PURPOSE: The importance of trace elements in their effect on the physiology and pathology of the central nervous system is well recognized. Changes in the concentrations of these elements in the brain could take place in pathological states. Recently, a greater emphasis has been given to the role of trace elements in the function of the nervous system both in normal and pathological conditions. The past experiments from animal demonstrate that Na+-K+-ATPase inhibition, particularly in the hippocampus, is involved in epileptogenicity. Zinc is the most potent inhibitor of Na+-K+-ATPase followed closely by copper. Zinc modulates the activity of glutamic acid decarboxylase, the rate limiting enzyme in the synthesis of -aminobutyric acid (GABA), which is a major inhibitory neurotransmitter. There are few reports of zinc and copper concentrations in normal CSF and in CSF from patients with neurological diseases. The aim of this study was designed to determine the zinc and copper concentrations and their correlation with protein in CSF of pediatric patients with neurologic disorders. METHODS: The study population was 43 patients who had admitted to Kang Nam Sacred Heart Hospital of Hallym University from March to June, 1996 due to high fever, headache, vomiting, and seizure. All patients were examined CSF study, 32 patients (group I) were showed abnormal CSF and seizure disorders including febrile convulsion and 11 patients (group II) were showed normal CSF and clinical symptoms of febrile illness. Zinc and copper concentrations in CSF were determined with flame atomic absorption spectrophotometry. In addition, CSF zinc and copper concentrations in normal CSF proteingroup (group A) and in increased CSF protein group (group B) were determined to investigate probability that the damaged blood-brain-barrier permits the passage of zinc and copper into the subarachnoid space. RESULTS: 1) The CSF zinc concentrations in group I and II were 9.40+/-6.18 and 7.39+/-5.48microgram/dl, and the CSF copper concentrations in group I and II were 4.86+/-7.07 and 2.93+/-1.45microgram/dl, respectively. There was no statistically significant difference in the CSF zinc and copper concentrations between the two groups. 2) The CSF zinc concentrations in group A and B were 7.21+/-4.96 and 11.24+/-7.32microgram/ dl, and the copper concentrations in group A and B were 3.31+/-2.15 and 5.59+/-9.46microgram/dl, respectively. There was no statistically significant difference in the CSF zinc and copper concentrations between the two groups. 3) There was a significant positive correlation between the CSF zinc and copper concentrations as well as between the CSF zinc and protein concentrations. But there was no significant correlation between the CSF copper and protein concentrations. CONCLUSIONS: There was no statiscally significant defference in the CSF zinc and copper concentrations between neurologic disorders and febrile diseases. Increased CSF zinc and copper concentrations in increased CSF protein groups were not found. But there were some correlation between zinc, copper, protein levels in CSF. These results do not support assumption that damaged BBB permits the passage of the zinc, copper into the subarachnoid space.