Abstract
This cross-sectional study investigates the association between lipoprotein(a) levels and risk of ventricular arrhythmias among patients with heart failure with reduced ejection fraction receiving implantable cardioverter-defibrillator therapy.
Introduction
The genetically determined lipoprotein(a) [Lp(a)] is associated with atherosclerotic cardiovascular disease and adverse myocardial remodeling.1,2,3 Elevated Lp(a) has been associated with incident heart failure and adverse cardiovascular outcomes,2,4,5 and myocardial fibrosis–related phenotypes.6 In patients with heart failure with reduced ejection fraction (HFrEF), however, it is not well established whether elevated Lp(a) identifies a subgroup at increased risk of ventricular tachyarrhythmias requiring implantable cardioverter-defibrillator (ICD) therapy. We evaluated the association between elevated Lp(a) and treated ventricular arrhythmias among adults with HFrEF undergoing ICD implantation.
Methods
This retrospective, multicenter, cross-sectional study across 3 US tertiary referral centers included adults aged 18 years or older with left ventricular ejection fraction of 40% or less who underwent ICD implantation for primary or secondary prevention between January 1, 1998, and January 1, 2025. Patients had to have at least 1 documented Lp(a) measurement (using immunoturbidimetric assays and categorized as <50 mg/dL vs ≥50 mg/dL) and ICD follow-up (to convert Lp(a) to milligrams per liter, multiply by 0.1). This study is reported following the STROBE reporting guideline and was exempt from Mayo Clinic Institutional Review Board review and informed consent because of the use of deidentified patient data..
The primary outcome was time to first appropriate ICD shock for ventricular tachyarrhythmias. Secondary outcomes were all-cause mortality and time to first appropriate ICD therapy (first occurrence of appropriate shock or antitachycardia pacing [ATP], with subsequent device therapy episodes not considered in this time-to-first-event analysis).
Baseline characteristics were compared using Wilcoxon rank-sum, χ2, or Fisher exact tests, as appropriate. Time-to-event outcomes were measured from ICD implantation to the first occurrence of the outcome or last available follow-up, with deaths before an arrhythmic event censored at death. We used unadjusted and multivariable Cox proportional hazards models for the primary outcome and analyzed secondary time-to-event outcomes using multivariable Cox proportional hazards models. ATP alone was not treated as an event or censoring event in the primary analysis. To assess overall arrhythmic burden beyond the first event, shock burden was evaluated using negative binomial regression with follow-up time as an offset. The level of statistical significance was deemed as a 2-sided P value < .05. Analyses were performed in R version 4.3.0 (R Project for Statistical Computing) from August until December 2025.
Results
Among 595 patients (median [IQR] age, 69 [58-77] years; 76.0% male), Lp(a) levels were less than 50 mg/dL and 50 mg/dL or greater in 405 and 190 patients, respectively. Median (IQR) follow-up was 2601 (1120-4560) days, approximately 7.1 years (Table 1).
Table 1. Baseline Characteristics.
| Characteristic | Patients, No. (%) (N = 595) | P value | |
|---|---|---|---|
| Lp(a) <50 mg/dL (n = 405) | Lp(a) ≥50 mg/dL (n = 190) | ||
| Demographics | |||
| Age, median (IQR), y | 69 (58-77) | 69 (57-77) | .74 |
| Sex | |||
| Female | 98 (24.2) | 45 (23.7) | .92 |
| Male | 307 (75.8) | 145 (76.3) | |
| Comorbidities | |||
| Diabetes | 167 (41.2) | 99 (52.1) | .01 |
| Hypertension | 166 (41.0) | 79 (41.6) | .93 |
| Ischemic cardiomyopathy | 253 (62.5) | 111 (58.4) | .37 |
| Obesity | 101 (24.9) | 54 (28.4) | .37 |
| Chronic kidney disease | 86 (21.2) | 64 (33.7) | .002 |
| Dyslipidemia | 329 (81.2) | 158 (83.2) | .65 |
| Sleep apnea | 117 (28.9) | 71 (37.4) | .047 |
| Thyroid disease | 62 (15.3) | 43 (22.6) | .04 |
| Atrial fibrillation or flutter | 225 (55.6) | 139 (73.2) | <.001 |
| History of stroke | 84 (20.7) | 55 (28.9) | .023 |
| Medications | |||
| β-Blocker | 338 (83.5) | 156 (82.1) | .73 |
| ACE inhibitor | 307 (75.8) | 149 (78.4) | .53 |
| Mineralocorticoid receptor antagonist | 203 (50.1) | 97 (51.1) | .86 |
| SGLT2 inhibitor | 65 (16.0) | 28 (14.7) | .72 |
| Loop diuretic | 308 (76.0) | 141 (74.2) | .68 |
| Amiodarone | 209 (51.6) | 102 (53.7) | .66 |
| Sotalol | 44 (10.9) | 22 (11.6) | .78 |
| Mexiletine | 45 (11.1) | 19 (10.0) | .78 |
| Dofetilide | 17 (4.2) | 6 (3.2) | .65 |
| Statin | 353 (87.2) | 165 (86.8) | .90 |
| Antiplatelet therapy | 330 (81.5) | 153 (80.5) | .82 |
| Apixaban | 99 (24.4) | 44 (23.2) | .76 |
| Rivaroxaban | 20 (4.9) | 9 (4.7) | >.99 |
| Dabigatran | 4 (1.0) | 0 | .31 |
| Warfarin | 194 (47.9) | 95 (50.0) | .66 |
| Clinical data, median (IQR) | |||
| LDL, mg/dL | 133 (107-164) | 144 (110-170) | .09 |
| CRP, mg/L | 6 (1-20) | 10 (4-25) | .07 |
| NT-proBNP, pg/mL | 3094 (1101-8455) | 2348 (707-8689) | .18 |
| HbA1c, % | 5.80 (5.40-6.70) | 5.70 (5.40-6.60) | .43 |
| Creatinine, mg/dL | 0.80 (0.66-0.90) | 0.80 (0.70-1.00) | .03 |
| EF, % | 25 (20-30) | 24 (19-30) | .05 |
| ICD purpose | |||
| Primary prevention | 351 (86.7) | 158 (83.2) | .26 |
| Secondary prevention | 54 (13.3) | 32 (16.8) | .26 |
| NYHA class | |||
| I | 74 (18.3) | 25 (13.2) | .31 |
| II | 189 (46.7) | 95 (50.0) | |
| III | 110 (27.2) | 49 (25.8) | |
| NYHA class IV | 28 (6.9) | 18 (9.5) | |
Abbreviations: ACE, angiotensin-converting enzyme; CRP, C-reactive protein; EF, ejection fraction; HbA1c, hemoglobin A1c; ICD, implantable cardioverter-defibrillator; LDL, low-density lipoprotein; Lp(a), lipoprotein(a); NT-proBNP, N-terminal pro-B-type natriuretic peptide; NYHA, New York Heart Association; SGLT2, sodium-glucose cotransporter 2.
SI conversion factors: To convert creatinine to micromoles per liter, multiply by 88.4; CRP to milligrams per liter, multiply by 10; HbA1c to proportion of total hemoglobin, multiply by 0.01; LDL to millimoles per liter, multiply by 0.0259; Lp(a) to milligrams per liter, multiply by 0.1.
Appropriate ICD shocks occurred in 49 patients (25.8%) with elevated Lp(a) and 39 patients (9.6%) with lower Lp(a) (unadjusted hazard ratio [HR], 3.10; 95% CI, 2.03-4.72; P < .001; adjusted HR [aHR], 2.95; 95% CI, 1.92-4.52; P < .001). No interaction was observed by ischemic vs nonischemic cardiomyopathy, whereas there was an association between elevated Lp(a) and appropriate shock among recipients of primary-prevention ICD (Table 2).
Table 2. Adjusted Associations With First Appropriate ICD Shock and ICD Therapy.
| Variable | Outcome 1 | Outcome 2 |
|---|---|---|
| First appropriate ICD, aHR (95% CI)a | ||
| Lp(a) ≥50 mg/dL | 2.95 (1.92-4.52) | 2.37 (1.65-3.40) |
| Age, per 1-year increase | 0.99 (0.98-1.00) | 0.99 (0.98-1.00) |
| Male sex | 1.43 (0.83-2.47) | 0.99 (0.64-1.52) |
| Ischemic cardiomyopathy | 0.81 (0.51-1.29) | 0.72 (0.49-1.07) |
| β-Blocker therapy | 1.39 (0.53-3.62) | 1.62 (0.72-3.63) |
| ACE inhibitor or ARB | 1.12 (0.52-2.44) | 1.04 (0.56-1.93) |
| Mineralocorticoid receptor antagonist | 1.06 (0.66-1.70) | 0.90 (0.61-1.35) |
| SGLT2 inhibitor | 0.91 (0.50-1.66) | 0.93 (0.56-1.54) |
| Antiarrhythmic therapy | 1.29 (0.76-2.17) | 1.67 (1.06-2.64) |
| Statin therapy | 0.85 (0.39-1.85) | 0.75 (0.40-1.38) |
| NYHA class III-IV, vs I-II | 2.33 (1.51-3.59) | 1.91 (1.32-2.75) |
| Minimum LVEF, per 1 percentage point increase | 0.97 (0.95-1.00) | 0.97 (0.95-0.99) |
| ICD primary vs secondary prevention | 1.34 (0.71-2.50) | 1.58 (0.96-2.60) |
| Interaction analyses for first appropriate ICD shock, No. b | ||
| Ischemic cardiomyopathy | ||
| No | 231 | 35 |
| Yes | 363 | 56 |
| ICD indication | ||
| Primary prevention | 508 | 76 |
| Secondary prevention | 86 | 15 |
Abbreviations: ACE, angiotensin-converting enzyme; aHR, adjusted hazard ratio; ARB, angiotensin II receptor blocker; ICD, implantable cardioverter-defibrillator; Lp(a), lipoprotein(a); LVEF, left ventricular ejection fraction; NYHA, New York Heart Association; SGLT2, sodium-glucose cotransporter 2.
SI conversion factor: To convert Lp(a) to milligrams per liter, multiply by 0.1.
The aHRs reflect associations with time to first appropriate ICD shock (outcome 1) or time to first appropriate ICD therapy (outcome 2) during follow-up. Adjusted for age, sex, ischemic cardiomyopathy, β-blocker therapy, ACE inhibitor or ARB, mineralocorticoid receptor antagonist, SGLT2 inhibitor, antiarrhythmic therapy, statin therapy, NYHA class, minimum LVEF, and ICD indication.
Outcomes are number of total patients (outcome 1) and events (outcome 2). P values for interaction were .23 for ischemic cardiomyopathy and .001 for ICD indication.
For the composite outcome of appropriate shock or ATP, events occurred in 60 patients (31.6%) with elevated Lp(a) and 62 patients (15.3%) with lower Lp(a) (aHR, 2.37; 95% CI, 1.65-3.40; P < .001). Elevated Lp(a) was associated with an increased rate of appropriate shocks (incidence rate ratio [iRR], 2.74; 95% CI, 1.48-5.21; P < .001), whereas among patients with at least 1 shock, recurrent shock burden did not differ significantly (iRR, 1.05; 95% CI, 0.72-1.53; P = .80). All-cause mortality was not significantly different between groups (aHR, 0.80; 95% CI, 0.60-1.08; P = .15).
Discussion
In this multicenter cohort study, elevated Lp(a) was associated with increased risk of treated ventricular arrhythmias, including ATP-treated events not resulting in shock, but did not translate into higher mortality. This may reflect effective termination of malignant arrhythmias by ICD therapy and the contribution of competing nonarrhythmic causes of death in HFrEF.
Our findings extend prior literature associating Lp(a) with incident heart failure, heart failure progression, and adverse cardiovascular outcomes.7,8,9 Although fibrosis data were not available, prior studies suggested that elevated Lp(a) may be associated with adverse myocardial remodeling, fibrosis, and inflammatory pathways that could contribute to arrhythmogenesis.6 The lack of interaction by ischemic cardiomyopathy importantly suggests that this association is not solely mediated by overt ischemic disease. Similarly, the association among recipients of primary-prevention ICD may reflect biomarker-associated risk before a fixed arrhythmogenic substrate is established, whereas in secondary-prevention patients the baseline arrhythmic substrate may attenuate the incremental prognostic value of Lp(a).
Limitations include retrospective design, residual confounding, and intersite variation in ICD programming. Similarly, because the analytic cohort was identified through database querying for complete inclusion criteria rather than stepwise exclusion from a broader source population, we could not compare included and nonincluded patients; therefore, selection bias related to missing variables cannot be excluded. These findings suggest that elevated Lp(a) may identify an arrhythmia-prone phenotype in HFrEF and support further prospective studies.
Data Sharing Statement
References
- 1.Reyes-Soffer G, Ginsberg HN, Berglund L, et al. ; American Heart Association Council on Arteriosclerosis, Thrombosis and Vascular Biology; Council on Cardiovascular Radiology and Intervention; and Council on Peripheral Vascular Disease . Lipoprotein(a): a genetically determined, causal, and prevalent risk factor for atherosclerotic cardiovascular disease: a scientific statement From the American Heart Association. Arterioscler Thromb Vasc Biol. 2022;42(1):e48-e60. doi: 10.1161/ATV.0000000000000147 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wong ND, Fan W, Hu X, et al. Lipoprotein(a) and long-term cardiovascular risk in a multi-ethnic pooled prospective cohort. J Am Coll Cardiol. 2024;83(16):1511-1525. doi: 10.1016/j.jacc.2024.02.031 [DOI] [PubMed] [Google Scholar]
- 3.Sheashaa H, Mousa H, Abbas MT, et al. Interaction between lipoprotein(a) and other lipid molecules: a review of the current literature. Biomolecules. 2025;15(2):162. doi: 10.3390/biom15020162 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Girard AS, Paulin A, Manikpurage HD, et al. Impact of lipoprotein(a) on valvular and cardiovascular outcomes in patients with calcific aortic valve stenosis. J Am Heart Assoc. 2025;14(6):e038955. doi: 10.1161/JAHA.124.038955 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Masson W, Barbagelata L, Lavalle-Cobo A, Corral P, Nogueira JP. Lipoprotein(a) and heart failure: a systematic review. Heart Fail Rev. 2023;28(6):1307-1314. doi: 10.1007/s10741-023-10333-2 [DOI] [PubMed] [Google Scholar]
- 6.Chehab O, Abdollahi A, Whelton SP, et al. Association of lipoprotein(a) levels with myocardial fibrosis in the multi-ethnic study of atherosclerosis. J Am Coll Cardiol. 2023;82(24):2280-2291. doi: 10.1016/j.jacc.2023.10.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nomura S, Guan W, Zhang Y, et al. Lipoprotein(a) and heart failure among Black and White participants in Atherosclerosis Risk in Communities Study, Framingham Offspring Study, and Multi-Ethnic Study of Atherosclerosis: the pooling project. J Am Heart Assoc. 2025;14(11):e038608. doi: 10.1161/JAHA.124.038608 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Steffen BT, Duprez D, Bertoni AG, Guan W, Tsai MY. Lp(a) [lipoprotein(a)]-related risk of heart failure is evident in Whites but not in other racial/ethnic groups. Arterioscler Thromb Vasc Biol. 2018;38(10):2498-2504. doi: 10.1161/ATVBAHA.118.311220 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Yadalam AK, Gangavelli A, Razavi AC, et al. Lipoprotein(a) levels and adverse outcomes in heart failure. J Card Fail. 2026;32(2):382-390. doi: 10.1016/j.cardfail.2025.03.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Sharing Statement
