It is not clear to date whether a first admission in heart failure (HF) marks a worse evolution in patients not previously diagnosed with HF ("de novo HF") than those already diagnosed as outpatients ("acutely decompensated HF"). The aim of the study was to analyze whether survival in patients admitted for de novo HF differs from the survival in those admitted for a first episode of decompensation but with a previous diagnosis of HF. This study includes an analysis of 1,728 patients admitted for decompensated HF during 9years. Readmissions and patients with left ventricular ejection fraction ≥50% were excluded (finally, 524 patients analyzed). We compared de novo HF (n=186) in patients not diagnosed with HF, although their structural heart disease was defined, versus acutely decompensated HF (n=338). The clinical profiles in both groups were similar. The de novo HF group more frequently presented with normal right ventricular function, with less presence of severe tricuspid regurgitation. The probability of survival was low in both groups. Thus, the median life in the de novo HF group was 2.1years and in the acutely decompensated HF group, 3.5years. There was a lower probability of long-term survival in the de novo HF group (p=0.035). The variables associated with mortality were age (p <0.0001), ischemic heart disease (p <0.0001), hypertension (p=0.009), obesity (p=0.025), diabetes (p=0.001), and N-terminal pro-brain natriuretic peptide at admission (p <0.0001). A higher glomerular filtration rate was associated with better survival (p=0.033). De novo HF was associated with a higher mortality than chronic HF with acute decompensation (hazard ratio 1.53, 95% confidence interval 1.03 to 2.27, p=0.036). In conclusion, the first admission for HF decompensation in patients with no previous diagnosis of HF identifies a subgroup of patients with higher long-term mortality.