A large number of patients with musculoskeletal injury were studied, in whom large reductions in red-cell mass (about 40 per cent) were usually associated with only slightly decreased values for hemoglobin and hematocrit in peripheral venous blood. This condition is referred to as stress anemia. Using ⁵¹chromium to measure the red-cell volume was the only reliable way to detect deficits in red-cell mass. Calculation of the red-cell volume from the ¹²⁵I iodinated plasma volume and the total-body hematocrit overestimated the red-cell volume; radioactive iodinated albumin was not reliable in estimating the deficits in red-cell mass in these patients.

The pathophysiological adaptations to the anemic hypoxia did not include an increase in either cardiac output or in red-cell production. The patients maintained their central red-cell volumes, but had decreased peripheral red-cell volumes (that is, flow to the extremities and skin). A decrease in affinity of the red cell for oxygen was associated with a significant increase in the peripheral red-cell 2,3 diphosphoglycerate, and this was associated with an increase in systemic arteriovenous difference of oxygen content. With correction of the red-cell-mass deficit by transfusion of washed non-frozen and previously frozen red cells, there was clinical improvement, reflecting an increase in circulation to the extremities.

Because red-cell-mass deficits recurred with or without apparent blood loss, blood-volume measurements were repeated in patients who clinically were not doing well. When washed red cells are used to correct the red-cell-mass deficit, the risks associated with transfusion are minimized.