Hemoglobin and Hematocrit Targets for the Anemia of Chronic Renal Failure: a Systematic Review of Randomized Trials.

Giovanni FM Strippoli, MD ¹, Jonathan C Craig, MD ¹, Carlo Manno, MD ² and Francesco P Shena, MD ².
1. Centre for Kidney Research, Cochrane Renal Group-Children's Hospital, Westmead, NSW, Australia
2. Dept. of Emergency and Organ Transplantation, University of Bari, Bari, Italy.

Anemia affects 60-80% of patients with renal impairment, influences quality of life and induces adaptive cardiovascular changes which are a risk factor for death. Treatment options are blood transfusion, erythropoietin (epo) and darbepoetin. There has been a consistent trend towards higher hemoglobin (Hb) and hematocrit (Htc) targets, but there is still debate whether this is appropriate since higher targets may lead to access thrombosis and hypertension and imply elevated costs for maintenance. We conducted a systematic review to assess the benefit and harms of low (Hb<12 g/dL-Htc<30%) and high (Hb>12 g/dL-Htc>30%) targets in pre and post dialysis patients receiving any treatment for anemia.

The Cochrane Controlled Trials Register, Medline (1966-2002), Embase (1988-2002) and article reference lists were searched for randomized controlled trials. Two reviewers independently assessed studies for inclusion criteria (randomised hemoglobin and/or hematocrit targets in patients with anemia of chronic renal failure) and extracted data. Data were analyzed by Cochrane RevMan and Metaview software (version 4.1) using a random effects model.

From 1665 abstracts retrieved, 43 full text articles were analyzed for eligibility. To date 18 have been included of which 14 relied only on surrogate outcomes (e.g. access thrombosis, seizures, hyperkalemia). Low and high Hb and Htc targets were achieved by different epo doses in 14 trials or by epo and placebo in 4. There was a higher mortality in the high Hb/Htc group (2 trials, 1379 patients, relative risk 1.7, 95% CIs 1.4-2.1, p<0.00001, heterogeneity c2=0.1, p=0.7), and a higher mean arterial pressure (3 trials, 43 patients, weighted mean difference 18.9 mmHg, 95% CIs 15.0-22.8, p<0.00001, heterogeneity c2=92.3, p<0.00001). Further analysis regarding these and other outcomes including quality of life will be possible upon completion of this review. Our preliminary data seem to favour low Hb and Htc targets in the treatment of anemia of chronic renal failure. These conclusions are subject to further ongoing data extraction and more data to be collected from individual authors.

Presented at the American Society of Nephrology 35th Annual Meeting & Scientific Exposition, Philadelphia, Pennsylvania, 30 Oct - 4 Nov 2002.

Correspondence
Giovanni Strippoli
Cochrane Renal Group
Centre for Kidney Research
The Children’s Hospital at Westmead
Locked Bag 4001
Westmead NSW 2145 Sydney Australia
Tel: +61 2 9845 1306
Fax: +6 12 9845 3038
Email1:Giovanni Strippoli
Email2:Giovanni Strippoli