https://orcid.org/0000-0002-2391-8397
Lipoprotein(a) [Lp(a)] is a macromolecular complex composed of low-density lipoprotein (LDL)-like particle containing apolipoprotein(b), linked by a disulphide bond to apolipoprotein(a).1 Circulating levels of Lp(a) are mainly genetically determined by the LPA gene locus, with an autosomal co-dominant pattern of inheritance. The Lp(a) is considered proatherogenic, proinflammatory, and potentially antifibrinolytic.1 Accumulating epidemiologic and genetic evidence consistently demonstrate the independent association between raised Lp(a) and cardiovascular risk, suggesting causal relationship.2 In different from statins that do not reduce Lp(a), PCSK9 monoclonal antibodies were shown to reduce up to 25% of Lp(a) concentration.2,3 Moreover, novel approaches to Lp(a)-lowering therapy are underway, involving apo(a) synthesis inhibitors that may reduce up to 70–80% of Lp(a).2
Recent guidelines advocate widescale screening of Lp(a), recommending that Lp(a) measurement should be considered at least once in each adult person’s lifetime.4 Nevertheless, at present, Lp(a) measurement is not widely adopted and significant variance exists worldwide regarding the incorporation of Lp(a) into patient care. Accordingly, our aim was to investigate the scope and patterns of Lp(a) measurement in a large health maintenance organization (HMO) and describe clinical characteristics associated with elevated Lp(a). We performed a retrospective nationwide analysis of all Lp(a) laboratory tests performed between the years 2015 and 2021 in Clalit Health Services, the largest HMO in Israel, providing healthcare for approximately 4.6 million insured members. The study was approved by the institutional ethics committee and the HMO data extraction committee.
Overall, 4539 individuals were tested nationwide for Lp(a) during study period. Current analysis was limited to the adult population ≥20 years (n = 3900). Lp(a) was measured using a particle-enhanced quantitative turbidimetric immunoassay [Tina-quant® Lipoprotein(a) Gen.2, Roche Diagnostics International Ltd]. The number of persons tested yearly increased progressively from 89 in the year 2015 to 1523 individuals in 2021 (Figure 1A). Mean age was 55 ± 15 years (39% females). Baseline (pre-treatment) LDL-cholesterol >190 mg/dL was evident in 21%. The Lp(a) level >125 nmol/L (∼50 mg/dL), an acceptable threshold for defining increased risk, was noted in 21.5% of the patients; the distribution of Lp(a) levels in study population was similar to reports from other countries (Figure 1B). Those with elevated Lp(a) (>125 vs. ≤125 nmol/L) were somewhat older (median age 58 vs. 55 years) with higher rates of hypertension (40 vs. 36%), hyperlipidaemia (85 vs. 74%), chronic kidney disease (5 vs. 3%), family history of pre-mature coronary disease (19 vs. 15%), and use of lipid-lowering medications (73 vs. 60%), but lower obesity rates (31 vs. 35%); in addition, peripheral artery disease (PAD; 11 vs. 8%), ischaemic heart disease (48 vs. 38%), and aortic valve stenosis (3.2 vs. 1.9%) were more common in those with elevated Lp(a); P < 0.05 for all comparisons. Of the patients with elevated Lp(a), 72% were treated with statin and/or ezetimibe and only 2% with PCSK9 inhibitors. Of those treated by statins, 4% were receiving low-intensity, 43% moderate-intensity, and 53% high-intensity statin. Elevated Lp(a) (>125 nmol/L) was more prevalent in patients receiving high-intensity statin (27%) than those receiving moderate intensity (23%) or low intensity (21%), P < 0.001.
Lipoprotein(a) level distribution and number of patients tested yearly. (A) Number of patients tested for lipoprotein(a) levels each year (2015–2021). (B) Distribution of lipoprotein(a) levels in the adult study population.
In a multivariable logistic regression model, prior myocardial infarction (odds ratio 1.38, 95% confidence interval 1.16–1.64), PAD (1.29, 1–1.68), peak LDL-cholesterol >190 mg/dL (1.35, 1.12–1.62), and the use of statins (1.54, 1.28–1.85) were independently associated with elevated Lp(a) levels >125 nmol/L, whereas an inverse association was seen with obesity (0.74, 0.62–0.88; Table 1). Of note, though limited and conflicting data are available, past studies have reported that saturated fatty acids consistently decrease Lp(a) concentrations5 and that diet-induced weight loss is accompanied by an increase in Lp(a) levels in obese individuals.6
Univariate and multivariate logistic regression models for the association between clinical variables and elevated lipoprotein(a) levela
| Variable | Odds ratio (univariate) | P-value | Odds ratio (multivariate) | P-value |
|---|---|---|---|---|
| Age (per 5 years) | 1.06 (1.03–1.09) | P < 0.001 | ||
| Gender (female) | 1.11 (0.95–1.30) | P = 0.177 | ||
| Obesity (BMI ≥ 30 kg/m2) | 0.84 (0.71–0.99) | P = 0.035 | 0.74 (0.62–0.88) | P < 0.001 |
| Family history of ASCVD | 1.33 (1.09–1.63) | P = 0.006 | ||
| Hypertension | 1.20 (1.02–1.40) | P = 0.024 | ||
| Baseline LDL-C >190 mg/dL | 1.48 (1.24–1.77) | P < 0.001 | 1.35 (1.12–1.62) | P = 0.001 |
| Diabetes mellitus | 0.99 (0.83–1.19) | P = 0.918 | ||
| Smoking (ever) | 1.09 (0.93–1.27) | P = 0.290 | ||
| Prior cancer | 1.24 (0.97–1.59) | P = 0.081 | ||
| Peripheral artery disease | 1.50 (1.17–1.93) | P = 0.002 | 1.29 (1.00–1.68) | P = 0.05 |
| Prior myocardial infarction | 1.53 (1.30–1.80) | P < 0.001 | 1.38 (1.16–1.64) | P < 0.001 |
| Ischaemic heart disease (overall) | 1.54 (1.32–1.80) | P < 0.001 | ||
| Prior stroke | 1.20 (0.90–1.60) | P = 0.216 | ||
| Aortic valve stenosis | 1.66 (1.05–2.63) | P = 0.031 | ||
| Creatinine ≥1.4 mg/dL | 1.61 (1.10–2.36) | P = 0.015 | ||
| Aspirin | 1.54 (1.32–1.80) | P < 0.001 | ||
| Statins | 1.75 (1.48–2.06) | P < 0.001 | 1.54 (1.28–1.85) | P < 0.001 |
| Variable | Odds ratio (univariate) | P-value | Odds ratio (multivariate) | P-value |
|---|---|---|---|---|
| Age (per 5 years) | 1.06 (1.03–1.09) | P < 0.001 | ||
| Gender (female) | 1.11 (0.95–1.30) | P = 0.177 | ||
| Obesity (BMI ≥ 30 kg/m2) | 0.84 (0.71–0.99) | P = 0.035 | 0.74 (0.62–0.88) | P < 0.001 |
| Family history of ASCVD | 1.33 (1.09–1.63) | P = 0.006 | ||
| Hypertension | 1.20 (1.02–1.40) | P = 0.024 | ||
| Baseline LDL-C >190 mg/dL | 1.48 (1.24–1.77) | P < 0.001 | 1.35 (1.12–1.62) | P = 0.001 |
| Diabetes mellitus | 0.99 (0.83–1.19) | P = 0.918 | ||
| Smoking (ever) | 1.09 (0.93–1.27) | P = 0.290 | ||
| Prior cancer | 1.24 (0.97–1.59) | P = 0.081 | ||
| Peripheral artery disease | 1.50 (1.17–1.93) | P = 0.002 | 1.29 (1.00–1.68) | P = 0.05 |
| Prior myocardial infarction | 1.53 (1.30–1.80) | P < 0.001 | 1.38 (1.16–1.64) | P < 0.001 |
| Ischaemic heart disease (overall) | 1.54 (1.32–1.80) | P < 0.001 | ||
| Prior stroke | 1.20 (0.90–1.60) | P = 0.216 | ||
| Aortic valve stenosis | 1.66 (1.05–2.63) | P = 0.031 | ||
| Creatinine ≥1.4 mg/dL | 1.61 (1.10–2.36) | P = 0.015 | ||
| Aspirin | 1.54 (1.32–1.80) | P < 0.001 | ||
| Statins | 1.75 (1.48–2.06) | P < 0.001 | 1.54 (1.28–1.85) | P < 0.001 |
ASCVD, atherosclerotic cardiovascular disease; BMI, body mass index; LDL-C, low-density lipoprotein cholesterol.
Lipoprotein(a) >125 nmol/L (∼50 mg/dL), a cutoff level defined in the literature as associated with increased cardiovascular risk.
Univariate and multivariate logistic regression models for the association between clinical variables and elevated lipoprotein(a) levela
| Variable | Odds ratio (univariate) | P-value | Odds ratio (multivariate) | P-value |
|---|---|---|---|---|
| Age (per 5 years) | 1.06 (1.03–1.09) | P < 0.001 | ||
| Gender (female) | 1.11 (0.95–1.30) | P = 0.177 | ||
| Obesity (BMI ≥ 30 kg/m2) | 0.84 (0.71–0.99) | P = 0.035 | 0.74 (0.62–0.88) | P < 0.001 |
| Family history of ASCVD | 1.33 (1.09–1.63) | P = 0.006 | ||
| Hypertension | 1.20 (1.02–1.40) | P = 0.024 | ||
| Baseline LDL-C >190 mg/dL | 1.48 (1.24–1.77) | P < 0.001 | 1.35 (1.12–1.62) | P = 0.001 |
| Diabetes mellitus | 0.99 (0.83–1.19) | P = 0.918 | ||
| Smoking (ever) | 1.09 (0.93–1.27) | P = 0.290 | ||
| Prior cancer | 1.24 (0.97–1.59) | P = 0.081 | ||
| Peripheral artery disease | 1.50 (1.17–1.93) | P = 0.002 | 1.29 (1.00–1.68) | P = 0.05 |
| Prior myocardial infarction | 1.53 (1.30–1.80) | P < 0.001 | 1.38 (1.16–1.64) | P < 0.001 |
| Ischaemic heart disease (overall) | 1.54 (1.32–1.80) | P < 0.001 | ||
| Prior stroke | 1.20 (0.90–1.60) | P = 0.216 | ||
| Aortic valve stenosis | 1.66 (1.05–2.63) | P = 0.031 | ||
| Creatinine ≥1.4 mg/dL | 1.61 (1.10–2.36) | P = 0.015 | ||
| Aspirin | 1.54 (1.32–1.80) | P < 0.001 | ||
| Statins | 1.75 (1.48–2.06) | P < 0.001 | 1.54 (1.28–1.85) | P < 0.001 |
| Variable | Odds ratio (univariate) | P-value | Odds ratio (multivariate) | P-value |
|---|---|---|---|---|
| Age (per 5 years) | 1.06 (1.03–1.09) | P < 0.001 | ||
| Gender (female) | 1.11 (0.95–1.30) | P = 0.177 | ||
| Obesity (BMI ≥ 30 kg/m2) | 0.84 (0.71–0.99) | P = 0.035 | 0.74 (0.62–0.88) | P < 0.001 |
| Family history of ASCVD | 1.33 (1.09–1.63) | P = 0.006 | ||
| Hypertension | 1.20 (1.02–1.40) | P = 0.024 | ||
| Baseline LDL-C >190 mg/dL | 1.48 (1.24–1.77) | P < 0.001 | 1.35 (1.12–1.62) | P = 0.001 |
| Diabetes mellitus | 0.99 (0.83–1.19) | P = 0.918 | ||
| Smoking (ever) | 1.09 (0.93–1.27) | P = 0.290 | ||
| Prior cancer | 1.24 (0.97–1.59) | P = 0.081 | ||
| Peripheral artery disease | 1.50 (1.17–1.93) | P = 0.002 | 1.29 (1.00–1.68) | P = 0.05 |
| Prior myocardial infarction | 1.53 (1.30–1.80) | P < 0.001 | 1.38 (1.16–1.64) | P < 0.001 |
| Ischaemic heart disease (overall) | 1.54 (1.32–1.80) | P < 0.001 | ||
| Prior stroke | 1.20 (0.90–1.60) | P = 0.216 | ||
| Aortic valve stenosis | 1.66 (1.05–2.63) | P = 0.031 | ||
| Creatinine ≥1.4 mg/dL | 1.61 (1.10–2.36) | P = 0.015 | ||
| Aspirin | 1.54 (1.32–1.80) | P < 0.001 | ||
| Statins | 1.75 (1.48–2.06) | P < 0.001 | 1.54 (1.28–1.85) | P < 0.001 |
ASCVD, atherosclerotic cardiovascular disease; BMI, body mass index; LDL-C, low-density lipoprotein cholesterol.
Lipoprotein(a) >125 nmol/L (∼50 mg/dL), a cutoff level defined in the literature as associated with increased cardiovascular risk.
Population screening for elevated Lp(a) may have the potential to improve cardiovascular disease risk prediction and help targeting Lp(a) lowering drugs to those with markedly elevated levels.7 Measurement of very high Lp(a) was shown to result in significant risk reclassification in both primary and secondary prevention settings.8 However, the magnitude of Lp(a) lowering that is required to produce clinically relevant beneficial effects is unknown,2 and proved therapies that specifically lowers Lp(a) and cardiovascular events are still under investigation. Furthermore, PCSK9 inhibitors which have the potential to reduce both LDL-cholesterol and to a lesser extent Lp(a) are vastly underused in patients with raised Lp(a), as demonstrated in the present analysis.
Although the frequency of Lp(a) measurement is increasing, the overall number of patients tested for Lp(a) is considerably low, with only 0.1% of the insured population tested in recent years by the largest HMO in Israel. Many hospitals and laboratories do not provide Lp(a) testing, and therefore tests are delivered to central laboratories, with the need for reimbursement. Even in patient populations which were targeted by earlier guidelines for Lp(a) measurement, the performance of tests for Lp(a) has been ‘rather an exception than a rule’.9 A recent report addressing public health aspects regarding Lp(a) suggests that for Lp(a) measurement to be more widely used, key barriers need to be addressed including: (i) lack of awareness on Lp(a), (ii) the perception that this measure may have limited clinical value, (iii) lack of data on the cardiovascular benefit of reducing Lp(a), and (iv) methodologies for measuring Lp(a) that address both standardization and harmonization.10 Laboratories reporting Lp(a) levels should mark with a ‘red flag’ elevated Lp(a) test results that are associated with increased risk in order to raise awareness of both patients and healthcare providers.
In conclusion, the reporting of Lp(a) level in real-life setting is increasing, though it is still measured in only a minority of the population. Presence of atherosclerotic cardiovascular disease and severe hypercholesterolaemia may identify patients who are more likely to be detected with elevated Lp(a) when tested; at present, the use of PCSK9 inhibitors in these patients is noticeably low. Implementation of recent recommendations for widescale Lp(a) measurement is challenging and may require a paradigm shift in the diagnosis and management of dyslipidaemias. Although evidence regarding the cardiovascular benefits of Lp(a) reduction is yet lacking, measurement of Lp(a) may help identify higher risk individuals in need for more aggressive preventive management.
Author contributions
B.Z. and A.A. contributed to the conception or design of the work. B.Z., A.A., and W.S. contributed to the acquisition, analysis, or interpretation of data for the work. B.Z. drafted the manuscript. All authors critically revised the manuscript, gave final approval, and agree to be accountable for all aspects of work ensuring integrity and accuracy.
Funding
None declared.
Data availability
The datasets generated during the current study are not publicly available due to privacy regulations. Data will be shared on reasonable request to the corresponding author.
References
Author notes
Conflict of interest: None declared.
Comments