1RCSI medical student
Royal College of Surgeons in Ireland Student Medical Journal 2011;4(1):46-52.
The aim of this literature review is to determine if prenatal administration of magnesium sulphate (MgSO4) provides neuroprotection in preterm infants. Data was analysed from five randomised controlled trials (MagNET, ACTOMgSO4, MAGPIE, PREMAG and BEAM). The data from each trial supported a correlation between MgSO4 and neuroprotection; however, only one trial was statistically significant – BEAM. Previously conducted systematic reviews and meta-analyses combined data from the trials and produced statistically significant results in favour of MgSO4 for neuroprotection. Studies suggest that MgSO4 acts as an NMDA (N-Methyl-D-Aspartate) receptor antagonist, reducing the neuronal damage secondary to increased intracellular calcium. Other studies suggest that it prevents neuronal insult by decreasing intrauterine inflammation. The challenges of using MgSO4 are with determining the therapeutic window, appropriate timing of administration, re-treatment possibilities, bias in tocolytic choices, serious maternal side effects (hypotension, tachycardia), and neonatal side effects. Further research is needed to determine the neuroprotective mechanisms, specific indications for MgSO4, optimum gestational age, timing of administration, dosing, and need for re-treatment. Follow-up trials should assess the long-term effects of MgSO4 on preterm infants. In conclusion, MgSO4 provides neuroprotection in preterm infants and likely improves their quality of life.
Key words: Magnesium sulphate, neuroprotection, preterm infants, randomised controlled trials, systematic review, meta-analysis.
Preterm birth and extremely low birth weight (less than 1,000g) are major risk factors for detrimental neurologic outcomes such as cerebral palsy.1-4 Cerebral palsy is a group of disorders of varying severity that results in abnormal movement and posture, which ultimately leads to limited activity.5,6 It is due to non-progressive brain damage that occurs in utero or in infancy, with a multitude of consequences including chronic disability along with medical, emotional and economic burdens.7,8 Lifetime costs include direct costs such as physician visits, hospital stays, medications, and home/vehicle alterations, while indirect costs include productivity costs. 9 The lifetime cost for an individual with cerebral palsy is approximately €860,000 for men and about €800,000 for women. 10
It is proven that the risk of neurological abnormalities increases with decreasing gestational age, and 25% of new cases of cerebral palsy occur in infants born at less than 34 weeks’ gestation.11-13 There has been an increase in survival rates among these preterm and low birth weight infants that can be attributed to improvements in perinatal and neonatal intensive care. 4 Since these children have a much higher risk of neurological deficits like cerebral palsy, while also having a much higher rate of survival, it is of the utmost importance to explore preventive measures, such as magnesium sulphate (MgSO4), treatment that may improve their quality of life. MgSO4 has two principal uses in obstetrics.
First, it can be used as seizure prophylaxis in pre-eclampsia and treatment of eclampsia. Second, it is an agent of tocolysis, whereby it delays preterm labour to facilitate administration of corticosteroids to be given for foetal maturation, patient transport, or successful treatment of reversible aetiologies of preterm labour.23,24 It is believed that MgSO4 also provides neuroprotection in preterm infants when given to mothers when labour is imminent. Observational studies by Nelson, Grether, and Schendel et al. reported such findings, which were later supported by several randomised controlled trials. 15,16 This report will discuss the findings of five randomised controlled trials pertaining to the correlation between antenatally administered MgSO4 and subsequent neuroprotection in preterm infants, possible mechanisms that allow MgSO4 to act as a neuroprotective agent, clinical challenges faced when using MgSO4 for neuroprotection, and important research that needs to be done in the future.
A literature search was performed using Ovid/Medline (1950 to February 2010) to identify randomised controlled trials and other published data associated with using MgSO4 for neuroprotection in the foetus. A variety of key words were used including “magnesium sulphate”, “neuroprotection”, and “preterm”. The “AND” function was often used to combine these terms with each other or with the names of known authors. Bibliographies of significant studies, meta-analyses, and systematic reviews were assessed for additional relevant data. Google Scholar was used to obtain full text articles when only abstracts could be found in Medline. The Cochrane database was also searched. Further information and published data was retrieved from clinicians who had been to conferences where oral presentations relating to the subject were given and from those who had been involved in the development of hospital protocols to administer MgSO4 to patients for the purpose of neuroprotection in preterm infants. The following principal outcome measures were extracted from the systematic reviews/meta-analyses: relative risk (cerebral palsy, gross motor dysfunction and paediatric mortality); absolute risk of cerebral palsy with MgSO4; and, number needed to treat. The relative risk is the ratio of the probability of developing cerebral palsy, gross motor dysfunction or paediatric death in the MgSO4-treated group versus a control group. When this ratio is less than 1, developing any of the previously listed adversities is less likely to occur in the treated group than in the control group. The opposite is true if the ratio is more than 1.
The absolute risk is the probability of developing cerebral palsy with MgSO4 or with a control, and is calculated without comparing the two groups. The number needed to treat is the number of patients who need to be treated with MgSO4 in order for just one patient to benefit. As the number needed to treat increases, the effectiveness of the MgSO4 decreases.
Systematic reviews/meta-analyses were included if they evaluated the following five randomised controlled trials: BEAM, MagNET, ACTOMgS04, MAGPIE and PREMAG.5,17-21 The studies had to investigate any differences in relative or absolute risk of cerebral palsy compared with control groups and they also had to calculate number needed to treat with MgSO4 in order to prevent one case of cerebral palsy. Only published studies were used to explain the possible neuroprotective mechanism of MgSO4, while published and non-published information was used to assess best practice measures. One reviewer evaluated and selected the literature reviews, meta-analyses and studies that would be included in this paper.
The results from the reviews/meta-analyses calculated relative risk (cerebral palsy, gross motor dysfunction, paediatric mortality) and number needed to treat with confidence intervals, while the absolute risk was calculated in percentages. The reviews used a mixture of the following analytical tools to examine and combine the data from the various trials: Mantel-Haenszel chi-squared model, Wilcoxon rank-sum test, chi-square test, Fisher’s exact test, examining the symmetry of funnel plots, and statistically by using the Egger test, Review Manager software (RevMan 2008), MIX software version 1.7, and SAS software, version 8.2. In this literature review, these values were analysed and tabulated, allowing simple comparison of results.
Three systematic reviews/meta-analyses met the inclusion criteria. These reviews analysed data from the following randomised controlled trials: BEAM, MagNET, ACTOMgS04, MAGPIE, and PREMAG.5,17-21 The characteristics of these trials are outlined in Table 1 and the findings are summarised in Table 2. All of the randomised controlled trials compared MgSO4 with placebo/other treatment in patients at risk for preterm delivery (gestational age <37 weeks), were all performed in the last 10 years, had a large number of participants, and drew conclusions on MgSO4 and subsequent neuroprotection.
There are five significant randomised controlled trials that could have a major impact on the use of MgSO4 for neuroprotection: the Magnesium and Neurologic Endpoints Trial [MagNET]; the Australasian Collaborative Trial of Magnesium Sulphate [ACTOMgSO4]; the Magnesium Sulphate for Prevention of Eclampsia [MAGPIE]; PREMAG; and, the Beneficial Effects of Antenatal Magnesium Sulphate [BEAM].5,17-21 Four of the trials revealed a trend of reduced rates in cerebral palsy in MgSO4-treated groups with no effect on total paediatric mortality.5,18-21 However, the results regarding decreased cerebral palsy were only statistically significant in the BEAM trial.5 The MagNET trial resulted in reduced rates of cerebral palsy in the magnesium-exposed group compared to the placebo group in its tocolytic arm, but its neuroprotective arm showed more cases of cerebral palsy in the magnesium group compared to the placebo group. 17 However, according to Mittendorf et al.: “this study was too small and the complication of cerebral palsy was too uncommon for meaningful statistical analysis”. 17 This trial also suggested a trend towards increased foetal/childhood death in MgSO4 groups; however, this was refuted in subsequent systematic reviews and meta-analyses. 14,22,23,30 Another statistically significant result found in the individual trials was decreased substantial gross motor function in the BEAM and ACTOMgSO4 studies. 5,18 When the data from these trials was combined through well conducted systematic reviews and meta-analyses, statistically significant results were attained that confirmed the neuroprotective role of MgSO4 therapy administered to women at risk of preterm delivery. 14,22,23 The following conclusions were made in the reviews/meta-analyses regarding MgSO4 administered to mothers at risk for preterm birth: it reduced the risk of cerebral palsy in their children; 14,22,23 it decreased the absolute risk of cerebral palsy compared to placebo; 22,14 on average, the number of people needed to treat to prevent one case of cerebral palsy was 54;14,22,23 and MgSO4 reduced the rate of substantial gross motor dysfunction in their children. 14,22 MgSO4 administration also had no statistically significant effect on paediatric mortality, 14,22,23 or other poor outcomes in the newborn period or in the first few years of life (e.g., blindness, deafness, developmental delay). 14,22 The outcomes investigated in the newborn period were Apgar scores less than 7 at five minutes, ongoing respiratory support, intraventricular haemorrhage, periventricular leukomalacia, convulsions, 14,22 respiratory distress syndrome, bronchopulmonary dysplasia, mechanical ventilation, and necrotising entercolitis (NEC), even though a drift toward increased NEC was found. 14 All of the reviews also found that the studies giving MgSO4 exclusively for neuroproection provided the most compelling evidence for reduced risk of cerebral palsy. While the major randomised controlled trials did not independently exhibit significant results, collectively they produced strong evidence that may persuade clinicians to use MgSO4 as a neuroprotective agent.
The exact mechanism by which MgSO4 provides neuroprotection is still unknown. Currently, there are two theories that describe how magnesium may inhibit neuronal damage, namely hypoxic-ischaemic damage and inflammatory damage. Cerebral palsy is thought to be a result of periventricular white matter damage that predominates in premature infants, especially those born before 32 weeks gestational age.24,25 Periventricular damage is illustrated by loss of oligodendrocytes (brain cells that myelinate or insulate nerves) and gain of astrocytes (cells involved in scarring).25 Hypoxic-ischaemic damage is a result of low oxygen and glucose supply, which ultimately leads to excessive glutamate release. 24 Glutamate stimulates the N-methyl-D-aspartate (NMDA) receptor, allowing a large influx of sodium and calcium into the neuron. 24 Intracellular calcium induces several enzymes that cause neuronal death, while reperfusion causes oxidative damage through free radicals. 24 MgSO4 is an NMDA receptor antagonist24,31 and NMDA antagonists have proven to be strong neuroprotectants in various animal models. 26 However, NMDA receptors are vital in certain aspects of brain development, which raises the issue that MgSO4 could have the potential to disrupt normal foetal brain development if given at specific stages in neurodevelopment. 24 It is important to remember that there is a strong correlation between spontaneous preterm birth and intrauterine inflammation. 27 The fact that preterm birth due to inflammation and cytokine production leads to neuronal insult has been shown in animal models. 28,29 Burd et al. investigated the explicit mechanisms responsible for the injury and found that injured foetal neurons in mice are capable of damaging other normal neurons. 25 They also found that foetal brains of mice exposed to lipopolysaccharide, a bacterial antigen that causes intrauterine inflammation, exhibited abnormal neuronal morphology with decreased dendritic processes, which can ultimately disrupt neuronal synaptic communication. 25 A subsequent animal study demonstrated that foetal brains subjected to inflammation that were later treated with MgSO4 did not display neuronal injury associated with fewer dendritic processes. 27 The medical community continues to face difficulties regarding best practice and antenatal use of MgSO4 despite years of its clinical use and significant findings from combined data. 30 Concerns arise in the context of appropriate dosing and timing of administration, tocolytic choice, maternal side effects, and infant side effects. Several studies and reviews suggest that high tocolytic doses of 50g or more30 increase paediatric mortality. 31-34 Although the major randomised controlled trials used differing dosing regimens, total dose remained low. The median total exposure to MgSO4 in the ACTOMgSO4 trial was less than 10.5g (4g bolus infusion with 2g/hour maintenance up to 24 hours) with a maximum allowable total dose of 28g; 30 total exposure in the PREMAG trial was 4g (single bolus infusion); 20 and, median total dose in the BEAM trial was 31.5g (6g bolus infusion with 2g/hour maintenance). 31 The low doses used in all of these trials showed a decrease in a subsequent diagnosis of cerebral palsy. There may have been even more improvement in cerebral palsy outcome in the lower dose trials versus the higher dose BEAM trial. 30 Determining the therapeutic window above which MgSO4 could be toxic to the foetus has proven difficult, since detecting magnesium levels in the infant can be unreliable. Babies delivered soon after magnesium infusions may have falsely high magnesium levels, whereas babies born after prolonged magnesium exposure may have falsely low magnesium levels.
Another clinical challenge arises when considering the optimum time to administer MgSO4 to mothers in preterm labour. The MagNET and BEAM trials used active preterm labour and cervical dilatation (>4cm and 4-8cm) as indications for treatment. 5,17 Women were eligible for treatment in the BEAM, ACTOMgSO4 and PREMAG trials if delivery was expected within 24 hours. 5,18,20 The Brigham and Women’s Hospital (BWH) in Boston, Massachusetts, has developed a new protocol regarding MgSO4 for neuroprotection, and their goal for initiating infusion is set for four hours prior to delivery. 35 However, there are difficulties in predicting when a woman will deliver, and the question of whether or not to re-treat (give another loading dose and maintenance infusions) emerges if the patient has not delivered within 24 hours of the original dose. Although the largest randomised controlled trial, the BEAM trial, 5 would continue MgSO4 infusion if six hours had passed since treatment stopped, there is not enough evidence to strongly support re-treatment. 35
There is controversy over the number of times a patient can be re-treated and the amount of magnesium to which a patient can safely be exposed.
Concern over biased use of MgSO4 for tocolysis also arises given its neuroprotective quality. Clinicians are faced with the dilemma of administering magnesium as the primary tocolytic instead of what is currently used, or to use MgSO4 simultaneously with the hospital’s preferred tocolytic, such as indomethacin or nifedipine.36 Combining MgSO4 and calcium channel blockers is especially challenging, as it may lead to serious maternal side effects such as hypotension. 36 The BWH has dealt with this issue in their protocol by discontinuing nifedipine and beginning infusion with magnesium when delivery is believed to occur within four hours.35 However, tocolysis and neuroprotection should be thought of separately, and all of the relevant data surrounding various tocolytics, along with individual patient traits, must be considered in order to choose the most suitable tocolytic.31 Maternal side effects are another important issue in antenatal MgSO4 use. Both reviews and independent studies reported a greater number of adverse side effects in MgSO4-treated groups compared to placebo-treated groups.
Minor adverse effects included flushing, nausea, vomiting, sweating, problems at injection site, lethargy and blurred vision, 5,14,18,31 and seemed to subside once treatment was finished. 37 More serious side effects such as hypotension and tachycardia were also seen, 14,18,37 and were increased by as much as 50% in the MgSO4 group compared to the placebo group. 14 MgSO4 therapy was rarely associated with severe side effects such as death, 5,14,18,20,31,37 cardiac and respiratory arrest, 5,14,18,20,31,37 pulmonary oedema, 5,14 postpartum haemorrhage, 14,18,20 and caesarean section. 5,14,18,20 Clinicians must also be cautious of fluid overload, which can lead to severe cardiovascular complications. 35 Infants exposed to MgSO4 may also have side effects. Although these were not statistically significant in trials and reviews, they are a clinical reality. It has been noted by several clinicians at BWH that these infants often emerge less vigorous than those who have not been exposed to magnesium, but this is short-lived. However, if this is a known, transient effect, clinicians may be less concerned when it occurs and this could subsequently cause neglect of serious medical problems that warrant aggressive treatment. There is no evidence to support this finding, but it could be an interesting focus of research in the future. The issues associated with using MgSO4 for neuroprotection are crucial in terms of best practice and should be carefully considered.
Supplementary studies must be performed to provide imperative information about antenatal use of MgSO4 for neuroprotection in preterm birth. More randomised controlled trials are needed to determine the optimum gestational age, timing of administration, dosing, need for re-treatment, 14,22,23 increased risk of NEC, 14 and the immediate effects on the newborn infant. There is need for follow-up of the infants included in both new and previously performed trials into later childhood, 14,22 since neurological outcomes such as cerebral palsy are sometimes not fully recognised until children are older. 22
Indications for MgSO4 therapy must be further investigated, 23,31 since there was a lack of consistency in patient characteristics among the five major trials. 23 Various indications for treatment ranged from pre-eclampsia to preterm labour and preterm premature rupture of membranes, and MgSO4 may affect these indications differently. 23 The mechanism by which MgSO4 provides neuroprotection24,31 to the human foetal brain must also be determined; however, this poses ethical, technical and financial difficulties. 24 If the mechanisms at work in individual infants could be determined, this would provide the potential to develop treatments that matched specific patient needs. 24 Finally, substantial information regarding serious maternal side effects should be obtained. For instance, only the BEAM trial looked into maternal pulmonary oedema, while the ACTOMgSO4 was the only trial to explore maternal tachycardia. 5,18
Neuroprotection in preterm infants should be considered as a healthcare priority, since there has been an increase in the survival of preterm infants who have an increased risk of neurological injury leading to debilitating outcomes.
Neurological problems such as cerebral palsy result in serious burdens faced by the diagnosed individual, their carers, and the healthcare system. Antenatal use of MgSO4 is a simple and economical way to relieve such burdens. Randomised controlled trials demonstrated a trend towards neuroprotection without subsequent neurological impairment or paediatric mortality when MgSO4 was administered to mothers in preterm labour. These findings were verified by large-scale systematic reviews and meta-analyses. The neuroprotective mechanisms employed by magnesium have yet to be determined and clinical trials are warranted to resolve the challenges posed by antenatal use of MgSO4. Although some centres have begun using MgSO4 for neuroprotection, future research is key to determine its optimal clinical use.
I would like to thank my supervisor, Dr Steven Ringer MD PhD, Director of Newborn Medicine at Brigham and Women’s Hospital in Boston, Massachusetts, and Dr Nicole Smith MD MPH, who explained the MgSO4 protocol used at BWH and provided further information from conferences she attended.
- Anderson P, Doyle LW. Neurobehavioural outcomes of school-age children born extremely low birth weight or very preterm in the 1990s. JAMA. 2003;289:3264-72.
- Bhutta AT, Cleves MA, Casey PH et al. Cognitive and behavioural outcomes of school-aged children who were born preterm: a meta-analysis. JAMA. 2002;288:728-37.
- Hack M, Taylor HG, Drotar D et al. Chronic conditions, function limitations and special health care needs of school-aged children born with extremely low birth weight in the 1990s. JAMA. 2005;294:318-25.
- Doyle LW; Victorian Infant Collaborative Study Group. Evaluation of neonatal intensive care for extremely low birth weight infants in Victoria over two decades: II. Efficiency. Pediatrics. 2004;113:510-4.
- Rouse DJ, Hirtz DG, Thom E et al. A randomised, controlled trial of magnesium sulphate for the prevention of cerebral palsy. N Engl J Med. 2008;359:895-905.
- Rosenbaum P, Paneth N, Leviton A et al. A report: the definition and classification of cerebral palsy April 2006. Dev Med Child Neurol. 2007;109(Suppl.):8-14.
- Executive Committee for the Definition of Cerebral Palsy. Proposed definition and classification of cerebral palsy, April 2005. Dev Med Child Neurol. 2005;47:571-6.
- Kuban KCK, Leviton A. Cerebral palsy. N Engl J Med. 1994;330:188-95.
- Honeycutt A, Dunlap L, Chen H, al Homsi G, Gross S, Schendel D. Economic costs associated with mental retardation, cerebral palsy, hearing loss, and vision impairment – United States 2003. MMWR. 2004;53:57-9.
- Kruse M, Michelsen SI, Flachs EM et al. Lifetime costs of cerebral palsy. Developmental Medicine & Child Neurology. 2009;51(8):622-8.
- Himpens E, Van den Broeck C, Oostra A et al. Prevalence, type, distribution, and severity of cerebral palsy in relation to gestational age: a meta-analytic review. Dev Med Child Neurol. 2008;50:334-40.
- Moster D, Lie RT, Markestad T. Long-term medical and social consequences of preterm birth. N Engl J Med. 2008:359:262-73.
- Himmelmann K, Hagberg G, Beckung E, Hagberg B, Uvebrant P. The changing panorama of cerebral palsy in Sweden: IX, prevalence and origin in the birth-year period 1995-1998. Acta Paediatr. 2005;94:287-94.
- Conde-Agudelo A, Romero R. Antenatal magnesium sulphate for the prevention of cerebral palsy in preterm infants less than 34 weeks’ gestation: a systematic review and meta-analysis. Am J Obstet Gynecol. 2009;200(6):596-609.
- Nelson KB, Grether JK. Can magnesium sulphate reduce the risk of cerebral palsy in very low birth weight infants? Pediatrics. 1995;95:263-9.
- Schendel DE, Berg CJ, Yeargin-Allsopp M et al. JAMA. 1996;276:1805-10.
- Mittendorf R, Dambrosia J, Pryde PG et al. Association between the use of antenatal magnesium sulphate in preterm labour and adverse health outcomes in infants; Am J Obstet Gynecol. 2002;186:1111-8.
- Crowther CA, Hiller JE, Doyle LW et al. Australasian Collaborative Trial of Magnesium Sulfate (ACTOMgSO4) Collaborative Group. Effect of magnesium sulphate given for neuroprotection before preterm birth: a randomised controlled trial. JAMA. 2003;290:2669-76.
- Magpie Trial Follow-Up Study Collaborative Group. The Magpie Trial: a randomised trial comparing magnesium sulphate with placebo for pre-eclampsia. Outcome for children at 18 months. BJOG. 2007;114:289-99.
- Marret S, Marpeau L, Zupan-Simunek V et al. Magnesium sulphate given before very preterm birth to protect infant brain: the randomised controlled PREMAG trial. BJOG. 2007; 121: 310-8.
- Marret S, Marpeau L, Benichou J. Benefit of magnesium sulphate given before very preterm birth to protect infant brain. Pediatrics. 2008;121:225-6.
- Doyle LW, Crowther CA, Middleton P, Marret S. Antenatal magnesium sulphate and neurologic outcome in preterm infants: a systematic review. Obstet Gynecol. 2009;113:1327-33.
- Constantine MM, Weiner SJ. Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) Maternal-Foetal Medicine Units Network (MFMU). Effect of antenatal exposure to magnesium sulphate on neuroprotection and mortality in preterm infants: a meta-analysis. Am J Obstet Gynecol. 2009;114:354-64.
- Degos V, Gauthier L, Mantz J, Gressens P. Neuroprotective strategies for the neonatal brain. Anesth Analg. 2008;106:1670-80.
- Burd I, Chai J, Gonzalez J et al. Beyond white matter damage; foetal neuronal injury in a mouse model of preterm birth. Am J Obstet Gynecol. 2009;201:279.e1-8.
- Johnston MV, Nakajima W, Hagberg H. Mechanisms of hypoxic neurodegeneration in the developing brain. Neuroscientist. 2002;8:212-20.
- Burd I, Breen K, Friedman A et al. Magnesium sulphate to prevent adverse neurological injury: providing biological evidence. Am J Obstet Gynecol. 2009;201(6):S3.
- Elovitz MA, Gonzalez J, Chai J et al. Preterm labour is insufficient to evoke foetal brain injury: activation of inflammatory pathways is an essential mechanism. Reprod Sci. 2008;15:181.
- Elovitz MA, Mrinalini C, Sammel MD. Elucidating the early signal transduction pathways leading to foetal brain injury in preterm birth. Pediatr Res. 2006;59:50-5.
- Pryde PG, Mittendorf R. Contemporary usage of obstetric magnesium sulphate. Obstet Gynecol. 2009;114:669-73.
- Mercer BM, Merlino AA. Society for Maternal-Foetal Medicine. Magnesium sulphate for preterm labour and preterm birth. Obstet Gynecol. 2009;114:650-68.
- Cox SM, Sherman ML, Leveno KJ. Randomised investigation of magnesium sulphate for prevention of preterm birth. Am J Obstet Gynecol. 1990;163:767-72.
- Mittendorf R, Covert R, Boman J et al. Is tocolytic magnesium sulphate associated with increased total paediatric mortality? Lancet. 1997;350:1517-8.
- Crowther CA, Hiller JE, Doyle LW. Magnesium sulphate for preventing preterm birth in threatened preterm labour. The Cochrane Database of Systematic Reviews 2002, Issue 4. Art. No.:CD001060. DOI: 10.1002/14651858.CD001060.
- Smith N, Acker D, Ringer S et al. Use of magnesium sulphate for neuroprotection. Staff Memo 2010;1-2.
- Rouse DJ. Magnesium sulfate for the prevention of cerebral palsy. Am J Obstet Gynecol. 2009;200(6):610-2.
- Doyle LW, Crowther CA, Middleton P et al. Magnesium sulphate for women at risk of preterm birth for neuroprotection of the foetus. The Cochrane Database of Systematic Reviews 2009, Issue 1. Art. No.: CD004661. DOI: 10.1002/14651858.CD004661.pub3.