The role of intraplaque neovascularization in recent and future ischemic stroke in patients with mild carotid stenosis
Article information
Abstract
Purpose
There is still insufficient evidence for predicting stroke risk in patients with mild carotid atherosclerotic stenosis. This study aimed to explore the association between carotid intraplaque neovascularization (IPN) in mild stenosis and ischemic stroke, using contrast-enhanced ultrasound (CEUS) imaging.
Methods
This retrospective observational study included 369 patients from July 2021 to March 2022. These patients were categorized as symptomatic or asymptomatic based on their recent history of ipsilateral ischemic stroke. Initial parameters of carotid plaques, such as IPN grading and contrast enhancement index, were assessed using B-mode ultrasonography and CEUS. The follow-up period lasted 12 months or until a newly-developed ischemic stroke occurred. Logistic regression models and Cox proportional-hazards regression models were employed to explore the associations between ultrasonic parameters and the incidence of recent and future ischemic strokes.
Results
In patients with mild stenosis, both increasing age and grade 2 carotid IPN were significant predictors of recent primary ischemic stroke. Furthermore, grade 2 carotid IPN independently predicted future ischemic strokes in both symptomatic and asymptomatic patients.
Conclusion
This study demonstrated that carotid IPN as detected by CEUS imaging holds potential as a useful non-invasive biomarker for predicting recent and future ischemic strokes in patients with mild carotid stenosis.
Introduction
Ischemic stroke is a major cause of disability and mortality worldwide. Carotid atherosclerosis is widely recognized as a contributing factor to ischemic stroke [1]. Traditionally, the degree of carotid stenosis has been the primary measure used to stratify risk in patients, with clinical surgical interventions such as carotid endarterectomy or stenting being recommended primarily for symptomatic patients with moderate to severe stenosis (50% to 99%) [2]. Increasing evidence, however, has shown that the rupture of vulnerable carotid plaques can lead to thrombus formation and subsequent distal embolization in intracranial arteries, regardless of the degree of stenosis [3,4]. Nevertheless, the significance of plaque vulnerability in patients with mild stenosis (<50%) in relation to ischemic stroke remains to be elucidated.
Intraplaque neovascularization (IPN) is a reliable indicator of carotid plaque vulnerability and has been proven to be an independent predictor of plaque rupture [5]. Contrast-enhanced ultrasound (CEUS) is a novel imaging technique that uses microbubble contrast agents as intravascular tracers. This method significantly enhances the visualization of the vessel lumen, free from artifacts, and allows for more precise delineation of the plaque surface compared to B-mode ultrasound, facilitating accurate evaluation of the degree of carotid stenosis [6-8]. Additionally, CEUS provides high-resolution visualization of IPN and enables precise assessment of IPN grading and enhancement within carotid plaques [5,9,10]. Notably, the assessment of IPN by CEUS has shown high consistency with the results of pathological examinations [11-15]. Previous studies involving small samples of patients who had recently experienced an ipsilateral ischemic stroke have demonstrated that IPN in carotid plaques, assessed by CEUS in cases of mild to severe stenosis, was an independent predictor for the recurrence of stroke and future cardiovascular events [16,17]. However, there have been few studies on the role of IPN in carotid plaques in the occurrence and recurrence of ipsilateral ischemic stroke in patients with mild carotid stenosis.
This study investigated IPN within carotid plaques using CEUS in both symptomatic and asymptomatic patients with mild stenosis. Additionally, it aimed to determine the association between neovascularization in carotid plaques and the occurrence of recent primary or future ipsilateral ischemic stroke in these patients.
Materials and Methods
Compliance with Ethical Standards
This study was approved by the local Ethics Committee of Shanghai General Hospital (No. [2022]028), and informed consent was obtained from all patients.
Study Design and Patient Selection
This retrospective case-control study enrolled 864 consecutive patients aged 38 years to 92 years, who were diagnosed with carotid plaques and underwent carotid B-mode ultrasound (US) and CEUS examinations, between July 2021 and March 2022.
The inclusion criteria were as follows: (1) the presence of at least one atherosclerotic plaque in the carotid artery ipsilateral to the intracranial ischemic lesion, with a plaque thickness of ≥2 mm [17]; (2) mild carotid artery stenosis (<50% stenosis degree) as determined by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria, assessed using CEUS examination [6-8,18,19]. The exclusion criteria included: (1) the presence of other thromboembolic sources such as vasculitis, fibrillation, and intracranial embolus; (2) transient ischemic attack (TIA), defined as an abnormal focal neurologic deficit lasting less than 24 hours, with no credible imaging evidence of an ischemic attack in the anterior cerebral circulation area ipsilateral to the carotid plaque; (3) poor image quality issues, including severe plaque calcification (type V according to the Gray-Weale-Nicolaides classification [20]), or incomplete clinical data; (4) previous neck irradiation or carotid surgery/stenting; (5) patients lost to follow-up, or those with a life expectancy of less than 1 year.
The included patients were divided into the asymptomatic group and the symptomatic group. The symptomatic group consisted of patients who were diagnosed with their first ipsilateral ischemic lesions in the carotid territory (anterior circulation) based on head computed tomography (CT) or magnetic resonance imaging (MRI) conducted within the previous 8 weeks [17], after excluding other thromboembolic sources [21,22]. The asymptomatic group included patients who showed no intracranial ischemic lesions on head CT/MRI images and exhibited no neurological symptoms.
Clinical Variables
The following variables were collected: (1) demographic data, including age, sex, and body mass index (BMI); (2) smoking history; (3) medical history, which encompasses hypertension, diabetes mellitus, and hyperlipidemia; and (4) drug use history, specifically the use of antihypertensive drugs, antidiabetic drugs, and statins.
US Study Protocol
B-mode US and CEUS examinations were conducted using an EPIQ-Elite scanner (Philips Medical System, Bothell, WA, USA) equipped with linear array L12-3 and L14-3 probes, respectively. Radiologists with over five years of experience in carotid US imaging performed all examinations. Standardized B-mode US and CEUS protocols and training were provided to all operators to ensure consistency in operation. During the examination, patients were required to maintain calm breathing while in a supine position. The common carotid artery, carotid bifurcation, and internal carotid artery were examined and documented in both longitudinal and transverse planes using B-mode US imaging. In the symptomatic group, the thickest plaque ipsilateral to the ischemic lesion was chosen as the target plaque for CEUS imaging and analysis, provided there were multiple plaques. For the symptomatic group, the thickest plaque on either side of the carotid arteries was selected as the target plaque.
CEUS was initiated with a 1.2 mL bolus injection of SonoVue microbubbles (Bracco Imaging S.p.A., Milan, Italy) as the contrast agent, followed by a 5 mL injection of normal saline via the peripheral vein. The mechanical index was adjusted to <0.1, and the gain and compression settings were also adjusted to ensure optimal imaging quality. CEUS videos were recorded in the longitudinal plane of the carotid plaque for a duration of 2 minutes, and all data were stored on a Philips medical system hard disk.
US Analysis of Carotid Plaques
US images and CEUS videos were analyzed offline using QLAB software (Philips Medical System). B-mode US parameters for both groups included plaque echogenicity, length, and thickness. Carotid IPN on CEUS imaging was semi-quantitatively graded based on the presence and location of microbubbles within the plaque. The IPN grading categories were defined as follows: grade 0, no visible microbubbles; grade 1, minimal microbubbles confined to the shoulder or adventitial side of the plaque; and grade 2, extensive intraplaque enhancement with microbubbles present within the core (Fig. 1) [23]. Additionally, when microbubbles were observed flowing from the carotid lumen into the plaque, this was also classified as grade 2 IPN.
Classification of carotid intraplaque neovascularization (IPN) grading.
Diagram (left), B-mode US image (middle) and CEUS image (right) of grade 0 (A), grade 1 (B), and grade 2 (C) are shown. The red solid lines outline the arterial walls. The white dotted lines outline the plaque lesions (R1). The white circles of 2 mm diameter outline the adjacent carotid lumina (R2). Red stars in the plaque depict intraplaque contrast microbubbles.
As shown in Fig. 1C, the regions of interest were manually delineated: the margin of the plaque was outlined and designated as R1, while R2 was defined as a circular region with a 2 mm diameter in the adjacent carotid lumen. Plaque enhancement was quantitatively analyzed using time-intensity curves on QLAB software. The basic intensity (BI) (pre-contrast) and peak intensity (PI) for both R1 and R2 were recorded. Enhanced intensity (EI) was calculated by subtracting the BI from the PI (EI=PI-BI). Additionally, the ratio of EI in the plaque to that in the arterial lumen was determined (EI ratio=EIR1/EIR2) [24].
Manual delineation and acquisition of US parameters from B-mode US and CEUS images and videos were carried out by two radiologists, each with five years of experience in carotid US. They were blinded to the patients’ histories. The assessment of the degree of carotid stenosis was conducted according to the NASCET criteria on CEUS images [18,19]. In cases of inconsistencies, the final decision was made by the senior radiologist, who has over 10 years of experience.
Follow-up and Endpoint
The primary outcome of the study was the occurrence of new ischemic anterior circulation lesions, confirmed by head CT/MRI imaging, and located ipsilaterally to the carotid plaque assessed by CEUS at the start of enrollment. Patients were monitored for 12 months or until a new ischemic stroke lesion was detected during the follow-up period. A new ischemic stroke lesion was defined as a new acute infarct on the ipsilateral side of the targeted carotid plaque, as determined by head CT/MRI imaging, while excluding any alternative thromboembolic sources [21,22,25].
Statistical Analysis
All statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Categorical data were presented as frequencies (percentages) and analyzed with either the Pearson chi-square test or the Fisher exact test, depending on appropriateness. Continuous data that followed a normal distribution were reported as means±standard deviations and analyzed using the independent-samples t-test. Risk factors for ischemic stroke were determined through both univariable and multivariable logistic regression analyses, with results reported as odds ratios (ORs) and 95% confidence intervals (CIs). Statistical significance was set at P<0.05. Cox proportional-hazards regression modeling was utilized to identify independent predictors of future ischemic stroke. The cumulative incidence of future stroke was compared across different IPN grades using the Kaplan-Meier method and the log-rank test.
Interobserver agreement on IPN grading was assessed by two independent radiologists, each blinded to the other's results. To evaluate intra-observer consistency, the data from 30 patients were reanalyzed by the same radiologist one month later, without referencing the initial results.
Results
Study Population and Baseline Clinical Characteristics
Among the 864 patients who underwent carotid CEUS, 495 were excluded for various reasons: involvement in other thromboembolic sources (n=56), poor image quality (n=91), incomplete clinical data (n=296), prior neck irradiation or ipsilateral carotid surgery/stenting (n=13), and either loss to follow-up or a life expectancy of less than one year (n=39) (Fig. 2). Consequently, 369 patients were included in the study, of which 279 (75.6%) were male. The mean age was 65.9±19.5 years, and the average BMI was 23.9±3.0 kg/m2. Clinically, 130 (35.2%) patients had diabetes mellitus, 195 (52.8%) had hypertension, 194 (52.6%) had hyperlipidemia, and 152 (41.1%) had a positive smoking history (Table 1).
Flowchart of patient selection and exclusion.
CEUS, contrast-enhanced ultrasound.
Baseline clinical characteristics in the symptomatic and asymptomatic groups
The symptomatic group (n=200) was significantly older than the asymptomatic group (n=169) (P=0.001). There were no significant differences between the groups in terms of sex, BMI, smoking history, medical history, or drug use history (all P>0.05).
US Parameters between Symptomatic and Asymptomatic Groups
Among the 369 patients with mild stenosis, IPN was detected in both the symptomatic group (97/200, 48.5%) and the asymptomatic group (69/169, 40.9%). The distribution of IPN grades among all included patients was as follows: grade 0, 203 patients (55.0%); grade 1, 33 patients (8.9%); and grade 2, 133 patients (36.1%). A significantly higher prevalence of grade 2 IPN plaques was observed in the symptomatic group (P=0.036). However, no significant differences were found in the proportions of grade 0 and grade 1 IPN plaques between the two groups (P=0.377) (Table 2).
Ultrasonographic parameters in the symptomatic and asymptomatic groups
The EI ratio, which reflects the relative degree of contrast enhancement within the plaque, was significantly higher in plaques with grade 2 IPN (P<0.001) (Supplementary Table 1). Additionally, a significantly higher EI ratio was observed in the symptomatic group (P=0.006) (Table 2). However, there were no significant differences in B-mode US parameters, including plaque length, thickness, echogenicity, and degree of stenosis, between the two groups (P>0.05) (Table 2). Furthermore, plaque thickness showed no correlation with the grade of carotid IPN (P>0.05) (Supplementary Table 1).
Multivariable Analysis of Risk Factors for Recent First-Time Ischemic Stroke
As shown in Table 3, the multivariable logistic regression model was developed using the results from the univariable analysis (Tables 1, 2) and traditional clinical risk factors [17]. After adjusting for traditional risk factors such as smoking history, hypertension, diabetes mellitus, and hyperlipidemia, age (OR, 1.04; 95% CI, 1.02 to 1.06; P=0.002), grade 2 IPN (OR, 1.33; 95% CI, 0.69 to 2.56; P=0.020), and the EI ratio (OR, 1.04; 95% CI, 1.01 to 1.07; P=0.029) were still significantly associated with recent primary ischemic stroke. This suggests that increasing age and grade 2 carotid IPN are significant predictors of recent primary ischemic stroke.
Multivariable analysis of risk factors for recent primary ischemic stroke
Predictive Value of IPN Grade for Future Ischemic Stroke
During the follow-up period, new ischemic anterior circulation lesions ipsilateral to the carotid plaques were identified in 22 patients through head CT/MRI imaging. Of these patients, 16 (72.7%) had a history of stroke. The prevalence of grades 0, 1, and 2 IPN among these patients was 22.7%, 9.1%, and 68.2%, respectively. As shown in Table 4, grade 2 IPN remained an independent predictor of future ischemic stroke in patients with mild carotid stenosis, regardless of their recent primary stroke history, as evidenced by multivariable analysis (hazard raatio, 4.31; 95% CI, 1.40 to 13.20; P=0.011). However, neither a positive stroke history nor the EI ratio was identified as an independent predictor of future ischemic stroke in these patients, according to both univariable and multivariable analyses.
Predictive factors for future ischemic stroke
The Kaplan-Meier analysis, using the log-rank test, demonstrated a significant association between grade 2 IPN and the future risk of ischemic stroke (P=0.004) (Fig. 3).
Kaplan-Meier survival curves based on carotid intraplaque neovascularization (IPN).
Survival analysis figures show the association between cumulative ischemic stroke event-free survival and carotid IPN for the symptomatic group (A), asymptomatic group (B), and all patients (C). The x-axis represents the time of follow-up in months. The y-axis represents the proportion of patients in whom ipsilateral stroke occurred.
Intra- and Inter-observer Consistency Analysis
Satisfactory reproducibility was observed in the analyses, with intra-and inter-observer consistency exhibiting values of 0.88 (95% CI, 2.32 to 2.45; P=0.010) and 0.85 (95% CI, 2.23 to 2.43; P=0.008), respectively.
Discussion
This study demonstrated that CEUS-assessed IPN grading plays a significant role in predicting the risk of ischemic stroke among patients with mild carotid stenosis, both symptomatic and asymptomatic. In the NASCET study, 40% of patients who experienced a stroke had less than 50% carotid stenosis [19]. Additionally, previous research has indicated that up to 20% of patients with embolic TIA or stroke exhibit mild carotid stenosis, with a 3-year recurrence rate for ipsilateral ischemic stroke as high as 8% [13-15]. Recent studies have also revealed that even with mild stenosis (<50%), there is a potentially high stroke risk when carotid plaques show characteristics of vulnerability [3,4,26]. Furthermore, the association between nonstenosing carotid plaques and a significant but unrecognized percentage of cryptogenic strokes has been highlighted [27]. Therefore, it is crucial to identify high-risk plaques in patients with mild carotid stenosis.
In this study, the symptomatic group with a recent primary ischemic stroke ipsilateral to carotid plaque was older, consistent with previous findings. However, no significant differences were observed in the medical histories of hypertension, diabetes mellitus, and hyperlipidemia between the two groups, likely due to medical treatment [3]. Hypertension is recognized as a significant risk factor for ischemic stroke in patients with more than 60% or 70% stenosis following carotid endarterectomy, according to earlier research [28]. However, this study included patients with only mild stenosis, unlike the aforementioned study. Additionally, the prevalence of diabetes mellitus and hyperlipidemia, as well as the history of antidiabetic and statin use, showed no significant differences, further indicating that the medical conditions of the patients with mild carotid stenosis were similar between the two groups in the present study.
IPN has been identified as a crucial indicator for vulnerable plaques, associated with ongoing inflammation and plaque rupture [5]. Additionally, a higher EI ratio was significantly associated with a higher carotid IPN grade and the occurrence of recent primary ischemic stroke. A previous study also found that increased plaque enhancement was significantly correlated with vulnerable features of carotid plaque, such as increased density and larger, more irregularly-shaped microvessels [9]. In study, the symptomatic group exhibited higher IPN grading (grade 2) and EI ratio (reflecting the relative degree of contrast enhancement within the plaque), suggesting that this group may have a higher risk of carotid plaque rupture. However, B-mode US parameters, including plaque length, thickness, and echogenicity, showed no significant differences between the symptomatic and asymptomatic groups. This lack of difference was likely because calcified plaques were excluded and the included plaques were smaller in size due to mild stenosis in this study.
In this study, grade 2 carotid IPN and a higher EI ratio were independently associated with a primary stroke in the symptomatic group, even after adjusting for traditional clinical risk factors. Additionally, even asymptomatic carotid plaques with mild stenosis posed a risk of rupture, potentially leading to future ischemic strokes [29]. Furthermore, grade 2 IPN independently predicted future ischemic stroke in both the symptomatic and asymptomatic groups. This suggests that higher carotid IPN grading could serve as a new biomarker for risk prediction in cases of mild carotid stenosis. However, neither a positive stroke history nor the EI ratio were predictors of future ischemic strokes, possibly due to the limited follow-up duration and the smaller size of carotid plaques in the patients included in this study.
In light of atherosclerosis being a systemic process, the vulnerability of carotid plaque has been recognized as an indicator of plaque instability throughout the entire arterial system [30]. This study may offer new insights into evaluating and predicting the vulnerability of the entire arterial system by visualizing IPN in carotid plaques using CEUS. Future studies should consider the IPN grade of plaques from multiple vascular beds when assessing the risk of stroke occurrence and recurrence. Although MRI demonstrates excellent capabilities in identifying plaque compositions, it measures the fibrous cap thickness and small structures within carotid plaque less reliably in cases of mild stenosis [31]. This study has identified carotid IPN on CEUS as a reproducible and reliable imaging biomarker, serving as a risk factor for recent primary strokes and a predictive factor for future ischemic strokes.
There were several limitations to the present study. First, although the findings indicate that carotid IPN is a promising non-invasive imaging biomarker for stroke risk stratification in mild carotid stenosis, this study was limited by its retrospective case-control design and the short follow-up period of 12 months. To confirm these findings and establish the clinical utility of carotid IPN in stroke prediction, prospective studies involving larger and more diverse populations, longer follow-up periods, and standardized imaging protocols are necessary. Second, the characteristics of the plaques assessed in this study were restricted in terms of size and echogenicity. Plaques thinner than 2 mm and those with severe plaque calcification were excluded due to the insufficient resolution of CEUS images. Future studies should develop new risk evaluation methods for these plaques. Additionally, the occurrence of TIA could not be localized in the anterior cerebral circulation area on the same side as the carotid plaque, which led to the exclusion of TIA patients from the study. However, TIA is an important factor that can significantly impact future stroke risk. More rigorous and comprehensive inclusion and exclusion criteria need to be established in future research. Finally, evaluations of carotid plaques using CEUS, CT, and MRI have been proven to identify the specific compositions of carotid plaques [5,32,33]. Furthermore, employing multiple imaging modalities could improve the predictive performance of risk assessments across different degrees of carotid stenosis.
In patients with mild stenosis, both increasing age and grade 2 carotid IPN were significant predictors of recent primary ischemic stroke. Additionally, carotid IPN can forecast the likelihood of future strokes in cases of mild carotid stenosis, regardless of whether the patient has a history of recent stroke. This study highlighted the potential of carotid IPN as a valuable non-invasive imaging biomarker for stroke risk stratification in mild carotid stenosis.
Notes
Author Contributions
Conceptualization: Wu R, Jia C. Data acquisition: Zhang L, Chen J, Jia C. Data analysis or interpretation: Zhang L, Gu S, Jia C. Drafting of the manuscript: Zhang L, Chen J, Gu S, Jia C. Critical revision of the manuscript: Wu R, Jia C. Approval of the final version of the manuscript: all authors.
No potential conflict of interest relevant to this article was reported.
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grants No. 82130057, 82071931, 82202176), and the National Key Research and Development Projects (2022YFC3602400).
Supplementary Material
Association of EI ratio and plaque thickness with carotid IPN grade (https://doi.org/10.14366/usg.24123).
References
Article information Continued
Notes
Key points
This study evaluated carotid plaque neovascularization in both symptomatic and asymptomatic groups with mild stenosis using contrast-enhanced ultrasonography. This method shows promise as a non-invasive imaging biomarker for stratifying stroke risk in patients with mild carotid stenosis. In individuals with mild stenosis, both increasing age and grade 2 carotid intraplaque neovascularization (IPN) were significant indicators of recent primary ischemic stroke. Furthermore, grade 2 carotid IPN was an independent predictor of future ischemic strokes in both symptomatic and asymptomatic patients.