Serum magnesium and major adverse cardiovascular events in patients with diabetic foot ulcers: a retrospective cohort study.

BACKGROUND AND AIMS: Patients with diabetic foot ulcers (DFU) have a high cardiovascular risk. We evaluated the association of baseline total serum magnesium (Mg) with major adverse cardiovascular events (MACE) without presuming a causal or therapeutic effect. METHODS: This single-center retrospective cohort included 1,176 patients with active DFU admitted from 2015 through 2022. The endpoint was the first MACE, defined as cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke. Cox models estimated hazard ratios (HRs) per 0.1 mmol/L and per standard deviation (SD) of Mg and by data-defined quartile groups. The extended model additionally adjusted for HbA1c, eGFR, albumin, log-transformed C-reactive protein, HDL-C, and LDL-C. Restricted cubic splines used four knots. Competing-risk, renal-function, complete-case, prior-cardiovascular-disease, time-specific, and propensity-score-matched analyses were performed. RESULTS: During a median follow-up of 3.6 years, 304 MACE events occurred, including 216 cardiovascular deaths and 88 combined nonfatal myocardial infarctions or strokes. In the extended model, each 0.1 mmol/L higher Mg was associated with lower MACE hazard (HR 0.87, 95% CI 0.80-0.95; p=0.002); the HR per SD was 0.83 (95% CI 0.74-0.94; p=0.002). The spline model supported an overall (p<0.001) and nonlinear (p<0.001) association. An exploratory two-change-point model selected 0.83 and 0.99 mmol/L, but profile-likelihood regions were broad (0.75-0.85 and 0.95-1.10 mmol/L). Results were directionally consistent in competing-risk and other sensitivity analyses. In 324 matched pairs, the high (>0.80 mmol/L) versus low (≤0.80 mmol/L) contrast was associated with lower MACE hazard (pair-stratified HR 0.60, 95% CI 0.39-0.92; p=0.020), whereas the continuous estimate was less precise (HR 0.85, 95% CI 0.71-1.01; p=0.070). CONCLUSIONS: Baseline serum Mg was independently associated with MACE risk in patients with DFU, with an exploratory nonlinear pattern. The data-derived change points are not therapeutic boundaries and require external prospective validation. These observational findings do not establish benefit from Mg supplementation.