The MAGPIE Trial

We’re going to start dedicating a few episodes to landmark trials of obstetrics and gynecology. Have a suggestion for other trials we should review? Let us know!

The MAGPIE Trial

Do Women with pre-eclampsia, and their babies, benefit from magnesium sulfate? The Magpie Trial: a randomized placebo-controlled trial 

MAGPIE = MAGnesium sulfate for Prevention of Eclampsia

Find the trial text available here from The Lancet (2002)

Some general background information 

  • Magpie Trial Collaborative Group - huge group that spanned many continents and countries.

  • Funded by lots of people, but ultimately coordinated by the Resource Centre for Randomised Trials at the Institute of Health Sciences in Oxford, UK

  • Why was the study done? 

    • For many decades, anticonvulsant drugs were used on women with pre-eclampsia with the thought that it could prevent eclamptic seizures. Of these, magnesium sulfate was one of the medications 

    • However, in 1998, a systematic review looked at four trials compared anticonvulsants with no anticonvulsants, and of these, magnesium sulfate was thought to be the most promising choice for eclampsia prevention

    • At the time of the study, while magnesium was starting to be used more and more to prevent seizures in those that had preeclampsia, there were still those that used other medications like diazepam, other benzos, phenytoin, barbiturates, etc. 

  • What was the research question? 

    • Do women with pre-eclampsia or their babies (or both) “do better” if they are given magnesium sulfate compared to placebo, regardless of whether treatment is started before or after delivery and irrespective of any previous anticonvulsant therapies

Methods 

  • Who participated, and when?

    • Initial pilot trial ran from 2/23-7/14/1998 

    • Actual trial was from 7/15/1998-11/29/2001. 

    • Study was conducted in 33 countries and spanned 6 continents

    • Eligibility:

      • If they had preeclampsia and uncertainty about whether to use magnesium 

      • Had not yet given birth or <24 hr postpartum 

      • BP had to be >140/90 mmHg x2 , with proteinuria 1+ or more 

      • Note: their definition of severe preeclampsia were a little different 

        • Severe: DBP >110 mmHg x2 or SBP > 170 mmHg x2 or proteinuria >3+

          • Or DBP >100, or SBP >150, and proteinuria >2+ and at least two signs or symptoms of imminent eclampsia  

    • Exclusion criteria:

      • Hypersensitivity to magnesium, hepatic coma due to renal failure, or myasthenia gravis 

      • If urine output was <25 mL/hr, then the dose was halved  

  • How was the study done? 

    • Patients were randomized and groups were balanced for severity of preeclampsia, gestation at randomization, delivered or not, if given anticonvulsant drugs before trial entry, if multiple pregnancy, and country

    • Randomized to either magnesium or placebo

      • Magnesium: 4 g magnesium loading dose over 10-15 minutes; maintenance was then followed by infusion over 24 hours of 1g/hr

        • If areas of low resource needed to do IM, then it was an initial 4 g IV combined with maintenance of 5 g magnesium IM injected into each buttock (10g total) every 4 hours for 24 hours 

      • Placebo: similar looking pack, but was saline in the same mL amounts 

    • Monitoring:

      • Monitored reflexes - if these were depressed, then mag dose was adjusted to prevent toxicity 

      • Checked reflexes and respirations q30 minutes 

      • If eclamptic seizure: trial had to be stopped; eclamptic rescue pack was given with two packs; so those that had no mag were given 4 g mag total, and those with mag previously were given another 2g  

    • Outcomes of the trial:

      • Primary: eclampsia and death of baby before discharge from hospital (for women who were randomized before delivery)  

      • Secondary: serious maternal morbidity (won’t go through all of them, but essentially resp depression, resp arrest, pneumonia, cardiac arrest, coagulopathy, renal failure, liver failure, pulmonary edema, cerebral hemorrhage), toxicity (ie. need for ca gluconate, stopping or reducing mag), and other side effects of mag sulfate (ie. n/v, flushing, drowsiness, confusion, etc)

        • If women were randomized before delivery: also looked at complications of labor and delivery, neonatal morbidity 

      • Other: LOS, admission to ICU, NICU LOS 

What were the results? 

  • 10,141 women were randomized at 175 hospitals in 33 countries

    • 47% in Africa, 27% in the Americas, 15% in Asia, 10% in Europe 

    • Data was available for 10136 women, follow-up was available for 10,110 

  • Baseline characteristics (ie. age, primiparity, systolic BP at entry, severe preeclampsia, other problems of preeclampsia, gestational age at entry) were not different at entry 

  • Outcomes

    • Primary outcomes

      • Significantly fewer eclamptic convulsions among women with mag sulfate then those in placebo (reduction by about 58%) 

        • (0.8% vs. 1.9% - small numbers 40 vs 96), RR 0.42

        • This effect was seen in patients who had severe preeclampsia and those without severe preeclampsia (RR for both is 0.42) 

      •  No difference in baby death for those randomized before birth 

    • Maternal mortality - also lower among women allocated to mag sulfate (0.2% vs. 0.4%), Relative risk reduction of 45% 

    • Secondary outcomes

      • No clear difference in any measure of maternal morbidity or in composite measures of serious morbidity  

        • There WAS significantly increase in side effects of mag sulfate in mag group( ie. flushing, n/v, muscle weakness, headache, hypotension, dizziness, etc) 

      • No difference in neonatal morbidity 

What was the impact of all this? 

  • This was a huge landmark study in Ob/Gyn because of how many people it included across many countries, in both resource rich and resource poor settings 

  • Their data was very good: high compliance, high completion of the study 

  • Clearly demonstrated that mag sulfate decrease the risk of eclampsia from preeclampsia

  • No increased ill effects on babies 

  • Also provided a regimen (4g then 1g/hr) as well as timing (24 hours) 

What about now? 

At our institution, we only treat preeclampsia with severe features with magnesium.

But in this study, they included all patients with preeclamptics. Why the change? 

  • When looking at the MAGPIE trial, specifically at the women enrolled from high-resource settings in the Western world, the reduction rate of eclamptic seizures was not statistically significant 

    • RR 0.67, 95% CI 0.19-2.37 

  • A quarter of women reported adverse effects with magnesium sulfate, primarily hot flashes, and rate of CS was increased by 5% when mag was used 

  • Two small randomized trials that allocated women with preeclampsia without severe features to either placebo or mag → no cases of eclampsia among women allocated to placebo and no significant difference in proportion of women that progressed to severe preeclampsia (but small size limits this) 

  • Finally, rate of seizures in preeclampsia with severe features w/ mag is 4x higher than those without severe features (4/200 vs 1/100) 

  • Also, with calculations of previous data, it appears that 129 women need to be treated to prevent 1 case of eclampsia, but in those with symptoms (ie. headache, blurred vision, etc), NNT is 36 

  • What does the practice bulletin say?

    • Evidence less clear about preeclampsia w/o SF and mag 

    • Can be individualized by institution or by physician  

What about dosing? 

  • MAGPIE

    • Dosing is 4g loading followed by 1g/hr 

    • But at different institutions, we have seen 6g loading, 2g/hr or 4g loading, 2g/hr 

    • What’s right??  

  • The case for more vs. less magnesium

    • There is sparse data about the therapeutic range, which is usually 4.8-9.6 mg/dL or 4-8 mEq/L, but accurate mg concentration clinically effective in prevention of eclampsia has not been established 

    • Higher infusion rates increase potential for toxicity, and infusion rates >2g/hr have been associated with increased perinatal mortality in systematic review of randomized trials 

    • Lower starting bolus may increase time to therapeutic dosing 

    • Then.. there’s the BEAM trial…

      • Mg for cerebral palsy prophylaxis for babies from 24-31 weeks gestation  - regimen was 6g bolus followed by 2g/hr 

      • So many institutions will try and get two birds with one stone (6g bolus, 2g/hr for both CP prophylaxis and seizure prophylaxis if <32 weeks) 

    • … As well as the ACTOMgSO4 trial 

      • Doen in Australia and New Zealand for mag for fetal neuroprotection

      • Their dose was 4g bolus and 2g/hr 

    • Institution dependent, but we have seen a lot of combinations of both! 

Take away points: 

  • Magnesium sulfate can help decrease eclamptic seizures in patients with preeclampsia 

  • We have very good date to suggest it can decrease this rate by about 50% 

  • However, mag is not a benign drug - 25% have side effects

  • You should dose the mag somewhere between 4-6g loading dose and 1-2g/hr, but this is institution dependent 

  • It is also institution dependent regarding if preeclampsia without severe features needs mag (in MAGPIE, had same amount of decrease in eclamptic seizures as preeclampsia w/ SF) 

  • However, the number of those that get eclamptic seizures overall is low, and even lower in those with preeclampsia w/o SF, so there can be an argument for increased toxicity/side effects with increased NNT for those with PEC w/o SF and therefore forego it