Does magnesium prevent kidney stones?

Not on its own, as far as the trials go. A systematic review of 28 trials for the American College of Physicians found moderate evidence that thiazides, citrates and allopurinol cut repeat stones in adults with several past calcium stones, citrates with a risk ratio of 0.25, and it does not name magnesium among them. The one double-blind trial of magnesium alone, magnesium hydroxide at 650 or 1,300 mg a day, was not shown to beat placebo. What does work is a combination pill, potassium-magnesium citrate, which cut new stones from 63.6% to 12.9% in 64 people over three years. Nobody has separated the magnesium from the citrate in it.

Unsettled. The magnesium question is open because every positive trial bundled it with something else. The citrate trials pooled by Cochrane included one potassium-magnesium citrate trial, and a 2025 meta-analysis pooled magnesium with thiazides and citrate. Neither gives magnesium its own number in the abstract, and we could not read the full texts.

Share of people who formed a new stone, percent

Raw percentages from one double-blind trial, so no interval is drawn. The pill held potassium, magnesium and citrate together, and the trial cannot say which of the three did the work. Source: card ev-mgks-06 below.

What the trials found

Stones counted, magnesium in a mixture

A Cochrane review of 7 trials in 477 adults with calcium stones found that citrate salts lowered new stone formation, with a risk ratio of 0.26 and a confidence interval from 0.10 to 0.68, across 7 studies and 324 participants. One of those trials, in 64 people, used potassium-magnesium citrate. The reviewers rated reporting quality moderate to poor, and the effect belongs to citrate as a class. The abstract gives no absolute risk, so we cannot turn 0.26 into a number of people.

That single potassium-magnesium citrate trial is the one with absolute numbers. Recurrent calcium oxalate stone formers took a daily dose of 42 mEq potassium, 21 mEq magnesium and 63 mEq citrate, or placebo, for up to three years. New stones appeared in 12.9% on the pill and 63.6% on placebo, a relative risk of treatment failure of 0.16, with an interval from 0.05 to 0.46.

A 2025 meta-analysis for Japanese guidelines pooled thiazides, citrate preparations, uric acid production inhibitors and magnesium preparations. As a group they cut recurrence of calcium oxalate stones with a risk ratio of 0.47, interval 0.35 to 0.63, on moderate strength of evidence. The abstract has no separate figure for magnesium.

Stones counted, magnesium alone

Myth. Plain magnesium tablets have not been shown to stop stones coming back. In a double-blind trial in people with recurrent calcium oxalate stones, chlorthalidone did significantly better than both placebo and magnesium hydroxide, and magnesium hydroxide was not shown to beat placebo. The abstract does not say how many people took part. An older meta-analysis of 14 trials found that only thiazide diuretics were shown to significantly reduce recurrence, and its authors said the trials were too varied to judge the other drugs properly.

Urine chemistry, not stones

Several trials measured urine instead of counting stones. In 90 calcium stone formers with high urine oxalate, 120 mg of magnesium citrate three times a day with meals for 8 weeks lowered 24-hour urine oxalate and calcium oxalate supersaturation against placebo. Magnesium oxide moved in the same direction without reaching significance. Lower urine oxalate is a step toward fewer stones, and it is not the same thing as fewer stones.

Diet did better than the supplement in a 12-week trial of 164 stone formers with high urine oxalate. A low-oxalate diet cut urine oxalate by a median 31.1%, against 16.0% for vitamin B6 with 400 mg of magnesium oxide, and only 99 of the 164 finished. A pilot in 8 people suggests magnesium taken with meals lowers urine oxalate more than magnesium taken fasting, over seven days and without a statistical test in the abstract. In 20 adults on five weeks of strict bed rest, potassium-magnesium citrate lowered calcium oxalate saturation in urine against placebo, which tells us about immobilised people more than about everyday life.

Kidney stone risk, relative to the comparison group

Meta-analyses of trialsSingle trialObservational data

Below 1 means fewer stones. None of the trial rows isolates magnesium. The cohort row is an association in men, and the NHANES row is odds of ever having had a stone, from a single survey. Sources: cards ev-mgks-01, 03, 05, 06, 11 and 14 below.

Magnesium from food

Among 45,619 US men followed for 14 years, those eating the most magnesium had a risk ratio of 0.71 for a first kidney stone against those eating the least, with an interval from 0.56 to 0.89. Among 96,245 younger US women followed for 8 years, magnesium intake was not linked to stone risk once other dietary factors were taken into account. In 51,336 adults of the EPIC-Oxford cohort, high magnesium intake went with fewer hospital-recorded stones, and the abstract gives no size for it. In US NHANES data on 19,271 adults, people eating 379 mg a day or more had odds of 0.70 of ever having had a stone, against those under 205 mg. Each extra 100 mg a day gave odds of 0.92, with an interval from 0.84 to 1.01 that crosses no effect. That survey asks about past stones, so the stone may have come before the diet. All four are observational, and people eating more magnesium also eat more fruit, fiber and potassium.

Who should be careful

Not for everyone. The citrate salts that do prevent stones are hard to stay on. In the Cochrane review, people taking them were 4.45 times as likely to drop out because of side effects, interval 1.28 to 15.50, across 4 trials and 271 people, mostly with stomach upset and once with a rash. The 2025 meta-analysis found the same pattern for drug therapy as a group, a risk ratio of 2.51 for dropping out over adverse reactions.

The NIH Office of Dietary Supplements sets the adult upper limit for magnesium from supplements and medicines at 350 mg a day, and between 65 and 350 mg for children depending on age. Trial doses were given as the compound, such as 650 mg of magnesium hydroxide, so how much elemental magnesium a stone regimen adds up to is a question to check with whoever prescribes it. The NIH also warns that magnesium toxicity becomes more likely with impaired kidney function or kidney failure, because the kidneys can no longer clear the excess. People who form stones and also have reduced kidney function should not start magnesium on their own.

What expert bodies say

The systematic review behind the American College of Physicians guidance lists thiazides, citrates and allopurinol as the options with moderate evidence, and does not name magnesium among them. Italian expert guidelines on diet for stone prevention advise older adults with stones to cut sodium and eat more potassium, magnesium and citrate, to lower urinary risk factors and also to protect bone and blood pressure. That advice is about food, and parts of it rest on expert opinion. The NIH magnesium fact sheet we searched has nothing on kidney stones as such, only the kidney cautions above.

How we searched

Searched: a local copy of PubMed on 7 October 2026, for magnesium, magnesium citrate or potassium-magnesium citrate with kidney stones, nephrolithiasis, urolithiasis or calcium oxalate (310 hits, top 40 screened), then magnesium with stones and diet or intake (53 hits) and magnesium hydroxide or oxide with stone recurrence or prevention (8 hits). We also searched the NIH supplement fact sheets, the US trial registry, the web for recent reviews, and Retraction Watch for every source cited. None was retracted.

Included: a Cochrane review, a review of 28 trials for the American College of Physicians, meta-analyses from 1999 and 2025, six randomized trials, three cohorts, one national survey, the NIH fact sheet and an Italian diet guideline.

Excluded: mineral water trials, where magnesium comes with calcium and bicarbonate and cannot be separated. Multi-ingredient supplements. Studies that used magnesium only as a urine test, or were about body weight, children's biomarkers or bowel disease. A 2026 narrative review we found on the web, because it was not a systematic synthesis.

What we read: abstracts. The men's cohort was not in our local copy, and its magnesium sentence was checked word for word through two separate requests to Europe PMC.

What we could not get: full texts of the Cochrane review, the American College of Physicians review and the 1999 and 2025 meta-analyses, which were paywalled. That is where any separate number for magnesium would sit. An older magnesium hydroxide trial from 1980 was not in our corpus.

What would change this answer

A placebo-controlled trial of magnesium citrate alone that counts stones rather than measuring urine, ideally in people with low urine magnesium or high urine oxalate. We found none, finished or running.
A trial that splits potassium-magnesium citrate into its parts, so the magnesium share is known.
Any randomized trial of magnesium from food with stones as the outcome. We found none.
The magnesium subgroup figures inside the four paywalled reviews, if they exist.
A cohort in women that agrees with the men's result. The one we have does not.

The rest of the nutrient, in one place. Magnesium: widely under-eaten, and widely over-promised → What it does, how much you need, who runs short, and what too much does, with the evidence level shown on every line.

The full guide

Magnesium: widely under-eaten, and widely over-promised

The same question for other foods (kidney stones)

Sources

ev-mgks-01 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis of RCTs · n = 477
New stone formation was significantly lower with citrate therapy compared to control (7 studies, 324 participants: RR 0.26, 95% CI 0.10 to 0.68).
Who: adults with calcium-containing kidney stones, mostly oxalate, treated at least 6 months; 1 of 7 trials (64 people) used potassium-magnesium citrate
Effect: new stone formation RR 0.26 (95% CI 0.10 to 0.68), 7 studies, 324 participants; reporting quality moderate to poor
Certainty: The effect belongs to citrate as a class; the magnesium contribution is not separated. Full text not in PMC.
Cochrane Database Syst Rev, 2015 · checked 2026-10-07 · we read the abstract
ev-mgks-02 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis of RCTs · n = 271
There were significantly more dropouts due to adverse events with citrate therapy compared to control (4 studies, 271 participants: RR 4.45, 95% CI 1.28 to 15.50).
Who: adults with calcium-containing kidney stones in RCTs of citrate salts, including potassium-magnesium citrate
Effect: dropouts due to adverse events RR 4.45 (95% CI 1.28 to 15.50), 4 studies, 271 participants
Certainty: Main side effects: gastrointestinal upset, one rash.
Cochrane Database Syst Rev, 2015 · checked 2026-10-07 · we read the abstract
ev-mgks-03 · Meta-analysis or systematic review · systematic review of RCTs with pooled estimates (AHRQ) · n = 28 trials
In patients with multiple past calcium stones, most of whom were receiving increased fluid intake, moderate-strength evidence showed that thiazides (RR, 0.52 [CI, 0.39 to 0.69]), citrates (RR, 0.25 [CI, 0.14 to 0.44]), and allopurinol (RR, 0.59 [CI, 0.42 to 0.84]) each further reduced composite stone recurrence risk compared with placebo or control
Who: adults with multiple past calcium stones, most also increasing fluid intake, in 28 RCTs
Effect: composite stone recurrence: thiazides RR 0.52 (0.39 to 0.69), citrates RR 0.25 (0.14 to 0.44), allopurinol RR 0.59 (0.42 to 0.84); magnesium not named in the quoted finding
Certainty: Magnesium is not named among the effective ones in the review conclusions; there are no details about magnesium RCTs in the abstract, the full text is not in PMC.
Ann Intern Med, 2013 · checked 2026-10-07 · we read the abstract
ev-mgks-04 · Meta-analysis or systematic review · meta-analysis of RCTs · n = 14 trials
Although only thiazide diuretics among the drug therapies were shown to significantly reduce stone recurrence, variability in study design and study population precluded adequate analysis of other drug therapies such as alkali citrate.
Who: calcium oxalate stone formers in 14 RCTs with 20 treatment arms (thiazides, allopurinol, phosphate, magnesium, alkali citrate)
Effect: overall drug therapy benefit P = 0.04, driven by thiazides (P = 0.02); magnesium not shown to significantly reduce recurrence
Certainty: Old meta-analysis; there is no effect size for magnesium in the abstract. The authors write that heterogeneity did not allow analysis of other preparations.
J Endourol, 1999 · checked 2026-10-07 · we read the abstract
ev-mgks-05 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs · n = —
Pharmacotherapies were shown to significantly reduce stone recurrence (risk ratio 0.47, 95% confidence interval 0.35-0.63).
Who: patients with calcium oxalate stones and abnormal urine chemistry in RCTs of thiazides, citrate preparations, uric acid production inhibitors and magnesium preparations
Effect: stone recurrence RR 0.47 (95% CI 0.35 to 0.63); adverse reactions leading to dropout RR 2.51 (1.09 to 5.75); strength of evidence moderate
Certainty: The effect is pooled across all preparations; there is no separate estimate for magnesium in the abstract. Full text not in PMC.
Int J Urol, 2025 · checked 2026-10-07 · we read the abstract
ev-mgks-06 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = 64
When compared with placebo, the relative risk of treatment failure for potassium-magnesium citrate was 0.16 (95% confidence interval 0.05 to 0.46).
Who: patients with recurrent calcium oxalate kidney stones given potassium-magnesium citrate (42 mEq potassium, 21 mEq magnesium, 63 mEq citrate daily) or placebo for up to 3 years
Effect: new calculi 12.9% vs 63.6%; relative risk of treatment failure 0.16 (95% CI 0.05 to 0.46)
Certainty: Combination: the contribution of magnesium is not separated from potassium and citrate. 21 mEq magnesium ≈ 255 mg per day.
J Urol, 1997 · checked 2026-10-07 · we read the abstract
ev-mgks-07 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = —
When the treatments were compared chlorthalidone was significantly better than the placebo or magnesium hydroxide (p less than 0.01).
Who: patients with recurrent calcium oxalate kidney stones randomized to chlorthalidone 25 or 50 mg, magnesium hydroxide 650 or 1,300 mg, or placebo
Effect: calculi fewer than predicted: placebo 56.1%, magnesium hydroxide 73.9% and 62.3%, chlorthalidone 90.1%; only chlorthalidone better than placebo (p < 0.01)
Certainty: The number of participants is not given in the abstract. The only RCT of magnesium as monotherapy with a "stones" endpoint.
J Urol, 1988 · checked 2026-10-07 · we read the abstract
ev-mgks-08 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = 90
After the intervention, both MgO and MgCit supplements decreased 24-U oxalate excretion (-8.13±16.45 in the MgO group and -16.99±18.02 in the MgCit group) and CaOx SS compared to the placebo, with the effects of MgCit reaching statistical significance (p=0.011 and p=0.010, respectively).
Who: calcium stone formers with idiopathic hyperoxaluria given 120 mg magnesium oxide, 120 mg magnesium citrate or placebo three times a day with meals; 76 analyzed
Effect: 24-h urine oxalate change -16.99 (MgCit) and -8.13 (MgO); MgCit vs placebo p = 0.011 for oxalate, p = 0.010 for supersaturation
Certainty: Surrogate endpoint (urine), not stones; 8 weeks.
Magnes Res, 2024 · checked 2026-10-07 · we read the abstract
ev-mgks-09 · Randomized controlled trial(s) · randomized controlled trial · n = 164
Significant differences were noted in median percent reduction in Ox24 values (-31.1% vs -16.0% vs -23.9%, p = 0.007) in the D, S, and DS groups, respectively.
Who: patients with calcium oxalate stones and idiopathic hyperoxaluria randomized to low-oxalate diet, 25 mg B6 plus 400 mg magnesium oxide, or both, for 12 weeks; 99 completed
Effect: median reduction in 24-h urine oxalate -31.1% (diet) vs -16.0% (supplement) vs -23.9% (both), p = 0.007
Certainty: High dropout (only 99 of 164 completed); surrogate endpoint.
J Endourol, 2021 · checked 2026-10-07 · we read the abstract
ev-mgks-10 · Randomized controlled trial(s) · randomized controlled pilot trial · n = 8
Patients with a prior history of calcium oxalate stone formation and isolated hyperoxaluria who took magnesium supplementation with meals were found to have a more substantial improvement in urinary parameters on 24-hour urinalysis compared to those who took magnesium supplementation while fasting.
Who: patients with prior calcium oxalate stones and isolated hyperoxaluria on a controlled diet, magnesium with meals vs fasting for 7 days
Effect: urine oxalate median -17.8 mg/day with meals vs average -8.5 mg/day fasting
Certainty: Eight people, seven days, no statistical comparison in the abstract.
Clin Nephrol, 2025 · checked 2026-10-07 · we read the abstract
ev-mgks-11 · Observational data · prospective cohort · n = 45,619
Other dietary factors showed the following multivariate RR among men in the highest quintile of intake compared with those in the lowest: magnesium, 0.71 (95% CI, 0.56 to 0.89; P = 0.01 for trend)
Who: 45,619 men without a history of kidney stones (Health Professionals Follow-up Study)
Effect: highest vs lowest quintile of magnesium intake RR 0.71 (95% CI 0.56 to 0.89)
Certainty: Outside the corpus snapshot. Observational data: magnesium travels together with potassium, fiber, fruit.
Taylor EN, Stampfer MJ, Curhan GC. J Am Soc Nephrol. 2004 · checked 2026-10-07 · we read the abstract
ev-mgks-12 · Observational data · prospective cohort · n = 96,245
The intakes of sodium, potassium, and magnesium were not independently associated with risk after adjusting for other dietary factors.
Who: female nurses aged 27 to 44 without kidney stones (Nurses' Health Study II)
Effect: magnesium not independently associated with incident symptomatic stones
Certainty: Discrepancy with the men of HPFS; the authors suggest dietary factors differ by age and sex.
Arch Intern Med, 2004 · checked 2026-10-07 · we read the abstract
ev-mgks-13 · Observational data · prospective cohort · n = 51,336
High intakes of fresh fruit, fibre from wholegrain cereals and magnesium were also associated with a lower risk of kidney stone formation.
Who: participants in the Oxford arm of EPIC, UK, with a wide range of diets including vegetarians
Effect: inverse association for magnesium intake; 303 new stone episodes; size for magnesium not in abstract
Certainty: HR for magnesium is not given in the abstract.
Eur J Epidemiol, 2014 · checked 2026-10-07 · we read the abstract
ev-mgks-14 · Observational data · cross-sectional survey analysis · n = 19,271
In the multivariable-adjusted regression analysis, those in the highest quartile of DMI compared to the lowest quartile (≥379 mg vs. <205 mg) had significantly reduced odds of prevalent KS (OR=0.70, 95% CI: 0.52-0.93, p=0.01).
Who: US adults in NHANES 2011-2018, 1,878 self-reported stone formers
Effect: highest vs lowest quartile OR 0.70 (95% CI 0.52 to 0.93); per 100 mg OR 0.92 (0.84 to 1.01)
Certainty: Cross-sectional: the stones could have been there before the diet change; causality not demonstrated.
R I Med J (2013), 2023 · checked 2026-10-07 · we read the abstract
ev-mgks-15 · Randomized controlled trial(s) · randomized double-blind placebo-controlled trial · n = 20
Despite this increase subjects treated with potassium-magnesium citrate demonstrated significant decreases in the relative saturation of calcium oxalate and in the concentration of undissociated uric acid compared to placebo.
Who: 20 normocalciuric adults on strict bed rest given potassium-magnesium citrate (42 mEq potassium, 21 mEq magnesium, 63 mEq citrate daily) or placebo
Effect: significant decrease in calcium oxalate relative saturation vs placebo
Certainty: A weightlessness and immobilization model, surrogate endpoint.
J Urol, 2007 · checked 2026-10-07 · we read the abstract
ev-mgks-16 · Position of an expert body · government fact sheet · n = —
The Tolerable Upper Intake Level for supplemental magnesium is 350 mg for adults, and it ranges from 65 to 350 mg for children and adolescents, depending on age.
Who: adults; children 65 to 350 mg depending on age
Effect: UL for supplemental magnesium 350 mg/day (adults)
Certainty: The limit applies to magnesium from supplements and medications, not from food. Doses in the RCTs are given as the compound, not as elemental magnesium; no comparison with the UL in the source.
NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
ev-mgks-17 · Position of an expert body · government fact sheet · n = —
The risk of magnesium toxicity increases with impaired renal function or kidney failure because the ability to remove excess magnesium is reduced or lost
Who: people with impaired renal function or kidney failure
Effect: higher risk of magnesium toxicity
Certainty: Relevant for people with stones and reduced kidney function.
NIH Office of Dietary Supplements, Magnesium Fact Sheet for Health Professionals · checked 2026-10-07 · we read the section
ev-mgks-18 · Position of an expert body · evidence-based guideline with expert opinion · n = —
A restriction of sodium intake in association with a higher intake of potassium, magnesium and citrate is advisable in order to reduce urinary risk factors for stone formation but also to prevent the loss of bone mass and the incidence of hypertension
Who: older patients with kidney stones
Effect: expert recommendation; no effect size
Certainty: Recommendation about food, not supplements; some statements are expert opinion.
Arch Ital Urol Androl, 2015 · checked 2026-10-07 · we read the abstract

How this page was made. Evidence was extracted from primary sources into cards, every number was re-checked against the source, and the text was written from those cards. Bioma Learn has no human medical reviewer at this time, and we say so rather than invent one. Methodology. This is information, not medical advice.