Theme Issue Article
Thrombin
generation, a function test of the haemostaticthrombotic system
H. Coenraad
Hemker, Raed Al Dieri, Erik De Smedt, Suzette Béguin
Synapse BV,
Cardiovascular Research Institute
Summary
By the use
of a fluorogenic thrombin substrate and continuous calibration of each
individual sample, it is now possible to obtain a thrombin generation (TG)
curve (or thrombogram) in plasma, with or without platelets, in an easy routine
procedure at high throughput and with an acceptable experimental error (<
5%). Evidence is growing that the parameters of the thrombogram, and notably
the area under the curve (endogenous thrombin potential, ETP), are useful in
assessing bleeding- or thrombotic risk and its modification by antithrombotic-
or haemostatic treatment. Available data strongly suggest that conditions
(congenital, acquired, drug-induced) that increase TG all cause
a thrombotic tendency and that conditions that decreaseTG prevent thrombosis
but, beyond a limit, cause bleeding. Diminution of TG is a common denominator
of all antithrombotic treatment, including anti-platelet drugs.The thrombogram
can also be used as a tool in the search for new antithrombotics and reflects the
haemorrhagic or thrombotic side effects of other drugs (e.g. oral
contraceptives).The thrombogram thus is a promising new approach to clinical
management of bleeding and thrombotic disease as well as a tool in drug
research and epidemiology. Our experience at this moment is insufficient,
however, to already clearly define its limits.
Keywords
Thrombin
generation, hypercoagulability, antithrombotics, haemophilia (Thromb
Haemost 2006; 96: 553–61)
Introduction
As we
know since Virchow, the occurrence and extent of thrombosis is determined by
the force of the local trigger, the local flow conditions and the readiness of
the blood to clot. For a long time this “coagulability” has been silently
identified with clotting times, as the name already indicates. However, more
than 95% of all thrombin forms after clotting has taken place (1), and one may well pose
the question why (2). Anyhow, it is there and is likely to serve a function. It
is therefore potentially interesting to assess this independent variable and investigate
its clinical usefulness. Independent variable, indeed, because the time that it
takes for clotting to start may or may not relate to the bulk amount of
thrombin formed. For example, tissue factor (TF)- induced clotting times
correlate with the amount of thrombin formed during anti-vitamin K treatment
but do not in heparin treatment, where a modified clotting time (the activated
partial thromboplastin time: aPTT) has to be used. Other antithrombotics (e.g.
dermatan sulphate) do not prolong any clotting time but they do decrease the
amount of thrombin formed. Clotting times and the amount of thrombin formed
thus are different things and both merit to be estimated. The reason that this
has not been done on a large scale was presumably for purely practical reasons–
doing a thrombin generation experiment with the methods available in the early
1980s was close to slave labour. Nevertheless, there were laboratories, such as
that of Barrowcliffe in
Noting
the interesting type of information that could be obtained, on the one hand we
applied it to problems such as the mode of action of heparins and the mechanism
of platelet procoagulant action, and on the other hand– and to make it simpler
– we automated it further and further. Over the last few years this has resulted
in a method that allows the thrombogram to be determined at high throughput in
the clinical laboratory (7) with a quite acceptable experimental error (<
5%). In so far as the present evidence goes, it certainly seems to be a welcome
extension of the possibilities of the coagulation laboratory.
It
does, indeed, produce information that the clotting time will not give.
Excessive thrombin formation, for example, always causes a thrombotic tendency
but hardly affects any clotting time. It also yields information that can not
be obtained from single factor determinations, because the extraordinary
complexity of the thrombin generating mechanism makes means that there is no
simple relation between factor concentrations and overall function. It is well
known, for example, that although haemophilia is a single-factor-disease, the
level of the missing factor is not a reliable indicator of the actual bleeding
tendency (8). This is probably because the state of the remainder of the
clotting system – including the platelets (9) – determines whether or not a
given haemophiliac can make sufficient use of the small residual amount of the
missing factor.
The place of
thrombin in H&T physiology, or textbook wisdom challenged
How
can thrombin be that important? Until yesterday – and until today in most
textbooks – the reigning paradigm of haemostasis states that firstly platelets
plug the wound and in a second stage thrombin is formed, which cements together
this platelet plug; the bleeding time reflects the first phase and the clotting
time the second. This view is based on the classical observation that
haemophiliacs have a normal bleeding time and a prolonged clotting time,
whereas thrombopenia (-pathy) results in a normal clotting time but a prolonged
bleeding time. There are – and have always been – observations that challenge
this "textbook view". Prolonged bleeding times have, for example,
been observed when TG is profoundly affected, as in severe overdosage of oral
anticoagulants (10) or heparin (11). Recently it has been found that mice with
defective platelet thrombin receptors show a prolonged bleeding time [see
further (12)]. The fact that thrombin is formed before a wound stops bleeding
is also in agreement with the old observation that products of thrombin action,
such as activated factors V and VIII, appear within seconds in the blood flowing
from a wound and diminish when heparin is administrated (13). The activity of
thrombin in such blood must be dependent upon platelet function because it is
inhibited by aspirin intake (14–17). Recently, early thrombin formation was
most convincingly demonstrated in the excellent experiments by the Furie group
[as reviewed in (18)]. They visualised exposed tissue factor, aggregated
platelets and fibrin at the site of a micro-trauma in a small vessel in the
mouse and saw fibrin appear as early as 10 – 15 seconds (sec) after the lesion.
An optimum concentration of tissue factor, as in the Quick time, clots plasma
in 12 sec, hence in these experiments thrombin must have appeared almost simultaneously
with the platelet aggregate!
It
thus appears that from the very beginning, platelets and plasma achieve
haemostasis in close cooperation. In fact, many experiments that support the
classical separation between platelet function and the clotting system were
conducted under conditions where one of the two was absent or inhibited, such
as platelets in anticoagulated blood (-plasma) or platelet-free plasma. As soon
as platelets and plasmatic coagulation factors are allowed to cooperate, they
invariably show early and strong mutual interaction. Thrombin is the most
potent platelet activator [see e.g. (12)] and the platelet membrane provides
procoagulant phospholipids (19) that allow explosive TG (20). This
platelet–clotting interaction, in which circulating tissue factor plays a key
role (21), is probably the most important positive feedback mechanism in
haemostasis and is responsible for the sudden arrest of blood flow from a
wound.
This
is not incompatible with the normal bleeding time in haemophiliacs. Early TG is
probably triggered by relatively high concentrations of tissue factor (21, 22)
and thus may be formed in a factor VIII and IX independent manner. The late
bleeding seen in haemophilia may well be related to the enhanced fibrinolysis
that ensues from insufficient activation of thrombin activated fibrinolysis
inhibitor (TAFI) (see below).
Not
only haemostasis but also thrombus growth requires thrombin generation.
Thrombus growth is due to a series of ligand- and receptor interactions that
have been studied in detail [(23) and references therein]. As soon as
experiments are carried out under conditions where thrombin can form, tissue
factor and thrombin appear to play an important role, especially at low
(venous) and intermediate (large artery) shear rates (12, 21, 24–27). At high
shear rates an effect of thrombin inhibition is not seen on initial thrombus
formation but becomes obvious when the experiment is prolonged (25). One of the
functions of the aggregate must be to serve as a niche in which a sufficient
concentration of thrombin can build up and diffuse into the surroundings. Where
the concentration of thrombin is high enough, new platelets will be recruited
and will be activated so that more thrombin will form [see also (28–31)].
Undoubtedly, several mechanisms of aggregation and plug- or thrombus- growth
exist in parallel. The relative importance of thrombin- and collagen-dependent
mechanisms has been shown to be influenced by the type of injury to a vessel
wall (22). As we will see below, clinical observations suggest that the
thrombin dependent mechanism must not be neglected.
Figure 1: The parameters of the thrombogram. A) Lag time (min), B) Peak height
(nM), C) Endogenous thrombin potential (ETP) (= area under the curve)(nM x
min), D) maximal rising slope (nM/min), E) time to peak (min).
Normal TG in
platelet poor plasma
To
assess the function of the thrombin generating system in a patient, one should
be able to measure the concentration of thrombin as a function of time and
space in a platelet aggregate at the site of a lesion. At the moment this is
technically impossible. A close approximation that is possible is measurement
of TG in PRP in the presence of elements of the vessel wall, i.e. tissue factor
(TF) and thrombomodulin (TM). Useful partial information can be obtained from
observations in platelet poor plasma (PPP).
Figure 2: Thrombin generation in contact activated
plasma, effect of heparin. (❑) Control; (X) 0.02 U/ml
unfractionated heparin; (❍) 0.03 U/ml
unfractionated heparin.
The
general form of any thrombogram is independent of the measuring system and the
experimental conditions. It starts with a lag phase, in which only minute
amounts of thrombin are formed, after which the full production starts with a
sudden burst of thrombin in nanomolar concentrations (Fig. 1). As has been known
for a long time, clotting occurs at the start of the explosive burst, i.e. at
the end of the lag time (1). The clotting time therefore is a good estimate of
the duration of the lag phase and vice versa. During the lag phase, also called
the initiation phase (32), hardly any observable amounts of thrombin form (33)
(Fig. 2) which, depending on the reaction conditions, activate one or more of
the factors V, VIII, XI and platelets and thus prepare the scene for the full
blown TG during the production (also called propagation-) phase (32). The
reaction mechanism in the lag phase and the production phase are different.
That is the reason why the lag- (=clotting-) time does not automatically
contain information on the amount of thrombin formed in the following peak
(Fig. 3).
Figure 3: Thrombin generation triggered with 25 pM
tissue factor, effect of heparin. (X) Control; (❍) 0.02 U/ml unfractionated
heparin; (●) 0.03 U/ml unfractionated
heparin.
There
is no sharp distinction between the production phase and the inactivation
phase. As soon as thrombin appears it is scavenged by the plasma antithrombins,
even during the lag phase. The velocity of inactivation increases proportionally
with the thrombin concentration. At the peak, thrombin generation and decay are
equally fast. Once past the peak, decay gains on prothrombin conversion, and
somewhere in the down slope the latter will stop. This can be shown by
calculating prothrombin conversion velocity from TG and the reaction constant
of thrombin – antithrombin interaction (34, 35).
It is
common knowledge that the pathway of thrombin formation is dependent upon the
type and the amount of trigger used. The physiological trigger is tissue factor
(TF). There is, however, no “physiological” concentration of tissue factor
because, in
vivo, it
is not a soluble reactant but a component of membranes that are large compared
to molecules. Reaction velocities are therefore determined by diffusion rather
than by chemical interaction. The same is true for the trigger of the protein C
system, thrombomodulin (TM). These two components, TF and TM, need to be added
in order to substitute for the (wounded or intact) vessel wall. For practical
reasons they are added in soluble form. The concentration of TF is chosen on basis
of the section of the reaction mechanism that one wants to study (see below),
that of TM so as to produce, in normal plasma, about ~ 50% inhibition of the
peak or of the ETP.
In
PPP, in the presence of ~4 μM procoagulant phospholipids and at high
concentrations of TF (>10 pM), the factor VIII, IX and XI -dependent
reactions are bypassed. Under the same conditions but at intermediate
concentrations (2–5 pM) of TF, TG is dependent upon the concentration of
factors VIII and IX, and at yet lower concentrations factor XI will start to
play a role (36). The amount of TM that inhibits TG by 50% decreases with
decreasing TF concentration. This stands to reason because TM needs to react with
thrombin before APC can be formed. At high TF concentration, TG can be so fast
that it is over before enough APC can be formed to delimit prothrombinase
action. Adding soluble TF and TM is about as close as we can come to the
in-vivo situation at the moment. However, it remains an artificial construct.
TF and TM are present in the same solution here, whereas in
vivo they
are membrane-bound in different compartments, TF in the wound, and TM on the
adjacent intact endothel ium.
Figure 4: Thrombin generation in PPP and in PRP,
effect of rTM (10 nM) and APC (10 nM). Fat lines: PPP, triggered with 5
pM of TF; lean lines PRP, triggered with 0.5 pM of TF. The small peak with the
same lag-time as the uninhibited TG is that with added TM. Addition of APC
increases the lag-time. Note that platelets appear to induce a resistance to
APC but not to TM.
Normal TG in
PRP, the roles of fibrin(-ogen)
TG in
PPP always requires procoagulant phospholipids to be present. At the
phospholipid-solute interface factors Xa and Va adsorb
to form prothrombinase (37) and factors IXa and VIIIa to form tenase (38). The
surface also serves to “funnel” the substrates (prothrombin and factor X) to
these enzyme complexes (39). If TG is observed in PPP without added
phospholipids, this is most often due to contamination with platelets or cell
remnants during plasma preparation. In carefully prepared plasma it indicated the
presence of procoagulant microparticles. In PRP, the platelets provide the
procoagulant phospholipids. Platelet activation that is important enough to
cause a prolonged rise of intracellular Ca++ makes
platelets lose the natural asymmetry of their plasma membrane (“scrambling”).
Thus, the outside of the platelet enriches in the procoagulant phospholipids
phosphatidyl serine (PS) and phosphatidyl ethanolamine (PE) that are normally found
at the inside (19). This turns the platelet surface into a two-dimensional
reaction compartment, in which adsorbed clotting factors generate thrombin far
more efficiently than in free solution (40, 41). Also, procoagulant vesicles
are shed from the
membrane of such activated
platelets. Hence, TG in PRP depends upon platelet activation equally as much as
on the plasmatic clotting factors. In vivo, there are two other sources of procoagulant phospholipids.
These are circulating procoagulant microparticles that tend to crowd on a
damaged site (42–44) as well as damaged or apoptotic cells that have lost the
normal asymmetry of their membrane (45, 46).
As
soon as a fibrin web forms, platelets, but also vesicles, adhere to its fibres
(47). This ensures that thrombin is generated on membrane structures that are
not able to move in solution. Under these conditions, it is not the chemical
reaction velocities, but rather the diffusional transport to and from the
membrane structures on the fibres which determine the velocity of TG. As
clotting takes place, TG velocity thus shifts from chemical to diffusional control
(48–50). When no fibrin can form, chemical kinetics remain
in force, so the bulk of thrombin is formed in an essentially different manner
in the presence and absence of fibrin. Another effect of the presence of
fibrinogen is that it attenuates the formation of α2macroglobulin-thrombin (α2M-thrombin) (Fig. 5).
Collagen
and thrombin are the classical triggers for the platelet procoagulant reaction
(19), so a role of the collagen receptors (GPIaIIa, GPVI) and thrombin
receptors (PAR 1, 4) was to be expected. Surprisingly, the fibrinogen receptor
GPIIb/IIIa was involved (51), not least because it revealed that drugs
conceived as aggregation inhibitors, such as abciximab, diminish TG in PRP.
Another
unexpected observation was that a clot, as such, provokes procoagulant activity
in platelets (52, 53). A snake venom enzyme such as Arvin does not play a role
on the platelet or the coagulation system apart from the fact that it clots
fibrinogen. Bringing about a clot with Arvin, quite unexpectedly, was found to
trigger TG in PRP but not in PPP. The underlying mechanism appeared to be that
polymerising fibrin interacts with vWF and modifies it in the same way as
adsorption onto collagen or shear stress does. The modified vWF then via GP1b
triggers the platelet procoagulant activity (54, 55). In this way fibrin takes
over the role of collagen as soon as platelets cover the collagen surface and
fibrin forms in the aggregate. Clotting apparently is not the closing act of
haemostasis but it is a dynamic part of the haemostatic process. This also
explains why thrombin becomes increasingly important the less collagen is
exposed (22).
This
fibrin-vWF-GP1b mechanism has been verified by showing that TG is diminished in
the PRP (but not in the PPP) of hypofibrinogenaemia, in mild von Willebrand
disease and in patients with Bernard-Soulier syndrome (in which functional GP1b
is congenitally lacking). It reveals a third function of vWF, that not only
acts as a “glue” for platelets onto matrix proteins and as a carrier of factor
VIII but, in PRP, also contributes to activation of the procoagulant function
of platelets and thus can be called a clotting factor in its own right.
Figure 5: Amidolytic activity in PPP with (left) and
without (right) fibrinogen, influence of heparin. Inset: Inhibition of thrombin
generation in % as a function of the heparin concentration (0.1, 0.2, 0.3 and 0,4 μg/ml).
TG and
thrombosis risk
The
pivotal role of thrombin in venous thrombotic disease goes undisputed [e.g.
(56)]. This is directly reflected in medical practice: treatment and prevention
of venous thrombosis require drugs that diminish thrombin activity in one way
or another. This is done by either impeding prothrombin synthesis (vitamin K
antagonists), increasing antithrombin activity (heparins), or by inhibiting thrombin
directly (hirudin and small-molecular-weight reversible inhibitors). Increased formation
of thrombin in PPP always induces a risk of venous thrombosis, whether it is
due to deficiency of antithrombin (57), an excess of prothrombin (58), or to
any other cause. Disorders in the protein C pathway (deficiency
of proteins S and C, factor VLeiden)
increase TG as such and becomes particularly obvious if the protein C
pathway is activated by thrombomodulin (Fig. 4). The thrombotic tendency induced
by oral contraceptives can be explained by the increase of TG caused by an
acquired resistance to activated protein C (APC) (59–61) possibly combined with
other changes (62, 63).
Particularly
interesting is the lupus anticoagulant. This antibody causes an increase of the
lag time of thrombin formation, i.e. of the clotting time, and is thus deemed “anticoagulant”.
It also brings about an important resistance to the activity of the protein C
system and hence an increased TG in the presence of thrombomodulin (64). This
again underlines that TG during the lag time is a different process from that
in the production phase but that it is the latter that governs the thrombotic
tendency. The “LAC paradox”, i.e. prolonged clotting times together with a thrombotic
tendency, thus disappears as soon as one resorts to TG.
The
role of the plasmatic thrombin generating system is much more evident in venous
than in arterial disease. There are arguments to surmise that thrombin plays a
part in arterial trombosis as well (25, 65–67). Excess amounts of factors II,
VIII and VII have been found to correlate with the occurrence of myocardial infarction.
Also, higher than normal levels of vWF increase TG and are a risk factor for
arterial thrombosis (68–74). In a sub-population of young stroke patients
(~30%), both TG in PRP and vWF have been shown to be significantly higher than normal
(75). Clinical trials have shown that vitamin K antagonists (76) as well as
heparins (77, 78) decrease the reoccurrence rate of myocardial infarction.
“Antiplatelet” drugs derive at least part of their effect by inhibiting TG in
PRP (14, 15, 51). Inversely, aspirin has been shown to
be beneficial also in venous thrombosis (79). Like the distinction between
primary and secondary haemostasis, the classical view that arterial thrombosis
is due to the platelet and venous thrombosis to clotting seems to become less
evident. Nevertheless, the precise role of thrombin generation in arterial
thrombotic disease remains enigmatic. We surmise that much can be learned from
the study of thrombin generation in PRP in relation to arterial disease.
TG and bleeding
It
has been demonstrated for the haemophilias (VIII, IX or XI) as well as for all
rare clotting factor deficiencies (II, V, VII, X, XII) that TG is diminished
and that a clinical bleeding tendency is seen as soon as TG drops below 20% of
normal (9, 36, 80, 81). In haemophilia A, not only does infusion of factor VIII
or administration of DDAVP augment TG (82), but also inhibitor bypassing therapy
with products containing prothrombin and/or factor VII (83–87). Any overshoot
of antithrombotic therapy (-prevention) invariably carries a risk of bleeding,
because antithrombotics act through diminution of TG.
Severe
thrombopenia (<50.000 μl-1) causes decreased TG as well as the Glanzmann (51)
and Bernard-Soulier (52) thrombopathies. As mentioned above, in von
Willebrand’s disease, TG in PRP is significantly impaired. The defect is much
more pronounced in PRP than in PPP, which indicates that it can not be
explained by the concomitant deficiency of factor VIII. The discrepancy between
TG in PPP and in PRP in von Willebrand disease has not yet been explored but is
a potentially useful diagnostic possibility.
One
of the reasons that a low ETP causes a bleeding tendency is likely due to the
fact that the amount of thrombin formed determines the amount of fibrinolysis
inhibitor TAFI that is generated (88). Low TG therefore causes increased
fibrinolysis. This is likely to be the explanation for the fibrinolytic
character of much of the bleeding in haemophilia and would explain the beneficial
effect of antifibrinolytic agents [e.g. (89)].
How can the
thrombin-forming capacity of blood be assessed?
Classically,
thrombograms were obtained by timed subsampling of aliquots from clotting blood
or plasma onto a solution of diluted bovine plasma (without Ca++) and
by assessing the thrombin concentration in the sample from the clotting time
observed (1). The procedure, which takes about one man-hour per curve, involves
a score of stopwatches and extraordinary dexterity. Subsampling onto a
chromogenic thrombin substrate (35) allows automatic recording of the sampling
time and semi-automatic measurement of the thrombin activities. In this way,
several parallel experiments can be done, which made it possible to recognise the
role of various platelet receptors and fibrinogen (51–53).
The
use of small MW substrates also introduces a complication because of the
exceptional way in which α2macroglobulin (α2M) inhibits thrombin:
After initial proteolytic interaction of thrombin with a “bait”-region,
thrombin is bound in a niche of α2M so that, through steric hindrance, no physiological
substrates can be split anymore. Small substrates can still be attacked, however.
The total amidolytic activity measured during TG is the sum of free thrombin
and the α2M-thrombin
complex (Fig. 5). The velocity of the formation of the complex is in good
approximation proportional to the concentration of free thrombin. Therefore,
the amount of α2M-thrombin
in serum is proportional to the integral of the thrombogram, i.e. to the ETP. A
simple algorithm, executable in any spreadsheet program, allows one to split the
curve of amidolytic activity into its free-thrombin and α2M-thrombin parts [see
further (90)]. In defibrinated plasma, the α2M-thrombin end level after
TG can be used to estimate the effect of APC. Rosing et al. (60) used this
approach to demonstrate the acquired APC-resistance induced by oral contraceptives.
The
next step in simplifying TG measurement was to add the chromogenic substrate
directly to the clotting plasma. This can not seriously be attempted with the
normal high affinity substrates because they bind thrombin so tightly that they
act as potent thrombin inhibitors, so that the normal thrombin generation mechanism,
with its many thrombin-driven feedback reactions, will be profoundly disturbed.
Once thrombin is formed, the substrate is converted very quickly so the decay
phase seen is due to substrate exhaustion and not to thrombin disappearance.
The resulting curves have a superficial similarity to TG curves but none of the
parameters of the real course of thrombin in time can be read from them.
Correct
measurement of the clotting function requires substrates that do not
significantly interfere with the normal prothrombin conversion process and have
such kinetic properties that, during the whole process, the
velocity of product formation is in good approximation proportional to
the prevailing concentration of thrombin. Chromogenic substrates of this type
are available and allow the monitoring of thrombin activity photometrically (91).
Although they do not interfere with normal prothrombin conversion, they will,
like any thrombin substrate, compete with the plasmatic antithrombins for the
active site of thrombin. Therefore, thrombin decay is always slowed down and the
measured ETP is consequently artificially augmented. The magnitude of this
effect can be easily calculated and adjusted for [the real ETP equals Km/(Km+S) times the observed ETP].
As OD
measurement is disturbed by light scattering, clot formation has to be avoided
when chromogenic substrates are used, either by defibrination or by adding
polymerization inhibitors. Defibrination can be carried out with suitable snake
venom enzymes (e.g. Arvin). However, platelets are removed together with the
fibrin, so chromogenic continuous methods are restricted to defibrinated PPP.
Defibrination abolishes the role of diffusion in TG and significantly changes
its kinetics (see above). As discussed, the effect of added heparin, for
example, is quite different in normal plasma and defibrinated plasma (Fig. 5).
Polymerisation inhibitors may interfere with normal prothrombin conversion and alter
the behaviour of platelets (unpublished results).
Fluorescence
can be measured in turbid media, so the use of fluorogenic substrates abolishes
the drawbacks of defibrination or inhibition of polymerisation (92). However,
it introduces other problems. Firstly, plasma adsorbs a significant and
variable amount of the light, so that the signal from plasma, even at 2:5 dilution, is on the mean ~ 65% of that in buffer. Worse is
the large variation of this property between apparently normal plasmas: ~ 15%.
This means that the reaction velocities measured in plasma can not be
standardised by comparison to buffer, or by comparison to another
plasma. Or, if thrombin activity in buffer is used as a standard [as for
example in (86, 93, 94)], thrombin generation will be ~ 30% underestimated and
~ 15% is added to the experimental variation
Using
the available substrates, the proportionality between reaction velocity and
product formation can not be maintained during the experiment and also the
fluorescent signal is not proportional to the concentration of fluorophore in
the solution. As a result, the same amount of thrombin activity causes a far greater
increase of signal at the beginning of the experiment than
towards the end. In other words,
the calibration factor is a function of the amount of fluorescent product
already present. Moreover, the readings are very sensitive to the colour of the
plasma, notably to minimal haemolysis (Fig. 6).
Figure 6: The fluorescence evolving from 100 nM
thrombin activity in three different plasmas. The stable thrombin activity was obtained
through addition of α2M-thrombin.
A)
The
three drawbacks of fluorescent methods have to be dealt with by continuous,
individual calibration of every sample. In this technique, a fixed amount of
constant thrombin activity is added to a parallel sample of the same plasma.
From the resulting curve, the calibration factor at any level of fluorescence
is read and the exact thrombin concentration in the sample where TG is taking
place can be calculated. Using continuous calibration significantly reduces the
intra-individual experimental error from >15% to ~5%.
Methods
that use fibrinogen as an indicator substance (95–97) are ideal for the
determination of the length of the initiation phase, i.e. the clotting time,
especially as no foreign substances have to be added. When the mechanical
properties of the clot serve as an indicator, an additional advantage is that
whole blood can be used directly (98). Alternatively, the turbidity of the clot
in plasma can be measured (99). The greatest drawback of fibrinogen- based
methods is that fibrinogen is consumed long before thrombin formation is over
so that no information can be obtained on the stages of thrombin generation
beyond the very beginning of the production (= propagation) phase. A second
drawback is that there is no known relation between mechanical properties or
turbidity and the activity of thrombin, so that even the information on this
first stage of TG can not be quantified in terms of thrombin concentrations.
This is all the more serious when curves are constructed that can easily be
mistaken for real thrombograms (100–102) as a result of mathematical
transformations that do not improve the content of information in any way. The
question remains whether the observed signals provide more useful information
than the clotting time does.
It is
perhaps now appropriate to remark that a multitude of different methods allow
scores of measurable properties of blood (-plasma) that change as a result of
the formation of thrombin and/or fibrin. Parameters can be obtained, e.g. with
unsuitable substrates, or can be derived from the mechanical properties of forming
fibrin (clotting times, thrombelastogram) or from its turbidity (wave form
analysis). Other parameters arise when data from suitable substrates are
inadequately calibrated etc. Often it will be possible to correlate such
changes to a clinical picture (103). There is a respectable clinical tradition
of linking misunderstood phenomena to clinical diagnosis (e.g. spider naevi
> liver cirrhosis). This may also extend to laboratory observations. Erythrocyte
sedimentation velocity started out as a pregnancy test. One can not deny the
use of such correlations. Nevertheless, laboratory variables, in some way resulting
from thrombin action, are better abandoned now that correct assessment of this physiological
function has become available.
Perspectives
From
the above it looks as if the thrombograms might be an overall function test of
the haemostatic-thrombotic system. In the present stage of development, the
focus is primarily to explore its possibilities. It will inevitably be followed
by a phase in which the merits of the test will be critically explored and the
limits of its use defined. A few domains in which we expect the test to be useful
are mentioned below. Future research will show which of the present projections
will be realized.
A: Hypercoagulability and the risk of thrombosis: Following
on from the observation that all known risk factors increase TG, one might
infer that increased TG is a risk factor as such, independent of its cause.
This is reinforced by the observation that TG increases with age in a
population without known risk factors (104, 105). The large inter-individual
variation of the ETP in the normal population (CV 16%) makes it not uncommon for
a normal person to generate twice as much thrombin as any other equally normal
person! Larger epidemiological studies seem justified to see whether such
differences influence the occurrence of thrombosis. A relation between TG in
PPP and venous thrombosis appears quite likely, especially when the function of
the protein C system is probed by adding TM. The subject of thrombin generation
in PRP and arterial thrombosis currently remains completely unexplored.
B: Control of antithrombotic therapy: At
the moment, different tests are being used for the control of different
antithrombotics. Diminution of TG is a common denominator of all
anticoagulation, including that by low-molecular-weight heparins and heparin
likes that hardly influence any clotting test. It also monitors the combined
effect of oral anticoagulation and heparins, as in the treatment of venous
thrombosis.
The
real challenge in this domain, however, is investigating the effect of anticoagulant
drugs and platelet inhibitors on TG in PRP. Does platelet inhibition enhance
the effect of heparin because it inhibits the release of platelet factor 4? Is
the relative APC-resistance of PRP enhanced by oral anticoagulation?; etc. etc. The (clinical-) pharmacology of antithrombotic
drugs acquires a new playground now that experiments on the isolated target
organ blood are possible.
C: Closely related to the previous item is the role
that TG may play in finding and testing of new antithrombotics. We dare
to surmise that a candidate drug that inhibits TG by about 40–60% for 24 hours
per day, will have an antithrombotic effect and will
not cause bleeding. As a screening test TG may significantly reduce the number
of experimental animals required to establish the efficacy of new drugs. It may
also be useful to monitor a new drug in human experiments and clinical trials. It
might even lead to the discovery of drugs that can hardly be found in another
way. One might dream, for example, of an orally available heparin-like compound
which acts via heparin cofactor II (HCII). Because there is only ~ 1 μM of HCII in plasma,
complete activation will never scavenge all the thrombin that can be formed
from prothrombin, presently ~2 μM, so overdosage is impossible. Such a drug could not
be detected with any clotting test but will immediately show its effect in a TG
experiment.
D: In haemophilia the discrepancy between factor VIII level
and clinical bleeding tendency remains to be solved. Pilot
experiments have shown that TG can be used for therapeutic decisions in
inhibitor bypassing therapy (83). If it could be proven that the level of TG –
in PPP or in PRP – indeed indicates the bleeding risk, not only would clinical
decision making be much easier, but one could also make a more economic use of very
costly preparations.
E: Diagnosis of von Willebrand disease: Because
vWF is required for the development of full procoagulant power in platelets, a
discrepancy between TG in PPP and PRP is systematically found in von Willebrand
disease. This could be explored as a means of diagnosis.
References
1. Biggs R, Macfarlane RG. Human Blood
Coagulation and its Disorders.
2. Mann KG, Brummel K, Butenas S. What is all that thrombin
for? J Thromb Haemost 2003; 1: 1504–14.
3. Barrowcliffe TW, Gray E, Kerry PJ, et al.
Triglyceride- rich lipoproteins are responsible for thrombin generation induced
by lipid peroxides. Thromb
Haemost 1984; 52: 7–10.
4. Barrowcliffe TW, Gutteridge JM, Dormandy TL. The effect of fatty-acid autoxidation products on blood coagulation.
Thromb Diath Haemorrh 1975; 33: 271–7.
5. Josso F, Monasterio de Sanchez J, Lavergne JM, et al.
Congenital abnormality of the prothrombin molecule (factor II) in four
siblings: prothrombin Barcelona. Blood 1971; 38: 9–16.
6. Josso F,
7. Hemker HC, Giesen P, AlDieri R, et al. The
calibrated automated thrombogram (CAT): a universal routine test for hyper- and
hypocoagulability. Pathophysiol Haemost Thromb 2002; 32: 249–53.
8. Aledort LM. Why thrombin generation? From bench to bedside. Pathophysiol Haemost Thromb 2003; 33:
2–3.
9. Siegemund T, Petros S, Siegemund A, et al. Thrombin generation
in severe haemophilia A and B: the endogenous thrombin potential in
platelet-rich plasma. Thromb Haemost 2003; 90: 781–6.
10. Marongiu F, Biondi G, Sorano GG, et al. Bleeding time is prolonged during oral anticoagulant
therapy. Thromb Res 1990; 59: 905–12.
11. Lavelle SM, MacIomhair M. Bleeding times and the
antithrombotic effects of high-dose aspirin, hirudin and heparins in the rat. Ir J Med Sci 1998; 167: 216–20.
12. Sambrano GR, Weiss EJ, Zheng YW, et al. Role of thrombin
signalling in platelets in haemostasis and thrombosis. Nature 2001; 413: 74–8.
13. Jensen AH, Beguin S, Josso F. Factor V and VIII
activation „in vivo“ during bleeding. Evidence of thrombin formation at the early stage of hemostasis.
Pathol Biol (
14. Kessels H, Beguin S, Andree H, et al. Measurement of
thrombin generation in whole blood--the effect of heparin and aspirin. Thromb Haemost 1994; 72: 78–83.
15. Musial J, Radwan J, Szczeklik A. Aspirin delays thrombin
generation in vitro through interaction with platelet phospholipids. Thromb Res 1997; 85: 367–8.
16. Musial J, Swadzba J, Jankowski M, et al. Thrombin generation
measured ex vivo following microvascular injury is increased in SLE
patients with antiphospholipidprotein antibodies. Thromb
Haemost 1997; 78: 1173–7.
17. Szczeklik A. Inhibition of thrombin generation by aspirin.
Thromb Haemost 1994; 72: 988–9.
18. Celi A, Merrill-Skoloff G, Gross P, et al. Thrombus formation:
direct real-time observation and digital analysis of thrombus assembly in a
living mouse by confocal and widefield intravital microscopy. J Thromb Haemost
2003; 1: 60–8.
19. Bevers EM, Comfurius P, van Rijn JL, et al. Generation
of prothrombin-converting activity and the exposure of phosphatidylserine at
the outer surface of platelets. Eur J Biochem 1982; 122:
429–36.
20. Beguin S, Lindhout T, Hemker HC. The effect of trace
amounts of tissue factor on thrombin generation in platelet rich plasma, its
inhibition by heparin. Thromb Haemost 1989; 61: 25–9.
22. Furie BC, Furie B. Tissue factor pathway vs. collagen pathway
for in
vivo platelet
activation. Blood Cells Mol Dis 2006.
23.
24. Fressinaud E,
25. Gast A, Tschopp TB,
Baumgartner HR. Thrombin plays a key role in late platelet thrombus growth
and/or stability. Effect of a specific thrombin inhibitor on thrombogenesis
induced by aortic subendothelium exposed to flowing rabbit blood. Arterioscler Thromb 1994; 14: 1466–74.
26. Inauen W, Baumgartner HR, Bombeli T, et al. Doseand shear
rate-dependent effects of heparin on thrombogenesis induced by rabbit aorta
subendothelium exposed to flowing human blood. Arteriosclerosis
1990; 10: 607–15.
27. Weiss EJ, Hamilton JR, Lease KE, et al. Protection against
thrombosis in mice lacking PAR3. Blood 2002; 100: 3240–4.
28. Anand M, Rajagopal K, Rajagopal KR.A model for the
formation and lysis of blood clots. Pathophysiol Haemost Thromb 2005; 34:
109–20.
29. Ermakova EA, Panteleev MA, Shnol EE. Blood coagulation and propagation of autowaves in flow.
Pathophysiol Haemost Thromb 2005; 34: 135–42.
30. Fogelson AL, Tania N. Coagulation under flow: the influence
of flow-mediated transport on the initiation and inhibition of coagulation.
Pathophysiol Haemost Thromb 2005; 34: 91–108.
31. Lobanov AI, Starozhilova TK. The effect of convective flows
on blood coagulation processes. Pathophysiol Haemost Thromb 2005; 34: 121–34.
32. Butenas S, DiLorenzo ME, Mann KG. Ultrasensitive
fluorogenic substrates for serine proteases. Thromb
Haemost 1997; 78: 1193–201.
33. Hemker HC, Beguin S. The mode of
action of heparins in vitro and in vivo. Adv Exp Med
Biol 1992; 313: 221–30.
34. Beguin S, Kessels H, Dol F, et al. The
consumption of antithrombin III during coagulation, its consequences for the
calculation of prothrombinase activity and the standardisation of heparin
activity. Thromb Haemost 1992; 68: 136–42.
35. Hemker HC, Willems GM, Beguin S. A computer assisted
method to obtain the prothrombin activation velocity in whole plasma
independent of thrombin decay processes. Thromb Haemost 1986;
56: 9–17.
36. Keularts IM, Zivelin A, Seligsohn U, et al. The role of factor XI in thrombin generation induced by low concentrations
of tissue factor. Thromb Haemost 2001; 85: 1060–5.
37. Hemker HC, Esnouf MP, Hemker PW, et al. Formation of
prothrombin converting activity. Nature 1967; 215: 248–51.
38. Hemker HC, Kahn MJ. Reaction
sequence of blood coagulation. Nature 1967; 215:
1201–2.
40. Nesheim ME, Taswell JB, Mann KG. The contribution of
bovine Factor V and
41. Rosing J, Tans G, Govers-Riemslag JW, et al. The role
of phospholipids and factor Va in the prothrombinase complex.
J Biol Chem 1980; 255: 274–83.
42. Falati S, Liu Q, Gross P, et al. Accumulation of
tissue factor into developing thrombi in vivo is
dependent upon microparticle P-selectin glycoprotein ligand 1 and platelet
P-selectin. J Exp Med 2003; 197: 1585–98.
44. Sim D, Flaumenhaft R, Furie B, et al. Interactions of
platelets, blood-borne tissue factor, and fibrin during arteriolar thrombus
formation in
vivo.
Microcirculation 2005; 12: 301–11.
45. Bombeli T, Karsan A, Tait JF, et al. Apoptotic
vascular endothelial cells become procoagulant. Blood 1997; 89:
2429–42.
46. Flynn PD, Byrne CD, Baglin TP, et al. Thrombin generation
by apoptotic vascular smooth muscle cells. Blood 1997; 89:
4378–84.
47. Siljander P, Carpen O, Lassila R. Platelet-derived microparticles
associate with fibrin during thrombosis. Blood 1996; 87:
4651–63.
48. Hemker HC, de Smedt E, Hemker PW. Mathematical and biological models of blood coagulation. A rebuttal. J Thromb Haemost 2006; 4: 710–1.
49. Hemker HC, E DES, Hemker PW. During coagulation, thrombin
generation shifts from chemical to diffusional control. J Thromb Haemost 2005;
3: 2399–400.
50. Mann KG, Brummel-Ziedins K, Orfeo T, et al. Models of
blood coagulation. Blood Cells Mol Dis 2006.
51. Reverter JC, Beguin S, Kessels H, et al. Inhibition of
platelet-mediated, tissue factor-induced thrombin generation by the mouse/human
chimeric 7E3 antibody. Potential implications for the effect
of c7E3 Fab treatment on acute thrombosis and „clinical restenosis“. J
Clin Invest 1996; 98: 863–74.
52. Beguin S, Keularts I, Al Dieri R, et al. Fibrin
polymerization is crucial for thrombin generation in platelet- rich plasma in a
VWF-GPIb-dependent process, defective in Bernard-Soulier syndrome. J Thromb
Haemost 2004; 2: 170–6.
53. Beguin S, Kumar R. Thrombin, fibrin and platelets: a
resonance loop in which von Willebrand factor is a necessary link. Thromb Haemost 1997; 78: 590–4.
54. Kumar R, Beguin S, Hemker HC. The influence of fibrinogen
and fibrin on thrombin generation—evidence for feedback activation of the
clotting system by clot bound thrombin. Thromb Haemost 1994;
72: 713–21.
55. Kumar R, Beguin S, Hemker HC. The
effect of fibrin clots and clot-bound thrombin on the development of platelet
procoagulant activity. Thromb Haemost 1995; 74: 962–8.
56. Nossel HL, Yudelman I, Canfield RE, et al. Measurement
of fibrinopeptide A in human blood. J Clin Invest 1974;
54: 43–53.
57. Wielders S, Mukherjee M, Michiels J, et al. The routine
determination of the endogenous thrombin potential, first results in different
forms of hyper- and hypocoagulability. Thromb Haemost 1997;
77: 629–36.
58. Kyrle PA, Mannhalter C, Beguin S, et al. Clinical studies
and thrombin generation in patients homozygous or heterozygous for the G20210A
mutation in the prothrombin gene. Arterioscler Thromb Vasc
Biol 1998; 18: 1287–91.
59. Nicolaes GA, Thomassen MC, Tans G, et al. Effect of
activated protein C on thrombin generation and on the thrombin potential in
plasma of normal and APCresistant individuals. Blood Coagul
Fibrinolysis 1997; 8: 28–38.
60. Rosing J, Tans G,
61. Rotteveel RC, Roozendaal KJ, Eijsman L, et al. The influence of oral contraceptives on the time-integral of thrombin
generation (thrombin potential). Thromb Haemost 1993;
70: 959–62.
62. Schubert L, Gibelli A, De Nicola P. Blood coagulation and
fibrinolysis in patients treated with an anovulatory drug with
estroprogestational action. Ann Ostet Ginecol Med Perinat
1968; 90: 48–58.
63. van Deijk WA, van Dam-Mieras MC, Muller AD, et al.
Evaluation of a coagulation assay determining the activity state of factor VII
in plasma. Haemostasis 1983; 13: 192–7.
64. Regnault V, Beguin S, Wahl D, et al. Thrombinography shows
acquired resistance to activated protein C in patients with lupus
anticoagulants. Thromb Haemost 2003; 89: 208–12.
65. Badimon L, Badimon JJ, Lassila R, et al. Thrombin regulation
of platelet interaction with damaged vessel wall and isolated collagen type I
at arterial flow conditions in a porcine model: effects of hirudins, heparin, and
calcium chelation. Blood 1991; 78: 423–34.
66. Badimon JJ, Lettino M, Toschi V, et al. Local
inhibition of tissue factor reduces the thrombogenicity of disrupted human
atherosclerotic plaques: effects of tissue factor pathway inhibitor on plaque
thrombogenicity under flow conditions. Circulation 1999; 99: 1780–7.
67. Heras M, Chesebro JH, Webster MW, et al. Hirudin, heparin,
and placebo during deep arterial injury in the pig. The in
vivo role of thrombin in platelet-mediated thrombosis. Circulation 1990; 82: 1476–84.
68. Giroud M, Dutrillaux F, Lemesle M, et al. Coagulation abnormalities
in lacunar and cortical ischemic stroke are quite different. Neurol
Res 1998; 20: 15–8.
69. Gurfinkel E, Altman R, Scazziota A, et al. Importance of
thrombosis and thrombolysis in silent ischaemia: comparison of patients with
acute myocardial infarction and unstable angina. Br Heart J
1994; 71: 151–5.
70. Haider AW, Andreotti F, Thompson GR, et al. Serum
lipoprotein(a) level is related to thrombin generation and spontaneous
intermittent coronary occlusion in patients with acute myocardial infarction. Circulation 1996; 94: 2072–6.
71.
72. Soskin P, Mossard JM, Arbogast R, et al. Variation in
von Willebrand's Factor according to the treatment of acute myocardial
infarction: physiopathological and clinical implications. Eur
Heart J 1994; 15: 479–82.
73. Soskin P, Wiesel ML, Mossard JM, et al. Potential thrombogenicity
of coronary angioplasty compared with coronary angiography. Blood Coagul
Fibrinolysis 1994; 5: 965–6.
74. Tanaka M, Suzuki A. Hemostatic abnormalities in acute
myocardial infarction as detected by specific blood markers. Thromb
Res 1994; 76: 289–98.
75. Faber CG, Lodder J, Kessels F, et al. Thrombin
generation in platelet-rich plasma as a tool for the detection of hypercoagulability
in young stroke patients. Pathophysiol Haemost Thromb 2003; 33: 52–8.
77. Neri Serneri GG, Modesti PA, Gensini GF, et al. Randomised
comparison of subcutaneous heparin, intravenous heparin, and aspirin in
unstable angina. Studio Epoorine Sottocutanea nell'Angina
Instobile (SESAIR) Refrattorie Group. Lancet 1995;
345: 1201–4.
78. Neri Serneri GG, Rovelli F, Gensini GF, et al.
Effectiveness of low-dose heparin in prevention of myocardial reinfarction. Lancet 1987; 1: 937–42.
79. Prevention of pulmonary embolism and deep vein thrombosis
with low dose aspirin: Pulmonary Embolism Prevention (PEP) trial. Lancet 2000;
355: 1295–302.
80. Al Dieri R, Peyvandi F, Santagostino E, et al. The thrombogram
in rare inherited coagulation disorders: its relation to clinical bleeding.
Thromb Haemost 2002; 88: 576–82.
81. Dargaud Y, Beguin S, Lienhart A, et al. Evaluation of
thrombin generating capacity in plasma from patients with haemophilia A and B.
Thromb Haemost 2005; 93: 475–80.
82. Keularts IM, Hamulyak K, Hemker HC, et al. The effect
of DDAVP infusion on thrombin generation in platelet-rich plasma of von
Willebrand type 1 and in mild haemophilia A patients.
Thromb Haemost 2000; 84: 638–42.
83. Dargaud Y, Lienhart A, Meunier S, et al. Major surgery
in a severe haemophilia A patient with high titre inhibitor: use of the
thrombin generation test in the therapeutic decision. Haemophilia 2005; 11:
552–8.
84. Hedner U. Dosing with recombinant factor viia based on
current evidence. Semin Hematol 2004; 41: 35–9.
85. Hedner U. Mechanism of action of recombinant activated
factor VII: an update. Semin Hematol 2006; 43: S105–7.
88. Lisman T, Mosnier LO, Lambert T, et al. Inhibition of
fibrinolysis by recombinant factor VIIa in plasma from patients with severe
hemophilia A. Blood 2002; 99: 175–9.
89. Lee AP, Boyle CA, Savidge GF, et al. Effectiveness in
controlling haemorrhage after dental scaling in people with haemophilia by
using tranexamic acid mouthwash. Br Dent J 2005; 198: 33–8; discussion 26.
90. Hemker HC, Beguin S. Thrombin generation in plasma:
its assessment via the endogenous thrombin potential. Thromb
Haemost 1995; 74: 134–8.
91. Hemker HC, Wielders S, Kessels H, et al. Continuous registration
of thrombin generation in plasma, its use for the determination of the thrombin
potential. Thromb Haemost 1993; 70: 617–24.
92. Hemker HC,
93. Varadi K, Negrier C, Berntorp E, et al. Monitoring the bioavailability of FEIBA with a thrombin
generation assay. J Thromb Haemost 2003; 1: 2374–80.
94. Varadi K,
95. Hartert H. Thrombelastography, a method for physical
analysis of blood coagulation. Z Gesamte Exp Med 1951; 117: 189–203.
96. Luddington RJ. Thrombelastography/thromboelastometry.
Clin Lab Haematol 2005; 27: 81–90.
97. Velik-Salchner C, Schnurer C, Fries D, et al. Normal values
for thrombelastography (ROTEM) and selected coagulation parameters in porcine blood.
Thromb Res 2006; 117: 597–602.
98. Rivard GE, Brummel-Ziedins KE, Mann KG, et al. Evaluation
of the profile of thrombin generation during the process of whole blood
clotting as assessed by thrombelastography. J Thromb Haemost 2005; 3: 2039–43.
99. Downey C, Kazmi R, Toh CH. Novel and diagnostically applicable
information from optical waveform analysis of blood coagulation in disseminated
intravascular coagulation. Br J Haematol 1997; 98: 68–73.
100. Hvitfeldt Poulsen L, Christiansen K, Sorensen B, et
al. Whole blood thrombelastographic coagulation profiles using minimal tissue
factor activation can display hypercoagulation in thrombosis-prone patients. Scand
J Clin Lab Invest 2006; 66: 329–36.
101. Ingerslev J, Poulsen LH, Sorensen B. Potential role of
the dynamic properties of whole blood coagulation in assessment of dosage
requirements in haemophilia. Haemophilia 2003; 9: 348–52.
102. Sorensen B, Johansen P, Christiansen K, et al. Whole
blood coagulation thrombelastographic profiles employing minimal tissue factor
activation. J Thromb Haemost 2003; 1: 551–8.
103. Barrowcliffe TW, Cattaneo M, Podda GM, et al. New approaches for measuring coagulation. Haemophilia 2006;
12 (Suppl 3): 76–81.
104. Hemker HC, Al Dieri R. Age-dependency of thrombin generation.
Thromb Haemost 2006; 95: 756–7.
105. Haidl H, Cimenti C, Leschnik B, et al. Age-dependency of
thrombin generation measured by means of calibrated automated thrombography
(CAT). Thromb Haemost 2006; 95: 772-5.