The Importance of Magnesium in Clinical Healthcare

Two physicians reviewed the medical literature on magnesium for practicing clinicians, covering how status is measured, how much people need, and how strong the evidence is across many conditions. Much of that evidence comes from patients with a diagnosis, from single studies or from observational data. They note that magnesium toxicity is rare, with loose stools usually the first sign of too much.

Why this study was done

The authors, two University of Alberta physicians, argue that nutrition gets little room in medical training, so doctors may miss mineral shortfalls. Magnesium is a cofactor in more than 300 enzyme systems. The paper pulls the clinical literature on it into one reference for practicing clinicians: how status is measured, how much people need, which conditions it has been linked to, and how strong the evidence is for each.

How it was run

This is a narrative review, described by the authors as a traditional integrated review, not a systematic review. They searched MEDLINE and PubMed, drew on books and conference proceedings, then followed references in those sources. Search terms combined magnesium with conditions such as osteoporosis, asthma, diabetes, eclampsia and cardiovascular disease. For each condition the authors assigned a level of evidence letter using the Evidence-Based Medicine Toolkit scale. No screening counts, risk of bias scoring or pooled statistics are reported, because the studies are not combined.

What they found

A 70 kg adult holds about 25 grams of magnesium: 53 percent in bone, 27 percent in muscle, 19 percent in soft tissue, under 1 percent in serum. Between 56 and 68 percent of Americans are estimated not to reach the recommended daily allowance. The biological half-life is about 1,000 hours, or 42 days. The Institute of Medicine upper tolerable limit for supplementation without side effects is 350 mg per day. People with kidney impairment face higher risk of harm.

Findings the authors graded as stronger evidence include:

• Migraine: oral magnesium reduced frequency, duration and intensity by 41 percent against 15.8 percent for placebo in one trial.
• Preeclampsia: magnesium sulfate cut the risk of recurrent seizures by 52 percent and halved the rate of eclampsia.
• Diabetes: in a 20-year study of 4,497 young adults, the highest quartile of magnesium intake had 47 percent lower diabetes incidence.
• Cardiovascular: the highest intake quartile had 77 percent lower sudden cardiac death, and 500 to 1,000 mg per day was linked to blood pressure falls of up to 5.6/2.8 mm Hg.

Several findings are mixed or negative. A review of magnesium for muscle cramps showed a trend towards benefit that did not reach significance, and a placebo-controlled trial at 360 mg per day found no benefit. A 2006 Cochrane review found not enough evidence to recommend magnesium for high blood pressure. The Women's Health Initiative found more wrist fractures in those with the highest magnesium levels.

Limitations

This is not a systematic review. There is no protocol, no screening numbers, no quality appraisal and no pooling, so the level of evidence letters reflect the authors' own judgment. Many claims rest on a single cited study, and some sections lean on animal or laboratory work rather than human trials. The authors note that limited primary literature exists on magnesium insufficiency and health outcomes. Percentages from observational cohorts show association, not cause. The authors declare no conflicts of interest and no industry funding is stated. The article is open access under a Creative Commons Attribution License.

What it means in practice

Treat this as a map of the magnesium literature written for clinicians, not as evidence in its own right. It shows which conditions have trial support and where the evidence is thin. It does not weigh study quality or resolve disagreement between trials, and the effect sizes quoted come from single studies chosen by the authors rather than a pooled analysis. Much of the clinical material concerns patients with an existing diagnosis, such as preeclampsia, diabetes or migraine, and those findings do not automatically extend to healthy people.