Arising from: Agarwal R et al. (2008) Does ferric gluconate lower epoetin requirements in hemodialysis patients with high ferritin levels? Nat Clin Pract Nephrol 4: 418–419

We read with interest Dr Rajiv Agarwal's Practice Point commentary1 on the Dialysis Patients' Response to IV Iron with Elevated Ferritin (DRIVE) II study results2 and his valuable insights into our study. In his commentary, Dr Agarwal criticized the statistical methods employed in DRIVE II, although the methods and analysis were predefined. We compared the end-of-study epoetin doses using an analysis of covariance test (ANCOVA), not an unpaired t-test,2 after deciding that there were no important departures from a normal distribution. As Dr Agarwal recommends, we have run a nonparametric ANCOVA test based on ranks; this test yields a significance value of P=0.002, which is even stronger than the P=0.017 reported in the paper.

Dr Agarwal also commented on our use of the last observation carried forward method.1,2 The use of the last observation carried forward technique is a very standard approach, and any method of data imputation is subject to criticism. We did not use repeated measures because our focus was on the final effect of intravenous iron at 12 weeks, not the journey leading to the end of study.

We concur with Dr Agarwal that longer studies assessing the efficacy and safety of intravenous iron are needed; the assumption that withholding intravenous iron and employing higher doses of epoetin is somehow safer should also be tested. At any given hemoglobin value, higher epoetin doses are associated with a greater risk of death in dialysis patients, and impaired epoetin responsiveness has also been linked with an increased risk of death.3,4 Use of intravenous iron has been shown to improve epoetin responsiveness and reduce epoetin requirements.5,6 The DRIVE and DRIVE II studies now extend that observation to our most difficult anemic dialysis patients.