Towards erythropoietin equations that estimate oxygen delivery rather than static hemoglobin targets.

Although we have known since the 19th century that oxygen tension affects erythrocyte production, we have only recently begun to understand many subtleties of erythropoietin physiology. The unanticipated increase in mortality associated with erythropoietin use found in recent randomized studies is prompting a reassessment of static hemoglobin targets. Hemoglobin levels in dialysis patients do not correlate with endogenous erythropoietin production and may be related to differences in oxygen delivery resulting from shifts in the oxygen-hemoglobin dissociation curve. The time may have arrived to develop more physiologic targets such as oxygen delivery that would mimic the natural response to hypoxia. There are several equations that already exist that can compensate for the effects of the concentration of inorganic and organic phosphates as well as pH, carbon dioxide, and temperature on the delivery of oxygen. However, since the shape and dispersion of the oxygen-hemoglobin dissociation curve may actually change in different disease states, more work is needed.