Objective: To assess the alterations in total serum magnesium (tsMg) and ionized serum magnesium (Mg(2+)) and their association with prognosis in critically ill patients.
Design and setting: Prospective, cohort study in the intensive care unit (ICU) of a university teaching hospital.
Patients: Adult patients admitted to the ICU without previous factors influencing magnesium homeostasis were included during a 6-month period.
Measurements and results: One hundred forty four patients were included. Mean age was 60.6+/-15.4 years; mean APACHE II score was 12.6+/-6.9. Blood samples were collected in the first 24 h after ICU admission and again on the second, third, and last days of stay in the ICU. At ICU admission 52.5% had total hypomagnesemia and 13.5% total hypermagnesemia; with respect to the Mg(2+) 9.7% showed ionized hypomagnesemia and 23.6% ionized hypermagnesemia. Patients who developed ionized hypermagnesemia had higher mortality than patients without ionized hypermagnesemia development (P=0.04). A moderate correlation between tsMg and Mg(2+) concentrations was found; however, a number of patients with total hypomagnesemia (69-85% during the study) had ionized normomagnesemia. The measure of agreement between tsMg and Mg(2+) levels was poor.
Conclusions: Magnesium alterations are frequently found in critically ill patients. The usually determined tsMg levels are not a reflection of Mg(2+) levels. Development of ionized hypermagnesemia is associated with prognosis.