The investigators undertook a randomized trial of magnesium sulfate (MgSO4) in 573 pregnant women at less than 33 weeks’ gestational age who were expected to deliver within 24 hours. They were seen in the years 1997–2003. Participants were randomly assigned to receive either a single 40-mL infusion of a solution containing 0.1 g/mL of MgSO4, totaling 4 g, or an isotonic saline placebo, over 30 minutes. The primary outcomes were neonatal mortality before discharge and neonatal cranial ultrasound findings of severe white matter injury (WMI) as evidenced by cystic periventricular leukomalacia, periventricular parenchymal hemorrhagic involvement defined as a large unilateral parenchymal hyperdensity, or a large single unilateral porencephalic cyst secondary to ischemic-hemorrhagic infarction. The actively treated and control women were similar in most respects. The median gestational age at entry to the trial was 30 weeks.
The trial was stopped after 6 years of enrollment when data from 688 infants were available. Rates of total mortality before discharge from hospital, severe WMI, and the combination of these outcomes all were lower in the MgSO4 group, but none of the differences was statistically significant. Overall neonatal mortality approximated 10%. Rates of neonatal death due to respiratory and/or neurological causes were similar for the actively treated women and placebo recipients. When all degrees of WMI were considered, the rate was lower in the MgSO4 group, as was the rate of nonparenchymal hemorrhage. No major adverse effects were noted in the MgSO4-treated women.
These findings suggest that MgSO4 infusion has neuroprotective effects when carried out before very preterm birth. Nevertheless, the evidence is not strong enough to recommend widespread use of MgSO4 in clinical practice at the present time. Treatments combining MgSO4 infusion with other neuroprotective measures should be further evaluated for their potential to further lower the risk of WMI and secondary disorders of brain development.