Inflammation
Elevated Lp(a) independently predicts incident atrial fibrillation regardless of inflammatory status, 365,899-participant UK Biobank study finds (Heart Rhythm 2025)
Original title: Lipoprotein(a) elevation independently associates with incident atrial fibrillation irrespective of inflammatory status
This retrospective cohort study analysed 365,899 UK Biobank participants without baseline atrial fibrillation (AF), testing whether Lp(a)'s association with incident AF is independent of systemic inflammation (defined as hs-CRP of 2 mg/L or above). Over a median 13.5-year follow-up, 25,048 participants (6.8%) developed AF. Both elevated Lp(a) (7.44% vs. 6.77% incidence, log-rank P < .001) and elevated hs-CRP (8.49% vs. 5.96%, log-rank P < .001) were associated with higher incident AF. In adjusted Cox models, Lp(a) of 125 nmol/L or above predicted higher AF risk regardless of inflammatory status (hs-CRP 2 mg/L or above: HR 1.12, 95% CI 1.06-1.19, P < .001; hs-CRP below 2 mg/L: HR 1.09, 95% CI 1.04-1.15, P = .001), with consistent results in propensity-matched and sensitivity analyses. This large biobank study establishes Lp(a) as an AF risk factor operating through pathways distinct from systemic inflammation.
Original abstract
Background: Atrial fibrillation (AF) significantly increases stroke, heart failure, and mortality risk. Elevated lipoprotein(a) (Lp[a]) is implicated in AF pathogenesis, but its relationship with systemic inflammation (assessed by high-sensitivity C-reactive protein [hs-CRP]) remains unclear.
Objective: We investigated whether the Lp(a)-AF association is independent of baseline inflammatory status.
Methods: In this retrospective cohort study, we analyzed 365,899 United Kingdom (UK) Biobank participants without baseline AF. Lp(a) was modeled as a continuous exposure (per 50 nmol/L increase) and categorically (≥125 nmol/L). Systemic inflammation was defined as hs-CRP ≥2 mg/L. Multivariable Cox proportional hazards models (adjusted for age, sex, cardiovascular risk factors, and comorbidities) and Fine-Gray competing risk analyses estimated adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for incident AF.
Results: Over a median 13.5-year follow-up, 25,048 (6.8%) incident AF cases occurred. Elevated Lp(a) (7.44% vs 6.77%, log-rank P < .001) and hs-CRP (8.49% vs 5.96%, log-rank P < .001) were associated with higher incident AF. Cox proportional hazard model adjusted for covariates, higher Lp(a) (≥125 nmol/L) was associated with increased AF risk regardless of hs-CRP levels for AF (hs-CRP ≥2 mg/L: HR, 1.12; 95% CI, 1.06-1.19; P < .001; hs-CRP <2 mg/L: HR, 1.09; 95% CI, 1.04-1.15; P = .001). Results remained consistent in propensity score matched cohorts and sensitivity analyses.
Conclusion: Lp(a) is an independent risk factor for AF, irrespective of baseline inflammatory status, as measured by hs-CRP.
Summary written by lp-a.org from the published abstract; figures as published. Page updated 17 August 2026. Methods.