Introduction

Cerebral small vessel disease (CSVD) is a frequent neurological condition that affects the elderly, it leads to strokes, dementia, mood swings, and gait abnormalities. The diagnosis of CSVD depends on imaging which can include white matter hyper intensities, microbleeds and lacunar ischemic strokes. However, the CSVD features may share some common characteristics with other types of stroke as border zone infarcts [1]. Border zone infarcts are ischemic lesions that occur in specific areas at the watershed zones between two main arterial territories. The most common causes of internal border zone (IBZ) infarctions are either hypoperfusion or embolism [2]. Despite the different etiologies yet both disorders can have a similar clinical presentation and share a common radiological picture of white matter lesions. Moreover, if IBZ infarctions developed bilaterally they can be misdiagnosed as CSVD. Therefore, given that the management pertaining to acute intervention and secondary prevention differs among both disorders, then it is essential to find handy tools for differentiation [3].

The aim of this study is to find radiological characteristics that distinguish CSVD with lacunar infarctions from IBZ infarctions regarding vasomotor reactivity and its correlation with MRI perfusion patterns, disease severity and outcome.

Methods

Sixty patients were recruited from the stroke unit of the University hospital from August 2019 through July 2021. A written informed consent was obtained from all participants or first of kin and the study was approved by local research ethics committee FWA 000017585 in 2019. All procedures performed in the study were in accordance with the ethical standards of the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This is a prospective cohort observational study. The study recruited 60 patients divided into 2 groups, 30 patients with lacunar infarction and 30 patients with IBZ infarction, all patients were above 18 years of age. Patients were included in the first group if they had clinical signs of lacunar syndrome [4], no large vessel disease or cardioembolic disease. Patients with IBZ infarction in the second group were included if their MRI showed unilateral lesions in the internal border zone area with ipsilateral internal carotid stenosis (with or without contralateral stenosis). Patients were excluded if they had intracranial hemorrhage or other types of stroke as large-artery atherosclerosis, cardioembolic, cryptogenic stroke, stroke of other determined etiology. Patients with silent white matter disease only without actual infarcts, patients who had rTPA were also excluded. All patients were subjected to complete history taking including personal data, past history and risk factors, detailed history of current illness, complete clinical picture and drug history, general and neurological examination. Stroke severity was assessed by National Institutes of Health Stroke Scale (NIHSS) on admission and discharge [5]. The modified Rankin Scale (mRS) was assessed on discharge and 3 months later to follow up their prognosis [6]. MRI protocol was performed, to investigate regional variations in perfusion in both grey matter and white matter. All patients were examined using a 1.5-T MR scanner (Achieva and Ingenia, Philips medical system, Eindhoven, Netherlands). To give the patients contrast, an 18- or 20-gauge IV catheter was inserted in the antecubital fossa before scanning. During the first run of a standard dose (0.1 mmol/kg) bolus of gadopentetate dimeglumine, a series of 60 DSCMRI pictures were taken at 1-s intervals. Absolute perfusion parameters such as cerebral blood volume (CBV), cerebral blood flow (CBF), and time to peak (TTP) were calculated by standard algorithms [7]. Carotid duplex was conducted to detect internal carotid stenosis, and transcranial color-coded duplex (TCCD) using phase array 2.4 Hz probe for evaluation of vasomotor reactivity (VMR) by measuring breath holding index (BHI). BHI = {[mean flow velocity (MFV) at the end of breath-holding- resting MFV]/resting MFV} × (100/seconds of breath-holding) [8]. VMR was assessed twice, first upon admission and second time 3 months after discharge.

Statistical analysis

Data analysis was done using IBM SPSS software package version 25.0 (Armonk, NY: IBM Corp). The normality test was done using Kolmogorov–Smirnov test. Quantitative data were described in mean and standard deviation in normally distributed variables and median and range and inter quartile range (IQR) in non-normally distributed ones and investigated using t test and Mann–Whitney U test, respectively, while qualitative data were described in frequency and percentage and tested with Chi-square test. Wilcoxon signed-rank test was used in the comparison of 2 related samples. Spearman correlation coefficient was used to evaluate the correlation between different variables. The significance of the obtained results was judged at P < 0.05.

Results

This study included 16 (53.3%) males and 14 (46.7%) females in lacunar infarction group in addition to 17 (56.7%) males and 13 (43.3%) females in the IBZ infarction group. The age of patients of the whole sample ranged from 48 to 85 years, with a mean age of 63.47 ± 7.637 in lacunar group and 61.30 ± 13.483 in the IBZ group. Most of patients had dyslipidemia (80% in lacunar group and 86.7% in border zone group), hypertension (63.3% in lacunar group and 57.7% in border zone group), while the least common risk factor was ischemic heart disease (13.3% in lacunar group and 26.7% in border zone group. There was no significant difference between groups as regards age, risk factors and number of risk factors (Table 1).

Table 1 Demographic data, risk factors and laboratory findings
Full size table

Comparing the severity of clinical presentation on admission using NIHSS, it was found that patients with border zone infarction had higher disability than patients with lacunar infarction: median NIHSS (IQ) 12 [6] vs. 7 [6], respectively, (p = 0.001) and on discharge 10 (5.25) vs. 4 [4], respectively, (p < 0.001). Although having different severity of disability (NIHSS) on admission, yet on measuring within-group improvement, both groups improved significantly on discharge (p≤ 0.001 in each group). However, inter-group comparison showed that lacunar group still had more significant improvement than IBZ group (p = 0.001) (Table 2).

Table 2 Comparison of NIHSS and mRS between lacunar infarctions and borderzone infarctions
Full size table

As for good outcome (mRS ≤ 2) on discharge, 89.9% of lacunar group showed good outcome, vs. 36.7% of IBZ group (p ≤ 0.001). At 3 months follow-up, 96.7% of the lacunar group had a good outcome vs. 53.3% of the IBZ group (p ≤0.001).

It is worth mentioning that within-group comparison showed significant improvement after 3 months (p≤ 0.001) in lacunar stroke compared to (p = 0.02) in IBZ (Table 2).

By investigating VMR in both groups, 56.7% of patients with lacunar stroke had impaired reactivity vs. 100% of patients in IBZ group (p < 0.001). At 3 months follow-up, this number dropped to 23.3% in lacunar group, but persisted as 100% in IBZ group (p < 0.001).

There was statistically significant impairment in all parameters in patients with IBZ infarction than patients with lacunar infarction which means longer TTP, decreased CBF and CBV (p < 0.001, 0.001, 0.023, respectively).

Logistic regression for the lacunar group showed a negative correlation between VMR at baseline and after 3 months with NIHSS (rho = -0.540, p = 0.002, rho = − 0.636, p < 0.001, respectively), which denotes that good VMR correlated with less stroke severity.

There was also a positive statistically significant correlation between inter-side difference CBF and NIHSS (rho = 0.399, p = 0.029), which means that reduced CBF was associated with severer stroke. Three months later mRS did not show significant correlation with any of the studied parameters.

On the other hand, logistic regression analysis for IBZ group could not be performed for VMR since 100% of patients were impaired. Therefore, logistic regression was performed for MRI perfusion parameters vs. degree of stenosis and showed a positive statistically significant correlation between inter-side difference TTP and CBF with degree of stenosis (rho = 0.501, p = 0.005, rho = 0.421, p = 0.021, respectively), denoting that reduced flow is associated with higher stenosis. While no correlation was detected for CBV. There was also no significant correlation between MRI perfusion parameters vs. NIHSS and mRS.

Discussion

In the current study, both IBZ infarctions and lacunar infarctions shared the same risk factors leading to atherosclerosis, whether small vessel or large vessel atherosclerosis, which is key factor in their pathogenesis. Atherosclerosis can cause many changes in the arterial wall leading to obstruction of proximal lumen. If the occlusion occurred in perforating arterioles smaller than 50 µm in diameter, lacunar infarctions will result [9], while if it occurred in leptomeningeal vessels and carotid arteries together with systemic hypotension, it would result in ischemic lesions situated along the watershed zones [10]. Additionally, both disorders share the MRI picture of white matter lesions that might occur bilaterally whether symmetrical or non-symmetrical. So, we aimed to differentiate between these 2 types of ischemic stroke because IBZ infarction can be misdiagnosed as SVD especially if it occurred bilaterally causing white matter lesions on MRI. In this study, the frequency of vascular risk factors in a descending order in both groups; dyslipidemia, hypertension, diabetes, smoking and IHD, and even the glycemic level did not differ significantly between groups. Similarly, many studies reported dyslipidemia to be the commonest risk factor [11, 12], while others stated that hypertension was the commonest [13].

Regarding clinical disability, on comparing NIHSS and mRS of the 2 groups, IBZ group had higher NIHSS on admission than lacunar group. At 3 months follow-up both groups showed significant improvement of mRS than on admission, yet IBZ group showed significantly less improvement than lacunar group. Likewise, bad prognosis of IBZ infarctions and good prognosis of lacunar infarctions were described by many studies [14, 15]. On admission, VMR was impaired in more than half of the patients in the lacunar group and in all those in the IBZ group. At 3 months follow-up 33.4% of patients improved in the lacunar group while impairment persisted in all patients of IBZ group. This agrees with many other studies who stated that perfusion reserve and VMR were dramatically reduced in patients complaining of carotid artery occlusions [16, 17]. In another study, VMR was related to early neurological deterioration in patients with lacunar infarction [18], white matter hyperintensities also were associated with less cerebral blood flow and more impairment of VMR than normal brain tissue, moreover it was reported that reduced VMR can predict progression from normal white matter to white matter hyperintensities in small vessel disease [19]. The current study suggests that VMR impairment in lacunar group is significantly correlated with severer stroke on presentation. However, we were not able to perform a correlation analysis between VMR and clinical severity in IBZ group as all the patients showed VMR impairment.

On comparing the 3 parameters (TTP, CBF, and CBV) between the 2 groups, impairment in all parameters was more in IBZ infarction than in lacunar infarction which means longer TTP, decreased CBF and CBV. This agrees with various studies stating that ICA stenosis causes hemodynamic disturbance thus causing more impairment in MRI perfusion parameters [3, 20]. We could also detect a correlation between inter-side difference of TTP and CBF with degree of stenosis, as the degree of stenosis increased, blood flow decreased ipsilaterally and the difference in TTP and CBF between the 2 hemispheres increased. Yet no such correlation was detected with CBV, mostly because CBV in some cases may actually show increased value to maintain cerebral auto regulation [21]. A previous study of CT perfusion in ICA occlusion found that CBF was not significantly different from that of the contralateral hemisphere, whereas ipsilateral CBV and TTP were in comparison with those in the contralateral hemisphere, substantially higher [20]. Meanwhile we identified a significant correlation between an increased inter-side difference CBF and stroke severity in lacunar group; other studies employed different elements of measurement as mean transit time (MTT) to reach the same correlation [22]. No significant correlation was found between VMR and MRI perfusion parameters in lacunar infarcts mostly due to absence of severe hemodynamic instability, unlike that which results from carotid stenosis in cases of IBZ infarction [23].

The study had some limitations as MRI brain perfusion could have been preferably repeated at 3 months follow-up and compared with the initial MRI to test for the possibility of recanalization and its effect on perfusion and prognosis. However, that was not feasible due to limited resources.

Conclusion

Although IBZ infarctions can simulate the MRI picture of multiple lacunar infarcts, yet the presence of severe impairment of VMR or reduced perfusion in white matter lesions is more in favor of IBZ infarction. Therefore, patients with IBZ infarctions tend to have worse prognosis and higher clinical disability.