Abstracts

Rationale: Lateralized periodic discharges (LPDs, also known as PLEDs) are a common EEG pattern in critically ill patients and associated with an increased risk of seizures. We aimed to establish an association between LPD amplitude and the risk of seizure on individual level.

Methods: This is a retrospective study of 20 patients who had LPDs while undergoing continuous EEG(cEEG) monitoring. 10 of these patients had seizures on their cEEG. LPDs were marked manually in each of the 20 EEG record for a 60-second epoch every 10 minutes for 1 hour and repeated every 8 hours in CURRY software (Compumedics Neuroscan, Charlotte, NC, USA). A set of variables for each epoch were abstracted that characterized specific electrographic features of the LPDs. These include absolute amplitude, relative amplitude, frequency, prevalence, sharpness, polarity, number of phases and plus feature. Analysis was performed with a mixed effects model with primary outcome measure of number of seizures per 8-hour EEG epoch with fixed effects being variables of interest and random effect being subject number.

A total of 127, 8-hour epochs were available in the group of 10 patients who had seizures. Of those epochs, 23 contained seizures and 100 contained LPDs. A total of 55, 8-hour epochs were contained in the group of 10 patients with LPDs without seizures. Of those epochs, 39 contained LPDs and 0 contained seizures. A total of 139 epochs (all epochs containing LPDs) were used for the first analysis, and then to confirm the findings only in patients who had both LPDs and seizures, 100 epochs were used for the second analysis.

Results: Overall, subdural hemorrhage (n=4) and infection (n=4) were the most common causes of LPDs. Mean duration of cEEG monitoring in the seizure group was longer than in the control group (134 vs. 50 hours).

Comparison of LPD Amplitude and other electrographic features between epochs with seizures (N=23) and epochs without seizures (N=116) for all 20 patients: EEG epochs from all 20 patients (10 with seizures and 10 without seizures) were included in this analysis. Absolute amplitude (p=0.04 corrected) and relative amplitude (p=0.04 uncorrected) of LPDs was significantly higher in the epochs that had seizures when compared to epochs without seizures.

Comparison of LPD Amplitude and other electrographic features between epochs with seizures (N=23) and epochs without seizures (N=77) for 10 patients with seizures: EEG epochs from only 10 patients who had seizures were included in this analysis. Absolute amplitude of LPDs showed significant difference at the epoch comparison level within the seizure group (p=0.04 corrected).

Conclusions: Higher LPD amplitude is associated with increased risk of seizures on an individual patient level. A decreasing amplitude is suggestive of decreasing cortical irritability, decreasing seizure risk, and may in fact be suggestive of decreasing ictal like behavior of LPDs.

Funding: Please list any funding that was received in support of this abstract.: NIH funding: NIH-NNDS R01-1NS111022.