New concept of stroke pathophysiology: the potential impact on stroke management

Traditionally, embolism and hypoperfusion are accepted as two independent stroke mechanisms, with the former predominating the stroke events. Henceforth, stroke prevention works towards antithrombotic therapy. The interplay of embolism and hypoperfusion has recently been proposed to be the mechanism responsible for most strokes, as a result of impaired clearance of emboli in hypoperfused brain tissue. It is expected that our treatment strategies will be geared towards combination therapy of antithrombosis and flow augmentation when this new concept of stroke pathophysiology is widely accepted by the clinical community.

Stroke is a common cause of death and disability; the best approach to managing stroke is prevention. Hence, the prevention of stroke remains a major clinical challenge for neurologists. Current management strategies usually involve treating, controlling or modifying specific stroke risk factors with appropriate changes in lifestyle, surgical intervention and pharmacological treatment. Well-documented modifiable risk factors, such as smoking, carotid-artery stenosis, dyslipidemia, atrial fibrillation (AF) and certain cardiac diseases (e.g., dilated cardiomyopathy, valvular heart disease and intracardiac congenital defects) are all related to an increased risk of thromboembolic stroke [1]. It implies that the management strategies for stroke prevention are largely focused on how to eliminate or reduce the chance of thrombus formation in the circulation, particularly targeting the cervico–cerebral arteries and the heart so as to minimize the risk of cerebral embolism.

Our treatment strategies have long been driven by our contemporary knowledge of the pathophysiology of a disease. A century ago, when spasmodic contraction of the middle-cerebral artery was believed to be a cause of transient ischemic attack and stroke, cervical sympathetic ganglionectomy or stellate ganglion blockade was the treatment of choice to overcome the angiospastic response. When hypoperfusion due to flow-limiting stenoses in the intracranial, carotid and vertebobasilar arteries was proposed as the ‘culprit’ for cerebral ischemia and stroke in the 1950s, preventive treatment was geared towards flow augmentation by surgical intervention in order to reperfuse the ischemic area. With the advent of computed tomography and magnetic resonance imaging in the past decades, thromboembolism gradually emerged to be the major cause of stroke and, henceforth, antithrombotic therapy became one of the mainstays of stroke prevention. It is likely that the management strategies for stroke will be altered if a new concept for the pathophysiology of stroke evolves.

The etiology of stroke is multifactorial and it is not easy to grasp a full picture of the stroke mechanism. In 1998, Caplan and his associates proposed the interactive role of hypoperfusion and embolism in promoting brain infarction as opposed to the traditional concept that the two mechanisms are independent causes of stroke in patients with cerebrovascular disease [2]. They reviewed reports of emboli monitoring with transcranial Doppler ultrasonography and studied data from four different situations of brain infarction that may be related to hypoperfusion:

•	Extracranial and intracranial arterial occlusive disease
•	Cerebrovascular reserve with carotid arterial occlusive disease
•	Collateral circulation in patients with thrombolytic therapy
•	Cardiac surgery

Their findings and hypothesis are summarized as follows. Firstly, microembolism is common in patients with severe symptomatic carotid stenosis and during and after cardiac surgery. For severe stenosis, the increase in load of microemboli can be explained by flow reduction at the stenosis, stimulating thrombus formation and embolization. For cardiac surgery, microemboli are frequently produced during clamping and cannulation of the aorta, as well as at the beginning and end of cardiac bypass surgery [3]. Secondly, there was close correlation between brain infarction and severe stenosis [4,5], compromised cerebrovascular reserve with severe stenosis [6], ineffective collateral circulation even after thrombolytic therapy [7], and hypoperfusion during cardiac surgery [3].

Having realized the significant role of embolism in stroke, they suspected that increased load of microemboli alone could not be the cause for brain infarction since severe extracranial stenosis as an active source of embolism is usually well tolerated by patients with good collateral circulation and normal cerebrovascular reserve, irrespective of proximal obstructive lesions. Furthermore, the severity and duration of cerebral hypofusion during cardiac surgery is predictive of ischemic stroke after surgery, albeit the procedure is prone to producing emboli [3]. The reason why brain infarction is only associated with reduced collateral blood flow, limited reserve and hypoperfusion during cardiac surgery must be elucidated. Thus they hypothesized that hypoperfusion and embolism must be interplayed in the stroke mechanism in which poor collateralization and marginal blood flow reduces the washout and clearance of the emboli that have entered the vascular bed of ischemic areas and obstructed the blood supply to the vulnerable brain tissue leading to infarction.

They further suggested that the brain border zone is a favorable destination for microemboli, as it is the most remote portions of the brain circulation that favor deposition of microemboli. In addition, infarction in the border zone, as traditionally believed, is not always attributable to hypoperfusion because in necropsy studies microemboli are frequently found within arteries in border zone [3,8,9]. Border-zone infarction is typically bilateral and is a common finding in brain imaging following chronic hypoperfusion and after cardiac arrest. As hypoperfusion and embolism often coexist in severe arterial occlusive disease, interplay of the two mechanisms probably promotes and enhances brain infarction by impaired washout of emboli.

In the years that followed, Sedlaczek and his colleagues presented additional evidence to support the concept of impaired washout of emboli in vascular territories of hypoperfusion. These include:

•

Demonstration of lateralization of subcortical cerebral infarction in the hypoperfused, unoperated hemisphere after angiography, which is a well-known procedure for emboli production [10] in four out of five patients with previous unilateral cerebral revascularization due to moyamoya. The findings are probably related to reduced washout of emboli in the hypoperfused hemisphere;

•

Demonstration of subcortical–cortical infarcts characteristic of embolic events in two patients with active embolic sources and venous occlusive disease. It is of note that diffuse lesions are typical of venous obstruction. These atypical stroke patterns are probably attributable to impaired washout of emboli due to reduced perfusion pressure and poor venous drainage in the occlusive venous disease [11].

With better understanding of the interactive role of hypoperfusion and embolism in stroke, it is not surprising to learn that the benefit of carotid endarterectomy was observed in those patients with internal carotid artery stenosis greater than 70% in the North American Symptomatic Carotid Surgery Trial and the European Carotid Surgery Trial, as it is likely a consequence of the removal of the severe stenosis which is both a predominant cause of flow reduction and a source of embolism.

Caplan and his associates concluded that the most important evidence supporting the new theory of stroke pathophysiology is the frequent distribution of infarction in border-zone regions. In the interplay of hypoperfusion and microembolism the reduced perfusion and pressure decrease washout and throughput of emboli, especially in remote border-zone areas of the cerebral circulation. In their opinion, the typical border-zone distribution of infarction can only be explained with the complementary role of these two mechanisms. Under the interaction of the two mechanisms, brain infarction is inevitable [12].

I find this hypothetical concept of stroke pathophysiology intriguing and compelling. It not only rekindles the important role of hypoperfusion in stroke, which I think has been undermined, but also substantiates my observations in stroke patients regarding the extracranial arterial blood flow volume (BFV) which is defined as the total BFV of two common carotid arteries (CCAs) and two vertebral arteries. I used this flow parameter in my previous studies as an index of total cerebral blood flow. In my observations, increased or high extracranial arterial BFV is closely related to the presence of effective intracranial collateral circulation [13], positive cerebral vasoreactivity [14], good functional outcome after stroke [15] and low risk of recurrent stroke [16]. In the presence of extracranial carotid occlusive disease, a BFV of 370 ml/min or more in the CCA or 120 ml/min or more in the vertebral artery is indicative of intracranial collateral circulation. An increased BFV in the CCA after acetazolamide administration is a positive sign of cerebral vasoreactivity. A poststroke extracranial arterial BFV over 600 ml/min is predicative of good functional outcome in 6 months and is associated with low risk of recurrent stroke. The former two observations indicate that high extracranial arterial BFV is a reflection of the integrity of cerebral protective mechanisms against stroke. The latter two observations underline the benefit of high extracranial arterial BFV after stroke. These data are in accordance with the newly proposed hypothesis. Effective protective mechanisms, as reflected by high total cerebral blood flow, facilitate emboli washout and ameliorate the adverse effect of cerebral embolism.

The new theory of stroke pathogenesis and the observations on extracranial arterial BFV may shed new light on stroke management with regard to risk assessment, prevention and treatment strategies. It may help to identify high-risk subgroups of stroke patients with a potential source of embolism aggravated by low extracranial arterial BFV. For example, patients with AF have a fivefold increase in the risk of stroke due to cardioembolism [17]. It is likely that AF patients with low extracranial arterial BFV may bear a higher risk of stroke compared with those with a high level of BFV. Combination therapy of flow augmentation and antithrombosis treatment may be evolved in stroke prevention and treatment strategies. With antithrombotic treatment utilizing antiplatelet agents such as aspirin, aspirin plus extended-release dipyridamole or clopidogrel in the absence of a specific contraindication still remaining as the mainstay of medical treatment in secondary stroke prevention [18], augmentation of cerebral blood flow may be achieved by upregulating the endothelial expression and/or activity with 3-hydroxy-3-methylglutaryl-CoA reductase inhibitors (statins), steroid hormones, nutrients and physical activity [19] or by surgical intervention such as carotid endarterectomy and angioplasty with or without stenting. Large prospective studies are warranted in the future to assess the efficacy of different combination treatment regimes when the new concept of stroke pathophysiology is widely accepted by the clinical community.