Autonomic Biomarkers Distinguish Epileptic versus Psychogenic Non-epileptic Seizures
Abstract number :
1.116
Submission category :
2. Translational Research / 2C. Biomarkers
Year :
2022
Submission ID :
2203979
Source :
www.aesnet.org
Presentation date :
12/3/2022 12:00:00 PM
Published date :
Nov 22, 2022, 05:22 AM
Authors :
Justin Ryan, PhD – SUNY Upstate Medical University; Kyle Wagner, BS – SUNY Upstate Medical University; Sushma Yerram, MD – University of Rochester Medical Center; Cathleen Concannon, MPH – University of Rochester Medical Center; Jennifer Lin, BS – University of Rochester Medical Center; Patrick Rooney, MD – University of Rochester Medical Center; Brian Hanrahan, MD – University of Rochester Medical Center; Victoria Titoff, MD – SUNY Upstate Medical University; Noreen Connolly, JD, MS, CCRC – University of Rochester Medical Center; Ramona Cranmer, BS – University of Rochester Medical Center; Natalia DeMaria, MS – SUNY Upstate Medical University; Xiajuan Xia, PhD – University of Rochester Medical Center; Betty Mykins, BS – University of Rochester Medical Center; Steven Erickson, BS – University of Rochester Medical Center; Jean-Phillippe Couderc, PhD, MBA – University of Rochester Medical Center; Giovanni Schiffito, MD – University of Rochester Medical Center; Dongliang Wang, PhD – SUNY Upstate Medical University; Giuseppe Erba, MD – University of Rochester Medical Center; David Auerbach, PhD – SUNY Upstate Medical University
Rationale: Thirty to 40% of patients whose seizures are not controlled by anti-seizure medications exhibit manifestations comparable to epileptic seizures (ES), but there are no EEG correlates, and thus are called psychogenic non-epileptic seizures (PNES). Due to limited access to EEG-monitoring and often inconclusive results, a non-invasive diagnostic tool is needed to optimize the clinical management. We evaluated the temporal evolution of ECG-based measures of autonomic function (heart rate, HR, and variability, HRV) to determine whether they predict and distinguish ES vs. PNES events.
Methods: This prospective cohort study includes consecutive patients admitted to the University of Rochester Epilepsy Monitoring Unit. Patients are 18 to 65 years, without therapies or co-morbidities associated with altered autonomic function. A typical ES or PNES event is recorded during admission. ECG analysts are blinded to ES/PNES diagnosis, and perform time, frequency, and nonlinear domain HRV analyses during specific non-seizure physiological states, and peri-ictally (150-minutes each pre-/post-ictally, overlapping 5-minute epochs, 1-minute sliding windows). Using logistic regression modeling of HR and HRV measures, we determined the sensitivity and specificity to predict an upcoming seizure and distinguish ES vs PNES events.
Results: There are 53 ES (60% female) and 46 PNES (80% female) patients. In each patient, total (SDNN, TP), vagal (RMSSD, SD1, HF), sympathetic (SD2), baroreflex (LF), and sympatho-vagal function (LF/HF) are higher at baseline on a seizure day ( >60 minutes before the event), compared to a non-seizure day at the same HR. Each baseline HRV measure predicts an upcoming seizure (ROC, 0.84–0.94). The temporal evolution and fluctuations in autonomic function significantly differ surrounding ES vs. PNES events. The pre-to-post-ictal change (Delta) in HR differ surrounding ES vs. PNES events. Post-ictal HR, SDNN, RMSSD, SD1, SD2, LF, HF, and TP differed for ES vs. PNES convulsive events. Delta HR and post-ictal HR, SDNN, RMSSD, LF, HF, and TP each distinguish ES vs. PNES convulsive events (ROC, 0.83-0.98). A logistic regression model with Delta and post-ictal HR provides the best prediction (92% sensitivity, 94% specificity, ROC 0.99).
Conclusions: HR and HRV are predictive markers for seizures and distinguish convulsive ES vs. PNES events. Results establish the framework for future studies to apply this diagnostic tool on outpatient recordings, particularly for populations with limited access to epilepsy monitoring units.
Funding: University of Rochester Provost Research Award, Central New York Foundation, CURE Epilepsy Cameron Boyce SUDEP Research Award
Translational Research