Admission hypomagnesemia--impact on mortality or morbidity in critically ill patients

No previous study exists to evaluate admission serum magnesium level as a predictor of morbidity or mortality. The aim of this study was to define the prevalence of admission hypomagnesemia in critically ill patients and to evaluate its relationship with organ dysfunction, length of stay, and mortality.

A retrospective study was done on 100 patients > or =16 years old, admitted to the medical-surgical intensive care unit (ICU) at the University Hospital over 2 years period. Observations were made on admission total serum magnesium level, a variety of lab tests related to magnesium, need for ventilator, duration of mechanical ventilation, hospital/ICU lengths of stay, and general patient demographics.

The serum magnesium level (normal value, 1.3-2.1 mEq/L) was measured at admission. At admission, 51% of patients had hypomagnesemia, 49% had normal magnesium levels. There was significant difference in mortality rate (55% vs 35%), the length of hospital (15.29 +/- 0.66 vs 12.81 +/- 0.91), or ICU (9.16 +/- 0.53 vs 5.71 +/- 0.55) stay between these two groups of patients (p < 0.05 for all). Hypomagnesemic patients more frequently had total hypocalcemia, hypokalemia, and hyponatremia. A total of 51 patients developed hypomagnesemia during their ICU stay; these patients had higher Acute Physiology And Chronic Health Evaluation II (APACHE II) (14.16 +/- 1.03 vs 10.80 +/- 0.94) and Sequential Organ Failure Assessment (SOFA; 10.89 +/- 0.90 vs 7.58 +/- 5.01) scores at admission (p < 0.01 for both), a higher maximum SOFA score during their ICU stay (14.75 +/- 0.73 vs 8.08 +/- 0.52, p < 0.01), a more need to ventilator (58.6% vs 41.4%, p < 0.05), and longer duration of mechanical ventilation (7.2 vs 4.7 day, p < 0.01) than the other patients. The ROC curve of SOFA score in the hypomagnesemia yields significantly better results than APACHE II. An increase of 5 units in the APACHE II or SOFA measured on admission increase relative probability of hypomagnesemia by a factor of 0.12 and 0.16 respectively.

Development of hypomagnesemia during an ICU stay is associated with guarded prognosis. Monitoring of serum magnesium levels may have prognostic, and perhaps therapeutic, implications.