288 Letters to the Editor Am J Epidemiol Vol. 154, No. 3, 2001 portions.” (Letter). Am J Epidemiol 2001;154:287. 2. Schaid DJ, Jacobsen SJ. Biased tests of association: compar- isons of allele frequencies when departing from Hardy- Weinberg proportions. Am J Epidemiol 1999;149:706–11. 3. Armitage P. Tests for linear trends in proportions and frequen- cies. Biometrics 1955;11:375–86. Daniel J. Schaid Steven J. Jacobsen Department of Health Sciences Research Mayo Clinic Rochester, MN 55905 RE: “LUNG CANCER AMONG INDUSTRIAL SAND WORKERS EXPOSED TO CRYSTALLINE SILICA” In a 22-year study in 18 plants of 4,027 domestic indus- trial sand workers exposed to crystalline silica, Steenland and Sanderson (1) observed a 60 percent excess of lung can- cer mortality in the exposed cohort compared with the US male population. This finding supports the 1997 judgment by the International Agency for Research on Cancer that inhaled crystalline silica from occupational sources is a car- cinogen (2). Indeed, as early as 1940, inhaled silica was shown to induce lung cancers in mice (3). However, the International Agency for Research on Cancer noted that the epidemiologic evidence was not entirely consistent and that different forms of silica might vary in carcinogenic efficacy. A possible cause of fluctua- tion in potency might be variable contamination of silica with a known carcinogen such as iron (4). In addition, inconsistency could reflect different levels of tissue iron in persons who inhale silica. Evidence is available to support each of these possibilities. Ghio et al. (5) observed that the ability of silica to stimu- late production of reactive oxygen radicals and induce lung inflammation was suppressed by the iron chelator desfer- rioxamine. Furthermore, in rats exposed to silica, exogenous iron contamination of the inhaled sand markedly enhanced such lung responses as leukocyte recruitment, macrophage synthesis of oxygen radicals and nitric oxide, and lipid per- oxidation (6). Moreover, after introduction of silica into the lower res- piratory tract, surface complexation of iron has been observed to occur on the phagocytosed particles (5). Subsequently, silica was instilled intratracheally in rats maintained on low- or high-iron diets. Compared with ani- mals fed high-iron diets, rats on the low amount of iron had diminished fibrotic injury (7). Iron loading of humans can occur via alimentary or respi- ratory routes. Ingestion of excessive amounts of red meat or of alcohol as well as several genetic disorders can enhance intestinal absorption of iron. Sources of inhaled iron include mainstream cigarette smoke, asbestos, and urban air partic- ulates (4). Thus, both the use of protective masks and mea- sures to reduce alimentary iron loading might be useful in minimizing the carcinogenic danger of silica for industrial sand workers. REFERENCES 1. Steenland K, Sanderson W. Lung cancer among industrial sand workers exposed to crystalline silica. Am J Epidemiol 2001; 153:695–703. 2. International Agency for Research on Cancer. Silica, some sil- icates, coal dust, and para-aramid fibrils. Monograph 68. Evaluation of carcinogenic risks to humans. Lyon, France: International Agency for Research on Cancer, 1997. 3. Campbell JA. Effects of precipitated silica and of iron oxide on the incidence of primary lung tumours in mice. Br Med J 1940;2:275–80. 4. Weinberg ED. The development of awareness of the carcino- genic hazard of inhaled iron. Oncol Res 1999;11:109–13. 5. Ghio AJ, Kennedy TP, Wharton AR et al. Role of surface com- plexed iron in oxidant generation and lung inflammation induced by silicates. Am J Physiol 1992;263:L511–18. 6. Castranova V, Vallyathan V, Ramsay DM, et al. Augmentation of pulmonary reactions to quartz inhalation by trace amounts of iron-containing particles. Environ Health Perspect 1997; 105(suppl 5):1319–24. 7. Ghio AJ, Jaskot RH, Hatch GE. Lung injury after silica instil- lation is associated with an accumulation of iron in rats. Am J Physiol 1994;267:L686–92. Eugene D. Weinberg Department of Biology and Program in Medical Sciences Indiana University Bloomington, IN 47405 Editor’s note: In accordance with Journal policy, Drs. Steenland and Sanderson were asked whether they wished to respond to this letter but chose not to do so. RE: ALUMINUM IN DRINKING WATER AND COGNITIVE DECLINE IN ELDERLY SUBJECTS: THE PAQUID COHORT Although the neurotoxicity of aluminum has been proved (1), the link between aluminum and the risk of Alzheimer’s disease is still debated (2). A related hypothesis has been put forward by Birchall and Chappell (3): silicon in water could be a protective factor against aluminum toxicity. We recently reported a significant association between the concentration of aluminum in drinking water and the incidence of dementia and Alzheimer’s disease (4). These results were based on a cohort of 3,777 elderly subjects fol- lowed up for 8 years. In the present study, we have evaluated in the Paquid cohort the association between aluminum in drinking water and the 8-year evolution of cognitive functions measured by the Mini-Mental State Examination score, a major predictor for dementia. The study has two main methodological inter- ests. First, the evolution of the Mini-Mental State by guest on October 14, 2011 aje.oxfordjournals.org Downloaded from