Journal of Atherosclerosis and Thrombosis
Online ISSN : 1880-3873
Print ISSN : 1340-3478
ISSN-L : 1340-3478
Original Article
Prevalence of Lipoprotein(a) Measurement and its Association with Arteriosclerosis in Asymptomatic Individuals in China
Ping-ting Yang, Li Tang, Hui-rong Guo, Yong-mei He, Yue-xiang Qin, Lei Yan, Zhen-xin Li, Ya-zhang Guo, Jian-gang Wang
Author information
Keywords: Lipoprotein(a), Atherosclerosis, Testing rate, Brachial-ankle pulse wave velocity (baPWV), Carotid intima-media thickness (cIMT)
JOURNAL OPEN ACCESS FULL-TEXT HTML

2025 Volume 32 Issue 4 Pages 513-524

Browse “Advance online publication” version
Details
Download PDF (692K)
Download citation RIS

(compatible with EndNote, Reference Manager, ProCite, RefWorks)

BIB TEX

(compatible with BibDesk, LaTeX)

Text
How to download citation
Contact us
Abstract

Aims: Lipoprotein(a) [Lp(a)] is an independent risk factor for atherosclerotic cardiovascular disease (ASCVD), and its level is genetically determined. Although guidelines and consensuses in various cardiovascular fields have emphasized the importance of Lp(a), screening for Lp(a) in China has not been well studied.

Methods: A cross-sectional study was conducted using a random sample of 30,000 medical examiners from each of the five health check-up centres. The distribution of Lp(a) was described for those who completed Lp(a) testing, and logistic regression modelling was used to evaluate the relationship between Lp(a) levels and vascular structure and function in the population who underwent carotid ultrasound and brachial‒ankle pulse wave velocity (baPWV) measurements.

Results: Lp(a) was measured in only 4400 (3.02%) of the 150,000 participants. Among those tested for Lp(a), the median concentration was 15.85 mg/dL. The proportion of participants with Lp(a) levels ≥ 30 mg/dL was 15.00%. Multiple logistic regression analysis revealed a significant correlation between Lp(a) and cIMT ≥ 1.0 mm (OR: 1.008, 95% CI: 1.001-1.014, P=0.020) and carotid artery plaques (OR: 1.010, 95% CI: 1.004-1.016, P=0.001) but no correlation with baPWV ≥ 1400 (OR: 0.999, 95% CI: 0.993-1.005, P=0.788) or baPWV ≥ 1800 (OR: 1.002, 95% CI: 0.993-1.011, P=0.634).

Conclusions: The detection rate of Lp(a) at health checkups is low, and Lp(a) is positively associated with cervical vascular sclerosis and plaque but not with baPWV. Therefore, the testing rate of Lp(a) and the awareness of the risk of vascular structural changes due to Lp(a) should be further improved.

Ya-zhang Guo and Jian-gang Wang contributed equally to this work.

Introduction

Lipoprotein(a) [Lp(a)] consists of low-density lipoprotein (LDL)-like particles and apolipoprotein(a). Lp(a) has significant polymorphisms originating from the variable length of the apolipoprotein(a) peptide chain. Unlike LDL, Lp(a) cannot be converted from very low-density lipoprotein (VLDL) or other lipoproteins and is a separate class of lipoproteins synthesized by the liver. Epidemiologic and observational studies have shown a potential causal relationship between elevated Lp(a) levels and increased risks of the major types of polyvascular disease, atherosclerotic cardiovascular disease (ASCVD)1-3) and aortic valve stenosis (AS)4, 5). According to research by the European Atherosclerosis Society (EAS), an increase in Lp(a) is independent of traditional risk factors6) and increases the risk of vascular disease7). Lp(a) levels are genetically determined, minimally affected by environmental factors, and relatively stable over a person’s lifetime. Thus, individuals with elevated Lp(a) levels can be identified by once measurement as being at increased risk of developing ASCVD or AS. This knowledge can help refine patients’ ASCVD risk and motivate both clinicians and patients to modify ASCVD risk factors more aggressively when appropriate. Several international societal guidelines recommend Lp(a) screening in individuals with a family or personal history of ASCVD, and some, including the European Atherosclerosis Society (EAC) and the Canadian Cardiovascular Society (CCS)8, 9), recommend universal screening in adults10). For the first time, the Chinese Guidelines for Lipid Management (2023) also suggest that a lifetime should include at least one Lp(a) test and that an Lp(a) ≥ 30 is the cut-off value for the risk incidence for management11). However, countries are currently not active in Lp(a) testing in the general population, with reported testing rates ranging from 0.25%-4.60%12, 13).

In the asymptomatic population, carotid ultrasonography and brachial-ankle pulse wave velocity (baPWV) are the two most commonly used noninvasive methods for evaluating subclinical atherosclerosis. Carotid ultrasound can be used to evaluate vascular structure by measuring the carotid intima‒media thickness (cIMT), and the baPWV reflects vascular function by measuring pulse wave conduction velocity. These two indicators have been confirmed to be closely related to the risk and mortality of ASCVD events. Evidence supports that high levels of Lp(a), especially those higher than 30 mg/dL, contribute to atherosclerosis14), but Lp(a) primarily affects vascular structure or function is unknown. Although some studies have independently associated Lp(a) with the development of vascular plaques and the deterioration of the baPWV, others have not shown this association. In China, Lp(a) levels in asymptomatic individuals have not been investigated in large population-based studies since the promulgation of the newest guidelines for lipid management, and the effects of Lp(a) levels on vascular structure and function have been controversial. Therefore, in view of the important role of Lp(a) in the development of ASCVD, this study attempted to quantify Lp(a) in asymptomatic individuals and explore the relationship between Lp(a) and vascular structure and function through the collection of multicentre health examination data, which will help guide effective ASCVD prevention.

Subjects and Methods

Subjects

The participants were recruited from five departments of the health management centres of general tertiary hospitals located in northern and southern China. Participants who met the following criteria were included in the study: (1) aged 18–80 years and (2) had basic physical examination data (blood pressure, liver function, kidney function, blood glucose, blood lipids, abdominal ultrasound and electrocardiograms, etc.). Those who experienced major physical and mental illnesses, including severe liver and kidney dysfunction, malignant tumours, or acute disease conditions, were excluded. A total of 30,000 eligible subjects were randomly selected from each research centre, with a total of 15,000 subjects enrolled in this study from January 1, 2023, to December 31, 2023. We divided the subjects into two groups according to whether they completed Lp(a) testing to explore the testing rate and distribution of Lp(a). Among the 4400 individuals who completed Lp(a) testing, 3577 completed carotid ultrasound and baPWV measurements. We investigated the impact of Lp(a) levels on vascular structure and function in these subjects. The physical examination protocol and the consent form for this study were reviewed by the institutional review boards of five hospitals and approved by the independent ethics committee of the Third Xiangya Hospital of Central South University (No. 24005). Written informed consent was provided in this study following the general recommendations of the Declaration of Helsinki. The study flowchart is shown in Fig.1.

Fig.1.

Flow Diagram

Measurements of the Carotid Intima–Media Thickness (cIMT) and Atherosclerotic Plaques

Carotid ultrasound was performed by experienced ultrasound imaging physicians. Quality control was conducted at all subcentres for operating physicians and equipment. The kappa values of the cIMT and carotid plaque measurements among doctors at each check-up centre were greater than 0.80. All procedures followed the expert consensus on some problems with cerebral and carotid vascular ultrasonography15). The cIMT of the far-field arterial wall was measured via two-dimensional grayscale imaging with the probe parallel to the vessel wall, the sound beam perpendicular to the wall, and a combination of longitudinal and transverse views. The cIMT of the far-field arterial wall was measured in the distal segment of the common carotid artery (1.0–1.5 mm below the level of the bifurcation) and/or the carotid artery bulb (the relatively bulging segment of the origin of the internal carotid artery), avoiding atherosclerotic plaques. The perpendicular distance from the upper edge of the intima to the upper edge of the periphery, which is the combined thickness of the intima and the intima of the vessel wall, was measured. Carotid arteriopathy was defined as the presence of an abnormal cIMT or carotid plaques. An abnormal cIMT was defined as a mean cIMT ≥ 1.0 mm. A carotid plaque was defined as meeting the criteria of a cIMT ≥ 1.5 mm or the protrusion of atherosclerosis into the lumen of the artery with a thickness ≥ 50% compared with the peripheral area16).

baPWV Measurement

The baPWV was measured with an automatic wave for the manalyzer (BP-203 RPE III; Omron Health Medical, Dalian, China). After a minimum rest of 5 min in the supine position, 4 cuffs were wrapped around the upper arms and ankles and connected to a plethysmographic sensor (volume pulse form) and oscillometric pressure sensor. The baPWV was calculated via the formula (La - Lb)/ΔTba, where La is the distance from the heart to the ankle, Lb is the distance from the heart to the brachium, and ΔTba is the transmission time between the brachial and posterior tibial artery waveforms. The measurements were performed twice, and the average values of the left-side and right-side assessments were calculated. Two trained technicians performed all the measurements17). Previous research proposes baPWV cut-off values of 1400 and 1800 cm/s (＜1400 for normal, ≥ 1400 and ＜1800 for borderline, and ≥ 1800 for abnormal)18).

Laboratory Measurements

All laboratory tests were performed by certified laboratory physicians from the central laboratory department of the hospital using standard protocols. The biochemical tests included venous fasting serum glucose (FSG), total cholesterol (TC), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), Lp(a), creatinine (Cr) and uric acid (UA) levels measured after an early morning fast of 12 hours. All five research centres used an immunoturbidimetric assay to measure the Lp(a) concentration. The principle of the test is that Lp(a) in the human body and the antibody in the reagent rapidly form an antigen‒antibody complex in a buffer, which results in a turbid reaction solution; the amount of Lp(a) in the sample can then be calculated based on a comparison with the calibrator. The detection equipment used included Hitachi 7600, Roche Cobas 800, and Beckman AU680 automatic detectors and an Lp(a) assay kit (Leadman, Beijing, China).

Definition of the Main Chronic Disease

Hypertension was defined as self-reported hypertension diagnosed by a physician, self-reported regular use of antihypertensive medications, or a systolic/diastolic blood pressure at recruitment ≥ 140/90 mmHg19).

Diabetes mellitus was defined as self-reported diabetes diagnosed by a physician, self-reported regular use of antidiabetic medications, or FSG at recruitment ≥ 7.0 mmol/L20).

Statistical Analyses

The data were analysed using the Statistical Package for Social Sciences (SPSS Inc., Chicago, IL, version 22.0 for Windows) and R version 4.1.1 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables are presented as the means±standard deviations (SDs), and categorical variables are reported as percentages (%) and numbers (n). Student’s t test was used to evaluate continuous variables with a normal distribution, and the Mann‒Whitney U test was used to evaluate variables with a skewed distribution, whereas categorical variables were analysed via Fisher’s exact test. Differences in baseline characteristics were examined between the Lp(a) measured group and the Lp(a) not measured group (Table 1) and between the Lp(a) ≥ 30 mg/dL group and the Lp(a) ＜30 mg/dL group (Table 2). The relationships between Lp(a) levels and arterial structure and function were assessed via logistic regression modelling in a population that had completed Lp(a) testing, carotid ultrasound and atherosclerosis measurements. Traditional cardiometabolic factors, including age, sex, body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), FSG, SCr, BUN, Ua, TC, LDL-C, HDL-C, hypertension and diabetes mellitus, were adjusted as covariates in the model. A two-sided p value ＜0.05 was considered statistically significant.

Table 1.Characteristics of participants with Lp(a) measured vs without Lp(a) measured (mean±SD, N%)

Characteristics Total Lp(a) Measured Lp(a) Not Measured t/χ2 P
N 150,000 4,400 145,600
Age (years) 43.66±12.77 50.20±10.71 43.46±12.78 40.872 ＜0.001
BMI (kg/m2) 24.01±3.55 25.02±3.44 23.98±3.55 19.041 ＜0.001
SBP (mmHg) 121.35±15.27 126.24±16.16 121.20±15.22 20.401 ＜0.001
DBP (mmHg) 73.96±10.88 77.29±11.56 73.86±10.85 19.451 ＜0.001
FSG (mmol/L) 5.35±1.21 5.54±1.68 5.34±1.20 7.542 ＜0.001
SCr (mmol/L) 67.45±19.27 67.22±26.12 67.45±19.02 -0.785 0.432
BUN (mmol/L) 4.69±1.26 5.01±1.33 4.68±1.26 16.420 ＜0.001
Ua (mmol/L) 333.64±88.23 343.48±88.89 333.34±88.19 7.491 ＜0.001
TC (mmol/L) 5.05±0.99 5.28±1.07 5.05±0.98 14.124 ＜0.001
TG (mmol/L) 1.72±1.77 2.14±2.28 1.72±1.75 12.329 ＜0.001
LDL-C (mmol/L) 2.94±0.82 2.92±0.82 2.94±0.82 -1.416 0.157
HDL-C (mmol/L) 1.33±0.31 1.32±0.30 1.33±0.31 -2.748 0.006
Male % (n) 50.79 (76,186) 58.64 (2,580) 50.55 (73,606) 111.639 ＜0.001
Hypertension % (n) 12.66 (18,992) 21.50 (946) 12.40 (18,046) 320.236 ＜0.001
Diabetes mellitus % (n) 7.68 (11,520) 14.41(634) 7.48 (10,886) 289.493 ＜0.001

Note: SD, standard deviance; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FSG, fasting serum glucose; SCr, serum creatinine; BUN, blood urea nitrogen; Ua, uric acid; TC, total cholesterol; TG, triglycerides; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol.

Table 2.Characteristics of the participants completing the examination of vascular structure and function by Lp(a) (mean±SD, N%)

Characteristics Total Lp(a) ＜30mg/dL Lp(a) ≥ 30mg/dL t/χ2 P
N 3,577 3045 532
Age (years) 51.93±9.38 51.67±9.41 53.45±9.11 -4.046 ＜0.001
BMI (kg/m2) 25.19±3.33 25.29±3.33 24.66±3.29 4.014 ＜0.001
SBP (mmHg) 127.2±16.29 127.27±16.19 126.81±16.88 0.601 0.548
DBP (mmHg) 78.01±11.62 78.22±11.66 76.80±11.32 2.593 0.010
FSG (mmol/L) 5.62±1.78 5.67±1.86 5.34±1.14 5.503 ＜0.001
SCr (mmol/L) 67.26±16.81 67.45±16.46 66.15±18.63 1.639 0.101
BUN (mmol/L) 5.06±1.33 5.01±1.32 5.07±1.34 -0.151 0.880
Ua (mmol/L) 345.48±88.16 347.01±88.19 336.74±87.51 2.477 0.013
TC (mmol/L) 5.29±1.04 5.25±1.02 5.52±1.13 -5.066 ＜0.001
TG (mmol/L) 2.19±2.30 2.29±2.45 1.66±0.95 10.356 ＜0.001
LDL-C (mmol/L) 2.94±0.81 2.89±0.79 3.22±0.85 -8.718 ＜0.001
HDL-C (mmol/L) 1.31±0.30 1.30±0.30 1.38±0.31 -6.247 ＜0.001
Lp(a)(mg/dL) 19.78±14.16 15.05±6.23 46.85±16.27 -44.502 ＜0.001
baPWV(cm/s) 1456.36±280.23 1457.01±281.71 1452.68±271.79 0.329 0.742
Male % (n) 60.33 (2158) 62.36 (1899) 48.68 (259) 35.414 ＜0.001
Hypertension % (n) 24.52 (877) 24.53 (747) 24.44 (130) 0.002 0.962
Diabetes mellitus % (n) 12.47 (446) 12.91 (393) 9.96 (53) 3.596 0.058
cIMT ≥ 1.0 mm % (n) 66.87 (2392) 65.98 (2009) 71.99 (383) 7.397 0.007
Carotid plaque % (n) 58.32 (2086) 57.34 (1746) 63.91 (340) 8.042 0.005

Note: SD, standard deviance; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; FSG, fasting serum glucose; SCr, serum creatinine; BUN, blood urea nitrogen; Ua, uric acid; TC, total cholesterol; TG, triglycerides; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; Lp(a), Lipoprotein (a); baPWV, brachial-ankle pulse wave velocity; cIMT, carotid intima-media thickness.

Results

Demographic Characteristics

A total of 150,000 subjects who met the inclusion criteria were randomly selected from five research centres. The mean age of the participants was 43.66±12.77 years, and 50.79% (n=76,186) were male. A total of 12.66% of the participants were diagnosed with hypertension, and 7.68% were diagnosed with diabetes mellitus. Of these, 4,400 (3.02%) had Lp(a) measured, and 145,600 had baseline lipid indices measured but not Lp(a). Table 1 shows the demographic and health characteristics of the subjects whose Lp(a) levels were measured and those whose Lp(a) levels were not measured. Subjects who had Lp(a) levels measured were more likely to be older; have a greater BMI; have greater blood pressure; have higher levels of FSG, BUN, Ua, TC and TG; be male; be hypertensive; and be diabetic.

Lp(a) Distributions

The distribution of Lp(a) concentrations was skewed to the right in the 4400 individuals in whom Lp(a) was measured (Fig.2 A). The median Lp(a) concentration in the entire population was 15.85 mg/dL. The prevalence of Lp(a) levels ≥ 30 was 15.00%. The median Lp(a) level was 17.30 mg/dL in females and 14.70 mg/dL in males. Using 30 mg/dL as the cut-off, 19.18% of females and 12.29% of males had elevated Lp(a) levels (Fig.2 B). The Lp(a) values increased with age and were highest in the 66- to 70-year-old group, with a mean of 22.65 mg/dL. The three groups aged 56–70 years had a mean Lp(a) concentration greater than 20 mg/dL, but the Lp(a) value decreased in the group aged 71–80 years. The pattern of distribution was similar in males and females (Fig.2 C, D). After the population was grouped according to hypertension status and diabetes status, the mean Lp(a) levels in the hypertension group and nonhypertension group were 18.89 mg/dL and 19.74 mg/dL, respectively, with no significant difference (P=0.781). The mean Lp(a) concentrations in the diabetes group and nondiabetes group were 18.45 mg/dL and 20.00 mg/dL, respectively, which were significantly different (P=0.014) (Fig.2 E).

Fig.2.

(A) Density of Lp(a) levels in the entire study population. (B) Density of Lp(a) levels in females and males. The median Lp(a) level and proportion of patients with an Lp(a) level ≥ 30 mg/dL are marked separately (B). Age-based mean fold plots of the trajectories of median Lp(a) levels in all subjects (C) and in each 5-year age group (except for the 18- to 20-year age group) stratified by sex (D). (E) Differences in Lp(a) between hypertensive and nonhypertensive populations and between diabetic and nondiabetic populations were compared.

Association of Lp(a) with Vascular Structure

Among the population whose Lp(a) level was measured, 3,577 people completed both vascular structure and function examinations, i.e., carotid ultrasound and baPWV measurements. The population was grouped according to Lp(a)＜30 mg/dL and Lp(a) ≥ 30 mg/dL (Table 2). The average age of this population was 51.93±9.38 years, and 60.33% were male. A total of 24.52% of the participants were diagnosed with hypertension, and 12.47% were diagnosed with diabetes mellitus. The mean values of Lp(a) in the two groups were 15.05 mg/dL and 46.85 mg/dL, respectively. For baPWV, the mean values were 1457.01 cm/s and 1452.68 cm/s, respectively, with no significant differences between the two groups (P=0.742). The proportions of patients with cIMT ≥ 1.0 mm in the Lp(a) ＜30 mg/dL and Lp(a) ≥ 30 mg/dL groups were 65.98% and 71.99%, respectively, with significant differences between the groups (P=0.007). The proportions of patients with carotid plaques were 57.34% and 63.91%, respectively, and the difference between the groups was significant (P=0.005). The incidence of hypertension or diabetes mellitus did not significantly differ between groups.

We used a logistic regression model to analyse the correlation between Lp(a) and vascular structure, and the results are shown in Table 3. We classified the carotid ultrasound results as vascular sclerosis (cIMT ≥ 1.0 mm) and the appearance of carotid plaques based on the intravascular intima–media thickness. Lp(a) was positively correlated with cIMT ≥ 1.0 mm (OR: 1.008, 95% CI: 1.001–1.014, P=0.020 in Model 3). No correlation was found between hypertension patients (OR: 1.015, 95% CI: 0.997-1.034, P=0.102 in Model 3) and nonhypertensive patients (OR: 1.007, 95% CI: 1.000-1.014, P=0.061 in Model 3). Lp(a) was positively correlated with cIMT ≥ 1.0 mm in diabetes mellitus patients (OR: 1.033, 95% CI: 1.004–1.063, P=0.027 in Model 3), but this correlation was not found in nondiabetic mellitus patients (OR: 1.006, 95% CI: 0.999–1.013, P=0.086 in Model 3). Lp(a) was positively correlated with carotid plaques (OR: 1.010, 95% CI: 1.004–1.016; P=0.001 in Model 3). The same correlation was found in nonhypertensive patients (OR: 1.010, 95% CI: 1.003–1.017, P=0.003 in Model 3), diabetes mellitus patients (OR: 1.026, 95% CI: 1.004–1.048, P=0.021 in Model 3) and patients without diabetes mellitus (OR: 1.009, 95% CI: 1.003–1.015, P=0.006 in Model 3).

Table 3.Associations of Lp (a) with carotid arteriopathy in different groups

Variables N (%) Unadjusted model Adjusted model 1 Adjusted model 2
OR (95% CI) P OR (95% CI) P OR (95% CI) P
cIMT ≥ 1.0 mm

All population

(n = 3,577)

2392 (66.87%)

1.01

(1.004, 1.015)

 ＜0.001

1.008

(1.002, 1.014)

 0.012

1.008

(1.001, 1.014)

 0.020

Hypertension

(n = 877)

741 (84.49%)

1.022

(1.005, 1.039)

 0.010

1.015

(0.998, 1.033)

 0.088

1.015

(0.997, 1.034)

 0.102

Non hypertension

(n = 2,700)

1651 (61.15%)

1.009

(1.003, 1.015)

 0.003

1.007

(1.000, 1.014)

 0.049

1.007

(1.000, 1.014)

 0.061

Diabetes mellitus

(n = 446)

370 (82.96%)

1.035

(1.008, 1.062)

 0.009

1.032

(1.004, 1.060)

 0.026

1.033

(1.004, 1.063)

 0.027

Non diabetes

mellitus(n = 3,131)

2022 (64.58%)

1.009

(1.003, 1.014)

 0.002

1.006

(1.000, 1.013)

 0.050

1.006

(0.999, 1.013)

 0.086
Carotid plaque

All population

(n = 3,577)

2086 (58.32%)

1.011

(1.006, 1.016)

 ＜0.001

1.011

(1.005, 1.016)

 ＜0.001

1.010

(1.004, 1.016)

 0.001

Hypertension

(n = 877)

673 (76.74%)

1.016

(1.003, 1.029)

 0.014

1.012

(0.998, 1.025)

 0.100

1.013

(0.999, 1.028)

 0.067

Nonhypertension

(n = 2,700)

1413 (52.33%)

1.010

(1.005, 1.016)

 ＜0.001

1.010

(1.004, 1.017)

 0.001

1.010

(1.003, 1.017)

 0.003

Diabetes mellitus

(n = 446)

336 (75.34%)

1.023

(1.004, 1.043)

 0.017

1.025

(1.004, 1.046)

 0.020

1.026

(1.004, 1.048)

 0.021

Nondiabetes

mellitus(n = 3,131)

1750 (55.89%)

1.010

(1.005, 1.016)

 ＜0.001

1.010

(1.004, 1.015)

 0.002

1.009

(1.003, 1.015)

 0.006

Model 1 was adjusted for age, sex, BMI, SBP, DBP.

Model 2 was adjusted for age, sex, BMI, SBP, DBP, FSG, SCr, BUN, Ua, TC, LDL-C, HDL-C, hypertension and diabetes mellitus. OR, odds ratio; CI, confidence interval; cIMT, carotid intima-media thickness.

Association of Lp(a) with Vascular Function

We used a logistic regression model to analyse the correlation between Lp(a) and vascular function, and the results are shown in Table 4. Vascular function was divided into borderline (baPWV ≥ 1400 cm/s) and abnormal (baPWV ≥ 1800 cm/s) according to the baPWV, and the analysis revealed that the Lp(a) level did not correlate with vascular function in either the borderline (OR: 0.999, 95% CI: 1.004–1.048, P=0.788 in Model 3) or abnormal (OR: 1.002, 95% CI: 0.993–1.011, P=0.634 in Model 3) group. The same results were found in hypertension and diabetes mellitus patients.

Table 4.Associations of Lp (a) with baPWV in different groups

Variables N (%) Unadjusted model Adjusted model 1 Adjusted model 2
OR (95% CI) P OR (95% CI) P OR (95% CI) P
baPWV ≥ 1400 cm/s

All population

(n = 3,577)

1818 (50.82%)

0.999

(0.995, 1.004)

 0.731

0.997

(0.991, 1.003)

 0.355

0.999

(0.993, 1.005)

 0.788

Hypertension

(n = 877)

708 (80.73%)

0.998

(0.987, 1.009)

 0.697

0.992

(0.979, 1.006)

 0.277

0.993

(0.978, 1.007)

 0.328

Nonhypertension

(n = 2,700)

1110 (41.11%)

1

(0.994, 1.005)

 0.912

0.998

(0.992, 1.005)

 0.615

1.001

(0.994, 1.008)

 0.804

Diabetes mellitus

(n = 446)

345 (77.35%)

0.999

(0.985, 1.013)

 0.877

0.997

(0.98, 1.013)

 0.681

1.004

(0.986, 1.022)

 0.685

Nondiabetes

mellitus(n = 3,131)

1473 (47.05%)

1

(0.995, 1.005)

 0.980

0.997

(0.991, 1.004)

 0.431

0.999

(0.992, 1.006)

 0.734
baPWV ≥ 1800 cm/s

All population

(n = 3,577)

395 (11.04%)

1.004

(0.997, 1.011)

 0.222

1

(0.992, 1.009)

 0.916

1.002

(0.993, 1.011)

 0.634

Hypertension

(n = 877)

234 (26.68%)

1.004

(0.994, 1.014)

 0.417

0.999

(0.988, 1.011)

 0.930

1.002

(0.99, 1.015)

 0.710

Nonhypertension

(n = 2,700)

161 (5.96%)

1.006

(0.995, 1.016)

 0.270

1.001

(0.989, 1.014)

 0.815

1.003

(0.99, 1.016)

 0.673

Diabetes mellitus

(n = 446)

121 (27.13%)

1.010

(0.997, 1.023)

 0.140

1.009

(0.993, 1.024)

 0.292

1.006

(0.988, 1.023)

 0.522

Nondiabetes

mellitus(n = 3,131)

274 (8.75%)

1.004

(0.995, 1.012)

 0.414

0.999

(0.989, 1.01)

 0.894

1.001

(0.99, 1.011)

 0.906

Model 1 was adjusted for age, sex, BMI, SBP, DBP.

Model 2 was adjusted for age, sex, BMI, SBP, DBP, FSG, SCr, BUN, Ua, TC, LDL-C, HDL-C, hypertension and diabetes mellitus. OR, odds ratio; CI, confidence interval.

Discussion

This study revealed that the prevalence of Lp(a) measurement was 3.02% in the 2023 asymptomatic physical examination population at Chinese health check-up centres, and the median Lp(a) level of those who completed Lp(a) testing was 15.85 mg/dL. The proportion of people with Lp(a) levels ≥ 30 was 15.00%. The median Lp(a) level was greater in females and elderly individuals. The Lp(a) level was lower in diabetes mellitus patients than in patients without diabetes mellitus. The Lp(a) level was significantly correlated with the incidence of cIMT ≥ 1.0 mm and carotid artery plaques, but it was not associated with the baPWV.

Earlier guidelines or consensus statements on Lp(a) had rather complex rules, suggesting that people with a high risk of ASCVD with concomitant disease should have Lp(a) measured once. These rules have changed with the 2019 ESC/EAS guidelines on dyslipidaemia, and the simpler recommendation of “Lp(a) measurement should be considered at least once in the lifetime of every adult” is much easier to follow21). The Chinese Guidelines for Lipid Management (2023) also specify this recommendation. However, knowledge of Lp(a) measurement in practice is limited, and this indicator is not included in routine examinations during outpatient or physical examinations. This study collected medical examination data from 150,000 people in five different regions of China from January 1, 2023, to December 31, 2023, and reported that the Lp(a) measurement rate was 3.02%. To our knowledge, this study is the first to report the prevalence of Lp(a) testing in asymptomatic physical examination populations.

Previous studies have focused mostly on cardiovascular disease or high-risk populations. Wilkinson et al.13) collected data on the prevalence of Lp(a) measurements in 2,710 patients with calcific aortic valve stenosis (CAVS) and 1,369 control patients without CAVS between January 2010 and February 2016 and reported that the prevalence of any Lp(a) measurement was 4.6% in patients with CAVS and 3.1% in controls, suggesting that Lp(a) measurements in patients with CAVS should be more widely obtained in clinical practice. Sturzebecher et al.22) analysed data from 4 million patient records in Germany in 2015 and 2018 and reported that 0.25% and 0.34% of patients with ASCVD, respectively, underwent Lp(a) testing. A retrospective, observational case‒control study of patients from 2017 to 2022 was conducted at Hartford Medical Center in the United States. Among the 241,220 patients in the health network, only 0.25% had their Lp(a) measured. Among 84,581 primary care or cardiac clinic patients, only 1.55% had their Lp(a) measured12). Another study of the frequency of Lp(a) testing at the University of Rochester Medical Center collected data from all patients who had at least one Lp(a) measurement in their electronic health records between January 2011 and August 2022. Approximately 11% of patients underwent more than one Lp(a) measurement, with an increasing number of patients being tested each year23). The above studies all suggest that Lp(a) testing is associated with more intensive prophylactic treatment and positively impacts clinical outcomes and survival. Therefore, promoting Lp(a) measurement in the general population to identify this high-risk factor would be helpful for cardiovascular risk management.

Previous studies focusing on Lp(a) and its role in atherosclerotic cardiovascular disease have been conducted only in Caucasian patients. Multiple large-scale independent investigations have revealed significant racial and ethnic differences in plasma Lp(a) levels and population distributions over the past four decades24). A personal record evaluating 126,634 participants from 36 prospective studies revealed 12% higher Lp(a) levels in females and 11% lower Lp(a) levels in diabetic mellitus patients7). This finding is consistent with the results of our study. Several studies have shown higher Lp(a) levels in females than in males, but there is no correlation or only a small correlation between endogenous hormones and postmenopausal hormone replacement therapy10, 25), suggesting that the difference may be greater at the genetic level. Some cohort studies of different populations and ethnic groups have shown that the Lp(a) level is inversely correlated with the risk of diabetes mellitus. This correlation is nonlinear, and the risk increases sharply only when the Lp(a) level is very low, reaching a plateau at moderate and high Lp(a) values26, 27). A similar phenomenon has been reported in the Chinese population28, 29). The plasma Lp(a) levels for all 23,998 individuals measured at the Karolinska University laboratories from 2003 to 2017 reached a median of 17 mg/dL30). Studies of Chinese disease populations have revealed that the median Lp(a) concentration is 18.83 mg/dL in patients with a history of myocardial infarction31), 30.27 mg/dL in patients with coronary artery disease32), 21.95 mg/dL in patients with stabilized coronary artery disease after percutaneous coronary intervention33), and 28.04 mg/dL in the Northern Community Chronic Disease Cohort34). The median Lp(a) concentration in the asymptomatic physical examination population in our study was 15.85 mg/dL, which was lower than that in the disease population. This value is closer to the 11.9 mg/dL reported in the community population of the Kailuan study35) and the 10.60 mg/dL reported in another study with a physical examination population36).

In this study, Lp(a) was found to be correlated with vascular structure, i.e., cIMT and carotid plaques, but not with vascular function, i.e., baPWV. The relationship between Lp(a) and atherosclerosis is controversial. Raitakari et al.37) suggested that Lp(a) was not correlated with either the structure or the function of the vessel. Sramek et al.38) reported no increase in cIMT in carotid or femoral arteries at high Lp(a) levels and that Lp(a) was not related to changes in the early atherosclerotic vascular wall of the carotid or femoral artery. Steffen et al.39) concluded that the association of Lp(a) with carotid plaques was race related and that Lp(a) levels may have a greater effect on plaque burden in whites than in blacks. Huffman et al.40) reported no association between Lp(a) concentration and atherosclerosis in South Asians aged 40 to 84 years without cardiovascular disease living in the United States. However, most studies have concluded that Lp(a) contributes to the development of atherosclerosis. Van et al.41) reported that in patients with symptomatic carotid artery stenosis, elevated Lp(a) levels were correlated with the degree of carotid artery stenosis. Xia et al.42) reported that Lp(a) is independently associated with increased levels of invasive neovascularization (IPN) in patients with carotid artery stenosis. A study of the coronary artery also revealed that Lp(a) was significantly related to the severity of coronary artery abnormalities43). In patients with advanced stable coronary artery disease, Lp(a) is related to the accelerated progression of coronary artery low attenuation plaque (necrotic core)44). An increase in Lp(a) is independently related to an increase in the percentage of atherosclerotic volume45).

The baPWV is a standard indicator of the stiffness of large arteries, a marker of vascular ageing and decreased arterial elasticity, and has predictive value for future ASCVD events. Previous studies reported a positive correlation between Lp(a) levels and the PWV46, 47). However, these studies were small in population size and were population specific. In a recent large-scale observational study, Lp(a) levels were not associated with arterial stiffness48). In a Mendelian randomized study involving 290,497 subjects, no evidence was found for a causal relationship between Lp(a) and the PWV49).

The formula to calculate the baPWV is (La - Lb)/DTba, which reflects the speed of pulse wave conduction; the stiffer the blood vessel is, the faster the pulse wave conduction. The stiffness of blood vessels increases with age and is associated mainly with hypertension, diabetes mellitus and unhealthy lifestyles18). Stiffening is primarily due to the ageing of elastic fibres in the blood vessel wall50). However, Lp(a) levels are genetically determined and have no direct effect on vascular stiffness. Lp(a) participates in atherosclerosis through inflammatory and thrombotic mechanisms51, 52) and has a direct effect on vascular structure, especially the vascular intima and media.

Compared with the familiar LDL-C, Lp(a) has received less attention, mainly because it has received attention later. Currently, an increasing number of studies are clarifying the role of Lp(a) as a risk factor for ASCVD. We have much work to do to increase the attention given to Lp(a). First, clinicians should focus on Lp(a) to clarify its significance and increase the detection rate of high-risk populations. We also need to increase the public’s awareness of Lp(a) through the popularization of science to allow patients to consciously and actively control related risk factors.

Limitations

This study has several limitations. First, the subcentres included in this study were all tertiary hospital medical check-up facilities in cities with moderate to high economic levels, which cannot fully reflect the true status of Lp(a) testing in China. Second, the effect on the measurement rate of Lp(a) in the medical check-up population is related not only to the presence of individual risk factors but also to the selection of physical examination items by the group. Third, this study was observational with limited potential for drawing causal inferences. Fourth, we minimized the rate of missing data by defining the inclusion criteria, which may have led to population bias. Fifth, incomplete data collection resulted in the absence of information on some medical and medication histories, such as history of ASCVD and lipid-lowering medication use. Finally, our study did not consider genetic variation, and plasma Lp(a) levels are largely determined by genetic factors; therefore, indicators such as lifestyle were not included in this study.

Conclusions

This is study the first to describe the prevalence of Lp(a) measurements in a large population in China. Although various guidelines and consensuses have emphasized the importance of Lp(a) testing and the relationship between Lp(a) and cardiovascular disease has been increasingly recognized, the current test remains subject to deficiencies. Our study also revealed that Lp(a) is associated with vascular structure, represented by the carotid artery, but not with vascular function, represented by the baPWV, in the asymptomatic physical examination population. An improved test of Lp(a) may identify people with underestimated cardiovascular risk who have normal conventional cardiovascular risk factors, and effective control of Lp(a) could ameliorate its effects on vascular structure.

Acknowledgements

The authors thank all the staff for their help.

Author Contributions

Ping-ting Yang: investigation, methodology, formal analysis, visualization, writing—original draft; Li Tang: investigation, formal analysis, methodology; Hui-rong Guo: investigation, methodology, writing—original draft; Yong-mei He: investigation; Yue-xiang Qin: supervision, writing—review and editing; Lei Yan: investigation; Zhen-xin Li: investigation; Ya-zhang Guo: investigation, validation, supervision, writing—editing; Jian-gang Wang: funding acquisition, conceptualization, methodology, validation, supervision, writing—review and editing.

Funding

This work was supported by funding from the Hunan Medical Scientific Research Project (W20243020), the Chinese Cardiovascular Association-ASCVD Fund (2023-CCA-ASCVD-018), the Project of State Key Clinical Department(Z2023058), and the National Natural Science Foundation of China (82303133).

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Disclosure and Competing Interests Statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References
 

This article is licensed under a Creative Commons [Attribution-NonCommercial-ShareAlike 4.0 International] license.
https://creativecommons.org/licenses/by-nc-sa/4.0/
feedback
 Top