Cerebral Autoregulation and Neurovascular Coupling in Acute and Chronic Stroke

So we still have no protocols for blood pressure management. You'd better hope your doctor guesses right about what to do. A wrong guess can be disastrous for you but won't affect your doctor at all. There are no consequences for your doctor not getting you 100% recovered.

 

,Cerebral Autoregulation and Neurovascular Coupling in Acute and Chronic Stroke

Lucy C. Beishon^1* and Jatinder S. Minhas^1,2

• ¹Department of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom
• ²National Institute for Health Research (NIHR) Leicester Biomedical Research Centre, British Heart Foundation Cardiovascular Research Centre, Glenfield Hospital, Leicester, United Kingdom

Introduction

Stroke is currently the second leading cause of death worldwide, and results in significant morbidity, and poorer quality of life for those affected (1). Stroke can be classified under two major sub-types: ischaemic and haemorrhagic. Ischaemic stroke accounts for ~70% of all stroke, and results from arterial occlusion, usually through embolism or small vessel thrombosis (2). Haemorrhagic stroke is a result of arterial rupture in the brain (2). However, the two sub-types frequently co-exist, with similar risk factors (e.g., hypertension), and overlap in pathological mechanisms (3).

Although stroke incidence has declined in high-income countries, it remains a prevalent issue amongst low and middle-income countries, disproportionately affecting a younger, working age population in these areas (1). The treatment of acute ischaemic stroke (AIS) has advanced over recent decades. Notably, the advent of both thrombolysis and mechanical thrombectomy has revolutionised the management of AIS, associated with reduced mortality, and improved functional outcome (2, 4, 5). Despite these advances, the management of haemorrhagic stroke has lagged behind, and treatment options are largely confined to reversal of anticoagulants and intensive blood pressure (BP) lowering (2). Conversely, in AIS, the target for BP management remains uncertain, and trials have largely shown equivalence (6, 7), or harm (8), associated with aggressive BP management strategies. To understand the mechanistic implications of BP lowering in AIS, studies have investigated the temporal changes in cerebral autoregulation (CA) following stroke (9). In healthy states, CA maintains a constant cerebral perfusion, despite fluctuations in systemic BP (10). However, the ability of the brain maintain CA may be compromised in the acute phase of stroke, increasing the vulnerability of the brain to hypoperfusion with intensive BP management strategies (11, 12). Conversely, surges in BP during this vulnerable phase may risk haemorrhagic transformation of the infarct, resulting in poorer outcomes (11, 12). Thus, understanding the temporal nature of CA in the acute phase of stroke could provide important mechanistic insights to guide BP management strategies in the clinical setting.

A related concept to CA is the physiological mechanism of neurovascular coupling (NVC). Under healthy conditions, neuronal activity is tightly coupled to cerebral blood flow (CBF), such that increases in neuronal activity will result in increases in CBF to ensure the metabolic demands of the brain are met. Intact NVC is integral to maintain optimal cognitive function, and thus may be an important physiological mechanism in the chronic or rehabilitation phase of stroke. The following sections consider the evidence to support a role for CA and NVC as important mechanistic factors in the acute and chronic phases of stroke, and the key clinical and research implications going forward.

CA

Dynamic cerebral autoregulation (dCA) has now been carefully characterised at rest in AIS (13), acute intracerebral haemorrhage (ICH) (14) and chronic stroke (15) states. Furthermore, several studies have modelled the relationship between arterial CO₂ (PaCO₂), cerebral blood flow and dynamic cerebral autoregulation (16, 17). Hypercapnia causes vasodilation and deteriorates CA, with hypocapnia conversely causing vasoconstriction and an improvement in CA status (16, 17). Meta-analyses, albeit with significant heterogeneity, have demonstrated transfer function analysis parameter [phase and autoregulation index (ARI)] impairment in large and small artery AIS, lower phase in ICH and “rebounding phase” in chronic stroke (13). Unfortunately, limitations of existing transcranial Doppler based haemodynamic studies include low assessment frequency post stroke [particularly lacking data in ultra-acute (hours) and medium to longer term (weeks to months)] and clarification of dCA “cut-points” for impairment. Until very recently, there was a lack of dCA data peri- mechanical thrombectomy (MT), however, recent studies have shown worse dCA in the first 24 h associated with higher rates of haemorrhagic transformation and lower rates of recanalization (18). Specific learnings from this data suggest incomplete recanalisation of large-vessel occlusion, with impaired autoregulation status confer complication—raising the importance of adequate blood pressure control in this context (18). Whilst there are confounders to consider when assessing dCA pre-, during or post- MT including blood pressure (19, 20), end-tidal carbon dioxide level (21) and mode of anaesthesia (22)—their behaviour and interactions are yet to be determined. Higher end-tidal CO₂ levels in those with incomplete recanalisation, especially beyond 72 h post large-vessel occlusion (LVO) is of significant interest (18). In ICH, the storey differs, with severe hypocapnia (low arterial CO₂ levels) associated with poor prognosis (23). Furthermore, lowering BP during acute hypertensive states during ICH, in the setting of low arterial CO₂ levels, leads to a greater risk of ischaemic lesions on MRI imaging (23). These differences in acute haemodynamics between stroke sub-types could be explained by nature of structural lesion (infarct vs. haematoma), existence of pre-existing chronic hypertension or differing responses to blood pressure lowering. Given personalised autoregulation-based BP targets are now possible in both a ward based stroke setting (24) and neurocritical care (25). Unfortunately, there still remains an inability to quantify the potential modulation of dCA by chronic hypertension before, during and immediately after acute stroke. The perceived “rightward shift” in the dCA curve is yet to be proven in acute (within 96 h) and sub-acute (7 to 14 days) contexts with ongoing hypertension or antihypertensive treatment being administered (26).

Recent advancements have further highlighted the need to recognise inter-subject variability (27) and responders vs. non-responders (28). There is evidence to suggest dCA impairment is greatest in regions with critically reduced perfusion (greatest volume of viable tissue), though dCA impairment can be present across the entire hemisphere to varying degrees (27). In ICH, through routinely obtained MRI scans in the first 7 days post-event, initial BP, nadir BP, and arterial CO₂ were independent predictors of diffusion-restricted lesion incidence (23). Pooled individual patient data meta-analyses from the ATACH-2 and MISTIE III trials demonstrated in a heterogeneous cohort of patients with ICH, diffusion-weighted imaging (DWI) lesions were associated with 2.5-fold heightened risk of stroke among ICH survivors—with elevated risk persisting for AIS but not for recurrent ICH (29). In order to determine whether ischaemic lesions noted on DWI are preventable, or indeed governed by therapeutic variation in BP approaches (30)—mechanistic dCA studies at time of BP lowering, with continuous end-tidal CO₂ measurement are needed, with MRI DWI assessment at 7 days. White matter ischaemic change may be attributed to by high blood pressure variability in addition to adverse adaptations of CA. In hypertensives without acute stroke disease, dCA (assessed using ARI) and CO₂ reactivity were not related to white matter lesions—however, relationships with duration of hypertension and nocturnal BP dipping were shown (31). Ultimately, there exists a complex interdependent relationship between acute and chronic hypertensive states, dCA, and chronic cerebrovascular ischaemic injury. Crucially, we have evidence to support the hypothesis that carbon dioxide change in the acute setting post-stroke may modify risk, through interaction with BP lowering and dCA status, increasing the ischaemic stroke risk post ICH (23).

In both AIS and ICH, there exist adverse pathophysiologically driven complications including vasogenic oedema and haematoma expansion, respectively. The behaviour of cerebrovascular tone (critical closing pressure, CrCP) and resistance (resistance area product, RAP) is less well-understood. There is debate as to the sensitivity of CrCP to variation in intracranial pressure (ICP) (32). However, the presence of a haematoma in ICH as compared to controls, during normocapnic and hypocapnic conditions, showed significant differences in CrCP and RAP (33). Beyond common indices of dCA, there is limited knowledge of tone and resistance parameters in acute cerebrovascular states as compared to the traumatic brain injury literature.

More at link.