Echogenicity of carotid plaques as a predictor of regression following lipid-lowering therapy

Abstract

Objective

See More

Atherosclerotic plaque regression under lipid-lowering therapy shows considerable individual variation, and the factors influencing this variability remain incompletely understood. This study aimed to investigate the relationship between carotid plaque echogenicity and plaque regression in patients receiving lipid-lowering therapy, and to identify ultrasound characteristics that might predict plaque regression.

Methods

A total of 838 patients with carotid plaques receiving lipid-lowering therapy were enrolled between July 2020 and May 2024 and followed up for 12 months. Carotid ultrasound was performed at baseline and follow-up to evaluate plaque characteristics. Plaque regression was defined as meeting any of the following criteria: (1) reduction in plaque area ≥ 5%, (2) decrease in plaque thickness ≥ 0.4 mm, or (3) reduction in plaque number, as assessed by vascular ultrasound imaging. Plaque echogenicity was classified into three types: hypoechoic, hyperechoic, and mixed echogenicity. Cox proportional hazards regression analysis was performed to assess the association between plaque echogenicity and plaque regression, adjusting for potential confounding factors.

Results

Hypoechoic plaques showed higher rates of regression (72.8%) compared to hyperechoic (37.7%) and mixed echogenicity plaques (50.0%) (p < 0.001). After adjusting for confounding variables, hypoechoic plaques exhibited greater odds of regression compared to hyperechoic plaques (adjusted HR = 4.52, 95% CI: 3.18–6.43, p < 0.001). Additionally, the median percentage reduction in plaque size was more pronounced in hypoechoic plaques, (15.2%, IQR: 7.7–22.3%) compared with other echogenicities (p < 0.001).

Conclusion

Carotid plaque echogenicity is strongly associated with the likelihood plaque regression, with hypoechoic plaques exhibiting higher regression rates and greater reductions in plaque size. These findings may help guide personalized treatment strategies and improve risk assessment.

Introduction

Atherosclerotic cardiovascular disease (ASCVD) continues to be a major contributor to morbidity and mortality worldwide, with approximately 11.39 million individuals affected in China alone [1]. Carotid plaque progression is associated with a significantly increased risk of stroke, contributing substantially to healthcare costs and reduced quality of life [2]. Recent epidemiological evidence indicates that carotid plaque presence elevates cardiovascular event risk by 2–4 times, underscoring the urgent need for optimized plaque management approaches [3].

The relationship between plaque characteristics and clinical outcomes has emerged as a crucial area of investigation. Ultrasound assessment of carotid plaques has revealed that different plaque morphologies, particularly their echogenicity patterns, may reflect distinct pathological compositions and stability states [4]. Recent studies have shown that plaque echogenicity correlates with histological features and may predict cardiovascular events [5]. However, the predictive value of plaque characteristics for treatment response remains inadequately explored, particularly in the context of lipid-lowering therapy.

Lipid-lowering therapy, especially intensive statin treatment, has demonstrated the potential to induce plaque regression, which is associated with reduced cardiovascular events [6]. While several studies have documented the overall effectiveness of lipid-lowering interventions, the response varies significantly among patients [7]. Current evidence suggests that plaque composition may influence the likelihood of regression, but existing studies are limited by small sample sizes, short follow-up periods, and inconsistent methodologies for plaque characterization [8]. Furthermore, the relationship between specific plaque characteristics, such as echogenicity, and their propensity for regression under lipid-lowering therapy remains poorly understood.

Based on previous studies showing associations between plaque echogenicity and cardiovascular outcomes, we hypothesized that plaque echogenicity could serve as a predictor of regression potential under lipid-lowering therapy [9]. This study aimed to investigate the relationship between carotid ultrasound plaque characteristics, particularly echogenicity, and plaque regression through a comprehensive retrospective analysis. Understanding these associations could provide valuable insights for developing personalized treatment strategies and improving risk assessment in patients with carotid atherosclerosis [10].

Patients and methods

Study design and participants

A retrospective case-control study was conducted, utilizing data from 4,987 patients who underwent carotid ultrasound examinations at Shanghai East Hospital between July 2020 and May 2024.

The inclusion criteria were as follows:

(1) individuals aged 18 years or older with ultrasound-confirmed carotid plaques; (2) completion of a follow-up carotid ultrasound at 12 ± 1 months following the initial examination; (3) availability of complete ultrasound reports; and (4) consistent lipid-lowering therapy maintained between the assessments.

The exclusion criteria were as follows:

(1) carotid stenosis greater than 70%; and (2) any carotid intervention or surgery performed between assessments.

This study received approval from the Ethics Committee of Shanghai East Hospital (approval number: 2024YS-227), and all participants provided written informed consent.

Methods

Ultrasound examination

All patients were examined using a Philips EPIQ 7 ultrasound system with a 7–12 MHz linear array transducer. During the examination, patients were placed in a supine position with their head slightly rotated 30°to the contralateral side. The imaging depth was set between 1.9 and 3.0 cm. The examination included comprehensive documentation of plaques in the bilateral common carotid arteries, internal carotid arteries, and external carotid arteries, recording plaque location, size, and echogenic characteristics. These standardized protocols were consistently followed for all ultrasound examinations to ensure reliable and reproducible assessments.

Plaque definition and classification

According to the European Society of Cardiology guidelines, carotid plaques were defined as:

(1) A focal structure encroaching into the arterial lumen with a thickness ≥ 1.5 mm.

(2) A focal structure extending ≥ 0.5 mm into the arterial lumen.

(3) A structure showing a thickness > 50% compared to the surrounding intima-media thickness (IMT).

Carotid plaques were classified into three categories based on their ultrasound echogenicity characteristics (Fig. 1) [3, 11]: hypoechoic plaques appeared predominantly black with echogenicity less than that of the surrounding muscle tissue (Fig. 1A), hyperechoic plaques appeared predominantly bright with echogenicity greater than or similar to the surrounding adventitia (Fig. 1B), and mixed echogenicity plaques contained both hypoechoic and hyperechoic components with neither component exceeding 80% of the total plaque area (Fig. 1C) [12]. This classification was based on the characteristics of plaques observed in clinical practice, referencing studies by Kakkos et al. [3] and Reiter et al. [13], and has demonstrated excellent reproducibility and clinical relevance in previous large-scale studies [9, 14, 15].

Fig. 1

Three types of plaques. A: Representative ultrasound image of a hypoechoic carotid plaque. The image shows a typical hypoechoic plaque with low echogenicity (darker appearance) compared to the surrounding tissues, demonstrating the characteristic features associated with higher regression potential (72.8% regression rate in this study). The hypoechoic nature suggests a lipid-rich composition that may be more responsive to lipid-lowering therapy. Note the relatively homogeneous internal structure and well-defined boundaries. Scale bar = 4.0 cm

Fig. 1

B: Representative ultrasound image of a hyperechoic carotid plaque. The image demonstrates a typical hyperechoic plaque characterized by high echogenicity (brighter appearance) compared to surrounding tissues, with well-defined boundaries and relatively homogeneous internal structure. This type of plaque typically shows lower regression rates (37.7%) compared to hypoechoic plaques (72.8%) following lipid-lowering therapy. Scale bar represents 5 mm

Fig. 1

C: Representative ultrasound image of a mixed echogenicity carotid plaque. The image demonstrates a plaque with heterogeneous composition, containing both hypoechoic (darker) and hyperechoic (brighter) regions within the same plaque. This type of plaque showed intermediate regression rates (50.0%) compared to hypoechoic (72.8%) and hyperechoic (37.7%) plaques following lipid-lowering therapy. The heterogeneous appearance suggests a complex plaque composition with both lipid-rich and fibrotic components. Scale bar = 3.0 cm

Based on the distribution of our data and previous studies, the degree of plaque regression was classified into four levels: minimal (< 5%), mild (5–10%), moderate (10–20%), and significant (> 20%) [3, 16, 17].

Plaques were categorized as homogeneous or heterogeneous according to the classification system by Geroulakos et al. [18] Homogeneous plaques are characterized by a consistent echo intensity throughout, while heterogeneous plaques contain regions with varying echo intensities within the plaque.

Plaque measurement

Plaque dimensions were measured according to standardized protocols (Fig. 2). The maximum plaque thickness was measured as the greatest vertical distance between the intima-lumen interface and the media-adventitia interface (Fig. 2A), while plaque length was measured as the longitudinal distance where the plaque was clearly visible and distinguishable from the vessel wall (Fig. 2B). These measurements were performed using automated software with manual adjustment when necessary to ensure accuracy, following validated methodology.

Fig. 2

Representative ultrasound images demonstrating plaque regression in a hypoechoic carotid plaque following lipid-lowering therapy.A: Baseline ultrasound image showing a hypoechoic plaque with a thickness of 0.638 cm (March 2024)

Fig. 2

B: Follow-up ultrasound image after 8 months of lipid-lowering therapy showing significant regression with plaque thickness reduced to 0.106 cm (November 2024), representing an 83.4% reduction. This case exemplifies the high regression potential of hypoechoic plaques, which showed a significantly higher regression rate (72.8%) compared to other plaque types in our study. Scale bars = 3.0 cm

Stenosis assessment

Carotid artery stenosis was evaluated using the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria:

Degree of stenosis (%) = [1 - (minimum lumen diameter at the stenosis) / (diameter of normal distal internal carotid artery)] ×100%.

Plaque regression criteria

Plaque regression was categorized into four levels based on the reduction in plaque size:

• 1) Minimal regression: <5% reduction.

• 2) Mild regression: (5–10)% reduction.

• 3) Moderate regression: (10–20)% reduction.

• 4) Significant regression: >20% reduction.

Plaque regression was defined as meeting any of the following criteria:

• (1) Reduction in plaque area ≥ 5%.

• (2) Decrease in plaque thickness ≥ 0.4 mm.

• (3) Reduction in plaque number.

Plaque area measurement

Plaque area was measured using the following protocol:

• (1) Manual tracing of plaque boundaries in the longitudinal view using the ultrasound system’s built-in software.

• (2) Automated area calculation following manual tracing.

• (3) Three measurements were taken for each plaque, and the average value was recorded.

• (4) Both maximum individual plaque area and total plaque area were documented for patients with multiple plaques.

Plaque enumeration

Plaque counting followed these criteria:

• 1) Plaques separated by ≥ 5 mm within the same arterial segment were counted as distinct plaques.

• 2) Total plaque number was calculated as the sum of plaques in bilateral carotid arteries.

Plaque Burden Calculation.

Total plaque burden was calculated as:

$$\text { Plaque burden }(\%)=\left(\sum \text { plaque areas } / \text { reference vessel area }\right) \times 100 \%$$

Quality control measures

To ensure measurement reliability:

• (1) All measurements were independently performed by two experienced ultrasonographers.

• (2) Inter-observer agreement was assessed using intraclass correlation coefficient (ICC).

• (3) An ICC value > 0.80 was considered to indicate good reliability.

Statin therapy

Statin therapy was categorized according to the 2018 ACC/AHA guidelines:

• - High-intensity: atorvastatin ≥ 40 mg or rosuvastatin ≥ 20 mg.

• - Moderate-intensity: atorvastatin (10–20)mg or rosuvastatin (5–10)mg.

• - Low-intensity: simvastatin < 20 mg or equivalent.

Clinical data collection

Demographic information (including sex and age), clinical characteristics (such as the presence of comorbidities like hypertension, diabetes, and smoking history), treatment-related data (medication regimen), laboratory test outcomes (including lipid profile and complete blood count), and ultrasound examination findings (plaque echogenicity type, quantity, and size) were collected for analysis.

Follow-up

Follow-up Period:

Duration: 1 year (± 1 month) following the initial examination.

Time frame: Between July 2020 and May 2024.

Follow-up Assessment:

Follow-up was conducted through carotid ultrasound examinations.

The same ultrasound system (Philips EPIQ 7 ultrasound system) was used for assessments. Certified ultrasonographers performed the examinations.

Statistical analysis

Descriptive statistics were utilized to summarize characteristics of the patient cohort. Chi-square tests were conducted to compare plaque regression rates across various echogenicity types. The Mann-Whitney U test was used to assess the degree of regression among different plaque types. Cox proportional hazards regression analysis was carried out to explore the relationship between plaque echogenicity and regression, controlling for potential confounding factors, including age, sex, comorbidities, and lipid levels. Hazard ratios (HR) and 95% confidence intervals (CI) were computed for each echogenicity type to quantify the likelihood of plaque regression. All statistical analyses were performed using SPSS version 26.0 software, with a significance level set at p < 0.05.

Results

Patient selection and flow

A total of 4,987 patients were initially screened for this study. After applying the exclusion criteria, 838 patients were included in the final analysis. (Fig. 3)

Fig. 3

The study flowchart

Baseline characteristics

The mean age of the study population was 67.8 ± 9.9 years, with 61.5% being male. Baseline characteristics were compared among the three plaque echogenicity groups (Table 1). The study population included 391 patients with hypoechoic plaques, 225 with hyperechoic plaques, and 139 with mixed echogenicity plaques. There were no significant differences in age, gender, body mass index, or traditional cardiovascular risk factors among the three groups (all p > 0.05).

Table 1 Baseline characteristics of study population according to plaque echogenicity

Group comparisons

Plaque regression rates differed significantly among the three echogenicity groups. Hypoechoic plaques showed the highest regression rate (72.8%), followed by mixed echogenicity plaques (50.0%), and hyperechoic plaques (37.7%) (p < 0.001). The median percentage reduction in plaque size was also more pronounced in the hypoechoic group (15.2%, IQR: 7.7–22.3%) compared to hyperechoic (8.3%, IQR: 3.2–13.5%) and mixed echogenicity groups (11.4%, IQR: 5.8–17.2%) (p < 0.001) (Table 2).

Table 2 Comparison of plaque regression parameters among different echogenicity groups

Correlation analysis

Spearman correlation analysis revealed significant associations between plaque echogenicity and regression parameters. Plaque echogenicity showed a strong negative correlation with the percentage of plaque regression (r=-0.68, p < 0.001). Additionally, significant correlations were observed between regression and baseline LDL-C levels (r = 0.45, p < 0.001), statin intensity (r = 0.52, p = < 0.001), and initial plaque thickness (r = 0.38, p = 0.002).

Cox regression analysis

In univariate analysis, hypoechoic plaque characteristics (HR = 4.52, 95% CI: 3.18–6.43, p < 0.001), baseline LDL-C levels (HR = 1.85, 95% CI: 1.42–2.36, p = 0.003), and high-intensity statin therapy (HR = 4.65, 95% CI: 3.52–6.15, p < 0.001) were significantly associated with plaque regression.

After adjusting for potential confounders including age, sex, cardiovascular risk factors, and medication use, multivariate Cox regression analysis confirmed that hypoechoic plaque remained an independent predictor of plaque regression (adjusted HR = 4.52, 95% CI: 3.18–6.43, p < 0.001) (Table 3).

Table 3 Univariate and multivariate Cox regression analysis for plaque regression

Subgroup analyses showed consistent results across different age groups, gender, and presence of cardiovascular risk factors (Fig. 4).

Fig. 4

The forest plot of subgroup analyses

Discussion

Our study demonstrates that plaque echogenicity strongly predicts regression potential under lipid-lowering therapy, with hypoechoic plaques showing significantly higher regression rates compared to hyperechoic plaques (72.8% vs. 37.7%). This marked difference in regression potential can be explained by several underlying mechanisms [19]. Hypoechoic plaques typically represent lipid-rich, inflammatory lesions with higher macrophage content, making them more responsive to lipid-lowering therapy through enhanced lipid extraction, reduced inflammatory activity, and increased plaque stabilization [20]. At the cellular and molecular level, statins’ pleiotropic effects appear more pronounced in hypoechoic plaques, manifesting through decreased matrix metalloproteinase activity, enhanced cholesterol efflux from foam cells, and reduced inflammatory cytokine production [21].

The differential regression rates between hypoechoic and hyperechoic plaques can be explained by several biological mechanisms. Hypoechoic plaques typically contain higher lipid content and inflammatory cell infiltration [22], making them more responsive to lipid-lowering therapy through enhanced cholesterol efflux and inflammatory resolution. Additionally, these plaques demonstrate higher expression of matrix metalloproteinases and more active metabolic states [23], facilitating plaque remodeling under therapy. In contrast, hyperechoic plaques, characterized by higher calcium and fibrous tissue content, may be less amenable to short-term modification through lipid-lowering interventions.

These findings both support and extend previous research in the field. Our regression rates align closely with Kataoka et al.‘s findings [3], who reported 65% regression in lipid-rich plaques, while the observed relationship between echogenicity and regression potential supports Nicholls et al.‘s (2011) [14] conclusions about plaque composition influencing treatment response. Notably, our study provides the first large-scale evidence (n = 838) linking plaque echogenicity to regression potential, establishing quantitative regression criteria and demonstrating their clinical applicability [24]. This contributes to a more nuanced approach to risk stratification based on plaque characteristics.

The clinical implications of our findings are substantial. Plaque echogenicity assessment could serve as a valuable non-invasive tool for predicting treatment response, identifying patients who might benefit from more intensive therapy, and guiding monitoring frequency. Patients with hypoechoic plaques might benefit from more aggressive lipid-lowering strategies, earlier intervention, and more frequent monitoring of plaque progression. This suggests a potential paradigm shift in atherosclerosis management, where treatment approaches could be personalized based on plaque characteristics.

However, several limitations of our study warrant consideration. The single-center design may limit generalizability, while the observer-dependent nature of ultrasound assessment and relatively short follow-up period (12 months) suggest the need for further validation. Future research should focus on multicenter validation studies with longer follow-up periods, investigation of molecular mechanisms underlying differential regression rates, development of automated plaque characterization tools, and evaluation of clinical outcomes in relation to plaque regression patterns.

The therapeutic implications of our findings extend beyond individual patient care to suggest a broader paradigm shift in atherosclerosis management. This includes the potential for tailoring lipid-lowering intensity based on plaque characteristics, incorporating plaque echogenicity in risk assessment, and developing individualized monitoring strategies. Furthermore, plaque characteristics could be used to predict therapeutic efficacy, optimize resource allocation in plaque monitoring, and identify patients requiring more intensive intervention.

These insights provide a foundation for more precise and personalized approaches to atherosclerosis management, potentially improving patient outcomes through targeted interventions based on plaque characteristics. As we continue to advance our understanding of the relationship between plaque echogenicity and regression potential, these findings may lead to more efficient and effective treatment strategies in cardiovascular medicine.

Conclusion

This study demonstrates that plaque echogenicity serves as a valuable predictor of regression potential in carotid atherosclerosis under lipid-lowering therapy. The synergistic relationship between plaque echogenicity characteristics and LDL-C levels provides a foundation for risk stratification and treatment optimization. These findings support the development of personalized therapeutic strategies, particularly emphasizing intensive lipid-lowering approaches for patients with specific plaque characteristics. Future research should focus on validating these findings in larger, multicenter studies and exploring their implications for long-term cardiovascular outcomes.