Abstracts

Rationale: Postictal generalized EEG suppression (PGES) may be associated with increased risk of sudden unexpected death in epilepsy (SUDEP). Artifact often precludes determination of PGES. Postictal tonic electromyographic activity (PTEMG) follows some focal to bilateral tonic-clonic seizures (FBTCS). The seizure-related characteristics preceding PTEMG have not been investigated. Acute hypoxia and hypercapnia occur with FBTCS. In humans, acute hypoxia lowers the resting motor threshold and shortens the cortical silent period as determined by transcranial magnetic pulse stimulation. There is time-dependent hypoxia-induced increase in motor corticospinal excitability and brainstem motor nuclei are depolarized. These findings suggest that PTEMG may be related to the severity of seizure-related respiratory dysfunction.We sought to characterize features of seizures and associated respiratory disturbances with PTEMG recorded in the epilepsy monitoring unit. Methods: We reviewed 145 FBTC seizures in 66 consecutive patients. PTEMG was analyzed with a minimum sampling rate of 512 Hz and filter settings 5-200 Hz. For other analyses, filter settings were 1-70 Hz.  Seizure duration, duration of tonic seizure component, PGES duration, seizure-associated SpO2 change and end-tidal CO2 were analyzed. The two-sided Mann-Whitney test compared data in FBTCS with and without PTEMG. Results: Of the 145 FBTCS, 90 had subsequent PTEMG (Figures 1,2), 36 had postictal EEG slowing without PTEMG (Figure2D), and artifact precluded PTMG determination in 19 (13%). PGES was present in 45, absent in 53, and artifact precluded determination of PGES in 47 (32%). Of the seizures with PTEMG, PGES was discernible in 45 (50%), absent in 18 (20%), and indeterminate in 27 (30%) patients due to artifact. PGES was not present in any of the seizures without PTEMG.Mean PTEMG duration was 22.3±10.292 seconds. Mean peak EMG frequency was 74.9±30.88Hz and peak power 76.4±103.4µV². Cessation of PTEMG coincided with appearance of postictal EEG slowing that was not manifest during the presence of PTEMG (Figure 1,2).  Mean SpO2 nadir was lower (p=0.005) with seizures followed by PTEMG than in those without PTEMG (72% vs. 77 %). Peak end-tidal CO2 was higher (p=0.05) in patients with PTEMG compared to those without PTEMG (63 mm Hg vs. 56 mm Hg). Mean PGES duration was 35.6±22.2 seconds. On linear regression analysis, PTEMG duration was associated with PGES duration (p<0.001). There was no significant association between PTEMG and seizure duration or duration of tonic seizure component. Conclusions: Seizures associated with PTEMG have a greater severity of respiratory dysfunction than those without. The duration of PTEMG is associated with PGES duration. PTEMG is readily discerned by visual analysis of the data at appropriate filter settings and is less prone to obscuration by movement artifact compared to PGES. PTEMG may be a better biomarker than PGES for assessment of SUDEP risk. Funding: Department of Neurology, University of California, Davis