Abstracts

Rationale: [18F]fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) is useful to identify epileptogenic focus as focal hypometabolism. However, FDG uptake in the brain depends on the level of epileptic activities. Frequent epileptiform spikes or ictal discharges increases metabolism in the epileptogenic focus, which may mask the interictal hypometabolism and make the interpretation difficult. FDG-PET in pediatric epilepsy shows less detectability of the epileptogenic focus as hypometabolism due to frequent seizures and hyperactivity of the epileptogenic focus. Metabolic changes on FDG-PET is also seen in the areas remote from the epileptogenic focus. Ictal hypermetabolism or hyperperfusion in the cerebellar vermis and the red nucleus has been reported in some cases1. This study was designed to evaluate association between FDG accumulation in the vermis and the red nucleus and detectability of the epileptogenic focus as hypometabolism on FDG-PET in pediatric epilepsy. Methods: Thirty-three children who underwent epilepsy surgery in our institution from July 2016 to December 2018 were enrolled in this study. All of them were age below 10 and had an epileptogenic lesion unilaterally. The cases were classified into hypometabolism or non-hypometabolism groups according to the metabolic state of the epileptogenic focus. Hypermetabolism in the vermis and metabolic laterality of the red nucleus were visually assessed and compared between these 2 groups. The maximum standardized uptake value (SUVmax) was measured and asymmetry index of the red nucleus and vermis-basal ganglia index ((SUVmax of the vermis) - (SUVmax of the contralateral basal ganglia)) / ((SUVmax of the vermis) + ((SUVmax of the contralateral basal ganglia)) * 200) were calculated. Receiver operating characteristic (ROC) curve analysis for detecting FDG-PET showing non-hypometabolism of the epileptogenic focus was performed for asymmetry index of the red nucleus and vermis-basal ganglia index. Results: Twenty-two cases (13 women, 9 men; mean age 2.1+-1.8 years) and 11 cases (6 women, 5 men; mean age 0.9+-1.2 years) were classified into hypometabolism and non-hypometabolism groups, respectively. In the non-hypometabolism group, age at surgery（P < 0.01）and age of seizure onset (P ＝ 0.037) were lower significantly and seizure frequency was tend to be higher (daily seizures in 90% and 64% cases of non-hypometabolism and hypometabolism groups, respectively) than those in the hypometabolism group. Hypermetabolism in the vermis was found frequently (67%) in the non-hypometabolism group but not (0%) in the hypometabolism group. Ipsilateral hypermetabolism of the red nucleus was seen more frequently in non-hypometabolism group (36%) than hypometabolism group (14%). The ROC curve for vermis-basal ganglia index demonstrates AUC of 0.793, with standard error of 0.081 (95% CI: 0.635, 0.952, P < 0.01) and for asymmetry index of the red nucleus demonstrates AUC of 0.603, with standard error of 0.105 (95% CI: 0.395, 0.812, P = 0.34). Conclusions: Hypermetabolism in the vermis or the ipsilateral red nucleus is associated with normo- or hypermetabolism of the epileptogenic focus. These observations may suggest hyperactivity of the epileptogenic focus including periictal acquisition of FDG-PET.Reference1. Bilo L, Meo R, de Leva MF, et al. Thalamic activation and cortical deactivation during typical absence status monitored using [18F]FDG-PET: a case report. Seizure 2010;19:198-201  Funding: The Japan Society for the Promotionof Science (KAKENHI grant No. 18K15574).* *Money paid to the institution