A prospective comparison of four techniques for measuring platelet-associated IgG.

We describe a prospective study comparing four different assays for PAIgG. Platelets from patients with a variety of thrombocytopenic disorders were collected into ACD, washed, and the PAIgG then measured using three assays for surface PAIgG. These included: (a) a direct binding assay using 125I-monoclonal anti-IgG (MoAb); (b) a direct binding assay using 125I-staphylococcal protein A (SPA); and (c) a two-stage assay. PAIgG also was measured using an assay for 'total' PAIgG following platelet lysis. The mean +/- SD number of molecules of IgG per platelet on washed platelets from 29 healthy, non-thrombocytopenic controls was: 86 +/- 80 (125I-MoAb); 94 +/- 96 (125I-SPA); 3520 +/- 1890 (two-stage surface assay); and 10,850 +/- 3720 (total PAIgG). A total of 62 different patients with idiopathic thrombocytopenic purpura or thrombocytopenia complicating systematic lupus erythematosus, and 73 different patients with 'non-immune' thrombocytopenia, were tested using each of the four assays. These 'non-immune' thrombocytopenic patients included patients with carcinoma, septicaemia, pre-eclampsia, chronic leukaemia, thrombotic thrombocytopenic purpura, haemolytic uraemic syndrome, acute leukaemia and myelodysplasia. All four assays gave similar results for both the immune and non-immune thrombocytopenic patients. The sensitivity of the assays for the most severely thrombocytopenic patients with immune thrombocytopenia was: MoAb 60%; SPA 88%; two-stage 82%; and 'total' PAIgG 88%. The specificity of the four assays in the non-immune thrombocytopenic patients was 57% 'total' PAIgG; 63% two-stage surface; 25% SPA; 38% MoAb.(ABSTRACT TRUNCATED AT 250 WORDS)