Integration of various brain signals can be used to determine cerebral autoregulation in neurocritical care patients to guide clinical management and to predict outcome. In this review, we will discuss current methodology of multimodal brain monitoring focusing on secondary 'reactivity indices' derived from various brain signals which are based on a 'moving correlation coefficient'. This algorithm was developed in order to analyze in a time dependent manner the degree of correlation between two factors within a time series where the number of paired observations is large. Of the various primary neuromonitoring sources which can be used to calculate reactivity indices, we will focus in this review on indices based on transcranial Doppler (TCD), intracranial pressure (ICP), brain tissue oxygenation (PbtO2) and near infrared spectroscopy (NIRS). Furthermore, we will demonstrate how reactivity indices can show transient changes of cerebral autoregulation and can be used to optimize management of arterial blood pressure (ABP) and cerebral perfusion pressure (CPP).

Methods to identify an autoregulation guided “optimal” cerebral perfusion pressure (CPPopt) for traumatic brain injury patients (TBI) have been reported through several studies. An important drawback of existing methodology is that CPPopt can be calculated only in ∼50–60% of the monitoring time. In this study, we hypothesized that the CPPopt yield and the continuity can be improved significantly through application of a multi-window and weighting calculation algorithm, without adversely affecting preservation of its prognostic value. Data of 526 severe TBI patients admitted between 2003 and 2015 were studied. The multi-window CPPopt calculation was based on automated curve fitting in pressure reactivity index (PRx)-CPP plots using data from 36 increasing length time windows (2–8 h). The resulting matrix of CPPopts was then averaged in a weighted manner. The yield, continuity, and stability of CPPopt were studied. The difference between patients' actual CPP and CPPopt (ΔCPP) was calculated and the association with outcome was analyzed. Overall, the multi-window method demonstrated more continuous and stable presentation of CPPopt in this cohort. The new method resulted in a mean (±SE) CPPopt yield of 94% ± 2.1%, as opposed to the previous single-window-based CPPopt yield of 51% ± 0.94%. The stability of CPPopt across the whole monitoring period was significantly improved by using the new algorithm (p < 0.001). The relationship between ΔCPP according to the multi-window algorithm and outcome was similar to that for CPPopt calculated on the basis of a single window. In conclusion, this study validates the use of a new multi-window and weighting algorithm for significant improvement of CPPopt yield in TBI patients. This methodological improvement is essential for its clinical application in future CPPopt trials.

Objective: Recently, a concept of an individually targeted level of cerebral perfusion pressure that aims to restore impaired cerebral vasoreactivity has been advocated after traumatic brain injury. The relationship between cerebral perfusion pressure and pressure reactivity index normally is supposed to have a U-shape with its minimum interpreted as the value of “optimal” cerebral perfusion pressure. The aim of this study is to investigate the relation between the absence of the optimal cerebral perfusion pressure curve and physiological variables, clinical factors, and interventions. Design: Retrospective analysis of prospectively collected data. Setting: Neurocritical care units in two university centers. Patients: Between May 2012 and December 2013, a total of 48 traumatic brain injury patients were studied with real-time annotation of predefined clinical events. Interventions None. Measurements and Main Results: All patients had continuous monitoring of arterial blood pressure, intracranial pressure, and cerebral perfusion pressure, with real-time calculations of pressure reactivity index and optimal cerebral perfusion pressure using ICM+ software (Cambridge Enterprise, University of Cambridge, Cambridge, UK). Selected clinical events were inserted on a daily basis, including changes in physiological variables, sedativeanalgesic drugs, vasoactive drugs, and medical/surgical therapies for intracranial hypertension. The collected data were divided into 4-hour periods, with the primary outcome being absence of the optimal cerebral perfusion pressure curve. For every period, mean values (± SDs) of arterial blood pressure, intracranial pressure, pressure reactivity index, and other physiological variables were calculated; clinical events were organized using predefined scales. In 28% of all 1,561 periods, an optimal cerebral perfusion pressure curve was absent. A generalized linear mixed model with binary logistic regression was fitted. Absence of slow arterial blood pressure waves (odds ratio, 2.7; p < 0.001), higher pressure reactivity index values (odds ratio, 2.9; p < 0.001), lower amount of sedative-analgesic drugs (odds ratio, 1.9; p = 0.03), higher vasoactive medication dose (odds ratio, 3.2; p = 0.02), no administration of maintenance neuromuscular blockers (odds ratio, 1.7; p < 0.01), and following decompressive craniectomy (odds ratio, 1.8; p < 0.01) were independently associated with optimal cerebral perfusion pressure curve absence. Conclusions: This study identified six factors that were independently associated with absence of optimal cerebral perfusion pressure curves.