Question explored with the scientific record
Is lithium good for humans
Lithium is a narrow therapeutic tool, not a general health tonic. It helps some people with bipolar disorder reduce suicide risk, but it damages kidneys and teeth in many who take it long-term.
The evidence here is almost entirely about lithium as a prescription drug for bipolar disorder, not as a trace mineral supplement. The strongest benefit shown is a meta-analysis from 2001 that found lower suicide rates during lithium treatment compared to periods without it [8]. But the studies in that analysis were small, mostly open-label (not blinded), and the comparison was often "on lithium" versus "off lithium" in the same patients, which confounds the effect of stopping treatment with the effect of the drug itself.
The harms are concrete and dose-dependent. A 2026 meta-analysis of nearly half a million people found that lithium users had an average eGFR (kidney function) about 11 mL/min lower than non-users [10]. The annual decline was small but real: about 0.13 mL/min per year faster [10]. A 2018 review reported that after 15 years on lithium, about 12% of patients develop overt nephrogenic diabetes insipidus (inability to concentrate urine) and 21-55% reach chronic kidney disease stage 3 [9]. End-stage kidney disease is rare but real, occurring in about 0.2-0.7% of new ESKD cases, with an average latency of 20 years [9]. A case report documented severe dental decay with loss of calcium, phosphorus, and magnesium from tooth dentin [6].
The efficacy data is surprisingly weak for a drug that has been used for decades. The LiTMUS trial (2013) found that adding moderate-dose lithium to optimized personalized treatment did not significantly improve clinical outcomes compared to optimized treatment alone [2]. Lithium monotherapy produced a good outcome in only 31% of patients in one study, while adding quetiapine, lamotrigine, or valproic acid raised that to 72% [5]. A 2020 systematic review found that the only consistent predictor of poor lithium outcome was a higher lifetime number of hospitalizations [7]. Rapid cycling also predicted poor response [7].
| Outcome | Lithium group | Comparison group | Source |
|---|---|---|---|
| Mean eGFR difference | -11.1 mL/min/1.73m² | Non-lithium controls | [10] |
| Annual eGFR decline | +0.13 mL/min/yr faster | Non-lithium controls | [10] |
| CKD stage 3 prevalence | 21-55% of long-term users | General population ~5-10% | [9] |
| NDI after 15 years | ~12% overt, ~54% impaired concentrating | - | [9] |
| Good outcome on monotherapy | 31% | 72% with combination | [5] |
| Suicide risk on vs off lithium | Lower on lithium | Higher off lithium | [8] |
The genetic studies here suggest that lithium response is partly heritable. Variants in the FKBP5 stress-response gene and the IMPA gene (involved in inositol metabolism) were associated with better or worse response [1, 4]. The Ser796 allele of the BCR gene was more common in non-responders [3]. This means lithium works well for some people and poorly for others, and we are only beginning to understand why.
For a person without bipolar disorder, there is no evidence in this retrieval that lithium provides any benefit. The animal LD50 data shows that lithium chloride is toxic at doses not far above therapeutic levels, with oral LD50 in rats ranging from 12.4 to 19.8 mg/kg depending on age [11]. The therapeutic index is narrow.
My call: lithium is a useful drug for a subset of bipolar patients who respond to it, with a real but manageable suicide risk reduction, but it carries substantial kidney and dental toxicity that accumulates over years. For anyone without bipolar disorder, there is no demonstrated benefit and clear potential for harm. Confidence: moderate for the bipolar population, high for the general population.
Sources used 11
-
Genomic Structure and Chromosomal Localization of a Humanmyo-Inositol Monophosphatase Gene (IMPA)
This study investigates the genomic structure and chromosomal localization of the human myo-inositol monophosphatase gene (IMPA), revealing its organization and a polymorphism in the 3-untranslated region that may influence lithium treatment response in bipolar disorder.
DOI: 10.1006/geno.1997.4862 -
Lithium Treatment Moderate-Dose Use Study (LiTMUS) for Bipolar Disorder: A Randomized Comparative Effectiveness Trial of Optimized Personalized Treatment With and Without Lithium
The Lithium Treatment Moderate-Dose Use Study (LiTMUS) found that adding moderate doses of lithium to optimized personalized treatment for bipolar disorder did not significantly improve clinical outcomes compared to optimized treatment alone, although it resulted in less exposur…
DOI: 10.1176/appi.ajp.2012.12060751 -
A possible association between missense polymorphism of the breakpoint cluster region gene and lithium prophylaxis in bipolar disorder
This study investigates the association between the Asn796Ser single-nucleotide polymorphism in the breakpoint cluster region gene and the efficacy of lithium prophylaxis in bipolar disorder patients, finding that the Ser796 allele is more prevalent in non-responders to lithium …
DOI: 10.1016/j.pnpbp.2007.08.010 -
Genes involved in stress response influence lithium efficacy in bipolar patients
This study investigates the influence of genetic variants in stress response-related genes on the efficacy of long-term lithium treatment in bipolar patients, revealing significant associations with specific polymorphisms and the impact of stressful life events.
DOI: 10.1111/bdi.12639 -
Prevalence and Characteristics of Mental Disorders Among Hospitalized Breast Cancer Patients in the United States
This study investigates the long-term treatment outcomes of 204 ambulatory patients diagnosed with bipolar disorder on lithium carbonate, revealing that supplementation with quetiapine, lamotrigine, or valproic acid significantly improves treatment outcomes, while lithium monoth…
DOI: 10.1016/j.jval.2016.08.535 -
Dentin decalcification during lithium treatment: case report
This case report investigates the severe dental decay and changes in tooth structure associated with lithium treatment in a 30-year-old woman with bipolar disorder, revealing significant alterations in dentin mineral composition and structure.
DOI: 10.1111/scd.12000 -
Socio-demographic and clinical predictors of outcome to long-term treatment with lithium in bipolar disorders: a systematic review of the contemporary literature and recommendations from the ISBD/IGSLI Task Force on treatment with lithium
A comprehensive qualitative systematic review (ISBD/IGSLI Task Force) of socio-demographic and clinical predictors of long-term lithium outcome in adults with bipolar disorder, finding few robust predictors and noting rapid cycling and higher prior hospitalizations as consistent…
DOI: 10.1186/s40345-020-00203-3 -
Lower suicide risk with long‐term lithium treatment in major affective illness: a meta‐analysis
This meta-analysis compares suicide rates in patients with major affective disorders during long-term lithium treatment versus periods without such treatment, finding a significant reduction in suicide risk associated with lithium maintenance therapy.
DOI: 10.1034/j.1600-0447.2001.00464.x -
Lithium and nephrotoxicity: a literature review of approaches to clinical management and risk stratification
A narrative literature review synthesizing evidence on lithium-associated nephrotoxicity, its risks, clinical management, and risk stratification to guide decisions for long-term lithium therapy in bipolar disorder.
DOI: 10.1186/s12882-018-1101-4 -
Lithium nephrotoxicity: a systematic review and meta-analysis of lithium versus non-lithium control studies in patients with affective disorders
Systematic review and meta-analysis finds lithium use in affective disorders is associated with lower eGFR and higher serum creatinine compared with non-lithium controls, but CKD risk evidence is highly heterogeneous and may reflect confounding rather than true lithium-induced n…
DOI: 10.1177/20451253261419633 -
Effect of Age and Route of Administration on LD50 of Lithium Chloride in the Rat *
This study investigates the lethal dose (LD50) of lithium chloride in male inbred rats of varying ages and administration routes, finding significant differences in toxicity based on age and route of administration.
DOI: 10.1111/j.1600-0773.1980.tb01571.x