Diagnostic Testing for von Willebrand Disease: Trends and Insights from North American Laboratories over the Last Decade

Abstract

Accurate diagnosis of von Willebrand disease (VWD) depends on the quality, precision, and variability of the laboratory assays. The North American Specialized Coagulation Laboratory Association (NASCOLA) is a provider of external quality assessment (EQA) for approximately 60 specialized coagulation laboratories in North America. In this report, NASCOLA EQA data from 2010 to 2021 are reviewed for trends in methodology and precision among various assays. In particular, recent ASH ISTH NHF WFH (American Society of Hematology, International Society on Thrombosis and Haemostasis, National Hemophilia Foundation, and World Hemophilia Federation) guidelines for diagnosis of VWD are reviewed in light of EQA data. In contrast to other geographic regions, laboratories in North America predominantly use three-assay screening panels (antigen, platelet-binding activity, and factor VIII [FVIII] activity) rather than four-assay panels (antigen, platelet-binding activity, FVIII activity, and collagen-binding activity). They also use latex immunoassays rather than chemiluminescence immunoassays, and the classic ristocetin cofactor (VWF:RCo) assay and monoclonal antibody (VWF:Ab) assay to assess VWF platelet-binding activity over newer recommended assays (VWF:GPIbM and VWF:GPIbR). Factors that may be influencing these North American practice patterns include lack of Food and Drug Administration approval of the VWF:GPIbM, VWF:GPIbR, collagen binding assays, and chemiluminescence methodologies, and the influence of the 2008 National Heart, Lung, and Blood Institute guidelines on laboratory practice. Lastly, systems-based solutions are urgently needed to improve the overall accuracy of laboratory testing for VWD by minimizing preanalytical variables and adopting assay standardization.

Laboratory Diagnosis of von Willebrand Disease (VWD): Geographical Perspectives

“…The ASH ISTH NHF WFH guidelines 50 feel it is more important to capture all type 2 cases, even if falsely capturing some type 1 cases, since findings “suggestive” of type 2 VWD usually result in further discriminatory testing, such as VWF multimers or genetic analysis. Interestingly, NASCOLA data 18 tended to confirm 0.7 as a preferred cut-off with assays dominant in North America, whereas UK NEQAS data seemed to favor 0.6 as the better cut-off. 17 CAP could not provide data on cut-off values due to sample issues.…”

“…16 18 VWF:GPIbM and VWF:GPIbR are performed by far fewer laboratories (15 and 6% in NASCOLA data, respectively). 18 VWF:CB is also performed by few laboratories (18%, NASCOLA data), 18 and this is exclusively by ELISA (CLIA based VWF assays not being available in North America).…”

“…The other programs did not specifically report on diagnostic errors, but may instead report on participant interpretation for separate individual assays (e.g., normal vs. abnormal). 16,17,18,19 Tests and Test Panels Used across Different Geographies…”

“…RCPAQAP data shows a varied mixture of assays in use, given all assays available according to manufacturer placement are NASCOLA and CAP data aligns to the fact that VWF:RCo and VWF:Ab are the only two FDA cleared/approved VWF activity assays, so these are respectively utilized by the majority of laboratories. 16,18 VWF:GPIbM and VWF:GPIbR are performed by far fewer laboratories (15 and 6% in NASCOLA data, respectively). 18 VWF:CB is also performed by few laboratories (18%, NASCOLA data), 18 and this is exclusively by ELISA (CLIA based VWF assays not being available in North America).…”

“…Most other EQA providers do not request an interpretative "diagnosis" based on the laboratory test panel used, instead characterizing individual results as "normal or abnormal." 16,17,18,19,55 In addition, where laboratories report a semiquantitative value limit (e.g., <10 U/dL; ratio <0.5), RCPAQAP will include such data in data analysis but converting such values to 1 value below the limit (i.e., 9 U/dL; ratio 0.4; for the above examples). 14,15 This is considered a valid approach within the context of assessing test limitations, as excluding such "poor performance" will artificially inflate the worth of these assays because of data omission.…”