Objective:To test ketamine infusion efficacy in the treatment of super-refractory status epilepticus (SRSE), we studied retrospectively SRSE patients who were treated with ketamine. Additionally, we studied the effect of high doses of ketamine on brain physiology as reflected by invasive multimodality monitoring (MMM).Methods:We studied a consecutive series of 68 SRSE patients who were admitted between 2009 and 2018, were treated with ketamine and monitored with scalp EEG. Eleven of these patients underwent MMM at the time of ketamine administration. We compared patients who had seizure cessation after ketamine initiation to those who did not.Results:Mean age was 53+/-18 years old, 46% of patients were female. Seizure burden decreased by at least 50% within 24 hours of starting ketamine in 55 (81%) patients, with complete cessation in 43 (63%). Average dose of ketamine infusion was 2.2+/-1.8 mg/kg/h, with median duration of 2 (1; 4) days. Average dose of midazolam was 1.0+/-0.8 mg/kg/h at the time of ketamine initiation and was started at a median of 0.4 (0.1; 1.0) days before ketamine. Using a generalized linear mixed effect model, ketamine was associated with stable mean arterial pressure (OR 1.39, 95% CI 1.38-1.40), and with decreased in vasopressor requirements over time. We found no effect on intracranial pressure, cerebral blood flow, and cerebral perfusion pressure.Conclusion:Ketamine treatment was associated with a decrease in seizure burden in patients with SRSE. Our data support the notion that high dose ketamine infusions are associated with decreased vasopressor requirements without increased intracranial pressure.Classification of Evidence:This study provides Class IV evidence that ketamine decreases seizures in patients with SRSE.

Objective:To better understand the heterogeneous population of patients with new-onset refractory status epilepticus (NORSE), we studied the most severe cases who presented with new-onset super refractory status epilepticus (NOSRSE).Methods:We report a retrospective case series of 26 adults admitted to the Columbia University Irving Medical Center Neurological Intensive Care Unit (NICU) from 2/2009-2/2016 with NOSRSE. We evaluated demographics, diagnostic studies, and treatment course. Outcomes were modified Rankin Score (mRS) at hospital discharge and most recent follow-up visit (minimum of 2 months post discharge), NICU and hospital length of stay, and long-term anti-epileptic drug (AED) use.Results:Of the 252 patients with refractory status epilepticus, 27/252 had NORSE and 26/27 of those had NOSRSE. Age was bimodally distributed with peaks at 27 and 63 years. The vast majority (96%) had an infectious and/or psychiatric prodrome. Etiology was cryptogenic in 73%, autoimmune in 19%, and infectious in 8%. Seven patients (27%) underwent brain biopsy, autopsy or both; 3 (12%) were diagnostic (herpes simplex encephalitis, candida encephalitis, and acute demyelinating encephalomyelitis). On discharge, 6 patients (23%) had good or fair outcome (mRS 0-3). Of the patients with long-term follow-up data (median 9 months, interquartile range 2-22 months), 12 patients (71%) had mRS 0-3.Conclusion:Among our cohort, nearly all patients with NORSE had NOSRSE. The majority were cryptogenic with few antibody-positive cases identified. Neuropathology was diagnostic in 12% of cases. Although only 23% of patients had good or fair outcome on discharge, 71% met these criteria at follow-up.

No abstract

No abstract

No abstract

Objective. Automatic detection of interictal epileptiform discharges (IEDs, short as ‘spikes’) from an epileptic brain can help predict seizure recurrence and support the diagnosis of epilepsy. Developing fast, reliable and robust detection methods for IEDs based on scalp or intracranial electroencephalogram (iEEG) may facilitate online seizure monitoring and closed-loop neurostimulation. Approach. We developed a new deep learning approach, which employs a long short-term memory network architecture (‘IEDnet’) and an auxiliary classifier generative adversarial network (AC-GAN), to train on both expert-annotated and augmented spike events from iEEG recordings of epilepsy patients. We validated our IEDnet with two real-world iEEG datasets, and compared IEDnet with the support vector machine (SVM) and random forest (RF) classifiers on their detection performances. Main results. IEDnet achieved excellent cross-validated detection performances in terms of both sensitivity and specificity, and outperformed SVM and RF. Synthetic spike samples augmented by AC-GAN further improved the detection performance. In addition, the performance of IEDnet was robust with respect to the sampling frequency and noise. Furthermore, we demonstrated the cross-institutional generalization ability of IEDnet while testing between two datasets. Significance. IEDnet achieves excellent detection performances in identifying interictal spikes. AC-GAN can produce augmented iEEG samples to improve supervised deep learning.

No abstract

Background and Aims: To examine whether Heart Rate Variability (HRV) measures can be used to detect Neurocardiogenic Injury (NCI).Methods: 326 consecutive admissions with aneurysmal subarachnoid hemorrhage (SAH) met criteria for the study. 56 of 326 subjects (17.2%) developed NCI which we defined by wall motion abnormality with ventricular dysfunction on transthoracic echocardiogram or cardiac troponin-I > 0.3 ng/mL without ECG evidence of coronary artery insufficiency. HRV measures (in time and frequency domains, as well as nonlinear technique of detrended fluctuation analysis) were calculated over the first 48 hours. We applied longitudinal multilevel linear regression to characterize the relationship of HRV measures with NCI and examine between group differences at baseline and over time.Results: There was decreased vagal activity in NCI subjects with a between group difference in Low/High Frequency Ratio (beta 3.42, SE 0.92, p=0.0002), with sympathovagal balance in favor of sympathetic nervous activity. All time-domain measures were decreased in SAH subjects with NCI. An ensemble machine learning approach translated these measures into a classification tool that demonstrated good discrimination using the area under the receiver operating characteristic