Abstracts

Rationale: Automatisms are a common semiology of focal epilepsies yet their origin remains a mystery. Early literature suggests that focal mesial temporal or neocortical structures alone are not directly responsible for generating automatisms and theorizes that large-area cortical involvement and/or subcortical propagation drive these clinical features. Previously, our group documented a variable relationship between the electrographic termination time of secondarily generalized seizures with both associated motor activity and peri-ictal desaturation recovery. These findings suggest that abnormal ictal processes occur after ostensible electrographic seizure termination, and that these phenomena are not the direct result of the apparent cortical ictal activity.  We evaluated this hypothesis more directly by evaluating automatisms during focal seizures with impaired awareness. Methods: We retrospectively reviewed EEG recordings of patients with pharmacoresistent epilepsy who underwent intracranial EEG monitoring with subdural grids, strips and/or depth electrodes, as part of their surgical evaluation between 2005 and 2015 at Columbia University Medical Center Epilepsy Monitoring Unit (EMU).  Patients met inclusion criteria if they were ³ 16-years-old and had at least one focal onset seizure with impaired awareness recorded on video and EEG. We included a maximum of 5 seizures per patient. Automatism and electrographic seizure onset and offset times were identified blinded to EEG by consensus of three epilepsy physicians. Seizures were further analyzed for the presence of an asynchronous offset and stuttering quality. We report on 7 cases of automatisms seen during the “post-ictal” period of secondarily generalized seizures.  Results: We analyzed 30 focal onset seizures with impaired awareness from 11 patients (7 women) with an average age of 32-years-old (16-61). Automatisms have a significantly different offset relative to the electrographic seizure offset with a mean termination time 12.2 seconds after electrographic seizure termination (SD 32.7s, p=0.0266).  A majority of automatisms (70%) terminated after electrographic seizure offset, and most patients (91%) had at least one automatism that terminated ³ 10 seconds before or after electrographic seizure offset.  No significant correlation, R2 = 0.02, was found between Engel outcome and the differential between automatism offset and electrographic seizure offset. Stuttering automatisms were seen in 10 out of 11 (91%) patients and 25 out of 30 (83%) seizures (see picture). Asynchronous seizure offsets were seen in 4 out of 10 (40%) patients and 9 out of 25 (36%) seizures.  Conclusions: Electrographic seizure termination does not accurately predict the cessation of automatisms, and a majority of automatisms persist after electrographic seizure termination.  Automatisms may be seen during the “post-ictal” period of secondarily generalized seizures. Asynchronous electrographic seizure offsets are common and underappreciated due to sampling limitations.  Persistent automatisms may be directly related to ongoing ictal activity not sampled.   Funding: not applicable