Effect of anterior communicating artery patency on the flow velocity in bilateral carotid artery stenosis after carotid endarterectomy

Abstract

Background:

The effect of anterior communicating artery (ACoA) patency on the flow velocity of the extracranial carotid arteries is unclear.

Methods:

A total of 285 patients with carotid artery stenosis were included between January 2019 and January 2021. All patients received unilateral carotid endarterectomy (CEA). The patients were classified into ACoA-patent (161) and ACoA-nonpatent (124) groups using digital subtraction angiography (DSA) and/or computed tomography angiography (CTA). The peak systolic velocity (PSV) and end-diastolic velocity (EDV) measured by carotid duplex ultrasonography (CDU) were compared between both groups, pre- and post-CEA.

Results:

There was no significant difference in the risk factors for cerebrovascular disease between the two groups. Within 1 week after CEA, the PSV and EDV on operative and nonoperative carotid (contralateral carotid in the same patient) arteries decreased significantly (both p < 0.01). Comparison of nonoperative carotid artery pre- and post-CEA between the two groups showed that post-CEA PSV and EDV in the ACoA-patent group were significantly lower than that of pre-CEA (PSV and EDV, t = 11.507 and 6.716, respectively, both p < 0.001) (according to the Society of Radiologists in Ultrasound Consensus Conference [SRUCC] PSV standard). There was no significant difference in the ACoA-nonpatent group (PSV: t = 1.924, p = 0.057; EDV: t = 1.237, p = 0.218). In the nonoperative carotid artery of the ACoA-patent group, the degree of stenosis assessed by CDU was inconsistent with that of DSA/CTA (κ = 0.294), whereas that in the ACoA-nonpatent group had a high consistency (κ = 0.982). Among 161 ACoA-patent cases, 68 showed overestimated stenosis.

Conclusions:

The patent ACoA increases PSV and EDV, causing an overestimation of carotid artery stenosis.

Background

Stroke is the leading cause of neurological damage in humans. Even transient ischemic attack (TIA) or mild cerebral ischemia may have different degrees of adverse effects on patients.^1,2 Carotid artery stenosis is an important risk factor for ischemic stroke. Carotid endarterectomy (CEA) is a classic treatment option for carotid artery stenosis revascularization, as it improves perfusion territory restoration and reduces the recurrence rate of stroke.^3–5 The peak systolic velocity (PSV) and end-diastolic velocity (EDV) measured by carotid duplex ultrasound (CDU), ⁶ a noninvasive method for evaluating the degree of atherosclerotic carotid artery stenosis, help follow up the improvement of blood flow after CEA revascularization. However, in the early 1990s, some studies reported that the diagnostic criteria of Doppler ultrasound overestimated the degree of stenosis for bilateral carotid artery stenosis.^7,8 Moreover, following the eversion technique for CEA, the degree of restenosis evaluated by CDU based on previous blood flow velocity criteria was overestimated. ⁹ However, the reason for this phenomenon is rarely mentioned in previous literature. In the clinical practice of our center, we have noticed that the anterior communicating artery patency will have a certain impact on the flow velocity measurement of the intracranial segment of the internal carotid artery (ICA). Therefore, we hypothesized that it would also affect the measurement of extracranial carotid artery velocity. Thus, we aimed to investigate the effect of the collateral circulation of the anterior communicating artery (ACoA) patency on flow velocity due to bilateral carotid artery stenosis using CDU.

Methods

The inclusion criteria were as follows: patients with bilateral carotid artery stenosis between January 2019 and January 2021, and at least 70–99% stenosis on one side and CEA surgery on the same side. The exclusion criteria were as follows: (1) severe stenosis of the basilar artery or bilateral vertebral artery (21 cases); (2) severe stenosis of any of the intracranial arteries at the level of the circle of Willis distal to the ICA (15 cases); and (3) lack of follow up with CDU after CEA (20 patients). Initially, 341 patients with bilateral carotid artery stenosis were continuously enrolled, and 285 cases were finally included in this study based on the exclusion criteria. The CEA side was selected according to the degree of carotid artery stenosis, signs, symptoms, and magnetic resonance angiography (MRA) results. Based on the PSV and EDV values, evaluated using CDU, ⁶ the degree of the nonoperative carotid artery was divided into < 50% (PSV < 125 cm/s; EDV < 40 cm/s), 50–69% (125 ⩽ PSV < 230 cm/s; 40 ⩽ EDV < 100 cm/s), and 70–99% (PSV ⩾ 230 cm/s; EDV ⩾ 100 cm/s) stenosis. All patients were classified based on their intracranial collateral circulations of ACoA (patent or not) by screening using transcranial color-coded Doppler (TCCD) or transcranial Doppler (TCD). This was confirmed using computed tomography angiography (CTA) or digital subtraction angiography (DSA). TCCD was used in cases with a good temporal window and TCD was used in cases with a poor temporal window in whom TCCD could not be performed.^10,11 Using CDU, bilateral measurements of PSV and EDV were recorded. The effect of collateral circulation via the ACoA was compared and analyzed for nonoperative carotid artery pre- and post-CEA (within 1 week).

An informed consent form was signed by each clinical attending physician and patient. The contents of the informed consent form included participation in surgery and some necessary examinations, such as CDU, TCCD/TCD, MRI, and CTA/DSA. The study protocol was approved by the institutional ethics committee of Xuanwu Hospital, Capital Medical University. Patient demographic data such as age, sex, and risk factors for the cerebrovascular disease were obtained from the database of the picture archiving and communication system (PACS). All data were de-identified before processing.

Ultrasound Methodology

CDU along with TCCD or TCD were used in this study and were mandated per protocol. An EPIQ5 (Philips Medical Systems, Bothell, WA, USA The Netherlands) CDU with a 3.0–12.0 MHz linear array transducer or an Ascendus (Hitachi, Tokyo, Japan) CDU with a 4.0–8.0 MHz micro-convex transducer and phased array probe with 1.0–5.0 MHz was used. All CDU examinations were performed by vascular ultrasonography trained doctors with > 5 years of experience. A standardized approach was used, with the patients placed in a supine position. The transducer was placed in a sagittal plane at the mid-cervical region and adjusted to optimize the visualization of the proximal ICA. The ultrasound transducer was adjusted to ⩽ 60° angles to measure PSV and EDV. The negative blood flow signal detected through the temporal acoustic window at the depth of 60~75 mm (related to the biparietal diameter) corresponds to anterior cerebral artery (ACA). When the temporal sound window has poor or no sound transmission, ACA can be checked through the eye window. When the ipsilateral carotid artery is compressed, the direction of ACA blood flow is reversed, which can confirm the accuracy of ACA detection and the existence of ACoA. ¹¹

Anterior communicating artery (ACoA)-patent and nonpatent identification

The ACoA-patent in this study refers to the following situations. First, DSA or CTA have confirmed that the ACoA and A1 segment of bilateral ACA exists or ACoA and A1 segments of the nonoperative ACA exist. Under this premise, when the nonoperative carotid artery is compressed, TCCD/TCD detected the reversal of the flow direction of the ACA and the decrease of the flow velocity of the middle cerebral artery (MCA) on the operative side, which indicates that the nonoperative carotid artery supplies blood to the operative cerebral hemisphere through the ACoA. While the ACoA is not patent, it means that when the nonoperative carotid artery is compressed, TCCD/TCD cannot detect the above conditions, indicating that the nonoperative carotid artery does not supply blood to operative ACA and MCA. ¹¹

DSA, CTA, and MRA

DSA, CTA, and MRA were mandated per protocol.

The DSA examination before CEA used a NEUROSTAR Plus/TOP double C-arm angiography machine (Siemens, Munich, Germany) to visualize the aortic arch; complete intra- and extracranial angiography was performed. Images were obtained from the frontal, lateral, and double oblique planes to clearly identify the lesion degree of the bilateral carotid artery, vertebral artery, and subclavian artery and to assess whether the ACoA was opened or not. The stenosis degree on both sides of the carotid artery was also recorded and analyzed.

CTA examinations were performed using a 128-section multidetector scanner (Ingenuity, Philips) with a slice thickness of 1.0 mm. A total of 50 mL Imeron 400, an iodine-based contrast agent (400 mg iodine per mL; Bracco Imaging, Konstanz, Germany), was administered intravenously followed by a saline flush. The degree of carotid stenosis on CTA was determined by experienced neuroradiologists after revisions of all maximum intensity projection (MIP) images in all planes (axial, coronal, and sagittal). All measurements were based on the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria.

MRI was performed using a Philips 1.5-T scanner for head examination; T1-weighted (T₁W), T2-weighted (T₂W), and diffusion-weighted imaging (DWI) scans were used to observe the sizes of cerebral ischemic lesions and correlate them with clinical signs and symptoms.

Surgical procedure

The CEA was performed according to the protocol we previously reported by neurosurgeons with > 10 years of CEA experience. ¹²

Statistical analyses

All statistical analyses were performed using IBM SPSS, Version 25.0 (IBM Corp., Armonk, NY, USA). The independent-sample t-test and the chi-squared test were used to compare the demographic data of patients, with results expressed as percentages of the component rates. The paired Student’s t-test was used to analyze the differences between the continuous variables PSV and EDV on both sides of the carotid artery stenosis before and after CEA. The results of blood velocity assessed using CDU are each expressed as mean ± SD. The independent-sample t-test was used to compare the differences between PSV and EDV on the side of the nonoperative carotid artery between the ACoA-patent and ACoA-nonpatent groups. The consistency between CDU and DSA/CTA was compared using a consistency test. All tests were two-tailed, and p < 0.05 was considered statistically significant.

Results

Comparison of cerebrovascular risk factors between the ACoA-patent and ACoA-nonpatent groups

The 285 included patients were aged 63.92 ± 7.2 years (range, 45–86 years). Men were 91.9% (262/285 cases) of the cohort. The following cerebrovascular risk factors were noted: hypertension (68.1%), diabetes mellitus (28.1%), coronary heart disease (13.7%), hyperlipidemia (49.5%), and smoking (82.4%). Hypertension, hyperlipidemia, and smoking were the main risk factors. There was no significant difference between the ACoA-patent and ACoA-nonpatent groups regarding baseline characteristics (Table 1).

Table 1.

Comparison of clinical data between ACoA-patent and ACoA-nonpatent groups.

Subjects	Total n = 285 (100%)	ACoA-patent n = 161 (56.5%)	ACoA-nonpatent n = 124 (43.5%)	t/χ2 value	p-value
Age, years	63.92 ± 7.2	63.7 ± 6.9	64.1 ± 7.6	0.560	0.576
Male	262 (91.9)	148 (91.9)	114 (91.9)	1.000	0.588
HP	194 (68.1)	113 (70.2)	81 (65.3)	0.442	0.228
HLD	141 (49.5)	79 (49.1)	62 (50.0)	0.905	0.485
DM	80 (28.1)	42 (26.1)	38 (30.6)	0.426	0.237
CHD	39 (13.7)	24 (14.9)	15 (12.1)	0.491	0.264
Smoking	235 (82.4)	135 (83.9)	100 (80.6)	0.531	0.291

Data are n (%) or mean ± SD.

ACoA, anterior communicating artery; CHD, coronary heart disease; DM, diabetes mellitus; HLD, hyperlipidemia; HP, hypertension.

Comparison of the PSV and EDV on both sides of carotid artery stenosis, pre- and post-CEA

Post-CEA revascularization, the PSV and EDV on the operative carotid artery returned to normal, wherein PSV decreased from 458.7 ± 125.2 cm/s to 82.2 ± 29.7 cm/s, and EDV decreased from 195.3 ± 72.8 cm/s to 28.5 ± 11.9 cm/s (both p < 0.01; Table 2). Similarly, the velocity on the nonoperative carotid artery was also reduced. PSV decreased from 156.3 ± 33.6 cm/s to 125.3 ± 27.2 cm/s, and EDV decreased from 63.3 ± 12.9 cm/s to 46.7 ± 10.2 cm/s (both p < 0.01; Table 2).

Table 2.

Comparison of blood flow velocity on operative-side and nonoperative carotid artery before and after carotid endarterectomy.

Classification (vessels)	Velocity (cm/s)	Pre-CEA	Post-CEA	t-value	p-value
Operative carotid artery (n = 285)	PSV	458.7 ± 125.2	82.2 ± 29.7	51.720	<0.001
	EDV	195.3 ± 72.8	28.5 ± 11.9	39.237	<0.001
Nonoperative carotid artery (n = 285)	PSV	156.3 ± 33.6	125.3 ± 27.2	10.117	<0.001
	EDV	63.3 ± 12.9	46.7 ± 10.2	6.490	<0.001

Data are mean ± SD.

CEA, carotid endarterectomy; EDV, end-diastolic velocity; PSV, peak systolic velocity.

Further, we compared the blood flow velocity between the ACoA-patent (161 cases) and the ACoA-nonpatent (124 cases) groups of the nonoperative carotid artery. The PSV (168.9 ± 36.7 cm/s) and EDV (70.7 ± 14.1 cm/s) pre-CEA in the ACoA-patent group were significantly higher than those post-CEA (PSV: 115.2 ± 25.2 cm/s; EDV: 42.8 ± 9.2 cm/s; both p < 0.01; Table 3). However, there was no difference in the PSV and EDV values pre- and post-CEA on the nonoperative carotid artery in the ACoA-nonpatent group (both p > 0.05; Table 3, Figures 1 and 2). There was a significant difference in PSV and EDV on the nonoperative carotid artery between the ACoA-patent and ACoA-nonpatent groups pre-CEA (PSV: t = 1.980, p = 0.049; EDV: t = 1.993, p = 0.047).

Table 3.

Comparison of before and after carotid endarterectomy blood flow velocity on the nonoperative carotid artery in ACoA-patent and ACoA-nonpatent groups.

Classification (vessels)	Velocity cm/s	Pre-CEA	Post-CEA	t-value	p-value
ACoA-nonpatent (n = 124)	PSV	140.0 ± 30.3 a	138.4 ± 29.7	1.924	0.057
	EDV	53.7 ± 11.6 b	51.7 ± 11.2	1.237	0.218
ACoA-patent (n = 161)	PSV	168.9 ± 36.7 a	115.2 ± 25.2	11.507	<0.001
	EDV	70.7 ± 14.1 b	42.8 ± 9.2	6.716	<0.001

^at = 1.980, p = 0.049.

^bt = 1.993, p = 0.047.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; EDV, end-diastolic velocity; PSV, peak systolic velocity.

Figure 1A.

CTA image showing that bilateral anterior cerebral arteries and anterior communicating arteries are well developed. The white arrow and circle indicate the ACoA.

ACoA, anterior communicating artery; CTA, computed tomography angiography; LACA, left anterior cerebral artery; RACA, right anterior cerebral artery.

Figure 2.

Comparison of PSV and EDV pre- and post-CEA of the nonoperative side carotid artery between the ACoA-patent and ACoA-nonpatent groups.

Data shown as mean ± SD.

***p < 0.01.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; EDV, end-diastolic velocity; NS, no significance; PSV, peak systolic velocity.

Comparison of degree of nonoperative carotid artery stenosis between CDU and DSA/CTA in the ACoA-patent and ACoA-nonpatent groups

In the nonoperative carotid artery of the ACoA-patent group, the degree of carotid artery stenosis assessed by CDU had a low consistency with DSA/CTA (κ = 0.294; Table 4), whereas that in the ACoA-nonpatent group had a high consistency (κ = 0.982; Table 5).

Table 4.

Comparison of consistency between degree of stenosis evaluated by PSV standard and DSA/CTA before carotid endarterectomy in the ACoA-patent group.

Stenosis	ACoA-patent PSV standard (pre-CEA)			Total
	< 50%	50–69%	70–99%
< 50%	69	0	0	69
50–69%	57	9	0	66
70–99%	2	9	15	26
Total	128	18	15	161

κ = 0.294.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; CTA, computed tomography angiography; DSA, digital subtraction angiography; PSV, peak systolic velocity.

Table 5.

Comparison of consistency between degree of stenosis evaluated by PSV standard and DSA/CTA before carotid endarterectomy in the ACoA-nonpatent group.

Stenosis	ACoA-nonpatent PSV standard (pre-CEA)			Total
	< 50%	50–69%	70–99%
< 50%	128	2	0	130
50–69%	0	16	1	17
70–99%	0	0	14	14
Total	128	18	15	161

κ = 0.982.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; CTA, computed tomography angiography; DSA, digital subtraction angiography; PSV, peak systolic velocity.

In the ACoA-patent group, there were 128 cases (128/161 cases, 79.5%) of mild stenosis assessed by DSA/CTA on the nonoperative carotid artery side; of these, 57 cases of mild stenosis were overestimated as moderate by CDU, and two cases were overestimated as severe. Thus, only 69 of these cases (69/161 cases, 42.9%) of mild stenosis were assessed by CDU. There were 15 cases (15/161, 9.3%) of severe stenosis assessed by DSA/CTA and 26 cases (26/161, 16.1%) of severe stenosis assessed by CDU. The inconsistency between the two assessments was due to the overestimation of severe stenosis in nine cases of moderate stenosis and two cases of mild stenosis (Table 4).

Discussion

Atherosclerotic carotid artery stenosis is a major cause of ischemic stroke. With an increasing degree of carotid artery stenosis, the incidence and recurrence rate of stroke increase significantly.^1,6 When patients have severe carotid artery stenosis ipsilaterally and contralateral ICA with ⩾ 70% stenosis or occlusion, the risk of TIA or stroke increases. A subgroup analysis in NASCET revealed that patients with bilateral carotid artery stenosis were more likely to experience perioperative stroke or death in the surgery group compared with those without contralateral stenosis. ¹³ Thus, it is important to evaluate the degree of preoperative carotid artery stenosis to accurately assess the risk and predict the need for CEA.^1,6,14 Therefore, standardizing cut-off values for parameters is of great significance for patient screening and evaluating clinical treatment. However, in the early 1990s, some studies^7–9 reported that diagnostic criteria assessed using CDU overestimated the stenosis degree for bilateral carotid artery stenosis.

In 1998, van Everdingen et al. ⁷ revealed that this overestimation was due to one side of the ICA being occluded and causing blood flow to increase on the other side of the ICA with carotid artery stenosis; this was confirmed by cerebral blood flow perfusion on MRI. Similarly, in 2000, Ray et al. ⁸ studied 131 patients with bilateral carotid artery stenosis who received CEA treatment on one side of carotid artery stenosis for 3 months. The study revealed that 42% (14/33 cases) of patients had overestimated degrees of stenosis on the contralateral ICA of the carotid artery stenosis based on PSV and EDV, assessed using duplex ultrasound. However, no study has described this mechanism in detail.

In this study, after CEA, the PSV and EDV of the nonoperative carotid artery were decreased (both p < 0.01, Table 2). These results were consistent with those of previous studies.^7,8 Based on examination results using TCD or TCCD, we further classified the patients into ACoA-patent and ACoA-nonpatent groups for comparison. We found that the blood flow velocity on the side of the nonoperative carotid artery in the ACoA-patent group was significantly decreased after revascularization (Table 3, Figure 1A–C, and Figure 2A and B). However, there was no significant difference in the ACoA-nonpatent group (Table 3, Figure 2). This result suggests that ACoA patency is an important influencing factor for the increase of PSV and EDV pre-CEA, and the degree of stenosis is overestimated on the side of the nonoperative carotid artery. Given the overestimation of blood flow velocity and stenosis by CDU, many scholars have studied this phenomenon and attempted to propose solutions.

Figure 1B.

Pre-CEA, the PSV is 271 cm/s and the EDV is 92.9 cm/s for the nonoperative carotid artery with the patent ACoA.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; EDV, end-diastolic velocity; PSV, peak systolic velocity.

Figure 1C.

Post-CEA, the PSV and EDV of the nonoperative carotid artery, respectively, decreased to 192 cm/s and 73.9 cm/s with the patent ACoA.

ACoA, anterior communicating artery; CEA, carotid endarterectomy; EDV, end-diastolic velocity; PSV, peak systolic velocity.

In 1995, AbuRahma et al. ¹⁵ compared different ultrasound criteria and noted that overestimation of ipsilateral stenosis can be avoided if the standard criteria are modified using an ICA PSV of ⩾ 4.5 kHz, relative to the PSV 145 cm/s for the diagnosis of ⩾ 50% stenosis. The AbuRahma article specifically looked at criteria with the contralateral disease. The intent of examining the criteria was similar to our study.

In 2021, Gornik et al. ¹⁶ reported that the primary criterion of the Society of Radiologists in Ultrasound Consensus Conference (SRUCC) ⁶ for ⩾ 50% ICA stenosis, a PSV value of ⩾ 125 cm/s, did not meet prespecified thresholds for adequate sensitivity, specificity, and accuracy (sensitivity, 97.8%; specificity, 64.2%; accuracy, 74.5%). The study found that the diagnostic criteria issued by the SRUCC (2003) ⁶ and results of duplex ultrasound and angiography showed a moderate agreement (κ = 0.42) with overestimation of the degree of stenosis for both moderate (50–69%) and severe (⩾ 70%) ICA lesions. They pointed out that the test performance was improved by increasing the PSV threshold to ⩾ 180 cm/s or by adding the additional parameter of ICA/common carotid artery (CCA) PSV ratio of ⩾ 2.0. The study by Gornik et al. was performed to validate the SRUCC criteria in a multicenter study. It also examined the data to determine if adjustments in the SRUCC criteria would provide better accuracy. They reported improved criteria.

The aforementioned studies focused on the criteria of ultrasound evaluation and did not consider the possible compensatory function from relatively mild stenosis to the severe side via the patent ACoA. The collateral circulation of the brain may be defined as the artery-to-artery anastomotic pathways that are capable, when needed, of supplying nutrient perfusion to a brain region whose primary source of blood flow has been reduced or compromised by disease. ACoA is part of it. The sufficiency of the collateral circulation has been convincingly established as a key factor influencing the likelihood of successful reperfusion and favorable clinical outcome.^17,18 In our study, patients were divided into two groups, ACoA-patent and ACoA-nonpatent, according to the TCCD or TCD and confirmed using CTA/DSA. The consistency between CDU and DSA/CTA was much higher (κ = 0.982) for patients with ACoA-nonpatent than for those with ACoA-patent (κ = 0.294) (Tables 4 and 5). The ACoA is one of the main cross-filling collateral circulations of the circle of Willis between the blood supplied by bilateral carotid artery stenosis. If the ACoA is patent, the blood supplied from the extracranial ICA with a lower degree of stenosis into the contralateral brain through the circle of Willis of the ACoA will increase the blood flow velocity of the carotid artery with a lower degree of stenosis. Therefore, the patent ACoA is responsible for the increased blood flow velocity and overestimation of the stenosis degree of the carotid artery on the side with relatively mild lesions. Our research confirms that this overestimation is based on the patent ACoA. When the interpretation of the contralateral carotid artery is based on CDU assessment before revascularization, the patient may receive an incorrect interpretation leading to an overestimation of disease and exposure to unnecessary surgical risks. Therefore, attention should be paid to the influence of ACoA on extracranial carotid flow velocity measurement, and reevaluation of the nonoperative side following the contralateral carotid surgery and prior additional surgery is recommended.

Study limitations

The retrospective nature of this study led to some limitations. First, the higher proportion of male patients and the lower of female patients is a limitation of this study. This could be attributed to the lower incidence of carotid artery stenosis since Chinese women rarely smoke. This is also the epidemiological situation of cerebrovascular disease in China. Second, the number of patients with bilateral severe stenosis (70–99% stenosis) is small. In future studies, we will further expand the sample size and establish criteria for evaluating the hemodynamic parameters of contralateral ICA stenosis in different degrees when one side of the carotid artery is severely stenotic or occlusive, to improve clinical treatment decisions.

Conclusion

If there is bilateral carotid artery stenosis, it is very important to combine the method of CDU with TCCD or TCD to evaluate and distinguish the degree of stenosis and the collateral circulation of the ACoA (patent or not). The patent ACoA may cause the velocity to rise on the side of the nonoperative carotid artery, and the degree of carotid artery stenosis will be overestimated.