Translating guidelines into clinical practice: benefits of an acute heart failure unit

Heart failure (HF) is a very prevalent condition, with high mortality, morbidity and costs, despite the development of effective drug therapy. Many patients still do not have access to appropriate treatment. Heart failure clinics have shown success in improving delivery of appropriate HF management, helping to reduce morbidity and mortality and costs.

To compare the quality of diagnosis and drug therapy at discharge in an unselected population of patients admitted to a medical ward of an urban central teaching hospital before and after the creation of a heart failure unit (HFU) and to evaluate its impact on long-term mortality and morbidity from all causes and from HF.

We compared a population of 153 patients, aged 75 +/- 12.8 years, 50.3% male, with a diagnosis of HF according to the guidelines of the European Society of Cardiology, NYHA classes II-IV (74% III or IV), admitted consecutively to a medical ward (MW) before the creation of an HFU (Group 1), with a population of 82 patients, aged 73 +/- 12.4 years, 62.2% male, admitted consecutively to the HFU for acute decompensated HF (Group 2). The prescription of diuretics, spironolactone, angiotensin-converting enzyme (ACE) inhibitors, beta-blockers, angiotensin receptor antagonists (ARAs) and warfarin in the two groups at discharge was analyzed. The endpoints of the study were: in-hospital mortality (IM), mortality and hospitalization due to all causes (P1), mortality and hospitalization due to HF (P2), all-cause mortality (P3), and mortality due to HF (P4). The follow-up duration was 18 months.

The two populations were not statistically different in gender, age or type of HF. HF diagnosis was correct in 67.7% of cases in Group 1, vs. 86% in Group 2. The increase in the prescription of beta-blockers in Group 2 compared with Group 1 was highly significant (4.9% vs. 36.0%, p = 0.000 in all-cause HF patients and 28.9% vs. 56.8%, p = 0.000, in HF due to systolic dysfunction), as was that of warfarin (16.7% vs. 32.6%, p = 0.003 in all-cause HF and 17.6% vs. 44.4%, p = 0.001, in HF due to systolic dysfunction). IM in the MW vs. HFU was 7.7% vs. 8.5% (p = 0.42). The probabilities for the different end-points in MW/HFU at 6, 12 and 18 months were: P1: 55/35% (p = 0.013), 68/46% (p = 0.0005) and 73/51% (p = 0.0004); P2: 26/19% (p = 0.1), 34/20% (p = 0.008) and 39/23% (p = 0.004); P3: 25.5/21% (p = 0.22), 31/30% (p = 0.44) and 38/35% (p = 0.32); P4: 16/19% (p = 0.28), 23/20% (p = 0.3) and 27/33% (p = 0.25).

These results are in agreement with reported morbidity and mortality from the syndrome after hospitalization. There is a tendency for better outcome in all the studied endpoints among the population treated in the HFU, despite greater severity of the disease, compared to those treated in the MW. A statistically significant improvement was found in the composite endpoints of mortality and hospitalization from all causes and due to HF, at 12 and 18 months. The results of the study indicate that the HFU had a better performance in reducing morbidity than mortality compared to the MW. As hospitalization accounts for 60% of HF costs, HFUs are expected to be cost effective.