Faculty Bibliography

Epilepsy surgery is gaining popularity for the treatment of children with intractable seizures in whom either a focal or extensive unilateral structural brain lesion is demonstrated. We evaluated the pre- and postoperative imaging findings in 29 patients (aged 22 days to 19 years) who underwent hemispherectomies, 12 total and 17 subtotal. Pathologic correlation was obtained in all cases. Preoperatively, positron emission tomography and electroencephalography demonstrated abnormalities in all of the 28 children studied, but frequently could not characterize the lesion. CT or MR or both demonstrated focal or unilateral lesions in only 19 of these but gave additional information regarding the nature of the lesion. Preoperative angiographic findings were abnormal in five of 17 patients studied and were particularly useful in the evaluation of the extent of abnormality in patients with Sturge-Weber syndrome. Postoperatively, CT and MR demonstrated early complications such as the development of epidural blood and fluid collections, parenchymal hemorrhage, infection, and early hydrocephalus. Postoperatively, MR demonstrated the early development of septations, the presence of subarachnoid hemorrhage, and/or the deposition of hemosiderin in four patients, findings that historically have been associated with the development of devastating clinical complications. From these data, a recommended protocol of radiologic evaluation for patients undergoing hemispherectomy has been established.

Twenty-three children with 4 clinical subtypes of cerebral palsy were studied using 2-deoxy-2(18F)fluoro-D-glucose (FDG) and positron emission tomography (PET). Subtypes included spastic quadriparesis (N = 6), spastic diplegia (N = 4), infantile hemiplegia (N = 8), and choreoathetosis (N = 5). FDG-PET images were correlated with magnetic resonance imaging or computed tomography. Although the location of glucose metabolic abnormalities corresponded, in general, to abnormalities of brain structure demonstrated by structural imaging studies, the distribution of metabolic impairment almost invariably extended beyond the region of anatomic involvement. The following observations in specific subtypes of cerebral palsy were determined with FDG-PET: (1) In spastic diplegic patients, PET revealed focal areas of cortical hypometabolism in the absence of apparent structural abnormality; (2) A relatively normal pattern of cortical metabolism was observed in most patients with choreoathetoid cerebral palsy, despite marked hypometabolism in the thalamus and lenticular nuclei; and (3) In patients with infantile hemiplegia, FDG-PET disclosed symmetric cerebellar glucose metabolism with absence of crossed cerebellar hypometabolism (diaschisis). This finding is contrary to the typical persistence of crossed cerebellar diaschisis in adult patients with acquired cerebral lesions and suggests metabolic recovery due to developmental plasticity. The possibility that FDG-PET may be clinically useful in identifying the cerebral palsy patient with potential learning handicap and in the study of functional recovery or sparing following brain injury should be explored further.

Previously, using positron emission tomography (PET), we showed that local cerebral metabolic rates for glucose (lCMRglc) in children undergo dynamic maturational trends before reaching adult values. In order to develop an animal model that can be used to explore the biological significance of the different segments of the lCMRglc maturational curve, we measured lCMRglc in kittens at various stages of postnatal development and in adult cats using quantitative [14C]2-deoxyglucose autoradiography. In the kitten, very low lCMRglc levels (0.14 to 0.53 mumol min-1 g-1) were seen during the first 15 days of life, with phylogenetically older brain regions being generally more metabolically mature than newer structures. After 15 days of age, many brain regions (particularly telencephalic structures) underwent sharp increases of lCMRglc to reach, or exceed, adult rates by 60 days. This developmental period (15 to 60 days) corresponds to the time of rapid synaptic proliferation known to occur in the cat. At 90 and 120 days, a slight decline in lCMRglc was observed, but this was followed by a second, larger peak occurring at about 180 days, when sexual maturation occurs in the cat. Only after 180 days did lCMRglc decrease to reach final adult values (0.21 to 2.04 mumol min-1 g-1). In general, there was good correlation between the metabolic maturation of various neuroanatomical regions and the emergence of behaviors mediated by the specific region. At least in the kitten visual cortex, which has been extensively studied with respect to developmental plasticity, the "critical period" corresponded to that portion of the lCMRglc maturational curve surrounding the 60-day metabolic peak. These normal maturational lCMRglc data will serve as baseline values with which to compare anatomical and metabolic plasticity changes induced by age-related lesions in the cat.