Factor VIII assay variability in postinfusion samples containing full length and B-domain deleted FVIII.

Abstract

Although the variability in factor VIII (FVIII):C measurement is well recognized, this has not been widely reported for post-FVIII infusion samples. Three samples from haemophilia A patients were distributed in a UK National External Quality Assessment Scheme survey, each after treatment with either ReFacto AF, Kogenate FS or Advate. Fifty-two UK haemophilia centres performed FVIII assays using one-stage (n = 46) and chromogenic (n = 10) assays. Centres calibrated assays with the local plasma standard and with ReFacto AF laboratory standard for the ReFacto AF sample. Chromogenic assays gave significantly higher results than one-stage assays (P < 0.0001, 32% difference) in the post-Kogenate sample but not in the post-ReFacto AF (11% higher by chromogenic assay, ns) or post-Advate samples (3% lower by chromogenic, ns) when assays were calibrated with plasma standards. Twenty centres used all Instrumentation Laboratory (IL)-activated partial thromboplastin time reagents (Synthasil)/IL deficient plasma/reference plasma) in the one-stage assay and 15 used all Siemens reagents (Actin FS/Siemens deficient plasma/reference plasma); this made a significant difference to results post-ReFacto AF (41% higher by IL reagents, P < 0.0001) and Advate (39% higher by IL reagents, P < 0.0001), but not Kogenate (7% higher by IL, ns) when calibrated with plasma standards. Differences between results obtained with different one-stage assay reagents for monitoring Advate have implications for dosing patients. Furthermore, there was considerable inter-laboratory variation as indicated by CVs in the range 15-26% for chromogenic assay and 12-19% for one-stage assay results. This study suggests that external quality assessment schemes should offer participation in post-FVIII infusion schemes where haemophilic patients are monitored.