Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

Elevation of serum magnesium may improve clinical outcome after aneursymal subarachnoid haemorrhage

Three Part Question

In a patient presenting with [aneurysmal subarachnoid haemorrhage] does [early elevation of serum magnesium by administration of intravenous magnesium sulphate] lead to a reduction in [subsequent vasospasm incidence, morbidity or mortality rates]

Clinical Scenario

A 40 year old female attends the emergency department as a standby call. Her partner gives you a history of acute headache followed by collapse and reduced conscious level. She is rapidly intubated and taken for brain CT, which confirms your suspected diagnosis of acute subarachnoid haemorrhage. The case is discussed with the neurosurgical team on call and transfer is arranged. You are keen to provide prophylaxis against future vasospasm and further brain injury. However, no oral nimodipine is available to go down the NG just inserted. By the time it is sourced the patient will likely have left the department. You are sure you have recently heard about the use of IV magnesium as a further preventative measure for vasospasm in SAH. Serum magnesium is 0.9mmol/L. While you wait for the anaesthetist to facilitate transfer, you resolve to find out for yourself....

Search Strategy

MEDLINE/EMBASE 1950-31/07/2010 using NHS evidence as an interface.
The Cochrane library was also searched directly for articles relevant to the three part question.
Limit to: Publication Year 1950-Current and (Clinical Queries Therapy sensitivity) and Humans and English Language

Search Outcome

60 papers were identified from medline and 196 from EMBASE total. 30 papers total directly addressed the three part question.
Two meta-analyses were found from 2009/2010. The 6 papers included in these reviews were subsequently discounted. One systematic review was discounted as abstract publication only. 19 papers were further discounted on abstract review.
This left 4 papers directly addressing the 3 part question.

1 Cochrane review was found with a subgroup analysis directly addressing the three part question

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Westermaier et al,
110 adult patients with confirmed aneurysmal SAH recruited <96 hours after SAH onset. Treatment group randomised to IV magnesium titrated to maintain serum level of 2.5-3mmol/L. Prospective, single blind, single centre RCT Delayed ischaemic infarction (DII)Event rate of 22% in the magnesium group compared to 51% in the control group. OR 0.28 (95% CI 0.12-0.64)Powered to detect a 30% reduction in symptomatic vasospasm between groups. This was extrapolated to provide an estimate to power for the primary outcome, DII. Degree of subjectivity to endpoints. Single centre study. Primary endpoint used as a surrogate parameter for functional outcome due to low numbers. Most patients not given current UK regimen (nimodipine) in addition to trial drug, therefore affecting external validity. No correction of serum hypomagnaesaemia in the control group.
Delayed cerebral ischaemia (DCI)Event rate of 37% in the magnesium group 66% in the control group. OR 0.3 (95% CI 0.13-0.66)
Glasgow outcome scoreMagnesium group had a median GOS of 4.5, with the control group scoring 4. p=0.84
MortalityEvent rate of 11% in the magnesium group compared to 19% in the control group. p=0.26
Wong et al,
Hong Kong
327 Aneurysmal SAH patients already receiving oral nimodipine, recruited and randomised to receive IV magnesium titrated to a serum concentration twice the baseline or saline placebo, within 48 hours of onset.Double blind, placebo controlled, multicentre PRCT with intention to treat analysis. Favourable outcome at 6 months (Glasgow Outcome Scale Extended 5-8)64% in the treatment group. 63% in the control group. (OR 1.0; 95% CI 0.7-1.6)
Clinical vasospasm25% in the treatment group. 18% in the control group. OR 1.5; 95% CI 0.9-2.5)
Ability to carry out Activities of Daily Living independently at 6 months (Barthel index >85)57% in the treatment group. 61% in the control group. OR 0.9 95% CI 0.6-1.4)
Zhao et al,
476 aneurysmal adult SAH patients included in prospective RCT's, with magnesium as the treatment arm in any form.Meta-analysisPoor outcome, defined as death, vegetative state or severe disabilityThe occurence of a poor outcome was less likely for the magnesium group than for control patients. OR 0.54 (95% CI 0.36 - 0.81)One study discounted in mortality analysis due to lack of available data. No clear definition of control group - was magnesium corrected to normal? Were all patients given nimodipine?
MortalityMortality rates were similiar. 14% in the magnesium group, 12% in the placebo group. OR 1.16 (95% CI 0.51 - 2.65)
Dorhout et al,
379 Adult patients with aneurysmal SAH exposed to intervention with magnesium versus control in addition to nimodipine in both groups, starting within 10 days of ictus. Systematic ReviewPoor outcome (death or dependence) RR 0.75 (95% CI 0.57 to 1.00)Sub group analysis of Cochrane review. One poor quality trial included, where randomisation mode not specified and data outcome incomplete.
Mortality at 3 monthsRR 0.93 (95% CI 0.61 to 1.41)
Secondary ischaemiaRR 0.66 (95% CI 0.45 to 0.96)
Ma et al,
513 patients receiving either intravenous magnesium sulphate or placebo in addition to nimodipine, within 4 days after SAH ictus. Meta-analysis of 6 prospective RCTs. Functional outcomeTreatment with magnesium was less likely to result in a poor outcome RR 0.62 (95%CI 0.46-0.83)In only one of the 6 trials included, was serum magnesium corrected to normal in the control group when found to be low. 2 of the 6 trials scored low quality according to the Jadad criteria.
Clincial evidence of delayed cerebral ischaemiaTreatment with magnesium was borderline less likely to result in delayed cerebral ischaemia RR 0.73 (95% CI 0.53-1.00)
Withdrawl for adverse effectsPatients receiving magnesium were more likely to suffer adverse effects requiring termination of the study medication RR 9.98 (95% CI 3.04-32.74)


Magnesium sulphate acts as an calcium channel blocker, with subsequent vasodilatory and neuroprotective properties. Hypomagnesaemia has also been reported in over 50% SAH patients and directly related to the occurrence of secondary ischaemia. The hypothesis that elevated serum levels may benefit acute SAH patients is therefore not surprising. Much research has been done in this area and trials are ongoing. For example, the MASH 2 study aims to recruit >1000 participants and results are expected within the next few years. Currently however, we are left with slightly mixed messages from the data. Comparison of magnesium against placebo alone in aneurysmal SAH patients has shown promise, reducing surrogate endpoint rates of infarct or secondary ischaemic event. However, comparison of magnesium therapy and conventional treatment (oral nimodipine) against conventional treatment alone is less convincing. Systematic reviews have shown improvement in outcome which teeters on the brink of statistical significance, but no benefit in mortality. A recent meta-analysis would suggest an improved outcome again, but offers no mortality data and demonstrates a high rate of adverse events. Finally, the most recent double blind RCT of high quality failed to show any significant difference between the two groups regarding outcome.

Editor Comment

ADL, activities of daily living; DCI, delayed cerebral ischaemia; DII, delayed ischaemic infarction; GOS Glasgow outcome scale; GOSE is Glasgow outcome scale; RCT, randomised controlled trial; SAH, subarachnoid haemorrhage.

Clinical Bottom Line

Prophylactic elevation of serum magnesium in aneurysmal SAH may improve outcome when given in addition to standard treatment. Current evidence is conflicting. Further multicentre randomised studies are ongoing to definitively address the question of benefit. If standard treatment (oral nimodipine) is unavailable, level 1 evidence exists to support the use of prophylactic magnesium in this patient group.

Level of Evidence

Level 1 - Recent well-done systematic review was considered or a study of high quality is available.


  1. Westermaier T, Stetter C, Vince G, et al. Prophylactic intravenous magnesium sulfate for treatment of aneurysmal subarachnoid haemorrhage: a randomized placebo controlled, clinical study Critical Care Medicine 2010;38(5):1284-1290.
  2. Wong GKC, Poon WS, Chan MTV et al. Intravenous Magnesium Sulphate for Aneurysmal Subarachnoid Haemorrhage (IMASH) Stroke 2010; 41(5): 921-926
  3. Zhao XD, Zhou YT, Zhang X et al. A meta analysis of treating subarachnoid hemorrhage with magnesium sulfate Journal of Clinical Neuroscience 2009;16(11):1394-1397
  4. Dorhout Mees S, Rinkel GJE, Feigin VL, et al. Calcium antagonists for aneurysmal subarachnoid haemorrhage. Cochrane Database of Systematic Reviews 2007; 3
  5. Ma L, Liu W-G, Zhang J-M et al. Magnesium sulphate in the management of patients with aneurysmal subarachnoid haemorrhage: a meta-analysis of prospective controlled trials. Brain Injury 2010; 24(5): 730-735