Local cerebral blood flow (CBF) measured by xenon-enhanced CT was correlated with the clinical findings in 21 patients with motor neuron disease (mean +/- SD age, 60 +/- 10 years). In 11 patients, CBF was also measured after intravenous injection of 2 mg thyrotropin releasing hormone (TRH). There were no significant differences in CBF of the cerebral cortex, basal ganglia, thalamus, and subcortical white matter with respect to age, duration of illness, or presence (n = 9) or absence (n = 12) of bulbar palsy. In patients with upper motor neuron disturbance (n = 10), however, local CBF in the frontal cortex was significantly lower (p < 0.05) than in those without it (n = 11). Within the frontal cortex, CBF reduction was marked in the upper posterolateral part (p < 0.05) that included the motor and premotor areas. Intravenous administration of TRH did not significantly alter the local CBF in any of the brain regions examined. An additional patient with motor neuron disease and severe dementia showed marked CBF reduction in the frontal lobe, which was in common with but much greater than the reduction in those exhibiting subcortical dementia (e.g., progressive supranuclear palsy). We conclude that in motor neuron disease patients with upper motor neuron involvement, the selective reduction of CBF in brain regions that include primary motor and premotor areas may reflect functional disturbance or neuronal degeneration in these regions. The ameliorating effect of TRH in patients with motor neuron disease, if any, appears not to be related to increases in local CBF or activation of brain function.