Accurate, early differentiation of dementias will become increasingly important as new therapies are introduced. Differential diagnosis by standard clinical criteria has limited accuracy. PET offers the potential to increase diagnostic accuracy. (18)F-FDG studies detect metabolic abnormalities in demented patients, but with limited specificity. PET also offers the ability to quantify other biochemical markers that can yield additional useful diagnostic information. We demonstrate that (+)-(11)C-dihydrotetrabenazine ((11)C-DTBZ) studies, which provide an index of nigrostriatal terminal density (distribution volume; DV), also provide a measure of transport (K(1)) that produces information comparable to the metabolic measure of (18)F-FDG. METHODS: Fifty-two patients and 19 control subjects underwent both (11)C-DTBZ and (18)F-FDG PET scans. Seven had the clinical diagnosis of frontotemporal dementia (FTD), 25 had Alzheimer's disease (AD), and 20 had dementia with Lewy bodies (DLB). DTBZ scans provided maps of K(1) and DV, whereas (18)F-FDG scans provided maps of glucose metabolism. Correlation analyses were performed between the different PET measures both within and across subjects. Discriminant analysis using logistic regression compared the performance of (11)C-DTBZ K(1) and (18)F-FDG in differentiating subject groups. Three experienced PET researchers participated in an interrater reliability study using both (11)C-DTBZ K(1) and (18)F-FDG images. RESULTS: Within-subject correspondence between (11)C-DTBZ K(1) and (18)F-FDG measures was high, with correlations averaging 0.92. Correlations between the (11)C-DTBZ DV and either K(1) or (18)F-FDG were far lower, averaging 0.37 and 0.31, respectively, indicating the much higher degree of similarity in information provided by K(1) and (18)F-FDG compared with the very different information provided by (11)C-DTBZ DV. Discriminant analysis demonstrated that (11)C-DTBZ K(1) and (18)F-FDG yielded similar levels of sensitivity and specificity for differentiating the subjects in this study. Including (11)C-DTBZ DV in addition to either K(1) or (18)F-FDG improved discrimination between groups. The raters classified PET scans nearly equivalently using K(1) and (18)F-FDG. CONCLUSION: Multiple PET measures, whether 2 parameters from a single PET study such as (11)C-DTBZ K(1) and DV, or 2 parameters from different studies such as (18)F-FDG and (11)C-DTBZ DV, offer complementary information useful for diagnosing dementias. K(1) and DV images generated from a single (11)C-DTBZ scan provide as much diagnostic information as 2-scan studies using both (11)C-DTBZ and (18)F-FDG.