Clinical and prognostic significance of serum magnesium concentration in patients with severe chronic congestive heart failure: The Promise Study

Eric J. Eichhorn, P. K. Tandon, Robert Dibianco, Gerald C. Timmis, Paul E. Fenster, James Shannon, Milton Packer, Study Investigators Promise Study Investigators

Research output: Contribution to journalArticle

66 Scopus citations


Objectives. The aim of this study was to determine the prognostic significance of alterations in serum magnesium in patients with moderate to severe congestive heart failure. Background. Reductions in serum magnesium have been postulated to play a role in promoting arrhythmias and to have an adverse impact on survival in congestive heart failure, although support for this postulate is lacking. Methods. Serum magnesium levels were measured in 1,068 patients enrolled in a survival study of class III or IV heart failure at the time of double-blind randomization to milrinone, a phosphodiesterase inhibitor, or placebo. All patients received conventional therapy with digoxin, diuretic drugs and a converting enzyme inhibitor throughout the trial. The median follow-up period was 6.1 months (range 1 day to 20 months). Results. Patients with high serum magnesium (defined as ≥1.9 mEq/liter, n = 242) were less likely to survive than were patients with a normal magnesium level (n = 627) (p < 0.05, risk ratio = 1.41). Patients with a low magnesium level (defined as ≤1.5 mEq/liter, n = 199) had no difference in survival compared with the group with a normal magnesium level (p = NS, risk ratio = 0.89). At baseline, the patients in the high magnesium group were older and had more severe functional and renal impairment. An analysis after adjustment for these variables demonstrated no difference in survival comparing the low, normal and high magnesium groups. Although the three groups had no difference in frequency of ventricular tachycardia, length of longest run or frequency of ventricular premature beats on baseline Holler monitoring, the group with hypomagnesemia had more frequent ventricular couplets. Conclusions. Serum magnesium does not appear to be an independent risk factor for either sudden death or death due to all causes in patients with moderate to severe heart failure. Hypomagnesemia is associated with an increase in the frequency of certain forms of ventricular ectopic activity, but this is not associated with an increase in clinical events. The higher mortality rate among the patients with hypermagnesemia is attributable to older age, more advanced heart failure and renal insufficiency.

Original languageEnglish (US)
Pages (from-to)634-640
Number of pages7
JournalJournal of the American College of Cardiology
Issue number3
StatePublished - Mar 1 1993


ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this