Improved survival with higher hematocrits: where is the evidence?

The Centers for Medicare and Medicaid Services (CMS) is reviewing Medicare coverage policy for recombinant human erythropoietin (epoetin) therapy. CMS officials are concerned that "Medicare spending on EPO may be higher than necessary without resulting in optimal patient benefit." Approximately 190 end-stage renal disease (ESRD) patients die each day-a mortality rate that has remained essentially unchanged since 1994-despite improvements in the "adequacy of dialysis, vascular access, and anemia management". To date, the evidence cited in support of a survival benefit of epoetin confuses the relationship between treatment response and outcomes with a causal effect of epoetin. A variety of mechanisms may account for a non-causal association between hematocrit and mortality can occur. We conclude that there is no basis for inferring the survival benefits (or detriments) of increasing a patient's hematocrit by adjusting the dosing of epoetin. Furthermore, we note that caution is required in administering large doses of epoetin to unresponsive patients in order to achieve the target hematocrit. A better understanding of the epoetin/survival relationship, well-grounded in science, is needed to provide a basis for CMS to improve its current epoetin policies, and may help to improve patient mortality.