Intravenous magnesium therapy in acute myocardial infarction (MI) has been shown to be beneficial in many studies. However, these effects are not consistent from one trial to another, and their clinical significance is often questionable. A total of 78 consecutive patients were included in a prospective placebo controlled, randomized study of the effects of intravenous magnesium sulfate infusion in acute MI. Of these, 52 (66%) received intravenous streptokinase and the rest belonged to a non-thrombolysed group. These patients were randomized to receive magnesium sulfate infusion (8 mmol over 5 min followed by 65 mmol over 24 h) or placebo. The end points were serious arrhythmias, left ventricular ejection fraction (LVEF) and death during hospital stay and at the end of 28 days. LVEF measured by radionuclide ventriculography at discharge (9 +/- 5 days) was similar in two groups (41 +/- 11 vs. 39 +/- 10; magnesium vs. placebo respectively, P = 0.40). However, the LVEF of the placebo-treated thrombolysed group was higher than the non-thrombolysed group (42 +/- 10 vs. 37 +/- 0, P = 0.02). Incidence of nonsustained ventricular tachycardia (NSVT) was higher with placebo than with magnesium (50% vs. 23% respectively, P < 0.02, odds ratio = 0.29; 95% confidence intervals (CI) 0.1-0.85). Mortality during hospital stay and at the end of 28 days was also similar in both the groups. Thus, the administration of magnesium infusion as an adjunct to thrombolytic therapy remains to be of questionable benefit.