Journal of the American College of Cardiology 2008 by the American College of Cardiology Foundation Published by Elsevier Inc.

Vol. 52, No. 10, 2008 ISSN 0735-1097/08/$34.00 doi:10.1016/j.jacc.2008.04.061

Heart Failure

Anemia and Mortality in Heart Failure Patients

A Systematic Review and Meta-Analysis
Hessel F. Groenveld, MD,* James L. Januzzi, MD, FACC, Kevin Damman, MD,* Jan van Wijngaarden, MD, PHD, Hans L. Hillege, MD, PHD,* Dirk J. van Veldhuisen, MD, PHD, FACC,* Peter van der Meer, MD, PHD Groningen and Deventer, the Netherlands; and Boston, Massachusetts
Objectives Background Methods
The aim of this study was to assess the effect of anemia on mortality in chronic heart failure (CHF). Anemia is frequently observed in patients with CHF, and evidence suggests that anemia might be associated with an increased mortality. A systematic literature search in MEDLINE (through November 2007) for English language articles was performed. In addition, a manual search was performed. We included cohort studies and retrospective secondary analyses of randomized controlled trials whose primary objective was to analyze the association between anemia and mortality in CHF. Of a total of 1,327 initial studies, we included 34 studies, comprising 153,180 patients. Information on study design, patient characteristics, outcome, and potential confounders were extracted. Anemia was dened by criteria used in the original articles. Of the 153,180 CHF patients, 37.2% were anemic. After a minimal follow-up of 6 months, 46.8% of anemic patients died compared with 29.5% of nonanemic patients. Crude mortality risk of anemia was odds ratio 1.96 (95% condence interval: 1.74 to 2.21, p 0.001). Lower baseline hemoglobin values were associated with increased crude mortality rates (r 0.396, p 0.025). Adjusted hazard ratios showed an increased adjusted risk for anemia (hazard ratio 1.46 [95% condence interval: 1.26 to 1.69, p 0.001]). Subgroup analysis showed no signicant difference between mortality risk of anemia in diastolic or systolic CHF. Anemia is associated with an increased risk of mortality in both systolic and diastolic CHF. Anemia should, therefore, be considered as a useful prognosticator, and therapeutic strategies aimed to increase hemoglobin levels in CHF should be investigated. (J Am Coll Cardiol 2008;52:81827) 2008 by the American College of Cardiology Foundation

Results

Conclusions

Anemia is frequently observed in patients with chronic heart failure (CHF) (15). Prevalence of anemia depends both on the severity of CHF and diagnostic criteria used to dene it, but may be as high as 50% in selected patient cohorts (3,6). Of note, anemia is not only prevalent in the CHF population, but several studies in different patient populations found an association with anemia, impaired cardiac function, more health care utilization, and morbidity (1,711). In addition, numerous studies have assessed associations between anemia and mortality in CHF. Although most studies have documented higher mortality rates in anemic CHF patients (1,3,1217), some
From the *Department of Cardiology, University Medical Center Groningen, Groningen, the Netherlands; Division of Cardiology, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts; and the Department of Cardiology, Deventer Hospital, Deventer, the Netherlands. Dr. Damman is supported by the Netherlands Heart Foundation (grant 2006B157). Dr. van Veldhuisen is a Clinical Established Investigator of the Netherlands Heart Foundation (D97-017). Dr. van der Meer is supported by the Dutch Scientic Organization (Rubicon grant: 825-07-011). Manuscript received February 19, 2008; revised manuscript received April 23, 2008, accepted April 28, 2008.

studies report the absence of an adverse effect of anemia on mortality in CHF (4,18 20). Therefore, better understanding of the risk associated with the presence of anemia is necessary. Elucidating the contribution of anemia to mortality may lead to a more accurate risk proling in CHF patients, especially since several treatment options for anemia exist including erythropoietin (EPO) and iron therapy (21,22). Ultimately, this may lead to more effective therapeutic strategies from a risk-benet perspective in heart failure patients. We, therefore, performed a rigorous systematic review and meta-analysis of published literature to more comprehensively assess the effect of anemia on mortality in CHF. In addition, we explored whether subgroups of anemic patients were at particularly high mortality risk. Methods Search strategy. We adhered to the MOOSE (Metaanalysis Of Observational Studies in Epidemiology) study guidelines as previously published (23). To identify all relevant

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827

Groenveld et al. Anemia and Mortality in Heart Failure

819

studies, we performed a literature search in MEDLINE accessed by PubMed (1966 through November 2007). Two of the authors (H.G. and P.v.d.M.) developed a search strategy. To identify heart failure patients and anemia, we used both the Medical Subject Heading term (MeSH) and text word search (Table 1). In addition, we identied potentially relevant studies using a manual search of references lists from all eligible studies and review articles. We consulted experts in the eld and searched the Institute for Scientic Information Web of Sciences for publications that cited key publications. Study selection. We included cohort studies and retrospective secondary analyses of randomized controlled trials whose primary objective was to analyze the association between anemia and mortality in CHF. Titles and abstracts of all articles were evaluated and rejected on initial screen if they: 1) included subjects other than CHF patients; 2) had no evaluation of hemoglobin (Hb) or hematocrit (Ht) levels; 3) did not include all-cause mortality as outcome; 4) were published only in abstract form; 5) included patients 18 years old; 6) had follow-up of 6 months; or 7) were published in another language than English. After obtaining full reports of candidate studies, the same reviewers independently assessed eligibility. Differences in data between the 2 reviewers were solved by rereviewing corresponding articles, and the nal set was agreed on by consensus. For studies in which not all data were published, or data was insufcient, authors were addressed and asked for additional information (3,4,12,19,24 26). Quality assessment and data abstraction. Each study was evaluated on quality according to the guidelines provided by the United States Preventive Task Force (27) and published recommendations (28). The following characteristics were assessed: 1) duration of follow-up 12 months; 2) reporting loss of follow-up; 3) adjustment of possible confounders in multivariate analysis; 4) denition of anemia; 5) full specication of outcome; 6) study sample representative for mentioned population; 7) full specication of clinical and demographic variables; 8) explanation of sample selection; 9) temporality (Hb measured at baseline, not at time of outcomes assessment); and 10) clear inclusion and exclusion criteria. Studies were graded as poor quality if they met 5
Terms Used in the Search Query Table 1
Text words Congestive heart failure, CHF, anemia, anaemia, hemoglobin, haemoglobin, Hb, Hgb, hematocrit, haematocrit, Ht, Hct MeSH terms Heart failure, congestive; anemia; hemoglobin; hematocrit Title and abstract search Hemoglobins Limits English language
CHF congestive heart failure; MeSH Medical Subject Heading.

criterion, fair if they met 5 to 7 Abbreviations and Acronyms criteria, and good if they met 8 criteria. CHF chronic heart failure Statistical analysis. Random CI condence interval effects meta-analysis was conEPO erythropoietin ducted to estimate the magniESP erythropoiesistude of risk associated with anestimulating protein mia and all-cause mortality, as Hb hemoglobin measured by crude mortality HR hazard ratio risks and unadjusted odds ratios Ht hematocrit (ORs). Where available, adjusted hazard ratios (HRs) were exOR odds ratio tracted from Cox regression WHO World Health analysis. Adjusted risk estimates Organization included those components as published in nal multivariate models for each study including confounding sociodemographic and clinical covariates. We performed tests of heterogeneity between studies using a standard chi-square test and I2 statistic. When heterogeneity was found to be signicant, pooled estimates based on random effects models were reported. To examine sources of heterogeneity, we conducted meta-regression analysis. Bias in published studies was assessed using a funnel plot of study results against study precision. We tested symmetry of the funnel plot by the Eggers test. Subgroup analysis was pre-dened for systolic versus diastolic CHF. Statistical signicance was set at p 0.05, and all statistical analyses were performed using Stata 9.0 (Stata Corp., College Station, Texas). Results Description of included studies. The electronic search retrieved 1,327 eligible studies. Nonelectrical search identied 4 additional studies. On initial screening, 1,142 were rejected, based on title. Of the 185 screened abstracts, 56 studies were retrieved for detailed evaluation. Eventually 34 studies were included in this meta-analysis (Fig. 1). Secondary analysis of the Val-HeFT (Valsartan Heart Failure Trial) study was described in 2 separate manuscripts: Anand et al. (29) and Maggioni et al. (5). Crude mortality data were not published by Anand et al. (29), and for this reason this study was excluded and the data of Maggioni et al. (5) was used. In addition, 1 article described the effect of anemia from 2 different studies (5). Both studies in this published report were analyzed separately for the metaanalysis. The 34 included studies described a total of 153,180 patients (1,320,24 26,30 41). Characteristics of the included studies are listed in Table 2, characteristics of study populations are listed in Online Tables 1 and 2. Ten different denitions of anemia were used in the included studies. The World Health Organization (WHO) denition (Hb 13.0 g/dl for males and Hb 12.0 g/dl for females) was used in the majority of the studies. Table 3 displays the different denitions of anemia in the included studies. Anemia, as dened in the original articles, was

Terms Used in the Search Query

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

820

Groenveld et al. Anemia and Mortality in Heart Failure

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827

Figure 1

Flow Chart for Study Selection

CHF chronic heart failure; Hb hemoglobin.

present in 37.2% of the CHF patients. Mean Hb from the studies ranged from 12.2 to 14.0 g/dl. Mean Ht values ranged from 36.6% to 42.7%. Meta-analysis ndings. Among the 34 studies, mean follow-up ranged from a minimum of 6 months to a maximum of 5 years. Crude mortality data were available in 33 studies. In one study, only the adjusted HR was available, whereas the unpublished crude mortality data were lost after hurricane Katrina (25). Analysis of the crude mortality revealed that 26,687 (46.8%) anemic patients died compared with 28,274 (29.5%) nonanemic patients. This translated into an unadjusted mortality risk of OR 1.96 (95% condence interval [CI]: 1.74 to 2.21, p 0.001) in anemic CHF patients compared with that in nonanemic CHF patients (Fig. 2). When the 2 largest studies (Go et al. [3] and Kosiborod et al. [4]) were excluded, the mortality risk of anemia was identical (OR: 1.95, 95% CI: 1.78 to 2.14, p 0.001). There was no evidence for publication bias. The funnel plot did not show asymmetry (Fig. 3), which was conrmed by the Eggers test (p 0.93). We found a

signicant heterogeneity between included studies (I2 92.4%, p 0.001). Meta-regression analysis was performed using sociodemographic variables, medical history, drug use, and laboratory data as parameters. In this sensitivity analysis, excluding studies not adhering to the WHO denition of anemia, no heterogeneity was observed (I2 36.5%, p NS), yet the mortality risk remained (OR: 2.22, 95% CI: 2.04 to 2.42, p 0.001). In addition, we found an inverse interstudy relationship between serum creatinine levels and the effect of anemia on mortality. The effect of anemia on mortality declined with higher serum creatinine levels. When assessing the gradual effect of baseline Hb values on mortality, we found a linear trend between baseline Hb measurements and mortality rates. Lower baseline Hb values were associated with increased annual mortality rates (r 0.396, p 0.025) (Fig. 4). Confouding factors. In 127,437 patients (83.1% of the total included patients), reported mortality risks were adjusted for several sociodemographic and clinical covariates. Potential confounders that were adjusted for in the original reports are shown in Table 4. In all but one study, age and renal function were considered as potential confounders (this study included 178 patients) (33). When mortality risk, estimated by adjusted HR, was combined, anemia remained an independent predictor of mortality HR 1.46 (95% CI: 1.26 to 1.69, p 0.001). Subgroup analysis. In the subgroup analysis, we assessed the difference between mortality risk of anemia in patients with systolic or diastolic CHF. Table 5 displays the denitions of systolic and diastolic CHF used in the original articles. Studies or substudies in patients with only diastolic CHF included 20,924 patients of which 5,957 (28.5%) patients died. In studies or substudies including only patients with systolic CHF, 40,025 patients were included, of which 12,423 (31.0%) died. The mortality risk associated with the presence of anemia was not signicantly different between patients with systolic CHF (OR: 1.96, 95% CI: 1.70 to 2.25, p 0.001) and diastolic CHF (OR: 2.09, 95% CI: 1.53 to 2.86, p 0.001). Discussion This is the rst meta-analysis to address the relationship between anemia and mortality in patients with CHF. In our analysis, which examined more than 150,000 subjects, anemia was frequently observed, found in over one-third of CHF patients. Presence of anemia in CHF patients is associated with an increased mortality risk in patients with systolic as well as diastolic heart failure. The adverse effect we found is substantial and signicant. When assessing the mortality risk by using multivariate analyses, anemia remains an independent risk factor for mortality in CHF patients. Prevalence. In prior studies, it is observed that CHF is often accompanied by anemia. A wide range of anemia prevalence in CHF has been reported, ranging from 7% to

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827 Study Characteristics Table 2 Study Characteristics
Follow-Up 33.4 months n 6,360 Patients CHF, LVEF 35%, age 21 to 81 yrs Referred for heart transplantation evaluation Discharge diagnosis of HF (ICD-9-CM) End Points Mortality

Groenveld et al. Anemia and Mortality in Heart Failure

821

Author (Ref. #), Year Al-Ahmad et al. (12), 2001

Major Exclusion Criteria Recent MI, unstable angina, sCR 2.5 mg/dl LVEF 40%, NYHA class III, no initial Hb, dry weight, or adequate follow-up Dialysis, death during index hospitalization, missing records Secondary anemia, hemodialysis, coronary intervention last 3 months

Study Design Secondary analysis of clinical trial (SOLVD) Retrospective, observational, single-center cohort study Retrospective, observational, multicenter cohort study Retrospective, observational, single-center cohort study Prospective observational, single-center cohort study Retrospective, observational, multicenter chart review Secondary analysis of multicenter clinical trial (PRAISE)

Study Good

Horwich et al. (9), 2002

5 yrs

1,061

Mortality

Good

McClellan et al. (38), 2002

2 yrs

633

Mortality

Good

Tanner et al. (32), 2002

6 months

193

Outpatient HF clinic at university hospital

Mortality, heart transplantation

Good

Kalra et al. (26), 2003

3 yrs

531

HF

Mortality

Fair

Kosiborod et al. (39), 2003

1 yr

2,281

Principal discharge diagnosis of HF (ICD-9-CM), age 65 yrs LVEF 30% and NYHA class IIIB or IV HF symptoms

Mortality, hospitalization

Severe AoS or MiS, HF due to acute illness, missing Htc values, in-hospital mortality MI; unstable angina; stroke or revascularization; severe pulmonary, renal, or hepatic disease, sCR 3.0 mg/dl

Good

Mozaffarian et al. (37), 2003

15 months

1,130

Mortality

Good

Kerzner et al. (40), 2003

25 months

367

Primary discharge diagnosis of HF

Mortality

Retrospective, observational, single-center chart review Prospective, observational, single-center, cohort study Secondary analysis of multicenter clinical trial (ELITE II) Secondary analysis of clinical trial (RENAISSANCE) Retrospective, observational, single-center cohort study Prospective observational, single-center cohort study Secondary analysis of a randomized clinical trial (Val-HeFT) Retrospective, observational, multicenter cohort study Prospective, observational, single-center cohort study

Fair

Szachniewicz et al. (33), 2003

529 days

176

Admitted for HF

Mortality

Anemia due to known secondary cause

Good

Sharma et al. (34), 2004

551 days

3,044

NYHA II to IV, LVEF 40%, age 60 yrs NYHA II to IV, LVEF 30%, age 18 to 55 yrs Mild-to-severe CHF

Mortality

Severe valvular disease, recent PTCA, CABG, MI, or unstable angina, sCR 220 mol/l Surgically correctable causes of HF, other serious illness, recent or planned revascularization MI, CVA, severe renal failure, valvular disease, isolated systolic dysfunction Missing Hb values

Good

Anand et al. (1), 2004

12.7 months

912

Mortality, hospitalization

Good

van der Meer et al. (30), 2004

1,100 days

74

Mortality

Good

Ezekowitz et al. (24), 2005

962 days

791

Referred for assessment and management of CHF Symptomatic HF (NYHA II), clinically stable

Mortality

Good

Maggioni et al. (5), Val-HeFT 2005

23 months

5,010

Mortality, time to death, rst morbid event

Rapidly deteriorating HF, acute MI, CAD likely to require intervention, unstable angina, recent CVA, sCR 2.5 mg/dl

Good

Maggioni et al. (5), IN-CHF 2005

1 yr

2,411

Diagnosis of HF according to European Society of Cardiology Advance HF, referred for cardiac Tx assessment

Mortality

Good

Gardner et al. (19), 2005

554 days

182

Mortality

Age 16 yrs, pregnancy, known malignancy

Fair

Continued on next page

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

822

Groenveld et al. Anemia and Mortality in Heart Failure

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827

Continued Table 2 Continued

Follow-Up 4 yrs n 508 Patients Suspected HF End Points Mortality Major Exclusion Criteria Study Design Retrospective, observational, multicenter cohort study Retrospective, observational, Single-center cohort study Retrospective, observational, multicenter cohort study Retrospective, observational, single-center cohort study Secondary analysis of clinical trial (CHARM) Retrospective, observational, single-center cohort study Secondary analysis of multicenter, prospective clinical trial Retrospective, observational, single-center cohort study Prospective, observational, single-center cohort study Retrospective, observational, single-center chart review Secondary analysis of clinical trial (COMET) Retrospective observational, longitudinal cohort study Retrospective, observational, single-center chart review Prospective, observational, single-center cohort study Secondary analysis of longitudinal observational study Prospective observational, single-center cohort study Study Fair

Author (Ref. #), Year Rosolova et al. (35), 2005

Grigorian et al. (14), 2006

1.5 yrs

205

Admitted to tertiary center for CHF

Mortality

LVEF 50%

Good

Kosiborod et al. (4), 2005

1 yr

50,405

Discharge diagnosis of HF

Mortality, hospitalization for HF Mortality

Patients without documented HF on admission, age 65 yrs, hemodialysis, severe Aos, MiS, ACS or myocarditis, LVEF 40%

Good

Ralli et al. (16), 2005

1 yr

264

Referred to a university hospital for HF management or heart Tx Symptomatic (NYHA II to IV) HF, for at least 4 weeks, age 18 yrs Symptomatic HF (NYHA II) patients undergoing cardiac catheterization Symptomatic HF (NYHA II to IV), LVEF 35% LVEF 40%, enrolled in chronic HF disease managemant program Admitted for symptomatic, previously unknown HF Discharge diagnosis of HF, without previous hospitalization for HF NYHA class II to IV and LVEF 35% Diagnosed with CHF

Good

OMeara et al. (36), 2006

37.7 months

2,653

Mortality

sCR 265 mol/l, symptomatic HT, MI, CVA

Good

Felker et al. (13), 2006

3.6 yrs

4,951

Mortality

Primary valvular heart disease or congenital heart disease Recent MI or CABG, unstable angina, anticipated cardiac surgery, sCR 3 mg/dl Missing baseline values of Hb, sCR

Fair

Terrovitis et al. (31), 2006

2 yrs

160

Mortality

Good

Hebert et al. (25), 2006

436 days

410

Mortality

Fair

Formiga et al. (18), 2006

1 yr

103

Mortality, hospitalization for HF Mortality

Chronic advanced disease, COPD, advanced renal failure hemodialysis, in-hospital mortality Insufcient evidence of new diagnosis of CHF

Good

Newton and Squire (15), 2006

1,257 days

528

Good

Komajda et al. (10), 2006

58 months

2,996

Mortality, hospitalization Mortality, hospitalization

Recent change of treatment, requirement of IV medication

Good

Go et al. (3), 2006

2 yrs

59,772

Good

Berry et al. (7), 2006

814 days

519

Diagnosed with CHF or a treatment for CHF LVEF 45%, CHF enrolled in a community-based program Diagnosed with HF, age 65 yrs

Mortality, hospitalization, HF hospitalization Mortality

Primary or secondary diagnosis of acute MI

Fair

de Silva et al. (42), 2006

531 days

955

Pregnancy, dialysis, disorders resulting in anemia Physical disability at hospital discharge, dementia, or severe cognitive impairment No complete follow-up

Fair

Maraldi et al. (11), 2006

12 months

567

Mortality, physical disability Mortality, hospitalization, ED visit

Fair

Elabbassi et al. (8), 2006

1.3 yrs

437

Systolic or diastolic CHF

Good

Continued on next page

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827 Continued Table 2 Continued
Follow-Up 5 yrs n 2,246 Patients Primary discharge of diagnosis of CHF End Points Mortality, hospitalization, ED visit Mortality

Groenveld et al. Anemia and Mortality in Heart Failure

823

Author (Ref. #), Year Varadarajan et al. (6), 2006

Major Exclusion Criteria CHF excluded with Framingham criteria, no ECG

Study Design Retrospective, observational, single-center, chart review Prospective, observational, single-center cohort study

Study Good

Schou et al. (17), 2007

30 months

345

Systolic HF (LVEF 45%)

Good

ACS acute coronary syndromes; AoS aortic valve stenosis; CABG coronary arterial bypass graft surgery; CAD coronary artery disease; CHARM Candesartan in Heart Failure trial; CHF chronic heart failure; COMET Carvedilol Or Metoprolol European Trial; COPD chronic obstructive pulmonary disease; CVA cerebral vascular accident; ECG electrocardiogram; ED emergency department; ELITE II Losartan Heart Failure Survival Study; F/U follow-up; Hb hemoglobin; HF heart failure; HT hypertension; Htc hematocrit; ICD-9-CM International Classication of Diseases, 9th Revision, Clinical Modication; IN-CHF Italian NetworkCongestive Heart Failure trial; IV intravenous; LVEF left ventricular ejection fraction; MI myocardial infarction; MiS mitral valve stenosis; NYHA New York Heart Association functional class; PRAISE Prospective Randomized Amlopidine Survival Evaluation trial; PTCA percutaneous transluminal coronary angioplasty; RENAISSANCE Randomized Etanercept North American Strategy to Study Antagonism of Cytokines; sCR serum creatinine; Tx heart transplantation; Val-HeFT Valsartan Heart Failure Trial.

over 50% (1 4,10,17,32,35). This range can be attributed to multiple factors, including the use of multiple denitions of anemia. The WHO criteria was used in the majority of studies. In addition, administrational codes or arbitrary boundary values to dene anemia were also used. Studies dening anemia with administrative codes (such as International Classication of Diseases, 9th Revision, codes) have clear limitations. The use of this coding system solely relies on physicians to consider anemia as a possible diagnosis leading to an underestimation of the prevalence. Therefore, we excluded in this meta-analysis studies without measurements of actual Hb or Ht values. This resulted in the exclusion of only one study (2). Results of the current meta-analysis clearly indicate the need for standardized denitions of anemia in CHF patients. The denition of anemia appeared to be a major source of heterogeneity. However, despite different denitions of anemia in the present meta-analysis, the association with mortality was robust, using different analysis strategies.
Denitions of Anemia Used in the Original Articles Table 3

Pathophysiology. It is tempting to speculate about the mechanisms behind the increased mortality risk observed in anemic CHF patients. Anemia may lead to an increased work load, resulting from an increased heart rate and stroke volume (32,42,43). In response to the increased workload, the heart undergoes remodeling, marked by left ventricular hypertrophy and dilation. This eventually may lead to CHF with an increased mortality risk. Also, renal failure, as a common comorbidity in CHF, might be implicated in the pathophysiology. CHF and renal failure are 2 entities that are often seen together with prevalences ranging from 20% to 40% (2,38,44 46). With meta-regression analysis, we found an inverse interstudy relationship between serum creatinine levels and the effect of anemia on mortality. However, this result does not suggest that the mortality risk declines with higher creatinine levels. It merely illustrates a statistical decreasing effect of anemia on mortality with increasing creatinine levels; the decline of the effect of anemia on mortality can be the result of an increased

Denitions of Anemia Used in the Original Articles

Author (Ref. #) Elabbassi et al. (8), Maraldi et al. (11), De Silva et al. (42), Go et al. (3), Komajda et al. (10), Formiga et al. (18), Hebert et al. (25), OMeara et al. (36), Felker et al. (13), Ralli et al. (16), Grigorian Shamagian et al. (14), Rosolova et al. (35), Gardner et al. (19), Ezekowitz et al. (24), van der Meer et al. (30), Kerzner et al. (40), Mozaffarian et al. (37), Schou et al. (17) Kalra et al. (26) Varadarajan et al. (6), Anand et al. (1), Szachniewicz et al. (33), Tanner et al. (32) Horwich et al. (9) Sharma et al. (34) Maggioni et al. (Val-HeFT) (5), Maggioni et al. (IN-CHF) (5)

Denition of Anemia WHO

Hb 11.5 g/dl Hb 12.0 g/dl Hb 12.3 g/dl Hb 13.0 g/dl Hb 11.0 g/dl in women and Hb 12.0 g/dl in men Hb 11.5 g/dl in women and 13.0 g/dl in men Ht 35% Ht 37% Ht 37% Ht 37% in women and 40% in men

Berry et al. (7), Newton and Squire (15)

Al-Ahmad et al. (12) Kosiborod et al. 2003 (39), McClellan et al. (38) Terrovitis et al. (31) Kosiborod et al. 2005 (4)

Ht hematocrit; WHO World Health Organization denition of anemia, Hb 12.0 g/dl in women, 13.0 in men; other abbreviations as in Table 2.

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

824

Groenveld et al. Anemia and Mortality in Heart Failure

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827

Figure 2

Risk of All-Cause Mortality of Anemic Versus Nonanemic CHF Patients

CHF chronic heart failure. CI condence interval.

mortality risk caused by renal failure (47). From a pathophysiological standpoint, CHF can also cause renal failure due to a decreased cardiac output reducing renal perfusion (i.e., forward failure); subsequent renal failure may lead to decreased endogenous EPO levels and may ultimately induce anemia, leading to an increased cardiac workload completing the vicious circle. Recently this has been called the cardio-renal-anemia syndrome (48). Findings from intervention studies. Considering the increased mortality risk caused by anemia, in heart failure as well as renal failure, trials have been designed in which patients receive erythropoiesis-stimulating proteins (ESPs) to increase Hb levels. The rst intervention study to address the efcacy of EPO in CHF patients was performed by Silverberg et al. (49) in 32 patients. Correction of anemia with EPO and intravenous iron led to a signicant increase in left ventricular ejection fraction and decrease in New York Heart Association functional class, which was reected by almost 90% reduction in the number of hospital-

Figure 3

Funnel Plot for Publication Bias

OR odds ratio.

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827

Groenveld et al. Anemia and Mortality in Heart Failure Denitions Diastolic HF ofUsed Systolic in the and Original Articles Denitions of Systolic and Table 5 Diastolic HF Used in the Original Articles

825

LVEF (%) Systolic CHF Mozaffarian et al. (37), Anand et al. (1) Komajda et al. (10), Terrovitis et al. (31) Sharma et al. (34), Ralli et al. (16), Maggioni et al. (5), Kerzner et al. (40), Kosiborod et al. (4), OMeara et al. (36), Felker et al. (13) Schou et al. (17) Diastolic CHF Kerzner et al. (40), Kosiborod et al. (4), OMeara et al. (36), Felker et al. (13) Grigorian Shamagian et al. (14)
Abbreviations as in Table 2.

30 35 40

45 40 50

Figure 4

Relationship Between Baseline Hemoglobin and Annual Mortality

The area of each circle is proportional to the sample size in each cohort. The center line shows the estimated mortality risk per year of lower baseline Hb values on a continuous scale. The dotted lines represent the 95% condence intervals.

izations. A similar study showed that EPO treatment signicantly increased peak oxygen consumption and exercise duration in patients with moderate-to-severe CHF (50). Recently, 2 substantially larger multicenter phase II studies evaluated the effects of darbepoetin (a long-acting ESP) on surrogate cardiovascular end points (51,52). Treatment with darbepoetin was safe and effectively raised Hb. Moreover, it signicantly improved clinical status; however, no signicant improvement in exercise tolerance could be observed. However, recently several studies showed a potentially harmful effect of ESP treatment in patients with kidney disease and malignancies. A meta-analysis in patients with cancer-associated anemia showed an increased risk in vePotential Confounders of Included Studies Table 4 Potential Confounders of Included Studies

nous thromboembolism and mortality associated with recombinant EPO and darbepoetin administration (53). Furthermore, concerns about the cardiovascular safety of ESP in patients with kidney disease have been raised. Two separate studies showed that patients targeted to a higher Hb level had an increased incidence of cardiovascular events (54,55). However, in these studies, no placebo groups were included. These studies were all performed in patients with severe renal failure, and only a minority of patients suffered from heart failure. Consequently, the results of the studies mentioned in the preceding text cannot be extrapolated to the CHF population. Larger randomized studies are clearly needed to determine the impact on morbidity and mortality in CHF. Currently, a large phase III morbidity and mortality trial, the RED-HF (Reduction of Events with Darbepoetin alfa in Heart Failure) trial (56), is being conducted to answer the question whether the use of ESPs in anemic heart failure patients is benecial. Besides ESP, intravenous iron treatment is another intervention modality in anemic CHF patients. Recently, 2 trials appeared show-

Author (Ref. #) Elabbassi et al. (8) Maraldi et al. (11) Berry et al. (7) Newton and Squire (15) Hebert et al. (25) Felker et al. (13) Kosiborod et al. (4) Maggioni-V et al. (5) Maggioni-I et al. (5) van der Meer et al. (30) Szachniewicz et al. (33) Mozaffarian et al. (37) Go et al. (3) Kalra et al. (26) Schou et al. (17)

Number of Patients 437 567 519 528 410 4,951 50,405 5,010 2,411 74 176 1,130 59,772 531 345

Age

Gender

Renal Function

Severity of Heart Failure

Medical History

Medication

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

826

Groenveld et al. Anemia and Mortality in Heart Failure

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827 2. Ezekowitz JA, McAlister FA, Armstrong PW. Anemia is common in heart failure and is associated with poor outcomes: insights from a cohort of 12 065 patients with new-onset heart failure. Circulation 2003;107:2235. 3. Go AS, Yang J, Ackerson LM, et al. Hemoglobin level, chronic kidney disease, and the risks of death and hospitalization in adults with chronic heart failure: the Anemia in Chronic Heart Failure: Outcomes and Resource Utilization (ANCHOR) study. Circulation 2006;113: 271323. 4. Kosiborod M, Curtis JP, Wang Y, et al. Anemia and outcomes in patients with heart failure: a study from the National Heart Care Project. Arch Intern Med 2005;165:2237 44. 5. Maggioni AP, Opasich C, Anand I, et al. Anemia in patients with heart failure: prevalence and prognostic role in a controlled trial and in clinical practice. J Card Fail 2005;11:91 8. 6. Varadarajan P, Gandhi S, Sharma S, Umakanthan B, Pai RG. Prognostic signicance of hemoglobin level in patients with congestive heart failure and normal ejection fraction. Clin Cardiol 2006;29: 444 9. 7. Berry C, Norrie J, Hogg K, Brett M, Stevenson K, McMurray JJ. The prevalence, nature, and importance of hematologic abnormalities in heart failure. Am Heart J 2006;151:131321. 8. Elabbassi W, Fraser M, Williams K, Cassan D, Haddad H. Prevalence and clinical implications of anemia in congestive heart failure patients followed at a specialized heart function clinic. Congest Heart Fail 2006;12:258 64. 9. Horwich TB, Fonarow GC, Hamilton MA, MacLellan WR, Borenstein J. Anemia is associated with worse symptoms, greater impairment in functional capacity and a signicant increase in mortality in patients with advanced heart failure. J Am Coll Cardiol 2002;39:1780 6. 10. Komajda M, Anker SD, Charlesworth A, et al. The impact of new onset anaemia on morbidity and mortality in chronic heart failure: results from COMET. Eur Heart J 2006;27:1440 6. 11. Maraldi C, Volpato S, Cesari M, et al. Anemia, physical disability, and survival in older patients with heart failure. J Card Fail 2006;12:5339. 12. Al-Ahmad A, Rand WM, Manjunath G, et al. Reduced kidney function and anemia as risk factors for mortality in patients with left ventricular dysfunction. J Am Coll Cardiol 2001;38:955 62. 13. Felker GM, Shaw LK, Stough WG, OConnor CM. Anemia in patients with heart failure and preserved systolic function. Am Heart J 2006;151:457 62. 14. Grigorian Shamagian L, Varela RA, Garcia-Acuna JM, Mazon RP, Virgos LA, Gonzalez-Juanatey JR. Anaemia is associated with higher mortality among patients with heart failure with preserved systolic function. Heart 2006;92:780 4. 15. Newton JD, Squire IB. Glucose and haemoglobin in the assessment of prognosis after rst hospitalisation for heart failure. Heart 2006;92: 1441 6. 16. Ralli S, Horwich TB, Fonarow GC. Relationship between anemia, cardiac troponin I, and B-type natriuretic peptide levels and mortality in patients with advanced heart failure. Am Heart J 2005;150:1220 7. 17. Schou M, Gustafsson F, Kistorp CN, Corell P, Kjaer A, Hildebrandt PR. Prognostic usefulness of anemia and N-terminal pro-brain natriuretic peptide in outpatients with systolic heart failure. Am J Cardiol 2007;100:1571 6. 18. Formiga F, Chivite D, Castaner O, Manito N, Ramon JM, Pujol R. Anemia in new-onset congestive heart failure inpatients admitted for acute decompensation. Eur J Intern Med 2006;17:179 84. 19. Gardner RS, Chong KS, Morton JJ, McDonagh TA. N-terminal brain natriuretic peptide, but not anemia, is a powerful predictor of mortality in advanced heart failure. J Card Fail 2005;11:S4753. 20. Silva RP, Barbosa PH, Kimura OS, et al. Prevalance of anemia and its association with cardio-renal syndrome. Int J Cardiol 2007;120:232 6. 21. van der Meer P, Voors AA, Lipsic E, van Gilst WH, van Veldhuisen DJ. Erythropoietin in cardiovascular diseases. Eur Heart J 2004;25: 28591. 22. van Veldhuisen DJ, McMurray JJ. Are erythropoietin stimulating proteins safe and efcacious in heart failure? Why we need an adequately powered randomised outcome trial. Eur J Heart Fail 2007;9:110 2. 23. Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group. JAMA 2000;283:2008 12.

ing that intravenous iron without ESPs increased Hb levels in patients with CHF. Moreover, they showed an improved exercise capacity and decreased NT-proBNP levels (57,58). Further research is needed to study the effect of intravenous iron on morbidity and mortality in anemic CHF patients. Study limitations. Limitations of meta-analyses include differences in designs and populations used in included trials, quality of studies, and heterogeneity (59). However, the quality of all studies was graded according to published guidelines and all met at least 7 of 12 criteria. Evaluation of anemia was only at baseline in all included studies. Tang et al. (60) recently reported a study with serial hemoglobin measurements and showed that persistent anemia has a larger effect on mortality than transient anemia. No evaluation of Hb was performed during the studies; thus, it remains unknown whether anemia in the included studies was persistent or transient. Furthermore, heterogeneity in our study is substantial and can be attributed to differences in duration of follow-up, type of patients included, study design, and denition of anemia. Denition of anemia was a major cause of heterogeneity. When mortality risk was assessed using studies that dened anemia by WHO criteria, heterogeneity was no longer signicant, while the increased mortality risk caused by anemia remained. Furthermore, our meta-analysis could not exclude residual confounding from studies in the adjusted analyses, although most studies adjusted for many known major confounders. However, this limitation emphasizes the inability of our analysis to prove causality. Previously, it has been argued that a larger meta-analysis (e.g., 1,000 events) is clinically more meaningful (59); our study, which includes over 50,000 events, is therefore of particular note. Besides the overall size, the individual studies in the present analysis were of considerable size, with 14 including more than 1,000 patients. These strengths make our ndings all the more rm. Conclusions Anemia is present in one-third of the CHF population and is an independent risk prognosticator for mortality in subjects so affected, irrespective of a systolic versus diastolic etiology of CHF. Further research is needed to assess the effect of correcting anemia in CHF patients.
Reprint requests and correspondence: Dr. Peter van der Meer, Cardiovascular Research Center, Massachusetts General Hospital, Charles River Plaza, 185 Cambridge Street, Boston, Massachusetts 02114. E-mail: pvandermeer@partners.org.

REFERENCES

1. Anand I, McMurray JJ, Whitmore J, et al. Anemia and its relationship to clinical outcome in heart failure. Circulation 2004;110:149 54.

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013

JACC Vol. 52, No. 10, 2008 September 2, 2008:81827 24. Ezekowitz JA, McAlister FA, Armstrong PW. The interaction among sex, hemoglobin and outcomes in a specialty heart failure clinic. Can J Cardiol 2005;21:16571. 25. Hebert K, Horswell R, Arcement L, Hare J, Stevenson L. The effect of anemia on mortality in indigent patients with mild-to-moderate chronic heart failure. Congest Heart Fail 2006;12:759. 26. Kalra PR, Collier T, Cowie MR, et al. Haemoglobin concentration and prognosis in new cases of heart failure. Lancet 2003;362:2112. 27. Harris RP, Helfand M, Woolf SH, et al. Current methods of the US Preventive Services Task Force: a review of the process. Am J Prev Med 2001;20:2135. 28. Altman DG. Systematic reviews of evaluations of prognostic variables. BMJ 2001;323:224 8. 29. Anand IS, Kuskowski MA, Rector TS, et al. Anemia and change in hemoglobin over time related to mortality and morbidity in patients with chronic heart failure: results from Val-HeFT. Circulation 2005; 112:11217. 30. van der Meer P, Voors AA, Lipsic E, Smilde TD, van Gilst WH, van Veldhuisen DJ. Prognostic value of plasma erythropoietin on mortality in patients with chronic heart failure. J Am Coll Cardiol 2004;44: 637. 31. Terrovitis JV, Anastasiou-Nana MI, Alexopoulos GP, et al. Prevalence and prognostic signicance of anemia in patients with congestive heart failure treated with standard vs high doses of enalapril. J Heart Lung Transplant 2006;25:333 8. 32. Tanner H, Moschovitis G, Kuster GM, et al. The prevalence of anemia in chronic heart failure. Int J Cardiol 2002;86:11521. 33. Szachniewicz J, Petruk-Kowalczyk J, Majda J, et al. Anaemia is an independent predictor of poor outcome in patients with chronic heart failure. Int J Cardiol 2003;90:303 8. 34. Sharma R, Francis DP, Pitt B, Poole-Wilson PA, Coats AJ, Anker SD. Haemoglobin predicts survival in patients with chronic heart failure: a substudy of the ELITE II trial. Eur Heart J 2004;25:1021 8. 35. Rosolova H, Cech J, Simon J, et al. Short to long term mortality of patients hospitalised with heart failure in the Czech Republica report from the EuroHeart Failure Survey. Eur J Heart Fail 2005;7: 780 3. 36. OMeara E, Clayton T, McEntegart MB, et al. Clinical correlates and consequences of anemia in a broad spectrum of patients with heart failure: results of the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity (CHARM) program. Circulation 2006;113:986 94. 37. Mozaffarian D, Nye R, Levy WC. Anemia predicts mortality in severe heart failure: the prospective randomized amlodipine survival evaluation (PRAISE). J Am Coll Cardiol 2003;41:19339. 38. McClellan WM, Flanders WD, Langston RD, Jurkovitz C, Presley R. Anemia and renal insufciency are independent risk factors for death among patients with congestive heart failure admitted to community hospitals: a population-based study. J Am Soc Nephrol 2002;13:1928 36. 39. Kosiborod M, Smith GL, Radford MJ, Foody JM, Krumholz HM. The prognostic importance of anemia in patients with heart failure. Am J Med 2003;114:1129. 40. Kerzner R, Gage BF, Freedland KE, Rich MW. Predictors of mortality in younger and older patients with heart failure and preserved or reduced left ventricular ejection fraction. Am Heart J 2003;146: 286 90. 41. de Silva R, Rigby AS, Witte KK, et al. Anemia, renal dysfunction, and their interaction in patients with chronic heart failure. Am J Cardiol 2006;98:391 8. 42. Akhter MW, Aronson D, Bitar F, et al. Effect of elevated admission serum creatinine and its worsening on outcome in hospitalized patients with decompensated heart failure. Am J Cardiol 2004;94:957 60. 43. Varat MA, Adolph RJ, Fowler NO. Cardiovascular effects of anemia. Am Heart J 1972;83:41526. 44. Dries DL, Exner DV, Domanski MJ, Greenberg B, Stevenson LW. The prognostic implications of renal insufciency in asymptomatic and

Groenveld et al. Anemia and Mortality in Heart Failure

827

45. 46. 47. 48. 49.

50.

51.

52. 53.

54. 55. 56. 57.

58.

59. 60.

symptomatic patients with left ventricular systolic dysfunction. J Am Coll Cardiol 2000;35:6819. Foley RN. Myocardial disease, anemia, and erythrocyte-stimulating proteins in chronic kidney disease. Rev Cardiovasc Med 2005;6 Suppl 3:S2734. Ruilope LM, van Veldhuisen DJ, Ritz E, Luscher TF. Renal function: the Cinderella of cardiovascular risk prole. J Am Coll Cardiol 2001;38:17827. Smith GL, Lichtman JH, Bracken MB, et al. Renal impairment and outcomes in heart failure: systematic review and meta-analysis. J Am Coll Cardiol 2006;47:198796. Silverberg DS, Wexler D, Iaina A. The importance of anemia and its correction in the management of severe congestive heart failure. Eur J Heart Fail 2002;4:681 6. Silverberg DS, Wexler D, Sheps D, et al. The effect of correction of mild anemia in severe, resistant congestive heart failure using subcutaneous erythropoietin and intravenous iron: a randomized controlled study. J Am Coll Cardiol 2001;37:1775 80. Palazzuoli A, Silverberg D, Iovine F, et al. Erythropoietin improves anemia exercise tolerance and renal function and reduces B-type natriuretic peptide and hospitalization in patients with heart failure and anemia. Am Heart J 2006;152:1096 115. van Veldhuisen DJ, Dickstein K, Cohen-Solal A, et al. Randomized, double-blind, placebo-controlled study to evaluate the effect of two dosing regimens of darbepoetin alfa in patients with heart failure and anaemia. Eur Heart J 2007;28:2208 16. Ghali JK, Anand IS, Abraham WT, et al. Randomized double-blind trial of darbepoetin alfa in patients with symptomatic heart failure and anemia. Circulation 2008;117:526 35. Bennett CL, Silver SM, Djulbegovic B, et al. Venous thromboembolism and mortality associated with recombinant erythropoietin and darbepoetin administration for the treatment of cancer-associated anemia. JAMA 2008;299:914 24. Drueke TB, Locatelli F, Clyne N, et al. Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med 2006;355:2071 84. Singh AK, Szczech L, Tang KL, et al. Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 2006;355: 208598. Young JB, Anand IS, Diaz R, et al. Reduction of events with darbepoetin alfa in Heart Failure (RED-HF) trial (abstr). J Card Fail 2006;12 Suppl 1:S77. Okonko DO, Grzeslo A, Witkowski T, et al. Effect of intravenous iron sucrose on exercise tolerance in anemic and nonanemic patients with symptomatic chronic heart failure and iron deciency FERRICHF: a randomized, controlled, observer-blinded trial. J Am Coll Cardiol 2008;51:10312. Toblli JE, Lombrana A, Duarte P, Di GF. Intravenous iron reduces NT-pro-brain natriuretic peptide in anemic patients with chronic heart failure and renal insufciency. J Am Coll Cardiol 2007;50: 1657 65. Flather MD, Farkouh ME, Pogue JM, Yusuf S. Strengths and limitations of meta-analysis: larger studies may be more reliable. Control Clin Trials 1997;18:568 79. Tang WH, Tong W, Jain A, Francis GS, Harris CM, Young JB. Evaluation and long-term prognosis of new-onset, transient, and persistent anemia in ambulatory patients with chronic heart failure. J Am Coll Cardiol 2008;51:569 76.

Key Words: heart failure y anemia y prognosis.

APPENDIX

For the Acknowledgments and supplementary Tables 1 and 2, please see the online version of this article.

Downloaded From: http://content.onlinejacc.org/ on 08/26/2013