Borderzone Infarction and Small Vessel Disease in a Sample of Egyptian Stroke Patients: Differences and Similarities
Keywords: Cerebrovascular diseases and cerebral circulation, cerebral Infarction, magnetic resonance imaging (MRI), perfusion, strokeKey Message: Borderzone infarctions could be clinically and radiologically differentiated from small vessel disease. This distinction is required in order to provide patients with the appropriate treatment.
Cerebral infarction is clinically categorized into several subtypes on the basis of size and location of the affected cerebral arteries and on pathogenesis. In the present study we are mainly concerned with border zone infarctions (BZI) and small vessel disease (SVD).
The anatomical location and general outline of white matter lesions of SVD are apparently similar to those of BZI. Both disorders show white matter hyperintensities (WMH) on MRI FLAIR sequence. And despite this apparent similarity, yet the underlying pathologies are quite different and thus the treatment is also supposed to be different.
Border zone infarcts manifest as WMH and are considered as ischemic lesions that occur at the junction between two main arterial territories. Hemodynamic impairment (HDI) as an underlying pathology has been widely accepted as a cause of BZI, but micro-emboli may also contribute to it. BZI might be associated with gradually progressive asymptomatic proximal arterial stenosis/occlusion and are commonly accidentally seen on MRI.
On the other hand, WMH of presumed vascular origin that are commonly seen in FLAIR sequence of MRI in older individuals are referred to as SVD. Demyelination, loss of oligodendrocytes and axonal damage, in addition to leakage of fluid from an impaired blood–brain barrier (BBB) were reported.
Hence, the current study aims at finding clinical and radiological points of differentiation between the two vascular disorders in a sample of Egyptian patients. This might have an impact on primary and secondary preventive measures.
This is a cross sectional (observational) study. Approval for conducting this study was obtained from Ain Shams University faculty of medicine research ethics committee.
A total of 52 patients, were recruited over a 2-year period, from the both stroke units of Ain Shams University hospital and Ain Shams Specialized hospital in Abbasia Square, Cairo, Egypt. All patients presented with acute or subacute ischemic stroke and their MRI showed cerebral white matter changes. Verbal informed consent was obtained from each subject prior to recruitment as indicated by the local ethics committee.
Patients with recent ischemic stroke, within 1 month, having cerebral white matter changes based on MRI.
All patients with cerebrovascular accidents who proved to have a territorial cerebral infarction as well as patients with border zone infarction in the posterior circulation of the brain.
Patients were considered to have BZI if their MRI showed periventricular WMH lesions, in the presence of extra or intra cranial arterial occlusion or stenosis of hemodynamic significance as shown by carotid duplex or MRI (≥50% stenosis). And were categorized as SVD if they had periventricular WMH lesions with no evidence of extra or intra cranial vascular stenosis or occlusion. Only diffuse atherosclerosis in MRA was accepted.
The following was done:
Relative difference of TTP among ROI was calculated by subtraction of TTP of each two homologous regions of interest.
WMH were graded according to Fazekas et al. and microbleeds by susceptibility weighted image (SWI) were classified according to the grading scale presented by Lee et al.
Statistical analysis was done on a personal commuter using SPSS© version 16th version Statistics (SPSS Inc., Chicago, USA). The Kolmogorov-Smirnov goodness of fit test was performed to test the normality of numerical data distribution. Continuous data were presented as mean and SD, whereas categorical data were presented as frequencies. For normally distributed numerical data, the independent-samples (unpaired) Student t test was used to compare the difference in the means between patients. For multiple intergroup comparisons, one-way analysis of variance (ANOVA) was used with application of the Scheffé test post hoc whenever a statistically significant difference was detected with one-way ANOVA. The Pearson Chi square test was used for comparison of categorical data. Fisher's exact test was used if >20% of the cells in any crosstabulation had an expected count of ≤5. P < 0.05 was considered statistically significant.
Forty-six patients completed the study: 29 with SVD and 17 with BZI. Demographic data showed significantly higher number of diabetic patients in BZI group (p = 0.047). Smoking, hypertension and recurrent stroke were more in BZI but not reaching significant difference [Table 1].
NIHSS was higher in BZI denoting more severe deficit than SVD (p = 0.004) [Table 1]. Clinical parameters showed higher incidence of syncope, limb shaking, and retinal claudication in BZI (47.1%) versus (20.7%) in SVD, being statistically significant for limb shaking (p = 0.049) [Figure 1].
BZI group was relatively more dyslipidemic than SVD on all parameters of lipid profile, not reaching statistical significance.
C-reactive protein was high in 50% of SVD and only 25% of BZI patients.
Radiologically, DWI showed more lacunar pattern in SVD while rosary pattern was commoner in BZI (p = 0.04). Symmetry of FLAIR lesions was highly significant in SVD as compared to BZI (p < 0.001). Microbleeds were observed in SVD significantly more than BZI (p = 0.048) [Table 2].
Perfusion study showed time to peak TPP difference significantly delayed in BZI group compared to SVD in all regions of interest denoting widespread decreased perfusion [Table 3]. [Figure 2] and [Figure 3] show radiological findings of two patients.
Aiming to highlight the differentiating points between SVD and BZI, this study showed that BZI has higher prevalence of vascular risk factors i.e., smoking, DM, hypertension, and recurrent strokes, than SVD. Similarly, Potter et. al., found age, hypertension, and diabetes to be commoner in what they called “asymmetric white matter lesions than in symmetric cases”. We suppose the terms they used stand for BZI and SVD.
These are the risk factors for atherosclerosis that might have led to arterial stenosis or occlusion associated with BZI thus leading to misery perfusion. Episodes of hypotension in the setting of arterial stenosis can result in those regions of low perfusion pressure that are prone to acute ischemia., resulting in BZI seen in watershed areas.,
C-reactive protein was higher in SVD, this might denote an underlying inflammatory process as previously suggested.
The typical symptoms of hypoperfusion as syncope, limb shaking, and retinal claudication, were found more in BZI, with limb shaking more significantly seen.
Syncope is a rare presentation of ischemic stroke in general, and although one of the manifestations of this hemodynamic compromise, yet it is uncommonly reported in this context., Kashiwazaki et al. in 2005, reported nine patients, and Welsh et al. 2012 described only 7 out of 204 patients presenting with syncope and having MRI of acute borderzone infarctions and occlusive carotid artery disease.,
Syncope at the onset is not routinely asked about, but in the present study it was specifically asked for, and this might explain the relatively higher incidence in this group of patients. It can denote widespread bilateral cerebral hypoperfusion at the onset of stroke.
Limb shaking was the commonest to be reported in BZI of the three studied symptoms. Movement disorder is not a usual manifestation of acute stroke. In the Lausanne stroke registry, 1% of stroke patients showed movement disorders.
In the current study, it was found in both groups, since SVD and BZI involve brain regions that might cause interruption of functionally related basal ganglia–thalamus–cortex circuits. and that might explain presence of limb shaking in both groups.
Some clinicians considered this symptom as a form of focal clonic symptomatic epilepsy,, but this was refuted by others who ascribed it to transient ischemia in patients with internal carotid occlusion.,
Retinal claudication is supposed to result from retinal hypoperfusion. It was the least reported symptom in both groups. It has been previously stated that 43% of patients presenting with transient monocular blindness had more than 70% ipsilateral carotid stenosis.
Rosary pattern was more common in BZI. It is considered to be the most sensitive indicator of hemodynamic failure according to Maddula et al. 2017. This hemodynamic impairment was further emphasized by our finding of widespread hypoperfusion in BZI.
Earlier studies reported similar findings of rosary pattern in internal BZI as compared with external BZI.,
On the other hand, lacunae were more prevalent among SVD. Lacunes are among the characteristic features of SVD as described by Wardlaw et al. 2013.
Patchy lesions were not a point of differentiation as they were nearly equally present in both groups.
FLAIR sequence showed highly significant preponderance of symmetrical lesions in SVD. The pathophysiology of SVD might explain this symmetry. SVD is rather a diffuse disorder involving cerebral microvasculature, leading to oxidative stress-induced endothelial dysfunction and alterations in the permeability of the BBB.
MB were found in all cases of SVD, with higher grades (2 and 3), reaching 62% of cases. While BZI showed significantly less MB. MB were reported to be one of the pathological features of SVD.
MB ranged from 25.7% to 62% in cases of lacunar infarcts in different studies, and were considered an indicator of advanced small artery disease of the brain.,
Perfusion was significantly impaired at all ROI in cases of BZI as compared to SVD. The internal border zone area lies in the territory of perforating medullary arteries from the pial arteries that are the most distal branches of the ICA, in addition, lenticulostriate arteries receive few collaterals thus perfusion pressure is expectedly low in this region.
So, in the presence of proximal arterial stenosis, misery perfusion has been described to evolve in serial stages starting by compromised cerebrovascular autoregulation and culminating to end-organ failure and stroke.
The term WMH has been used in some studies to describe SVD and BZI with no clear discrimination. That is why cerebral blood flow in cases of white matter hyperintensities or leukoaraiosis showed controversial results among different studies. When it was within normal limits, WMH were ascribed to microvascular abnormalities or SVD and if hypoperfusion was present then BZI was considered.
Finally, we can conclude that differentiating between BZI and SVD has an important impact on proper management. The general outline is common for both conditions, including treatment of modifiable risk factors and secondary prevention by antiplatelet agents. However, when it comes to targeting the underlying pathology, treatment strategies might differ. Regarding BZI, causes of brain hypoperfusion need to be treated with carotid endarterectomy or carotid artery stenting in case of significant carotid stenosis. Also, we have to be very cautious with lowering blood pressure as this might predispose to further BZI.
On the other hand, there are no established therapeutic guidelines for either preventing or treating SVD. Similar to other stroke subtypes, thrombolysis, antiplatelets, and statins are currently used for treatment of patients with SVD. However, it has been reported that bleeding risk increases in patients with extensive WMH and MB after thrombolysis therapy. Future studies are required to weigh the benefits and risks of thrombolysis in these patients. As for secondary prevention, although the risk of bleeding with dual antiplatelet therapy (aspirin plus clopidogrel) was identified with SVD, yet still the appropriate antiplatelets were not determined. The efficacy of drugs such as cilostazol or triflusal, which are known to be associated with less frequent bleeding than aspirin, need further investigation.
In the current study, we could conclude that a differentiation between SVD and IBZI could be made on clinical and radiological basis. A history of limb shaking, retinal claudication or syncope, with DWI showing asymmetrical rosary pattern of WMH, few microbleeds and markedly impaired perfusion favors the diagnosis of BZI. While the presence of raised C-reactive protein, lacunae, symmetrical WMH and multiple microbleeds, with minimal or no perfusion deficit goes with the diagnosis of SVD.
Declaration of patient consent
The authors certify that they have obtained all appropriate patient consent forms. In the form, the patient(s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3]
[Table 1], [Table 2], [Table 3]