Interpretation of serum magnesium concentrations in the presence of the systemic inflammatory response: a large patient cohort study.

The present study aimed to quantify the relationship between the systemic inflammatory response (SIR), measured by C-reactive protein (CRP) and albumin, and serum magnesium concentrations. Retrospective interrogation of laboratory computer databases at 4 centres between 1st October 2018 to 31st March 2019 provided results from patients in whom serum CRP and albumin had been measured together with serum magnesium concentrations. Analyte results were categorized into groups according to CRP and albumin. Patients (n = 9,624) were grouped according to serum magnesium concentrations: within normal range (0.75-1.10 mmol/L) (n = 6,313) (65%), low (< 0.75 mmol/L) (n = 3,152) (33%), or high (> 1.10) (n = 159) (2%). In those patients with a serum magnesium concentration below the normal range and a normal CRP, median serum magnesium concentrations were 0.70 (0.65-0.72) mmol/L, 0.68 (0.60-0.71) mmol/L, and 0.67 (0.59-0.72) mmol/L in patients with serum albumin concentrations ≥ 35 g/L, 25- <35, and < 25 g/L respectively. This represents a 4.2% change in serum magnesium concentrations among patients within the low serum magnesium group between mild inflammation and severe inflammation categories (p = 0.086). In those patients with a normal serum albumin concentration who were within the low serum magnesium category, serum magnesium concentrations were 0.70 (0.65-0.72) mmol/L, 0.69 (0.63-0.72) mmol/L, and 0.68 (0.63-0.71) mmol/L for CRP concentrations < 10 mg/dL, 11-79 mg/dL and > 80 mg/dL respectively. This represents a 2.9% change in serum magnesium concentrations for patients with normal serum albumin concentration between mild - severe inflammation (p = 0.109). Across a large, real-world clinical population, serum magnesium concentrations demonstrated limited variation in relation to the magnitude of the systemic inflammatory response and serum albumin concentration. Routine albumin adjustment of serum magnesium measurements may offer limited additional value as any observed differences in the present study were small and unlikely to be meaningful in clinical practice.