Abstract DP032: Delayed Cerebral Ischemia Alters Neurophysiologic-Autoregulatory Coupling in Subarachnoid Hemorrhage

Introduction: Delayed cerebral ischemia (DCI), a major complication of aneurysmal subarachnoid hemorrhage (aSAH), has been linked to impaired autoregulation. Disruptions in cerebral blood flow and metabolism may also manifest as impaired neurophysiology on EEG. However, the relationship between autoregulation and EEG biomarkers in aSAH remains poorly defined. This study aims to characterize this relationship and how DCI modifies it. Methods: In this study, DCI was defined as a new neurologic deterioration or infarction that could not be explained by other causes. DCI diagnosis was determined retrospectively by the consensus of at least two neurointensivists. Cerebral autoregulation was assessed by quantifying responses in regional oxygen saturation to changes in mean arterial pressure (MAP). Alpha-delta ratio (ADR) values were derived from spectral analysis of continuous EEG recordings, averaged across hemispheres, and plotted against 5-mmHg MAP bins for DCI and non-DCI patients. Inverse parabolic curves were fitted to the resulting scatter plots using quadratic regressions; the peaks and widths of these curves were compared to each other and to population-derived autoregulatory limits. A likelihood ratio test was used to determine whether including a MAP-DCI interaction term improves model fit when predicting ADR. Results: Seventy-seven patients, including 37 who developed DCI, with a median overlapping EEG-hemodynamic recording length of 113.4 hours (IQR 68.9–166.1) were included in the analysis. In patients with DCI, the ADR–MAP curve peaked at a higher MAP value (148 mmHg vs. 108 mmHg) and was broader (243.5 mmHg vs. 112.97 mmHg at half maximum) compared to non-DCI patients (Figs. 1&2). In DCI patients, the peak exceeded the median autoregulatory range (93.5–127.8 mmHg), whereas in non-DCI patients, the peak remained within its respective range (85.2–117.6 mmHg). Including the DCI–MAP interaction significantly improved model fit for predicting ADR ( χ 2 = 153.84, df = 3, p < 0.001; Table 1). Conclusions: Patients who developed DCI demonstrated maximum ADR values at MAPs higher than those of non-DCI patients and beyond the autoregulatory range. These findings suggest that DCI alters the relationship between autoregulation and EEG-based neurophysiology, revealing a novel physiological link that may inform new approaches to monitoring and preventing DCI.