RESUMO
AIMS: Heart failure (HF) management is complicated by difficulties in clinical assessment. Biomarkers may help guide HF management, but the correspondence between clinical evaluation and biomarker serum levels has hardly been studied. We investigated the correlation between biomarkers and clinical signs and symptoms, the influence of patient characteristics and comorbidities on New York Heart Association (NYHA) classification and the effect of using biomarkers on clinical evaluation. METHODS AND RESULTS: This post-hoc analysis comprised 622 patients (77 ± 8 years, 76 % NYHA class ≥3, 80 % LVEF ≤45 %) participating in TIME-CHF, randomising patients to either NT-proBNP-guided or symptom-guided therapy. Biomarker measurements and clinical evaluation were performed at baseline and after 1, 3, 6, 12 and 18 months. NT-proBNP, GDF-15, hs-TnT and to a lesser extent hs-CRP and cystatin-C were weakly correlated to NYHA, oedema, jugular vein distension and orthopnoea (ρ-range: 0.12-0.33; p < 0.01). NT-proBNP correlated more strongly to NYHA class in the NT-proBNP-guided group compared with the symptom-guided group. NYHA class was significantly influenced by age, body mass index, anaemia, and the presence of two or more comorbidities. CONCLUSION: In HF, biomarkers correlate only weakly with clinical signs and symptoms. NYHA classification is influenced by several comorbidities and patient characteristics. Clinical judgement seems to be influenced by a clinician's awareness of NT-proBNP concentrations.