A Novel U-Shaped Association Between Serum Magnesium on Admission and 28-Day In-hospital All-Cause Mortality in the Pediatric Intensive Care Unit.

Objective: Our purpose is to evaluate whether serum magnesium when entering the ICU is related to 28-day in-hospital all-cause mortality in the pediatric ICU. Methods: We used the PIC database to conduct a retrospective analysis to investigate the first-time serum magnesium levels of 10,033 critically ill children admitted to the pediatric ICU, and analyzed association between serum magnesium and all-cause mortality. Smoothing spline plots, subgroup analysis and segmented multivariate logistic regression analysis were conducted to estimate the relative risk between serum magnesium and all-cause mortality. The shape of the curve was used to describe the relationship between magnesium and 28-day in-hospital mortality. Results: There is a non-linear relationship between serum magnesium and 28-day in-hospital all-cause mortality. The U-type relationship between serum magnesium and all-cause mortality was observed. The optimal range of serum magnesium with the lowest risk of mortality was 0.74-0.93 mmol/L. As the serum magnesium level reaches the turning point (0.74 mmol/L), the risk of death decreases by 60% for every 0.1 mmol/L increase in serum magnesium; when the serum magnesium level exceeds 0.93, an increase of 0.1 mmol/L increases the risk of death by 38 %. Conclusion: Serum magnesium has a U-shaped relationship with 28-day in-hospital all-cause mortality. Both low and high serum magnesium can increase the risk of death. The best serum magnesium range when the risk of death is the lowest is 0.74-0.93 mmol/L.