Abstracts

Can Diffusion Weighted Imaging (DWI) be used as a marker for LPDs associated with seizures on continuous electroencephalography?

Abstract number : 43
Submission category : 3. Neurophysiology / 3B. ICU EEG
Year : 2020
Submission ID : 2422392
Source : www.aesnet.org
Presentation date : 12/5/2020 9:07:12 AM
Published date : Nov 21, 2020, 02:24 AM

Authors :
Ahmad Yusuf-Solaiman, University of Michigan; Alaa Seddika - Damascus University; Temenuzhka Mihaylova - University of Michigan;;


Rationale:
Ictal-interictal continuum periodic patterns are encountered with increasing frequency as continuous video EEG monitoring (cEEG) is adopted more widely for seizure detection and prognostication. Lateralized periodic discharges (LPDs) in particular may independently impact outcome, and there is mounting evidence that their presence exacerbates underlying acute brain injury and may indicate poor prognosis. Acute MRI changes due to seizures often suggest combinations of cytotoxic and vasogenic edema, but it is unclear why only certain patients have shown MRI changes. There is paucity of literature on the relationship between DWI changes and LPDs and it is unclear which patients with LPDs will have associated seizures.
Method:
This was a retrospective chart review analysis, to investigate whether critically ill patients with LPDs on cEEG have corresponding DWI MR changes, and if such association can serve as a marker for increased risk of seizures. We reviewed records for demographics, medical history, presence and location of LPDs, presence and location of DWI changes, and development of seizures in all patients who were monitored with cEEG at MMHS from 8/1/2009 to 7/1/2019. The cEEG monitoring was with mean 4 days duration. We included patients with and without structural brain abnormalities who had neuroimaging during the same admission.
Results:
Out initial chart search using search words “LPDs and EEG” identified a total of 431 patients from whom 208 patients were excluded from final analysis due to database search errors (duplicate reports, wrong diagnostic codes, routine EEGs). From the 223 patients (F/M: 106/117 (47.5% F: 52.5% M, mean age 59 years), who were identified to have LPDs on cEEG, only 136 patients had brain MRI during the same admission. DWI abnormalities were seen in 47 patients (47/136 or 34.5 %). In this group, the LPDs and DWI changes were co-localized in 37 patients (78.7%). We found a structural abnormality corresponding to the LPD location in 11/47 (23.4%).   Only 27 patients (57%) who had DWI changes were found to have seizures and 24 of them (88.8%) had co-localization of LPDs and DWI changes with seizure onset.  20/47 with DWI changes did not have seizures (43%).
Conclusion:
Our retrospective chart review showed that 34.5% of the patients who had LPDs on cEEG also had DWI changes, and in the majority of them, 78.7%, the LPDs and the MR changes were co-localized. 57% of the patients with DWI changes had seizures and in the majority of them, 88.8%, the seizure onset was in the same location as the LPDs and DWI changes. These preliminary data sets indicate that restricted diffusion changes on brain MRI can be used as a marker to identify critically ill patients with LPDs who are likely to have associated seizures. Utilizing this information to guide physicians on which patients need to be treated more aggressively could have a significant impact on clinical outcomes.
Funding:
:None
FIGURES
Figure 1
Neurophysiology