To cite this article: Lawrie AS, Green L, Mackie IJ, Liesner R, Machin SJ, Peyvandi F. Factor XIII -an under diagnosed deficiency -are we using the right assays? J Thromb Haemost 2010; 8: 2478-82.Summary. Background: The clot solubility test is the most widely used method for detection of factor (F)XIII deficiency. However, it will only detect severe deficiencies; consequently mild deficiencies and heterozygous states are probably under diagnosed. Objective: As an alternative first-line screening test, we assessed an automated quantitative ammonia release assay (QARA). Patients/methods: Inter-assay imprecision was evaluated with commercial normal and pathological control plasmas (10 replicates on each of 5 days). Using the QARA and other commercial assays a comparative assessment of congenital (FXIII range < 1-70 u dL, n = 9) and acquired (n = 43) deficiencies was made. We also investigated the prevalence of acquired deficiencies in hospitalized patients using residual samples from adult patients (n = 1004) and from a paediatric intensive care unit (ICU, n = 56). Results: Assay imprecision was acceptably low (normal control: mean 86.6 u dL ), a second chromogenic assay, the FXIII-A and FXIII A+B-subunit ELISA. We found that 21% of samples from adult patients had FXIII levels < 70 u dL )1 (mean normal ± 2 SD 73-161 u dL) with 6% < 50 u dL. Within the paediatric ICU samples, 52% were < 70 u dL . Conclusions: Our data demonstrates that the automated assay is sensitive, highly reproducible and the results from clinical samples suggest that acquired FXIII deficiency is a relatively common phenomenon in hospital patients after surgery and in ICU.

Introduction of the International Normalized Ratio (INR) has improved the standardization of laboratory control of oral anticoagulant therapy (OAT). However, it has been reported that misleading INR results can be obtained from OAT patients with lupus anticoagulant (LA). To investigate this claim, we studied 35 OAT patients, 14 of whom had anti‐phospholipid syndrome (APS) with a documented LA. Attainment of anticoagulation was confirmed by chromogenic assay of factor VII and factor X. Prothrombin times were performed using eight thromboplastins (five derived from rabbit brain, two recombinant human tissue factor and one made from human placenta) with an International Sensitivity Index (ISI) of <1.40. When using the thromboplastin manufacturers' ISI there was a significant difference (ANOVA, P<0.0001) between INR results obtained with the eight reagents for both APS (average CV = 12.4%) and non‐APS (average CV = 12.5%) patient groups. Variation using the eight thromboplastins was assessed by calculating the CV for each sample; these values were then pooled for each patient group to give the average CV for all samples with all reagents for the two patient groups. Results for both patient groups exhibited markedly reduced variation (APS group average CV = 6.5%, non‐APS group average CV = 5.8%) when locally assigned ISI values were employed in the calculation of INRs. Our data does not support the suggestion that the INR may not reflect the true level of anticoagulation in the long‐term warfarin‐treated patient, in whom lupus anticoagulant was detected. However, there was strong evidence that thromboplastin use should be restricted to those clot detection systems for which the reagent's manufacturer has assigned an ISI, or local ISI assignment must be undertaken. The inappropriate use of a generic (i.e. optical or mechanical clot detection system without regard to specific analyser type) ISI value can lead to ambiguous results.

Summary. This study assessed the inter-laboratory imprecision associated with the measurement of the endogenous thrombin potential (ETP). The initial studies used techniques that had evolved in each of the participating laboratories. Samples from normal healthy subjects (n ¼ 10), two patients receiving coumarin therapy [International Normalized Ratio 2AE0 and 4AE0] and a further two subjects receiving treatment with unfractionated heparin (antiXa 0AE07 iu/ml and 0AE31 iu/ml) were assayed relative to a lyophilized normal plasma that had arbitrarily been assigned a potency of 100%. Considerable variation in potency estimates was observed between the centres, although individual laboratories using fully automated techniques achieved acceptable levels of imprecision as assessed by the coefficient of variation (CV) (intra-assay CV < 9AE5%, inter-assay CV < 12AE5%). A second study to assess a similar range of samples, using a standardized assay protocol and incorporating appraisal of two chromogenic substrates, CBS.0068 or Pefachrom Ò TG, demonstrated markedly improved agreement in potency estimates between centres and good correlation (r > 0AE96) between the chromogenic substrates. Our data demonstrates that an automated ETP method can be standardized between laboratories and suitable levels of imprecision achieved, using different analysers (COBAS Mira at two centres and an ACL-300R) and two thrombin substrates. This indicates that more widespread use of ETP measurements in clinical laboratories is feasible.

No abstract

No abstract