episodes and 34 hours of interictal episodes. Correlation of the eight HRV features were monitored for predicting seizures by using multi- variate statistical process control (MSPC). Results: All of the fourteen seizures could be predicted at least one minute prior to the seizure onset. The false positive rate were much higher for the sleep period (3.9/hour) than for the awake period (0.27/hour). Most of the false positives were associated with body motion and alteration of sleep stage. The total false positive rate, whose cause could not be identified, was 0.53/hour. Conclusion: The proposed ECG-based seizure prediction method demonstrated competitive sensitivity with electroencephalography- based methods. We are presently developing a mobile seizure prediction system based on a wearable RRI sensor and prediction algorithm on smartphone application. The system may be useful for closed-loop seizure treatment system. doi:10.1016/j.jns.2017.08.3769 1534 WCN17-2393 SHIFT 4 - EPILEPSY Chronic kidney disease and Electroencephalogram (EEG) abnormalities: What factors lead to poor patient prognosis? M. Monif a , U. Seneviratne b . a Level 4, Department of Neurology, Royal Melbourne Hospital, Parkville- 3050, Vic, Neurology, 3050, Australia; b Department of Neuroscience, Monash Medical Center, Neurosciences, 246 Clayton Rd- Clayton, Victoria, 3268, Australia Background: Chronic kidney disease (CKD) is a leading cause of morbidity and mortality worldwide. Individuals with CKD are at increased risk of metabolic encephalopathy and seizures. Objective: We report a retrospective analysis of 44 consecutive patients with CKD referred for an EEG due to altered conscious state. Patients and Methods/Material and Methods: We performed unadjusted logistic regression to test the association of predictors (age at EEG, eGFR, serum urea and EEG category- encephalopathic patterns: type 1; seizure, rhythmic or periodic patterns: type 2) with the outcome. The outcome was defined as ‘favourable’ (discharge to home or rehabilitation) versus ‘unfavourable’ (death, palliation, or discharge to nursing home). Results: The average age of our cohort was 62.06 ±16.3 years (55% male, 45% female). The most common aetiology of CKD was diabetes (56.8%) followed by glomerulonephritis (22.7%). The average creatinine was 500±325.76 umol/L, and average urea of 17.42 ± 8.76 mmol/L. In 65% of individuals the final diagnosis was encephalopathy (n=26), followed by non-convulsive status epilepticus (NCSE) (15%, n=6). Notably, 27% of patients had an ‘unfavourable outcome’. Only age and EEG category were signifi- cantly associated with poor outcomes. With increasing age, the odds of unfavourable outcome increased (p=0.027; OR = 1.07, 95% CI 1.01 to 1.15). Type 2 EEG category was associated with increasing odds of an unfavourable outcome (p=0.01; OR 9.5 (1.64-54.99)). Conclusion: Our study reveals that older age and epileptiform/ rhythmic/periodic EEG patterns are associated with a poor prognosis irrespective of the degree of renal impairment. Altered conscious state in patients with CKD is due to NCSE in a fair proportion. doi:10.1016/j.jns.2017.08.3770 1535 WCN17-1195 SHIFT 4 - EPILEPSY The relationship between neuropsychological functioning and mesial temporal lobe volumes determined by mri volumetry in hippocampal sclerosis international league against epilepsy type1 T. Mukaino a , T. Uehara a , K. Hashiguchi b , S.O. Suzuki c , J. Kira a . a Kyushu University, Neurology, Fukuoka, Japan; b Kyushu University, Neurosurgery, Fukuoka, Japan; c Kyushu University, Basic Medicine, Fukuoka, Japan Background: The international consensus classification for hippo- campal sclerosis (HS) has been proposed in 2013, but no research has reported the relationship between neuropsychological function- ing and brain structural changes in patients with pathologically confirmed HS according to the classification. Objective: To investigate the association between neuropsycholog- ical measures and volumes of mesial temporal lobe structures in patients with unilateral mesial temporal lobe epilepsy (MTLE) with HS International League Against Epilepsy (ILAE) type 1. Patients and Methods/Material and Methods: The medical records of patients who underwent epilepsy surgery between 2007 and 2016 were retrospectively reviewed. Twenty-one consecutive unilateral MTLE patients confirmed with HS ILAE type 1 were selected for analysis. We used clinical data, neuropsychological indexes (WAIS-III, WMS-R) and 3D-volume T1-weighted magnetic resonance imaging obtained for presurgical evaluation. Volumetry of hippocampus, subiculum and entorhinal cortices was performed with FreeSurfer (ver6.0). Results: There were 12 patients with left HS (LHS) and 9 with right HS (RHS). As a group, patients with LHS had a longer disease duration (p b 0.05) than RHS. Volumes of mesial temporal lobe structures were significantly smaller on the side ipsilateral to seizure focus, except for left subiculum and left entorhinal cortex. Combined volume of left and right hippocampi were significantly smaller in patients with RHS. Stepwise multiple regression analysis indicated there is a significant relationship between combined bilateral hippocampal volumes and visual memory (p=0.017). Left subiculum volumes were also significantly related to performance IQ (p=0.04). Conclusion: Volume changes of mesial temporal lobe structures are good markers of neuropsychological functioning in MTLE with HS ILAE type 1. doi:10.1016/j.jns.2017.08.3771 1536 WCN17-1292 SHIFT 4 - EPILEPSY Invasive 3D source localization by wide-band electroencephalography findings T. Murai a , M. Matsuhashi b , T. Hitomi c , K. Kobayashi a , A. Shimotake d , M. Inouchi e , T. Kikuchi f , K. Yoshida f , T. Kunieda g , R. Matsumoto a , R. Takahashi a , A. Ikeda d . a Kyoto University Graduate School of Medicine, Neurology, Kyoto, Japan; b Kyoto University Graduate School of Medicine, Research and Educational Unit of Leaders for Integrated Medical system- Human Brain Research Center, Kyoto, Japan; c Kyoto University Graduate School of Medicine, Clinical Laboratory Medicine, Kyoto, Japan; d Kyoto University Graduate School of Medicine, Epilepsy- Movement disorders and Physiology, Kyoto, Japan; e Kyoto University Graduate School of Medicine, Respiratory Care and Sleep Control Medicine, Kyoto, Japan; f Kyoto University Graduate School of Medicine, Neurosurgery, Kyoto, Japan; g Ehime Unversity, Neurosurgery, Matsuyama, Japan Abstracts / Journal of the Neurological Sciences 381 (2017) 374–560 555