An antiplatelet effect of magnesium has been demonstrated in vitro and ex vivo, and this effect may be advantageous in patients with acute myocardial infarction to inhibit reocclusion after coronary angioplasty or thrombolysis. We investigated the antithrombotic in vivo effect of intravenous magnesium in a placebo-controlled, blinded study in 46 male Wistar rats. Thrombus formation was induced by standardized arteriotomy of the femoral artery and partial inversion of the vessel wall to produce a thrombogenic area. The vessel was transilluminated and visualized dynamically by in vivo microscopy. Thrombus area was measured every minute for 30 minutes after removal of the vessel clamp by image analysis techniques, and the number of visible emboli was registered. Animals were randomized into three groups: groups 1 and 2 received saline (control group, n = 15) or MgSO4 (Mg-early group, n = 15), respectively, during the entire infusion period. In group 3 intravenous saline was given during preparation of the arteriotomy followed by infusion of MgSO4 (Mg-late group, n = 16) from 10 minutes after removal of the vessel clamp. Thrombus area was significantly reduced by 75% in the Mg-early group (P < .005) but not in the Mg-late group compared with the control group. Mean number of emboli was reduced during magnesium infusion. The serum magnesium level increased to 2.2 (2.1 to 2.5) and 3.5 (3.0 to 4.2) mmol/L after infusion in the Mg-late and the Mg-early group, respectively. In the present study, intravenous infusion of MgSO4 significantly reduced thrombus formation in vivo but only when it was given before reperfusion. The antithrombotic effect of magnesium may be utilized in patients with acute myocardial infarction to reduce the rate of reocclusion. Magnesium infusion may also be of value in elective arterial angioplasty, but this option has not been investigated in clinical trials. However, correct timing of magnesium administration is critical to obtain an efficient antithrombotic effect.