Cardiovascular · MAG-CV-01 · v2
Several randomised trials that measured blood pressure directly found that oral magnesium lowered it, and no trial found magnesium raising it. But the reductions cluster in people who were low in magnesium and/or already hypertensive [1][14][2][17], while carefully done trials in healthier or normal-magnesium people found no effect [6][4][5]. One trial saw a fall only in participants who started with higher blood pressure [3]. The largest reported drop [1] is far bigger than anything else in the evidence, and two of the strongest positive trials come from the same research group [1][2]. So the plain, unqualified statement is only partly borne out: the effect looks real but modest and dependent on who is treated.
Multiple RCTs measuring blood pressure as a direct endpoint report significant reductions with oral magnesium in hypomagnesaemic and/or hypertensive populations: diabetic hypertensive adults with low serum magnesium [1], thiazide-treated hypertensive women [14], metabolically obese normal-weight hypomagnesaemic individuals [2], and normomagnesaemic metabolic-syndrome patients [17]. No trial in the corpus shows magnesium raising blood pressure. Counterbalancing this, well-conducted trials in normomagnesaemic or lower-risk groups are null: a magnesium glycinate ambulatory trial in untreated elevated BP [6], a 24-week arterial-stiffness trial [4], and a prediabetes trial [5]. An overweight normomagnesaemic trial found no overall difference but a signal confined to those with baseline SBP >= 140 and DBP >= 90 [3], and a CKD trial reached only borderline significance (P=0.053 systolic, P=0.071 diastolic) [7]. The contested point: The division is whether the effect generalises or is confined to selected populations: positives concentrate in hypomagnesaemic and/or hypertensive groups [1][14][2][17] while unselected or normomagnesaemic trials are null [6][4][5], and [3] finds a reduction only in the higher-baseline-BP subgroup. Magnitude is also contested, with [1]'s -20.4 mmHg systolic fall standing far outside the null ambulatory result of [6].
All 18 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Magnesium, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The claim is unqualified but the corpus effect is population-conditional: significant reductions appear in hypomagnesaemic/hypertensive groups [1][14][2], while well-designed trials in normomagnesaemic or lower-risk populations are null [6][4][5], and [3] explicitly finds an effect only in the higher-baseline-BP subgroup. The general wording overreaches.
Two of the strongest positives share a single research group (Guerrero-Romero and Rodríguez-Morán, [1] and [2]), raising single-lab and small-study effect concerns; [1]'s -20.4 mmHg SBP is an implausibly large outlier relative to the null ambulatory trial [6].
Where it remains uncertain. Well-conducted trials in normomagnesaemic or lower-risk populations are null [6][4][5], and [3] finds an effect only where baseline BP is high, so the unqualified claim overreaches.
Aleatory
Several estimates come from small samples (e.g. [17] n=24, [6] n=59), so play of chance around modest true effects is large.
Epistemic
Effects appear in hypomagnesaemic/hypertensive groups [1][14][2][17] but not normomagnesaemic/lower-risk groups [6][4][5]; whether the claim holds generally is unknown.
Inferential
[1]'s -20.4 mmHg systolic reduction is far larger than other trials and the null [6], making a pooled magnitude hard to infer.
The load-bearing papers from a corpus of 18, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Strongest single positive result, but in a specific hypomagnesaemic diabetic hypertensive population
“SBP (-20.4 versus -4.7 mm Hg, P=0.03) and DBP (-8.7 versus -1.2 mm Hg, P=0.02) showed significant decreases in the magnesium group compared to the placebo group.”
Panel auditInternal validity: medium · scope of inference: Diabetic hypertensive adults with hypomagnesaemia on captopril, not on diuretics.Key limitations: hypomagnesaemic subjects only · diabetic hypertensive population · small sample
Positive RCT in hypomagnesaemic metabolically obese individuals; supports population-conditional effect
“changes in the mean of systolic (-2.1 vs. 3.9% mmHg, p <0.05) and diastolic (-3.8 vs. 7.5% mmHg, p <0.05) blood pressures were significantly lower in the subjects who received MgCl2 compared with individuals in the control group.”
Panel auditInternal validity: medium · scope of inference: Metabolically obese, normal-weight individuals with hypomagnesemia.Key limitations: hypomagnesemic subjects only · small sample · percent-change reporting
Directly establishes the conditional pattern: null overall, positive only at higher baseline BP
“These results suggested that magnesium supplementation does not reduce BP in normo-magnesemic nondiabetic overweight people. However, it appears that magnesium supplementation may lower BP in healthy adults with higher BP.”
Panel auditInternal validity: medium · scope of inference: Normo-magnesemic nondiabetic overweight Korean adults.Key limitations: normo-magnesemic population · post-hoc subgroup for high-BP effect · 12-week duration
Directionally favourable but non-significant even in hypomagnesaemic CKD
Panel auditInternal validity: high · scope of inference: Hypomagnesemic pre-diabetic obese patients with mild-to-moderate CKD.Key limitations: BP secondary endpoint · borderline non-significant trend · 3-month duration
4 of 18 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
A genuine, directionally consistent but modest and population-conditional effect, offset by several credible nulls, places the unqualified claim in the Uncertain band.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Population moderators of the BP effect. Positives concentrate in hypomagnesaemic/hypertensive groups [1][14][2] while normomagnesaemic/unselected trials are null [6][4][5], so who benefits is unresolved. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A large blinded placebo-controlled RCT stratified by baseline serum magnesium and baseline BP, pre-specifying subgroup analyses.
Open question: Does oral magnesium lower blood pressure in normomagnesaemic, lower-risk adults, or only in magnesium-deficient and/or hypertensive people? Why it is still open: Direct-endpoint trials split cleanly: positive in hypomagnesaemic/hypertensive groups [1][14][2][17] but null in normomagnesaemic/unselected groups [6][4][5].
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 18 Sep 2026 · protocol v0.19