Discussions
In healthy individuals, a balance between procoagulant and
anti-haemostatic factors maintains proper haemostasis, thus avoiding
abnormal bleeding or coagulation. Although this process involves in many
types of cells and organs, the liver produces most of the proteins that
regulate hemostasis [7]. The liver can
synthesize coagulation factors, coagulation inhibitors, and fibrinolytic
proteins and is therefore essential for
hemostasis[6]. Hepatic dysfunction
inhibits the synthesis of most coagulation factors. Furthermore, the
malnourished patients with liver disease, especially the active drinkers
have low levels of vitamin K, which reduces the produced amount of
procoagulation factors II, VII, IX, and X and anticoagulation factors,
protein C and protein S [7]. In
addition, thrombocytopenia (platelet counts <150,000/µL) is
one of the mostly reported complications in patients with chronic liver
disease. According to reports, up to 76% of cirrhotic patients have
this complication [8] while
fibrinolysis occurs in as many as 30% of cirrhotic patients
[9]. Although the patient had a normal
level of platelets, he had a history of epistaxis and was treated with
nadroparin calcium (0.6 ml, twice a day). His routine urine test
revealed hematuria (Red 78.8/μl). The clinical manifestation showed that
this patient had an increased risk of bleeding. Besides, deep venous
thrombosis (DVT) or pulmonary embolism (PE) has been reported to have an
incidence of 0.5%-6.3% [6]. In
addition, bacterial infections and chronic inflammation can affect the
synthesis of these endogenous coagulation factors and change blood flow.
The patient’s blood pressure was stable and his hemodynamics was stable,
too. However, this patient had a high infection index, which indicated a
high coagulating state. The
computer tomography pulmonary angiography (CTPA) showed he was at high
risk of multiple small branches embolism of the left pulmonary artery.
Although the patients with liver disease have a higher risk of bleeding,
they may benefit from preventive or therapeutic anticoagulation.
However, due to the lack of studies, it is challenging to make clinical
decisions on the optimal dose, duration, monitoring, or selection of
anticoagulant, and more importantly, to understand the clear clinical
benefit and safety [6].
The effect of LMWH on thrombin generation (TG) in the patients with
different phases of cirrhosis was evaluated in vitro. Mean antithrombin
(AT) levels and the endogenous thrombin potential (ETP) ratio were
evaluated in cirrhotic patients and healthy controls. Compared to
control groups, the mean AT levels and ETP ratio in all cirrhotic groups
were dramatically decreased (the 0.35 ETP ratios cirrhotic and control
groups were 0.26 ± 0.1 and 0.48 ± 0.1, respectively, P <
0.001). In addition, the decreasing trends of AT levels and ETP ratio
were linearly correlated with the severity of liver disease. Complete TG
inhibition by LMWH at 0.7U anti-Xa mL concentration was observed in 9
out of 30 patients with advanced cirrhosis (Child Pugh B and C), whereas
complete TG inhibition was only observed in 1/10 controls
[10]. This study suggests that the
patients with cirrhosis have increased response to LMWH anticoagulants.
In addition, as the liver disease gets more severe, the response to LMWH
anticoagulants is increased. Depending on the severity of liver disease,
the dose of LMWH might need to be adjusted to prevent excessive
anticoagulation and possible bleeding complications.
The Child-Turcotte-Pugh (CTP) grading system is widely used to evaluate
liver function in patients with chronic liver diseases
[11,
12]. According to CTP system, this
patient had a CTP rating of CTP A with a score of 5. Low molecular
weight heparin sodium at the dose of 0.6 ml (Fraxiparine) was injected
subcutaneously every 12 h. On the second day of admission, he presented
epistaxis and hematuria (Red 78.8/μL). Unfortunately, there was no study
on dose adjustment in patients with different grades of CTP. According
to clinical experience, we changed the dose
of LMWH to 0.4 ml with the
subcutaneous administration every 12 h. After the dose adjustment, the
patient did not have obvious epistaxis and hematuria any more. Thus, for
this patient, an adjusted dose of 0.4 ml LMWH was more suitable.
Oral anticoagulants should be started as soon as possible and preferably
on the same day as the parenteral anticoagulant. Now, NOACs are commonly
used for the prevention and treatment of multiple indications, including
venous thromboembolism (VTE) and stroke prophylaxis in atrial
fibrillation. In addition, the American College of Clinical Pharmacy
guidelines have suggested using NOACs as first-line treatment for PE
[13]. NOACs were compared to VKAs in
a variety of conditions, and the results indicated that NOACs had equal
or even better treatment efficacy for VTE than VKAs. In addition,
according to the pooled data, NOACs caused fewer bleeding complications
than VKAs. However, to date, the patients with liver disease have been
excluded in all the reported prospective studies on NOACs. The
incidences of thrombosis and bleeding events in patients with cirrhosis
were higher than those in healthy individuals. The risks and benefits of
using anticoagulation in patients with cirrhosis need to be carefully
evaluated.
A meta-analysis with 152,116 patients in 29 RCT studies (with an average
follow-up period of 16 months) evaluated drug-induced liver injury
(DILI) by NOACs and conventional anticoagulants. The result showed that
NOAC did not cause a higher risk of DILI (RR 0.90, 95% CI 0.72 to 1.13,
I2=0%) [12].
In a retrospective cohort study, therapeutic anticoagulation was
administered to cirrhotic patients for more than 3 years to treat
thrombosis or prevent stroke in patients with atrial fibrillation.
During the study period, 27 cirrhotic patients (11 patients with CTPA,
12 patients with CTPB and 4 patients with CTPC) were treated with a NOAC
(apixaban 5mg BID with or without a 10mg BID loading dose; rivaroxaban
15 mg daily with or without a 20 mg daily loading dose) and 18 patients
(7 patients with CTPA, 9 patients with CTPB, and 2 patients with CTPC)
were treated with VKA (warfarin dosed to an INR target of 2-3 or an INR
1 unit above baseline) or low molecular weight heparin (LMWH, 1 mg/kg
BID or 1.5 mg/kg daily). Both patient groups with the treatment of NOAC
and conventional anticoagulants had similar results in the total
bleeding events (8 NOAC vs. 10 other, p=0.12). However, in the patient
group with the treatment of NOAC, the major bleeding episodes were
significantly fewer, (1(4%) vs. 5 (28%), p = 0.03)
[14], which might be due to the low
dose of NOAC. A regular dose of NOAC may have similar safety as
traditional anticoagulants in the treatment of cirrhosis. In another
3-year retrospective study, 39 patients with cirrhosis (CTP A and CTP B)
were treated with anticoagulation therapy (20 patients treated by NOAC,
apixaban 5 mg twice daily or rivaroxaban 20 mg daily, and 19 patients
treated by traditional anticoagulation, LMWH 1 mg/kg twice daily or
warfarin variable with INR). The treatment results of both patient
groups were evaluated and compared. Compared to traditional
anticoagulation, a regular dose of NOAC had similar safety
characteristics in patients with cirrhosis
[15].
Hepatic dysfunction affects not only the metabolism of oral
anticoagulant, but also the coagulation function in moderate and severe
patients. A study by Dagmar Kubitza reported that moderate (but not
mild) liver insufficiency caused the reduction in systemic clearance of
rivaroxaban after a single 10 mg dose, resulting in the higher exposure
and improved pharmacodynamic effects of rivaroxaban
[16]. Another study by Jochen Graff
reported that in subjects with moderate liver insufficiency (i.e.
Child-Pugh classification B), the area under the plasma
concentration-time curve (AUC) increased by 2.27 fold for rivaroxaban
(10 mg single dose), 1.09 fold for apixaban (5 mg single dose), 4.8%
for edoxaban (15 mg single dose), and 5.6% for dabigatran (150 mg
single dose) [17]. Specific marker
limitations for impaired liver function for rivaroxaban, apixaban and
dabigatran are according to both the Child-Pugh classification and
liver-related exclusion criteria used in pivotal clinical trials.
Due to the therapeutic dose of LMWH, this patient was at a higher risk
of bleeding. There is a significant individual variation in the efficacy
of warfarin; therefore, the dose should be individualized according to
blood drug concentration
monitoring. Perhaps apixaban is
the best choice for this patient, but it is not yet available in China.
Following the instructions, the patient was given rivaroxaban 15 mg, bid
20 mg (three weeks prior), and qd (three weeks after). During the
follow-up, he did not show symptoms such as bleeding from gums and
gastrointestinal tract. After three weeks of the treatment with
rivaroxaban, the patient’s blood routine tests showed normal, without
any side effects.