For this, we partly have to thank the French epidemiologists who performed a series of long-term surveys of the incidence of AD and drinking water. The biggest one was an eight-year study tracking the onset of dementia in everyone over the age of 65 in an area of southwest France. Within this general area, many local boroughs have their own water supplies.
What the researchers found seemed to be clear-cut. In the boroughs with relatively high levels of natural aluminium in the water (more than 0.1 mg/L), there was a more than doubling of the number of AD cases
(Am J Epidemiol, 2000; 152: 59-66).
However, the small print of the data wasn't quite as simple as that. The same researchers had previously found that there were two other key factors, without which the aluminium-AD connection no longer showed up. One factor was the acidity of the water. In areas where the water was alkaline, aluminium had no effect. Equally, where there were high natural levels of silica in the water, high aluminium not only had no effect, but actually seemed to protect against AD (Epidemiology, 1996; 7: 281-5).
Given apparently paradoxical results like these, the fact that there's been so much to-ing and fro-ing over the dangers of aluminium is hardly surprising. Added to that, it's known that non-prescription antacids contain vastly more aluminium than any water supply does and, yet, chronic antacid addicts appear to be at no greater risk of developing Alzheimer's than anyone else (Regul Toxicol Pharmacol, 2001; 33: 66-79).
The current thinking is that aluminium in the water supply must be more easily taken up by the body (Brain Res Bull, 2001; 55: 187-96).
Aluminium is probably not the only metal to be involved in Alzheimer's. Copper, zinc and iron have also been fingered as possible culprits (J Biol Inorg Chem, 2004; 9: 954-60), as has mercury (see Viewpoint). In these cases, the evidence comes from either laboratory cell cultures, animal studies or autopsies of AD patients rather than from surveys of living people. Nevertheless, experts are becoming iconvinced that even small quantities of these four metals are involved in AD. The metals are thought to create an oxidative inflammatory response in the brain, causing the brain cells to clump into the disease's characteristic neural plaques (J Alzheimers Dis, 2004; 6: 291-301).
The inflammation theory is further bolstered by the finding that AD sufferers tend to have high levels of homocysteine, a naturally occurring amino acid. It is best known as a marker for heart disease, but it's also believed to be toxic to brain cells, probably through an inflammatory mechanism (Am J Clin Nutr, 2005; 82: 636-43).