Haemophilia (2008), 14 (Suppl.
3), 76–82
Diagnosis of factor VIII deficiency
B. VERBRUGGEN,* P. MEIJER, I. NOVÁKOVA,* and W. VAN HEERDE*
*Radboud University Nijmegen Medical Centre, Central Laboratory for Haematology, Nijmegen; and Foundation of the
European Concerted Action on Thrombophilia (ECAT), Leiden, the Netherlands
Summary. The correct diagnosis of factor VIII defi-
ciency and the assessment of severity of the disease
are essential for a patient-tailored treatment strategy.
An optimal diagnostic procedure comprises sensitive
and specific screening methods and factor VIII
activity assays. Different screening reagents show
variable characteristics and receiver operator char-
acteristic curves are presented showing the relation
between sensitivity and specificity of eleven activated
partial thromboplastin time reagents. The details of
Introduction
Factor VIII (FVIII) deficiency, a laboratory-based
diagnosis, includes all the clinical situations where
FVIII clotting activity (FVIII:C) is lower than the
normal values and may lead to clinical bleeding
complications.
An FVIII:C assay has to be performed when a
patient is suspected of having haemophilia or being
carrier of the disease or when a screening clotting test
is prolonged. Assays to monitor treatment and to
establish concentrate activities have recently and
extensively been reviewed [1,2] and are beyond the
scope of this publication and will not be further
discussed.
The correct diagnosis of FVIII deficiency is a
prerequisite for a patient-tailored treatment strategy.
It is extremely important to discriminate between
severe (FVIII:C < 1 IU dL)1), moderate (FVIII:C
between 1 and 5 IU dL)1) and mild haemophilia
(FVIII:C between 5 and 25 IU dL)1). Besides, the
correct identification of patients with sub-haemo-
Correspondence: B. Verbruggen, Radboud University Nijmegen
Medical Centre, Central Laboratory for Haematology, Nijmegen,
the Netherlands.
Tel.: +31 24 3614796; fax: +31 24 3568408;
e-mail: H.verbnuggen@chl.uncm.nl
The authors stated that they had no interests which might be
perceived as posing a conflict or bias.
Accepted after revision 15 February 2008
76
the three methods for factor VIII activity assay, one-
stage and two-stage assay and chromogenic assays,
are discussed. The chromogenic assay seems to be
more sensitive than the one-stage assay with regard
to the detection of severe haemophilia. Discrepant
results obtained with one-stage and two-stage assays
are reviewed and discussed.
Keywords: assay, deficiency, factor VIII, laboratory
philia and carriers of haemophilia is important as
these patients need to be protected against bleeding
complications during surgical interventions [3]. In
this review, the practical problems that arise with the
laboratory diagnosis of FVIII deficiency will be
discussed.
Screening tests for detection of FVIII:C
deficiency
Activated Partial Thromboplastin Time (aPTT)
While hereditary haemophilia is detected by direct
measurement of FVIII:C, non-hereditary haemo-
philia (especially the mild forms in de novo muta-
tions) will mostly be identified by a screening test for
bleeding problems using aPTT assays. Nowadays,
more than 35 aPTT reagents are available but only a
few publications are available that investigate the
sensitivity and specificity, to detect FVIII deficiency,
of these reagents [4–6].
We evaluated eleven aPTT reagents that were most
frequently used in external quality surveys for their
potential to detect FVIII deficiency: Cephascreen,
aPTT Kaolin, PTT-LA, PTTa and PTT Reagent, all
from Stago; Synthasil and aPTT-SP from IL; Actin
FSL, Actin FS and Pathromtin SL from Dade Behring
(Siemens); Platelin LS from Biomerieux (Trinity
Biotech). In total, 30 samples were analysed, 15
normal samples and 15 samples from patients with
haemophilia As FVIII:C was between 20 and
 2008 The Authors
Journal compilation  2008 Blackwell Publishing Ltd
 DIAGNOSIS OF FVIII DEFICIENCY 77

(a) Cephascreen Platelin * SL test does not have any diagnostic values at all. All
tests have a reasonable sensitivity and specificity for
1.00
30 IU dL)1 FVIII:C activity (Fig. 1a). However, a
0.90 number of tests lack sensitivity and/or specificity for
0.80 detecting 50 IU dL)1 FVIII:C activity (Fig. 1b). A
number of reagents show batch variability to some
0.70
degree which means that one cannot fully rely on the
Sensitivity

0.60 data of a single batch.

0.50 Besides these characteristics of the test, the pre-
analytical treatment of samples and the quality of
0.40
liquid handling within the laboratory are determi-
0.30 nants of quality of testing. Guidelines for optimal
0.20 performance are provided by the Clinical and Lab-
oratory Standards Institute (CLSI, Wayne, USA) in
0.10
the H21-A5 protocol describing collection, trans-
0.00 port, and processing of blood specimens for testing
0.00 0.20 0.40 0.60 0.80 1.00 plasma-based coagulation assays and molecular
1-specificity haemostasis.
As most tests are sensitive for low FVIII levels and
(b) Cephascreen Kaolin as (mild) FVIII deficiencies are the most common
1.00 abnormality of the classical coagulation pathway,
FVIII has to be quantified whenever an abnormal
0.90
aPTT is found.
0.80
0.70 Thrombin generation test and thromboelastography
Sensitivity

0.60
0.50
0.40
0.30
0.20
0.10
0.00
0.00 0.20 0.40 0.60 0.80 1.00
1-specificity
Fig. 1. (a) Receiver operator characteristics (ROC) curves for the
ability of the reagents to detect 30 IU dL)1 (1a) and 50 IU dL)1
(1b) FVIII:C activities. On the Y-axis the sensitivity and on the
X-axis 1-specificity. (b) Every individual point on a curve repre-
sents the correlation between sensitivity and specificity at a certain
cut-off value of the reagent. Only the extreme curves are shown.
60 IU dL)1, with a proven gene mutation. Receiver
operator characteristics (ROC) curves were con-
structed (Fig 1a,b, only the extremes are shown) for
the ability of the reagents to detect 30 IU dL)1 and
50 IU dL)1 FVIII:C activity representing mild and
borderline FVIII deficiency, respectively. The more a
curve is located to the left upper edge of the diagram
the better are both the sensitivity and specificity. A
diagonal line from (0;0) to (1;1) indicates that the
The thrombin generation time (TGT) was originally
described by Macfarlane and Biggs [7] for whole
blood and by Pitney and Dacy [8] for plasma. Since
then, a vast number of variants have been pub-
lished, all differing in detail from one another. In
the more modern versions of the test, chromogenic
and fluorogenic substrates are used in the reaction
mixture allowing continuous monitoring of throm-
bin generation. The test has recently been reviewed
by Barrowcliffe [2]. The TGT seems of limited value
for the diagnosis of haemophilia A because of the
low sensitivity and specificity especially for moder-
ate and mild FVIII deficiency. However, the test
may be useful to discriminate between different
phenotypes within the population of severe haemo-
philia A patients [9–11]. In addition, the test may
be useful as a global haemostasis test to monitor
haemophilia A patients, who are being treated with
FVIII products, to predict the outcome of treatment
[12,13].
Thromboelastography [14] and other types of
waveform analyses [15] are whole blood assays that
measure speed of clot formation and strength of the
clot. The assays are started by addition of calcium
and/or tissue factor or celite to citrated blood
samples. The assays show variable results for fac-
tor-deficient plasmas [16], lack specificity but may be
useful for monitoring of treatment of haemophilia A

 2008 The Authors

Journal compilation  2008 Blackwell Publishing Ltd Haemophilia (2008), 14 (Suppl. 3), 76–82
78 B. VERBRUGGEN et al.
patients. The methods have recently been reviewed
extensively [2].
FVIII assays
FVIII concentrations can be quantified by bio-
assays measuring activity and immunological assays
measuring cross-reacting material (CRM). There are
three bioassays available: (i) one-stage clotting assay,
(ii) two-stage clotting assay and (iii) assay with
chromogenic substrates.
The most widely used immunological assay now-
adays is the ELISA-based method. We analysed FVIII
CRM with ELISA and FVIII:C activity with the one-
stage assay in 60 patients with mild to moderate
haemophilia A (Fig. 2). There is a big variation in
ratio between activity and CRM and we concluded
that the CRM assay does not contribute to the
classification of haemophilia A patients.
Two-stage FVIII:C activity assay
The two-stage FVIII:C activity assay was developed
by Biggs in 1955 [17]. In this test, serum of a normal
donor and diluted patient plasma are incubated with
phospholipids and calcium, which results in the
formation of prothrombinase. This is proportional to
the FVIII concentration and can be quantified by sub
sampling into normal plasma and measuring the
clotting time. The details and limitations of the test
have recently been reviewed extensively [2]. The
160
120
FVIII CRM (IU/dl)
80
40
0 Reference plasma
0 40 80
FVIII:C activity (IU/dL)
Fig. 2. The relationship between the data of the FVIII assay of
cross reactive material (CRM) with ELISA and the data of FVIII:C
assay with the one-stage assay in plasma of 60 hemophilia A pa-
tients with a missense mutation in the FVIII gene. The curve from
(0;0) to (160;160) represents the line of identity.
Haemophilia (2008), 14 (Suppl. 3), 76–82
main advantage of the method is that it does not need
expensive reagents (e.g. FVIII deficient plasma or
chromogenic substrates) and that it is not influenced
by pre-activated coagulation factors in the sample.
However, no commercial reagents are available for
this assay and it is difficult to automate. At present,
the two-stage clotting assay is largely replaced by the
chromogenic assay that has more or less the same
characteristics.
One-stage FVIII:C activity assay
The one-stage FVIII:C assay is the most widely used
method for the determination of FVIII clotting activ-
ity and detection of FVIII:C deficiency. The test was
first described by Langell in 1953 [18]. Nowadays, it
is performed identically to the original method and
has been extensively reviewed by Over [19] and more
recently by Mannucci [20]. The test is based upon the
ability of a highly diluted patient sample to shorten
the clotting time in a medium that additionally
consists of FVIII deficient plasma, an aPTT reagent
(phospholipids and a negatively charged surface or
liquid contact activator) and calcium. Because all
clotting factors are present in excess of FVIII, the
clotting time of the mixture is mainly affected by the
FVIII:C activity resulting in a linear relation between
both clotting times (linear) and standard dilutions
(logarithmic). An important prerequisite for this assay
is parallelism of these curves between standard
dilutions and dilutions of patient plasma. Non-paral-
lelism indicates the presence of inhibiting or otherwise
disturbing factors and the results of these assays are
not appropriate. Consequently, FVIII:C activity
assays always have to be performed in at least three
different dilutions to check parallelism. However, this
is a time-consuming and expensive procedure. As a
practical alternative in our laboratory, we assay more
than one dilution in only previously unknown
patients with FVIII:C < 50 IU dL)1.
A number of variables can influence the assay and
may contribute to an undesirably high intra- and
inter-laboratory coefficient of variation and/or de-
creased accuracy of the results. These items will
shortly be discussed.
120 160
In all types of assays (one- and two-stage assays and
chromogenic assay), the FVIII:C activity of patient
plasma is read from a calibration curve of reference
plasma representing the correlation between clot-
ting time or extinction (linear) and FVIII:C activ-
ity (logarithmic). An absolute requirement in the
 2008 The Authors
Journal compilation  2008 Blackwell Publishing Ltd

performance of the FVIII:C assay is the identical
composition of the type of sample, the type of FVIII
(native or recombinant) and citrate concentration in
reference and patient plasma. (the so-called Ôlike vs.
likeÕ principle). The FVIII:C activity of a given
plasma is expressed in IU dL)1, where 100 IU is
arbitrarily defined as the activity that is present in
1 dL of pooled plasma. To be sure of potency, as
close as possible to 100 IU dL)1, at least 50 healthy
donors aged 18–65 years have to be included,
selected to represent all blood groups and both sexes
equally. Aberrations in the potency of the pool will
result in over- or underestimation of the FVIII levels.
It is strongly advised to calibrate the reference
plasma against the international standard for FVIII
of the World Health Organization (WHO) that can
be obtained from the National Institute for Biolog-
ical Standards and Control (Potters Bar, UK). The
calibration is performed by parallel measurement of
WHO standard and local standard in at least four
dilutions. The potency of the local standard can be
derived from these assays when both curves show
parallelism [20].
While final dilutions of patient plasma samples and
standard dilutions can be made with a buffer
solution, predilutions of the standard sample have
to be made with FVIII deficient plasma [21]. Stan-
dard samples, prediluted with buffer, will give a
matrix in the assay medium that will differ from that
of the patient sample. This will result in prolonged
clotting times of the prediluted low activity standard
samples and therefore FVIII:C activity results in the
patient sample will be falsely high.
Factor VIII-deficient plasma
An important feature of the FVIII:C assay is the
quality of the FVIII-deficient plasma [22]. In the past,
congenital-deficient plasmas derived from severely
affected patients were mostly used in the assays.
Nowadays, artificially depleted plasmas (immunode-
pleted, chemically depleted) are almost exclusively
used as substrate plasma.
The most important condition for deficient sub-
strate plasma is the absolute absence of FVIII.
However, in our experience, part of the commer-
cially available immunodepleted FVIII-deficient plas-
mas is not fully deficient. As a result, the reference
curve will flatten in the range 0–3 IU dL)1 FVIII:C
activity and FVIII:C activity in this window cannot
be determined reliably and so severe haemophilia A
(FVIII:C activity<1 IU dL)1) cannot be diagnosed. It
is strongly advised to check the residual FVIII:C
activity of FVIII-deficient plasma and reject all
 2008 The Authors
Journal compilation  2008 Blackwell Publishing Ltd
DIAGNOSIS OF FVIII DEFICIENCY 79
batches with FVIII:C residual activity equal to or
more than 1 IU dL)1.
A second requirement is the excess of all factors
except FVIII. Shortage of one of the other clotting
factors may influence the FVIII:C activity data and
may lead to non-parallelism between reference and
patient plasma. Concerning this aspect, a certificate
of quality from the manufacturer or provider of the
deficient plasma may be very helpful. Otherwise, one
has to check every new batch of substrate plasma for
these requirements.
As von Willebrand Factor (VWF) is the natural
carrier protein of FVIII, the presence of VWF in
FVIII-deficient plasma is of importance. In FVIII:C
assays, markedly lower activities were found when
using VWF-depleted-deficient plasma in comparison
with VWF-containing plasmas [22].
aPTT reagent
Limited data comparing the sensitivity, reproducibil-
ity and accuracy of various aPTT reagents used in the
FVIII assay are available. Cinotti et al. [21] used five
different reagents (four from DADE and one from
IL) in highly standardized test procedures to analyse
three different plasma samples with 0.9, 8.6 and
43 IU FVIIIdL)1. Good correlation was found
between different reagents for the two highest
concentrations but results were more variable for
the samples with very low activity. The problems in
the very low range of FVIII levels were also clear
from a survey by the United Kingdom National
External Quality Assessment Service (UKNEQUAS),
where approximately 30% of the participating lab-
oratories obtained results in the range between 1 and
3 IU dL)1 with plasma of a severe haemophiliac
[23]. These variations were of course not only due to
the reagent variations but were the sum of all
possible errors. To clarify the role of the type of
reagent in identifying severe haemophiliacs, we
analysed five samples from severe haemophiliacs
and five samples from moderate–severe haemophil-
iacs with eleven different reagents that were most
frequently used in the FVIII:C assays. The results
(Table 1) show that 2/11 reagents (reagents 1 and 9)
were not able to detect FVIII:C activity <1 IU dL)1
indicating that these reagents cannot be used to
detect severe haemophilia. For the samples from the
moderate–severe patients, there was a ratio of about
3 between the lowest and the highest result of the
different reagents. From these figures, it can be
concluded that the diagnosis of severe haemophilia
with the one-stage assay may be very difficult unless
an appropriate aPTT reagent is used and the other
Haemophilia (2008), 14 (Suppl. 3), 76–82
80 B. VERBRUGGEN et al.

Table 1. FVIII:C activity assay with one-stage method of five samples from severe haemophilia A patients (1, 5, 7, 8 and10) and five
samples from moderate haemophilia A patients (2, 3, 4, 6 and 9). Eleven different aPTT reagents were used in the FVIII:C assays.
Reagents

Sample Mutation gene type 1 2 3 4 5 6 7 8 9 10 11

1 inv. Intron 1 1.4 0.4 0.4 0.3 0.9 0.1 0.3 0.8 1.4 0.3 0.5
2 ex10 Tyr 476 stop C fi G 6.4 4.1 5.1 4.3 5.4 3.8 4.6 4.6 6.8 5.2 5.8
3 ex7 Hs256Arg A fi G 2.5 1.4 1.5 1.4 2.1 1.2 1.3 1.9 3.6 1.7 1.8
4 ex14 Asp121461u C fi G 5.4 3.9 4.5 3.7 5.0 3.8 4.1 4.6 8.3 5.0 4.9
5 inv. Intron 22 1.3 0.4 0.4 0.4 0.9 0.1 0.2 0.8 2.0 0.4 0.4
6 ex23 Arg2163Cys C fi T 1.8 0.8 0.9 0.9 1.3 0.5 0.8 1.5 3.0 1.0 1.0
7 ex14 Arg795stop C fi T nd 0.4 0.7 0.3 0.6 0.1 0.3 0.9 1.3 nd 0.5
8 inv. Intron 1 1.7 1.0 1.0 0.8 0.9 0.5 0.6 0.9 1.8 0.7 0.9
9 ex7 His256Arg A fi G 2.8 1.7 2.0 1.3 3.0 1.6 1.9 2.9 1.9 2.1 2.6
10 inv. Intron 22+/) 2.1 1.1 0.4 1.0 1.9 1.0 1.3 2.0 1.6 1.2 1.6
The results of the assays are expressed as IU dL)1.
Reagents: 1, STA Cephascreen (Stago); 2, STA APTT Kaolin(Stago); 3, PTT-LA (Stago); 4, PTT a (Stago); 5, PTT Reagent (Stago); 6,
Synthasil (HemosIL); 7, APTT-SP (HemosIL); 8, Dade Actin FSL Activated PTT Reagent (Dade Behring); 9, Dade Actin FS Activated
PTT Reagent (Dade Behring); 10, Pathromtin SL (Dade Behring); 11, Platelin LS (Biomerieux); NO, Not Determined.

test requirements are optimally met. To our under-
standing, it may be better to diagnose severe
haemophilia with a chromogenic assay.
Chromogenic assay
The chromogenic assay of FVIII:C is recommended
by the FVIII and FIX subcommittee of the Scientific
and Standardization Committee (SSC) of the Inter-
national Society of Thrombosis and Haemostasis
(ISTH) as a reference method for FVIII:C activity
assay both in plasma and in FVIII concentrates. The
first report on chromogenic FVIII assay was pub-
lished in 1975 [24] and further elaborated in the
following decade [25]. The reaction mixture consists
of purified activated factor IX (FIXa), purified factor
X (FX), phospholipids, calcium, a chromogenic
substrate sensitive for activated factor X (FXa),
thrombin to activate FVIII and a highly diluted
reference plasma or patient plasma as a FVIII source.
FX is activated by FIXa, and this step is accelerated
by FVIIIa, approximately 100 000-fold [26]. FXa
hydrolyses an FXa-specific chromogenic substrate,
resulting in the release of the chromophoric group
para-Nitro Aniline (pNA). The extinction can be
read in a photometer at 405 nm and is directly
proportional to the FVIII concentration. The method
can be easily executed with automatic analysers.
The results obtained with the amidolytic assay and
the one-stage clotting assay highly correlate with
each other [27,28]. No significant differences in the
results between the two methods were detected in the
blood of donors with normal or high FVIII levels, in
patients with a chronic liver disease and in normal
persons with high FVIII levels after treatment with
desmopressin. However, in about 1/3 of the patients
with mild haemophilia caused by a single-point
mutation, a discrepancy is described between the
one-stage and two-stage assay (both the two-stage
clotting assay and chromogenic assay) [29]. Usually,
the activity with the chromogenic assay is lower
compared with the results of the one-stage assay and
at least 17 different mutations have been identified so
far with discrepant assay data. Most mutations
appear in the interface between the subunits of the
FVIII molecule [29]. The discrepancy can, at least
partly, be explained by increased instability of the
mutated molecule. In the chromogenic assay, this
renders lower FVIII:C activity because of the longer
incubation time as compared with that of the one-
stage assay. Recently, mutations have been described
which result in the converse discrepancy [30,31].
The chromogenic method has some advantages
over the one-stage clotting assay. The assay is not
influenced by the presence of lupus anticoagulants
and thus can be used to distinguish between FVIII
inhibitors and lupus anticoagulants in patients with
acquired deficiency of FVIII [32]. Additionally, the
assay is not influenced by clotting-specific reagents
like deficient plasma and aPTT reagent, or by pre-
activation of FVIII and other clotting factors and has
a lower analytical variation and is more accurate
especially in the lower range [27]. A method based
on commercially available reagents has been recently
described that claims a detection limit of
0.15 IU dL)1 [33].
At our laboratory, a method has been developed
for improved discrimination between severe and

 2008 The Authors

Haemophilia (2008), 14 (Suppl. 3), 76–82 Journal compilation  2008 Blackwell Publishing Ltd

Fig. 3. The relationship of the Factor VIII:C activity and the
between-laboratory variation (CV) in the ECAT external quality
assessment programme (n = 150–200) for the period 2005–2007.
moderate FVIII levels using chromogenic assays in
samples concentrated by selective protein filtration.
In 43 samples of haemophilia A patients with
clinically severe phenotype (0–2 IU dL)1 FVIII:C
with one-stage assay), FVIII:C was measured with
chromogenic methods before and after concentra-
tion. All samples were concentrated at least 2.5
times. In 31 samples, FVIII levels were <1 IU dL)1
both before and after concentration. In four samples,
the levels increased from <1 IU dL)1 before to 1–
2 IU dL)1 after concentration. Both groups of
patients thus appeared to be true severe haemophilia
A patients. Eight samples had a moderate level of 1–
2 IU dL)1; five of these were <1 IU dL)1 after
concentration indicating a true severe plasma level
and three samples were >3 IU dL)1 after concentra-
tion suggesting an original moderate plasma level.
Quality control
To ensure the reliability of the laboratory test results,
it is important to have an appropriate internal and
external quality control system. Data from external
quality assessment programmes show a considerable
between-laboratory variation for FVIII:C testing
[34,35]. From the external assessment programme
of the ECAT Foundation, we have evaluated the
between-laboratory variation for the period 2005–
2007 (Fig. 3). This evaluation shows a J-shape
pattern with the lowest variation (10–15%) in the
reference range, and a strong increase of the varia-
tion at FVIII:C levels below 20 IU dL)1. Also at high
FVIII:C levels, there is a tendency to higher variation.
A variety of factors may influence the between-
laboratory variation, such as calibration, source of
deficient plasma, type of activator and the equipment
 2008 The Authors
Journal compilation  2008 Blackwell Publishing Ltd
DIAGNOSIS OF FVIII DEFICIENCY 81
used [34,36]. It shows the need for further standard-
ization as well as the need for critical evaluation of
the performance of the local test system.
Conclusion
Each step in the assessment of FVIII deficiency
(screening, diagnosis, classification) has its specific
conditions and limitations and requires a method
that best fits the specific demands concerning sensi-
tivity, specificity, accuracy and reproducibility. One-
stage and chromogenic FVIII:C assays mostly show
good correlation but clinically important discrepan-
cies are observed in mild haemophilia patients with
missense mutations. Very low FVIII:C activities can
best be diagnosed by chromogenic assays.
References
1 Barrowcliffe TW. Standardization of FVIII & FIX
assays. Haemophilia 2003; 9: 397–402 (Review).
2 Barrowcliffe TW. Monitoring haemophilia severity and
treatment: new or old laboratory tests? Haemophilia
2004; 10(Suppl. 4): 109–14 (Review).
3 Plug I, Mauser-Bunschoten EP, Bröcker-Vriends AH
et al. Bleeding in carriers of hemophilia. Blood 2006;
108: 52–6.
4 Lawrie AS, Kitchen S, Purdy G, Mackie IJ, Preston FE,
Machin SJ. Assessment of Actin FS and Actin FSL
sensitivity to specific clotting factor deficiencies. Clin
Lab Haematol 1998; 20: 179–86.
5 Shetty S, Ghosh K, Mohanty D. Comparison of four
commercially available activated partial thrombo-
plastin time reagents using a semi-automated coagu-
lometer. Blood Coagul Fibrinolysis 2003; 14: 493–7.
6 Hales SC, Johnson GS, Wagner D. Comparison of six
activated partial thromboplastin time reagents: intrin-
sic system factorsÕ sensitivity and responsiveness. Clin
Lab Sci 1990; 3: 194–6.
7 Macfarlane RG, Biggs R. A thrombin generation test;
the application in haemophilia and thrombocytopenia.
J Clin Pathol 1953; 6: 3–8.
8 Pitney WR, Dacie JV. A simple method of studying the
generation of thrombin in recalcified plasma; applica-
tion in the investigation of haemophilia. J Clin Pathol
1953; 6: 9–14.
9 Beltrán-Miranda CP, Khan A, Jaloma-Cruz AR, Laffan
MA. Thrombin generation and phenotypic correlation
in haemophilia A. Haemophilia 2005; 11: 326–34.
10 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.
11 Van den Berg HM, De Groot PH, Fischer K. Pheno-
typic heterogeneity in severe hemophilia. J Thromb
Haemost 2007; 5(Suppl. 1): 151–6.
Haemophilia (2008), 14 (Suppl. 3), 76–82
82 B. VERBRUGGEN et al.
12 Lewis SJ, Stephens E, Florou G et al. Measurement of
global haemostasis in severe haemophilia A following
factor VIII infusion. Br J Haematol 2007; 138: 775–
82.
13 McIntosh JH, Owens D, Lee CA, Raut S, Barrowcliffe
TW. A modified thrombin generation test for the
measurement of factor VIII concentrates. J Thromb
Haemost 2003; 1: 1005–11.
14 Hartert H. Das Thromboelastogramm. Verhandlungen
der Deutschen Gessellschaft für innere Medizin 1952;
58: 562.
15 Ingerslev J, Poulsen LH, Sørensen B. Potential role of
the dynamic properties of whole blood coagulation in
assessment of dosage requirements in haemophilia.
Haemophilia 2003; 9: 348–52.
16 Nielsen VG, Cohen BM, Cohen E. Effects of coagula-
tion factor deficiency on plasma coagulation kinetics
determined via thrombelastography: critical roles of
fibrinogen and factors II, VII, X and XII. Acta Anaes-
thesiol Scand 2005; 49: 222–31.
17 Biggs R, Eveling J, Richards G. The assay of anti-
haemophilic-globulin activity. Br J Haematol 1955; 1:
20–34.
18 Langdell RD, Wagner RH, Brinkhous KM. Effect of
antihemophilic factor on one-stage clotting tests; a
presumptive test for hemophilia and a simple one-stage
antihemophilic factor assay procedure. J Lab Clin Med
1953; 41: 637–47.
19 Over J. Methodology of the one-stage assay of Factor
VIII (VIII:C). Scand J Haematol Suppl 1984; 41: 13–
24.
20 Mannucci PM, Tripodi A. Factor VIII Clotting Activ-
ity. Laboratory Techniques in Thrombosis. A Manual.
2nd revised edition of ECAT Assay Procedures 1999:
107–13.
21 Cinotti S, Paladino E, Morfini M. Accuracy of FVIII: C
assay by one-stage method can be improved using
hemophilic plasma as diluent. J Thromb Haemost
2006; 4: 828–33.
22 Barrowcliffe TW, Tubbs JE, Wong MY. Evaluation of
factor VIII deficient plasmas. Thromb Haemost 1993;
70: 433–7.
23 Preston FE, Kitchen S, Jennings I, Woods TA, Makris
M. SSC/ISTH classification of hemophilia A: can
hemophilia center laboratories achieve the new crite-
ria? J Thromb Haemost 2004; 2: 271–4.
Haemophilia (2008), 14 (Suppl. 3), 76–82
24 Seghatchian MJ, Miller-Anderson M, Howarth DJ,
Stirling Y, Brozovic M. Proceedings: automated tech-
niques for estimating factor VIII levels in concentrate
preparations. Thromb Diath Haemorrh 1975; 34: 611.
25 Rosén S. Assay of factor VIII:C with a chromogenic
substrate. Scand J Haematol Suppl 1984; 40: 139–45.
26 Fay PJ. Activation of factor VIII and mechanisms of
cofactor action. Blood Rev 2004; 18: 1–15 (Review).
27 Tripodi A, Mannucci PM. Factor VIII activity as
measured by an amidolytic assay compared with a one-
stage clotting assay. Am J Clin Pathol 1986; 86: 341–4.
28 Chandler WL, Ferrell C, Lee J, Tun T, Kha H. Com-
parison of three methods for measuring factor VIII
levels in plasma. Am J Clin Pathol 2003; 120: 34–9.
29 Rodgers SE, Duncan EM, Barbulescu DM, Quinn DM,
Lloyd JV. In vitro kinetics of factor VIII activity in
patients with mild haemophilia A and a discrepancy
between one-stage and two-stage factor VIII assay re-
sults. Br J Haematol 2007; 136: 138–45.
30 Lyall H, Hill M, Westby J, Grimley C, Dolan G.
Tyr346 – >Cys mutation results in factor VIII:C assay
discrepancy and a normal bleeding phenotype – is this
mild haemophilia A? Haemophilia 2008; 14: 78–80.
31 Trossaërt M, Regnault V, Sigaud M, Boisseau P,
Fressinaud E, Lecompte T. Mild haemophilia a with
factor VIII assay discrepancy: using thrombin genera-
tion assay to assess. J Thromb Haemost 2007; 6: 486–
493.
32 Blanco AN, Alcira Peirano A, Grosso SH, Gennari LC,
Pérez Bianco R, Lazzari MA. A chromogenic substrate
method for detecting and titrating anti-factor VIII
antibodies in the presence of lupus anticoagulant.
Haematologica 2002; 87: 271–8.
33 Yatuv R, Dayan I, Baru M. A modified chromogenic
assay for the measurement of very low levels of factor
VIII activity (FVIII:C). Haemophilia 2006; 12: 253–7.
34 Arkin CF, Bovill EG, Brandt JT, Rock WA, Triplett
DA. Factors affecting the performance of factor VIII
coagulant activity assays. Results of proficiency surveys
of the College of American Pathologists. Arch Pathol
Lab Med 1992; 116: 908–15.
35 Preston FE, Kitchen S. Quality control and factor VIII
assays. Haemophilia 1998; 4: 651–3.
36 Brandt JT, Triplett DA, Musgrave K et al. Factor VIII
assays. Assessment of variables. Arch Pathol Lab Med
1988; 112: 7–12.
 2008 The Authors
Journal compilation  2008 Blackwell Publishing Ltd