Continuous Assessment of “Optimal” Cerebral Perfusion Pressure in Traumatic Brain Injury: A Cohort Study of Feasibility, Reliability, and Relation to Outcome

Abstract

Background: Guidelines recommend maintaining cerebral perfusion pressure (CPP) between 60 and 70 mmHg in patients with severe traumatic brain injury (TBI), but acknowledge that optimal CPP may vary depending on cerebral blood flow autoregulation. Previous retrospective studies suggest that targeting CPP where the pressure reactivity index (PRx) is optimized (CPP opt ) may be associated with improved recovery. Methods: We performed a retrospective cohort study involving TBI patients who underwent PRx monitoring to assess issues of feasibility relevant to future interventional studies: (1) the proportion of time that CPP opt could be detected; (2) inter-observer variability in CPP opt determination; and (3) agreement between manual and automated CPP opt estimates. CPP opt was determined for consecutive 6-h epochs during the first week following TBI. Sixty PRx-CPP tracings were randomly selected and independently reviewed by six critical care professionals. We also assessed whether greater deviation between actual CPP and CPP opt (ΔCPP) was associated with poor outcomes using multivariable models. Results: In 71 patients, CPP opt could be manually determined in 985 of 1173 (84%) epochs. Inter-observer agreement for detectability was moderate (kappa 0.46, 0.23–0.68). In cases where there was consensus that it could be determined, agreement for the specific CPP opt value was excellent (weighted kappa 0.96, 0.91–1.00). Automated CPP opt was within 5 mmHg of manually determined CPP opt in 93% of epochs. Lower PRx was predictive of better recovery, but there was no association between ΔCPP and outcome. Percentage time spent below CPP opt increased over time among patients with poor outcomes (p = 0.03). This effect was magnified in patients with impaired autoregulation (defined as PRx > 0.2; p = 0.003). Conclusion: Prospective interventional clinical trials with regular determination of CPP opt and corresponding adjustment of CPP goals are feasible, but measures to maximize consistency in CPP opt determination are necessary. Although we could not confirm a clear association between ΔCPP and outcome, time spent below CPP opt may be particularly harmful, especially when autoregulation is impaired.