Carotid stiffness quantified through shear wave elastography and pulse wave velocity by ultrafast ultrasound for assessing cardiovascular risk

Introduction

Atherosclerosis, a chronic arterial inflammation characterized by lipid accumulation, fibrotic remodeling, and progressive calcification, is the pathology underlying coronary heart disease and stroke (1-3). Notably, the development of this insidious condition often begins in early life, even among low-risk younger individuals (4). Vascular stiffness is an early precursor of adverse vascular remodeling and is closely linked to cardiovascular (CV) risk. Clinically, cardiovascular risk factors (CVRFs) are commonly used to assess CV risk because they can independently predict CV and cerebrovascular event risks (5). However, the recent Progression of Early Subclinical Atherosclerosis (PESA) study confirmed that even in populations without conventional CVRFs, approximately 60% of individuals have early vascular dysfunction (6). Thus, more reliable and newer methods are warranted to evaluate early vascular stiffness and stratify CV risk at an early stage.

Carotid intima-media thickness (cIMT), an early marker of vascular injury and an independent predictor of CV risk, is strongly associated with the occurrence of CV events (7). cIMT can be safely, straightforwardly, and noninvasively measured using ultrasound; however, this method is limited by a lack of a standardized measurement scheme, resulting in poor repeatability and accuracy (8). CV risk stratification based on cIMT is most applicable to middle-aged and older individuals (typically ≥60 years) (9). The 2023 European Society of Cardiology (ESC) Guidelines (10) stated that the current evidence is insufficient to indicate whether cIMT has significant clinical value in CV risk assessment in younger populations, reducing its clinical applicability in general.

Real-time shear wave elastography (RT-SWE) is grounded in the principle that tissue hardness is positively correlated with shear wave propagation velocity. It involves inducing shear waves within tissues via acoustic pressure waves and capturing their propagation characteristics using high-speed acquisition sequences, thereby enabling an accurate and quantitative assessment of tissue elasticity (11). In clinical applications, RT-SWE has demonstrated unique diagnostic value across various critical domains, such as the classification of liver fibrosis (12), differentiation between benign and malignant nodules (13), evaluation of plaque stability (14), and quantitative analysis of carotid wall elasticity (15). However, in theory, due to the continuous blood flow within the vascular lumen, the carotid artery is not an ideal elastic body for assessment based on RT-SWE (16). As such, whether RT-SWE can accurately quantify carotid wall elasticity remains unclear. Nevertheless, some studies have confirmed that RT-SWE exhibits good reliability for measuring vascular stiffness using Young’s modulus (E) (17).

Pulse wave velocity (PWV) is a gold-standard modality for assessing arterial stiffness in atherogenesis (18). Ultrafast PWV (ufPWV), a novel PWV technique, involves directly photographing the propagation of a pulse wave with ultrafast speed in real time (>2,000 frames/s). It can assess early CV risk in patients, even in the absence of major CVRFs (19). Notably, the Moens-Korteweg equation reveals that PWV is theoretically and directly correlated with the square root of arterial wall elasticity (20). However, its applicability in the carotid wall has not been validated in vivo.

Therefore, in this study, we utilized RT-SWE and ufPWV to quantify and comprehensively analyze the relationship between the elasticity and stiffness of the carotid wall in vivo. We also evaluated the predictive role of combined RT-SWE and ufPWV models for early CV risk assessment. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0414/rc).

Methods

Compliance with ethical standards

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of the Affiliated Hospital of Nanjing University of Chinese Medicine (No. 2022NL-056-02) and individual consent for this retrospective analysis was waived.

Study participants

We collected data from 159 patients who underwent clinical and laboratory examinations at the Affiliated Hospital of Nanjing University of Chinese Medicine from May 2023 to December 2023. Their cIMT, ufPWV, and RT-SWE parameters were measured on the same day during a physical examination. Data collected, such as age, sex, systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting blood sugar (FBG), triglycerides (TG), total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL), and creatinine (Cr), were recorded. The exclusion criteria were as follows: (I) incomplete clinical, laboratory, or imaging data (n=19); (II) invalid ufPWV or RT-SWE measurements (n=13, detailed in the ufPWV and RT-SWE protocols); (III) CV events occurring in the last 6 months (n=2); (IV) autoimmune diseases, pregnancy status, cancer, or any disease possibly affecting life expectancy (n=15); and (V) any medication for hypertension, dyslipidemia, or diabetes management in the previous 6 months (n=22; Figure 1). Finally, 88 patients (55.3% of all recruited patients) were included in further analysis and divided into two groups (Figure 1). In the control group, we included patients without hypertension, diabetes, dyslipidemia, and a current smoking habit, according to the ESC guidelines (21): (I) untreated SBP <140 mmHg and DBP <90 mmHg; (II) untreated FBG <7.0 mmol/L; and (III) untreated TC <6.2 mmol/L, LDL cholesterol <4.1 mmol/L, and HDL cholesterol ≥1.0 mmol/L. In the CV risk group, we included patients with hypertension, diabetes, dyslipidemia, or a current smoking habit: (I) untreated SBP ≥140 mmHg and/or DBP ≥90 mmHg; (II) untreated FBG ≥7.0 mmol/L or diabetes; and (III) untreated TC ≥6.2 mmol/L, LDL cholesterol ≥4.1 mmol/L, HDL cholesterol <1.0 mmol/L, or their combination.

Figure 1 Flow diagram of study cohort selection. In total, 159 patients were recruited for clinical and laboratory examination. cIMT, ufPWV, and RT-SWE were measured. After implementation of the inclusion and exclusion criteria, the remaining 88 patients were divided into control (n=20) and CV risk (n=68) groups. cIMT, carotid intima-media thickness; CV, cardiovascular; DBP, diastolic blood pressure; FBG, fasting blood sugar; HDL, high-density lipoprotein; LDL, low-density lipoprotein; RT-SWE, real-time shear wave elastography; SBP, systolic blood pressure; TC, total cholesterol; ufPWV, ultrafast pulse wave velocity.

cIMT and ufPWV measurements

The Aixplorer ultrasound system (Supersonic Imagine, Aix-en-Provence, France) and the Aixplorer cIMT automatic measurement system (ACAMS; Supersonic Imagine, Aix-en-Provence, France; Figure 2) equipped with the SL10-2 probe were used to detect and assess cIMT and ufPWV parameters (22). Each patient was placed in a supine position with their head slightly tilted back. The common carotid artery (CCA) was scanned along its long axis, avoiding the jugular vein and any plaques. Both the anterior and posterior walls of the CCA were displayed simultaneously, and a straight, clear segment of the posterior wall of the carotid artery was selected. The sampling frame width of the region of interest (ROI) was set to 10 mm, and the software program automatically drew two white dotted lines to track the intima and media line of the CCA and calculate the average cIMT within the ROI. If the overlap rate was ≥70% (19) (Figure 2A,2D), then the measurement was considered valid. Three repeated measurements were taken to obtain the average value. Finally, the average value of both sides was used as the final cIMT measurement.

Figure 2 Illustration of carotid ultrasound parameter measurement procedure. (A) Measuring cIMT with ACAMS. ACAMS automatically draws two white dotted lines to track the CCA intima and media lines and obtains the average cIMT of the ROI, with 86% fitting. (B) PWV-BS and PWV-ES measured through ufPWV. In the PWV mode, two red lines automatically track the anterior and posterior walls of the CCA. PWV-BS and PWV-ES values are calculated in the ROI of Δ± ≤2.0 m/s separately. (C) E_mean, E_min, E_max, and E_sd measured through RT-SWE. In SWE mode, the freeze button is pressed when the blue-green color of the ROI is stable, and trace mode is used to describe the intima-media in the posterior wall of the CCA. The system automatically calculates four RT-SWE parameters: E_mean, E_min, E_max, and E_sd. (D) A local magnification of the automatic measurement of cIMT. (E) Measurement of partially enlarged drawings of PWV-BS and PWV-ES. (F) Partial enlarged view tracing the intima-media in the posterior wall of the CCA. ACAMS, Aixplorer cIMT automatic measurement system; CCA, common carotid artery; cIMT, carotid intima-media thickness; PWV, pulse wave velocity; PWV-BS, pulse wave velocity at the beginning of systole; PWV-ES, pulse wave velocity at the end of systole; ROI, region of interest; RT-SWE, real-time shear wave elastography; ufPWV, ultrafast pulse wave velocity.

We previously confirmed that ufPWV has good stability and repeatability for assessing carotid wall stiffness (22). All ufPWV acquisitions were carried out in the longitudinal plane of the CCA, avoiding plaque areas. Each patient was asked to hold their breath for 5 s, and a stable image was automatically obtained by pressing the PWV button. After a stable ufPWV image was obtained, the ROI (3.0 cm × 3.0 cm) was moved to cover both the anterior and posterior walls of the carotid artery, with two red lines automatically tracking these CCA walls. The system automatically calculated the PWV at the beginning of systole (PWV-BS) and end of systole (PWV-ES) for the ROI; Δ± ≤2.0 m/s was considered a valid measurement (19) (Figure 2B,2E). Invalid ufPWV measurements were attributed to (I) failure to calculate PWV-BS, PWV-ES, or both; and (II) Δ± >2.0 m/s (Figure S1A,S1B). Three effective ufPWV measurements were averaged, and the average from both sides was used as the final ufPWV value. According to the Moens-Korteweg equation (20): PWV² = Eh / (2Rρ), where E is elastic modulus, h is wall thickness, R is vessel radius, and ρ is blood density.

RT-SWE measurements

After the acquisition of ufPWV parameters, we again used the Aixplorer ultrasound system with the linear array probe (SL10-2) to measure the patients’ RT-SWE parameters (Figure 2C,2F). The patients were asked to hold their breath. We then pressed the SWE key in the longitudinal plane of the CCA and placed the ROI on the straight section of the CCA, covering both the anterior and posterior walls. When the ROI was filled with blue-green color symbolizing elasticity, we pressed the freeze key when the image was stable. When the intima-media in the posterior wall of CCA was thickest, we used trace mode to outline this area, and the system automatically calculated four RT-SWE parameters: E_mean, E_min, E_max, and E_sd. Invalid RT-SWE measurements were attributed to two factors: (I) improper recording of E_mean, E_max, and E_min in the intimal-medial region, resulting in E_sd =0 kPa; and (II) unstable image of the ROI, leading to excessive variance with E_sd >10 kPa (Figure S2A,S2B). Three effective RT-SWE measurements were averaged, and the average from both sides was used as the final arterial wall elasticity value. The Young’s modulus (E) was automatically calculated by the ultrasound system based on the shear wave propagation speed: E = 3ρ × Vs², where ρ is tissue density and Vs is shear wave velocity.

All ultrasound examinations were performed by a single experienced sonographer with 15 years of vascular ultrasound experience. Intra-observer reproducibility, assessed in 20 randomly selected subjects using the intraclass correlation coefficient (ICC), was good to excellent across all parameters (ICCs: 0.676–0.878), with elasticity indices demonstrating particularly high reliability (ICCs: 0.732–0.878). All ICC values were statistically significant (P<0.001 for all; see Table S1 for details). Data analysis was performed by two independent researchers who were blinded to group information.

Statistical analysis

Statistical analysis was performed using SPSS (version 27.0; SPSS, Chicago, IL, USA). Continuous variables are presented as mean ± standard deviation, and categorical variables as counts and proportions. Differences among normally distributed continuous variables, nonnormally distributed continuous variables, and categorical variables were analyzed using independent-sample t, Mann-Whitney U, and Chi-squared tests, respectively. Levene’s test for equality of variances was also performed. If the P value was <0.05, the corrected t-test P value was reported. Between-group comparisons with adjustment for age were performed using analysis of covariance (ANCOVA). Pearson correlation coefficients were used to analyze the relationships among cIMT, ufPWV, RT-SWE, and baseline characteristics (excluding sex, given its binary nature). Intra-observer reproducibility was assessed in 20 randomly selected subjects using the ICC, which demonstrated good to excellent reliability for all measured parameters.

To systematically evaluate the impact of different variables on the prediction of CV risk, we first conducted univariate logistic regression analysis and calculated the odds ratio (OR) and its 95% confidence interval (CI) for each variable. Subsequently, multivariate logistic regression analysis demonstrated that the combined use of PWV-ES and RT-SWE parameters significantly enhance the accuracy of CV risk prediction. Based on this, the regression model for CV risk prediction can be expressed as follows:

z=C+BPES[1]

z=C+BPES+BSE[2]

Eq. [1] is the regression equation calculated for the model using PWV-ES parameters alone, and Eq. [2] is that calculated for the model using combined PWV-ES and RT-SWE parameters. Here, B represents the regression coefficient, C the constant, ES the PWV-ES, and E the RT-SWE parameter. BP and B_S represent the regression coefficients of ufPWV and RT-SWE, respectively. Subsequently, the z values were calculated based on Eqs. [1] and [2], and the categorical variables (control and CV risk groups) were converted into continuous variable z to calculate CV risk probability. A receiver operating characteristic (ROC) curve (based on risk probability z values) was constructed to compare the diagnostic accuracy of univariate ufPWV and RT-SWE. A multivariate model combining RT-SWE and ufPWV was constructed to compare the diagnostic accuracy of single ufPWV parameters with that of the combined multivariate model using both RT-SWE and ufPWV parameters. Their difference in diagnostic power was intuitively demonstrated using ROC curve analysis, along with the area under the curve (AUC) and 95% CI. A P value of <0.05 was considered to indicate statistical significance. Sensitivity analyses were performed using blood pressure-normalized PWV indices, including PWV-BS/DBP and PWV-ES/SBP ratios, with between-group differences evaluated using independent-samples t-tests.

Results

Clinical, laboratory, and ultrasound characteristics

In the CV risk group, age, SBP, and DBP, as well as levels of FBG, TG, TC, LDL, and Cr, were significantly higher than those in the control group (all P<0.05; Table 1). In contrast, no statistically significant differences were observed between the groups with respect to sex distribution (P=0.114) or HDL levels (P=0.053). The cIMT, PWV-ES, E_max, E_sd, and E_mean were significantly higher in the CV risk group than in the control group (all P<0.05; P=0.017 for E_mean; Table 2). However, there were no statistically significant differences in PWV-BS (P=0.125) or E_min (P=0.431) between the two groups (Table 2). To account for the potential confounding effect of blood pressure on PWV, sensitivity analyses were performed using PWV-BS/DBP and PWV-ES/SBP ratios. As shown in Table S2, no significant between-group differences were observed in either ratio (P=0.077 and P=0.381, respectively).

Correlation between ultrasound characteristics and major CVRFs

Figure 3 illustrates pairwise correlations between ultrasound parameters and major CVRFs. Age was significantly correlated with cIMT (r=0.522), PWV-BS (r=0.338), PWV-ES (r=0.314), E_max (r=0.245), and E_sd (r=0.299; all P<0.05; Figure 3A-3D,3G,3H) but not with E_mean (P=0.187) or E_min (P=0.855; Figure 3E,3F and Table S3). Furthermore, significant between-group differences were observed in cIMT, PWV-ES, E_max, E_sd (all P<0.001), and E_mean (P=0.017) but not in PWV-BS (P=0.125) or E_min (P=0.431; Figure 3I and Table 2). Nevertheless, after adjustments for age using ANCOVA, only E_max (P=0.009), E_mean (P=0.047), and PWV-ES (P=0.046) remained significantly associated with CV risk (Figure 3J and Table 3).

Figure 3 Analysis of the correlation between ultrasound characteristics and major cardiovascular risk factors, and comparison of ultrasound features among groups. (A) Correlation between major CVRFs. (B-H) Pearson correlation analysis between age and (B) cIMT (r=0.522, P<0.001), (C) PWV-BS (r=0.338, P=0.001), (D) PWV-ES (r=0.314, P=0.003), (E) E_mean (r=0.142, P=0.187), (F) E_min (r=−0.020, P=0.855), (G) E_max (r=0.245, P=0.022), and (H) E_sd (r=0.299, P=0.005). (I) cIMT, PWV-BS, PWV-ES, E_mean, E_min, E_max, and E_sd compared between control (n=20) and CV risk (n=68) groups. (J) After adjustments for age (ANCOVA), the major CVRFs, cIMT, PWV-BS, PWV-ES, E_mean, E_min, E_max, and E_sd, compared between the control (n=20) and CV risk (n=68) groups. *, P<0.05; ***, P<0.001. ANCOVA, analysis of covariance; cIMT, carotid intima-media thickness; Cr, creatinine; CV, cardiovascular; CVRFs, cardiovascular risk factors; DBP, diastolic blood pressure; FBG, fasting blood sugar; HDL, high-density lipoprotein; LDL, low-density lipoprotein; PWV-BS, pulse wave velocity at the beginning of systole; PWV-ES, pulse wave velocity at the end of systole; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides.

cIMT showed weak positive correlations with SBP (r=0.259) and FBG (r=0.278; all P<0.05). PWV-ES was positively associated with SBP (r=0.357), DBP (r=0.465), and Cr (r=0.218), and negatively correlated with HDL (r=−0.234; all P<0.05). Notably, among the RT-SWE parameters, E_max and E_sd demonstrated broader associations with major CVRFs compared to E_mean and E_min (Figure 4). Specifically, E_max exhibited significant correlations with TC (r=0.251) and HDL (r=0.276; all P<0.05; Figure 4B), while E_sd showed a weak negative correlation with Cr (r=−0.220; P=0.039; Figure 4D). In contrast, E_mean and E_min were only marginally correlated with HDL (r=0.233; P=0.029; Figure 4A) and Cr (r=0.228; P=0.032; Figure 4C), respectively (Table S3).

Figure 4 Independent association between major CVRFs and the elastic modulus parameters (A) E_mean, (B) E_max, (C) E_min, and (D) E_sd. Cr, creatinine; CVRFs, cardiovascular risk factors; DBP, diastolic blood pressure; FBG, fasting blood sugar; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides.

Predictive role of increased ufPWV and RT-SWE parameters for CV risk assessment

PWV-ES, E_mean, E_max, and E_sd were all significantly associated with CV risk, with AUC values ranging from 0.654 to 0.753 (all P<0.05). Among these parameters, PWV-ES demonstrated the highest predictive performance, yielding an AUC of 0.753. In contrast, PWV-BS and E_min did not exhibit significant predictive value for CV risk (all P>0.05; Table 4 and Figure 5A).

Figure 5 ROC curves for CV risk prediction. (A) ROC curve for evaluating CV risk prediction based on ufPWV and RT-SWE parameters. (B) ROC curve for evaluating CV risk prediction based on the combined E_mean, E_max, E_sd, and PWV-ES model. CV, cardiovascular; PWV-BS, pulse wave velocity at the beginning of systole; PWV-ES, pulse wave velocity at the end of systole; ROC, receiver operating characteristic; RT-SWE, real-time shear wave elastography; ufPWV, ultrafast pulse wave velocity.

RT-SWE + ufPWV for CV risk assessment

Table 5 demonstrates the combination of ufPWV and RT-SWE parameter models to further enhance the predictive effect of assessing the risk of early CV. When E_mean, E_max, and E_sd were respectively introduced into the PWV-ES model, the predictive ability of the model improved to varying degrees. The AUC of the PWV-ES + E_mean model was 0.796 (95% CI: 0.700–0.892); the AUC of the PWV-ES + E_max model was 0.839 (95% CI: 0.753–0.925); and the AUC of the PWV-ES + E_sd model was 0.843 (95% CI: 0.756–0.931). The AUC of the final comprehensive model (PWV-ES + E_mean + E_max) reached 0.863 (95% CI: 0.779–0.947), which was significantly improved by approximately 14.6% compared to the single PWV-ES model (0.753; 95% CI: 0.640–0.866) (Table 5 and Figure 5B).

Discussion

In this study, we noted that carotid wall elasticity, quantified through RT-SWE, is strongly associated with carotid stiffness measured through ufPWV in vivo. The present study assessed plaque-free CCA wall biomechanics for CV risk stratification, rather than directly evaluating atherosclerotic plaques. More importantly, we found that the elastic indices have different degrees of dependence on age: after adjusting for age, the differences in E_mean, E_max and PWV-ES remained significant between the control group and the CV risk group, while the difference in E_sd was no longer significant. This may indicate that E_sd represents the regional heterogeneity and local dispersion of arterial elasticity, which tends to gradually increase with physiological aging alone due to mild, diffuse, and heterogeneous changes in the arterial wall. In contrast, E_max, E_mean, and PWV-ES reflect the overall stiffness and elastic behavior of the vessel, which are more closely associated with pathological vascular remodeling driven by CVRFs rather than normal senescence. Therefore, after accounting for age-related vascular changes, E_max, E_mean, and PWV-ES parameters are still effective in discriminating CV risk. Furthermore, the combined model of ufPWV and RT-SWE shows a significant improvement in predictive performance and discriminatory ability compared with using either parameter alone, suggesting that integrating elastic information from different dimensions can provide a more comprehensive assessment of the biomechanical state of blood vessels.

From a biomechanical perspective, RT-SWE and ufPWV both assess vascular elasticity but reflect distinct mechanical properties. While RT-SWE directly measures local wall stiffness (elastic modulus) in the longitudinal direction, ufPWV indirectly reflects bulk arterial stiffness in the circumferential direction. As described in the Methods section, the Moens-Korteweg equation establishes a theoretical relationship between these two approaches. Notably, arterial vessels are anisotropic, and their elasticity can be divided into two main directions: circumferential and longitudinal (23), supporting their combined use for comprehensive vascular biomechanical evaluation. It is generally agreed that ufPWV is an effective tool for evaluating carotid artery stiffness. The results of this study show that PWV-ES is the single indicator with the strongest predictive ability (AUC =0.753), and PWV-ES, E_mean, and E_max all remained significant after adjusting for age. This suggests from a methodological perspective that both circumferential elasticity (represented by PWV-ES) and longitudinal elasticity (represented by E_mean and E_max) reflect pathological vascular changes that are independent of normal aging, thereby providing reliable information for CV risk stratification. Therefore, we attempted to integrate parameters reflecting circumferential elasticity (PWV-ES) and longitudinal elasticity (E_mean, E_max) to construct a combined model. The diagnostic performance of this model (AUC =0.863) was significantly better than that of the single PWV-ES indicator (AUC =0.753), confirming the synergistic value of multi-dimensional mechanical assessment. However, overly inclusive parameter combinations increase the risk of chance findings and reduced replicability. Furthermore, given the high theoretical correlation between E and PWV, as well as correlations among RT-SWE parameters, the incremental information provided by the combined model may be partially overlapping. Thus, the optimal parameter panel and independent contribution of each index still require validation in larger cohorts.

RT-SWE has significant technical advantages over traditional strain elastography. It can aid in quantitatively assessing the elastic characteristics of the carotid artery wall and carotid plaques. RT-SWE induces shear wave propagation through acoustic radiation pulses (24,25), directly measures tissue elastic modulus, and overcomes the limitations of traditional techniques, which require manual compression by the operator and qualitative color coding (26). This technique offers high operator independence and good repeatability, enabling the precise quantification of the longitudinal elasticity of the arterial wall and providing an objective indicator for vascular stiffness assessment. Although absolute values of elastic moduli vary across studies due to differences in equipment, protocols, and patient populations, consistent associations between carotid elastic parameters and CVRFs have been widely reported. In the case of Behcet’s disease and CVRFs other than hypertension, RT-SWE detection shows that the carotid wall stiffness values are significantly higher than those in the control group (27). This finding is highly consistent with the observations of this study. E_max (r=0.245; P=0.022) and E_sd (r=0.299; P=0.005) were positively correlated with age and significantly higher in the CV risk group than in the control group (all P<0.05). In contrast, although E_mean did not show age dependence, it could still effectively distinguish between the CV risk group and the control group (P=0.017). E_min, on the other hand, was not related to age and failed to show a statistically significant difference between the two groups (P>0.05). After age adjustment using ANCOVA, both E_mean and E_max remained significantly associated with CV risk and showed comparable reliability. E_sd did not retain significance after age adjustment and largely reflects measurement variability rather than genuine arterial stiffness. This indicates that not all elasticity indices are suitable for CV risk assessment, and suggests that among the numerous elasticity parameters, E_mean and E_max may be a more stable and specific marker for CV risk. In addition, RT-SWE has unique value in evaluating plaque stability. E is significantly lower for asymptomatic progressive and symptomatic plaques than for asymptomatic stable plaques (28). Thus, this technique has crucial clinical application value in CV risk stratification and plaque stability assessment.

As a noninvasive technique for assessing arterial stiffness, ufPWV has recently demonstrated significant value in the early diagnosis of vascular dysfunction and carotid artery diseases. By quantifying parameters such as PWV-BS and PWV-ES, it sensitively reflects the arterial wall’s elastic function. The current results confirmed that both PWV-BS (r=0.338) and PWV-ES (r=0.314) are significantly positively correlated with age (P<0.05)—consistent with previous reports: PWV-BS and PWV-ES both demonstrate a significant upward trend with age (22). This finding confirmed that ufPWV provides reliable biomechanical indicators for the early identification of age-related vascular stiffness. Our previous studies demonstrated that PWV-ES was more pathophysiologically relevant and reliable than PWV-BS, as it effectively detected progressive arterial stiffening with elevated blood pressure and was independently associated with SBP and DBP, while PWV-BS failed to show significant intergroup differences (29). PWV-ES showed better reliability and stronger associations with CV risk compared with PWV-BS, which may be attributed to more stable measurement during end-systole (30). Notably, PWV-ES was significantly higher in the CV risk group than in the control group (P<0.001), corroborating the conclusion of Zhu et al. (19) and further verifying the sensitivity of ufPWV. Even in the absence of traditional CVRFs, early vascular dysfunction can be detected through PWV-ES. ufPWV can also be used to identify decreased carotid artery elasticity in patients with prehypertension (29) and subclinical vascular dysfunction in young patients with type 1 diabetes (31), demonstrating its broad clinical application prospects. However, its strong correlation with age and the change in predictive power after adjusting for age suggest that the influence of age should be taken into account in clinical application. In addition to age, blood pressure is also a critical confounding factor. Sensitivity analyses based on blood pressure-normalized PWV ratios further demonstrated that intergroup blood pressure differences were the main cause of the initial intergroup disparity in PWV-ES.

As the gold standard for CV risk stratification, the degree of cIMT thickening is significantly positively correlated with disease risk (32). From a pathological mechanism perspective, cIMT reflects vascular morphological changes, whereas RT-SWE and ufPWV provide biomechanical assessment evidence from the longitudinal and circumferential elasticity, respectively. We found that the PWV-ES, E_max, E_sd (all P<0.001), and E_mean (P=0.017) in the CV risk group were all higher than those in the control group, which verified the value of the multi-parameter combined assessment proposed by Yu et al. (33). Consistent with previous findings in chronic kidney disease patients (34), changes in cIMT may lag behind PWV-ES measured through ufPWV, with approximately one-third of the CV risk population having abnormal PWV-ES but normal cIMT (22). However, in the present study, both cIMT and PWV-ES in the CV risk group were significantly higher (all P<0.001), possibly because of lesion formation accelerated by high-risk factors such as diabetes. Moreover, Guo et al. (35) demonstrated that RT-SWE and ufPWV can assess carotid artery elasticity in patients with multiple sclerosis. In the current study, we found that their combined application in patients with CV risk was more effective. Therefore, cIMT, RT-SWE, and ufPWV provide structural, longitudinal, and circumferential elasticity parameters, respectively, and their combined use can enable multidimensional and precise CV risk stratification.

Limitations

Our research has several limitations. First, CV risk development is associated with many risk factors not considered in this study; for instance, we did not investigate patients who are overweight. Second, we excluded patients who had taken medications for hypertension, dyslipidemia, or diabetes, as well as those who had experienced CV events within the previous 6 months. This led to a lack of individuals with subclinical or advanced CV disease; therefore, our RT-SWE results may not have been comprehensive enough. Third, our RT-SWE data were obtained only from one cohort, and a single probe (SL10-2) was used. The general applicability of carotid CV risk in different cohorts and with different probes warrants verification. Fourth, the control group was relatively small (n=20) due to the strict exclusion of individuals with any CVRFs. This limitation may have implications for the generalizability of our findings. Therefore, some of our observations warrant further confirmation in larger, multicenter cohorts to enhance the statistical robustness and generalizability of the results. Fifth, the incremental information provided by RT-SWE parameters in the joint model may have some overlap. Therefore, their independent contributions to the model performance still need to be further verified through studies with larger samples. Finally, because of the anisotropy of blood vessels, combining longitudinal and transverse RT-SWE can improve the accuracy of the results. Further technical and clinical studies are needed to optimize and standardize the methodology, and to enhance its pathophysiological relevance, diagnostic performance and predictive value through standardized protocols and integration with other biomarkers and imaging modalities.

Conclusions

Carotid elasticity quantified through RT-SWE is strongly associated with carotid stiffness quantified through ufPWV, and measurements from both modalities indicate elevated levels of major CVRFs. Combined models incorporating RT-SWE and ufPWV parameters may serve as a promising noninvasive tool for stratifying CV risk. Notably, E_mean and E_max remains independently associated with CV risk after adjustment for age, supporting its incremental value beyond vascular aging.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0414/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0414/dss

Funding: This study was supported by the Research Project on Health Care for Cadres in Jiangsu Province (Nos. BJ23007 and BJ25010) and the Research Project of Jiangsu Province Hospital of Chinese Medicine (No. Y2021CX27).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2026-1-0414/coif). All authors report that this study was supported by the Research Project on Health Care for Cadres in Jiangsu Province (Nos. BJ23007 and BJ25010) and the Research Project of Jiangsu Province Hospital of Chinese Medicine (No. Y2021CX27). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of the Affiliated Hospital of Nanjing University of Chinese Medicine (No. 2022NL-056-02) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Wu J, He S, Song Z, Chen S, Lin X, Sun H, Zhou P, Peng Q, Du S, Zheng S, Liu X. Macrophage polarization states in atherosclerosis. Front Immunol 2023;14:1185587. [Crossref] [PubMed]
2. Jebari-Benslaiman S, Galicia-García U, Larrea-Sebal A, Olaetxea JR, Alloza I, Vandenbroeck K, Benito-Vicente A, Martín C. Pathophysiology of Atherosclerosis. Int J Mol Sci 2022;23:3346. [Crossref] [PubMed]
3. Ajoolabady A, Pratico D, Lin L, Mantzoros CS, Bahijri S, Tuomilehto J, Ren J. Inflammation in atherosclerosis: pathophysiology and mechanisms. Cell Death Dis 2024;15:817. [Crossref] [PubMed]
4. Fernández-Friera L, Fuster V, López-Melgar B, Oliva B, García-Ruiz JM, Mendiguren J, Bueno H, Pocock S, Ibáñez B, Fernández-Ortiz A, Sanz J. Normal LDL-Cholesterol Levels Are Associated With Subclinical Atherosclerosis in the Absence of Risk Factors. J Am Coll Cardiol 2017;70:2979-91. [Crossref] [PubMed]
5. Zhu Z, Chen L, Liu W, Wu Y, Zou C, Zhang X, He S, Wang Y, Shen B, Ma X, Gao H, Luan Y, Huang H. Carotid stiffening predicts cardiovascular risk stratification in mid-life: non-invasive quantification with ultrafast ultrasound imaging. Ultrasonography 2022;41:462-72. [Crossref] [PubMed]
6. Zhu Z, Chen L, Shen B, Liu W, Zou C, Wang Y, Ma X, Gao H, Xu D, Wu Y, Huang H. Predicting cardiovascular risk stratification in apparently healthy population by using noninvasive ultrafast ultrasound imaging. Acad Radiol 2024;31:3944-55. [Crossref] [PubMed]
7. Robert C, Tan WY, Ling LH, Hilal S. Association of arterial structure and function with incident cardiovascular diseases and cognitive decline. Alzheimers Dement (Amst) 2025;17:e70069. [Crossref] [PubMed]
8. Li X, Cokkinos D, Gadani S, Rafailidis V, Aschwanden M, Levitin A, Szaflarski D, Kirksey L, Staub D, Partovi S. Advanced ultrasound techniques in arterial diseases. Int J Cardiovasc Imaging 2022;38:1711-21. [Crossref] [PubMed]
9. Nasir K, Ziffer JA, Cainzos-Achirica M, Ali SS, Feldman DI, Arias L, Saxena A, Feldman T, Cury R, Budoff MJ, Fialkow J. The Miami Heart Study (MiHeart) at Baptist Health South Florida, A prospective study of subclinical cardiovascular disease and emerging cardiovascular risk factors in asymptomatic young and middle-aged adults: The Miami Heart Study: Rationale and Design. Am J Prev Cardiol 2021;7:100202. [Crossref] [PubMed]
10. Marx N, Federici M, Schütt K, Müller-Wieland D, Ajjan RA, Antunes MJ, et al. 2023 ESC Guidelines for the management of cardiovascular disease in patients with diabetes. Eur Heart J 2023;44:4043-140. Erratum in: Eur Heart J 2023;44:5060. [Crossref]
11. Piscaglia F, Salvatore V, Mulazzani L, Cantisani V, Schiavone C. Ultrasound Shear Wave Elastography for Liver Disease. A Critical Appraisal of the Many Actors on the Stage. Ultraschall Med 2016;37:1-5.
12. Jiang X, Li L, Xue HY. The impact of body position and exercise on the measurement of liver Young's modulus by real-time shear wave elastography. Technol Health Care 2022;30:445-54. [Crossref] [PubMed]
13. Mena G, Montalvo A, Ubidia M, Olmedo J, Guerrero A, Leon-Rojas JE. Elastography of the thyroid nodule, cut-off points between benign and malignant lesions for strain, 2D shear wave real time and point shear wave: a correlation with pathology, ACR TIRADS and Alpha Score. Front Endocrinol (Lausanne) 2023;14:1182557. [Crossref] [PubMed]
14. Cheng L, Zheng S, Zhang J, Wang F, Liu X, Zhang L, Chen Z, Cheng Y, Zhang W, Li Y, He W. Multimodal ultrasound-based carotid plaque risk biomarkers predict poor functional outcome in patients with ischemic stroke or TIA. BMC Neurol 2023;23:13. [Crossref] [PubMed]
15. Kavvadas D, Rafailidis V, Partovi S, Tegos T, Kallia Z, Savvoulidis P, Papamitsou T, Prassopoulos P. Shear Wave Elastography for Carotid Artery Stiffness: Ready for Prime Time? Diagnostics (Basel) 2025;15:303. [Crossref] [PubMed]
16. Zhang XY, Zhang D, Han LZ, Pan YS, Wei Q, Lv WZ, Dietrich CF, Wang ZY, Cui XW. Predicting Malignancy of Thyroid Micronodules: Radiomics Analysis Based on Two Types of Ultrasound Elastography Images. Acad Radiol 2023;30:2156-68. [Crossref] [PubMed]
17. Gülşen F, Samanci C, Memis Durmaz ES, Durmaz E, Tel C, Gencturk M, Ağırman A. Brachial Artery Wall Stiffness Assessment by Shear Wave Elastography: A Promising New Diagnostic Tool for Endothelial Dysfunction Detection. J Ultrasound Med 2018;37:1977-83. [Crossref] [PubMed]
18. Boutouyrie P, Chowienczyk P, Humphrey JD, Mitchell GF. Arterial Stiffness and Cardiovascular Risk in Hypertension. Circ Res 2021;128:864-86. [Crossref] [PubMed]
19. Zhu ZQ, Chen LS, Jiang XZ, Wu YY, Zou C, Luan Y, Gao H, Dai P, Ma XH, Wu LL, Sun HJ, Wang YP, Zou F, Liu FM, Huang H. Absent atherosclerotic risk factors are associated with carotid stiffening quantified with ultrafast ultrasound imaging. Eur Radiol 2021;31:3195-206. [Crossref] [PubMed]
20. Whitlock MC, Hundley WG. Noninvasive Imaging of Flow and Vascular Function in Disease of the Aorta. JACC Cardiovasc Imaging 2015;8:1094-106. [Crossref] [PubMed]
21. Mach F, Baigent C, Catapano AL, Koskinas KC, Casula M, Badimon L, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J 2020;41:111-88. [Crossref] [PubMed]
22. Zhu ZQ, Chen LS, Wang H, Liu FM, Luan Y, Wu LL, Liu N, Wang P, Huang H. Carotid stiffness and atherosclerotic risk: non-invasive quantification with ultrafast ultrasound pulse wave velocity. Eur Radiol 2019;29:1507-17. [Crossref] [PubMed]
23. Li Z, Du L, Wang F, Luo X. Assessment of the arterial stiffness in patients with acute ischemic stroke using longitudinal elasticity modulus measurements obtained with Shear Wave Elastography. Med Ultrason 2016;18:182-9. [Crossref] [PubMed]
24. Couade M, Pernot M, Prada C, Messas E, Emmerich J, Bruneval P, Criton A, Fink M, Tanter M. Quantitative assessment of arterial wall biomechanical properties using shear wave imaging. Ultrasound Med Biol 2010;36:1662-76. [Crossref] [PubMed]
25. Maksuti E, Widman E, Larsson D, Urban MW, Larsson M, Bjällmark A. Arterial Stiffness Estimation by Shear Wave Elastography: Validation in Phantoms with Mechanical Testing. Ultrasound Med Biol 2016;42:308-21. [Crossref] [PubMed]
26. Cosgrove DO, Berg WA, Doré CJ, Skyba DM, Henry JP, Gay J, Cohen-Bacrie C. Shear wave elastography for breast masses is highly reproducible. Eur Radiol 2012;22:1023-32. [Crossref] [PubMed]
27. Pruijssen JT, de Korte CL, Voss I, Hansen HHG. Vascular Shear Wave Elastography in Atherosclerotic Arteries: A Systematic Review. Ultrasound Med Biol 2020;46:2145-63. [Crossref] [PubMed]
28. Školoudík D, Kešnerová P, Vomáčka J, Hrbáč T, Netuka D, Forostyak S, Roubec M, Herzig R, Belšan T. Shear-Wave Elastography Enables Identification of Unstable Carotid Plaque. Ultrasound Med Biol 2021;47:1704-10. [Crossref] [PubMed]
29. Ma X, Zhu Z, Wang Y, Shen B, Jiang X, Liu W, Wu Y, Zou C, Luan Y, Gao H, Huang H. Quantifying carotid stiffness in a pre-hypertensive population with ultrafast ultrasound imaging. Ultrasonography 2023;42:89-99. [Crossref] [PubMed]
30. Goudot G, Mirault T, Khider L, Pedreira O, Cheng C, Porée J, Gruest M, Jeunemaître X, Pernot M, Messas E. Carotid Stiffness Assessment With Ultrafast Ultrasound Imaging in Case of Bicuspid Aortic Valve. Front Physiol 2019;10:1330. [Crossref] [PubMed]
31. De Biasio MJ, Furman M, Clarke A, Hui W, Elia Y, Baranger J, Villemain O, Mertens L, Mahmud FH. Abnormal vascular thickness and stiffness in young adults with type 1 diabetes: new insights from cutting-edge ultrasound modalities. Cardiovasc Diabetol 2024;23:178. [Crossref] [PubMed]
32. Ramírez-Terán AL, Vega-Memije ME, Torres-Tamayo M, Martínez-Alvarado MR. Carotid intima-media thickness in patients with psoriasis with and without metabolic syndrome. Arch Cardiol Mex 2022;92:305-11. [Crossref] [PubMed]
33. Yu P, Wang Y, Liu T, Yang L, Li D, Zhu J. Assessment of common carotid artery wall stiffness using shear wave elastography: a promising technique for evaluating active and inactive Behçet's disease. Quant Imaging Med Surg 2025;15:2175-82. [Crossref] [PubMed]
34. Huang H, Zhu Z, Wang H, Ma X, Liu W, Wu Y, Zou C, Wang Y, Shen B, Ge W, Gao H, Luan Y, Jiang X. Quantifying carotid stiffness in chronic kidney disease using ultrafast ultrasound imaging. Quant Imaging Med Surg 2024;14:75-85. [Crossref] [PubMed]
35. Guo S, Gu C, Sun L, Qi Z, Wang B. Evaluation of Carotid Stiffness in Metabolic Syndrome by Real-Time Shear Wave Elasticity Imaging and Ultrafast Pulse Wave Velocity. Ultrasound Med Biol 2024;50:1280-6. [Crossref] [PubMed]