Dementia is becoming a major health problem as the population of the Northern Hemisphere ages. Early differential diagnosis between normal cognitive decline and dementia is particularly difficult. If psychometric evaluation can contribute to the diagnosis, quantitative cerebral functional imaging would play an important role. We therefore proposed, first, to constitute a normative dataset that could later be used to identify subjects at risk for neurodegenerative processes and, second, to describe the risk of abnormal global cerebral blood flow (gCBF) by defining categories based on the standardized cutoff scores of a young, healthy population (T-score). METHODS: Of a total of 203 healthy volunteers, 187 were included in the protocol, which included evaluation of medical history, neurologic and neuropsychologic status, and body composition; analysis of blood; and measurement of gCBF by means of (133)Xe SPECT. RESULTS: With ANOVA analysis using age and sex as between-subject factors and gCBF as a within-subject factor, a significantly higher gCBF was found in women than in men. In addition, a linear reduction as a function of age was observed for both sexes (-0.3%/y). T-score was determined for the 18- to 28-y-old age group, for whom gCBF was found to be 46.7 +/- 5.1 mL/min/100 g tissue in men and 49.0 +/- 5.0 mL/min/100 g tissue in women. The age-dependent decrease could thus be expressed in T-scores and, in the 29- to 38-y-old, 39- to 48-y-old, and >48-y-old age groups, averaged -0.63, -1.29, and -1.92, respectively, in men and -0.63, -0.83, and-2.40, respectively, in women. Cognitive performance, body composition, and blood analysis revealed the expected significant effects from sex and age. CONCLUSION: The large-scale reference database of gCBF measurements constituted from a healthy, well-controlled population enabled age and sex stratification, which showed significant differences between the sexes and a significant decline as a function of age. T-scores were determined and warrant further studies on the prospective identification of early dementia by (133)Xe SPECT in elderly individuals.