**3. Epidemiological studies of lithium in Alzheimer's disease**

The results of population studies of an association between lithium and dementia were reviewed by Donix and Bauer [26]. Data from large cohort and most case–control studies suggest an association between lithium treatment and dementia risk reduction or reduced dementia severity.

In their publication in 2005, Dunn et al. [27] reported that among 19,328 participants selected from a General Practice Research Database, more subjects with dementia were treated with Li compared to control subjects without dementia. However, mood disorders are the most frequent indication for Li treatment and also belong to the strongest risk factors for dementia, and as this study did not control for compliance/optimal treatment, it may have simply detected the increased risk of dementia in mood disorders. Terao et al. [28] investigated clinical records of 1423 outpatients at a university psychiatric department and compared patient treated with lithium to age- and the gender-matched control group who had never been prescribed with lithium. Patients who had previously received lithium and/or were currently on lithium had significantly better Mini Mental State Examination (MMSE) scores than the control patients. Nunes et al. [29] studied the occurrence of AD in 66 elderly BD patients assessed during euthymia, receiving long-term lithium therapy, and in 48 age-matched patients who were not recently taking lithium. The percentage of patients with dementia was 19% in the first group and 7% in the second group. The diagnosis of the AD was made in three patients (5%) receiving lithium and in 16 patients (33%) who were not taking lithium, which suggests that lithium treatment may reduce the prevalence of AD in patients with bipolar disorder. Angst et al. [30] studied subjects with bipolar disorder (N = 220) and major depressive disorder (N = 186) followed from 1965 to 1985, receiving long-term treatment with lithium, clozapine, or antidepressants. In the whole group, the prevalence of dementia showed a significant association with age. However, when an analysis of the 88 patients with dementia was performed, the association with age was lost, and there was a trend to an inverse correlation between lithium administration and the severity of dementia.

amyloid-beta pathology. Both these effects are thought to be obtained by lithium inhibition of the GSK-3 [16, 17]. About the GSK-3 inhibition by lithium, the effect of this ion on autophagy regulation should also be indicated, the signaling pathway of which is associated with the mammalian

The PI pathway plays a role in signal transduction mechanisms connected with the action of multiple neurotransmitters. Lithium significantly influences this pathway which resulted in the inositol-depletion hypothesis of lithium action, as an essential therapeutic mechanism in mood disorders. Lithium inhibits the inositol monophosphatase (IMPase) and many other phases of the PI pathway [12]. The effect of lithium on the PI pathway is also connected with

Protein kinase C (PKC) is an enzyme associated with the PI pathway and plays a role in the action of many neurotransmitters and other cellular mechanisms. It has been found that lithium inhibits the activity of PKC that may contribute to its regulation of intracellular signaling

Bcl-2 is a protein playing a significant role in cellular resilience and plasticity, among others, by inhibiting apoptosis. Experimental studies demonstrated an increase of Bcl-2 in the brain by lithium treatment. Enhancing by lithium the expression of Bcl-2-associated athanogene (bag-1) augments the antiapoptotic effect, by mitigating glucocorticoid receptor nuclear translocation [12]. Morris and Berk [19] suggested some additional mechanisms of lithium action which may be important in the treatment of AD, such as the effects on mitochondrial function, calcium homeostasis, inflammation, microglial activation, glutamate excitotoxicity, and oxidative

Lithium produces a significant increase in mitochondrial performance in human brain tissue, the main factors of such effect being the inhibition of GSK-3 and activation of mTOR [20]. This cation also desensitizes brain mitochondria to the damaging effects of calcium influx [21] and increases mitochondrial levels of Bcl-2 [22]. As peripheral and neuro-inflammation, together with the chronic activation of microglia, constitutes an important element in the development of the AD, there is evidence that lithium can ameliorate various aspects involved in the pro-inflammatory response. These include the generation of tumor necrosis factor-alpha and interleukin-1 beta by microglia, and this effect is obtained via the inhibition of GSK-3 [23]. Lithium also exhibits a protective effect against the development of glutamate neurotoxicity, which is a consequence of chronic microglial activation, and this effect is due to the upregulation of BDNF [24]. In clinical conditions, lithium administration causes a decrease in markers

stress. Most of these processes are connected with the mechanisms described above.

of oxidative stress such as catalase and superoxide dismutase [25].

**3. Epidemiological studies of lithium in Alzheimer's disease**

The results of population studies of an association between lithium and dementia were reviewed by Donix and Bauer [26]. Data from large cohort and most case–control studies suggest an association between lithium treatment and dementia risk reduction or reduced dementia severity.

enhancing autophagy by the mTOR-independent pathway [18].

target of rapamycin (mTOR) [18].

82 Alzheimer's Disease - The 21st Century Challenge

and increasing neuroplasticity [12].

Two papers coming from the University of Copenhagen employed the Danish nationwide register of lithium prescriptions. In the first one, a comparison was made for the diagnosis of dementia or AD between 16,238 persons who had purchased lithium at least once during inpatient or outpatient treatment and 1,487,177 persons from the general population who had never bought lithium. Those who had bought lithium at least once had the 1.5-fold higher rate of dementia than the persons not taking lithium. However, those who continued treatment with lithium had the rate of dementia decreased to the same level as that for the general population. Such a decrease was exclusive to lithium because persons receiving anticonvulsant drugs had the risk of dementia increased with the duration of treatment [31]. The second study followed up 4856 patients which received a diagnosis of a manic or mixed episode or bipolar disorder at their first psychiatric contact for the period of 1995–2005 (103.6/10000 person-years). The percentages of patients receiving given drug were as follows: lithium 50.4%, anticonvulsants 36.7%, antidepressants 88.1%, and antipsychotics 80.3%. During the follow-up period, 216 patients were diagnosed with dementia. It was found that a reduced rate of dementia in BD patients was connected with long-term treatment with lithium. On the other hand, such a phenomenon was not observed with continued treatment with anticonvulsants, antidepressants, or antipsychotics [32].

In 2015, Gerhard et al. [33] examined the association of lithium and dementia risk in a large claim-based US cohort of publicly insured older adults with bipolar disorder (n = 41,931), including individuals ≥50 years who did not receive dementia-related services during the prior year. Each follow-up day was classified by past-year cumulative duration of lithium use.


across several Texas counties. 6180 water samples from public wells since 2007 were obtained, and changes in AD mortality rates were calculated by subtracting aggregated age-adjusted mortality rates between 2000 and 2006 from those between 2009 and 2015. The authors found that the age-adjusted AD mortality rate was significantly (+27%) increased over time. Changes in AD mortality were negatively correlated with trace lithium levels, and statistical significance was maintained after controlling for most risk factors except for physical inactivity, obesity, and type 2 diabetes. Furthermore, the prevalence of obesity and type 2 diabetes positively correlated with changes in AD mortality but also negatively correlated with trace lithium in drinking water. The results suggest that trace lithium in water may be negatively linked with changes in AD mortality, as well as obesity and type 2 diabetes, which are important risk factors for AD. The chronological arrangement of epidemiological studies on lithium and dementia is presented

Lithium and Alzheimer's Disease: Experimental, Epidemiological, and Clinical Findings

http://dx.doi.org/10.5772/intechopen.74239

85

In 2008, Macdonald et al. [36] first attempted to assess the safety and feasibility of prescribing long-term lithium (up to 1 year) to 22 elderly people with mild to moderate Alzheimer's disease (AD) in an open-label study. A comparison group not receiving lithium therapy was matched for cognition and age. The mean duration of treatment for 14 patients who discontinued prematurely was 16 weeks and for those continuing treatment at the end of the study was 39 weeks. The reason for discontinuation in three patients was possible side effects which disappeared on stopping therapy. The intensity of side effects did not differ between patients discontinuing therapy and the subjects remaining in the study. Two patients receiving lithium died; however, in neither case the treatment with lithium was related to the cause of death. The lithium and non-lithium groups were not different as to deaths, drop outs, or change in MMSE. In 2009, the first randomized lithium trial in patients with mild AD appeared [37]. Seventyone patients were randomized to receive either lithium (0.5–0.8 mmol/l) (n = 33) or a placebo for (n = 38) 10 weeks. The results obtained showed that there were no differences as to global cognitive performance, as measured by the ADAS-Cog subscale, depressive symptoms, as well as plasma activity of GSK-3 and disease biomarker concentrations in the cerebrospinal fluid (CSF), between lithium and placebo groups [42]. However, interesting results were obtained by an analysis of a single site subsample (Tübingen) containing 27 patients, 13 of which were randomized to lithium and 14 to placebo. In AD patients treated with lithium, in comparison to placebo-treated patients, a significant increase of BDNF serum levels and a significant decrease of cognitive impairment measured by the ADAS-Cog sum scores, inversely

Two Brazilian studies performed in 2011 and 2013 brought about some promising results. Forlenza et al. [39] employed lithium in placebo-controlled trial of 45 patients with amnestic mild cognitive impairment (MCI), randomized to lithium (n = 24) or placebo (n = 21) for 12 months. They found that lithium treatment (0.25–0.5 mmol/l) was associated with significantly better performance on the cognitive subscale of the Alzheimer's Disease Assessment

in **Table 1**.

**4. Clinical studies of lithium in MCI and AD**

correlated with lithium serum concentration, were found [38].

**Table 1.** Epidemiological studies of lithium and dementia.

Compared with nonuse, 301–365 days of lithium exposure was associated with significantly reduced dementia risk No corresponding association was observed for shorter lithium exposures or any exposure to anticonvulsants that may suggest that long-term lithium treatment may reduce dementia risk in older adults with bipolar disorder.

Recently, two papers appeared studying a relationship between lithium in drinking water and dementia. Kessing et al. [34] performed a Danish nationwide, case–control research, studying an association between the municipality of residence and measurements of lithium in drinking water. The data were obtained from all patients between 50 and 90 years of age who had a diagnosis of dementia during hospitalization, from 1970 to 2013. A total of 73, 731 patients with dementia and 733, 653 controls were included in the study. Lithium exposure was statistically significantly different between patients with a diagnosis of dementia and controls, and a nonlinear association was observed. Compared with individuals exposed to 2.0–5.0 μg/L, the incidence rate ratio of dementia was decreased in those exposed to more than 15.0 μg/L and 10.1–15.0 μg/L and increased with 5.1–10.0 μg/L. Similar patterns were found for Alzheimer's disease and vascular dementia as outcomes. In the second study, Fajardo et al. [35] examined the relationship between trace levels of lithium in drinking water and changes in AD mortality across several Texas counties. 6180 water samples from public wells since 2007 were obtained, and changes in AD mortality rates were calculated by subtracting aggregated age-adjusted mortality rates between 2000 and 2006 from those between 2009 and 2015. The authors found that the age-adjusted AD mortality rate was significantly (+27%) increased over time. Changes in AD mortality were negatively correlated with trace lithium levels, and statistical significance was maintained after controlling for most risk factors except for physical inactivity, obesity, and type 2 diabetes. Furthermore, the prevalence of obesity and type 2 diabetes positively correlated with changes in AD mortality but also negatively correlated with trace lithium in drinking water. The results suggest that trace lithium in water may be negatively linked with changes in AD mortality, as well as obesity and type 2 diabetes, which are important risk factors for AD.

The chronological arrangement of epidemiological studies on lithium and dementia is presented in **Table 1**.
