Lithium and Alzheimer’s Disease: What New Research Suggests

Patients have started asking me about lithium ever since a recent Nature paper pulled together a striking story. In human brain tissue, lithium levels were lower in the prefrontal cortex of people with mild cognitive impairment and Alzheimer’s disease, and lithium appeared to accumulate inside amyloid plaques. In that study, lithium was the only metal the investigators found to be significantly reduced already at the mild cognitive impairment stage, while serum lithium levels were not significantly different from controls.  It suggests the phenomenon may not simply reflect low dietary intake. Instead, it raises the possibility that lithium handling within the brain changes early in the disease process. Cause or effect? 

The human data make the animal work more interesting than the average mouse study. The researchers examined postmortem tissue from people with no cognitive impairment, mild cognitive impairment, and Alzheimer’s disease. They found lower lithium levels in the prefrontal cortex in MCI and Alzheimer’s disease but not in the cerebellum, a region relatively spared in Alzheimer’s pathology. They also replicated the finding in an independent cohort. When they examined plaque-containing tissue more closely, lithium appeared highly concentrated within amyloid plaques, while lithium levels in the surrounding plaque-free brain fraction were reduced. Lower lithium in that non-plaque fraction correlated with worse episodic memory, semantic memory, and global cognition. In other words, the finding was not simply that lithium levels were lower overall. It suggested lithium might be getting sequestered away from where it is biologically needed.

The mouse experiments then explored what happens when brain lithium is depleted. The investigators created a lithium-deficient diet that reduced cortical lithium by roughly 50 percent without otherwise altering nutrition. In Alzheimer’s mouse models, that depletion worsened amyloid deposition, increased phosphorylated tau, activated inflammatory microglia, and accelerated cognitive decline. In aging wild-type mice, lithium depletion increased Aβ42 ratios and impaired memory in ways that resembled accelerated brain aging. The signal appeared across pathology, behavior, and molecular biology.

The investigators also examined how lithium affects cellular signaling. Single-nucleus RNA sequencing showed lithium deficiency altered gene expression across multiple brain cell types, including neurons, oligodendrocytes, astrocytes, and microglia. Transcriptional signatures overlapped with Alzheimer’s-associated inflammatory states and included enrichment for known Alzheimer’s risk genes such as APOE and TREM2, and other genes linked with tau-phosphorylation. (Parts of underlying Alzheimer’s pathology.

The treatment experiments are what captured the most attention. The researchers used lithium orotate rather than the more familiar lithium carbonate. Their rationale was mechanistic: lithium carbonate appeared to bind more avidly to amyloid fibrils and oligomers, potentially becoming sequestered within plaques. Lithium orotate, a more weakly dissociated salt, appeared less prone to plaque binding and more effective at restoring lithium in the plaque-free brain fraction. In mouse models, low-dose lithium orotate reduced plaque burden, reduced phosphorylated tau, improved synaptic markers, dampened neuroinflammation, and improved performance on memory tasks.

The dose used in the study is where much of the online discussion becomes misleading. The mice were not exposed to anything resembling psychiatric or OTC supplement lithium treatment levels. The lithium concentration in drinking water was 4.3 μEq/L, which corresponds to approximately ~30 micrograms of elemental lithium per liter of water. When translated into a rough human-equivalent exposure, that corresponds to roughly 30–50 micrograms of elemental lithium per day, orders of magnitude below conventional psychiatric lithium therapy or OTC supplements.

For context, prescription lithium carbonate used for bipolar disorder typically provides about 50 to 200 milligrams of elemental lithium per day.  At the extremely low exposures of micrograms of elemental lithium, the pharmacologic toxicity profile associated with therapeutic lithium,effects on kidney function, thyroid function, and electrolyte balance, would not be expected to occur.

Interestingly, exposures in this microgram range are not unusual in the real world. Lithium occurs naturally in groundwater as it leaches from rock formations. Concentrations in municipal drinking water vary widely but commonly fall between about 1 and 200 micrograms per liter. With typical daily water intake of one to two liters, that corresponds to roughly 5 to 100 micrograms of lithium intake per day in some regions. Epidemiologic studies have suggested that populations living in areas with moderately higher lithium concentrations in drinking water sometimes show lower suicide rates and, in some analyses, lower dementia incidence. These ecological studies cannot establish causation, but they have helped generate the hypothesis that lithium may function as a trace neuroactive element.

Importantly, lithium in drinking water exists as inorganic salts derived from geological sources. Lithium orotate does not occur naturally in water; it is a synthetic compound. The mouse study used the orotate salt because of its biochemical behavior in amyloid-rich tissue, not because it mimics natural water exposure.

This raises the practical question patients are now asking: if someone is at elevated risk for Alzheimer’s disease, or already has early dementia, is there meaningful risk in consuming lithium in the tens-of-micrograms range?

Based on current toxicity data, there is little indication that exposures in the 30–50 microgram per day range would pose significant physiological risk in otherwise healthy adults. Those exposures overlap with lithium intake levels that occur naturally in some drinking waters and remain thousands of times lower than prescription lithium doses. At such tiny levels, routine laboratory monitoring is unlikely to reveal measurable serum lithium concentrations, and clinically significant kidney or thyroid effects would be biologically implausible. That said, absence of evidence is not the same as proof of safety, and I would monitor anyone who takes it upon themselves to take lithium supplement whether in microgram dose or mg dose.

The larger issue is not safety but uncertainty about benefit. What happened in mouse models may or may not translate to the human brain. The human portion of the Nature study showed an association between lower brain lithium and Alzheimer’s pathology, but it did not prove that restoring lithium levels will prevent or treat disease. We still need those studies to better understand if this translates, as well as proper dosing.

For patients already facing progressive cognitive decline, the risk-benefit calculation sometimes looks different than it does for healthy individuals seeking longevity interventions. When exposures remain in the tens-of-micrograms range, similar to what some people consume through water alone, the theoretical risk appears extremely small. Whether that exposure meaningfully alters Alzheimer’s biology in humans remains unknown. But the practical problem:

At present, there is no pharmaceutical-grade or supplement grade lithium-orotate  approved by regulatory agencies such as the FDA or EMA or even OTC lithium orotate supplement in the microgram dose that would mimic the microdose exposure in the mouse experiment from the Nature study. Lithium orotate is sold in some countries as a dietary supplement, typically containing 1-10 milligrams of elemental lithium per capsule, which is about 100+ times higher than the exposure range discussed in the mouse study. At that dosage one does need to start considering possible lithium monitoring and toxicity and side effects as there have been published reports of toxicity from excess supplements. What if more isn’t better? If the mouse data is true it is possible that microgram dosing is a sweet spot and mg dosing is not helpful at best or harmful at worst.

Ultimately, what the new research provides is biological plausibility. The most important next step would be carefully designed human trials testing physiological-range (microgram doses found in water) lithium OROTATE (rather than the other salts used in all other studies to date like carbonate) exposure in biomarker-confirmed early Alzheimer’s disease. Until then, any use of lithium for neuroprotection remains an experiment conducted in the gray space between intriguing science and definitive evidence. I’m surprised a CPG brand has not yet created a microgram dose lithium orotate to match the dosing in this study.

Alongside next-generation drugs targeting multiple biological pathways, there is growing interest in multimodal prevention and reversal strategies for Alzheimer’s disease. Pharmaceutical approaches being studied include oral medications, like Alzeon, an oral oligomer inhibitor designed to prevent formation of the most neurotoxic amyloid species, particularly in APOE4 carriers; tau-lowering therapies such as BIIB080 that reduce tau production using antisense technology; and microglial-targeted immunotherapies such as AL002 that activate TREM2 signaling to enhance clearance of pathological proteins and regulate neuroinflammation. 

Metabolic approaches are also promising but not yet conclusive, including GLP-1 receptor agonists that may improve neuronal insulin signaling and reduce neurodegeneration. If anyone with high risk or actual cognitive decline has an indication to be on a GLP-1 such as diabetes, pre diabetes, insulin resistance, OSA (itself a risk for cognitive decline), or obesity, I certainly reach for one of these medications early with my patients.

Lifestyle-based intervention trials are also very interesting and where I focus a lot of my clinical practice. A randomized controlled trial led by Dean Ornish tested an intensive multimodal program combining a whole-food plant-based diet, structured exercise, stress reduction, sleep optimization, social engagement, and targeted supplements including omega-3 fish oil, curcumin, and Lion’s Mane mushroom, alongside aggressive cardiometabolic risk reduction; participants in the intervention group showed significant improvements in cognition and favorable changes in plasma Alzheimer biomarkers compared with usual care.  These findings align with a growing systems-biology view of Alzheimer’s disease as a network disorder involving amyloid, tau, neuroinflammation, vascular dysfunction, and metabolic impairment, suggesting that future treatment may combine targeted drugs with comprehensive lifestyle interventions to modify disease progression. Lithium orotate may end up being part of this multi-modal approach and I am eager for more studies to enlighten us. Certainly one can test their water levels for lithium minerals for the time being. If you can find a glass bottle microplastic free spring water then enjoy the natural occurring lithium in there!

This information is provided for educational and informational purposes only and is not intended as medical advice. It should not be used to diagnose, treat, cure, or prevent any disease. Clinical decisions should be made in consultation with a qualified healthcare professional who can consider an individual’s specific medical history and circumstances. The therapies and research discussed may be investigational and not yet approved for routine clinical use.

References

Nature. Alzheimer’s Disease Research Collection.
https://www.nature.com/immersive/alzheimers-disease/index.html

Kessing LV, et al. Association between lithium in drinking water and the incidence of dementia.
https://pubmed.ncbi.nlm.nih.gov/28832877/

Mauer S, et al. Lithium levels and neuroprotection in Alzheimer’s disease models.
https://pmc.ncbi.nlm.nih.gov/articles/PMC5710473/

Kessing LV, et al. Lithium in drinking water and suicide rates. JAMA Psychiatry.
https://jamanetwork.com/journals/jamapsychiatry/fullarticle/2649275

Fajardo VA, et al. Trace lithium and health outcomes: a systematic review. PLOS Medicine.
https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1003941

Clinical research on lithium and neurodegenerative disease risk.
https://pubmed.ncbi.nlm.nih.gov/35512382/

Nunes MA, et al. Lithium and Alzheimer’s disease: mechanistic insights and clinical implications.
https://pubmed.ncbi.nlm.nih.gov/21525519/

About the Author:

Dr. Julia Loewenthal is an internist and geriatrician at Lutanen Health specializing in integrative primary care and healthy aging. Board certified in Internal and Geriatric Medicine, she is an Assistant Professor of Medicine at Harvard Medical School and a faculty fellow with the Osher Collaborative for Integrative Medicine. Her work combines clinical depth, compassion, and mind-body expertise to support lifelong wellness.

Dr. Julia Loewenthal, MD

March 16, 2026