Abstract
It has been proposed that subject-specific intracranial pressure (ICP) thresholds can be feasibly derived using the relationship between cerebrovascular reactivity and ICP. Such individualized intracranial pressure (iICP) thresholds have been suggested to have more robust associations with long-term outcomes of post-traumatic brain injury (TBI) than current guideline-based thresholds. However, both existing works have derived iICP using solely the pressure reactivity index (PRx) and a threshold of +0.20. Therefore, the goal of this study was to validate prior works and compare various cerebrovascular reactivity indices for their utility in deriving iICP. A custom iICP derivation algorithm was developed. Then, using existing archived human datasets from the Winnipeg Acute TBI Database, iICP thresholds were derived using three cerebrovascular reactivity indices: PRx, the pulse amplitude index (PAx), and the RAC (correlation (R) between the pulse amplitude of ICP (A) and cerebral perfusion pressure (C)). The yield of iICP derivation was found to vary significantly, depending on the cerebrovascular reactivity index and threshold used. A logistic regression analysis was then used to evaluate and compare the abilities of each index-derived iICP to predict the 6-month outcome. Among all index-threshold combinations tested, only PRx > 0 was able to produce an iICP that was able to outperform guideline-based ICP thresholds. PRx-based iICP seems to be superior to both PAx- and RAC-based iICP for predicting long-term outcomes. However, further work is needed to identify the ideal cerebrovascular reactivity thresholds for iICP derivation.