Trans fatty acid blood levels and HFpEF phenotype: from the Aldo-DHF RCT

Abstract

Background: Prognosis in HFpEF is determined by risk factor control and treatment of comorbidities. Industrially processed TFA (IP‐TFA) from partially hydrogenated oils have been linked to altered lipoprotein metabolism, endothelial dysfunction, increased biomarkers of inflammation and increased NTproBNP. In patients with heart failure with preserved ejection fraction (HFpEF), associations of TFA blood levels with patient characteristics are unknown. Purpose: To evaluate associations of blood TFA with cardiovascular risk factors, aerobic capacity and cardiac function in patients with HFpEF. Methods: This is a secondary analysis from the Aldo‐DHF‐RCT. From 422 patients, individual blood TFA were analyzed at baseline in n=404 using the HS‐Omega‐3‐Index® methodology. Patient characteristics were; 67±8 years, 53% female, NYHA II/III (87/13%), ejection fraction ≥50%, E/e' 7.1±1.5; median NT‐proBNP 158 ng/L (IQR 82‐298). Multiple linear regression analyses, using sex and age as covariates, were used to describe associations of TFA with metabolic phenotype, functional capacity, echocardiographic markers for left ventricular diastolic function (LVDF), and neurohumoral activation at baseline and after 12‐months‐follow‐up (12mFU). To account for randomization group, all analyses were repeated as sensitivity analysis with group as covariate. A significance level of α=5% was used for all tests. As all tests were hypothesis generating without confirmatory interpretation, no correction was applied to counteract the problem of multiple comparisons. Results: Higher blood levels of the naturally occurring TFA C16:1n‐7t were broadly associated with a more favorable lipid profile, lower body weight/central adiposity, lower white blood cell count and lower biochemical markers of non‐alcoholic fatty liver disease at baseline/12mFU. Conversely, blood levels of the IP‐TFA C18:1n9t were directly associated with lipid risk markers [triglycerides (β=19.7, p\{{\textbackslash}textless}0,001), non‐HDL‐C (β=7.9, p=0,001), and LDL‐C (β=5.4, p=0,011)]. The two IP‐TFA C18:2n6 isomers C18:2n6tt and C18:2n6ct were positively associated with HbA1c [(β=14.6, p=0,003) and (β=4.2, p=0,014) respectively]. The IP‐TFA C18:2n6tt/‐ct isomers were associated with lower submaximal aerobic capacity (distance covered in the 6MWT) at baseline/12mFU. No significant association was found between TFA blood levels and left ventricular filling pressures, left ventricular relaxation or neurohumoral activation. Significant effects of group allocation (spironolactone +/‐) were found for the 12mFU outcomes systolic/ diastolic blood pressure (all p\{{\textbackslash}textless}0.001), heart rate, E/e' and HbA1c. Conclusions: In HFpEF patients, higher blood levels of industrially processed TFA, but not of the TFA C16:1n‐7t in full fat dairy and meat, were associated with a higher risk phenotype and lower aerobic capacity. Our findings support efforts to remove IP‐TFA from the food supply for improving risk factor control in HFpEF patients.