Hemoglobin Targets and Blood Transfusions in Hemodialysis Patients without Symptomatic Cardiac Disease Receiving Erythropoietin Therapy

Abstract

Background and objectives: Optimal hemoglobin targets for chronic kidney disease patients receiving erythropoiesis-stimulating agents remain controversial. The effects of different hemoglobin targets on blood transfusion requirements have not been well characterized, despite their relevance to clinical decision-making.

Design, setting, participants, & measurements: Five hundred ninety-six incident hemodialysis patients without symptomatic cardiac disease were randomly assigned to hemoglobin targets of 9.5 to 11.5 g/dl or 13.5 to 14.5 g/dl for 96 wk using epoetin alfa as primary therapy and changes in left ventricular structure as the primary outcome (previously reported). Patients were masked to treatment assignment. Blood transfusion data were prospectively collected at 4-wk intervals.

Results: The mean age and prior duration of dialysis therapy of the study population were 50.8 and 0.8 yr, respectively. Previously reported mortality was similar in low and high-target subjects, at 4.7 (95% confidence interval 3.0, 7.3) and 3.1 (1.8, 5.4) per hundred patient years, respectively. Transfusion rates were 0.66 (0.59, 0.74) units of blood per year in low and 0.26 (0.22, 0.32) in high-target subjects (P < 0.0001). Hemoglobin level at transfusion (7.7 [7.5, 7.9]) versus 8.1 [7.6, 8.5] g/dl) were similar with both groups. High hemoglobin target was a significant predictor of time to first transfusion independent of baseline associations (hazard ratio = 0.42; 95% confidence interval = 0.26 − 0.67).

Conclusions: In hemodialysis patients with comparatively low mortality risks, normal hemoglobin targets may reduce the need for transfusions.

Anemia is highly prevalent in ESRD patients, and inadequate renal erythropoietin production is a contributing cause (1,2). Although erythropoietin-stimulating agents have been used to treat anemia in patients with chronic kidney disease (CKD) for almost 2 decades, optimal hemoglobin targets remain unclear, and concerns remain about the safety of normal hemoglobin levels as a target level for erythropoietin therapy. In this regard, the U.S. Food and Drug Administration (FDA) recently issued an alert to healthcare professionals about revisions to the product label for erythropoiesis-stimulating agents in patients with CKD (3). Notable features of this label change were the recommended hemoglobin target range of 10 to 12 g/dl and the removal of all quality of life claims, with the exception of improved exercise tolerance and functional ability. In addition, among patients exhibiting inadequate hemoglobin responses to appropriate dose titrations, it was recommended that future dosing should target hemoglobin levels that minimize transfusions requirements.

In day-to-day practice, the decision to administer transfusions in patients with CKD is heavily influenced by hemoglobin levels. It is possible that higher hemoglobin targets than those recommended by the FDA will limit transfusion requirements. However few long-term, randomized, controlled clinical trials have formally compared transfusion requirements at different hemoglobin targets in hemodialysis patients. This is important, not only from the perspective of transfusion-related infection (4), but also because transfusions can have deleterious immunomodulatory effects in patients in whom renal transplantation is a consideration (5).

We enrolled 596 incident hemodialysis patients without symptomatic cardiac disease in a randomized, controlled trial that compared a normal hemoglobin target to partial correction of anemia, with epoetin alfa as the erythropoiesis-stimulating agent. Changes in cardiac structure constituted the primary study outcome and no difference was observed between the two groups (6). Data on blood transfusions were prospectively collected, although transfusion rates were not a prespecified secondary outcome. In this paper we examine the hypothesis that normal hemoglobin targets reduce transfusion rates compared with conventional targets.

Materials and Methods

Design

The design, methods, and sample size assumptions of the study have been reported previously (6).

Patients were randomly assigned to one of the following hemoglobin targets: 9.5 to 11.5 g/dl (“low target”) and 13.5 to 14.5 g/dl (“high target”). Patients were masked to treatment assignment, as were their doctors. However doctors had access to clinic hemoglobin values when managing the patients. Mean hemoglobin levels at the end of the initial 24-wk titration phase were 13.3 and 10.9 g/dl, respectively. During the maintenance phase, from weeks 24 to 96, corresponding mean hemoglobin levels were 13.1 and 10.8 g/dl.

The study was centrally coordinated from St. John's, Canada for Canadian patients and Manchester, England for European patients. Randomization was performed at the coordinating centers with an interactive voice randomization telephone system using permuted blocks stratified by concurrent epoetin use and sex.

Study Population

Inclusion criteria were as follows: age ≥18 yr, inception of maintenance hemodialysis within the previous 3 to 18 mo, predialysis hemoglobin between 8 and 12 g/dl, left ventricular volume index <100 ml/m², and predialysis diastolic BP <100 mmHg. Exclusion criteria were as follows: clinical evidence or history of symptomatic cardiac failure or ischemic heart disease; daily prednisone dose ≤10 mg; medical conditions likely to reduce epoetin responsiveness, including uncorrected iron deficiency; concurrent malignancy; blood transfusion in the preceding month; therapy with cytotoxic agents; seizure in the preceding year; hypersensitivity to intravenous iron; and current pregnancy or breastfeeding.

Description of Study Procedures

Laboratory tests were measured centrally by Quest Diagnostics (Van Nuys, California, and Heston, United Kingdom). Hemoglobin was measured weekly for 24 wk and biweekly thereafter. With the high target, the treatment goal was increments of 0.5 to 1.0 g/dl every 2 wk, until achieving stability between 13.5 and 14.5 g/dl. Other treatment goals included predialysis diastolic BP between 70 and 90 mmHg; urea reduction ratio >67%, and transferrin saturation ≥20%.

After random treatment assignment, patients assigned to the low target remained on their prestudy epoetin dose. Patients with the higher target received a 25% dose escalation, or an initial dose of 150 U/kg/wk if naïve to epoetin. In both groups, when hemoglobin levels deviated from target, epoetin doses were changed by 25% of the previous dose or 25 U/kg. For each patient, a standardized form was faxed weekly from the study sites to the coordinating center showing hemoglobin and epoetin dose, BP, and transferrin saturation levels. In response, treatment recommendations were faxed weekly from the coordinating center to the study sites. Initially, the choice of subcutaneous or intravenous epoetin alfa administration was discretionary; concerns about pure red cell aplasia (7) lead to a study amendment on August 22, 2002 limiting administration to the intravenous route. The last patient completed the study in May 2003.

Patients were questioned about blood transfusions and patient notes were examined at 4-wk intervals throughout the study. Specifically, responses were recorded for the following data elements: Did the patient receive any red blood cell transfusions in the preceding 4 wk? If the previous response was affirmative, what were the dates of transfusion, the pretransfusion hemoglobin levels, the nature of the transfusion [packed red blood cells (92.5% of units transfused) or whole blood (7.5% of units transfused)], and how many units were transfused? Adverse events occurring within 4 wk before the first transfusion were classified by the World Health Adverse Reactions Terminology (8) preferred term and aggregated as follows:

Hemorrhage: epistaxis, gastric ulcer hemorrhagic, gastrointestinal hemorrhage, hematoma, hematuria, hemoptysis, nose hemorrhage, rectal hemorrhage, hemothorax, oral hemorrhage, peptic ulcer hemorrhagic, vaginal hemorrhage, and cystitis hemorrhagic.

Infection: fever, herpes zoster, infection, bacterial infection, fungal infection, influenza-like symptoms, peritonitis, pneumonia, sinusitis, and tooth caries

Vascular access problems: arterio-venous fistula loss or thrombosis, device related complications, permanent dialysis catheter loss, and thrombosis.

Surgical intervention

Anemia and related symptoms: anemia, asthenia, fatigue, and malaise.

Cardiovascular: BP fluctuation, cardiac failure, chest pain, coronary artery disorder, dizziness, hypertension, hypotension, myocardial infarction, non-site-specific vascular disorder, palpitations, pericarditis, peripheral gangrene, pulmonary edema, and vascular disorder

Respiratory: coughing, cyanosis, dyspnea, and atrial fibrillation.

Gastrointestinal: abdominal pain, anorexia, ascites, ulcerative colitis, diarrhea, gastric ulcer, hepatic cirrhosis, intestinal obstruction, nausea, oesophagitis, and vomiting.

Musculoskeletal: arthralgia, arthritis, arthropathy, back pain, bone disorder, fall, fracture pathologic, injury, leg pain, myalgia, skeletal pain, and anklosing spondylitis.

Skin: folliculitis, pruritus, purpura, rash, skin disorder, and skin ulceration

Neurologic: cerebellar infarction, cerebral atrophy, cerebrovascular disorder, coma, confusion, gait abnormal, headache, hearing decreased, insomnia, ischial neuralgia, somnolence, and abnormal vision.

Miscellaneous: acidosis, allergic reaction, anxiety, aggravated diabetes mellitus, dysuria, hydronephrosis, hyperkalemia, hyperparathyroidism, hypoglycemia, nail disorder, non-site-specific embolism, thrombosis, edema, generalized edema, peripheral edema, pain, renal cyst, thrombocytopenia, thrombosis, transplant rejection, Wegener's granulomatosis, weight decrease.

Sample Size Estimate

The sample size needed to detect a 15% difference between treatment groups in the primary outcome (left ventricular cavity volume index) was calculated as 166 per treatment group, given a two-tailed significance of 0.05, a power of 0.90, and an SD of the percentage change in left ventricular cavity volume index of 42% (6). With an expected dropout rate of 40%, primarily as a result of transplantation, 277 patients were required for each treatment group.

Analysis

Red cell transfusion rates per year were calculated, with the total number of units of red blood cells and the total duration of follow-up constituting the numerator and denominator for each subject, respectively. Poisson regression was used for between-target rate comparisons. In addition, because individuals with heavy transfusion requirements could potentially bias comparisons, we also compared proportions receiving at least one transfusion, as well as transfusion-free survival with proportional hazards regression. Finally, we hypothesized that threshold hemoglobin levels for transfusion could differ between high and low targets. To assess the sensitivity of the study findings to this possibility, we performed supplementary analyses in which transfusion outcomes were predicated on specific hemoglobin levels (≤9 g/dl, ≤8 g/dl, and ≤7 g/dl). Multivariate models to determine independent factors predictive of transfusion rates and first transfusion event were created. Details of the models are contained in Table 3.

Table 3.

Multivariate analysis: Baseline associations of transfusion^a

Parameter	Transfusions as Rates		Time to First Transfusion
	Rate Ratio	P	Hazards Ratio	P
Target 13.5 to 14.5 g/dl	0.36 (0.29, 0.46)	<0.0001	0.42 (0.26, 0.67)	0.0003
Hemoglobin ≤11.1 g/dl	1.93 (1.54, 2.43)	<0.0001	1.48 (0.93, 2.34)	0.01
Epoetin dose >6000 U/wk	2.28 (1.85, 2.82)	<0.0001	1.63 (1.03, 2.56)	0.04
Transferrin saturation ≤32.0%	1.87 (1.51, 2.32)	<0.0001	1.55 (0.99, 2.41)	0.05
Age >51.5 yr	0.85 (0.69, 1.06)	0.14	0.79 (0.50, 1.24)	0.30
Female sex	0.59 (0.47, 0.73)	<0.0001	0.89 (0.57, 1.40)	0.62
Non-White race	1.02 (0.75, 1.38)	0.91	0.73 (0.35, 1.50)	0.39
Dialysis duration >9.0 mo	1.85 (1.50, 2.28)	<0.0001	1.56 (1.00, 2.42)	0.049
Body mass index >25.5 kg/m2	0.71 (0.58, 0.87)	0.001	0.84 (0.54, 1.31)	0.45
Renal disease due to diabetes	1.16 (0.89, 1.52)	0.28	1.21 (0.69, 2.14)	0.50
Europe	1.89 (1.44, 2.49)	<0.000	1.11 (0.65, 1.90)	0.69
Dialysis catheter	2.01 (1.32, 3.07)	0.001	1.38 (0.56, 3.40)	0.48
Serum albumin ≤4.0 mg/dl	1.37 (1.09, 1.72)	0.007	1.63 (0.98, 2.72)	0.06

^a95% confidence intervals are reported in parentheses. Poisson regression was used to calculate rate ratios for transfusion and proportional hazards regression to compute hazards ratio for the time to first transfusion. All of the variables shown in the first column were included simultaneously in multivariate models. Median values of the overall study population were used to categorize hemoglobin, epoetin dose, transferrin saturation, dialysis duration, age, body mass index, and serum albumin.

Results

Five hundred ninety-six patients were enrolled in 95 treatment centers in 10 countries between February 2000 and June 2001. Table 1 compares baseline characteristics by random hemoglobin target assignment (6). Baseline characteristics were similar except for the older age of high-target subjects (52.2 versus 49.4 yr). As dictated by the study design, initial on-study epoetin doses were greater in high-target subjects (7009 versus 6183 U/wk).

Table 1.

Baseline characteristics, compared by target hemoglobin level^a

Variable	Target 9.5 to 11.5 g/dln = 300	Target 13.5 to 14.5 g/dln = 296	P
Hemoglobin (g/dl)	11.0 (10.8, 11.1)	11.0 (10.9, 11.2)	0.54
Epoetin dose (units per week)	6183 (5698, 6667)	7009 (6528, 7490)	0.02
Transferrin saturation (%)	36.8 (34.9, 38.8)	35.8 (33.8, 37.7)	0.44
Age	49.4 (47.7, 51.2)	52.2 (50.4, 53.9)	0.03
Female sex (%)	39.7	39.5	0.97
Race	0.24
White	87.7	91.2
Black	5.7	4.4
Asian	4.3	1.7
other	2.3	2.7
Dialysis duration (months)	10.2 (9.6, 10.8)	10.0 (9.4, 10.5)	0.58
Body mass index (kg/m2)	26.3 (25.7, 26.9)	26.5 (25.9, 27.1)	0.78
Country (%)
Austria	3.0	4.1
Belgium	3.3	4.7
Canada	32.0	27.7
France	3.0	2.7
Germany	11.0	12.5
Greece	2.3	1.4
Hungary	10.7	12.2
Poland	22.3	22.6
Spain	3.0	3.0
United Kingdom	9.3	9.1
Primary cause of renal disease	0.54
glomerulonephritis	29.0	28.4
diabetes	16.7	18.9
hypertension	9.3	6.8
polycystic kidney disease	7.7	10.5
other/unknown	37.3	35.5
Dialysis access (%)	0.21
fistula	82.7	85.8
graft	5.0	6.1
catheter	12.3	8.11
Serum albumin (mg/dl)	4.0 (3.9, 4.0)	4.0 (3.9, 4.0)	0.88

^aData are reported either as percent or as means (95% confidence intervals). The χ² test and analysis of variance were used for between-target comparisons.

There were 20 deaths in low-target and 13 deaths in high-target subjects (6), equivalent to rates of 4.7 (95% confidence interval 3.0, 7.3) and 3.1 (1.8, 5.4) per hundred patient years, respectively (P = 0.25).

The average number of transfusions per patient during the course of the study and the transfusion rate differed by random treatment assignment (Table 2). In contrast, duration of follow up and hemoglobin levels immediately preceding transfusion were similar. The proportions receiving a blood cell transfusion during the course of the study (9.1% with the high target versus 19.3% with the low target) and transfusion-free survival times (hazards ratio 0.40 for the high target, P = 0.0002) differed by random treatment assignment (Table 2). In contrast, hemoglobin levels immediately preceding transfusion levels were similar in both groups. The proportions receiving at least one transfusion with pretransfusion hemoglobin levels ≤9 g/dl (7.8% versus 18.0%, P = 0.0002), ≤8 g/dl (5.1% versus 15.7%, P < 0.0001) and ≤7 g/dl (3.0% versus 7.7%, P = 0.0122) were also lower in the group whose target hemoglobin level was 13.5 to 14.5 g/dl.

Table 2.

Effect of hemoglobin target on transfusions^a

Parameter	Target Hemoglobin		P
	9.5 to 11.5 g/dl	13.5 to 14.5 g/dl
Transfusions as rates
transfusions per patient	0.94 (0.62, 1.26)	0.37 (0.20, 0.37)	0.00
follow-up (yr)	1.42 (1.35, 1.49)	1.41 (1.34, 1.48)	0.82
transfusion rate (per patient per year)	0.66 (0.59, 0.74)	0.26 (0.22, 0.32)	<0.0001
transfusion rate ratio	1 (reference category)	0.40 (0.32, 0.50)
pretransfusion hemoglobin (g/dl)	7.7 (7.5, 7.9)	8.1 (7.6, 8.5)	0.09
Time to first transfusion
proportion transfused (%)	19.3	9.1	0.0004
pretransfusion hemoglobin (g/dl)	7.6 (7.3, 8.0)	8.2 (7.7, 8.7)	0.12
transfusion hazards ratio	1 (reference category)	0.46 (0.29, 0.72)	0.0007

^a95% confidence intervals are reported in parentheses. Poisson regression was used for comparison of transfusion rates and proportional hazards regression to compute hazards ratio for the time to first transfusion.

The association between hemoglobin target and transfusion rates persisted when adjustment was made for other baseline characteristics (Table 3). Other multivariate associations of transfusion included baseline hemoglobin level, epoetin dose, transferrin saturation, sex, dialysis duration, body mass index, European or Canadian study site, dialysis vascular access, and serum albumin level. Similarly, with identical adjustment strategies to those used in Table 3, adjusted hazards ratios for transfusion with hemoglobin levels ≤9, ≤8 and ≤7 g/dl were 0.38 (0.23, 0.63, P = 0.0002), 0.28 (0.16, 0.51, P < 0.0001), and 0.40 (0.18, 0.88, P = 0.023).

The adverse events occurring within 4 wk before the first transfusion were diverse and similar (except for gastrointestinal) across the low- and high-target groups (Table 4).

Table 4.

Adverse events occurring within 4 wk before the first transfusion, classified by World Health Adverse Reactions Terminology preferred term and grouped by clinical manifestation

Clinical Manifestation	Target 9.5 to 11.5g/dl 58 Patients		13.5 to 14.5 g/d 27 Patients
	N	%	N	%
Hemorrhage	22	9.9	11	13.8
Infection	18	8.1	5	6.3
Vascular access problems	9	4.0	4	5.0
Surgery	6	2.7	2	2.5
Anemia/symptoms	8	3.6	4	5.0
Cardiovascular	23	10.3	9	11.3
Respiratory	10	4.5	2	2.5
Gastrointestinal	25	11.2	1	1.3
Musculoskeletal	30	13.5	13	16.3
Skin	28	12.6	10	12.5
Neurologic	12	5.8	8	10.0
Miscellaneous	31	13.9	11	13.8
Total	223	100.0	80	100.0

Discussion

Randomized controlled trials comparing a normal hemoglobin target to partial corrections of anemia with epoetin have varied in the stage of CKD disease of patients enrolled, primary outcomes assessed, statistical power to compare major clinical outcomes, and research methodology. Nonetheless, signals have emerged to suggest that higher hemoglobin targets may be harmful. Clinical events related to higher hemoglobin targets in some trials have included higher vascular access thrombosis (9), higher BP or greater requirements for antihypertensives (6,10,11), cerebrovascular events (6), cardiovascular events (9,10), earlier need for renal replacement therapy (12), and higher mortality (9,10). These adverse outcomes have not been seen uniformly across studies, and several outcomes have had marginal statistical significance and wide confidence intervals for effect estimates. However higher vascular access thrombosis and hypertension have been observed in the early studies comparing no treatment of anemia with partial correction using erythropoietin (13).

Controversy exists concerning target hemoglobin levels for erythropoiesis stimulating agents therapy in CKD. At the joint meeting of the FDA Cardiovascular and Renal Drugs Advisory Committee and the Drug Safety and Risk Management Advisory Committee, members were divided on whether a specific target hemoglobin should be set (14). The European Medicines Agency stipulated a uniform target hemoglobin range for all patients with CKD of 10 to 12 g/dl, with a warning not to exceed a concentration of 12 g/dl (15). It noted that trials with high-target hemoglobin concentrations “have not shown significant benefits attributable to the administration of epoetins to increase hemoglobin concentrations beyond the level necessary to control symptoms of anemia and to avoid blood transfusion.” However, this study of “healthy” patients starting hemodialysis, with a sample size similar to CREATE (12), clearly demonstrates that higher hemoglobin targets reduce the need for blood transfusions.

Transfusion was a common event in this study, despite the selection of a comparatively healthy population of hemodialysis patients and the requirement of a transfusion-free interval before study entry. Transfusions were between two and three times more frequent with low compared with high hemoglobin targets, a disparity that could not be explained by differences in baseline characteristics or hemoglobin levels precipitating transfusion.

Blood transfusions are labor-intensive and have intrinsic risks, both short-term and long-term (4), including transfusion-associated infection, systemic iron overload (16,17), and stimulation of antibodies to HLA, which can jeopardize future renal transplantation prospects (5). Few, if any, long-term studies have formally quantified the effect of different hemoglobin targets on transfusion use in CKD populations in the last decade. This is surprising, because registry data indicate that red blood cell transfusion remains a comparatively common event in latter-day dialysis populations (18). The U.S. Normal Hematocrit Study reported a significantly lower proportion transfused (21%) in a high-hematocrit group compared with low-hematocrit patients (31%) over 29 mo (9), but it differs from our study in several respects. First, in direct contradistinction to our study, overt cardiac disease was an absolute inclusion criterion, it preceded ours by almost a decade, and was performed in an era in which iron management was considerably less aggressive than currently. Finally, compared with our study, epoetin doses were approximately two times higher with the low hematocrit target and three times higher with the high hematocrit target.

Some of the limitations of this study are worth considering. Transfusion was not the primary study outcome of this trial. However, transfusions were carefully recorded, in a prospective manner and at regular intervals, so that the precision of the outcome is likely to be high. Treating physicians were masked to treatment assignment but not masked to ongoing hemoglobin levels during the trial. Although there is no definitive answer to the hypothesis that threshold hemoglobin levels stimulating transfusion could differ because treating physicians knew assigned hemoglobin targets, analysis of pretransfusion hemoglobin levels and transfusion at specific hemoglobin levels did not support this hypothesis. Patients, in contrast, were masked, a design feature that tends to lessen the possibility that patient-related biases could explain the findings. By design, we studied hemodialysis patients without overt cardiac disease. About 50% of incident dialysis patients in Canada have not had symptomatic cardiac failure or coronary artery disease on starting dialysis (19). The generalizability of our findings to other populations with CKD is not certain. One would anticipate fewer transfusions in non-dialysis-dependent CKD patients, and more in hemodialysis patients with more comorbid illnesses. The high dropout rate was anticipated because patients such as those in our study are usually referred for and then wait for renal transplantation. This was taken into account in the sample size estimate.

We conclude that in incident hemodialysis patients without symptomatic cardiac disease in which the mortality rates are low, normal hemoglobin targets lower blood transfusion requirements.

Members of the Canadian European Study Group

EPO-INT-68 Independent Data Monitoring Committee Members.

L.J. Wei, Boston, Massachusetts; M.-M. Samama, Paris, France; P. Ivanovich, Chicago, Illinois; M.A. Pfeffer, Boston, Massachusetts

Principal Investigator/Site.

W. Hoerl, Wien, Austria; H.-K. Stummvoll, Linz, Austria; G. Mayer, Innsbruck, Austria; H. Graf, Wien, Austria; H. Holzer, Graz, Austria; Y. Vanrenterghem, Leuven, Belgium; M. Jadoul, Bruxelles, Belgium; P. Parfrey, St. John's, Canada; P. Barre, Montréal, Canada; A. Levin, Vancouver, Canada; P. Cartier, Montréal, Canada; N. Muirhead, London, Canada; A. Fine, Winnipeg, Canada; B. Murphy, Calgary, Canada; S. Handa, St. John's, Canada; P. Campbell, Edmonton, Canada; V. Pichette, Montreal, Canada; S. Tobe, Toronto, Canada; C. Lok, Toronto, Canada; D. Kates, Kelowna, Canada; D. Holland, Kingston, Canada; G. Karr, Penticton, Canada; G. Pylpchuk, Saskatoon, Canada; G. Wu, Mississauga, Canada; M. Vasilevsky, Montreal, Canada; E. Carisle, Hamilton, Canada; E.R. Gagne, Fleurimont, Canada; W. Callaghan, Windsor, Canada; G. Soltys, Greenfield Park, Canada; P. Tam, Scarborough, Canada; R. Turcot, Trios-Rivieres, Canada; M. Berrall, Toronto, Canada; J. Zacharias, Winnipeg, Canada; S. Donnelly, Toronto, Canada; G. London, Fleury-Merogis, France; A. London, Aulnay sous Bois, France; F.P. Wambergue, Lille, France; H. Geiger, Frankfurt, Germany; V. Kliem, Hann Muenden, Germany; R. Winkler, Rostock, Germany; B. Kraemer, Regensburg, Germany; H. Schiffl, Munich, Germany; R. Brunkhorst, Hannover, Germany; D. Seybold, Bayreuth, Germany; M. Hilfenhaus, Langenhagen, Germany; D. Schaumann, Hameln, Germany; R. Goetz, Bad Windsheim, Germany; P. Roch, Regensburg, Germany; H.-P. Brasche, Ludwigshafen, Germany; V. Wizemann, Giessen, Germany; K. Bittner, Ansbach, Germany; K. Appen, Hamburg, Germany; B. Schroeder, Bad Toelz, Germany; W. Schropp, Munich, Germany; D. O'Donoghue, Salford, England; I. MacDougall, London, England; G. Warwick Leicester, England; M. Raftery, London, England; K. Farrington, Stevenage, England; J. Kwan, Carshalton, England; P. Conlon, Dublin, Ireland; G. Mellotte, Dublin, Ireland; K. Siamopoulos, Ioannina, Greece; N. Tsaparas, Crete, Greece; D. Tsakiris, Veria, Greece; V. Vargemezis, Alexandroupolis, Greece; S. Ferenczi, Gyor, Hungary; S. Gorogh, Kisvarda, Hungary; I. Kulcsar, Szombathely, Hungary; L. Locsey, Debrecen, Hungary; K. Akocsi, Veszprem, Hungary; I. Solt, Szekesfehervar, Hungary; O. Arkossy, Budapest, Hungary; E. Kiss, Szeged, Hungary; J. Manitius, Bydgoszcz, Poland; B. Ruthowski, Gdansk, Poland; A. Wiecek, Katowice, Poland; W. Sulowicz, Krakow, Poland; A. Ksiazek, Lublin, Poland; S. Czekalski, Poznan, Poland; M. Klinger, Wroclaw, Poland; M. Mysliwiec, Bialystok, Poland; H. Augustyniak-Bartosik, Milicz, Poland; Imiela, Warszawa-Miedzylesie, Poland; A. Sydor, Tarnow, Poland; R. Rudka, Bytom, Poland; M. Kiersztejn, Chrzanow, Poland; R. Wnuk, Oswiecim, Poland; A. Milkowsk, Krakow, Poland; F. Valderrabano, Madrid, Spain; P. Aljama, Còrdoba, Spain; H. Alcocer, Valencia, Spain; A. Purroy, Pamplona, Spain.

Funding

The Canadian-European Normalization of Hemoglobin with Erythropoietin Trial was funded by Johnson and Johnson Pharmaceutical Research and Development.

Role of Funding Source

The study sponsor identified the participating centers, monitored the data collection, and entered the data in a central database.

R.N.F. and P.S.P., the co-principal investigators, designed the trial, applied for funding to Johnson and Johnson, coordinated the study, analyzed the data, and wrote the report.

Conflict of Interest Statement

P.S.P. has received research support and is an academic advisor to companies that make erythropoietin products—Ortho Biotech, Amgen, and Roche. R.N.F. has received research support and honoraria from Ortho Biotech and honoraria from Affymax, Amgen, Ortho Biotech, and Roche. B.M.C. has received research support and honoraria from Ortho Biotech. P.S.P. declares that he had full access to all of the data in the study and had final responsibility for the decision to submit for publication.

Disclosures

None.