Background: The treatment goal for patients with immune thrombocytopenia (ITP) is to raise platelet counts to levels that minimize or stop bleeding. Thrombopoietin receptor agonists (TPO-RAs) have been successfully and extensively employed as second-line therapy for ITP. TPO-RAs, however, have a small but significant increase in the risk of thrombosis.Aim: The aim of this study was to elucidate the mechanisms involved in the procoagulant effect of TPO-RAs.Methods: This is a prospective, observational and transversal study. Eighty-two patients with chronic primary ITP, 40 without treatment for at least six months (UT-ITP) and 42 responders to TPO-RA therapy (64.3% with eltrombopag and 35.7 % with romiplostim) were recruited. One hundred and twelve healthy participants were also included.ROTEM® (naTEM test: only recalcification) was performed on platelet rich plasma adjusted to a platelet count of 25 x 109/L. Clotting time (CT, time from start of measurement until 2 mm of amplitude [in seconds], alpha angle, which reflects the rate of fibrin polymerisation (tangent to the curve at 2 mm amplitude [in degrees]), maximum clot firmness, which reflects the maximum tensile strength of the thrombus (MCF, [in mm]) and LI60, which describes the percentage of maximum clot strength present at 60 min (in %), were recorded.Surface exposure of phosphatidylserine (PS), active caspase-3, -8 or -9 and prothrombinase complex binding to platelets were assessed by flow cytometry.Plasma and platelet levels of PAI-1 were determined by ELISA (eBioscience Ltd., Hatfield, United Kingdom). The effect of TPO and romiplostim on PAI-1 content of MEG-01 cells was evaluated by Western blot. Three MEG-01 cell cultures were initiated simultaneously: control without drugs and treated with either TPO (100 ng/mL) or romiplostim (53 μg/mL). Samples were collected at the start and after 24, 48 and 72 hours to determine the PAI-1 content.The statistical analysis was performed using SPSS 9.0 software (SPSS Inc., Chicago, Illinois, USA).Results: The ROTEM® studies showed significant differences in the dynamics of clot formation when comparing the control with ITP samples.There was a delay in clot formation in the UT-ITP group, as observed by a prolonged CT [expressed as median (p25-p75): control: 516 (490- 633) s; UT-ITP: 938 (914-1348) s, p<0.001], and a diminished alpha angle (mean±SD; control: 61.7±5.6 degrees; UT-ITP: 49.2±7.3 degrees, p<0.05). Nevertheless, samples from patients with UT-ITP reached the same MCF as those from healthy controls (control: 45.3±2.4 mm; UT-ITP: 46.9±3.7 mm). On the other hand, patients with ITP undergoing TPO-RA therapy presented an initial clot formation similar to that of the control group [expressed as median (p25-p75): CT, 672 (598-928) s; alpha angle, 55.8±5.8 degrees] but achieved a higher MCF (53.1±4.5 mm, p<0.05) and a reduced clot lysis after 60 min (control: 91.8±4.0%; UT-ITP: 93.7±4.0%, TPO-RA ITP: 97.6±1.7, p<0.05).Higher values of MCF observed with platelets from ITP patients treated with TPO-RAs might be a consequence of their augmented apoptosis signs: platelets from this group exposed more PS than controls and this situation was accompanied by an increased activity of caspases-3,7, -8 and -9 (Figure 1 A and B). Moreover, platelets from ITP patients on treatment with TPO-RAs bound more prothrombinase complex than platelets from UT-ITP patients and healthy controls (Figure 1 C).Reduced clot lysis observed in ITP patients treated with TPO-RA was due, at least in part, to increased plasma and platelet levels of PAI-1 (Table 1).Increase in platelet content of PAI-1 might be the result of the effect of TPO-RAs during megakaryopoiesis since treatments of MEG-01 cells with TPO or romiplostim induced a 3-fold increase in their endogenous PAI-1 content after an incubation period of 48 hs.Conclusion: The patients with ITP undergoing TPO-RAs therapy presented a procoagulant profile due to the formation of a more fibrinolysis-resistant clot because of increased platelet and plasma PAI-1 levels.Moreover, platelets from this group of patients showed more signs of apoptosis that causes a higher exposure of PS and, consequently, a larger surface for the binding of the prothrombinase complex.Work supported by grant from FIS-FEDER PI15/01457. NB holds a Miguel Servet II (FIS-FEDER CP14/00024). [Display omitted] DisclosuresÁlvarez-Roman:SOBI: Consultancy; NovoNordisk: Consultancy; Shire: Consultancy. Jimenez-Yuste:Grifols: Consultancy, Research Funding; Octapharma: Consultancy, Research Funding; CSL Behring: Consultancy; Bayer: Consultancy, Research Funding; Roche: Consultancy, Research Funding; Shire: Consultancy, Research Funding; Sobi: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; NovoNordisk: Consultancy, Research Funding. Butta:FIS-Fondos FEDER: Research Funding.