Data on the concentrations and distributions of aluminum (Al) in Alzheimer's disease (AD) specimens reported by various laboratories are difficult to compare for several reasons. These include different microanalytical techniques, specimen preparation procedures, artifacts, and detection limits. The fundamental information necessary to interpret and to quantitatively compare data, image data in particular, has been lacking in many published results. Sample populations have not been selected to a consistent standard of disease definition. As a result, not only has it been impossible to clarify the role of Al in the etiology of the disease, but it has been impossible even to determine the conditions under which it may have a role. To examine the issue of artifacts, rat-brain homogenates were prepared using extremely stringent methods to avoid Al contamination. These homogenates were doped with aluminum, giving a range of Al concentrations up to 200 ppm. The samples were analyzed by graphite furnace atomic absorption spectrometry (AA) and electron beam X-ray microanalysis (EPMA). By AA, the concentration of Al in pure rat-brain homogenate was found to be an order of magnitude lower than most published values, but concentrations threefold to tenfold greater than the average for each sample set were found. Good agreement between AA and EPMA was obtained by taking into account, by spectrum modeling, the total sample composition and thickness for EPMA measurements.
