CHOIR

From Wiki Journal Club
Jump to: navigation, search
Singh, AK et al. "Correction of Anemia with Epoetin Alfa in Chronic Kidney Disease". The New England Journal of Medicine. 2006. 355(20):2085-2098.
PubMedFull textPDF

Clinical Question

In patients with non-dialysis–dependent CKD and anemia, how does erythropoetin treatment targeting a hemoglobin >13.5 g/dL compare with a hemoglobin >11.3 g/dL in reducing CV events including death, MI, hospitalizations for CHF, and stroke?

Bottom Line

Erythropoietin therapy targeting a higher hemoglobin level of 13.5 g/dL, as compared with 11.3 g/dL, was associated with a higher risk of death and hospitalizations for CHF among patients with non-dialysis–dependent CKD and anemia.

Major Points

The Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) trial randomized 1,432 patients with non-dialysis–dependent CKD and anemia to achieve a target hemoglobin (Hgb) of either 13.5 or 11.3 g/dL. CHOIR was terminated early because there was no evidence of CV benefit in the high-Hgb group. In addition, there were a higher number of events in the high-Hgb arm, primarily due to nonsignificant trends towards a higher risk of death and of hospitalization for CHF. Improvements in quality of life were similar in both arms.

The CHOIR, CREATE, and TREAT trials investigated target hemoglobin levels in patients with non-dialysis–dependent CKD, while the Normal Hematocrit Trial obtained data in hemodialysis patients. Overall, the overwhelming evidence from these trials suggest that targeting normal or near-normal Hgb values confer no significant clinical benefit but may in fact be associated with increased harm. Based on these trials, the K/DOQI Anemia Workgroup updated their 2007 guidelines to recommended that the Hgb target not exceed 13 g/dL.

Design

  • Multicenter, open-label, parallel-group, randomized, controlled trial
  • N=1,432 patients with CKD
    • High-Hgb goal, target 13.5 g/dL (n=715)
    • Low-Hgb goal, target 11.3 g/dL (n=717)
  • Primary outcome: composite of death, MI, hospitalization for CHF without renal replacement therapy, or stroke
  • Setting: 130 centers in the US
  • Enrollment: 2003-2005, terminated early in May 2005 at second interim analysis
  • Median follow-up: 16 mos
  • Analysis: Intention-to-treat

Population

Inclusion Criteria

  • Age ≥18 years
  • CKD, defined by estimated GFR of 15-50 ml/min/1.73m2BSA by MDRD
  • Hgb <11.0 g/dL

Exclusion Criteria

  • CKD receiving dialysis
  • Uncontrolled HTN
  • Active GI bleed
  • Iron-overload state
  • History of frequent transfusions in prior 6 mos
  • Refractory iron deficiency anemia
  • Active cancer
  • Previous therapy with EPO
  • Unstable angina

Baseline Characteristics

Comparisons are high- vs. low-Hgb groups.

  • Mean age: 66 years
  • Female: 55%
  • BMI: 30 kg/m2
  • SBP: 136
  • DBP: 71
  • Race
    • White: 61.7%
    • Black: 29%
    • American Indian or Alaskan Native: 0.25%
    • Other: 5.8%
  • Hispanic: 13%
  • History of smoking: 46%
  • Cause of CKD:
    • DM:49%
    • HTN: 28.7%
    • Other: 22.5%
  • HTN: 95.8% vs. 93.2% (P=0.03)
  • MI: 16.4% vs. 15.0%
  • CABG: 17.4% vs. 13.5% (P=0.05)
  • PCI: 10.9% vs. 11.9%
  • CHF: 24.4% vs. 22.9%
  • AF: 9.4% vs. 8.6%
  • Stroke: 9.8% vs. 10.0%
  • PAD: 16.4% vs. 16.4%
  • Hgb: 10.1 g/dL
  • Hct: 31.4%
  • Transferrin saturation: 24.9%
  • Ferritin: 173.5 ng/dL
  • CrCl by Cockcroft-Gault: 36.9 mL/min/1.73 m2
  • GFR by MDRD: 27.6 mL/min
  • Albumin: 3.8 g/dL
  • Urine protein:creatinine: 1.6

Baseline Medications

  • ACE inhibitor: 36.8%
  • ARB: 28.3%
  • ACE inhibitor and ARB: 9%
  • ß-blocker: 47.3%
  • Anti-platelet: 43.9%
  • Statin: 52.6%
  • Iron:
    • IV: 2.1%
    • Oral: 26.6%

Interventions

  • Randomized to high-Hgb group (target 13.5 g/dL) vs. low-Hgb group (target 11.3 g/dL)
    • Mean change in Hgb level from baseline to final: 2.5 g/dL vs. 1.2 g/dL (P<0.001)
  • Weekly or biweekly subcutaneous administrations of epoetin alfa
    • Mean dose required to maintain target level: 11,215 vs. 6,276 U/wk
  • Quality of life assessed using following scales, with higher scores indicating better health:
    • Linear Analogue Self-Assessment (LASA): scores range [0-100]
    • Kidney Disease Questionnaire (KDQ): scores range [4-35]
    • Medical Outcomes Study 36-item Short-Form Health Survey (SF-36): scores for each subscale range [0-100]

Outcomes

Comparisons are high-Hgb vs. low-Hgb groups.

Primary Outcome

Composite of death, MI, hospitalization for CHF, or stroke
17.5% vs. 13.5% (HR 1.34; 95% CI 1.03-1.74; P=0.03)

Secondary Outcomes

Death
7.3% vs. 5.0% (HR 1.48; 95% CI 0.97-2.27; P=0.07)
Hospitalization for CHF without renal replacement therapy
9.0% vs. 6.6% (HR 1.41; 95% CI 0.97-2.05; P=0.07)
MI
2.5% vs. 2.8% (HR 0.91; 95% CI 0.48-1.73; P=0.78)
Stroke
1.7% vs. 1.7% (HR 1.01; 95% CI 0.45-2.25; P=0.98)
Any renal replacement therapy
21.7% vs. 18.7% (HR 1.19; 95% CI 0.94-1.49; P=0.15)
Hospitalization for renal replacement therapy
13.8% vs. 11.3% (HR 1.25; 95% CI 0.93-1.68; P=0.13)
Hospitalization for CV causes
32.6% vs. 27.5% (HR 1.23; 95% CI 1.01-1.48; P=0.03)
Hospitalization for any cause
51.6% vs. 46.6% (HR 1.18; 95% CI 1.02-1.37; P=0.03)

Adverse Events

Any serious adverse event
54.8% vs. 48.5% (P=0.02)
Any serious adverse event associated with epoetin alfa
1.5% vs. 0.4% (P=0.05)
CHF
11.2% vs. 7.4% (P=0.02)
MI
1.5 vs. 2.8% (P=0.09)

Criticisms

  • Even though the study population was randomized, the high Hgb group had more comorbid conditions, including a greater percentage of HTN, history of CABG, and severe CHF.
  • The high-Hgb group was not able to reach the target Hgb level despite higher doses of erythropoietin.
  • The trial was open-label, not double-blinded.

Funding

Supported by Ortho Biotech Clinical Affairs and Johnson & Johnson Pharmaceutical Research and Development, and authors had multiple financial disclosures.