Cerebrovascular pressure reactivity and intracranial pressure are associated with neurologic outcome after hypoxic-ischemic brain injury.

Abstract

We evaluated the association of physiological parameters measured by intracranial multimodality neuromonitoring with neurologic outcome in a consecutive series of patients with hypoxic-ischemic brain injury (HIBI). We retrospectively identified all patients with HIBI who underwent combined invasive intracranial pressure (ICP) and brain tissue oxygen (P_btO₂) monitoring over a 3 year period. Cerebrovascular pressure reactivity index (PRx) was calculated continuously as a surrogate of cerebral autoregulation. Favorable outcome was defined as recovery of consciousness (Glasgow Coma Scale motor score = 6). Differences in mean ICP, PRx and P_btO₂ for the entire monitoring period across outcomes were measured. Logistic regression and area under receiver operating characteristic (AUROC) curve were used to assess the association of each monitoring parameter with neurologic outcome. We analyzed data from 36 patients. Most (89%) had an antecedent sudden cardiac arrest. Favorable outcome occurred in 8 (22%) patients. ICP and PRx were higher in patients with unfavorable outcome (ICP: 26 ± 4.1 mmHg vs 7.5 ± 2 mmHg, p = 0.0002; PRx: 0.51 ± 0.05 vs 0.11 ± 0.05, p < 0.0001). There was no significant difference in P_btO₂ between groups (unfavorable: 20 ± 2.4 mmHg vs favorable: 25 ± 1.5 mmHg, p = 0.12). Both ICP (AUROC 0.84, 95%CI 0.72-0.98, p = 0.003) and PRx (AUROC 0.94, 95%CI 0.85-1, p = 0.0002) discriminated between favorable and unfavorable outcome, in contrast to P_btO₂, (AUROC 0.59, 95%CI 0.39-0.78, p = 0.52). ICP > 15 mmHg, PRx > 0.2, and P_btO₂ < 18 mmHg had sensitivity/specificity of 68%/100%, 89%/88%, and 40%/100% respectively for discriminating outcomes. Cerebrovascular pressure reactivity and intracranial pressure appear to be associated with neurologic outcome in patients with HIBI.