|
W.
Nieuwenhuizen, Ph.D.,
TNO-PG Gaubius Laboratory, P.O. Box 2215, 2301 CE
Leiden, The Netherlands
 Download as a PDF
here
Abstract
Information about the status of the haemostatic balance can be
derived from the end products of coagulation and fibrinolysis, i.e.
soluble fibrin and fibrin degradation products (FbDP), such as
D-dimer, respectively. Assays for FbDP have been available for more
than 15 years. Examples are semi-quantitative latex
agglutination assays, and quantitative enzyme immunoassays
for D-dimer. It is known that the use of serum can lead to
erroneous and even false-positive or false-negative results. Little
is known about the use of the assays in the Netherlands (type of
test; serum or plasma as sample; requested by which specialism; for
which indications; cito assay or not; numbers of assays). To
collect such data we sent out questionnaires to 116 clinical
centres. On the basis of the responses received from 82 centres we
can conclude that the vast majority of the centres (76) use
semi-quantitative latex tests. Of these 76 centres 59 used plasma
samples (28000 tests/year); and 17 used serum (4800 test/year). The
assays are done at the request of gynaecologists, internists,
intensive care units and cardiologists for a variety of indications
such as DIC, pregnancy complications, DVT and PE. In most cases
cito assays were involved.
The D-dimer assays are discussed with special reference to
standardisation, (biochemical) specificity, reproducibility, and
the reasons why serum can cause erroneous results.
Introduction
Under normal conditions there is equilibrium between on the one
hand the activity of the coagulation system, and on the other hand
the activity of the fibrinolytic system. Disturbance of this
equilibrium, designated as the haemostatic balance, can cause
bleedings when the fibrinolytic system is relatively more active
than the coagulation system, and can cause thrombotic
phenomena when coagulation is more active than fibrinolysis.
An activated coagulation system leads to thrombin formation. The
thrombin formed converts fibrinogen to fibrin, and activates factor
XIII to factor XIIIa, which cross-links the fibrin formed.
Crosslinked fibrin is insoluble.
An activated fibrinolytic system (often as a reaction to the
activation of coagulation) yields active plasmin. Plasmin degrades
the insoluble, cross-linked fibrin to soluble (still cross-linked)
degradation products (FbDP) such as �D-dimer�.
In its most elementary form the haemostatic balance can thus be
described as the equilibrium between fibrin formation and fibrin
degradation. As a result measurements of derivatives of fibrinogen,
such as D-dimer can help in the diagnosis of disturbances in the
haemostatic balance.
Fibrin(ogen) derivatives have been measured for many years by
using a variety of methods (1). Elevated concentrations of FbDPs
have been found in serum of patients with venous thrombo-embolism
and disseminated intravascular coagulation (DIC), but also in the
case of trauma, surgical procedures, infections, malignancies, and
sickle-cell anaemia (2-12). It was postulated that increased FbDP
concentrations confirm the diagnosis of diseases which are
characterized by an activated coagulation system, and that normal
FbDP concentrations exclude such diseases (8,12).
The observation that FbDP concentrations rise during
thrombolytic therapy of acute myocardial infarction patients, and
subsequently decrease again, suggested that the measurement of
FbDPs could be used to monitor the effect of the therapy
(13,14).
In general a test for a disease, which is characterized by an
activated coagulation, is clinically useful when that test excludes
the patients without the disease (specificity) and confirms the
diagnosis in patients with the disease (sensitivity).
However, the lack of specificity and accuracy of serum FbDP
assays has caused a lot of confusion with regard to the value of
FbDPs as markers for haemostatic disturbances. Most FbDP assays
were traditionally performed with serum samples, since the assays
were usually based on polyclonal antibodies which, in plasma, would
cross-react with the huge excess of fibrinogen. As will be
elaborated in the discussion section the use of serum samples is a
source of artifacts. For almost twenty years now, however, FbDP
assays have been available, which are based on monoclonal
antibodies, which do not cross-react with fibrinogen; and those
assays can thus be carried out on plasma samples without those
risks of artifacts. But the modern assays for D-dimer also have
their limitations, which will be discussed below.
Several types of assay are available e.g. semi-quantitative
latex agglutination assays, and quantitative enzyme
immunoassays. Relatively little is known about their use in the
Netherlands.
One purpose of the present study was to investigate which types
of assay are used, and how frequently; by which specialisms and for
which indications; and whether or not the assays are carried out as
cito assays. The impression existed that serum samples are still
being used on a quite considerable scale, notwithstanding the fact
that it has been known for many years (15) that serum can be a
source of erroneous results in assays for fibrin degradation
products (16). A second purpose of this study was to investigate
whether this notion has led to the abandonment of the use of serum
samples.
Design of the study
A questionnaire was sent to 123 addresses in 116 different
clinical centres (some centres have more than one laboratory)
throughout the Netherlands.
The following questions were asked:
-
Do you use as a sample serum, plasma or blood?
-
What is the principle of the assay you use: ELISA, latex
agglutination or an other principle?
-
Is the assay used quantitative or semi-quantitative?
-
Requested by which specialism?
-
For which indication/question is the assay used ?
-
Is the assay used in cito situations or not?
-
What is the approximate number of assays which you performed in
the past month?
Results
The centres were asked to return the completed forms within two
weeks. After some weeks those who had not responded were contacted
by phone.
Of the 123 laboratories approached (in the 116 centres) a total
of 94 responded either via the completed questionnaire or by phone.
Of those who reacted only by phone ten indicated that they did not,
not anymore, or not yet use the assays; five indicated that they
did not have time to respond; the remaining were repeatedly
unreachable by phone. A total of 82 (71% of all centres) completed
forms were received.
The 82 forms showed:
-
76 centres use a semi-quantitative latex agglutination test
-
2 centres use a quantitative ELISA
-
3 centres use a quantitative or semi-quantitative version of a
filtration-type assay
-
1 centre does not use fibrin degradation products assays
-
4 centres use more than one type of assay
Of the 76 centres which perform semi-quantitative latex
agglutination assays, 59 indicated that they use plasma samples;
the remaining 17 use serum. Extrapolation of the number of assays
performed by each centre during the last month indicates that
28000, and 4800 plasma and serum assays, respectively, are
performed per year by using a latex agglutination assay.
The majority (49 of the returned forms) of the centres indicated
the use of cito determinations.
As can be concluded from the above the use of other types of
assays (ELISA, filtration, turbidimetric) is relatively small.
The aforementioned data plus the requesting specialisms and the
indications for which the assays are used are summarized in the
table.
It is obvious that by far most assays are requested by
gynaecologists, internists, intensive care units (ICU) and
cardiologists; and that DIC, pregnancy complications, DVT and
pulmonary embolism are the most important indications.
Discussion
It can be derived from the table that 39000 assays are performed
on a yearly basis. When all types of assays are taken together,
this figure is rather modest. A possible reason for this may be
derived from remarks made on the forms by several of the centres.
One important reason mentioned was the lack of standardization.
Results obtained for the same sample by using different assay
methods may vary by a factor of ten in absolute value. Several
reasons can be mentioned for these differences:
the terms fibrin degradation products, and even more so D-dimer
suggest a molecular uniform identity of the fibrinolysis products
measured. This is misleading. Degradation products are highly
heterogeneous, and this heterogeneity varies from patient to
patient. The enormously long fibrin stands in a blood clot consist
of a very large number of fibrin monomer subunits, which are kept
together by among others crosslinks (iso-peptide bonds). Each
subunit consists of a central E-domain, flanked by two D-domains.
The crosslinks are situated between the D-domains of two
neighboring subunits. Hence a pattern is performed which can be
represented as: (= DED = DED = DED =)n (in which = stands for the
crosslinks).
The plasmin digestion (fibrinolysis) of the fibrin strands is a
random process which occurs via proteolytic cleavage of the
subunits. This leads to the formation of fragments of those
strands. These fragments are soluble, and are composed of a very
variable (even large) number of subunits, which contain the motive
designated as D-dimer, but are not D-dimer. Examples are ED = DED =
DED and ED = DED = DED = DED = D, whereas the term D-dimer is
limited to D = D.
The various assays used are based on different (mostly
monoclonal) antibodies. These antibodies have different
reactivities with the various fibrin degradation products
comprising D-dimers. Each manufacturer uses his own calibrator, and
assigns a concentration to it.
On the basis of the above considerations it will be clear that
standardisation of FbDP assays is not simple. Yet, the situation is
not hopeless. This author has (in his capacity as chairman of the
Fibrinogen Subcommittee of the Scientific and Standardization
Committee (SSC) of the International Society of Thrombosis and
Haemostasis (ISTH)) made a successful attempt at generating a
reference material. Although this is not an international standard,
it enables manufacturers and users to compare the results obtained
with the various assays.
Some users have the impression that the lower the numerical
results obtained with a particular assay for normal plasma samples,
the more specific the particular assay. From what is said above
about calibration, it will be clear that this is a mistake: the
numerical values found are the direct result of the values assigned
by the manufacturers to the calibrators.
It has become clear that the diagnosis of DVT or pulmonary
embolism cannot be made on the basis of a positive D-dimer assay.
This is not surprising considering the fact that fibrin degradation
products can occur in all cases of (even local) fibrin formation.
It has been shown in a number of studies that a negative result can
be used as an exclusion criteria for the occurrence of DVT (16).
However, negative results of latex agglutination assays appear less
suitable as exclusion criteria than normal results of quantitative
ELISAs. These conclusions are mostly based on studies on reasonably
large numbers of clinical samples. A possible reason for the
greater value of ELISAs could be the lower lowest detection limit
of the ELISAs. Recently we came across another possible reason for
the lesser value of latex agglutination assays for the exclusion of
thrombosis. We found (in at least one latex agglutination test kit)
that the batch-to-batch variation of the detection limit is at
least a factor of 2. This variation is not indicated by the
manufacturer in the insert. The consequences of this for a
particular study are obvious, when different batches of such test
kits are used, and it is assumed that the FbDP concentrations at
which they become positive, are identical.
Although in many cases the assays for degradation products have
undoubtedly added to the diagnostic potential, it is quite likely
that the aforementioned points have not helped to implement assays
for fibrin degradation products in the clinical routine.
Finally this:
Serum samples should not be used. As early as 1985 Gaffney (15)
emphasised that the use of serum causes erroneous results and
artifacts (16). Too high or even false-positive results can be
found, when anti-coagulant degradation products are present
(17-19); in the case of an abnormal fibrinogen with a reduced
coagulability ; or when a patient is anticoagulated with heparin.
Another possible cause is the partial lysis of the clot during
serum preparation, not only in hyperfibrinolytic patients,
but even in normals (20).
Also too low or false-negative results are possible. Some
degradation products coagulate (18,21) or adsorb to the clot
(15,20,22).
In the coming years existing assays will be improved and new
improved assays will become available. It is our intention to
repeat this study after a couple of years to monitor whether the
clinical acceptance of fibrin degradation products assays has
increased.
Acknowledgements
The author acknowledges Mrs Jongsma for collecting the data, and
Prof.Dr. P. Brakman, Dr. F. Haverkate and Dr. E. Brommer for
critically reading the manuscript, and their constructive
comments.
References
1. Donati MB.
Assays for fibrinogen/fibrin degradation products in biological
fluids: some methodological aspects. Thromb Diath Haemorrh 1975;
34: 652-662.
2. Ruckley CV,
Das PC, Leitch AG, Donaldson AA, Copland WA, Redpath AT, Scott P,
Cash JD. Serum fibrin/fibrinogen degradation products associated
with postoperative pulmonary embolus and venous thrombosis. Br Med
J 1970; 4: 395-398.
3. Hedner U,
Nilsson IM. Clinical experience with determination of fibrinogen
degradation products. Acta Med Scand 1971; 189: 471-477.
4. Wood EH,
Prentice CRM, McNicol GP. Association of fibrinogen-fibrin- related
antigen (F.R. antigen) with postoperative deep vein thrombosis and
systemic complications. Lancet 1972; i: 166-169.
5. Gallus AS,
Hirsch J, Gent M. Relevance of preoperative and postoperative blood
tests to postoperative leg vein thrombosis. Lancet 1973; ii:
805-809.
6. Gurewich V,
Hume M, Patrick M. The laboratory diagnosis of venous
thromboembolic disease by measurement of fibrinogen/fibrin
degradation products and fibrin monomer. Chest 1973; 64:
585-590.
7. Cooke ED,
Gordon YB, Bowcock SA, Sola CM, Pilcher MF, Chard T, Ibbotson RM,
Ainsworth ME. Serum fibrin(ogen) degradation products in diagnosis
of deep vein thrombosis and pulmonary embolism after hip surgery.
Lancet 1974; ii: 51-54.
8. Clayton JK,
Anderson JA, McNicol GP. Preoperative prediction of postoperative
deep vein thrombosis. Br Med J 1976; 2:910-912.
9. Leslie J,
Langler D, Serjeant G, Serjeant B, Desai P, Gordon Y. Coagulation
changes during the steady state in homozygous sickle-cell disease
in Jamaica. Br J Haematol 1975; 30: 159-166.
10. Bick RL.
Disseminated intravascular coagulation: a clinical/laboratory study
of 48 patients. Ann NY Acad Sci 1981; 370: 843.
11. Kies MS,
Posch JJ, Giolma JP, Rubin RM. Hemostatic function in cancer
patients. Cancer 1980; 46: 831.
12. Tibbut DA,
Chesterman CN, Allington MJ, Williams EW, Faulkner T. Measurement
of fibrinogen-fibrin related antigen in serum as aid to diagnosis
of deep vein thrombosis in outpatients. Br Med J 1975; 43: 493.
13. Gaffney PJ,
Chesterman GN, Allington MJ. Plasma fibrinogen and its fragments
during streptokinase treatment. Br J Haematol 1974; 26:
285-293.
14. Francis CW,
Marder VJ, Barlow GH. Plasmid degradation of cross-linked fibrin.
Biochim Biophys Acta 1973; 295: 308-313.
15. Gaffney PJ,
Perry MJ. Unreliability of current serum fibrin degradation product
(FDP) assays. Thromb Haemostas 1985; 53: 301-302.
16. Kroneman H,
Nieuwenhuizen W, Knot EAR. Monoclonal antibody-based plasma assays
for fibrin(ogen) and derivatives, and their clinical relevance.
Blood Coagul Fibrinol 1990; 1: 91-111.
17. Haverkate F,
Timan G, Nieuwenhuizen W. Anticlotting properties of fragments D
from human fibrinogen and fibrin. Eur J Clin Invest 1979; 9:
253-255.
18.
Nieuwenhuizen W, Gravesen M. Anticoagulant and calcium-binding
properties of high molecular weight derivatives of human
fibrinogen, produced by plasmin (fragments X). Biochim Biophys Acta
1981; 668: 81-88.
19.
Nieuwenhuizen W, Voskuilen M, Hermans J. Anticoagulant and
calcium-binding properties of high molecular weight derivatives of
human fibrinogen (fragments Y). Biochim Biophys Acta 1982; 708:
313-316.
20. Koopman J,
Haverkate F, Koppert PW, Nieuwenhuizen W, Brommer EJP, van der Werf
WGC. New immunoassay for fibrin-fibrinogen degradation products in
plasma using a monoclonal antibody. J Lab Clin Med 1987; 109:
75-84.
21. Marder VJ,
Shulman NR. High molecular weight derivatives of human fibrinogen
produced by plasmin II. Mechanism of their anticoagulant activity.
J Biol Chem 1969; 244: 2120-2124.
22. Niewiarowski
S, Stewart GJ, Marder VJ. Formation of highly ordered polymers of
fibrinogen and fibrin degradation products. Biochim Biophys Acta
1970; 221: 326-341.
Correspondence should be directed to:
W. Nieuwenhuizen, Ph.D.,
TNO-PG Gaubius Laboratory,
P.O. Box 2215, 2301
CE Leiden, The Netherlands
|