Non-invasive cerebral oxygenation and blood pressure neuromonitoring characterizes dynamic cerebrovascular autoregulation following cardiac arrest.

BACKGROUND AND PURPOSE: Cerebrovascular autoregulation (CVAR) is impaired following cardiac arrest. We describe near-infrared spectroscopy (NIRS)-derived CVAR parameters following cardiac arrest and characterize their evolution over time.

METHODS

This was a retrospective analysis of a prospectively collected clinical and physiological cardiac arrest data repository at a single academic medical center between March 2022 and October 2023. Eligible patients were connected to neuromonitoring within 72 h of cardiac arrest, had greater than four hours of monitoring, and had no confounding neurologic diagnosis. NIRS derived cerebral oxygenation index (COx) and limits of autoregulation were computed using the linear correlation between time-synchronized regional cerebral oxygenation (rSO2) and continuous mean arterial pressure (MAP). CVAR limits [inclusive of optimal MAP (MAPOPT), lower limit of autoregulation (LLA), and upper limit of autoregulation (ULA)] were determined using a fitted COx versus MAP parabolic curve with a COx threshold of 0.3.

RESULTS

Of 60 eligible patients, 56 underwent neuromonitoring, and 55 were included in this analysis. Mean (standard deviation [SD]) age was 53.4 (15.4) years, and the majority had high severity of post-cardiac arrest syndrome. The median (interquartile range) time from cardiac arrest to first recorded COx was 14.2 (10.1-21.2) h; median monitoring duration was 49.1 (26.6-85.3) h. Mean (SD) MAPOPT and limits of autoregulation were 86.0 (13.1) mmHg and 74.6 (11.3)-96.8 (13.0) mmHg, respectively. The LLA was ≥65 mmHg in 85.5% (n = 47) of patients. All patients had at least one epoch of impaired or dysfunctional CVAR. Thirteen patients (23.6%) survived to hospital discharge. Survivors had a higher MAP (91.6 vs 82.7 mmHg, p = 0.039) and lower COx (0.05 vs 0.15, p = 0.041) than non-survivors.

CONCLUSION

Impaired CVAR was common and dynamic over at least 120 h following cardiac arrest. If a population-based MAP threshold of ≥65 mmHg were applied to our cohort, it may have resulted in significant time burden below the LLA. While it was feasible to determine CVAR limits, further work is needed to identify if CVAR-based treatment approaches influence outcomes.