Background Continuity Distinguishes Radiological Brain Injury Severity in Post-cardiac Arrest Patients with Generalized Periodic Discharges
Abstract number :
1.252
Submission category :
3. Neurophysiology / 3B. ICU EEG
Year :
2024
Submission ID :
771
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Urs Fisch, MD, PhD – Brigham and Women’s Hospital, Harvard Medical School
Joseph Ho, * – Brigham and Women’s Hospital, Harvard Medical School
Benjamin Scirica, MD – Brigham and Women's Hospital, Harvard Medical School
Samuel Snider, MD – Brigham and Women’s Hospital, Harvard Medical School
Liangge Hsu, MD – Brigham and Women’s Hospital, Harvard Medical School
Jong Woo Lee, MD, PhD – Brigham and Women's Hospital
Rationale: Multimodal neuroprognostication is essential for predicting outcomes after cardiac arrest and the significance of its components remains under active investigation. Generalized periodic discharges (GPD) in the EEG occur in about 8%-27% of post-cardiac arrest patients1-3. Despite the high prevalence of GPD, their pathophysiology is not fully understood, and their specificity for poor outcomes is debated with some studies indicating an 80-100% mortality associated with GPD1-3. We investigated whether brain injury severity quantified by MRI in post-cardiac arrest patients with GPD varies depending on the EEG background continuity.
Methods: In this single-center retrospective study, we identified post-cardiac arrest patients older than 18 years who underwent EEG monitoring between 2016 and 2021 from a prospectively collected EEG database. Patients were included if they had EEG results describing GPD with continuous or nearly continuous background (≥90 continuity; cont. background) or discontinuous, burst-suppressed, or suppressed backgrounds (< 90% continuity; noncont. background), with available post-cardiac arrest MRI data. Brain injury was assessed using whole-brain quantitative apparent diffusion coefficient (ADC) mapping. A good outcome was defined as a cerebral performance score (CPC) of 1-2 at 6 months. Statistical analyses were performed using Wilcoxon rank sum exact tests and Chi-squared tests.
Results: 28 patients (8 females; median age 59 years, interquartile range [IQR] 49.3 – 56.5) met the inclusion criteria, with 9 patients in the cont. background group and 19 in the noncont. background group. Good vs. poor outcome was significantly associated with whole brain ADC values (median 1023, IQR 944 - 1048 vs. median 835, IQR 791 – 1003; p = 0.01) and cont. background activity (67% vs.11%; p = 0.009). Likewise, coma recovery to follow commands at discharge was significantly associated with whole brain ADC values (median 1067, IQR 948-1092 vs. median 829, IQR 790-982; p = 0.002) and cont. background activity (p = 0.002). Whole-brain ADC values were significantly lower in the noncont. background group compared to the cont. background group (median 841, IQR 798 – 986 vs. median 1009, IQR 931 – 1092; p = 0.042). Any GPD with triphasic morphology were more frequent in patients with cont. background compared to those with noncont. backgrounds (77% vs. 32%, p = 0.059).
Conclusions: Our findings suggest that in post-cardiac arrest patients with GPD, the severity of brain injury quantified by MRI significantly differs based on whether patients have continuous or noncontinuous EEG background activity. The association between GPD with triphasic morphology and EEG background continuity calls for further investigation. We conclude that EEG background activity and discharge morphology may be important co-determinants when evaluating GPD in the EEG of post-cardiac arrest patients.
References
1. Beretta S, et al., Neurology. 2018;91(23):e2153-e2162.
2. Renzel R, et al., Clin Neurophysiol.2017;128(1):147-152.
3. Ruijter BJ, et al., Epilepsia.2015;56(11):1845-54.
Funding: This work was supported by the Bangerter-Rhyner Foundation, Switzerland, and the Andre and Michel Bouriez Foundation, Switzerland.
Neurophysiology