Magnesium sulphate has antiarrhythmic and antithrombotic properties, a coronary and systemic vasodilating action, a direct myocardial protective effect in experimental and clinical models of ischemia-reperfusion injury. Two meta-analyses have pooled the results of several small studies that had analyzed the effect of controlled hypermagnesiemia in acute myocardial infarction before the advent of thrombolytic and antithrombotic therapies. The results have shown a more than 50% mortality reduction, with a minimum estimated benefit of about 30%, and a reduction in ventricular arrhythmias of about 50%. In LIMIT-2, a double-blind trial of 2,316 patients where magnesium was administered as a 8 mMol bolus followed by a 24-hour infusion of 65 mMol, a 24% reduction in mortality was observed. However, these data have not been confirmed in the more than 58,000 patients of the ISIS-4 trial. In this study magnesium, at the same dose of the LIMIT trial, did not reduce 5-week mortality, neither in the general population (7.64% versus 7.24% in control patients, p = n.s.) nor in specific subgroups. The results of ISIS-4 have excluded the routine use of magnesium sulphate in acute myocardial infarction in the era of fibrinolysis and aspirin, beta-blockers and ACE-inhibitors. Nevertheless, magnesium administration could still be considered in certain clinical situations, such as 1) the presence of contraindications to fibrinolysis and aspirin, 2) the treatment of ventricular tachyarrhythmias unresponsive (or as an alternative) to lidocaine, 3) severe hypertension when beta-blockers are not indicated.