No abstract

In recognition of the importance of von Willebrand factor (vWF) testing in the diagnosis of von Willebrand disease (vWD), the United Kingdom National External Quality Assessment Scheme for Blood Coagulation regularly distributes samples for determination of vWF:antigen (vWF:Ag). Data from 10 separate surveys performed between 2001 and 2005 are reviewed. These include results from ~200 different centers, of which 55% are within the United Kingdom and the remainder are from other countries. During the period of the surveys, the use of immunoelectrophoresis for determination of vWF:Ag practically disappeared and was largely replaced by latex agglutination assays. The coefficient of variation (CV) of results in different centers was approximately 15 to 20% for most vWF:Ag techniques, with CVs of approximately 7% for a fluorescence-based assay. Several different techniques were used for determination of vWF ristocetin cofactor activity (vWF:RCo), all of which were associated with poor agreement among centers as indicated by CVs of 40 to 50%. Several centers calculated the ratio of vWF:Ag/vWF:RCo but with variable success. Ratios compatible with either type 1 or type 2 vWD were obtained on samples from subjects with type 1 vWD, as well as on samples from subjects with genetically confirmed type 2 vWD. Overall, our data show that laboratory testing for vWD remains problematic. It remains to be seen whether newer techniques will offer consistently improved precision.

Global hemostasis devices are currently being employed in operating rooms to assess the bleeding risk and outcomes for patients undergoing surgery. Two devices currently available are the TEG (Thromboelastograph; Haemoscope Corp., Niles, IL) and the ROTEM (Rotation Thromboelastometer; Pentapharm GmbH, Munich, Germany). Both measure the speed of clot formation, the strength of the clot when formed, and clot fibrinolysis kinetics. The two devices use different parameters so no cross comparisons of results can be made. The devices are usually operated by a member of the operating team and not a laboratory scientist; thus their testing and performance is generally not laboratory controlled, despite quality control being required to ensure reliable results. The UK National External Quality Assessment Scheme (NEQAS) for Blood Coagulation has undertaken a series of exercises evaluating the provision of External Quality Assessment (EQA) material for these devices. A series of four studies have taken place using lyophilized plasmas as the test material. Up to 18 TEG users and 10 ROTEM users have been involved in testing two samples per study, for a total of eight samples tested. The samples were normal plasmas, factor VIII or XI deficient samples, or normal plasmas spiked with heparin. The precision of the tests varied greatly for both devices, with coefficients of variances ranging from 7.1 to 39.9% for TEG and 7.0 to 83.6% for ROTEM. Some centers returned results that were sufficiently different from those obtained by other participants to predict alterations in patient management decisions. Our data indicate that regular EQA/proficiency testing is needed for these devices.

We report the results of external quality assessment exercises in which 60 to 120 centers performed factor VIII (FVIII) inhibitor testing on a series of samples over a 13-year period. Samples from seven different subjects were distributed for analysis comprising the following: four different subjects with severe hemophilia A with antibodies following replacement therapy, one subject with acquired hemophilia A and antibodies to FVIII, one subject with normal FVIII and an easily detected lupus anticoagulant, and one subject with mild hemophilia A and a difficult-to-detect lupus anticoagulant but without antibodies to FVIII. In all of the surveys the results obtained in different centers analyzing the same sample varied to an extent that would influence patient management decisions. In the UK National External Quality Assessment Scheme surveys reported here, there was considerable interlaboratory variation in the results of FVIII inhibitor testing that did not improve over the survey period. The coefficient of variation of results in different centers was between 33% and 106% in samples from patients with severe congenital hemophilia A. In some cases, results were affected by assay components. For one plasma, the mean FVIII inhibitor results in centers using one source of normal plasma was 3.9 Bethesda unit (BU)/mL compared with a mean of 5.7 BU/mL in centers using a different normal plasma source ( P = 0.04). Our data indicate that the detection of FVIII inhibitors is not the same in different centers, and the degree of variability noted makes it likely that assay variability has contributed to the lack of international consensus in relation to the real incidence of FVIII inhibitors in different clinical settings. Improvements in assay standardization are urgently needed.

Introduction