Antenatal magnesium sulphate to prevent cerebral palsy in very preterm infants

was also the value observed in the largest trial). For each individual assumption, the required sample size fell from the 11 776 calculated by Huusom et al., and the Z-value for the monitoring boundary was less than the value of 2.75 observed in the meta-analysis in two of the three scenarios. Hence, under these assumptions the treatment estimate would cross the monitoring boundary for two of the three scenarios, i.e. no further trials would be required to establish magnesium benefit. Minor changes in all three assumptions simultaneously (Scenario D, Table 1), which is a more reasonable approach than the conservative values of Huusom et al., result in a required information size of 4973—fewer than the 5357 participants already included in the current Cochrane meta-analysis. The possibility that the uptake of antenatal magnesium therapy might be slow is not surprising—it took over 20 years for the world to accept antenatal corticosteroids, which are known to improve fetal outcome. We are more optimistic about magnesium sulphate, however. The recent approval by the National Health and Medical Research Council of Australia of guidelines for antenatal magnesium sulphate (www.nhmrc.gov.au/publications/synopses/cp128 syn.htm) indicates that policy makers at the highest level in Australia have been convinced of its worth in preventing cerebral palsy. The American College of Obstetricians and Gynecologists concluded that antenatal magnesium sulphate reduces the risk of cerebral palsy among children who survive very preterm birth. Across the USA and France many centres now routinely use antenatal magnesium sulphate for fetal neuroprotection. Given the consistent and arguably compelling randomised trial data that are currently available, the statistical handwringing of Huusom et al. only do a disservice to fetuses destined to be born very preterm.