From arteries to age: an intrinsic impact of Circle of Willis on first-ever acute ischemic stroke in the oldest-old

doi:10.21203/rs.3.rs-4930342/v1

Background: This study aimed to describe the characteristics of CoW configuration in the oldest-old patients with first-ever acute ischemic stroke, and investigate the association of CoW with the first-onset age of acute ischemic stroke in the elderly population.

Methods: The patients with first-ever acute ischemic stroke were consecutively enrolled between March 2022 and June 2024. All patients underwent MRA or CTA of brain. According to their first-onset age, they were divided into two groups (oldest-old group, aged ≥80 years; elderly group, aged 60-79 years). Demographic and clinical data were collected, and CoW scores and types were independently assessed by two experienced neurologists.

Results: The oldest-old group showed greater Anterior communicating artery (AcoA) and total CoW scores and a higher ratio of complete CoW as compared with the elderly group. Poor CoW was associated with earlier first-onset age, and higher AcoA score was associated with later first-onset age, even after adjustments by conventional risk factors. In the female cohort, the oldest-old subset showed high scores of AcoA, Posterior communicating artery (PcoA) and total CoW, with all difference strength significantly greater than those in all subjects. Additionally, in the female cohort, first-onset age was significantly related with AcoA score, total CoW score and poor CoW, whether non-adjusted or adjusted.

Conclusions: The oldest-old patients with first-onset ischemic stroke have a higher CoW score and better morphology than the elderly patients, with females more significant. Females with poor CoW should be recommended for risk stratification of ischemic stroke to generate more effective and precise strategies for stroke prevention and healthy aging.

Circle of Willis

oldest-old

ischemic stroke

first onset

Age is the most significant non-modifiable risk factor for acute ischemic stroke. With increasing expectancy of life, the majority of the patients with acute ischemic stroke will be very elderly subjects. An epidemiological study based on the National Acute Stroke Israeli Survey (NASIS) revealed that the affected proportion of the oldest-old patients with first-ever ischemic stroke among the general population increased from 18.3% in 2004 to 19.9% in 2007 and 24.5% in 2010^[1]. Interestingly, this “special” cohort for the elderly patients with first-ever ischemic stroke showed a distinctive clinical profile. Compared with the patients with first onset ischemic stroke at the age of 65-84 years, the patients aged ≥85 years demonstrate a higher prevalence of atrial fibrillation, large artery atherosclerosis^[2], and major adverse cardiovascular events^[3]. According to the existing knowledge for the pathogenesis of cerebrovascular disease, worse and more vascular risk factors mean earlier onset of ischemic stroke in the subjects^[4]. However, from the perspective of onset age, high vascular risk factors indeed did not accelerate the onset of acute ischemic stroke in the very elderly population with first-onset ischemic stroke. Therefore, we reasonably assume that there might be some protective factors combating those age-related irreversibly pathological changes deriving from the traditional vascular risk factors, and delaying onset age finally in very elderly population.

The circle of Willis (CoW) is an anastomotic arterial network located on the base of the brain, which creates a polygonal‐shaped connection between the internal carotid and vertebrobasilar systems (anterior and posterior circulation of the brain, respectively)^[5]. The presence of communicating arteries in CoW is related to potential autoregulation capacity and sufficient blood flow. In cases of occlusion, where cerebral arteries are blocked and the brain blood flow is insufficient for normal cerebral metabolism, the CoW integrity becomes even more important. The general population has considerable anatomic variations in CoW morphology and configuration, with numerous studies highlighting a higher prevalence of incomplete CoW in patients with acute ischemic stroke (ranged from 34% to 82.4%) compared to healthy controls^[6]. Considerable published studies have shown that the Cow integrity and hypoplasia played a vital role in stroke severity and had a significant impact on stroke outcome, with patients achieving more favorable results when communicating arteries were present, especially in cases of distal internal carotid T-form occlusion^{[7, 8]}. A recent meta-analysis study revealed that the presence of any variation in CoW was 1.38 times more likely to develop ischemic stroke as compared to the patent CoW, and clearly pointed out that the presence of hypoplasia or incompleteness in a posterior or anterior communicating artery was a contributing factor for the development of ischemic stroke^[9]. The above information hints that having a complete CoW or “better” configuration has some kind of protective effect on brain and facilitates the delayed- or late-onset of acute ischemic stroke in the elderly subjects. However, so far, there have been no studies directly analyzing the trend towards later first-onset age with better CoW variants in elderly stroke patients.

In order to investigate the underlying protective factors accounting for the late onset of first-ever acute ischemic stroke in elderly patients, and to further test the relevance of the onset age of acute ischemic stroke with CoW configuration, in the present study we initiated a retrospective data analysis. Patients with first-onset ischemic stroke who underwent magnetic resonance angiography (MRA) or computed tomography angiography (CTA) of brain were consecutively enrolled between March 2022 and June 2024. According to their first-onset age, they were divided into two groups (the oldest-old group, aged ≥80 years; and the elderly group, aged 60-79 years). Demographic and clinical data were collected, and CoW scores and types were independently assessed by two experienced neurologists. Characteristics of CoW configuration in two groups were described and regression tests were used to determine whether CoW configuration was related to first-onset age in elderly patients. In the next exploratory analyses, the influence on the association of CoW with first-onset age by gender was investigated.

The study was approved by the ethics committee of Bethune International Peace Hospital. All patients or their legal representatives gave written informed consent before being included in this study.

2.1 Collection of clinical data

The patients with first-ever acute ischemic stroke, who were treated at the Department of Neurology, Bethune International Peace Hospital, were consecutively enrolled in this study between March 2022 and June 2024. According to their first-onset age, they were divided into two age groups (the oldest-old group, aged ≥80 years; and the elderly group, aged 60-79 years). The demographic and clinical data of all patients were collected, such as age, gender, current smoking, main comorbidity (including atrial fibrillation, diabetes, hypertension, hyperlipidemia, coronary heart disease and hyperhomocysteinemia), and corresponding medication history (antiplatelets, oral anticoagulants, statins, antihypertensives, antidiabetics). In addition, the National Institutes of Health Stroke Scale (NIHSS) score at admission, time from onset to admission was collected in all cases. The inclusion criteria were as follows: (1) age ≥ 60 years; (2) initial clinical onset; (3) time from onset to admission ≤ 7 days; (4) acute cerebral infarction diagnosed with diffusion-weighted imaging (DWI) of head magnetic resonance imaging (MRI) performed within 24 h after admission; and (5) ability to cooperate with various examinations at admission, such as head MRI + MRA or CTA, Doppler vascular ultrasonography of the neck, and laboratory tests. The exclusion criteria were as follows: (1) previous history of stroke including cerebral infarction, transient ischemic attack and cerebral hemorrhage; and (2) other intracranial diseases, such as traumatic brain injury, intracranial tumors, arteriovenous malformations, venous sinus thrombosis, hydrocephalus, multiple sclerosis, and congenital abnormalities.

2.2 Collection of CoW data

All enrolled patients underwent a plain head MRI scan (axial T1-FLAIR, T2WI, and T2-FIAIR), diffusion-weighted imaging (DWI), and 3-dimensional time-of-flight (3D-TOF) MRA or CTA of brain within 72 hours after admission. Two experienced neurologists (Lian-Yuan Feng and Xue-Li Wang) masked to clinical characteristics conducted a blinded and independent evaluation of all MRA or CTA images to assess the CoW. The disagreements were resolved through discussion in the case of discrepancy. The fetal variant of the posterior communicating artery (PcoA) was defined when patients had visible PcoA but without P1 (segment 1 of posterior communicating artery). CoW score assessment was performed according to the method proposed by Xu et al.^[10], wherein the CoW score ranged from 0 to 6. Anterior communicating artery (AcoA) score was scored 0 when the AcoA or unilateral A1 segment of the anterior cerebral artery was absent; scored 1 with an intact AcoA and ipsilateral A1 diameter was ≤ 50% of that of the contralateral A1; and scored 2 when the AcoA was intact and both A1 diameters were > 50% of that of the contralateral A1. For the PcoA score, 0 points were allocated when the PcoA or P1 segment of the posterior cerebral artery was absent; 1 point when the diameter of PcoA was ≤ 50% of that of the ipsilateral P1; and 2 points when the diameter of PcoA was > 50% of that of the ipsilateral P1. A total CoW score ranging from 0 to 6 was calculated with a formula of AcoA score (0-2) + left PcoA score (0-2) + right PcoA score (0-2). A poor CoW was defined as a CoW score of 0–2, and a non-poor CoW was determined when a CoW score was of 3-6, per case. Furthermore, CoW variants were classified into four types (based on the classification method of Hartkamp et al.^[11]). An “incomplete” CoW was considered when any artery of the CoW was absent or dysplasia (vessel diameter < 0.8mm in diameter). Type I was defined as complete CoW-presence and no dysplasia of all constituent vessels of the CoW; type II: complete anterior half and incomplete posterior half of the CoW (presence and no dysplasia of the constituent vessels of the anterior half of the CoW, and absence or dysplasia of at least 1 vessel of the posterior half of the CoW); type III: incomplete anterior half and complete posterior half of the CoW (absence or dysplasia of at least 1 vessel of the anterior half of the CoW, and presence and no dysplasia of the constitute vessels of the posterior half of the CoW); and type IV: incomplete anterior and posterior halves of the CoW (absence or dysplasia of at least 1 vessel of the anterior half of the CoW, and absence or dysplasia of at least 1 vessel of the posterior half of the CoW).

2.3 Statistical Methods

Data analyses were performed using SPSS (version 24, IBM, Armonk, NY, USA). Patients were categorized into two groups: the oldest-old group (aged ≥80 years) and the elderly group (aged 60–79 years). The continuous variables were first assessed for normal distribution using the one-sample Kolmogorov–Smirnov test. For the descriptive analyses, continuous variables are presented as the median and interquartile range (IQR) for non-normal distribution, and mean ± standard deviation (SD) for normal distribution. Categorical variables are described as frequencies and proportions. Comparisons of continuous variables between two groups were performed using the independent-sample t-test or Mann–Whitney U-test, as applicable. The comparisons of categorical variables between groups were performed using Fisher’s exact test or the χ ² -test.

We used linear regression analysis to look for the associations for first-onset age, including total CoW score, AcoA score, PcoA score, poor CoW, fetal CoW and incomplete CoW type. The nonparametric tests were also performed to investigate the association of the age group with CoW variables mentioned above. Multivariable regression analyses were performed with adjustments for sex, current smoking and comorbidities (in model 1), and medication history (in model 2). Using the same tactics, in the different sex cohorts, the clinical and demographic data were compared, and the associations of first-onset age and CoW parameters were investigated. The statistical threshold was fixed at P<0.05.

3.1 Comparison of demographic and clinical data between the oldest-old group and the elderly group

A total of 357 acute first-onset ischemic stroke (MR-DWI positive) patients aged ≥60 years with MRA and/or CTA images were screened and finally, 276 cases were included in the analysis. Forty-seven cases were excluded due to unclear stroke history or TIA/cerebral infarct-mimic history and 12 cases were excluded either because of insufficient image quality due to excessive head movement or other imaging artifacts. A total of 22 was excluded because of the existing comorbidity, including chronic subdural hematoma (3), hypophyseal adenoma (2), cavernous hemangioma (3), arteriovenous malformations (1), venous sinus thrombosis(1), hydrocephalus (2), Alzheimer’s disease (7), mitochondrial encephalopathy (1), suspected drug abuse (1) and suspected cerebral autosomal dominant arteriopathy (1).

Among 276 cases enrolled, 188 (68.1%) were 60-79 years (the elderly group), and 88 (31.9%) were 80 years or older (the oldest-old group). Comparisons of demographic characteristics, vascular risk factors, previous medication, and clinical parameters between two groups are shown in Table 1. In the elderly group, the median first-onset age is 70 years, with 84 years in the oldest-old group. The oldest-old group with first-ever ischemic stroke showed a higher percentage of females (47.1% vs. 31.4%). In terms of vascular risk factors, oldest-old patients have a higher ratio of cardiac risk factors including medical history of atrial fibrillation and coronary heart disease, when compared with elderly patients. There were no significant differences in the percentage of current smoking between the two groups. In comparison to the elderly patients, more oldest-old patients took antiplatelets prior to admission (38.6% vs. 28.2%). There were no significant disparities between the two groups in the percentage of previous statin use in all patients (23.9% vs. 27.1%), oral anticoagulant use in atrial fibrillation patients (40% vs. 60%), antihypertensive agent use in hypertension patients (87.5% vs. 78.1%), and antidiabetic agent use in diabetes patients (77.8% vs. 84.8%). Regarding clinical parameters, the oldest-old group showed higher NIHSS scores at admission compared with the elderly group, and there were no differences between the two groups in time from onset to admission.

Table 1. Baseline characteristics of stroke patients

Characteristic	All (n =276)	Elderly (n=188)	Oldest-old (n=88)	P value*
Demographic factors
Age, year，median, (IQR)	73.5 (67.0-81.0)	70.0 (65.0-74.0)	84.0 (81.0-88.0)	<0.001
Male sex, n (%)	175 (63.4)	129 (68.6)	46 (52.3)	0.009
Vascular risk factors, n (%)
Atrial fibrillation	25 (9.1)	10 (5.3)	15 (17.1)	0.002
Diabetes mellitus	106 (38.4)	79 (40.2)	27 (30.7)	0.071
Arterial hypertension	184 (66.7)	128 (68.1)	56 (63.6)	0.465
Hyperlipidemia	86 (31.2)	60 (32.2)	26 (29.5)	0.692
Coronary heart disease	43 (15.6)	20 (10.6)	23 (26.1)	0.001
Hyperhomocysteinemia	54 (19.6)	34 (18.1)	20 (22.7)	0.365
Current smoking	22 (8.0)	16 (8.5)	6 (6.8)	0.629
Previous medication, n (%)
Antiplatelets	87 (31.5)	53 (28.2)	34 (38.6)	0.082
Oral anticoagulants	12 (4.3)	6 (3.2)	6 (6.8)	0.169
Statins	72 (26.1)	51 (27.1)	21 (23.9)	0.565
Antihypertensives	149 (54.0)	100 (53.2)	49 (55.7)	0.699
Antidiabetics	88 (31.9)	67 (35.4)	21 (23.9)	0.050
Clinical parameters, median, (IQR)
NIHSS at admission	2.0 (1.0-4.0)	2.0 (1.0-3.0)	2.5 (2.0-5.0)	0.001
Time onset to admission, hour	16.5 (12.0-40.0)	17.0 (12.0-39.5)	14.0 (12.0-48.0)	0.775

Abbreviations: IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale. * Elderly group vs Oldest-old group.

3.2 Differences in CoW score and type between the oldest-old group and the elderly group

Table 2 represents the characteristics of CoW variants in overall cases and the two groups. The distribution details of the total CoW score (0-6), AcoA score (0-2) and PcoA score (0-4) were also summarized in Figure 1. There was a significant difference in the proportion of total CoW score from 0 to 6 between the oldest-old group and the elderly group (p=0.013), and the distribution disparities of AcoA score (0-2) and PcoA score (left half + right half, 0-4) were also observed between the two groups (p=0.003, 0.007, respectively). There was a trend towards a higher CoW score in the oldest-old group compared with the elderly group, and especially the AcoA and total CoW score in the oldest-old group were significantly greater than those in the elderly group (p<0.001, p=0.001, respectively). In the cohort of all cases, 76.1% (210/276) had poor CoW variants (termed as total CoW score 0-2), and the poor CoW was less frequently seen in the oldest-old group (oldest-old vs. elderly, 64.8% vs. 81.9%, p=0.002). In terms of another CoW variant assessment in which CoW variants were classified into four types, there was no difference in the distribution of four variant types of CoW configuration between the two groups (shown in Figure 1), whereas a higher ratio of complete CoW (type I) was found in the oldest-old group as compared with the elderly group. There was no difference in the percentage of fetal PcoA between the two groups.

Table 2. CoW characteristics of stroke patients

Characteristic	All (n =276)	Elderly (n=188)	Oldest-old (n=88)	P value*
Total CoW score，median, (IQR)	2 (1-2)	2 (2-2)	2 (2-3.75)	<0.001
Total CoW score, n (%)				0.013
Score 0	39 (14.1)	31 (16.5)	8 (9.1)
Score 1	36 (13.0)	30 (16.0)	6 (6.8)
Score 2	135 (48.9)	92 (48.9)	43 (48.9)
Score 3	21 (7.6)	12 (6.4)	9 (10.2)
Score 4	37 (13.4)	18 (9.6)	19 (21.6)
Score 5	5 (1.8)	4 (2.1)	1 (1.1)
Score 6	3 (1.1)	1 (0.5)	2 (2.2)
AcoA score，median, (IQR)	2(1-2)	2 (1-2)	2 (2-2)	0.001
AcoA score, n (%)				0.003
Score 0	53 (19.2)	43(22.9)	10 (11.4)
Score 1	49 (17.8)	39 (20.7)	10 (11.4)
Score 2	174 (63.0)	106 (56.4)	68 (77.3)
PcoA score，median, (IQR)	0 (0-1)	0 (0-1)	0 (0-2)	0.061
PcoA score, n (%)				0.007
Score 0	188 (68.1)	134 (71.3)	54 (61.4)
Score 1	31 (11.2)	24 (12.8)	7 (8.0)
Score 2	44 (15.9)	20 (10.6)	24 (27.3)
Score 3	8 (2.9)	7 (3.7)	1 (1.1)
Score 4	5 (1.8)	3 (1.6)	2 (2.3)
Poor CoW (TcoW score 0-2), n (%)	210 (76.1)	154 (81.9)	57 (64.8)	0.002
CoW type, n (%)				0.065
Type I (complete CoW)	26 (9.4)	13 (6.9)	13 (14.8)
Type II (incomplete AcoW only)	197 (71.4)	133 (70.7)	64 (72.7)
Type III (incomplete PcoW only)	14 (5.1)	12 (6.4)	2 (2.3)
Type IV (incomplete CoW)	39 (14.1)	30 (16.0)	9 (10.2)
Fetal PcoA, n (%)	27 (9.6)	19 (10.1)	8 (9.1)	0.791

Abbreviations: IQR, interquartile range; CoW, circle of Willis; AcoA, anterior communicating artery; PcoA, posterior communicating artery; AcoW, anterior CoW; PcoW, posterior CoW. * Elderly group vs Oldest-old group.

3.3 Association of age category with CoW characteristics

Previous studies suggested that the outcome and severity of ischemic stroke were correlated with CoW score, complete CoW and fetal PcoA. In the present study, logistic regression tests were performed to investigate the influences of these CoW characteristics on the categories of first-onset age (60-79 years as early-onset, ≥80 years as late-onset, in all elderly patients). Our data suggested that AcoA score, total CoW score and complete CoW were positively related with the late onset of the elderly patients, and as expected, poor CoW was negatively associated with late onset (shown in Table 3). Adjustment for sex, current smoking and comorbidities (including atrial fibrillation, diabetes, hypertension, hyperlipidemia, coronary heart disease and hyperhomocysteinemia) did not alter the direction and strength of the association of late-onset with AcoA score, total CoW score and poor CoW (model 1). Whereas, no significant association between late-onset with PcoA score was observed after adjustment in model 1. After further adjustment (model 2) for variables in model 1 and medication history (antiplatelets, Statins, oral anticoagulants, antihypertensives and antidiabetics), we found a similar result as for model 1.

Table 3. The association of age category with CoW parameters

	Univariable analysis		Multivariable analysis (model 1)		Multivariable analysis (model 2)
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
AcoA score	0.561 (0.389–0.810)	0.002	0.584 (0.394–0.864)	0.007	0.566 (0.373–0.859)	0.008
PcoA score	0.791 (0.615–1.016)	0.066	0.794 (0.609–1.037)	0.090	0.083 (0.632–1.093)	0.185
Total CoW score	0.709 (0.578–0.870)	0.001	0.722 (0.580–0.900)	0.004	0.738 (0.588–0.926)	0.009
Poor CoW	2.588 (1.462–4.583)	0.001	2.692 (1.463–4.956)	0.001	2.428 (1.293–4.560)	0.006
Complete CoW	0.429 (0.190–0.968)	0.042	0.546 (0.225–1.326)	0.182	0.474 (0.189–1.191)	0.112
Fetal PcoA	1.124 (0.472–2.678)	0.791	1.011 (0.399-–2.564)	0.982	1.201 (0.450–3.207)	0.714

Abbreviations: OR, odds ratio; CI, confidence interval; AcoA, anterior communicating artery; PcoA, posterior communicating artery; CoW, circle of Willis. Model 1: adjusted for hypertension history, diabetes mellitus history, coronary heart disease history, atrial fibrillation history, hyperlipidemia, hyperhomocysteinemia, and current smoking. Model 2: adjusted for factors in model 1 + medication history (antiplatelets, Statins, oral anticoagulants, antihypertensives and antidiabetics).

3.4 Association of first-onset age with CoW characteristics

There were no associations of first-onset age with traditional vascular risk factors (including current smoking, atrial fibrillation, diabetes, hypertension, hyperlipidemia, coronary heart disease and hyperhomocysteinemia) and medication history. Using regression tests, we found that in elderly stroke patients, poor CoW was associated with earlier first-onset age (OR=-2.671, 95% CI= -5.140–-0.203, p=0.034), and higher AcoA score was associated with later first-onset age (OR=1.665, 95% CI=0.343-2.9873, p=0.014). After adjustment for model 1 or model 2 (described above), there was still a significant association between first-onset age with whether poor CoW or AcoA score. First-onset age in elderly ischemic stroke was not associated with PcoA score, total CoW score, complete CoW and fetal PcoA (shown in Table 4).

Table 4. The association of first-onset age with CoW parameters

	Univariable analysis		Multivariable analysis (model 1)		Multivariable analysis (model 2)
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
AcoA score	1.665 (0.343–2.987)	0.014	1.367 (0.059–2.675)	0.041	1.336 (0.031–2.640)	0.045
PcoA score	0.148 (-0.939–1.236)	0.789	0.179 (-0.882–1.240)	0.740	-0.025 (-1.085–1.035)	0.963
Total CoW score	0.713 (-0.110–1.537)	0.089	0.621 (-0.193–1.435)	0.134	0.497 (-0.321–1.315)	0.233
Poor CoW	-2.671 (-5.140–-0.203)	0.034	-2.629 (-5.033–-0.226)	0.032	-2.144 (-4.588–0.269)	0.081
Complete CoW	1.324 (-2.307–4.955)	0.473	0.586 (-3.023–4.196)	0.320	0.605 (-2.989–4.199)	0.741
Fetal PcoA	-2.181 (-5.745–1.383)	0.229	-2.013 (-5.539–1.513)	0.262	-2.306 (-5.817–1.204)	0.197

Abbreviations: OR, odds ratio; CI, confidence interval; AcoA, anterior communicating artery; PcoA, posterior communicating artery; CoW, circle of Willis. Model 1: adjusted for hypertension history, diabetes mellitus history, coronary heart disease history, atrial fibrillation history, hyperlipidemia, hyperhomocysteinemia, and current smoking. Model 2: adjusted for factors in model 1 + medication history (antiplatelets, Statins, oral anticoagulants, antihypertensives and antidiabetics).

3.5 Influence on the association of CoW with first-onset age by gender

Substantial evidence reveals that both CoW configuration and stroke condition differ in gender. Thus, we further investigated the differences in CoW score and type between the oldest-old and the elderly subjects with first-ever ischemic stroke according to their gender. The demographic, clinical, and CoW characteristics of male and female cohorts from two groups are presented in Table 5. In the male cohort, the oldest-old subset showed a higher frequency of atrial fibrillation history, as well as a lower frequency of diabetes history and previous antidiabetics use, compared with the elderly subset. In the female cohort, a higher frequency of coronary heart disease history was found in the oldest-old subset. Moreover, a higher NIHSS score at admission in the oldest-old subset was found in the male cohort, which was not shown in the female cohort.

Table 5. CoW characteristics of male and female cohorts

	Male				Female
Characteristic	Elderly (n=129)	Oldest-old (n=46)	P value	Elderly (n=59)	Oldest-old (n=42)	P value
Demographic factors
Age, year, median, (IQR)	70(64.5-74)	85(81-88)	-	70(65-73)	84(81-87.25)	-
Vascular risk factors, n (%)
Atrial fibrillation	7(5.4)	9(19.6)	0.006	3(5.1)	6(14.3)	0.158
Diabetes mellitus	52(40.3)	8(17.4)	0.006	27(45.8)	19(45.2)	1.000
Arterial hypertension	86(66.7)	28(60.9)	0.478	42(71.2)	28(66.7)	0.666
Dyslipidemia	37(28.7)	14(30.4)	0.851	23(39.0)	12(28.6)	0.298
Coronary heart disease	13(10.1)	9(19.6)	0.120	7(11.9)	14(33.3)	0.012
Hyperhomocysteinemia	24(18.6)	10(21.7)	0.667	10(16.9)	10(23.8)	0.452
Current smoking	16(12.4)	5(10.9)	0.803	0(0.0)	1(2.4)	0.416
Previous medication, n (%)
Antiplatelets	35(27.1)	17(37.0)	0.260	18(30.5)	17(40.5)	0.396
Oral anticoagulants	5(3.9)	4(8.7)	0.245	1(1.7)	2(4.8)	0.569
Statins	32(24.8)	8(17.4)	0.319	19(32.2)	13(31.0)	0.535
Antihypertensives	72(55.8)	24(52.2)	0.399	28(47.5)	25(59.5)	0.312
Antidiabetics	43(33.3)	6(13.0)	0.012	24(40.7)	15(35.7)	0.681
Clinical parameters, median, (IQR)
NIHSS at admission	2(1-3)	2(2-4.25)	0.015	2(1-4)	3(1.75-5)	0.145
Time onset to admission, hour	20(12-40)	14(10.75-48)	0.749	14(9-24)	14(12-48)	0.187
Cow parameters
AcoA score, median, (IQR)	2 (1-2)	2 (1-2)	0.316	2 (0-2)	2 (2-2)	<0.001
PcoA score, median, (IQR)	0 (0-1)	0 (0-1.25)	0.634	1 (0-2)	2(0-2)	0.033
Total CoW score, median, (IQR)	2(1-2)	2(2-3)	0.437	2 (2-4)	2(2-4)	<0.001
Poor CoW (0-2), n (%)	105 (81.4)	33 (71.7)	0.170	49 (83.1)	24 (57.1)	0.004
Fetal PcoA, n (%)	9(7.0)	4(8.7)	0.460	10(16.9)	4(9.5)	0.385
Total CoW score, n (%)			0.629			<0.001
Score 0	19 (14.7)	8 (17.4)		12 (20.3 )	0 (0.0)
Score 1	21 (16.3)	5 (10.9)		9 (15.3)	1 (2.4)
Score 2	64 (49.6)	20 (43.5)		28 (47.5)	23 (54.8)
Score 3	9 (7.0)	4 (8.7)		3 (5.1)	5 (11.9)
Score 4	13 (10.1)	8 (17.4)		5(8.5)	11 (26.2)
Score 5	2 (1.6)	0 (0.0)		2 (3.4)	1 (2.4)
Score 6	1 (0.8)	1 (2.2)		0 (0.0)	1 (2.4)
CoW type, n (%)			0.225			0.006
Type I	7 (5.5)	6 (13.0)		6 (10.2)	7 (16.7)
Type II	95 (73.6)	30 (65.2)		38 (64.4)	34 (81.0)
Type III	8 (6.2)	1 (2.2)		4 (6.8)	1(2.4)
Type IV	19 (14.7)	9 (19.6)		11 (18.6)	0 (0.0)

Abbreviations: IQR, interquartile range; CoW, circle of Willis; AcoA, anterior communicating artery; PcoA, posterior communicating artery; AcoW, anterior CoW; PcoW, posterior CoW; NIHSS, National Institutes of Health Stroke Scale.

Interestingly, there were no differences in the AcoW score, PcoW score, total CoW score, poor CoW and fetal PcoA between oldest-old subset and elderly subset in the male cohort. However, in the female cohort, the oldest-old subset showed high scores of AcoA, PcoA and total CoW, with difference strength significantly greater than those in all subjects (male+female, shown in Table 2, Table 5). Meanwhile, within the female cohort, a lower frequency of poor CoW was found in the oldest-old subset compared with the elderly subset, with a greater ratio gap of poor CoW between two age subsets in the female cohort. Significant distribution differences of the total CoW score, AcoW score, PcoW score and CoW type between the oldest-old patients and the elderly patients were found in the female cohort, whereas not in the male cohort (Figure 1).

There was no association between first-onset age and AcoA score, total CoW score and poor CoA in the male cohort. However, in the female cohort, first-onset age was significantly related with AcoA score, total CoW score and poor CoW whether non-adjusted or adjusted for model 1 and model 2 (shown in Table 6).

Table 6. The association of first-onset age with CoW parameters in male and female cohorts

	Male		Female
Univariable analysis	OR (95% CI)	P-value	OR (95% CI)	P-value
AcoA score	0.771 (-0.855–2.397)	0.351	3.050 (0.768–5.331)	0.009
Total Cow score	-0.131 (-1.176–0.913)	0.804	1.889 (0.572–3.206)	0.005
Poor CoW	-0.611 (-3.848–2.626)	0.710	-4.795 (-8.643–-0.948)	0.015
Multivariable analysis (model 1)
AcoA score	0.690 (-0.916–2.297)	0.397	2.891 (0.543–5.239)	0.016
Total Cow score	-0.102 (-1.143–0.939)	0.847	1.979 (0.643–3.314)	0.004
Poor CoW	-0.681 (-3.859–2.497)	0.673	-5.543 (-9.429–-1.656)	0.006
Multivariable analysis (model 2)
AcoA score	0.662 (-0.948–2.271)	0.418	2.984 (0.564–5.404)	0.016
Total Cow score	-0.321 (-1.373–0.730)	0.547	1.887 (0.509–3.266)	0.008
Poor CoW	0.323 (-2.926–3.572)	0.845	-5.140 (-9.122–-1.159)	0.012

Abbreviations: OR, odds ratio; CI, confidence interval; AcoA, anterior communicating artery; PcoA, posterior communicating artery; CoW, circle of Willis. Model 1: adjusted for hypertension history, diabetes mellitus history, coronary heart disease history, atrial fibrillation history, hyperlipidemia, hyperhomocysteinemia, and current smoking. Model 2: adjusted for factors in model 1 + medication history (antiplatelets, Statins, oral anticoagulants, antihypertensives and antidiabetics).

The occurrence and severity of ischemic stroke is strongly associated with vascular risk factors^[12]. Timely diagnosis and control of cerebrovascular risk factors is pivotal for the primary and secondary prevention of ischemic stroke. Aging, hypertension, diabetes mellitus and atrial fibrillation are the most common risk factors for acute cerebrovascular events. However, late-onset stroke patients seem to be immune to or escape from the influence of those risk factors. It hints that there might be some underlying protective factors or mechanisms in the oldest-old patients with first-onset acute ischemic stroke. Numerous studies suggest that CoW variants impact the patterns of ischemic pathology, the outcome and severity of occlusion, and subsequent neurosurgical approaches, in varying degrees, due to its morphological and functional characteristics^[13]. Therefore, in the present article, we proposed a hypothesis that “late-onset” oldest-old subjects with acute ischemic stroke may have better CoW configuration in comparison to “early-onset” elderly subjects. By two assessment methods (CoW score and CoW style), we found that the oldest-old group is more likely to demonstrate a high CoW score and complete CoW style, with even more significant in females.

It is well-known that stroke is a heterogeneous, multifactorial disease regulated by modifiable and non-modifiable risk factors^[14], and the determination of the feature of a risk factor for ischemic stroke may contribute to more effective prevention. Our data and multiple lines of evidence support the poor CoW as a risk factor for ischemic stroke. However, the kind of risk factor for poor CoW, modifiable or non-modifiable, should be interpreted with some caution. A previous study investigating the MR imaging CoW differences between healthy children and adults, revealed that the CoW is more symmetric in children, with non-visualized segments more frequently in adults^[15]. From an evolutionary point of view, the results mentioned above were interpreted that CoW got more incomplete and asymmetric with age. Nonetheless, in the elderly population, it is unlikely that the CoW steers evolution and solely develops to compensate for insufficient blood flow at the physiological background of artery occlusions. Furthermore, from a physiological point of view, some scholars believe that the communicating arteries are too small for effective collateral blood supply. In view of these information, we hold an opinion that the poor Cow should be more deemed as a “non-modifiable” risk factor for ischemic stroke. Earlier detection for poor CoW cases and timely launching stricter control for other modifiable risk factors in those cases appears to be particularly important, which could promote the late-onset or never-onset for ischemic stroke in the elderly population.

According to our data, gender deserves more attention in the elderly. In the present study, the oldest-old female subjects showed higher scores for total CoW, AcoW half and PcoW half than the elderly female subjects, and a greater ratio gap of poor CoW between the two age groups in females was found in comparison to that in males. In line with our findings, sex difference in CoW variants was also observed in other age groups. Chopra T et al. found a higher percentage of males with an incomplete CoW compared with females (p < 0.0001) in children^[16]. Kızılgöz V et al. revealed that in the group younger than 40 years old, there was a significant sex difference regarding the distribution of the complete, partially complete, and incomplete circle groups, and a higher percentage of complete circles were observed in females^[17]. Combined with our study, this information hinted that: as a whole, the female population has an innate advantage in CoW morphology and configuration in comparison to the male population; as an individual, a female elderly subject with poor CoW variant is more susceptible to early-onset acute ischemic stroke. Female poor CoW should be considered in risk stratification, and thus for precision medicine, a female elderly subject with poor CoW should be recommended more early intervention and stricter management for the known cerebrovascular risk factors, such as atrial fibrillation, diabetes, hypertension, hyperlipidemia, coronary heart disease, et al.

Multiple lines of evidence demonstrated that women were more affected by stroke than men. It is reported that worse functional outcome and higher mortality among women was observed in the United States^[18]. A study based on Latin American Stroke Registry reached the same conclusions^[19]. In the Australian stroke clinical registry, women were more likely to be deceased up to 1 disability-adjusted life year after stroke versus men^[20]. A meta‐analysis of population‐based studies found that stroke tended to be more severe in women, with a 1‐month case fatality of 24.7% compared with 19.7% for men^[21]. The reasons for these sex disparities have been attributed to longer life expectancy, less prophylactic application, more severe strokes, and likely lower quality of care. Another possible explanation is the physiological differences between men and women, such as hormonal effects. The present study showed a trend towards lower scores for AcoW, left PcoW half, right PcoW half, as well as total CoW, in female elderly patients with first-onset ischemic stroke compared with the male. It offers another physiological clue that CoW configuration may be an underlying physiological variable for sex disparities in stroke. Such a detrimental effect of poor CoW on female stroke patients should be confirmed systematically in the future.

The oldest-old, those over 80 years, are the fastest growing segment of the population, and represent a distinct group, both clinically and pathologically, compared to younger old individuals. The oldest-old with first-ever ischemic stroke deserves to be taken as a specific group in analyses, not only due to being at the outermost extreme of human life, but also because of its distinct clinical feature^[22]. More understanding of stroke characteristics and potential contributing factors is important for investing the mechanism of healthy aging and corresponding prevention strategies. Thus, investigating the association of CoW morphology with first-onset age of ischemic stroke may provide a novel window for stroke prevention and healthy aging. After all, the neuroimaging data in the oldest-old stroke subjects is more plentiful than in healthy aging individuals.

In summary, our study shows that the oldest-old patients with first-onset ischemic stroke have a higher CoW score and better morphology than the elderly patients, with females more significant. To our knowledge, it is the first work focusing on the oldest-old patients with first-onset ischemic stroke and investigating the association of onset age with CoW in the elderly population, it needs to be verified in future independent studies with a larger sample size. Our results help to improve our understanding of the underlying mechanisms of late-onset ischemic stroke in the oldest-old patients. More importantly, we propose that poor CoW in females should be considered in risk stratification for ischemic stroke in order to provide more effective and precise prevention and treatment.

CoW, circle of Willis; AcoA, anterior communicating artery; PcoA, posterior communicating artery; NIHSS, National Institutes of Health Stroke Scale; MRI, magnetic resonance imaging; MRA, magnetic resonance angiography; CTA, computed tomography angiography; OR, odds ratio; CI, confidence interval; IQR, interquartile range.

Acknowledgements

We acknowledge the patients of Bethune International Peace Hospital who gave us consent to obtain this information.

Authors’ contributions

XFC– conception, design of work, acquisition, analysis, interpretation of data, drafted and substantively revised the manuscript, SSJ– analysis, interpretation of data, drafted and substantively revised the manuscript, JXZ– analysis, interpretation of data, drafted and substantively revised the manuscript, MML and LPC– design of work, acquisition of data, drafted and substantively revised the manuscript, ZHG and HYM– analysis, interpretation of data, drafted and substantively revised the manuscript, QHW–interpretation of data, drafted and substantively revised the manuscript. All authors read and approved the final manuscript.

Funding

The study was supported by the Scientific Research Project of Hebei Provincial Administration of Traditional Chinese Medicine (No. 2024258 to Xiaofang Cheng), the Health Commission of Hebei Province (No. 20240528 and No. 20180915 to Shusheng Jiao). The funding project had no role in the design of the study and collection, analysis, and interpretation of data and no role in writing the manuscript.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Written informed consent was obtained from all participants or legal authorized representative to participate in the study. Ethical approval was obtained from Bethune International Peace Hospital. All methods and procedures were carried out in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Author details

¹Department of Neurology, Bethune International Peace Hospital, Shijiazhuang, Hebei, China.

²Shijiazhuang 10th Cadre Rest Center of Hebei Military Region, Shijiazhuang, Hebei, China.

³Department of Neurology, Daping Hospital, Army Medical University, Chongqing, China

Gur A Y, Tanne D, Bornstein N M, et al. Stroke in the very elderly: characteristics and outcome in patients aged >/=85 years with a first-ever ischemic stroke[J]. Neuroepidemiology, 2012,39(1):57-62.
Auriel E, Gur A Y, Uralev O, et al. Characteristics of first ever ischemic stroke in the very elderly: profile of vascular risk factors and clinical outcome[J]. Clin Neurol Neurosurg, 2011,113(8):654-657.
Sposato L A, Lam M, Allen B, et al. First-ever ischemic stroke and increased risk of incident heart disease in older adults[J]. Neurology, 2020,94(15):e1559-e1570.
Jillella D V, Crawford S, Lopez R, et al. Vascular Risk Factor Prevalence and Trends in Native Americans with Ischemic Stroke[J]. J Stroke Cerebrovasc Dis, 2022,31(6):106467.
Ayre J R, Bazira P J, Abumattar M, et al. A new classification system for the anatomical variations of the human circle of Willis: A systematic review[J]. J Anat, 2022,240(6):1187-1204.
Jones J D, Castanho P, Bazira P, et al. Anatomical variations of the circle of Willis and their prevalence, with a focus on the posterior communicating artery: A literature review and meta-analysis[J]. Clin Anat, 2021,34(7):978-990.
Sirvinskas A, Lengvenis G, Ledas G, et al. Circle of Willis Configuration and Thrombus Localization Impact on Ischemic Stroke Patient Outcomes: A Systematic Review[J]. Medicina (Kaunas), 2023,59(12).
Lin E, Kamel H, Gupta A, et al. Incomplete circle of Willis variants and stroke outcome[J]. Eur J Radiol, 2022,153:110383.
Oumer M, Alemayehu M, Muche A. Association between circle of Willis and ischemic stroke: a systematic review and meta-analysis[J]. BMC Neurosci, 2021,22(1):3.
Xu M, Wu Q, Cheng Y, et al. Circle of Willis Morphology in Primary Intracerebral Hemorrhage[J]. Transl Stroke Res, 2022,13(5):736-744.
Krabbe-Hartkamp M J, van der Grond J, de Leeuw F E, et al. Circle of Willis: morphologic variation on three-dimensional time-of-flight MR angiograms[J]. Radiology, 1998,207(1):103-111.
Sigurdsson S, Aspelund T, Kjartansson O, et al. Cerebrovascular Risk-Factors of Prevalent and Incident Brain Infarcts in the General Population: The AGES-Reykjavik Study[J]. Stroke, 2022,53(4):1199-1206.
De Caro J, Ciacciarelli A, Tessitore A, et al. Variants of the circle of Willis in ischemic stroke patients[J]. J Neurol, 2021,268(10):3799-3807.
Mishra A, Malik R, Hachiya T, et al. Stroke genetics informs drug discovery and risk prediction across ancestries[J]. Nature, 2022,611(7934):115-123.
Guilliams K P, Gupta N, Srinivasan S, et al. MR Imaging Differences in the Circle of Willis between Healthy Children and Adults[J]. AJNR Am J Neuroradiol, 2021,42(11):2062-2069.
Chopra T, Neuberger I, Prince E, et al. Age-related changes in the completeness of the circle of Willis in children[J]. Childs Nerv Syst, 2022,38(6):1181-1184.
Kizilgoz V, Kantarci M, Kahraman S. Evaluation of Circle of Willis variants using magnetic resonance angiography[J]. Sci Rep, 2022,12(1):17611.
Girijala R L, Sohrabji F, Bush R L. Sex differences in stroke: Review of current knowledge and evidence[J]. Vasc Med, 2017,22(2):135-145.
Arauz A, Serrano F, Ameriso S F, et al. Sex Differences Among Participants in the Latin American Stroke Registry[J]. J Am Heart Assoc, 2020,9(4):e13903.
Phan H T, Gall S L, Blizzard C L, et al. Sex Differences in Care and Long-Term Mortality After Stroke: Australian Stroke Clinical Registry[J]. J Womens Health (Larchmt), 2019,28(5):712-720.
Barker-Collo S, Bennett D A, Krishnamurthi R V, et al. Sex Differences in Stroke Incidence, Prevalence, Mortality and Disability-Adjusted Life Years: Results from the Global Burden of Disease Study 2013[J]. Neuroepidemiology, 2015,45(3):203-214.
Marini C, Baldassarre M, Russo T, et al. Burden of first-ever ischemic stroke in the oldest old: evidence from a population-based study[J]. Neurology, 2004,62(1):77-81.

No competing interests reported.

From arteries to age: an intrinsic impact of Circle of Willis on first-ever acute ischemic stroke in the oldest-old

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and methods

3. Results

4. Discussion

5. Conclusions

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1