Abstract

In 1992 we recommended that, barring contraindications, magnesium should be given to all patients with a suspected myocardial infarction.1 A meta-analysis of 7 randomised controlled trials (1300 patients in all) had suggested that prompt intravenous infusion of magnesium could reduce early mortality by half.2 Then the larger LIMIT-2 trial involving 2316 patients found that early mortality was reduced by about a quarter among those allocated magnesium.3 It was this trial that prompted our recommendations. Last month, however, the results of a very much larger trial, this time involving over 58 000 patients, suggested that magnesium was of no benefit.4 In this article we review the place of magnesium in the treatment of patients with myocardial infarction, using the review to illustrate the differences between meta-analysis and megatrials.