F84. Sensitivity of persyst seizure detection for different electrographic seizure patterns in patients with status epilepticus. (May 2018)

• Record Type: Journal Article
• Authors: Zorlu, Musab; Chuang, David; Kettani, Hind; Zarnegar, Reza
• Is Part Of: Clinical neurophysiology. Volume 129(2018)Supplement 1
• Page Start: e98
• Publication Date: 2018-05
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2018.04.247
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Introduction: With the rising use of continuous EEG monitoring, the ability to detect seizures rapidly and treat is becoming a standard of care. The issue for many institutions is the availability of an EEG reader to review EEG studies continuously. Persyst is a seizure detection software that allows health care professionals not trained in EEG to identify seizures rapidly at the bedside. We aim to determine if different electrographic patterns of status epilepticus (SE) affects detection rate by Persyst. Methods: Continuous EEG monitoring reports at New York Presbyterian Queens Hospital from January 1st, 2015 to December 31st, 2016 with the phrase "status epilepticus" were selected. EEG patterns were categorized by board certified epileptologists as (1) SE associated with repetitive focal seizures (SERFS) or (2) SE associated with the Ictal-Interictal continuum (SEIIC). SERFS were defined as focal origin with a clear ictal onset and offset and a definable inter-seizure interval. SEIIC were defined as having association with characteristic features in the ictal-interictal continuum such as periodic discharges (generalized, lateralized, bilateral independent or stimulus-induced) and/or rhythmic delta activity. The archived raw EEG for each case was analyzed using Persyst Software. If at least one seizure was detected, it was classified as positive. If there were zero seizure detections, then it was classified as negative. Cases which had insufficient raw EEG data available were excluded from this study. Statistical analysis was performed using Chi-Square test. Results: A total of 11 patients were identified with status epilepticus. Of the 11 patients, 5 were classified as SERFS and 6 were classified as SEIIC. Following analysis using Persyst, 5/5 patients with SERFS and 2/6 patients with SEIIC had positive seizure detections. Persyst was significantly more sensitive in detecting seizures in SERFS than in SIIC ( p = 0.02). Conclusion: This study showed that Persyst is highly sensitive in detecting SERFS but ineffective in SEIIC. These findings are consistent with prior study on this topic. This study, although small in sample size, further adds to the literature showing that the electrographic pattern of status epilepticus may influence the sensitivity of seizure detection using automated seizure detection software such as Persyst. Future studies with a larger sample size would be helpful to further clarify this hypothesis.