Longitudinal association between N-terminal B-type natriuretic peptide, anxiety and social support in patients with HFpEF: results from the multicentre randomized controlled Aldo-DHF trial

Background Higher plasma levels of natriuretic peptides (NPs) have been associated with reduced anxiety in experimental research and a number of patient samples. As NP levels are elevated in heart failure patients, we investigate whether this elevation is related to anxiety in patients with heart failure with preserved ejection fraction (HFpEF). Methods Post-hoc regression and mediation analyses were conducted, using data of 422 patients with HFpEF from the randomized, placebo-controlled, double-blinded, two-armed, multicentre aldosterone in diastolic heart failure trial, testing associations and their mediators between the N-terminal B-type natriuretic peptide (NT-proBNP) and anxiety at baseline and over 12-month follow-up. Anxiety was measured by the Hospital Anxiety and Depression Scale (HADS), social support by the ENRICHD Social Support Inventory and physical functioning by the Short Form 36 Health Survey. Results The mean age of the study population was 66.8 ± 7.6 years, 47.6% were male and 86.0% had NYHA class II. NT-proBNP showed a weak negative correlation with HADS anxiety scores at baseline (r = − 0.087; p = 0.092), which was significant (r = − 0.165; p = 0.028) in men but not in women. NT-proBNP also tended to predict lower anxiety at 12-months in men. On the other hand, higher anxiety at baseline was associated with lower NT-proBNP scores 12 months later (r = − 0.116; p = 0.026). All associations lost significance in multivariate regression for age, perceived social support (ESSI), physical function (SF-36) and study arm. Mediation analyses revealed that social support acts as a full mediator for the link between NT-proBNP levels and anxiety. Conclusion The mechanisms linking NT-proBNP to anxiety may be more complex than originally assumed. While effects of NT-proBNP on anxiety may be mediated by perceived social support, there may be an additional negative effect of anxiety on NT-proBNP. Future research should consider this possible bi-directionality of the association and assess the potential influence of gender, social support, oxytocin and vagal tone on the interaction of anxiety and natriuretic peptide levels. Trial Registrationhttp://www.controlled-trials.com (ISRCTN94726526) on 07/11/2006. Eudra-CT-number: 2006–002,605-31.