OBJECTIVES: To evaluate physician-determined worsening heart failure (PD-WHF) in patients admitted with acute heart failure (AHF). METHODS: The PROTECT pilot study evaluated rolofylline, an adenosine A(1) receptor antagonist, versus placebo in patients with AHF and renal impairment. Signs and symptoms of heart failure (HF) and diuretic administration were prospectively recorded daily for 7 days and patients were followed for 60 days. Patients were categorized into three groups: (A) PD-WHF, based on worsening symptoms and signs of HF and need for additional intravenous (IV) or mechanical therapy (n = 29); (B) increased IV diuretic therapy without PD-WHF (n = 61), and (C) neither PD-WHF nor increase in IV diuretic dose (n = 211). RESULTS: Patients in group A had slower resolution of dyspnea, longer mean (+/-SD) length of hospitalization (13.8 +/- 6.8 vs. 10.5 +/- 8.5 and 9.3 +/- 5.9 days in groups B and C, respectively; p < 0.05 for both), and higher 60-day death and cardiovascular or renal readmission rates [49.7 (95% confidence interval: 33.1-69.1) vs. 37.3 (26.4-50.9) vs. 19.5% (14.7-25.6) in groups B and C, respectively]. PD-WHF was a strong independent predictor of length of stay and 60-day death and cardiovascular or renal readmission. CONCLUSIONS: PD-WHF may be an indicator of short-term risk and treatment efficacy in AHF.