Evaluation and Long-Term Prognosis of New-Onset, Transient, and Persistent Anemia in Ambulatory Patients with Chronic Heart Failure

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Journal of the American College of Cardiology Vol. 51, No. 5, 2008 © 2008 by the American College of Cardiology Foundation ISSN 0735-1097/08/$34.00 Published by Elsevier Inc. doi:10.1016/j.jacc.2007.07.094

Heart Failure

Evaluation and Long-Term Prognosis of New-Onset, Transient, and Persistent Anemia in Ambulatory Patients With Chronic Heart Failure

W. H. Wilson Tang, MD, FACC,* Wilson Tong, MSC,* Anil Jain, MD,†‡ Gary S. Francis, MD, FACC,* C. Martin Harris, MD, MBA,‡ James B. Young, MD, FACC* Cleveland, Ohio

Objectives This study sought to determine the characteristics and long-term prognosis of anemia in ambulatory patients with chronic heart failure.

Background Anemia is prevalent in heart failure, and may portend poor outcomes.

Methods We reviewed 6,159 consecutive outpatients with chronic stable heart failure at baseline, short-term (3-month) follow-up, and long-term (6-month) follow-up between 2001 and 2006. Clinical, demographic, laboratory, and echocardiographic data were reviewed from electronic medical records. Mortality rates were determined from 6-month follow-up to end of study period.

Results Prevalence of anemia (hemoglobin [Hb] Ͻ12 g/dl for men, Ͻ11 g/dl for women) was 17.2% in our cohort. Diabetes, B-natriuretic peptide, left ventricular ejection fraction, and estimated glomerular ﬁltration rate were independent predictors of baseline anemia. Documented evaluation of anemia was found in only 3% of all anemic patients, and better in internal medicine than in cardiology clinics. At 6-month follow-up, new-onset anemia developed in 16% of patients without prior anemia, whereas 43% patients with anemia at baseline had resolution of their hemoglobin levels. Higher total mortality rates were evident in patients with persistent anemia (58% vs. 31%, p Ͻ 0.0001) or with incident anemia (45% vs. 31%, p Ͻ 0.0001) compared with those with without anemia at 6 months.

Conclusions These observations in a broad unselected outpatient cohort suggest that anemia in patients with heart failure is under-recognized and underevaluated. However, resolution of anemia was evident in up to 43% of patients who presented initially with anemia, and did not pose greater long-term risk for all-cause mortality. However, the presence of persistent anemia conferred poorest survival in patients with heart failure when compared with that of incident, resolved, or no anemia. (J Am Coll Cardiol 2008;51:569–76) © 2008 by the American College of Cardiology Foundation

Anemia has become increasingly recognized as an important applicability of their observations (8). Furthermore, data factor in the development and progression of heart failure. regarding the changes in Hb levels in patients over time and Anemia occurs with surprisingly high prevalence and por- the impact on prognoses remain limited. tends a poor prognosis (1–3). In a cohort of 1,061 patients with advanced heart failure referred for transplant evalua- See page 577 tion, lower hemoglobin (Hb) was associated with greater functional impairment, worse exercise capacity, and in- Because of the implementation of an electronic medical creased mortality (4). However, most analyses have been record system in our practice since 2001, we have the ability limited to data sets from hospitalized patients (5), post-hoc to characterize anemia in an unselected cohort of ambula- analyses of clinical trials with primarily baseline measure- tory patients with a clinical diagnosis of chronic heart failure ments (6), or interrogation of large administrative databases seen in internal medicine and cardiology clinics. The objec- (7). Bias of ascertainment in these populations can limit the tive of this paper is to determine the prevalence and characteristics of anemia in ambulatory patients with From the *Department of Cardiovascular Medicine, †Department of Internal chronic heart failure and to identify the impact of changes in Medicine, and ‡Information Technology Division, Cleveland Clinic, Cleveland, anemia status over time to long-term survival across patients Ohio. Drs. Tang, Francis, and Young are consultants for Amgen, Inc. Manuscript received May 14, 2007; revised manuscript received July 20, 2007, with varying degrees of renal insufﬁciency and left ventric- accepted July 23, 2007. ular systolic function. 570 Tang et al. JACC Vol. 51, No. 5, 2008 Anemia in Ambulatory Heart Failure Patients February 5, 2008:569–76

Abbreviations Methods (including iron studies, reticulocyte count, vitamin B12 and and Acronyms folate levels, and gastrointestinal bleeding workup) were Study population. The Cleve- identified. angiotensin- land Clinic Institutional Review ؍ ACE Short-term follow-up was defined as the next available converting enzyme Board approved this study proto- Hb measurement within 3 months from baseline, which -B-type natriuretic col. We reviewed our heart fail ؍ BNP results in a mean 0.6 Ϯ 0.8 months of follow-up. Patients peptide ure registry compiled from an must also have had consecutive Hb measurements within estimated electronic medical record system ؍ eGFR glomerular filtration rate 100 days of the 6-month follow-up to be included in the (EpicCare, Epic Systems Corp., .(long-term follow-up (mean follow-up at 5.3 Ϯ 1.8 months ؍ Hb hemoglobin Madison, Wisconsin), identified Incidence of (new-onset) anemia was defined as the pres- International 6,159 consecutive ambulatory ؍ ICD-9 Classification of Diseases- ence of anemia at short-term (3-month) or long-term patients seen at the Cleveland Ninth Revision (6-month) follow-up in those patients without anemia at Clinic internal medicine or car- -left ventricular baseline. Resolution of anemia was classified as normaliza ؍ LVEF diology outpatient clinics with a ejection fraction tion of Hb levels (Ն12 g/dl for men and Ն11 g/dl for documented clinical diagnosis of mean corpuscular women) with at least improvement in Ն0.5 g/dl at ؍ MCV heart failure, and performed a volume follow-up in those who presented with anemia at baseline. complete blood count test be- Persistent anemia was said to be present if patients had tween January 2001 and April anemia at both baseline and at short-term or long-term 2006. The EpicCare system provides a choice of common follow-up. Proportions of incidence, resolution, and persis- diagnoses with corresponding International Classification of tence of anemia were determined relative to the total Diseases-Ninth Revision (ICD-9) codes to document clin- number of patients at follow-up. Changes in Hb levels over ical problems in clinical chart notes. We therefore con- the course of follow-up within the study period were also firmed the diagnosis by performing a search for the ICD-9 determined. codes related to heart failure to establish our study cohort: All echocardiograms and laboratory testing were per- 242.9, 398.91, 402.01, 402.11, 402.91, 404.01, 404.03-4, formed at the Cleveland Clinic and reported in EpicCare. 404.13, 404.91, 404.93, 425.1, 425.4-5, 425.7-9, 428.0-1, Echocardiographic left atrial diameter and area, interven- and 428.9. Patients must also have had documented left tricular septal thickness, posterior wall thickness, left ven- ventricular ejection fraction (LVEF) data within 100 days of tricular end-diastolic diameter, and LVEF were collected. the baseline Hb blood analysis for inclusion in the study. The LVEF was determined by the Simpson rule, whereas We excluded patients with a history of congenital heart left atrial area was determined from the area–length method disease, malignancies or inherited or acquired anemia as a gross surrogate of diastolic function (10). The LV mass (ICD-9 code 282.x to 285.x), or documented active gastro- was calculated using the American Society of Echocardiog- intestinal bleeding at the time of blood draw. If a patient raphy equation, defined as 0.8 ([(interventricular septal was seen at both internal medicine and cardiology clinics, he thickness ϩ posterior wall thickness ϩ left ventricular or she would be classified as being followed up in the end-diastolic diameter)3 Ϫ left ventricular end-diastolic cardiology clinic. diameter3] ϩ 0.6) adjusted for gender (10). Standard LV Data synthesis. Clinical, demographic, laboratory, and diastolic filling pattern was graded on a scale from I to IV documented primary and secondary diagnoses in the elec- based on mitral inflow E- and A-wave velocity profiles tronic medical record were reviewed. All clinical data were described previously (11,12). Mitral regurgitation and tri- matched to within 100 days of baseline Hb blood analysis, cuspid regurgitation severity was evaluated on a scale from 0 with the first occurrence defined as baseline. Body mass to 4ϩ, with 0 representing normal and 4ϩ signifying severe index and body surface area were derived from height and regurgitation by color Doppler inflow patterns. Right ven- weight measurements. Estimated glomerular filtration rate tricular systolic pressure was determined from estimated (eGFR) was calculated through the Modification of Diet in jugular venous pressure and maximum tricuspid regurgita- Renal Disease equation (9). Long-term all-cause mortality tion velocity using the modified Bernoulli equation. data were determined from electronic medical records and Statistical analyses. Results are presented in percentages validated by the social security death index. Death and (for dichotomous variables) or mean Ϯ standard deviation cardiac transplantation dates were used in survival analysis (for continuous variables). Univariable analyses were per- (censoring transplantation), and all patients were followed formed using 2-tailed chi-square and Student t tests, re- up until June 20, 2007. spectively. Multivariable logistic regression with forward We defined anemia using a conservative criteria: Hb Ͻ12 stepwise selection was performed to determine the predic- g/dl for men and Ͻ11 g/dl for women. Standard criteria for tors of baseline prevalence of anemia in our cohort (variables microcytic (mean corpuscular volume [MCV] Ͻ82 fl) and included baseline age, gender, hypertension, diabetes, macrocytic (MCV Ͼ98 fl) anemia were used. For those with MCV, medications (angiotensin-converting enzyme [ACE] anemia identified by blood work, diagnostic ICD-9 coding inhibitors/angiotensin receptor blockers, beta-blockers, loop for anemia (280.x to 281.x), and standard evaluation tests diuretics), B-type natriuretic peptide (BNP), eGFR, LVEF, JACC Vol. 51, No. 5, 2008 Tang et al. 571 February 5, 2008:569–76 Anemia in Ambulatory Heart Failure Patients

Baseline Characteristics and left atrial area). Trends for the prevalence of anemia Table 1 Baseline Characteristics across different predictors were assessed with the Cochran- Armitage test. Linear univariate correlation assessed rela- No Anemia Anemia p Value (1,058 ؍ n) (5,101 ؍ tions between changes in Hb and changes in clinical status. Baseline Variable (n Multivariate correlation to changes in Hb used a stepwise Clinical data Age (yrs) 65 Ϯ 14 69 Ϯ 14 Ͻ0.0001 model followed by standard least-squares fit. Survival was Gender (% male) 67 64 0.046 analyzed between patients with incident, resolved, or per- Race (% Caucasian) 82 79 0.007 sistent anemia versus those without anemia using Kaplan- Heart rate (beats/min) 74 Ϯ 13 75 Ϯ 13 0.10 Meier analysis. Long-term mortality data over a mean Body mass index (kg/m2) 30 Ϯ 729Ϯ 7 0.23 follow-up of 3.9 Ϯ 2.2 years were used to calculate time to Systolic blood pressure 120 Ϯ 23 117 Ϯ 24 0.037 death or end of follow-up (censoring cardiac transplanta- (mm Hg) tion) in survival analysis. Multivariate Cox proportional Diabetes mellitus (%) 25 35 Ͻ0.0001 hazards model was used to calculate risk ratios for indepen- Hypertension (%) 43 39 0.019 dent predictors of mortality with incremental increases in Myocardial infarction (%) 43 44 0.36 Mortality rate (%) 26 47 Ͻ0.0001 continuous variables. A p value of Ͻ0.05 was considered Laboratory data statistically significant. All statistical analyses were per- Hemoglobin (g/dl) 13.9 Ϯ 1.4 10.4 Ϯ 1.1 Ͻ0.0001 formed using SAS version 9.1 and JMP version 5.1 (SAS Mean corpuscular 90 Ϯ 688Ϯ 8 Ͻ0.0001 Institute, Cary, North Carolina). volume (fl) B-type natriuretic peptide 529 Ϯ 702 962 Ϯ 1,007 Ͻ0.0001 (pg/ml) Results Blood urea nitrogen (mg/dl) 24 Ϯ 14 38 Ϯ 24 Ͻ0.0001 Total cholesterol (mg/dl) 183 Ϯ 51 154 Ϯ 45 Ͻ0.0001 A total of 6,159 outpatients with a clinical diagnosis of heart Low-density lipoprotein 105 Ϯ 40 84 Ϯ 33 Ͻ0.0001 failure met the study inclusion and exclusion criteria (in- cholesterol (mg/dl) cluding 3,360 and 2,796 patients initially presenting in Triglycerides (mg/dl) 171 Ϯ 166 148 Ϯ 141 0.0005 internal medicine and cardiology clinics, respectively). The Sodium (mEq/l) 140 Ϯ 13 139 Ϯ 14 Ͻ0.0001 mean baseline Hb level was 14 Ϯ 2 g/dl in men and 13 Ϯ Serum creatinine (mg/dl) 1.20 Ϯ 0.74 1.79 Ϯ 1.36 Ͻ0.0001 2 g/dl in women in the overall study cohort. The majority of Estimated GFR 72 Ϯ 27 57 Ϯ 36 Ͻ0.0001 2 our patients in the cohort had been treated with ACE (ml/min/1.73 m ) Medications inhibitors, beta-blockers, and loop diuretics at baseline Iron supplementation (%) 7 21 Ͻ0.0001 (Table 1). The median LVEF was 30%, and median plasma Erythropoiesis stimulating 18Ͻ0.0001 BNP level was 325 pg/ml for those with available data at agents (%) baseline (n ϭ 2,128). Among the overall study cohort, 1,393 ACE inhibitor/ARB (%) 72/21 65/21 Ͻ0.0001 patients (23%) had complete blood count data available at Beta-blocker (%) 77 69 Ͻ0.0001 both short-term (3-month) and long-term (6-month) Calcium channel blocker (%) 22 24 0.032 follow-up time points (Fig. 1). Digitalis (%) 39 30 Ͻ0.0001 Prevalence of anemia. At baseline, the prevalence of ane- Nitrates (%) 37 44 Ͻ0.0001 mia was 17.2% (17.8% in men, 16.0% in women, n ϭ 6,159) Statin (%) 55 52 0.072 Loop diuretic (%) 80 86 Ͻ0.0001 (Fig. 1). At follow-up, the prevalence of anemia at baseline ϭ Thiazide diuretic (%) 18 24 Ͻ0.0001 in the subgroup was 23.2% (n 1,393), similarly when Echocardiographic data assessed at 3- and 6-month follow-up. Baseline character- LV ejection fraction (%-U) 34 Ϯ 17 35 Ϯ 17 0.025 istics between patients with versus without anemia are LV end-diastolic 6.4 Ϯ 1.0 6.4 Ϯ 1.2 0.080 shown in Table 1. Patients with anemia were older, had diameter (cm) more prevalent diabetes mellitus and heart failure medica- LA area (cm2) 27 Ϯ 929Ϯ 9 Ͻ0.0001 tion usage; had higher mortality rates, BNP, and blood urea Estimated LV mass (g) 281 Ϯ 105 281 Ϯ 102 0.93 nitrogen; and had lower eGFR, low-density lipoprotein, LV wall motion score 1.8 Ϯ 0.6 1.8 Ϯ 0.6 0.69 Ն Ͻ and more severe diastolic dysfunction (both diastolic stage Diastolic stage II (%) 46 60 0.0001 MR severity Ն3ϩ (%) 17 20 0.010 and left atrial area as surrogate measure). A higher preva- RV systolic pressure 39 Ϯ 14 43 Ϯ 14 Ͻ0.0001 lence of anemia was associated with male gender, particu- (mm Hg) larly at age Ͼ75 years (Fig. 2). Patients with moderately TR severity Ն3ϩ (%) 10 20 Ͻ0.0001 impaired renal function (eGFR Ͻ60 ml/min/1.73 m2), Ն ACE ϭ angiotensin-converting enzyme; ARB ϭ angiotensin receptor blocker; GFR ϭ glomerular more preserved LVEF above median ( 30%), and above filtration rate; LA ϭ left atrial; LV ϭ left ventricular; MR ϭ mitral regurgitation; RV ϭ right median plasma levels of BNP (Ͼ325 pg/ml) had signifi- ventricular; TR ϭ tricuspid regurgitation. cantly higher prevalence of anemia (Fig. 3). These factors were found to be independent predictors of the prevalence Patterns of diagnostic evaluation and treatment of ane- of anemia in heart failure patients after multivariate logistic mia. The majority (71.6%) were normocytic (MCV 82 to regression analyses (Table 2). 98 fl). Microcytic (MCV Ͻ82 fl) and macrocytic (MCV 572 Tang et al. JACC Vol. 51, No. 5, 2008 Anemia in Ambulatory Heart Failure Patients February 5, 2008:569–76

anemia, Ϫ2.59 Ϯ 1.4 g/dl for those with incident anemia, and Ϫ0.04 Ϯ 1.4 g/dl for those with persistent anemia. All cohorts, except the persistent anemia cohort (p ϭ 0.77), had significant changes in Hb at 6-month follow-up from baseline (p Ͻ 0.0001 for the other 3 subgroups). Changes in clinical and echocardiographic variables identified from Table 1 were not correlated with changes in Hb (r Ͻ 0.10 for all). However, patients with LVEF Ն30% (median LVEF), BNP Ͼ325 pg/ml (median plasma levels of BNP), eGFR Ͻ60 ml/min/1.73 m2, or diabetes mellitus were more likely to have persistent anemia rather than no anemia or transient anemia (Fig. 3). Patients with eGFR Ͻ60 ml/min/1.73 m2 also had a higher rate of incident anemia. Long-term prognosis. During a mean follow-up of 3.9 Ϯ Figure 1 Prevalence and Resolution of Anemia in the Study Population 2.2 years, 1,827 patients (30%) died. Incremental changes in Hb levels were independently attributed to higher This is a flow diagram describing subject selection according to anemia status likelihood of survival after adjusting for LVEF and eGFR in the total population (n ϭ 6,159) as well as in the subgroup with follow-up hemoglobin levels (n ϭ 1,393). (Table 3). At baseline, patients presenting with anemia had a significantly worse prognosis than those without anemia, with an estimated 18.6% increased absolute risk Ͼ98 fl) anemia were observed in 19.6% and 8.8% of of death at 3 years (overall mortality 47% vs. 26%, p Ͻ patients, respectively, with more microcytic anemia in 0.0001) (Fig. 4). In the subset of patients with follow-up women (24.9% vs. 17.0%, p ϭ 0.0001), whereas macrocytic Hb levels, patients with persistent anemia still had the anemia had similar prevalence between genders (6.9% vs. worst survival relative to those with new-incident, resolved, 9.7%, p ϭ 0.082). Using ICD-9 codes and standard or no anemia at short-term follow-up (chi-square ϭ 81.2) laboratory workup orders for anemia, only 3.0% of patients (Fig. 4). At 6 months, patients with persistent anemia or with heart failure had documented anemia evaluation. Of incident anemia had higher all-cause mortality risk than interest, anemia was detected and evaluated more often in patients without anemia at baseline (all-cause mortality 58% patients presenting to internal medicine clinics than cardi- vs. 45% vs. 31%, respectively) (Table 4). The differenti- ology clinics (18.1% vs. 16.1% from laboratory data, p Ͻ ation in survival between these 4 different cohorts became 0.0001; 3.6% vs. 2.3% from documentation of diagnosis in more significant at 6 months (Fig. 4) (chi-square ϭ 81.2) problem list, p ϭ 0.0002), although patients seen in internal than at 3 months, further showing the greater differen- medicine clinics were older, were more often women, and tiation in survival over time (Table 4). were more likely to have preserved LV function. As expected, patients with anemia were more likely to be treated with iron supplementation (21% vs. 7%, p Ͻ 0.0001) or erythropoiesis-stimulating agents (8% vs. 1%, p Ͻ 0.0001). Careful chart review showed that those on anemia treatment in the nonanemia group had a prior diagnosis of anemia or renal insufficiency in their medical records (data not shown). However, nearly 70% of patients who fulfilled the criteria for anemia did not have a documented follow-up complete blood count in the electronic medical record within the 6 months of follow-up after the initial low Hb level. Changes in anemia status and Hb levels over time. Follow-up patients had consecutive Hb data at 3- and 6-month follow-up (n ϭ 1,393). In the 6-month follow-up cohort, 210 patients (15.1% of cohort, or 19.6% of patients with no anemia at baseline) developed new-onset anemia, 143 patients (10.3% of cohort, or 43.0% of patients with anemia at baseline) had resolution of their Hb levels, and 180 patients Figure 2 Prevalence of Baseline (12.9% of cohort, or 55.7% of patients with anemia at Anemia Stratified by Age and Gender baseline) had persistent anemia (Fig. 1). Mean changes in Bar graph illustrating the prevalence of anemia according Hb levels over 6 months were Ϫ0.32 Ϯ 1.4 g/dl for those to age cohorts and differences in rates between men and women. without anemia, ϩ2.44 Ϯ 1.7 g/dl for those with resolved JACC Vol. 51, No. 5, 2008 Tang et al. 573 February 5, 2008:569–76 Anemia in Ambulatory Heart Failure Patients

Figure 3 Anemia Characteristics According to Patient Subgroups

Bar graphs illustrating rates of persistent, transient, incident, or no anemia according to subgroups above versus below median values, including (A) baseline estimated glomerular ﬁltration rates (eGFR) (median 60 ml/min/1.73 m2); (B) baseline left ventricular ejection fraction (LVEF) (median 30%); (C) baseline B-type natriuretic peptide levels (BNP) (median 325 pg/ml); and (D) baseline diabetes status.

Discussion in patients with mild-to-moderate heart failure confers a higher risk for mortality compared with patients with The key finding in our single-center heart failure cohort is resolved, new-onset, or no anemia. that anemia in patients with chronic heart failure has been Although many heart failure studies have shown that underrecognized and underevaluated in clinical practice. anemia (often determined at a single time point) was Meanwhile, careful follow-up in patients with sequential associated with worse prognosis (3,13–16) , few have ex- evaluation of their Hb levels showed that a relatively large plored the impact of longitudinal changes in anemia status cohort (up to 43% of patients with anemia detected in a as well as the consequences of these changes on clinical single time point at baseline) may proceed with resolution of outcomes. Our findings complement prior observations their Hb to their nonanemic levels over time. Particularly from post hoc analysis of the Val-HeFT (Valsartan in Heart reassuring was the fact that these patients with so-called transient anemia may show equivalent long-term outcomes Multivariatefor Independent Risk Prediction Ratios of All-Cause Mortality Multivariate Risk Ratios when compared with those without any anemia. Our data Table 3 for Independent Prediction of All-Cause Mortality also confirmed prior studies showing that persistent anemia MultivariateRegression for Nominal Independent Logistic Prediction of Anemia Multivariate Risk Ratio Baseline Variable Increment (95% CI) p Value Multivariate Nominal Logistic Table 2 Hemoglobin (g/dl) ϩ1 0.85 (0.77–0.94) 0.0016 Regression for Independent Prediction of Anemia Age (yrs) ϩ10 1.29 (1.11–1.50) 0.0005 Baseline Variable Odds Ratio (95% CI) p Value Gender (male/female) ϩ1 (male) 1.86 (1.25–2.83) 0.0019 BNP Ͼ325 pg/ml 4.38 (2.67–7.41) Ͻ0.0001 Diabetes (yes/no) ϩ1 (yes) 1.36 (0.95–1.93) 0.090 eGFR Ͻ60 ml/min/1.73 m2 3.04 (1.97–4.72) Ͻ0.0001 eGFR (ml/min/1.73 m2) ϩ10 0.92 (0.86–0.98) 0.011 Diabetes mellitus 2.14 (1.39–3.28) 0.0005 LVEF (%-U) ϩ10 0.93 (0.82–1.05) 0.22 LVEF Ͻ30% 0.81 (0.70–0.92) 0.0015 LA area (cm2) ϩ10 1.34 (1.13–1.55) 0.0014 Mean corpuscular volume Յ90 fl 1.85 (1.22–2.78) 0.004 BNP (pg/ml) ϩ100 1.03 (1.01–1.05) 0.0014 Male 1.62 (1.02–2.63) 0.047 Systolic blood pressure ϩ10 0.90 (0.83–0.98) 0.011 Age Ͼ65 yrs 1.20 (0.78–1.86) 0.41 (mm Hg) ϩ Hypertension 0.83 (0.55–1.26) 0.39 Heart rate (beats/min) 10 1.07 (0.94–1.20) 0.30 ϩ LA area Ͼ25 cm2 1.03 (0.66–1.59) 0.91 Sodium (mEq/l) 10 0.88 (0.56–1.42) 0.59 Total cholesterol (mg/dl) ϩ10 0.96 (0.92–1.01) 0.095 BNP ϭ B-type natriuretic peptide; CI ϭ confidence interval; eGFR ϭ estimated glomerular filtration rate; LA ϭ left atrial; LVEF ϭ left ventricular ejection fraction. Abbreviations as in Table 2. 574 Tang et al. JACC Vol. 51, No. 5, 2008 Anemia in Ambulatory Heart Failure Patients February 5, 2008:569–76

dilutional anemia) and neurohormonal antagonism. Indeed, a recent publication has also challenged the adverse contri- bution of a diagnosis of anemia at the time of new-onset diagnosis of heart failure to long-term clinical outcomes because the prognostic value of anemia became less apparent when adjusted for other known prognostic variables (age, diastolic blood pressure, serum creatinine, New York Heart Association functional class, and left ventricular function) (18). We were unable to identify reliable predictors of transient versus persistent anemia, therefore careful moni- toring and evaluation of sequential Hb levels over time remains the recommended approach. In our overall study cohort of 6,159 patients, the 17% to 23% prevalence of anemia is consistent with that reported in the literature (4). The majority of patients with anemia (70%) showed normocytic indices, which may influence the perceived need for a more aggressive workup. Up to one-fifth of our anemia patients had microcytic anemia, and many patients may have specific reasons for anemia (such as iron deficiency anemia, thalassemia, renal insufficiency, or occult gastrointestinal bleeding) even with normal erythrocyte volumes. The most likely explanation is still the fact that Hb levels have not been commonly recognized as a relevant measurement in the management of heart failure until recently. This is particularly apparent in cardiology clinics, where the primary focus of managing heart conditions has distracted from evaluating or even establishing a diagnosis of anemia. The reported incidence of new-onset anemia (12% to Figure 4 Kaplan-Meier Analysis of All-Cause Mortality According to Anemia Status 15% within 6 months) in patients with heart failure

Stratified according to (A) presence or absence of anemia at deserved further discussion and investigation. Excessive baseline and to (B) persistent, transient, incident, or no anemia at follow-up. ACE inhibition may suppress the degradation of inhibitors of hematopoiesis, resulting in iatrogenic anemia (19). The majority of our patient population had already Failure Trial) and COMET (Carvedilol or Metoprolol been receiving ACE inhibitors for some duration, there- Evaluation Trial), showing patients with the largest de- fore the impact of an ACE inhibitor-induced reduction crease in Hb levels fared the worse in hospitalization, in Hb levels may not be as relevant (17,20). Nevertheless, morbidity, and mortality (17). The fact that a substantial these data were comparable with the 14.1% incidence of proportion of patients had transient anemia (despite the fact anemia at year 1 from COMET using less stringent that only 21% were on iron supplementation and 8% were anemia criteria (Hb Ͻ13 g/dl for men and Ͻ12 g/dl for on erythropoiesis stimulating agents) may imply the hetero- women) (21). Regardless, the number of patients who geneity of the patient population when anemia was deter- may still have persistent anemia or may have developed mined only at a single time point. Furthermore, our data new-onset anemia within 3 to 6 months of initial showed that patients with no or transient anemia have evaluation (approximately 30%) implies that sequential survival rates significantly better than those of patients with evaluation of complete blood count may be clinically persistent anemia. This finding may imply that a single- useful. As expected, anemia is also more common in time-point evaluation of Hb may not reflect the true patients with diabetes mellitus, who may also have more prognostic value of low Hb levels because resolution of cardiac and renal compromise. Patients with higher anemia may occur especially when the development of plasma BNP levels also were expected to have low Hb anemia at that time point was influenced by other concom- levels (22,23). Better understanding of the factors leading itant and potentially reversible conditions. Because only a to the development of anemia in heart failure will likely minority of patients received treatment for anemia, it is provide potential insights into disease mechanisms and conceivable that anemia in a substantial number of patients potential therapeutic targets. In particular, whether spe- may have resolved after standard heart failure management, cific therapy targeting optimal Hb levels (such as the use particularly with better control of fluid status (leading to less of erythropoiesis-stimulating agents) can benefit long- JACC Vol. 51, No. 5, 2008 Tang et al. 575 February 5, 2008:569–76 Anemia in Ambulatory Heart Failure Patients

Mortality Risk Ratios by Anemia Status at Baseline and Follow-Up Table 4 Mortality Risk Ratios by Anemia Status at Baseline and Follow-Up

Subgroup 1 Subgroup 2 Short-Term Follow-Up Long-Term Follow-Up Comparison Comparison Risk Ratio (95% CI) Chi-Square Risk Ratio (95% CI) Chi-Square Persistent anemia No anemia at baseline and follow-up 1.68 (1.37–2.06) 23.4* 2.51 (1.99–3.14) 54.4* Persistent anemia No anemia at baseline 1.61 (1.32–1.96) 20.8* 2.19 (1.75–2.71) 41.8* Persistent anemia Resolved anemia 1.73 (1.03–3.14) 4.4† 2.43 (1.74–3.48) 27.6* Persistent anemia Incident anemia 1.30 (0.98–1.75) 3.2‡ 1.33 (1.01–1.76) 4.0† Incident anemia No anemia at baseline and follow-up 1.29 (0.99–1.68) 3.5‡ 1.89 (1.48–2.38) 25.0* Incident anemia Resolved anemia 1.35 (0.78–2.49) 1.1‡ 1.82 (1.29–2.61) 11.8§ Resolved anemia No anemia at baseline and follow-up 0.98 (0.54–1.61) 0.1‡ 0.98 (0.73–1.36) 0.1‡ Resolved anemia No anemia at baseline 0.93 (0.52–1.53) 0.1‡ 0.90 (0.65–1.20) 0.5‡

*p Ͻ 0.0001. †p Ͻ 0.05. ‡p Ͼ 0.05. §p Ͻ 0.001. CI ϭ confidence interval. term survival and affect short-term clinical improvement follow-up durations may have limited the sample size of (24–26) remains to be proven in large multicenter ran- the follow-up subgroups, and can introduce selection bias domized controlled trials, particularly with recent studies in favor of those for whom follow-up evaluation were in the chronic kidney disease population that have failed deemed necessary (which may have explained the higher to show significant benefits with the use of erythropoietin prevalence of anemia at follow-up). The practice patterns therapy to achieve higher Hb targets (27,28). may underestimate the true rates of evaluation because The strength of our data from electronic medical patients may seek care outside our hospital system, records lies in our ability to easily track our patient cohort although the majority of the patients in this analysis have over time in real clinical practices. This study shows the received long-term follow-up in internal medicine and power of using electronic medical records as a unique and cardiology clinics. It is also difficult to adjust for varia- novel way to further our understanding of practice tions in treatment effects based on sequential Hb levels patterns as well as disease characteristics and outcomes. because the small subgroup of patients undergoing iron or Detailed information such as anemia subtypes (such as erythropoietin treatment were initiated on therapy only microcytic vs. macrocytic), concomitant conditions (such after the follow-up time interval, and a definitive diag- as renal insufficiency), and diagnostic workup patterns (or nosis of iron deficiency was not available in a large the lack of follow-up evaluation) can be determined number of individuals with anemia at baseline even with directly from the clinical laboratory data of the electronic medical records. The striking contrast between the doc- microcytic anemia. Anemia with normal versus low iron umented diagnosis of anemia and the documented Hb may have different relationships to concurrent hemody- levels can illustrate some advantages of this research namic measures. Regression of the mean may have strategy over administrative database interrogation. In contributed to our observations, particularly with the Ͻ turn, future quality assurance measures can be imple- higher resolution rate in patients with eGFR 60 ml/ 2 mented directly into the electronic medical record system min/1.73 m , as the prevalence and incidence of anemia to improve diagnostic and coding accuracy via reminder were substantially higher in this subgroup. Further stud- systems or automatic coding algorithms. ies are needed to determine how to improve the diagnos- Our study is limited by its retrospective design, and the tic workup of anemia for this patient population and how mandate to have an available complete blood count to identify those who may benefit from specific therapies assessment (at baseline and especially at pre-defined targeting anemia associated with heart failure. intervals) may not represent the true prevalence of an unselected heart failure population. Furthermore, a single cutoff value (even with gender-specific definitions) may Conclusions limit the reliability of the estimation of the prevalence and incidence rates by values that were borderline, even In our single-center cohort of ambulatory patients with a when an Ն0.5 g/dl increase in Hb levels was instituted to clinical diagnosis of heart failure with available complete increase the reliability of assessing the changes in the blood count assessment, anemia was underrecognized and follow-up Hb levels. The lack of consistent objective underevaluated, especially in cardiology clinics. Resolution measurements of fluid overload as well as physician- of anemia was evident in up to 43% of patients who assessed or self-reported functional status (such as New presented initially with anemia, and did not pose greater York Heart Association functional classification) at spe- long-term mortality risks. However, the presence of persis- cific time intervals may have limited the accuracy of tent anemia conferred poorer survival in patients with heart determining concomitant changes in clinical status with failure when compared with that of incident, resolved, or no changes in Hb over time. The arbitrary criteria set for the anemia. 576 Tang et al. JACC Vol. 51, No. 5, 2008 Anemia in Ambulatory Heart Failure Patients February 5, 2008:569–76

13. Go AS, Yang J, Ackerson LM, et al. Hemoglobin level, chronic kidney Reprint requests and correspondence: Dr. W. H. Wilson Tang, disease, and the risks of death and hospitalization in adults with Department of Cardiovascular Medicine, Cleveland Clinic, 9500 chronic heart failure: the Anemia in Chronic Heart Failure: Outcomes Euclid Avenue, Desk F25, Cleveland, Ohio 44195. E-mail: tangw@ and Resource Utilization (ANCHOR) study. Circulation 2006;113: 2713–23. ccf.org. 14. Tang YD, Katz SD. Anemia in chronic heart failure: prevalence, etiology, clinical correlates, and treatment options. Circulation 2006; 113:2454–61. REFERENCES 15. O’Meara E, Murphy C, McMurray JJ. Anemia and heart failure. Curr Heart Fail Rep 2004;1:176–82. 1. Sandgren PE, Murray AM, Herzog CA, et al. Anemia and new-onset 16. Lindenfeld J. Prevalence of anemia and effects on mortality in patients congestive heart failure in the general Medicare population. J Card Fail with heart failure. Am Heart J 2005;149:391–401. 2005;11:99–105. 17. Anand IS, Kuskowski MA, Rector TS, et al. Anemia and change in 2. Silverberg DS, Wexler D, Iaina A. The importance of anemia and its hemoglobin over time related to mortality and morbidity in patients correction in the management of severe congestive heart failure. Eur with chronic heart failure: results from Val-HeFT. Circulation 2005; J Heart Fail 2002;4:681–6. 112:1121–7. 3. Maggioni AP, Opasich C, Anand I, et al. Anemia in patients with 18. Kalra PR, Collier T, Cowie MR, et al. Haemoglobin concentration heart failure: prevalence and prognostic role in a controlled trial and in and prognosis in new cases of heart failure. Lancet 2003;362:211–2. clinical practice. J Card Fail 2005;11:91–8. 19. van der Meer P, Lipsic E, Westenbrink BD, et al. Levels of 4. Horwich TB, Fonarow GC, Hamilton MA, MacLellan WR, Boren- hematopoiesis inhibitor N-acetyl-seryl-aspartyl-lysyl-proline partially stein J. Anemia is associated with worse symptoms, greater impairment explain the occurrence of anemia in heart failure. Circulation 2005; in functional capacity and a significant increase in mortality in patients 112:1743–7. with advanced heart failure. J Am Coll Cardiol 2002;39:1780–6. 20. Ishani A, Weinhandl E, Zhao Z, et al. Angiotensin-converting 5. Ezekowitz JA, McAlister FA, Armstrong PW. Anemia is common in enzyme inhibitor as a risk factor for the development of anemia, and heart failure and is associated with poor outcomes: insights from a the impact of incident anemia on mortality in patients with left cohort of 12,065 patients with new-onset heart failure. Circulation ventricular dysfunction. J Am Coll Cardiol 2005;45:391–9. 2003;107:223–5. 21. Komajda M, Anker SD, Charlesworth A, et al. The impact of new 6. Mozaffarian D, Nye R, Levy WC. Anemia predicts mortality in severe onset anaemia on morbidity and mortality in chronic heart failure: heart failure. The Prospective Randomized Amlodipine Survival Eval- results from COMET. Eur Heart J 2006;27:1440–6. uation (PRAISE). J Am Coll Cardiol 2003;41:1933–9. 22. Wold Knudsen C, Vik-Mo H, Omland T. Blood haemoglobin is an 7. Kosiborod M, Smith GL, Radford MJ, Foody JM, Krumholz HM. independent predictor of B-type natriuretic peptide (BNP). Clin Sci The prognostic importance of anemia in patients with heart failure. (Lond) 2005;109:69–74. Am J Med 2003;114:112–9. 23. Tsuji H, Nishino N, Kimura Y, et al. Haemoglobin level influences 8. McCarthy PM, Young JB. Heart Failure: A Combined Medical and plasma brain natriuretic peptide concentration. Acta Cardiol 2004;59: Surgical Approach. New York, NY: Blackwell, 2007. 527–31. 9. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more 24. Silverberg DS, Wexler D, Sheps D, et al. The effect of correction of accurate method to estimate glomerular filtration rate from serum mild anemia in severe, resistant congestive heart failure using subcu- creatinine: a new prediction equation. Modification of Diet in Renal taneous erythropoietin and intravenous iron: a randomized controlled Disease Study Group. Ann Intern Med 1999;130:461–70. study. J Am Coll Cardiol 2001;37:1775–80. 10. Lang RM, Bierig M, Devereux RB, et al. Recommendations for 25. Silverberg DS, Wexler D, Blum M, et al. Aggressive therapy of chamber quantification: a report from the American Society of congestive heart failure and associated chronic renal failure with Echocardiography’s Guidelines and Standards Committee and the medications and correction of anemia stops or slows the progression of Chamber Quantification Writing Group, developed in conjunction both diseases. Perit Dial Int 2001;21:S236–40. with the European Association of Echocardiography, a branch of the 26. Mancini DM, Katz SD, Lang CC, LaManca J, Hudaihed A, Androne European Society of Cardiology. J Am Soc Echocardiogr 2005;18: AS. Effect of erythropoietin on exercise capacity in patients with 1440–63. moderate to severe chronic heart failure. Circulation 2003;107:294–9. 11. Garcia MJ, Thomas JD, Klein AL. New Doppler echocardiographic 27. Drueke TB, Locatelli F, Clyne N, et al. Normalization of hemoglobin applications for the study of diastolic function. J Am Coll Cardiol level in patients with chronic kidney disease and anemia. N Engl 1998;32:865–75. J Med 2006;355:2071–84. 12. Khouri SJ, Maly GT, Suh DD, Walsh TE. A practical approach to the 28. Singh AK, Szczech L, Tang KL, et al. Correction of anemia with echocardiographic evaluation of diastolic function. J Am Soc Echo- epoetin alfa in chronic kidney disease. N Engl J Med 2006;355: cardiogr 2004;17:290–7. 2085–98.