ABSTRACT
BACKGROUND: To investigate the interaction of clinical characteristics with disease characterising parameters in heart failure with preserved ejection fraction (HFpEF). Methods and results In the multicenter, randomized, placebo-controlled, double-blinded, Aldo-DHF trial investigating the effects of spironolactone on exercise capacity (peakVO2) and diastolic function (E/e') n=422 patients with HFpEF (age 67 ± 8 years, 52% females, LVEF 67 ± 8%) were included. After multiple adjustment, higher age was significantly related to reduced peakVO2, and to increased E/e', NT-proBNP, LAVI as well as LVMI (all p<0.05). Female gender (p<0.001), CAD (p=0.002), BMI (p<0.001), sleep apnoea (p=0.02), and chronotropic incompetence (CI, p=0.002) were related to lower peakVO2 values. Higher pulse pressure (p=0.04), lower heart rates (p=0.03), CI (p=0.03) and beta-blocker treatment (p=0.001) were associated with higher E/e'. BMI correlated inversely (p=0.03), whereas atrial fibrillation (p<0.001), lower haemoglobin levels (p<0.001), CI (p=0.02), and beta-blocker treatment (p<0.001) were associated with higher NT-proBNP. After multiple adjustment for demographic and clinical variables peakVO2 was not significantly associated with E/e' (r=+0.01, p=0.87), logNT-proBNP (r=0.09, p=0.08), LAVI (r=+0.03, p=0.55), and LVMI (r=+0.05, p=0.37). The associations of E/e' with logNT-proBNP (r=0.21, p<0.001), LAVI (r=+0.29, p<0.001) and LVMI (r=0.09, p=0.06) were detectable also after multiple adjustment. CONCLUSIONS: Demographic and clinical characteristics differentially interact with exercise capacity, resting left ventricular filling index, neurohumoral activation, and left atrial and ventricular remodelling in HFpEF. Exercise intolerance in HFpEF is multi-factorial and therapeutic approaches addressing exercise capacity should therefore not only aim to improve single pathological mechanisms. REGISTRATION: ISRCTN94726526 (http://www.controlled-trials.com), Eudra-CT-number 2006-002605-31.