T cell depletion with antithymocyte globulins (ATG) can be complicated by thrombopenia and hypercoagulability. The underlying mechanism is still unclear. We found that binding of ATG to platelets caused platelet aggregation, α-granule release, membrane phosphatidylserine exposure and the rapid release of procoagulant platelet microvesicles (MV). Platelet activation and MV release were complement-dependent and required membrane insertion of C5b-8 but not stable lytic pore formation by C5b-9. ATG also activated platelets via binding to the low-affinity Fc gamma receptor FcγRII. However, only complement inhibition but not blockade of FcγRII prevented MV release and subsequent thrombin activation in plasma. In 19 hematopoietic stem cell and kidney transplant patients, ATG treatment resulted in thrombopenia and increased plasma levels of d-dimer and thrombin-antithrombin complexes. Flow cytometric analysis of complement fragments on platelet MV in patient plasma confirmed dose-dependent complement activation by ATG. However, the rapid rise in MV numbers observed in vitro was not seen during ATG treatment. In vitro experiments suggested that this was due to adherence of C3b-tagged MV to red blood cells via complement receptor CR1. These data suggest a clinically relevant link between complement activation and thrombin generation and offer a potential mechanism underlying ATG-induced hypercoagulability.

The diagnosis of sideroblastic anemia is based on bone marrow aspiration, and the detection of ring sideroblasts (RS) in iron staining. The finding of laboratory parameters to approach this diagnosis still remains a great challenge. In this study, we analyzed the value of a specific erythrogram pattern from peripheral blood, produced by the ADVIA120 cell counter, to predict sideroblastic changes in the bone marrow. In a two step-design study, we first showed that 32/38 consecutive patients reporting > or =15% RS had such a pattern in the erythrogram. In the second step, we prospectively identified over a period of 32 months 21 patients with this typical erythrogram; 20/21 had > or =15% RS in the bone marrow. Hence, by this validation, we confirm that the erythrogram is highly predictive of RS in the bone marrow. The interpretation of the erythrogram should become daily practice in hematology to improve the efficacy to detect sideroblastic changes.