Benefits
Brain magnesium elevation: shown in rats, never measured in people
The brain-magnesium claim rests on animal work. In the 2010 Neuron study by Slutsky and colleagues, rats given magnesium L-threonate had higher cerebrospinal fluid magnesium and learned better than rats given other magnesium salts. No human study has measured brain or cerebrospinal fluid magnesium after taking Magtein®, so the blood-brain barrier claim has not been tested in people. The word only is also not settled even in animals: a 2026 rat comparison, co-authored by a company that sells a competing magnesium salt, reported that magnesium acetyl-taurate raised brain and cerebrospinal fluid magnesium more than magnesium L-threonate did. Brain magnesium genuinely matters for synaptic plasticity and NMDA receptor function, but that is general biology rather than a demonstrated effect of this product in a person.
Cognitive performance and working memory: small gains in one 6 week trial
One 6 week randomised trial of Magtein® at 2 g/day in 100 adults aged 18 to 45 found a small improvement over placebo on the NIH Toolbox total cognition composite (p = 0.043), on the list sorting working memory subtest (p = 0.033) and on an aim and reaction time task (p = 0.031). The NIH fluid composite (p = 0.277), the crystallised composite (p = 0.111) and Raven's matrices (p = 0.953) showed no difference from placebo. The reported 7.5 year reduction in cognitive age is a derived, exploratory figure: it converts a 2.24 point group difference on a composite change score using an assumed decline of 0.3 points a year. The trial was funded by Threotech, which owns Magtein®. The often quoted 9 year reversal of cognitive age is not from Magtein®: it comes from a 2016 trial of MMFS-01, sold as ClariMem® by Neurocentria, in 44 adults aged 50 to 70 with memory and concentration complaints, where estimated brain age fell 9.4 years on treatment and 0.8 years on placebo.
Sleep: one questionnaire improved, tracked sleep did not
In the 6 week Magtein® trial the sleep result was mixed. PROMIS sleep-related impairment improved against placebo (p = 0.043), but PROMIS sleep disturbance did not (p = 0.316), the Restorative Sleep Questionnaire did not (p = 0.439), and the Oura ring sleep measures showed no group difference at all. A group difference in sleep disturbance did appear in a subgroup with more severe sleep problems (p = 0.031), which is an exploratory result. Resting heart rate fell (p = 0.030) and heart rate variability rose (p = 0.036); those are autonomic readings, not sleep outcomes. A separate 21 day trial in 80 adults aged 35 to 55 taking 1 g/day magnesium L-threonate reported better deep and REM sleep scores and better daytime functioning than placebo, with the supplement group holding steady while placebo declined. AIDP, the Magtein® distributor, commissioned and funded that trial, owns the patents and paid three of its authors, the trial was registered only after it had finished, and a 2025 corrigendum was needed to disclose those interests.
Stress and anxiety: not demonstrated in these trials
The stress case for Magtein® is general magnesium biology, not a Magtein® result. In the 6 week Magtein® trial, self-reported general wellbeing did not differ from placebo (p = 0.436); the only stress-related signals were a lower resting heart rate and higher heart rate variability, which are autonomic readings rather than anxiety or mood scores. In the 2016 magnesium L-threonate trial in older adults, both groups improved so much on the sleep and anxiety scales that the authors reported the effect on anxiety could not be determined. None of the cited trials measured cortisol. At 145 mg of elemental magnesium a day this product also supplies less than half the adult reference intake for magnesium, which is 310 to 420 mg depending on age and sex.
Mechanism of action
Proposed GLUT transporter route into the brain (rodent evidence)
This mechanism was worked out in rodents and cell models, not in people. L-threonate is proposed to enter the brain via glucose transporters (GLUT1 and GLUT3) carrying magnesium with it, and rats given magnesium L-threonate had higher cerebrospinal fluid magnesium than rats given other magnesium salts. In those animals the raised brain magnesium reduced NMDA receptor over-activation, supported long-term potentiation and increased synapse density. Whether any of this happens in a person taking 2 g/day has not been measured, and a 2026 rat comparison, co-authored by a company selling a competing magnesium salt, reported that magnesium acetyl-taurate reached the brain better than magnesium L-threonate, so the route is not unique to this salt.
Clinical trials
Randomised, double-blind, placebo-controlled parallel-arm trial of Magtein® at 2 g/day for 6 weeks in 100 adults aged 18 to 45 recruited for self-reported dissatisfaction with their sleep. Outcomes: NIH Toolbox cognition battery, Raven's Progressive Matrices 2, PROMIS sleep questionnaires, a reaction time task, and Oura ring heart rate and sleep data. Lopresti and Smith, Frontiers in Nutrition 2025.
100 adults aged 18 to 45 with self-reported poor sleep; they were not selected for cognitive complaints.
Magtein® beat placebo on the NIH Toolbox total cognition composite (p = 0.043), the list sorting working memory subtest (p = 0.033) and an aim task measuring reaction time and hand-eye coordination (p = 0.031). The 7.5 year reduction in cognitive age is a derived, exploratory conversion of a 2.24 point group difference on a composite change score. The NIH fluid composite, the crystallised composite and Raven's matrices were unchanged. Of the sleep measures, only PROMIS sleep-related impairment improved (p = 0.043); sleep disturbance, restorative sleep and every Oura ring sleep reading were no different from placebo. Resting heart rate fell and heart rate variability rose. Funded by Threotech Inc., the owner of Magtein®, which was also involved in designing the study.
12 week randomised, double-blind, placebo-controlled trial in adults aged 50 to 70 with self-reported memory and concentration complaints: 51 were randomised and 44 completed. The product was MMFS-01, sold as ClariMem® by Neurocentria; it is magnesium L-threonate but it is not Magtein®. Liu G, Weinger JG, Lu ZL, Xue F, Sadeghpour S, Journal of Alzheimer's Disease 2016;49(4):971-90.
Adults aged 50 to 70 with self-reported memory and concentration complaints plus anxiety and sleep difficulty; 51 randomised, 44 analysed (23 on treatment, 21 on placebo).
Overall cognitive ability improved against placebo (p = 0.003, Cohen's d = 0.91) in the 44 people who completed 12 weeks. The widely quoted 9 year cognitive age improvement is a derived figure: estimated brain age in the treated group fell 9.4 years by week 12 while the placebo group fell 0.8 years, and the paper reports no between-group p-value for brain age. Effects on sleep and anxiety could not be determined because the placebo group improved as much. Funded by Neurocentria Inc. and designed jointly with the sponsor, and not independently replicated.