Prognostic value of elevated lipoprotein (a) in patients with acute coronary syndromes: a systematic review and meta-analysis

Background Elevated lipoprotein (a) level was recognized as an independent risk factor for significant adverse cardiovascular events in acute coronary syndrome (ACS) patients. Despite this recognition, the consensus in the literature regarding the prognostic significance of elevated lipoprotein (a) in ACS was also limited. Consequently, we conducted a thorough systematic review and meta-analysis to evaluate the prognostic relevance of elevated lipoprotein (a) level in individuals diagnosed with ACS. Methods and results A thorough literature review was conducted by systematically searching PubMed, Embase, and Cochrane databases until September 2023. This review specifically examined cohort studies exploring the prognostic implications of elevated lipoprotein (a) level in relation to major adverse cardiovascular events (MACE), including death, stroke, non-fatal myocardial infarction (MI), and coronary revascularization, in patients with ACS. The meta-analysis utilized aggregated multivariable hazard ratios (HR) and their respective 95% confidence intervals (CI) to evaluate prognostic implications between high and low lipoprotein (a) levels [the cut-off of high lipoprotein (a) level varies from 12.5 to 60 mg/dl]. Among 18,168 patients in the identified studies, elevated lipoprotein (a) was independently associated with increased MACE risk (HR 1.26; 95% CI: 1.17–1.35, P < 0.00001) and all-cause mortality (HR 1.36; 95% CI: 1.05–1.76, P = 0.02) in ACS patients. In summary, elevated lipoprotein (a) levels independently forecast MACE and all-cause mortality in ACS patients. Assessing lipoprotein (a) levels appears promising for risk stratification in ACS, offering valuable insights for tailoring secondary prevention strategies. Systematic Review Registration PROSPERO (CRD42023476543).


Introduction
Acute coronary syndromes (ACS) encompasses ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (NSTEMI), and unstable angina (UA) (1)(2)(3)(4).They are common in older adults (5) , and yet the incident has increased in younger people (6, 7).Annually, over 7 million individuals worldwide receive ACS diagnoses (8,9).Roughly 5% succumb before hospital discharge (2,(8)(9)(10).Subsequent ACS survivors frequently face major adverse cardiovascular events (MACE), which encompass recurrent ischemic incidents and mortality (11)(12)(13).Despite advancements in medical therapy, the one-year incidence of MACE post-ACS has risen to 9.2% (14).Identifying a predictive indicator for MACE after ACS is imperative for improved prognostic outcomes.Recently, lipoprotein(a) [Lp (a)] has emerged as an independent risk factor linked to MACE following ACS (15)(16)(17)(18)(19). Nevertheless, conflicting evidence surrounds the prognostic significance of elevated blood Lp (a) levels in ACS patients.For instance, the FORTIAM investigation (20), a multicenter cohort study in Spain involving 1,371 acute myocardial infarction (AMI) patients across 15 hospitals, revealed a poorer prognosis in those with elevated Lp (a) levels at admission.Consistent findings were reported by Andrea Kallmeyer et al. and Si-qi Yang et al. (21,22).Conversely, a Vietnam-based observational cohort study (23) yielded disparate results, suggesting no correlation between Lp (a) levels ≥ 50 mg/ dl during AMI and MACE or all-cause mortality.No comprehensive systematic review or meta-analysis has yet assessed the prognostic significance of elevated Lp (a) levels in ACS patients regarding MACE and all-cause mortality.This meta-analysis aims to investigate the prognostic relevance of baseline blood lipoprotein(a) levels in predicting MACE and allcause mortality among ACS patients.

Identification of eligible studies
Eligible studies met the following criteria: (1) utilization of randomized controlled, cohort, or case-control designs; (2) examination of adults with diagnosed ACS; (3) primary focus on comparing MACE and/or all-cause mortality between ACS patients with high and low Lp (a) levels; (4) assessment of at least one MACE (e.g., death, stroke, non-fatal MI, or coronary revascularization) with concurrent evaluation of all-cause mortality; and (5) availability of sufficient data to compute the Hazard Ratio (HR).Exclusions comprised reviews, letters, editorial comments, case reports, conference abstracts, pediatric articles, unpublished works, and non-English publications.
According to ACS criteria, the syndrome comprises UA, NSTEMI, and STEMI.Therefore, studies that focused on stable angina, chronic coronary artery disease (CAD) were also excluded.

Data extraction
Two independent investigators (G.C.W and M.Y.X) performed data extraction.Discrepancies were resolved through intervention by a third investigator (S.T.L or D.X.J) to establish consensus.Extracted data from the studies included primary author, publication year, study duration, location, design, ACS types, sample size, follow-up duration, participant age, body mass index (BMI), high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c) levels, history of prior medical therapy, diabetes prevalence, Lp (a) threshold, MACE, and mortality-related outcomes.For studies reporting continuous variables as median with range or interquartile range, a validated mathematical approach was employed to derive mean and standard deviation (21)(22)(23)(24)(25)(26)(27)(28).

Quality assessment
Study quality was assessed using the Newcastle-Ottawa Scale (NOS) (29), categorizing studies scoring seven to nine points as high quality (30).Two independent reviewers evaluated evidence quality in eligible studies, resolving discrepancies through consensus discussions.

Statistical analysis
Evidence synthesis was performed using Review Manager version 5.4.1 (Cochrane Collaboration, Oxford, UK).Survival variables were assessed using Hazard Ratios (HR), and metrics were presented with corresponding 95% confidence intervals (CIs).Heterogeneity assessment utilized the inconsistency index (I 2 ) (31), where I 2 values exceeding 50% indicate significant heterogeneity.In instances of significant heterogeneity (I 2 > 50%), a random-effects model estimated combined HR; otherwise, a fixed-effect model was applied (31).Furthermore, we conducted one-way sensitivity analyses to assess the influence of included studies on combined outcomes showing substantial heterogeneity.Publication bias was evaluated visually using funnel plots generated with Review Manager 5.4.1 (Cochrane Collaboration, Oxford, UK) and statistically through Egger's regression tests (32) implemented with Stata version 15.0 (Stata Corp, College Station, TX, USA) for outcomes with 10 or more included studies.A P-value < 0.05 indicated statistically significant publication bias.

Literature search and study characteristics
The schematic representation of the methodical search and selection procedure is depicted in Figure 1.A comprehensive literature search across PubMed (n = 1,165), Embase (n = 1,713), and Cochrane (n = 172) databases identified a total of 3,050 pertinent articles.Following the elimination of duplicate publications, a meticulous review of titles and abstracts resulted in the inclusion of 2024 articles for further consideration.Ultimately, 18 full-text articles, encompassing a total of 18,168 patients (13,843 male, 4,325 female), were selected for the comprehensive pooled analysis (20-23, 25-28, [33][34][35][36][37][38][39][40][41][42].Of these articles, 5 were prospective cohort studies (21,23,34,35,40), 10 were retrospective cohort studies (20, 22, 25, 26, 28, 33, 36-39), and 3 were observational cohorts (4,27,41).The encompassed investigations in this analysis spanned the period from 2009 to 2023.The sample sizes exhibited considerable variability, ranging from 88 to 2007 participants, resulting in a cumulative cohort of 18,168 patients.The duration of follow-up varied notably, extending from 2.93 to 66 months.It is noteworthy that the defined cutoff value for Lp (a) demonstrated heterogeneity across the studies incorporated in this review.Table 1 comprehensively outlines the distinctive characteristics, level of evidence, and quality assessments of each included study.The median (range) quality score, indicative of methodological rigor, was determined to be 7 (5)(6)(7)(8)(9).Notably, 14 studies were acknowledged for their high quality, indicating the strength of their methodologies and findings within the scope of this comprehensive analysis (21, 23, 26-28, 33-40, 42).The quality assessment details of all eligible are presented in Supplementary Table S3.

Impact of elevated lipoprotein (a) on major adverse cardiovascular events
Fourteen studies (20-23, 25, 26, 28, 33, 36-41) investigated the prognostic significance of elevated Lp (a) levels in relation to MACE (Figure 2A).The meta-analysis utilizing a random-effects model revealed a pooled HR of 1.26 (95% CI: 1.17-1.35)for high compared to low category of Lp (a) levels, demonstrating a statistically significant association.However, notable heterogeneity was observed (I 2 = 88%; P < 0.00001).A visual inspection of the funnel plot did not reveal any significant publication bias (Figure 2B).Nevertheless, Egger's test yielded a statistically significant result (P < 0.0001).Sensitivity analyses were performed by excluding two studies, both conducted by the same author (26, 28).The removal of these studies resulted in a discernible impact on the original pooled effect sizes of MACE, suggesting that the observed heterogeneity may be attributed to these specific studies.

Impact of elevated lipoprotein (a) on allcause mortality
Seven studies (22,23,25,27,33,34,40) were included in the analysis to investigate the relationship between elevated Lp (a) levels and the risk of all-cause mortality (Figure 3A).Utilizing a random-effects model, the pooled Hazard Ratio (HR) for allcause mortality was 1.36 (95% CI: 1.05-1.76)when comparing high to low category of Lp (a) levels.There was no significant heterogeneity observed (I 2 = 49%; P = 0.02).Examination of the funnel plot revealed no significant publication bias (Figure 3B), and Egger's test did not show statistical significance (P = 0.518).Sensitivity analyses, involving the removal of each study one at a time, demonstrated minimal alterations to the original pooled effect sizes for all-cause mortality.This supported the stability of the results.In summary, the comprehensive analysis of these seven studies indicates a statistically significant association between elevated Lp (a) levels and an increased risk of all-cause mortality.The findings are robust, as evidenced by the lack of publication bias, low heterogeneity, and consistent results in sensitivity analyses.

Subgroup analysis
In addition, we conducted a subgroup analysis to assess the correlation between elevated Lp (a) levels and the occurrence of MACE, as well as all-cause mortality specifically within the subgroup of patients diagnosed with ACS.We divided the patients into two subgroups according to the study design (prospective cohort vs. retrospective cohort), follow-up duration (≥3 years vs. < 3 years), Lp (a) level (Lp (a) > 30 mg/dl vs. Lp (a) threshold ≤30 mg/dl), area (Asia vs. Europe), and divided into 3 subgroups according to the types of ACS (ACS vs. AMI vs. STEMI).The findings from the subgroup analysis reveal a strong correlation between elevated Lp (a) levels and an increased risk of MACE across all subgroups, with the exception of the STEMI subgroup (HR 1.02, 95% CI: 1.00-1.04,P = 0.02), but the difference of all-cause mortality was not statistically significant in subgroup with prospective cohort (HR 0.94 95%CI: 0.50-1.78,P = 0.86), follow-up duration < 3 years (HR 1.23 95%CI: 0.82-1.86,P = 0.32), Europe area (HR 1.34 95%CI: 0.72-2.49,P = 0.36), and Lp (a) threshold > 30 mg/dl (HR 0.84 95%CI: 0.36-1.95,P = 0.69).

Discussion
ACS stands as a predominant contributor to global morbidity and mortality, acting as a primary catalyst for coronary heart disease-associated hospitalizations and fatalities, thereby imposing a substantial disease burden (43).Lp (a), comprising apolipoprotein (a), apolipoprotein B-100, and lipid components  Frontiers in Cardiovascular Medicine like cholesterol, phospholipids, and triglycerides (44), has been identified as a promoter of atherosclerosis, inflammation, and thrombosis, emerging as an independent risk factor for ACS (45,46).The 2019 ESC/EAS guidelines for dyslipidemia management advocate for at least one Lp (a) level assessment for every adult during their lifetime (47).The determination of Lp (a) levels may aid in identifying patients necessitating more intensive therapeutic interventions in ACS.Future investigations are imperative to discern whether reducing Lp (a) levels can confer cardiovascular benefits to ACS patients.The precise mechanisms driving Lp (a)'s predictive value in ACS patients remain enigmatic.Plausibly, Lp (a) exhibits dual pathogenicity, manifesting proinflammatory and antifibrinolytic effects (48).An alternative rationale suggests elevated Lp (a) may compromise endothelial and anticoagulant functions by promoting endothelial dysfunction and increasing phospholipid oxidation (49,50).
Ongoing research delves into the potential role of Lp (a) in risk stratification and residual risk modification (51), prompting the present meta-analysis.The primary findings of this meta-analysis underscore that an elevated Lp (a) level serves as an independent predictor of MACE and all-cause mortality in ACS patients.Individuals with high Lp (a) levels exhibited approximately 26% and 36% higher risks of MACE (HR 1.26, 95% CI: 1.17-1.35,P < 0.00001) and all-cause mortality (HR 1.36, 95% CI: 1.05-1.76,P = 0.02), respectively, compared to those with low Lp (a) levels.Notably, the findings in subgroup with STEMI for MACE were not significantly different in this paper, which may be due to the shorter follow-up periods in studies with STEMI.In addition, most of the studies about STEMI were large-scale and retrospective.Furthermore, the inclusion criteria varied slightly according to different studies (e.g., a Japan-based study in 2019 (28) included patients with index STEMI, while others did not make the declaration).On the other hand, the different results in subgroup analysis for all-cause mortality suggested that the prognostic value of Lp (a) level for all-cause mortality is associated with the study design, follow-up duration, and patients' races (from different areas and with various Lp (a) levels).It is presumed that the results in subgroup analysis did not show statistical significance, maybe due to the smaller samples with longer follow-up durations (n = 4,885 for <3 years vs. n = 1,475 for ≥3 years), which suggests that the prognostic value of high Lp (a) levels for all-cause mortality may not be present in the short term.The differences in Lp (a) thresholds among various areas may result from genetic racial differences (52).All of which may lead to the differences among studies.This meta-analysis represents a groundbreaking investigation, confirming Lp (a)'s prognostic significance in predicting MACE and all-cause mortality among ACS patients.However, some evidence demonstrates that cathepsin s, soluble LOX-1, and LDL-electronegativity, which have been implicated in the pathogenesis of atherosclerotic cardiovascular disease, have been related with prognosis in patients with ACS (54-56).All of these biomarkers, elevated during ACS, not measured and analysed in the included studies, could be associated with worse outcomes in ACS patients, may limit the interpretation of results.Additionally, several considerations merit attention.Firstly, the absence of individuallevel data may introduce variability in pooled outcomes based on patient characteristics.Secondly, diverse cutoff values for elevated blood Lp (a) levels across studies hinder the establishment of a standardized threshold for Lp (a) elevation.Thirdly, the heterogeneity in pooled outcomes may stem from ACS subgroup, study design, follow-up duration, Lp (a) measurement methods, and cutoff values.The prognostic role of Lp (a) using continuous data couldn't be assessed due to insufficient information.Additionally, the exclusion of randomized controlled trials was

Conclusions
This meta-analysis underscores that elevated Lp (a) levels independently predict MACE and all-cause mortality in ACS patients.Because of minimal influence from lifestyle, diet, and medical interventions, Lp (a) measurement has the potential to enhance risk stratification in ACS.Future well-designed studies are crucial to investigate variations in Lp (a) prognostic significance among ACS subgroups.Furthermore, these investigations should conclusively establish Lp (a)'s role in refining risk stratification and modifying residual risk in ACSdiagnosed individuals.

FIGURE 1 Flowchart
FIGURE 1Flowchart of the systematic search and selection process.
necessary due to data limitations.Moreover, besides cathepsin s, soluble LOX-1, and LDL-electronegativity, the ACS, as acute inflammation, has potential to increase the level of Lp (a) and thus affect its measurement.Considering the potential confounders and above limitations, results of the present pooled analysis should be interpreted with caution.In spite of several limitations of our study, we confirmed that an elevated Lp (a) level is an independent predictor of MACE and all-cause mortality in ACS patients.The funnel plots and Egger's test examination revealed no significant publication bias, which makes our results more robust.Our evidencebased analysis validated previous studies reporting the different MACE outcomes in high level Lp (a) vs. low level patients with ACS(21, 25, 28,(36)(37)(38).More well-designed, large-scale prospective randomized studies with longterm follow-up are needed to further compare the MACE, allcause mortality in high level vs. low level of Lp (a) patients with ACS.

FIGURE 2 (
FIGURE 2(A) Forest plots of major adverse cardiovascular events outcomes.(B) Funnel plots of major adverse cardiovascular events outcomes.

TABLE 1
Baseline characteristics of include studies and quality assessment.