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

Prophylactic magnesium is not indicated in myocardial infarction

Three Part Question

In [patients with suspected acute myocardial infarction] is [magnesium] effective at [reducing the incidence of ventricular fibrillation]?

Clinical Scenario

You see a 50 year old man with a 2 hour history of cardiac chest pain and an ECG suggestive of acute myocardial infarction. You decide to thrombolyse. The cardiology registrar suggests that you also give IV Magnesium to reduce the incidence of ventricular fibrillation. You wonder whether there is any evidence to support this.

Search Strategy

Medline 1966-11/00 using the OVID interface.
[(exp myocardial infarction OR myocardial OR AND (exp magnesium sulfate OR magnesium OR magnesium OR exp magnesium OR exp OR exp magnesium chloride OR magnesium AND (exp arrythmia OR OR OR exp ventricular fibrillation OR ventricular OR ( OR (exp mortality/ or] AND maximally senstive RCT filter LIMIT to human AND english.

Search Outcome

103 papers found of which 86 were irrelevant and 12 of insufficient quality for inclusion. The remaining 5 papers are shown in the table.

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Abraham AS et al,
94 patients with proven MI 2.4 g MgSO4 daily for 3 days vs glucosePRCTIncidence of:Analysed by group sequential design (interim analysis)
Ventricular Triplets8 vs 13% p=NS
R-on-T0 vs 2% p=NS
VT7 vs 15% p=NS
VF0 vs 4% p=NS
Total of above 14 vs 34% p=0.05
Roffe C et al,
2316 patients with suspected MI 8 mmol MgSO4 stat and 65mmol over 24hrs vs equal volume of salinePRCTOdds ratio (95% CI)Clinical significance of arrhythmias not described
VF0.74(0.46,1.20) P=NS
VT0.87(0.63,1.20) P=NS
SVT0.69(0.38,1.26) P=NS
AF0.92(0.69,1.23) P=NS
Heart block1.17(0.83,1.65) P=NS
Sinus Bradycardia1.38(1.03,1.85) p=0.02
Bhargava B et al,
78 patients with proven MI 73 mmol MgSO4 over 24hrs vs salinePRCTIncidence of :Small numbers
Sustained VT10 vs 20% p=NS
Nonsustained VT23 vs 50% p<0.02
VF5 vs 8% p=NS
SVT0 vs 6% p=NS
Bradycardia5 vs 3% p=NS
Asystole0 vs 3% p=NS
Mortality at 28 daysNone
In hospital mortality7.5 vs 8% p=NS
ISIS-4 investigators,
58,050 patients 80 mmol Mg over 24 h vs no infusionPRCTIncidence of :
VF 3.5 vs 3.8%
other cardiac arrest3.2 vs 2.9%
2nd or 3rd degree heart block3.9 vs 3.7% 0.01 < p <0.05
Heart failure17.8 vs 16.6% p<0.001
Cardiogenic Shock4.6 vs 4.1% p<0.01
profound hypotension16.8 vs 15.1% p<0.0001
5 week mortality7.64 vs 7.24% p=NS
Gyamlani G et al,
100 patients with proven MI 50 mmol Mg in 1st 24hr then 12 mmol Mg in next 24hr vs glucosePRCTIncidence of:Small numbers
SVT2 vs 8%p=NS
Sustained VT2 vs 10%p=NS
Nonsustained VT4 vs 12%p=NS
VF0 vs 4%p=NS
Total arrhythmias8 vs 34%p<0.01
Mortality4 vs 20%p<0.05


A number of small studies published have suggested that magnesium therapy significantly improves mortality following myocardial infarction. While the two larger studies show a trend to reduction in the incidence of VF but also demonstrates that this benefit is outweighed by an increased incidence of detrimental effects.

Clinical Bottom Line

Routine prophylactic magnesium in patients with myocardial infarction is not indicated.


  1. Abraham AS, Rosenmann D, Kramer M et al. Magnesium in the prevention of lethal arrhythmias in acute Myocardial Infarction. Arch Int Med 1987;147:753-5.
  2. Roffe C, Fletcher S, Woods KL. Investigation of the effects of intravenous magnesium sulphate on cardiac rhythm in acute myocardial infarction. Br Heart J 1994;71:141-5.
  3. Bhargava B, Chandra S, Agarwal VV et al, Adjunctive magnesium infusion therapy in acute myocardial infarction. Int J Cardiol 1995;52:95-9.
  4. Anonymous. ISIS 4: A randomised factorial trial assessing early oral captopril, oral mononitrate, and intravenous magnesium sulphate in 58050 patients with suspected acute myocardial infarction. Lancet 1995;345:669-85.
  5. Gyamlani G, Parikh C, Kulkarni AG et al. Benefits of magnesium in acute myocardial infarction : Timing is crucial. Am Heart J 2000;139:703.