White matter hyperintensities had a correlation with the cerebral perfusion level, but no correlation with the severity of large vessel stenosis in the anterior circulation

Abstract Objective The contribution of large vessel stenosis to the development of white matter hyperintensities (WMHs) has not been fully elucidated. This study aims to explore the correlation between ipsilateral white matter hyperintensities (WMHs) and the severity of large vessel stenosis in the anterior circulation and cerebral perfusion level, as well as analyze the factors influencing WMHs. Methods A cross‐sectional study of 150 patients with unilateral anterior circulation large vessel stenosis of ≥50% was conducted. The severity of ipsilateral WMHs was assessed by Fazekas scale on T2‐weighted image and/or fluid‐attenuated inversion recovery MR imaging, vascular stenosis severity was evaluated on computed tomography angiography images, and the level of cerebral perfusion was rated according to a staging system for abnormal cerebral perfusion based on CTP results. The relationships between the stenosis severity, cerebral perfusion level and ipsilateral WMHs severity were analyzed. A multivariate logistic regression analysis was performed to determine the factors independently influencing WMHs. Results Among 150 patients (mean age, 63.12 ± 10.55 years), there was a statistically significant positive correlation between cerebral perfusion level and the severity of DWMHs and PWMHs (Gamma = 0.561, p < .001; Gamma = 0.600, p < .001), and a positive correlation between cerebral perfusion level and the severity of vascular stenosis (Gamma = 0.495, p < .001).While, there was no statistically significant correlation between the severity of vascular stenosis and the severity of DWMHs and PWMHs (Gamma = 0.188, p = .08; Gamma = 0.196, p = .06). The multivariate logistic regression analysis results demonstrated that age (OR = 1.047, 95% CI 1.003–1.093; p = .035), stroke/TIA history (OR = 2.880, 95% CI 1.154–7.190; p = .023) and stage II of cerebral perfusion (OR = 2.880, 95% CI 1.154–7.190; p = .023) were independent influencing factors on ipsilateral DWMHs. Age (OR = 1.051, 95% CI 1.009–1.094; p = .018), and stage II of cerebral perfusion (OR = 12.871, 95% CI 3.576‐46.322; p < .001) were factors independently influencing ipsilateral PWMHs. Conclusion White matter hyperintensities may be attributed to cerebral hypoperfusion secondary to vascular stenosis but not directly to the severity of stenosis in the large vessels of anterior circulation. Moreover, longitudinal studies with sequential imaging exams may further reveal the impact of cerebral perfusion secondary to vascular stenosis on the development and progression of WMHs.

Although the pathogenesis of WMHs remains unclear, cerebral ischemia and hypoperfusion may be important mechanisms for the occurrence of WMHs (Wong et al., 2019). From a hemodynamics standpoint, intracranial and extracranial large vessels are located in the "upstream" of small cerebral vessels. The local abnormal blood flow caused by stenosis of large cerebral vessels will be transmitted to the distal end, which may lead to chronic hypoperfusion injury of small cerebral vessels and brain parenchyma, thus resulting in WMHs . The influence of the severity, distribution, and number of vascular stenosis on WMHs has been evaluated in many studies. The results demonstrate that WMHs are related to stenosis of large arteries, but this conclusion remains controversial (Baradaran et al., 2017;Duan et al., 2018;Nam et al., 2017;Park et al., 2015) .It is considered that hypoperfusion plays an important role in WMHs caused by vascular stenosis (Stewart et al., 2021;Wang et al., 2020;Wong et al., 2019), but there is a lack of research on the mechanism of cerebral perfusion after vascular stenosis affecting WMHs. CT perfusion (CTP) technique can be employed to semi-quantitatively obtain the blood perfusion volume of local brain tissues and accurately reflect the changes in the blood perfusion volume of brain tissues. Ren et al. (2021) and Yin et al. (2018) proposed a staging system for abnormal cerebral perfusion based on CTP results, which can distinguish different perfusion states and provide guidance for clinicians to select proper therapeutic regi-mens and evaluate the prognosis for patients with cerebral ischemia.
This staging system was adopted in this study to evaluate the cerebral perfusion level. In this study, we aimed to investigate the relationships between the anterior circulation large vessel stenosis severity, cerebral perfusion level and ipsilateral WMHs severity, we hypothesized that the severity of WMHs may be related to cerebral hypoperfusion secondary to vascular stenosis. This may improve our understanding of how vascular stenosis affects the presence or progression of WMHs and may provide an important theoretical basis for the prophylaxis and treatment of WMHs.

Research subjects
Patients who were admitted to the Neurology Department of the Affil-

Evaluation of anterior circulation large vessel stenosis
Head and neck CTA examinations were performed with 256-slice spiral CT (Philips iCT). The scanning was performed under the following parameters: Tube voltage of 120 kV, effective milliampere-second of 200 mAs, collimator width of 64 mm × 0.6 mm, the pitch of 1.0, rotation speed of 0.4 s, original image acquisition slice thickness of 1.0 mm, reconstruction slice thickness of 0.5 mm, and slice spacing of 0.5 mm.
Based on the calculation method related to warfarin-aspirin symptomatic intracranial disease (WASID) (Feng et al., 2021), the severity of stenosis for intracranial arteries = [1 -(the diameter of the artery at the most severe stenosis site/the diameter of the normal artery)] × 100%. According to the standard in the North American Symptomatic Carotid Endarterectomy Trial (NASCET) (Feng et al., 2021), the severity of stenosis for extracranial arteries = [(the diameter of the artery above the stenosis site -the diameter of nonstenosis at the most severe stenosis site)/the diameter of the artery above the stenosis site] × 100%. According to the severity of stenosis, the patients were assigned into three groups: moderate, 50% to 69%; severe, 70% to 99%; occlusion, 100%.

Evaluation of cerebral perfusion
The 256-slice spiral CT (Philips iCT) was used to perform scanning examinations. The 6-cm layer adjacent to the basal ganglia was

Statistical analysis
Statistical analysis was performed by the SPSS 26.0 software. Normally distributed data were expressed as mean ± standard deviation (x ± s), and an independent sample t-test was performed to compare these data between groups. The data that did not conform to a normal distribution were expressed by median (quartile). The classification and counting data were expressed by the number of cases (percentage), and the χ 2 test was performed to compare these data between groups. The relationships between the stenosis severity, cerebral perfusion level and ipsilateral WMHs severity were tested by Gamma correlation analysis, as they are both ordinal level variables. Logistic regression analysis enabled the analysis of factors influencing WMHs.

F I G U R E 1
The enrollment flowchart.
The Wilcoxon signed-rank test was applied to evaluate the differences in WMH scores between the ipsilateral hemisphere of vascular stenosis and the contralateral hemisphere. p < .05 indicated the threshold for a statistically significant difference.

Demographic and clinical characteristics of patients
According to the inclusion and exclusion criteria (refer to the enrollment flow chart for details; Figure 1), a total of 150 patients (with an average age of 63.12 ± 10.55 years old) were included in this study.  Table 2). Among them, the proportion of patients with a smoking history, hypertension, stroke/TIA history or stage II of cerebral perfusion in the moderate-severe DWMH group was significantly higher than in the nonmild DWMH group (p < .05) ( Table 2).
Additionally, patients with moderate-severe PWMH were older and had a higher rate of hypertension, stroke/TIA history, and stage II of cerebral perfusion than patients with nonmild PWMH (p < .05) ( Table 2).

Correlations of the ipsilateral WMH severity with the severity of anterior circulation large vessel stenosis and cerebral perfusion level
According to univariable analysis, there was no difference in the severity of stenosis and site of stenosis between patients in the nonmild DWMH and moderate-severe DWMH groups, as well as in the nonmild PWMH and moderate-severe PWMH groups (p > .05), and the proportion of stage II patients was significantly higher in the moderate-severe DWMH and moderate-severe PWMH groups than in the nonmild DWMH and nonmild PWMH groups (p < .05) ( Table 2). The relationships between the severity of vascular stenosis, cerebral perfusion level, and ipsilateral WMHs severity were tested by Gamma correlation analysis. The results showed that there was a statistically significant positive correlation between cerebral perfusion level and the severity of DWMHs and PWMHs (Gamma = 0.561, p < .001; Gamma = 0.600, p < .001), and a positive correlation between cerebral perfusion level and the severity of vascular stenosis (Gamma = 0.495, p < .001).
While, there was no statistically significant correlation between the severity of vascular stenosis and the severity of DWMHs and PWMHs (Gamma = 0.188, p = .08; Gamma = 0.196, p = .06). Figures 2 and 3 showed images of two cases.

Logistic regression analysis of factors influencing WMHs
In the univariate analysis, age, smoking history, hypertension, stroke/TIA history, Hcy, and cerebral perfusion level were strongly associated with ipsilateral DWMHs. Since p < .05 for all of these variables, they were included as covariates in the subsequent binary

Severity of WMHs on the ipsilateral and contralateral sides of vascular stenosis
The Wilcoxon signed-rank test on WMH severity between the ipsilateral and contralateral sides of the stenosis revealed that the scores of DWMHs and PWMHs were higher in the ipsilateral hemisphere of stenosis than in the contralateral hemisphere (p < .001 for both;

DISCUSSION
In this study, we investigated whether the severity of vascular stenosis and the level of cerebral perfusion were associated with ipsilateral WMH severity. The results revealed that there was a statistically significant positive correlation between cerebral perfusion level and the severity of DWMHs and PWMHs. However, the severity of vascular stenosis was not significantly associated with DWMH and PWMH severity. Logistic regression analysis further confirmed that cerebral hypoperfusion was an independent influencing factor on ipsilateral DWMHs and PWMHs.
In the past, WMHs were generally assumed to be associated with small vascular lesions in the brain. However, nowadays, more academics believe that there is a relationship between WMHs and large vessels in the brain. Wang  Existing studies on the relationship between WMHs and cerebral macrovascular stenosis focus on stenosis severity, stenosis distribution, and the number of stenoses, while few studies directly refer to cerebral perfusion. Bahrani et al. (2017) and Chen et al. (2018) reported that WMHs increased with decrements in the local cere- CBV is defined as the volume occupied by intravascular blood within a particular quantity of brain tissue (including capillaries and large vessels perfusion reserve. Similarly, TTP is defined as the time duration from the beginning of injection of contrast media to reaching the peak concentration in the ROI. The prolonged TTP is mainly caused by the slow blood flow or collateral blood supply (Tao et al., 2022;Yang et al., 2022). Comprehensive analysis of these parameters contributes to an in-depth insight into the blood supply in brain tissues. Based on the staging system proposed by Yin et al. (2018), abnormal cerebral perfusion can be divided into two stages. In Stage I 1 , there is a change in the cerebral blood flow velocity, but no compensatory dilatation of local cerebral microvessels. In Stage I 2 , there is compensatory dilatation in local cerebral microvessels. In Stage II 1 , there is a decrease in CBF and local astrocyte swelling due to ischemia, which begins to compress local microvessels. In Stage II 2 , there is significant astrocyte swelling, which compresses, narrows or blocks local microvessels, thus inducing impairment of the local microcirculation. Cerebral hypoperfusion may cause changes in downstream structures by inducing cell energy imbalance, oxidative stress, inflammatory reaction, endoplasmic reticulum stress, mitochondrial dysfunction and other mechanisms. This would result in endothelial dysfunction, damage of the blood-brain barrier (BBB), glial activation, and apoptosis, thus leading to the progression of WMHs (Huang et al., 2018;Rajeev et al., 2022).
In this study, we discovered that DWMHs and PWMHs were more severe on the ipsilateral side of the stenosis than on the contralateral side. This may be related to their anatomy since DWMHs are supplied by deep penetrating arteries branching from the intracranial soft meningeal arteries. These arteries have relatively short vascular pathways and it is difficult for ischemia to occur in this region when the collateral circulation is good or perfusion is normal. Thus, ischemic changes only happen when perfusion is inadequate or when cerebral blood flow self-regulation is reduced. Blood supply to the PWMHs is derived from the choroidal artery of the inferior ventricular artery or from the terminal branches of the striatum arteriosus, which has a sparse or absent anastomosis with vessels originating from the cerebral surface (Song et al., 2021). This study revealed that besides cerebral hypoperfusion, WMHs also correlated with age. These are consistent with the findings in some previous studies (Vedala et al., 2019;Wang et al., 2021). The changes in white matter blood supply caused by arteriosclerosis during aging may be one of the factors contributing to the strong correlation between age and WMHs, and most white matter nerve fibers are myelinated and their lengths shorten dramatically during normal aging, so aging can cause demyelination changes (Yu et al., 2018). Stroke/TIA history was significantly associated with ipsilateral DWMHs, probably because they share the same risk factors and are both associated with cerebral underperfusion (Holmegaard et al., 2018).
The current findings on the impact of cerebral hypoperfusion on the severity of WMHs may have clinical implications for the management of patients with vascular stenosis. Improving cerebral perfusion could be highly beneficial to patients with WMHs and cerebrovascular stenosis. It has been demonstrated in a prospective study (Chuang et al., 2011) that the cerebral perfusion and WMH severity of patients can be optimized after internal carotid artery stent implantation. In clinical work, patients with vascular stenosis and ipsilateral WMH should be given high priority, and we can perform further CTP examinations so as to provide a reliable basis for the treatment of WMHs. However, these conclusions necessitate further verification and validation in future prospective studies.
Limitations of our study: First, this study is a single-center retrospective study, which could not evaluate the causality of the association between cerebral perfusion and WMHs. Therefore, it is still necessary to conduct multicenter, large-sample and prospective studies to further verify the findings reported in this study. However, the results of the relationships between the severity of large vessel stenosis, the level of cerebral perfusion, and the severity of ipsilateral WMHs may provide important clues for the direction of future research in this area. Second, most patients in our study were symptomatic, a heterogeneous group with different presentations, so sample selection bias was inevitable, a multicenter study including both symptomatic and asymptomatic patients would reduce this error. Third, many patients without stenosis may also have WMHs, and since few patients without stenosis are clinically assessed for cerebral perfusion, patients without stenosis were excluded from this study. Our study has some strengths.
First, we combine CTA with CTP, and the vascular stenosis can be more fully represented morphologically and functionally. Second, two experienced radiologists assessed the severity of stenosis, WMH, and cerebral perfusion without knowledge of clinical demographics, which minimized the expected bias. Third, this is a retrospective study, and it is much easier and cheaper to conduct a review of medical records, and valuable information can often be found this way.
In summary, white matter hyperintensities may be attributed to cerebral hypoperfusion secondary to vascular stenosis but not directly to the severity of stenosis in the large vessels of anterior circulation.
We need to actively treat and intervene in factors that cause cerebral hypoperfusion in order to delay the development of WMHs. However, further investigations are required to verify these findings and identify potential pathophysiological mechanisms.

AUTHOR CONTRIBUTIONS
FF, RD, WK, and HY conceived and designed the study. FF, WK, and HY acquired the data. FF, RD, WK, and HY analyzed and interpreted the data. All authors contributed to the writing of the manuscript, read and approved the final version.