Anticoagulation bridging: the substitution of sintrom by heparin in the periprocedural period: the Brug study

There are 400.000 persons in the Netherlands using vitamin K antagonist (VKA)
daily to lower their risk at thrombus formation, both in venous and arterial
thrombosis. Indications for VKA therapy are:
- Primary and secondary prevention of venous thrombosis
- Prevention of thrombus growth
- Prevention of thrombus formation in case of mechanical heart valves
- Prevention of thrombus formation as a result of atrial fibrillation
- Complementing the prevention of thromboemboli for patients with arterial
pathologies
VKA limits the generation of fully active vitamin K dependent coagulation
clotting factors, viz. coagulation factor II, VII, IX and X. This inhibitory
effect offers protection against thrombus formation en thrombus growth, but is
also responsible for the most important side effect of VKA, bleeding diathesis.

Approximately 10 % of the total population of VKA users has to undergo an
invasive intervention on a yearly basis. In some of these interventions the
patient is treated with bridging therapy. In case of perioperative
(anti)coagulation management bridging is defined as being a temporary switch of
approximately 7 to 11 days to short-acting anticoagulants such as low-molecular
weight heparin (LMWH) or unfractionated heparin (UFH) during the period in
which the VKA is discontinued. VKA is discontinued because of the bleeding risk
it is associated with during invasive interventions. The purpose of the
bridging therapy is to limit the period a patient is insufficiently
anticoagulated to a minimum. One of the properties of VKA is that they have a
long half-life time compared to heparin-like compounds. This results in a
residual anticoagulant effect for a prolonged period of time after
discontinuation of VKA. The day of the intervention the patient should have an
INR, which is the measure of VKA efficacy, equal to its normal value of 1 in
order to minimize the risk of bleeding as a result of the intervention. This
means that the patient needs to stop taking VKA 3 to 5 days before the
intervention. During these 3 to 5 days the patient is at elevated risk of
thrombosis due to insufficient anticoagulation. Therefore the patient is
treated with short-acting heparins, which can be stopped one day before the
intervention. In the Netherlands acenocoumarol as VKA is mainly prescribed.
Acenocoumarol has a half-life time of 8 to 14 hours.

There is widespread consensus in literature about VKA therapy not needed to be
discontinued for small (surgical) interventions, such as dental procedures or
small dermatological excision . The physician can decide to bridge the surgical
patient with LMWH according to the CBO guidelines, when the patient has an
increased bleeding risk around the intervention and a high risk of venous
and/or arterial thrombosis . Patients with a low or moderate risk of thrombosis
can discontinue VKA treatment without bridging to LMWH perioperatively.
different researchers report that patients treated according to this guideline
have a risk between 0.0 % and 2.0 % at thrombosis and 0.5 % - 20 % to acquire a
(major) bleeding.

There is not much scientific evidence available to justify the use of bridging
therapy. There is lack of large, randomized studies because these studies are
not feasible on ethical grounds. Currently there are only results available
from one-armed, poorly executed observational studies of case series to measure
the effect of the bridging in respect to bleedings and thromboses.
Consequently, the difficult consideration between risk of bleeding or
thrombosis can*t be supported by strong evidence. In 2008 the American College
of Chest Physicians (ACCP) published a review of the until then published
studies. Based upon this review they came to the nowadays generally accepted
guidelines. These guidelines are also adopted by the CBO. The question whether
a VKA patient has to be bridged perioperatively is answered by these guidelines
by making an individual consideration of the risk of bleeding and thrombosis
and subsequently use this to define the bridging strategy. Patient
stratification of the thrombosis risk is based upon the indication for VKA
therapy and comorbidity. This stratification is however not validated. The
stratification of the bleeding risk is determined by using multiple conditions
such as the planned intervention, age, INR value, medication use (for example
NSAIDs) and comorbidity (for example diabetes mellitus, lever function,
aneurysma). Unfortunately an algorithm is missing for a physician to calculate
the bleeding risk of the planned intervention for this particular patient,
inevitably resulting in misclassification. The lack of understanding the
relationship between the used bridging strategy and perioperative bleeding and
thrombosis forces us to search for alternatives, in order to develop an
instrument that enhances the safety of VKA users that are bridged by using
short-acting anticoagulants.

Philips researchers together with the Maastricht University have started a
project to gain more insight in the molecular and biological processes that are
taking place during bridging therapy, document biological variations associated
with these processes and develop tests that support the physicians to prescribe
a more individualized bridging therapy in order to lower the incidence of
bridging-related thrombosis and bleeding. The first step of this project is to
observe the bridging period of 30 patients that need to undergo a (small)
intervention in the MUMC+ for 9 days by daily blood measurements (via
venipuncture), starting 3 days before the intervention. In addition the
patients are followed 3 days before until 30 days after the intervention
documenting bleeding and/or thrombosis complications. The results of our
computer model will be compared to the patient outcomes using this data. In
addition the data will be used to investigate possible associations between
bleeding and thrombosis risk and biological variation.

The computer model of the human coagulation system currently contains the
coagulation cascade and the pharmacokinetic and pharmacodynamic (PK/PD) of VKA
and heparin-like compounds. The coagulation cascade contains the
protein-protein interactions taking place after a disturbance of the hemostatic
balance (for example after an injury of the vascular wall) and the formation of
a fibrin network which is an important part of the formed thrombus. The
computer model includes the intrinsic and extrinsic pathway of the coagulation
cascade. The computer model is able to calculate the effect that VKA and
heparin-like compounds have on the coagulation cascade. Besides this,it is
possible with the model to simulate diverse test conditions, for example the
INR/PT, aPTT and TGA.

The computer model is, as of yet, largely optimized (and validated) by means of
laboratory measurements and plasma from healthy individuals. Using the here
proposed study we would like to test the validity of the computer model in
clinical practice and to improve the computer model. In the end, we will use
the data from this study in combination with the improved model to design a
larger randomized trial more efficiently in order to compare the functioning of
our model compared to the common bridging guidelines in terms of postoperative
thromboses and perioperative bleedings.

We have set prior to the design and implementation of this research a primary
and a secondary goal.
Primary goal:
Testing the validity of the computer model of the coagulation cascade and the
related PK / PD model of VKA and heparin for use in clinical practice.
Secondary goal:
The computer model is now partially optimized with respect to the simulation of
the TGA, INR and aPTT. Further optimization with respect to the clinical
applicability of the model for bridging from VKA to low molecular weight
heparin and a local adaptation to the blood tests used in MUMC+ is required.

The determination, using this computer model, of which patients, treated
according to the standard bridging regimen, are at high or low risk of
developing bleeding complications or thrombosis is explicitly not an objective
of the present study. This study can be seen as a pilot study for a similar,
larger study.

The present study is a mono-center validation study, performed in patients on
chronic vitamin K antagonists who have to undergo minor surgery. This study was
initiated to validate the computer model of the coagulation cascade, developed
by Philips Research. Currently the model is constructed using information
obtained from blood that was provided by the blood bank. We can reach further
alignment of the PK / PD models, currently based on literature study, if we
would have blood of perioperatively bridged patients. In order to achieve our
goals we will daily gather blood samples from the participant during and after
the bridging period. Concentrations of various proteins related to the
coagulation cascade will be measured in the blood. Relative simple measurements
of blood proteins that are sensitive to changing concentrations of heparin and
VKA are performed, as well as measurements of key blood proteins that influence
the INR, aPTT/anti FXa and TGA and measurements of the INR, aPTT/anti FXa and
TGA themselves. In addition, measurements of the SNPs related to the VKORC1 and
CYP2C9 will be performed. VKORC1 is responsible for the production of enzymes
that reduce oxidized vitamin K, making it available again for the full
formation of clotting factors. CYP2C9 is an enzyme that contributes to the
metabolism of VKA. These measurements may be included as variables in our
model. Because we want to link patient outcomes to the outcomes of our model,
we will follow participants for 30 days and document severe / not severe
bleedings and thromboembolisms. In summary, we have access to the following
information of all participants:

a) Blood samples from patients every day during the entire bridging period of 3
days before to 5 days after surgery.
o The plasma measurements: FII, FV, FVII, FVIII, FIX, FX, FXI, FXII,
Fibrinogen, protC, prot, Protze, AT
o INR, aPTT, TGA, anti FXa
o Hb / Ht
o thrombocyte count
o Spare citraatplasma
b)The following measurements will be determined once using the blood samples of
the first day.
o SNP of VKORC1 and CYP2C9
o Standard clinical renal function
o Standard clinical liver function
c)Follow-up until 30 days after the intervention regarding
perioperative-related bleedings and thromboses via the electronic hospital
documentation system.
d)The following patient characteristics, through the electronic hospital
documentation system.
o Age
o Gender
o Length
o Weight
o Medication use: dosage, administration schedule and INR therapeutic target
range
o Medical indication of planned intervention and expected bleeding risk (low,
moderate, high)
o use of red blood cell (RBC) concentrate, thrombocyte concentrate and fresh
frozen plasma (FFP)

The interventions necessary for our research do not conflict in any way with
the treatment the patient has to undergo. The patient undergoes surgery at the
time scheduled by the treating physician and all blood samples are taken as in
patients who are not included in the research. The participants bridging regime
is made according to the current CBO guidelines. Patients undergo 2 or 3 blood
samples (venipunctures) in the context of bridging. The difference with the
normal procedure is that 9 (6 to 7 extra) blood samples (venipunctures) are
taken and per sample 23ml blood is taken instead of the usual 4.5 ml. The
amount of extra blood is approximately 210 ml. When an INR is to be determined
in the context of the treatment (2 to 3) the nurse will draw an additional
amount of 4.5 ml blood and deliver the blood at the Trombosedienst Maastricht.
A venipuncture can be experienced as annoying but gives little discomfort,
except that the insertion site should be covered for some time with a bandage.
The risk of a (severe) bleeding is present, especially since the patient is
using anticoagulants. To avoid unnecessary burden, blood samples will be taken
at home by a BIG registered nurse, who is competent to staunch bleeding or
refer the patient to a general practitioner. The nurse makes appointments with
the individual participant during business hours. With the obtained blood, the
lab tests will be performed as described in the study design. To be able to
follow the patient for any bleeding or thrombosis we use electronic patient
records. Permission for use will be asked. The total duration of the study per
participant is 9 days.