This retrospective study enrolled 85 consecutive patients who were diagnosed with carotid stenosis using carotid artery ultrasonography (US) in Guangdong General Hospital, China, from January 2017 to January 2020. The inclusion criteria were as follows: (1) carotid stenosis confirmed using carotid artery ultrasonography, (2) absence of clinical contraindications for CEUS, and (3) ≥ 18 years of age. The exclusion criteria were: (1) having undergone previous carotid endarterectomy, (2) no outcome data concerning Lp(a), and (3) declining to be involved initially or in follow-up. All patients underwent CEUS after being diagnosed with carotid stenosis using carotid artery ultrasonography. Clinical histories, along with demographic and clinical data, were recorded for all patients at admission. All patients provided written informed consent. This retrospective investigation was approved by the local institutional review board as well as performed in accordance with the Declaration of Helsinki.

All fasting venous blood samples were collected during hospital admission before undergoing CEUS. Serum Lp(a) levels were measured through a murine monoclonal antibody (E022-1-1, Bioroyee, Beijing, China) involving latex turbidimetric method. Cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), apolipoprotein(apo) A and apo B levels were determined by chemiluminescence method using an auto-analyser. Friedwald's formula (TC -HDL-C-TG/2.2) was used to calculate LDL-C^(F).

Carotid US examinations were performed in all participants at screened visit. Focal structures protruding into carotid lumen with a height > 1.5 mm or 50% intima-media thickness was defined as atherosclerotic plaques [13]. Maximun vertical distance from the top of plaque to adventitia interface of lumen was measured as maximum internal carotid artery (ICA) plaque thickness. Total maximum plaque thickness was the cumulative total by bilateral maximum ICA plaques thickness. The degree of ICA stenosis was divided into 4 groups: mild, < 50%; moderate, 50–69%; severe, 70–99% [14]; and occlusion, without signal of blood flow.

Carotid CEUS examinations were performed by a researcher who is blinded to the patients’ histories and characteristics, using a GE Vivid E95 or Philips IU elite diasonograph contrast model and a high-frequency superficial probe. CEUS was performed with an ultrasound contrast agent, SonoVue. An initial bolus injection of 1.6 mL of contrast agent was quickly administered into the median cubital vein in 2–3 s, immediately followed by 3 mL of 0.9% normal saline solution at the same speed. Ultrasound cine-loops were then recorded over 15–30 s. The images at 3 s before and 5 min after contrast agent was introduced into the carotid artery lumen were stored for real-time dynamic analysis. IPN grade was determined using CEUS grade as follows: 0, no visible microbubbles in the plaque; 1, minimal microbubbles restricted to adventitial side or shoulder of the plaque; or 2, microbubbles spread all over the plaque [15]. We stratified participants into one of two groups based on their CEUS grade, that is, a CEUS grade on both sides that added up to greater than or equal to 2 was used to define an IPN group whereas a CEUS grade on both sides that added up to fewer than 2 was used to define a no IPN group.

All descriptive data consistent with normal distribution are expressed as mean value ± standard deviation, with the rest expressed as median (interquartile range). Discrete data are presented as frequencies and percentages. A Student’s t-test was used to evaluate continuous variables showing a normal distribution and a Mann–Whitney U-test was used to evaluate variables that show a skewed distribution, while categorial variables were performed using Fisher’s exact test. Differences in baseline characteristics were examined between IPN group versus No IPN group (Table 1) and Lp(a) ≥ 300 mg/L vs < 300 mg/L(Table 2). Spearman’s correlation analysis was further performed to analyse the relationship between Lp(a) and maximum ICA plaque thickness, total ICA maximum plaque thickness, degree of carotid stenosis and ICA occlusion, respectively. Univariate binary logistic regression analyses were used to determine the association between Lp(a) and IPN grade (IPN group vs No IPN group). OR and 95% CI was reported. Multivariate binary logistic regression analyses were then performed to evaluate the association between Lp(a) concentration and IPN grade (IPN group vs No IPN group), with stepwise adjustment for covariates including age, sex and comorbid conditions (model 1), total cholesterol, triglyceride, low-density lipoprotein cholesterol calculated by Friedwald's formula, high-density lipoprotein cholesterol, apolipoprotein A and apolipoprotein B (model 2), maximum ICA plaque thickness and total ICA maximum plaque thickness, degree of carotid stenosis and ICA occlusion (model 3). Meanwhile, participants were divided into two groups according to Lp(a) concentration (≥ 300 mg/L and < 300 mg/L). Univariate and multivariate binary logistic regression analyses were then performed to evaluate the association with IPN grade, adjusting for model 1, model 2 and model 3. All analyses were performed with SPSS version 22.0 for Windows, and a two-sided P value of less than 0.05 was considered to indicate statistical significance.

The characteristics of the participants are shown in Table 1 (IPN group vs No IPN group) and Table 2 (Lp(a) ≥ 300 mg/L vs < 300 mg/L). All 85 consecutive patients, that we had complete data for the binary logistic regression analysis, were seen in Guangdong General Hospital from January 2017 to January 2020. The mean age of the population was 68.48 ± 8.78 years. More than 70% of the participants were male, with 92.9% of the participants receiving statin therapy. Approximately 20% were smokers, with a similar proportion suffering from diabetes mellitus and cerebral infarction. Slightly fewer than half of the participants had coronary heart disease, and 70% had hypertension. Half of the patients (51.8%) admitted for cardiovascular ischemia-related reasons, including the symptom of chest pain or shortness of breath. While other 35.3% patients admitted for cerebral ischemia-related reasons including dizzy, blurred vision or unilateral limb weakness and 8.2% patients suffered from asymptomatic carotid stenosis. The IPN group tended to have greater maximum ICA plaque thickness and total maximum ICA plaque thickness compared to No IPN group, while compared to < 300 mg/L group, it was not significant in the Lp(a) ≥ 300 mg/L group. And Lp(a) ≥ 300 mg/L group have lower TG and higher HDL-C level than the Lp(a) < 300 mg/L group.

Spearman’s correlation analysis was performed to evaluate whether Lp(a) was associate with ICA plaque thickness or total ICA plaque thickness, degree of ICA stenosis, ICA occlusion. Neither the ICA plaque thickness (correlation coefficient = 0.205, P value = 0.060) nor the total ICA plaque thickness (correlation coefficient = 0.158, P value = 0.149) was correlated with plasm Lp(a) concentration. Likewise, both of the degree of ICA stenosis (correlation coefficient = 0.118, P value = 0.281) and ICA occlusion (correlation coefficient = − 0.027, P value = 0.806) were also not correlated with Lp(a) concentration. After being divided into two groups based on Lp(a) concentration, there were similarly not association between the ICA plaque thickness, total ICA plaque thickness, degree of ICA stenosis, ICA occlusion and whether Lp(a) ≥ 300 mg/L or not.(Table 3).

Table 4a shows that Lp(a), in the univariate analysis, was a significant predictor of IPN in carotid stenosis patients, with per 100 mg/L increasing associated with 1.238-fold higher hazard (95% CI 1.020, 1.503, P value = 0.031) of total IPN ≥ 2. After stepwise adjusting for covariates, per 100 mg/L increasing associated with 1.305-fold higher hazard (95% CI 1.045, 1.628, P value = 0.019) of total IPN ≥ 2.

Lp(a) ≥ 300 mg/L group, in the unadjusted model, was associated with 2.828-fold higher hazard (95% CI 1.055, 7.580, P value = 0.039) of total IPN ≥ 2 vs Lp(a) < 300 mg/L. After stepwise adjusting for other lipid parameters (Model 2) and maximum plaque thickness, total carotid maximum plaque thickness, degree of carotid stenosis, ICA occlusion (Model 3), the association did not reach statistical difference. (Table 4b).