Randomised trial of magnesium in in-hospital cardiac arrest. Duke Internal Medicine Housestaff.


Journal Article

BACKGROUND: The apparent benefit of magnesium in acute myocardial infarction, and the persistently poor outcome after cardiac arrest, have led to use of magnesium in cardiopulmonary resuscitation. Because few data on its use in cardiac arrest were available, we undertook a randomised placebo-controlled trial (MAGIC trial). METHODS: Patients treated for cardiac arrest by the Duke Hospital code team were randomly assigned intravenous magnesium (2 g [8 mmoles] bolus, followed by 8 g [32 mmoles] over 24 h; 76 patients) or placebo (80 patients). Only patients in intensive care or general wards were eligible; those whose cardiac arrest occurred in emergency, operating, or recovery rooms were excluded. The primary endpoint was return of spontaneous circulation, defined as attainment of any measurable blood pressure or palpable pulse for at least 1 h after cardiac arrest. The secondary endpoints were survival to 24 h, survival to hospital discharge, and neurological outcome. Analysis was by intention to treat. FINDINGS: There were no significant differences between the magnesium and placebo groups in the proportion with return of spontaneous circulation (41 [54%] vs 48 [60%], p = 0.44), survival to 24 h (33 [43%] vs 40 [50%], p = 0.41), survival to hospital discharge (16 [21%] vs 17 [21%], p = 0.98), or Glasgow coma score (median 15 in both). INTERPRETATION: Empirical magnesium supplementation did not improve the rate of successful resuscitation, survival to 24 h, or survival to hospital discharge overall or in any subpopulation of patients with in-hospital cardiac arrest.

Full Text

Duke Authors

Cited Authors

  • Thel, MC; Armstrong, AL; McNulty, SE; Califf, RM; O'Connor, CM

Published Date

  • November 1, 1997

Published In

Volume / Issue

  • 350 / 9087

Start / End Page

  • 1272 - 1276

PubMed ID

  • 9357406

Pubmed Central ID

  • 9357406

International Standard Serial Number (ISSN)

  • 0140-6736

Digital Object Identifier (DOI)

  • 10.1016/s0140-6736(97)05048-4


  • eng

Conference Location

  • England