Journal Logo


Knowledge of Blood Group Decreases von Willebrand Factor Panel Testing in Children

Archer, Natasha M.1; Forbes, Peter W.2; Brugnara, Carlo3

Author Information
doi: 10.1097/HS9.0000000000000003
  • Open

Von Willebrand factor (VWF) and Factor VIII:C testing are often initiated in the evaluation of pediatric patients with a concern for a bleeding disorder. Subjects with O blood group are known to have significantly lower baseline von Willebrand factor (VWF) values than subjects with non-O blood groups and thus are more likely to fall within a low VWF category (low VWF or disease) than any other blood group.1 Repeat VWF panel (VWF antigen [VWF:Ag], VWF ristocetin cofactor [VWF:RCo], and Factor VIII:C drawn on same date and at same time) testing has been advocated to help improve clinician diagnostic ability,2 but there is little evidence to suggest that it does. In subjects with low VWF, knowledge of blood group might not only be useful in decreasing overall VWF panel testing but also help in the diagnosis of patients with true bleeding disorders. We aimed to determine if the timing of blood group testing influenced the number of VWF panels performed.

We conducted a retrospective cohort study of all pediatric subjects age 0 to 18 years at the time of their first VWF panel and who also had a blood group in our electronic medical record system between June 1990 and October 2015. Subjects with hemophilia were excluded. Subjects with greater than 7 panels (n = 10) were excluded because of the assumption that greater than 7 panels was less likely related to bleeding than inflammation. Subjects with more than 1 blood group in our system (secondary to bone marrow transplant) were also excluded. The second of 2 panels done on the same day was excluded as it was part of a DDAVP (desmopressin) challenge and often above baseline values.

Subjects were placed into 3 final diagnosis categories (normal, low VWF, and disease) based on the results of panel testing extracted from the electronic medical record. To be classified as normal, all VWF:Ag and VWF:RCo values had to fall within the normal range (≥50 IU/dL), while low VWF (≥30 IU/dL and <50 IU/dL) or disease (<30 IU/dL) needed at least 1 or more panels within their respective ranges to fall within the category. If a subject had panels that fell within the low VWF and disease value ranges, they were categorized as disease.

Factor VIII:C, VWF functional (VWF:RCo) and antigenic (VWF:Ag) were measured on Siemens BCX and BCS-XP analyzers using Innovance VWF Ac Kit and VWF Ag Kit (Siemens Healthcare Diagnostics, Marburg, Germany).

The primary study outcome was number of VWF panels tested. Additional variables collected included age, sex, blood group, test date of blood group, VWF:Ag, VWF:RCo and FVIII:C values, and test date of VWF panel. Statistical analysis used SAS version 9.3 (SAS Institute Inc., North Carolina, USA). The Kruskal-Wallis test was used when comparing the 3 disease groups. The Wilcoxon 2-sample test was used to compare number of panels between groups defined in terms of the timing of their initial VWF panel. Fisher's exact test was used to compare groups on categorical variables.

During the study period, 2497 VWF panels were tested in 1610 pediatric subjects (median age at first panel 9.3 years, range 0–18 years) (Table 1). Participants were most often tested for personal and/or family history of bleeding or bleeding disorder or abnormal coagulation study, usually prolonged partial thromboplastin. Most children (1074, 67%) had only 1 panel sent. A total of 85 children were diagnosed with VWD and 344 were diagnosed with low VWF (Table 1). Of those characterized by low VWF or VWD, 77% and 74%, respectively, have O blood group, while only 51% of subjects in the normal group have O blood group. In subjects with VWD, those who had their blood group identified before or on the same day as their initial VWF panel had on average a total of 2.48 (SD 1.05) panels performed compared with an average of 3.54 (SD 1.65) panels for those with a blood group tested after the initial VWF panel (P = 0.003) (Table 2). Subjects with normal or low VWF values also had 25% to 30% fewer VWF panels tested if their blood group was known at the time of initial VWF panel. The difference between VWF values in subjects with a blood group before or same day versus after the initial panel was not statistically significant (P > 0.38).

Table 1:
Baseline Characteristics of Study Population
Table 2:
Mean Number of Panels by Final Diagnosis Category, Blood Group and Timing of Blood Group

Subjects with O blood group have approximately 75% of values of the other blood groups. The proposed mechanism for this is decreased clearance of VWF secondary to AB antigens.3,4 Klarmann et al further specified that the difference between O blood group and non-O blood group only begins to manifest itself in childhood as the ABO system develops.5 The difference between O blood group and non-O blood group VWF:Ag and VWF:RCo values is also present in healthy children with no personal or family history of bleeding.6

While the above studies have shown a relationship between ABO blood group and VWF values, it is not standard practice to check ABO blood group with an initial VWF panel.7 Based on our study results, knowing the blood group prior to the initial VWF panel results in fewer VWF panels performed. Thus, pediatric subjects age 0 to 18 years undergo less intensive VWF panel testing if their blood group is determined before or on the same day as the initial VWF testing irrespective of blood group (1.36 vs 2.35, P < 0.0001). Table 2 shows the number of panels by final diagnosis category, blood group, and timing of blood group. As hypothesized, these data show that subjects have fewer panels performed when they are identified as having O blood group before or on the same day as their initial VWF panel compared with after for all final diagnosis categories.

Using the average number of panels when blood group is known at initial VWF panel testing (1.36 [SD .78]), approximately 12% or 302 panels (drawn from 536 children), could have potentially been avoided if the blood group had been determined with the initial VWF panel. This not only would lead to cost savings but decreased phlebotomy visits and missed days from work and school.

While this study shows that knowledge of blood group results in less diagnostic VWF panel testing, it is still unclear whether blood group, especially in low VWF, predicts outcome. Future prospective studies should address whether O blood group subjects with low VWF or VWD are less or more likely to bleed than non-O blood group with low VWF or VWD. This would further support reduced testing in patients identified as having blood group O before or on the same day as initial VWF panel testing. In addition, understanding the bleeding phenotype associated with ABO blood group, especially in patients with low VWF, will help clinicians manage patients after they have been diagnosed.

This study suggests that subjects being tested for VWD should have their blood group determined at their initial assessment, an intervention that might mitigate unnecessary testing. This study is retrospective. While we do not know the exact rationale behind each repeat VWF panel or whether the repeat testing influenced the subject's clinical outcome, these data support the addition of a blood group to the initial laboratory evaluation of VWD. We recommend blood group testing be added to the initial diagnosis and management of VWD guidelines.


1. Miller CH, Haff E, Platt SJ, et al Measurement of von Willebrand factor activity: relative effects of ABO blood type and race. J Thromb Haemost 2003;1:2191–2197.
2. Ng C, Motto DG, Di Paola J. Diagnostic approach to von Willebrand disease. Blood 2015;125:2029–2037.
3. Davies JA, Collins PW, Hathaway LS, et al von Willebrand factor: evidence for variable clearance in vivo according to Y/C1584 phenotype and ABO blood group. J Thromb Haemost 2008;6:97–103.
4. Gallinaro L, Cattini MG, Sztukowska M, et al A shorter von Willebrand factor survival in O blood group subjects explains how ABO determinants influence plasma von Willebrand factor. Blood 2008;111:3540–3545.
5. Klarmann D, Eggert C, Geisen C, et al Association of ABO(H) and I blood group system development with von Willebrand factor and Factor VIII plasma levels in children and adolescents. Transfusion 2010;50:1571–1580.
6. Akin M, Balkan C, Karapinar DY, et al The influence of the ABO blood type on the distribution of von Willebrand factor in healthy children with no bleeding symptoms. Clin Appl Thromb Hemost 2012;18:316–319.
7. 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 Haematology. Br J Haematol 2014;167:453–465.
Copyright © 2017 The Authors. Published by Wolters Kluwer Health Inc., on behalf of the European Hematology Association.