Abstracts

Rationale: It is estimated that 76% of individuals with anti-NMDA receptor encephalitis (anti-NMDARE) have seizures early in their clinical course (Dalmau J, Lancet Neurol 2008;7(12):1091-1098); however, the clinical approach to treatment of seizures is understudied. The primary aim of this study is to identify seizure phenotype and electroencephalographic (EEG) findings in anti-NMDARE patients to guide treatment strategies. We investigate the clinical features and timing of seizure management. We also sought to determine if specific electrographic features are associated with persistent seizures after acute presentation amongst patients with anti-NMDARE. Methods: We reviewed medical records of children and adults with anti-NMDARE between 2010 and 2014 who were part of The Rare Epilepsy of New York City (RENYC) database. This database includes medical records for patients with a visit associated with ICD9 code 345 (epilepsy), 779.0 (neonatal convulsions), or 780.39 (convulsions) from five academic medical centers in New York City from 2010 - 2014 (Weill Cornell Medicine, Columbia University Irving Medical Center, New York University Langone School of Medicine, Icahn School of Medicine at Mount Sinai, and Montefiore Medical Center). This database was queried for the search term “NMDA.” Medical records for identified patients were then reviewed to confirm definite diagnosis of anti-NMDARE according to consensus criteria (Graus F, Lancet Neurol 2016;15(4):391-404). Information regarding clinical seizures and electrographic features were extracted from chart review. EEG reports were reviewed to assess background organization, generalized or focal slowing, epileptiform activity, seizures, and sleep architecture. Patient cohorts were divided into those with persistent seizures (ongoing seizures after two weeks from date of presentation) and those with acute seizures only (i.e., seizures ceased within two weeks of clinical presentation). The data were then collated and analyzed using STATA (STATACorp) and association tested using Fisher’s exact T-test and Wilcoxon Rank sum. Results: Thirty-eight patients had definite anti-NMDARE. Of these patients, 32 (84%) had electroclinical or electrographic seizures, typically occurring a median of four days after initial symptoms (IQR 3-6 days) (Table 1). Frontal lobe-onset focal seizures with impaired awareness were the most common. Median time from hospital presentation to seizure cessation was 9 days (IQR 1.5-51 days). 14/38 (36.8%) patients had electro-clinical status epilepticus. Treatment involved a median of 3 (IQR 2-4) anti-epileptic seizure drugs (AEDs), most commonly levetiracetam, and continuous infusions (midazolam, ketamine, propofol, pentobarbital) were administered to 18/38 (47%) subjects. EEG was obtained in 37/38 (97%), of which 35 included simultaneous video. The first EEG was significantly different from subsequent EEGs in terms of frequency amplitude gradient, epileptiform activity, sleep features, and seizures captured (Table 2). No particular electrographic features were associated with persistence of seizures after the initial 2-week period. Conclusions: In our patient cohort, the seizures associated with anti-NMDARE are characterized primarily by focal seizures originating in the frontal lobes and frequently require treatment with multiple AEDs. Repeated EEGs are warranted given that the first EEG often does not capture the full extent of electrographic pathology. Most patients no longer had seizures after 9 days, thus further investigation as to the ideal duration for AED therapy is needed. Funding: NIH-NSADA K12 NS066274