Predicting the Risk of Stroke and Delirium During Cardiac Surgery Using a Novel EEG Based Index of Interhemispheric Synchronization

Introduction: Neurological complications after surgery (e.g. stroke, delirium and cognitive dysfunction) remain a major concern despite advancements in surgical and anaesthetic techniques and technology. The aim of this study was to evaluate whether a novel electroencephalogram (EEG)-based index of interhemispheric synchronization, between two prefrontal EEG channels, could predict stroke or delirium following cardiac surgery. Methods: Two-channel EEG was analyzed in 803 patients undergoing cardiac surgery, without prior documentation of stroke. 31 patients suffered from stroke upon waking up from surgery (stroke group), for another 48 patients post-operative delirium was documented (delirium group) and the rest (724 patients) did not suffer from a documented neurological complication (no dysfunction group). A novel EEG-based index, the Lateral Interconnection Ratio (LIR) index, of frontal interhemispheric synchronization was calculated every 10 seconds and compared among the three patient groups over five key periods during surgery: start, pre-bypass, on-bypass, post-bypass, and end of surgery. The LIR index was analyzed in a blinded manner, without knowledge of patient group affiliation, and only compared among groups at the final analysis step. Results: The LIR index of patients with stroke decreased by 0.17±0.11 (mean±95% CI) from the start of surgery period to the immediate post bypass period in comparison to its relative stability of -0.03±0.02 in the no dysfunction group (p<0.0001). The LIR index of patients with delirium decreased from the start of surgery period to the end of surgery period by 0.20±0.13 in comparison to its relative stability of 0.02±0.02 in the no dysfunction group (p≈0.001). Summary: LIR index drops may provide an intraoperative alert to the occurrence of brain injury. The timing of LIR drops (post-bypass or end of surgery) may provide hints regarding the pathophysiology of the injury. For even better results it might be advisable to continue the monitoring during the post-operative period, especially while the patient is still anesthetized in the ICU, as there is continued risk for stroke and delirium also after surgery ends. Funding: The original study (Clinicaltrials.gov identifier: NCT02976584) was supported by the Bangerter-Rhyner Foundation and the Clinical Trials Unit of the University Hospital of Bern, University of Bern, Switzerland. Drs Hight & Kaiser were supported by the Department of Anaesthesiology and Pain Medicine, Bern University Hospital, University of Bern, Bern, Switzerland. Declaration of Interest: GS and DBS are spouses. GS founded several companies in the field of EEG analysis and DBS consulted one of these companies up to ~1 year ago. However, the present study is not related to these companies and their technology. It is based on developments in GS academic laboratory at Rambam Health Care Campus. The other authors have no conflict of interest to declare. Ethical Approval: The ethics committee of the canton of Bern, Switzerland, approved the prospective observational study (KEK#210/15). The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for observational studies was used to guide the methods of the study and to structure the manuscript