It is generally thought that the plasma increase in factor VIII (FVIII) after desmopressin (dDAVP) infusion is related to the plasma increase in von Willebrand factor (vWF), which is the plasma carrier for FVIII. The aim of this study was to evaluate the FVIII and vWF responses in patients with type 2N vWD, characterized by the mild FVIII deficiency related to markedly decreased affinity of vWF for FVIII. At different times after one intravenous dose of dDAVP (0.3 or 0.4 microgram/kg) we measured the FVIII coagulant activity, FVIII antigen, vWF antigen and ristocetin cofactor activity, in eight patients with either Arg91Gln or Arg53Trp amino acid substitution in mature vWF. In all the patients, whatever their mutation, the dDAVP infusion resulted in a 2.3 +/- 0.7-fold increase of vWF and a variable rise (9.5 +/- 7.7 times) of FVIII, whereas the vWF capacity to bind FVIII was not improved. The FVIII response was more transient than vWF response, and FVIII half disappearance time was evaluated to be approximately 3 h. The data indicate that the stabilizing effect of vWF on FVIII is not responsible for the FVIII increase induced by dDAVP. The clinical implication of this study is that, in the 2N vWD patients, dDAVP may be a useful prophylactic or curative treatment when the test dose has been shown to be effective to reach a haemostatic FVIII level.

The patients with inherited bleeding diathesis related to quantitative, structural, and/or functional abnormalities of von Willebrand factor (vWF) are said to have von Willebrand's disease (vWD). We report here the clinical and laboratory features of a 50-year-old woman with a life- long history of excessive bleeding. Her particular laboratory data are factor VIII (FVIII) deficiency, subnormal bleeding time, and the presence of all plasma and platelet vWF multimers in normal amounts. Infused with FVIII/vWF concentrate, she showed a persistent increase in FVIII that led us to discard hemophilia A carrier or “acquired hemophilia” diagnoses. vWF devoid of FVIII purified from normal and patient's plasma by immunoaffinity on anti-vWF monoclonal antibody (MoAb) was immobilized onto polystyrene tubes that were further incubated with purified normal FVIII. The bound FVIII was evidenced using radiolabeled anti-FVIII MoAb. The data showed that the patient's vWF, in contrast to vWF purified from normal plasma, was unable to bind FVIII. Furthermore, no inhibitor of FVIII/vWF interaction was evidenced in incubating purified normal vWF with the patient's plasma before the addition of FVIII and anti-FVIII MoAb. These results support the concept that the bleeding diathesis of this patient appears to be due mainly to her abnormal vWF preventing FVIII/vWF interaction. This abnormality, which is not yet described in present classification of vWD, could be considered as a new variant of vWD.

A plasma von Willebrand factor (vWf) defect limited to its failure to bind factor VIII (FVIII) was previously characterized in a woman with FVIII deficiency and normal primary haemostasis. By using in vitro tests we found a similar pattern in three siblings of another family previously thought to be affected with mild haemophilia A. Furthermore, a decrease in vWf ability to bind FVIII was found in the parents and the brother of the three patients. This decrease was consistent with heterozygous expression of a recessive vWf gene abnormality. FVIII deficiency was corrected by infusion with a vWf concentrate almost devoid of FVIII coagulant activity. FVIII recovery and half-life thus obtained showed that this treatment was more effective than a FVIII infusion performed by way of comparison. These results indicate that this vWf defect may account for FVIII deficiency in patients without the usual laboratory and clinical features of von Willebrand's disease. Changes in therapy and genetic counselling following the new diagnosis in this family emphasize the need to search for such a vWf defect in patients in whom FVIII deficiency is not obviously X-linked.

SummaryA systematic study of the levels of FVIII antigen and activity was done in 133 haemophiliacs. No measurable antigen was demonstrated in the 60 severe haemophiliacs, with the exception of 3 patients with levels ranging between 1.5 and 4.5 U/dl, which corresponded to a dramatic FVIII deficiency. The situation was more complex with the 73 moderate and mild haemophiliacs: 39 of them (53.4%) had a partial, concordant deficiency of both the antigen and the procoagulant activity (1- and 2-stage methods), likely corresponding to a decrease in the synthesis of normal FVIII. The conclusion for the other 34 patients, was a qualitative abnormality of FVIII, the levels of antigen in comparison with the procoagulant activity (1-stage method) appearing to be either very reduced (n = 6) or even nil (n = 8), or on the contrary very much higher (n = 20) or normal. For 11 patients in this last category, we found a clear discrepancy between the procoagulant activity levels obtained with the 2 different techniques, the 1-stage levels being higher than the 2-stage levels. This discrepancy which was stable with restudy on multiple occasions and found in different members of the same families was remedied when vWF was absent in one-stage assay. This suggests that we have identified a variant of haemophilia A with an inherited abnormality of FVIII characterized by an in vitro vWF-dependent expression of procoagulant activity.

A new ion-exchange chromatographic procedure has been developed to produce a highly purified factor VIII (FVIII) concentrate from plasma cryoprecipitate. Solubilized cryoprecipitate, after adsorption on aluminium hydroxide and cold precipitation, was treated with 0.3% tri(n-butyl)phosphate and 1% Tween 80 at 25 degrees C for at least 8 h to inactivate lipid-enveloped viruses. The fraction was then loaded onto a column packed with DEAE-Fractogel TSK 650 M and chromatographed. Most proteins and TnBP-Tween 80 flowed through the gel unretarded. FVIII:c, which bound to the gel, was eluted by increasing the ionic strength, then was directly filter-sterilized without ultrafiltration or addition of a protein stabilizer. Chromatographic recovery of FVIII:c was 80-90%. After freeze-drying, FVIII:c was at a concentration of 42.5 +/- 9.5 IU/ml and had a specific activity of 175.4 +/- 37.8 IU/mg (n = 40), corresponding to a purification factor of over 12,000 from plasma. The typical yield of the freeze-dried FVIII:c from cryoprecipitate was 55-65%. FVIII:c was stable for over 24 h at room temperature in the liquid state. The mean content of fibrinogen and immunoglobulin G was only 65 and 100 mg/l, respectively, corresponding to 1.4 and 2.3 mg/1,000 IU FVIII:c. This concentrate, which is much purer than traditional FVIII concentrates, has been found to be well tolerated and effective in clinical treatment of hemophilia A patients.