Download Coagulation Disorders and Portal Vein Thrombosis in Cirrhosis

Coagulation Disorders and Portal Vein
Thrombosis in Cirrhosis
Stephen H. Caldwell, MD, Zachary Henry, MD, and Nicolas Intagliata, MD
University of Virginia
Charlottesville, VA
Key Concepts
• Despite prolongation of the prothrombin time, patients with
cirrhosis often have decreased protein C and other changes
leading to a rebalanced or even hypercoagulable state.
• Conventional tests, like INR, are not adequate measures of
hemostasis in cirrhotic patients and other tests like
Endogenous Thrombin Production (ETP) and
Thromboelastography (TEG) need greater standardization and
further translational study to better assess their clinical utility.
• Cirrhosis patients can develop hyperfibrinolysis and related
bleeding complications. Diagnostic tests for this condition are
not easily or widely available. A high index of suspicion is
needed as anti-fibrinolytic agents such as Aminocaproic Acid
may be beneficial.
• Portal vein thrombosis has an estimated annual incidence of
5-15% in patients with cirrhosis and has been associated with
increased risk of decompensation, worse overall survival, and
worse post-transplant outcomes.
• Anti-coagulation with vitamin K antagonists or low molecular
weight heparin has been shown to be safe and should be
considered in patients with portal vein thrombosis, whether
acute or chronic, due to the favorable impact on transplant
outcomes and overall survival. TIPS is also an option in
selected cases and has the advantage of decreasing the risk
of variceal bleeding.
• Prophylactic anti-coagulation with heparin products is safe
and should be considered in hospitalized cirrhotic patients for
the prevention of venous thromboembolism. It has also shown
benefit in non-hospitalized compensated cirrhotic patients for
prevention of portal vein thrombosis. In the latter setting, the
benefit appears to extend beyond simply preventing portal vein
thrombosis although further studies are needed to confirm
these quite remarkable findings.
Recent studies have resulted in a paradigm shift in how
cirrhotic coagulopathy is regarded. The concept of ‘autoanticoagulation’ in cirrhosis is now discredited and the
complexity of the bleeding and clotting diathesis associated
with liver disease has become increasingly apparent and
clinically relevant in surprising, paradoxical, and often
perplexing ways.
Laboratory Assessment of the Clotting Cascade
in Cirrhosis
As measured by more physiological laboratory
techniques, it is now evident that stable cirrhosis is
hematologically characterized by relatively preserved
hemostatic mechanisms. CIrrhosis may even be associated
with a hypercoagulable state despite prolongation of the
prothrombin time (PT) and INR (international normalized
ratio) and/or thrombocytopenia.1 Key laboratory methods
include the thrombin generation assay (endogenous
American Association for the Study of Liver Diseases
159
New Treatments in Liver Disease
thrombin production or ETP). This test measures the
capacity of blood to produce thrombin or activated factor II
(which then converts fibrinogen to fibrin clot). Global
clotting assays, like thromboelastography (TEG, Rotem,
Sonorheometry), measure the physical properties of actual
clot formation and subsequent dissolution by physiological
fibrinolysis. Prolongation of the PT/INR indicates
decreased liver synthesis of pro-coagulant factors, but these
changes are offset by impaired hepatic synthesis of anticoagulant factors such as protein C, S and antithrombin
(AT) and other changes in relevant factors synthesized by
the vascular endothelial cells.
Protein C and Thrombomodulin
Impaired hepatic synthesis of protein C combined with
normal levels of endothelial-derived thrombomodulin (a
protein C co-factor – see below) and increased levels of
endothelial-derived factor VIII and von Willebrand factor,
lead to enhanced thrombin (activated factor II) production
and enhanced aspects of platelet function such as
adhesion.1,2,3 Thrombomodulin (TM) has played a key role
in recent advances in laboratory testing. TM is a
transmembrane glycoprotein which exists mainly in the
endothelium (less so in a soluble form) which promotes
protein C activation. Its use in thrombin generation assays
permits more physiological study of the pro- and anticoagulant systems in cirrhosis. Normally, activated protein
C (APC) together with TM and additional co-factors
including protein S and negatively charged phospholipids,
inactivates factors V and VIII thereby limiting thrombin
production (thus promoting anticoagulation).
Blood sample testing with the addition of a
physiological amount of TM reveals that when a normal
amount of protein C is present, thrombin production is
limited by the presence of APC. On the other hand, in
cirrhosis, low levels of hepatic-derived protein C and high
levels of endothelial-derived factor VIII augment thrombin
production, which can then convert fibrinogen to fibrin
clot. Notably, past series have shown fibrinogen levels are
often preserved or only mildly decreased in cirrhosis.4
These compensatory changes thus offset diminished
hepatic synthesis of pro-coagulant factors, preserve the
hemostatic pathway of clot formation, and can result in a
net hypercoagulable state.5
As a result, hemostatic mechanisms in stable cirrhosis
are best seen as essentially rebalanced albeit in a fragile state
that can be tilted in either direction depending on complex
exogenous factors.6 Bleeding problems are dominant in
day-to-day practice, especially among hospitalized patients.
However, thrombotic processes can dominate and may
ultimately propel the compensated cirrhotic liver into a
state of atrophy and failure.7,8 Bleeding is favored by a
number of common superimposed conditions including
infection (via release of endogenous heparinoids from
endothelial cells) and renal failure (by dysfunctional uremic
platelets).9,10 In addition, a state of hyperfibrinolysis may
occur (see below) characterized by premature clot
dissolution. Pressure driven systems, such as portal
collaterals, may also complicate the disorder and warrant
careful consideration of the ill-effects of excessive volume
expansion on portal pressure.11 Thrombotic issues may
dominate as a result of changes in the over-all balance of
pro- and anti- coagulant pathways and especially by stasis
of blood flow in which latent hypercoagulability may
dominate.12 In addition, the production of microparticles
(phospholipid vesicles) by the sick liver may cause a
dynamic and very little explored stimulant of
coagulation.13,14
The Precarious Balance: Accelerated
Intravascular Coagulation and Fibrinolysis
(AICF)?
Although the term ‘DIC’, is sometimes used loosely to
describe the coagulopathy of liver disease especially in
decompensated patients, the disorder often lacks features of
classical DIC such as rapidly declining platelets and
fibrinogen and decreased FVIII. The term ‘AICF’ has been
proposed as a more accurate descriptor of cirrhotic
coagulopathy characterized by low but usually stable
platelet levels, mildly decreased or normal fibrinogen,
decreased pro- and anti-coagulant factors, increased Factor
VIII and von Willebrand factor, increased clot degradation
products and increased fibrinolytic (clot dissolution)
capacity.15,16
Hyperfibrinolysis
Increased fibrinolytic capacity is one of the earliest
observations regarding coagulopathy in liver disease as
evident from Dr EW Goodpasture’s 1914 study describing
the increased fibrinolytic capacity of blood in liver disease.
Sixty years later, Boks et al demonstrated the difficulty in
predicting bleeding in both acute and decompensated
chronic liver disease based on conventional measures of
hemostasis among hospitalized patients.4 In that study,
mucosal-type (non-variceal) bleeding could only be
predicted by factoring in a measure of fibrinolysis or clot
dissolution. The clinical syndrome of hyperfibrinolysis is
perhaps most evident in the decompensated cirrhotic where
it is characterized by mucosal and puncture wound oozing
American Association for the Study of Liver Diseases
160
New Treatments in Liver Disease
and, sometimes delayed, post-procedure bleeding. It is
estimated to be clinically apparent in around 5-10% of
decompensated cirrhotics.17 The mechanism is uncertain
but abnormal metabolism of TAFI (Thrombin Activatable
Fibrinolysis Inhibitor) has been implicated18 as has the
presence of ascites. Expressed as ‘Global Fibrinolytic
Capacity’, a step-wise increase is evident in progressively
more severe liver disease.19 The clinical importance of
recognizing the condition lies in the potential role of antifibrinolytics, such as Amicar (aminocaproic acid), to slow
clot dissolution. It is perhaps worth mentioning also that
most body cavities appear to have fibrinolytic capacity
including the oral cavity, bladder and peritoneal cavity.
Thromboelastography (TEG – see below) is probably the
most convincing laboratory test to confirm the diagnosis of
hyperfibrinolysis but its limited availability and variation in
performance necessitate the continued reliance on clinical
suspicion.
Laboratory Studies
As discussed above, advances in clinical laboratory
testing have led the way to a more practical understanding
of the hemostatic mechanisms in clinical liver disease.12 As
a result, there has been increasing recognition of the
conventional INR in cirrhosis as a narrowly focused assay
with significant inter-laboratory variation (depending, as
shown initially by J Trotter, on commercially available
thromboplastins). However, many of the recent studies,
especially those involving the ETP test (endogenous
thrombin production) have been performed in stable outpatients without strong clinical/translational studies in
decompensated in-patients. Thromboelastography (TEG)
has also re-emerged as a leading candidate for a global
measure in cirrhosis. Indeed, there is extensive published
experience and precedent in defining hypercoagulability
based on measured parameters of the TEG such as rate of
clot formation and maximal clot strength.20 Despite this
situation, it is also apparent that there is significant intercenter variation in nuances of the test and different types of
elastography appear to vary in their performance and ability
to detect hyperfibrinolysis (unpublished experience).
Clearly, clinically relevant translational studies are needed
to refine this approach and understand the strengths and
limitations of various approaches. At our center, we have
come to rely more on platelet levels and fibrinogen levels as
stop gap measures of hemostasis and much less on the
potentially misleading INR while we anxiously await the
results of ongoing studies.12
DVT in Cirrhosis
Based on the foregoing discussion, it is not surprising
that hospitalized cirrhosis patients are at risk for peripheral
deep venous thrombosis with an estimated incidence of
around 1-2%.21,22 Conventional DVT prophylaxis is felt to
be safe in such patients in the absence of bleeding although
efficacy measures remain uncertain due to changes in for
example heparin co-factors (liver-derived AT) which have
not been fully investigated and may cause resistance to
heparin activity.23
Portal Vein Thrombosis (PVT) in Cirrhosis—
Clinical Significance
Among cirrhotic patients without HCC (which
increases the risk), PVT is estimated to develop in 5% to
15% of patients annually with a prevalence of 25% or
more.24 Assessment of the existing literature is complicated
by existence of only a few prospective studies. In addition,
the clinical spectrum of the disorder is broad and ranges
from partial to occlusive PVT, overtly symptomatic to
subclinical, localized or extensive, acute versus chronic with
cavernous transformation (cavernoma) and concurrent
presence or absence of variable risk factors like Factor V
Leiden and Prothrombin mutation (which are infrequent
but potentially significant prothrombotic (thrombophilic)
factors in these patients and can occur in about 10-15% of
cases).25 However, slow portal vein blood flow (<15 cm/s)
appears to be the most significant risk for PVT.26 The
situation warrants future consideration of beta blocker
effects which are known to decrease portal flow. It is also
interesting to speculate on the potential role of liver derived
pro-coagulant microparticles (see above). The association
of PVT in cirrhosis with smaller liver size, ascites and
hydrothorax, recurrent variceal bleeding, increased ChildPugh/MELD scores and especially the effect on survival
convincingly argue for its clinical significance.27,28
Moreover, PVT has a significant impact on the technical
aspects of transplant surgery and a complicated but
detrimental effect on outcomes: PVT pre-transplant shifts
the survival benefit of transplant such that the net benefit of
transplant on survival is evident only at higher pretransplant MELD.29
Clinical Evaluation and Therapy in Cirrhotic PVT
The discovery of PVT in a cirrhosis patient, usually
detected first by screening ultrasound, should be first
confirmed by cross-sectional venous phase imaging with
CT or MRI. When confirmed, a hypercoagulable workup
including testing for FV Leiden, prothrombin 20210, and
American Association for the Study of Liver Diseases
161
New Treatments in Liver Disease
Table: Key considerations in newly recognized Portal Vein Thrombosis in cirrhosis
1. Recall that slow portal flow (<15 cm/s) is one of the major risks for PVT
2. Exclude non-cirrhotic PVT (not considered here)
3. Confirm PVT and assess extent with cross sectional imaging
- Partial PVT or occlusive
- Are intra-hepatic portal branches evident?
- Is there extension into the mesenteric system?
- Are large porto-systemic collaterals present which promote slow portal flow?
4. Are symptoms evident such as ascites, variceal bleeding, abdominal pain or overall deterioration (although often silent, about 1/2 or more of patients experience some
symptoms)
5. Assess genetic thrombophilic factors which may affect the duration of subsequent therapy: Factor V Leiden, Prothrombin mutation, MTHFR (individual factor levels can also be
measured but are difficult to interpret in cirrhosis, anti-cardiolipins may also be measured but their relevance in this setting is uncertain)
6. Assess for varices and undertake banding if high risk unless TIPS is planned
7. Consider TIPS (and thrombectomy) in selected cases if portal vein branches are evident and in the absence of cavernous transformation
6. Anti-coagulation therapy: risk-benefit remains to be established but is probably most-favorable at this time in transplant candidates. However, there is emerging data on the
overall favorable effects on hepatic function of prophylactic anticoagulation in stable cirrhosis so increasing potential role in PVT
- Coumadin: studied in a number of series, target INR generally 2-3, obvious challenges in monitoring and dosing adjustments, generally available and inexpensive
- Low molecular weight heparin (LMWR): optimal dosing and possible resistance (due to low anti-thrombin) have not been well studied. Appears effective in
prophylactic dosing (see paper by Villa et al). Costs may be a problem and inconvenience of injection is a drawback
- Newer agents (oral Xa inhibitors) are expected to undergo study
MTHFR (methylenetetrahydrofolate reductase) mutations
in order to assess fixed thrombotic risks may help to guide
long-term therapy with anti-coagulants.30 The utility of
individual factor levels (Protein C, S and AT) in this setting
is less clear as their interpretation in cirrhosis is challenging
although this clearly requires further investigation. Early
endoscopy should be undertaken with band ligation if
significant varices are evident (unless perhaps if TIPS portal
decompression is planned).
With the increasing awareness of PVT effects on
morbidity, survival and transplantation, efforts to reverse
the process with systemic anti-coagulation are increasingly
accepted. Anti-coagulation with coumadin or low
molecular weight heparin results in recanalization in 4060% of patients (compared to very infrequent spontaneous
recanalization) and appears to have a favorable impact on
transplant outcomes.31,32 Close clinical follow-up in these
patients is challenging but essential. The potential role of
newer anti-coagulants such as factor Xa inhibitors warrants
consideration, but also requires caution in listed patients
due to issues of reversibility. Although there are only a few
studies, TIPS (usually along with post-procedure anticoagulation) is a viable option in these patients even with
occlusive PVT provided intrahepatic portal branches can be
visualized.33 This approach has the advantage of offering
varices decompression but carries technical challenges in
some patients and increased risk of hepatic decompensation
with higher MELD scores.
Should We Anti-coagulate Stable Cirrhosis
Without PVT?
Based on the problems associated with the occurrence
of PVT and the pro-coagulant changes of cirrhosis,
prospective, prophylactic anti-coagulant studies have been
undertaken aimed primarily at preventing PVT with
surprising results. In a study of enoxaparin in patients with
PVT presenting with esophageal variceal bleeding (anticoagulation started after varices eradication with banding),
Amitrano et al noted no benefit in terms of variceal rebleeding but a surprising improvement in survival in the
treated group.34 These results have been remarkably
advanced in a subsequent provocative, prospective,
controlled, and partially blinded trial by Villa et al.35 70
cirrhosis patients without PVT were randomized to
prophylactic dosing of enoxaparin (34) versus no treatment
(36) for 48 weeks with an additional year of follow-up
consisting of q3mos US and q6mos CT. The divergence of
the curves for not only PVT but also for episodes of
decompensation and actuarial survival is impressive. The
differences were attributed in part to decreased gut
microbial translocation and improved gut perfusion.
However, a possible intrahepatic mechanism is suggested by
other work. This includes the proposed role of intrahepatic
microvascular thrombosis in parenchymal extinction put
forward by Wanless et al as a mechanism of liver atrophy in
cirrhosis, work by Anstee and Thursz et al showing
thrombin to activate fibrosis pathways in stellate cells via
the PAR-1 receptor (PAR: protease activated receptor) and
unpublished work by Y Ikura et al showing evidence of
intra-hepatic activated platelets in advanced cirrhosis.
American Association for the Study of Liver Diseases
162
New Treatments in Liver Disease
These complementary pathways need to be revisited in
future studies of anti-coagulation therapy in cirrhosis.
6.
7.
Conclusion
The ‘coagulopathy of cirrhosis’ is extraordinarily
dynamic and complex. Our understanding of this condition
has dramatically evolved over the last decade. Studies now
show that clinicians must look beyond the notion that
conventional laboratory tests accurately convey bleeding or
clotting risk in these patients. The liver produces nearly all
of the proteins needed to form clots at sites of damaged
vessel walls (coagulation/hemostasis), but also produces the
majority of proteins that govern this cascade, by providing
inhibitory and fibrinolytic control (anticoagulation). When
this balance is disturbed, pathology in the form of bleeding
or thrombosis is clinically observed. In cirrhosis, the
imbalance is mitigated by compensatory mechanisms
although consequences of portal hypertension lead to
disruption of this tenuous balance. As our understanding of
this process evolves, developing laboratory tests that allow
the clinician to better predict the patient’s relative risk to
bleed or clot is an essential goal to guide therapy. Another
very interesting area of research in this field includes
examining the role of microparticles in coagulation and
fibrinolysis in cirrhosis patients. New oral anticoagulation
medications remain to be explored in cirrhosis patients but
hold promise as a means of preventing deterioration of the
injured organ. As our understanding of the complex
hemostatic state of cirrhosis continues to advance, our
ability to prevent and predict clinical sequelae of bleeding
and clotting will undoubtedly improve.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
References
1.
2.
3.
4.
5.
Tripodi A, Mannucci PM. The coagulopathy of chronic
liver disease. N Engl J Med. 2011 Jul 14;365(2):147-56.
Lisman T et al. Elevated levels of von Willebrand Factor
in cirrhosis support platelet adhesion despite reduced
functional capacity. Hepatology 2006;44:53-61.
Biguzzi E et al. Endothelial protein C receptor plasma
levels increase in chronic liver disease, while
thrombomodulin plasma levels increase only in
hepatocellular carcinoma. Thromb Res. 2007;120:28993.
Boks AL et al., Hemostasis and fibrinolysis in severe liver
failure and their relation to hemorrhage. Hepatol,
1986;6:79-86.
Tripodi A et al. Evidence that low protein C contributes
to the procoagulant imbalance in cirrhosis. J Hepatol.
2013 Apr 11. pii: S0168-8278.
18.
19.
20.
21.
Lisman T, Porte RJ. Rebalanced hemostasis in patients
with liver disease: evidence and clinical consequences.
Blood. 2010;116:878-85.
Shah NL et al. A clinical survey of bleeding, thrombosis,
and blood product use in decompensated cirrhosis
patients. Ann Hepatol. 2012;11:686-90.
Anstee QM et al. Parenchymal extinction: coagulation
and hepatic fibrogenesis.Clin Liver Dis. 2009;13:117-26.
Montalto P et al. Bacterial infection in cirrhosis impairs
coagulation by a heparin effect: a prospective study. J
Hepatol. 2002;37:463-70.
Caldwell SH et al. Coagulation disorders and hemostasis
in liver disease: pathophysiology and critical assessment
of current management. Hepatology. 2006;44:1039-46.
Massicotte L et al. Effects of phlebotomy and
phenylephrine infusion on portal venous pressure and
systemic hemodynamics during liver transplantation.
Transplantation. 2010;89:920-7.
Northup P, Caldwell S. Coagulation in Liver Disease: A
Guide for the Clinician. Clin Gastroenterol Hepatol.
2013 Mar 16. doi:pii: S1542-3565(13)00317-0.
10.1016/j.cgh.2013.02.026. [Epub ahead of print]
Kornek M et al. Circulating microparticles as diseasespecific biomarkers of severity of inflammation in
patients with hepatitis C or nonalcoholic steatohepatitis.
Gastro 2012;143:448-58.
Lacroix R et al. Leukocyte- and endothelial-derived
microparticles: a circulating source for fibrinolysis.
Haematologica. 2012;97:1864-72.
Joist JH. AICF and DIC in liver cirrhosis: expressions of
a hypercoagulable state. Am J Gastroenterol.
1999;94:2801-3.
Ben-Ari Z et al. Disseminated intravascular coagulation
in liver cirrhosis: fact or fiction? Am J Gastroenterol.
1999;94:2977-82.
Hu KQ et al Hyperfibrinolytic activity in hospitalized
cirrhotic patients in a referral liver unit. Am J
Gastroenterol. 2001;96:1581-6.
Colucci M et al. Deficiency of thrombin activatable
fibrinolysis inhibitor in cirrhosis is associated with
increased plasma fibrinolysis. Hepatology. 2003;38:2307.
Rijken DC et al. Evidence for an enhanced fibrinolytic
capacity in cirrhosis as measured with two different
global fibrinolysis tests. J Thromb Haemost.
2012;10:2116-22.
Ben-Ari Z et al Hypercoagulability in patients with
primary biliary cirrhosis and primary sclerosing
cholangitis evaluated by thrombelastography. J Hepatol.
1997;26:554-9.
Northup PG et al. Coagulopathy does not fully protect
hospitalized cirrhosis patients from peripheral venous
thromboembolism. Am J Gastroenterol. 2006;101:15248.
American Association for the Study of Liver Diseases
163
New Treatments in Liver Disease
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Gulley D et al. Deep vein thrombosis and pulmonary
embolism in cirrhosis patients. Dig Dis Sci.
2008;53:3012-7.
Intagliata NM et al. Prophylactic anticoagulation for
venous thromboembolism in hospitalized cirrhosis
patients is not associated with high rates of
gastrointestinal bleeding. Liver Int. 2013 May 13. doi:
10.1111/liv.12211.
Tsochatzis EA et al. Systematic review: portal vein
thrombosis in cirrhosis. Aliment Pharm Ther.
2010;31:366-74.
Garcia-Pagan JC, Valla DC. Portal vein thrombosis: a
predictable milestone in cirrhosis? J Hepatol.
2009;51:632-4.
Zocco MA et al. Thrombotic risk factors in patients with
liver cirrhosis: correlation with MELD scoring system
and portal vein thrombosis development. J Hepatol.
2009;51:682-9.
Amitrano L et al. Risk factors and clinical presentation
of portal vein thrombosis in patients with liver cirrhosis.
J Hepatol. 2004;40:736-41.
Englesbe MJ et al. Portal vein thrombosis and survival in
patients with cirrhosis. Liver Transpl. 2010;16:83-90.
Englesbe MJ et al. Portal vein thrombosis and liver
transplant survival benefit. Liver Transpl. 2010;16:9991005.
Deleve LD et al. Vascular disorders of the Liver. Hepatol
2009;49:1729-64.
Francoz C et al. Splanchnic vein thrombosis in
candidates for liver transplantation: usefulness of
screening and anticoagulation. Gut. 2005;54:691-7.
Delgado MG et al Efficacy and safety of anticoagulation
on patients with cirrhosis and portal vein thrombosis.
Clin Gastroenterol Hepatol. 2012;10:776-83.
Senzolo M et al. Prospective evaluation of
anticoagulation and transjugular intrahepatic
portosystemic shunt for the management of portal vein
thrombosis in cirrhosis. Liver International 2012;32:91927.
Amitrano L et al. Splanchnic vein thrombosis and
variceal rebleeding in patients with cirrhosis. Eur J
Gastroenterol Hepatol. 2012;24:1381-5.
Villa E et al. Enoxaparin prevents portal vein thrombosis
and liver decompensation in patients with advanced
cirrhosis. Gastroenterology. 2012;143:1253-60.
American Association for the Study of Liver Diseases
164