The effect of EEG on assessment of cerebral autoregulation

Thesis

Cerebral blood flow (CBF) is determined by vascular resistance, mean arterial blood pressure (MABP) and intracranial pressure. The energy requirement of brain cells means CBF has to be maintained constant despite changes in pressures. Cerebral autoregulation is the mechanism which keeps the CBF constant for a range of variations in pressure. Autoregulation can be impaired after stroke, trauma or anaesthesia. Following the impairment of autoregulation, such as under intensive care conditions, previous studies have found that Electroencephalography (EEG) signals also change in their frequency content in a manner that corresponds to decreases in CBF. These studies have suggested that decreasing levels of blood flow change the metabolic and electrical activity of cortical neurons. Assessment of autoregulation status is essential to help to reduce or to prevent any further damage or cell death. With the availability of techniques to continuously monitor and to record physiological variables such as blood pressure, blood flow rate, it is possible to make continuous assessments of autoregulation. There are many methods to assess autoregulation and the most commonly used methods include transfer function analysis (TFA) and the autoregulation index (ARI) method. As EEG has the advantage that it can be recorded in real time, it offers the promise of potentially providing an additional non-invasive real time autoregulation assessment, given that we have some understanding of interpreting how exactly changes in EEG reflect the status of blood flow. Though there is strong clinical evidence of the relationship between EEG frequency change and CBF changes there are no model based investigations or methods that could explain the mechanism of how both are coupled. In order to attempt a study of this link between EEG and CBF, we devised an artificial reduction in CBF levels, by increasing the resistance of large arteries in steps, using an existing haemodynamic model, which is capable of handling simultaneous changes in ABP, CO2 and neural input (EEG). We generated CBF using our haemodynamic model, by feeding in experimental data for ABP, CO2 and synthetic EEG, for every step reduction in large artery resistance. Using the autoregulation assessment methods TFA and ARI, we studied how the EEG behaviour affects the assessment of autoregulation. Results show that EEG input appears to make autoregulation ’better’ by a small fraction under the circumstances of the model and data we have used. There is, however, high variability between different subject data. The implication from this study is that, on average EEG input increases the measure of autoregulation slightly, but the variability between subjects means that more, controlled, data are required to validate fully this study’s observations.

Authors: Narinahally Krishnamurthy, SK

• DOI: 10.5287/ora-vybmzaznx
• Type of award: MSc by Research
• Level of award: Masters
• Awarding institution: University of Oxford
• Language: English
• Keywords: Assessment of Autoregulation; Cerebral Autoregulation; Autoregulation
• Subjects: Biomedical engineering; Cerebral Autoregulation