Normal vs. Low Hematocrit Values in Patients with Cardiac Disease Receiving Hemodialysis and Epoetin

Abstract & Commentary

Synopsis: The one-year and two-year mortality rates were 7% higher in the normal-hematocrit group than in the low-hematocrit group.

Source: Besarab A, et al. N Engl J Med 1998;339:584-590.

Anemia is a common problem in chronic renal failure, and occurs primarily due to a deficiency of erythropoietin production in the kidney. Hence the majority of dialysis patients in the United States are treated with recombinant human erythropoietin (epoetin) to stimulate erythropoiesis and thus to partially correct the anemia. Prior to 1993, in a random sample of almost 1000 patients at 188 U.S. hemodialysis units, Besarab and colleagues found that almost 70% of the patients had hematocrits of 27-33%, 15% had hematocrits below 27%, and 16% had hematocrits above 33% (normal ranges for hematocrits are 37-48% for women, and 42-52% for men). Cardiac disease is the most common cause of death in the dialysis population and partial correction of anemia is associated with many benefits, including reduction left ventricular hypertrophy and improvement in exercise tolerance. Would a normal hematocrit lead to better mortality outcomes in dialysis patients with pre-existing cardiac disease? The present trial was designed to answer this question.

This 51-center study was a randomized, prospective, open-label trial that included 1233 hemodialysis patients with congestive heart failure or ischemic heart disease. All patients had hematocrits of 27- 33% and had received epoetin during the four weeks prior to enrollment—an average of three times weekly intravenously. Exclusion criteria included a diastolic blood pressure of 100 mmHg or more and severe cardiac disability (NYHA Class IV).

Six hundred eighteen patients were randomized to the normal-hematocrit group (goal hematocrit 42% ± 3%) and 615 patients to the low-hematocrit group (goal hematocrit 30% ± 3%). The primary end point was the length of time to death or a first nonfatal myocardial infarction. All adverse events were recorded, in addition to hemodialysis adequacy using Kt/V, and quality of life outcomes. The planned duration of the study was three years, but safety concerns prompted an early termination of the study (median follow-up 14 months).

The baseline characteristics of the normal- and low-hematocrit groups were similar. The normal-hematocrit group required about three times as much epoetin as before the study to maintain their hematocrits at 42%. After 29 months, there were 183 deaths and 19 first nonfatal myocardial infarctions in the normal-hematocrit group, compared with 150 deaths and 14 nonfatal myocardial infarctions in the low-hematocrit group (risk ratio for normal-hematocrit group as compared with low hematocrit group, 1.3). There were no baseline characteristics that explained the differences in outcomes in the two groups. The causes of death were similar in both groups, with cardiovascular deaths accounting for 67% of cases. The patients in the normal-hematocrit group had a decline in the adequacy of dialysis and received intravenous iron-dextran more often than those in the low-hematocrit group.

The one- and two-year mortality rates were 7% higher in the normal-hematocrit group than in the low-hematocrit group. However, in both the normal- and low-hematocrit study groups, there was a 30% reduction in the risk of death for each 10% increase in hematocrit for all patients.

Comment by Kamaljit sethi, MD, FACP

How much epoetin is enough? Doubtlessly, recombinant human erythropoietin has vastly improved the quality of life of hemodialysis patients. The question now is, could it become too much of a good thing.

End stage renal disease and anemia treatment are expensive. There are approximately 200,000 dialysis patients in the United States. The Medicare cost for dialysis in 1995 was approximately $10 billion, and epoetin costs are about $1 billion annually. The mean hematocrit in the Medicare End Stage Renal Disease Program was 31% in 1995. The National Kidney Foundation Dialysis Outcome Quality Initiative Anemia Work Group guidelines published in 1997 recommended a target hematocrit of 33-36%, using intravenous iron to optimize red cell production.

It would seem that if improving anemia was beneficial in reducing cardiac mortality, then normalizing hematocrits would be even better. Surprisingly, the answer seems to be no, if one looks at this study. Could higher iron doses and possible iron-overload related infections in the normal hematocrit group (range, 39-45%) be the reason for the higher mortality, or could it be because the dialysis adequacy declined in the normal-hematocrit group? Should we settle for a new hematocrit goal of 27-33%?

The following represents a reasonable framework for the management of anemia in patients with congestive heart failure or ischemic heart disease receiving hemodialysis:

• Administration of epoetin to raise hematocrit to 39% - 45% is not recommended.

• The goal hematocrit should be 33-36%, using epoetin and intravenous iron as needed.

• The route of administration of epoetin should be intravenous or subcutaneous, depending on whichever route allows the smallest dose to be given.

• Iron stores or deficiency should be assessed using transferrin saturation and ferritin levels, and ferritin levels should not exceed 500 mg/dL.