Background: The optimal scheme of thromboprophylaxis in bariatric surgery remains uncertain, because clinical practice is different between countries and randomized trials are lacking. Objectives: The primary objective of this randomized multicenter study was to determine the optimal regimen of enoxaparin providing an antifactor Xa peak activity between .3 and .5 IU/mL at equilibrium and to evaluate the course of procoagulant microparticles (MPs). Setting: University hospital. Methods: A total of 164 patients scheduled for gastric bypass were allocated to 3 groups (A, B, and C) of enoxaparin treatment (4000, 6000, or 2 x 4000 IU, respectively). Antifactor Xa activity was measured before and 4 hours after each injection from DO to D2. Doppler screening of the lower limbs was performed at D1, D9, and D30. Bleeding (BE) and thrombotic events (TE) were recorded during the first postoperative month. Total MPs were measured at DO, D9, and D30. MPs of leucocyte, platelet, and granulocyte origin were assessed in one third of the patients from each group. The 3 groups were compared by ANOVA. Results: A total of 135 patients were analyzed. The equilibrium of antifactor Xa peak levels was obtained 52 hours after the presurgery injection and 12.8%, 56.4%, and 27.3% of the patients reached the target in groups A, B, and C, respectively (P <.001). No TE was detected. BE occurred in 1, 2, and 6 patients in groups A, B, and C, respectively). Total MPs remained unchanged over time. While no significant variation was observed in the other groups, platelet GP1 b(+)-MPs increased (P=.01) at D9 in group C, suggesting an incomplete control of anticoagulation leading to cell activation and procoagulant MP release that was confirmed by the higher MP levels measured at D30 (P =.04). CD66(+)-MPs were also highly elevated at J9 and D30 in group C indicating a granulocyte contribution. Conclusions: This study shows that a single dose of enoxaparin 6000 IU/d allowed most of the patients to reach the target range of antifactor Xa activity without increasing the bleeding risk, with the most likely efficient reduction of procoagulant MPs. (Surg Obes Relat Dis 2015;0:000-000.)