Optimizing Management of Medically Responsive Epilepsy

State of the Art and Challenges in Epilepsy—A Narrative Review

Epilepsy is a common condition worldwide, with approximately 50 million people suffering from it. A single seizure does not mean epilepsy; almost 10% of the population can have a seizure during their lifetime. In particular, there are many other central nervous system disorders other than epilepsy in which seizures occur, either transiently or as a comorbid condition. The impact of seizures and epilepsy is, therefore, widespread and easily underestimated. It is estimated that about 70% of patients with epilepsy could be seizure-free if correctly diagnosed and treated. However, for patients with epilepsy, quality of life is influenced not only by seizure control but also by antiepileptic drug-adverse reactions, access to education, mood, employment, and transportation.

Expert-Level Intracranial Electroencephalogram Ictal Pattern Detection by a Deep Learning Neural Network

Background: Decision-making in epilepsy surgery is strongly connected to the interpretation of the intracranial EEG (iEEG). Although deep learning approaches have demonstrated efficiency in processing extracranial EEG, few studies have addressed iEEG seizure detection, in part due to the small number of seizures per patient typically available from intracranial investigations. This study aims to evaluate the efficiency of deep learning methodology in detecting iEEG seizures using a large dataset of ictal patterns collected from epilepsy patients implanted with a responsive neurostimulation system (RNS). Methods: Five thousand two hundred and twenty-six ictal events were collected from 22 patients implanted with RNS. A convolutional neural network (CNN) architecture was created to provide personalized seizure annotations for each patient. Accuracy of seizure identification was tested in two scenarios: patients with seizures occurring following a period of chronic recording (scenario 1) and patients with seizures occurring immediately following implantation (scenario 2). The accuracy of the CNN in identifying RNS-recorded iEEG ictal patterns was evaluated against human neurophysiology expertise. Statistical performance was assessed via the area-under-precision-recall curve (AUPRC). Results: In scenario 1, the CNN achieved a maximum mean binary classification AUPRC of 0.84 ± 0.19 (95%CI, 0.72-0.93) and mean regression accuracy of 6.3 ± 1.0 s (95%CI, 4.3-8.5 s) at 30 seed samples. In scenario 2, maximum mean AUPRC was 0.80 ± 0.19 (95%CI, 0.68-0.91) and mean regression accuracy was 6.3 ± 0.9 s (95%CI, 4.8-8.3 s) at 20 seed samples. We obtained near-maximum accuracies at seed size of 10 in both scenarios. CNN classification failures can be explained by ictal electro-decrements, brief seizures, single-channel ictal patterns, highly concentrated interictal activity, changes in the sleep-wake cycle, and progressive modulation of electrographic ictal features. Conclusions: We developed a deep learning neural network that performs personalized detection of RNS-derived ictal patterns with expert-level accuracy. These results suggest the potential for automated techniques to significantly improve the management of closed-loop brain stimulation, including during the initial period of recording when the device is otherwise naïve to a given patient's seizures.

Adverse Skin Reactions with Antiepileptic Drugs Using Korea Adverse Event Reporting System Database, 2008-2017

BACKGROUND

Severe and life-threatening drug eruptions include drug reaction with eosinophilia and systemic symptoms (DRESS), Stevens-Johnson syndrome (SJS), and toxic epidermal necrolysis (TEN). One class of medications that has been highly associated with such drug eruptions is antiepileptic drugs (AEDs). We attempt to investigate drug eruptions associated with AEDs as a class, as well as with individual AEDs, in Korea.

METHODS

We used the Korea Institute of Drug Safety and Risk Management - Korea Adverse Event Reporting System (KIDS-KAERS) database, a nationwide database of adverse events reports, between January 2008 and December 2017 to investigate the reporting count of all drug eruptions and calculated the ratio of DRESS/SJS/TEN reports for each AED.

RESULTS

Among a total of 2,942 reports, most were of rash/urticaria (2,702, 91.8%), followed by those of DRESS (109, 3.7%), SJS (106, 3.6%), and TEN (25, 0.85%). The common causative AEDs were lamotrigine (699, 23.8%), valproic acid (677, 23%), carbamazepine (512, 17.4%), oxcarbazepine (320, 10.9%), levetiracetam (181, 6.2%), and phenytoin (158, 5.4%). In limited to severe drug eruptions (DRESS, SJS, and TEN; total 241 reports), the causative AEDs were carbamazepine (117, 48.8%), lamotrigine (57, 23.8%), valproic acid (20, 8.3%), phenytoin (15, 6.3%), and oxcarbazepine (10, 4.2%). When comparing aromatic AED with non-aromatic AED, aromatic AEDs were more likely to be associated with severe drug eruption (aromatic AEDs: 204/1,793 versus non-aromatic AEDs: 37/1,149; OR, 3.86; 95% CI, 2.7-5.5). Death was reported in 7 cases; DRESS was the most commonly reported adverse event (n = 5), and lamotrigine was the most common causative AED (n = 5).

CONCLUSION

Although most cutaneous drug eruptions in this study were rash or urticaria, approximately 8% of reports were of severe or life-threatening adverse drug reactions, such as SJS, TEN, or DRESS. When hypersensitivity skin reactions occurred, aromatic AEDs were associated with 4 fold the risk of SJS/TEN/DRESS compared with non-aromatic AEDs. Our findings further emphasize that high risk AEDs should be prescribed under careful monitoring, and early detection and prompt interventions are needed to prevent severe complications.