ObjectiveNew York Heart Association (NYHA) functional class plays a central role in heart failure (HF) assessment but might be unreliable in mild presentations. We compared objective measures of HF functional evaluation between patients classified as NYHA I and II in the Rede Brasileira de Estudos em Insuficiência Cardíaca (ReBIC)-1 Trial.MethodsThe ReBIC-1 Trial included outpatients with stable HF with reduced ejection fraction. All patients had simultaneous protocol-defined assessment of NYHA class, 6 min walk test (6MWT), N-terminal pro-brain natriuretic peptide (NT-proBNP) levels and patient’s self-perception of dyspnoea using a Visual Analogue Scale (VAS, range 0–100).ResultsOf 188 included patients with HF, 122 (65%) were classified as NYHA I and 66 (35%) as NYHA II at baseline. Although NYHA class I patients had lower dyspnoea VAS Scores (median 16 (IQR, 4–30) for class I vs 27.5 (11–49) for class II, p=0.001), overlap between classes was substantial (density overlap=60%). A similar profile was observed for NT-proBNP levels (620 pg/mL (248–1333) vs 778 (421–1737), p=0.015; overlap=78%) and for 6MWT distance (400 m (330–466) vs 351 m (286–408), p=0.028; overlap=64%). Among NYHA class I patients, 19%–34% had one marker of HF severity (VAS Score >30 points, 6MWT <300 m or NT-proBNP levels >1000 pg/mL) and 6%–10% had two of them. Temporal change in functional class was not accompanied by variation on dyspnoea VAS (p=0.14).ConclusionsMost patients classified as NYHA classes I and II had similar self-perception of their limitation, objective physical capabilities and levels of natriuretic peptides. These results suggest the NYHA classification poorly discriminates patients with mild HF.