A Cochrane review included 7 trials involving 972 patients. Magnesium sulphate was associated with a substantial reduction in the recurrence of seizures when compared to phenytoin (OR 0.34, 95% CI 0.24 to 0.49; 6 studies, n=972). The trend in maternal mortality also favours magnesium sulphate, but this difference is not statistically significant (RR 0.50, 95% CI 0.24 to 1.05; 3 studies, n=847). There were reductions in the risk of pneumonia (RR 0.44, 95% CI 0.24 to 0.79; 1 study), ventilation (RR 0.68, 95% CI 0.50 to 0.91; 1 study) and admission to an intensive care unit (RR 0.67, 95% CI 0.50 to 0.89; 1 study) associated with the use of magnesium sulphate rather than phenytoin. For the baby, magnesium sulphate was associated with fewer admissions to a special care baby unit (SCBU) (RR 0.73, 95% CI 0.58 to 0.91; 1 study, 518 babies) and fewer babies who died or were in SCBU for more than seven days (RR 0.77, 95% CI 0.63 to 0.95; 1 study, 643 babies) than phenytoin. There was no clear difference in perinatal deaths (RR 0.85, 95% CI 0.67 to 1.09; 2 studies, 665 babies).
Another Cochrane review on Mg versus lytic cocktail (abstract , review ) included 3 studies with a total of 397 subjects. Mg was better than lytic cocktail at preventing further seizures (RR 0.06, 95% CI 0.03 to 0.12; 3 studies, n=397; NNT 2, 95% CI 2 to 3) and was associated with fever maternal deaths (RR 0.14, 95% CI 0.03 to 0.59; 3 studies, n=397). Magnesium sulphate was also associated with less respiratory depression (RR 0.12, 95% CI 0.02 to 0.91; 2 studies, n=198), less coma (RR 0.04, 95% CI 0.00 to 0.74; 1 study,n=108), and less pneumonia (RR 0.20, 95% CI 0.06 to 0.67; 2 studies, n=307). There was no clear difference in the RR for any death of the baby (RR 0.35, 95% CI 0.05 to 2.38; 2 studies, 177 babies).
A third Cochrane review (abstract , review ) included 7 trials involving 1441 women. Most of the data were from trials of good quality; however, the allocated treatments could not be blinded after randomisation. Mg was associated with a reduction in maternal death when compared to diazepam (6 trials, n=1336; relative risk [RR] 0.59, 95% CI 0.37–0.94) and with a reduction in the risk recurrence of further fits (7 trials, n=1441; RR 0.44, 95% CI 0.34–0.57). There were few differences in any other measures of outcome, except for fewer Apgar scores less than seven at five minutes (2 trials, 597 babies; RR 0.72, 95% CI 0.55 to 0.94) and fewer babies with a length of stay in special care baby unit more than seven days (3 trials, 631 babies; RR 0.66, 95% CI 0.46 to 0.95) associated with Mg. There is no clear evidence of any other effects on maternal morbidity, or on perinatal morbidity or mortality.