Correspondence Boston University Medical Center, Renal Section, EBRC 504, 650 Albany Street, Boston, MA 02118, USA

Email ldember@bu.edu

Original article

Singh AK et al. (2006) Correction of anemia with epoetin alfa in chronic kidney disease. N Engl J Med 355: 2085–2098   PubMed

Synopsis

Background

There are no definitive data to indicate the optimal hemoglobin level in patients with anemia caused by chronic kidney disease (CKD).

Objective

To establish whether correction of anemia to a hemoglobin level of 135 g/l (13.5 g/dl) lowers the risk of adverse cardiovascular outcomes and death in patients with CKD, compared with correction to 113 g/l (11.3 g/dl).

Design

The open-label, randomized, multicenter US CHOIR (Correction of Hemoglobin and Outcomes in Renal Insufficiency) study recruited adults with anemia (hemoglobin level <110 g/l [<11.0 g/dl]) and CKD (glomerular filtration rate 15–50 ml/min/1.73 m², as estimated by the modification of diet in renal disease equation). Individuals with iron overload, active gastrointestinal bleeding, or uncontrolled hypertension were among those excluded.

Intervention

Participants received sufficient subcutaneous epoetin alfa to achieve a randomly assigned target hemoglobin level of either 135 g/l or 113 g/l. Maintenance doses were administered fortnightly. For patients who did not experience an end point, follow-up was censored at completion of, or withdrawal from, the study.

Outcome measures

A composite of death, myocardial infarction, stroke, and hospital admission for congestive heart failure (not involving renal replacement therapy) was the primary end point. Quality of life (QOL) was a secondary end point.

Results

Of the 1,432 patients enrolled, 715 were assigned to the high hemoglobin target and 717 were assigned to the low hemoglobin target. The study was terminated early because interim analysis indicated that it was unlikely that a beneficial effect of the higher target would be found. Overall median follow-up was 16 months. A total of 222 patients (15.5%) experienced a primary end point. Of the remaining 1,210 patients, 661 (46.2%) completed 36 months of follow-up and 549 (38.3%) withdrew from the study early. Reasons for early withdrawal did not differ significantly between the hemoglobin target groups. The groups also had largely similar baseline demographic and clinical characteristics. Mean hemoglobin levels during the study were 113 g/l in the low-target group and 126 g/l (12.6 g/dl) in the high-target group. In the intention-to-treat analysis, there was a significantly higher rate of primary end point events in the patients assigned the high target than in those assigned the low target (17.5% [125/715] vs 13.5% [97/717]; hazard ratio 1.34, 95% CI 1.03–1.74; P = 0.03). The end points included 65 deaths (29.3%), 101 hospital admissions for congestive heart failure (45.5%), 25 myocardial infarctions (11.3%) and 23 strokes (10.4%). Improvement in QOL, as evaluated by the LASA, KDQ and SF-36 instruments, was generally similar in both groups. The rate of serious adverse events was significantly higher in the high-hemoglobin group than in the low-hemoglobin group (54.8% [376/686] vs 48.5% [334/688]; P = 0.02).

Conclusion

Patients with anemia caused by CKD are at higher risk of death or cardiovascular events when the target hemoglobin level is 135 g/l rather than 113 g/l.

Commentary

The CHOIR study highlights the importance of randomized controlled trials in determining optimal treatment approaches for patients with CKD. The trial was undertaken with the expectation that the higher hemoglobin target would be associated with better clinical outcomes; however, the opposite effect was found. The Cardiovascular Risk Reduction by Early Anemia Treatment with Epoetin Beta (CREATE) trial, results of which were published in the same issue of the New England Journal of Medicine, also compared high (130–150 g/l [13.0–15.0 g/dl]) and low (105–115 g/l [10.5–11.5 g/dl]) hemoglobin targets in patients with stage 3 and 4 CKD.¹ In this trial, the high hemoglobin target was associated with improved QOL, but it did not reduce the risk of a first cardiovascular event or affect left ventricular mass index, and it was associated with a more rapid progression to dialysis. Despite some limitations of the CHOIR and CREATE trials (neither was blinded; the dropout rate was high in CHOIR; and rates of the primary end point components were lower than anticipated in CREATE), their findings are compelling.

The apparently erroneous expectation that a normal hemoglobin target would result in improved clinical outcomes was based on observational cohort studies in both predialysis patients and those on dialysis,^{2, 3} and perhaps also on the general intuition that 'normal' should be better than 'subnormal'. Like the CHOIR and CREATE studies, earlier randomized controlled trials in patients on hemodialysis^{4, 5} and patients with advanced CKD⁶ found no benefit of high hemoglobin targets. These results have, however, been viewed as either not highly generalizable because enrollment was confined to patients with underlying heart disease,⁴ or as nondefinitive because sample sizes were small⁶ or intermediate end points were used.^{5, 6} There are multiple possible reasons for discrepant results between observational studies and large randomized trials. Most importantly, observational studies cannot distinguish indicators of risk from causal factors.

Hemoglobin levels in the US end-stage renal disease population have steadily increased since recombinant erythropoietin became available. The mean hemoglobin concentration is currently 120 g/l (12.0 g/dl), and 20% of patients have concentrations greater than 130 g/l (13.0 g/dl).⁷ The CHOIR and CREATE studies, together with previous randomized controlled trials in patients on dialysis, indicate that we should aim for only partial correction of anemia in patients with CKD, and that 115 g/l (11.5 g/dl) is a reasonable upper limit for a hemoglobin target. It is important to recognize that, because of the difficulty of maintaining hemoglobin concentrations within a narrow window, reducing the upper limits of hemoglobin targets is likely to increase the proportion of patients with hemoglobin concentrations less than 110 g/l (11.0 g/dl), and perhaps even the proportion with levels below 100 g/l (10.0 g/dl). The ongoing Trial to Reduce Cardiovascular Events with Aranesp Therapy (TREAT), which allows a hemoglobin level as low as 90 g/l (9.0 g/dl) in one treatment arm, should provide important information about acceptable lower limits.

The CHOIR and CREATE trials have received substantial attention from nephrology societies, the Centers for Medicare and Medicaid Services, the US Congress, the pharmaceutical industry, and dialysis providers. It is likely that the hemoglobin targets recommended in clinical practice guidelines will be lowered in response to the findings of these trials, and reimbursement policies for erythropoiesis-stimulating agents might be altered.

Acknowledgments

The synopsis was written by Chloë Harman, Associate Editor, Nature Clinical Practice.

References

1. Drueke TB et al. (2006) Normalization of hemoglobin level in patients with chronic kidney disease and anemia. N Engl J Med 355: 2071–2084 | Article | PubMed | ISI | ChemPort |
2. Jurkovitz CT et al. (2003) Association of high serum creatinine and anemia increases the risk of coronary events: results from the prospective community-based atherosclerosis risk in communities (ARIC) study. J Am Soc Nephrol 14: 2919–2925 | Article | PubMed | ISI | ChemPort |
3. Volkova N and Arab L (2006) Evidence-based systematic literature review of hemoglobin/hematocrit and all-cause mortality in dialysis patients. Am J Kidney Dis 47: 24–36 | Article | PubMed |
4. Besarab A et al. (1998) 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 339: 584–590 | Article | PubMed | ISI | ChemPort |
5. Parfrey PS et al. (2005) Double-blind comparison of full and partial anemia correction in incident hemodialysis patients without symptomatic heart disease. J Am Soc Nephrol 16: 2180–2189 | Article | PubMed | ISI |
6. Levin A et al. (2005) Canadian randomized trial of hemoglobin maintenance to prevent or delay left ventricular mass growth in patients with CKD. Am J Kidney Dis 46: 799–811 | Article | PubMed | ISI | ChemPort |
7. US Renal Data System (2006) Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. Bethesda, MD: National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Disease

Contact the journal about this article