- Email: [email protected]
Relation of Serum Uric Acid Levels and Outcomes Among Patients Hospitalized for Worsening Heart Failure With Reduced Ejection Fraction (from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study With Tolvaptan Trial)
Relation of Serum Uric Acid Levels and Outcomes Among Patients Hospitalized for Worsening Heart Failure With Reduced Ejection Fraction (from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study With Tolvaptan Trial) Muthiah Vaduganathan, MD, MPHa,*, Stephen J. Greene, MDb, Andrew P. Ambrosy, MDc, Robert J. Mentz, MDd, Haris P. Subacius, MAe, Ovidiu Chioncel, MDf, Aldo P. Maggioni, MDg, Karl Swedberg, MDh, Faiez Zannad, MDi, Marvin A. Konstam, MDj, Michele Senni, MDk, Michael M. Givertz, MDl, Javed Butler, MD, MPHm, and Mihai Gheorghiade, MDb, on behalf of the EVEREST trial investigators We investigated the clinical profiles associated with serum uric acid (sUA) levels in a large cohort of patients hospitalized for worsening chronic heart failure with ejection fraction (EF) £40%, with specific focus on gender, race, and renal function based interactions. In 3,955 of 4,133 patients (96%) with baseline sUA data, clinical characteristics and outcomes were compared across sUA quartiles. The primary end points were all-cause mortality and a composite of cardiovascular mortality or heart failure hospitalization. Interaction analyses were performed for gender, race, and baseline renal function. Median follow-up was 9.9 months. Mean sUA was 9.1 – 2.8 mg/dl and was higher in men than in women (9.3 – 2.7 vs 8.7 – 3.0 mg/dl, p <0.001) and in blacks than in whites (10.0 – 2.7 vs 9.0 – 2.8 mg/dl, p <0.001). Higher sUA was associated with lower systolic blood pressure and EF, higher natriuretic peptides, and more impaired renal function. After accounting for 24 baseline covariates, in patients with enrollment estimated glomerular filtration rate ‡30 ml/min/1.73 m2, sUA was strongly associated with increased all-cause mortality (hazard ratio 1.44, 95% confidence interval 1.22 to 1.69, p <0.001) and the composite end point (hazard ratio 1.44, 95% confidence interval 1.26 to 1.64, p <0.001). However, in patients with estimated glomerular filtration rate <30 ml/min/1.73 m2, sUA was not related with either end point (both p >0.4). Adjusted interaction analyses for gender, race, and admission allopurinol use were not significant. In conclusion, sUA is commonly elevated in patients hospitalized for worsening chronic heart failure and reduced EF, especially in men and blacks. The prognostic use of sUA differs by baseline renal function, suggesting different biologic and pathophysiologic significance of sUA among those with and without significant renal dysfunction. Ó 2014 Elsevier Inc. All rights reserved. (Am J Cardiol 2014;114:1713e1721) Hospitalization for worsening chronic heart failure (WCHF) is a unique entity, distinguished by acute perturbations in clinical, neurohormonal, and laboratory indexes.1 Serum uric acid (sUA) levels have been shown to fluctuate with WCHF hospitalizations in large, prospectivelyfollowed ambulatory patients with heart failure (HF).2 Consistently, studies have suggested that sUA is a marker of adverse prognosis in the setting of WCHF.3e11 sUA may
be strongly influenced by impaired renal function,12 race,13 gender,14 and diuretic therapy. The Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST)15e17 trial database provides detailed, longitudinal, patient-level data on sUA and other clinical parameters during and after hospitalization for WCHF. Thus, we aimed to evaluate the independent association between sUA at the time of enrollment and clinical characteristics and
a Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts; bCenter for Cardiovascular Innovation, Northwestern University Feinberg School of Medicine, Chicago, Illinois; cDepartment of Medicine, Stanford University School of Medicine, Stanford, California; d Duke University Medical Center, Durham, North Carolina; eDivision of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois; fInstitute of Emergency for Cardiovascular Diseases “Prof. C.C. Iliescu”, Cardiology, Bucharest, Romania; gResearch Center of the Italian Association of Hospital Cardiologists (ANMCO), Florence, Italy; hUniversity of Gothenburg, Gothenburg, Sweden; iDepartment of Cardiology, Nancy University, Nancy, France; jTufts Medical Center, Boston, Massachusetts; kDipartimento
Cardiovascolare, Azienda Ospedaliera Papa Giovanni XXIII, Bergamo, Italy; lCardiovascular Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts; and mDivision of Cardiology, Emory University School of Medicine, Atlanta, Georgia. Manuscript received July 21, 2014; revised manuscript received and accepted September 2, 2014. Otsuka Inc. (Rockville, Maryland) provided financial and material support for the EVEREST trial. Database management was performed by the sponsor. See page 1720 for disclosure information. *Corresponding author: Tel: (þ832) 725-7222; fax: (þ617) 726-6861. E-mail address: [email protected] (M. Vaduganathan).
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.09.008
www.ajconline.org
1714
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Overall study design and analytical cohort selection. LVEF ¼ left ventricular ejection fraction; NYHA ¼ New York Heart Association.
Figure 2. Distribution and descriptive statistics of sUA levels. The primary predictor variable of baseline sUA was normally distributed with a mean and median of approximately 9 mg/dl. The black curve represents the normal density distribution and the gray curve represents the kernel density distribution for this sample. IQR ¼ interquartile range.
postdischarge outcomes in patients hospitalized for WCHF with reduced ejection fraction (EF). Methods The study design16 and primary results15,17 of the EVEREST trial have been previously described. In brief, EVEREST was a global, multicenter, double-blinded, placebo-controlled randomized trial examining tolvaptan, an oral vasopressin-2 receptor antagonist. Patients eligible for enrollment were 18 years of age, hospitalized for WCHF
with New York Heart Association III-IV functional status, EF 40% and presenting with 2 signs or symptoms of volume overload. Relevant exclusion criteria include serum creatinine >3.5 mg/dl; subjects currently treated with hemofiltration or dialysis; refractory, end-stage HF; or a life expectancy <6 months. The ethics committee and institutional review board of each participating trial center approved the study protocol. Patients were randomized to receive oral tolvaptan 30 mg fixed once daily or matching placebo within 48 hours of hospital admission and was continued for at least 60 days. Laboratory samples were
Heart Failure/Uric Acid and Hospitalized Heart Failure
1715
Table 1 Baseline characteristics by serum uric acid quartile Variable
Serum uric acid, (mg/dL), meanSD Serum uric acid, (mg/dL), range Tolvaptan assignment Age (years), meanSD Male Race Non-Hispanic White Black Asian Hispanic Other Region Eastern Europe North America South America Western Europe Weight (kg), meanSD Systolic blood pressure (mmHg), meanSD Diastolic blood pressure (mmHg), meanSD Heart rate (bpm), meanSD Jugular venous distension 10 cm Rales Peripheral edema* Dyspnea Ejection fraction (%), meanSD Blood urea nitrogen (mg/dL), median (IQR) Creatinine (mg/dL), median (IQR) Estimated GFR (mL/min/1.73 m2), meanSD † Estimated GFR<30 (mL/min/1.73 m2) † Sodium (mEq/L), median (IQR) B-type natriuretic peptide (pg/ml), median (IQR) N-terminal pro-B-type natriuretic peptide (pg/ml) median (IQR) Albumin (g/dL), median (IQR) QRS (ms), median (IQR) Atrial fibrillation Previous heart failure hospitalization Ischemic heart failure etiology NYHA class IV Coronary artery disease Previous myocardial infarction Hypertension Hypercholesterolemia Diabetes Chronic kidney disease Peripheral vascular disease Previous coronary bypass Previous percutaneous coronary intervention Automatic implantable cardioverter-defibrillator Chronic obstructive pulmonary disease Baseline medication use Allopurinol Diuretics
Baseline Serum Uric Acid Quartiles
p-value
I (n¼967)
II (n¼1,012)
III (n¼977)
5.81 1.2-7.0 514 (53.2%) 67.110.7 637 (65.9%)
80.5 7.1-8.8 494 (48.8%) 66.111.8 754 (74.5%)
9.80.6 8.9-10.8 464 (47.5%) 65.412 774 (79.2%)
12.91.8 10.9-21.0 504 (50.5%) 64.512.7 780 (78.1%)
869 42 1 40 16
876 63 1 59 14
836 73 2 47 21
816 107 5 45 31
(89.9%) (4.3%) (0.1%) (4.1%) (1.7%)
(86.6%) (6.2%) (0.1%) (5.8%) (1.4%)
(85.6%) (7.5%) (0.2%) (4.8%) (2.1%)
IV (n¼999)
0.073 <0.001 <0.001 <0.001
(81.7%) (10.7%) (0.5%) (4.5%) (3.1%) <0.001
494 (51.1%) 205 (21.2%) 144 (14.9%) 124 (12.8%) 79.818.2 124.920.5 74.312.5 78.814.8 196 (20.7%) 801 (83.7%) 778 (81.3%) 882 (92.3%) 29.67.8
452 (44.7%) 268 (26.5%) 172 (17%) 120 (11.9%) 8218.1 121.719.6 7312.4 79.215.5 248 (24.8%) 823 (82.1%) 789 (78.7%) 897 (89.5%) 28.27.8
379 (38.8%) 298 (30.5%) 171 (17.5%) 129 (13.2%) 8518.1 120.319.6 7312.9 80.316.4 291 (30.4%) 772 (79.9%) 767 (79.3%) 891 (92.2%) 27.18
245 (24.5%) 415 (41.5%) 189 (18.9%) 150 (15%) 86.220.4 115.217.9 70.412.7 80.715.7 308 (32.1%) 774 (79.8%) 783 (80.6%) 874 (90.4%) 25.48.2
<0.001 <0.001 <0.001 0.019 <0.001 0.083 0.450 0.078 <0.001
20 (17-27) 1 (0.9-1.3)
24 (19-31) 1.2 (1-1.4)
27 (21-35) 1.3 (1.1-1.6)
35 (26-48) 1.5 (1.3-1.9)
<0.001 <0.001
65.821.9 58 (6.1%) 140 (138-143) 481 (194-1137) 4070 (1650-8824) 3.8 (3.5-4.1)
60.320.1 53 (5.3%) 140 (138-142) 538.5 (254-1284) 3641 (1661-7468) 3.8 (3.4-4.1)
53.818.9 88 (9.2%) 140 (137-143) 746 (316-1474) 4256 (2241-8662) 3.8 (3.4-4.2)
44.417.4 223 (22.7%) 139 (136-142) 1106.5 (543-2126) 6800 (3642-12837) 3.7 (3.3-4)
<0.001 <0.001 <0.001
<0.001
121 258 740 659 323 689 485 693 450 379 137 205 159 141 125 73
121 277 783 653 371 725 518 717 491 387 216 223 207 174 127 98
123 311 786 641 401 687 506 693 473 363 280 191 209 181 152 100
125 288 810 614 468 685 492 705 499 401 421 209 253 205 173 124
<0.001 0.060 0.019 0.031 <0.001 0.423 0.678 0.958 0.549 0.563 <0.001 0.585 <0.001 0.006 0.007 0.004
(95-144) (26.7%) (76.7%) (68.9%) (33.4%) (71.4%) (50.2%) (71.7%) (46.9%) (39.2%) (14.2%) (21.2%) (16.4%) (14.6%) (12.9%) (7.5%)
149 (15.4%) 931 (96.5%)
(97-150) (27.4%) (77.7%) (65.4%) (36.7%) (71.6%) (51.2%) (70.8%) (48.8%) (38.2%) (21.3%) (22.1%) (20.5%) (17.2%) (12.5%) (9.7%)
110 (10.9%) 987 (97.7%)
(97-152) (31.8%) (80.9%) (66.4%) (41%) (70.4%) (51.8%) (70.9%) (48.6%) (37.2%) (28.7%) (19.5%) (21.4%) (18.5%) (15.6%) (10.2%)
96 (9.8%) 951 (97.4%)
(100-155) (28.8%) (81.5%) (62.5%) (47%) (68.6%) (49.3%) (70.6%) (50.2%) (40.1%) (42.2%) (21%) (25.3%) (20.5%) (17.3%) (12.4%)
85 (8.5%) 976 (98%)
<0.001 <0.001
<0.001 0.166 (continued)
1716
The American Journal of Cardiology (www.ajconline.org)
Table 1 (continued) Variable
Baseline Serum Uric Acid Quartiles I (n¼967)
ACEI/ ARB Beta-Blockers Mineralocorticoid receptor antagonists Digoxin Intravenous Inotropes Statin
823 667 511 438 33 345
II (n¼1,012)
(85.3%) (69.1%) (53%) (45.4%) (3.4%) (35.8%)
880 725 573 496 32 368
p-value
III (n¼977)
(87.1%) (71.8%) (56.7%) (49.1%) (3.2%) (36.4%)
837 706 550 493 44 319
(85.8%) (72.3%) (56.4%) (50.5%) (4.5%) (32.7%)
IV (n¼999) 796 696 516 489 71 339
(79.9%) (69.9%) (51.8%) (49.1%) (7.1%) (34%)
<0.001 0.346 0.065 0.132 <0.001 0.289
ACEI ¼ angiotensin converting enzyme inhibitor; ARB ¼ angiotensin II receptor blocker; GFR ¼ glomerular filtration rate; IQR ¼ interquartile range; NYHA ¼ New York Heart Association; SD ¼ standard deviation. * Peripheral edema was defined as slight/ moderate/ marked pedal or sacral edema. † Glomerular filtration rate estimated by Cockcroft-Gault formula. Table 2 Causes of death and rehospitalization by baseline serum uric acid quartile Baseline Serum Uric Acid Quartiles
Serum uric acid, (mg/dL), mean (SD) Serum uric acid, (mg/dL), range Primary Endpoints All-cause mortality CV mortality þ HF hospitalization Secondary Endpoints CV mortality CV mortality þ CV hospitalization Worsening HF * HF mortality HF hospitalization MI mortality MI hospitalization Stroke mortality Stroke hospitalization
p-value
I (n¼967)
II (n¼1,012)
III (n¼977)
IV (n¼999)
5.81 1.2-7.0
80.5 7.1-8.8
9.80.6 8.9-10.8
12.91.8 10.9-21.0
180 (18.6%) 293 (30.3%)
212 (20.9%) 362 (35.8%)
257 (26.3%) 425 (43.5%)
391 (39.1%) 547 (54.8%)
<0.001 <0.001
136 354 256 70 192 5 9 4 12
158 445 308 81 245 8 11 8 15
211 478 371 102 283 7 16 6 14
296 603 496 179 377 7 6 5 8
<0.001 <0.001 <0.001 <0.001 <0.001 0.900 0.153 0.713 0.502
(14.1%) (36.6%) (26.5%) (7.2%) (19.9%) (0.5%) (0.9%) (0.4%) (1.2%)
(15.6%) (44%) (30.4%) (8%) (24.2%) (0.8%) (1.1%) (0.8%) (1.5%)
(21.6%) (48.9%) (38%) (10.4%) (29%) (0.7%) (1.6%) (0.6%) (1.4%)
(29.6%) (60.4%) (49.6%) (17.9%) (37.7%) (0.7%) (0.6%) (0.5%) (0.8%)
CV ¼ cardiovascular; HF ¼ heart failure; MI ¼ myocardial infarction. * Worsening heart failure was defined as death from heart failure, hospitalization for heart failure, or unscheduled medical office visit for heart failure.
collected, processed, and cross validated across 5 central facilities. sUA (mg/dl) was measured at the time of study enrollment (baseline, up to 48 hours after admission) and every 4 to 8 weeks up to 112 weeks postdischarge. There was little evidence for nonlinearity in the relation between sUA and clinical end points, across a physiological range of sUA. Enrollment sUA levels were divided into quartiles and effect sizes are presented in reference to quartile 1 (lowest). For complete interaction analyses, sUA was treated as a continuous function, and effect sizes are presented per 5 mg/dl increase in sUA. Serial postdischarge sUA levels are presented by presence or absence of the primary end points. The overall study design and final analytical cohort selection are displayed in Figure 1. Demographic characteristics including self-reported race, signs and/or symptoms of HF, vital signs, laboratory parameters, initial electrocardiographic findings, medical history, and admission medications were compared across quartiles of baseline sUA levels. Baseline characteristics and
clinical outcomes were also presented separately for men, women, whites, and blacks. An independent blinded adjudication committee determined the specific causes of death and reasons for rehospitalization. The present post hoc analysis used the same 2 co-primary end points as the overall EVEREST trial: all-cause mortality (ACM) and a composite end point of cardiovascular mortality or HF hospitalization. Secondary end points included other causes of death and rehospitalization, worsening HF (defined as death, hospitalization, or unplanned office visit for HF), and combined cardiovascular mortality and rehospitalization. Median follow-up in the EVEREST trial was 9.9 months (interquartile range 5.3 to 16.1 months). Continuous variables are expressed as mean SD if normally distributed and as median (interquartile range) if non-normally distributed. Categorical variables are expressed as number (%). Outcomes were assessed as time to first event using Cox proportional hazard models. KaplanMeier curves by sUA quartile were constructed for the both primary end points and compared using log-rank tests. The
Heart Failure/Uric Acid and Hospitalized Heart Failure
1717
Figure 3. Kaplan Meier curves. Event curves for ACM (A) and composite cardiovascular mortality and HF hospitalizations (B). Times to events were compared using log-rank tests.
proportional hazards assumption (by Kolmogorov-type supremum tests for nonproportionality) was upheld. Effect sizes were reported as hazard ratios (HR) with 95% confidence intervals (CI). Multivariate models included 24 prespecified covariates including tolvaptan treatment assignment, demographic characteristics (age, gender, and region of origin), vital signs on admission (supine systolic blood pressure), laboratory testing (EF, serum sodium, blood urea nitrogen, and B-type
natriuretic peptide), initial admission electrocardiogram (QRS duration and presence of atrial fibrillation), clinical characteristics (ischemic HF origin, coronary artery disease, diabetes, hypertension, chronic obstructive pulmonary disease, chronic kidney disease [CKD], and New York Heart Association class IV), and baseline medication use (allopurinol, angiotensin converting enzyme-inhibitors, angiotensin II receptor blockers, b blockers, mineralocorticoid receptor antagonists, digoxin, and intravenous inotropes).
1718
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Forest plots. Univariate subgroup analyses for ACM (A) and composite cardiovascular (CV) mortality and HF hospitalizations (B). The p values are displayed for each interaction term.
Figure 5. Changes in sUA levels over time. Baseline sUA was measured within 48 hours of hospital admission. The overall time course of mean sUA levels over the 112 weeks follow-up has been stratified by the presence and absence of ACM (A) and composite cardiovascular (CV) mortality and HF hospitalizations (B).
Multiple imputation procedures (fully conditional specification method) were used to impute any missing covariate data (w28% for natriuretic peptides, 4% for QRS duration, 2% for jugular venous distension, and 2% for all other variables). No evidence of significant collinearity between baseline sUA and the covariate set was detected. Separate interaction analyses were performed for gender, race, estimated glomerular filtration rate (eGFR) <30 ml/min/1.73 m2 (by the Modification of Diet in Renal Disease formula), and admission allopurinol use. This cutoff in eGFR was selected a priori because sUA does not appear to predict further declines in renal function in stage IV and V CKD18 and may be differentially associated with clinical end points in this population
compared with preserved or mildly impaired renal function.19,20 Because tolvaptan is known to increase sUA as early as day 1 of hospitalization,15,17 treatment assignment was also included as an interaction term. All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, North Carolina). Results Of the 4,133 patients enrolled in the EVEREST program, 178 patients had missing baseline sUA levels (4.3%) and were excluded. The remaining cohort (n ¼ 3,955) was divided into sUA quartiles Q1 (n ¼ 967, range 1.2 to 7.0 mg/dl), Q2 (n ¼ 1,012, range 7.1 to 8.8 mg/dl),
Heart Failure/Uric Acid and Hospitalized Heart Failure
Q3 (n ¼ 977, range 8.9 to 10.8 mg/dl), and Q4 (n ¼ 999, range 10.9 to 21.0 mg/dl). sUA was normally distributed with a mean of 9.1 2.8 mg/dl and median of 8.8 (interquartile range 7.1 to 10.9 mg/dl; Figure 2). sUA was higher in men than in women (9.3 2.7 vs 8.7 3.0 mg/dl; Supplementary Figure 1) and higher in blacks than in whites (10.0 2.7 vs 9.0 2.8 mg/dl; Supplementary Figure 2). Table 1 presents the baseline characteristics by sUA quartiles. Patients in the highest sUA quartile were more likely to be younger, men, black, and from North America (all comparisons p <0.001). On admission, higher sUA was associated with lower systolic and diastolic blood pressures (p <0.001), higher heart rates (p ¼ 0.02), and lower EF (p <0.001). Patients in the highest sUA quartiles had more jugular venous distension and elevated natriuretic peptides (both p <0.001). These patients more frequently had a history of stage IV and V CKD with elevated blood urea nitrogen and serum creatinine on admission (p <0.001). Higher sUA was associated with less ischemic HF origin (p ¼ 0.03), New York Heart Association class IV symptoms (p <0.001), and increased previous coronary artery bypass grafting and percutaneous coronary interventions (p <0.001). Allopurinol was used in 440 patients at admission, ranging from 15.4% in quartile 1 to 8.5% in quartile 4 (p <0.001). Admission angiotensin converting enzyme-inhibitor and angiotensin II receptor blocker use was less frequently reported in higher sUA quartiles, whereas intravenous inotrope use was higher (p <0.001 for both). Similar patterns and trends were observed across gender and race subgroups (Supplementary Tables 1 to 4). There were a total of 1,040 ACM events and 1,627 composite end points (cardiovascular mortality or HF hospitalization) over the median follow-up timeframe of 9.9 months. Rates of ACM increased with increasing sUA levels in a step-wise fashion from quartile 1 (18.6%) to quartile 4 (39.1%, p <0.001; Table 2). Similarly, rates of the composite end point increased from 30.3% to 54.8% (p <0.001). Analysis of secondary end points revealed similar trends of increasing event rates with higher sUA levels for cardiovascular mortality, worsening HF, HF mortality, and hospitalization (p <0.001 for all end points). Rates of myocardial infarction, stroke morbidity, or mortality did not differ by sUA quartile. Times to first event were also significantly different by the Kaplan-Meier method across sUA quartiles for ACM (Figure 3; log rank p <0.001) and the composite end point (Figure 3; log rank p ¼ 0.004). Compared with the lowest quartile of sUA, the adjusted hazard increased with higher sUA levels for ACM (Q2: HR 1.14, 95% CI 0.96 to 1.34, p ¼ 0.129; Q3: HR 1.22, 95% CI 1.02 to 1.46, p ¼ 0.033; Q4: HR 1.33, 95% CI 1.10 to 1.61, p ¼ 0.004) and the composite end point (Q2: HR 1.10, 95% CI 0.96 to 1.25, p ¼ 0.185; Q3: HR 1.23, 95% CI 1.07 to 1.43, p ¼ 0.005; Q4: HR 1.38, 95% CI 1.18 to 1.62, p <0.001). Tolvaptan assignment did not modify the association between sUA and the primary outcomes, ACM (interaction term p ¼ 0.31), and the composite end point (interaction term p ¼ 0.45). Thus, even in a placeborestricted subgroup, sUA (per 5 mg/dl increase) was robustly predictive of excess ACM (HR 1.75, 95% CI 1.51 to 2.02, p <0.001) and the composite end point (HR 1.69, 95% CI 1.50 to 1.90, p <0.001). Supplementary Figure 3
1719
depicts the individual adjusted HR for increasing deciles of sUA (p ¼ 0.007 for ACM trend and p <0.001 for composite end point trend). Figure 4 summarizes the unadjusted interaction analysis for gender, race, renal function, and allopurinol use. In the final multivariate analysis, only renal function significantly influenced the relation between sUA and clinical end points. In patients with enrollment eGFR 30 ml/min/1.73 m2, sUA (per 5 mg/dl increase) was strongly associated with increased hazard of ACM (HR 1.44, 95% CI 1.22 to 1.69, p <0.001) and the composite end point (HR 1.44, 95% CI 1.26 to 1.64; p <0.001). However, in patients with eGFR <30 ml/min/1.73 m2, sUA (per 5 mg/dl increase) was not related to neither ACM (HR 0.93, 95% CI 0.78 to 1.10, p ¼ 0.393) nor the composite end point (HR 1.00, 95% CI 0.87 to 1.16, p ¼ 0.978). When examining the temporal trends, sUA levels continued to decrease postdischarge to approximately 8 mg/ dl at 112 weeks follow-up. sUA was consistently higher in patients who experienced postdischarge clinical events than in those who did not; these differences persisted into the late postdischarge period (Figure 5). Discussion In this large contemporary cohort of patients hospitalized for WCHF, approximately half of patients had sUA levels >9.0 mg/dl. sUA levels decrease during hospitalization and into the postdischarge period, consistently across race and gender subgroups. sUA was higher in younger patients, men, blacks, and in patients enrolled in North American centers. Despite observed differences across these demographic subsets, gender and race did not modulate the association between sUA and postdischarge outcomes. Higher sUA was associated with markers of congestion, poor renal function, lower EF, and more advanced functional class. After accounting for the baseline differences in risk profiles and multiple interaction terms, baseline sUA was associated with worse clinical end points in patients with eGFR 30 ml/min/1.73 m2 at the time of enrollment, but not in those with eGFR <30 ml/min/1.73 m2. To our knowledge, this is the largest study to date to evaluate the prognostic use of sUA in patients hospitalized for WCHF. Our data from the EVEREST trial provide a number of advantages to the current literature: (1) Comprehensive characterization of a high-risk cohort with a high number of postdischarge events; (2) Global, multicenter patient sampling; (3) Robust multivariate analysis including 24 clinical and laboratory parameters with limited missing data; (4) Reliable assessment of sUA in a centralized core laboratory; (5) Serial, longitudinal measurement of sUA levels; and (6) Specific interaction analyses for gender, race, and renal function. Based on our study findings, sUA represents a strong, independent prognostic variable in patients with WCHF with relatively preserved baseline renal function. In a recent meta-analysis, incremental elevations in sUA (by 1 mg/dL) increased the odds of incident HF by w19% in the general population and increased risk of ACM by w4% in patients with existing chronic HF.21 These findings have been extended to patients admitted for WCHF with
1720
The American Journal of Cardiology (www.ajconline.org)
sUA measured in the emergency department,6 during hospitalization3,4,6e9,11 and predischarge5,10 strongly correlated with in-hospital and postdischarge outcomes. A retrospective population-based study from Israel including 8,246 patients showed that the addition of laboratory parameters including sUA improved 1-year mortality risk prediction.8 In 11,681 men enrolled in the Multiple Risk Factor Intervention Trial, WCHF admissions and increased diuretic requirement were associated with increased sUA levels, which improved with hospital discharge and diuretic discontinuation.2 The decrease in mean sUA throughout the postdischarge period seen in our study may be partially explained by increased survivorship in patients with low sUA levels. Most of these studies are limited by smaller numbers of patients from single institutions, incomplete statistical accounting, and short-term follow-up. Enhanced xanthine oxidase activity, in response to hypoxia and inflammation, appears to be a major source of myocardial and vascular oxidative stress in patients with HF. Consistently, sUA is significantly associated with several inflammatory markers, including C-reactive protein and interleukin-6 in chronic HF.22 Interestingly, UA itself is associated with enhanced free-radical scavenging and endothelium-dependent vasodilation in HF.23 However, sUA may be a simple byproduct of xanthine oxidase enzymatic upregulation, produced along with multiple reactive oxygen species. Thus, sUA may serve as an important index of impaired oxidative metabolism, which at least partially mediates myocardial dysfunction and reduced functional capacity.24 sUA was initially believed to represent simply a marker of poor renal function.12 However, our data suggest that sUA has differential prognostic use based on in-hospital eGFR. These findings corroborate data from a propensityscore matched analysis from the Beta-Blocker Evaluation of Survival Trial showing that hyperuricemia was associated with ACM and HF hospitalization only in patients with chronic HF without CKD, but not in those with CKD.20 This suggests that elevated sUA is associated with adverse outcomes, when it is a result of increased production rather than decreased clearance alone.20 In fact, limited data have demonstrated that sUA is at least partially secreted by the failing myocardium in HF. In a small Japanese cohort, the transcardiac gradient in sUA directly sampled from the aortic root and coronary sinus increased with severity of HF.25 Most sUA-lowering therapies (uricosuric) without xanthine oxidase inhibition have not been shown to improve clinical severity, functional capacity, hemodynamics, and prognosis of patients with HF in large, prospectively conducted clinical studies.26 Accumulating data, supporting the prognostic use of sUA in HF have made xanthine oxidaseinhibition an attractive therapeutic target. Treatment with allopurinol in a large Israeli cohort of patients with chronic HF was independently associated with improved survival at median follow-up of 1.4 years.27 In contrast, the Oxypurinol Compared With Placebo for Class III-IV NYHA Congestive Heart Failure trial showed that treatment with oxypurinol did not influence composite morbidity and mortality at 24 weeks.28 A recent meta-analysis of 11 trials and over 20,000 subjects followed for 2 years showed that sUA
changes during pharmacologic treatment did not predict cardiovascular outcomes.29 We await the results of the recently completed EXACT-HF (Xanthine Oxidase Inhibition for Hyperuricemic Heart Failure Patients; NCT00987415) trial, which is a multicenter randomized 24-week trial of allopurinol in hyperuricemic (sUA 9.5 mg/dl) patients with chronic HF with reduced EF.30 The post hoc nature of the present study makes the findings vulnerable to selection bias and confounding from measured and unmeasured parameters. Robust multivariate modeling may not be sufficient to account for the vast differences in clinical profiles of patients with differing sUA levels. The EVEREST population represents a highly selected cohort and notably excluded patients with severe CKD (serum creatinine >3.5 mg/dl). In WCHF, treatment with diuretics has a known effect on sUA levels; however, our multivariate models did not account for this parameter given that >95% of patients were on diuretics before or during hospitalization. Data regarding specific in-hospital diuretic dosages and postdischarge allopurinol use were not available for analysis. More data are required regarding the prognostic use of sUA in HF with preserved EF and new onset HF. Despite these limitations, this simple, inexpensive, widely available, reliably measured metric may serve as a valuable predictive biomarker, beyond traditional markers of prognosis. The differences in sUA levels and attendant prognostic use in important patient subgroups may inform selection in future clinical trials. Our hypothesisgenerating work calls for a prospective, randomized, controlled trial evaluating xanthine oxidase-inhibition in the high-risk population of patients hospitalized for WCHF. Disclosures The author Haris P. Subacius conducted all final analyses for this manuscript with funding from the Center for Cardiovascular Innovation (Northwestern University Feinberg School of Medicine, Chicago, Illinois). The authors had full access to the data, take responsibility for its integrity, and had complete control and authority over manuscript preparation and the decision to publish. Dr. Mentz has received research support from Gilead and honoraria from Thoratec, HeartWare, BMS, and Novartis. Dr. Chioncel has received research support from Abbott, Servier, and Vifor and served as member on the Steering Committee of studies sponsored by Novartis. Dr. Maggioni served as member on the Steering Committee of studies sponsored by Bayer, Novartis, Otsuka, Cardiorentis, and Abbott Vascular. Dr. Zannad has had principal relation with Cardiorenal Diagnostics, served as a consultant for Air Liquide, St. Jude’s Medical, Boston Scientific, Servier, Novartis, and participated as a committee member for Takeda Pharmaceuticals, BIOTRONIK, Janssen, ResMed, Bayer, and Pfizer. Dr. Konstam discloses research support and/or consulting fees from Otsuka, Amgen, Johnson & Johnson Services, Inc., and Novartis. Dr. Givertz discloses funding related to the NIH/NHLBI HF Network. Dr. Butler has received research support from the National Institutes of Health, European Union, Health Resources and Services Administration, U.S. Food and Drug Administration, and served as a consultant for Amgen, Bayer, Celladon,
Heart Failure/Uric Acid and Hospitalized Heart Failure
Gambro, GE Healthcare, Janssen, Medtronic, Novartis, Ono, Relypsa, and Trevena. Dr. Gheorghiade has been a consultant for Abbott Laboratories, Astellas, AstraZeneca, Bayer HealthCare AG, Corthera, Cytokinetics, Debiopharm S.A., Errekappa Terapeutici, GlaxoSmithKline, Ikaria, Johnson & Johnson, Medtronic, Merck, Novartis Pharma AG, Otsuka Pharmaceuticals, Palatin Technologies, PeriCor Therapeutics, Protein Design Laboratories, Sanofi-Aventis, Sigma Tau, Solvay Pharmaceuticals, Takeda Pharmaceutical, and Trevena Therapeutics. All other authors have no conflicts of interest to declare. Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.09.008. 1. Gheorghiade M, Pang PS, Ambrosy AP, Lan G, Schmidt P, Filippatos G, Konstam M, Swedberg K, Cook T, Traver B, Maggioni A, Burnett J, Grinfeld L, Udelson J, Zannad F. A comprehensive, longitudinal description of the in-hospital and post-discharge clinical, laboratory, and neurohormonal course of patients with heart failure who die or are re-hospitalized within 90 days: analysis from the EVEREST trial. Heart Fail Rev 2012;17:485e509. 2. Misra D, Zhu Y, Zhang Y, Choi HK. The independent impact of congestive heart failure status and diuretic use on serum uric acid among men with a high cardiovascular risk profile: a prospective longitudinal study. Semin Arthritis Rheum 2011;41:471e476. 3. Alimonda AL, Nunez J, Nunez E, Husser O, Sanchis J, Bodi V, Minana G, Robles R, Mainar L, Merlos P, Darmofal H, Llacer A. Hyperuricemia in acute heart failure. More than a simple spectator? Eur J Intern Med 2009;20:74e79. 4. Cengel A, Turkoglu S, Turfan M, Boyaci B. Serum uric acid levels as a predictor of in-hospital death in patients hospitalized for decompensated heart failure. Acta Cardiol 2005;60:489e492. 5. Hamaguchi S, Furumoto T, Tsuchihashi-Makaya M, Goto K, Goto D, Yokota T, Kinugawa S, Yokoshiki H, Takeshita A, Tsutsui H; Investigators J-C. Hyperuricemia predicts adverse outcomes in patients with heart failure. Int J Cardiol 2011;151:143e147. 6. Henry-Okafor Q, Collins SP, Jenkins CA, Miller KF, Maron DJ, Naftilan AJ, Weintraub N, Fermann GJ, McPherson J, Menon S, Sawyer DB, Storrow AB. Relationship between uric acid levels and diagnostic and prognostic outcomes in acute heart failure. Open Biomark J 2012;5:9e15. 7. Malek F, Ostadal P, Parenica J, Jarkovsky J, Vitovec J, Widimsky P, Linhart A, Fedorco M, Coufal Z, Miklik R, Kruger A, Vondrakova D, Spinar J. Uric acid, allopurinol therapy, and mortality in patients with acute heart failureeresults of the Acute HEart FAilure Database registry. J Crit Care 2012;27:737.e11e737.e24. 8. Novack V, Pencina M, Zahger D, Fuchs L, Nevzorov R, Jotkowitz A, Porath A. Routine laboratory results and thirty day and one-year mortality risk following hospitalization with acute decompensated heart failure. PLoS One 2010;5:e12184. 9. Park HS, Kim H, Sohn JH, Shin HW, Cho YK, Yoon HJ, Nam CW, Hur SH, Kim YN, Kim KB, Park HJ. Combination of uric acid and NT-ProBNP: a more useful prognostic marker for short-term clinical outcomes in patients with acute heart failure. Korean J Intern Med 2010;25:253e259. 10. Pascual-Figal DA, Hurtado-Martinez JA, Redondo B, Antolinos MJ, Ruiperez JA, Valdes M. Hyperuricaemia and long-term outcome after hospital discharge in acute heart failure patients. Eur J Heart Fail 2007;9:518e524. 11. Wasserman A, Shnell M, Boursi B, Guzner-Gur H. Prognostic significance of serum uric acid in patients admitted to the Department of Medicine. Am J Med Sci 2010;339:15e21. 12. Tian Y, Chen Y, Deng B, Liu G, Ji ZG, Zhao QZ, Zhen YZ, Gao YQ, Tian L, Wang L, Ji LS, Ma GP, Liu KS, Liu C. Serum uric acid as an index of impaired renal function in congestive heart failure. J Geriatr Cardiol 2012;9:137e142.
1721
13. Maynard JW, McAdams-Demarco MA, Law A, Kao L, Gelber AC, Coresh J, Baer AN. Racial differences in gout incidence in a population-based cohort: Atherosclerosis Risk in Communities Study. Am J Epidemiol 2014;179:576e583. 14. Puig JG, Michan AD, Jimenez ML, Perez de Ayala C, Mateos FA, Capitan CF, de Miguel E, Gijon JB. Female gout. Clinical spectrum and uric acid metabolism. Arch Intern Med 1991;151:726e732. 15. Gheorghiade M, Konstam MA, Burnett JC Jr, Grinfeld L, Maggioni AP, Swedberg K, Udelson JE, Zannad F, Cook T, Ouyang J, Zimmer C, Orlandi C. Short-term clinical effects of tolvaptan, an oral vasopressin antagonist, in patients hospitalized for heart failure: the EVEREST Clinical Status Trials. JAMA 2007;297:1332e1343. 16. Gheorghiade M, Orlandi C, Burnett JC, Demets D, Grinfeld L, Maggioni A, Swedberg K, Udelson JE, Zannad F, Zimmer C, Konstam MA. Rationale and design of the multicenter, randomized, doubleblind, placebo-controlled study to evaluate the Efficacy of Vasopressin antagonism in Heart Failure: Outcome Study with Tolvaptan (EVEREST). J Card Fail 2005;11:260e269. 17. Konstam MA, Gheorghiade M, Burnett JC Jr, Grinfeld L, Maggioni AP, Swedberg K, Udelson JE, Zannad F, Cook T, Ouyang J, Zimmer C, Orlandi C. Effects of oral tolvaptan in patients hospitalized for worsening heart failure: the EVEREST Outcome Trial. JAMA 2007;297:1319e1331. 18. Nacak H, van Diepen M, de Goeij MC, Rotmans JI, Dekker FW. Uric acid: association with rate of renal function decline and time until start of dialysis in incident pre-dialysis patients. BMC Nephrol 2014;15:91. 19. Sato T, Yamauchi H, Suzuki S, Yoshihisa A, Yamaki T, Sugimoto K, Kunii H, Nakazato K, Suzuki H, Saitoh S, Takeishi Y. Distinct prognostic factors in patients with chronic heart failure and chronic kidney disease. Int Heart J 2013;54:311e317. 20. Filippatos GS, Ahmed MI, Gladden JD, Mujib M, Aban IB, Love TE, Sanders PW, Pitt B, Anker SD, Ahmed A. Hyperuricaemia, chronic kidney disease, and outcomes in heart failure: potential mechanistic insights from epidemiological data. Eur Heart J 2011;32:712e720. 21. Huang H, Huang B, Li Y, Huang Y, Li J, Yao H, Jing X, Chen J, Wang J. Uric acid and risk of heart failure: a systematic review and metaanalysis. Eur J Heart Fail 2014;16:15e24. 22. Leyva F, Anker SD, Godsland IF, Teixeira M, Hellewell PG, Kox WJ, Poole-Wilson PA, Coats AJ. Uric acid in chronic heart failure: a marker of chronic inflammation. Eur Heart J 1998;19:1814e1822. 23. Alcaino H, Greig D, Chiong M, Verdejo H, Miranda R, Concepcion R, Vukasovic JL, Diaz-Araya G, Mellado R, Garcia L, Salas D, Gonzalez L, Godoy I, Castro P, Lavandero S. Serum uric acid correlates with extracellular superoxide dismutase activity in patients with chronic heart failure. Eur J Heart Fail 2008;10:646e651. 24. Leyva F, Anker S, Swan JW, Godsland IF, Wingrove CS, Chua TP, Stevenson JC, Coats AJ. Serum uric acid as an index of impaired oxidative metabolism in chronic heart failure. Eur Heart J 1997;18: 858e865. 25. Sakai H, Tsutamoto T, Tsutsui T, Tanaka T, Ishikawa C, Horie M. Serum level of uric acid, partly secreted from the failing heart, is a prognostic marker in patients with congestive heart failure. Circ J 2006;70:1006e1011. 26. Ogino K, Kato M, Furuse Y, Kinugasa Y, Ishida K, Osaki S, Kinugawa T, Igawa O, Hisatome I, Shigemasa C, Anker SD, Doehner W. Uric acid-lowering treatment with benzbromarone in patients with heart failure: a double-blind placebo-controlled crossover preliminary study. Circ Heart Fail 2010;3:73e81. 27. Gotsman I, Keren A, Lotan C, Zwas DR. Changes in uric acid levels and allopurinol use in chronic heart failure: association with improved survival. J Card Fail 2012;18:694e701. 28. Hare JM, Mangal B, Brown J, Fisher C Jr, Freudenberger R, Colucci WS, Mann DL, Liu P, Givertz MM, Schwarz RP; Investigators O-C. Impact of oxypurinol in patients with symptomatic heart failure. Results of the OPT-CHF study. J Am Coll Cardiol 2008;51:2301e2309. 29. Savarese G, Ferri C, Trimarco B, Rosano G, Dellegrottaglie S, Losco T, Casaretti L, D’Amore C, Gambardella F, Prastaro M, Rengo G, Leosco D, Perrone-Filardi P. Changes in serum uric acid levels and cardiovascular events: a meta-analysis. Nutr Metab Cardiovasc Dis 2013;23:707e714. 30. Givertz MM, Mann DL, Lee KL, Ibarra JC, Velazquez EJ, Hernandez AF, Mascette AM, Braunwald E. Xanthine oxidase inhibition for hyperuricemic heart failure patients: design and rationale of the EXACT-HF study. Circ Heart Fail 2013;6:862e868.
be strongly influenced by impaired renal function,12 race,13 gender,14 and diuretic therapy. The Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST)15e17 trial database provides detailed, longitudinal, patient-level data on sUA and other clinical parameters during and after hospitalization for WCHF. Thus, we aimed to evaluate the independent association between sUA at the time of enrollment and clinical characteristics and
a Department of Medicine, Massachusetts General Hospital, Boston, Massachusetts; bCenter for Cardiovascular Innovation, Northwestern University Feinberg School of Medicine, Chicago, Illinois; cDepartment of Medicine, Stanford University School of Medicine, Stanford, California; d Duke University Medical Center, Durham, North Carolina; eDivision of Cardiology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, Illinois; fInstitute of Emergency for Cardiovascular Diseases “Prof. C.C. Iliescu”, Cardiology, Bucharest, Romania; gResearch Center of the Italian Association of Hospital Cardiologists (ANMCO), Florence, Italy; hUniversity of Gothenburg, Gothenburg, Sweden; iDepartment of Cardiology, Nancy University, Nancy, France; jTufts Medical Center, Boston, Massachusetts; kDipartimento
Cardiovascolare, Azienda Ospedaliera Papa Giovanni XXIII, Bergamo, Italy; lCardiovascular Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts; and mDivision of Cardiology, Emory University School of Medicine, Atlanta, Georgia. Manuscript received July 21, 2014; revised manuscript received and accepted September 2, 2014. Otsuka Inc. (Rockville, Maryland) provided financial and material support for the EVEREST trial. Database management was performed by the sponsor. See page 1720 for disclosure information. *Corresponding author: Tel: (þ832) 725-7222; fax: (þ617) 726-6861. E-mail address: [email protected] (M. Vaduganathan).
0002-9149/14/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjcard.2014.09.008
www.ajconline.org
1714
The American Journal of Cardiology (www.ajconline.org)
Figure 1. Overall study design and analytical cohort selection. LVEF ¼ left ventricular ejection fraction; NYHA ¼ New York Heart Association.
Figure 2. Distribution and descriptive statistics of sUA levels. The primary predictor variable of baseline sUA was normally distributed with a mean and median of approximately 9 mg/dl. The black curve represents the normal density distribution and the gray curve represents the kernel density distribution for this sample. IQR ¼ interquartile range.
postdischarge outcomes in patients hospitalized for WCHF with reduced ejection fraction (EF). Methods The study design16 and primary results15,17 of the EVEREST trial have been previously described. In brief, EVEREST was a global, multicenter, double-blinded, placebo-controlled randomized trial examining tolvaptan, an oral vasopressin-2 receptor antagonist. Patients eligible for enrollment were 18 years of age, hospitalized for WCHF
with New York Heart Association III-IV functional status, EF 40% and presenting with 2 signs or symptoms of volume overload. Relevant exclusion criteria include serum creatinine >3.5 mg/dl; subjects currently treated with hemofiltration or dialysis; refractory, end-stage HF; or a life expectancy <6 months. The ethics committee and institutional review board of each participating trial center approved the study protocol. Patients were randomized to receive oral tolvaptan 30 mg fixed once daily or matching placebo within 48 hours of hospital admission and was continued for at least 60 days. Laboratory samples were
Heart Failure/Uric Acid and Hospitalized Heart Failure
1715
Table 1 Baseline characteristics by serum uric acid quartile Variable
Serum uric acid, (mg/dL), meanSD Serum uric acid, (mg/dL), range Tolvaptan assignment Age (years), meanSD Male Race Non-Hispanic White Black Asian Hispanic Other Region Eastern Europe North America South America Western Europe Weight (kg), meanSD Systolic blood pressure (mmHg), meanSD Diastolic blood pressure (mmHg), meanSD Heart rate (bpm), meanSD Jugular venous distension 10 cm Rales Peripheral edema* Dyspnea Ejection fraction (%), meanSD Blood urea nitrogen (mg/dL), median (IQR) Creatinine (mg/dL), median (IQR) Estimated GFR (mL/min/1.73 m2), meanSD † Estimated GFR<30 (mL/min/1.73 m2) † Sodium (mEq/L), median (IQR) B-type natriuretic peptide (pg/ml), median (IQR) N-terminal pro-B-type natriuretic peptide (pg/ml) median (IQR) Albumin (g/dL), median (IQR) QRS (ms), median (IQR) Atrial fibrillation Previous heart failure hospitalization Ischemic heart failure etiology NYHA class IV Coronary artery disease Previous myocardial infarction Hypertension Hypercholesterolemia Diabetes Chronic kidney disease Peripheral vascular disease Previous coronary bypass Previous percutaneous coronary intervention Automatic implantable cardioverter-defibrillator Chronic obstructive pulmonary disease Baseline medication use Allopurinol Diuretics
Baseline Serum Uric Acid Quartiles
p-value
I (n¼967)
II (n¼1,012)
III (n¼977)
5.81 1.2-7.0 514 (53.2%) 67.110.7 637 (65.9%)
80.5 7.1-8.8 494 (48.8%) 66.111.8 754 (74.5%)
9.80.6 8.9-10.8 464 (47.5%) 65.412 774 (79.2%)
12.91.8 10.9-21.0 504 (50.5%) 64.512.7 780 (78.1%)
869 42 1 40 16
876 63 1 59 14
836 73 2 47 21
816 107 5 45 31
(89.9%) (4.3%) (0.1%) (4.1%) (1.7%)
(86.6%) (6.2%) (0.1%) (5.8%) (1.4%)
(85.6%) (7.5%) (0.2%) (4.8%) (2.1%)
IV (n¼999)
0.073 <0.001 <0.001 <0.001
(81.7%) (10.7%) (0.5%) (4.5%) (3.1%) <0.001
494 (51.1%) 205 (21.2%) 144 (14.9%) 124 (12.8%) 79.818.2 124.920.5 74.312.5 78.814.8 196 (20.7%) 801 (83.7%) 778 (81.3%) 882 (92.3%) 29.67.8
452 (44.7%) 268 (26.5%) 172 (17%) 120 (11.9%) 8218.1 121.719.6 7312.4 79.215.5 248 (24.8%) 823 (82.1%) 789 (78.7%) 897 (89.5%) 28.27.8
379 (38.8%) 298 (30.5%) 171 (17.5%) 129 (13.2%) 8518.1 120.319.6 7312.9 80.316.4 291 (30.4%) 772 (79.9%) 767 (79.3%) 891 (92.2%) 27.18
245 (24.5%) 415 (41.5%) 189 (18.9%) 150 (15%) 86.220.4 115.217.9 70.412.7 80.715.7 308 (32.1%) 774 (79.8%) 783 (80.6%) 874 (90.4%) 25.48.2
<0.001 <0.001 <0.001 0.019 <0.001 0.083 0.450 0.078 <0.001
20 (17-27) 1 (0.9-1.3)
24 (19-31) 1.2 (1-1.4)
27 (21-35) 1.3 (1.1-1.6)
35 (26-48) 1.5 (1.3-1.9)
<0.001 <0.001
65.821.9 58 (6.1%) 140 (138-143) 481 (194-1137) 4070 (1650-8824) 3.8 (3.5-4.1)
60.320.1 53 (5.3%) 140 (138-142) 538.5 (254-1284) 3641 (1661-7468) 3.8 (3.4-4.1)
53.818.9 88 (9.2%) 140 (137-143) 746 (316-1474) 4256 (2241-8662) 3.8 (3.4-4.2)
44.417.4 223 (22.7%) 139 (136-142) 1106.5 (543-2126) 6800 (3642-12837) 3.7 (3.3-4)
<0.001 <0.001 <0.001
<0.001
121 258 740 659 323 689 485 693 450 379 137 205 159 141 125 73
121 277 783 653 371 725 518 717 491 387 216 223 207 174 127 98
123 311 786 641 401 687 506 693 473 363 280 191 209 181 152 100
125 288 810 614 468 685 492 705 499 401 421 209 253 205 173 124
<0.001 0.060 0.019 0.031 <0.001 0.423 0.678 0.958 0.549 0.563 <0.001 0.585 <0.001 0.006 0.007 0.004
(95-144) (26.7%) (76.7%) (68.9%) (33.4%) (71.4%) (50.2%) (71.7%) (46.9%) (39.2%) (14.2%) (21.2%) (16.4%) (14.6%) (12.9%) (7.5%)
149 (15.4%) 931 (96.5%)
(97-150) (27.4%) (77.7%) (65.4%) (36.7%) (71.6%) (51.2%) (70.8%) (48.8%) (38.2%) (21.3%) (22.1%) (20.5%) (17.2%) (12.5%) (9.7%)
110 (10.9%) 987 (97.7%)
(97-152) (31.8%) (80.9%) (66.4%) (41%) (70.4%) (51.8%) (70.9%) (48.6%) (37.2%) (28.7%) (19.5%) (21.4%) (18.5%) (15.6%) (10.2%)
96 (9.8%) 951 (97.4%)
(100-155) (28.8%) (81.5%) (62.5%) (47%) (68.6%) (49.3%) (70.6%) (50.2%) (40.1%) (42.2%) (21%) (25.3%) (20.5%) (17.3%) (12.4%)
85 (8.5%) 976 (98%)
<0.001 <0.001
<0.001 0.166 (continued)
1716
The American Journal of Cardiology (www.ajconline.org)
Table 1 (continued) Variable
Baseline Serum Uric Acid Quartiles I (n¼967)
ACEI/ ARB Beta-Blockers Mineralocorticoid receptor antagonists Digoxin Intravenous Inotropes Statin
823 667 511 438 33 345
II (n¼1,012)
(85.3%) (69.1%) (53%) (45.4%) (3.4%) (35.8%)
880 725 573 496 32 368
p-value
III (n¼977)
(87.1%) (71.8%) (56.7%) (49.1%) (3.2%) (36.4%)
837 706 550 493 44 319
(85.8%) (72.3%) (56.4%) (50.5%) (4.5%) (32.7%)
IV (n¼999) 796 696 516 489 71 339
(79.9%) (69.9%) (51.8%) (49.1%) (7.1%) (34%)
<0.001 0.346 0.065 0.132 <0.001 0.289
ACEI ¼ angiotensin converting enzyme inhibitor; ARB ¼ angiotensin II receptor blocker; GFR ¼ glomerular filtration rate; IQR ¼ interquartile range; NYHA ¼ New York Heart Association; SD ¼ standard deviation. * Peripheral edema was defined as slight/ moderate/ marked pedal or sacral edema. † Glomerular filtration rate estimated by Cockcroft-Gault formula. Table 2 Causes of death and rehospitalization by baseline serum uric acid quartile Baseline Serum Uric Acid Quartiles
Serum uric acid, (mg/dL), mean (SD) Serum uric acid, (mg/dL), range Primary Endpoints All-cause mortality CV mortality þ HF hospitalization Secondary Endpoints CV mortality CV mortality þ CV hospitalization Worsening HF * HF mortality HF hospitalization MI mortality MI hospitalization Stroke mortality Stroke hospitalization
p-value
I (n¼967)
II (n¼1,012)
III (n¼977)
IV (n¼999)
5.81 1.2-7.0
80.5 7.1-8.8
9.80.6 8.9-10.8
12.91.8 10.9-21.0
180 (18.6%) 293 (30.3%)
212 (20.9%) 362 (35.8%)
257 (26.3%) 425 (43.5%)
391 (39.1%) 547 (54.8%)
<0.001 <0.001
136 354 256 70 192 5 9 4 12
158 445 308 81 245 8 11 8 15
211 478 371 102 283 7 16 6 14
296 603 496 179 377 7 6 5 8
<0.001 <0.001 <0.001 <0.001 <0.001 0.900 0.153 0.713 0.502
(14.1%) (36.6%) (26.5%) (7.2%) (19.9%) (0.5%) (0.9%) (0.4%) (1.2%)
(15.6%) (44%) (30.4%) (8%) (24.2%) (0.8%) (1.1%) (0.8%) (1.5%)
(21.6%) (48.9%) (38%) (10.4%) (29%) (0.7%) (1.6%) (0.6%) (1.4%)
(29.6%) (60.4%) (49.6%) (17.9%) (37.7%) (0.7%) (0.6%) (0.5%) (0.8%)
CV ¼ cardiovascular; HF ¼ heart failure; MI ¼ myocardial infarction. * Worsening heart failure was defined as death from heart failure, hospitalization for heart failure, or unscheduled medical office visit for heart failure.
collected, processed, and cross validated across 5 central facilities. sUA (mg/dl) was measured at the time of study enrollment (baseline, up to 48 hours after admission) and every 4 to 8 weeks up to 112 weeks postdischarge. There was little evidence for nonlinearity in the relation between sUA and clinical end points, across a physiological range of sUA. Enrollment sUA levels were divided into quartiles and effect sizes are presented in reference to quartile 1 (lowest). For complete interaction analyses, sUA was treated as a continuous function, and effect sizes are presented per 5 mg/dl increase in sUA. Serial postdischarge sUA levels are presented by presence or absence of the primary end points. The overall study design and final analytical cohort selection are displayed in Figure 1. Demographic characteristics including self-reported race, signs and/or symptoms of HF, vital signs, laboratory parameters, initial electrocardiographic findings, medical history, and admission medications were compared across quartiles of baseline sUA levels. Baseline characteristics and
clinical outcomes were also presented separately for men, women, whites, and blacks. An independent blinded adjudication committee determined the specific causes of death and reasons for rehospitalization. The present post hoc analysis used the same 2 co-primary end points as the overall EVEREST trial: all-cause mortality (ACM) and a composite end point of cardiovascular mortality or HF hospitalization. Secondary end points included other causes of death and rehospitalization, worsening HF (defined as death, hospitalization, or unplanned office visit for HF), and combined cardiovascular mortality and rehospitalization. Median follow-up in the EVEREST trial was 9.9 months (interquartile range 5.3 to 16.1 months). Continuous variables are expressed as mean SD if normally distributed and as median (interquartile range) if non-normally distributed. Categorical variables are expressed as number (%). Outcomes were assessed as time to first event using Cox proportional hazard models. KaplanMeier curves by sUA quartile were constructed for the both primary end points and compared using log-rank tests. The
Heart Failure/Uric Acid and Hospitalized Heart Failure
1717
Figure 3. Kaplan Meier curves. Event curves for ACM (A) and composite cardiovascular mortality and HF hospitalizations (B). Times to events were compared using log-rank tests.
proportional hazards assumption (by Kolmogorov-type supremum tests for nonproportionality) was upheld. Effect sizes were reported as hazard ratios (HR) with 95% confidence intervals (CI). Multivariate models included 24 prespecified covariates including tolvaptan treatment assignment, demographic characteristics (age, gender, and region of origin), vital signs on admission (supine systolic blood pressure), laboratory testing (EF, serum sodium, blood urea nitrogen, and B-type
natriuretic peptide), initial admission electrocardiogram (QRS duration and presence of atrial fibrillation), clinical characteristics (ischemic HF origin, coronary artery disease, diabetes, hypertension, chronic obstructive pulmonary disease, chronic kidney disease [CKD], and New York Heart Association class IV), and baseline medication use (allopurinol, angiotensin converting enzyme-inhibitors, angiotensin II receptor blockers, b blockers, mineralocorticoid receptor antagonists, digoxin, and intravenous inotropes).
1718
The American Journal of Cardiology (www.ajconline.org)
Figure 4. Forest plots. Univariate subgroup analyses for ACM (A) and composite cardiovascular (CV) mortality and HF hospitalizations (B). The p values are displayed for each interaction term.
Figure 5. Changes in sUA levels over time. Baseline sUA was measured within 48 hours of hospital admission. The overall time course of mean sUA levels over the 112 weeks follow-up has been stratified by the presence and absence of ACM (A) and composite cardiovascular (CV) mortality and HF hospitalizations (B).
Multiple imputation procedures (fully conditional specification method) were used to impute any missing covariate data (w28% for natriuretic peptides, 4% for QRS duration, 2% for jugular venous distension, and 2% for all other variables). No evidence of significant collinearity between baseline sUA and the covariate set was detected. Separate interaction analyses were performed for gender, race, estimated glomerular filtration rate (eGFR) <30 ml/min/1.73 m2 (by the Modification of Diet in Renal Disease formula), and admission allopurinol use. This cutoff in eGFR was selected a priori because sUA does not appear to predict further declines in renal function in stage IV and V CKD18 and may be differentially associated with clinical end points in this population
compared with preserved or mildly impaired renal function.19,20 Because tolvaptan is known to increase sUA as early as day 1 of hospitalization,15,17 treatment assignment was also included as an interaction term. All statistical analyses were performed using SAS version 9.2 (SAS Institute Inc, Cary, North Carolina). Results Of the 4,133 patients enrolled in the EVEREST program, 178 patients had missing baseline sUA levels (4.3%) and were excluded. The remaining cohort (n ¼ 3,955) was divided into sUA quartiles Q1 (n ¼ 967, range 1.2 to 7.0 mg/dl), Q2 (n ¼ 1,012, range 7.1 to 8.8 mg/dl),
Heart Failure/Uric Acid and Hospitalized Heart Failure
Q3 (n ¼ 977, range 8.9 to 10.8 mg/dl), and Q4 (n ¼ 999, range 10.9 to 21.0 mg/dl). sUA was normally distributed with a mean of 9.1 2.8 mg/dl and median of 8.8 (interquartile range 7.1 to 10.9 mg/dl; Figure 2). sUA was higher in men than in women (9.3 2.7 vs 8.7 3.0 mg/dl; Supplementary Figure 1) and higher in blacks than in whites (10.0 2.7 vs 9.0 2.8 mg/dl; Supplementary Figure 2). Table 1 presents the baseline characteristics by sUA quartiles. Patients in the highest sUA quartile were more likely to be younger, men, black, and from North America (all comparisons p <0.001). On admission, higher sUA was associated with lower systolic and diastolic blood pressures (p <0.001), higher heart rates (p ¼ 0.02), and lower EF (p <0.001). Patients in the highest sUA quartiles had more jugular venous distension and elevated natriuretic peptides (both p <0.001). These patients more frequently had a history of stage IV and V CKD with elevated blood urea nitrogen and serum creatinine on admission (p <0.001). Higher sUA was associated with less ischemic HF origin (p ¼ 0.03), New York Heart Association class IV symptoms (p <0.001), and increased previous coronary artery bypass grafting and percutaneous coronary interventions (p <0.001). Allopurinol was used in 440 patients at admission, ranging from 15.4% in quartile 1 to 8.5% in quartile 4 (p <0.001). Admission angiotensin converting enzyme-inhibitor and angiotensin II receptor blocker use was less frequently reported in higher sUA quartiles, whereas intravenous inotrope use was higher (p <0.001 for both). Similar patterns and trends were observed across gender and race subgroups (Supplementary Tables 1 to 4). There were a total of 1,040 ACM events and 1,627 composite end points (cardiovascular mortality or HF hospitalization) over the median follow-up timeframe of 9.9 months. Rates of ACM increased with increasing sUA levels in a step-wise fashion from quartile 1 (18.6%) to quartile 4 (39.1%, p <0.001; Table 2). Similarly, rates of the composite end point increased from 30.3% to 54.8% (p <0.001). Analysis of secondary end points revealed similar trends of increasing event rates with higher sUA levels for cardiovascular mortality, worsening HF, HF mortality, and hospitalization (p <0.001 for all end points). Rates of myocardial infarction, stroke morbidity, or mortality did not differ by sUA quartile. Times to first event were also significantly different by the Kaplan-Meier method across sUA quartiles for ACM (Figure 3; log rank p <0.001) and the composite end point (Figure 3; log rank p ¼ 0.004). Compared with the lowest quartile of sUA, the adjusted hazard increased with higher sUA levels for ACM (Q2: HR 1.14, 95% CI 0.96 to 1.34, p ¼ 0.129; Q3: HR 1.22, 95% CI 1.02 to 1.46, p ¼ 0.033; Q4: HR 1.33, 95% CI 1.10 to 1.61, p ¼ 0.004) and the composite end point (Q2: HR 1.10, 95% CI 0.96 to 1.25, p ¼ 0.185; Q3: HR 1.23, 95% CI 1.07 to 1.43, p ¼ 0.005; Q4: HR 1.38, 95% CI 1.18 to 1.62, p <0.001). Tolvaptan assignment did not modify the association between sUA and the primary outcomes, ACM (interaction term p ¼ 0.31), and the composite end point (interaction term p ¼ 0.45). Thus, even in a placeborestricted subgroup, sUA (per 5 mg/dl increase) was robustly predictive of excess ACM (HR 1.75, 95% CI 1.51 to 2.02, p <0.001) and the composite end point (HR 1.69, 95% CI 1.50 to 1.90, p <0.001). Supplementary Figure 3
1719
depicts the individual adjusted HR for increasing deciles of sUA (p ¼ 0.007 for ACM trend and p <0.001 for composite end point trend). Figure 4 summarizes the unadjusted interaction analysis for gender, race, renal function, and allopurinol use. In the final multivariate analysis, only renal function significantly influenced the relation between sUA and clinical end points. In patients with enrollment eGFR 30 ml/min/1.73 m2, sUA (per 5 mg/dl increase) was strongly associated with increased hazard of ACM (HR 1.44, 95% CI 1.22 to 1.69, p <0.001) and the composite end point (HR 1.44, 95% CI 1.26 to 1.64; p <0.001). However, in patients with eGFR <30 ml/min/1.73 m2, sUA (per 5 mg/dl increase) was not related to neither ACM (HR 0.93, 95% CI 0.78 to 1.10, p ¼ 0.393) nor the composite end point (HR 1.00, 95% CI 0.87 to 1.16, p ¼ 0.978). When examining the temporal trends, sUA levels continued to decrease postdischarge to approximately 8 mg/ dl at 112 weeks follow-up. sUA was consistently higher in patients who experienced postdischarge clinical events than in those who did not; these differences persisted into the late postdischarge period (Figure 5). Discussion In this large contemporary cohort of patients hospitalized for WCHF, approximately half of patients had sUA levels >9.0 mg/dl. sUA levels decrease during hospitalization and into the postdischarge period, consistently across race and gender subgroups. sUA was higher in younger patients, men, blacks, and in patients enrolled in North American centers. Despite observed differences across these demographic subsets, gender and race did not modulate the association between sUA and postdischarge outcomes. Higher sUA was associated with markers of congestion, poor renal function, lower EF, and more advanced functional class. After accounting for the baseline differences in risk profiles and multiple interaction terms, baseline sUA was associated with worse clinical end points in patients with eGFR 30 ml/min/1.73 m2 at the time of enrollment, but not in those with eGFR <30 ml/min/1.73 m2. To our knowledge, this is the largest study to date to evaluate the prognostic use of sUA in patients hospitalized for WCHF. Our data from the EVEREST trial provide a number of advantages to the current literature: (1) Comprehensive characterization of a high-risk cohort with a high number of postdischarge events; (2) Global, multicenter patient sampling; (3) Robust multivariate analysis including 24 clinical and laboratory parameters with limited missing data; (4) Reliable assessment of sUA in a centralized core laboratory; (5) Serial, longitudinal measurement of sUA levels; and (6) Specific interaction analyses for gender, race, and renal function. Based on our study findings, sUA represents a strong, independent prognostic variable in patients with WCHF with relatively preserved baseline renal function. In a recent meta-analysis, incremental elevations in sUA (by 1 mg/dL) increased the odds of incident HF by w19% in the general population and increased risk of ACM by w4% in patients with existing chronic HF.21 These findings have been extended to patients admitted for WCHF with
1720
The American Journal of Cardiology (www.ajconline.org)
sUA measured in the emergency department,6 during hospitalization3,4,6e9,11 and predischarge5,10 strongly correlated with in-hospital and postdischarge outcomes. A retrospective population-based study from Israel including 8,246 patients showed that the addition of laboratory parameters including sUA improved 1-year mortality risk prediction.8 In 11,681 men enrolled in the Multiple Risk Factor Intervention Trial, WCHF admissions and increased diuretic requirement were associated with increased sUA levels, which improved with hospital discharge and diuretic discontinuation.2 The decrease in mean sUA throughout the postdischarge period seen in our study may be partially explained by increased survivorship in patients with low sUA levels. Most of these studies are limited by smaller numbers of patients from single institutions, incomplete statistical accounting, and short-term follow-up. Enhanced xanthine oxidase activity, in response to hypoxia and inflammation, appears to be a major source of myocardial and vascular oxidative stress in patients with HF. Consistently, sUA is significantly associated with several inflammatory markers, including C-reactive protein and interleukin-6 in chronic HF.22 Interestingly, UA itself is associated with enhanced free-radical scavenging and endothelium-dependent vasodilation in HF.23 However, sUA may be a simple byproduct of xanthine oxidase enzymatic upregulation, produced along with multiple reactive oxygen species. Thus, sUA may serve as an important index of impaired oxidative metabolism, which at least partially mediates myocardial dysfunction and reduced functional capacity.24 sUA was initially believed to represent simply a marker of poor renal function.12 However, our data suggest that sUA has differential prognostic use based on in-hospital eGFR. These findings corroborate data from a propensityscore matched analysis from the Beta-Blocker Evaluation of Survival Trial showing that hyperuricemia was associated with ACM and HF hospitalization only in patients with chronic HF without CKD, but not in those with CKD.20 This suggests that elevated sUA is associated with adverse outcomes, when it is a result of increased production rather than decreased clearance alone.20 In fact, limited data have demonstrated that sUA is at least partially secreted by the failing myocardium in HF. In a small Japanese cohort, the transcardiac gradient in sUA directly sampled from the aortic root and coronary sinus increased with severity of HF.25 Most sUA-lowering therapies (uricosuric) without xanthine oxidase inhibition have not been shown to improve clinical severity, functional capacity, hemodynamics, and prognosis of patients with HF in large, prospectively conducted clinical studies.26 Accumulating data, supporting the prognostic use of sUA in HF have made xanthine oxidaseinhibition an attractive therapeutic target. Treatment with allopurinol in a large Israeli cohort of patients with chronic HF was independently associated with improved survival at median follow-up of 1.4 years.27 In contrast, the Oxypurinol Compared With Placebo for Class III-IV NYHA Congestive Heart Failure trial showed that treatment with oxypurinol did not influence composite morbidity and mortality at 24 weeks.28 A recent meta-analysis of 11 trials and over 20,000 subjects followed for 2 years showed that sUA
changes during pharmacologic treatment did not predict cardiovascular outcomes.29 We await the results of the recently completed EXACT-HF (Xanthine Oxidase Inhibition for Hyperuricemic Heart Failure Patients; NCT00987415) trial, which is a multicenter randomized 24-week trial of allopurinol in hyperuricemic (sUA 9.5 mg/dl) patients with chronic HF with reduced EF.30 The post hoc nature of the present study makes the findings vulnerable to selection bias and confounding from measured and unmeasured parameters. Robust multivariate modeling may not be sufficient to account for the vast differences in clinical profiles of patients with differing sUA levels. The EVEREST population represents a highly selected cohort and notably excluded patients with severe CKD (serum creatinine >3.5 mg/dl). In WCHF, treatment with diuretics has a known effect on sUA levels; however, our multivariate models did not account for this parameter given that >95% of patients were on diuretics before or during hospitalization. Data regarding specific in-hospital diuretic dosages and postdischarge allopurinol use were not available for analysis. More data are required regarding the prognostic use of sUA in HF with preserved EF and new onset HF. Despite these limitations, this simple, inexpensive, widely available, reliably measured metric may serve as a valuable predictive biomarker, beyond traditional markers of prognosis. The differences in sUA levels and attendant prognostic use in important patient subgroups may inform selection in future clinical trials. Our hypothesisgenerating work calls for a prospective, randomized, controlled trial evaluating xanthine oxidase-inhibition in the high-risk population of patients hospitalized for WCHF. Disclosures The author Haris P. Subacius conducted all final analyses for this manuscript with funding from the Center for Cardiovascular Innovation (Northwestern University Feinberg School of Medicine, Chicago, Illinois). The authors had full access to the data, take responsibility for its integrity, and had complete control and authority over manuscript preparation and the decision to publish. Dr. Mentz has received research support from Gilead and honoraria from Thoratec, HeartWare, BMS, and Novartis. Dr. Chioncel has received research support from Abbott, Servier, and Vifor and served as member on the Steering Committee of studies sponsored by Novartis. Dr. Maggioni served as member on the Steering Committee of studies sponsored by Bayer, Novartis, Otsuka, Cardiorentis, and Abbott Vascular. Dr. Zannad has had principal relation with Cardiorenal Diagnostics, served as a consultant for Air Liquide, St. Jude’s Medical, Boston Scientific, Servier, Novartis, and participated as a committee member for Takeda Pharmaceuticals, BIOTRONIK, Janssen, ResMed, Bayer, and Pfizer. Dr. Konstam discloses research support and/or consulting fees from Otsuka, Amgen, Johnson & Johnson Services, Inc., and Novartis. Dr. Givertz discloses funding related to the NIH/NHLBI HF Network. Dr. Butler has received research support from the National Institutes of Health, European Union, Health Resources and Services Administration, U.S. Food and Drug Administration, and served as a consultant for Amgen, Bayer, Celladon,
Heart Failure/Uric Acid and Hospitalized Heart Failure
Gambro, GE Healthcare, Janssen, Medtronic, Novartis, Ono, Relypsa, and Trevena. Dr. Gheorghiade has been a consultant for Abbott Laboratories, Astellas, AstraZeneca, Bayer HealthCare AG, Corthera, Cytokinetics, Debiopharm S.A., Errekappa Terapeutici, GlaxoSmithKline, Ikaria, Johnson & Johnson, Medtronic, Merck, Novartis Pharma AG, Otsuka Pharmaceuticals, Palatin Technologies, PeriCor Therapeutics, Protein Design Laboratories, Sanofi-Aventis, Sigma Tau, Solvay Pharmaceuticals, Takeda Pharmaceutical, and Trevena Therapeutics. All other authors have no conflicts of interest to declare. Supplementary Data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. amjcard.2014.09.008. 1. Gheorghiade M, Pang PS, Ambrosy AP, Lan G, Schmidt P, Filippatos G, Konstam M, Swedberg K, Cook T, Traver B, Maggioni A, Burnett J, Grinfeld L, Udelson J, Zannad F. A comprehensive, longitudinal description of the in-hospital and post-discharge clinical, laboratory, and neurohormonal course of patients with heart failure who die or are re-hospitalized within 90 days: analysis from the EVEREST trial. Heart Fail Rev 2012;17:485e509. 2. Misra D, Zhu Y, Zhang Y, Choi HK. The independent impact of congestive heart failure status and diuretic use on serum uric acid among men with a high cardiovascular risk profile: a prospective longitudinal study. Semin Arthritis Rheum 2011;41:471e476. 3. Alimonda AL, Nunez J, Nunez E, Husser O, Sanchis J, Bodi V, Minana G, Robles R, Mainar L, Merlos P, Darmofal H, Llacer A. Hyperuricemia in acute heart failure. More than a simple spectator? Eur J Intern Med 2009;20:74e79. 4. Cengel A, Turkoglu S, Turfan M, Boyaci B. Serum uric acid levels as a predictor of in-hospital death in patients hospitalized for decompensated heart failure. Acta Cardiol 2005;60:489e492. 5. Hamaguchi S, Furumoto T, Tsuchihashi-Makaya M, Goto K, Goto D, Yokota T, Kinugawa S, Yokoshiki H, Takeshita A, Tsutsui H; Investigators J-C. Hyperuricemia predicts adverse outcomes in patients with heart failure. Int J Cardiol 2011;151:143e147. 6. Henry-Okafor Q, Collins SP, Jenkins CA, Miller KF, Maron DJ, Naftilan AJ, Weintraub N, Fermann GJ, McPherson J, Menon S, Sawyer DB, Storrow AB. Relationship between uric acid levels and diagnostic and prognostic outcomes in acute heart failure. Open Biomark J 2012;5:9e15. 7. Malek F, Ostadal P, Parenica J, Jarkovsky J, Vitovec J, Widimsky P, Linhart A, Fedorco M, Coufal Z, Miklik R, Kruger A, Vondrakova D, Spinar J. Uric acid, allopurinol therapy, and mortality in patients with acute heart failureeresults of the Acute HEart FAilure Database registry. J Crit Care 2012;27:737.e11e737.e24. 8. Novack V, Pencina M, Zahger D, Fuchs L, Nevzorov R, Jotkowitz A, Porath A. Routine laboratory results and thirty day and one-year mortality risk following hospitalization with acute decompensated heart failure. PLoS One 2010;5:e12184. 9. Park HS, Kim H, Sohn JH, Shin HW, Cho YK, Yoon HJ, Nam CW, Hur SH, Kim YN, Kim KB, Park HJ. Combination of uric acid and NT-ProBNP: a more useful prognostic marker for short-term clinical outcomes in patients with acute heart failure. Korean J Intern Med 2010;25:253e259. 10. Pascual-Figal DA, Hurtado-Martinez JA, Redondo B, Antolinos MJ, Ruiperez JA, Valdes M. Hyperuricaemia and long-term outcome after hospital discharge in acute heart failure patients. Eur J Heart Fail 2007;9:518e524. 11. Wasserman A, Shnell M, Boursi B, Guzner-Gur H. Prognostic significance of serum uric acid in patients admitted to the Department of Medicine. Am J Med Sci 2010;339:15e21. 12. Tian Y, Chen Y, Deng B, Liu G, Ji ZG, Zhao QZ, Zhen YZ, Gao YQ, Tian L, Wang L, Ji LS, Ma GP, Liu KS, Liu C. Serum uric acid as an index of impaired renal function in congestive heart failure. J Geriatr Cardiol 2012;9:137e142.
1721
13. Maynard JW, McAdams-Demarco MA, Law A, Kao L, Gelber AC, Coresh J, Baer AN. Racial differences in gout incidence in a population-based cohort: Atherosclerosis Risk in Communities Study. Am J Epidemiol 2014;179:576e583. 14. Puig JG, Michan AD, Jimenez ML, Perez de Ayala C, Mateos FA, Capitan CF, de Miguel E, Gijon JB. Female gout. Clinical spectrum and uric acid metabolism. Arch Intern Med 1991;151:726e732. 15. Gheorghiade M, Konstam MA, Burnett JC Jr, Grinfeld L, Maggioni AP, Swedberg K, Udelson JE, Zannad F, Cook T, Ouyang J, Zimmer C, Orlandi C. Short-term clinical effects of tolvaptan, an oral vasopressin antagonist, in patients hospitalized for heart failure: the EVEREST Clinical Status Trials. JAMA 2007;297:1332e1343. 16. Gheorghiade M, Orlandi C, Burnett JC, Demets D, Grinfeld L, Maggioni A, Swedberg K, Udelson JE, Zannad F, Zimmer C, Konstam MA. Rationale and design of the multicenter, randomized, doubleblind, placebo-controlled study to evaluate the Efficacy of Vasopressin antagonism in Heart Failure: Outcome Study with Tolvaptan (EVEREST). J Card Fail 2005;11:260e269. 17. Konstam MA, Gheorghiade M, Burnett JC Jr, Grinfeld L, Maggioni AP, Swedberg K, Udelson JE, Zannad F, Cook T, Ouyang J, Zimmer C, Orlandi C. Effects of oral tolvaptan in patients hospitalized for worsening heart failure: the EVEREST Outcome Trial. JAMA 2007;297:1319e1331. 18. Nacak H, van Diepen M, de Goeij MC, Rotmans JI, Dekker FW. Uric acid: association with rate of renal function decline and time until start of dialysis in incident pre-dialysis patients. BMC Nephrol 2014;15:91. 19. Sato T, Yamauchi H, Suzuki S, Yoshihisa A, Yamaki T, Sugimoto K, Kunii H, Nakazato K, Suzuki H, Saitoh S, Takeishi Y. Distinct prognostic factors in patients with chronic heart failure and chronic kidney disease. Int Heart J 2013;54:311e317. 20. Filippatos GS, Ahmed MI, Gladden JD, Mujib M, Aban IB, Love TE, Sanders PW, Pitt B, Anker SD, Ahmed A. Hyperuricaemia, chronic kidney disease, and outcomes in heart failure: potential mechanistic insights from epidemiological data. Eur Heart J 2011;32:712e720. 21. Huang H, Huang B, Li Y, Huang Y, Li J, Yao H, Jing X, Chen J, Wang J. Uric acid and risk of heart failure: a systematic review and metaanalysis. Eur J Heart Fail 2014;16:15e24. 22. Leyva F, Anker SD, Godsland IF, Teixeira M, Hellewell PG, Kox WJ, Poole-Wilson PA, Coats AJ. Uric acid in chronic heart failure: a marker of chronic inflammation. Eur Heart J 1998;19:1814e1822. 23. Alcaino H, Greig D, Chiong M, Verdejo H, Miranda R, Concepcion R, Vukasovic JL, Diaz-Araya G, Mellado R, Garcia L, Salas D, Gonzalez L, Godoy I, Castro P, Lavandero S. Serum uric acid correlates with extracellular superoxide dismutase activity in patients with chronic heart failure. Eur J Heart Fail 2008;10:646e651. 24. Leyva F, Anker S, Swan JW, Godsland IF, Wingrove CS, Chua TP, Stevenson JC, Coats AJ. Serum uric acid as an index of impaired oxidative metabolism in chronic heart failure. Eur Heart J 1997;18: 858e865. 25. Sakai H, Tsutamoto T, Tsutsui T, Tanaka T, Ishikawa C, Horie M. Serum level of uric acid, partly secreted from the failing heart, is a prognostic marker in patients with congestive heart failure. Circ J 2006;70:1006e1011. 26. Ogino K, Kato M, Furuse Y, Kinugasa Y, Ishida K, Osaki S, Kinugawa T, Igawa O, Hisatome I, Shigemasa C, Anker SD, Doehner W. Uric acid-lowering treatment with benzbromarone in patients with heart failure: a double-blind placebo-controlled crossover preliminary study. Circ Heart Fail 2010;3:73e81. 27. Gotsman I, Keren A, Lotan C, Zwas DR. Changes in uric acid levels and allopurinol use in chronic heart failure: association with improved survival. J Card Fail 2012;18:694e701. 28. Hare JM, Mangal B, Brown J, Fisher C Jr, Freudenberger R, Colucci WS, Mann DL, Liu P, Givertz MM, Schwarz RP; Investigators O-C. Impact of oxypurinol in patients with symptomatic heart failure. Results of the OPT-CHF study. J Am Coll Cardiol 2008;51:2301e2309. 29. Savarese G, Ferri C, Trimarco B, Rosano G, Dellegrottaglie S, Losco T, Casaretti L, D’Amore C, Gambardella F, Prastaro M, Rengo G, Leosco D, Perrone-Filardi P. Changes in serum uric acid levels and cardiovascular events: a meta-analysis. Nutr Metab Cardiovasc Dis 2013;23:707e714. 30. Givertz MM, Mann DL, Lee KL, Ibarra JC, Velazquez EJ, Hernandez AF, Mascette AM, Braunwald E. Xanthine oxidase inhibition for hyperuricemic heart failure patients: design and rationale of the EXACT-HF study. Circ Heart Fail 2013;6:862e868.