Abstract & Commentary
Synopsis: Magnesium sulfate used to treat preterm labor is associated with a high rate of adverse effects.
Source: Mittendorf R, et al. Am J Obstet Gynecol. 2002;186:1111-1118.
In June a report emerged in the American Journal of Obstetrics and Gynecology that should get every obstetrical provider’s attention. Mittendorf and colleagues randomized patients in preterm labor with cervical dilatation of less than 5 cm to having either magnesium sulfate in a standard dose (4 g bolus followed by 2-3 g/h) or to another tocolytic of the provider’s choice (tocolytic arm). If the patients had a cervical dilatation of 5 cm or more, they were assigned to have one 4-g dose of magnesium sulfate or sterile saline (prevention arm).
The results were eye opening. Adverse neonatal outcome was defined as death, intraventricular hemorrhage (IVH), paraventricular leukomalacia (PVL), or cerebral palsy (CP). In the tocolytic arm, 16 of 46 infants (26%) exposed to magnesium sulfate had adverse outcomes vs 9 of 46 infants (18%) exposed to other tocolytics. In the "prevention" arm, there were 11 adverse effects in 30 infants (37%) in the magnesium sulfate group compared with 6 of 29 (21%) in the saline group. Neither result was statistically significant, but when the magnesium sulfate-exposed infants from both arms were combined and compared against those mothers did not get magnesium sulfate, the results approached significance (32% vs 19%; P = .04).
Perhaps the most telling finding was the fact that the adverse effects outcomes bore a dose-effect relationship with the magnesium sulfate. The higher the serum level in mother and infant, and the larger the total dose given, the greater the chance of an adverse effect. With multivariate analysis, this effect, along with the effect of clinical chorioamnionitis, were the only ones remaining statistically significant after the variables of low birth weight and vaginal bleeding were weeded out.
Comment by John C. Hobbins, MD
In 1995, an article appeared that suggested that magnesium sulfate had a protective effect against IVH in very low birth weight babies.1 Certainly, there is good evidence that magnesium sulfate is the drug of choice in the treatment of pre-eclampsia and eclampsia, and that these results seemed to represent yet another bonus to patients receiving this drug. Interestingly, some even suggested that magnesium sulfate be given to everyone who was about to deliver before 32 weeks of gestation.
In many centers, including ours, magnesium sulfate has been the most commonly used tocolytic agent because: 1) it seems to decrease uterine contractions; 2) it is widely available; and 3) obstetricians have had substantial experience with its administration and in handling its side effects. However, its efficacy has not been rigorously tested in the same way as beta mimetic agents, which have been found in randomized trials to have little benefit in prolonging pregnancy.
So what to do? Should we go back to beta mimetics with their uncomfortable side effects and negligible effect (by the way, ritodrine is the only FDA-approved tocolytic)? Or—should we move on to prostaglandin synthase inhibitors (indomethacin or sulindac) with their potential neonatal effects, or calcium channel blockers, a largely untested group of drugs? The answers are simply not available. However, one thing has always been clear, when given a choice of doing "something" or doing "nothing," rarely is "nothing" chosen.
The findings of the above study pertain to magnesium sulfate as a tocolytic and should not be equated with magnesium sulfate as a therapeutic agent in pre-eclampsia. In the latter case there seems to be adequate evidence regarding its benefits, which can be compared favorably with its potential risk. Nevertheless, since the neonatal effects seem to be dose related, when using magnesium sulfate for any reason "less may be better" if one is considering how much to use and how long to keep it going.
Dr. Hobbins is Professor and Chief of Obstetrics, University of Colorado Health Sciences Center, Denver.
1. Nelson K, et al. Pediatrics. 1995;95:263-269.