Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

A population-based study of the sex-specific associations between apolipoprotein B and incidence of atrial fibrillation

Wei, Yingying ; Wang, Lintao ; Zhang, Chao LU ; Xu, Biao ; Johnson, Linda S. LU ; Engström, Gunnar LU and Bao, Xue LU (2026) In Lipids in Health and Disease 25(1).
Abstract

Background: Apolipoprotein B (apoB) is a well-known risk factor for atherosclerosis. However, studies examining its relation to atrial fibrillation (AF) have produced conflicting results and suggested possible sex-specific differences. This study investigated the sex-specific associations between serum apoB concentrations and incident AF and offer insight into the inconsistencies in previous research. Methods: A prospective analysis of 26,803 participants without pre-existing AF was performed using data from the Malmö Diet and Cancer Study. Sex-specific associations between apoB and AF were assessed using multivariable Cox proportional hazards models. To ensure the robustness of the results, several sensitivity analyses, such as... (More)

Background: Apolipoprotein B (apoB) is a well-known risk factor for atherosclerosis. However, studies examining its relation to atrial fibrillation (AF) have produced conflicting results and suggested possible sex-specific differences. This study investigated the sex-specific associations between serum apoB concentrations and incident AF and offer insight into the inconsistencies in previous research. Methods: A prospective analysis of 26,803 participants without pre-existing AF was performed using data from the Malmö Diet and Cancer Study. Sex-specific associations between apoB and AF were assessed using multivariable Cox proportional hazards models. To ensure the robustness of the results, several sensitivity analyses, such as restricted cubic spline modeling, competing risks regression, alternative adjustment strategies, subgroup analyses, follow-up time restrictions, and multiple imputation for missing data, were conducted. Results: For median follow-up periods of 21.2 and 24.8 years in men and women, respectively, 2,768 and 2,968 incident cases of AF were recorded, respectively. Among women, unadjusted models showed a strong positive association between apoB and AF, with the highest versus lowest quartile showing a hazard ratio (HR) of 1.65 (95% confidence interval [CI] 1.49–1.84; P for trend < 0.0001). The association became non-significant after age adjustment (P for trend = 0.09), and was reversed after multivariable adjustment (HR 0.77, 95% CI 0.69–0.86; P for trend < 0.0001). Sensitivity analyses consistently supported a significant linear inverse association in women. No significant link between apoB and AF was detected in the male population. Most sensitivity analyses were similarly null except for the restricted cubic spline analysis which suggested a borderline non-linear association (P for effect = 0.04, P for nonlinearity = 0.05). Post-hoc analysis suggested an inverse association at lower apoB concentrations (P = 0.0012) (≤ 100 mg/dL: HR per standard deviation 0.90, 95% CI 0.85–0.96). Conclusions: Results show sex-specific observational links between apoB concentrations and risk of AF. In women, higher apoB levels were linearly inversely associated with AF, whereas in men, the association was borderline non-linear, with inverse effects seen mainly at lower apoB concentrations. These sex differences in AF susceptibility may partly reflect underlying atrial electrophysiological variations and hormonal influences, though whether these factors directly mediate the apoB-AF association remains speculative.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Apolipoprotein B, Atrial fibrillation, Cohort studies, Sex differences
in
Lipids in Health and Disease
volume
25
issue
1
article number
93
publisher
BioMed Central (BMC)
external identifiers
  • scopus:105034539690
  • pmid:41731549
ISSN
1476-511X
DOI
10.1186/s12944-026-02905-6
language
English
LU publication?
yes
additional info
Publisher Copyright: © The Author(s) 2026.
id
5b43633c-3abe-4903-bca4-3ba796dec99b
date added to LUP
2026-05-26 14:56:17
date last changed
2026-06-09 15:53:33
@article{5b43633c-3abe-4903-bca4-3ba796dec99b,
  abstract     = {{<p>Background: Apolipoprotein B (apoB) is a well-known risk factor for atherosclerosis. However, studies examining its relation to atrial fibrillation (AF) have produced conflicting results and suggested possible sex-specific differences. This study investigated the sex-specific associations between serum apoB concentrations and incident AF and offer insight into the inconsistencies in previous research. Methods: A prospective analysis of 26,803 participants without pre-existing AF was performed using data from the Malmö Diet and Cancer Study. Sex-specific associations between apoB and AF were assessed using multivariable Cox proportional hazards models. To ensure the robustness of the results, several sensitivity analyses, such as restricted cubic spline modeling, competing risks regression, alternative adjustment strategies, subgroup analyses, follow-up time restrictions, and multiple imputation for missing data, were conducted. Results: For median follow-up periods of 21.2 and 24.8 years in men and women, respectively, 2,768 and 2,968 incident cases of AF were recorded, respectively. Among women, unadjusted models showed a strong positive association between apoB and AF, with the highest versus lowest quartile showing a hazard ratio (HR) of 1.65 (95% confidence interval [CI] 1.49–1.84; P for trend &lt; 0.0001). The association became non-significant after age adjustment (P for trend = 0.09), and was reversed after multivariable adjustment (HR 0.77, 95% CI 0.69–0.86; P for trend &lt; 0.0001). Sensitivity analyses consistently supported a significant linear inverse association in women. No significant link between apoB and AF was detected in the male population. Most sensitivity analyses were similarly null except for the restricted cubic spline analysis which suggested a borderline non-linear association (P for effect = 0.04, P for nonlinearity = 0.05). Post-hoc analysis suggested an inverse association at lower apoB concentrations (P = 0.0012) (≤ 100 mg/dL: HR per standard deviation 0.90, 95% CI 0.85–0.96). Conclusions: Results show sex-specific observational links between apoB concentrations and risk of AF. In women, higher apoB levels were linearly inversely associated with AF, whereas in men, the association was borderline non-linear, with inverse effects seen mainly at lower apoB concentrations. These sex differences in AF susceptibility may partly reflect underlying atrial electrophysiological variations and hormonal influences, though whether these factors directly mediate the apoB-AF association remains speculative.</p>}},
  author       = {{Wei, Yingying and Wang, Lintao and Zhang, Chao and Xu, Biao and Johnson, Linda S. and Engström, Gunnar and Bao, Xue}},
  issn         = {{1476-511X}},
  keywords     = {{Apolipoprotein B; Atrial fibrillation; Cohort studies; Sex differences}},
  language     = {{eng}},
  number       = {{1}},
  publisher    = {{BioMed Central (BMC)}},
  series       = {{Lipids in Health and Disease}},
  title        = {{A population-based study of the sex-specific associations between apolipoprotein B and incidence of atrial fibrillation}},
  url          = {{http://dx.doi.org/10.1186/s12944-026-02905-6}},
  doi          = {{10.1186/s12944-026-02905-6}},
  volume       = {{25}},
  year         = {{2026}},
}