, small,
randomised studies suggest anticoagulation has in clinical practice. However, high-­quality
a potential role in reducing the risk of hepatic data to support the optimal approach in
decompensation and liver disease progression. this patient group is lacking due to histor-
In this review, current knowledge of the efficacy ical misunderstanding of the complex
and safety of anticoagulation in patients with changes in haemostasis in patients with
cirrhosis is presented. Concerns related to cirrhosis, concerns regarding the increased
portal hypertension, thrombocytopenia and baseline bleeding risk and that most oral
liver transplantation are discussed, with an anticoagulants are at least partially metab-
approach to the initiation and management of olised by the liver. These concerns have
anticoagulation in atrial fibrillation and venous contributed to the exclusion of patients
thromboembolism provided. with liver disease from the seminal
randomised controlled trials of direct
oral anticoagulants (DOACs) for stroke
INTRODUCTION prevention in AF and treatment of DVT/
It is increasingly recognised that patients PE. In this review, the available evidence
with cirrhosis are at high risk of throm- is summarised and a practical approach to
botic events. Atrial fibrillation (AF) is anticoagulation in patients with cirrhosis
common with a reported prevalence is provided.
© Author(s) (or their employer(s)) ranging from 6% to 14% in patients with
2025. No commercial re-­use. See cirrhosis.1 The risk of deep vein throm- Haemostasis in patients with cirrhosis
rights and permissions. Published
by BMJ Group. bosis (DVT) and pulmonary embolism Patients with cirrhosis commonly have
(PE) is increased two-­ fold,2 and portal prolonged prothrombin time (PT) and
To cite: Roberts L. Frontline
Gastroenterology Epub ahead
vein thrombosis (PVT), the most common thrombocytopenia. Historically, this
of print: [please include Day thrombotic manifestation, has an annual was erroneously thought to indicate a
Month Year]. doi:10.1136/ incidence of 3% to 24%.3 Thus, consider- bleeding tendency and/or that patients
flgastro-2024-102994
ation of anticoagulation arises frequently were ‘auto-­ anticoagulated’. Cirrhosis

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 1

 Hepatology

results in complex haemostatic changes due to (1) the risk of AF.1 Additionally, alcohol increases the risk

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
reduced liver synthesis of procoagulants and antico- of AF via electrical, autonomic and structural remod-
agulants, thrombopoietin, a disintegrin and metal- elling, in addition to associations with obesity and
loproteinase with a thrombospondin type 1 motif, hypertension.1 Of note, patients with cirrhosis have
member 13 (ADAMTS13), profibrinolytic and anti- a higher risk of stroke, compared with those without
fibrinolytic proteins; (2) portal hypertension leading cirrhosis, and increased mortality.20
to hypersplenism with thrombocytopenia and contrib- Observational studies suggest anticoagulation is
uting to (3) endothelial activation with increased associated with a reduced risk of stroke and mortality
Factor (F) VIII, von Willebrand factor (VWF) and in patients with Child-­Pugh A and B cirrhosis.20 There
plasminogen activator inhibitor-­1 (PAI-­1) and platelet is a paucity of data in patients with Child-­ Pugh C

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
activation.4 Routine laboratory tests do not measure cirrhosis. Anticoagulation for stroke prevention has
these complex changes, with the PT reflecting reduc- not been consistently associated with an increased
tions in procoagulant factors (FII, FV, FVII, FX) only. risk of bleeding.20 21 This may in part be explained by
Global haemostatic assays demonstrate overall rebal- anticoagulation reducing the risk of portal hyperten-
anced haemostasis with features of hypercoagula- sive bleeding but increasing non-­portal hypertensive
bility, particularly in those with compensated disease. bleeding (as seen in studies of PVT discussed below).
For example, thrombin generation measured in the Additionally, selection bias and/or residual confounding
presence of a protein C activator (eg, thrombomod- associated with the observational and predominantly
ulin) shows patients with cirrhosis have increased retrospective study design may further contribute to
thrombin-­generating capacity compared with healthy this finding. Of note, major bleeding was frequent in
controls, even in those with decompensated disease.5 patients with cirrhosis both on and off anticoagulants
Increased VWF appears to compensate for thrombocy- (6% to 20%), consistent with the known increased risk
topenia with additional evidence of platelet activation of bleeding associated with portal hypertension. For
in patients with cirrhosis.6 7 Clot lysis assays suggest comparison, major bleeding occurred in 5%–6% of
reduced fibrinolysis in patients with liver disease, with those randomised to either warfarin or DOAC in the
additional evidence of reduced clot permeability.8 9 It seminal AF studies (which largely excluded patients
is now recognised that neither prolonged PT/interna- with cirrhosis).22
tional normalised ratio (INR) nor thrombocytopenia
predicts bleeding in patients with cirrhosis.4 10–12 Portal vein thrombosis
Further details of the changes in haemostasis in patients PVT is the most frequent presentation of venous
with cirrhosis are available elsewhere.4 thromboembolism (VTE) in patients with cirrhosis,
with a reported annual incidence in prospective studies
Baseline bleeding risk in patients with cirrhosis ranging from 3%–24%, with incidence increasing in
Patients with cirrhosis are at increased risk of bleeding, parallel with the severity of liver disease.3 Involvement
with rates of 2%–5% per year reported, with most of mesenteric and/or splenic veins is less common,
bleeding events secondary to portal hypertension.13 14 and the approach to management is similar to those
The risk of portal hypertensive bleeding is mediated with isolated PVT. Severity of portal hypertension is
by portal pressures with no evidence that haemostasis the other key factor predisposing to PVT.10 17 The aim
contributes to this risk.14 Prolonged PT/INR and/or of treatment is to prevent thrombosis progression and
thrombocytopenia do not predict bleeding in patients to facilitate recanalisation of the portal vein to mini-
with cirrhosis.11 14 15 Indeed, thrombocytopenia is mise technical difficulties at future transplantation.
associated with an increased risk of PVT and repre- There are reports of spontaneous regression of PVT,
sents a marker of severity of portal hypertension.16 17 particularly with non-­occlusive presentations.23 Meta-­
analysis of predominantly small, retrospective obser-
EVIDENCE TO SUPPORT THE USE OF vational studies reports anticoagulation reduces the
ANTICOAGULATION IN PATIENTS WITH risk of thrombosis progression (OR, 0.141; 95% CI
CIRRHOSIS 0.06 to 0.31) and improves recanalisation (OR,
Atrial fibrillation 4.8; 95% CI 2.7 to 8.7).24 There was no difference
AF is commonly encountered in patients with cirrhosis, in overall bleeding rates between those treated with
with a large meta-­analysis reporting a prevalence of anticoagulants or not (11%), but with a reduction in
5% (95% CI 2.8% to 8.6%).18 Increasing liver disease variceal bleeding (OR, 0.232; 95% CI 0.06 to 0.94)
severity is associated with an increased prevalence of across the four studies reporting this outcome.24 A
AF, with 20% of those with Model for End-­Stage Liver later individual patient data (n=500) meta-­analysis of
Disease (MELD) >30 found to have AF compared five studies reported reduced mortality associated with
with 3.7% in those with MELD of ≤10.19 Underlying anticoagulation (adjusted sub-­ distribution HR 0.59;
aetiology of cirrhosis also influences the risk of AF. 95% CI 0.49 to 0.70), independent of severity of PVT
Steatotic liver disease is associated with obesity, hyper- and subsequent recanalisation.25 In this analysis, an
tension and heart failure, all of which also increase increased risk of non-­portal hypertensive bleeding was

2 Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994

 Hepatology

reported (9.7% vs 1.7%) with a non-­significant reduc- with possible reduced all-­cause mortality and reduced

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
tion in portal hypertensive bleeds (9.3% vs 13.9%) in major bleeding risk. Apixaban and rivaroxaban are
those treated with anticoagulants compared with those the most commonly reported DOACs evaluated in
without, respectively.25 patients with cirrhosis, with limited observational data
suggesting a lower bleeding risk with apixaban.28 29 It
Prevention of hepatic decompensation is important to acknowledge additional limitations due
A small randomised controlled trial of daily prophy- to observational design, including selection bias and
lactic enoxaparin (40 mg) versus no intervention in 80 residual confounding effects.
participants with decompensated cirrhosis (Child-­Pugh A longitudinal, nationwide claims analysis in the
B7 – C10) reported enoxaparin was associated with USA of patients with AF (mean CHA2DS2VASc score,

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
reduced hepatic decompensation, PVT and improved 4.2) and cirrhosis (n=32 487) found 45% were
survival.26 The benefit of enoxaparin persisted beyond prescribed oral anticoagulants within 180 days of AF
the 48week- duration of intervention, and of note, diagnosis.28 Those prescribed anticoagulants were less
there were no bleeding events reported. A subsequent likely to have decompensated disease (11%) compared
placebo-­ controlled randomised controlled trial of with those without a prescription (19%). Over time,
daily prophylactic rivaroxaban (10 mg) in patients with warfarin prescribing reduced and DOAC prescribing
decompensated cirrhosis (published in abstract form) increased (from 19% prescribed DOACs in 2012 to
confirmed a reduction in portal hypertensive compli- 78% in 2019 of all oral anticoagulant prescription).
cations (HR 0.47, 95% CI 0.22 to 0.98).27 However, A similar prescribing pattern was seen in those with
in this study, non-­portal hypertensive bleeding events both compensated and decompensated disease.28 A
were more frequent (26.6%). These data suggest anti- further study in the UK examined persistence with oral
coagulation may benefit patients with cirrhosis by anticoagulants among those with AF and liver disease
slowing disease progression even in the absence of a (n=3167).30 A minority of the cohort had cirrhosis
conventional indication for anticoagulation. (n=376) and were predominantly prescribed DOACs
(65%). At 3 years, only 39% and 18% continued
CHOICE OF ANTICOAGULANT DOAC or VAK, respectively. The low persistence rate
In the absence of cirrhosis, DOACs are now the antico- was only partially explained by clinical events, with
agulant of choice for most patients with non-­valvular <25% having either ischaemic or bleeding events
AF and VTE. Direct thrombin inhibitor, dabigatran, in the 180 days preceding discontinuation.30 This is
and direct Factor Xa inhibitors, apixaban, edoxaban vastly lower than persistence rates in non-­liver popula-
and rivaroxaban have all been demonstrated to be tions31 and warrants further exploration.
non-­inferior to warfarin for both stroke prevention
in AF and treatment of DVT/PE, with a reduced risk Deep vein thrombosis/pulmonary embolism
of major bleeding in those without liver disease. Addi- A retrospective cohort study using claims data from the
tional advantages over warfarin/vitamin K antagonists USA compared DOACs to warfarin for the treatment of
(VKAs) include predictable pharmacokinetics obviating DVT/PE with a composite primary outcome of hospi-
the need for routine drug monitoring, fixed dosing, talisation for recurrent VTE or bleeding in patients with
rapid onset and offset of action, along with fewer liver disease.32 Subgroup analyses of 2449 patients with
food and drug interactions. The characteristics of both cirrhosis (±decompensation) revealed DOACs were asso-
parenteral and oral anticoagulants available for VTE ciated with a reduction in the composite primary outcome
management are summarised in table 1. All DOACs compared with warfarin (HR, 0.64; 95% CI 0.43 to 0.96
have variable non-­ renal metabolism, and therefore and HR, 0.44; 95% CI 0.26 to 0.73 for all patients with
patients with cirrhosis may experience altered pharma- cirrhosis and those with decompensation, respectively). On
cokinetics and/or clinical effects due to impaired liver further analysis of the whole study population (including
synthesis of coagulation factors, altered plasma protein a majority without cirrhosis, n=6028), the difference in
binding, reduced liver metabolism and/or renal clear- primary outcome was driven by a reduction in hospitalisa-
ance in hepatorenal syndrome. Pharmacokinetic data tion due to bleeding with no difference in hospitalisation
are summarised elsewhere21 with a focus on clinical for recurrent VTE.32 There was no difference in bleeding
data in this review. outcomes between apixaban and rivaroxaban.

Atrial fibrillation Portal vein thrombosis

Observational studies comparing the efficacy and For patients presenting acutely unwell, particularly with
safety of DOACs compared with warfarin are summa- mesenteric vasculature involvement, low molecular
rised in table 2. Only a minority of patients with weight heparin (LMWH) may be the preferred initial anti-
cirrhosis were included, with limited information coagulant in view of its short, predictable half-­life. There
regarding disease severity and a lack of data pertaining are limited data evaluating DOACs versus LMWH/VKA
to those with Child-­ Pugh C disease. However, the for the management of PVT. Meta-­analyses and systematic
studies suggest similar efficacy of DOACs to warfarin reviews of small, predominantly retrospective studies have

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 3

Hepatology

Table 1 Characteristics of anticoagulants available to treat venous thromboembolism

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Vitamin K
antagonist LMWH Fondaparinux Dabigatran Apixaban Edoxaban Rivaroxaban
Dosing for VTE Variable once Weight-­based, Weight-­based, 150 mg twice Twice daily 60 mg once Twice daily for 3
daily, once daily or 7.5 mg once daily daily daily weeks then once
based on INR twice daily daily
Initiation for VTE Overlap with Twice daily Weight-­based, once ≥5 days LMWH 10 mg twice ≥5 days LMWH 15 mg twice daily
LMWH until INR preferred daily then switch to daily for 7 days then switch to for 3 weeks
>2 dabigatran edoxaban
Subsequent dosing Based on INR Consider switch No change Continue same 5 mg twice daily Continue same 20 mg daily from

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
for VTE to once daily if dose from day 8. dose day 22.
continuing beyond Consider ↓ Consider ↓ to
acute phase to 2.5 mg at 10 mg after
6 months for 3–6 months for
secondary secondary VTE
prevention prevention
Dose modification Based on INR Dose reduction if <50 kg, 5 mg once Y to 110 mg n/a Y to 30 mg daily n/a
CrCl <30 mL/min daily twice daily for selected
>100 kg, 10 mg for selected patients†
once daily patients*
Drug monitoring Y N N N N N N
Hepatobiliary Predominant Minimal Minimal 20% 73% 50% 65%
& intestinal
metabolism
Renal excretion Minimal Predominant Predominant 80% 27% 50% 66%
CKD Y ↓dose Do not use Do not use Limited Limited Limited experience
CrCl, 15–30 mL/min Y Avoid Do not use Do not use experience experience Do not use
<15 mL/min Do not use Do not use
Reversal agent Y Partial N Y Y, restricted use N Y, restricted use in
in some regions some regions
Adapted from Northup et al with permission from Wolters Kluwer Health. The Creative Commons license does not apply to this content. Use of the
material in any format is prohibited without written permission from the publisher, Wolters Kluwer Health. Please contact permissions@lww.com for
further information.
*Dabigatran 110 mg two times per day in patients >80 years of age and those also on verapamil. Consider dose reduction in patients aged 75–80 years,
with moderate renal impairment (CrCl 30–50 mL/min), with gastritis, oesophagitis or gastro-­oesophageal reflux or at increased risk of bleeding.
† Edoxaban 30 mg once daily for patients with CrCl 15–50 mL/min or <60 kg or taking concomitant P-­glycoprotein (PgP) inhibitors.
CKD, chronic kidney disease; CrCl, creatinine clearance; INR, international normalised ratio; LMWH, low molecular weight heparin; N, no; n/a, not
applicable; VTE, venous thromboembolism; Y, yes.

been attempted.33 34 However, all analyses include a now-­ score is 2 or more, the annual risk of stroke outweighs
retracted randomised controlled trial35 and are limited by the risk of bleeding, and anticoagulation is recom-
the low quality of the source data, including retrospec- mended. For men, with a score of 1, anticoagulation
tive observational design, small patient numbers, highly should be considered with a review of bleeding risk
selective inclusion criteria and a suboptimal LMWH/ and patient preferences. Guidance documents suggest
VKA comparator (eg, with inadequate target INR, lack of raising the threshold for anticoagulation in patients
overlap with LMWH, lack of evaluation of time in thera- with cirrhosis based on expert opinion, as summarised
peutic range).21 Bearing this in mind, the limited published in table 3 and figure 1. Of note, there is very limited
data suggest DOACs are associated with at least equiva- evaluation of anticoagulation in patients with AF
lent rates of portal vein recanalisation without an increase and Child-­Pugh C cirrhosis; thus, an individualised,
in bleeding risk compared with LMWH/VKA. Further patient-­centred approach is suggested.
well-­designed studies are required to confirm the role of
DOACs in this setting.
Deep vein thrombosis/pulmonary embolism
APPROACH TO PATIENTS Patients with acute DVT/PE should be offered anti-
General considerations coagulation due to the risk of thrombosis extension,
Atrial fibrillation embolisation and mortality without treatment.38
In patients without cirrhosis, the need for anticoagula- These risks are unlikely to be different in patients with
tion is generally evaluated using a validated risk strati- cirrhosis, and anticoagulation would be recommended
fication tool, CHA2DS2VASc (figure 1).36 37 Where the in most scenarios (see table 3 and figure 2). Where clot

4 Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994

 Hepatology

Table 2 Direct oral anticoagulants compared with warfarin for stroke prevention in AF in cirrhosis

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Study Design, location Population Child-­Pugh score Efficacy Bleeding
Lee et al, 201950 Subgroup analysis of patients n=768 Not reported Ischaemic stroke: DOACs ↔ DOACs ↔ warfarin
with cirrhosis in retrospective (equates to 2% warfarin Hospitalisation for major
cohort study in South Korea of whole study HR, 0.62; 95% CI 0.25 to 1.46 bleeding:
population) HR, 0.59; 95% CI 0.30
to 1.12
Intracranial haemorrhage:
HR, 0.48; 95% CI, 0.15
to 1.40

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
NB 58% of patients
received dosing
Menichelli et al, Meta-­analysis of seven n=40 809 with AF Not reported Ischaemic stroke: DOACs ↔ DOACs ↓ vs warfarin
202051 studies (five retrospective and advanced liver warfarin Major bleeding:
cohort studies, one disease or cirrhosis HR, 0.89, 95% CI 0.62 to 1.27 HR, 0.63; 95% CI 0.43
prospective cohort study and (NB only 6.3% with All-­cause death to 0.93
one RCT) cirrhosis) DOACs ↓ vs warfarin Intracranial haemorrhage:
HR, 0.77; 95% CI 0.61 to 0.96 HR, 0.49; 95% CI 0.40
to 0.59
Lawal et al, 202352 Retrospective cohort study Subgroup of Not reported Ischaemic stroke/systemic Major bleeding
in USA n=2940 with AF embolism: DOACs ↓ vs warfarin
and cirrhosis DOACs ↔ warfarin HR, 0.70; 95% CI 0.53
HR, 0.90; 95% CI 0.51 to 1.57 to 0.93
All-­cause death:
DOACs ↓ vs warfarin
HR, 0.75; 95% CI 0.62 to 0.91
Baylo et al, 202353 RCT: dabigatran 110 mg n=56 Child-­Pugh A, 43%; Ischaemic stroke/systemic Major bleeding: dabigatran
twice daily vs warfarin in B, 57% embolism: no events ↔ warfarin, 0% vs 4%
Ukraine
Simon et al, 202429 Retrospective propensity-­ n=5704 Not reported Ischaemic stroke warfarin ↔ rivaroxaban ↑ vs apixaban
matched cohort study in USA apixaban Major bleeding
HR, 1.09 (0.51 to 2.35) HR, 1.47 (1.11 to 1.94)
rivaroxaban ↔ apixaban warfarin ↑ vs apixaban
HR, 0.88 (0.37 to 2.06) Major bleeding:
All-­cause mortality: warfarin ↔ HR, 1.38 (1.03 to 1.84)
apixaban Intracranial bleeding:
HR, 1.1 (0.9 to 1.35) HR, 2.85 (1.24 to 6.59)
rivaroxaban ↔ apixaban
HR, 1.09 (0.91 to 1.31)
Adapted from Carlin et al 21 with permission from Elsevier.
↓, reduced; ↔, equivalent; ↑, increased; AF, atrial fibrillation; DOAC, direct oral anticoagulant; MB, major bleeding.

burden is minimal, for example isolated distal DVT approach to anticoagulation in figure 2. A treatment
or subsegmental PE, surveillance may be considered a duration of at least 6 months is recommended. Due to
reasonable alternative, particularly if there are addi- the high risk of PVT recurrence, extended anticoagula-
tional concerns for bleeding. tion is suggested for liver transplant candidates.21 39 40

Portal vein thrombosis

Patients presenting with symptomatic PVT should be Bleeding risk assessment
offered anticoagulation. As PVT is frequently detected An assessment of bleeding risk should be under-
on imaging performed for other reasons (eg, hepato- taken prior to initiating anticoagulation to identify
cellular carcinoma surveillance or during acute illness), modifiable bleeding risk factors such as use of anti-
the extent and location of thrombosis should be consid- platelet agents, non-­steroidal anti-­inflammatory drugs,
ered for those without symptoms. There are reports anaemia, alcohol consumption and poorly controlled
of spontaneous regression of non-­occlusive (<50%) hypertension. There are several tools available to aid
PVT, particularly involving intrahepatic branches, in this assessment,41 but none are specifically validated
and surveillance with repeat imaging at 3 months in patients with cirrhosis. Such tools should be used
is suggested.10 Treatment should be initiated in the to identify and address modifiable risk factors for
event of progressive thrombosis. Guidance recommen- bleeding (when appropriate) rather than to withhold
dations are summarised in table 3 with a suggested anticoagulation for those with a high bleeding score.

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 5

 Hepatology

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Figure 1 Approach to anticoagulation for non-­valvular atrial fibrillation in patients with cirrhosis Child-­Pugh A or B as proposed by the
International Society of Thrombosis and Haemostasis guidance.21 54–56 Created in BioRender. Roberts, L. (2025) https://BioRender.com/mmlmhsh.
CVA, cerebrovascular accident; DOAC, direct oral anticoagulant; F, female; LAAO, left atrial appendage occlusion; M, male; SE, systemic embolism;
TIA, transient ischaemic attack; VKA, vitamin K antagonist.

Choice of anticoagulant anticoagulation initiation is not mandatory; however,

In patients without cirrhosis, DOACs are now interval endoscopic surveillance should continue
preferred for both stroke prevention in non-­valvular as per existing guidance recommendations.10 39 For
AF and DVT/PE due to at least equivalent efficacy and patients with acute portal hypertensive bleeding
the lower risk of bleeding. In patients with cirrhosis, it who then require anticoagulation for concurrent
is reasonable to extrapolate this approach to patients new AF, DVT/PE or PVT, anticoagulation should be
with Child A and B cirrhosis. The limited data in deferred until haemostasis is achieved. LMWH may
patients with Child-­Pugh C means an individualised be preferred initially, particularly where further endo-
approach is needed, and LMWH with/without VKA scopic variceal band ligation is planned, as it can then
(provided the baseline PT/INR is not prolonged) will be simply delayed/omitted on the day of procedure.42
usually be preferred. In patients with VTE undergoing As discussed above, anticoagulation is associated with
frequent interventions, such as endoscopic variceal a reduced risk of portal hypertensive bleeding. Addi-
band ligation or regular paracentesis, LMWH may tionally, anticoagulation is not associated with worse
be preferred due to its short, predictable half-­ life, clinical outcomes in the event of gastrointestinal
enabling the last dose to be administered 24 hours bleeding.43
before the planned procedure. Of note, there is no
data to support LMWH as a long-­term strategy for
stroke prevention in AF. Prolonged baseline prothrombin time/INR
Prolonged PT/INR does not predict bleeding in
Specific considerations in patients with cirrhosis patients with cirrhosis and represents a marker of liver
Presence of Varices disease severity.4 It should therefore be considered in
In the absence of active bleeding, varices should not this context and may preclude the use of DOACs, given
be considered a contraindication to anticoagula- the limited data in decompensated liver disease. VKA
tion. In patients without prior bleeding, the use of a should also be avoided in those with a prolonged base-
non-­selective beta blocker should be considered for line PT/INR, as there is limited data available to guide
primary prevention. Evaluation of varices prior to adjusting the target INR. LMWH would generally be

6 Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994

 Hepatology

Table 3 Summary of guideline recommendations for anticoagulation in patients with cirrhosis and atrial fibrillation, deep vein

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
thrombosis/pulmonary embolism or portal vein thrombosis
Atrial fibrillation
Guidance group ISTH 202421 ACC/AHA 202336 AGA 202157 EHRA 202158
When to initiate Child-­Pugh A/B ‘at increased risk of Consider for ‘higher Baseline assessment for
♂ CHA2DS2VASc≥2 thromboembolism’ CHA2DS2VASc scores’ prior thromboembolism
♀ CHA2DS2VASc≥3 Reasonable to avoid and previous bleeding (incl
Consider for ♂ in Child-­Pugh C, lower variceal bleeding) in the
CHA2DS2VASc of 1, and CHA2DS2VASc absence of coagulopathy/
♀ 2. low platelet count
Child-­Pugh C individualise

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Choice of agent Child-­Pugh A/B Child-­Pugh A DOACs No recommendation Child-­Pugh A DOACs
DOAC preferred to VKA Child-­Pugh B apixaban, Child-­Pugh B apixaban,
dabigatran or edoxaban in dabigatran, edoxaban
preference to VKA
Other Consider LAAO if high risk ‘Highest bleeding risk’ →
and anticoagulation failed consider alternate strategy
or contraindicated

Deep vein thrombosis

Guidance group ISTH 2024 21 EASL 2022 12

Choice of agent Child-­Pugh A: DOAC or LMWH/VKA Child-­Pugh A: DOAC or LMWH/VKA

Child-­Pugh B: DOAC or LMWH/VKA* Child-­Pugh B: LMWH, caution with DOAC
Child-­Pugh C: LMWH alone or LMWH/VKA* Child-­Pugh C: LMWH (UFH in renal failure)

*if normal baseline INR

Portal vein thrombosis (PVT)

Guidance group VALDIG 202444 ISTH 2024 21 BAVENO VII 2022 39 AASLD 2021 10 AGA 2019 40
Treatment indications ► Recent (<6 months) Symptomatic & ► Recent (<6 months) ► Recent (<6 months) ► Acute PVT with
completely or partially asymptomatic with thrombosis of PV trunk main PVT or SMV complete occlusion
occlusive (>50%) progression ► Symptomatic PVT thrombosis 1. Chronic PVT
thrombosis of the main ► Potential LT candidate ► Ischaemic symptoms with:Progression of
portal vein ► Minimally occlusive thrombus or SMV
► Symptomatic PVT PV that is rapidly involvement
(independent of extent) progressing or with 2. History of bowel
► Progression of SMV compromise ischaemia
PVT without 3. Inherited
anticoagulation thrombophilia
► Extension to the 4. Patients awaiting liver
superior mesenteric transplant
vein
► Any PVT in LT
candidate
Choice of agent For symptomatic, LMWH Child-­Pugh A/B: DOAC or LMWH, VKA or DOAC Initiate with LMWH Initiate with LMWH or UFH
initially with VKA preferred LMWH/VKA Maintain with LMWH, VKA Maintain with LMWH, VKA
for those on LT list Child-­Pugh C: LMWH only or DOAC or DOAC (for Child-­Pugh
(provided normal baseline or bridge to VKA if normal Use DOACs with caution in A only)
INR) baseline PT/INR Child-­Pugh B and avoid in
Child-­Pugh A: DOAC Child-­Pugh C
Child-­Pugh B: use DOAC
with caution
Duration Uncertain due to high At least 6 months Until recanalisation and at At least 6 months
risk of recurrence on Continue long term if a least 6 months
withdrawal transplant candidate Extend anticoagulation for
Case-­by-­case decision to those on transplant list and
extend in non-­transplant consider extension for all
candidates with regular others based on thrombosis
bleeding risk assessment vs bleeding risk
♂, males; ♀, females.
AASLD, American Association for the Study of Liver Diseases; ACC, American College of Cardiology; AGA, American Gastroenterological Association; AHA, American Heart Association;
BAVENO VII, Baveno VII Faculty; DOAC, direct oral anticoagulant; EASL, European Association for the Study of the Liver; EHRA, European Heart Rhythm Association; INR, international
normalised ratio; ISTH, International Society of Thrombosis and Haemostasis; LMWH, low molecular weight heparin; LT, liver transplant; PVT, portal vein thrombosis; SMV, superior
mesenteric vein; VALDIG, Vascular Liver Disease Group; VKA, vitamin K antagonist.

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 7

 Hepatology

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
Figure 2 Approach to the management of venous thromboembolism in patients with cirrhosis. DVT, deep vein thrombosis; INR, international
normalised ratio; LMWH, low molecular weight heparin; PE, pulmonary embolism; PVT, portal vein thrombosis; VKA, vitamin K antagonist.

preferred for the management of VTE in those with It should not be considered an absolute contraindi-
Child-­Pugh C disease and/or a prolonged PT/INR. cation to anticoagulation, particularly for patients
with acute symptomatic VTE or those at high risk
Thrombocytopenia of stroke.21 Alternate causes for thrombocytopenia
Thrombocytopenia is common in those with cirrhosis, should be considered, for example, haematinic defi-
with most having a platelet count below the lower ciency, particularly for patients with a platelet count
limit of normal. Of note, severe thrombocytopenia
of <30×109 /L, and addressed appropriately. A case-­
(platelet count <50×109 /L) is less frequent in ambu-
by-­case decision should be made based on the risk of
latory patients but may be more commonly encoun-
tered in those admitted to hospital with acute illness. thrombosis embolisation and/or extension, bleeding
Platelet counts of <30×109 /L are rarely encoun- risk factors and patient preference. For patients with
tered.11 Of note, thrombocytopenia is a marker of the acute DVT/PE, anticoagulation should be offered
severity of portal hypertension and is associated with due to the risk of thrombosis progression/embolisa-
an increased risk of PVT.16 17 While thrombocytopenia tion unless there is very limited volume thrombosis,
is a risk factor for bleeding in unselected (non-­liver) for example, distal DVT or isolated subsegmental
patients commencing anticoagulation, it has not been PE, where initial surveillance may be considered.21
found to predict bleeding in patients with cirrhosis. Similarly for liver transplant candidates with recent

8 Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994

 Hepatology

occlusive PVT or progressive thrombosis, anticoagu- Supplemental material This content has been supplied by the
author(s). It has not been vetted by BMJ Publishing Group

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
lation should be offered. Limited (BMJ) and may not have been peer-­reviewed. Any
opinions or recommendations discussed are solely those of
Liver transplant candidates the author(s) and are not endorsed by BMJ. BMJ disclaims
The approach to management of anticoagulation at all liability and responsibility arising from any reliance placed
on the content. Where the content includes any translated
the time of transplant needs to be considered prior to material, BMJ does not warrant the accuracy and reliability of
listing. While LMWH and DOACs are short-­ acting the translations (including but not limited to local regulations,
with potential minimal residual activity by the time clinical guidelines, terminology, drug names and drug dosages),
and is not responsible for any error and/or omissions arising
transplant proceeds, reversibility remains a concern. from translation and adaptation or otherwise.
VKAs are readily reversed with intravenous vitamin K

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
ORCID iD
and prothrombin complex concentrate and remain the Lara Roberts http://orcid.org/0000-0003-3871-8491
preferred agent for those listed for transplant with a
normal baseline PT/INR.44 LMWH is suggested as the
preferred agent for those with a prolonged PT/INR. REFERENCES
1 Vandenberk B, Altieri MH, Liu H, et al. Review article:
There are antidotes available for dabigatran (idaruci-
diagnosis, pathophysiology and management of atrial
zumab) as well as apixaban and rivaroxaban (andexanet fibrillation in cirrhosis and portal hypertension. Aliment
alpha). However, there is limited data regarding the Pharmacol Ther 2023;57:290–303.
use of andexanet alpha in urgent/emergency surgery 2 Ambrosino P, Tarantino L, Di Minno G, et al. The risk of
and none to the author’s knowledge in liver transplan- venous thromboembolism in patients with cirrhosis. Thromb
Haemost 2017;117:139–48.
tation. Furthermore, the randomised controlled trial 3 Senzolo M, Garcia-­Tsao G, García-­Pagán JC. Current
of andexanet alpha in the context of intracranial haem- knowledge and management of portal vein thrombosis in
orrhage raised concerns regarding increased arterial cirrhosis. J Hepatol 2021;75:442–53.
thrombotic events.45 It should therefore be avoided in 4 Lisman T, Hernandez‐Gea V, Magnusson M, et al. The
concept of rebalanced hemostasis in patients with liver
the liver transplant setting. Some have suggested the
disease: Communication from the ISTH SSC working group
time interval between liver transplant availability and on hemostatic management of patients with liver disease. J
proceeding with transplantation allows continued use Thromb Haemost 2021;19:1116–22.
of Factor Xa inhibitors with limited published experi- 5 Fisher C, Patel VC, Stoy SH, et al. Balanced haemostasis
ence of this approach.1 46 47 Similarly, there is limited with both hypo- and hyper-­coagulable features in critically
ill patients with acute-­on-­chronic-­liver failure. J Crit Care
experience of dabigatran reversal with idarucizumab 2018;43:54–60.
at liver transplantation, although a prospective trial in 6 Lisman T, Bongers TN, Adelmeijer J, et al. Elevated levels of
bleeding and emergency surgery demonstrated efficacy von Willebrand Factor in cirrhosis support platelet adhesion
and safety.48 49 despite reduced functional capacity. Hepatology 2006;44:53–
61.
7 Lecchi A, Tosetti G, Ghali C, et al. Comprehensive
SUMMARY investigation of platelet function in patients with cirrhosis.
Thromb Res 2024;237:64–70.
Patients with cirrhosis are at increased risk of AF, PVT 8 Lisman T, Leebeek FW, Mosnier LO, et al. Thrombin-­
and DVT/PE, where anticoagulation should be offered activatable fibrinolysis inhibitor deficiency in cirrhosis is not
as part of standard care. Despite the exclusion of associated with increased plasma fibrinolysis. Gastroenterology
patients with cirrhosis from the seminal randomised 2001;121:131–9.
9 Hugenholtz GCG, Macrae F, Adelmeijer J, et al. Procoagulant
controlled trials of DOACs, they appear to be an
changes in fibrin clot structure in patients with cirrhosis
acceptable and likely safer option compared with VKA are associated with oxidative modifications of fibrinogen. J
for AF and DVT/PE for patients with Child-­Pugh A Thromb Haemost 2016;14:1054–66.
and B disease. Further prospective studies are required 10 Northup PG, Garcia-­Pagan JC, Garcia-­Tsao G, et al. Vascular
to evaluate the efficacy and safety of DOACs for PVT Liver Disorders, Portal Vein Thrombosis, and Procedural
Bleeding in Patients With Liver Disease: 2020 Practice
and of anticoagulation broadly in patients with Child-­ Guidance by the American Association for the Study of Liver
Pugh C disease. Diseases. Hepatology 2021;73:366–413.
11 Roberts LN, Lisman T, Stanworth S, et al. Periprocedural
Contributors LR researched and wrote the paper and is the management of abnormal coagulation parameters and
guarantor. thrombocytopenia in patients with cirrhosis: Guidance from
Funding The authors have not declared a specific grant for this the SSC of the ISTH. J Thromb Haemost 2022;20:39–47.
research from any funding agency in the public, commercial or 12 Villa E, Bianchini M, Blasi A, et al. EASL Clinical Practice
not-­for-­profit sectors. Guidelines on prevention and management of bleeding and
Competing interests LR has received speaker fees from thrombosis in patients with cirrhosis. J Hepatol 2022;76:1151–
Bayer (manufacturer of rivaroxaban) and Viatris (marketer of 84.
fondaparinux). 13 Roberts LN, Bernal W. Incidence of Bleeding and Thrombosis
Patient consent for publication Not applicable. in Patients with Liver Disease. Semin Thromb Hemost
2020;46:656–64.
Ethics approval Not applicable. 14 Basili S, Raparelli V, Napoleone L, et al. Platelet Count Does
Provenance and peer review Commissioned; internally peer Not Predict Bleeding in Cirrhotic Patients: Results from the
reviewed. PRO-­LIVER Study. Am J Gastroenterol 2018;113:368–75.

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 9

 Hepatology
15 Intagliata NM, Rahimi RS, Higuera-­de-­la-­Tijera F, et al. choice: effectiveness and safety of direct oral anticoagulants

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
Procedural-­Related Bleeding in Hospitalized Patients With versus warfarin. Res Pract Thromb Haemost 2024;8:102293.
Liver Disease (PROC-­BLeeD): An International, Prospective, 33 Koh JH, Liew ZH, Ng GK, et al. Efficacy and safety of direct
Multicenter Observational Study. Gastroenterology oral anticoagulants versus vitamin K antagonist for portal vein
2023;165:717–32. thrombosis in cirrhosis: A systematic review and meta-­analysis.
16 Violi F, Corazza G, Caldwell S, et al. Incidence and Recurrence Dig Liver Dis 2022;54:56–62.
of Portal Vein Thrombosis in Cirrhotic Patients. Thromb 34 Ng CH, Tan DJH, Nistala KRY, et al. A network meta-­analysis
Haemost 2019;119:496–9. of direct oral anticoagulants for portal vein thrombosis in
17 Turon F, Driever EG, Baiges A, et al. Predicting portal cirrhosis. Hepatol Int 2021;15:1196–206.
thrombosis in cirrhosis: A prospective study of clinical, 35 Hanafy AS, Abd-­Elsalam S, Dawoud MM. Retraction notice
ultrasonographic and hemostatic factors. J Hepatol to “Randomized controlled trial of rivaroxaban versus
2021;75:1367–76. warfarin in the management of acute non-­neoplastic portal

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
18 Chokesuwattanaskul R, Thongprayoon C, Bathini T, et al. vein thrombosis” [Vascular Pharmacology 113 (2019) 86-­91].
Epidemiology of atrial fibrillation in patients with cirrhosis Vascul Pharmacol 2023;153:107142.
and clinical significance: a meta-­analysis. Eur J Gastroenterol 36 Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/
Hepatol 2019;31:514–9. ACCP/HRS Guideline for the Diagnosis and Management
19 Huang WA, Dunipace EA, Sorg JM, et al. Liver Disease as a of Atrial Fibrillation: A Report of the American College of
Predictor of New-­Onset Atrial Fibrillation. J Am Heart Assoc Cardiology/American Heart Association Joint Committee on
2018;7:e008703. Clinical Practice Guidelines. Circulation 2024;149:e1–156.
20 Chokesuwattanaskul R, Thongprayoon C, Bathini T, et al. 37 Hindricks G, Potpara T, Dagres N, et al. 2020 ESC Guidelines
Efficacy and safety of anticoagulation for atrial fibrillation in for the diagnosis and management of atrial fibrillation
patients with cirrhosis: A systematic review and meta-­analysis. developed in collaboration with the European Association
Dig Liver Dis 2019;51:489–95. for Cardio-­Thoracic Surgery (EACTS): The Task Force for
21 Carlin S, Cuker A, Gatt A, et al. Anticoagulation for stroke the diagnosis and management of atrial fibrillation of the
prevention in atrial fibrillation and treatment of venous European Society of Cardiology (ESC) Developed with
thromboembolism and portal vein thrombosis in cirrhosis: the special contribution of the European Heart Rhythm
guidance from the SSC of the ISTH. J Thromb Haemost Association (EHRA) of the ESC. Eur Heart J 2021;42:373–
2024;22:2653–69. 498.
22 Ruff CT, Giugliano RP, Braunwald E, et al. Comparison of the 38 Cox C, Roberts LN. Basics of diagnosis and treatment of
efficacy and safety of new oral anticoagulants with warfarin in venous thromboembolism. J Thromb Haemost 2025;23:1185–
patients with atrial fibrillation: a meta-­analysis of randomised 202.
trials. The Lancet 2014;383:955–62. 39 de Franchis R, Bosch J, Garcia-­Tsao G, et al. Baveno VII
23 Nery F, Chevret S, Condat B, et al. Causes and consequences - Renewing consensus in portal hypertension. J Hepatol
of portal vein thrombosis in 1,243 patients with cirrhosis: 2022;76:959–74.
results of a longitudinal study. Hepatology 2015;61:660–7. 40 O’Leary JG, Greenberg CS, Patton HM, et al. AGA Clinical
24 Loffredo L, Pastori D, Farcomeni A, et al. Effects of Practice Update: Coagulation in Cirrhosis. Gastroenterology
Anticoagulants in Patients With Cirrhosis and Portal Vein 2019;157:34–43.
Thrombosis: A Systematic Review and Meta-­analysis. 41 den Exter PL, Woller SC, Robert‐Ebadi H, et al. Management
Gastroenterology 2017;153:480–7. of bleeding risk in patients who receive anticoagulant therapy
25 Guerrero A, Campo LD, Piscaglia F, et al. Anticoagulation for venous thromboembolism: Communication from the ISTH
improves survival in patients with cirrhosis and portal vein SSC Subcommittee on Predictive and Diagnostic Variables in
thrombosis: The IMPORTAL competing-­risk meta-­analysis. J Thrombotic Disease. J Thromb Haemost 2022;20:1910–9.
Hepatol 2023;79:69–78. 42 Bianchini M, Cavani G, Bonaccorso A, et al. Low molecular
26 Villa E, Cammà C, Marietta M, et al. Enoxaparin prevents weight heparin does not increase bleeding and mortality post-­
portal vein thrombosis and liver decompensation in patients endoscopic variceal band ligation in cirrhotic patients. Liver Int
with advanced cirrhosis. Gastroenterology 2012;143:1253– 2018;38:1253–62.
60. 43 Cerini F, Gonzalez JM, Torres F, et al. Impact of
27 Sanchez AP, Turon F, Martinez J, et al. Rivaroxaban improves anticoagulation on upper-­gastrointestinal bleeding in cirrhosis.
survival and decompensation in cirrhotic patients with A retrospective multicenter study. Hepatology 2015;62:575–
moderate liver dysfunction. Double-­blind, placebo-­controlled 83.
trial. J Hepatol 2023;78:S2–3. 44 Elkrief L, Hernandez-­Gea V, Senzolo M, et al. Portal vein
28 Simon TG, Schneeweiss S, Singer DE, et al. Prescribing thrombosis: diagnosis, management, and endpoints for future
Trends of Oral Anticoagulants in US Patients With Cirrhosis clinical studies. Lancet Gastroenterol Hepatol 2024;9:859–83.
and Nonvalvular Atrial Fibrillation. J Am Heart Assoc 45 Connolly SJ, Sharma M, Cohen AT, et al. Andexanet for Factor
2023;12:e026863. Xa Inhibitor-­Associated Acute Intracerebral Hemorrhage. N
29 Simon TG, Singer DE, Zhang Y, et al. Comparative Engl J Med 2024;390:1745–55.
Effectiveness and Safety of Apixaban, Rivaroxaban, and 46 Dibiasi C, Wiegele M, Gratz J. Orthotopic liver transplantation
Warfarin in Patients With Cirrhosis and Atrial Fibrillation : A in a patient receiving apixaban for portal vein thrombosis. J
Nationwide Cohort Study. Ann Intern Med 2024;177:1028– Clin Anesth 2019;56:41–2.
38. 47 Vuilleumier PH, Nagler M, Beldi G, et al. Orthotopic Liver
30 Douros A, Cui Y, Platt RW, et al. Utilization and long-­term Transplant in a Patient Anticoagulated With Rivaroxaban: A
persistence of direct oral anticoagulants among patients with Case Report. A A Pract 2019;13:54–7.
nonvalvular atrial fibrillation and liver disease. Br J Clin 48 Williams C, Stewart E, Conzen KD, et al. Dabigatran Reversal
Pharmacol 2022;88:994–1009. With Idarucizumab in 2 Patients With Portal Vein Thrombosis
31 Lund J, Saunders CL, Edwards D, et al. Anticoagulation trends Undergoing Orthotopic Liver Transplantation. Semin
in adults aged 65 years and over with atrial fibrillation: a Cardiothorac Vasc Anesth 2021;25:200–7.
cohort study. Open Heart 2021;8:e001737. 49 Pollack CV Jr, Reilly PA, van Ryn J, et al. Idarucizumab for
32 Lawal OD, Aronow HD, Hume AL, et al. Venous Dabigatran Reversal - Full Cohort Analysis. N Engl J Med
thromboembolism, chronic liver disease and anticoagulant 2017;377:431–41.

10 Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994

 Hepatology
50 Lee S-­R, Lee H-­J, Choi E-­K, et al. Direct Oral Anticoagulants 55 Lip GYH, Nieuwlaat R, Pisters R, et al. Refining clinical risk

Frontline Gastroenterol: first published as 10.1136/flgastro-2024-102994 on 24 May 2025. Downloaded from http://fg.bmj.com/ on June 18, 2025 at AIIMS Jodhpur.
in Patients With Atrial Fibrillation and Liver Disease. J Am Coll stratification for predicting stroke and thromboembolism in
Cardiol 2019;73:3295–308. atrial fibrillation using a novel risk factor-­based approach: the
51 Menichelli D, Ronca V, Di Rocco A, et al. Direct oral euro heart survey on atrial fibrillation. Chest 2010;137:263–
anticoagulants and advanced liver disease: A systematic review 72.
and meta‐analysis. Eur J Clin Investigation 2021;51:e13397. 56 Friberg L, Rosenqvist M, Lip GYH. Evaluation of risk
52 Lawal OD, Aronow HD, Shobayo F, et al. Comparative stratification schemes for ischaemic stroke and bleeding in
Effectiveness and Safety of Direct Oral Anticoagulants and
182 678 patients with atrial fibrillation: the Swedish Atrial
Warfarin in Patients With Atrial Fibrillation and Chronic
Fibrillation cohort study. Eur Heart J 2012;33:1500–10.
Liver Disease: A Nationwide Cohort Study. Circulation
57 O’Shea RS, Davitkov P, Ko CW, et al. AGA Clinical Practice
2023;147:782–94.
53 Baylo A, Cherniavskyi V, Reshotko D. Assessment of the Guideline on the Management of Coagulation Disorders in

Protected by copyright, including for uses related to text and data mining, AI training, and similar technologies.
efficiency and safety of anti-­coagulation therapy in patients Patients With Cirrhosis. Gastroenterology 2021;161:1615–
with liver cirrhosis and atrial fibrillation. Clin Exp Hepatol 27.
2023;9:265–71. 58 Steffel J, Collins R, Antz M, et al. 2021 European Heart
54 Kirchhof P, Benussi S, Kotecha D, et al. 2016 ESC Guidelines Rhythm Association Practical Guide on the Use of Non-­
for the management of atrial fibrillation developed in Vitamin K Antagonist Oral Anticoagulants in Patients with
collaboration with EACTS. Eur Heart J 2016;37:2893–962. Atrial Fibrillation. Europace 2021;23:1612–76.

Roberts L. Frontline Gastroenterology 2025;0:1–11. doi:10.1136/flgastro-2024-102994 11