The importance of effect mechanism in the design and interpretation of clinical trials: the role of magnesium in acute myocardial infarction

Abstract

The design and interpretation of randomized clinical trials and of meta-analyses of trials should be informed by a biologically plausible hypothesis of treatment effect. Without some insight on likely mechanism, trial conditions may not be optimum to allow a true treatment effect to be detected. Judgments on mechanism underpin decisions on the appropriateness of pooling studies in meta-analysis. Where statistical heterogeneity of trial results is found, the possibility of true biological effect modification can only be assessed by considering potential treatment mechanisms. These can then be tested in carefully designed laboratory models. Meta-analysis of 12 randomized controlled trials of intravenous Mg(2+) in acute myocardial infarction gives a null effect (odds ratio 1.02, 95% CI 0.96 to 1.08) with a fixed effects model, but with strong evidence of heterogeneity (P <.0001) due to a single large study in which Mg(2+) was generally given late and after fibrinolytic treatment. A random effects model gives a pooled odds ratio 0.61 (95% CI 0.43 to 0.87, P = 0.006). Laboratory models show that timing of Mg(2+) administration before or after reperfusion critically determines whether myocardial protection occurs.