Sage Journals: Discover world-class research

Abstract

Background

Accurate diagnosis of Von Willebrand Disease (VWD) is essential for effective management, but no single test can definitively confirm it due to the disease's complexity and the limitations of current methods. Tests measuring Von Willebrand Factor (VWF) binding activity, such as the platelet-dependent VWF activity test (VWF:Ab), have been developed, but their application in Latin American populations remains underexplored. This study aimed to evaluate the performance of the VWF:Ab test compared to the Ristocetin Cofactor test (VWF:GPIbR) in a Colombian population with suspected or confirmed VWD.

Methods

A diagnostic cross-sectional and retrospective study was conducted, including adults referred for hematological evaluation due to a personal or family history of bleeding disorders. Samples were analyzed using both VWF:GPIbR and VWF:Ab tests (HemosIL^®) simultaneously. Diagnostic performance was assessed through sensitivity, specificity, predictive values. Additionally, correlation analyses were performed, and the level of agreement at various diagnostic thresholds was assessed.

Results

125 patients were included, 88% of whom were women. A strong positive correlation was found between both tests (ICC = 0.96, p < 0.001). VWF:Ab had a specificity of 98.8% and sensitivity of 71.4%. Degree of agreement for 89.6% threshold was Normal (Greater than 50 IU/dL), Pathological (Less than 50.0 IU/dL). Notably, 7.2% of samples were classified as unclassifiable by VWF:Ab, highlighting potential limitations in qualitative defect detection

Conclusion

VWF:Ab exhibited high specificity and a strong correlation with VWF:GPIbR. Nevertheless, the differences in subtype classification and the identification of unclassifiable samples underscore the necessity of using both tests in a complementary manner.

Keywords

Von Willebrand disease VWF:Ab VWF:GPIbR Von Willebrand factor diagnostic performance Colombia

Introduction

Von Willebrand disease (VWD) is a bleeding disorder caused by a quantitative or qualitative deficiency of von Willebrand factor (VWF), a multimeric protein essential for hemostasis.^1,2 VWF plays a key role in mediating platelet adhesion to sites of vascular injury through its binding to collagen and to platelet glycoprotein Ibα (GPIbα), promoting platelet aggregation and stabilizing factor VIII (FVIII) in circulation through non-covalent binding.^3,4 VWD may be inherited or acquired, and the VWF gene is highly polymorphic, contributing to substantial interindividual variability in plasma levels of the protein. Phenotypically, VWD is classified into three major types: type 1 (partial quantitative deficiency), type 3 (complete deficiency), and type 2 (qualitative functional defects).^5,6

Type 1 accounts for approximately 80% of diagnosed cases and is characterized by mucocutaneous bleeding and a VWF:GPIbR to VWF:Ag ratio greater than 0.7.⁷ Type 2 involves qualitative defects and is defined by a ratio below 0.7. It is further subdivided into: type 2A, marked by loss of high molecular weight multimers; type 2B, with increased affinity of VWF for GPIbα, enhanced platelet aggregation, and frequent thrombocytopenia; type 2 M, with reduced binding of VWF to GPIbα or collagen (2M-CB variant); and type 2N, caused by impaired binding of VWF to FVIII6. Type 3 represents the most severe form of the disease, with near-complete absence of VWF and markedly reduced FVIII levels (<5 IU/dL).⁸

Understanding and diagnosing VWD is inherently complex. No single test can definitively confirm VWF functionality. This necessitates a combination of laboratory tests, including a primary screening and diagnostic panel, as well as complementary second-level or secondary panel tests for diagnostic confirmation and disease classification.⁹ At the local level, we analyzed primary screening tests such as Von Willebrand antigen (VWF:Ag), Von Willebrand activity (VWF:Ab), Ristocetin cofactor (VWF:GPIbR), activated partial thromboplastin time (aPTT), and Factor VIII (FVIII:C). Although numerous assays are available to quantify VWF levels, there remains a pressing need for greater expertise in interpreting these tests.

Indeterminate results often lead to misdiagnoses, while variations in test sensitivity highlight the necessity of updating methodologies and standardizing protocols. The traditional VWF assay, which uses ristocetin to expose the A1 domain of VWF and facilitate binding to GPIbα on platelets, has significant limitations. It exhibits a high coefficient of variation and low sensitivity, complicating the detection of low VWF levels.^10–12 Moreover, false positives can occur in individuals with the p.D1472H variant in the VWF gene, which is prevalent among African Americans.¹³ Conversely, the VWF antigen test measures VWF levels in plasma but can fluctuate within the same individual due to environmental factors such as stress, hormonal cycles, and exercise.^14–16

Several VWF activity assays are available that quantify its activity in plasma, some of which do not require ristocetin. According to the 2021 international guidelines from ASH, ISTH, NHF, and WFH,¹⁷ it is recommended to use newer assays that measure VWF binding activity, such as VWF:CB, VWF:GPIbM, and VWF:GPIbR, instead of the traditional VWF:Rco assay (automated or manual) for the diagnosis of VWD. The VWF:GPIbR assay uses a recombinant fragment of platelet glycoprotein Ib alpha (GP1bα) and still requires ristocetin to promote the interaction between VWF and GP1bα. In contrast, the VWF:GPIbM assay employs a gain-of-function mutation in the GP1bα fragment, enabling spontaneous binding to the A1 domain of VWF without the need for ristocetin, thereby reducing assay variability and enhancing reproducibility.^5,18 These newer assays more accurately assess functional VWF activity and reduce the risk of false-positive or false-negative results, particularly in individuals carrying specific VWF variants, such as p.D1472H, that may interfere with ristocetin-dependent methods like VWF:GPIbR.¹³

Despite these recommendations, there is a significant knowledge gap regarding the practical implementation and comparative effectiveness of these new assays in Latin populations, such as the Colombian population. Incomplete or ambiguous classifications, limited treatment options and complicated decision-making for scheduled and emergency surgical procedures. Timely classification, on the other hand, supports the inclusion of other family members with a high clinical suspicion and familial relation due to the high penetrance of the disease, or at least becomes a risk factor to consider.^3,7 In this context, the study aimed to evaluate the analytical performance of the platelet-dependent von Willebrand factor activity test (VWF:Ab, HemosIL^®) in comparison with the ristocetin-dependent assay (VWF:GPIbR, HemosIL^®) for measuring plasma VWF activity levels. The latter is currently used in our clinical laboratory in Bogotá, Colombia, as part of the primary hemostasis evaluation panel. This study seeks to contribute to the optimization of laboratory protocols for VWF testing across Colombia, in accordance with recent recommendations from international scientific societies.

Materials and Methods

Study Design

A cross-sectional study was conducted to evaluate the analytic performance of the VWF:Ab assay. Data were retrospectively collected between December 2021 and January 2023 from a national private laboratory in Bogotá, Colombia. Diagnostic evaluation typically begins with the organization of a primary test panel, which includes VWF:Ag, VWF:Ab, VWF:GPIbR, aPTT, and FVIII:C. The methods and results of this study are reported in accordance with the STARD guidelines for diagnostic accuracy studies.¹⁹

Samples

This retrospective study used data retrieved from the medical records of patients aged over 18 years who were referred for hematological evaluation due to a personal or family history of bleeding, or because they were undergoing assessment for suspected VWD. Only patients with available results for both the index test (VWF:Ab) and the reference standard (VWF:GPIbR) were included. No missing data were present for these tests. Samples were excluded if they were icteric, lipemic, hemolyzed, coagulated, or lacked accompanying clinical data. Pregnant women and patients with complications during sample collection were also excluded. The estimated sample size was 114, based on an expected specificity of 96%, a 95% confidence level, a standard error of 4%, and a prevalence of 20.4% of abnormal results in the reference test, as observed at the study site.

Study Variables and Data Sources

Information on age, sex, and test indications was collected for all participants. Test indications included bleeding disorders, suspected VWD, hereditary conditions such as Ehlers-Danlos syndrome and factor VIII deficiency, as well as acquired disorders including acquired hemophilia, Henoch-Schönlein purpura, idiopathic thrombocytopenic purpura (ITP), thrombohemorrhagic syndrome, other unspecified coagulation defects, bleeding of unknown origin, unspecified anemia or thrombocytopenia, and hematologic malignancies. Data were also collected for the hematological panel, including aPTT, VWF:Ag, VWF:GPIbR, VWF:Ab, and FVIII:C.

All data were obtained from two primary sources: laboratory test results were retrieved from the institution's Laboratory Information System (LIS), while test indications and clinical histories were obtained from the electronic medical record system. To ensure participant confidentiality, all records were anonymized and a consecutive code was assigned to each participant for statistical analysis, separate from any identifying information

Diagnostic Tests

The reference test was VWF:GPIbR, an automated immunoturbidimetric assay provided by HemosIL^®, which employs latex particle amplification to measure VWF activity in citrated plasma. Turbidity changes resulting from the agglutination of latex particles are detected, with a recombinant fragment of platelet glycoprotein (rGP1bα) serving as the VWF receptor. This fragment is linked to the latex particles via a monoclonal antibody, allowing interaction with the patient's VWF in the presence of ristocetin. The extent of agglutination, measured as a decrease in transmitted light, reflects VWF activity. For, the detection threshold is 4.4%, with reference values ranging from 48.2% to 201.9% for blood type O, and from 60.8% to 239.8% for blood types A, B, and AB.^20,21

The test under evaluation was the automated VWF:Ab test (HemosIL^®), which also uses latex-enhanced immunoturbidimetry to quantify VWF activity. A monoclonal antibody bound to the latex reagent, specifically targeting the platelet-binding site of VWF, reacts with the analyte in the patient's plasma. The degree of agglutination, measured by the reduction in transmitted light, is directly proportional to VWF activity. For the ACL TOP 750 coagulation analyzer series, the detection threshold is 3.2%, with reference ranges of 40.3–125.9% for blood group O and 48.8–163.4% for groups A, B, and AB stated in the insert²²; although the guidelines for the diagnosis of von Willebrand's disease have eliminated the need for interpretation according to blood group based on clinical evidence of bleeding and an objective value for diagnostic inclusion.¹⁷ Both assays were performed in parallel, and personnel conducting the VWF:Ab measurements were blinded to the VWF: GPIbR results.

The additional tests included in the study panel are described below:

The VWF:Ag assay (HemosIL^®) quantifies antigen levels using latex particle-enhanced immunoturbidimetry, with agglutination measured as a decrease in transmitted light.²³

The FVIII:C assay (HemosIL^®) measures coagulation activity by evaluating the correction of clotting time in factor VIII-deficient plasma after the addition of patient plasma. Results are derived from a calibration curve, with a reference range of 50%–150%.²⁴

Activated partial thromboplastin time (aPTT) was measured using the SynthASil^® reagent (Instrumentation Laboratory, HemosIL^®), a silica-based reagent designed to assess the intrinsic and common coagulation pathways. This test evaluates the functionality of clotting factors VIII, IX, XI, and XII, as well as factors of the common pathway, and is sensitive to deficiencies or inhibitors affecting these components.²⁵

All measurements (VWF:GPIbR, VWF:Ab, and VWF:Ag) were performed simultaneously using ACL TOP 750 analyzers (Instrumentation Laboratory, HemosIL^®), with reagents from the same manufacturer. This approach was intended to ensure consistency of results, minimize pre-analytical variability, and allow for direct comparison between methods.

Statistical Analysis

Categorical variables were described using absolute and relative frequencies, while quantitative variables were summarized as medians and interquartile ranges (IQR) after assessing their distribution with the Shapiro–Wilk test. The relationship between VWF:GPIbR and VWF:Ab was evaluated using two approaches for continuous variables: (1) Spearman's correlation and (2) intraclass correlation coefficient (ICC). A p-value < 0.05 was considered statistically significant.

Subsequently, both VWF:GPIbR and VWF:Ab were categorized according to three cut-off strategies:

Scenario 1 – Laboratory cut-offs, which included three categories: Normal (≥50.0%), Borderline (VWF:GPIbR 48.2-49.99%; VWF:Ab 40.3-49.99%), and Pathological (<48.2% for VWF:GPIbR; < 40.3% for VWF:Ab); Scenario 2 – Manufacturer insert cut-offs, which defined only two categories: Normal (VWF:GPIbR ≥48.2% ; VWF:Ab ≥40.3%) and Pathological (below those thresholds); and Scenario 3 – Guideline-based cut-offs, based on the 2021 ASH/ISTH/NHF/WFH recommendations,¹⁷ with Normal defined as ≥50.0% and Pathological as <50.0% for both assays.

The borderline category in Scenario 1 represents values falling between the manufacturer's cut-off and the international guideline threshold, a range that may require closer clinical correlation or follow-up testing to determine whether an abnormal result reflects a true underlying defect or is influenced by preanalytical or biological variability.

The use of three classification scenarios in this study reflects the ongoing lack of international consensus regarding optimal diagnostic thresholds for VWF activity assays. While assay manufacturers provide cut-off values based on analytical validation, these often differ from those recommended by clinical guidelines, such as the 2021 ASH/ISTH/NHF/WFH consensus.¹⁷ In routine laboratory practice, local protocols may incorporate modified thresholds or intermediate “borderline” categories to account for analytical variability and preanalytical factors. By evaluating all three scenarios, manufacturer insert cut-offs, locally adapted laboratory cut-offs, and international guideline-based thresholds, this study aimed to capture the range of real-world interpretations and assess how each approach influences agreement between assays and the classification of patients. This approach highlights the need for greater harmonization of diagnostic criteria, particularly in settings with variable access to second-tier confirmatory testing.

Agreement between the categorical classifications was assessed using the Kappa (κ) statistic. The analytical performance of the VWF:Ab assay was evaluated by calculating sensitivity, specificity, and predictive values, with 95% confidence intervals (CI95%) reported for each metric. Finally, participants with confirmed VWD were subclassified as type 1 if the VWF:Ab/VWF:Ag ratio was ≥ 0.7, and as type 2 if the ratio was <0.7. All statistical analyses were performed using Stata software, version 13.0.

This study was approved by the Ethics Committee of Fundación Universitaria Sanitas and Clínica Colsanitas (CEIFUS 1610–22, 2022).

Results

A total of 125 patient samples from a retrospective cohort with suspected or previously diagnosed VWD were included, of whom 88% were women (Figure 1). The primary indication for hematological testing was the presence of bleeding disorders, including coagulopathies, spontaneous bleeding, uterine bleeding, thrombocytopenia, and factor VIII deficiency. Overall, values from the hematological panel were within reference ranges for the evaluated population (Table 1).

Figure 1.

Flow Diagram of Patient Selection.

Table 1.

Clinical and Demographic Characteristics of the 125 Patients Included in the Study.

Variable	Median (IQR)N = 125
Age, years	40 (29−47)
Gender*
Females	110 (88.0)
Males	15 (12.0)
Indication for the test panel*
1. Bleeding disorders	44 (35.2%)
2. Von Willebrand Disease	20 (16.0%)
3. Hereditary or acquired syndromes	28 (22.4%)
4. Others	31 (24.8%)
5. No data	2 (1.6%)
aPTT, (seconds)	34.1 (31.9−37.0)
Factor VIII, (%)	60.8 (43−85.9)
VWF:Ag, (%)	84.9 (63.7−112.0)
VWF:Ab, (%)	69.7 (50−95.7)
VWF:GPIbR, (%)	63.8 (43.1−89.6)

*Absolute frequency (relative frequency).

Hereditary or acquired syndromes: Hemato-oncological diseases, Idiopathic Thrombocytopenic Purpura, Venous Thromboembolic Disease, Ehlers-Danlos Syndrome. Others: Unspecified anemia, Unspecified thrombocytopenia, Leukocyte disorders, unspecified, Eosinophilia, Thrombophilia, etc.

A statistically significant positive linear correlation was observed between the VWF:Ab and VWF:GPIbR assays (p < 0.001; Figure 2A). The Bland–Altman plot (Figure 2B) showed that most paired values fell within the limits of agreement, although four samples exhibited extreme differences exceeding 150 IU/dL. The individual ICC was 0.92 (95% CI: 0.89-0.94), and the mean ICC was 0.96 (95% CI: 0.94-0.97), both statistically significant (p < 0.001). A similarly strong correlation (r = 0.93) was observed between VWF:Ab and VWF:Ag.

Figure 2.

Correlation of the VWF:Ab and VWF:GPIbR Test. A) Spearman correlation between VWF:Ab and VWF:GPIbR test, (B) Bland-Altman plot showing agreement between the two tests.

Table 2 summarizes the agreement between VWF:Ab and VWF:GPIbR across three classification scenarios. Overall agreement exceeded 80% in all cases, with the highest observed in the guidelines-based scenario (89.6%). The Kappa coefficient indicated moderate agreement for both the laboratory (κ = 0.57) and insert cut-offs (κ = 0.55), while substantial agreement was reached only under the guideline-based classification (κ = 0.75). All comparisons were statistically significant (p < 0.001). These findings suggest that the choice of cut-off strategy influences the level of concordance between the assays, with alignment to international guideline thresholds yielding the highest consistency.

Table 2.

Degree of Agreement Between the VWF:GPIbR Test and the VWF:Ab Test.

Scenario	Agreement	Expected Agreement	Kappa (k)	Standard Error	P Value
Laboratory cut offs	80.8%	54.95%	0.57	0.069	<0.001
Insert cut offs	83.2%	62.53%	0.55	0.079	<0.001
Guidelines cut offs recommendations	89.6%	58.27%	0.75	0.087	<0.001

Sensitivity varied notably depending on the cut-off strategy, ranging from 47.5% (95% CI: 31.5-63.9) using manufacturer insert values to 71.4% (95% CI: 55.4-84.3) when applying guideline-based thresholds. In contrast, specificity remained consistently high across all scenarios, reaching 100% in both the laboratory and insert-based classifications, and 98.8% (95% CI: 93.5-100) under guideline-based criteria. Positive predictive values were also high across the board (≥96.8%), whereas negative predictive values ranged from 80.2% to 89.1% (Table 3). These findings confirm the assay's strong capacity to correctly identify individuals without abnormal VWF activity, while highlighting a trade-off between sensitivity and threshold selection. The guideline-based scenario yielded the highest sensitivity with minimal compromise in specificity, supporting its potential utility in clinical decision-making where underdiagnosis may carry significant consequences.

Table 3.

Analytical Performance of the VWF:Ab Test.

Scenario	Sensitivity, (%; 95% CI)	Specificity (%; 95% CI)	NPV (%; 95% CI)	PPV (%; 95% CI)
Laboratory cut offs	65.5 (45.7−82.1)	100 (95.6–100)	89.1 (80.9−94.7)	100 (82.4–100)
Insert cut offs	47.5 (31.5−63.9)	100 (95.8–100)	80.2 (71.3−87.3)	100 (82.4–100)
Guidelines cut offs recommendations	71.4 (55.4−84.3)	98.8 (93.5–100)	87.2 (78.8−93.2)	96.8 (83.3−99.9)

NPV: Negative predictive value. PPV: Positive predictive value . CI: Confidence Interval

Based on the results obtained using the international guideline-based thresholds, 42 pathological cases were identified with the VWF:Ab assay and 31 with the VWF:GPIbR assay. Among these, only four individuals had no personal or family history of bleeding or prior diagnosis of VWD. Of these four, three are presented with clinical contexts suggestive of underlying conditions potentially associated with acquired forms of VWD.

Regarding the complementary tests, aPTT levels were analyzed in individuals classified as pathological under each of the three evaluated scenarios. Although aPTT is dependent on multiple pre-analytical variables and directly on the amount of FVIII present in the sample, it is also known that it is part of the initial approach tests for coagulopathies, so following the clinical study algorithms, this test is included as an analysis variable for the identification of the prolongation or not of a coagulation time prior to the study of specific diagnostic tests for Von Willebrand's disease. Among patients subclassified using the VWF:Ab/VWF:Ag ratio, those with a ratio <0.7, indicative of Type 2 VWD, showed slightly higher median aPTT values compared to those with ratios ≥0.7, classified as Type 1. However, the overall aPTT levels were comparable between both groups, suggesting limited discriminatory power of this parameter alone. In patients with abnormal VWF:GPIbR or VWF:Ab values, a general trend toward lower FVIII levels was observed in those with a reduced proportion of VWF activity relative to antigen (VWF:Ag), regardless of the cut-off criterion applied (laboratory values, assay-specific thresholds, or clinical guidelines) (Table 4).

Table 4.

aPTT (Seconds) and Factor VIII (%) Levels in Patients with Pathological Values of VWF:GPIbR or VWF:Ab.

Scenario	Ratio*	VWF:GPIbR			VWF:Ab
Scenario	Ratio*	n	Factor VIII (%) Median (IQR)	aPTT (Seconds) Median (IQR)	n	aPTT (Seconds) Median (IQR)	Factor VIII (%) Median (IQR)
Laboratory cut offs	<0.7	24	45.9 (30.1 – 63.8)	35.5 (32.1 – 39.5)	9 (47.4)	38.6 (34.4 – 41.1)	26.1 (25.4 – 44.8)
Laboratory cut offs	≥0.7	16	35.7 (28.8 – 44.6)	35.3 (32.9 – 38.5)	10 (52.6)	34.6 (32.5 – 38.4)	39.8 (29.6 – 45.4)
Insert cut offs	<0.7	24	45.9 (30.1 – 63.8)	35.5 (32.1 – 39.5)	9 (47.4)	38.6 (34.4 – 41.1)	26.1 (25.4 – 44.8)
Insert cut offs	≥0.7	16	35.7 (28.8 – 44.6)	35.3 (32.9 – 38.5)	10 (52.6)	34.6 (32.5- 38.4)	39.8 (29.6 – 45.4)
Guidelines cut offs recommendations	<0.7	26	48.4 (31.6 – 62.9)	34.8 (32.2 – 39.3)	10 (32.3)	37.5 (31.3 – 41.1)	34.5 (25.4 – 47.9)
Guidelines cut offs recommendations	≥0.7	16	45.9 (30.1 – 63.8)	35.3 (32.9 – 38.5)	21 (67.7)	35.2 (33.6–37)	39.7 (29.6 – 45.4)

IQR: Interquartile Range, Ratio defined as VWF:Activity (VWF:GPIbR or VWF:Ab) to VWF:Ag. Reference range for aPTT: 25.1–36.5 s, Reference range for Factor VIII: 50–150% (SynthASil^®, Instrumentation Laboratory).

Among the 20 patients with a prior diagnosis of VWD originally classified using the ristocetin-dependent activity test, the VWF:GPIbR/VWF:Ag ratio was below 0.7 in 20% of cases (n = 4), identifying them as Type 2 VWD. When classification was performed with the new activity assay (VWF:Ab/VWF:Ag ratio), 80% (n = 16) of patients were identified as Type 2% and 20% as Type 1. In contrast, using the VWF:GPIbR ratio, 55% (n = 11) were classified as Type 2% and 45% (n = 9) as Type 1. These findings suggest that the ristocetin-independent assay may have a greater ability to detect qualitative VWF defects.

Additionally, nine cases were categorized as not classifiable. These results were likely influenced by various confounding factors, including prior anticoagulation related to thrombosis or malignancy, gynecological conditions, or the need to differentiate between Type 2N VWD and mild hemophilia. In some cases, acquired forms of VWD were suspected.

Discussion

This study compared the performance of the VWF:Ab assay, which uses monoclonal antibodies, with the ristocetin-dependent VWF:GPIbR assay in the preliminary evaluation of VWD in a Colombian population. The findings demonstrated a strong correlation and substantial agreement between both assays, particularly when using guideline-based thresholds. The VWF:Ab test exhibited consistently high specificity and positive predictive value across all cut-off scenarios. However, sensitivity remained modest, varying according to the classification strategy applied, and was highest when international guideline thresholds were used.

Despite these similarities in performance, it's crucial to understand their methodological differences. Although both are latex particle-enhanced immunoturbidimetric assays, the VWF:GPIbR and VWF:Ab tests are based on distinct principles. In the VWF:GPIbR assay, VWF activity is measured indirectly through an increase in turbidity caused by latex agglutination. This occurs when a recombinant fragment of platelet glycoprotein (rGP1bα), bound to latex particles via a specific monoclonal antibody, interacts with the patient's VWF in the presence of ristocetin.^26,27 In contrast, the VWF:Ab assay measures activity directly, with a monoclonal anti-VWF antibody binding to the platelet-binding site of VWF in plasma, leading to latex agglutination. In both assays, the degree of agglutination, and therefore VWF activity, is determined by the reduction in light transmission due to aggregate formation.²⁸ The distinctions between these VWF:GPIbR and VWF:Ab tests may arise from variations in measurement conditions, differences in method accuracy, or specific outliers that require further investigation. Additionally, based on diagnostic experience, it has been observed that initial tests often produce expected results in many cases. However, there is biological variability, where VWF levels increase in response to factors such as illness, pregnancy, or hormone use. Retesting may be necessary in cases with a strong clinical suspicion.⁶

These findings support our study's results, which indicate a strong agreement between the two tests used to classify VWD, a well-documented phenomenon in the literature.^10,18,29,30 Despite being based on different measurement principles, the newer assays generally show good correlation with the traditional VWF:GPIbR method. The VWF:Ab is particularly relevant in populations with the D1472H polymorphism, which is present in 67% African Americans and 17% Caucasians.^3,31 This polymorphism can lead to an underestimation of VWF:GPIbR tests. Genetic studies have revealed that the Colombian population has a genetic admixture of 60%–70% European ancestry, 20%–30% Native American ancestry, and 10%–20% African ancestry.³² This genetic diversity underscores the importance of utilizing tests like VWF:Ab in our population to ensure diagnostic accuracy and prevent underestimating VWF activity due to genetic polymorphisms present in the population.

However, it is important to acknowledge the limitations and discrepancies that emerged from the assays. According to the ASH ISTH NHF WFH 2021 guidelines,¹⁷ the new assays have a lower coefficient of variation and better repeatability compared to VWF:RCo. However, the studies did not include many patients of African descent, so they may not accurately represent the bleeding risk or the presence of VWF variants (for example, the D1472H variant) in that population. This could lead to potential overdiagnosis, which is considered harmful in patients with VWF. The positive predictive value was relatively low, indicating that a positive von Willebrand activity test result does not always confirm the presence of the disease. Therefore, it is necessary to consider the variables that impact von Willebrand's results, as well as confirm the results by controlling these modifiable variables. On the other hand, the negative predictive value was 100%, ensuring that a negative result effectively rules out the disease.

In addition, there were notable differences in classifying VWD between the two tests. The VWF:Ab test identified more Type 2 cases compared to VWF:GPIbR. This may be due to the varying sensitivity of tests for qualitative VWF defects. Additionally, the VWF:Ab test identified a greater number of samples as unclassifiable (7.2%) compared to the VWF:GPIbR. This suggests potential limitations in the ability of the VWF:Ab test to classify certain cases, or it may be associated with the possible overdiagnosis of the VWF:GPIbR. Some cases were not consistent among the primary diagnostic tests for VWD; others required diagnostic differentiation between Type 2N VWD and mild hemophilia through more specialized complementary tests or otherwise the bleeds were presumably associated with causes other than congenital hemorrhagic coagulopathy. These unclassifiable cases could be due to the test's difficulty in detecting subtle qualitative defects or differentiating between various VWD subtypes, highlighting an inherent limitation in the technique used.^13,28,33 This indicates that, in some situations, the test may lack the capacity to properly discriminate between disease subtypes, justifying the need for complementary tests or alternative methods to achieve a more accurate classification.

Furthermore, both biological and pre-analytical factors must be carefully considered when interpreting VWF results. VWF levels can fluctuate within the same individual over time, influenced by stress, hormonal cycles (eg, estrogen variations during the menstrual cycle), ovulation, or physical activity.^6,14,15 As an acute-phase reactant, VWF may also increase in response to inflammation or infection, leading to normal results in some VWD patients.^2–4^,6 Certain medications can cause transient increases, potentially masking an underlying deficiency.^2–4 Pre-analytical issues, such as platelet contamination and inadequate sample refrigeration (2°–8 °C), can alter VWF activity and its multimeric structure,⁵ while errors during collection or processing may yield inaccurate results.⁶ The varying sensitivity and specificity of these tests emphasize the necessity of employing multiple diagnostic methods for accurate classification of VWD. Combining results from different tests offers a more comprehensive understanding of VWF functionality, enhancing diagnostic precision and the classification of VWD subtypes. In particular, the prolonged aPTT observed in patients with Type 1 and Type 2 VWD, compared to normal individuals, reinforces the importance of a thorough diagnostic panel to assess VWF function and ensure reliable diagnosis.⁷

Integrating multiple tests provides a broader and more reliable assessment of VWF function, minimizing the risk of misdiagnosis when results are ambiguous or inconsistent.^5,7,8 Unclassifiable cases further underscore the need for additional research and the development of new diagnostic tools with enhanced accuracy and reproducibility. Borderline levels present a unique challenge due to the natural fluctuations in VWF throughout life, underscoring the need for a comprehensive panel of study variables. It's essential to recognize the complementary nature of differing measurements over time and correlate them with family and personal medical history, recognizing that hemorrhagic coagulopathies are similar in their clinical presentation, so that support from highly accurate phenotypic tests is essential for diagnostic accuracy. Severe cases are easier to diagnose with available platforms that measure significantly decreased values with a high specificity.¹⁸ A notable strength of this study is the achievement of the target sample size and an adequate level of representativeness, which closely approximates real-world conditions in line with the panel's guidelines. However, by limiting the study to participants over 18 years old, the findings are not generalizable to pediatric populations. Additionally, it's important to note that environmental and biological factors affecting VWF levels were not controlled for, which may have contributed to variability in the results.

In conclusion, the VWF:Ab demonstrated a higher specificity compared to the VWF:GPIbR. However, the discrepancies in subtype classification and the identification of unclassifiable samples underscore the need to use both tests in a complementary manner to ensure a thorough and accurate evaluation of VWD. The introduction of new diagnostic methods and the revision of clinical laboratory protocols are critical to enhancing the diagnosis and management of this complex coagulopathy. This study aims to support the update of diagnostic protocols in the reference clinical laboratory, aligning with recent international guidelines published in 2021.¹⁷ These updates will not only expand our understanding of the disease but also lead to more precise diagnoses, thereby improving treatment options and enhancing patient outcomes and safety. Building confidence among healthcare specialists in the reliability of laboratory processes is key to ensuring effective medical interventions, ultimately stabilizing or improving patients’ bleeding conditions.

Footnotes

ORCID iDs

Claudia C. Colmenares-Mejia

Natalia Gomez-Lopera

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Clinica Colsanitas.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

Doshi

Rogers

Whitworth

, et al. Utility of repeat testing in the evaluation for von Willebrand disease in pediatric patients. J Thromb Haemost. 2019;17(11):1838–1847. 10.1111/jth.14591

Weyand

Flood

. Von Willebrand disease: Current status of diagnosis and management. Hematol Oncol Clin North Am. 2021;35(6):1085–1101. 10.1016/j.hoc.2021.07.004

Baronciani

Peyvandi

. How we make an accurate diagnosis of von Willebrand disease. Thromb Res. 2020;196:579–589. 10.1016/j.thromres.2019.07.010

Harris

Pelletier

Marin

Winter

. Von Willebrand factor and disease: A review for laboratory professionals. Crit Rev Clin Lab Sci. 2022;59(4):241–256. 10.1080/10408363.2021.2014781

Sharma

Haberichter

. New advances in the diagnosis of von Willebrand disease. Hematology 2019;2019(1):596–600. 10.1182/hematology.2019000064

Seidizadeh

Baronciani

Pagliari

, et al. Phenotypic and genetic characterizations of the Milan cohort of von Willebrand disease type 2. Blood Adv. 2022;6(13):4031–4040. 10.1182/bloodadvances.2022007216

Seidizadeh

Eikenboom

JCJ

Denis

Flood

James

Lenting

. von Willebrand disease. Nat Rev Dis Primer. 2024;10(1):1–51. 10.1038/s41572-024-00536-8

Bergamasco

Moride

Truong Berthoz

Özen

Tzivelekis

. Von Willebrand disease epidemiology, burden of illness and management: A systematic review. J Blood Med 2023;Volume 14:189–208. 10.2147/JBM.S389241.

Sadler

Castaman

O’Donnell

. von Willebrand disease and von Willebrand factor. Haemophilia. 2022;28(S4):11–17. 10.1111/hae.14547

10.

Lai

S-W

Chang

C-Y

Cheng

S-N

S-H

Lai

C-Y

Chen

Y-C

. A comparative evaluation of an automated functional assay for Von Willebrand factor activity in type 1 Von Willebrand disease. Int J Gen Med 2021;Volume 14:5167–5174. 10.2147/IJGM.S321605.

11.

Vangenechten

Mayger

Smejkal

, et al. A comparative analysis of different automated von Willebrand factor glycoprotein Ib-binding activity assays in well typed von Willebrand disease patients. J Thromb Haemost. 2018;16(7):1268–1277. 10.1111/jth.14145

12.

Christopherson

Haberichter

Flood

, et al. Ristocetin dependent cofactor activity in von Willebrand disease diagnosis: Limitations of relying on a single measure. Res Pract Thromb Haemost. 2022;6:e12807. 10.1002/rth2.12807

13.

Hayward

CPM

Moffat

Graf

. Technological advances in diagnostic testing for von Willebrand disease: New approaches and challenges. Int J Lab Hematol. 2014;36(3):334–340. 10.1111/ijlh.12220

14.

Brown

White

Friedberg

, et al. Elevated von Willebrand factor levels during heavy menstrual bleeding episodes limit the diagnostic utility for von Willebrand disease. Res Pract Thromb Haemost. 2021;5:e12513. 10.1002/rth2.12513

15.

Swystun

Lillicrap

. Genetic regulation of plasma von Willebrand factor levels in health and disease. J Thromb Haemost. 2018;16(12):2375–2390. 10.1111/jth.14304

16.

Van Loon

Sonneveld

MAH

Praet

SFE

De Maat

MPM

Leebeek

FWG

. Performance Related Factors Are the Main Determinants of the von Willebrand Factor Response to Exhaustive Physical Exercise. PLoS ONE. 2014;9:e91687. 10.1371/journal.pone.0091687

17.

James

Connell

Ameer

, et al. ASH ISTH NHF WFH 2021 guidelines on the diagnosis of von Willebrand disease. Blood Adv. 2021;5(1):280–300. 10.1182/bloodadvances.2020003265

18.

Kalot

Husainat

Abughanimeh

, et al. Laboratory assays of VWF activity and use of desmopressin trials in the diagnosis of VWD: A systematic review and meta-analysis. Blood Adv. 2022;6(12):3735–3745. 10.1182/bloodadvances.2021005431

19.

Altman

Bossuyt

PMM

. Estudios de precisión diagnóstica (STARD) y pronóstica (REMARK). Med Clínica. 2005;125:49–55. 10.1016/S0025-7753(05)72210-7

20.

Laporte

Tuffigo

Ryman

, et al. HemosIL VWF:GPIbR assay has a greater sensitivity than VWF:RCo technique to detect acquired von Willebrand syndrome in myeloproliferative neoplasms. Thromb Haemost. 2022;122(10):1673–1682. 10.1055/a-1806-9972

21.

HemosIL. 2020. Von Willebrand Factor Ristocetin cofactor. Instrumentation Laboratory. Retrieved from http://www.ilww.com.

22.

Instrumentation Laboratory Company. HemosIL. (2020). Von Willebrand Factor Activity- 0020004700 Protocol. Instrumentation Laboratory. Retrieved from http://www.ilww.com.

23.

HemosIL. (2020). von Willebrand Factor Antigen – 0020002300. Instrumentation Laboratory. Retrieved from http://www.ilww.com.

24.

HemosIL. 2020. Factor VIII deficient plasma – 0020012800. Instrumentation Laboratory. Retrieved from http://www.ilww.com.

25.

HemosIL. 2017. LAB Tromboplastina Parcial Activada PTT SynthASil – 0020006800. Instrumentation Laboratory. Retrieved from http://www.ilww.com.

26.

Geisen

Zieger

Nakamura

, et al. Comparison of Von Willebrand factor (VWF) activity VWF:Ac with VWF ristocetin cofactor activity VWF:RCo. Thromb Res. 2014;134(2):246–250. 10.1016/j.thromres.2014.04.033

27.

Favaloro

Oliver

Mohammed

Vong

. Comparative assessment of von Willebrand factor multimers vs activity for von Willebrand disease using modern contemporary methodologies. Haemophilia. 2020;26(3):503–512. 10.1111/hae.13957

28.

Colpani

Baronciani

Stufano

, et al. A comparative study in patients with type 2 von Willebrand disease using 4 different platelet-dependent von Willebrand factor assays. Res Pract Thromb Haemost. 2023;7(100139):1–10. 10.1016/j.rpth.2023.100139

29.

Roberts

Flood

. Laboratory diagnosis of Von Willebrand disease. Int J Lab Hematol. 2015;37(S1):11–17. 10.1111/ijlh.12345

30.

Graf

Moffat

Carlino

, et al. Evaluation of an automated method for measuring von W illebrand factor activity in clinical samples without ristocetin. Int J Lab Hematol. 2014;36(3):341–351. 10.1111/ijlh.12218

31.

Flood

Gill

Morateck

, et al. Common VWF exon 28 polymorphisms in African Americans affecting the VWF activity assay by ristocetin cofactor. Blood. 2010;116(2):280–286. 10.1182/blood-2009-10-249102

32.

Adhikari

Mendoza-Revilla

Chacon-Duque

Fuentes-Guajardo

Ruiz-Linares

. Admixture in Latin America. Curr Opin Genet Dev. 2016;41:106–114. 10.1016/j.gde.2016.09.003

33.

Quiroga

Goycoolea

Belmont

, et al. Quantitative impact of using different criteria for the laboratory diagnosis of type 1 von Willebrand disease. J Thromb Haemost. 2014;12(8):1238–1243. 10.1111/jth.12594

Diagnostic Performance of Von Willebrand Activity Versus Ristocetin Cofactor Tests in the Primary Evaluation of Von Willebrand Disease in Colombia

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Materials and Methods

Study Design

Samples

Study Variables and Data Sources

Diagnostic Tests

Statistical Analysis

Results

Discussion

Footnotes

ORCID iDs

Funding

Declaration of Conflicting Interests

References