Cases were selected from the Department of Neurosurgery, Shengjing Hospital of China Medical University. The degree of ICAS and the imaging manifestations of main intracranial and extracranial vessels were determined by CTA examination. The stenosis site of internal carotid artery (ICA) was located at the bifurcation of common carotid artery (CCA) in enrolled cases. The degree of ICAS was calculated according to the North American Symptomatic Carotid Endarterectomy Trial (NASCET) Steering Committee (1991) criteria. Cases with unilateral ICAS that had an ipsilateral stenosis degree ≥50% with a contralateral ICAS degree <50% and bilateral stenosis (cases with bilateral ICAS ≥50%) were included, where a difference in the degree of stenosis ≥10% was defined as a severe stenosis (on the worse side), and <10% was defined as similar bilateral stenosis. Cases with intracranial vasogenic diseases, such as moyamoya disease, vascular malformation, and intracranial aneurysm, were excluded. A total of 97 cases of ICAS between June 2019 and June 2021 met the criteria, which included 75 unilateral and 22 bilateral ICAS cases. This study was approved by the Ethics Committee of Shengjing Hospital of China Medical University.

All enrolled patients underwent TCD for the detection of major intracranial and extracranial vessels that included the CCA, ICA, middle cerebral artery (MCA), A1, posterior cerebral artery (PCA), ophthalmic artery (OA), supratrochlear artery (StrA) and siphonage of internal carotid artery (CS). A digital dual-channel, multi-depth TCD instrument and pulse Doppler probe (frequency 1.6 and 4.0 Hz; MVU6203, Delica, China) were used. Bilateral MCAs, A1 branches and PCAs were detected in the supine temporal window, and bilateral OAs and CSs were detected in the orbital window (detection depth: MCA, 40–75 mm; A1 branch, 60–75 mm; PCA, 55–75 mm; OA, 40–50 mm; CS, 40–75 mm). Bilateral CCAs, proximal ICAs, and StrAs (at the inner canthus) were detected in the supine position with a 4.0 Hz probe. In addition, cervical compression was used to determine the opening of the ipsilateral posterior communicating artery (PcoA). The blood flow direction, systolic and end diastolic blood flow velocity were measured, and the mean blood flow velocity (Vm) and pulse index (PI) were calculated (blood flow direction: toward the probe, CCA, MCA, PCA, OA, StrA and CS; deviate away from probe: ICA and A1).

The A1 was classified into three categories according to the imaging characteristics of bilateral A1 in CTA, namely: symmetrical A1; unilateral slender A1 (the diameter of the ipsilateral A1 is less than two-thirds of that of the contralateral one); and unilateral A1 that was not visualized on imaging.

The unilateral and bilateral stenosis cases were first grouped according to the degree of ICAS that was classified according to the CTA by NASCET criteria. Unilateral stenosis cases were classified into five groups according to the degree of stenosis that were 50–69%, 70–79%, 80–89%, 90%–near complete occlusion, and complete occlusion. According to the group, the Vm, blood flow direction, PI value, wall shear stress (WSS) of the A1, and imaging manifestations of A1 on both sides were recorded. Finally, based on the imaging manifestations of A1, the proportion of the degree of ICAS under different manifestations of A1, the Vm of A1 and MCA on both sides, blood flow direction and PI value were recorded.

When the patients were grouped according to the degree of ICAS and A1 performance, the age of the patients, Vm of each vessel and PI values were represented by the mean ± standard deviation. The paired sample t-test (two-tailed) statistical method was used to calculate the statistical differences in Vm and PI values between the ipsilateral stenosis vessels and the contralateral vessels. Multiple comparisons between groups were performed using the Bonferroni method. The Chi-square independence test was used to compare the differences in the case distribution of ICAS in different manifestations of A1. SPSS 26.0 software was used for statistical analysis, and statistical significance was considered with a P value < 0.05.

The average age of the 97 cases enrolled was of 64 ± 7 years, with 82 males and 15 females. Of the 75 unilateral ICAS cases, 15 were classified into in the 50–69% degree of stenosis group, 10 in the 70–79% group, 22 in the 80–89% group, 19 in the 90%–near complete occlusion group, and 9 in the complete occlusion group. Of the 22 bilateral ICAS cases, 16 were considered unilateral severe stenosis and six were similar bilateral stenosis (Table 1).

In the case of the unilateral ICAS, difference in hemodynamics of the ICA bilaterally was obviously related to the improvement of the degree of stenosis. In intracranial vessels, the hemodynamic difference of A1 bilaterally, that initially increased but then decreased with the increased degree of ICAS, was observed to be the most statistically significant (Figures 1, 2). When the degree of stenosis was 50–69%, there was no statistically significant difference observed in the Vm of A1 bilaterally (P > 0.05). When the degree of stenosis was 70–79%, the Vm of the A1 on the stenotic side was lower than that of the contralateral side (P < 0.05). When the degree of stenosis was 80–89%, the difference observed in the Vm of A1 bilaterally was the most obvious (P < 0.001), and the direction of blood flow was toward the probe in 15% of these patients. When the degree of stenosis was 90% to near complete occlusion, the difference observed in the Vm of A1 bilaterally decreased (P < 0.05). In 31.58% of these cases, the direction of blood flow in the A1 on the stenotic side was toward the probe. When an ICA was completely occluded, there was no statistical difference observed in the Vm of A1 bilaterally (P > 0.05), and the direction of blood flow in the A1 of the stenotic side was toward the probe in 33.33% of these cases. It should be noted that when the degree of internal carotid artery stenosis reached over 80%, the PI value of the stenotic MCA decreased when compared with that of the contralateral side, and the opening of the OA and StrA secondary collateral circulation gradually increased. When the ipsilateral internal carotid artery was completely occluded, the OA on the stenotic side of all enrolled cases showed a reverse direction of blood flow. In addition, we found that as the degree of stenosis increased, the proportion of ipsilateral PcoA opened and supplied blood to the anterior circulation increased gradually (Figure 3A). When multiple comparisons were made between groups according to the degree of stenosis, the significance level of the mean difference of A1 was obvious (Table 2).

In the case of the bilateral ICAS, Vm in the A1 of the stenotic more severe side was significantly lower than that of the contralateral side (P < 0.001), except for the difference in the hemodynamics when the CCA and ICA were compared. In 31.25% of the cases, the direction of blood flow of the A1 on the stenotic side was toward the probe. In 56.25% of the cases, the PcoA on the worse side was open and supplied to the ipsilateral anterior circulation. In addition, both the OA and StrA secondary collateral circulation showed varying degrees of opening in cases with both severe and relatively mild stenosis (Figures 4A–C). However, there was no statistically significant difference in the hemodynamics of bilateral intracranial and extracranial vessels in the group with similar degrees of bilateral stenosis (Figures 4D–F).

In addition, we calculated and compared the WSS (Poiseuille’s Law) (Malek et al., 1999) of A1 on both sides. As shown in Figures 5A,B, in cases of unilateral ICAS, although the ratio of WSS of A1 on the stenotic side to that of A1 on the contralateral side increased with the increase of stenosis degree, the difference was not statistically significant (p > 0.05). In cases of the bilateral ICAS, the WSS of A1 on the side with more severe stenosis was lower than that on the other side (p < 0.05) (Figure 5C), while in cases with similar bilateral stenosis, there was no significant difference in the WSS of A1 on both sides (p > 0.05) (Figure 5D).

In cases of unilateral ICAS, the manifestations of A1 observed on CTA bilaterally varied according to the degree of ICAS. When the degree of stenosis was 50–69%, most A1 images were symmetrical. When the degree of stenosis was 70–79%, bilateral A1 symmetry observed on imaging decreased. When the degree of stenosis was in the range of 80–89%, majority of A1 on the stenotic side presented as slender or non-visualized. When the degree of stenosis was 90% to near complete occlusion, the proportion of A1 on the stenotic side that were slender or non-visualized was 68.42%. However, when the ICA was completely occluded, the proportion of occluded A1 that presented as slender or non-visualized was 77.78% (Figure 3B). In the bilateral ICAS cases, 68.75% of A1 on the severe stenotic side presented as slender or non-visualized (Figure 4A), while symmetric imaging was mainly observed in cases with similar degrees of bilateral stenosis (Figure 4D).

As per the CTA manifestations of the unilateral A1 on the stenotic side, hemodynamic changes of A1 and MCA on both sides were observed. Among them, there were 22 cases of non-visualized unilateral A1 on the stenotic side, and seven cases (31.82%) had a degree of stenosis of 80–89%. In 17 of these cases, the direction of blood flow in the A1 on the stenotic side deviated from the probe (Figure 6A), and the difference in Vm between A1 on the stenotic and contralateral side was statistically significant (P < 0.001). There were a total of 38 cases of unilateral slender A1 on the stenotic side, among which 11 had a degree of stenosis of 80–89% (28.95%), and 9 had 90% to near complete occlusion (23.68%). In 10 of these cases, the direction of blood flow in the stenotic A1 was toward the probe (Figure 6B), while the other 28 cases deviated from the probe (Figure 6C). There was a statistically significant difference in Vm between the cases where the blood in the A1 on the stenotic and contralateral flowed in the direction away from the probe (P < 0.001). In addition, regardless of the direction of blood flow in the A1 on the stenotic side, there were statistically significant differences in the Vm and PI value for the MCA on both sides (P < 0.05). There were 26 cases with symmetrical A1 imaging, with the highest proportion of 10 cases in unilateral ICAS with a degree of stenosis of 50–69% (38.46%). In 21 cases, the direction of blood flow of the A1 on the stenotic side deviated away from the probe, and there was no statistically significant difference between the Vm and PI value of A1 and MCA bilaterally (P > 0.05; Table 3).

According to the symmetry or asymmetry (slender or non-visualized) of A1 on both sides, it can be seen that the degree of unilateral ICAS was usually more than 70% when A1 on the side of ICAS is slender or non-visualized (P < 0.01). The Vm in the A1 on the stenotic side was significantly lower than that on the contralateral side if the direction blood flow in A1 deviated from the probe (P < 0.001), regardless of whether the A1 on the stenotic side was slender or non-visualized.

Our study found that the hemodynamic changes in A1 on the side of ICAS were related to the degree of stenosis, and the CTA manifestations of A1 may be caused by its hemodynamics. Although CTA examination is an effective method to evaluate ICAS, in clinical practice (Osephson et al., 2004), cervical CTA or ultrasound examination is often not performed promptly in patients with ischemia, which may omit the consideration of the condition of neck vessels. Previous studies have shown that asymmetry of the intracranial segment of the ICA suggests severe stenosis in an ICA (Naggara et al., 2008). In cases of severe ICAS, a high proportion of unilateral A1 dysplasia or underdevelopment was observed on CTA that was attributed to the anatomic variation of the circle of Willis in previous studies (Waaijer et al., 2007; Varga et al., 2019). Although these results are consistent with our study, the hemodynamics of A1 have not yet been further studied. Through our TCD studies, it was found that when the A1 was slender or non-visualized on CTA, it was owing to the difference in hemodynamics of two sides of A1 caused by intracranial hemodynamic changes after ICAS rather than from stenosis or lack of blood flow. Therefore, clinicians should conduct a more comprehensive assessment of the CTA findings in combination with assessment of the hemodynamics when judging the pathogenesis of cerebral ischemia patients (Suwanwela et al., 2002; Tsivgoulis et al., 2007) to find the potential causes of ischemia, such as severe stenosis of ICA. In our study, the appearance of an A1 on CTA as slender or non-visualized, and a decreased Vm when compared with that on the contralateral side often predicted a severe ipsilateral ICAS (≥70%). Furthermore, it is suggested to also pay attention to the opening of the secondary collateral circulation, such as in the OA or StrA, for a more comprehensive judgment.

It is also important to note that the degree of stenosis in the ICA did not affect the Vm of the MCA on the stenotic side, which indicated that the hemodynamic changes caused by ICAS did not affect the distal vessels (Shakur et al., 2014, 2015). However, when the degree of ICAS was >80%, the PI value of the ipsilateral MCA decreased significantly when compared with that of the contralateral MCA; this suggested that when ipsilateral blood flow decreased due to ICAS, the terminal capillaries opened, vascular resistance decreased, and the PI value appeared to “dynamically decline.” Further research is needed to confirm this.

The limitation of this study is that there are no dynamic intracranial vascular changes, and longer observation and related hemodynamic mechanism research should be conducted in the future.