# Magnesium forms Rating: Proven — Replicated results in human randomized trials Type: Compound Updated: 2026-10-14 Canonical: https://heretic.bio/articles/magnesium-types.html > Magnesium is required for over 600 enzymatic reactions and deficiency is widespread, but the salt form changes everything: glycinate and citrate absorb well, oxide barely does, and threonate is the only form with randomized-trial evidence for brain effects. Sleep trials are positive but modest; in one 155-person study 81% saw no clinically meaningful benefit, and meta-analyses show no lipid effect beyond a small HDL rise. A proven nutrient with real but modest trial effects. > Magnesium is the most critical mineral most people are deficient in. Over 600 enzymatic reactions require it, including ATP production, DNA synthesis, and muscle/nerve function. However, the form matters enormously — bioavailability varies 5-10x between forms, and different forms serve different therapeutic purposes. ## The Deficiency Crisis ### Why Deficiency Is Universal * **Soil depletion:** Modern agriculture has stripped magnesium from soils * **Processed foods:** Refined grains, sugars, and processed foods are magnesium-free * **Medications:** PPIs, diuretics, antibiotics, and proton pump inhibitors deplete magnesium * **Stress:** Both physical and psychological stress increase magnesium excretion * **Alcohol:** Regular alcohol consumption significantly depletes magnesium * **Aging:** Renal function declines; absorption decreases ### Prevalence * Actual deficiency (including tissue deficiency) may be 70-80% of population * Standard serum blood tests miss 99% of deficiencies (only measures serum, not tissue magnesium) ## Magnesium Forms Compared ### Quick Reference Table Form | Bioavailability | Best For | Key Notes ---|---|---|--- **Glycinate** | High | Sleep, anxiety, relaxation | Most calming; glycine enhances sleep **Citrate** | High (~90%) | General supplementation, constipation | Good absorption, mild laxative effect **Malate** | High | Energy, fibromyalgia, muscle pain | Malic acid supports ATP production **Threonate** | Moderate-High | Cognitive function, brain health | Crosses blood-brain barrier **Bisglycinate** | Very High | Gentle on stomach, bioavailability | Chelated, no laxative effect **Taurate** | High | Cardiovascular, calming | Taurine + magnesium **Orotate** | Moderate | Heart health, performance | Supports DNA/RNA synthesis **Sulfate** | Variable | Topical, Epsom salts | Poor oral absorption **Oxide** | Low (~4%) | Laxative only | Avoid for supplementation **Chloride** | Moderate | Stomach acid support | Good for digestion issues ### Detailed Form Analysis #### Magnesium Glycinate * **Absorption:** Excellent (glycine is a small amino acid that enhances absorption) * **Mechanism:** Glycine is an inhibitory neurotransmitter; promotes calmness * **Best applications:** * Anxiety reduction * Muscle relaxation * Restless leg syndrome * **Note:** Most bioavailable oral form; least likely to cause diarrhea #### Magnesium Citrate * **Absorption:** Very good (~90% in studies) * **Mechanism:** Citrate is an organic acid, readily absorbed * **Best applications:** * General supplementation * Constipation relief (osmotic laxative effect) * Kidney stone prevention (citrate binds calcium) * **Note:** Good default choice; may cause loose stools in some #### Magnesium Malate * **Absorption:** High * **Mechanism:** Malic acid is a key player in Krebs cycle / ATP production * **Best applications:** * Chronic fatigue * Fibromyalgia * Muscle pain/soreness * Athletic performance #### Magnesium Threonate (L-Threonate) * **Absorption:** Moderate-high * **Mechanism:** Specifically designed to cross blood-brain barrier * **Best applications:** * Cognitive function * Memory improvement * Neuroprotection * Brain fog * **Note:** Newer form; more expensive but promising for brain health #### Magnesium Bisglycinate * **Absorption:** Very high; gentle on GI * **Mechanism:** Double-chelated form; glycine on both sides * **Best applications:** * Sensitive stomach * People who get diarrhea from other forms * General supplementation * **Note:** Best for long-term daily use #### Magnesium Taurate * **Absorption:** High * **Mechanism:** Taurine supports heart function; both are calming * **Best applications:** * Cardiovascular health * Blood pressure support * Arrhythmia prevention * Anxiety with heart palpitations * **Note:** Excellent for stress + heart support #### Magnesium Orotate * **Absorption:** Moderate * **Mechanism:** Orotic acid supports DNA/RNA synthesis * **Best applications:** * Athletic performance * Heart failure (historically used) * Recovery from exercise * **Note:** Less common; orotic acid has some cardiovascular research #### Magnesium Oxide — AVOID * **Absorption:** Very poor (~4%) * **Mechanism:** Inorganic salt; poorly absorbed * **Best applications:** Only as osmotic laxative (Milk of Magnesia) * **Note:** Cheap but nearly useless as supplement ## Testing Methods Method | What It Measures | Limitations ---|---|--- **Serum magnesium** | Blood levels only | Misses 99% of deficiencies (body maintains serum at expense of tissue) **RBC magnesium** | Intracellular levels | Better than serum; still not perfect **EXA (Exactech) test** | Tissue magnesium | Dried blood spot; measures tissue levels **Sublingual epithelial test** | Tissue levels | More accurate for tissue deficiency **Urinary magnesium** | Excretion levels | Useful for determining if supplementing is helping **Recommendation:** Don't rely on serum magnesium. Use RBC magnesium or EXA test for accurate assessment. ## Doses in studies ### Uses * **Form:** Glycinate or bisglycinate * **Timing:** Evening (supports sleep; magnesium helps activate parasympathetic nervous system) * **Duration:** Ongoing/long-term ### Doses in studies * **Form:** Glycinate * **Stack:** Can combine with glycine (3g), apigenin (50mg), or magnesium threonate * **Note:** Start lower; increase as needed ### Doses in studies * **Form:** Glycinate or taurate * **Stack:** Can combine with L-theanine (200-400mg) * **Note:** Higher doses may be needed during acute stress ### Athletic Performance / Muscle Recovery * **Form:** Malate or citrate * **Note:** Athletes need 10-20% more magnesium due to losses in sweat ### Fibromyalgia / Chronic Pain * **Form:** Malate * **Duration:** May take 4-6 weeks for full effect * **Evidence:** Small but promising studies (e.g., Abraham & Myers 1995) ### Cognitive Enhancement * **Form:** Threonate * **Note:** Take in morning; threonate may be stimulating for some ### Cardiovascular Support * **Form:** Taurate or citrate * **Stack:** Consider with CoQ10 (100-200 mg) for synergistic effect ## Recent RCT Evidence (2024–2026) — Form-Specific Updates The 2024–2026 literature has narrowed in on whether **specific forms** beat inexpensive generic salts. Several trials are large, registered, and placebo-controlled, allowing an honest read on effect sizes. ### Magnesium L-Threonate (MgT, Magtein®) — Sleep and Cognition * **Hausenblas et al. 2024 — Sleep RCT ([PMC11381753](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11381753/), PMID [39252819](https://pubmed.ncbi.nlm.nih.gov/39252819/), _Sleep Med X_):** 80 adults aged 35–55 with self-reported sleep problems, randomized to **1 g/day MgT (≈ 75 mg elemental Mg) vs placebo for 21 days**. ISI, LSEQ, and Restorative Sleep Questionnaire all improved significantly; Oura-ring-derived objective sleep was mixed (placebo declined on some metrics). Mood and daytime functioning improved. Funded by AIDP (Magtein manufacturer) — interpret accordingly. * **Source:** https://pmc.ncbi.nlm.nih.gov/articles/[PMC11381753](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11381753/) * **Lopresti & Smith 2026 — Cognitive + sleep RCT (_Front Nutr_ 12:1729164, [PMC12832366](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12832366/), PMID [41601871](https://pubmed.ncbi.nlm.nih.gov/41601871/)):** Adult RCT of Magtein® measuring NIH Toolbox cognitive battery. Improved composite cognitive ability in older adults with anxiety, poor sleep, and self-reported memory complaints — but the composite was author-derived, not a validated global cognition measure. Secondary sleep outcomes also favoured MgT. * **Source:** https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2025.1729164/full * **2026 Athletic-Performance RCT ([NCT07015047](https://clinicaltrials.gov/study/NCT07015047)):** Collegiate-athlete trial evaluating MgT on sleep architecture, autonomic recovery, and cognitive function (in progress). * **Honest read:** Effect sizes for MgT are real but modest. It is the only well-studied oral form with documented brain-magnesium elevations in animals (Slutsky 2010) and reproducible cognitive/sleep effects in human RCTs. It is **not** a panacea. ### Magnesium Bisglycinate — Sleep RCT * **Schuster et al. 2025 — Bisglycinate Sleep RCT (_Nature and Science of Sleep_ 17:2027–2040, [PMC12412596](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12412596/)):** 155 adults reporting poor sleep, randomized to **250 mg elemental Mg as bisglycinate vs placebo for 4 weeks**. ISI improved significantly vs placebo; effect was **small (d ≈ 0.2)** and ~81% of participants did not achieve a clinically meaningful benefit. Benefit was concentrated in low-dietary-Mg participants. * **Source:** https://pmc.ncbi.nlm.nih.gov/articles/[PMC12412596](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12412596/) * **Hausenblas re-analysis 2024 caveat:** Objective Oura data in the Hausenblas MgT trial showed the placebo group **worsened** over time rather than MgT dramatically improving — this is a common critique across recent Mg trials: small subjective gains, weak objective signals. ### Magnesium and Cognitive Health — Systematic Review (2024) * **Liu et al. 2024 — Meta-analysis of Mg and cognition (_Adv Nutr_ 15(8):100272):** Systematic review and meta-analysis covering RCTs and cohort studies on Mg forms, supplements, dietary intake, and biomarkers against cognitive outcomes. Reinforces an inverse association between higher Mg status and cognitive decline risk; effect sizes remain modest for oral supplements. * **Source:** https://www.sciencedirect.com/science/article/pii/S2161831324001066 ### Magnesium and Cardiovascular Disease — Updated Reviews (2024) * **Source:** https://www.mdpi.com/2072-6643/16/23/4223 * **Hariri et al. 2025 — Lipid-profile meta-analysis (_Nutr J_ 24:24):** Updated systematic review and dose-response meta-analysis of RCTs (27 trials; doses 20–548 mg/day; 4–24 weeks; populations: diabetic, MetS, PCOS, NAFLD, obese/overweight, healthy, hemodialysis, coronary disease, hypertension). Primary pooled estimates: * **TC** (21 studies, 24 datasets): MD +0.34 mg/dL (95% CI −1.75 to +2.43) — **no effect**. * **TG** (23 studies, 26 datasets): **no effect**. * **LDL-C** (20 studies, 24 datasets): **no effect**. * **HDL-C** (25 studies, 29 datasets): **MD +1.24 mg/dL (significant)** — small but consistent HDL-C increase. * **Key subgroup signals:** HDL-C benefit emerges only in studies ≥84 days; normomagnesemic participants show blunted response (renal homeostasis limits uptake); chloride and organic-acid salts drive most positive signals; inorganic oxide trials (Mortazavi hemodialysis, Razzaghi diabetic foot) account for some TC/LDL-C reductions. No dose-response on absolute mg/day. * **Mechanistic anchor:** ↑ LCAT activity (raises HDL-C); ↑ desaturase activity → ↑ EPA/DHA from linoleic acid (also HDL-C); ↓ HMG-CoA reductase; improved insulin signaling. * **Source:** https://link.springer.com/article/10.1186/s12937-025-01085-w (DOI [10.1186/s12937-025-01085-w](https://doi.org/10.1186/s12937-025-01085-w); PDF archived at Sources/Papers/doi_10.1186_s12937-025-01085-w.pdf) * **Honest read:** Magnesium supplementation is **not** an effective lipid-lowering intervention in the general adult population. The clinically meaningful signal is a small HDL-C increase after ≥12 weeks, particularly in deficient subjects on bioavailable salts. Do not market magnesium as a "cholesterol fix." ### Magnesium and Malate — Fibromyalgia * The malate/fibromyalgia evidence base remains weak. Boulis et al. 2021 literature review found mainly negative correlation studies and small intervention trials. A 2024 rat-model study (Xu et al., _J Nanobiotechnology_ 22:368) showed magnesium malate–modified calcium phosphate bone cement enhanced bone repair but is **non-supplemental** ; not applicable to oral malate dosing for fibromyalgia. No 2024–2026 placebo-controlled RCT has validated oral malate for fibromyalgia pain endpoints. The 1995 Abraham & Myers protocol remains the canonical clinical reference; benefit appears to require 4–6 weeks minimum and may be limited to a subgroup with measurable deficiency. ### Bottom Line on Form Choice Recent evidence does **not** overturn the older hierarchy (avoid oxide; glycinate/bisglycinate/citrate are cheap workhorses; threonate is the only well-evidenced brain-targeted form; malate's fibromyalgia edge is anecdote-strength). What 2024–2026 data add: 1. **Bisglycinate** sleep effects are real but small — do not present as a "cure" for insomnia. 2. **Threonate** has the best-funded cognitive/sleep RCT pipeline but modest effect sizes and manufacturer funding bias. 3. **Magnesium status (dietary + serum)** matters more than the specific salt for cardiovascular and mortality endpoints. ## Drug Interactions Medication | Interaction ---|--- **Proton pump inhibitors (PPIs)** | Severely reduce magnesium absorption; long-term use causes deficiency **Diuretics** | Increase magnesium excretion **Antibiotics (tetracyclines, quinolones)** | Magnesium binds to them; take 2-4 hours apart **Bisphosphonates** | Magnesium reduces absorption; take 2 hours apart **ACE inhibitors** | May increase magnesium levels; monitor **Digoxin** | Magnesium enhances effect; may need lower dose ## Sources 1. PMC11381753 (PMC11381753) 2. PubMed 39252819 (PubMed 39252819) 3. PMC12832366 (PMC12832366) 4. PubMed 41601871 (PubMed 41601871) 5. NCT07015047 (NCT07015047) 6. PMC12412596 (PMC12412596) 7. DOI 10.1186/s12937-025-01085-w (DOI 10.1186/s12937-025-01085-w) 8. PubMed 28588366 (PubMed 28588366) 9. PubMed 28655598 (PubMed 28655598) 10. PubMed 20152124 (PubMed 20152124) 11. PMC12535714 (PMC12535714)