[Skip to Content]
[Skip to Content Landing]
August 1988

The Effect of Intravenous Magnesium Therapy on Serum and Urine Levels of Potassium, Calcium, and Sodium in Patients With Ischemic Heart Disease, With and Without Acute Myocardial Infarction

Author Affiliations

From the Medical Department P/Chest Clinic, Bispebjerg Hospital and the Departments of Medical Cardiology, Medical Gastroenterology (Drs Aurup, Breum, and Cintin), and Clinical Chemistry (Drs Rasmussen and McNair), Hvidovre Hospital, University of Copenhagen.

Arch Intern Med. 1988;148(8):1801-1805. doi:10.1001/archinte.1988.00380080081022

• Serum concentrations of magnesium, potassium, calcium, and sodium were determined on admission of 224 patients to the hospital and after 2, 4, and 6 days in hospital; all were admitted to the hospital with suspected acute myocardial infarction (AMI). On admission, the patients were randomly allocated to 48 hours of treatment with magnesium intravenously or placebo. One hundred twenty-three patients had AMI (of whom 53 [43%] were treated with magnesium) and 101 had their suspected AMI disproven (of whom 51 [50%] were treated with magnesium). In a supplementary study, serum and urine levels of magnesium, potassium, calcium, and sodium, together with serum levels of parathyroid hormone, were determined before and after intravenous magnesium treatment in six patients with AMI and six patients with ischemic heart disease but without AMI. In both studies, magnesium therapy was associated with significant alterations in extracellular ion homeostasis. Serum concentrations of potassium decreased during the initial days of hospitalization in the patients treated with placebo, but increased slightly in the patients treated with magnesium infusions. These increments in the serum concentrations of magnesium and potassium correlated significantly. The increase in the serum concentration of potassium after magnesium infusions was due to a reduced renal potassium excretion level (from 71.3 to 49.4 mmol/24 h), indicating the existence of a divalent-monovalent cation exchange mechanism in the nephron. This hypothesis was supported by the observation that renal sodium excretion likewise decreased after magnesium infusions (from 83.2 to 59.2 mmol/24 h). Serum concentration of calcium decreased significantly after magnesium treatment (from 2.35 mmol/L on admission to 2.15 mmol/L after 24 hours in the hospital) in the AMI group, in contrast to the placebotreated patients, where no significant fluctuations in serum concentration of calcium were detected during the initial six days. This decrease in serum concentration of calcium was due to a marked increase in renal calcium excretion (from 3.43 mmol/24 h before to 6.59 mmol/24 h after magnesium infusion). A correlation between increments in serum magnesium concentration and decrements in serum calcium concentration was detected. No change in serum levels of parathyroid hormone was found before and after magnesium infusions. Both serum and urine levels of magnesium significantly increased after magnesium treatment to levels above the upper normal limits (serum magnesium concentration increased from 0.81 to 1.21 mmol/L, urine magnesium excretion levels from 3.57 to 16.57 mmol/24 h for both serum and urine changes). No changes were found in serum sodium concentration. It is concluded that the magnesium infusions significantly influence serum levels of potassium and calcium by changes in the renal threshold for tubular reabsorption. Theoretically, such changes in extracellular ion homeostasis are suppressant of arrhythmias, and it is suggested that this might be one mechanism for the antiarrhythmic effect of magnesium therapy in patients with AMI.

(Arch Intern Med 1988;148:1801-1805)