Abstract
Background and aims:
Plasma low-density lipoprotein (LDL)-cholesterol (LDL-C) and lipoprotein(a) (Lp(a)) are both associated with coronary heart disease (CHD). This study investigated whether elevated plasma Lp(a) concentration was associated with increased CHD risk when LDL-C was low (≤ 100 mg/dL) in individuals not on statin therapy.
Methods:
Participants from the Multi-Ethnic Study of Atherosclerosis (MESA) (n= 4,585) were categorized into four groups: Group 1: LDL-C ≤ 100 mg/dL, Lp(a) < 50 mg/dL; Group 2: LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL; Group 3: LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL; and Group 4: LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL. The relationship of Lp(a) and LDL-C with time to CHD events was assessed with Kaplan Meier curves and multivariable Cox proportional hazard models.
Results:
Participants were followed for a mean of 13.4 years and a total of 315 CHD events occurred. Compared to participants with LDL-C ≤ 100 mg/dL and Lp(a) < 50 mg/dL, those with LDL-C > 100 mg/dL and Lp(a) < 50 mg/dL (Group 2) demonstrated no increased risk for CHD events (HR: 0.92; 95% CI: 0.69, 1.21). However, participants with LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 3) and those with LDL-C > 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 4) exhibited significantly increased risk of CHD events compared to Group 1 (HR: 1.83; 95% CI: 1.02, 3.27) and (HR: 1.61; 95% CI: 1.15, 2.26), respectively.
Conclusions:
When Lp(a) was elevated, risk of CHD events increased, regardless of baseline LDL-C.
Keywords: Lipoprotein(a), primary prevention, low-density lipoprotein-cholesterol, atherosclerosis, cardiovascular risk, coronary heart disease
Introduction
Multiple lines of evidence have established that apolipoprotein B (apoB)-containing lipoproteins are causally associated with the development of atherosclerosis.1 Typically, low-density lipoprotein (LDL) is the most abundant apoB-containing lipoprotein in plasma and serves as the primary driver of atherosclerotic plaque formation.2 The rate at which atherosclerotic plaques develop and grow is related to both the concentration and duration of (e.g., cumulative) exposure to atherogenic lipoproteins.3, 4 Thus, lifestyle and pharmacological interventions that reduce plasma LDL-cholesterol (LDL-C) lead to lower atherosclerotic cardiovascular disease (ASCVD) risk.3
Lipoprotein(a) (Lp(a)) is another apoB-containing lipoprotein that has been shown through epidemiologic and genetic association studies to be causally associated with coronary heart disease (CHD).5-7 The lipid composition of Lp(a) particles is similar to LDL, but these particles are characterized by the presence of a glycoprotein termed apolipoprotein(a) (apo(a)) covalently linked to the apoB moiety.8 Interestingly, the apo(a) moiety of Lp(a) shares structural homology with plasminogen, a proenzyme of plasmin, and may thus competitively inhibit plasminogen activation and contribute to thrombosis.9-11 Additionally, apo(a) contains lysine binding sites, which serve as a repository for plasma oxidized phospholipids facilitating attachment to the endothelium and promotion of inflammation and atherogenesis, respectively.11 Thus, beyond its cholesterol payload, there are several other features of Lp(a) that may also contribute to its association with CHD and residual risk even when LDL-C is well controlled.11 Current evidence suggests that Lp(a) is independently associated with CHD, peripheral arterial disease, and calcific aortic valve stenosis.5, 11 The relationship between Lp(a) and stroke is less clear, but almost certainly less pronounced than for CHD.12, 13 Lp(a) related risk is continuous across its plasma levels; however, at values ≥ 50 mg/dL, ASCVD risk becomes clinically significant.9, 14 As such, this threshold is considered a risk-enhancing factor by national guidelines.15 Furthermore, Lp(a) is the most common inherited form of dyslipidemia, with an estimated 1.4 billion individuals worldwide with an Lp(a) ≥ 50 mg/dL.16
Elevated Lp(a) may be a source of residual atherosclerotic risk, but its implications are unclear among individuals free of ASCVD who have well controlled LDL-C levels (≤ 100 mg/dL). Moreover, a recent analysis in patients with established ASCVD found that the use of aggressive add-on lipid lowering therapy once LDL-C levels were well treated was only beneficial if Lp(a) was elevated.17 To investigate this issue amongst a primary prevention population, we compared time to incident CHD events stratified according to both LDL-C and Lp(a) in a large multi-ethnic cohort free from clinical cardiovascular disease at baseline.
Patients and methods
Study population
Data used in this study were obtained from the Multi-Ethnic Study of Atherosclerosis (MESA) (https://www.mesa-nhlbi.org) in accordance with their published data access policies, including an approved written proposal. The methodology of the Multi-Ethnic Study of Atherosclerosis (MESA) has been previously described.18 In brief, MESA recruited a diverse cohort of men and women from the year 2000 to 2002, 45 to 84 years of age, from 6 communities in the United States to better understand the natural history of atherosclerosis.18 In total, MESA enrolled 6,814 participants, all free of ASCVD at recruitment. Individuals without documented Lp(a) (n=2,155) were first excluded from the study. Then, individuals without calculated LDL-C (n=7) or missing LDL-C (n=46) were excluded. Lastly, those with missing CHD event data (n=21) were excluded from the study (Figure 1).
Figure 1. Flow Diagram of Study Population.
From the total 6,814 MESA participants, 2,155 were initially excluded for missing Lp(a). Individuals without calculated LDL-C (n=7) or missing LDL-C (n=46) were subsequently excluded. Lastly, those with missing CHD event data (n=21) were excluded from the study.
Our study included 4,585 individuals with documented LDL-C, Lp(a), and CHD event data, who were not on statin, fibrate, or niacin therapy from the ancillary MESA Apolipoprotein study. Institutional Review Board approval was obtained at all participating institutions, and written informed consent was obtained from all participants.
Demographic and baseline characteristics
Questionnaires were used to gather demographic data as well as cigarette use, categorized as never, former, and current (at least one cigarette consumed in the past 30 days). Body mass index (BMI) was quantified by dividing weight (kg) by height squared (m2) and treated as a continuous variable. Hypertension was present in participants with a systolic blood pressure ≥ 140 mmHg, or diastolic blood pressure ≥ 90 mmHg, or if taking medications for hypertension. Diabetes mellitus was present in participants with a fasting serum glucose ≥ 126 mg/dL, self-reported diagnosis of diabetes, or if taking medications for diabetes mellitus. All covariate data were obtained from the MESA baseline exam (2000 to 2002).
Laboratory measurements
Participants provided serum samples in EDTA-anticoagulant tubes that were stored at −70° C for laboratory assessment. Overnight fasting serum samples were obtained to measure total cholesterol, high-density lipoprotein (HDL)-cholesterol (HDL-C), and triglycerides at Collaborative Studies Clinical Laboratory (Minneapolis, MN). The cholesterol oxidase method (Roche Diagnostic, Indianapolis, IN) was used to measure total cholesterol and HDL-C. A triglyceride GB reagent was used to measure triglycerides (Roche Diagnostic, Indianapolis, IN). In samples with triglycerides below 400 mg/dL, the Friedewald equation was used to calculate LDL-C.19 Individuals with triglycerides above 400 mg/dL did not have calculated LDL-C; and thus, were excluded (n=7). Plasma Lp(a) mass was measured with a latex-enhanced turbidimetric immunoassay (Denka Seiken, Tokyo, Japan) by Health Diagnostics Laboratory (Richmond, Virginia).20 This type of immunoassay measures Lp(a) mass using a polyclonal antibody and is apo(a) isoform dependent.21 Lp(a) was measured 10-11 years after sample storage and samples were never thawed. All lipid data were obtained from the baseline exam.
Clinical endpoints
Study participants were followed from 2000 to 2018, for an average of 13.4 +/− 3.52 years. Time to incident CHD events, defined as centrally adjudicated myocardial infarction, resuscitated cardiac arrest, and CHD death, were documented. Myocardial infarction was present if biomarkers were twice the upper limit of normal, Q waves were present on electrocardiogram, or if there were a combination of chest pain symptoms, ST changes on electrocardiogram, and elevated biomarkers one to two times the upper limit of normal. A diagnosis of resuscitated cardiac arrest was made in those who recovered post cardiopulmonary resuscitation. Death due to CHD was identified in individuals who either had chest pain within 72 hours, a myocardial infraction within 28 days, or a history of CHD prior to death. Additionally, the diagnosis of CHD death required ruling out any known cause of non-cardiac death. For each participant, telephone interviews were conducted at 9- to 12-month intervals to obtain data regarding interim hospitalizations, outpatient cardiovascular procedures and diagnoses, and deaths. For outpatient deaths, interviews were obtained from next of kin. Self-reported follow-up data were confirmed with medical records, autopsy reports, and death certificates. In addition to telephone interviews, data were abstracted from hospital records through trained staff, including outpatient records, lab tests, electrocardiograms, echocardiograms, catheterization reports, and other procedures and imaging. All data were then independently reviewed by two physicians, and any disagreements were adjudicated. If there were continued differences in classification, a full review committee made the final decision.
Statistical analysis
The study population was divided into four groups, as follows:
Group 1: LDL-C ≤ 100 mg/dL, Lp(a) < 50 mg/dL
Group 2: LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL
Group 3: LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL
Group 4: LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL
Normally distributed continuous variables were presented as mean and standard deviation (SD), while continuous variables with a skewed distribution were presented as median and interquartile range. Categorical variables were presented as count and percentage. Baseline characteristics of participants across the four groups were compared using a chi-square test for categorical variables, an analysis of variance test for continuous variables, and a Kruskal-Wallis test for continuous variables with a non-normal distribution. Lp(a) and LDL-C were treated as both continuous and categorical variables, with Lp(a) log transformed (LogLp(a)) when treated as a continuous variable. Elevated Lp(a) was defined as ≥ 50 mg/dL, the threshold suggested in the 2018 American Heart Association (AHA) cholesterol guideline that qualifies Lp(a) as an ASCVD risk-enhancing factor.22 Elevated LDL-C was defined as > 100 mg/dL, a commonly used threshold in primary prevention.23
A Kaplan-Meier plot was used to measure time-to-CHD across Lp(a) quintiles. Additionally, the relationship between LogLp(a) and the composite CHD endpoint was measured with a fully adjusted multivariable Cox proportional hazards model that included age, sex, race/ethnicity, BMI, HDL-C, LDL-C, hypertension, hypertension medication use, diabetes, and cigarette use.
Kaplan-Meier curves and log-rank tests were used to assess time-to-CHD across the four groups. Incidence rates for each of the four groups were calculated and reported per 1,000 person-years. Hazard ratios for the composite CHD endpoint for each of the groups were calculated using three different multivariable Cox proportional hazards models: Model 1 unadjusted; model 2 adjusted for age, sex, and race/ethnicity; and model 3 adjusted for model 2 plus BMI, HDL-C, hypertension, hypertension medication use, diabetes, and cigarette use. SAS version 9.4 and R Statistical Software, packages survival and survminer, were used to perform all statistical analyses and a 2-sided p value < 0.05 was used to determine statistical significance.
Results
Baseline characteristics of study population
The mean age of the entire cohort (52.5% female) was 61.9 years of which 36.6% were White, 12.0% were Chinese American, 28.8% were African American, and 22.7% were Hispanic race/ethnicity. Baseline characteristics of each group and total population are shown in Table 1. Overall, groups were well balanced by risk factor status. There were no significant differences in age or cigarette use among the four groups. Female sex, African American race/ethnicity, BMI, and hypertension were more prevalent in groups with Lp(a) ≥ 50 mg/dL.
Table 1.
Baseline Characteristics by Lp(a) and LDL-C groups
| Lp(a) < 50 mg/dL | Lp(a) ≥ 50 mg/dL | Total | p value | |||
|---|---|---|---|---|---|---|
| Baseline Characteristics |
LDL-C ≤ 100 mg/dL Group 1 (n=1,104) |
LDL-C > 100 mg/dL Group 2 (n=2,602) |
LDL-C ≤ 100 mg/dL Group 3 (n=130) |
LDL-C > 100 mg/dL Group 4 (n=749) |
n=4,585 | |
| Age, yrs (SD) | 61.9 (10.8) | 61.9 (10.3) | 62.2 (11.2) | 62.0 (10.1) | 61.9 (10.4) | 0.990 |
| Female (%) | 559 (50.6) | 1,314 (50.5) | 79 (60.8) | 455 (60.8) | 2,407 (52.5) | <0.001 |
| Race/ethnicity (%) | <0.001 | |||||
| White | 403 (36.5) | 1,025 (39.4) | 33 (25.4) | 217 (29.0) | 1,678 (36.6) | |
| Chinese American | 149 (13.5) | 346 (13.3) | 2 (1.5) | 52 (6.9) | 549 (12.0) | |
| African American | 296 (26.8) | 585 (22.5) | 85 (65.4) | 353 (47.1) | 1,319 (28.8) | |
| Hispanic | 256 (23.2) | 646 (24.8) | 10 (7.7) | 127 (17.0) | 1,039 (22.7) | |
| Hypertension (%) | 501 (45.4) | 1,121 (43.1) | 67 (51.5) | 378 (50.5) | 2,067 (45.1) | 0.002 |
| Diabetes (%) | 159 (14.4) | 260 (10.0) | 17 (13.1) | 98 (13.1) | 534 (11.7) | 0.001 |
| Cigarette use (%) | Former: 407 (36.9) Current: 152 (13.8) |
Former: 959 (37.0) Current: 314 (12.1) |
Former: 42 (32.8) Current: 22 (17.2) |
Former: 255 (34.2) Current: 100 (13.4) |
Former: 1,663 (36.4) Current: 588 (12.9) |
0.358 |
| BMI, kg/m2 (SD) | 27.7 (5.7) | 28.2 (5.3) | 28.3 (5.9) | 28.7 (5.6) | 28.2 (5.5) | 0.003 |
| HDL-C, mg/dL (SD) | 52.2 (17.1) | 50.1 (13.7) | 58.3 (20.1) | 52.9 (14.8) | 51.3 (15.1) | <0.001 |
| LDL-C, mg/dL (SD) | 82.8 (14.6) | 131.5 (23.1) | 84.2 (15.7) | 139.4 (27.0) | 119.7 (31.4) | <0.001 |
| Lp(a), mg/dL (IQR) | 11.1 (5.1-22.2) | 14.3 (7.4-25.6) | 71.3 (57.8-86.6) | 77.8 (61.6-98.3) | 18.0 (8.2-40.0) | <0.001 |
| LogLp(a), U (SD) | 2.3 (1.1) | 2.5 (1.0) | 4.3 (0.3) | 4.4 (0.3) | 2.8 (1.2) | <0.001 |
Values displayed are mean (standard deviation) for normally distributed continuous variables, median (interquartile range) for continuous variables with a skewed distribution, or total number (%) for categorical variables. Lp(a) = lipoprotein(a); BMI = body mass index; HDL-C = high-density lipoprotein-cholesterol; LDL-C = low-density lipoprotein-cholesterol; yrs = years; SD = standard deviation, IQR = interquartile range.
The Kaplan-Meier plot in Figure 2 displays Lp(a) quintile groups, with clear divergence in CHD risk in those belonging to the 5th quintile, corresponding to an Lp(a) > 48.2 mg/dL (log-rank test, p=0.02). In the entire cohort, LogLp(a) was independently associated with CHD risk in a fully adjusted multivariable Cox proportional hazards model that included LDL-C (HR: 1.16; 95% CI: 1.05, 1.30). CHD incident rates per 1,000 person-years for the four groups were as follows: Group 1, 5.1; Group 2, 4.5; Group 3, 8.5; and Group 4, 7.0. Figure 3 presents the Kaplan-Meier curves, illustrating the probability of survival free from CHD over an average follow up duration of 13.4 years for each of the four groups: Group 1: LDL-C ≤ 100 mg/dL, Lp(a) < 50 mg/dL; Group 2: LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL; Group 3: LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL; and Group 4: LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL. The Kaplan-Meier plot depicts shorter time to CHD events in participants with Lp(a) ≥ 50 mg/dL (log-rank test, p<0.004). Results from the Cox proportional hazards model using LDL-C ≤ 100 mg/dL and Lp(a) < 50 mg/dL (Group 1) as a reference are depicted in Table 2. Compared to the reference group (LDL-C ≤ 100 mg/dL and Lp(a) < 50 mg/dL), those with LDL-C > 100 mg/dL and Lp(a) < 50 mg/dL (Group 2) demonstrated no increase in risk for CHD events (HR: 0.92; 95% CI: 0.69, 1.21). However, those with LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 3) or LDL-C > 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 4) exhibited a statistically significant increase in risk of CHD events compared to the reference group (HR: 1.83; 95% CI: 1.02, 3.27) and (HR: 1.61; 95% CI: 1.15, 2.26), respectively in a fully adjusted model.
Figure 2. Incident Coronary Heart Disease Events According to Lp(a) Quintiles.
Kaplan-Meier curves of incident CHD events by Lp(a) quintiles. Individuals in the highest quintile have a lower probability of survival free from CHD. Lp(a) = lipoprotein(a); CHD = coronary heart disease.
Figure 3. Incident Coronary Heart Disease Events According to LDL-C and Lp(a) Groups.
Kaplan-Meier curves with 95% CI limits of incident CHD events with LDL-C ≤ 100 mg/dL and Lp(a) < 50 mg/dL used as the reference group. In individuals with an optimal LDL-C (≤ 100 mg/dL), elevated Lp(a) results in a lower probability of survival free from CHD. Lp(a) = lipoprotein(a); LDL-C = low-density lipoprotein-cholesterol; CHD = coronary heart disease.
Table 2.
Risk for Coronary Heart Disease Events According to LDL-C and Lp(a) Groups
| Risk Groups | Events/Total (%) |
Incidence Rate (per 1,000 p-yrs) |
HR (95% CI), p value (Model 1) |
HR (95% CI), p value (Model 2) |
HR (95% CI), p value (Model 3) |
|---|---|---|---|---|---|
| Group 1 (LDL-C ≤ 100 mg/dL, Lp(a) < 50 mg/dL) | 74/1104 (6.7) | 5.1 | 1.00 (reference), NA | 1.00 (reference), NA | 1.00 (reference), NA |
| Group 2 (LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL) | 158/2602 (6.1) | 4.5 | 0.88 (0.67, 1.16), 0.349 | 0.89 (0.67, 1.17), 0.391 | 0.92 (0.69, 1.21), 0.540 |
| Group 3 (LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 14/130 (10.8) | 8.5 | 1.68 (0.95, 2.97), 0.075 | 1.73 (0.97, 3.10), 0.064 | 1.83 (1.02, 3.27), 0.043 |
| Group 4 (LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 69/749 (9.2) | 7.0 | 1.38 (0.99, 1.91), 0.056 | 1.52 (1.09, 2.12), 0.014 | 1.61 (1.15, 2.26), 0.006 |
Lp(a) ≥ 50 mg/dL was associated with significant CHD risk when LDL-C was ≤ 100 mg/dL. Multivariable Cox proportional hazards model 1 unadjusted; model 2 adjusted for age, sex, race/ethnicity; and model 3 adjusted for model 2 + BMI, HDL-C, hypertension, hypertension medication use, diabetes, and cigarette use. P-yrs= person-years; HR = hazard ratio; Lp(a) = lipoprotein(a); LDL-C = low-density lipoprotein-cholesterol.
Results from the Cox proportional hazards model using LDL-C > 100 mg/dL and Lp(a) < 50 mg/dL (Group 2) or LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 3) as a reference are displayed in Table 3. Compared to those with LDL-C > 100 mg/dL and Lp(a) < 50 mg/dL (Group 2), those with LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 3) or LDL-C > 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 4) exhibited a statistically significant increase in risk of CHD events compared to the reference group (HR: 1.99; 95% CI: 1.14, 3.49) and (HR: 1.76; 95% CI: 1.31, 2.36), respectively in a fully adjusted model. However, compared to those with LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 (Group 3), those with LDL-C > 100 mg/dL and Lp(a) ≥ 50 mg/dL (Group 4) demonstrated no increase in risk for CHD events (HR: 0.88; 95% CI: 0.49, 1.58).
Table 3.
Group Comparisons of Risk for Coronary Heart Disease Events
| Risk Group | Events/Total | Reference Risk Group | Events/Total | HR (95% CI), p value |
|---|---|---|---|---|
| Group 3 (LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 14/130 | Group 2 (LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL) | 158/2602 | 1.99 (1.14, 3.49), 0.016 |
| Group 4 (LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 69/749 | Group 2 (LDL-C > 100 mg/dL, Lp(a) < 50 mg/dL) | 158/2602 | 1.76 (1.31, 2.36), <0.001 |
| Group 4 (LDL-C > 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 69/749 | Group 3 (LDL-C ≤ 100 mg/dL, Lp(a) ≥ 50 mg/dL) | 14/130 | 0.88 (0.49, 1.58), 0.672 |
Lp(a) ≥ 50 mg/dL was associated with significantly higher CHD risk irrespective of LDL-C. Multivariable Cox proportional hazards model adjusted for age, sex, race/ethnicity, BMI, HDL-C, hypertension, hypertension medication use, diabetes, and cigarette use. HR = hazard ratio; Lp(a) = lipoprotein(a); LDL-C = low-density lipoprotein-cholesterol.
Discussion
Observational and genetic epidemiology data support Lp(a) as a causal risk factor for CHD.1, 11 Whether elevated Lp(a) in the context of low LDL-C is also associated with greater CHD risk in primary prevention remains unknown. The purpose of the current study was to examine this question in a population free of clinical CHD to better understand whether elevated Lp(a) poses increased risk in those who present with optimal LDL-C in primary prevention.
The primary finding from our analysis of 4,585 MESA participants followed over the course of a median 13.4 years was that elevation of Lp(a) increased CHD risk even when LDL-C was in a range that is generally considered optimal. This risk was amplified after adjusting for traditional cardiovascular risk factors and was similar in those with optimal and elevated LDL-C. Not surprisingly, we also found that elevation of both LDL-C and Lp(a) was associated with increased CHD risk compared to the reference group. However, no increased risk of CHD was found in the group with elevated LDL-C when Lp(a) was < 50 mg/dL. Overall, our findings suggest that increased risk of CHD events is present when Lp(a) is elevated, irrespective of baseline LDL-C (even when optimal) in the primary prevention setting.
Analyses from large cohorts have demonstrated a continuous and independent relationship between Lp(a) and CHD outcomes.24 Few studies in primary prevention have examined whether this level of risk is consistent across LDL-C values, or if the risk is attenuated when LDL-C is low. The Women’s Health Study found that increased cardiovascular disease risk associated with elevated Lp(a) was no longer present when LDL-C was low.25 However, in this study, elevated LDL-C was defined as above median (121.4 mg/dL), which has less clinical relevance.25 In another study with two population cohorts, the European Prospective Investigation of Cancer-Norfolk and Copenhagen City Heart Study, Verbeek et al. investigated the relationship between Lp(a) and LDL-C on cardiovascular outcomes.26 Participants from this analysis were excluded if they were non-Caucasian or had a history of myocardial infarction or stroke.26 The authors found that in individuals with LDL-C below 2.5 mmol/L (100 mg/dL), elevation of Lp(a) above the 80th percentile was not associated with a significant increase in cardiovascular disease compared to individuals with an Lp(a) below the 80th percentile, HR of 1.11 (95% CI: 0.77 – 1.59) in European Prospective Investigation of Cancer-Norfolk and 1.08 (95% CI: 0.85-1.38) in Copenhagen City Heart Study.26 The authors replicated the analysis using an Lp(a) cut off of 50 mg/dL and had similar results.26 Contrary to the prior studies, we found in a multi-ethnic study that Lp(a) associated CHD risk persisted when LDL-C was ≤ 100 mg/dL. In the European Prospective Investigation of Cancer-Norfolk and Copenhagen City Heart Study, there were relatively few participants in the low LDL-C groups, which may explain the lack of association between elevated Lp(a) and cardiovascular events when LDL-C was below 2.5 mmol/L.26 Additionally, in the study by Verbeek et al., participants with a history of transient ischemic attack, anginal symptoms, heart failure, or cardiovascular revascularization procedures, were included.26 Finally, in the analysis by Verbeek et al., LDL-C was corrected for Lp(a) mass, which has recently been shown to be inaccurate, as the contribution of Lp(a) cholesterol is much more variable than previously thought.26
Our data have similar findings to research investigating the association between Lp(a) and LDL-C in individuals with established CHD. In a cross-sectional study of 3,449 patients diagnosed with CHD, the authors scored coronary artery lesions and found that the correlation between Lp(a) and coronary artery lesion severity was stronger when LDL-C was below 100 mg/dL compared to above 100 mg/dL.27 In a recent post hoc study from ODYSSEY Outcomes trial, the efficacy of PCSK9 inhibitor treatment was assessed in individuals presenting with the acute coronary syndrome who had an LDL-C below 70 mg/dL compared to an LDL-C above 70 mg/dL.17 The authors found that in patients with an LDL-C below 70 mg/dL, there was a benefit to PCSK9 inhibitor therapy, but only when Lp(a) was elevated (defined as above median of the study population).17 This observation may be due to the Lp(a) lowering effect of PCSK9 inhibitors, which has been independently associated with reduced cardiovascular events.28
For risk assessment in primary prevention, the American College of Cardiology (ACC)-AHA Risk Calculator can be used to assess individual 10-year cardiovascular disease risk.29 It is recommended that individuals in all risk categories receive education on positive lifestyle changes, while those with higher risk be treated with lipid-lowering therapy in addition to emphasizing a healthy lifestyle.30 Additionally, there are risk-enhancing factors that can be evaluated in those in the borderline or intermediate risk group, of which Lp(a) is one of them, that allow for a shared decision approach to initiating lipid-lowering treatment.30 Similarly, the European Atherosclerosis Society (EAS) advocates for the incorporation of Lp(a) in cardiovascular risk prediction calculators, with an additional recommendation of measuring Lp(a) in all adults at least once in a lifetime.6 The present study lends credence to Lp(a) as a risk-enhancer as individuals with an Lp(a) ≥ 50 mg/dL demonstrated greater CHD risk even when LDL-C was low. These findings are clinically important since there is equipoise related to management of individuals free of cardiovascular disease who present with an optimal LDL-C level, but an elevated Lp(a) who are not already on lipid-lowering therapy. Our findings suggest that more aggressive preventive measures may be appropriate for these individuals. Indeed, the 2022 EAS Lp(a) consensus statement endorses the intensification of other cardiovascular risk factors for those with elevated Lp(a), including management of blood pressure, diabetes, LDL-C, and lifestyle factors. Additionally, there are emerging Lp(a) specific lowering therapies that show promise; however, these therapies have not been studied in primary prevention.
Study Limitations
Our study has several limitations. First, the cohort is comprised of relatively young participants with a low number of cardiovascular events in the group with LDL-C ≤ 100 mg/dL and Lp(a) ≥ 50 mg/dL. Second, the observational nature of the study prevents any causal inferences regarding the relationship of LDL-C, Lp(a), and CHD. Third, although model adjustments were made for traditional cardiovascular risk factors, residual confounding may still exist. Lastly, selection bias due to differential loss to follow up could have occurred in this longitudinal study.
Conclusions
Elevated Lp(a) is an important risk factor in primary prevention, even when LDL-C is optimal. Thus, individuals with elevated Lp(a), despite well controlled LDL-C without the use of statin therapy, may benefit from more aggressive preventive interventions.
Acknowledgements:
The authors thank the other investigators, the staff, and the participants of the MESA study for their valuable contributions. A full list of participating MESA investigators and institutions can be found at http://www.mesa-nhlbi.org.
Financial Support
This research was supported by contracts 75N92020D00001, HHSN268201500003I, N01-HC-95159, 75N92020D00005, N01-HC-95160, 75N92020D00002, N01-HC-95161, 75N92020D00003, N01-HC-95162, 75N92020D00006, N01-HC-95163, 75N92020D00004, N01-HC-95164, 75N92020D00007, N01-HC-95165, N01-HC-95166, N01-HC-95167, N01-HC-95168 and N01-HC-95169 from the National Heart, Lung, and Blood Institute, and by grants UL1-TR-000040, UL1-TR-001079, and UL1-TR-001420 from the National Center for Advancing Translational Sciences (NCATS).
Conflict of Interest
Dr. Shapiro has participated in scientific advisory boards with the following entities: Amgen; Novartis; Novo Nordisk; and has served as a consultant for Regeneron.
Dr. Rikhi and Dr. Ashburn are supported by the National Heart, Lung, And Blood Institute of the National Institutes of Health under Award Number T32HL076132. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
The remaining authors have nothing to disclose.
References
- [1].Ference BA, Ginsberg HN, Graham I, et al. , Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel, Eur Heart J, 2017;38:2459–2472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Shapiro MD and Fazio S, Apolipoprotein B-containing lipoproteins and atherosclerotic cardiovascular disease, F1000Res, 2017;6:134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Ference BA, Graham I, Tokgozoglu L, et al. , Impact of Lipids on Cardiovascular Health: JACC Health Promotion Series, J Am Coll Cardiol, 2018;72:1141–1156. [DOI] [PubMed] [Google Scholar]
- [4].Shapiro MD and Bhatt DL, "Cholesterol-Years" for ASCVD Risk Prediction and Treatment, J Am Coll Cardiol, 2020;76:1517–1520. [DOI] [PubMed] [Google Scholar]
- [5].Gencer B, Kronenberg F, Stroes ES, et al. , Lipoprotein(a): the revenant, Eur Heart J, 2017;38:1553–1560. [DOI] [PubMed] [Google Scholar]
- [6].Kronenberg F, Mora S, Stroes ESG, et al. , Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement, Eur Heart J, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Arsenault BJ and Kamstrup PR, Lipoprotein(a) and cardiovascular and valvular diseases: A genetic epidemiological perspective, Atherosclerosis, 2022;349:7–16. [DOI] [PubMed] [Google Scholar]
- [8].Thanassoulis G, Screening for High Lipoprotein(a), Circulation, 2019;139:1493–1496. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Saeed A, Kinoush S and Virani SS, Lipoprotein (a): Recent Updates on a Unique Lipoprotein, Curr Atheroscler Rep, 2021;23:41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Spence JD and Koschinsky M, Mechanisms of lipoprotein(a) pathogenicity: prothrombotic, proatherosclerotic, or both?, Arterioscler Thromb Vasc Biol, 2012;32:1550–1551. [DOI] [PubMed] [Google Scholar]
- [11].Tsimikas S, A Test in Context: Lipoprotein(a): Diagnosis, Prognosis, Controversies, and Emerging Therapies, J Am Coll Cardiol, 2017;69:692–711. [DOI] [PubMed] [Google Scholar]
- [12].Langsted A, Nordestgaard BG and Kamstrup PR, Elevated Lipoprotein(a) and Risk of Ischemic Stroke, J Am Coll Cardiol, 2019;74:54–66. [DOI] [PubMed] [Google Scholar]
- [13].Ballantyne CM, Lipoprotein(a) and Risk for Stroke and Myocardial Infarction: Why Aren't We Screening?, J Am Coll Cardiol, 2019;74:67–69. [DOI] [PubMed] [Google Scholar]
- [14].Marcovina SM and Shapiro MD, Measurement of Lipoprotein(a): A Once in a Lifetime Opportunity, J Am Coll Cardiol, 2022;79:629–631. [DOI] [PubMed] [Google Scholar]
- [15].Grundy SM, Stone NJ, Bailey AL, et al. , 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines, Circulation, 2019;139:e1082–e1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Enas EA, Varkey B, Dharmarajan TS, et al. , Lipoprotein(a): An independent, genetic, and causal factor for cardiovascular disease and acute myocardial infarction, Indian Heart J, 2019;71:99–112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Schwartz GG, Szarek M, Bittner VA, et al. , Lipoprotein(a) and Benefit of PCSK9 Inhibition in Patients With Nominally Controlled LDL Cholesterol, J Am Coll Cardiol, 2021;78:421–433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Bild DE, Bluemke DA, Burke GL, et al. , Multi-Ethnic Study of Atherosclerosis: objectives and design, Am J Epidemiol, 2002;156:871–881. [DOI] [PubMed] [Google Scholar]
- [19].Friedewald WT, Levy RI and Fredrickson DS, Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge, Clin Chem, 1972;18:499–502. [PubMed] [Google Scholar]
- [20].Marcovina SM, Albers JJ, Scanu AM, et al. , Use of a reference material proposed by the International Federation of Clinical Chemistry and Laboratory Medicine to evaluate analytical methods for the determination of plasma lipoprotein(a), Clin Chem, 2000;46:1956–1967. [PubMed] [Google Scholar]
- [21].Tsimikas S, Fazio S, Viney NJ, et al. , Relationship of lipoprotein(a) molar concentrations and mass according to lipoprotein(a) thresholds and apolipoprotein(a) isoform size, J Clin Lipidol, 2018;12:1313–1323. [DOI] [PubMed] [Google Scholar]
- [22].Grundy SM, Stone NJ and Guideline Writing Committee for the Cholesterol, G, 2018 Cholesterol Clinical Practice Guidelines: Synopsis of the 2018 American Heart Association/American College of Cardiology/Multisociety Cholesterol Guideline, Ann Intern Med, 2019;170:779–783. [DOI] [PubMed] [Google Scholar]
- [23].Kim EJ and Wierzbicki AS, Cardiovascular prevention: Frontiers in lipid guidelines, Clin Med (Lond), 2020;20:36–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Emerging Risk Factors C, Erqou S, Kaptoge S, et al. , Lipoprotein(a) concentration and the risk of coronary heart disease, stroke, and nonvascular mortality, JAMA, 2009;302:412–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Suk Danik J, Rifai N, Buring JE, et al. , Lipoprotein(a), measured with an assay independent of apolipoprotein(a) isoform size, and risk of future cardiovascular events among initially healthy women, JAMA, 2006;296:1363–1370. [DOI] [PubMed] [Google Scholar]
- [26].Verbeek R, Hoogeveen RM, Langsted A, et al. , Cardiovascular disease risk associated with elevated lipoprotein(a) attenuates at low low-density lipoprotein cholesterol levels in a primary prevention setting, Eur Heart J, 2018;39:2589–2596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Li C, Chen Q, Zhang M, et al. , The correlation between lipoprotein(a) and coronary atherosclerotic lesion is stronger than LDL-C, when LDL-C is less than 104 mg/dL, BMC Cardiovasc Disord, 2021;21:41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Bittner VA, Szarek M, Aylward PE, et al. , Effect of Alirocumab on Lipoprotein(a) and Cardiovascular Risk After Acute Coronary Syndrome, J Am Coll Cardiol, 2020;75:133–144. [DOI] [PubMed] [Google Scholar]
- [29].Goff DC Jr., Lloyd-Jones DM, Bennett G, et al. , 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines, Circulation, 2014;129:S49–73. [DOI] [PubMed] [Google Scholar]
- [30].Michos ED, McEvoy JW and Blumenthal RS, Lipid Management for the Prevention of Atherosclerotic Cardiovascular Disease, N Engl J Med, 2019;381:1557–1567. [DOI] [PubMed] [Google Scholar]