Chronic inflammation is associated with a broad spectrum of neurodegenerative diseases of aging. Included are such disorders as Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis, the Parkinson-dementia complex of Guam, all of the tauopathies, and age-related macular degeneration. Also included are such peripheral conditions as osteoarthritis, rheumatoid arthritis, atherosclerosis, and myocardial infarction. Inflammation is a two-edged sword. In acute situations, or at low levels, it deals with the abnormality and promotes healing. When chronically sustained at high levels, it can seriously damage viable host tissue. We describe this latter phenomenon as autotoxicity to distinguish it from autoimmunity. The latter involves a lymphocyte-directed attack against self proteins. Autotoxicity, on the other hand, is determined by the concentration and degree of activation of tissue-based monocytic phagocytes. Microglial cells are the brain representatives of the monocyte phagocytic system. Biochemically, the intensity of their activation is related to a spectrum of inflammatory mediators generated by a variety of local cells. The known spectrum includes, but is not limited to, prostaglandins, pentraxins, complement components, anaphylotoxins, cytokines, chemokines, proteases, protease inhibitors, adhesion molecules, and free radicals. This spectrum offers a huge variety of targets for new anti-inflammatory agents. It has been suggested, largely on the basis of transgenic mouse models, that stimulating inflammation rather than inhibiting it can be beneficial in such diseases as AD. If this were the case, administration of NSAIDs, or other anti-inflammatory drugs, would be expected to exacerbate conditions such as AD, PD, and atherosclerosis. However, epidemiological evidence overwhelmingly demonstrates that the reverse is true. This indicates that, at least in these diseases, the inflammation is harmful. So far, advantage has not been taken of opportunities indicated by these epidemiological studies to treat AD and PD with appropriate anti-inflammatory agents. Based on this evidence, classical NSAIDs are the most logical choice. Dosage, though, must be sufficient to combat the inflammation. Analysis of mRNA levels of inflammatory mediators indicates that the intensity of inflammation is considerably higher in AD hippocampus and in PD substantia nigra than in osteoarthritic joints. Thus, full therapeutic doses of NSAIDs, or combinations of anti-inflammatory agents, are needed to achieve the suggested neurological benefits.