Atrial fibrillation (AF) is an increasingly prevalent cardiac arrhythmia that is significantly associated with mortality and morbidity from stroke, heart failure (HF), dementia, and hospitalizations.¹ Despite advances in the pharmacological and procedural treatment of AF, primary prevention strategies remain relatively limited.

Lipoprotein(a) (Lp[a]) is attributed to atherosclerotic cardiovascular disease through proposed mechanisms related to its proatherogenic, prothrombotic, and proinflammatory inclinations.² Epidemiologic, genome‑wide associations and Mendelian randomization data provide evidence for a causal role of elevated Lp(a) levels in the development of atherosclerotic cardiovascular disease, independent of other risk factors.³

There is increasing evidence that Lp(a) and body mass index (BMI) are independent risk factors for AF.^4,5 Indeed, the synergistic effects of Lp(a) and BMI may be greater than the sum of their separate effects. Thus, exploration of interactions may enable identification of more susceptible subgroups of patients.

In this issue of Polish Archives of Internal Medicine, Guo et al⁶ sought to determine the combined influence of BMI and Lp(a) on AF. In their retrospective cohort study from a single center, the authors found out that a reduction in Lp(a) and BMI, individually or in combination, was independently associated with a reduced risk of AF. Individuals with lower Lp(a) levels and lower BMI had a comparatively lower rate of stroke, HF, and major adverse cardiovascular events. Following adjustment for age and sex, decreased Lp(a) levels alone and decreased Lp(a) levels and BMI in combination were independently associated with a significant reduction in stroke risk. The authors conducted a Mendelian randomization mediation analysis to assess the direct and indirect associations between Lp(a) (predictor variable [X]) and AF‑related events (outcome variable [Y] with BMI [Z] as a mediator). There was no significant mediating effect of the Lp(a) level or BMI, which indicated an independent causal mechanism involving Lp(a) levels, different BMI categories, and the risk of AF.

There has been a significant increase in studies exploring the relationship between Lp(a) and the risk of AF, albeit with conflicting results. A large retrospective cohort study conducted by Awad et al⁴ reported findings indicating that an elevated Lp(a) level (>50 mg/dl) is an independent risk factor for incident AF, while Aronis et al⁷ applied multivariable‑adjusted Cox models in the community‑based ARIC (Atherosclerosis Risk in Communities) study cohort, and did not observe any association between high Lp(a) levels and incident AF. In contrast, Garg et al⁸ observed an inverse association between Lp(a) levels and incidence of AF in a population derived from the MESA (Multi‑Ethnic Study of Atherosclerosis).

Mendelian randomization has emerged as an effective tool to identify relevant associations between biomarkers and phenotypes, and is well suited to the evaluation of Lp(a) levels, as they are predominantly determined by genetic factors.⁹ The results from the Mendelian randomization analysis conducted by Guo et al⁶ support an independent causal mechanism linking Lp(a) levels and incidence of AF. This is consistent with other studies, including that by Mohammadi‑Shemirani et al,¹⁰ where the authors reported that 15 genetic variants within 500 kb of the LPA gene associated with Lp(a) concentrations in the United Kingdom Biobank were associated with an increasing risk of AF in both investigated cohorts.^11,12 Furthermore, a subgroup analysis showed no modification of the effect of either observed or genetically predicted Lp(a) on AF to prevalent ischemic heart disease or aortic valve stenosis status. Of note, Singh et al¹³ conducted a meta‑analysis of Mendelian randomization studies that demonstrated ethnic differences, where elevated Lp(a) levels were associated with an increased AF risk in the European cohort, and a low AF risk in the Chinese population. The discrepancy in findings between the ethnic groups potentially explains the observations reported by Awad et al,⁴ where the investigated population was predominantly derived from individuals of European descent.

The mechanisms underlying the association between the elevated Lp(a) levels, higher BMI, and AF are not yet fully elucidated, although several potential hypotheses have been proposed. Higher BMI is associated with increased myocardial and vascular stiffness,¹⁴ while elevated Lp(a) is associated with prothrombotic effect and hypercoagulable status, which is implicated in atrial remodeling and fibrosis.¹⁵ Both elevated Lp(a) and higher BMI contribute to an increased systemic inflammatory burden, which affects atrial electrophysiology and atrial structure, including vascular injury, endothelial dysfunction, and arteriosclerosis.

The study by Guo et al⁶ has several limitations, many of which have already been acknowledged by the authors. First, the external validity of the findings is limited, as the clinical data were derived from a single center. Second, although incident events were prospectively collated, baseline data were collected retrospectively. As a result, certain relevant variables, such as ischemic heart disease and valvular heart disease, were not adjusted for in the analysis. Third, there was no ethnic description of the population and thus, ethnic differences were not explored, which is important given ethic differences in clinical epidemiology and cardiovascular conditions.^16,17 Lastly, the data were acquired from patients receiving care within a health care system, which potentially introduces selection bias in view of the likely better health care accessibility and greater likelihood of comorbidities relative to the general population. In view of the above, the potential extrapolation of the findings has to be approached with caution, especially in the context of primary prevention of AF, where consideration of asymptomatic individuals in community‑based settings is crucial.

Further research is needed to advance our understanding of the association between Lp(a), BMI, and AF. Studies with robust methodologies and significant longitudinal periods are necessary to explore the impact of dynamic changes in Lp(a) and BMI on AF development. Also, incorporation of Lp(a) and BMI into multifactorial clinical prediction models may have the potential to augment risk stratification and earlier identification of AF. Next, the standardization of Lp(a) measurements and establishing clinically relevant thresholds and significant trends are essential for eventual translation into clinical practice. Lastly, future studies should proactively incorporate a greater range of diversity in the studied populations to determine whether the association of Lp(a), BMI, and AF are consistent across varying ethnical and environmental contexts.