Hyperuricemia treatment in acute heart failure patients does not improve their long‐term prognosis: A propensity score matched analysis from the AHEAD registry

Background Hyperuricemia is associated with a poorer prognosis in heart failure (HF) patients. Benefits of hyperuricemia treatment with allopurinol have not yet been confirmed in clinical practice. The aim of our work was to assess the benefit of allopurinol treatment in a large cohort of HF patients. Methods The prospective acute heart failure registry (AHEAD) was used to select 3160 hospitalized patients with a known level of uric acid (UA) who were discharged in a stable condition. Hyperuricemia was defined as UA ≥500 μmoL/L and/or allopurinol treatment at admission. The patients were classified into three groups: without hyperuricemia, with treated hyperuricemia, and with untreated hyperuricemia at discharge. Two‐ and five‐year all‐cause mortality were defined as endpoints. Patients without hyperuricemia, unlike those with hyperuricemia, had a higher left ventricular ejection fraction, a better renal function, and higher hemoglobin levels, had less frequently diabetes mellitus and atrial fibrillation, and showed better tolerance to treatment with angiotensin‐converting enzyme inhibitors/angiotensin receptor blockers and/or beta‐blockers. Results In a primary analysis, the patients without hyperuricemia had the highest survival rate. After using the propensity score to set up comparable groups, the patients without hyperuricemia had a similar 5‐year survival rate as those with untreated hyperuricemia (42.0% vs 39.7%, P = 0.362) whereas those with treated hyperuricemia had a poorer prognosis (32.4% survival rate, P = 0.006 vs non‐hyperuricemia group and P = 0.073 vs untreated group). Conclusion Hyperuricemia was associated with an unfavorable cardiovascular risk profile in HF patients. Treatment with low doses of allopurinol did not improve the prognosis of HF patients.


| INTRODUCTION
Heart failure (HF) affects over 23 million people worldwide. Despite improvements in both pharmacological and non-pharmacological treatment, its prognosis remains rather poor: patients hospitalized for acute heart failure have 1-and 5-year mortality rates 32% and 60%, respectively. 1 The current European Society of Cardiology (ESC) guidelines for HF highlight the importance of comorbidities that might influence the patients' prognosis and quality of life. 2 Hyperuricemia might be an extremely interesting culprit in HF and cardiovascular disease. Several publications have recently identified hyperuricemia as an independent unfavorable prognostic factor in HF patients. 3,4 Elevated levels of uric acid (UA) have been reported in up to 50% of HF patients, which is in stark contrast to 2% to 18% of people in a healthy population. 5 The pathophysiological association between elevated levels of UA and an unfavorable prognosis of HF patients has not been clarified yet. UA is the end product of purine metabolism. In a two-step reaction, the enzyme xanthine oxidase (XO) catalyzes the oxidation of hypoxanthine to xanthine and the oxidation of xanthine to UA. During this process, a highly reactive superoxide radical (O 2 − ) -one of the reactive oxygen species (ROS) -is produced. There are several possible contributors to elevated levels of UA in HF, including increased activity of XO or decreased renal excretion of UA. 5 Diuretic therapy can also contribute to elevated levels of UA. 6 The elevation of UA levels itself can have a negative prognostic influence, but the up-regulation of XO activity is probably a significant contributor, too. Elevated UA levels are associated with a reduction in nitric oxide (NO) levels and with endothelial dysfunction. 7 On the other hand, UA has an antioxidant effect and might counteract the effect of ROS. 8,9 To some extent, hyperuricemia is a marker of upregulation of XO activity, which subsequently leads to an increase in oxidative stress.
Oxidative stress plays an important negative role in the pathophysiology of HF: it causes myocardial fibrosis, 10 left ventricular remodeling, 11 decreased myocardial contractility, 12 and diastolic dysfunction. 13 In a mouse model, XO inhibition by allopurinol delayed heart failure progression. 14 In HF patients, inhibition of XO by allopurinol had a beneficial effect in terms of improved endothelial dysfunction, 15 decreased levels of B-type natriuretic peptide 16 and improved left ventricular contractility. 17,18 On the other hand, two smaller randomized trials in patients with left ventricular systolic dysfunction did not prove a positive clinical effect of hyperuricemia treatment with allopurinol. 8,19 Hyperuricemia is therefore a marker of poor prognosis in HF patients. There are experimental data on the beneficial effect of XO inhibition by allopurinol but there is no evidence on the benefit of this treatment in clinical practice.
Using the propensity score, we examined data from the AHEAD registry to assess whether treatment of hyperuricemia with allopurinol is associated with improvements in middle and long-term mortality in acute HF patients.

| Study population
The AHEAD registry, with inclusion and exclusion criteria, was described in detail in previous publications. 20,21 In brief, patients hospitalized with acute HF in 15 centers across a region with approximately 3 million inhabitants in the Czech Republic were consecutively enrolled between 2006 and 2012 and followed after their discharge.
A total of 7318 records on hospitalizations for acute HF were prospectively collected in the registry. In our analyses, we only included records on HF patients who were hospitalized for the first time for HF, regardless of whether they had de novo heart failure or acute decompensation of pre-existing chronic HF, and on which follow-up data were available. We excluded patients with unknown UA levels, those who died during their hospital stay or were transferred elsewhere in an unstable condition, those who were resuscitated before hospital admission and those with pulmonary embolism ( Figure 1).
Mortality data were obtained from the centralized database of the Ministry of Health of the Czech Republic. At the time of evaluation, all patients were followed up for at least 5 years. Based on a previous study, prognostically unfavorable hyperuricemia was defined as UA ≥500 μmoL/L (8.41 mg/dL). 22 The patients were classified into three groups: those without hyperuricemia (UA < 500 μmoL/L and not treated with allopurinol at admission), those with hyperuricemia and treated with allopurinol (UA > 500 μmoL/L and/or treated with allopurinol at admission; treated with allopurinol at discharge) and those with untreated hyperuricemia (UA > 500 μmoL/L and/or treated with allopurinol at admission; not treated with allopurinol at discharge).
The 24-month all-cause mortality was the primary endpoint and the 5-year all-cause mortality was the secondary endpoint. This study was carried out in accordance with the 1975 Declaration of Helsinki and was approved by the Ethics Committee of the University Hospital Brno (Brno, Czech Republic). All patients were enrolled in the study upon providing written informed consent.

| Statistical methods
Baseline patient characteristics were described using absolute and relative frequencies for categorical variables and median values supplemented with 5th-95th percentile range for continuous variables.
Factors previously demonstrated as significant for the prognosis were evaluated (Table 1).
Statistical significance of differences in patient characteristics according to UA levels and hyperuricemia treatment (no hyperuricemia, hyperuricemia treated, hyperuricemia untreated) were tested

| DISCUSSION
The present study highlights three extremely important points regarding the care for heart failure patients: (a) hyperuricemia was a predictor of a poorer long-term prognosis, (b) treatment with allopurinol, that is, xanthine oxidase inhibitor, had no influence on the patients' prognosis, (c) after setting up comparable groups using the propensity score, no difference in long-term all-cause mortality was observed between patients without hyperuricemia and those with untreated hyperuricemia. Two years and later after discharge, the mortality was significantly worse in patients with treated hyperuricemia than in those without hyperuricemia; the difference was not significant when compared treated hyperuricemia patients to patients with untreated hyperuricemia.
The admission characteristics of patients clearly show that patients without hyperuricemia had a more favorable profile of prognostic factors: higher left ventricular ejection fraction, better renal function, higher hemoglobin levels, lower occurrence of diabetes mellitus, and atrial fibrillation, better tolerance of treatment with ACE-Is/ARBs and/or beta-blockers. All of these parameters were associated with a better long-term prognosis. Additionally, NT-proBNP levels, which were only available in 30% of evaluated patients, were also lowest in patients without hyperuricemia, and highest in patients with untreated hyperuricemia. Our results suggest that elevated UA levels themselves are an indicator of a more serious condition in terms of prognosis rather than a risk factor which should be treated with the intention to improve the long-term prognosis.
T A B L E 1 Patient characteristics before propensity score matching 49% and those with EF ≥50% (heart failure with preserved ejection fraction) were involved, was the advantage of our study. Moreover, the patients were followed up for a minimum of 5 years.

| CONCLUSION
Hyperuricemia was associated with an unfavorable cardiovascular risk profile in HF patients. Treatment of hyperuricemia with low doses of allopurinol did not improve the long-term prognosis of HF patients.

ACKNOWLEDGEMENT
We would like to thank Lenka Snajdrova, MSc, Ph.D., a specialist in the Office for Development and Project Support of the Masaryk University, for her professional English grammar editing.

CONFLICT OF INTEREST
The authors declare no potential conflict of interests. F I G U R E 3 Kaplan-Meier estimate of 5-year overall survival in patients with acute heart failure according to hyperuricemia and its treatment (after propensity score matching)