Search

Patients with chronic idiopathic thrombocytopenic purpura (CITP) have been described to have bleeding times (B.Ts) that were shorter than would be predicted by their platelet counts. This phenomenon was explained by the presence in CITP of a young platelet population with increased hemostatic competence (NEJM 287:155, ’72). In contradistinction, we have observed patients with CITP to have a bleeding tendency at platelet counts >75,000/cu mm. We therefore evaluated B.Ts and platelet arachidonic acid (AA) metabolism in 7 patients with CITP who demonstrated increased amounts of platelet associated IgG (PAIgG >3fg per platelet) and compared them to 20 healthy controls. 3/7 patients with CITP and platelet counts of >75,000/cu mm demonstrated marked prolongations in their B.Ts. (10’, 12’ and 14’, normal

The complexity of tests used to measure PAIgG has diminished its effectiveness as a measure of platelet bound antibody. We recently reported a simplified radial immunodiffusion technique to quantify PAIgG using washed solubilized platelets and low level IgG immunoplates (Behring-Calbiochem). Three days were required for diffusion to reach completion. Normal values ranged from 1.6 - 7.0 fg IgG/platelet. 92% of patients with immune thrombocytopenia (ITP) had values which exceeded the mean by greater than 2 S.D. units. To further simplify and hasten the measurement, we have developed a nephelometric method. 0.25 × 108 platelets, obtained from 10-30 ml of blood collected in EDTA are solubilized using 1% triton X-100 plus sonication. Nephelometric buffer containing polyethylene glycol is added followed by anti-human IgG. Relative light scattering units thus generated can be converted to ng of IgG by comparison with standards. Results of PAIgG levels can thus be made available within several hours of blood collection.These results are in excellent agreement with PAIgG determined by radial immunodiffusion. Our cumulative experience indicates a mean normal control value of 3.9 ± 1.9 fg IgG/platelet. 90% (40/45) of patients with immune thrombocytopenia exceed the mean by greater than 2 S.D. units. Elevated values are also seen in SLE patients with (50%) or without (25%) thrombocytopenia, 50% of patients with sepsis and thrombocytopenia, 80% of patients with IgG myeloma, and 20% of patients with thrombocytopenias of diverse etiologies.

The diagnosis and management of immune thrombocytopenia requires an effective method for the quantitation of platelet- associated IgG (PAIgG). Although several methods have been recently devised for this purpose, the search continues for a method which will prove to be accurate and reproducible, and yet simple enough to be utilized by the general clinical hemostasis laboratory. We now describe the development of an assay for PAIgG which we believe satisfies these criteria.Human platelets were isolated by differential centrifugation, washed five times in Tris-HCl pH 7.4 or phosphate buffered saline pH 7.4, both containing 5mM EDTA, and disrupted in 0.1M sodium acetate buffer pH 5.0 containing 1% (v/v) Triton X-100. Rabbit antibody specific for γ chains of human IgG (purified IgG fraction from DAKO) was carbamylated such that its isoelectric point was decreased to roughly 5.0. Total PAIgG was measured by electrophoresis at pH 5.0 of Triton X-100 disrupted washed platelets against the carbamylated rabbit anti-human IgG (γ) in 1% agarose gels containing 0.5% Triton X-100.The mean total PAIgG for 8 normals, was 5.7 fg/platelet (range 4.9 to 7.0). This technique is applicable to the quantitation of alloantibodies bound to normal platelets (indirect assay) as well as the quantitation of PAIgG on platelets of individuals with immune thrombocytopenic states, such as ITP.

Employing the biological property of staphylococcal protein A to bind human IgG a test system has been devised to detect platelet associated IgG (PAIg) by quantitating platelet retention in an agarose column with immobilised protein A.Platelets were washed free of plasma and resuspended in Tyrode’s buffer with EDTA and 0.1% BSA. Platelet counts were performed on the suspension prior to and following passage through a miniature protein A column and the percentage of platelets retained calculated. An indirect test for serum platelet antibody utilised washed normal platelets following their incubation in patient serum.Less than 30% normal platelets or normal platelets incubated in normal serum were retained in the affinity column. Similarly in four samples from patients with thrombocytopenia of nonimmune origin there was less than 15% retention in the column. Platelets from two of five patients with probable ITP (all treated with corticosteroids) showed 85% retention in the column. The remaining patients fell within the normal range. The sera from three patients with proven platelet antibody using other techniques produced 65-80% retention in an indirect test. Four other sera from possible ITP patients (with comparatively high platelet counts) showed less than 30% retention. Specificity for PAIg was suggested by inhibition of retention from 80% to less than 30% following pretreatment of the colurrai with purified IgG or normal serum.The optimum conditions for the technique are being studied and its specificity and sensitivity ascertained in patients with thrombocytopenia of various kinds. If these preliminary results are extended, such a simple technique may prove of diagnostic value.