Fat and fat-free mass causally and independently linked to incident AF

Body composition and atrial fibrillation: a Mendelian randomization study.

Literature - Tikkanen E, Gustafsson S, Knowles JW et al., - Eur Heart J. 2019 Feb 5. doi: 10.1093/eurheartj/ehz003. [

Introduction and methods

Epidemiological evidence suggests that obesity increases the risk of atrial fibrillation (AF), possibly through promoting inflammation, oxidative stress, autonomic dysfunction, cardiac fibrosis, insulin resistance and hypertension [1-4]. Alternatively, it is hypothesized that via increased hemodynamic and musculoskeletal load, obesity leads to atrial hypertrophy and larger atrial mass, and ultimately to higher risk of AF [4].

These effects could, however, also be a consequence of increased body size per se, independent of obesity. Fat-free mass (mostly muscle mass) has in fact recently been suggested to be the key anthropometric driver of elevated AF risk [5,6]. But, since different body composition traits are highly inter-correlated, it is difficult to assess their causality in AF development based on observational data.

A mendelian randomization (MR) study using genetic variants as proxy for body mass index (BMI) supported the causal role of obesity in AF [7]. BMI, however, does not distinguish between fat and muscle mass, and is therefore not an ideal measure to address the role of different anthropometric aspects. This study therefore aimed to evaluate the potential causal roles of fat-free mass and fat mass in AF development. Genetic instruments for MR analysis were obtained based on data from the UK Biobank, a large, prospective cohort study [8]. Among other physical measurements, in 2006 -2010 participants had their fat-free mass and fat mass assessed using bioelectrical impedance technique (n=492.441 and n=491.643, respectively). Median follow-up was 6.1 years (IQR: 5.4-6.7 years, 3.005.107 person-years at risk), in which 10.852 incident AF cases occurred. Mean age at baseline was 56.5 years (SD: 8.1 years).

Main results

Observational analyses

  • Fat-free mass showed a stronger association with incident AF (HR: 1.77, 95%CI: 1.72-1.83) than fat mass (HR: 1.40, 95%CI: 1.37-1.43) per SD increase in fat-free mass and fat mass, respectively. Adjusting for additional covariates attenuated these associations slightly (HR: 1.70, 95%CI: 1.65-1.76 and HR: 1.34, 95%CI: 1.31-1.37, respectively).
  • Differences in body composition were noted between males and females, thus a potential interaction of sex with these associations was investigated. Females showed a stronger association of fat mass with incident AF (HR: 1.49, 95%CI: 1.44-1.54) than males (HR: 1.35, 95%CI: 1.31-1.39, P-interaction 1.2x10^-5). No interaction was seen for gender with the association between fat-free mass and AF (HR: 1.75, 95%CI: 1.65-1.85 for females and HR: 1:77, 95%CI: 1.71-1.84 for males, P-interaction: 0.26).

Mendelian randomization analyses

  • Univariate MR analyses corroborated the causal effects of fat-free mass and fat mass on AF (causal risk ratio: 1.55, 95%CI: 1.38-1.75 and 1:30, 95%CI: 1.17-1.45 per SD increases in fat-free mass and fat mass, respectively).
  • Multivariate MR analysis showed that the potential causal effects of fat-free mass and fat mass on AF were independent of each other.
  • Adjusting for fat mass attenuated the effect of fat-free mass as compared with the univariate analysis (causal risk ratio: 1.37, 95%CI: 1.06-1.75). Adjusting for fat-free mass did not substantially change the effect of fat-mass (causal risk ratio: 1.28, 95%CI: 1.03-1.58).
  • MR analysis also showed larger effect of fat mass on AF in females than in males (causal risk ratio: 1.30, 95%CI: 1.22-1.39 and 1.21, 95%CI: 1.15-1.29, respectively), while the effects of fat-free mass on AF were similar for women and men.


The MR analyses in this study suggest that both fat-free mass and fat mass are causally and independently associated with incident AF. Both observational and MR analyses suggest that the association of fat-free mass with AF is similar for men and women, while fat mass was more strongly associated with AF in women. The effect size of the relation between fat-free mass and AF was smaller in MR analyses than in observational data, thus, the latter are likely to be confounded.

Editorial comment

Trenkwalder and Schunkert [9] also note that lean and fat mass are closely related, and thus their effect on incident AF is hard to disentangle. Both can be mechanistically linked to incident AF: elevated lean mass via an increased volume load of the heart, and elevated fat mass via processes such as inflammation, oxidative stress and lipid deposits.

About the application of mendelian randomization to test for a causal relationship, they note that it is a prerequisite that the genetic variant that is used as a proxy for the trait, can only have a single immediate effect on the trait that is being evaluated. This is relevant in light of the study by Tikkanen et al., and they wonder whether the observed finding is definitive? Trenkwalder and Schunkert state that body composition is not the ideal trait for MR analyses. Ideally, there should be no alternative pathway from the genetic variant to the outcome. However, some of the genetic variants used by Tikkanen et al. affected both fat and fat-free mass, to varying degrees. Moreover, the primary cellular effects of several variants are largely unknown. It is therefore feasible that the used genetic variants affect various pathways that lead to a change in body composition, and ultimately to AF. Tikkanen et al. addressed these limitations by using the MR-Egger technique, but also this technique cannot rule pleiotropy completely.

Unfortunately, the current paper did not assess some of the measurable potential intermediary effects that link genetic variants with AF risk, such as atrial size, heart rate, left ventricular diastolic or systolic function. Thus, this MR analysis cannot provide as definitive conclusions as some MR analyses of other traits. However, the results of the study by Tikkanen et al. do, however, contribute to the notion that obtaining an optimal weight may represent an important component in AF prevention, as it combines reducing both fat and fat-free mass. The study data also suggest that ‘lean body mass should be kept lean by not overeagerly shaping the body to high muscle (or fat-free) mass.’ Trenkwalder and Schunkert conclude: ‘Taken together, as complex as the etiology of AF appears to be, its prevention is just as complex.’ (…) ‘Only the successful execution of an integrated approach may rigorously reduce AF risk.’


1. Aune D, Sen A, Schlesinger S et al. Body mass index, abdominal fatness, fat mass and the risk of atrial fibrillation: a systematic review and dose-response meta-analysis of prospective

studies. Eur J Epidemiol 2017;32:181–192.

2. Frost L, Benjamin EJ, Fenger-Grøn M et al. Body fat, body fat distribution, lean body mass and atrial fibrillation and flutter. A Danish cohort study. Obesity 2014;22:1546–1552.

3. Karas MG, Yee LM, Biggs ML, et al. Measures of body size and composition and risk of incident atrial fibrillation in older people: the cardiovascular health study. Am J Epidemiol 2016;183:998–1007.

4. Nattel S. Atrial fibrillation and body composition: is it fat or lean that ultimately determines the risk? J Am Coll Cardiol 2017;69:2498–2501.

5. Fenger-Gron M, Overvad K, Tjonneland A, Frost L. Lean body mass is the predominant anthropometric risk factor for atrial fibrillation. J Am Coll Cardiol 2017; 69:2488–2497.

6. Azarbal F, Stefanick ML, Assimes TL, et al. Lean body mass and risk of incident atrial fibrillation in post-menopausal women. Eur Heart J 2016;37:1606–1613.

7. Chatterjee NA, Giulianini F, Geelhoed B et al. Genetic obesity and the risk of atrial fibrillation: causal estimates from Mendelian randomization. Circulation 2017;135: 741–754.

8. UK Biobank. ukbiobank.ac.uk. https://www.ukbiobank.ac.uk/ (10 January 2019).

9. Trenkwalder T and Schunkert H. Risk of atrial fibrillation in big people under the magnifying glass of G. J. Mendel. Eur Heart 2019, ehz037, https://doi.org/10.1093/eurheartj/ehz037

Find this article online at Eur Heart J.

Facebook Comments


We’re glad to see you’re enjoying PACE-CME…
but how about a more personalized experience?

Register for free