The New England Journal of Medicine August 28, 2008

A Randomized, Controlled Trial of Magnesium Sulfate for the Prevention of Cerebral Palsy

Dwight J. Rouse, Deborah G. Hirtz, Elizabeth Thom, et al.

Bottom Line

In pregnant women at imminent risk of early preterm delivery, antenatal magnesium sulfate did not significantly reduce the primary composite outcome of death or moderate-to-severe cerebral palsy, but it did significantly decrease the risk of moderate or severe cerebral palsy among surviving infants.

Key Findings

1. The primary composite outcome (stillbirth, infant death by 1 year, or moderate/severe cerebral palsy by 2 years) occurred in 11.3% of the magnesium sulfate group and 11.7% of the placebo group, showing no significant difference (RR 0.97; 95% CI, 0.77 to 1.23) [2.1.1].
2. In a prespecified secondary analysis, the incidence of moderate or severe cerebral palsy was significantly lower in the magnesium sulfate group compared to the placebo group (1.9% vs. 3.5%; RR 0.55; 95% CI, 0.32 to 0.95).
3. The risk of death did not differ significantly between the two groups (9.5% in the magnesium sulfate group vs. 8.5% in the placebo group; RR 1.12; 95% CI, 0.85 to 1.47).
4. No life-threatening maternal events occurred in either group, demonstrating the maternal safety of the magnesium sulfate regimen.

Study Design

Design
Multicenter RCT
Double-Blind
Sample
2,241
Patients
Duration
2 yr
Median
Setting
Multicenter, US
Population Pregnant women with singleton or twin gestations at imminent risk for delivery between 24 and 31 weeks of gestation
Intervention Intravenous magnesium sulfate (6-g bolus followed by a constant infusion of 2 g per hour)
Comparator Matching intravenous placebo
Outcome Composite of stillbirth or infant death by 1 year of corrected age, or moderate or severe cerebral palsy at or beyond 2 years of corrected age

Study Limitations

The trial failed to show a statistically significant reduction in its primary composite outcome, meaning the main evidence for efficacy relies on a prespecified secondary analysis of its individual components.
The overall absolute incidence of moderate to severe cerebral palsy was relatively low, necessitating a large number needed to treat (NNT of approximately 63) to prevent one case of cerebral palsy.
There was a slight, though statistically non-significant, numerical increase in infant mortality in the magnesium sulfate group (9.5% vs. 8.5%), which occasionally raises theoretical concerns regarding competing risks.
The trial utilized a relatively high dose of magnesium sulfate (6-g bolus followed by 2 g/hr) compared to other international neuroprotection trials, leaving some debate over the optimal dosing regimen.

Clinical Significance

The BEAM trial provided definitive, large-scale evidence that antenatal magnesium sulfate offers fetal neuroprotection when administered before early preterm delivery. Despite missing its primary composite endpoint, the robust 45% relative risk reduction in moderate or severe cerebral palsy fundamentally shifted obstetric practice. Leading organizations, including ACOG and SMFM, subsequently issued guidelines recommending the administration of magnesium sulfate to pregnant women at high risk for imminent preterm birth to reduce the incidence and severity of cerebral palsy.

Historical Context

Prior to the BEAM trial, retrospective data (such as the 1995 Nelson and Grether case-control study) and secondary analyses of earlier trials (like the Magpie trial for preeclampsia) suggested that maternal magnesium sulfate administration might protect the fetal brain and reduce the risk of cerebral palsy in premature infants. However, definitive, adequately powered randomized controlled trials specifically designed to evaluate this neuroprotective effect were lacking. BEAM, alongside other contemporary trials like ACTOMgSO4 and PREMAG, aimed to resolve this clinical question and successfully established magnesium sulfate as standard of care for fetal neuroprotection.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Magnesium sulfate is commonly used in obstetrics for both eclampsia seizure prophylaxis and, as shown in this trial, fetal neuroprotection. What is the proposed mechanism by which magnesium sulfate protects the developing preterm fetal brain from cerebral palsy?

Key Response

MgSO4 acts as an NMDA receptor antagonist, blocking glutamate-induced excitotoxicity, stabilizing cell membranes, and increasing cerebral blood flow, thereby preventing calcium influx and subsequent hypoxic-ischemic damage in the vulnerable preterm infant brain.

Resident
Resident

A patient presents in active labor at 28 weeks gestation and you plan to initiate magnesium sulfate for fetal neuroprotection based on the BEAM trial. What is the standard dosing regimen, and how does the management differ if delivery does not occur within 24 hours of administration?

Key Response

The BEAM trial used a 6-gram loading dose followed by a 2-gram/hour continuous infusion. Clinically, if delivery does not occur within 24 hours, the infusion is typically discontinued to avoid maternal toxicity and prolonged fetal exposure, which can lead to bone abnormalities, and can be restarted if imminent preterm delivery re-emerges.

Fellow
Fellow

The primary composite outcome of death or moderate-to-severe cerebral palsy in the BEAM trial was not significantly different between the magnesium and placebo groups. How do you interpret the divergent results between the primary composite outcome and the secondary outcome of CP among survivors, and what role does the concept of competing risks play here?

Key Response

The primary outcome was heavily influenced by neonatal mortality, which was slightly higher in the MgSO4 group, diluting the protective effect seen in the CP outcome. Fellows must recognize that analyzing CP alone among survivors introduces potential survival bias, but since the mortality difference was not statistically significant, the reduction in CP is accepted as a true neuroprotective benefit.

Attending
Attending

When counseling parents at 26 weeks gestation who are experiencing preterm labor, how do you explain the number needed to treat (NNT) for magnesium sulfate to prevent one case of cerebral palsy, and how do you weigh this against potential maternal side effects and the lack of mortality benefit?

Key Response

The NNT to prevent one case of moderate-to-severe CP is roughly 46 to 63 depending on gestational age. Attendings must balance this relatively high NNT with the severity of CP and the transient, manageable nature of maternal side effects like flushing and nausea, framing the intervention as low-risk and high-reward to facilitate shared decision-making.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The BEAM trial utilized a composite primary outcome combining death and cerebral palsy, which yielded a null result, while the secondary outcome of CP in survivors was significant. From a trial design perspective, what are the statistical and epidemiological trade-offs of using a composite outcome combining a highly frequent terminal event like death with a less frequent morbid event like CP in preterm populations?

Key Response

Using a composite outcome increases statistical power and addresses competing risks, since death precludes a CP diagnosis. However, if the intervention only affects the less frequent outcome (CP) and not the frequent outcome (death), the composite measure can statistically mask a true therapeutic effect, a classic pitfall in perinatal trial design.

Journal Editor
Journal Editor

If reviewing this manuscript today, how would you critically evaluate the authors' decision to highlight a positive secondary outcome (reduction in CP) given the negative primary composite outcome, and what specific caveats would you require in the abstract to prevent 'spin'?

Key Response

Editors must be vigilant against over-emphasizing secondary outcomes when the primary fails. A rigorous editor would require the abstract to explicitly state the primary outcome was null and frame the secondary findings as conditional, ensuring readers understand the risk of type I error from multiple comparisons and survival bias.

Guideline Committee
Guideline Committee

Current ACOG guidelines recommend magnesium sulfate for fetal neuroprotection before 32 weeks of gestation based heavily on the BEAM trial. Given that the trial's primary outcome was negative, what criteria and additional evidence justified elevating this secondary finding to a strong, Level A guideline recommendation?

Key Response

Guideline committees rarely base strong recommendations on a secondary outcome of a single trial. However, the BEAM trial was pooled with other trials in Cochrane meta-analyses, which consistently demonstrated a significant reduction in CP without an increase in mortality across multiple cohorts, providing the Level A evidence required for current ACOG recommendations.

Clinical Landscape

Noteworthy Related Trials

2002

Magpie Trial

n = 10,141 · Lancet

Tested

Magnesium sulfate (IV or IM)

Population

Women with pre-eclampsia

Comparator

Placebo

Endpoint

Eclampsia and perinatal death

Key result: Magnesium sulfate halved the risk of eclampsia and reduced the risk of maternal death without significant adverse effects on the baby.
2003

ACTOMgSO4 Trial

n = 1,062 · JAMA

Tested

Magnesium sulfate (4g IV bolus, then 1g/h)

Population

Women at risk of preterm birth < 30 weeks gestation

Comparator

Placebo

Endpoint

Death or cerebral palsy in the child at 2 years of age

Key result: Magnesium sulfate reduced the risk of substantial gross motor dysfunction and cerebral palsy.
2007

PREMAG Trial

n = 573 · Pediatrics

Tested

Magnesium sulfate (4g IV bolus)

Population

Women expected to deliver at < 33 weeks gestation

Comparator

Placebo

Endpoint

Severe white matter damage, death, or cerebral palsy

Key result: Showed a trend toward reduced cerebral palsy and significantly decreased the rate of gross motor dysfunction.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis