Laboratory Tests for VWF Level and Activity
Before embarking on this rich history perspective, it is important to first overview
the tests that are used to diagnose and type VWD. The greater the number of such tests
that are performed and the better the tests that are performed, the more accurate
the diagnosis and characterization of VWD. The most common tests are also summarized
in [Table 1 ].
Table 1
A summary of tests used to diagnose VWD
Test
Abbreviation
Utility in VWD
Comments
Factor VIII coagulant activity
FVIII:C
VWD is due to loss of VWF or activity. Since VWF is a carrier of FVIII, and protects
FVIII from proteolysis, a reduction in VWF or its activity is accompanied by a reduction
in FVIII and its activity. FVIII:C is also proportionally low in type 2N VWD, where
VWF:FVIII binding is impaired
Essential test for VWD diagnostics
VWF antigen
VWF:Ag
Measure of the level of VWF protein
Essential test for VWD diagnostics
VWF glycoprotein Ib binding activity
VWF:GPIbB
Generic term for an assay able to assess the VWF activity related to binding its platelet
receptor, namely, glycoprotein Ib (GPIb). Can be performed in several ways (see below)
Essential test for VWD diagnostics
VWF ristocetin cofactor
VWF:RCo
One of the original VWF:GPIbB assays, measuring ristocetin-aided platelet agglutination
These assays are largely interchangeable; all laboratories should perform at least
one of these assays for the diagnosis of VWD
VWF GPIb binding using recombinant GPIb
VWF:GPIbR
A second-generation VWF:GPIbB assay that uses recombinant GPIb coupled to an inert
particle (e.g., latex or magnetic beads) to measure ristocetin-aided particle agglutination
or binding
VWF GPIb binding using recombinant mutated GPIb
VWF:GPIbM
A third-generation VWF:GPIbB assay that uses recombinant GPIb with gain-of-function
mutation(s) coupled to an inert particle (e.g., latex or plastic well) to measure
mutation-aided particle agglutination or binding
VWF collagen binding activity
VWF:CB
An assay that measures the ability of VWF to bind to collagen (which is a subendothelial
matrix protein)
Essential test for VWD diagnostics, especially to correctly identify types 2A, 2B,
2M VWD
VWF factor VIII binding activity
VWF:FVIIIB
An assay that measures the ability of VWF to bind to FVIII
An important assay in VWD diagnostics, especially for correctly identifying type 2N
(including discriminating 2N VWD from hemophilia A)
Ristocetin-induced platelet aggregation/agglutination
RIPA
An agglutination assay performed to identify/exclude gain-of-function mutations in
VWF (i.e., 2B VWD) or GPIb (i.e., platelet type [PT] VWD). Performed as part of platelet
function studies using a platelet aggregometer
An important assay in VWD diagnostics, especially for correctly identifying or excluding
type 2B or PT-VWD
VWF multimers
VWF:mult
VWF multimers can be separated according to size using agarose gels; with particular
agarose gel, the individual bands of multimers of the same size can be visualized
An important assay in VWD diagnostics, especially when laboratories use only a three-test
panel. Can often be omitted in laboratories using a four-test panel including an optimized
VWF:CB
Abbreviations: GPIb, (platelet) glycoprotein Ib; VWD, von Willebrand disease; VWF,
von Willebrand factor.
Factor VIII
First, it is important to determine the FVIII level in any patient being investigated
for VWD. The FVIII level is typically assessed using a FVIII activity assay, for which
there are two main options—a chromogenic assay and a clot-based “coagulant” assay
(FVIII:C).[7 ]
[8 ] The clot-based “coagulant” assay (FVIII:C) is easier to perform, and easier to automate,
and thus more often performed than the chromogenic assay, and is also sufficient for
investigation of VWD. Essentially, this assay can be performed on any automated hemostasis
analyzer enabling clot detection, provided there are reagents available to perform
the assay on said analyzer. FVIII:C levels are reported in percent of normal (%) or
in units/mL (U/mL) or in units/dL (U/dL), with U/dL and % reflecting identical values.
If the assay calibrant is linked to an international standard, the units may be reported
in international units (IU). The normal reference range for FVIII:C should be determined
for the specific method in use in any given laboratory, but would approximate 50 to
200 U/dL (or %), which is the actual range in use in our laboratory.
von Willebrand Factor “Level”
Second, the plasma level (or “quantity”) of VWF should be assessed, with this most
typically performed using an immunological assay employing antibodies against VWF
to capture and quantify VWF.[3 ]
[4 ]
[9 ] Since VWF in these assays reflects the “antigen” against which the antibody was
developed and is reactive against, these assays are called VWF antigen (VWF:Ag) assays.
These are several methodological options, with Laurel gel rocket (or electroimmunodiffusion;
EID) and radioimmunoassay (using radioisotopic detection) representing historical
methods, but enzyme-linked immunosorbent assays (ELISA) and latex-immuno assay (LIA)
representing the main contemporary options. For readership interest, and since this
is a historical article, an example of a Laurel gel rocket VWF:Ag assay performed
in our laboratory in the mid-1980s is shown in [Fig. 1 ]. The LIA method can essentially be performed on any automated hemostasis analyzer
that is enabled with turbidimetric detection, provided there are reagents available
to perform the assay on said instrument. A newer methodology based on chemiluminescence
immunoassay (“CLIA”) is also available in some locations with certain instrumentation,
as performed on an AcuStar instrument (Werfen). Levels of VWF:Ag are reported in the
same units as used for FVIII:C. The normal reference range for VWF:Ag should be determined
for the specific method in use in any given laboratory, but would again approximate
50 to 200 U/dL (or %), which is the actual range in use in our laboratory.
Fig. 1 A Laurel gel rocket (electro-immunodiffusion; EID) assay performed for von Willebrand
factor antigen (VWF:Ag) in 1986. Shown at top is the gel, with rockets migrating upward.
The level of plasma VWF:Ag is related to the height of the rocket. A calibration curve
was hand drawn from the rocket heights of a series of dilutions for pool normal plasma
(PNP), and patient values extrapolated from this curve. Since the patient values were
unknown, it was usual to perform tests at two dilution points to ensure at least one
point on the curve; the results could then be averaged, or in the case where values
were very high or very low, then the value fitting within the curve would be taken.
von Willebrand Factor Glycoprotein Ib Binding Activity
Third, the activity of VWF that measures platelet GPIb binding should be assessed.
Historically, this was assessed using an assay called the VWF ristocetin cofactor
(VWF:RCo) assay.[10 ] Ristocetin was originally developed as an antibiotic; although effective as an antibiotic,
use in humans identified that ristocetin also caused thrombocytopenia. It was later
identified that this was since ristocetin bound to plasma VWF, and caused its “unfolding,”
which in turn uncovered a normally cryptic epitope on VWF that otherwise bound to
platelet GPIb. In vivo, this would cause platelet agglutination and clearance, and
thus thrombocytopenia. This adverse in vivo effect of ristocetin led to its eventual
use to measure VWF GPIb binding in vitro, as facilitated by ristocetin.[10 ] The VWF:RCo assay can be performed on fresh platelets washed free of endogenous
VWF; however, to enable assays to be used over a longer term, it became usual to either
fix the platelets (using formalin) or to freeze-dry (or lyophilize) the platelets.
The original VWF:RCo assays were either performed by qualitative visual assessment
of platelet clumping or quantified using a platelet aggregometer. Later, the VWF:RCo
assays were automated, and can now essentially be performed on several (but not all)
automated hemostasis analyzers, provided the right turbidimetric parameters and reagents
are available for use on that instrument.[10 ]
[11 ] Several manufacturers can provide reagents to enable VWF:RCo testing, with at least
two systems designed to be automated (Siemens and Diagnostica Stago).
Because classical VWF:RCo was associated with poor reproducibility, and/or poor low
VWF level sensitivity, several commercial and research alternatives were developed.
For example, it is possible to replace the platelets, as used in a classical VWF:RCo
assay, with either latex or magnetic beads that have been coated with recombinant
GPIb, and these assays are called VWF:GPIbR assays.[3 ]
[11 ]
[12 ] These assays still require ristocetin to unfold the plasma VWF and enable its capture
by the recombinant GPIb. There is only a single manufacturer of commercial VWF:GPIbR
assays by latex (LIA) or by CLIA method, namely, Werfen.
Alternately, latex beads can be coated with mutated recombinant forms of GPIb that
reflect gain of VWF function mutations; these mutated forms bind to VWF spontaneously,
and thus the assays do not require ristocetin. These assays are abbreviated as VWF:GPIbM.
There is only a single commercially available latex-based (i.e., LIA) VWF:GPIbM assay
(Siemens), which is marketed as “VWF Ac.”[12 ] An ELISA version of VWF:GPIbM can also be developed, provided there is access to
suitable reagents.[13 ]
All the GPIb-binding assays measure VWF:GPIb binding (VWF:GPIbB), but they do so differently,
with different reagents and detection methods. Thus, it is inevitable that there will
be some differences in the VWF:GPIbB that they each detect.[14 ] Nevertheless, these three assay types (VWF:RCo, VWF:GPIbR, and VWF:GPIbM) are largely
interchangeable and yield similar values for the same tested sample in the majority
of tested samples, as, for example, can be evidenced in modern external quality assessment
(EQA) exercises.[15 ]
[16 ]
[17 ] However, there remain differing opinions around which assays are best for diagnosis/classification
of VWD. The latest VWD diagnosis guidelines,[18 ] for example, recommend the newer VWF:GPIbM and VWF:GPIbR assays over classical VWF:RCo,
because they would be expected to show better reproducibility (or lower assay variation)
as well as improved detection of low levels of VWF. Moreover, these guidelines also
suggest VWF:GPIbM might be preferred over VWF:GPIbR, since use of ristocetin can identify
false low VWF:RCo levels in some normal individuals, most notable in African Americans
with particular VWF polymorphisms that prevent ristocetin from binding to their VWF.[19 ] However, these polymorphisms may not be a problem for VWF:GPIb testing in geographic
areas in which the polymorphisms are absent or rare,[17 ] and there exists evidence that certain VWF:GPIbR assays (notably the CLIA method)
have better reproducibility and low-level VWF sensitivity detection than VWF:GPIbM.[15 ]
[16 ]
[17 ] Also, commercial availability and regulatory clearance of assays is an important
consideration, and not all assays are available in all locations. Finally, some type
2 VWD cases may show VWF dysfunction detectable by VWF:GPIbR and not by VWF:GPIbM
(and potentially vice versa),[15 ]
[16 ]
[17 ] so laboratories need to weigh up all these factors to make the best assay choice
of VWF:GPIbB assay, as available for their geographic locality, which may also be
predicated by what instrumentation they use.
von Willebrand Factor Collagen-Binding Activity
The VWF:CB is an under-appreciated assay that is less widely available than VWF:GPIbB
assays. Availability may also be constrained by lack of regulatory clearance. For
example, no VWF:CB is regularly cleared or approved for use by the U.S. FDA (Food
and Drug Administration), and so very few U.S. laboratories would perform this assay.
As previously noted for VWF:GPIbB assays, where several versions are available, the
situation for VWF:CB is similar and perhaps even more complex. For VWF:GPIbB, the
choices are VWF:RCo (several manufacturers), VWF:GPIbR (one manufacturer, Werfen),
and VWF:GPIbM (one manufacturer, Siemens). For VWF:CB, the main choices are either
ELISA (several manufacturers or in-house assays) or CLIA (one manufacturer, Werfen).[3 ]
[20 ] For ELISA, the presence of several manufacturers or in-house assays (what U.S. laboratories
call laboratory-developed tests [LDTs]), all potentially using different collagen
sources (from different animals, from different anatomical locations [e.g., skin,
tendon], different collagen extraction methods, different coated collagen concentrations,
different plastic plates with different binding properties) makes for very varied
assay utility. This is explored in far greater detail later; suffice to say for the
moment that different VWF:CB assays can yield great variation between methods, and
some users may find the assay problematic, whereas other users will indicate the VWF:CB
to be an invaluable tool in VWD diagnostics, provided that you use an “optimized”
assay.[21 ]
[22 ]
von Willebrand Factor:Factor VIII Binding Activity
This assay assesses the binding of FVIII to VWF, and is important for diagnosis/exclusion
of type 2N VWD.[23 ]
[24 ] There is only one commercial option (Diagnostica Stago). It is possible to use an
in-house (or LDT) version of VWF:FVIII binding (VWF:FVIIIB), but like the in-house
VWF:CB, it needs to be properly “optimized.”[23 ]
[24 ] Very few laboratories perform this assay. As an example for Australia, all laboratories
that test for VWD will likely perform a FVIII:C, a VWF:Ag (most typically by LIA,
and less so by CLIA), and a VWF:GPIbB, with method depending on preference and available
instrumentation (i.e., VWF:RCo or VWF:GPIbR or VWF:GPIbM). Around 50% of Australian
laboratories testing for VWD perform VWF:CB, mostly using a variety of ELISA methods,
and increasingly by CLIA. But less than 5% of Australian laboratories testing for
VWD would perform a VWF:FVIIIB assay.
Ristocetin-Induced Platelet Aggregation/Agglutination
The ristocetin-induced platelet aggregation/agglutination (RIPA) assay is typically
performed as part of a platelet function assay using an aggregometer.[25 ] The assay utilizes ristocetin (like VWF:RCo), but markedly differs from the VWF:RCo
assay. In VWF:RCo, a fixed concentration of platelets is mixed with a fixed concentration
of ristocetin, and then mixed with a dilution of test plasma, which makes the VWF
activity the assay-limiting step, and enables quantitation of VWF:RCo activity. In
RIPA, patient platelet-rich plasma is mixed with varying concentrations of ristocetin
to determine the sensitivity of the patient's VWF/platelets to ristocetin. Platelets
aggregate according to the level of VWF and platelets, and their ristocetin sensitivity.
The RIPA is primarily used to assess for the presence of type 2B VWD, reflecting hyper-adhesive
VWF, or the presence of platelet type (PT-) VWD (also called pseudo-VWD), which reflects
hyper-adhesive GPIb. Thus, RIPA is expected to show best sensitivity to ristocetin
(i.e., lead to platelet aggregation at low concentrations of ristocetin) in 2B and
PT-VWD. In contrast, in type 2A and severe type 1 VWD, platelet aggregation will occur
only with high concentrations of ristocetin. Type 3 VWD, with an absence of VWF, will
not show any aggregation to any concentration of ristocetin.[26 ]
von Willebrand Factor Multimers
As mentioned, plasma VWF is constructed in vivo as a multimeric protein, which begins
with inter-subunit carboxyl termini disulphide bond formation of “pre-VWF” protein.
This initial “tail-to-tail” dimerization is eventually further processed by additional
multimerization of pro-VWF dimers that involves another round of disulphide bond formation
near the amino-termini of the subunit.[5 ] In vivo biosynthesis of VWF is limited to endothelial cells and megakaryocytes.
After “construction,” mature VWF exits in the plasma as a series of oligomers containing
a variable number of subunits, ranging from a minimum of 2 to a maximum of 40, with
the largest multimers having molecular weights in excess of 20,000 kDa.
As also previously mentioned, the largest multimers have the greatest overall adhesive
or functional ability since these contain the greatest number of overall binding sites
for GPIb, collagen and FVIII. In my teachings, I often refer to VWF as being like
“sticky string.” The longer the VWF molecule (i.e., highest number of multimers),
the greater the adhesion of this VWF to platelets (i.e., GPIb) and to damaged sub-endothelium
(i.e., collagen), thereby the longest “sticky string” being able to arrest platelets
at the site of injury and lead to more stable thrombus formation.
It is possible to visualize the multimeric pattern of plasma VWF by performing VWF
multimer analysis. In the past, this used to be done using radio-labeled VWF and gel
electrophoresis with X-ray visualization ([Fig. 2 ]). These assays used to take upward of 2 to 4 days to perform for a total of 15 or
so lanes, which meant 13 patient samples after inclusion of normal and type HMW-deficient
2A/2B VWD controls. The other problem with VWF multimers is that most laboratories
are required to perform in-house assays (i.e., LDTs), and there is considerable variability
in performance of these between laboratories. The outcome is that these assays suffer
from poor accuracy, as has been highlighted in several reports. In one study, from
the North American Specialized Coagulation Laboratory Association (NASCOLA), 5% (4/81)
of participants incorrectly reported loss of HMWM in normal samples, 18% (14/80) incorrectly
reported loss of HMWM in type 1 VWD samples, and 22% (8/36) incorrectly reported a
normal multimer pattern in type 2 VWD samples; overall, 14.7% of VWF multimer survey
responses were in error.[27 ] In another EQA report, from the European Concerted Action on Thrombosis and Disabilities
Foundation (later to become ECAT, for external quality control for assays and tests),
VWF multimer testing was associated with 10 to 23% error rates for normal samples,
a 33% error rate for abnormal samples, and a 35 to 52% error rate for type 1 VWD.[28 ]
Fig. 2 An example of von Willebrand factor (VWF) multimer analysis using radio-isotopic
visualization of multimers (from 1991). Shown is the multimers for a 15-lane gel,
with interpretation shown in the table. A pool normal plasma (PNP) and a type 2A von
Willebrand disease sample would be used as controls, and then the interpretation based
on the multimer patterns. Normal samples would show retention of all multimers with
intensity of bands close to the normal control. Type 1 VWD (T1) would be represented
by retention of all multimers with intensity of bands significantly less than the
normal control. Type 2 VWD samples would show loss of high-molecular-weight multimers
(HMWM) and usually also intermediate-molecular-weight multimers (IMWM), with retention
of only low-molecular-weight multimers (LMWM). Lanes 2, 5, 6, 8, 11, 12, 14, and 15
were interpreted as normal (Nor); lanes 7, 9, and 10 as type 1 (T1), with lane 10
sample also identified as severe (sev); finally, lanes 3, 4, and 13 show loss of IMWM
and HMWM, and thus type 2A VWD.
The situation has potentially improved in 2023, since a semiautomated VWF multimer
assay method is now available.[29 ]
[30 ]
[31 ]
[32 ] Using the 11-lane system, nine patient samples and two controls can be run over
a single day with approximately 4 hours of hands on time. The system can clearly show
the loss or retention of HMWM and also intermediate MWM (IMWM), and thus is sufficiently
powered to help correctly characterize the majority of VWD cases ([Fig. 3 ]), for example, distinguishing samples with loss of HMWM (and potentially IMWM) (being
type 2A, 2B, or PT-VWD), from samples without loss of HMWM/IMWM (being type 1, 2M,
2N VWD, or normal samples), from samples without VWF (i.e., type 3 VWD). Because the
system uses a fixed agarose gel concentration, it cannot distinguish the triplet structure
of individual multimers, or the individual HMWMs; so, for fine detail VWF work, in-house
methods would still be required.
Fig. 3 An example of modern VWF multimer analysis using the Sebia system. Shown are results
for an 11-lane gel, with controls run in lane 1 (Dade Standard Human Plasma [pool
normal plasma]) with VWF results all close to 100 U/dL and assay ratios all close
to 1.0) and in lane 11 (a type 2A VWD with loss of high- and intermediate-molecular-weight
multimers (HMWM, IMWM). Lanes 2 and 4 show two type 1 VWD samples with retention of
all multimers, and slight loss of VWF, but normal assay ratios. Lane 5 shows a borderline
normal/low VWF sample, again with retention of all multimers and normal assay ratios.
Lane 3 shows a type 2M VWD sample, with reduced VWF:GPIbR/Ag ratio, reflecting a loss
of VWF:GPIb binding, but without loss of HMWM or IMWM (and normal VWF:CB/Ag ratio).
Lanes 6 and 9, respectively, show a type 2A and 2B VWD sample, with loss of HMWM and
IMWM. Lane 8 shows a platelet-type (PT-) VWD sample, with normal level of VWF:Ag,
but low VWF:GPIbR /Ag and VWF:Ag/CB ratio reflective of the loss of HMWM. Red font
numbers represent abnormal values for that test or parameter. LMWM, low-molecular-weight
multimers.
[Table 1 ] provides a summary of the assays used to diagnose/characterize VWD.
A History of the von Willebrand Factor Collagen-Binding Assay
In this section, I will provide a linear chronology of the use of the VWF:CB by various
workers over the assay's near 40-year life time. The main portion of the content was
identified using a PubMed search of “von Willebrand factor” “collagen binding” assay,
with citations downloaded according to publication date (earliest to latest). The
search was updated on January 9, 2023, to ensure the capture date was complete until
the end of 2022. [Fig. 4 ] shows the timeline of captured citations per year. Although the first citation was
for a 1984 study, this described collagen binding in a research study, and the first
description of a collagen-binding assay for diagnosis of VWD was in 1986.[35 ] I apologize for the length of this section, but I think it is useful to document
this chronology for the STH history series. Readers may opt to omit this section entirely
and go directly to the “Discussion” section if they wish to just read a summary of
the chronology and the main utility of the VWF:CB.
Fig. 4 Timeline of publications mentioning (“von Willebrand factor” “collagen binding” assay)
in PubMed. This search term will not capture all publications reporting on the von
Willebrand factor collagen-binding (VWF:CB) assay, as used for diagnosis of von Willebrand
disease (VWD), being restricted to the term appearing in the abstract or title. Moreover,
only around 50% of these publications actually refer to the VWF:CB, as used for diagnosis
of VWD; some alternatively refer to collagen binding in research studies evaluating
the function of VWF, or platelets. The first paper referring to a VWF:CB for use in
VWD was in 1986 by Brown and Bosak. As can be seen, the VWF:CB was reported only in
a small number of papers in the 1990s, with increasing use reported over subsequent
decades.
Early History—The 1990s
To my knowledge, the VWF:CB was originally reported by Brown and Bosak, in 1986, now
some 36 years ago.[35 ] The VWF:CB was then coevaluated with the VWF:RCo assay, regarding its ability to
both detect VWD and identify and discriminate potential type 2 VWD subtypes. Although
this study was a relative landmark, being the first description of the potential utility
of the VWF:CB in VWD, the data reported, and the conclusions raised, did little to
endorse the continued use of the VWF:CB in VWD diagnostics. The study comprised a
very small group of VWD patients and normal individuals, and used only a single source
of collagen to coat ELISA plates. The comparative data between the VWF:CB and VWF:RCo
were not strikingly dissimilar, and in hindsight, the VWF:CB utilized was not really
optimized for discrimination of HMWM VWF. Indeed, the authors saw the VWF:CB as a
potential replacement to the “problematic” and time-consuming VWF:RCo, rather than
as a supplementary assay. The VWF:CB story might well have ended there, except that
several other workers decided to more extensively test the utility of this novel assay.
I was one of these workers, and was certainly tempted even by the possibility that
an ELISA-based VWF:CB could potentially replace our laborious and highly variable
platelet aggregometer-based VWF:RCo assay. We published our first study on the VWF:CB
in 1991.[36 ] We tested several different collagen preparations, but finally settled on one particular
preparation that contained a mixture of type I/III collagen. We also tested various
concentrations of collagen for coating the ELISA plates. We assessed a panel of 42
normal individual samples, and an additional 39 samples from patients identified not
to have VWD, and 53 samples from patients with VWD (37 type 1, and 16 type 2A/2B).
Of interest, we identified that in our hands, the VWF:CB was much more sensitive to
loss of HMWM VWF in type 2 VWD than was our VWF:RCo. This alone would have proved
its worth for replacement of VWF:RCo (or inclusion in a wider test panel), since the
VWF:RCo/Ag ratio was occasionally normal in type 2 VWD (i.e., yielded false functional
concordance), which might lead to false misdiagnosis of type 2 as type 1 VWD, something
we also consistently see in RCPAQAP surveys, even today.[15 ]
[16 ]
[17 ]
[33 ] Of course, the VWF:CB was to prove even more useful over time.
Naturally, there were many others who started to report on “collagen-binding activity
assays” in those early years. For example, Perret and colleagues attempted to create
a VWF:RCo-like assay using collagen.[37 ] They found that insoluble collagen from bovine aorta was able to agglutinate washed
fixed platelets in the presence of VWF, similarly with the VWF:RCo assay. Their “VWF-collagen
cofactor” activity assay showed good correlation (r = 0.91) with their VWF:RCo assay using 65 plasma samples containing various concentrations
of VWF. It is not surprising that this assay never took off, since in essence it was
similar to VWF:RCo, and just as complex and time consuming. In another early VWF “collagen-binding
assay” exploration, collagen-mediated VWF absorption was investigated.[38 ] In this assay, plasma samples were incubated with aliquots of a collagen suspension,
and following ultracentrifugation, values for (residual) VWF:Ag were assessed in the
supernatant using a standard VWF:Ag ELISA, and compared with non–collagen-incubated
samples. This was a time-consuming assay requiring ultracentrifugation, and thus similarly
did not take off.
In terms of those who worked up versions of VWF:CB by ELISA, a smattering of individuals
were developing variations of VWF:CB ELISA, or were using VWF:CB to characterize VWD
or VWF or VWF concentrate, or exploring the VWF:CB in various disease states. For
example, in 1989, Lawrie et al[39 ] used a VWF:CB as part of a panel of tests (also including VWF:Ag, VWF:RCo, and VWF
multimers) to characterize “FVIII concentrates,” then used for hemophilia A treatment,
for potential use as VWD treatment (note that some of these “FVIII concentrates” were
in fact also VWF concentrates). They found a significant correlation between VWF:RCo
and VWF:CB for most concentrates, and based on VWF activity/Ag ratios, determined
those concentrates most likely to be useful for symptomatic treatment of severe VWD
patients. In 1994, Niesvizky et al[40 ] used a VWF:CB employing type III collagen to investigate 20 patients with end-stage
renal disease. They also assessed VWF:Ag and VWF:RCo. They found increased levels
of VWF:Ag in their patients, before and after dialysis. For VWF:RCo, values were elevated
in patients who had not undergone dialysis, but values were similar to control values
in patients undergoing dialysis. Interestingly, VWF:CB values were similar to controls
in patients who had not undergone dialysis. Of course, a relevant question in relation
to this and all other studies using the VWF:CB is “was this an optimized VWF:CB”?
The next relevant study was from our laboratory,[41 ] and evaluated the potential utility of the VWF:CB in monitoring of desmopressin
(DDAVP) therapy. We were then still using the same assay we described in 1991.[36 ] We evaluated the response of seven patients with VWD (four type 1, three type 2A)
to DDAVP, administered using a standard protocol, and assessed levels of FVIII:C,
FVIII antigen, VWF:Ag, VWF:RCo, VWF:CB, and VWF multimers, as well as performing skin
bleeding times (SBT) prior to, and at sequential time points following, DDAVP administration.
All patients showed an initial incremental increase in VWF and FVIII levels using
all assays, and some showed some correction in SBT. Although the absolute levels of
VWF and FVIII varied between patients, the VWF:CB was found to provide consistently
the greatest proportional incremental increases (i.e., x-fold) compared with baseline
(pre-DDAVP) levels. Accordingly, we consistently observed an increase in the VWF:CB/Ag
ratio for all patients evaluated. This suggested that our specific VWF:CB bound preferentially
to higher molecular weight (i.e., more functionally active) forms of VWF than did
our VWF:RCo, and we therefore proposed that the use of the VWF:CB (and VWF:CB/Ag ratio)
might provide a basis for more accurate estimation of a patient's functional responsiveness
to DDAVP therapy in future studies.
Also in 1994,[42 ] van Genderen et al described an interesting case of an AVWS in a patient with non-Hodgkin
lymphoma with a selective inhibitor directed against collagen, and as detected using
their VWF:CB and mixing studies. There was no apparent inhibitor detected by their
VWF:RCo assay. This pointed to the fact that the assays were identifying different
VWF activities, and also pointed to the utility of the VWF:CB in supplement to the
VWF:RCo.
In 1995, we used the VWF:CB to investigate patients with neurofibromatosis type 1.[43 ] In response to 1 and 2 μg/mL collagen, these patients expressed an attenuated rate
of aggregation, aggregation lag phase and ATP release, as well as requiring higher
collagen concentrations to attain threshold aggregation response. VWF:CB values were
also reduced in these patients compared with controls. Thus, as a group, patients
with neurofibromatosis type 1 displayed defective platelet function characterized
by in vitro evidence of impaired responsiveness to collagen. Also in 1995,[44 ] Thomas et al used a VWF:CB to identify increased VWF activity in newborns and infants.
They also reported elevated levels of VWF:Ag, but VWF:CB values were higher. They
thus suggested caution be exercised when interpreting laboratory data and diagnosing
VWD in newborns and young infants, which warranted the use of age-specific reference
ranges. They also proposed that efficient hemostasis observed during early neonatal
life may in part be due to the increased ability of VWF to interact with collagen.
In 1996,[45 ] van Genderen et al used the VWF:CB and VWF:RCo to investigate the relationship between
platelet count and large VWF multimers in the plasma of 36 patients with essential
thrombocythemia (ET) and 26 patients with reactive thrombocytosis (RT). In both ET
and RT patients, an inverse relationship could be established between platelet count
and large VWF multimers in plasma as well in relatively decreased VWF:RCo/Ag and VWF:CB/Ag
ratios. A normalization of the platelet count was accompanied by restoration of a
normal plasma VWF multimeric distribution. They concluded that these data suggested
that increasing numbers of platelets circulating in blood result in increased removal
of large VWF multimers from plasma. Also in 1996,[46 ] Fischer et al used the VWF:CB to investigate plasma-derived and recombinant VWF.
Heparin affinity chromatography was used to isolate VWF polymers with different degrees
of multimerization. Analysis of VWF:CB and platelet aggregation revealed that these
activities increased with increasing degree of VWF multimerization. Thus, the relationship
of the VWF:CB to preferentially detect HMWM was confirmed.
In 1997, Zieger et al[47 ] included a VWF:CB to characterize a new variant of type 2M VWD with supranormal
VWF multimers in plasma similar to those seen in normal plasma after desmopressin
infusion. Clinically, the patients presented with bleeding symptoms and expressed
reduced laboratory values for VWF:Ag, FVII:C, VWF:RCo, and VWF:CB. Also in 1997, Chang
and Aronson[48 ] investigated the VWF activity of various plasma-derived VWF preparations using VWF:CB
and VWF:RCo, in addition to VWF:Ag. They found that some preparations had higher VWF:RCo/Ag
and VWF:CB/Ag ratios than that found in normal plasma. VWF:RCo and VWF:CB activities
were tightly correlated (r = 0.95). Ultracentrifugal analysis was used to compare the size distribution of VWF:Ag,
VWF:RCo, and VWF:CB. The VWF:CB/Ag ratio decreased with decreasing VWF size. They
concluded that assignment of potency to VWF-containing preparations utilizing the
VWF:CB might be more precise and as accurate as with the traditional VWF:RCo assay.
Also in 1997, van Genderen and colleagues[49 ] extended on their earlier 1996 study[45 ] to investigate whether the decrease in large VWF multimers in plasma with increasing
platelet counts was the consequence of increased turnover of large VWF multimers in
vivo. They measured the half-life times of endogenously released VWF:Ag and VWF:CB
following DDAVP in nine ET patients and nine control subjects. Also, the half-life
times of VWF:Ag and VWF:CB were measured in four ET patients after cytoreduction of
the increased platelet count to normal or nearly normal values. Estimated half-life
times of VWF:Ag did not differ between ET patients and normal, but estimated half-life
times of VWF:CB were significantly lower in ET patients as compared with normal individuals.
After cytoreduction of the increased platelet count to (nearly) normal values in all
four ET patients, the half-life time of VWF:CB increased significantly. They concluded
that their data suggested that platelets might play a role in the homeostasis of circulating
VWF. Finally for 1997, the year recorded my first review on VWD and VWF assays in
its diagnosis.[50 ] This review was written from the perspective of the relative contribution of different
assays to the diagnosis of VWD. The review also attempted to clarify some of the issues
that led to confusion around the relative roles of different assays used to diagnose
VWD. It also tried to put some perspective into the potential relative contribution
of VWF:CB versus VWF:RCo.
In 1998, Mohri et al[51 ] described another interesting case of AVWS, this one in myeloma, with an autoantibody
inhibiting VWF activity, this time against both VWF:GPIbB (using VWF:RCo) and VWF:CB,
using type I collagen. Also in 1998, Siekmann et al[52 ] investigated the use of various collagens for VWF:CB testing, and found optimal
results using a pepsin-digested type III collagen from human placenta covalently immobilized
on a microtiter plate. Their VWF:CB data corresponded to the degree of VWF multimerization
and proposed their assay would be useful for both clinical diagnosis and for the measurement
of VWF functional activity in factor concentrates; moreover, in certain applications,
the VWF:CB might also represent a suitable replacement for VWF:RCo. This study confirmed
the observation we had made earlier that not all VWF:CB assays are the same.[36 ] In the same year, Fischer et al[53 ] also described an assay using type III collagen immobilized covalently on ELISA
plates. Assay plates were simple to prepare and remained stable at 4 and −20 °C for
at least 2 months. Sample testing confirmed proportionally low levels of VWF:CB/Ag
in samples lacking HMWM, while higher VWF:CB/Ag values were obtained for samples containing
these multimers. Furthermore, the VWF:CB/Ag ratio sensitively reflected the functional
and structural intactness of the VWF molecules for all analyzed samples. Monoclonal
antibody directed to the region within the A1 domain of VWF which interacts with the
glycoprotein Ib completely inhibited VWF:RCo, while VWF:CB was not affected. Thus,
VWF:CB and VWF:RCo clearly represented separate, noninterchangeable functional parameters
of VWF. The authors concluded that their newly described method for the immobilization
of collagen onto microtiter plates was suitable for the determination of VWF:CB, and
that VWF:CB/Ag ratio would simplify the detection and classification of patients with
VWD and assist in quality control during the purification of normal VWF. The exploration
of the potential utility of the VWF:CB for measuring VWF activity in VWF concentrates
continued in 1998, with the study from Ramasamy et al,[54 ] who flagged the recent recommendation by the European Pharmacopoeia for the characterization
of factor VIII/VWF concentrates to utilize the VWF:CB. They optimized their VWF:CB
to decrease reagent variability and to allow for interlaboratory comparison. A study
of clinical samples of patients with VWD was performed to establish that a ratio of
VWF:Ag/CB antigen ratio >3.7 was associated with loss of HMWM and a decrease in biological
activity, whereas a ratio <1.4 was associated with normal multimeric distribution.
The VWF:CB was also used to monitor changes in biological activity of VWF during the
manufacture of concentrates. They used the opposite ratio to VWF:CB/Ag, as used by
most other workers at that time. Also in 1998, Fischer et al[55 ] added to their earlier study[53 ] to compare findings with various VWF activity assays and using VWF purified from
normal human plasma and then separated into three fractions containing high, medium,
and low-molecular-weight VWF multimers. The VWF fractions were tested for VWF:Ag,
VWF:RCo, VWF:CB, and a monoclonal antibody-binding ELISA (mAB-binding ELISA), based
on VWF binding to an immobilized monoclonal antibody directed to the GPIb-binding
region within the A1 domain of VWF. They reported that the three different fractions
of VWF showed a correlation between multimer size and VWF:RCo/Ag and VWF:CB/Ag, but
that results obtained with the mAB-binding ELISA showed identical levels of mAB-binding/Ag,
without regard for the multimer size present in the tested fraction. Their results
therefore suggested that in the case of structurally normal VWF, the mAB-binding ELISA
reflected the concentration of VWF:Ag rather than VWF function. They also felt that
it was feasible that while the mAB-binding ELISA might show reduced levels for abnormal
VWF protein, structurally altered within the A1 domain of VWF, as might be found in
some patients with type 2 VWD, this assay did not appear to be suitable for functional
analysis of structurally intact VWF. This is actually a very important distinction,
and was to prove true in subsequent studies (refer to later sections). In 2023, this
mAB-binding ELISA would be called VWF:Ab, according to the latest nomenclature recommendation.[56 ]
The year 1999 also saw quite a few reports on the VWF:CB. Barington and Kaersgaard[57 ] used the VWF:CB to correlate with VWF multimers in a new process for large-scale
production of VWF concentrate. Fischer continued to report on the potential utility
of the VWF:CB in potential therapy with recombinant VWF.[58 ] Several studies were reported using the VWF:CB to investigate VWD in dogs,[59 ]
[60 ] including responsiveness to DDAVP. We reported our first series of RCPAQAP surveys
on VWF/VWD, then with a small number of participants (n = 25).[61 ] This would prove to be the first of a long series of reports where we highlight
the value of the VWF:CB via EQA findings. We explored the outcomes using a set of
10 plasmas, including a normal plasma pool (in duplicate), a 50% dilution of this
pool (in duplicate), a normal individual (x1), a severe type 1 VWD individual (x1),
two type 2B VWD individuals (unrelated donors), a type 3 VWD (x1), and a 2A VWD individual
(x1). Laboratories were asked to perform all tests available to them to establish
a laboratory diagnosis of VWD, and then to comment on the possibility or otherwise
of VWD. Overall findings indicated a wide variation in test practice, in the effectiveness
of various test procedures in detecting VWD, and in the ability of various composite
test panels to identify type 2 VWD subtypes. First, while all laboratories (n = 25) performed tests for FVIII:C activity, VWF:Ag and at least one functional VWF
assay (VWF:RCo; n = 23), and/or the VWF:CB (n = 12), only three laboratories performed VWF:multimer analysis. Second, for the three
quantitative VWF assays, 10/25 (40%) laboratories performed all three, whereas 15/25
(60%) performed only two (VWF:Ag and VWF:RCo, n = 13; VWF:Ag and VWF:CBA, n = 2). Third, a variety of assay methodologies were evident for VWF:Ag (ELISA, EID,
LIA, and VIDAS assay) and VWF:RCo (platelet agglutination/“aggregometry” and a “functional
VWF:RCo alternative” ELISA assay—this being an early version of VWF:Ab). Between method
analysis for the quantitative VWF assays showed that the VWF:RCo yielded the greatest
degree of interlaboratory assay variation, and had the poorest overall performance
with respect to sensitivity to low levels of VWF. The VWF:CB also performed better
than VWF:RCo in terms of ability to detect functional VWF “discordance” (i.e., type
2 VWD). Within VWF:Ag method, analysis showed that the EID assay procedure was associated
with the greatest variation in assay results, while the EID and LIA test methods both
showed poorer sensitivity at low VWF levels compared with the ELISA method. Within
the VWF:RCo assay procedure, greatest variation in assay results and poorest sensitivity
to low VWF levels was obtained using the agglutination method; however, the agglutination
procedure showed better performance than the “functional VWF:RCo alternative” ELISA
[VWF:Ab] assay in identifying type 2 VWD plasma samples. Finally, despite identified
variations, most laboratories appeared to understand the complexities involved in
the VWD-diagnostic process, and made appropriate diagnostic predictions regarding
tested samples. From a total possible 246 interpretation events, laboratories in most
cases correctly identified normal samples as normal (67/75 events = 89%), and VWD
samples as derived from individuals with VWD (117/121 events = 97%). Moreover, when
VWD was suggested by laboratory findings, laboratories usually correctly predicted
the general subtype of VWD present (96/109 events = 88%). When “misinterpretations”
occurred, these could often be linked to the test panels utilized by laboratories.
That is, laboratories using the VWF:Ag and VWF:RCo combination were more likely to
incorrectly identify samples derived from type 2 VWD patients as being type 1, type
1 VWD patients as being type 2, and normal plasma samples as potentially derived from
patients with VWD, compared with those using the VWF:Ag and VWF:CB. It remains of
interest to me that the essential findings from this inaugural RCPAQAP report would
continue to be found in subsequent reports, including those most recent.[15 ]
[16 ]
[17 ]
The 1999 VWF:CB reports continued, with Gerritsen et al[62 ] using the VWF:CB to assay levels of plasma VWF-cleavage protease (later to be identified
as ADAMTS13, a disintegrin and metalloproteinase with a thrombospondin type 1 motif,
member 13) as a tool for the diagnosis of thrombotic thrombocytopenic purpura (TTP).
Prior methods by immunoblotting were time-intensive and cumbersome. They therefore
developed a new functional assay based on the preferential binding of HMWM forms of
VWF to collagen. In this assay, the diluted plasma sample to be tested is added to
normal human plasma in which protease activity had been abolished (i.e., TTP sample).
The VWF present in the protease-depleted plasma is digested by the VWF-cleaving protease
(i.e., ADAMTS13) in the test plasma. The proteolytic degradation leads to low-molecular-weight
forms of VWF, which show impaired binding in the VWF:CB (they used human collagen
type III). Testing of plasma from patients with TTP and hemolytic-uremic syndrome
(HUS) showed that the assay could be used to distinguish between the two syndromes.
The presence of an inhibitor could be detected by carrying out the test after incubation
of normal human plasma with the patient plasma sample, thus enabling differentiation
of patients with familial TTP from those with nonfamilial TTP. This marked the start
of the wider use of residual VWF:CB as a marker of ADAMTS13 activity.
Mauz-Körholz et al[63 ] used a VWF:CB, along with VWF:RCo, VWF:Ag, and FVIII:C to explore DDAVP treatment
in a child with 2M VWD, with good increments in all 2 hours post-DDAVP, and concluding
that in some type 2M VWD cases, DDAVP administration might be an effective treatment
in cases of elective surgery, dispensing with VWF replacement by pooled blood products.
Kertzscher et al[64 ] explored the potential utility of the VWF:CB to characterize platelet VWF content
in 24 patients with various forms of VWD. No platelet VWF:Ag or VWF:CB was detectable
in type 3 patients (n = 4). In contrast, six out of seven patients with type 2 VWD had normal or increased
VWF levels. Two type 1 patients (out of n = 13) with low VWF levels in platelets had no increased bleeding tendency. In two
other individuals with normal amounts of VWF in platelets and low plasmatic VWF and
FVIII:C, more frequent bleeding episodes reflecting the low plasmatic levels were
observed in a long-term follow-up. The authors concluded that in their patients, bleeding
history corresponded to plasmatic levels of FVIII:C and VWF. Finally, the decade ended
with my second review on VWF/VWD, again from the perspective of laboratory testing
for VWD, and where I concluded that selection of an appropriate test panel is a critical
component for the proper diagnosis and classification of VWD.[65 ]
The Next Century History—The 2000s
The 2000s started with another report from our laboratory, this one attempting to
clarify that the utility of a VWF:CB depended on the source of collagen.[21 ] I evaluated 21 different collagen preparations for their ability to both detect
VWD and discriminate different VWD subtypes (i.e., type 1 vs. 2A/2B). Collagen preparations
were tested at a range of concentrations and included types I, III, and IV, collagen,
and various mixtures of these, as aqueous supplied preparations and/or reconstituted
from bulk lyophilized stock. Tissue sources for collagens ranged from human placenta
to calf skin to equine tendon. Three of the collagen preparations tested did not support
VWF binding in an ELISA process (and were therefore unable to facilitate VWD diagnosis).
The ability of the remaining preparations to detect VWF was variable, as was their
ability to discriminate VWD subtypes. Detection of VWF and discrimination of VWD subtypes
were not mutually inclusive. Thus, some collagen preparations provided excellent detection
systems for VWF, but comparatively poorer discrimination of type 2 VWD, while others
provided good to acceptable detection and discrimination. Subtype discrimination was
also dependent on the collagen concentration, and some batch-to-batch variation was
evident with some preparations (particularly type I collagens). Overall, best discrimination
was typically achieved with type I/III collagen mixtures, or type III collagen preparations
(where effectiveness was highly dependent on concentration used). Good discrimination
was also achieved with a commercial type III collagen-based VWF:CB method. Results
of the various “VWF:CB assays” are also compared with those VWF:RCo assay (by platelet
agglutination) and that using a commercial “VWF:RCo alternative/activity” ELISA procedure
(i.e., an early VWF:Ab assay). These VWF:Ab methodologies tended to be less sensitive
to VWF discordance identified in type 2 VWD when compared with that detected by the
majority of the VWF:CB procedures.
Another report from our laboratory soon followed, this one evaluating a novel sulfatide-binding
assay for VWF, which was able to detect VWD, but seemed to offer no utility to discriminate
VWD subtypes.[66 ] This was in contrast to both VWF:RCo and VWF:CB assessed at the same time in the
same samples. Dean et al[67 ] then reported the use of both the VWF:CB and the platelet function analyzer (PFA)-100
in pediatric patients. They concluded that the PFA-100 was a better screening test
for VWD than the SBT. They compared the VWF:CB as a functional test for VWF against
the more routinely used VWF:RCo. They found that the VWF:CB detected 43/49 (88%) subjects
with definite types 1, 2, or 3 VWD, performing as well as the VWF:RCo, which detected
42/48 (88%). They also showed that, used in conjunction with VWF:Ag levels, the VWF:CB
might be useful in the classification of VWD subtypes.
Two more reports from our laboratory followed. The first directly compared several
commercial VWF activity options for the identification and characterization of VWD,[68 ] and the second compared VWF:CB testing against various constructed or commercial
VWF:Ab options.[69 ] At that time, two VWF:CB assay methods were commercially available, as was a monoclonal
antibody (MAB)-based ELISA (now called VWF: Ab) by then reported to correlate with
a standard VWF:RCo assay. This VWF:Ab assay was then marketed as a VWF:activity assay
and was available in two assay version formats. These four VWF-activity options were
directly compared with each other and in-house VWF:CB ELISAs for their ability to
detect VWD, and discriminate qualitative VWD defects.[68 ]
[69 ] The two MAB-based systems detected VWD but could not specifically identify qualitative
VWF defects, although the recently modified “Mark II” kit was more effective for the
latter compared with the original “Mark I” kit. All VWF:CB methods, including in-house
and commercial, also effectively detected VWD but differed in their ability to identify
qualitative VWF defects. Effectiveness was highest using our in-house reference VWF:CB
(based on a type I/III collagen mix product from equine tendon), the then available
Gradipore VWF:CB (which also used equine tendon-derived collagen), or an in-house
VWF:CB method using type III human collagen at a relatively low concentration (1 or
3 μg/mL, without covalent linkage). The then available IMMUNO VWF:CB seemed to be
the least effective among all the compared VWF:CB methods for the detection of qualitative
VWF defects. In the second study,[69 ] we adapted several anti-VWF MAB developed at Westmead for use in in-house ELISA
assays to assess their utility for VWD diagnosis and subtype discrimination, and compared
these assays with other assay systems. Thus, our in-house VWF:CB, VWF:RCo by agglutination,
the commercial VWF:Ab assay, and in-house VWF:Ab assays were directly compared for
their ability to discriminate type 1 from type 2A/2B VWD samples. All VWF:Ab assay
systems effectively measured VWF and could confirm a diagnosis of VWD, as well as
exhibiting some VWD-subtype discriminatory capabilities. However, better evidence
of VWF-discordance in type 2 VWD was usually achieved using the VWF:RCo (agglutination)
assay, and best performance in VWF-discordance was consistently observed using the
VWF:CB assay. In conclusion, the VWF:CB assay proved to offer the best diagnostic
predictive tool for a type 2 VWD defect, while VWF:Ab-based systems appeared to be
least effective in this regard.
The second of our RCPAQAP VWF/VWD studies was also published in the year 2000.[70 ] This study reported an evaluation of current laboratory practice for the diagnosis
of VWD by means of a multi-laboratory (n = 19) survey, and results compared with our earlier survey.[61 ] Samples comprised a new set of seven plasmas: a type 3 VWD, a type 2B VWD, a moderate
type 1 VWD/hemophilia A combined defect, a normal individual, a mild type 1 VWD, a
type 2A/2M VWD, and a type 2N VWD. Overall, many findings confirmed those reported
earlier (including between-method analysis, within-method analysis, inter-laboratory
assay variation, sensitivity to low levels of VWF, detection of functional VWF “discordance,”
and appropriateness of diagnostic predictions). Novel findings included: (1) although
VWF:CB activity performed better than VWF:RCo in the identification of functional
discordance in type 2B VWD, both assays performed equally in the identification of
discordance in the type 2A/2M VWD; (2) most laboratories failed to identify the type
2N VWD as a potential type 2N VWD utilizing VWF:Ag and FVIII:C testing as a screening
process. This latest survey was followed up by a dry workshop attended by over 45
scientists from Australia and New Zealand, and comprising representatives from most
survey participants. Discussion covered many topics including the effect of blood
group, the role (if any) of the SBT, the role of the PFA-100, confirmatory and additional
tests, and the possibility of restricting testing to specialized centers. Consensus
was reached on the following points: (1) diagnosis of VWD requires both clinical and
laboratory assessment; (2) testing should comprise FVIII:C, VWF:Ag, and either/or
both VWF:RCo and VWF:CB; (3) laboratory results should be reviewed in the light of
clinical findings; and (4) confirmatory repeat testing should be performed on a sample
taken 6 weeks later.
Federici et al[71 ] then published a short report on VWF:RCo/Ag and VWF:CB/Ag for a rapid diagnosis
of type 2 VWD, including a comparison of four different assays. These comprised an
in-house VWF:RCo, an in-house VWF:CB, a commercial VWF:CB (Immuno-Baxter), and a commercial
“VWF activity” assay (i.e., a VWF:Ab). The latter two were the same as those we previously
evaluated in the Australian studies.[68 ]
[69 ] The in-house VWF:CB and VWF:RCo assays were found to be better for VWD diagnosis
than either of the commercial methods, and the VWF:Ab the least useful, confirming
the findings of our earlier studies.
Casonato and colleagues fuelled the discussion of whether the VWF:CB could replace
the VWF:RCo in their 2021 study of 10 type 2A and 12 type 2B VWD patients, together
with 30 type 1 VWD patients with reduced platelet VWF content.[72 ] In both 2A and 2B VWD, VWF:CB and VWF:RCo were decreased, but reductions of VWF:CB
were more consistent. The difference was more evident when values were expressed as
a ratio, where the VWF:CB/Ag ratio was always below 0.2, while that for VWF:RCo was
greater than 0.4, and in no patient was the VWF:CB/Ag ratio higher than VWF:RCo/Ag.
In contrast, in type 1 VWD, the decrease in VWF:CB was similar to that seen in VWF:RCo
with the ratios always within the normal range. The authors also assessed DDAVP response
in type 2A and 2B VWD patients. The differences between the VWF:CB and VWF:RCo were
even more evident after DDAVP, and in type 2A VWD, even though large multimers were
persistently decreased, VWF:RCo was normalized, while VWF:CB remained defective. These
findings clearly indicated to the authors that VWF:CB detected the absence of large
and intermediate VWF multimers better than VWF:RCo. Hence, they suggested adding VWF:CB
to the panel of tests employed in the diagnosis of VWD. Moreover, owing to the difficulty
in performing VWF:RCo and its low reproducibility, they suggested that, when necessary,
VWF:CB may be substituted for VWF:RCo.
Our own study of VWF:CB and PFA-100 testing in DDAVP-treated VWD patients soon followed.[73 ] From a panel of 125 patients undergoing evaluation for clinical hemostatic defects,
29/30 samples from patients with VWD (17/18 type 1, 1/1 type 3, 3/3 type 2A, 7/7 type
2B, and 1/1 PT-VWD) gave prolonged closure times (CTs) using the collagen/epinephrine
(C/Epi) cartridge. For these patients, correction of an initially prolonged CT by
DDAVP was accompanied by normalization of VWF:Ag, VWF:CB, and VWF:RCo in type 1 VWD
(n = 5). In an individual with type 2A VWD, DDAVP normalized VWF:Ag and VWF:RCo, but
had no apparent effect on the baseline maximally prolonged CT. In an individual with
type 2B VWD, VWF concentrate also normalized VWF:Ag and VWF:RCo, but similarly had
no apparent effect on the baseline maximally prolonged CT. VWF:CB did not normalize
for either of these two individuals, potentially suggesting that normalization of
VWF:CB might be required for normalization of CT. This concept was supported by correlation
analysis undertaken between CT and various VWF parameters. Among these, VWF:CB held
the strongest relationship in our dataset, which showed an inverse progressive rise
in CT for falling VWF:CB. Based on our results, we concluded that the PFA-100 was
highly sensitive to the presence of VWD, and might thus provide a valuable screening
test for VWD. Furthermore, the combined utility of the PFA-100 and VWF:CB as markers
of DDAVP responsiveness may prove to be simple, quick but powerful, predictors for
its clinical efficacy.
A study on VWF test parameters, including VWF:Ag, VWF:CB, VWF:RCo, and VWF multimers,
in patients with uremia followed,[74 ] and failed to identify a correlation with uremic bleeding. Calibration of the 4th
International Standard (97/586) for FVIII and VWF followed, with the attempted inclusion
of a new parameter, the VWF:CB, to complement the usual FVIII:C, FVIII:Ag, VWF:Ag,
and VWF:RCo values.[75 ] Unfortunately, excessive inter-laboratory variability and a low number of estimates
(n = 6) precluded the assignment of a potency for VWF:CB! The function of VWF in children
with diarrhea-associated HUS was investigated by Sutor et al.[76 ] These children had a reduced VWF:CB/Ag ratio, despite high levels of VWF:Ag (mean:
253 U/dL). They proposed that the very high concentration of plasma VWF:Ag in HUS
probably reflected endothelial cell damage or irritation. The presence of dysfunctional
VWF might be caused either by a primary (due to enterohemorrhagic Escherichia coli ) or secondary (due to consumption of functionally active VWF) process. This abnormality
was not evident as structural anomaly by multimer analysis.
Kallas and Talpsep[77 ] evaluated a VWF:CB, based on type III collagen, for potential utility in VWD diagnosis
and DDAVP response monitoring. The assay correlated with VWF:RCo, low VWF:CB/Ag ratios
were observed in type 2 VWD, and VWF:CB/Ag ratios increased post DDAVP, consistent
with release of HMWM. They concluded that their findings suggested the VWF:CB assay
to be a useful test for measuring the functional activity of VWF in plasma samples,
VWF concentrates, as well as for estimating the outcome of treatment.
Mannucci et al[78 ] used a VWF:CB to evaluate changes in the VWF-cleaving protease (i.e., ADAMTS13)
in health and disease. They identified an inverse relation between low VWF-protease
and high plasma levels of VWF:Ag and VWF:CB activity, and that low plasma levels of
VWF-cleaving protease were not a specific beacon of TTP, because the protease was
also low in several physiological and pathologic conditions. Váradi et al used a VWF:CB
to explore thrombin-mediated in vitro processing of pro-VWF.[79 ]
In 2002, Riddell et al used a VWF:CB to explore type 2M VWD, and compared this with
VWF:RCo.[80 ] They analyzed a group of 32 patients with type 2 VWD (25 patients with type 2M,
six with type 2A, and one with type 2B) versus 22 normal control subjects. VWF:RCo/Ag
and VWF:CB/Ag ratios were compared between the patient and control groups. In the
six patients with type 2A VWD, both VWF:RCo/Ag ratios and VWF:CBA/Ag ratios were discordant
(≤0.7). In the 25 type 2M VWD patients, the VWF:CBA/Ag ratios were concordant (> 0.7),
but the VWF:RCo/Ag ratios were discordant (≤0.7) compared with control subjects. Thus,
VWF:RCo/Ag ratios were discordant in both type 2M and 2A VWD patient groups indicating
a functional abnormality. However, VWF:CB/Ag ratios were discordant in the type 2A
VWD group but not in the type 2M VWD group. Their study showed that VWF:CB is sensitive
to functional variants associated with the loss of HMWM (i.e., types 2A and 2B VWD),
but was unable to discriminate defective platelet-binding VWD variants with normal
multimeric patterns such as type 2M VWD. They concluded that the VWF:CB assay should
be used in association with rather than as a replacement for the VWF:RCo assay. However,
what this study also showed, but not expressed in the abstract, was that combination
of VWF:RCo and VWF:CB could be used to discriminate between 2A and 2M VWD.
Turecek et al continued to explore the VWF:CB in recombinant pro-VWF processing.[81 ] Linder et al used a VWF:CB as part of a study to explore the attachment of whole
blood platelets on extracellular matrix under flow conditions in preterm infants.[82 ] Saenko et al described the development and application of a surface plasmon resonance-based
VWF:CB assay.[83 ] Their assay correlated with an ELISA-based VWF:CB, but was more accurate and reproducible.
They used their assay to explore VWF concentrates during production. They suggested
their assay to be a useful tool in the development of industrial virus-inactivation
procedures, allowing preservation of VWF activity and achieving the maximal therapeutic
efficacy of FVIII/VWF concentrates.
Michiels et al explored the DDAVP response in patients with type 1 or 2 VWD from a
diagnostic and therapeutic perspective.[84 ] They evaluated whether VWF:CB, VWF multimeric analysis, and the response to intravenous
DDAVP could correctly diagnose and classify congenital VWD in 24 probands with mild
to moderate type 1 VWD, 6 probands with severe VWD type 1, and 12 probands with type
2 VWD. Neugebauer et al compared two VWF:CB assays with different binding affinities
for low, medium, and high VWF multimers.[85 ] They concluded that the assay with pepsin-digested collagen (human, type III) that
was covalently linked to pre-activated microtiter plates revealed a higher affinity
for low and medium VWF multimers, whereas the assay with collagen fibrils (equine,
type I) that were adsorbed to microtiter plates predominantly bound high VWF multimers.
Turecek et al performed a comparative study on VWF:CB by ELISA versus VWF:RCo for
the detection of functional VWF activity.[86 ] Measurement of functional VWF:CB activity could be performed with substantially
higher interassay reproducibility than VWF:RCo. Both assay systems could be used for
diagnosis and subtyping of VWD, but their VWF:CB was more sensitive than VWF:RCo.
The analysis of VWF multimers in different fractions obtained by affinity chromatography
on heparin Sepharose showed that VWF activity measured both with VWF:RCo and VWF:CB
correlated with the degree of VWF multimerization. They suggested that measurement
of VWF:RCo could be replaced by the more reliable VWF:CB, which appeared not only
to be more sensitive and easier to carry out but was also found to have a higher reproducibility
and allow better standardization.
Also in 2002, Budde et al used the VWF:CB as part of a test panel including VWF:Ag,
VWF:RCo, and VWF multimers, to characterize 303 patients with VWD.[87 ] I wrote my first review focused on the VWF:CB.[88 ] Finkelstein et al used the VWF:CB as part of a study to explore platelet deposition
on extracellular matrix under flow conditions in preterm neonatal sepsis.[89 ] Remuzzi et al used the VWF:CB as part of a study into TTP versus HUS to help address
the question of whether ADAMTS13 levels can be used to distinguish TTP/HUS.[90 ] Hubbard et al again attempted to incorporate VWF:CB values into an international
standard, this time the 1st International Standard for von Willebrand Factor concentrate
(00/514).[91 ] As per the previous attempt for the plasma standard,[75 ] large interlaboratory variability of estimates precluded the assignment of a value
for VWF:CB. Rick et al described the clinical use of a rapid VWF:CB to evaluate VWF-cleaving
protease (i.e., ADAMTS13) in patients with TTP.[92 ] They reported 97.5% concordance between the VWF:CB and a VWF multimer gel assay.
The VWF:CB identified low VWF-protease activity in 78% of patients who had a clinical
syndrome consistent with TTP/HUS and in 2 of 10 sick controls, giving it a positive
predictive value of 0.94. The VWF:CB also detected inhibitors of VWF-protease in 26
of 29 patients (90%) with TTP/HUS and low protease activity levels. The authors concluded
the VWF:CB to be a useful clinical assay for examining VWF-protease activity and detecting
associated inhibitors.
Paczuski assessed the effect of collagen source and coating conditions on the ability
of a VWF:CB to diagnose VWD, and reported that modification of coating conditions
and the use of an alkaline buffer permitted the use of a relatively low concentration
of collagen, and that this procedure was useful in the diagnosis of VWD and to distinguish
between types 1 and 2 VWD.[93 ] Sciahbasi et al used a VWF:CB to help explore the prothrombotic response to coronary
angioplasty in patients with unstable angina and raised C-reactive protein (CRP).[94 ] After angioplasty, virtually all patients with unstable angina and raised preprocedural
CRP showed increased VWF:Ag, VWF:CB, and CRP. Their data suggested that such changes
may contribute to the worse prognosis of unstable patients with raised indices of
inflammation. Haley et al explored the effect of ABO blood group on the VWF:CB, to
show similar changes to VWF:RCo, with lower levels in O-blood group.[95 ] The potential role of the VWF:CB in assessing VWF activity in VWF concentrates was
further explored,[96 ] as was the utility of the VWF:CB to measure VWF-cleaving protease (ADAMTS13) activity.[97 ]
[98 ]
[99 ]
In 2003, Vincentelli et al[100 ] used the VWF:CB to explore AVWS in aortic stenosis, to identify decreased VWF:CB
with loss of HMWM, and reduced platelet function, in 67 to 92% of patients with severe
aortic stenosis and results correlating significantly with the severity of valve stenosis.
The use of the VWF:CB in measuring VWF-cleaving protease (ADAMTS13) activity continued
with a comparison of methods by Studt et al.[101 ] Miller et al[102 ] reassessed changes in VWF level and activity (including the VWF:CB) according to
ABO blood group and race.
By 2004, a Chinese group had also evaluated a VWF:CB by ELISA to conclude that the
measurement of the functional activity of VWF by VWF:RCo or RIPA could be replaced
by the more reliable VWF:CB.[103 ] In 2023, I no longer think of replacement of VWF:RCo (and never thought the VWF:CB
could replace RIPA), but rather the VWF:CB has supplementary utility in VWD diagnostics.
In Japan, Sakai et al used the VWF:CB to help characterize another AVWS due to autoantibodies[104 ]; in this case, the antibodies were not functionally inhibitory by either VWF:CB
or VWF:RCo, and so VWF clearance was proposed as the likely mechanism for the AVWS.
Another study utilizing the VWF:CB to help characterize VWF concentrates was performed
by Lethagen et al.[105 ] VWF:RCo and VWF:CB correlated with each other, but not with VWF:Ag, and there were
substantial differences noted between the six concentrates assessed. Yet more studies
performed on VWF-cleaving protease (ADAMTS13) measured by residual VWF:CB activity
by Gao et al.[106 ]
[107 ] The authors concluded that measurement of the VWF-cleaving protease activity using
residual VWF:CB activity was a simple and rapid method for diagnosing TTP, and that
the VWF-cleaving protease activity in patients with TTP was markedly lower than those
of patients with tumors. The VWF:CB continued to make inroads into transfusion practice
when Burnouf et al used the VWF:CB to help assess the content and functional activity
of VWF in apheresis plasma.[108 ] They concluded that the VWF:Ag, VWF:RCo activity, and the 11th to 15th group of
VWF multimers were well preserved in all plasma units from each of the five apheresis
procedures, but that the VWF:CB activity and the percentage of multimers greater than
15 in apheresis plasma were less than that in normal plasma pools and differed slightly
among procedures. The subtle inference here is that the VWF:CB correlated better with
HMWM than did the VWF:RCo. Hubbard and Heath had another attempt at assigning a VWF:CB
value to a WHO (5th) International Standard, this time with success![109 ] Tripodi et al conducted an international collaborative study on ADAMTS13 testing
involving 11 methods including the VWF:CB.[110 ] They concluded that overall, the best performance was observed for three methods
measuring cleaved VWF by ristocetin cofactor, collagen binding, and immunoblotting
of degraded multimers of VWF substrate, respectively. Casonato et al used the VWF:CB
as part of a test panel to assess a new variant of VWD (L1446P; 4337T- > C).[111 ]
2005 saw another study on the effect of the ABO blood group on VWF:CB (plus VWF:Ag
and VWF:RCo), this one performed by Chng et al.[112 ] We published our third RCPAQAP VWF/VWD survey paper,[113 ] this one reporting on findings with 37 plasma samples which include 9 normal samples,
4 type 1 VWD samples, 8 type 2 VWD samples (2A × 3, 2B × 3, 2M × 1, and 2N × 1), and
4 type 3 VWD samples. Similar to earlier, laboratories performing the VWF:CB performed
better than those that did not. On average, type 1 VWD plasma was misidentified as
type 2 VWD plasma in 11% of cases, and laboratories that performed the VWF:RCo without
performing the VWF:CB were six times more likely to make such an error than those
that did perform the VWF:CB. Similarly, type 2 VWD plasma samples were misidentified
as type 1 or type 3 VWD in an average of 20% of cases, and laboratories that performed
the VWF:RCo without the VWF:CB were three times more likely to make such an error
than those that performed the VWF:CB. Finally, normal plasma was misidentified as
VWD plasma in an average of 5% of cases, and laboratories that performed the VWF:RCo
without the VWF:CB were 10 times more likely to make such an error than those that
performed the VWF:CB. We concluded that laboratories were generally proficient in
their testing for VWD and that diagnostic error rates are substantially reduced when
test panels are more comprehensive and include the VWF:CB.
Mühlhausen et al[114 ] used the VWF:CB (plus VWF:Ag and multimers) to help characterize 10 patients with
glycogen storage disease type Ia, concluding that these showed a type 1 AVWS-type
pattern. Hellström-Westas et al assessed ADAMTS13, VWF:Ag, and VWF:CB in premature
infants and reported lowered ADAMTS13 and increased VWF:Ag and VWF:CB compared with
term infants.[115 ] Reiter et al reported another study on ADAMTS13 activity measured using a VWF:CB.[116 ] Trasi et al used the VWF:CB as part of a test panel to investigate the prevalence
and spectrum of VWD in western India.[117 ] Of 796 patients with bleeding manifestations, 58 were diagnosed with VWD. The majority
of patients were type 3 (59.5%) with severe clinical manifestations, ∼18% were type
1 VWD and the prevalence of qualitative variants (i.e., type 2 VWD) was 19% (2A, 9.52%;
2B, 4.76%; 2M, 1.2%; 2N, 3.6%). They concluded that the high prevalence of type 3
and the low prevalence of type 1 VWD were in contrast to western reports, suggesting
low awareness of VWD and underdiagnosis of mild cases in India.
Callan et al[118 ] performed a DDAVP study utilizing the VWF:CB on dogs with type 1 VWD, while Heseltine
et al[119 ] used the VWF:CB to help assess the effects of levothyroxine administration on dogs
with VWD. Rojnuckarin et al[120 ] used the VWF:CB to help characterize VWF levels and activity in Thais Health, also
looking at the ABO blood group and age, while O'Donnell et al[121 ] used the VWF:CB to help investigate a rare Bombay phenotype blood group associated
with reduced plasma VWF levels and an increased susceptibility to ADAMTS13 proteolysis.
Penas et al used the VWF:CB to help address the diagnostic question of 2A versus 2M
VWD.[122 ] The authors investigated 21 patients who presented with low plasma levels of VWF:Ag
and VWF:RCo, reduced VWF:RCo/Ag, and the presence of all sizes of multimers. The results
for VWF:CB varied depending on the type of collagen used. Genetic analysis showed
these type 2M patients had the mutation R1374C, and that a high frequency of the R1374C
mutation was observed in northwestern Spain (Galicia). The VWF:CB (with type I collagen)
assay was unable to discriminate defective platelet binding of the R1374C VWF, confirming
that VWF:CB cannot substitute for VWF:RCo, and both should be tested when diagnosing
VWD. Liu et al reported on a Chinese study assessing ADAMTS13 activity using residual
VWF:CB activity.[123 ] We reported our own evaluation of VWF parameters according to blood group, age,
and sex.[124 ]
In 2006, another study using the VWF:CB to assess VWD in dogs was reported, to assess
the potential for type 2 VWD. Sabino et al showed that the inclusion of the VWF:CB
resulted in reclassification of 5% of those previously identified as type 1 to type
2 VWD.[125 ] Couto et al evaluated dogs for possible VWD or platelet function defects using the
PFA-100 combined with the VWF:CB (and VWF:Ag).[126 ] More Chinese studies assessing ADAMTS13 activity using residual VWF:CB activity
were reported by Lu et al[127 ] and Liu et al.[128 ] Iwaki et al used the VWF:CB to help assess VWF activity in a mouse model of cholesterol-driven
atherosclerosis.[129 ] Lisman et al used the VWF:CB (with the VWF:RCo) to show reduced VWF activity in
liver cirrhosis, despite elevation of VWF:Ag.[130 ] Meijer and Haverkate[28 ] reported on their EQA findings for VWF/VWD on behalf of the European Concerted Action
on Thrombosis and Disabilities Foundation, the forerunner to the ECAT program. Normal
samples were interpreted correctly by the majority of the participants. However, type
1 VWD samples were wrongly interpreted by 20 to 40% of participants, which was mainly
caused by a false discordance in the VWF:RCo/Ag ratio. The between-laboratory variation
was also higher for VWF:RCo than VWF:CB (respectively, 20–40% and 17–29%). Casonato
et al[131 ] used the VWF:CB as an aide to diagnosis and follow-up of nine patients with TTP.
VWF:CB was able to detect the absence or decrease of large VWF multimers better than
VWF:RCo; moreover, VWF:CB was defective when large VWF multimers persisted to be decreased,
in contrast with what was observed with VWF:RCo. They concluded the VWF:CB to be a
simple test that appears to be useful, together with clinical symptoms and reduced
platelet count, for the diagnosis and follow-up of TTP. Sucker et al[132 ] used the VWF:CB to help characterize a new commercial VWF activity test from Werfen
(would now be termed VWF:Ab). Using 300 samples, including some with VWD, they reported
good correlation with VWF activities determined as VWF:RCo (r = 0.88), VWF:CB (r = 0.93), and VWF:GPIbB ELISA (r = 0.91). The comparability of results obtained by the new “HemosIL” assay and VWF:GPIbB
ELISA were excellent, whereas more variance was evident with VWF:RCo and VWF:CB. However,
the HemosIL assay failed to indicate a loss of HMWM; so multimeric analysis was suggested
as the procedure of choice for the differentiation of functional defects.
Shelat et al[133 ] used the VWF:CB assay in comparison with GST-VWF73 and FRETS-VWF73 assays to assess
ADAMTS13 and inhibitory autoantibodies in TTP. Shenkman et al[134 ] evaluated the VWF:CB in comparison to a new flow-based method for differentiation
between inherited and acquired TTP. Discussion on the utility of the VWF:CB in types
2A, 2B, and 2M VWD continued, with Baronciani et al[135 ] commenting on several previously published studies. I published a review highlighting
the efficacy of the PFA-100 and VWF:CB activity as coupled strategies for laboratory
monitoring of therapy in VWD.[136 ] Budde et al[137 ] undertook a comparative analysis and classification of VWF/FVIII concentrates, and
the potential impact on treatment of patients with VWD, utilizing a comprehensive
approach including the VWF:CB. They reported that both VWF:RCo and VWF:CB correlated
well with the HMW VWF content of the assessed products, and that these products differed
markedly based on relative VWF activity. Popov et al[138 ] reported on the performance and clinical utility of a commercial VWF:CB assay for
laboratory diagnosis of VWD, to conclude that the assay provided reliable results
and proved useful for laboratory diagnosis of VWD. We reported an interesting case
of AVWS, and the potential utility of the VWF:CB to the identification of functionally
inhibiting autoantibodies to VWF.[139 ]
In 2007, Lo et al[140 ] used the VWF:CB (with the VWF:RCo) to show relatively increased VWF activity after
off-pump coronary artery bypass graft surgery. Meiring et al[141 ] reported on the performance and utility of their in-house VWF:CB for the cost-effective
laboratory diagnosis of VWD, and recommending this assay together with VWF:RCo in
the diagnostic workup of VWD. Gudmundsdottir et al[142 ] used a VWF:CB (with VWF:RCo) in a study to investigate risk of excessive bleeding
in patients with marginally low VWF and mild platelet dysfunction. Starke et al[143 ] evaluated the clinical utility of various ADAMTS13 activity, antigen, and autoantibody
assays, including the VWF:CB, in TTP, to confirm the utility of newer assays. Davies
et al[144 ] used the VWF:CB to study the effect of the Y/C1584 change in VWF on in vivo protein
level and function and interaction with ABO blood group. Veyradier et al[145 ] evaluated VWF:CB using a new commercial kit with type III collagen in type 2 VWD.
Our laboratory used the VWF:CB as part of a large panel of VWF tests to compare the
pharmacokinetics of two VWF concentrates in people with VWD for a randomized crossover,
multicenter study.[146 ] In brief, VWF:CB and VWF:RCo values tended to follow VWF multimer patterns, and
functional VWF appeared to be better in the newer product. Feys et al[147 ] used a VWF:CB assay method to assess ADAMTS13 activity to antigen ratio in various
physiological and pathological conditions associated with an increased risk of thrombosis.
Wadanoli et al[148 ] utilized a VWF:CB assay to help evaluate the efficacy of a novel VWF antagonist
(GPG-290) to prevent coronary thrombosis without prolongation of bleeding time. Another
study using the VWF:CB to assess ADAMTS13 activity in TTP, this time human immunodeficiency
virus related, was reported by Gunther et al.[149 ] I published my second review on the VWF:CB,[150 ] and a separate article proposing the use of VWF:CB/Ag and VWF:RCo/Ag ratios to identify
specific VWF activity in VWF concentrates.[151 ]
In 2008, Marcucci et al[152 ] reported on VWF:Ag, VWF:CB, and ADAMTS13 activity to help assess residual platelet
reactivity in high-risk coronary patients on antiplatelet treatment. A lower ADAMTS13
activity was present in patients with VWF:Ag and VWF:CB in the upper tertile. Lu et
al[153 ] published another Chinese study on ADAMTS13 activity measured by residual VWF:CB,
this time in patients with chronic renal diseases. Tiede et al[154 ] used the VWF:CB as part of a large panel of tests to investigate patients with AVWS.
A combination of VWF:Ag < 50 IU/dL, VWF:RCo/Ag ratio < 0.7, and VWF:CB/Ag ratio < 0.8
yielded a sensitivity of 86% for AVWS diagnosis. Gallinaro et al[155 ] used a VWF:CB as part of a panel to show that shorter VWF survival in O blood group
subjects explains how ABO determinants influence plasma VWF. Geisen et al[156 ] used a VWF:CB as part of a panel to identify AVWS in patients with ventricular assist
devices (VADs) and to explain nonsurgical bleeding in these patients. The loss of
large multimers was paralleled by the reduction in VWF:CB and VWF:RCo. Chung et al[157 ] used a VWF:CB to help explore the degradation of circulating VWF and its regulator
ADAMTS13 in anthrax consumptive coagulopathy. Hanebutt et al[158 ] utilized the VWF:CB (together with FVIII:C, VWF:Ag, PFA-100) to help evaluate DDAVP
effects on hemostasis in children with congenital bleeding disorders. They concluded
that DDAVP was effective in most, but not all patients, and so DDAVP testing was recommended
to determine the individual hemostatic response. Lara-García et al[159 ] used the VWF:CB to help explain postoperative bleeding in retired racing greyhounds.
Guerin et al[160 ] published a rebuttal to our earlier study on the utility of the VWF:CB to identify
functionally inhibiting autoantibodies to VWF.[139 ] Burnouf et al[161 ] used a VWF:CB to help explore the properties of a concentrated minipool solvent-detergent–treated
cryoprecipitate. They reported the feasibility of preparing virally inactivated cryoprecipitate
minipools depleted of isoagglutinins and enriched in functional FVIII, VWF, and clottable
fibrinogen. Prohaska et al[162 ] used the VWF:CB with the PFA-100 to help investigate platelet dysfunction in patients
with aortic valve disease. Song et al[163 ] explored the relationship between ADAMTS13 activity and VWF:CB in a study that aimed
to explore platelet dynamics and relation to platelet count in patients with consumptive
coagulopathy. The platelet count itself was not correlated with ADAMTS13 activity
or VWF:CB; however, the rate of decline of log-scaled platelet count did correlate
with both ADAMTS13 activity and VWF:CB. Burgess and Wood[164 ] reported the validation of a VWF:Ag ELISA and a newly developed VWF:CB in yet another
study on dog VWD. Le Tourneau et al[165 ] used a VWF:CB to assess functional impairment of VWF in hypertrophic cardiomyopathy.
Platelet adhesion time, VWF:CB/Ag ratio, and the percentage of HMWM all correlated
closely and independently with the magnitude of outflow obstruction.
In 2009, there was continued interest in the differential patterns expressed in 2A
versus 2M VWD in relation to VWF:CB/Ag versus VWF:RCo/Ag, and thus their potential
to help characterize 2M.[166 ]
[167 ]
[168 ] The utility of the VWF:CB in measuring ADAMTS13 continued to be explored.[169 ] I continued to promote the VWF:CB and DDAVP response data for assisting diagnosis
of VWD and characterizing of subtypes.[170 ] Larkin et al[171 ] reported on severe Plasmodium falciparum malaria, noting this was associated with circulating ultra-large VWF multimers, raised
VWF:CB, and ADAMTS13 inhibition, with correlation of these markers. We published some
findings around DDAVP therapy acting as a diagnostic aide for functional identification
and characterization of VWD, and the combined use of VWF:CB and VWF:RCo to facilitate
this.[172 ] In brief, (1) type 1 VWD displayed generally good absolute and relative rises in
all test parameters, although relative rises were greatest for FVIII:C and VWF:CB,
and CB/Ag ratio increases overshadowed those for RCo/Ag; (2) type 2A VWD patients
showed good absolute and relative rises in both FVIII:C and VWF:Ag, but poor absolute
rises in both VWF:CB and VWF:RCo; although small rises in both CB/Ag and RCo/Ag were
also observed, both ratios tended to remain below 0.7; (3) finally, type 2 M VWD patients
generally showed good absolute and relative rises in FVIII:C, VWF:Ag, and VWF:CB,
but a poor absolute and relative rise in VWF:RCo; thus, there were good rises in CB/Ag
ratios but little change in RCo/Ag, which tended to remain below 0.7. We proposed
future multicenter prospective investigations to validate these findings and to investigate
their therapeutic implications.
Another study exploring VWF level and activity (VWF:RCo and VWF:CB) in plasma products
was reported.[173 ] Pérez-Rodríguez et al[174 ] included the VWF:CB in their study of autosomal dominant C1149R VWD, previously
characterized as type 1 VWD, to be recharacterized as type 2A VWD. Varadi et al[175 ] included the VWF:CB in their study of species-dependent variability of ADAMTS13-mediated
proteolysis of human recombinant VWF. Udvardy et al[176 ] included the VWF:CB (and VWF:RCo) in their evaluation of densitometric curves of
VWF multimers to describe the degree of multimerization. We published a study exploring
the potential supplementary utility of combined PFA-100 and functional VWF testing
for the laboratory assessment of DDAVP and VWF concentrate therapy in VWD.[177 ] In brief, both therapies tended to normalize VWF test parameters and PFA-100 CTs
in type 1 VWD, with the level of correction in CTs related to the level of normalization
of VWF, particularly the VWF:CB. However, although occasional correction of CTs was
observed in type 2A or type 2M VWD, these did not in general normalize PFA-100 CTs,
either with DDAVP or VWF concentrate therapy. In these patients, improvement in CTs
was more likely in those in whom VWF:CB values normalized or when VWF:CB/VWF:Ag ratios
normalized. This study confirmed the strong relationship between the presenting levels
of plasma VWF, especially the VWF:CB, and PFA-100 CTs, and that the supplementary
combination of PFA-100 and VWF:CB testing might provide added clinical utility to
current broadly applied testing strategies limited primarily to VWF:Ag, VWF:RCo, and
FVIII:C. Burgess et al[178 ]
[179 ] published two studies utilizing the VWF:CB and PFA-100 to investigate dog VWD.
The decade ended with a landmark study in which Riddell et al[180 ] reported on the characterization of W1745C and S1783A as novel mutations causing
defective collagen binding in the A3 domain of VWF. In other words, these cases could
be classified as VWF:CB defective type 2M VWD cases. Their findings demonstrated that
mutations causing an abnormality in the binding of VWF to collagen may contribute
to clinically significant bleeding symptoms, and they proposed that isolated collagen-binding
defects should be classified as a distinct subtype of VWD.
The Next Decade—The 2010s
Fuchs et al[181 ] reported on novel flow-based measurements of VWF function, namely, binding to collagen
and platelet adhesion under physiological shear. We published a study that included
VWF:CB to investigate short-activated partial thromboplastin times (APTTs).[182 ] Mori et al[183 ] evaluated VWF activity in FVIII/VWF concentrates with a newly described automated
VWF activity (IL) test (i.e., VWF:Ab). The assay had good correlation with values
determined by VWF:Ag, VWF:RCo, and VWF:CB. However, comparison of means using the
Wilcoxon test proved no significant variation with VWF:Ag, but displayed a slight
variation with VWF:RCo and VWF:CB. Bongers et al[184 ] included the VWF:CB in a study investigating reduced ADAMTS13 in children with severe
meningococcal sepsis, and its association with severity and outcome. In the acute
phase, both ADAMTS13 activity and antigen were decreased and VWF:CB and VWF:Ag levels
were strongly increased. ADAMTS13 activity and VWF:CB were both correlated with the
severity of the disease. Wang et al[185 ] evaluated a new method for detecting ADAMTS13 activity alongside their historical
VWF:CB. Casonato et al[186 ] used the VWF:CB as a surrogate for VWF multimers to explore the reduced survival
of type 2B VWF, which occurred irrespective of large multimer representation or thrombocytopenia.
Wang et al[187 ] reported another study of ADAMTS13 activity utilizing a VWF:CB. Crow et al[188 ] and Meyer et al[189 ] utilized the VWF:CB to investigate AVWS in VAD recipients. I published my evaluation
of seven commercial VWF:CB assays for the discrimination of types 1 and 2A/2B VWD.[22 ] In brief, these VWF:CB assays varied in their ability to discriminate types 1 and
2A/2B VWD, and the optimal VWF:CB/Ag ratio at which optimal discrimination occurred
which also differed between assays, with some improvements observed with some (but
not all) assays following a harmonization process that aimed to correct for different
calibrator effects. Assay variability also compromised assay utility in some test
occasions. I suggested that future standardization and improvements in some commercial
VWF:CB assays are needed before the VWF:CB assay could be more fully and globally
utilized for discrimination of VWD types in diagnostic laboratories. Frontroth et
al[190 ] included a VWF:CB in their prospective study of low-dose ristocetin-induced platelet
aggregation (RIPA) to identify type 2B VWD and PT-VWD in children. van Loon et al[191 ] included the VWF:CB in a study investigating the effect of genetic variations in
syntaxin-binding protein-5 and syntaxin-2 on VWF concentration and cardiovascular
risk. Some alleles were associated with higher VWF plasma levels and activity, whereas
others were associated with lower VWF plasma levels and activity. Shapovalov and Vitkovskiĭ[192 ] investigated VWF:CB and platelet aggregation in Russian patients with frostbite.
In 2011, another study relating VWF activity including VWF:CB to the risk for cardiovascular
disease was reported by van Schie et al[193 ]; this one related to variations in the VWF gene. Chen et al[194 ] investigated the possibility that N-acetylcysteine (NAC), which reduces the size
and activity of VWF in human plasma and mice, could represent a feasible therapy for
TTP. In vitro, NAC reduced soluble plasma-type VWF multimers in a concentration-dependent
manner and rapidly degraded ultra large VWF multimer strings extruded from activated
endothelial cells. NAC also inhibited VWF-dependent platelet aggregation and collagen
binding. Qin et al[195 ] investigated the phenotype and genotype analysis of three Chinese pedigrees with
VWD utilizing the VWF:CB as part of a VWF test panel. Karger et al[196 ] included a VWF:CB to characterize a patient with type 2B-like AVWS. Yango et al[197 ] used the VWF:CB as part of a VWF test panel to investigate hypothyroidism, and showing
reduction in VWF:Ag, VWF:CB, and FVIII:C in these patients. Solomon et al[198 ] utilized the VWF:CB to investigate type 2A-like AVWS caused by aortic valve disease,
which was corrected following valve surgery. Steinlechner et al[199 ] used the VWF:CB and the PFA-100 to show that patients with severe aortic valve stenosis
(AVS) and impaired platelet function benefit from preoperative desmopressin infusion.
Tauer et al[200 ] utilized the PFA-100 and a VWF:CB to evaluate the DDAVP effect on primary hemostasis
in pediatric patients with aspirin-like defect as hereditary thrombocytopathy. Topf
et al[201 ] reported an evaluation of a modified thromboelastography assay for the screening
of VWD, using patients previously diagnosed with VWD using a test panel including
the VWF:CB. Sosothikul et al[202 ] included the VWF:CB in their study of reference values for thrombotic markers in
children.
Chandler et al[27 ] reported on VWF assay proficiency testing from the perspective of the North American
Specialized Coagulation Laboratory Association (NASCOLA). They reported increasing
use of the VWF:CB in their participant laboratories. Overall interpretation error
rates ranged from 3% for normal samples, 28% for type 1 VWD, to a staggering 60% for
type 2 VWD. Type 2 VWD samples were correctly identified by all laboratories using
VWF:CB/Ag ratios but by only one-third of laboratories using VWF:RCo/Ag or VWF:Ab/Ag
ratios. In 2009, only 27% (12/45) of laboratories performed VWF multimer analysis,
with error rates ranging from 7 to 22%. Lotta et al[203 ] reported on the platelet reactive conformation and multimeric pattern of VWF in
acquired TTP during acute disease and remission. VWF:Ag, VWF:RCo, VWF:CB, and VWF
multimeric pattern were investigated in patients with acquired TTP during acute disease,
remission or both. VWF:Ag was higher in TTP patients than in controls. Larger VWF
multimers were frequently lacking in acute TTP patients, who displayed ultra-large
multimers at remission. The degree of loss of larger VWF multimers correlated with
the degree of abnormality of hemoglobin, platelet counts, and serum lactate dehydrogenase
(LDH) and was associated with low levels of both VWF:RCo/Ag and VWF:CB/Ag ratios.
Pérez-Rodríguez et al[204 ] included VWF:CB testing in their study of AVWS and mitral valve prosthesis leakage.
FVIII:C, VWF:RCo, and VWF:CB were all considerably elevated before surgery, with disproportionate
VWF:RCo/Ag and VWF:CB/Ag ratios seen along with the loss of large VWF multimers. Following
surgery, all parameters were markedly increased and the ratios and multimeric VWF
profile became normal. Heilmann et al[205 ] also used the VWF:CB as part of a panel to assess AVWS as an early-onset problem
in VAD patients. Diagnosis of AVWS was based on reduced VWF:RCo/Ag, VWF:CB/Ag, and
multimeric analysis. No patient had an AVWS prior to VAD implantation. An AVWS was
identified already in the very early postoperative period, that is, in almost all
patients on the first day and in all patients on the third day. The AVWS was also
detected in the majority of patients in the further course. Nine of all 17 patients
suffered bleeding complications and required a total of 25 interventions due to hemorrhages.
Dieckmann et al[206 ] used the VWF:CB to provide evidence for acute vascular toxicity of cisplatin-based
chemotherapy in patients with germ cell tumor. Levels of both VWF:Ag and VWF:CB increased
significantly upon initiation of therapy, suggesting early vascular toxicity and endothelial
cell activation.
In 2012, Heilmann et al[207 ] utilized the VWF:CB to help investigate AVWS in patients with extracorporeal life
support (ECLS). They analyzed 32 patients with ECLS and 19 of them without support.
They used VWF:RCo/Ag and VWF:CB/Ag ratios to diagnose AVWS, alongside multimeric analysis.
Reduced VWF:RCo/Ag ratios were identified in 28 ECLS patients, while reduced VWF:CB/Ag
ratios were identified in 31 patients; HMWMs were missing in the same 31 patients.
Thus, 31/32 ECLS patients presented with AVWS, and 22/32 patients suffered from bleeding
complications. AVWS was not detectable in any analyzed patient without support. Choi
et al[208 ] utilized a VWF:CB to help characterize a Korean patient with type 2A VWD and a Gly1609Arg
missense mutation in the VWF gene. Weiss et al[209 ] used high-resolution multimer analysis, the VWF:CB and the PFA-100, to detect type
2A VWD in patients previously showing a normal VWF:RCo/Ag. Czúcz et al[210 ] utilized a VWF:CB as part of a panel of tests to investigate endothelial cell function
in patients with hereditary angioedema.
We published our evaluation into the differential sensitivity of VWF “activity” assays
(VWF:RCo, VWF:CB, VWF:Ab) to large and small VWF molecular weight forms.[211 ] The data showed that the VWF:CB and VWF:RCo assays had higher sensitivity to the
loss of HMW VWF than did the VWF:Ab assay. Moreover, within-method analysis identified
better HMW VWF sensitivity of some VWF:CB assays than of others, with all VWF:CB assays
still showing better sensitivity than the VWF:Ab assay. Differences were also identified
between VWF:RCo methodologies on the basis of either platelet aggregometry or as performed
on automated analyzers. We felt that our results had significant clinical implications
for the diagnosis of VWD and monitoring of its therapy, as well as for the future
diagnosis and therapy monitoring of TTP. Flood et al[212 ] published their findings using a type VI collagen VWF:CB assay. Healthy controls
and index VWD cases were analyzed for VWF:Ag, VWF:RCo activity, and VWF:CB with types
I, III, and VI collagen. VWF gene sequencing was performed for all subjects. Two healthy controls and one type
1 VWD subject were heterozygous for an A1 domain sequence variation, R1399H, and displayed
a selective decreased binding to type VI collagen but not types I and III. Expression
of recombinant 1399H VWF resulted in absent binding to type VI collagen. Two other
VWF A1 domain mutations, S1387I and Q1402P, displayed diminished binding to type VI
collagen. An 11 amino acid deletion in the A1 domain also abrogated binding to type
VI collagen. The authors concluded that the VWF:CB may be useful in diagnosis of VWD,
as a decreased VWF:CB/VWF:Ag ratio may reflect specific loss of collagen-binding ability.
However, mutations that exclusively affect type VI collagen binding may be associated
with bleeding, yet missed by current testing. Flood et al[213 ] also undertook a comparison of type I, type III, and type VI VWF:CB assays in the
diagnosis of VWD. The mean VWF:CB in healthy controls was similar and highly correlated
for types I, III, and VI collagen (means: 1.2–1.3). In type 1 VWD subjects, VWF:CB
was also similar to VWF:Ag with similar mean VWF:CB/VWF:Ag ratios for types I, III,
and VI collagen (means: 1.1–1.3). For type 2A and 2B subjects, VWF:CB was uniformly
low, with mean ratios of 0.62 and 0.7 for type I collagen, 0.38 and 0.4 for type III
collagen, and 0.5 and 0.47 for type VI collagen. The authors concluded that the low
VWF:CB in type 2A and 2B subjects suggested that VWF:CB may also supplement analysis
of multimer distribution.
Koyama et al[214 ] explored VWF test patterns (VWF:Ag, VWF:CB, VWF multimers) and ADAMTS13 in patients
with obstructive sleep apnea (OSA). Hassan et al[215 ] explored potential VWD among 30 Malay patients with menorrhagia. Using the APTT,
FVIII:C, VWF:Ag, VWF activity (VWF:Ab), and VWF:CB, VWD was diagnosed in four patients
(13.3%), three with type 1 and one with type 2M VWD. Jiang et al[216 ]
[217 ] included the VWF:CB in their evaluation of VWD among their Chinese cohort of patients.
Wiegand et al[218 ] included VWF:CB to help investigate the bleeding diathesis in 15 patients with Noonan
syndrome. Nine patients displayed a relevant bleeding diathesis and complained of
easy bruising; three reported spontaneous gum bleeding. Five patients had pulmonary
valve stenosis, three of who had loss of HMWM and reduced VWF:CB, suggesting AVWS.
Montilla et al[219 ] used a VWF:CB to help investigate the effect of polyphosphate on VWF activity. We
published a novel VWF:CB method using flow cytometry.[220 ] Al-Awadhi et al included a VWF:CB in their assessment of smoking, VWF, and ADAMTS13
in healthy males.[221 ] Acute smokers had significantly higher levels of VWF:CB activity and ADAMTS13 antigen
and activity levels compared with smokers at rest.
In 2013, Legendre et al[222 ] identified several mutations in the A3 domain of VWF that induced combined qualitative
and quantitative defects in the protein, including decreased binding to types I and
III collagen. I helped to establish and characterize a novel VWF:CB assay for the
measurement of VWF activity.[223 ] Soares et al[224 ]
[225 ] employed a VWF:CB to help characterize ADAMTS13 and VWF level and activity in children
with cyanotic congenital heart disease. Hugenholtz et al[226 ] included a VWF:CB to highlight an unbalance between VWF and ADAMTS13 in acute liver
failure; in these patients, VWF:Ag was raised, but VWF:CB and VWF:RCo/Ag ratio was
reduced, as was ADAMTS13 and the proportion of HMWM. Colombatti et al[227 ] included the VWF:CB in a study of coagulation activation in children with sickle
cell disease, as associated with cerebral small vessel vasculopathy. Hu et al[228 ] used a VWF:CB to help investigate nonsurgical bleeding in heart failure patients
supported by continuous-flow left VAD. Kraisin et al[229 ] used a VWF:CB to evaluate reduced ADAMTS13 activity in a malaria-like model. Mazur
et al[230 ] included a VWF:CB to explore postoperative draining and blood product usage with
coronary artery bypass grafting.
In 2014, van Loon et al[231 ] included the VWF:CB to study performance-related factors as the main determinants
of the VWF response to exhaustive physical exercise. VWF:Ag, VWF:CB, and ADAMTS13
activity all increased after exhaustive exercise, with this increase strongly determined
by physical fitness level and the intensity of the exercise. Rau et al[232 ] used a VWF:CB to investigate VWF activity in a model of in vitro hemodilution. Kajdácsi
et al[233 ] used a VWF:CB to help study endothelial cell activation during edematous attacks
of hereditary angioedema types I and II. Levels of several endothelial cell activation
markers, including VWF:Ag and VWF:CB, significantly increased during such attacks.
Goel et al[234 ] used the VWF:CB to help measure ADAMTS13 activity, alongside other methods, to investigate
ADAMTS13 deficiency in patients with noncirrhotic portal hypertension. I wrote a historical
review on VWD, including the VWF:CB and other VWF assays.[10 ] Mancini et al[235 ] utilized the VWF:CB in addition to a FRETS-VWF73 assay to show that the FRETS-VWF73
better reflects ADAMTS13 proteolytic activity in acquired TTP. Quiroga et al[236 ] utilized the VWF:CB as part of a panel of tests to investigate type 1 VWD in Chile.
We reported the next in our series of RCPQAP EQA for VWF/VWD.[33 ] In brief, a set of 29 plasma samples comprising both “quantitative” VWF deficiency
(“type 1 and 3 VWD”) versus “qualitative” defects (“type 2 VWD”) were tested in a
cross-laboratory setting. Different VWF assays and activity/antigen ratios show different
utility in VWD and type identification. VWD identification errors were often linked
to high inter-laboratory test variation and result misinterpretation (i.e., laboratories
failed to correctly interpret their own test panel findings). Moderate quantitative
VWF-deficient samples were misinterpreted as qualitative defects on 30/334 occasions
(9% error rate); 17% of these errors were due to laboratories misinterpreting their
own data, which was instead consistent with quantitative deficiencies. Conversely,
while qualitative VWF defects were misinterpreted as quantitative deficiencies at
a similar error rate (∼9%), this was more often due to laboratories misinterpreting
their data (∼50% of errors). For test-associated errors, VWF:RCo was associated with
the highest variability and error rate, which was at least twice that using VWF:CB.
Waldow et al[237 ] used a VWF:CB, the PFA-100, and multimeric analysis to help explore AVWS in adult
patients with congenital heart disease, and Loeffelbein et al[238 ] utilized a VWF:CB to help explore shear-stress induced AVWS in children with congenital
heart disease with and without stenosis; here, VWF:CB was lower in the stenosis group
as was the VWF:CB/Ag; after intervention, VWF parameters normalized rapidly within
the first 24 hours after the procedure and showed no group difference. Our laboratory
published a study comparing several automated VWF antigen and activity assays.[14 ] We included a large sample test set (n = 600), and also evaluated DDAVP responsiveness plus differential sensitivity to
high-molecular-weight VWF. We reported that VWF:Ag results from different methods
were respectively largely comparable, although some notable differences were evident,
including one high false normal VWF:Ag value (105 U/dL) on a type 3 VWD sample, possibly
due to heterophile antibody interference in the latex-based CS-5100 methodology. VWF:RCo
versus VWF:GPIbR was also largely comparable. VWF:GPIbM was largely comparable to
VWF:RCo and GPIbR, but VWF:CB showed discrepant findings to VWF:RCo, VWF:GPIbR, and
VWF:GPIbM with some patients, most notably patients with type 2M VWD. We concluded
(1) VWF:Ag on different platforms were largely interchangeable, as were VWF:RCo versus
VWF:GPIbR on different platforms, although occasional (some potentially important)
differences may be present, and manufacturer-recommended methods may otherwise require
some assay optimization; (2) VWF:RCo, VWF:GPIbR, and VWF:GPIbM were also largely interchangeable,
except for occasional differences that may also relate to assay design (differing
optimizations); (3) VWF:CB provides an additional activity to supplement VWF:RCo,
VWF:GPIbR, and VWF:GPIbM activity assays, and is therefore not interchangeable with
any of the VWF:GPIbB assays.
In 2015, Kalbhenn et al[239 ] included the VWF:CB in an evaluation of AVWS due to extracorporeal membrane oxygenation
(ECMO) support. The diagnosis of AVWS was based on the VWF:CB/Ag ratio and VWF multimeric
analysis. Bleeding episodes were also monitored. All 18 patients supported with ECMO
developed AVWS, which was identified within 24 hours of ECMO implantation. In 17/18
patients, bleeding complications arose requiring transfusions of blood, fresh frozen
plasma (FFP), and/or platelet concentrates. The AVWS was reversed after ECMO explantation.
The authors concluded that making an early diagnosis of AVWS and providing appropriate
treatment may reduce the incidence of life-threatening bleeding in ECMO. Bürgin-Maunder
et al[240 ] utilized the VWF:CB (with VWF:Ag, and VWF multimers) to assess whether moderate
dietary supplementation with omega-3 fatty acids impacted plasma VWF profile in mildly
hypertensive subjects. These subjects often have raised VWF levels, but supplementation
did not impact VWF level or activity. Caspar et al[241 ] included a VWF:CB to evaluate the effects of transcutaneous aortic valve implantation
(TAVI) on aortic valve disease–related hemostatic disorders involving VWF. They prospectively
enrolled 49 consecutive patients with severe AVS addressed for TAVI. At baseline,
a significant link between VWF abnormalities and the severity of AVS was evident,
and negatively associated with VWF:Ag, VWF:RCo, and VWF:CB, and consistent with AVWS
(VWF:CB/Ag < 0.7). One week after TAVI, all VWF levels increased and VWF:CB/Ag normalized.
Hugenholtz et al[242 ] investigated ADAMTS13 and VWF level and activity (VWF:RCo, VWF:CB, and VWF multimers)
during lung transplantation, and reported an unbalance promoting the development of
hyperactive primary hemostasis. Ferhat-Hamida et al[243 ] described the contribution of the VWF:CB in the range of tests for the diagnosis
and classification of VWD. They reported that a comparison between VWF:CB and VWF:RCo
showed good correlation for all types of VWD except for type 2, while comparison between
VWF:CB and multimer pattern showed good concordance for all types of VWD. They concluded
that the VWF:CB could be a good alternative to VWF:RCo for the diagnosis of quantitative
deficiencies of VWF and could also replace VWF multimers. van Meegeren et al[244 ] included the VWF:CB in their panel of VWF tests to help phenotype genetically confirmed
type 2N VWD patients. Bartoli et al[245 ] included the VWF:CB to help assess the effect of ADAMTS13 inhibition on VWF activity
using doxycycline during supraphysiological shear stress.
In 2016, Periayah et al[246 ] included the VWF:CB to evaluate VWD in Malaysia. Qu et al[247 ] included the VWF:CB to assess the diagnostic value of plasma levels, activities,
and their ratios of VWF and ADAMTS13 in patients with cerebral infarction. They found
an association between reduced levels of VWF:CB/Ag, ADAMTS13/VWF:Ag, and ADAMTS13/VWF:RCo
ratios and cerebral infarction. They suggested that increased levels of VWF and reduced
levels of ADAMTS13 activity may contribute to the pathogenesis of cerebral infarction.
Feys et al[248 ] included the VWF:CB in their study investigating the potential contribution of VWF:GPIbα
interactions to persistent aggregate formation in apheresis platelet concentrates.
Our laboratory reported our evaluation of the three VWF test panel using the chemiluminescent-based
assay AcuStar system for identification of, and therapy monitoring in, VWD.[249 ] This test system was compared with previously evaluated and validated test systems
including VWF:RCo on CS-5100, the new Siemens VWF:GPIbM on CS-5100, and VWF:Ag and
VWF:CB assays performed by automated ELISA. We employed a large total sample test
set (n = 535) including plasma samples from individuals with and without VWD, some on treatment,
normal plasmas, and normal and pathological controls. We also evaluated DDAVP responsiveness,
plus differential sensitivity to reduction in HMWM. The chemiluminescent test panel
(VWF:Ag, VWF:RCo, VWF:CB) showed good comparability to similar assays performed by
alternate methods, and broadly similar data for identification of VWD, provisional
VWD-type identification, DDAVP and VWD therapy, and HMWM sensitivity, although some
notable differences were evident. The chemiluminescent system showed best low-level
VWF sensitivity, and lowest interassay variability, compared with all other systems.
Nagy et al[250 ] included the VWF:CB to help determine that circulating osteoprotegerin levels are
associated with non–O blood groups. Muthiah et al[251 ] utilized the VWF:CB to help evaluate whether longitudinal changes in hemostatic
parameters and reduced pulsatility contributed to nonsurgical bleeding in patients
with centrifugal continuous-flow left VADs. Bleeding events occurred in 14/28 (50%)
patients. VWF profile impairment (VWF:CB/Ag < 0.8) was demonstrated in 89% of patients
at D30, with subsequent recovery but further deterioration after D180. Bleeding was
associated with elevated preimplant sGPVI, and pulsatility was associated with higher
VWF:CB/Ag and a trend to less bleeding. The residual VWF:CB level continued to be
used in China as a measure of ADAMTS13 activity, as reported by Sun et al,[252 ]
[253 ] for various patient groups with prothrombotic status (atherosclerosis, diabetes,
acute promyelocytic leukemia, cancer and sepsis, for a total of 260 cases) and in
patients with hematologic malignancies before and after treatment. de Maat et al[254 ] included the VWF:CB in their investigation into the biological variation of hemostasis
variables in thrombosis and bleeding.
In 2017, I published my third review focused on the VWF:CB,[255 ] as well as a methodological study.[20 ] Ezigbo et al[256 ] included the VWF:CB to help characterize VWD in the Nigerian population. Zhang et
al[257 ] included the VWF:CB in their study of VWF level and activity in healthy Chinese.
Heilmann et al[258 ] included the VWF:CB in their study of AVWS in 74 patients on long-term support with
the VAD HeartMate II. Abnormally low values of VWF:RCo/Ag and VWF:CB/Ag were found
in 69 and 97% of blood samples, respectively. Only 10/181 multimer analyses showed
a normal pattern. The VWF:CB/Ag ratio correlated with the multimer patterns, whereas
the VWF:RCo/VWF:Ag ratio seemed to be less sensitive for AVWS. Wang et al[259 ] included the VWF:CB in their evaluation into ABO blood group, age and gender on
plasma FVIII:C, fibrinogen, VWF, and ADAMTS13 in a Chinese population. Frank et al[260 ] studied 21 consecutive patients with AVS before and 6 to 18 months after valve surgery.
They assessed PFA-100 CTs, FVIII:C, VWF multimers, VWF:Ag, VWF:RCo, VWF:CB, and VWF:CB/Ag
ratio. Large VWF multimers were strongly reduced in all patients with AVS, while all
controls had normal multimers. Collagen/ADP (C/ADP) CTs were prolonged in patients
with AVS compared with the controls and the VWF:CB/Ag ratio was pathological in 20/21
patients but normal in all controls. After surgery, VWF multimers normalized in all
patients, C/ADP CTs shortened, and VWF:CB/Ag ratio strongly improved, normalizing
in 14 of 17 patients. Thus, all consecutive patients with severe AVS had an AVWS,
and the combination of C/ADP CT and VWF:CB/Ag ratio detected the AVWS in all patients.
Kumar et al[261 ] included the VWF:CB in their evaluation of plasma ADAMTS13 activity (reduced) and
VWF antigen and activity (both elevated) in patients with subarachnoid hemorrhage.
Cluster analysis also demonstrated that patients with higher VWF:Ag and VWF:CB and/or
lower ADAMTS13 activity might be at risk of increased mortality. Cibor et al[262 ] investigated levels and activities of VWF and ADAMTS13 in inflammatory bowel diseases
(IBDs). They concluded that complex VWF-ADAMTS13-mediated mechanisms disturbed hemostasis
in IBD; a reduced VWF:CB was a risk factor for bleeding, while a lower ADAMTS13 level
combined with an elevated VWF:Ag could predispose to thrombosis. Doruelo et al[263 ] included the VWF:CB in their clinical and laboratory study into phenotype variability
in type 2M VWD. Michiels et al[264 ] included the VWF:CB in their study into VWD and DDAVP. Chan et al[265 ] included the VWF:CB in their study on shear stress-induced total blood trauma in
multiple species. Lavin et al[266 ] included the VWF:CB in their study into the clinical phenotype and pathophysiology
underlying low VWF levels.
In 2018, Stufano et al[267 ] reported their comparative evaluation of the AcuStar chemiluminescence VWF:CB to
the Stago ELISA VWF:CB. Geisen et al[268 ] included the VWF:CB in their investigation into platelet secretion defects and AVWS
in patients with VAD. All 198 VAD patients developed AVWS. As soon as the VAD was
explanted, the AVWS disappeared within hours. AVWS was less severe in the HeartMate
III patients than in the HeartMate II patients. The HeartMate III patients also had
fewer bleeding symptoms. Jousselme et al[269 ] also undertook a comparison of the automated AcuStar chemiluminescent assay versus
the Stago ELISA assay using VWD plasma from patients previously diagnosed through
VWF molecular analysis. Discrepancies of VWF:CB/Ag ratio were observed in type 2M–2A-like
VWD, a finding that we reported in 2016.[34 ] The residual VWF:CB assay was still being used in China to determine ADAMTS13 activity.[270 ] Alharbi et al[271 ] included the VWF:CB in their assessment of ABO blood group effect on VWF tests in
healthy Saudi blood donors. Russell et al[272 ] included the VWF:CB in their evaluation of coagulopathy, endothelial cell damage,
and mortality after severe pediatric trauma, showing lowest plasma ADAMTS13 activity
in patients who died from their injuries. Zayat et al[273 ] used the VWF:CB in their evaluation of survival in HeartMate II patients. Sherazi
et al[274 ] included the VWF:CB in their prospective analysis of bleeding events in left VAD
patients. We published our findings on the differential sensitivity of different VWF
activity assays to reduced VWF molecular weight forms,[275 ] in essence an update of the study we reported in 2012.[211 ] In brief, sensitivity for reduction of HMW was highest for VWF:CB and VWF:GPIbM,
intermediate for VWF:RCo and VWF:GPIbR, and lowest for VWF:Ab, findings similar to
those we previously reported.[211 ] We again felt that our results held significant clinical implications for diagnosis
and therapy monitoring of VWD, as well as potential future diagnosis and therapy monitoring
of TTP. Atiq et al[276 ] included the VWF:CB in their study on comorbidities associated with higher VWF levels,
and the age-related increase of VWF in VWD. Torkildsen et al[277 ] included the VWF:CB in their comparison of multiple thawing techniques on thaw time
and stability of hemostatic proteins in canine plasma products. Lasom et al[278 ] included a VWF:CB in their evaluation of coronary stenosis in type 2 diabetes mellitus
patients. Palyu et al[279 ] included the VWF:CB in their study into cirrhotic patients with stable disease or
acute decompensation (AD). VWF:Ag, VWF:RCo, and VWF:CB were elevated in both cirrhotic
groups, with 24/54 AD patients showing presence of ultra-large VWF multimers, with
this also associated with low ADAMTS13 activity and high CRP levels. Oliveira et al[280 ] compared three nonautomated VWF:CB assays, two by ELISA and the other by flow cytometry.
The ELISA assays could differentiate 2A and 2M VWD. Pelland-Marcotte et al[281 ] included the VWF:CB to investigate AVWS in children with idiopathic pulmonary arterial
hypertension. Overall, 8/14 children had mild to moderate bleeding symptoms and/or
laboratory abnormalities in keeping with AVWS. Normalization of the hemostatic defects
following lung transplantation and lack of family history of bleeding attested to
the acquired nature of their defects. Icheva et al[282 ] included the VWF:CB in their study into AVWS in congenital heart disease surgery.
In 2019, Staley et al[283 ] included the VWF:CB in their study into clinical factors and biomarkers that predicted
outcome in patients with immune-mediated TTP. Slobodianuk et al[284 ] reported defective VWF:CB and increased bleeding in a murine model of VWD affecting
collagen IV binding. Moonla et al[285 ] reported the bleeding symptoms and VWF levels in their VWD cohort, representing
a 30-year experience in a tertiary-care center. Low VWF:RCo, VWF:GPIbM, and VWF:CB
were all able to predict increased bleeding risk. Colling et al[286 ] included the VWF:CB in their in vitro assessment of VWF in cryoprecipitate FVIII/VWF
concentrate, and recombinant VWF. Pechmann et al[287 ] included the VWF:Ag and VWF:CB in their study reporting increased VWF parameters
in children with febrile seizures, concluding that especially VWF:CB may serve as
additional biomarker in the diagnosis of febrile seizures. Kumar et al[288 ] included VWF:Ag and VWF:CB (both increased) and ADAMTS13 activity (decreased) in
their study of patients with traumatic brain injury. Yan et al[289 ] reported the establishment of a flow cytometric immunobead assay to detecting plasma
VWF activity (VWF:GPIbR-like assay) and showed its clinical application in the prognosis
of ischemic stroke, as compared with VWF:Ag, VWF:GPIbR, and VWF:CB. Kubicki et al[290 ] included the VWF:CB in their investigation into AVWS in 39 pediatric patients during
mechanical circulatory support (MCS) (extracorporeal life support, n = 13; ECMO, n = 5; and VAD, n = 12). All children developed AVWS during MCS, usually during the early postoperative
course, but no AVWS after device explantation. The authors detected a loss of HMWM,
decreased VWF:CB/Ag ratios, and reduced VWF:CB levels; 20/30 patients experienced
bleeding complications, and 53% of them required surgical revision. There were no
deaths due to bleeding during support. Authors concluded that AVWS prevalence in pediatric
patients on MCS is 100% regardless of the types of devices used, but that the bleeding
propensity varies widely. Coghill et al[291 ] compared several commercially available VAD using a bench-top method and exampled
the effect on various VWF parameters, with loss of VWF:Ag, VWF:CB, and HMWM evident
with their use. Klaeske et al[292 ] investigated AVWS in HeartMate 3 patients compared with HeartWare VAD (HVAD), showing
reduction of HMWM, VWF:Ag, VWF activity, and VWF:CB in patients with both devices,
but that the defects were more severe with the HVAD device. McBride et al[293 ] included the VWF:CB in their investigation into primary hemostatic function in dogs
with acute kidney injury (AKI), and reported a type 2 VWD-like picture (low VWF:CB/Ag)
in these dogs.
