High resolution magnetic resonance imaging in pathogenesis diagnosis of single lenticulostriate infarction with nonstenotic middle cerebral artery, a retrospective study

Stroke is an important public health problem throughout the world with major mortality and severe morbidity. The accurate identification of stroke etiology is important for guiding patient management and prognosis [1]. Lenticulostriate infarction is ischemia in the territory supplied by the deep perforating branches of the middle cerebral artery (MCA). Lenticulostriate arteries supply the basal ganglia and most of the internal capsule [2]. There are two different vascular pathogenesis in single lenticulostriate infarction with nonstenotic MCA: 1) branch atheromatous disease (BAD), atheromatous plaque of MCA at the orifice of lenticulostriate arteries and 2) non-branch atheromatous small vessel disease, histologically characterized by lipohyalinotic degeneration of lenticulostriate arteries themselves [3]. BAD was proposed by Caplan in 1989 [4]. It was identified as a new pathogenesis caused by an occlusion or stenosis at the origin of a deep penetrating artery of the brain. Non-branch atheromatous small vessel disease is different from BAD showed lower mortality and higher risk of death when with anticoagulant therapy [1].

However, it is usually difficult to distinguish BAD from non-branch atheromatous small vessel disease for practicing clinicians. Conventional imaging techniques, magnetic resonance angiography (MRA), computed tomography angiogram (CTA), and digital subtraction angiography (DSA) can’t identify the two pathogenesis in nonstenotic arteries because those only show the information of the lumen, not the vessel wall.

High-resolution magnetic resonance imaging (HR-MRI) could display the features of arterial wall, and recent studies showed that it was also an effective tool for identifying plaque of intracranial arteries [5‐7]. In this study, we aimed to determine whether HR-MRI can identify early stage atherosclerotic plaque in patients with nonstenotic MCA on MRA, and investigate the clinical distinction of the two pathogenesis BAD and non-branch atheromatous small vessel disease.

It is usually difficult to distinguish BAD and non-branch atheromatous small vessel disease for practicing clinicians. Microatheromas have been reported to be the most common underlying mechanism of symptomatic lacunar infarction in autopsy studies [10]. However, traditional angiography examinations, such as MRA, CTA and DSA, cannot identify BAD. Recent studies have reported HR-MRI is an effective and directive tool for the evaluation of intracranial atherosclerotic plaques. In our study, we found that as many as 46.9% of acute lenticulostriate stroke patients with nonstenotic MCA on MRA demonstrated HR-MRI identified MCA plaques. This result is consistent with previous studies about MCA plaques in patients with lacunar infarction [11‐13].

Previously identification of BAD is mainly based on the neuroimaging features on traditional brain MRI. However, different studies used different methods to define BAD. Yamamoto et al. defined BAD of the lenticulostriate stroke as infarcts more than 10 mm in diameter on axial slice and visible on 3 or more axial slices [3]. Jeong et al. defined BAD as lesions visible on 4 or more axial slices [14]. Nah et al. considered that infarction lesion involvement of the lowest portion of the basal ganglia was an indication of BAD [9]. A study from China, 312 patients with lacunar infarction who had normal MCA on MRA enrolled, suggested that infarct lesion size may be used as a method to identify BAD [15]. According to the studies mentioned above, BAD tended to have larger infarcts and lesions were closer to the orifice of the perforating artery. However, these studies lacked directive atherosclerotic plaques evidences on HR-MRI. In our study, we showed that patients with HR-MRI identified plaque had larger infarction lesions (P < 0.05) and tended to present proximal lesions (P = 0.055) and ≧3 lesion slices (P = 0.055), which was consistent with the past studies about imaging features of BAD. Lenticulostriate artery originate from the initial segment of MCA, BAD tend to lead large proximal infarction lesion we think it was caused by plaque within parent artery (MCA) blocking the lenticulostriate artery orifice (Fig. 3a). But we can’t directly show the plaque and the orifice of perforator branches in HR-MRI due to the perivascular space or Virchow-Robin spaces. Kim et al. found that located plaque, and plaque length no related to large lenticulostriate infarction. And considered that it was due to the variation of branching pattern of perforators, which occlusion of a single lenticulostriate artery produced a variety of infarct sizes [16, 17]. In fact, both of the two pathogenesis can lead to the large infarction, we need a better way to show the details of lenticulostriate artery from orifice to ending, which can be used to distinguish the two pathogenesis.

However, the frequencies of proximal infarction lesions were not significantly different between patients with plaque and without plaque. We speculated that this result was associated with some patients had microatheroma in the orifice of the lenticulostriate artery itself (Fig. 3). In this hypothesis, patients could have proximal infarction lesions, but without HR-MRI identified MCA plaque. Different from a group lenticulostriate arteries occluded by MCA atherosclerotic plaque, this group patients could present small infarction lesions. So we furtherly speculated that microatheroma in the orifice of the lenticulostriate artery itself was the possible pathogenesis of small proximal infarction lesion. Unfortunately, microatheroma in the orifice of the lenticulostriate artery itself could not identified directly by using HR-MRI. Besides, the branching patterns of MCA perforators vary greatly in terms of type [18, 19] and infarction size and location may also be related to these variations.

What’s more, in our study 10 patients’ infarction lengths were more than 20 mm which can defined as large lenticulostriate infarction, and of these 6 patients with HR-MRI identified plaque. In the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) criteria, lenticulostriate infarction which length less than 15 mm was belong to the smell-vessel occlusion subtype [20]. But several studies found nearly half of patients with small vessel disease suffered infarction more than 15 mm [16], and small vessel disease can lead to an infarction with length even more than 20 mm because of occlusion of multiple perforating arterioles [21]. New ischemic stroke classifications included lesions of up to 20 mm in diameter [22] or even abandoned the rule of length of infarction [23, 24].

Theoretically, non-branch atheromatous small vessel disease is more strongly associated with hypertension and leukoaraiosis. Nevertheless, Yamamoto et al. found that there were no significant differences in prevalence of hypertension, diabetes mellitus, and hyperlipidemia between non-branch atheromatous small vessel disease and BAD [3]. In our study, we showed higher prevalence of diabetes mellitus in patients with MCA plaques than patients without MCA plaques. However, there were no differences in terms of other risk factors of atherosclerosis and significant leukoaraiosis between two groups.

Microanatomy studies suggested that most of penetrating arteries arose dorsally from the upper part of MCA wall [19]. We subdivided patients with MCA plaque into two groups: plaque location involving dorsally upper part and plaque location not involving dorsally upper part. However, there were no significant differences in terms of imaging features and severity of stroke between the two groups. We attributed this partly to the limited number of enrolled patients in our study and partly to the variations of branching patterns of MCA perforators [19].

Pathogenesis diagnosis of ischemic stroke is important for patient management. However, it is usually difficult to distinguish BAD from non-branch atheromatous small vessel disease for lenticulostriate infarction patients with nonstenotic MCA by using traditional imaging examinations. In our study, we demonstrated high frequency of MCA atheromatous plaques visualized on HR-MRI in patients with nonstenotic MCA on MRA. Patients with HR-MRI identified plaque presented larger infarction lesions and more proximal lesions than patients without plaque, which were consistent with imaging features of BAD. HR-MRI was an important and effective tool for identifying stroke etiology, especially for patients with normal MCA by MRA. We strongly suggest that HR-MRI is particularly important for large proximal lenticulostriate infarction patients with nonstenotic MCA.