Average whole brain metabolic rates, when corrected for brain atrophy, are similar between patients with Alzheimer's disease (AD) and age-matched controls. To elucidate the relationship between reduced cognitive function and cerebral metabolism in patients with AD, we hypothesized that the absolute amount of glucose used by the entire brain may prove to be a more reliable indicator of the disease than metabolic rates calculated for a unit of brain weight. Twenty patients with the probable diagnosis of AD and 17 similarly aged controls underwent 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) studies as well as magnetic resonance imaging (MRI) within a few days of each other. Average metabolic rates, when corrected for atrophy, were 3.91 +/- 1.02 and 4.43 +/- 0.87 (mg of glucose per 100 cc brain tissue per min +/- s.d.) respectively for AD patients and controls. Two other indices were determined, atrophy weighted total brain metabolism (calculated by multiplying the brain volume, determined by MR analysis, by the average metabolic rate) and absolute whole brain metabolism (calculated by multiplying the brain volume by the average metabolic rate corrected for atrophy). The former showed a very significant difference between the two groups (29.96 +/- 7.90 for AD patients compared to 39.1 +/- 7.0 for controls, p < 0.001). Atrophy weighted total brain metabolism also correlated very well with mini mental status exam (MMSE) scores (r = 0.59, p < 0.01). Absolute whole brain metabolism was significantly different between AD and control groups and correlated well with MMSE. These data demonstrate that although the metabolic rate per unit weight of the brain is unchanged in AD compared to controls, atrophy weighted total brain metabolism and absolute whole brain metabolism are significantly affected. Both indices may prove to be a sensitive correlate for cognitive dysfunction in AD.