Hospital-acquired dysmagnesemia and mortality in critically ill patients: data from MIMIC-III database.

This study aimed to determine the relationship between hospital-acquired dysmagnesemia and in-hospital mortality in critically ill patients. A retrospective cohort study was conducted on critically ill patients who had normal serum magnesium levels on admission. Data were extracted from the Multiparameter Intelligent Monitoring in Intensive Care III database. The normal range of serum magnesium was 1.6-2.6 mg/dL. In-hospital serum magnesium levels were categorized based on the occurrence of hospital-acquired hypomagnesemia and hypermagnesemia. Hospital-acquired hypomagnesemia and hypermagnesemia in the same patient were defined as a patient with the lowest level of serum magnesium of <1.6 mg/dL and the highest level of serum magnesium of >2.6 mg/dL, respectively. The in-hospital outcomes were collected. The findings revealed that 27.2% of patients developed hospital-acquired dysmagnesemia. In-hospital mortalities were 8.8% in patients with persistently normal serum magnesium levels, 12.2% in patients with hospital-acquired hypomagnesemia only, 18.4% in patients with hospital-acquired hypermagnesemia only, and 20.6% in patients with both hospital-acquired hypomagnesemia and hypermagnesemia. Compared to patients with persistently normal serum magnesium in hospital, those with hospital-acquired hypermagnesemia only [odds ratio (OR) = 1.346, P < 0.001] and those with both hospital-acquired hypomagnesemia and hypermagnesemia (OR = 1.333, P = 0.001) were significantly associated with higher in-hospital mortality. Hospital-acquired dysmagnesemia was common among critically ill patients. Hospital-acquired dysmagnesemia, especially hospital-acquired hypermagnesemia, was significantly associated with increased in-hospital mortality in critically ill patients.