The use of magnesium in the treatment of acute myocardial infarction remains controversial despite preliminary experimental evidence that magnesium plays a beneficial role as a regulator of thrombosis. This study examines whether oral magnesium treatment inhibits platelet-dependent thrombosis (PDT) in patients with coronary artery disease (CAD). In a randomized prospective, double-blind, crossover, and placebo-controlled study, 42 patients with CAD (37 men, 5 women, mean age 68 +/- 9 years) on aspirin received either magnesium oxide tablets (800 to 1,200 mg/day) or placebo for 3 months (phase 1) followed by a 4-week wash-out period, and the crossover treatment for 3 months (phase 2). PDT, platelet aggregation, platelet P-selectin flow cytometry, monocyte tissue factor procoagulant activity (TF-PCA), and adhesion molecule density were assessed before and after each phase. PDT was evaluated by an ex vivo perfusion model using the Badimon chamber. Median PDT was significantly reduced by 35% in patients who received magnesium versus placebo (delta change from baseline -24 vs 26 mm2/mm; p = 0.02, respectively). There was no significant effect of magnesium treatment on platelet aggregation, P-selectin expression, monocyte TF-PCA, or adhesion molecules. Oral magnesium treatment inhibited PDT in patients with stable CAD. This effect appears to be independent of platelet aggregation or P-selectin expression, and is evident despite aspirin therapy. These findings suggest a potential mechanism whereby magnesium may beneficially alter outcomes in patients with CAD.