Anemia E Insuficiencia Cardiaca

Curriculum in Cardiology

Prevalence of anemia and effects on mortality in patients with heart failure JoAnn Lindenfeld, MD, FACC Denver, Colo

Background In patients with chronic kidney disease, the adverse cardiovascular effects of anemia have been well established. New data are emerging to suggest anemia may represent an important treatable cause of cardiac morbidity and mortality in patients with heart failure. To improve the understanding of the problem of anemia in heart failure, it is important to assess the factors that influence the prevalence of anemia and to assess the consistency of the association of anemia and mortality in various populations of patients with heart failure. Methods

A systematic review of the literature was conducted by performing detailed searches of MEDLINE and EMBASE, searching the bibliographies of the articles retrieved during the database search, and conferring with heart-failure experts involved in clinical trials.Twenty-eight publications from 26 studies that evaluated anemia prevalence with or without effects on mortality in patients with heart failure were identified. The definition of anemia used in each study was tabulated along with pertinent patient characteristics, the prevalence of anemia, and the association between anemia and mortality.

Results Anemia is common among patients with heart failure. The prevalence of anemia increases with increasing severity of heart failure, declining renal function, and increasing age. Anemia is consistently associated with poorer survival in all patient populations, but there are substantial differences in the patient populations and definition of anemia. Conclusions To clarify the prognostic relationship of anemia in patients with heart failure, a standard definition of anemia should be adopted accounting for the menopausal status of women. Age, severity of heart failure, evaluation of kidney function, important comorbidities, and use of angiotensin-converting-enzyme inhibitors should be included, and correctable causes of anemia should be excluded. Inclusion of these factors should allow better definition of the relationship between anemia and prognosis in patients with heart failure. (Am Heart J 2005;149:391-401.) Anemia and heart failure are each common medical problems that occur with increasing prevalence as patients age. At least 3.4 million Americans are estimated to be anemic,1,2 and nearly 5 million Americans have heart failure.3 Although anemia is clearly more prevalent in people with heart failure than in the general population, estimates of the actual prevalence of anemia among patients with heart failure have varied widely. The role that anemia plays in exacerbating cardiac disease in patients with end-stage renal disease has been clear for some time, but only recently has a potential relationship of anemia to long-term prognosis been

From the Cardiac Transplantation Program and the Center for Women’s Health Research, University of Colorado Health Sciences Center, Denver, Colo. Support for this literature review and preparation of the manuscript provided by Amgen Inc. Submitted May 25, 2004; accepted August 26, 2004. Reprint requests: JoAnn Lindenfeld, MD, FACC, Cardiac Transplantation Program, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Campus Box B130, Denver, CO 80262. E-mail: [email protected] 0002-8703/$ - see front matter n 2005, Elsevier Inc. All rights reserved. doi:10.1016/j.ahj.2004.08.039

described in patients with heart failure.4-12 The purposes of this paper are to define the prevalence of anemia in patients with heart failure, to describe the factors associated with an increased prevalence of anemia, to describe the relationship between anemia and mortality, and to suggest definitions and data that are critical in defining the relationship of anemia and prognosis in patients with heart failure.

Methods Search strategy This systematic review of the literature was designed to identify and review studies evaluating the prevalence and prognostic import of anemia in patients with heart failure. The search included the US National Library of Medicine database (MEDLINE) and the Excerpta Medica database (EMBASE). The bibliographies of the articles retrieved during the database search were also individually reviewed for articles matching the search criteria. The MEDLINE search included articles from January 1966 through August of 2004. The EMBASE search included January 1974 through August of 2004.

Selection criteria Studies were accepted if the population was defined, and the prevalence and prognosis of anemia were reported. Abstracts

American Heart Journal March 2005

392 Lindenfeld

Table I. Prevalence of anemia in patients with heart failure

Publication year

Number of patients

2002

152 584

Kosiborod

2003

2281

Connecticut

McClellan

2002

633

Atlanta, Ge

Hospital discharge population studies Szachniewicz 2003

176

Poland

Author/study name Medicare population studies Herzog

Study location

United States

Population in which prevalence was measured

Patients with heart failure from the 1998 (5%) General Medicare database (end-stage renal disease patients were excluded) Medicare patients 65 y of age or older who had been admitted with a principal discharge diagnosis of heart failure Medicare patients who had been admitted to the hospital with a primary discharge diagnosis of heart failure

Definition of anemia

Mean age (y)

Sex (% male)

ICD-9 codes

65-80+

NA

Hematocrit V37%

79 F 8 y

37% male

Hematocrit b40%

75.7 y

40% male

Patients admitted to the hospital with a diagnosis of heart failure

Hemoglobin b12.0 g/dL

63 y

68% male

United Kingdom

Hospital admissions/ discharges

Hemoglobin V11 g/dL

NA

NA

Alberta, Canada

A population-based cohort of patients with new-onset heart failure discharged from 138 acute-care hospitals Hospital discharges

ICD-9 codes

78 y

45% male

Cromie

2002

269

Ezekowitz

2003

12 065

Cleland/ EuroHeart Failure Survey

2003

9971

24 European countries

Wisniacki

2001

201

United Kingdom

Patients older than 65 y who had been admitted to the hospital with heart failure

Hemoglobin b12 g/dL

Wexler

2004

338

Israel

Patients admitted to hospital with a primary diagnosis of congestive heart failure

Hemoglobin b12 g/dL

76 y

60% male

Outpatient clinic population studies Tanner 2002

193

Switzerland

Patients from a tertiary heart failure outpatient clinic

Hemoglobin b12.0 g/dL

54 y

82% male

Cleveland, Ohio

Outpatient clinic–consecutive patients with standard ICD-9 diagnosis of heart failure and CBC evaluation Outpatients with heart failure at Veteran’s Administration Medical Center

Hemoglobin V12 g/dL, males; V11 g/ dL, females Hemoglobin b12.0 g/dL

62 y

64% male

Anemic patients 74.8 F 11.2 y; nonanemic patients 70.5 F 11.9 y

97% male

Tang

2003

2011

Hussein

2003

604

United States

Hemoglobin b11.0 g/dL

51% of females and 30% of males were N75 y N65 y

53% male

NA

American Heart Journal Volume 149, Number 3

NYHA class

Lindenfeld 393

Mean LVEF (%)

Comorbidities

Etiology of LV dysfunction (ischemic heart disease [%])

Mean hemoglobin (g/dL)

Serum creatinine (mg%)

Prevalence estimate

NA

NA

Chronic kidney disease (not end-stage renal disease) 13.6%,. Mean number of comorbidities was 13.9 F 3.1

NA

NA

NA

28%

NA

V20% in 11%; 20%-40% in 24%; N40% in 38% 38.4%

Diabetes 37%; hypertension 60%; renal insufficiency 20%; COPD 26%

NA

z2.0 mg/dL in 32%

Mean hematocrit 38%

48%

Chronic kidney disease 38%; hypertension 66%; diabetes 44%

NA

1.46 mg/dL

Mean hematocrit 36.6%

69.7%

9% NYHA class I; 46% NYHA class II; 29% NYHA class III; 16% NYHA class IV

42%

Diabetes 27%; hypertension 8%

62%

1.2 mg/dL

14.0 g/dL

100% NYHA class IV NA

NA

Renal impairment 44%

NA

NA

NA

10%; 0% in patients with NYHA class I; 9% in class II; 10% in class III; 21% in class IV 14.4%

NA

Diabetes 26%, hypertension 35%, hyperlipidemia 4% ; COPD 30%; chronic renal insufficiency 19%

42%

NA

NA

17%

36% NYHA class I, 37% NYHA class II, 26% NYHA class III/IV NYHA class I-IV

b40% in 51% of males, 28% of females NA

Diabetes 27%, hypertension 53%

NA

z150 Mmol/L in 16% and z200 Mmol/L in 7%

NA

18% males; 23% females

NA

NA

NA

NA

20% NYHA class III; 81% NYHA class IV

NA

Diabetes 36%, hypertension 65%, hyperlipidemia 61%

77%

1.7 F 1.1 mg%

12.0 F 1.8 g/dL

49.8%; 0% in patients with NYHA class I; 36.4% in patients with NYHA class II; 52.0% in class III; 65.9% in class IV 52.4%

7% NYHA class I; 36% NYHA class II; 41% NYHA class III; 16% NYHA class IV

29%

Diabetes 19%; hypertension 31%; dyslipidemia 43%

41%

111 with 106 125 131

NA

NA

NA

NA

NA

14 g/dL in males; 13 g/dL in females

NA

Anemic patients 47%; nonanemic patients 44%

NA

NA

Anemic patients 1.37 F 0.50 mg/dL; nonanemic patients 1.42 F 0.56 mg/dL

NA

NA

Mmol/L in patients NYHA class I; Mmol/L in class II; Mmol/L in class III; Mmol/L in class IV

14.2 g/dL in patients with NYHA class I; 14.0 g/dL in class II; 13.6 g/dL in class III; 13.8 g/dL in class IV

15%; 7% in patients with NYHA class I; 9% in class II; 17% in class III; 26% in class IV 16.1%

21%

(continued on next page)

American Heart Journal March 2005

394 Lindenfeld

Table 1. continued

Author/study name

Publication year

Outpatient clinic population studies Horwich 2002

Number of patients

Study location

1061

Los Angeles, Calif

Population in which prevalence was measured

Patients referred for heart transplantation evaluation (excluded patients with LVEF N40%, NYHA class b3, or those without an initial hemoglobin level, bdryQ weight, or adequate follow-up data) Patients with heart failure referred to a rapid-access clinic Patients with chronic heart failure attending the heart failure clinic of the Royal Brompton Hospital Patients with chronic heart failure referred to heart failure outpatient clinic

Definition of anemia

Mean age (y)

Sex (% male)

Hemoglobin b13 g/dL, males; b12 g/dL, females

16-72 y

77% male

Hemoglobin b13 g/dL

76 y

54% male

Hemoglobin b13 g/dL

61 y

100% male

Hemoglobin b12 g/dL

70.1 F 11.1 y

79% male

Kalra

2003

552

United Kingdom

Kalra

2002

93

United Kingdom

Silverberg

2000

142

Israel

Androne

2003

196

New York, NY

Patients with advanced heart failure referred to a transplant clinic

Hematocrit b41% males; b38% females

51.5 y

77.5% male

Golden 2002

2002

239 157

Hematocrit L25 Hemoglobin b12.1 q/dL

68.6% male

2002

Outpatient heart failure population Outpatient HF

65.1

Bolger

North America London

NA

NA

2002

5010

International

Clinical trial participants— patients had LVEF b40%

62.7 y

80% male

Anand/ RENAISSANCE

2004

912

North America

Multicenter, double-blind, placebo-controlled clinical trial

Hemoglobin b12 g/dL, males; b11 g/dL, females Hemoglobin V12 g/dL

62 y

78% male

Anker/Pitt/ Sharma ELITE II

2002/ 2000/ 2004

3044

International —289 centers in 46 countries

Clinical trial participants— patients had LVEF V40%

Hemoglobin b12.5 g/dL

72 y

69% male

Packer/ COPERNICUS

2001

2289

334 centers in 21 countries

Double-blind, randomized clinical trial participants

Hemoglobin b12.5 g/dL

63 y

80% male

Mozaffarian/ PRAISE

2003

1130

Clinical trial participants with severe heart failure—patients had LVEF V30%; NYHA class IIIB or IV

Hematocrit V37.6%

65 y

76% male

Clinical trials Maggioni/ ValHeFT

American Heart Journal Volume 149, Number 3

NYHA class

Lindenfeld 395

Mean LVEF (%)

Comorbidities

Etiology of LV dysfunction (ischemic heart disease [%])

Serum creatinine (mg%)

Mean hemoglobin (g/dL)

Prevalence estimate

35% NYHA class III; 65% NYHA class IV

22%

NA

50%

NA

13.6 g/dL

30%

92% NYHA class III or IV

NA

NA

NA

NA

13.3 g/dL

36%

2.4 F 0.1

31%

NA

62%

117 F 5 Mmol/L

13.7 F 0.2 g/dL

39%

8% NYHA class I; 18% NYHA class II; 27% NYHA class III; 47% NYHA class IV

32.5 F 12.2

Diabetes 28%; hypertension 64%; dyslipidemia 72%

74%

1.6 F 1.1 mg/dL

11.9 F 1.5 g/dL

NA

26%

NA

NA

II-IV

26.6%

Diabetes 46%

Mean = 2.5

30%

NA

59.4% ischemia NA

NA

NA

55.6%; 9.1% in patients with NYHA class I; 19.2% in patients with NYHA class II; 52.6% in class III; 79.1% in patients with NYHA class IV 61%; 33% in patients with NYHA class II; 68% in patients with NYHA class IV 27.6%

NA

NA

29%

1.65 mg/dL males; 1.17 mg/dL females

12.4 g/dL males; 11.9 g/dL females

NYHA class II-IV

23% NYHA class II; 47% NYHA class IIIa; 30% NYHA class IIIb/IV NYHA class 2.5 F 0.6; 52% NYHA class II; 43% NYHA class III; 5% NYHA class IV

9.0%

22%

NA

NA

NA

13.8 F 1.6 g/dL

31%

Diabetes 24%; hypertension 49%; COPD 6% in patients with hemoglobin b12.5 g/dL; 8% in patients with hemoglobin 12.5-13.9 g/dL; 9% in patients with hemoglobin 14.0-15.0; g/dL; 10% in patients with hemoglobin N15.0; g/dL

79%

104 F 29 Mmol/L

14.0 g/dL

67%

134 Mmol/L

64%

1.4 F 0.5 mg/dL

19.9%

81% NYHA class IIIB; 19% NYHA class IV

21%

Diabetes 37%

12%

16.9%; 27.5% females; 12.0% males

19%

13.9 g/dL

20%

(continued on next page)

American Heart Journal March 2005

396 Lindenfeld

Table 1. continued

Author/study name

Publication year

Outpatient clinic population studies Al-Ahmed/ 2001 SOLVD

Felker/ OPTIME-CHF

2003

Number of patients

Population in which prevalence was measured

Study location

6563

906

United States

Clinical trial participants— patients had LVEF V35%; excluded patients with serum Cr N2.5 mg/dL, recent MI, unstable angina, severe pulmonary disease, uncontrollable hypertension, major cerebrovascular disease, or suspected renal artery stenosis Clinical trial participants with decompensated heart failure, significantly depressed systolic function

Figure 1

Mean age (y)

Sex (% male)

Hematocrit V39%

60 y

86% male

Hemoglobin b13 g/dL, males; b12 g/dL, females

65 y

NA

Figure 2

70

90

60

80

50 40 30 20 10 0 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105

Mean Age (years)

Prevalence of Anemia (%)

Prevalence of Anemia (%)

Definition of anemia

70 60

5

50

18

40

23 24

30

The prevalence of anemia vs mean age of subjects.

Anand

20 10 0

4-6,8,10-16,20-31

17

l

l

l

l

l

l

I

II

IIIa

III

IIIb + IV

IV

NYHA Class

were accepted if prevalence of anemia, definition of anemia, and patient population were described.

Data abstraction Two reviewers evaluated the studies retrieved and agreed upon the final studies included. Twenty-eight publications from 26 studies that evaluated anemia prevalence in patients with heart failure were identified and grouped by population studied. Three publications were used to garner data from a single study—Elite II.13 - 15 The definition of anemia used in each study was tabulated along with pertinent patient characteristics, such as mean age, New York Heart Association (NYHA) functional class, mean serum creatinine level, etiology of heart failure, prevalence of comorbidities, and the prevalence of anemia. The association of anemia with mortality was tabulated in a second table.

Results Twenty-six studies from 28 publications reported the prevalence of anemia in patients with heart failure and

The prevalence of anemia vs NYHA class.6,16,20,21,26,27

included a total of 203 600 patients. Three studies evaluated Medicare populations, 6 studies evaluated patients discharged from hospital, 10 studies evaluated outpatient populations, and 7 studies included patients from clinical research studies (Table I).5 - 7,9-29 The definitions of anemia used in these studies varied significantly. Two studies defined anemia by ICD-9 codes.5,17 Five studies defined anemia as a hematocrit of b35%, 37%, 37.6%, 39%, 40%, and 41%.4,7,10 - 12,27 The remaining 19 studies defined anemia as a hemoglobin concentration of b11 to 13 g/dL. Four of the 22 studies defined anemia differently in men and women, using a lower value of hemoglobin or hematocrit in women.22,23,27,28 The prevalence of anemia varied considerably, from 9% to 69.7%, based on the definition of anemia used and the patient population studied. In

American Heart Journal Volume 149, Number 3

NYHA class

Lindenfeld 397

Mean LVEF (%)

Etiology of LV dysfunction (ischemic heart disease [%])

Comorbidities

Mean hemoglobin (g/dL)

Serum creatinine (mg%)

Prevalence estimate

46% NYHA class I; 42% NYHA class II; 12% NYHA class III/IV

27%

Diabetes 19%; hypertension 39%

79%

1.18 mg/dL

Mean hematocrit 43%

22%

92% NYHA class III or IV

23%

NA

NA

NA

12.6 F 1.8 g/dL

49%

Figure 3

Figure 4

Prevalence of Anemia (%)

Prevalence of Anemia (%)

80 70 60 50 40 30 20

80 70 60 50 40 30 20 10 0 0

10

10

20

30

40

50

60

Prevalence of Diabetes (%)

0 0

5

10

15

20

25

30

35

40

45

50

Percentage of Patients With Chronic Renal Insufficiency

The prevalence of anemia vs percentage of subjects with diabetes.4,5,7,10 - 12,16,19,21,26,30

The prevalence of anemia vs percentage of subjects with chronic kidney disease.5,10,11,17,18,30

19 of 26 studies, mean age was reported, and the relationship between mean age and prevalence of anemia is demonstrated in Figure 1. In general, increasing age was associated with an increasing prevalence of anemia. An exception to this relationship was seen in the study of Androne et al who reported a 61% prevalence of anemia in patients with a mean age of 51.5 years. However, these were patients from a referral population patients with advanced heart failure.27 The association between increasing NYHA class and increasing prevalence of anemia from 6 studies is shown in Figure 2. The prevalence of anemia does appear to increase as NYHA class worsens. The frequencies of comorbidities such as chronic kidney disease, diabetes, and hypertension are shown in Table I. Only 6 studies reported the prevalence of

chronic kidney disease.5,10,11,17,18,30 There appears to be an association between an increasing prevalence of chronic kidney disease and an increasing prevalence of anemia (Figure 3). Cromie reported a 44% incidence of chronic kidney disease, but only a 17% prevalence of anemia, much lower than in other studies.18 However, the definition of anemia in this study was a hemoglobin of b11 g/dL, the lowest cutoff in any of the studies. The prevalence of diabetes was reported in 12 studies, of which 11 reported a mean prevalence of anemia for all subjects. Figure 4 demonstrates the relationship between prevalence of diabetes and prevalence of anemia. Although the data are scattered, it appears that the prevalence of anemia increases with an increasing prevalence of diabetes. Only 2 studies reported the prevalence of anemia by sex. In Elite II, the prevalence of anemia was 27.5% in women and 12% in men using a hemoglobin of

American Heart Journal March 2005

398 Lindenfeld

Table II. Association of anemia with mortality Author Medicare populations studies Kosiborod McClellan Hospital discharge population studies Szachniewicz Ezekowitz Outpatient clinic population studies Horwich Kalra Golden Bolger Clinical trials Sharma Mozaffarian Al-Ahmad Felker Anand

Follow-up

Outcome (mortality)*

1y 12 mo

1% decrease in Hct = 2% increase in mortality 1% increase in Hct = 1.6% decrease in mortality

mean = 529 d mean = 573 d

anemia vs no anemia HR = 2.61 (CI 1.05-6.47) anemia vs no anemia HR = 1.34 (CI 1.24-1.46)

1y median = 3 y mean = 23.1 mo mean = 31 mo

1 g/dL decrease in Hb = 16% increase in mortality no association with mortality decrease in Hct single most powerful predictor of mortality 1 g/dL decrease in Hb = 38% increase in mortality

median = 551 d mean = 15 mo mean = 33.4 mo 60 d mean = 12.7 mo

Nonlinear mortality that increases above and below Hb = 14.5 g/dL Hct only associated with mortality in the lowest quintile 1% decrease in Hct = 2.7% increase in mortality 1 g/dL increase in Hb = 11% decrease in mortality and rehospitalization 1 g/dL increase in Hb = 15.8% decrease in mortality

Hb, hemoglobin; Hct, hematocrit; HR, hazard ratio. *Cox proportional hazards model in all studies.

b12.5 g/dL as the definition of anemia.13 - 15 In the study by Cleland et al, the prevalence of anemia was 23% in women and 18% in men with anemia defined as a hemoglobin of b11 g/dL.19 Both studies included a very elderly population. Two studies evaluated the potential underlying causes of anemia in patients with heart failure. Cromie et al evaluated 269 patients requiring acute admission to a single hospital in Scotland for heart failure between June of 1998 and December of 2000.18 Anemia defined as a hemoglobin of b11 g/dL was found in 14.4% of patients. There were no patients with iron, B12, or folate deficiencies. Ezekowitz et al evaluated a populationbased cohort of patients with new-onset heart failure discharged from 138 acute care hospitals in Alberta, Canada.5 Using ICD-9 codes, 17% of all patients had anemia and 58% of these (9.9% of all patients) were felt to have anemia of chronic disease. The remainder had identifiable causes of anemia. Both studies emphasized the strong association of chronic renal insufficiency with anemia.5,18 Only 1 study reported the prevalence of true anemia vs anemia due to hemodilution. Androne et al reported that 46% of 37 anemic patients with advanced heart failure had a normal red blood cell volume with increased plasma volume (hemodilution), whereas the remaining patients had a true anemia.27 The patients with hemodilution had considerably poorer survival than those patients with true anemia. Table II describes the 13 studies that reported the relationship of anemia to mortality. All studies used a Cox proportional hazards model. All but 1 study reported increased mortality associated with anemia. Six studies reported a linear association between increasing mortality and decreasing hemoglobin or

Figure 5 +40% +38% +36% +34% +32% +30% +28% +26% +24% +22% +20% +18% +16% +12% Mortality +8% +6% +4% -3%

-2%

-1%

+2%

1%

2%

3%

Hct (%)

1.0

Hb (g/dL)

-

-1.0g/dL

-.05

-2%

0.5

-4% -6% -8% Mortality -12% -14% -16% -18% -20% -22% -24% -26%

Changes in mortality associated with changes in hemoglobin (Hb) in grams per decaliter or hematocrit (Hct) in percent.4,6,8 - 12,31 Filled circle, mortality; filled square, mortality and rehospitalization.

hematocrit.4,6,8 - 11,31 One short-term study demonstrated a linear association between lower hemoglobin and increased mortality and rehospitalization.8 These studies

American Heart Journal Volume 149, Number 3

are presented in Figure 5. Data from the Elite II trial reported a nonlinear relationship of anemia and mortality with mortality increasing above or below a hemoglobin of 14.5.15 Mozaffarian et al reported an association of mortality with anemia only in patients in the lowest hematocrit quintile.12 In the study of Bolger et al, anemia lost predictive value for mortality when cytokine measurements were added to the multivariate analysis.6 The 3 remaining studies reported a significantly increased mortality in anemic patients vs those without anemia.5,7,16

Discussion The prevalence of anemia in patients with heart failure varies enormously among studies. This marked variation is due to a number of factors, including differences in the definition of anemia, substantial differences in populations studied, a lack of information about correctable causes of anemia, and no control for differences in true anemia vs anemia due to hemodilution. An important factor limiting the assessment of the prevalence of anemia is the lack of a uniform definition of anemia. Any determination of the prevalence of anemia depends heavily on the precise definition of anemia used—and the definitions of anemia found in our literature review varied widely. Some studies employed the World Health Organization (WHO) definition (hemoglobin b13 g/dL in men and postmenopausal women; b12 g/dL in premenopausal women).32 Still other studies used different criteria, such as hematocrit levels or the ICD-9 codes used for hospital admissions, to define anemia. The National Kidney Foundation definition (hemoglobin V12.5 g/dL in men and postmenopausal women; V11 g/dL in premenopausal women) was not used in any of these studies.33 These differences make comparisons among studies difficult. Overall, studies that identified patients with anemia by ICD-9 codes reported lower frequencies of anemia than similar studies that used laboratory values. Anemia definitions should take women’s menopausal status into consideration—and the same criteria used for men should be applied to postmenopausal women.32,33 However, several studies have used anemia definitions with a lower hemoglobin cutoff for women, even when (because of their ages) the majority of women in the study would have been postmenopausal.8,19,24,25,33 Thus, it is likely that many studies have underestimated the prevalence of anemia in postmenopausal women with heart failure. However, other studies may have overestimated the prevalence of anemia in women compared with men for another reason. In most studies of heart failure, women are generally older than men and the prevalence of anemia increases with age in the general population.8,34 For example, although Cleland reported a 23% prevalence of anemia in women compared with 18% in men,

Lindenfeld 399

51% of the women were older than 75 years compared with only 30% of the men. Although there is no single best definition of anemia, it would be helpful if studies adopted the WHO or National Kidney Foundation definitions, both of which define anemia by menopausal status in women.32,33 These data highlight the importance of correcting for age and menopausal status when comparing the importance of anemia on survival in patients with heart failure. Another important cause of variation in the reported prevalence of anemia in patients with heart failure is the difference in populations. As shown in Figures 1–4, the prevalence of anemia in these patients increases with increasing age, more severe heart failure as estimated by NYHA classification, and increasing prevalence of important comorbidities such as diabetes and chronic kidney disease. Although anemia in patients with heart failure is often attributed to bchronic disease,Q it seems likely that a substantial amount of anemia may be related to chronic kidney disease and, possibly, to diabetes.5,18 Certainly, these data should be reported in all studies evaluating the prognostic import of anemia in these patients. The prevalence of reversible causes of anemia in patients with heart failure has been poorly defined. At least 2 studies suggest that 40% to 50% of patients have a reversible cause not related to heart failure.4,16 Increasing age is likely to be associated with a higher incidence of reversible causes of anemia, such as iron deficiency and B12 deficiency.35 - 38 Indeed, macrocytosis and normal B12 levels may not exclude anemia due to B12 deficiency, especially in an elderly white population.35 - 38 If correctable etiologic factors for anemia not associated with heart failure are not excluded, the relationship between anemia and heart failure will be difficult to accurately assess. The actual prevalence of true anemia and anemia due to hemodilution in patients with heart failure has been defined only in a small population of patients.27 However, it appears that anemia due to hemodilution may account for half of these patients and is associated with a poorer prognosis than true anemia.27 Unfortunately, at present, no simple test applicable to large populations exists to separate these patients. Finally, as ACE-I therapy has been associated with a modest decrease in hemoglobin, the use of ACE-I should be reported in all studies.39,40 Twelve of 13 studies have shown that anemia is associated with increased mortality in patients with heart failure (Table II). In the majority of the studies the association was a linear one, with decreasing hemoglobin or hematocrit associated with increasing mortality. However, all of these studies use different definitions of anemia, report different populations, and lack consistency in the correction for other factors such as diabetes and chronic renal insufficiency. Furthermore, none of

American Heart Journal March 2005

400 Lindenfeld

the studies that report on the association of anemia and mortality has consistently excluded correctable causes of anemia. Although there is substantial variation in the reported prevalence of anemia and its effect on mortality in patients with heart failure, some general conclusions can be derived from this analysis. Anemia is common among patients with heart failure, and the prevalence of anemia increases with the severity of heart failure, decline in renal function, and with increased age of the population and possibly with associated comorbidities such as diabetes. Anemia is more common among women than among men (although definitions of anemia vary), but it is not certain if anemia is more common in postmenopausal women when the prevalence of anemia is corrected for age. Anemia is associated with an increased mortality and the more severe the anemia the higher the mortality. However, these studies lack uniform definitions of anemia and patient populations. Future studies of anemia in patients with heart failure should use either the WHO or National Kidney Foundation definitions of anemia and should report anemia by sex along with important comorbidities such as renal function and prevalence of diabetes and the use of ACE-I. A standard evaluation for correctable causes of anemia should be carried out. Inclusion of these factors should allow a much more accurate estimation of the association of anemia with prognosis in patients with heart failure.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

References 1. National Center for Health Statistics. Fast-stats A to Z. In: Centers for Disease Control and Prevention US Department of Health and Human Services. Hyattsville (Md). Available at: http://www. cdc.gov/nchs/fastats/anemia.htm. Accessed April 1, 2003. 2. U.S. Census Bureau population estimates for July 1, 1996. Available at: http://doe.state.wy.us/lmi/0197/0197a3.htm. Accessed January 3, 2005. 3. http://www.americanheart.org; 2004. 4. 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. 5. 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;107:223 - 5. 6. Bolger AP, Wolfram D, Sharma R, et al. Anaemia in chronic heart failure: the relationship to inflammatory cytokine expression and prognostic importance. Circulation 2002;106:S2-S570. 7. Golden JS, Fallick CC, Josephson SR, et al. Baseline hematocrit predicts clinical outcomes in heart failure. Circulation 2002;106: S2-S680. 8. Felker GM, Gattis WA, Leimberger JD, et al. Usefulness of anemia as a predictor of death and rehospitalization in patients with decompensated heart failure. Am J Cardiol 2003;92:625 - 8. 9. Horwich TB, Fonarow GC, Hamilton MA, et al. Anemia is associated with worse symptoms, greater impairment in functional capacity and

20. 21. 22. 23. 24.

25.

26.

27. 28.

a significant increase in mortality in patients with advanced heart failure. J Am Coll Cardiol 2002;39:1780 - 6. Kosiborod M, Smith GL, Radford MJ, et al. The prognostic importance of anemia in patients with heart failure. Am J Med 2003;114:112 - 9. McClellan WM, Flanders WD, Langston RD, et al. Anemia and renal insufficiency 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. 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;41:1933 - 9. Anker SD, Sharma R, Francis D, et al. Patients with chronic heart failure (CHF) in the ELITE II Trial. Circulation 2002; (Suppl)106:472. Pitt B, Poole-Wilson PA, Segal R, et al. Effect of losartan compared with captopril on mortality in patients with symptomatic heart failure: randomised trial—the Losartan Heart Failure Survival Study ELITE II. Lancet 2000;355:1582 - 7. Sharma R, Francis DP, Pitt B, et al. Haemoglobin predicts survival in patients with chronic heart failure: a substudy of the ELITE II trial. Eur Heart J 2004;25:1021 - 8. 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. Herzog CA, Li S, Guo H, et al. The prevalence of CHF, chronic kidney disease, anemia, and multiple comorbid conditions in the Medicare population. J Card Fail 2002;8(Suppl):S63. Cromie N, Lee C, Struthers AD. Anaemia in chronic heart failure: what is its frequency in the UK and its underlying causes? Heart 2002;87:377 - 8. Cleland JG, Swedberg K, Follath F, et al. The EuroHeart Failure survey programme—a survey on the quality of care among patients with heart failure in Europe: Part 1. Patient characteristics and diagnosis. Eur Heart J 2003;24:442 - 63. Wisniacki N, Aimson P, Lyle M. Is anemia a cause of heart failure in the elderly? Heart 2001;85(Suppl):P4. Tanner H, Moschovitis G, Kuster GM, et al. The prevalence of anemia in chronic heart failure. Int J Cardiol 2002;86:115 - 21. Tang WHW, Miller H, Partin M, et al. Anemia in heart failure: a single-center experience. J Am Coll Cardiol 2003;41:157A. Hussein SJ, Jain R, Shlipak MG, et al. Chronic heart failure is not an independent cause of anemia. J Card Fail 2003;9:564. Kalra PR, Bolger AP, Francis DP, et al. Effect of anemia on exercise tolerance in chronic heart failure in men. Am J Cardiol 2003;91:888 - 91. Kalra PR, Collier T, Cowie MR, et al. Haemoglobin concentration and prognosis in new cases of heart failure. Lancet 2003;362: 211 - 2. Silverberg DS, Wexler D, Blum M, et al. The use of subcutaneous erythropoietin and intravenous iron for the treatment of the anemia of severe, resistant congestive heart failure improves cardiac and renal function and functional cardiac class, and markedly reduces hospitalizations. J Am Coll Cardiol 2000;35:1737 - 44. Androne AS, Katz SD, Lund L, et al. Hemodilution is common in patients with advanced heart failure. Circulation 2003;107:226 - 9. Maggioni AP, Anand I, Gottlieb SO, et al. Effects of valsartan on morbidity and mortality in patients with heart failure not receiving angiotensin-converting enzyme inhibitors. J Am Coll Cardiol 2002;40:1414 - 21.

American Heart Journal Volume 149, Number 3

29. Packer M, Fowler MB, Roecker EB, et al. Effect of carvedilol on the morbidity of patients with severe chronic heart failure: results of the Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS) study. Circulation 2002;106:2194 - 9. 30. Wexler D, Silverberg D, Sheps D, et al. Prevalence of anemia in patients admitted to hospital with a primary diagnosis of congestive heart failure. Int J Cardiol 2004;96:79 - 87. 31. Anand I, McMurray JJ, Whitmore J, et al. Anemia and its relationship to clinical outcome in heart failure. Circulation 2004;110:149 - 54. 32. World Health Organization. Iron deficiency anaemia assessment, prevention and control: a guide for programme managers. Geneva (Switzerland): World Health Organization; 2001. 33. National Kidney Foundation. National Kidney Foundation Kidney Disease Outcomes Quality Initiative. Clinical practice guidelines for anemia of chronic kidney disease: 2000 Update. Am J Kidney Dis 2001;37(Suppl 1):S182-S238. 34. Lindenfeld J, Fiske KS, Stevens BR, et al. Age, but not sex, influences the measurement of ejection fraction in elderly patients hospitalized for heart failure. J Card Fail 2003;9:100 - 6.

Lindenfeld 401

35. Vaccarino V, Chen YT, Wang Y, et al. Sex differences in the clinical care and outcomes of congestive heart failure in the elderly. Am Heart J 1999;138:835 - 42. 36. Pennypacker LC, Allen RH, Kelly JP, et al. High prevalence of cobalamin deficiency in elderly outpatients. J Am Geriatr Soc 1992;40:1197 - 204. 37. Lindenbaum J, Rosenberg IH, Wilson PW, et al. Prevalence of cobalamin deficiency in the Framingham elderly population. Am J Clin Nutr 1994;60:2 - 11. 38. Joosten E, van den Berg A, Riezler R, et al. Metabolic evidence that deficiencies of vitamin B-12 (cobalamin), folate, and vitamin B-6 occur commonly in elderly people. Am J Clin Nutr 1993;58: 468 - 76. 39. Chatterjee B, Nydegger UE, Mohacsi P. Serum erythropoietin in heart failure patients treated with ACE-inhibitors or AT(1) antagonists. Eur J Heart Fail 2000;2:393 - 8. 40. Usalan C, Erdem Y, Caglar M, et al. Effect of enalapril on exaggerated erythropoietin response to phlebotomy in erythrocytosic renal transplant patients. Nephrol Dial Transplant 1998;13:2884 - 9.