Evid Based Med: first published as 10.1136/ebm.1996.1.44 on 1 February 1996. Downloaded from Magnesium sulphate was superior to diazepam or phenyioin for eclampsia

The Eclampsia Trial Collaborative Group. treatment of additional convulsions. scores < 7 at 1 minute and fewer spe- Which anticonvukantfur women with Diazepam packages included taper- cial care nursery use days for babies eclampsia? Evidence from the Col- ing doses and MgS packages in- of women given MgS. MgS also led laborative Eclampsia Trial. Lancet. cluded calcium gluconate for MgS to fewer recurrences of convulsions 1995fun 10;34S:1455-63. toxicity. The MgS arm was started than phenytoin (5.7% vs 17.1 %, P < with an intravenous (IV) dose and 0.001) (NNT 9 (CI 6 to 14), RRR then either oral or IV MgS (with IV 67% (CI 48% to 79%)}*. A strong Objective diazepam used only for immediate trend existed toward lower maternal To compare magnesium sulphate control of seizures). The diazepam mortality with MgS than phenytoin (MgS) with diazepam or phenytoin and phenytoin arms were adminis- (2.6% vs 5.2%, P = 0.06), as well as less for reducing recurrent seizures and tered by IV MgS was compared with need for ventilation (14.9% vs 22.5%, maternal death in eclampsia. diazepam in 905 women and with P = 0.007), less pneumonia (3.9% vs phenytoin in 775 women. 8,8%, P = 0.005), less use of intensive Design care facilities (16.7% vs25.1%,P = Randomised controlled trial with Main Outcome Measures 0.004), and less infant mortality and concealed allocation. Recurrence of convulsions and ma- morbidity. ternal death. Setting Conclusions 23 centres in 8 countries compared Main Results Magnesium sulphate was superior to MgS with diazepam; 4 centres in 2 MgS led to fewer recurrences of con- diazepam or phenytoin for reducing countries compared MgS with vulsions than diazepam (13.2% vs recurrent convulsions in eclampsia, phenytoin. 27.9%, P < 0.001). {This absolute risk reduction (ARR) of 14.7% converts to including a strong trend toward re- Patients a number of eclamptic women who duced maternal mortality. 1687 women (mean age 22 y, 65% need to be treated (NNT) with MgS Sources of funding: Overseas Development primiparous) with eclampsia who had (rather than diazepam) to prevent 1 Administi-ation; World Health Organisation; Wellcome Trust. no contraindications to study drugs. recurrent convulsion of 7 (95% CI 5 to For article reprint Dr. L. Duley, Perinatal

Follow-up was > 99%, 11); the relative risk reduction (RRR) http://ebm.bmj.com/ was 53% (CI 37% to 64%)}*. MgS Trials Service, NPEU, Radcliffe Infimittry, Oxford 0X2 6HE, England, UK FAX 44- Intervention also led to a trend in fewer maternal 1865-126-360. Study centres chose the 2nd drug to deaths than diazepam (3.8% vs 5.1%, compare with MgS. The packages P= 0.3 l)f. There were no significant *Numbers calculated from data in article, appeared identical and included in- differences in serious maternal mor- structions and a loading dose, 24 bidity, intensive care, or infant mortal- f Data obtained from principal investigator. hours of maintenance therapy, and ity or morbidity except for fewer Apgar on September 23, 2021 by guest. Protected copyright. Commentary The study by the Eclampsia Trial Collabo- has existed about how to reduce the inci- trial comparisons. The study also showed a rative Group is a remarkable achievement. dence of convulsions after a first seizure. reduction in maternal mortality at die bor- Mercifully, eclampsia occurs in only 1 in 2000 MgS is commonly used in North America der of conventional statistical significance. births in Europe, although the incidence in but rarely in Europe. This huge multina- No reduction in perinatal mortality was developing countries is considerabiy higher. tional study may be seen as an example of shown—a nonsignificant trend toward in- Although rare, eclampsia is very important how to carry out effectively a worldwide creased perinatal death was observed when because hypertensive disorders are now the study that involves developing countries, MgS was compared with diazepam, and a main direct cause of maternal death in die in this instance 9 countries throughout nonsignificant trend toward decreased United Kingdom, documented at 1.8% in a South America, Africa, and the Indian perinatal mortality was observed when recent report (1). subcontinent. The co-ordinators did a phenytoin was compared widi MgS. MgS can reverse distal cerebral arterial brilliant job, and adherence to the proto- But the main message is clear. MgS now vasoconstriction and prevent endothelial col follow-up was remarkable (> 99%). must be the treatment of choice in eclampsia, damage there, probably by release of The study clearly shows that MgS (given unless the woman has had an adverse reac- prostacyclin, which results in the inhibition as a loading dose of 4 g intravenously for S tion to MgS (very rare). This holds true of platelet clumping. Pre-eclampsia and min followed by either an infusion of i g/h whether die convulsion occurs before or af- eclampsia, however, are multisystem diseases, for 24 h or 5 g intramuscularly every 4 h) ter delivery (2 0% of first seizures occur after and we do not know whether MgS has any was more effective than diazepam or delivery). Moreover, I would encourage any- beneficial effect on any other organs. phenytoin. MgS significandy reduced the one interested in clinical trials to read this Until now, a transatlantic controversy incidence of recurrent convulsions in both Continued on page 45 44 Therapeutics Evidence-Based Medicine January/February 1996