[Effect of serum magnesium on mortality in intensive care unit patients with heart failure: a retrospective analysis...

To analyze the effect of serum magnesium on 90-day all-cause mortality in heart failure patients admitted to the intensive care unit (ICU).

A retrospective cohort study was conducted using data from the Medical Information Mart for Intensive Care- III (MIMIC- III) database. Heart failure patients admitted to the ICU of Beth Israel Deaconess Medical Center between 2001 and 2012 were selected. Demographic characteristics, comorbidities, initial vital signs, and laboratory data were extracted. The outcome measure was 90-day all-cause mortality after ICU admission. Patients were divided into non-renal failure and renal failure groups based on the presence of renal failure. Univariate Logistic regression was used to analyze the relationship between potential confounding factors and the 90-day all-cause mortality. Multivariate Logistic regression was used to analyze the independent effect of serum magnesium on the 90-day all-cause mortality. A threshold effect analysis identified an inflection point for blood urea nitrogen (BUN) at 530 mg/L, and an interaction analysis examined the effect of BUN on the relationship between serum magnesium and the 90-day all-cause mortality.

A total of 1 162 ICU patients with heart failure were included, among which 695 were in the non-renal failure group and 467 in the renal failure group; 317 patients died (27.3%). Univariate analysis showed that serum magnesium was significantly associated with mortality in the renal failure group [odds ratio (OR) = 1.994, 95% confidence interval (95%CI) was 1.223-3.252, P < 0.05], but not in the non-renal failure group (OR = 1.098, 95%CI was 0.700-1.722, P > 0.05). Multivariate analysis showed that after adjusting for all selected confounding factors, serum magnesium was not significantly associated with mortality in the renal failure group (OR = 1.053, 95%CI was 0.519-2.132, P > 0.05), while in the non-renal failure group, serum magnesium showed a protective effect (OR = 0.460, 95%CI was 0.239-0.885, P < 0.05). The interaction analysis between BUN and serum magnesium showed that after adjusting for all selected confounding factors, there was an interaction between BUN and serum magnesium in the non-renal failure group (interaction P values were < 0.05), with a significant association between serum magnesium and mortality in the high BUN subgroup (OR = 0.082, 95%CI was 0.016-0.406, P < 0.05), but no significant association in the low BUN subgroup (OR = 0.946, 95%CI was 0.466-1.918, P > 0.05); there was no interaction between BUN and serum magnesium in the renal failure group (interaction P values were > 0.05), and there was no significant association between serum magnesium and mortality in both the high BUN subgroup and the low BUN subgroup (all P > 0.05).

Renal function and BUN levels significantly modify the relationship between serum magnesium and 90-day all-cause mortality in heart failure patients admitted to the ICU. The mortality significantly decreases with the increase of serum magnesium in non-renal failure patients with high BUN, while the mortality shows no significant change with serum magnesium concentration in non-renal failure with low BUN. In renal failure patients, no significant change in mortality occurs with varying serum magnesium concentration, regardless of BUN levels.