CHOIR Silenced as Findings Show Increased Risk of CVD Outcomes/Death

Caroline Cassels

April 24, 2006

April 24, 2006

Chicago, IL - A major study investigating whether exceeding current recommended hemoglobin targets improves outcomes in anemic patients with chronic kidney disease (CKD) has been terminated due to findings of an increased risk of death and cardiovascular hospitalization in patients assigned to achieve an investigational target hemoglobin of 13.5 g/dL.

The Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) study principal investigators were Dr Ajay Singh (Brigham and Women's Hospital, Boston, MA) and Dr Donal Reddan (University College Galway, Ireland). Singh presented the results for the first time here at the National Kidney Foundation 2006 Spring Clinical Meeting.

The study's primary hypothesis, said Singh, was that anemia correction to 13.5 g/dL with a once-weekly dosing of epoetin alfa (Procrit, Ortho Biotech) in patients with CKD would decrease mortality and cardiovascular morbidity, but this turned out not to be the case.

"Targeting of hemoglobin to 13.5 g/dL carries an increased risk in CKD patients compared with a hemoglobin target of 11.3 g/dL, the current standard of care as outlined by the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines. Our strong recommendation is to target and maintain a hemoglobin in the range of 11 to 12 g/dL in all CKD patients," Singh said.

Raising the target

Anemia, the result of both erythropoietin and iron deficiencies, is present in at least 45% of patients with stage 3 CKD, Singh said. Treatment of anemia improves well-being and exercise tolerance in these patients and reduces the need for transfusion, he noted.

"However, I think we also recognize that the recommended target hemoglobin for patients with CKD must still be established based on evidence," he said. Currently KDOQI guidelines that recommend a target of 11 to 12 g/dL are based on consensus, "and most of us in the clinic aim for this."

Reddan agreed and told renalwire , "These data support the current guideline target of 11 to 12 g/dL but raise concerns about going above that from a safety perspective."

As for evidence of efficacy, the current KDOQI guidelines are largely based on expert consensus and observational data, highlighting the need for randomized controlled trials to determine optimal targets.

Singh also noted that while evidence from observational studies suggests treatment of anemia to increase hemoglobin levels is associated with progressive benefits in hard end points, including CVD mortality, there have been "important glitches" in this hypothesis in treatment studies to date.

Results of the Normalization of Hematocrit study, published in 1998, suggested that treating patients with end-stage renal disease and high cardiovascular risk to a target hematocrit of 42% could be harmful [ 1 ]. In 2005, resultsfrom another randomized study [ 2 ], the Canadian European Study, showed no benefit on left-ventricular-volume index of anemia correction to a higher target (13.5-14.5 g/dL vs 9.5-11.5 g/dL) in patients on hemodialysis, although, Singh noted, mortality was not an end point in this trial.

Further randomized studies to look at this issue in patients with CKD include the Cardiovascular Risk Reduction by Early Anemia Treatment with Epoetin beta (CREATE), and CHOIR, the current study. CREATE results, presented in June 2005 at the European Renal Association/European Dialysis and Transplant Association (ERA/EDTA) XLII Congress and reported by renalwire at that time, showed no significant impact on CVD events with a higher-target hemoglobin vs standard target values in CKD patients.

According to Dr Jeffrey Berns (University of Pennsylvania School of Medicine, Philadelphia), anemia section editor for, the CHOIR finding is disturbing but, given the results of the Normalization of Hematocrit study, not surprising.

"This study highlights the need for properly conducted prospective clinical trials. Other large ongoing studies in CKD patients will provide additional useful information. In the meantime, the general consensus seems to be that while patients with CKD treated with erythropoietin-simulating agents should have hemoglobin levels maintained at about 11 g/dL, they should not be maintained above 13 g/dL, even thoughthese higher levels may be associated with a better sense of well-being and quality-of-life assessments in the short term," he said.

Launched in April 2002, CHOIR, an open-label, multicenter study, included 130 US sites and randomized a total of 1432 patients. A total of 715 and 717 subjects were assigned to achieve target hemoglobin levels of 13.5 or 11.3 g/dL, respectively.

All subjects were treated with epoetin alfa. Initial dosing was 10000 units by subcutaneous injection once weekly for three weeks. Beyond three weeks, the dose was adjusted based on hemoglobin response, to a maximum of 20000 units. Not surprisingly, said Singh, there was a much higher use of epoetin alfa in the 13.5-g/dL target group.

Subjects were monitored weekly for a median follow-up time of 16 months.

Death, CHF hospitalization end-point drivers

The study's primary outcome was a composite end point of death and cardiovascular outcomes, including stroke, myocardial infarction (MI), and congestive-heart-failure (CHF) hospitalization.


CHF hospitalization was characterized as congestive heart failure requiring admission during which the patient received intravenous therapy with inotropes, diuretics, or vasodilators. However, said Singh, if the hospitalization involved renal-replacement therapy it was not defined as a CHF hospitalization.

Inclusion criteria included CKD patients who had a hemoglobin of less than 11 g/dL, were over age 18, and had a estimated glomerular filtration rate (eGFR) of >15 mL/min and <50 mL/min.

Patients were excluded from the study if they had uncontrolled hypertension, iron overload, unstable angina pectoris, or refractory iron-deficiency anemia (TSAT <20%) or if they had received erythropoietin within three months of study entry.

A total of 222 composite end point events occurred up to study termination in May 2005 on the recommendation of the data safety monitoring board. Of these, 125 were in the 13.5-g/dL arm and 97 in the 11.3-g/dL arm.

"CHF hospitalization and death, rather than MI and stroke, appeared to be driving the composite end point," said Singh.

Risk associated with investigational 13.5-g/dL target hemoglobin vs target of 11.3 g/dL

End point
Hazard ratio
95% CI
Death, stroke, MI, CHF

To download table as a slide, click on slide logo below

Although there were a larger number of deaths (52 vs 36 at study termination) in the 13.5-g/dL study arm, this did not reach statistical significance, said Singh. Causes of death were primarily cardiovascular-related in both groups.

Over time, said Singh, the lower-goal arm of the study achieved its target of 11.3 g/dL. However, the upper-goal arm fell short of its 13.5-g/dL target and came in at 13 g/dL.

Stay tuned...

Large numbers of secondary end points, including all-cause mortality, renal-replacement therapy, and data on health-related quality of life and functional-status indicators were also included as part of CHOIR study. These and other findings, said Singh, will be published in due course.


  1. Besarab A, Bolton WK, Browne JK, et al. The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin. N Engl J Med 1998; 339:584-590.

  2. Parfrey P, Foley RN, Wittreich BH, et al for the Canadian European Study. Double-blind comparison of full and partial anemia correction in incident hemodialysis patients without symptomatic heart disease. J Am Soc Nephrol 2005; 16:2180-2189.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: