Summary: Purpose:We sought to establish a noninvasive method for focus diagnosis of epilepsy.Methods: We examined the use of multichannel nearinfrared spectroscopy (NIRS). It is known as a noninvasive method of functional brain mapping. We monitored cerebral blood volume change with NIRS during long-term EEG monitoring of epilepsy in 32 cases with intractable epilepsies to diagnose the epileptogenic focus.Results: In 96% of cases, NIRS showed significant hyperperfusion in the side of seizure foci, whereas ictal SPECT showed hyperperfusion in 69% of cases.Conclusions: Ictal NIRS is a reliable method to evaluate the focus side in epilepsy, especially when it is coupled with ictal SPECT.

Summary:  Purpose: For the past 20 years (1978–1997), a series of 100 cases of uncontrolled epilepsy had surgery in our department under the stated standard for surgical indications and were followed up for 2–22 years after surgery. Methods: We evaluated 70 cases of temporal lobectomy, 20 cases of neocortical focal resection, and 10 cases of corpus callosotomy. Results: Analysis of postoperative seizure control showed that 78 cases were class 1 or 2 (no or rare seizures), 14 cases were class 3 (worthwhile improvement), and eight cases were class 4 (no improvement). As generally accepted, temporal lobectomy was the most effective operative procedure, yielding excellent or good results in 87% of the 70 cases so treated. Among the nine cases in whom various psychiatric symptoms developed after surgery, four cases showed neurotic and five cases psychotic symptoms. Patients with psychosis had delusions of various types as a core symptom, combined with other symptoms such as anxiety, irritability, aggression, and depressive state. In two patients with psychosis who had episodes of delusions in the interictal phase before surgery, the symptoms were extremely resistant. The full IQ score of the Wechsler Adult Intelligence Scale–Revised (WAIS‐R) was increased after temporal lobectomy in 75% of the cases (p < 0.01; n = 44). The general MQ score in 31 cases, however, showed a 50:50 split between increase and decrease postoperatively. In correlation with the dominance of language by the Wada test, the general MQ score in the 15 cases of nondominant temporal lobe resection showed a significant increase (p < 0.05); whereas the MQ in the 16 cases of dominant‐side operation did not change. Conclusions: Our new test, in which hippocampal stimulation and supraspan learning are combined, seems to be efficient for estimating the postoperative outcome of memory function.

PLATFORM AND POSTER SESSIONS 57 with a chromosomal dcletion. Thcse non-deletion AS patients are caused by paternal uniparental disomy (UPD), imprinting mutation (IM), or loss-or-function mutations of the UBE3A gene, cach of which predisposes different recurrence risk. To elucidate molecular etiology of non-dclction AS patients, we investigated 34 AS patients without a chromosomal deletion.Me/hod.s: Thirty sporadic AS patients, and 4 familial AS patients (2 families of 2 sibs) were enrolled to the study. The diagnosis of AS was based on Williams' criteria (Williams et al., Am J Med Genet 1995, 56: 237). Genomic DNA was extracted from peripheral blood by a standard procedure. DNA mcthylation tcst at SNRPN locus and genotyping using 7 highly informative PCR-based polymorphisms within 15q I Iq I3 were carried out to identify UPD and IM. When both UPD and IM were ruled out, the patients were classified :LS non-UPD, non-IM. For thcsc non-UPD, non-1M paticnts, UBE3A mutations were screened by PCR-SSCP analysis using 10 sets ofprimcrs covering all coding exons.Results: Among 30 sporadic patients, I UPD and 3 IM patients were identified, and the remaining 26 patients were classified as non-UPD, non-IM. Among 4 familial patients, 2 sibs from I family were detected as IM, whcrcas 2 sibs from another family were classified as non-UPD, non-IM. No UBE3A mutations were identified within 26 sporadic and 2 familial non-UPD, non-IM patients. Conclusion: Threc molecular classes were identified for noiideletion AS patients. Therefore, the underlying genetic mechanism was dcmonstratcd to be complex for AS patients without a chromosomal deletion. Combination of the DNA methylation test and PCR-based polymorphisms was sufficient to detect UPD and IM patients. Because recurrence risk is low for UPD and high lor IM, systematic molecular investigation including the DNA methylation test and PCR-based polymorphisms should bc donc for non-delction AS paticnts for genetic counscling purpose. A majority of non-deletion patients were classified as noii-UPD, non-1M. Although, approximate 30 % of non-UPD, non-

Summary:Purpose: For the past 20 years (1978)(1979)(1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997), a series of 100 cases of uncontrolled epilepsy had surgery in our department under the stated standard for surgical indications and were followed up for 2-22 years after surgery.Methods: We evaluated 70 cases of temporal lobectomy, 20 cases of neocortical focal resection, and 10 cases of corpus callosotomy.Results: Analysis of postoperative seizure control showed that 78 cases were class 1 or 2 (no or rare seizures), 14 cases were class 3 (worthwhile improvement), and eight cases were class 4 (no improvement). As generally accepted, temporal lobectomy was the most effective operative procedure, yielding excellent or good results in 87% of the 70 cases so treated. Among the nine cases in whom various psychiatric symptoms developed after surgery, four cases showed neurotic and five cases psychotic symptoms. Patients with psychosis had delusions of various types as a core symptom, combined with other symptoms such as anxiety, irritability, aggression, and depressive state. In two patients with psychosis who had episodes of delusions in the interictal phase before surgery, the symptoms were extremely resistant. The full IQ score of the Wechsler Adult Intelligence Scale-Revised (WAIS-R) was increased after temporal lobectomy in 75% of the cases (p < 0.01; n ‫ס‬ 44). The general MQ score in 31 cases, however, showed a 50:50 split between increase and decrease postoperatively. In correlation with the dominance of language by the Wada test, the general MQ score in the 15 cases of nondominant temporal lobe resection showed a significant increase (p < 0.05); whereas the MQ in the 16 cases of dominant-side operation did not change.Conclusions: Our new test, in which hippocampal stimulation and supraspan learning are combined, seems to be efficient for estimating the postoperative outcome of memory function.