An examination of ischaemic penumbra in the spontaneously hypertensive stroke-prone rat (SHRSP) using the MRI perfusion-diffusion mismatch model

Stroke accounts for 9% of all deaths worldwide and is a major cause of severe disability (Donnan et al, 2008). Following ischaemic stroke, the penumbra represents tissue which is hypoperfused and functionally impaired but is not yet irreversibly damaged. However, the penumbra has a finite lifespan and will proceed to infarction in the absence of swift reperfusion. Therefore, the identification and potential salvage of penumbral tissue in acute ischaemic stroke is the ultimate goal for both clinicians and experimental stroke researchers. Positron emission tomography (PET) is the ‘gold standard’ imaging modality for identifying the penumbra, but the complex logistics of PET limit its widespread use. Magnetic Resonance Imaging (MRI) is widely used for penumbra imaging in both clinical and pre-clinical research. The MRI perfusion-diffusion mismatch model provides an approximation of the penumbra, where diffusion weighted imaging (DWI) identifies the core of ischaemic injury and perfusion weighted imaging (PWI) reveals the perfusion deficit. The mismatch between the DWI and PWI provides a measure of penumbral tissue. However, there is no consensus on the perfusion and diffusion thresholds used to identify mismatch tissue in clinical and preclinical stroke research. Furthermore, in rodent stroke models differences in the evolution of ischaemic injury between strains may limit the use of a single set of threshold values. Therefore, the first aim of this thesis was to establish strain specific perfusion and diffusion thresholds to compare penumbra volume in the clinically relevant spontaneously hypertensive stroke-prone rat (SHRSP) and the normotensive control strain, Wistar-Kyoto (WKY) using 3 different methods. The SHRSP strain is characterised by the progressive development of severe hypertension which is followed by a tendency to spontaneous stroke and an increased sensitivity to experimental stroke. Experimental stroke was induced by permanent middle cerebral artery occlusion (MCAO) by the intraluminal filament method. DWI and PWI were obtained every hour from 1-4 hours post-MCAO. Strain-specific diffusion and perfusion thresholds were established from final infarct at 24 hours post-MCAO, as defined by T2 weighted imaging. The calculated ADC thresholds were comparable between the strains but the absolute perfusion threshold was significantly higher in SHRSP compared to WKY. This may be indicative of an increased sensitivity to ischaemia in the hypertensive strain. Furthermore, application of these thresholds to the acute MRI data revealed that the volume of ischaemic injury and the perfusion deficit were significantly larger in SHRSP compared to WKY and this was also reflected in the significantly larger infarct volume observed in SHRSP at 24 hours post-MCAO. Interestingly, there was evidence of a temporal increase in the volume of the perfusion deficit in SHRSP and WKY. This may indicate that there is a progressive failure of collateral blood supply in both strains following stroke. Penumbra volume was then assessed in SHRSP and WKY rats using the mismatch method and also indirectly by examining the growth of the volume of ADC derived ischaemic injury. Mismatch volume was determined by arithmetic subtraction of the volume of ischaemic injury from the volume of perfusion deficit (volumetric method) and also by manual delineation of mismatch on each of 6 coronal slices (spatial method). There was a limited volume of mismatch tissue in either strain from as early as 1 hour post-MCAO and the volumetric method generated smaller mismatch volumes than the spatial mismatch method. Mismatch volume was comparable in SHRSP and WKY from 1-4 hours post-MCAO. Penumbra was also determined retrospectively by subtracting the volume of ischaemic injury at each time point from final infarct volume. Using this method, penumbra volume was significantly larger in WKY compared to SHRSP at 30 minutes post-MCAO but penumbra volume was comparable at all later time points. This suggests that there is reduced potential for tissue salvage in SHRSP compared to WKY within the first hour following MCAO but from 1 hour onwards, there is limited potential for penumbra salvage in both strains. In addition, there was evidence of ‘negative’ mismatch tissue in SHRSP and WKY rats, where the ADC derived lesion expanded beyond the boundary of the perfusion deficit. The volume of negative mismatch tissue was comparable between the strains and was persistent over the 4 hour time course. This phenomenon may arise from the spread of toxic mediators from the ischaemic core. Oxidative stress is a major mediator of cellular injury following ischaemic stroke and reactive oxygen species, like superoxide, have multiple deleterious effects on the components of the neurovascular unit. It is well established that NADPH oxidase is the principal source of superoxide in acute ischaemic stroke and is therefore a target for potential neuroprotective strategies (Moskowitz et al, 2010). Consequently, the second aim of this thesis was to evaluate the potential neuroprotective effect of NADPH oxidase inhibition with low and high dose apocynin following permanent or transient ischaemia. Rats were administered apocynin at a dose of 5mg/kg or 30mg/kg or vehicle, at 5 minutes post-MCAO. Apocynin treatment had no significant effect on infarct volume or functional outcome at 24 hours following permanent MCAO in WKY rats. However, both low and high dose apocynin treatment significantly reduced infarct volume at 72 hours post-MCAO by 60% following 1 hour of ischaemia in Sprague-Dawley rats. Furthermore, functional outcome was improved in the low dose apocynin treated group, although this did not reach the level of statistical significance. On the basis of these results, low dose apocynin treatment was evaluated in SHRSP rats following 1 hour of ischaemia. However, apocynin treatment had no effect on the acute evolution of ischaemic injury and failed to improve stroke outcome, where the mortality rate was high in both the apocynin treated and the vehicle treated group. The conflicting effects of apocynin may be attributable to a differential expression of NADPH oxidase subunits in normotensive and hypertensive rat strains. These findings may also explain the failure of neuroprotective drugs to translate from bench to bedside, as therapies which are neuroprotective in young healthy animals may not demonstrate the same efficacy in animal models with stroke co-morbidities. Therefore, potential therapeutic strategies should be extensively evaluated in animal models with stroke risk factors before proceeding to clinical trial