Abstract 274: Identifying the Optimal Cerebral Perfusion Pressure Following Hypoxic-Ischemic Brain Injury

Introduction: The optimal cerebral perfusion pressure (CPP) following hypoxic-ischemic brain injury (HIBI) is currently unknown. We retrospectively analyzed intracranial monitoring data from a cohort of patients with HIBI to identify a threshold level for CPP that optimizes cerebrovascular pressure reactivity (a surrogate for CA) while limiting the risk of intracranial hypertension and brain tissue hypoxia. Hypothesis: We hypothesized that higher CPP values would be associated with improved cerebrovascular pressure reactivity. Methods: ICP, brain tissue oxygen (P bt O 2 ), MAP, and CPP (defined as MAP - ICP) were recorded continuously and time synchronized for all patients using a bedside monitor (CNS Monitor, Moberg Research). Pressure Reactivity Index (PRx) was calculated as the time varying correlation between MAP and ICP over 5 min intervals updated every minute. The degree of CA impairment (defined as % time PRx > 0.2) was plotted against MAP and CPP, respectively. The relationships between ICP and P bt O 2 versus CPP, as well as ICP and P bt O 2 versus % time PRx > 0.2, were similarly calculated. Results: We analyzed 37 patients (33 cardiac arrest, 4 prolonged hypoxia) with HIBI who underwent intracranial neuromonitoring over a 3 year period. Lower CPP values were associated with higher degrees of CA impairment. The cumulative burden of elevated PRx was significantly lower for CPP values above a cutoff of 85 mmHg compared to lower CPP values (p < 0.001, Wilcoxon rank sum test). A similar cutoff for MAP could not be identified, although lower MAP values were also associated with greater CA impairment. Intracranial hypertension (ICP > 20 mmHg) and brain hypoxia (P bt O2 < 20 mmHg) were both associated with CA impairment (p < 0.001 and p < 0.001, respectively, Wilcoxon rank sum test). Conclusions: Higher CPP and MAP values appear to be associated with improved CA after HIBI. Given that CA impairment is associated with both intracranial hypertension and brain hypoxia, our work reaffirms the notion that higher blood pressure targets may improve outcome after HIBI. The identification of a distinct CPP cutoff for CA optimization suggests that targeting CPP instead of MAP may be advantageous, although further work is required to clarify this issue.

Abstract 12: Cerebrovascular Pressure Reactivity Predicts Outcome in Diffuse Hypoxic-Ischemic Brain Injury

Introduction: Outcome prediction after diffuse hypoxic-ischemic brain injury (HIBI), which most commonly occurs after sudden cardiac arrest, remains fraught with uncertainty. Cerebral autoregulation (CA) normally limits fluctuations in cerebral blood flow through pressure induced changes in cerebral arteriolar diameter. CA is disrupted after brain injury, and the degree of autoregulatory failure correlates with the severity of brain injury. CA measurements may thus enable more accurate outcome prediction after HIBI. Hypothesis: We hypothesized that the degree of impairment in cerebrovascular pressure reactivity, one of the principle mechanisms of CA, correlates with the extent of HIBI and can predict outcome. Methods: Intracranial pressure (ICP) and mean arterial pressure (MAP) were continuously recorded for all subjects. The Pressure Reactivity Index (PRx) was calculated as a moving linear correlation coefficient between ICP and MAP for each 30 sec data epoch. PRx is positive when ICP increases in concert with MAP and implies impaired pressure reactivity. For each subject, both the mean PRx for the entire recording session as well as the proportion of PRx values > 0.2 (previously found to predict poor outcome in TBI) were calculated. Results: We included 22 consecutive patients with HIBI who underwent invasive ICP monitoring in this single center observational study between 02/2016-02/2018. Median age was 41 (range 20-77), 36% (8/22) survived to hospital discharge, and 18% (4/22) had return of consciousness. The median time of ICP monitoring was 101 hr (range 8-381 hr). HIBI was due to cardiac arrest in 19/22 patients and prolonged hypoxia in 3/22. Mean PRx was significantly higher in patients without return of consciousness (0.49 ± 0.05 vs. 0.036 ± 0.032, p = 0.001), as was the proportion of PRx values > 0.2 (0.73 ± 0.037 vs 0.34 ± 0.058, p < 0.001). All patients with return of consciousness had mean PRx < 0.2. Using this cutoff value, PRx had a 94% sensitivity and 100% specificity for predicting death or vegetative state. Conclusions: Continuous pressure reactivity measurements correlate with outcome after HIBI, and may be a useful adjunct for neuroprognostication. Future studies are needed to validate these findings in larger cohorts.