Is It Feasible To Manipulate Arterial Carbon Dioxide Levels To Improve Impaired Cerebral Autoregulation In Acute Haemorrhagic Stroke?

Background: Cerebral Autoregulation (CA) is usually defined as the tendency of cerebral blood flow (CBF) to remain approximately constant despite changes in arterial blood pressure (BP) within the range of 50 to 170mmHg. Dynamic CA (dCA) can be estimated from the transient response of CBF to rapid changes in BP. Spontaneous acute intracerebral haemorrhage (ICH) presents a devastating cerebral event with high morbidity and mortality. The mainstay of treatment remains BP control, which relies on a functioning CA. CA has been shown to be impaired in acute ICH. Arterial partial pressure of carbon dioxide (PaCO2) has a strong influence on dCA and other cardio- and cerebro-vascular variables. Understanding the dynamic CA response to physiological manoeuvres, such as exercise and changes in respiratory patterns, has often been confounded by simultaneous changes in PaCO2. Hypercapnia leads to vasodilation of cerebral vessels and overall causes deterioration in CA. Conversely, hypocapnia has a vasoconstrictive effect, improving CA. Aim: The aim of this thesis is to comprehensively model the cerebral haemodynamic response to the entire physiological range of PaCO2 in order to safely permit the assessment of feasibility of clinical translation of a CO2-based intervention into a cohort of acute ICH patients with impaired autoregulation. Objectives: This thesis objectives are to: i) determine the natural history and prognostic implications of CA impairment in acute ICH; ii) determine if the use of current CO2 measurement techniques leads to significant differences in CO2-related systemic and cerebrovascular parameters; iii) determine the most appropriate method of initiating and maintaining hypocapnia; iv) determine if key cerebral haemodynamic parameters including autoregulation index (ARI), arterial BP (BP), heart rate (HR), critical closing pressure (CrCP) and resistance-area product (RAP) follow a logistic non-linear model similar to those described for cerebral blood flow velocity (CBFV); v) investigate sex differences in cerebral haemodynamics across the physiological range of PaCO2 and vi) determine whether hypocapnia (via hyperventilation) in acute ICH may improve CA and consequently clinical outcome. Methods: CA was measured in healthy and acute ICH patients by transcranial Doppler ultrasound assessment of middle cerebral artery velocities alongside continuous non-invasive monitoring of BP. Results: 45 healthy volunteers underwent a multi-step protocol inducing hypo- and hyper-capnia and 12 acute ICH patients underwent a metronome based hypocapnic intervention at <48 hours and 10-14 days. The thesis results demonstrated i) the aforementioned parameters follow a logistic curve relationship; ii) CBFV is lower and dCA is impaired in acute ICH as compared to healthy controls; iii) different EtCO2 measurement techniques do lead to physiological changes and differences in parameter values; iv) dysautoregulation can be minimised by continuous metronome use during hyperventilation-induced hypocapnia; v) logistic curve parameters are influenced by sex and vi) dCA can be improved in acute ICH using a CA targeted CO2-based interventional manoeuvre. Conclusion: This thesis presents a new paradigm for assessment of the interaction of CO2 and dCA and its potential clinical applications. In addition, original findings include improved understanding of CO2 focussed physiological measurement protocol design, comprehensive clarification of cerebral haemodynamics in ICH, optimisation of hypocapnia induction, demonstration of novel sex differences during PaCO2 change and via accumulation of the aforementioned knowledge, the safety and feasibility of a novel CA targeted CO2-based interventional manoeuvre in acute ICH.