Abstract
Background and aim
Obese individuals have a high risk of developing heart failure independent of concomitant diseases like diabetes mellitus or arterial hypertension. Hormones secreted by adipose tissue, so-called adipokines such as retinol binding protein 4 (RBP4) or fatty-acid binding protein 4 (FABP4), might constitute a link between obesity and cardiomyopathy. RBP4 is known to be associated with left ventricular hypertrophy, whereas FABP 4 has been reported to display anti- as well as pro-hypertrophic properties.
Purpose
The aim of the present study was to investigate the association of FABP4 and RBP4 with cardiac magnet resonance imaging (MRI) parameters.
Methods
We investigated plasma FABP4 and plasma RBP4 levels in 210 patients (70% men, mean age 63 years) who underwent an MRI examination and tested whether FAPB4 and RBP4 levels were associated with the following MRI parameters using linear cox regression: left ventricular end-diastolic diameter (LVEDD), interventricular septal (IVS) wall thickness, posterior wall thickness (PWT), relative wall thickness (RWT), left ventricular mass index (LVMI), and left ventricular ejection fraction (LVEF). An unadjusted model and a model adjusted for age, sex, body mass index, diabetes mellitus, arterial hypertension, smoking, presence of coronary artery disease, and heart rate were used.
Results
FABP4 levels were associated with higher PWT (p=0.0044, est. = 0.4) and RWT (p=0.0011, est. = 0.02). The associations were significant in the fully adjusted model (for PWT p=0.013, for RWT p=0.0043). We observed a non-linear, U-shaped relationship between FABP4 and LVEDD, RVEDD, and LVMI. FABP4 levels below the median were associated with lower LVEDD (p=0.0239, est. = −2.7), RVEDD (p=0.0127, est. = −3.3) and LVMI (p<0.001, est. = −7.9). The association were significant in the fully adjusted model for LVMI only (p=0.0016). The associations between FABP4 levels above the median and LVEDD, RVEDD, and LVMI tended to be positive but were not statistically significant.
In the unadjusted model, RBP4 was associated with higher IVS (p=0.005, est. = 0.59), PWT (p=0.0135, est. = 0.37), and RWT (p=0.036, est. = 0.02); these associations were not significant in the fully adjusted model. Interestingly, an interaction was observed between age and RBP4 in the association with IVS (p interaction = 0.01). Higher RBP4 and older age might have a stronger effect on IVS than age and RBP4 alone (see Figure).
Conclusion
Anti- and pro-hypertrophic associations between FABP4 and left ventricular size have been reported. Our results suggest that the pro- and anti-hypertrophic effects of FABP4 are dependent on FABP4 concentrations: lower FABP4 levels have an anti-hypertrophic effect, whereas higher levels have a pro-hypertrophic effect.
RBP4 was associated with parameters of cardiac hypertrophy determined by MRI, suggesting a possible link between obesity and cardiomyopathy in an age dependent manner.
Cardiometabolic Correlates and Heritability of Fetuin-A, Retinol-Binding Protein 4, and Fatty-Acid Binding Protein 4 in the Framingham Heart Study
