Additive effect of increased urea and creatinine levels at admission on long-term prognosis in patients with decompensated heart failure

Background - Increased level of creatinine is a powerful risk marker in decompensated heart failure (DHF). Our objective was to evaluate the long-term prognostic role of early detection of renal dysfunction (RD), defined by abnormal levels of urea and/or creatinine, in patients with DHF. Patients and methods - Two hundred and forty-one patients admitted for DHF were prospectively included. The cut-off of urea and creatinine were selected using ROC curves for predicting combined events (death or rehospitalization for DHF). The mean follow-up was 366 +/- 482 days. Results - The mean age were 65.4 +/- 11.6 years (64% male, 42.3% ischemic etiology), and 44.4% had events. The area under ROC curves to predicting events for urea and creatinine was 0.59 and 0.57, respectively. The cut-off, sensitivity and specificity were: urea 55 mg/dL, 57% and 63%; creatinine 117 mg/dL, 58% and 62%, respectively. RD was identified in 144 (60.4%) subjects, 82 had elevated both markers, 29 with only increased levels of creatinine, and 33 with only abnormal levels of urea. RD groups had more frequently a previous diagnosis of HF (89 vs 78%, p=0.041) and peripheral hypoperfusion (12.5 vs 4.1%, p=0.020), and they showed lower LVEF (36.4 +/- 17.2% vs 41.1 +/- 19.6%, p=0.05) and higher pro-BNP (8.681 +/- 9010 pg/mL vs 2943 +/- 2690 pg/mL, p< 0.001) than those without RD. Eighteen-month freeDHF rehospitalization survival in patients with and without RD was 35% and 60% (p=0.0086). The variables significantly associated with events were RD (1.8, p< 0.001; CI 95%=1.1-2.7) and previous diagnosis of HF (HR=1.9, CI 95%=1.1-3.5). Conclusion - The combined use of urea and creatinine improve the early detection of RD in patients with DHF. This finding was a strong long-term prognostic predictor.