Acknowledgment

This continuing medical education (CME) activity is supported by an educational grant from Amgen , Thousand Oaks, California.

Disclosures

Dr. Friedewald has received honoraria from and is a speaker for Novartis, East Hanover, New Jersey. Dr. Emmett has no relevant relationships to disclose. Dr. McCullough has no relevant relationships to disclose. Dr. Yancy has received honoraria for research, consulting, speaking from GlaxoSmithKline, Research Triangle, North Carolina; Medtronic, Minneapolis, Minnesota; and Scios, Sunnyvale, California. Dr. Yancy has received honoraria for consulting from Novartis. Dr. Roberts has received honoraria from and is a speaker for AstraZeneca, Wilmington, Delaware; Merck, Whitehouse Station, New Jersey; Novartis; and Schering-Plough, Kenilworth, New Jersey.

Disclosures

Dr. Friedewald has received honoraria from and is a speaker for Novartis, East Hanover, New Jersey. Dr. Emmett has no relevant relationships to disclose. Dr. McCullough has no relevant relationships to disclose. Dr. Yancy has received honoraria for research, consulting, speaking from GlaxoSmithKline, Research Triangle, North Carolina; Medtronic, Minneapolis, Minnesota; and Scios, Sunnyvale, California. Dr. Yancy has received honoraria for consulting from Novartis. Dr. Roberts has received honoraria from and is a speaker for AstraZeneca, Wilmington, Delaware; Merck, Whitehouse Station, New Jersey; Novartis; and Schering-Plough, Kenilworth, New Jersey.

Objectives

Upon completion of this activity, the physician should be able to:

• 1
  

  Evaluate anemia in patients with cardiovascular (CV) disease.

  
  
• 2
  

  Assess the significance of anemia in patients with CV disease.

  
  
• 3
  

  Prescribe blood transfusion or erythropoietin therapy when appropriate in patients with CV disease.

Introduction

The risk for coronary artery disease (CAD) is associated with hemoglobin (Hgb) >17.0 and <9.0 g/dl in the general population. In patients with chronic heart failure (HF), anemia is common and is associated with a poor prognosis. With the aging population in the United States and globally, in whom anemia is more common, this association is increasing in importance in the management of cardiac patients. Despite the important relation between anemia and CV disease, many aspects of the cause, pathophysiology, and therapy, including when to treat and not to treat anemia, are unresolved.

Dr. Friedewald: How do you define anemia ?

Dr. McCullough: The National Kidney Foundation and World Health Organization define anemia as a hemoglobin (Hgb) <12 g/dl in men and postmenopausal women and <11 g/dl in premenopausal women.

Dr. Yancy: Other factors must be considered in the diagnosis of anemia. For example, age affects Hgb and hydration affects hematocrit (Hct).

Dr. Roberts: The Hct is decreased in almost everyone greater than age 90 years.

Dr. McCullough: With aging, there is a decrease in the testosterone level, which stimulates bone marrow red blood cell production.

Dr. Emmett: There also are age-related decreases in the glomerular filtration rate (GFR) and erythropoietin blood levels. Thus, although 90-year-old individuals often have Hgb levels meeting the criteria for anemia, they may not be abnormal for that age.

Dr. Roberts: Does Hgb correlate with longevity?

Dr. Yancy: We do not know. Blood hyperviscosity shortens longevity, so having a “super level” of Hgb may be disadvantageous. CV mortality increases with progression of renal disease, which is associated with anemia.

Dr. McCullough: The National Kidney Foundation has screened about 60,000 people for renal disease and has found that there are 3 independent factors for increased mortality: decreased renal function (measured by the GFR), kidney injury (measured by proteinuria), and anemia . The association of anemia with CV disease is also important. PubMed lists over 3,000 papers on this topic, most of them published in recent years. For example, in the RENAISSANCE (Randomized Etanercept North American Strategy to Study Antagonism of Cytokines) trial, which tested an endothelin receptor blocker in chronic heart failure (HF), serial Hgb levels and imaging studies correlated symptom improvement with increased Hgb and ejection fraction and decreased left ventricular (LV) mass index. Thus, anemia appears to be a barometer of decreased LV function.

Dr. Yancy: In HF, changes in the Hgb as small as 0.5 g/dl can profoundly change patient outcomes.

Dr. Emmett: Without prospective studies, however, it is not known whether changes in Hgb are merely markers or true causes of changes in cardiac function.

Dr. Roberts: How does obesity relate to anemia?

Dr. McCullough: Obesity drives the sleep apnea syndrome, in which chronic hypoxia causes the kidneys to increase erythropoietin production, stimulating bone marrow erythrocyte production. Thus, Hgb is often higher in obese patients, as well as in patients with advanced chronic obstructive pulmonary disease and other causes of hypoxia.

Dr. Friedewald: Does altitude affect the definition of anemia?

Dr. Yancy: Yes, because decreased arterial partial oxygen tension (pO ₂ ) is a major stimulus to erythropoietin production.

Dr. Friedewald: Is there a formula for correcting for altitude in the definition of anemia?

Dr. McCullough: No.

Dr. Emmett: The effect of altitude on Hgb is probably related to arterial oxygen tension in the kidney. Other determinants of erythropoietin production are perfusion extraction ratios and the state of arterial dilation or constriction.

Dr. McCullough: The most important factor to be excluded before attributing anemia to renal disease or other forms of chronic disease is gastrointestinal bleeding. As a general guide, iron saturation <20% or a ferritin level <100 mg/ml is a “red flag” that gastrointestinal bleeding could be the cause of anemia.

Dr. Yancy: What is the value in measuring the mean corpuscular volume (MCV)?

Dr. Emmett: The MCV is increased in deficiencies of vitamin B ₁₂ and folic acid in certain types of anemia, especially hemolytic anemia. Such forms of anemia should be excluded before diagnosing anemia due to chronic disease or advanced age.

Dr. Friedewald: What is the role of dietar y iron as a cause of anemia?

Dr. Yancy: Elemental dietary iron is difficult to absorb in any form, including animal and plant proteins, which is why menstruating and postpartum women have difficulty correcting iron deficiency. Contrary to popular belief, animal protein is not a better source of iron.

Dr. Friedewald: Is anemia a risk factor for CV disease?

Dr. Yancy: The phrase “risk factor” refers to an entity that makes a specific disease process more likely to occur. Anemia adds to the risk of the complications of atherosclerosis (i.e., stroke, myocardial infarction) in patients with major risk factors such as hypertension and dyslipidemia. Because anemia is so intertwined with other disease processes, however, its exact cause and effect is unclear. Thus, while anemia is clearly a risk marker , because it tracks disease, I do not know whether it is a risk factor that promotes CV disease in the absence of other facilitating conditions.

Dr. McCullough: I prefer the term risk state , which differs from risk factor or even risk marker in that it implies that there is inherent clustering with other disease states. Anemia clusters with other disease states such as chronic kidney disease, HF, and diabetes mellitus, so thinking in terms of risk states forces us to be more open minded in understanding that there may be unknown confounders accompanying a patient’s anemia. Rather, biomarkers such as anemia might be considered epiphenomena , or measurable barometers of underlying processes that are not understood.

Dr. Emmett: We must be careful when using the phrase “anemia of chronic disease” because anemia of chronic renal failure is different from the anemia of chronic inflammation or chronic infection. Anemia of renal disease is much more erythropoietin dependent and therefore responds to erythropoietin therapy. Anemia secondary to inflammation and perhaps HF, however, are abnormalities of transfer of iron into Hgb and red cell creation and are more difficult to treat.

Dr. Roberts: What is the effect of body weight on the Hgb?

Dr. McCullough: Body fat composition does not appear to have much influence on the serum Hgb level, other than obesity-related sleep apnea and chronic hypoxia. Body water volume, however, does affect serum Hgb. There are studies of blood volume suggesting that in patients with HF, only a portion of the decrease in serum Hgb is a manifestation of hemodilution. For that reason, anemia does not entirely go away when ideal blood volumes are restored in patients with HF.

Dr. Friedewald: What is the relation between diabetes mellitus and anemia?

Dr. McCullough: The connection between diabetes mellitus and anemia involves several different mechanisms. One connection is with chronic kidney disease, which is common in patients with diabetes mellitus. Another mechanism is decreased bone marrow response to sex hormones, especially testosterone, in patients with diabetes mellitus.

Dr. Yancy: Another possible cause of anemia in diabetes mellitus is autonomic neuropathy, causing decreased signal generation for renin release, which in turn reduces the serum level of erythropoietin.

Dr. McCullough: The renin-angiotensin system also is probably a factor in anemia in diabetic patients. For example, patients who receive angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin-receptor blockers have slightly lower Hgb levels. My understanding of the mechanism is that drugs that suppress the renin-angiotensin system impair stimuli to renal release of erythropoietin.

Dr. Emmett: This observation was first made in patients after kidney transplantation. There is a population of kidney transplant patients who develop relatively severe polycythemia following transplant. Although the exact mechanism is unclear, it is believed that post-transplant polycythemia is related to pretransplant decreased perfusion of the native kidneys. It has been observed, however, that this form of polycythemia can be treated with ACEIs, causing dramatic reductions in the Hct through a reduction in erythropoietin.

Dr. Roberts: Do diuretics affect the Hgb level?

Dr. McCullough: I am not aware of any studies, although a large water volume constriction would theoretically raise the Hgb level.

Dr. Emmett: Patients on diuretics, however, have higher blood renin levels, and thus perhaps produce more erythropoietin or have increased responsiveness to erythropoietin.

Dr. Friedewald: Does anemia accelerate atherosclerosis?

Dr. McCullough: We do not know whether anemia accelerates atherosclerotic plaque development. Patients with anemia, however, do have earlier development of first myocardial infarction, first stroke, and CV death, when adjusted for all of the other baseline risk factors for CV disease.

Dr. Roberts: Does raising the Hgb level decrease the incidence of those events?

Dr. McCullough: We do not know.

Dr. Friedewald: How does chronic kidney disease cause anemia?

Dr. Emmett: The definition of chronic kidney disease is predicated on a decreased GFR. Using only GFR as the parameter, corrected for body surface area, stage 1 kidney disease is defined by a (normal) GFR >90 ml/min/1.73 M ² when accompanied by some other evidence of kidney disease, such as proteinuria or hypertension. Stage 2 kidney disease is defined by an actual reduction in the GFR to 60 to 90 ml/min/1.73 M ² .

Dr. Friedewald: Which abnormality is first detected in chronic kidney disease, proteinuria or reduced GFR?

Dr. Emmett: In many patients, such as persons with diabetes mellitus, proteinuria develops first. In other forms of kidney disease, however, a decreased GFR alone may be the first manifestation.

Dr. McCullough: In young persons, age 20 to 30 years, the kidneys have a GFR of 120 to 130 ml/min, which is usually estimated according to a multivariable equation using serum creatinine, weight, age, gender, and race. There follows a “normal” age-related decline in GFR, but it remains above 60 ml/min, generally falling to a level of about 70 ml/min by age 80 years. Thus, the National Kidney Foundation says that a patient with a GFR <60 ml/min, regardless of any other information about the patient, probably has chronic kidney disease. A GFR <60 ml/min is a surrogate for reduced renal functional parenchymal mass. This is important because in all epidemiologic studies, there is a break point somewhere below the GFR of 60 ml/min, in which increased rates of CV events and death occur.

Dr. Roberts: What is the role of serum creatinine alone, versus the GFR, for assessing renal function?

Dr. McCullough: The serum creatinine is a deceptive measurement because its level is based on total body muscle mass. Creatinine is derived from creatine, which is derived from muscle, so persons with small muscle mass cannot have a greatly elevated creatinine and therefore chronic kidney disease can go undiagnosed using that measurement alone. The standard calculation of GFR is only an estimate of the true GFR. The estimated GFR, however, has a bias of about 6 to 8 ml/min compared to the real biochemical measurement, with considerable laboratory variation above 60 ml/min.

Dr. Yancy: The premise of creatinine clearance is that it is an “ideal” substance to measure because it is neither filtered nor secreted. Serum creatinine is always in steady state with the skeletal muscle mass. However, cachectic individuals have significantly less creatine and therefore lower serum creatinine levels. In contrast, persons with large amounts of muscle mass such as body builders can have normal creatinine levels as high as 1.6 mg/ml.

Dr. McCullough: The oldest equation for GFR, the Cockcroft-Gault formula, had body weight in the denominator, which was not a significant problem in past years when so many more people had normal body weight. With our current obesity pandemic, however, the body weight has become an important consideration. The MDRD (Modification of Diet in Renal Disease) equation does not include body weight; rather, it uses body surface area, and standardizes to 1.73 m ² , which would be a “normal” body surface area in men or women.

Dr. Emmett: Many clinical laboratories use the MDRD equation, and major errors can occur if you do not take this correction for normal body surface area into account.

Dr. McCullough: Cystatin C is another blood test for renal function. This is a cysteine protease inhibitor produced by all nucleated cells in the body, freely filtered by the kidney, and 100% reabsorbed. Compared with serum creatinine, it is a better measure of renal function, but it is not widely available.

Dr. Yancy: Cystatin C has become an important metric in research. We are following it in several large, ongoing HF trials because there is so much less plasticity to it.

Dr. Emmett: If you are interested in the GFR, then measure the GFR. I do not agree with the practice of basing conclusions on indirect measurements of GFR. You would not accept a measure of cardiac output as some blood test that you can “kind of” correlate with cardiac output rather than measuring the true cardiac output.

Dr. Friedewald: What are the mechanisms of anemia in patients with CV disease?

Dr. McCullough: Although uncomplicated ischemic heart disease has no direct association with anemia, chronic HF has such an association. Among patients with heart failure, 1 large sample size of 130,000 Medicare patients found that 32% have anemia, and among those with anemia, only 26% have chronic kidney disease. Thus, HF alone is often associated with anemia.

Dr. Friedewald: Why?

Dr. McCullough: One theory is that HF is associated with elevation of multiple cytokines, which are inflammatory factors affecting the production of red blood cells in the bone marrow by making the bone marrow resistant to erythropoietin levels.

Dr. Yancy: Oxidative stress also may play a role. Oxidative stress is tightly associated with HF and is perhaps more important in some persons than others, independent of renal disease.

Dr. Friedewald: Please define oxidative stress .

Dr. Yancy: Oxidative stress is a biochemical variable within the CV milieu related to disproportionate elevations of factors like nicotine adenine dinucleotide and nicotinamide adenine dinucleotide phosphate. These variables in the vascular system are equivalent to rust in pipes. Oxidative stress in the bone marrow and the kidneys may be a link between anemia and heart failure.

Dr. Roberts: You are saying that abnormalities of certain measurable substances fall under the umbrella of “oxidative stress.”

Dr. Yancy: That is partly correct, because HF is much more than a mechanical problem. It is also a biochemical problem that involves neurohormonal activation and its classic cascades including inflammatory mediators, development of tumor necrosis–α and primitive cytokines, and the reactive oxygen species. Thus, there is a whole environment of risk-promoting circumstances in HF.

Dr. Roberts: Without the cardiac mechanical abnormalities of HF, however, all these other factors would be absent.

Dr. Yancy: That is correct. But all of these other factors do not necessarily track with the severity of left ventricular dysfunction. HF is an odd state in which neurohormonal substances can be disproportionately increased compared to the amount of LV dysfunction. Conversely, there can be profound LV dysfunction without much increase in these substances.

Dr. Roberts: Does HF severity correlate with the degree of anemia?

Dr. McCullough: There is a weak relation. As HF spontaneously improves, Hgb levels rise, and as HF spontaneously worsens, Hgb levels fall. This suggests the presence of neurohormonal oxidative stress and inflammatory changes with possible mediators that comprise triangular signals among cardiac function, the bone marrow, and the kidneys.

Dr. Friedewald: Does the degree of anemia in HF correlate with renal function?

Dr. McCullough: Yes, there is a correlation with renal function, but many patients with anemia and heart failure have normal renal function.

Dr. Yancy: We do not know precisely why the HF patient gets anemia. We only know that it occurs, and that it occurs frequently.

Dr. Friedewald: Let’s discuss diabetes mellitus and anemia.

Dr. Emmett: The most important determinant of anemia of both type 1 and type 2 diabetes mellitus is kidney disease. The anemia is due to decreased renal erythropoietin (EPO) release.

Dr. Friedewald: Are there abnormal forms of EPO?

Dr. Emmett: EPO is synthesized as a protein that has to be cleaved to function appropriately. There may be clinical disturbances due to abnormal postsynthetic alterations in the structure of EPO, but this is unproven.

Dr. Roberts: Does exogenous insulin have an effect on Hgb?

Dr. Emmett: I do not know.

Dr. Friedewald: Does anemia improve with control of diabetes mellitus?

Dr. Emmett: Renal dysfunction can be stabilized in the diabetic patient who is aggressively managed, whether it is with weight reduction, tighter glucose control, or blood pressure control. When those treatment goals are achieved, the Hgb is stabilized.

Dr. McCullough: Microalbuminuria improves with weight loss in patients with diabetes mellitus. The explanation is that obesity is a proinflammatory state, and circulating inflammatory factors probably contribute to ongoing glomerular injury at the level of the endothelium, thereby causing microalbuminuria.

Dr. Yancy: There also is a theory that hyperfiltration is the initial injury to the kidney in diabetes mellitus. This is associated with an exaggerated GFR and mesangial hypertrophy. Once established, it is difficult to cause regression of these abnormalities, although renal function can be stabilized by treatment with drugs such as ACEIs.

Dr. Roberts: How do you know?

Dr. Emmett: Early type 1 diabetes mellitus is associated with a “supra-normal elevated” GFR and enlarged kidneys. Elevated GFR in diabetes mellitus is associated with worse outcomes. The presumed mechanism is that elevated GFR is related to a hyperfiltration state. The glomeruli are “pounded” by high pressure, and over time, this causes glomerular scarring. With aggressive early treatment of type 1 diabetes mellitus, the GFR falls, which we believe improves outcomes. This is analogous to treating a diabetic patient with an ACE inhibitor that reduces intraglomerular pressures and the GFR. This stabilizes or perhaps even regresses structural damage.

Dr. Yancy: There is a fascinating parallel between glomerular mesangial hypertrophy and LV hypertrophy. They both may respond to the same growth stimuli and to the same treatment interventions.

Dr. McCullough: Renal fibrosis and cardiac fibrosis are probably 2 tightly linked processes, and aldosterone is most often regarded as the common primary driver.

Dr. Yancy: There could be a common injury profile for the renal tubule cell and cardiac myocyte.

Dr. Roberts: Does anemia portend a worse prognosis in patients with diabetes mellitus?

Dr. McCullough: Yes, anemia is associated with a worse prognosis.

Dr. Roberts: Is anemia also associated with a worse prognosis in patients with HF?

Dr. Yancy: Yes, but what this means is uncertain because we have only observational data and no clinical trials designed to test outcomes related to anemia in patients with HF.

Dr. McCullough: Anemia appears to be a risk multiplier or a risk amplifier in patients with HF.

Dr. Friedewald: Let’s examine the effect of anemia on specific cardiac conditions, starting with chronic angina pectoris.

Dr. McCullough: Blood Hgb level that is sufficiently low to reduce oxygen delivery to the myocardium, in the presence of normal LV function, can contribute to a reduced threshold for angina pectoris. The American Society of Anesthesiologists, which has basic guidelines and a position paper on this topic, states that in people with normal LV function, a hemoglobin <6 g/dl causes impaired tissue perfusion. Above that level, the anginal determinant of tissue perfusion is the cardiac output, not the oxygen-carrying capacity. Thus, it is important that we not attribute symptoms of angina pectoris to anemia unless the Hgb is profoundly low.

Dr. Roberts: Transfusion is indicated when the Hgb is <6 g/dl?

Dr. McCullough: That is what the American Society of Anesthesiologists recommends. At that level, tissue perfusion throughout the body can be impaired. A restrictive approach is recommended, however, because Hgb raised by transfusion into the 7 to 9 g/dl range carries better survival than transfusion to higher Hgb levels.

Dr. Friedewald: What about treatment of anemia in patients with acute coronary syndrome?

Dr. McCullough: In acute coronary syndrome the same concept of restrictive transfusion is advocated. This applies not only to patients with acute coronary syndrome, but to all patients who are ill in the hospital, who have worse survival and other adverse CV outcomes with transfusion.

Dr. Friedewald: Including patients undergoing coronary interventions?

Dr. McCullough: Yes, and the message is clear: transfusion in the setting of acute coronary artery disease worsens the prognosis . Another factor is that transfusion of red blood cells more than 2 weeks old causes worse outcomes than transfusion of newer red blood cells. As red blood cells age, they may undergo oxidative stress or other changes harmful to transfusion recipients.

Dr. Friedewald: Your recommendation would be to use blood as fresh as possible?

Dr. McCullough: I do not have a recommendation. It is just an anecdotal, retrospective finding right now. The main point is that transfusion in general has adverse effects on patients with ischemic heart disease. And transfusion that increases the Hgb from <6 to 9 mg/dl carries less risk than increasing it to >10 mg/dl.

Dr. Friedewald: What is the role of blood transfusion in patients with anemia who are undergoing elective coronary intervention?

Dr. McCullough: Raising the hemoglobin does not improve important CV outcomes in terms of readmission rates or death. Thus, there is no guideline stating that erythropoietin should be given to raise the Hgb in patients having elective coronary intervention.

Dr. Friedewald: What about treating anemia after cardiac surgery?

Dr. Yancy: There is no set level of Hgb for when anemia should be treated. Rather, the indication for treating anemia is evidence that the anemia is retarding clinical recovery. There is an increased risk of thrombosis when increasing the Hgb in patients who are already prothrombotic. Increased platelet and blood viscosity and increased platelet adherence are thrombotic risks inherent to administering blood products.

Dr. McCullough: There is a general trend, in both coronary artery surgery and surgery in general, to avoid transfusion except in instances of impaired oxygen delivery due to anemia, such as cerebral or myocardial ischemia.

Dr. Roberts: Do you discharge the anemic patient on iron after coronary artery bypass surgery?

Dr. McCullough: Yes, if the iron saturation is <20% and the ferritin levels are <100 ng/ml caused by blood loss from surgery.

Dr. Roberts: What is the CV effect of very low Hgb in patients with sickle-cell anemia?

Dr. Yancy: Patients with sickle-cell disease who survive to adulthood typically have substantial LV hypertrophy and small vessel disease.

Dr. Emmett: And renal disease.

Dr. Yancy: We have not discussed the anemia of end-stage renal disease associated with cardiomyopathy, which is ventricular hypertrophy with concomitant myocardial fibrosis. This is a ventricular diastolic abnormality in which there is a tight relationship between changes in ventricular mass and end-stage renal disease.

Dr. Friedewald: Is the LV hypertrophy in this type of cardiomyopathy concentric?

Dr. Yancy: Yes.

Dr. Emmett: And it can regress when the anemia is corrected.

Dr. McCullough: Unless the LV mass index is not sufficiently increased, increasing the Hgb has little effect. When the LV mass index is significantly increased, however, the LV wall becomes thinner with treatment of anemia.

Dr. Friedewald: What is the effect of anemia in the patient with acute decompensated HF?

Dr. Yancy: In the setting of acute HF, anemia is an independent factor for increased risk of in-hospital mortality and also is associated with increased length of stay.

Dr. Roberts: Should those patients be transfused?

Dr. Yancy: We do not know; probably not.

Dr. Friedewald: How should iron deficiency anemia in cardiac patients be treated?

Dr. Yancy: The first step is to diagnose the cause of the anemia. Too often, because supplemental therapy is so easy, anemia treatment is prescribed without first trying to determine the etiology.

Dr. McCullough: Normal upper and lower endoscopies, negative stool tests for blood, and absence of gastrointestinal (GI) bleeding history is sufficient evaluation. Some would say that you also have to look for small bowel sources of blood loss. For treatment, ferrous sulfate, which is cheap and does not require a prescription, is 1 way to start. However, it is not well absorbed by the GI tract and does not contain a very high level of elemental iron. A prescription form of iron that I commonly use is ferrous fumarate. It is important to remember that an acidic environment helps iron absorption, so a common strategy is to use it in combination with ascorbic acid.

Dr. Roberts: What about people who donate blood on a regular basis?

Dr. Emmett: If they do not ingest an adequate amount of iron in their diet they become iron deficient.

Dr. McCullough: A rule we follow in our clinic is that if the iron saturation is >20% and the ferritin level is >100 ng/ml, the patient is not iron deficient, because there are adequate amounts of iron that the bone marrow can draw upon. Iron saturations of 10% to 15% can make patients feel fatigued, without a decreased hemoglobin level. Somehow, iron itself is linked to a general feeling of how patients feel.

Dr. Yancy: In any treatment strategy for HF there is substantial disconnect between how patients subjectively feel versus measurable outcomes. When we give agents to selectively increase the Hgb, patients have a better perception of their quality of life and even have measurable increases in their functional capacity. But we do not know whether that translates to reduced hospitalization or mortality.

Dr. Friedewald: If they feel better in the interim, is it not a reason for treating?

Dr. Yancy: Our experience in patients with HF challenges that premise. When you look at surrogate markers of the responsiveness to an agent and those surrogate markers are subjective or they are functional, there may be reasons independent of what causes disease to progress that explain why those surrogate markers improve; surrogates are only helpful if they are linked with improved outcomes.

Dr. McCullough: Nine of 10 persons who are well say, “I want to feel better and at the expense of a shorter life.” For patients diagnosed with a terminal condition, however, that figure reverses. This has been called “time trade-off,” and is the reason why cancer patients agree to chemotherapy.

Dr. Yancy: In HF, however, the time trade-off is the opposite, so that patients who have been suffering with HF will surrender incremental increases in longevity for even modest improvements in their quality of life. Their feeling of wellness scale is worse than patients with advanced cancer.

Dr. Roberts: That is because breathlessness is terribly frightening.

Dr. Friedewald: What should cardiologists know about erythrocyte-stimulating proteins?

Dr. McCullough: The kidney produces EPO. It stimulates EPO receptors, which are on hemangioblasts in the bone marrow. EPO receptors are not only on hemangioblasts but they are on some of the primordial cells in the bone marrow. Thus, EPO itself not only stimulates hemangioblasts to produce mature red blood cells but also produces a greater number of endothelial-drive progenitor cells and other forms of stem cells that circulate into the systemic vasculature. EPO has been sequenced and cloned, and it can be made by recombinant techniques by E. coli and manufactured as an injectable product, either intravenously or subcutaneously. It is given once every week or once every couple of weeks in a bolus to boost the production of red blood cells. Measurement of the EPO blood level, however, is not useful.

Dr. Emmett: The best indication for EPO is in a person who is EPO deficient, either absolutely or relative to their level of Hct.

Dr. Yancy: Erythrocyte-stimulating proteins are only indicated in the setting of renal disease and anemia and there are no independent CV conditions where chronic administration of EPO is indicated.

Dr. Yancy: If we go back to HF, there are some concerted efforts under way to see whether there is a select benefit in using EPO analogues. If there is, then we have to address whether there is value in using them in addition to already extensive medical treatment strategies, including device therapy. Also, its cost effectiveness has to be considered. Thus, we cannot restrict our discussion to the relative benefits of improving the Hgb in response to EPO analogues, but we have to add a cost model into this equation and put it into the context of risks and benefits, especially if the benefits are not directly measurable as an improvement in mortality and a reduction in hospitalization. I think that is going to be a critical friction point if we are to take EPO analogue use to a CV population.

Dr. McCullough: Generally we do not give patients therapies that hasten their demise, despite the time trade-off we mentioned.

Dr. Friedewald: Thank you.