Alternative regimens of magnesium sulfate for treatment of preeclampsia and eclampsia: a systematic review of non-randomized studies.

INTRODUCTION The optimal dosing regimen of magnesium sulfate for treating preeclampsia and eclampsia is unclear. Evidence from the Cochrane review of randomized controlled trials (RCTs) was inconclusive due to lack of relevant data. MATERIAL AND METHODS To complement the evidence from the Cochrane review, we assessed available data from non-randomized studies on the comparative efficacy and safety of alternative magnesium sulfate regimens for the management of preeclampsia and eclampsia. Sources included Medline, EMBASE, Popline, CINAHL, Global Health Library, African Index Medicus, Biological abstract, BIOSIS and reference lists of eligible studies. We selected non-randomized study designs including quasi-RCTs, cohort, case-control and cross-sectional studies that compared magnesium sulfate regimens in women with preeclampsia or eclampsia. RESULTS Of 6178 citations identified, 248 were reviewed in full text and five studies of low to very low quality were included. Compared with standard regimens, lower-dose regimens appeared equally as good in terms of preventing seizures [odds ratio (OR) 1.02, 95% confidence interval (CI) 0.46-2.28, 899 women, four studies], maternal morbidity (OR 0.47, 95%CI 0.32-0.71, 796 women, three studies), and fetal and/or neonatal mortality (OR 0.87, 95%CI 0.38-2.00, 800 women, four studies). Comparison of loading dose only with maintenance dose regimens showed no differences in seizure rates (OR 0.99, 95%CI 0.22-4.50, 146 women, two studies), maternal morbidity (OR 0.53, 95%CI 0.15-1.93, 146 women, two studies), maternal mortality (OR 0.63, 95%CI 0.05-7.50, 146 women, two studies), and fetal and/or neonatal mortality (OR 0.49, 95%CI 0.23-1.03, 146 women, two studies). CONCLUSION Lower-dose and loading dose-only regimens could be as safe and efficacious as standard regimens; however, this evidence comes from low to very low quality studies and further high quality studies are needed.