N-terminal pro-brain natriuretic peptide (NT–pro-BNP) and blood urea nitrogen (BUN) predict outcomes in patients with heart failure (HF). However, it is unknown whether NT–pro-BNP is a better prognostic marker than BUN in patients hospitalized with HF. Chart reviews were performed on 257 consecutively hospitalized patients with HF whose NT–pro-BNP levels were drawn at the time of admission. The ability of NT–pro-BNP and BUN to predict the primary end point (death or readmission <30 days after discharge) was determined. Seventy-three patients (28%) reached the primary end point. Patients who reached the primary end point had significantly higher NT–pro-BNP and BUN levels. On multivariate regression analysis, the predictive values of BUN and NT–pro-BNP were very similar: the hazard ratio for NT–pro-BNP greater than the median was 1.81 (p = 0.044), and the hazard ratio for BUN greater than the median was 1.83 (p = 0.039). Analysis of the associations between NT–pro-BNP, BUN, and 30-day death or readmission as end points showed that BUN is a better predictor of outcomes (hazard ratio 3.15, p = 0.012) than NT–pro-BNP (hazard ratio 1.44, p = 0.399). In conclusion, in patients admitted to hospitals with HF, BUN is at least an equal prognosticator of HF rehospitalization or death as NT–pro-BNP. BUN outperforms NT–pro-BNP in predicting mortality in patients with advanced HF. If admitting physicians are confident that the diagnosis of HF is correct, then admission NT–pro-BNP adds little to clinical management.