More studies are needed before giving magnesium sulphate for pre-eclampsia

Editor—Clark et al argue that magnesium sulphate reduces the risk of eclampsia and should therefore be given to all women with severe pre-eclampsia.¹ The study they refer to in support of this view has reported outcome for 685 women out of 822 randomised. A more reliable guide to clinical practice is a systematic review of all relevant trials. Four trials have now compared magnesium sulphate with placebo or no anticonvulsant,² and the relative risk for the overall effect on eclampsia is 0.33 (95% confidence interval 0.11 to 1.02). Although this suggests that magnesium sulphate may reduce the risk of eclampsia, it is little help in estimating the size of such an effect. For example, the confidence interval includes everything from a 90% reduction in risk to a 2% increase. The number of events is also small (3 v 11), and there are many examples where high hopes based on this level of evidence have been dashed by the results of larger trials. For example, in the first trial of fetal movement counting, the numbers of stillbirths were 3 v 12. In the subsequent large trial the numbers were 99 v 100.³

Even if magnesium sulphate does prevent eclampsia, before it can be safely recommended we need to know whether, overall, it does more good than harm. As discussed by Idama et al,⁴ data from case-control studies have suggested that in utero exposure to magnesium sulphate may protect very low birthweight babies from cerebral palsy. This hypothesis awaits confirmation in randomised trials, however. What they fail to mention is a trial of magnesium sulphate tocolysis that was stopped because of an excess of deaths among babies exposed to magnesium sulphate.⁵ Although this trial was small, it is a timely warning that magnesium sulphate is not innocuous. We need to know far more about its short term effects on the woman and her child, and we need information about possible effects on long term development of the exposed children. Clinicians have plenty of reason to be uncertain about the role of magnesium sulphate for women with severe pre-eclampsia. Those who wish to contribute to the further evaluation of magnesium sulphate are welcome to join the collaboration on the magpie trial (magpie@ndm.ox.ac.uk).