SummaryOne of the frequently proposed mechanisms for pregnancy losses refers to uteroplacental thrombosis. However the contribution of classical thrombotic risk factors remains questionable and, if real, does not account for a large number of pregnancy losses. The aim of this study was to investigate the presence of circulating procoagulant microparticles, a new marker of cell activation already associated with various prothrombotic clinical settings. Microparticles were assessed by an original prothrombinase assay on platelet depleted plasma obtained from 74 women with a history of pregnancy loss without apparent cause and 50 controls. Patients were studied at least 2 months after the last obstetrical event and were classified into 2 groups: 49 women with at least 3 consecutive spontaneous abortions at or before the 10th postmenstrual week and 25 with at least one fetal death beyond the 10th postmenstrual week. Among the 74 patients, 41 had increased levels of circulating microparticles, 29 belonging to the group of early pregnancy loss (59%) and 12 to the group of late pregnancy loss (48%). The high prevalence of increased levels of procoagulant microparticles in both groups makes this new marker very promising for the understanding, follow up and therapeutical handling of pregnancy loss.

SUMMARYAntiphospholipid antibodies (APL) are usually detected using solid-phase immunoassays, where cardiolipin is the most common capture antigen. Phosholipids are believed to adopt a monolayer organization when coated onto polystyrene after evaporation of the solvent. However, bilayer phospholipids are probably those evidenced as microparticles or cell fragments circulating in vivo under various pathological circumstances. The surface density of monolayer phospholipids on polystyrene is six times lower than that of bilayer phospholipids. In order to assess the influence of phospholipid organization on the detection of APL, we prepared glass microspheres coated with bilayer phospholipids (cardiolipin, phosphatidylcholine, cholesterol). Such lipospheres enabled us to study the binding of antibodies in 1:100 diluted plasma samples from patients with anti-cardiolipin antibodies of ¹ M annexin V and with a large excess of liposomes of the same composition. The absence of correlation between liposphere and ELISA results suggests that different subgroups of antibodies are detected depending on the method. The detection of APL using bilayer phospholipids is an original assay and may represent a more physiopathological approach to the specificity of APL.