# Serum magnesium is associated with osteoporosis risk in postmenopausal women: a retrospective study and risk-prediction model

## TL;DR

Low magnesium levels in blood are linked to higher osteoporosis risk in postmenopausal women, and a new model helps predict this risk using multiple factors.

## Contribution

A new risk-prediction model for osteoporosis in postmenopausal women integrating serum magnesium and clinical markers.

## Key findings

- Lower serum magnesium levels were independently associated with increased odds of osteoporosis.
- A multi-parameter model incorporating BMI, age, magnesium, phosphorus, and bone markers showed good calibration for predicting osteoporosis risk.
- Osteoporotic fractures were linked to lower magnesium and phosphorus levels in postmenopausal women.

## Abstract

To investigate the association between serum magnesium (Mg) levels and osteoporosis (OP)/osteoporotic fractures in postmenopausal women, and to develop a multi-parameter model for predicting OP risk.

This retrospective study included 496 hospitalized postmenopausal women from January 1, 2021 to June 30, 2025. Patients were categorized into the OP group (T-score ≤ −2.5) and the non-OP group (T-score>−2.5). Clinical characteristics, bone mineral density (BMD), and laboratory indices were compared between groups. and additional comparisons were performed between patients with and without osteoporotic fractures. Receiver operating characteristic (ROC) curves were used to evaluate the discrimination of Mg and phosphorus (P) for OP. Binary logistic regression was performed with OP as the outcome. Variables selected by least absolute shrinkage and selection operator (LASSO) regression were used to construct a nomogram, which was evaluated using calibration curves and decision curve analysis (DCA).

Among 496 participants, 208 were classified as non-OP and 288 as OP. Osteoporotic fractures were identified in 31 patients, while 465 had no osteoporotic fractures. patients with OP were older, had lower body mass index (BMI), lower BMD, and lower serum Mg and P levels. ROC analysis showed modest discrimination for Mg (AUC 0.602, 95% CI 0.551–0.652), with an optimal cut-off of 0.845 mmol/L. After adjustment for potential confounders, Mg ≤ 0.845 mmol/L was independently associated with higher odds of OP (OR 1.940, 95% CI 1.312–2.868; P < 0.05). β-C-terminal telopeptide of type I collagen (β-CTx) (OR 2.229, 95% CI 1.034–4.809), age (OR 1.067, 95% CI 1.042–1.092), and BMI (OR 0.894, 95% CI 0.846–0.945) were also independently associated with OP (all P < 0.05). Patients with osteoporotic fractures were older and had lower BMD and lower Mg and P levels than those without fractures. A nomogram incorporating BMI, age, Mg, P, β-CTx, and procollagen type I N-terminal propeptide (P1NP) showed acceptable calibration and potential net clinical benefit.

Lower serum Mg levels were associated with OP and osteoporotic fractures in hospitalized postmenopausal women. Mg may support OP risk stratification, and a multi-parameter model integrating clinical and bone metabolism–related markers may further improve risk assessment.