Clinical features and imaging markers of small vessel disease in symptomatic acute subcortical cerebral microinfarcts

Cerebral microinfarcts (CMIs) are small lesions presumed to be of ischemic origin [1]. The lesions, small and often not visible to the naked eyes on autopsy, have been found incidentally in older individuals with cognitive impairment or other manifestations of cerebrovascular disease, indicating damage to brain structure and a poor clinical outcome [2, 3].

CMIs in the acute stage can be detected as small incidental diffusion-weighted imaging (DWI) lesions. DWI has been verified with high sensitivity in detecting very small infarcts, including those 1–2 mm in diameter in any brain parenchymal location [4]. Due to the blooming effects, the maximum diameter of acute DWI might overestimate the true infarct size [5]. The defined diameter of acute CMI varied among studies, from a few mm to ≤ 10 mm [6]. The majority of studies on incidental small DWI lesions did not report their actual size and usually adopt a cut-off with a maximum axial diameter of less than 5 mm [7, 8]. Based on available evidence, experts propose a size criterion of less than 5 mm for an acute cerebral microinfarct (A-CMI) [1]. However, confusion arises when diagnosing a small DWI lesion in the sub-cortex of ischemic stroke patients. If perforating arterioles are affected, it will cause a recent small subcortical infarct (RSSI), formerly termed ‘acute lacunar stroke’. The STandards for ReportIng Vascular changes on nEuroimaging (STRIVE) criteria for RSSI applied an upper size cutoff of 20 mm, a criterion designed to identify lacunar infarcts in their acute stage, without providing a lower size cutoff [9]. As currently defined, RSSI does not have a lower size boundary, and the smallest DWI infarcts, which we term acute subcortical cerebral microinfarcts (As-CMI), might have clinical implications distinct from RSSI imaging characteristics of small DWI hyperintense lesions in vivo have been reported in patients with cerebral amyloid angiopathy, stroke, or memory loss [10‐14]. These studies reported that the small DWI hyperintense lesions were associated with magnetic resonance imaging markers of cerebral small vessel disease (SVD), such as white matter hyperintensities, cerebral microbleeds, and enlarged perivascular spaces, but the association between different sizes of small subcortical DWI lesions and SVD neuroimaging markers has not been thoroughly investigated. Besides in the sub-cortex, the acute microinfarct could act as lacunar stroke syndrome, but whether it has similar symptoms/signs distribution or better outcome compared to larger RSSI is still unknown. A better understanding of the varied clinical features of As-CMI and larger RSSI could extend emerging literature from previous studies and provide further evidence concerning the diagnostic classification of acute SVD.

Hence, we aimed to clarify the distinct characteristics of As-CMI as compared to the larger size of RSSI regarding vascular risk factors, clinical manifestation, radiological markers of SVD distribution, and outcomes.

Of the 584 patients finally included, 23 (3.9%) were defined as As-CMI. Five patients in the Larger RSSI group had multiple DWI small lesions.

To the best of our knowledge, the present study is the first to examine As-CMI in a large series of consecutive patients in whom DWI has also shown RSSI. In our current study, we found that As-CMI was identified in 3.9% of RSSI patients. We found a similar distribution among risk factors, common symptoms/signs as well as lesion location. Secondly, we assessed the association between As-CMI and SVD markers and observed that the As-CMI was associated with lacunes and a higher total SVD burden. Thirdly, we demonstrated that patients with As-CMI did not have a better functional outcome compared to patients with Larger RSSI.

The mean age in our study was 62.2 years which is congruent with previous reports on the Asian race [23, 24]. Nonetheless, the mean age of previous reports on subcortical lacunes in the Caucasian race was higher [25, 26]. This may be due to the different population study design (i.e. in terms of age range), classification of lacunes and/or lacunar infarction, and race. In addition, a higher frequency of a history of coronary artery disease (CAD) was observed in subjects with As-CMI compared to the larger RSSI group. The epidemiology of SVD and its consequences on the heart is not well understood [27]. It could be explained by a higher mean age in the As-CMI group that could cause an age-related high incidence of CAD. Furthermore, the vascular anatomy of the heart and the brain is quite similar, with arteries on the surface and penetrating arteries that provide tissue perfusion [28]. Thus, further studies are required to investigate the association between small vessel disease in the heart and brain and CAD.

In our series, dichotomization of RSSI according to the axial diameter (< / ≥ 5 mm) did not result in quite different regional distributions and symptoms/signs frequencies. In patients with symptomatic As-CMI, nearly half of the lesions were located in the basal ganglia, followed by the thalamus, centrum semiovale (CSO), and brainstem. In addition, hemiparalysis was the most common neurological deficit, followed by central facial/lingual palsy and hemidysesthesia. These are common symptoms in acute lacunar stroke according to the widely used OCSP classification [19]. In other words, the clinical symptoms/signs poorly discriminate the size of subcortical infarcts. Our findings showed that even the small lesion size of As-CMI could cause overt neurological symptoms. Noticeably, As-CMI could also contribute to other low-frequency symptoms such as dysphagia, aphasia, or ataxia.

Another novel finding was the association between subcortical As-CMI and the presence of multiple lacunes and a high total SVD burden. Previous studies reported a higher SVD burden in patients with RSSI or As-CMI who presented with possible vascular cognitive impairment [7, 29]. However, the relation between different RSSI sizes and SVD burden was not explored. As-CMI located in the white matter mostly occurs in relationship to distal branches of perforating arteries [30]. The increased frequency of lacunes and SVD burden in As-CMI showed that As-CMI might be a significant sign of active small vessel disease. The anatomical and hemodynamic characteristics of these smaller and distal branches of medullary or lenticulostriate arteries may partly explain an increased vulnerability of these territories to hypoperfusion injury and other pathophysiological mechanisms such as arteriosclerosis, endothelial injury, or blood–brain barrier dysfunction [30, 31]. Interestingly, in a novel whole-brain vessel-wall MRI study, patients with symptomatic single subcortical infarction but without relevant MCA disease were included. Although lesion size was not measured in the study, it is found that superiorly distributed MCA plaques at the lenticulostriate arteries origin (LSA) are associated with morphological changes in the LSA [32]. Whether the smaller size of the lesion would reflect worse morphological changes in the supplying artery and cause heavier SVD damage needs to be verified.

In long-term follow-up, our data did not show that As-CMI patients had a better functional outcome and recovery. The mRS score we applied is good at measuring the degree of disability or dependence in the daily activities of patients after stroke. However, the majority of lacunar infarcts would cause minor stroke-related deficits. Among lacunar strokes, the mRS might not be a subtle tool to evaluate the degree of rehabilitation. A previous study indicated brain frailty was associated with a worse cognitive score with a stronger effect in lacunar stroke [33]. As-CMI was found to increase odds of lacunes and SVD burden, implying a correlation with a worse cognitive outcome that merits further investigation.

Our study has several strengths such as the large hospital-based sample with MR-DWI, comprehensive evaluation of clinical symptoms/signs, and the assessment of SVD markers and outcomes. However, we would like to acknowledge some limitations. First, there is an ongoing discussion on whether different RSSI lesion sizes are associated with a distinct etiology. Thus, the present study excluded patients with potential embolic sources or undetermined causative mechanisms to focus on small vessel arteriopathy. Second, cerebral microbleeds were not included in the total SVD burden evaluation since a large proportion of patients did not undergo gradient-echo T2*-weighted imaging. Third, because of its retrospective nature, we cannot avoid selection bias. Also, the number of As-CMI is relatively small and until now the evaluation of imaging markers of As-CMI and SVD are mainly based on neurologists.

In conclusion, we add to the increasing body of evidence that patients with As-CMI had a non-specific clinical profile but a higher burden of SVD, indicating As-CMI might be s sign of more severe small vascular injury. Whether its vascular features are associated with worse cognitive outcomes requires further investigation.