The (F-18) fluorodeoxyglucose (FDG) technique to measure local cerebral metabolic rate for glucose (LCMRglu) is well accepted and widely used by many institutions around the world. A large number of studies has been carried out in normal volunteers and patients with a variety of CNS disorders. Several investigators have noted that no significant age-related changes in cerebral glucose use occur with normal aging. Some important and interesting findings have been revealed following sensory, motor, visual, and auditory stimulations. Functional imaging with FDG in certain neurologic disorders has dramatically improved our understanding of their underlying pathophysiologic phenomena. Some abnormalities detected on the positron emission tomography (PET) images have no corresponding changes on either x-ray computed tomograms (XCT) or magnetic resonance images (MRI). In patients with Alzheimer's disease, primary sensorimotor, visual, and cerebellar metabolic activity appears relatively preserved. In contrast, parietal, temporal, and to some degree, frontal glucose metabolism is significantly diminished even in the early stages of the disease. Patients with Huntington's disease and those at risk of developing this disorder have a typical pattern of diminished CMRglu in the caudate nuclei and putamen. In patients with stroke, PET images with FDG have demonstrated abnormal findings earlier than either XCT or MRI and with a wider topographic distribution. FDG scans have revealed interictal zones of decreased LCMRglu in approximately 70% of patients with partial epilepsy. The location of the area of hypometabolism corresponds to the site of the epileptic focus as determined by electroencephalography and microscopic examination of the resected tissue. Ictal scans during partial seizures demonstrate areas of hypermetabolism corresponding to the sites of seizure onset and spread. Several investigators have reported relative hypofrontal CMRglu in patients with schizophrenia. In our center, FDG scans from patients with schizophrenia were successfully differentiated from those obtained in normal controls. Finally, our preliminary data (using PET, XCT, and MRI) in patients with CNS disorders indicate that MRI provides excellent delineation of the structural abnormalities. It may prove to be superior to XCT in the evaluation of certain diseases such as cerebral ischemia and infarcts, head injury, tumors, and white matter lesions. Metabolic imaging with FDG provides functional information not obtainable with either MRI or NMR spectroscopy. Therefore, PET studies will play a complementary role to the anatomic imaging in the management of patients with CNS disorders.