Skip to main content

Table 5 Individual studies reporting the association between other metals and dementia

From: Environmental risk factors for dementia: a systematic review

Study

Exposure

Sample description

N

Methodology/design

Finding

Grade

Rondeau et al., 2009 [39]

Aluminium

PAQUID: A community-based cohort of 3,777 elderly people aged ≥65 years in SW France.

1925 individuals of whom 461 developed clinically diagnosed dementia (364 AD).

Prospective cohort study: Cox PH models. Mean levels of aluminium in drinking water over the previous decade based on current residential location was linked to incident dementia over 15 years follow up.

Highest:lowest quartile of aluminium in drinking water was associated with an increased risk of dementia and AD (multivariable-adjusted HR, 95 % CI: 2.34, 1.03-5.32; 3.04, 1.32-6.97).

A

Flaten, 1990 [40]

Aluminium

The Norwegian population – mortality data provided by the Central Bureau of Statistics of Norway.

5,642 male and 9,085 female dementia deaths were recorded 1974–1983. The denominator population is estimated at 40.6 million person-years. The population of Norway was approximately 4 million people during this period.

Cross-sectional study: Pearson’s correlation and relative risk. Four water samples (one per season) were collected in 1982–3 from 384 waterworks and analysed. Small municipalities were aggregated to regions of at least 10,000 inhabitants. Dementia was ascertained from death certification (any mention).

There was a correlation between the aluminium content of drinking water at municipality level and dementia death rates in four time periods from 1969–83 in men and women (P < 0.025 and <0.005) but not with PD or ALS death rates. This pattern was less clear at county level.

B

Highest:lowest tertile of aluminium concentration was associated with an increased risk of dementia death in men and women (RR, 95 % CI: 1.32, 1.20-1.46; 1.42, 1.32-1.54).

Forbes et al., 1995 [41]

Aluminium

Males from the Ontario Longitudinal Study of Aging dying between 1984 and 1991.

3161 patients with AD or young-onset dementia.

Cross-sectional study: Poisson regression. Water quality was based on 30 years’ residential history. Dementia was identified from death certificates (underlying cause).

Highest:lowest tertile of aluminium concentration was associated with an increased risk of AD and AD plus young-onset dementia (RR, 95 % CI: 2.42, 1.42-4.11; 1.96, 1.15-3.32). Effect sizes were greater in analyses restricted to individuals aged ≥75 years.

B

Total sample not stated.

Some models showed a lower risk of dementia in the middle tertile of aluminium concentrations compared to the lowest.

Neri & Hewitt, 1991 [44]

Aluminium

Patients discharged from general hospitals in Ontario, Canada in 1986.

2344 people aged ≥55 years with a diagnosis of AD or young-onset dementia recorded and 2232 people, matched for age and sex, with a non-psychiatric diagnosis recorded.

Cross-sectional study (case–control). Aluminium concentration in drinking water based on residential location was compared with dementia diagnoses based on hospital discharge statistics.

The authors report a dose–response pattern of association (OR for quartiles 2–4 compared to lowest, 95 % CI: 1.13, 0.55-2.29; 1.26, 0.61-2.59; 1.46, 0.71-2.99) but all CIs (not reported but calculated from data reported in paper) include unity.

B

Gillette-Guyonnet et al., 2005 [57]

Aluminium

Toulouse subset of the EPIDOS study cohort of women aged ≥75 years.

1462 women from one centre of whom 60 developed clinically diagnosed AD. 323 had normal cognitive function and the remainder were lost to follow up.

Prospective cohort study: logistic regression. Water consumption based on self-report (at baseline and follow up) was combined with local tap water composition data. Dementia was clinically diagnosed.

It is not clear if aluminium consumption was included in logistic regression models and not found to be a predictor of dementia or if it wasn’t included.

B

Martyn et al., 1997 [46]

Aluminium

Participants were selected from CT records of eight neuroradiology centres in the UK.

A total of 872 men (106 AD, 99 other dementia, 226 brain cancer, 441 other disease of the nervous system)

Cross-sectional study (case–control): logistic regression. Average levels of aluminium in drinking water based on residential history (after age 25 years) was related to diagnosis based on hospital records.

There were no associations identified between aluminium concentrations over three time periods (from age 25 years to diagnosis; from age 25 years to 10 years before diagnosis; and the ten years preceding diagnosis) with each of the three comparison groups.

B

McLachlan et al., 1996 [43]

Aluminium

Participants were selected from a brain tissue bank in Ontario, Canada.

296 participants had AD, 89 had mixed dementia, 125 had no histopathological abnormalities, and 170 had other conditions (HD, schizophrenia, MS, multiple infarcts, CJD, and other neurodegenerative diseases).

Cross-sectional study (case–control). Average aluminium concentrations in domestic water supply of residential location at death (10 year residential history was available for a subsample) were compared to dementia status (neuropathologically diagnosed).

Aluminium concentration >100 μg/L compared to <100 μg/L was associated with an increased risk of AD and dementia (OR, 95 % CI: 1.7, 1.2-2.6; 1.7, 1.2-2.5).

B

Effect sizes were larger for the subgroup analysis based on 10-year residential history (2.6, 1.2-5.7; 2.5, 1.2-5.3).

Frecker, 1991 [42]

Aluminium

Mortality records in Bonavista Bay, Newfoundland, 1985–86.

191 dementia deaths in 1985, 208 deaths in 1986

Cross-sectional study. Place of birth of all individuals dying with dementia was identified and associated with drinking water samples at those locations from 1986.

The area with the highest dementia mortality (37.5 % in 1985 and 68.8 % in 1986) also had the highest aluminium concentrations in drinking water. This association was not assessed for statistical significance, but was argued to not be confounded by age, sex or place of residence stated on death certificate.

B

Emard et al., 1994 [37]

Aluminium

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

44 individuals with AD were born in areas with lower than average concentration of aluminium; 41 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Forster et al., 1995 [47]

Aluminium

Patients were drawn from a specialist service in the north of England. Controls were randomly selected from the general population.

109 people with clinically diagnosed young-onset AD and 109 age- and sex-matched controls.

Cross-sectional study (case–control). Aluminium levels in drinking water were determined according to residential history (longest residence in the preceding 10 years plus birthplace for a subset of 80 cases and controls) at local authority district level.

There was no evidence of an association between aluminium concentration in drinking water and dementia risk (e.g. aluminium >149 mg/L OR, 95 % CI: recent residence 1.0, 0.41-2.43; birthplace 1.1, 0.38-3.35).

B

Taylor et al., 1995 [48]

Aluminium

Same cohort as Forster et al., 1995 [47]

Water samples were obtained for 214 addresses of the 218 cases and controls.

Cross-sectional study (case–control). Aluminium concentration in water samples drawn from the place of residence at which they had lived longest within 10 years prior to onset of dementia (or equivalent date for controls).

There were no differences in aluminium concentrations in samples for cases or controls (P = 0.60).

B

Gauthier et al., 2000 [107]

Aluminium

Random sample of individuals living in SLSJ aged ≥70 years from the provincial health plan of Quebec in 1994.

68 participants with clinically-diagnosed AD and 68 age- and sex-matched controls.

Cross-sectional study (case–control). Aluminium concentration in drinking water was sampled four times in 1995–6 in 54 municipalities of SLSJ. Long-term exposure was estimated based on residential history.

Of all variables involved in the speciation of aluminium in drinking water (total Al, total dissolved Al, total monomeric Al, organic monomeric Al, inorganic monomeric Al, polymeric Al, as well as the main monomeric inorganic forms) based on recent and long-term exposure, only recent exposure to monomeric organic aluminium was associated with AD at conventional levels (OR, 95 % CI 2.67, 1.04-6.90).

B

Shen et al., 2014 [49]

Aluminium

26 provinces and 3 municipal districts of mainland China.

Not specified.

Cross-sectional study. Soil chemical levels in 1990 were related to AD mortality 1991–2000.

Higher aluminium levels in soil were associated with reduced AD mortality (highest:lowest group RR, 95 % CI: 0.267, 0.265-0.268).

B

Vogt, 1986 [45]

Aluminium

Mortality data from the Central Bureau of Statistics (1969–1983) and dementia patients within 112 psychiatric nursing homes (as collected by Norwegian Institute for Gerontology in 1982).

Not specified.

Cross-sectional study. Drinking water quality and acidification of lakes was related to standardized dementia mortality rates.

Highest:lowest zone of aluminium concentration was associated with a raised standardised dementia mortality rate (per 10,000 inhabitants per year: 48.3 vs 32.4). The effect size was greater in women (59.4 vs 38.5) than men (36.9 vs 26.1).

B

Wettstein et al., 1991 [108]

Aluminium

Residents of two Zurich city districts.

775 men and women aged 82–85 years who had lived in that area for at least 15 years.

Cross-sectional study. Aluminium concentrations were measured in drinking water. The dementia outcome was measured using the truncated MMSE.

No significant difference was observed between the participant groups on mnestic or naming subscores (P = 0.962 and P = 0.567).

C

Civita, Fiorucci, & Mie, 2001 [109]

Aluminium

AD mortality in municipalities within 20 km of Alba based on national mortality statistics. Census data were used as a reference population.

Not stated.

Cross-sectional study. A map of the location of deaths with AD was compared to levels of aluminium released from 1 kg of clay at various sampling sites.

They found an increased SMR in the municipality compared to the province or the whole of Italy. They also noted that the areas with highest dementia mortality had the highest aluminium concentrations in water.

C

Gillette-Guyonnet et al., 2005 [57]

Calcium

Toulouse subset of the EPIDOS study cohort of women aged ≥75 years.

1462 women from one centre of whom 60 developed clinically diagnosed AD. 323 had normal cognitive function and the remainder were lost to follow up.

Prospective cohort study: logistic regression. Water consumption based on self-report (at baseline and follow up) was combined with local tap water composition data. Dementia was clinically diagnosed.

Women who developed AD showed a decrease in daily calcium intake at follow up. It is not clear if calcium consumption was included in logistic regression models and not found to be a predictor of dementia or if it wasn’t included.

B

Emard et al., 1994 [37]

Cobalt

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

21 individuals with AD were born in areas with lower than average concentration of cobalt; 20 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Emard et al., 1994 [37]

Copper

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

18 individuals with AD were born in areas with lower than average concentration of copper; 30 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Shen et al., 2014 [49]

Copper

26 provinces and 3 municipal districts of mainland China.

Not specified.

Cross-sectional study. Soil chemical levels in 1990 were related to AD mortality 1991–2000.

Copper concentration correlated with annual AD mortality after three outlier provinces were removed (r = 0.449, P = 0.021).

B

Loef & Walach, 2012 [24]

Copper

Systematic review of studies relating copper to AD from 11 databases.

101 studies:

Systematic review.

Of relevant evidence that is reviewed, the authors conclude that “In summary, the current trials provide no conclusive evidence that depletion or supplementation of Cu is beneficial for AD… [t]he specific outcomes for Cu are more conflicting; while evidence suggests that the systemic Cu level is increased in patients with AD, further research is needed to define the alterations of Cu in the brain during AD.” (p.6).

-

2 meta analyses, 2 systematic reviews, 11 RCTs, 2 prospective studies, 3 cross-sectional studies, 45 case–control studies, 30 autopsy studies,5 uncontrolled studies, 1 case study.

Emard et al., 1994 [37]

Iron

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

16 individuals with AD were born in areas with lower than average concentration of iron; 35 were born in areas with higher than average concentrations. This difference was statistically significant (P < 0.05).

B

Shen et al., 2014 [49]

Iron

26 provinces and 3 municipal districts of mainland China.

Not specified.

Cross-sectional study. Soil chemical levels in 1990 were related to AD mortality 1991–2000.

Higher iron levels in soil were associated with increased AD mortality (highest:lowest group RR, 95 % CI: 1.248, 1.245-1.251). Iron concentration correlated with annual AD mortality after three outlier provinces were removed (r = 0.537, P = 0.007).

B

Loef & Walach, 2012 [24]

Iron

Systematic review of studies relating iron to AD from 11 databases.

101 studies:

Systematic review.

“In summary, the current trials provide no conclusive evidence that depletion or supplementation of … Fe is beneficial for AD… Fe has been consistently found at elevated levels in the brains of AD sufferers by both autopsy and case–control studies.” (p.6).

-

2 meta analyses, 2 systematic reviews, 11 RCTs, 2 prospective studies, 3 cross-sectional studies, 45 case–control studies, 30 autopsy studies,5 uncontrolled studies, 1 case study.

Emard et al., 1994 [37]

Manganese

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

12 individuals with AD were born in areas with lower than average concentration of manganese; 35 were born in areas with higher than average concentrations. This difference was statistically significant (P < 0.05).

B

Emard et al., 1994 [37]

Molybdenum

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

29 individuals with AD were born in areas with lower than average concentration of molybdenum; 16 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Emard et al., 1994 [37]

Nickel

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

27 individuals with AD were born in areas with lower than average concentration of nickel; 29 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Emard et al., 1994 [37]

Uranium

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

10 individuals with AD were born in areas with lower than average concentration of uranium; 9 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Emard et al., 1994 [37]

Zinc

The IMAGE Project covering the population of Saguenay-Lac-Saint-Jean (SLSJ), Québec.

129 individuals with AD (clinically diagnosed by standard medical services) who were born in SLSJ.

Cross-sectional study: principal components analysis. Samples of aquatic sediment were analysed for geochemical variables. AD cases were identified from a registry.

29 individuals with AD were born in areas with lower than average concentration of aluminium; 30 were born in areas with higher than average concentrations. This difference is not statistically significant.

B

Shen et al., 2014 [49]

Zinc

26 provinces and 3 municipal districts of mainland China.

Not specified.

Cross-sectional study. Soil chemical levels in 1990 were related to AD mortality 1991–2000.

Higher zinc levels in soil were associated with increased AD mortality (highest:lowest group RR, 95 % CI: 2.289, 2.275-2.303).

B

  1. AD Alzheimer’s dementia, ALS amyotrophic lateral sclerosis, CI confidence interval, CJD Creutzfeld-Jakob Disease, CT computed tomography, HD Huntington’s disease, HR hazard ratio, MCI mild cognitive impairment, MMSE mini-mental state examination, MS multiple sclerosis, PD Parkinson’s disease, PH proportional hazards, RCT randomized, controlled trial, RR relative risk, SD standard deviation, SEM standard error of the mean, SMR Standardised Mortality Ratio
  2. Figures in italics indicate data not reported but which have been calculated from data in the paper