Antiplatelet therapy for primary prevention of stroke in patients with carotid web: a prospective cohort study

Introduction

A carotid web is a membrane-like defect that usually originates from the posterior wall of carotid bulb (1). By altering local hemodynamics and inducing blood stasis, carotid webs promote in-situ thrombosis and subsequent distal embolization (2). Although relatively rare, it is an increasingly recognized etiology for cryptogenic stroke, especially in younger patients (3). The prevalence of carotid web was reported to range from 8 to 37% in young patients with cryptogenic stroke (4, 5), compared to less than 1% in non-stroke populations (2, 6). Symptomatic carotid webs have a high stroke recurrence rate of 17 to 56% (7, 8), making secondary stroke prevention essential.

The optimal therapy for carotid web is still unclear (9). Current prevention guidelines recommend antiplatelet therapy for patients with carotid web in the distribution of ischemic stroke or transient ischemic attack (TIA), if no other causes are found, based on case series (10). However, the effect of antiplatelet therapy on asymptomatic carotid webs is unknown. Therefore, we conducted this prospective cohort study to evaluate the efficacy and safety of single antiplatelet therapy as primary prevention of stroke in asymptomatic carotid web.

Methods

Participants

We conducted a prospective observational cohort study at the Affiliated Hospital of Qingdao University. Patients were prospectively enrolled between June 2017 and June 2022 after a carotid web was identified on CT angiography (CTA). The clinical indications for CTA included vascular abnormalities detected by ultrasound during annual physical examinations, preoperative assessment, and patients with dizziness, headache, syncope or vertigo. Images of CTA were independently reviewed by two neuroradiologists. A carotid web was defined as a thin, membrane-like filling defect along the posterior wall of the carotid bulb or the origin of the internal carotid artery on oblique sagittal images, with a corresponding thin septum on axial images (5, 11). Patients with no history of cerebral ischemic events and no cerebral infarction on baseline magnetic resonance imaging (MRI) were classified as having asymptomatic carotid webs.

The study was approved by the institutional Ethics committee of the Affiliated Hospital of Qingdao University (no. 2017-102) and conducted in accordance with the Declaration of Helsinki and relevant Chinese regulations on clinical research. Written informed consent to participate was obtained from all participants.

Study design

The prescription of aspirin was evaluated by the treating physicians based on an individualized clinical assessment, including ASCVD risk, the degree of carotid stenosis, and the clinical assessment of the carotid web. Patients receiving single antiplatelet therapy (aspirin 100 mg/day) were included in antiplatelet cohort, otherwise, they were included in the non-antiplatelet cohort. Patients receiving anticoagulant therapy, switching to carotid revascularization, discontinuing antiplatelet therapy or those with other etiologies of stroke, such as atrial fibrillation, large artery atherosclerosis, cerebral small vessel disease or connective tissue disorders were also excluded. Two patients who underwent carotid revascularization at the time the carotid web was identified were excluded.

Baseline clinical characteristics including age, sex, and atherosclerotic risk factors were extracted from the medical records at enrollment. Dyslipidaemia was defined as elevated total cholesterol, low-density lipoprotein (LDL) cholesterol, or triglycerides, or a low level of high-density lipoprotein (HDL) cholesterol (12). Carotid stenosis were graded using CTA images (13). The 10-year risk for atherosclerotic cardiovascular disease (ASCVD) was assessed according to 2021 European Society of Cardiology (ESC) Guidelines on cardiovascular disease prevention (14).

Outcomes

Patients were prospectively followed at 3-month intervals by telephone or face-to-face interview. The primary outcome was cerebral ischemic stroke, defined as new-onset neurological deficits with corresponding acute cerebral infarction on MRI in the vascular territory of the ipsilateral carotid web. Follow-up was calculated from the index CTA to the occurrence of the primary outcome, the last documented clinical contact, or the end of the study period, as applicable. Patients who did not experience the primary outcome were censored at their last documented contact. Adverse reaction associated with antiplatelet therapy were also recorded to evaluate safety.

Statistical analyses

Statistical analyses were conducted using R software (version 4.3.1, Vienna, Austria). Categorical variables were analyzed with the Chi-square test or Fisher’s exact test. Age was summarized as the median with range, and compared by Mann–Whitney U test depending on data distribution. Standardized mean differences (SMD) were calculated to assess baseline covariate imbalances between the two cohorts.

Given the limited sample size and low event rate in this rare disease cohort, we conducted Cox regression with Firth’s penalized maximum likelihood method to evaluate the effect of antiplatelet therapy on stroke (15). Multivariable Cox regression with Firth’s penalized maximum likelihood method was adjusted for ASCVD risk and carotid stenosis. Results were expressed as hazard ratios (HR) with 95% confidence intervals (CIs). To address potential confounding from observed baseline imbalances, we conducted a sensitivity analysis by incorporating a propensity score as an additional covariate in the Firth penalized Cox regression. Propensity scores were estimated using a logistic regression model, incorporating baseline variables selected based on clinical relevance and the magnitude of baseline imbalance. Incidence rates were calculated as events per 100 person-years, with exact 95% confidence intervals computed using the Poisson distribution. To characterize the statistical precision of the survival analysis, the approximate detectable HR was estimated using the Schoenfeld approximation based on the observed number of outcome events and the observed exposure distribution, assuming 80% power and a two-sided α of 0.05. p < 0.05 was considered significant.

Results

A total of 123 patients with asymptomatic web were initially identified. We excluded 14 patients (2 receiving oral anticoagulation, 2 undergoing carotid revascularization, and 10 with other causes of stroke). Of the remaining 109 eligible patients, 4 (3.7%) were lost to follow-up. These four patients were not included in the analysis because reliable follow-up information was insufficient to define a valid censoring time. Ultimately, 105 patients were included in the final analysis. During a median follow-up of 31 months (range from 9 to 89 months), 13 patients (12.4%, 13/105) experienced an ischemic stroke (Figure 1). Clinical and Imaging characteristics of patients with stroke during follow-up were present in Supplementary Table S1. No deaths occurred during follow-up.

Baseline clinical characteristics of patients are presented in Table 1. The median age of patients with asymptomatic carotid web was 61 years old (range from 19 to 80 years), with 41.0% (43/105) aged older than 65 years. As for ASCVD risk, 27 patients (25.7%) were classified as high risk, with 8 patients (7.6%) as very high risk. Besides, hypertension was the most common atherosclerotic risk factor (43.8%, 46/105). Only 5 patients (4.8%) were detected with over 50% stenosis of carotid, including 2 patients with over 70% stenosis. Bilateral carotid web were observed in 17 patients (16.2%).

Table 1. Comparison of characteristics between antiplatelet and non-antiplatelet cohorts in patients with carotid webs.

Variables were present as n (%) except age.

Antiplatelet therapy

Patients in the antiplatelet cohort were significantly older than those in the non-antiplatelet cohort (median age, 67 vs. 58 years, p = 0.001, Table 1). Hypertension was more common in the antiplatelet cohort (64.5% vs. 35.1%, p = 0.006). In addition, the rate of patients with high or very high ASCVD risk was significantly higher in antiplatelet cohort (58.1% vs. 23.0%, p = 0.001).

Primary prevention of ischemic stroke

During the follow-up, stroke occurred in 3 of 31 patients in the antiplatelet group and 10 of 74 patients in the non-antiplatelet group. The corresponding incidence rates were 3.51 (95% CI = 0.72–10.25) and 5.16 (95% CI = 2.47–9.50) per 100 person-years, respectively. Univariable Firth-penalized Cox regression yielded an HR of 0.755 (95% CI = 0.215–2.653, p = 0.642, Figure 2). After adjustment for ASCVD risk and carotid stenosis, the HR was 0.665 (95% CI = 0.167–2.655; p = 0.553, Table 2). In the propensity score-adjusted sensitivity analysis, the HR was 0.982 (5% CI = 0.246–3.916, p = 0.979, Table 2).

Table 2. Association between antiplatelet therapy and stroke in asymptomatic carotid webs.

HR, hazard ratio; CI, confidence interval; ASCVD, atherosclerotic cardiovascular disease. All analyses were conducted by using Firth’s penalized likelihood Cox regression.

Propensity score derived from age, hypertension, and stenosis.

Regarding adverse effects, one patient reported gingival bleeding, and one patient reported hematochezia. Both patients continued antiplatelet therapy despite these mild bleeding events, and their symptoms resolved within days.

Primary prevention in patients without high ASCVD risk

In the stratified analysis of 70 patients without high ASCVD risk, 13 patients (18.6%) received antiplatelet therapy. During a median follow-up of 31 months (range from 12 to 66 months), stroke occurred in 1 patient (7.7%, 1/13) in the antiplatelet cohort and 8 patients (14.0%, 8/57) in the non-antiplatelet cohort (Table 3). Consistent with the primary analysis, no significant association was detected between antiplatelet therapy and stroke in this subgroup (HR = 0.671, 95% CI = 0.108–4.191, p = 0.638, Table 2).

Table 3. Stratified analysis in patients without high ASCVD risk.

Variables were present as n (%) except age. NA, not applicable.

Discussion

The optimal management of asymptomatic carotid webs remains uncertain. In this prospective cohort study, ischemic stroke occurred less frequently with single antiplatelet therapy, although the association was not statistically significant. Given the small number of outcome events and the wide confidence intervals, the estimate remained compatible with both potential benefit and harm.

The incidence of ipsilateral ischemic stroke in the non-antiplatelet group was 5.16 per 100 person-years. The ipsilateral distribution and isolated or multifocal cortical pattern of all 13 events, together with the exclusion of major competing stroke etiologies, support a thromboembolic mechanism related to the carotid web. This finding raises concern that some patients with asymptomatic carotid web may remain at clinically relevant risk. While current guidelines lack a consensus on the management of asymptomatic carotid webs, surveys indicate that approximately 39% of neurologists and neurointerventionalists prefer antiplatelet therapy for these patients (16).

In this prospective cohort, stroke incidence was numerically lower with single antiplatelet therapy, but estimates were imprecise and compatible with both benefit and harm. Beyond antiplatelet therapy, oral anticoagulation has been considered a theoretical alternative, given the presumed stasis-induced thrombogenesis that similar to atrial fibrillation. Nevertheless, clinical evidence supporting anticoagulant use for asymptomatic carotid webs is currently lacking (9). To eliminate confounding, two patients receiving oral anticoagulants at baseline were explicitly excluded from our cohort. Multi-arm clinical studies are warranted to compare the efficacy and safety of anticoagulants versus antiplatelet agents. Additionally, carotid revascularization has been investigated as a structural treatment for carotid webs, primarily in symptomatic patients. A recent meta-analysis of observational studies indicate that revascularization is associated with a lower risk of recurrent stroke in symptomatic carotid webs, compared to medical management (17). However, these findings cannot be directly extrapolated to asymptomatic carotid webs. Furthermore, while no severe adverse effects were observed in our cohort, the potential hemorrhagic complications of antiplatelet agents should be carefully weighed against their benefits, particularly during long-term primary prevention. Information on longitudinal antihypertensive treatment and blood pressure control was not systematically available, and residual confounding related to concomitant cardiovascular therapy cannot be excluded.

Consistent with previous reports, concurrent carotid stenosis of ≥50% was uncommon in our cohort, occurring in only 4.8% of patients (2). Carotid webs may enlarge the recirculation zone, resulting in disturbed flow and focal blood stasis in the post-web space, which progressively promotes in-situ thrombosis (18). Based on this hemodynamic mechanism, risk stratification using specific anatomic and angioarchitectural features may facilitate individualized treatment. Previous cross-sectional studies indicate that specific morphological parameters, such as the length of the web and its relationship to the carotid bifurcation, are associated with an increased risk of stroke (19–21). However, reliable morphological criteria for risk stratification in asymptomatic carotid web have not been established. Notably, asymptomatic re-thrombosis in patients with carotid web-related stroke indicated that structured imaging surveillance may be useful in stroke prevention (22).

Our study has several limitations. First, based on the 13 observed events and the observed distribution of antiplatelet exposure, the approximate detectable HR at 80% power and a two-sided α of 0.05 was ≤0.18 or ≥5.49, indicating that only very large treatment effects could have been reliably detected. Therefore, the absence of statistical significance should not be interpreted as evidence of no treatment effect. Larger multicenter studies with more outcome events are needed to provide more precise estimates. Second, the non-randomized observational design introduces inherent confounding by indication, as patients with higher ASCVD risk or concurrent mild stenosis were more likely to be prescribed single antiplatelet therapy. While multivariable Cox regression with Firth’s penalized maximum likelihood method, propensity score-adjusted sensitivity analyses, and subgroup analyses yielded consistent results, unmeasured confounding cannot be entirely excluded. Third, we did not stratify the effect of antiplatelet therapy based on detailed anatomic and angioarchitectural features, which requires further investigation. Forth, subclinical thrombus was not evaluated by duplex or CTA during follow-up, which may underestimate the event burden. Finally, considering the presumed lifelong cumulative stroke risk associated with carotid webs, our median follow-up of 31 months may be insufficient to fully capture long-term clinical outcomes.

Conclusion

In this prospective cohort study, stroke incidence was numerically lower with single antiplatelet therapy, but estimates were imprecise and compatible with both benefit and harm. Further large-scale, multicenter, randomised controlled trials are warranted to establish optimal management guidelines for asymptomatic webs.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fneur.2026.1955675/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Stefano Maria Priola, Health Sciences North, Canada

Reviewed by: Junzo Nakao, Osaka Medical and Pharmaceutical University, Japan

Pablo Ros-Arlanzón, Hospital General Universitario de Alicante, Spain