Controversial data have been reported with regard to the diagnostic value of the positron emission tomographic (PET) measurement of striatal glucose consumption (rCMRGlc) in chorea-free subjects at risk of Huntington's disease (HD). For further clarification of this issue we measured striatal and cerebellar rCMRGlc in 27 chorea-free subjects at risk of HD, 20 patients with manifest HD and 20 control subjects, using PET and 18F-fluorodeoxyglucose. In 6 of the at-risk subjects cerebellar ratios of striatal rCMRGlc were decreased below the corresponding 99% confidence limit determined in the controls. This indicates that the PET measurement of rCMRGlc may, indeed, be valuable in establishing the diagnosis of incipient HD in presymptomatic at-risk subjects.