Considerable controversy continues over the optimal hematocrit target
for dialysis patients being treated with recombinant erythropoietin. R
ecent short-term studies have demonstrated a significant improvement i
n brain function when hematocrit is 42% compared with when it is 31%.
Questions regarding the safety of long-term maintenance of a normal he
matocrit have been raised; however, this is in part because of the ear
ly termination of the Normal Hematocrit Cardiac Trial, in which there
was an increased occurrence of death or nonfatal myocardial infarction
in patients randomly assigned to the normal hematocrit group. For the
present, a target hematocrit of 36% seems reasonable and safe.