Learning Objectives:After participating in this continuing education activity, the provider should be better able to:

1. Cite current evidence for the use of magnesium sulfate for fetal neuroprotection.

2. Appraise risks to the fetus and the mother related to magnesium sulfate.

3. Describe appropriate use of magnesium sulfate in clinical practice for prevention of neonatal cerebral palsy and death.

Preterm birth, defined as birth before 37 weeks' gestational age (GA), can be subcategorized as extremely preterm (GA < 28 weeks), very preterm (GA 28-32 weeks), and moderate to late preterm (GA 32-37 weeks). In the United States, the preterm birth rate was 10% in 2018, with 2.75% of infants born very preterm.1 Decreased GA at birth is related to increased mortality and morbidity rates, with preterm birth as the second leading cause of infant mortality.2

Cerebral palsy (CP) is the most common motor disability in childhood. Impacting approximately 1 in 323 children, CP is associated with both preterm birth and low birth weight.3 CP is a permanent disorder of movement, muscle tone, and posture that results in limitation of activity. Although CP is most commonly associated with prematurity, the etiology is multifactorial and can include any disturbance of the developing fetal or neonatal brain (eg, preterm birth, intrauterine growth restriction, multiple gestation, placental vascular disease, abruption, chorioamnionitis, perinatal stroke, congenital anomalies, birth asphyxia, and untreated severe maternal hypothyroidism).4 The immature brain is particularly prone to excitotoxicity, and inflammation is strongly implicated in the pathogenesis of CP.

In 1995, Nelson and Grether5 first reported that magnesium sulfate reduced CP in preterm very-low-birth-weight infants (< 1500 g). This finding proved controversial, as some reports corroborated the findings6 and others failed to show benefit.7 The neuroprotective action of magnesium sulfate may involve the following hypothesized mechanisms: (1) prevention of the actions of excitatory neurotransmitters (eg, glutamate) at N-methyl-D-aspartate receptors by acting as an endogenous calcium channel agonist at neuronal synapses, or/and (2) downregulation of pro-inflammatory and oxidative injury pathways.8

Currently, the American College of Obstetricians and Gynecologists (ACOG) and the Society for Maternal-Fetal Medicine support the short-term (<48 hours) use of magnesium sulfate for fetal neuroprotection before anticipated early preterm (GA <32 weeks) delivery. ACOG also recommends that physicians develop specific guidelines regarding inclusion criteria, treatment regimens, concurrent tocolysis, and monitoring.9

In this review, we provide an overview of the current evidence, risks and benefits, clinical considerations, and recommended protocols for neuroprotective magnesium sulfate administration.

Evidence of Benefits and Efficacy

Several randomized trials and meta-analyses have evaluated the neuroprotective effects of magnesium sulfate administered to women at risk of preterm birth. A 2009 Cochrane systematic review provides an overview of this evidence with an evaluation of 5 major randomized trials (Table 1).10 Three major trials were designed to assess the neuroprotective benefits of magnesium sulfate: (1) the Australian Collaborative Trial of Magnesium Sulfate (ACTOMgSO4),11 (2) the Beneficial Effects of Antenatal Magnesium Sulfate (BEAM) trial,12 and (3) the PREMAG trial.13 A fourth trial, the Magnesium and Neurologic Endpoints (MagNET) trial, was designed with a tocolytic arm and a neuroprotective arm; the subgroup of women in the neuroprotective arm was considered for the review.14 The fifth trial, the Magnesium Sulfate for Prevention of Eclampsia (MAGPIE) trial, evaluated the efficacy of magnesium sulfate for the prevention of eclampsia and death of the fetus.15

Table 1. Summary of Clinical Trials for Magnesium Sulfate as Fetal Neuroprotection

The ACTOMgSO4 trial, outlined in Table 1, showed no significant difference in death, CP, or composite of death or CP between treated and untreated offspring.11 The investigators found a significantly lower rate of substantial motor dysfunction at 2 corrected years of age in children who had been treated with magnesium [3.4% vs 6.6%; relative risk (RR), 0.51; 95% confidence interval (CI), 0.29-0.91]. The combined rate of death or substantial motor dysfunction at the corrected age of 2 years was also lower in the magnesium group (17% vs 22.7%; Table 1). There was no difference in necrotizing enterocolitis (NEC) between groups (RR, 0.96; 95% CI, 0.59-1.57). This led the authors to conclude that magnesium sulfate, when given immediately before very preterm birth, may improve pediatric outcomes without causing harmful effects.

The PREMAG trial, also outlined in Table 1, demonstrated decreased risk for gross motor dysfunction [odds ratio (OR) 0.62; 95% CI 0.41-0.93], death, CP, and cognitive dysfunction in the group of infants exposed to magnesium (OR, 0.68; 95% CI, 0.47-1.0), and they did not observe any major maternal adverse effects.13

Finally, the BEAM trial, the details of which are in Table 1, demonstrated no difference in the primary outcome of death or CP, but did show significant difference in reduction of moderate to severe CP by 2 years in the magnesium group.12 Lastly, there was no difference in all neonatal outcomes, including NEC (RR, 1.27; 95% CI, 0.97-1.66).

Most recently, Zeng et al16 published a meta-analysis of 11 studies (6 randomized controlled trials and 5 cohort studies) of magnesium sulfate for neuroprotection that included 18,655 preterm infants. Primary outcomes included fetal death, CP, intraventricular hemorrhage, and periventricular leukomalacia. A significant decrease in moderate to severe CP in the magnesium group was seen (OR, 0.61; 95% CI, 0.42-0.89; 4 studies; 7092 neonates). There was no difference in infant mortality (OR, 0.92; 95% CI, 0.77-1.11; 8 studies; 17,046 neonates). Rates of total mortality, death under 28 days or greater than 28 days, and death after discharge all showed a reduction in preterm infants exposed to magnesium sulfate; however, this failed to reach statistical significance. No statistically significant difference in risk of NEC between neonates exposed to magnesium sulfate compared with placebo was seen.

A Cochrane review concluded that antenatal magnesium sulfate therapy given to women at risk of preterm birth substantially reduced the risk of CP in their child (RR, 0.68; CI, 0.54-0.87; 5 trials; 6145 infants).10 The rate of substantial gross motor dysfunction was also significantly reduced (RR, 0.61; 95% CI, 0.44-0.85; 4 trials; 5980 infants). Importantly, the combined outcome of "death or CP" was not significantly reduced (RR, 0.94; 95% CI, 0.78-1.12), raising concerns that the lack of significance could be attributed to the increased risk of fetal death in those exposed to magnesium sulfate therapy. However, the "death or CP" combined outcome was significantly reduced in a subgroup analysis of the trials designed to specifically assess the neuroprotective effect of magnesium sulfate (RR, 0.85; 95% CI, 0.74-0.98), providing reassurance that decreased occurrence of CP is not a result of an increase in fetal death. The number of women needed to treat to benefit one baby is 63 (95% CI, 43-87). Of note, changes in dosing-including differences in loading dose (4 g vs 6 g) and maintenance dose (0 g vs 3 g/h)-did not affect CP risk, as there was a significant reduction in CP in any loading and any maintenance group (RR, 0.68; 95% CI, 0.51-0.91; 3 trials; 5292 infants).10 No significant difference was found between treatment groups for other outcomes within these varying dosing regimens.10

From the existing data, it is clear that magnesium has some neuroprotective benefits. However, many specific questions regarding dosing and timing remain. To address these, multiple secondary analyses of the above trials have been performed. Palatnik et al,17 using data from 936 infants whose mothers were exposed to magnesium versus placebo in the BEAM trial, attempted to determine whether magnesium sulfate infusion at the time of delivery or umbilical cord blood magnesium concentration is associated with outcomes of CP or death diagnosed by 2 years of age. The authors found no difference in the outcomes "CP or death" (OR, 0.87; 95% CI, 0.53-1.41) or "CP alone" (OR, 0.56; 95% CI, 0.23-1.33) between those with magnesium infusion at time of delivery and those without. Additionally, no association was found in the level of cord blood magnesium and the outcomes of "CP or death" (OR, 1.04; 95% CI, 0.78-1.52) or "CP alone" (OR, 1.29; 95% CI, 0.57-2.87). The authors concluded that neither magnesium sulfate infusion at the time of delivery nor magnesium cord blood concentration is associated with an increased risk of CP or death.

McPherson et al18 attempted to assess the impact of total magnesium duration on adverse neonatal outcomes in a secondary analysis of the BEAM trial. The study included 933 fetuses at 24 to 31 weeks' GA, subdividing them by the duration of magnesium exposure (<12 hours, 12-18 hours, and >18 hours). After adjusting for GA at delivery, there was no difference in CP or death between the groups, using the group with less than 12 hours' exposure as the reference group (adjusted OR, 1.03; 95% CI, 0.60-1.77 for 12-18 hours, and adjusted OR, 1.08; 95% CI, 0.57-2.03 for greater than 18 hours). Maternal adverse drug effects (eg, pulmonary edema or respiratory depression) and neonatal morbidities (eg, NEC, respiratory distress syndrome, bronchopulmonary dysplasia, and intraventricular hemorrhage) were similar across groups. Although the duration of magnesium sulfate infusion was not associated with risk of death or CP, optimal treatment durations for maximal neuroprotection remain unknown.

Fetal Risks

Magnesium sulfate was previously thought to alter the neurologic status of the neonate; however, randomized trials have raised no concerns about short-term adverse effects of antenatal exposure on the neonate, and no increased neonatal care or resuscitation is required.10 Another systematic review of 18 studies (2 randomized controlled trials, 12 prospective observational studies, 1 prospective cohort study, and 3 retrospective cohort studies) including 932 women evaluated the effect of magnesium sulfate administration on intrapartum fetal heart rate (FHR). Magnesium sulfate exposure was associated with a lower baseline FHR, decreased FHR variability, and reduced reactivity without an increase in decelerative patterns; however, changes were small and were deemed clinically insignificant.19 In neonates whose serum levels exceeded 4 mEq/L, as is typical with larger doses of magnesium for eclampsia prevention, neonates can present with symptoms of hypermagnesemia: apnea or hypoventilation, weakness or hypotonia, reduced deep tendon reflexes, or coma.20

In the case of neuroprotective magnesium, none of the seminal trials reported a need for resuscitation at birth, and therapy did not significantly affect the incidence of Apgar scores less than 7 at 5 minutes, hypotonia, or need for ventilatory support.10 There were also no differences in neonatal morbidity, including seizures, respiratory distress syndrome, bronchopulmonary dysplasia, or NEC. These findings were supported by a secondary analysis of the BEAM trial, which found no association between magnesium exposure and supplemental oxygen requirement, hypotension, hypotonicity, intraventricular hemorrhage, and death.21 Further, the MAGnesium sulfate for fetal neuroprotection to prevent Cerebral Palsy (MAG-CP) implementation project in Canada found a decreased risk of intensive resuscitation associated with the magnesium sulfate group compared with no magnesium (adjusted OR, 0.63; 95% CI, 0.54-0.73) or magnesium for other indications besides neuroprotection (adjusted OR, 0.81; 95% CI, 0.66-0.99).22

Several additional retrospective cohort studies evaluated the risk of spontaneous intestinal perforation (SIP), NEC, and death among infants exposed to antenatal magnesium sulfate therapy for neuroprotection (Table 2). SIP and NEC are separate and common gastrointestinal complications of premature and extremely low-birth-weight neonates, and they are associated with disruption of intestinal flora, motility, and/or mucosal integrity.23 Rattray et al24 found that GA was strongly associated with SIP (P < .01) and that higher antenatal magnesium dosage was associated with increased risk of SIP and death in extremely low-birth-weight infants, particularly those under 25 weeks' gestation. Conversely, a large multicenter retrospective cohort study of 11,789 infants admitted to neonatal intensive care units evaluated outcomes in the first 21 days after birth in infants exposed and unexposed to antenatal magnesium sulfate therapy.25 This study showed no difference in risk of SIP (OR, 1.08; 95% CI, 0.91-1.29) in addition to finding decreased mortality in the first 21 days after birth (OR, 0.76; 95% CI, 0.70-0.83).25 In another secondary analysis of BEAM, Kamyar et al24 showed that antenatal magnesium sulfate exposure was not associated with NEC in infants under 28 weeks' gestation, but neonates under 26 weeks' GA with magnesium exposure had increased odds ratio of death or severe NEC (adjusted OR, 1.90; 95% CI, 1.12-3.22).

Table 2. Summary of Studies Evaluating Magnesium Sulfate and Pediatric Gastrointestinal Outcomes

Overall, the evidence fails to demonstrate a significant association between magnesium exposure and NEC or SIP. The data for increased rates of NEC and SIP in neonates under 26 weeks' GA or who are extremely low birth weight warrant extra consideration in clinical scenarios for treating women at borderline viability.

Pediatric Outcomes

Short-term pediatric outcomes were evaluated by a meta-analysis in a 2009 Cochrane review.26 No reduction in risk of blindness, deafness, or developmental delay was noted among infants receiving magnesium for neuroprotection. Additionally, magnesium exposure did not decrease the frequency of adverse effects (eg, Apgar score <7 at 5 minutes, intraventricular hemorrhage, periventricular leukomalacia, neonatal seizures, or need for ongoing respiratory support). GA at randomization (<30 weeks vs 32-34 weeks) did not demonstrate a significant difference between exposed and unexposed patients.

Authors of the ACTOMgSO4 trial followed up with 77% of their cohort to evaluate long-term outcomes at school age after magnesium sulfate administration.27 Data were collected on 669 children at school age (6-11 years, mean age 8.4 years), corrected for prematurity. The authors found no significant difference in CP, abnormal motor function, or cognitive, neurologic, and behavioral outcomes between exposed and unexposed children. In utero magnesium exposure was associated with a trend toward improved survival (RR, 0.80; 95% CI, 0.62-1.03). The earlier finding of reduced gross motor dysfunction did not translate to an overall reduction in the severity of CP at school age.

Another trial, PREMAG, also detailed long-term follow-up data in 73% of their cohort, including 431 school-aged children (7-14 years, mean age 11 years).28 The authors reported no significant difference in neuromotor, cognitive, or language ability between exposed and unexposed children. The magnesium-exposed children had fewer cases of motor dysfunction, qualitative behavioral disorders, school grade repetition, and specific educational needs. Furthermore, overall better parental perception of child health was noted by parents of children who were magnesium-exposed. However, none of these differences were significant.

Maternal Side Effects

The side effects of magnesium sulfate are well-studied, given its widespread use in the prevention of eclampsia. Magnesium therapy can reduce serum calcium levels, rarely causing symptomatic hypocalcemia (hyperreflexia, altered mental status, and cardiac conduction abnormalities). Conversely, symptoms of magnesium toxicity include absent patellar reflex, respiratory depression, cardiac arrest, and respiratory arrest.11,15

The previously discussed 2009 Cochrane review demonstrated no significant difference in severe maternal outcomes, including death, cardiac arrest, or respiratory arrest between the magnesium sulfate and placebo control groups; only maternal hypotension and tachycardia reached statistical significance in the exposed group.10 Still, women assigned to the magnesium group were significantly more likely to experience side effects that led to therapy cessation (RR, 3.26; 95% CI, 2.46-4.31; 3 trials; 4847 women).10

Monitoring of urine output and deep tendon reflexes is required in patients receiving magnesium sulfate therapy. Calcium gluconate is administered in the setting of magnesium toxicity. A rare side effect of magnesium administration is pulmonary edema, so contraindications to magnesium therapy include severe acute pulmonary diseases such as adult respiratory distress syndrome (ARDS), pulmonary hypertension, or pulmonary edema. Monitoring of maternal serum magnesium levels should be considered in the setting of renal impairment.

Use of Magnesium Sulfate in Clinical Practice

An approach to the use of magnesium sulfate for neuroprotection before preterm birth is outlined later (Table 3). This is based on evidence cited earlier and is consistent with the most recent 2010 ACOG guideline, which states that physicians choosing to administer neuroprotective magnesium should develop protocols in accordance with one of the larger randomized trials.9

Table 3. Recommended ACOG Guidelines for use of Neuroprotective Magnesium

Women are considered candidates for treatment if they are at high risk of imminent preterm birth within 24 hours. This may include women with recent preterm premature rupture of membranes (PPROM), preterm labor (with cervical dilation between 4 and 8 cm) with intact membranes, or planned medically or obstetrically indicated preterm delivery (eg, fetal growth restriction, recurrent periodic decelerations, or other concerns with the patient or her fetus[es]).

Contraindications to the use of magnesium sulfate include the following: myasthenia gravis, myocardial compromise or cardiac conduction defects, impaired renal function, and acute pulmonary disease. Magnesium sulfate administration can precipitate a severe myasthenic crisis by neuromuscular blockade. Precaution must be exercised for patients with hypertension on a calcium channel blocker due to risk of hypocalcemia and cardiac events. Magnesium also alters cardiac conduction at high concentrations and should be avoided in patients with a history of cardiac ischemia. Magnesium sulfate is solely eliminated by the kidneys and can accumulate to cause magnesium toxicity in women with renal insufficiency (eg, urinary flow <100 mL/4 h); in these women, maintenance dosing should be adjusted or eliminated, although loading dose can remain the same due to volume of distribution remaining unaltered.

The randomized trials included pregnancies between 24 and 32 weeks' GA. Although the upper limit of GA has not been studied, a meta-analysis showed no difference in benefit for fetuses at GA less than 32 to 34 weeks compared with those at GA less than 30 weeks.29 For women at risk for imminent delivery at the lower limit of viability, it is recommended to confer with the neonatology team and jointly counsel families on potential management strategies. Magnesium sulfate may be administered for neuroprotection if families opt for neonatal interventions.

Data are limited regarding optimal maternal loading and maintenance doses, and it is an essential area for future study. The seminal trials (Table 1) favored a loading bolus of 4 g over 30 minutes with or without a maintenance infusion of 1 g/h; however, ACOG does not provide specific guidelines for dosing.9

In terms of timing of administration, magnesium should be given to women who are at high risk of imminent delivery within 24 hours. Because this assessment may be somewhat subjective, care should be taken to avoid administration of therapy to women with only threatened preterm labor or PPROM without preterm labor. Current data do not specify the optimal duration of therapy before delivery. Emergent or expeditious delivery should not be delayed to administer or complete the administration of magnesium sulfate. Magnesium sulfate therapy should be discontinued when the infant is delivered or limited to 24 hours, even if delivery has not occurred, as this was the duration of therapy studied in the seminal trials (Table 1). Magnesium should also be discontinued if the patient has significant complications from the magnesium sulfate infusion.

Data on retreatment with magnesium sulfate for neuroprotection are also limited, as only the BEAM trial allowed retreatment.12 Retreatment decisions should be made individually based on factors such as GA, time since initial treatment, and indication and urgency for delivery. Cumulative dose does not appear to be associated with variability in risk reduction for CP, risk of other neonatal morbidities, or risk of maternal adverse drug events.18

Special Considerations

Magnesium sulfate administration can result in increased maternal side effects when concurrently given with [beta]-agonists or calcium channel blockers (eg, nifedipine). Although data are sparse, limited evidence suggests possible increased risk of symptomatic hypocalcemia, hypotension, and cardiac suppression.15,26 Importantly, magnesium sulfate should not be chosen for tocolysis solely based on fetal neuroprotection, as magnesium sulfate is not an effective tocolytic agent.

In patients with preeclampsia or eclampsia, treatment for and prophylaxis against seizures supersedes the use of magnesium sulfate for neuroprotection. Preeclampsia protocol should be used if indicated, and magnesium sulfate therapy should not be delayed; magnesium therapy should not be discontinued if the patient was already receiving it for neuroprotection.

Conclusion

Preterm birth is a significant risk factor for CP. There is mixed evidence for a small but significant correlation between magnesium sulfate administration and decreased risk of CP, intraventricular hemorrhage, and fetal death. An evidence-based treatment protocol for the use of magnesium sulfate in anticipated preterm delivery within 24 hours is recommended when indicated for women at less than 32 weeks of gestation and for a maximum of 48 hours of therapy per ACOG, although specific guidelines for dosage and timing have not been outlined.9 No association has been found for the risk of maternal or neonatal adverse drug events, including spontaneous neonatal intestinal perforation and NEC. Notably, magnesium sulfate may interact with [beta]-agonists and calcium channel blockers, but they are not a contraindication; also, treatment for preeclampsia or eclampsia takes precedence over the neuroprotective magnesium protocol. A patient-centered discussion of the risks and benefits remains essential in the treatment process.

Practice Pearls

* Strong evidence supports magnesium sulfate administration to reduce the risk of CP in the case of imminent preterm birth.

* At less than 32 weeks' GA with expected delivery within 24 hours, magnesium sulfate should be administered for neuroprotection.

* Optimal dosage and timing are not clear, but seminal studies recommend a loading dose of 4 g and a maintenance dose of 1 g/h.

* Contraindications to use of magnesium sulfate include myasthenia gravis, myocardial compromise or cardiac conduction defects, and acute pulmonary disease.

* Delivery should not be delayed to administer magnesium sulfate.

* Monitor maternal vital signs and FHR per existing magnesium protocol.

REFERENCES