RESEARCH ARTICLE

A JH
Bleeding spectrum in children with moderate or severe von
Willebrand disease: Relevance of pediatric-specific bleeding
Yvonne V. Sanders,1 Karin Fijnvandraat,2 Johan Boender,1 Evelien P. Mauser-Bunschoten,3 Johanna G. van der Bom,4,5
Joke de Meris,6 Frans J. Smiers,7 Bernd Granzen,8 Paul Brons,9 Rienk Y.J. Tamminga,10 Marjon H. Cnossen,11
Frank W.G. Leebeek,1* and the WiN Study Group

The bleeding phenotype of children with von Willebrand disease (VWD) needs to be characterized in detail
to facilitate diagnosis during childhood and aid in the planning and assessment of treatment strategies. The
objective was to evaluate the occurrence, type, and severity of bleeding in a large cohort of children with
moderate and severe VWD. We included 113 children (aged 0–16 years) with Type 1 (n 5 60), 2 (n 5 44), and 3
(n 5 9) VWD with von Willebrand factor (VWF) antigen and/or VWF ristocetin cofactor levels  30 U/dL from a
nation-wide cross-sectional study (“Willebrand in the Netherlands” study). Bleeding severity and frequency
were determined using the International Society on Thrombosis and Hemostasis-Bleeding Assessment Tool
(ISTH-BAT) with supplementary pediatric-specific bleeding symptoms (umbilical stump bleeding,
cephalohematoma, cheek hematoma, conjunctival bleeding, postcircumcision and postvenipuncture
bleeding). We found that all 26 postmenarche girls experienced menorrhagia. Other common bleedings
were cutaneous (81%), oropharyngeal (64%), prolonged bleeding from minor wounds (58%), and epistaxis
(56%). Pediatric-specific bleeding symptoms were present in 44% of patients. ISTH-BAT bleeding score was
higher in index cases than in affected family members (median, 12.0 vs. 6.5, P < 0.001), higher in Type 3 VWD
than in Type 2 or 1 (17.0 vs. 10.5 or 6.5, P < 0.001) and higher in children with severe (<10 U/dL) than moderate
VWD (10–30 U/dL) (11.0 vs. 7.0, P < 0.001). Frequency of any bleeding, epistaxis, and oral cavity was higher in
types 2 and 3 than in Type 1 VWD and was associated with VWF levels. We conclude that pediatric-specific
bleeding symptoms occurred in a large proportion of children with moderate or severe VWD and should be
included when evaluating children for VWD.
Am. J. Hematol. 90:1142–1148, 2015. V
C 2015 Wiley Periodicals, Inc.

䊏 Introduction
von Willebrand disease (VWD) is the most common inherited bleeding disorder and is caused by quantitative or qualitative defects in von Wil-
lebrand factor (VWF) [1,2]. Although patients frequently suffer from mucocutaneous bleeding, clinical expression is heterogeneous, with a large
inter- and intraindividual variability of symptoms [3,4]. In general, the frequency and severity of bleeding is associated with the type of VWD and
the level of functional VWF.
VWD is classified into three types according to the current guidelines of the International Society on Thrombosis and Hemostasis (ISTH) [5].
The most common type (75%), Type 1 VWD, is a quantitative partial deficiency of VWF, which is usually characterized by mild bleeding symp-
toms. Type 2 VWD accounts for 20% of patients and is described as a qualitative defect of VWF. It is subdivided into four categories: 2A, 2B,

Additional Supporting Information may be found in the online version of this article.
1
Department of Hematology, Erasmus University Medical Center, Rotterdam, The Netherlands; 2Department of Pediatric Hematology, Emma Children’s Hospital, Aca-
demic Medical Center, Amsterdam, The Netherlands; 3Van Creveldkliniek/Department of Benign Hematology, University Medical Center Utrecht, Utrecht, The Nether-
lands; 4Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The Netherlands; 5Jon J Van Rood Center for Clinical Transfusion Medicine,
Sanquin Research, Leiden, The Netherlands; 6Netherlands Hemophilia Society, Leiden, The Netherlands; 7Department of Pediatric Hematology, Willem-Alexander
Children’s Hospital, Leiden University Medical Center, Leiden, The Netherlands; 8Department of Pediatric Hematology, Maastricht University Medical Center, Maas-
tricht, The Netherlands; 9Department of Pediatric Hematology, Radboud University Medical Center, Nijmegen, The Netherlands; 10Department of Pediatric Hematology,
University Medical Center Groningen, Groningen, The Netherlands; 11Department of Pediatric Hematology, Erasmus University Medical Center-Sophia Children’s Hos-
pital, Rotterdam, The Netherlands
Conflict of interest: F.W.G. Leebeek received research support from CSL Behring for performing the WiN-study and an unrestricted research grant of Baxter.
Y.V. Sanders has been a teacher on educational activities of Baxter. M.H. Cnossen has received unrestricted research/educational funding for various projects
from the following companies as well as travel fees: Pfizer, Baxter, Bayer Schering Pharma, Novo Nordisk, CSL Behring, and Novartis. K. Fijnvandraat is a
member of the European Hemophilia Treatment and Strategy Board sponsored by Baxter, has received unrestricted research grants from CSL Behring, Novo
Nordisk, and Bayer, and has given lectures at educational symposiums organized by Pfizer, Bayer, Novo Nordisk, and Baxter. E.P. Mauser-Bunschoten received
research/educational support from CSL Behring, Bayer, Baxter, Novo Nordisk, Pfizer, Biovitrum, Griffols, and Sanquin. J.G. van der Bom has received unre-
stricted research/educational funding for various projects from the following companies: Bayer Schering Pharma, Baxter, CSL Behring, Novo Nordisk, and
Pfizer. In addition, she has been a consultant to Baxter and Pfizer, and she has been a teacher on educational activities of Bayer Schering Pharma. None of the
other authors has a conflict of interest to declare. The WiN-study, including the current study, was supported by CSL Behring (unrestricted grant).
*Correspondence to: Frank W.G. Leebeek, Erasmus University Medical Center, PO Box 2040, 3000 CA, Rotterdam, The Netherlands. E-mail: f.leebeek@eras-
musmc.nl
Received for publication: 27 July 2015; Revised: 15 September 2015; Accepted: 15 September 2015
Am. J. Hematol. 90:1142–1148, 2015.
Published online: 16 September 2015 in Wiley Online Library (wileyonlinelibrary.com).
DOI: 10.1002/ajh.24195

C 2015 Wiley Periodicals, Inc.

V

1142 American Journal of Hematology, Vol. 90, No. 12, December 2015 doi:10.1002/ajh.24195
RESEARCH ARTICLE
2M, and 2N, based on functional VWF differences. Although clinical
symptoms vary between patients, in general, the phenotype is more
severe than in Type 1 VWD. Type 3 VWD is rarely diagnosed, but
clinically the most severe type and characterized by a virtually com-
plete absence of VWF.
Because of their young age, children generally lack exposure to
hemostatic challenges and may, therefore, not yet manifest bleeding
symptoms, even when affected with VWD. Because healthy children
may also suffer from mucocutaneous bleeding, such as epistaxis,
bruising, and oropharyngeal bleeding [6–8], bleeding during child-
hood does not necessarily reflect the presence of a bleeding disorder.
Therefore, diagnosing VWD can be especially difficult in younger age
groups [2].
In recent years, bleeding scores (BS) have been developed that
aim to discriminate between individuals with and without VWD by
quantifying the number and severity of various bleeding symptoms
[9–11]. In patients with a known diagnosis, BS can also be used to
assess the bleeding phenotype and can, therefore, be a valuable tool
to aid in the planning and assessment of treatment strategies [12].
For children, a pediatric bleeding questionnaire, including pediatric-
specific bleeding symptoms was developed for this purpose, of
which the ISTH-Bleeding Assessment Tool (ISTH-BAT) is currently
the most recommended [11,13]. Studies have demonstrated the
value of a pediatric BS in diagnosis of mild VWD and mild bleed-
ing disorder, but the tool has never been used to analyze frequency
and severity of bleeding symptoms in children with more severe
VWD [13–15].
Therefore, we assessed the overall bleeding phenotype in a large
cohort of children with moderate and severe VWD from the
“Willebrand in the Netherlands” (WiN) study and evaluated the value
of the ISTH-BAT with regard to this specific group.
䊏 Patients and Methods
Patients. Children from the nation-wide cross-sectional WiN study who had
been diagnosed earlier with Type 1, 2, or 3 VWD were included if the following
inclusion criteria were met: (1) a hemorrhagic diathesis or a family history of
VWD and (2) historically lowest VWF plasma levels  30 U/dL (VWF antigen
[VWF:Ag] and/or VWF ristocetin cofactor) and/or FVIII:C plasma levels  40 U/
dL [4,16,17]. Patients were excluded if they were known to have other hemostatic
disorders. In total, 140 children (aged < 16 years) were included between 2007 and
2009 in the study. This cohort was approached by a trained physician (Y.V.S.)
between 2013 and 2014 to analyze bleeding pattern using the new ISTH-BAT dur-
ing an interview by telephone. Ten of 13 centers participating in the WiN-study
contributed patients; the remainder do not provide care to pediatric patients with
bleeding disorders. The Medical Ethical Committees in all participating centers
approved the study; all participants gave informed consent. This study was also
approved by the ISTH-BAT SSC (chair F. Rodeghiero).
Bleeding questionnaires and assessment methods. The ISTH-BAT systematically
evaluates the severity and frequency of 13 specific bleeding symptoms and
pediatric-specific bleeding symptoms (umbilical stump bleeding, cephalohematoma,
cheek hematoma caused by sucking during breast or bottle feeding, conjunctival
hemorrhage, excessive postcircumcision bleeding, and excessive postvenipuncture
bleeding). Because none of the girls included had been pregnant, we did not score
postpartum hemorrhage in this pediatric cohort. Based on treatment for the specific
bleeding, the bleeding episodes are scored on a scale that ranges from 0 to 4 points.
The cut-off value of the ISTH-BAT for an abnormal BS was recently determined as
3 for children [18].
During the interview, we also collected information on the use of desmopressin
or replacement therapy with clotting factor concentrate. The time investment of the
interview was 15–45 min per child. An overall BS was calculated by adding the
scores for all individual bleeding symptoms. The BS was also calculated by a second
physician (J.B.), who had access to a written transcript of the interviews. If a patient
had multiple pediatric-specific bleeding symptoms, the most severe episode of
pediatric-specific bleeding was scored.
According to the ISTH-BAT, the absence of bleeding symptoms after surgery
or dental extraction were not scored if patients had received prophylactic treat-
ment with desmopressin or replacement therapy before the surgical intervention
or dental extraction [4,19]. However, if bleeding did occur during intervention
despite this prophylactic treatment, we did score this bleeding according to the
ISTH-BAT.
doi:10.1002/ajh.24195
Bleeding Phenotype of Pediatric VWD Patients
The ISTH-BAT incorporates six pediatric-specific bleeding symptoms. For this
study, we also collected information on four additional pediatric-specific bleeding
symptoms (bleeding from fetal scalp electrode, excessive bruising after birth, exces-
sive bleeding after heel prick test, and postvaccination bleeding) to explore the
occurrence and severity of these bleedings and to be sure no pediatric-specific
bleeding symptoms were missed. The additional symptoms were not used in the
calculation of the overall BS.
Although the BS is based mainly on the bleeding severity, the ISTH-BAT also
allows collection of the frequency of nontrivial bleeding symptoms. For this study,
we collected information on the estimated frequency of epistaxis, bruising, bleeding
from minor wounds, hematuria, gastrointestinal bleeding, oral cavity bleeding, mus-
cle hematomas, hemarthrosis, and any bleeding.
Definitions. Moderate VWD was defined as VWF levels (VWF:Ag and/or
VWF:CB and/or VWF:Act) 10–30 U/dL and/or FVIII:C 20–40 U/dL, and severe
VWD as VWF levels (VWF:Ag and/or VWF:CB and/or VWF:Act) 10 U/dL and/
or FVIII:C 20 U/dL. Classification of VWD was based on the current ISTH crite-
ria and was done as previously reported in the WiN-study [4,5,20,21]. Patients
were classified as moderate or severe VWD based on centrally measured VWF and
FVIII levels. If blood was not obtained centrally, historically measured VWD
parameters were used [4]. Because of restrictions imposed by the medical ethical
committee, we were allowed to obtain a blood sample from children only if veni-
puncture was performed for other clinical purposes by the treating physician. A
blood sample for central analysis of the FVIII/VWF parameters was obtained in
42% of children.
The first patient in a family referred to a medical specialist because of hemor-
rhagic diathesis and diagnosed with VWD was classified as the index case. A
patient diagnosed with VWD because of a previously diagnosed family member
with VWD was classified as an affected family member.
The time span between first bleeding and time of first consultation for bleeding
was defined as “patient’s delay.” The time span between time of first consultation
for bleeding and diagnosis was defined as “doctor’s delay.”
Laboratory measurements of VWD. Patients’ historically measured VWF param-
eters were determined earlier in local Hemophilia Treatment Centers. Blood collec-
tion and laboratory measurements were performed as described previously [4]. All
VWD assays, including VWF:Ag, VWF:CB, VWF:Act, FVIII:C, and multimer anal-
ysis were performed centrally at the Erasmus University Medical Center, Rotter-
dam, if a blood sample was available.
Statistical methods. Descriptive statistics for categorical data are presented
as numbers and percentages (n, %) and for continuous variables as median and
25%–75% interquartile ranges (IQR) or ranges in case of age.
The chi-squared test or the Fisher’s exact test was used to compare the preva-
lence of bleeding symptoms between groups. Because BS was skewed, we used
Mann–Whitney U or Kruskal–Wallis tests to analyze differences in BS between
groups. We performed linear regression analysis to model the association between
BS and age. The association between bleeding frequency and VWF levels was calcu-
lated using Spearman’s correlation coefficient. Statistical analyses were performed
with SPSS for Windows, version 21.0 (SPSS Inc., Chicago, IL). A P-value <0.05
was considered statistically significant.
䊏 Results
Patients
Between 2007 and 2009, 140 children (aged 0–16 years) with VWD
were included in the WiN-study. All parents or caretakers of these chil-
dren were approached separately by telephone for an additional inter-
view between March 2013 and May 2014. Twenty-seven of these
children could not be included in this study because of different reasons
(unreachable and no contact after several attempts in n 5 20, declined in
n 5 5, and language barrier in n 5 2). The majority (82%) of the inter-
views were done with parents only, 13% by the parents together with
the patient, and the remaining 5% with the patient alone. In Table I, the
characteristics of the 113 included children are reported.
Bleeding symptoms and required treatment
All 26 (100%) postmenarche girls had menorrhagia, and a large
proportion (21 of 26, 81%) required treatment (Fig. 1), most fre-
quently by a combination of antifibrinolytics and hormonal therapy
(nine of 26, 35%). Notably, two girls (with Type 2A and Type 3
VWD) had required emergency hospitalization and blood transfu-
sions because of acute excessive menstrual bleeding.
A large proportion of patients (91 of 113, 81%) experienced clini-
cally significant cutaneous bleeding, although in the majority of cases,
American Journal of Hematology, Vol. 90, No. 12, December 2015 1143
Sanders et al.
TABLE I. Baseline Patient and Laboratory Characteristics (n 5 113)
Characteristics
Age (years), median (range)
Male sex, n (%)
VWD type, n (%)
1 60 (53)
2 44 (39)
2A
2B
2M
2N
3 9 (8)
Blood group O, n (%)a
Index cases, n (%)
Centrally measured VWF:Ag (IU/dL)b
Centrally measured VWF:CB (IU/dL)b
Centrally measured VWF:Act (IU/dL)b
Centrally measured FVIII:C (IU/dL)b
Levels are presented as median (25%–75% interquartile range).
a
n 5 79 based on availability; 34 missing.
b
n 5 48, based on patients for whom plasma was available centrally and
patients who used desmopressin or replacement therapy 72 hrs before
blood sampling were excluded.
VWD, von Willebrand disease; VWF:Ag, von Willebrand factor antigen;
VWF:CB, von Willebrand factor collagen binding; VWF:Act, von Willebrand
factor activity; FVIII:C: factor VIII coagulant activity.
no specific bleeding treatment was required (68 of 91, 75%) (Fig. 1).
Frequently observed symptoms were oropharyngeal bleeding (72 of
113, 64%), prolonged bleeding from minor wounds (65 of 113, 58%),
and epistaxis (63 of 113, 56%). Despite having received prophylactic
treatment with replacement therapy, bleedings were still reported after
60% (nine of 15) of surgeries and 38% (six of 16) of dental
extractions.
Applied therapy included antifibrinolytics, surgical hemostasis, des-
mopressin, and replacement therapy (Fig. 1).
Bleeding symptoms and required treatment according
to type of VWD
Epistaxis was more frequent in Type 2 and 3 VWD than in
Type 1 VWD patients. Hemarthrosis and oropharyngeal bleeding
were most prevalent in Type 3 VWD patients (Supporting Informa-
tion Fig. S1A). The proportion of epistaxis, oropharyngeal bleeding,
joint bleeds, and cutaneous bleeding requiring treatment differed sig-
nificantly between Type 1, 2, and 3 VWD (Supporting Information
Fig. S1B).
Compared with affected family members, index cases suffered
more frequently from epistaxis (26 of 35, 74% vs. 37 of 78, 47%,
P 5 0.008), prolonged bleeding from minor wounds (25 of 35, 71%
vs. 40 of 78, 51%, P 5 0.045), postsurgical bleeding (nine of 18, 50%
vs. six of 29, 21%, P 5 0.036), gastrointestinal bleeding (seven of 35,
20% vs. four of 78, 5%, P 5 0.033), and joint bleeds (10 of 35, 29%
vs. 10 of 78, 13%, P 5 0.043). The need for treatment for bleeding
symptoms did not differ between index cases and affected family
members (data not shown).
Prevalence of pediatric-specific bleedings and required
treatment
Pediatric-specific bleeding symptoms included in the ISTH-BAT
occurred in 44% (50 of 113) of patients, occurring more frequently in
Type 3 VWD patients than Type 1 or 2 patients (78% vs. 48% or
32% (P 5 0.028) (Table II). This did not differ between index cases
and affected family members: 19 of 35, 54% vs. 31 of 78, 40%,
P 5 0.150. The proportion of patients with at least one pediatric-
1144 American Journal of Hematology, Vol. 90, No. 12, December 2015
RESEARCH ARTICLE
specific bleeding symptom increased to 52% (59 of 113) when the
additional pediatric-specific bleedings were included.
The most frequently reported pediatric-specific bleeding was post-
13 (4–22) venipuncture bleeding (38 of 113, 34%), followed by postvaccination
68 (60) bleeding (33 of 113, 29%) and postcircumcision bleeding (one of
four, 25% of patients who underwent circumcision). A cephalohema-
toma was reported in 10 of 113 (9%) of children. Five (50%) of them
33
were born by vacuum delivery, one by emergency caesarean section,
10 and four spontaneously. After being born by vacuum delivery, 71% of
1 children (five of seven) had a cephalohematoma (P < 0.001).
0
Age at presentation in children with VWD
46 (58)
35 (31) The first bleeding occurred at a median age of 1.5 years (IQR 0.8–2.0).
21 (9–27) Index cases and patients with Type 3 VWD experienced their first bleed-
9 (4–27) ing episode at the youngest age and were the first to consult a physician
13 (2–27) and diagnosed with VWD (Supporting Information Table S1). Median
39 (23–53)
age at first consultation for bleeding was 3.0 years (1.0–7.0).
Median patient’s delay was 0 years (0–1.0) in index cases in com-
parison with 1.0 year (0–5.5) in affected family members, P 5 0.054.
We could not demonstrate a statistically significant difference in
patient delay between Type 1, 2, and 3 VWD: 1.0 (0–6.0) year vs. 0.3
(0–2.0) year vs. 0.1 year (0–1), respectively, P 5 0.285. The majority
of affected family members had already been diagnosed with VWD
before they first consulted a doctor in relation to a bleeding episode
(78%, 47 of 60 patients who have ever consulted a physician because
of bleeding).
Presenting and diagnosing symptoms
The first bleeding was cutaneous bleeding in the majority of
patients (32 of 101, 32%; 12 missing), followed by epistaxis (31 of
101, 31%) (Supporting Information Fig. S2). This pattern differed
between Type 1, 2, and 3 VWD patients (P 5 0.031). In Type 2 and 3
patients, the first bleeding was mainly from mucocutaneous origin
(68% and 50%), and Type 1 VWD patients mainly presented with
hematomas (45%). A quarter of patients that consulted a doctor for
the first time suffered from epistaxis (26 of 95, 27%; 18 missing),
20% (19 of 95) for oropharyngeal bleeding, and 18% (17 of 95) for
cutaneous bleeding.
Frequency of bleeding symptoms
The distribution of frequency of any bleeding is shown in Fig. 2A.
The frequency of any bleeding was higher in Type 3 VWD than in Type
1 VWD (P 5 0.01), and higher in Type 2 VWD than in Type 1 VWD
(P 5 0.003). Frequency of any bleeding was associated with VWF:Ag
(Spearman’s q 5 20.29, P 5 0.05), VWF:Act (q 5 20.34, P 5 0.02),
VWF:CB (q 5 20.37, P 5 0.01), and FVIII:C (q 5 20.30, P 5 0.04); see
Fig. 2B.
Frequencies of epistaxis and oral cavity bleeding were also higher
in Type 3 and Type 2 VWD than in Type 1 VWD (data not shown).
An association similar to that of any bleeding was observed for oral
cavity bleeding (q 5 20.31 to 20.47, P < 0.05). Epistaxis frequency
was associated with VWF:Act, VWF:CB, and FVIII:C (q 5 20.33 to
20.38, P < 0.05) but not VWF:Ag. Bruising was very frequent, occur-
ring weekly or even daily in the majority of patients, regardless of
VWD type or VWF levels.
Bleeding scores
In the total group of children with VWD, median BS (ISTH-BAT)
was 9.0 (IQR 5.0–12.5). BS did not differ between boys and girls: 9.0
(6.0–11.8) vs. 8.0 (4.0–14.5), P 5 0.963. Type 3 patients had the high-
est BS (17.0 [11.5–25.5]) compared with that of Type 2 patients (10.5
[6.3–12.8]) or Type 1 patients (6.5 [4.0–10.0]), P < 0.001. BS did not
differ between Type 2A and Type 2B VWD. Index cases had a higher
doi:10.1002/ajh.24195
RESEARCH ARTICLE Bleeding Phenotype of Pediatric VWD Patients

Figure 1. Prevalence and treatment of bleeding symptoms. The prevalence (left) and percentage of received treatments (right) for the bleeding symptoms
are shown. Hematuria did not occur in this cohort of children with VWD. The colors of the bars represent the different bleedings scores according to the
ISTH-BAT: in white, a bleeding score of 1 or 2, in gray, a bleeding score of 3, and in black a bleeding score of 4. *Percentage of children that underwent sur-
gery (n5 15) or tooth extraction (n5 16), or percentage of girls who menstruate (n5 26).

BS than affected family members: 12.0 (8.0–15.0) vs. 6.5 (4.0–11.0), children had a BS below the recently determined cut-off (<3) [18].
P < 0.001. Severe VWD patients had a higher BS than moderate These children had a median age of 11.5 and were mainly affected
VWD patients: 11.0 (6.0–15.0) vs. 7.0 (4.0–10.0), P < 0.001. Twelve family members with Type 1 VWD (10 of 12).

doi:10.1002/ajh.24195 American Journal of Hematology, Vol. 90, No. 12, December 2015 1145
Sanders et al. RESEARCH ARTICLE

TABLE II. Prevalence of Pediatric-Specific Bleeding Symptoms

P-value (P-for trend

Total cohort Type 1 VWD Type 2 VWD Type 3 VWD between different
(n5113) (n560) (n544) (n59) types of VWD)

Pediatric items in ISTH-BAT, specified under

the item “Other bleedings”
Umbilical stump bleeding, n (%) 7 (6.2) 6 (10.0) 1 (2.3) 0 (0) 0.197
Cephalohematoma, n (%) 10 (8.8) 5 (8.3) 4 (9.1) 1 (11.1) 0.961
Cheek hematoma caused by sucking, n (%) 5 (4.4) 3 (5.0) 0 (0) 2 (22.2) 0.012
Conjunctival hemorrhage, n (%) 2 (1.8) 1 (1.7) 0 (0) 1 (11.1) 0.070
Postcircumcision bleeding, n (%) 1 (0.9) 0 of 1 (0) 0 of 2 (0) 1 of 1 (100) –
Postvenipuncture bleeding, n (%) 38 (33.6) 23 (38.3) 11 (25.0) 4 (44.4) 0.282
Total no. of children with pediatric-specific 50 (44.2) 29 (48.3) 14 (31.8) 7 (77.8) 0.026
bleeding symptoms, n (%)
Bleeding score for pediatric-specific 0 (0–6) 0 (0–6) 0 (0–3) 1 (0–4) 0.010
bleeding symptoms, median (range)
Additional pediatric-specific bleeding
not incorporated in ISTH-BAT
Postvaccination bleeding, n (%) 33 (29.2) 16 (26.7) 14 (31.8) 3 (33.3) 0.816
Bleeding from fetal scalp electrode, n (%) 2 (1.8) 1 (1.7) 1 (2.3) 0 (0) 0.891
Excessive bruising after birth, n (%) 10 (8.8) 6 (10.0) 3 (6.8) 1 (11.1) 0.827
Excessive bleeding after heel prick test, n (%) 15 (13.3) 10 (16.7) 5 (11.4) 0 (0) 0.347
Total no. of children with all pediatric- 59 (52.2) 30 (50.0) 22 (50.0) 7 (77.8) 0.278
specific bleeding symptoms (ISTH-BAT
plus additional), n (%)

BS increased with age; a 1-year age increase was associated with
0.4-point increase in BS (95% confidence interval [CI] 0.2–0.7). A 1-
year increase in age was associated with a 0.5-point increase in BS
(CI 0.2–0.8) in Type 1 patients, a 0.3 point (CI 20.1 to 0.6) in Type
2 patients, and 0.7 points (CI 21.0 to 2.4) in Type 3 patients.
As described earlier, 50 children had bleeding symptoms in the
additional bleeding symptom category “others” of the ISTH-BAT.
Scores for bleedings in this category, including umbilical stump bleed-
ing, cephalohematoma, cheek hematoma caused by sucking during
breast or bottle feeding, conjunctival hemorrhage, excessive postcircum-
cision bleeding, and postvenipuncture bleeding, ranged from 1 to 4.

Figure 2. Frequency of any bleeding. (A) Proportion of VWD patients who
reported any bleeding less than once a year, once a year to nearly once
every 2 months, once every 2 months to almost every week, and weekly
or more. Bleeding frequency was higher in Type 3 (n5 9, P5 0.01) and
Type 2 (n5 44, P5 0.003) than in Type 1 (n5 60) VWD patients. (B) Associa-
tion between frequency of any bleeding and VWF:Act and FVIII:C in all
patients for whom plasma was available and who had not used desmo-
pressin or replacement therapy 72 hrs (n5 48).
䊏 Discussion
In this large pediatric cohort of 113 moderate or severe VWD
patients from the WiN-study, the majority (44%) experienced
pediatric-specific bleeding symptoms incorporated in the ISTH-BAT.
In addition, all 26 postmenarche girls had menorrhagia. As expected,
Type 3 VWD children had a more severe bleeding phenotype than
Type 2 or Type 1 VWD children, with earlier presentation in child-
hood. Type 3 patients also suffered more from joint bleeds as well as
oropharyngeal bleeding and more treatment moments with desmopres-
sin, coagulation factor replacement therapy, or blood transfusion. The
frequencies of epistaxis, oral cavity bleeding, and any bleeding were
higher in Type 2 and 3 VWD than in Type 1 VWD. Bleeding fre-
quency was also moderately associated with VWF levels and FVIII:C
levels. A significantly higher BS was observed in index cases compared
with affected family members and in older children compared with
younger children.
Mild bleeding symptoms like epistaxis and bruising are common
in healthy children with prevalences of 40% and 25% [6,22]. Recently,
Mauer et al. [23] collected bleeding histories from 500 healthy adults
and reported prevalences of 25% for epistaxis, 18% for easy bruising,
18% for prolonged bleeding after a tooth extraction, and 47% for
menorrhagia. In our pediatric cohort of moderate and severe VWD,
significantly higher proportions of children of 50% up to 100% suf-
fered from epistaxis, cutaneous bleeding, oropharyngeal bleeding, and
menorrhagia as is supported by other studies [22].
Hemostatic challenges, such as surgery, dental extractions, or men-
struation, occur less frequently in children, and the incidence
increases with age [15,24,25]. However, if children with VWD are
challenged, a large proportion suffers from bleeding as a result, a

1146 American Journal of Hematology, Vol. 90, No. 12, December 2015 doi:10.1002/ajh.24195
RESEARCH ARTICLE Bleeding Phenotype of Pediatric VWD Patients

finding that was seen in our cohort as well as in other pediatric The strength of our study is that it covers a large group of mod-
VWD cohorts [15]. Even after prophylactic treatment with replace- erate and severe pediatric VWD patients with various types of
ment therapy or desmopressin a large proportion of moderately or VWD. In addition, The ISTH-BAT has, thus far, only been eval-
severely affected children still suffered from bleedings. Recently, it has uated in healthy children and in children with mild VWD [14,18];
been demonstrated that clinical bleeding outcome of VWD and so, this article is the first to study the ISTH-BAT in a large group
patients that need prophylactic treatment can be predicted by the of moderate and severe pediatric VWD patients with various types
Tosetto BS [9,12]. It would be of interest to study prospectively if a of VWD.
BS can also predict bleeding in children and can be used to prevent In conclusion, pediatric-specific bleedings included in the ISTH-
bleedings after surgery and dental extraction in VWD patients. BAT and also other pediatric-specific bleedings occur in a large pro-
In several cohorts, BS have been shown to discriminate between portion of children with moderate to severe VWD. A BS with supple-
individuals with and without VWD or other bleeding disorders and mentary pediatric-specific bleeding symptoms is, therefore, of
to indicate the severity of the bleeding phenotype [4,9,14,26–28]. A important value in the diagnosis and management of VWD in chil-
few years ago, a Pediatric Bleeding Questionnaire (PBQ) was devel- dren because it may predict bleeding phenotype and necessity of ther-
oped by Bowman et al. [13] and evaluated in healthy children and in apeutic interventions.
a large cohort of children that were referred for evaluation of bleed-
ing diathesis or preoperative screening. The PBQ was similar to the 䊏 Acknowledgments
adult version of the BS but supplemented with pediatric-specific
bleeding symptoms, like umbilical stump bleeding, cephalohematoma, The WiN study was supported by research funding from Dutch
postcircumcision bleeding, postvenipuncture bleeding, and macro- Hemophilia Foundation (Stichting Hemophilia) and CSL Behring
scopic hematuria. In their cohort of 151 children, six were diagnosed (unrestricted grant). The authors thank Dr. E.M. de Wee and all
with VWD. None of these children presented with pediatric-specific hemophilia nurses for including participating patients.
bleeding symptoms. However, cheek hematoma and conjunctival
hemorrhage were not evaluated in this study. Also, in similar cohorts 䊏 Author Contributions
from the United States (n 5 104, eight diagnosed with Type 1 VWD)
F.W.G. Leebeek, K. Fijnvandraat, and M.H. Cnossen designed the
and from Germany (n 5 100, 12 Type 1 VWD and 11 Type 2 VWD),
research, analyzed and interpreted data, and wrote the manuscript.
no pediatric-specific bleedings were reported [14,29]. In 2010, the
Y.V. Sanders performed research, analyzed and interpreted data, and
PBQ was administered in 100 Canadian children with VWD, mostly
wrote the manuscript. J.G. van der Bom and E.P. Mauser-Bunschoten
Type 1 VWD patients. Pediatric-specific bleedings were reported in
designed the research, analyzed and interpreted data, and critically
17% of these children [15]. In our more severely affected VWD
reviewed the manuscript. J. Boender, J. de Meris, F.J. Smiers, B. Gran-
cohort, in which only patients with VWF levels < 0.30 U/dL were
zen, P. Brons, and R.Y.J. Tamminga analyzed and interpreted data
included, pediatric-specific bleedings occurred in 44% of children,
and critically reviewed the manuscript. All authors gave their consent
mainly in Type 3 VWD patients (78%), but also in Type 2 (32%) and
to the final version of the manuscript.
moderate to severe Type 1 (48%) patients. These bleedings also
occurred in children with a low overall BS. Adding the supplementary
Appendix : Complete List of WIN Study Group Members and
pediatric-specific bleeding symptoms increased the percentage of chil-
Participating Hospitals
dren reporting bleeding to 52. Importantly, 29% of patients experi-
enced bleeding after vaccination, indicating that VWD patients
should receive special care when being vaccinated. This demonstrates • Academic Medical Center, K. Fijnvandraat and M. Coppens
that pediatric-specific bleedings are definitely important to determine Amsterdam
the bleeding phenotype and may help to discriminate children with • VU University Medical Center, A. Kors
and without VWD. Amsterdam
• Netherlands Hemophilia Society J. de Meris
BS were obtained at the time of the interview, not time of VWD • Amphia Hospital, Breda
diagnosis. The threshold for treatments like desmopressin or replace- • Catharina Hospital, Eindhoven
ment therapy is lower for bleeding patients with a known diagnosis • Maxima Medical Center, M.R. Nijziel
of VWD than an undiagnosed bleeding patient. Our patients have Eindhoven
• University Medical Center Gro- K. Meijer and R.Y.J. Tamminga
had easier access to treatment like desmopressin and replacement
ningen, Groningen
therapy that may not have been given to a patient prior to diagnosis. • Kennemer Gasthuis, Haarlem
As a result, this may have led to inflation of the BS, and, therefore, • HagaZiekenhuis, The Hague P.F. Ypma
this study does not reflect the BS as a diagnostic tool but rather as a • Leiden University Medical Center, J.G. van der Bom, H.C.J.
tool to establish the phenotype of children with a known diagnosis of Leiden Eikenboom, and F.J.W. Smiers
• Maastricht University Medical B. Granzen and K. Hamulyak
VWD. Center, Maastricht
For this study, we obtained data from a cohort of patients with • Radboud University Medical P. Brons and B.A.P. Laros-van
well-defined, certain diagnosis of VWD (i.e., historically lowest VWF Center, Nijmegen Gorkom
levels  30 U/dL). Although this does not allow a direct comparison • Erasmus University Medical F.W.G. Leebeek (principal investi-
Center, Rotterdam gator), M.H. Cnossen, Y.V.
with the studies by Bowman et al. (VWF levels 28–44 U/dL), Biss
Sanders, and J. Boender
et al., Bidlingmaier et al., and Marcus et al. (VWF levels 5–50 U/dL), • Van Creveldkliniek, University E.P. Mauser-Bunschoten (chairman
our inclusion criteria are similar to the diagnostic criteria used in sev- Medical Center, Utrecht steering committee)
eral national guidelines, including those currently used in the United
States and the United Kingdom [13–15,29–31].

䊏 References 2. Bowman M, Hopman WM, Rapson D, et al.

1. Rodeghiero F, Castaman G, Dini E. Epidemio-
logical investigation of the prevalence of von
Willebrand’s disease. Blood 1987; 69:454–459.
The prevalence of symptomatic von Willebrand
3. James AH. More than menorrhagia: A review of
the obstetric and gynaecological manifestations
disease in primary care practice. J Thromb Hae- of von Willebrand disease. Thromb Res 2007;
most 2010; 8:213–216. 120: S17–S20.

doi:10.1002/ajh.24195 American Journal of Hematology, Vol. 90, No. 12, December 2015 1147
Sanders et al.
4. De Wee EM, Sanders YV, Mauser-Bunschoten
EP, et al. Determinants of bleeding phenotype
in adult patients with moderate or severe von
Willebrand disease. Thromb Haemost 2012;
108:683–692.
5. Sadler JE, Budde U, Eikenboom JC, et al.
Update on the pathophysiology and classifica-
tion of von Willebrand disease: A report of the
Subcommittee on von Willebrand Factor.
J Thromb Haemost 2006; 4:2103–2114.
6. Katsanis E, Luke KH, Hsu E, et al. Prevalence
and significance of mild bleeding disorders in
children with recurrent epistaxis. J Pediatr 1988;
113:73–76.
7. Huq FY, Kulkarni A, Agbim EC, et al. Changes
in the levels of factor VIII and von Willebrand
factor in the puerperium. Haemophilia 2012; 18:
241–245.
8. Knol HM, Kemperman RF, Kluin-Nelemans
HC, et al. Haemostatic variables during normal
menstrual cycle. A systematic review. Thromb
Haemost 2012; 107:22–29.
9. Tosetto A, Rodeghiero F, Castaman G, et al. A
quantitative analysis of bleeding symptoms in
type 1 von Willebrand disease: Results from a
multicenter European study (MCMDM-1
VWD). J Thromb Haemost 2006; 4:766–773.
10. Rodeghiero F, Castaman G, Tosetto A, et al.
The discriminant power of bleeding history for
the diagnosis of type 1 von Willebrand disease:
An international, multicenter study. J Thromb
Haemost 2005; 3:2619–2626.
11. Rodeghiero F, Tosetto A, Abshire T, et al.
ISTH/SSC bleeding assessment tool: A standar-
dized questionnaire and a proposal for a new
bleeding score for inherited bleeding disorders.
J Thromb Haemost 2010; 8:2063–2065.
12. Federici AB, Bucciarelli P, Castaman G, et al.
The bleeding score predicts clinical outcomes
and replacement therapy in adults with von
Willebrand disease. Blood 2014; 123:4037–4044.
13. Bowman M, Riddel J, Rand ML, et al. Evalua-
tion of the diagnostic utility for von Willebrand
1148 American Journal of Hematology, Vol. 90, No. 12, December 2015
disease of a pediatric bleeding questionnaire.
J Thromb Haemost 2009; 7:1418–1421.
14. Bidlingmaier C, Grote V, Budde U, et al. Pro-
spective evaluation of a pediatric bleeding ques-
tionnaire and the ISTH bleeding assessment tool
in children and parents in routine clinical prac-
tice. J Thromb Haemost 2012; 10:1335–1341.
15. Biss TT, Blanchette VS, Clark DS, et al. Quanti-
tation of bleeding symptoms in children with
von Willebrand disease: Use of a standardized
pediatric bleeding questionnaire. J Thromb Hae-
most 2010; 8:950–956.
16. de Wee EM, Leebeek FW, Eikenboom JC. Diag-
nosis and management of von Willebrand dis-
ease in The Netherlands. Semin Thromb
Hemost 2011; 37:480–487.
17. Sanders YV, Eikenboom J, de Wee EM, et al.
Reduced prevalence of arterial thrombosis in
von Willebrand disease. J Thromb Haemost
2013; 11:845–854.
18. Elbatarny M, Mollah S, Grabell J, et al. Normal
range of bleeding scores for the ISTH-BAT:
Adult and pediatric data from the merging pro-
ject. Haemophilia 2014; 20:831–835.
19. Tosetto A, Castaman G, Rodeghiero F. Bleeding
scores in inherited bleeding disorders: Clinical
or research tools? Haemophilia 2008; 14:415–
422.
20. Federici AB. Clinical diagnosis of von Willebrand
disease. Haemophilia 2004; 10(Suppl 4):169–176.
21. de Wee EM, Mauser-Bunschoten EP, Van Der
Bom JG, et al. Health-related quality of life
among adult patients with moderate and severe
von Willebrand disease. J Thromb Haemost
2010; 8:1492–1499.
22. Nosek-Cenkowska B, Cheang MS, Pizzi NJ,
et al. Bleeding/bruising symptomatology in chil-
dren with and without bleeding disorders.
Thromb Haemost 1991; 65:237–241.
23. Mauer AC, Khazanov NA, Levenkova N, et al.
Impact of sex, age, race, ethnicity and aspirin
use on bleeding symptoms in healthy adults.
J Thromb Haemost 2011; 9:100–108.
RESEARCH ARTICLE
24. Montgomery RR. von Willebrand disease–The
relevance of history. J Thromb Haemost 2005;
3:2617–2618.
25. Bowman M, Hopman WM, Rapson D, et al. A
prospective evaluation of the prevalence of
symptomatic von Willebrand disease (VWD) in
a pediatric primary care population. Pediatr
Blood Cancer 2010; 55:171–173.
26. Bowman M, Mundell G, Grabell J, et al. Genera-
tion and validation of the Condensed
MCMDM-1VWD Bleeding Questionnaire for
von Willebrand disease. J Thromb Haemost
2008; 6:2062–2066.
27. Castaman G, Federici AB, Tosetto A, et al. Differ-
ent bleeding risk in type 2A and 2M von Wille-
brand disease: A 2-year prospective study in 107
patients. J Thromb Haemost 2012; 10:632–638.
28. Boelaars MF, Peters M, Fijnvandraat K. Evalua-
tion of a self-administrated pediatric bleeding
questionnaire measuring bleeding severity in chil-
dren. Thromb Haemost 2012; 108:1006–1007.
29. Marcus PD, Nire KG, Grooms L, et al. The
power of a standardized bleeding score in diag-
nosing paediatric type 1 von Willebrand’s dis-
ease and platelet function defects. Haemophilia
2011; 17:223–227.
30. Nichols WL, Hultin MB, James AH, et al. von Wil-
lebrand disease (VWD): Evidence-based diagnosis
and management guidelines, the National Heart,
Lung, and Blood Institute (NHLBI) Expert Panel
report (USA). Haemophilia 2008; 14:171–232.
31. Laffan MA, Lester W, O’Donnell JS, et al. The
diagnosis and management of von Willebrand
disease: A United Kingdom Haemophilia Centre
Doctors Organization guideline approved by the
British Committee for Standards in Haematol-
ogy. Br J Haematol 2014; 167:453–465.
doi:10.1002/ajh.24195