Lower limit of assay sensitivity: an under-recognised and significant problem in von Willebrand disease identification and classification.

Unlabelled: von Willebrand disease (VWD) is the most common inherited bleeding ailment, and is characterised by low levels of, or abnormal function in, the plasma protein von Willebrand factor (VWF). However, the laboratory testing process is problematic because of both the heterogeneity of VWD and the limitations in the tests used to identify reduced or abnormal VWF.

Objective: This study reports on the lower levels of sensitivity for the different assays used in the diagnostic process for VWD and their significance in the diagnostic identification and classification ofVWD.

Methods: The RCPA Haematology QAP is an international external quality assurance (EQA) program that includes VWF/VWD testing within one of its special haemostasis modules. Over the past 10 years, over 50 samples have been distributed to participants, including five samples devoid of VWF and derived from either true Type 3 VWD patients or else from commerciallypurchased VWF deficient plasma. Samples were tested blind by study participants, who report back both numerical values (for VWF and Factor VIII:C) and an interpretation regarding whether or not VWD is suggested by laboratory findings, and if so, the probable VWD subtype.

Results: Returned data indicates that the lower level of sensitivity (LLS) tends to be around 5-10U/dL for Factor VIII:C, VWF antigen(VWF:Ag), VWFcollagen binding (VWF:CB), and VWF 'activity' (VWF:Act), but canreach 20U/dL or more for VWF ristocetin cofactor (VWF:RCo). There does not appear to be any improvement over the past decade despite ongoing automation of methodology, and indeed, automation does not seem to provide better LLS performance.

Conclusions: Limitations in the LLS of VWD testing have significant implications in terms of the identification and classification of an individual's VWD, given that these laboratory assays are used to identify VWD andhelp characterise functional VWF discordance, and that the majority of severe VWD subtypes have levels of VWF below 20U/dL. Thus, laboratories will sometimes be unable to distinguish whether VWF deficient samples derive from Type 3 VWD or severe Type 1 VWD or even Type 2 VWD.