We report six cases of protein S deficiency secondary to varicella. Five cases were complicated by thrombotic and vascular events, namely purpura fulminans and necrotic vasculitis, deep vein thrombosis and stroke. Two cases were associated with protein C deficiency and one case revealed a heterozygous factor XII deficiency. The underlying mechanism of this acquired protein S deficiency is unclear but could be related to a direct effect of zoster virus.

SummaryPreviously developed murine monoclonal antibodies (MAbs) to human β2-glycoprotein I (β2GPI), a plasma protein required for the binding of anti-phospholipid antibodies, were studied for anti-platelet reactivity and influence on platelet function. The six MAbs (IgG1 isotype) tested interacted with both intact and fixed platelets in a β2GPI-dependent manner. Carbamylated β2GPI was still recognized by MAbs but was unable to mediate platelet- antibody binding. MAbs induced aggregation and secretion responses of platelets in platelet-rich plasma (PRP) and whole blood, provided subthreshold concentrations of weak agonists (i.e. ADP or adrenaline) were added. When aggregation in PRP was evaluated by a counting technique instead of turbidometrically, the sole addition of MAbs led to a rapid fall in single platelets. Triggering gel-filtered platelets with MAbs together with β2GPI, but not its carbamylated form, led to platelet activation after a lag time, as monitored by aggregometry, measurements of ATP and β-thromboglobulin secretion and calcium mobilization. F(ab')2 fragments of one of the MAbs failed to activate platelets but inhibited the responses to the whole antibody. This process thus depends on MAbs binding to platelets through both Fab and Fc domains, as confirmed by the suppression of platelet responses upon pretreatment with the anti-FcγRII MAb IV.3. Aggregation and secretion induced by MAbs plus β2GPI did not require exogenous fibrinogen and were variably inhibited in the presence of acetyl salicylic acid, apyrase or Ca2+, depending on the concentrations used for the two proteins. We propose that platelet FcγRII crosslinking that follows a platelet-antibody interaction via the platelet binding protein, β2GPI, may be a new mechanism by which anti-β2 GPI antibodies activate platelets and/or synergize with weak agonists.

The lipid-binding inhibitor of coagulation, beta 2-glycoprotein I (beta 2GPI), has been shown to form the antigen to which some autoantibodies against anionic phospholipids (aPL) are directed. Six murine monoclonal antibodies (MAbs) of the IgG1 isotype were raised against human beta 2GPI and could be subdivided into three groups on the basis of mutual competition experiments. MAbs 9G1 and 8C3 (group A) markedly inhibited the binding of immunoglobulins from aPL-positive sera to beta 2GPI-coated wells. Using a lipid-based solid-phase radioimmunoassay, the MAbs interacted with both anionic phospholipids and phosphatidylethanolamine, but not phosphatidylcholine, in a beta 2GPI-dependent manner. A cross-reaction between beta 2GPI from several (including bovine) species was seen with one of the MAbs (9G1). All six MAbs induced dose-dependent prolongation of the DAPTT, DRVVT, KCT and TTI clotting times of human plasma, whereas 9G1 was the sole antibody to be inhibitory with plasma from bovine origin. Synergistic inhibitory effects were observed with MAbs used in pairs provided that they did not compete with each other for beta 2GPI binding. The anticoagulant activity of the MAbs was fully neutralized by the addition of freeze-thawed platelets. The MAbs described here resemble lupus anticoagulants in several respects which makes them valuable to study the involvement of beta 2GPI in the autoimmune thrombotic pathophysiology.

This retrospective, open-label, non-comparative study evaluated continuous infusion of recombinant factor VIII (ReFacto), B-domain deleted recombinant FVIII (BDDrFVIII), in patients with haemophilia A undergoing surgery and requiring >5 consecutive days of treatment. Sixteen patients from eight centres underwent a total of 20 procedures. Haemostatic outcome was assessed as 'excellent' or 'good' in 75% of procedures, and target FVIII:C levels were maintained throughout the continuous infusion period. The reported volume of blood loss during surgery was also within the normal range for non-haemophilic patients for the type of surgery performed. Red blood cell transfusions were required to balance excessive blood loss during BDDrFVIII continuous infusion in eight (40%) procedures (seven patients), five with bleeding or requiring volume replacement and three to treat anaemia secondary to blood loss. Non-serious adverse events considered by investigators as possibly or probably related to BDDrFVIII continuous infusion were infrequent (n = 5) considering the duration of treatment (n =239 cumulative days of continuous infusion), and all of these were mild-to-moderate in severity. No thromboembolic complications were reported except for one case of thrombophlebitis occurring at the infusion site. Only two patients (four events) experienced serious adverse bleeding; BDDrFVIII was otherwise well-tolerated. These data show that continuous infusion of BDDrFVIII provides reliable haemostasis and is an effective and well-tolerated regimen for patients with haemophilia A undergoing surgery.