OBJECTIVE To compare the effect of prolonged maternal intravenous MgSO4 administration on amniotic fluid and serum concentrations of magnesium over time in preterm labor patients. METHODS Patients at 24-34 weeks of singleton gestation who presented with contractions (> 8 in 60 minutes) underwent amniocentesis to rule out intrauterine infection after signing an informed consent form. Some of these women who were clinically judged to have preterm labor received intravenous MgSO4: a 4-g loading dose followed by a 2 g/h maintenance dose. For technical reasons some patients had amniocentesis performed before initiation of MgSO4 (controls), while others had the procedure during tocolytic therapy (study subjects). Duration of treatment until amniocentesis was recorded, and blood samples were drawn at the time of amniocentesis. Maternal serum and amniotic fluid magnesium levels were measured using a colorimetric end point method. Data were evaluated using the Student t test and linear regression analysis. RESULTS Mean magnesium levels in maternal serum rose from 1.74 +/- 0.2 mg/dL in controls to 4.01 +/- 0.4 mg/dL in the study group. Mean magnesium levels in Mean magnesium levels in amniotic fluid were 1.41 +/- 0.18 mg/dL in the controls versus 2.28 +/- 0.53 mg/dL in the treatment group. Duration of MgSO4 treatment ranged from 3 to 22 hours. Amniotic fluid magnesium concentrations increased significantly during therapy (correlation coefficient = 0.89; p < 0.001), while maternal serum levels remained stable over time (correlation coefficient between maternal serum levels and time = -0.39; p=0.34). CONCLUSIONS Although maternal serum magnesium levels remained stable with intravenous MgSO4 therapy, concentrations continued to rise in amniotic fluid over time. However, amniotic fluid magnesium levels never exceeded maternal serum concentrations during the study period.