TTP remains enigmatic both in terms of etiology and management. The most recent approach is aggressive plasma exchange (PE) employing cryopoor plasma for replacement, based on the pathogenetic relevance given to exceedingly large Von Willebrand (VWF) multimers in the determination of the syndrome with normalization during remission. PE with fresh frozen plasma (FFP) is better than FFP infusion as shown by a recent Canadian study, supporting the theory that to treat TTP an offending circulating agent needs to be removed from the patient's plasma in contrast to the hypothesis that a missing factor is to be given along with FFP. A more recent hypothesis is supported by the results of studies published by the end of 1998 [Moake J, Chintagumpala M, Turner N et al. Blood 1994;84:490-97; Moake J, McPherson PD. Am J Med 1989;87: 3-9N] which would show that TTP is mediated by auto-antibodies to VWF-cleaving protease, or is the result of deficiency of the protease ascribed to abnormalities in its production, function or survival. Plasmapheresis without plasma infusion is relatively ineffective perhaps because it does not increase the protease activity. Cascade filtration (CF) is the autologous counterpart of plasmapheresis. It has been used by our group since 1980 to remove from patients plasma macromolecules such as VWF, fibrinogen, LDL and circulatory immune complexes (CIC). After secondary filtration, the autologous plasma has a composition which is very similar to that of allogeneic plasma after cryoprecipitation, a product which used in the management of TTP. Based on this knowledge, in 1994 we began to use CF in the treatment of TTP patients. In the beginning (7 patients) CF was combined with a decreasing number of conventional PEs using allogeneic plasma for substitution. Lately only CF with some plasma supplementation has been used in the last 9 cases. From a clinical point of view our 16 patients achieved remission after a number of treatments (11 +/- 7) that compares sufficiently well with those required by our historical control group of 47 cases (14 +/- 13). Of course the patient's exposure to allogenic plasma was significantly lower for patients in the CF only group (1.4 +/- 1.2 plasma U/session) compared to the PE + CF group (4.4 +/- 2.3 plasma U/session) or for the controls treated by PE only (10.8 +/- 4.6 plasma U/session). There were no deaths in the CF or PE + CF groups and no untoward effect was observed. On the contrary there were 5 deaths (1 on the day of presentation) in the PE group, and 1 HBV and 2 HCV infections as well as 4 severe allergic reactions to plasma proteins (or passive antibody infusion). We conclude that CF is presently the best treatment to offer to patients suffering from sporadic TTP and that CF may contribute to expanding the knowledge of the pathogenetic mechanisms of this uncommon multisystem disorder.