The effects of changing hematocrit (Hct) on the rate of circulatory oxygen (O2) delivery were modeled analytically to describe transfusion of 0.5-3.0 units of packed red blood cells (pRBC, 300 ml/unit, 60% Hct) to anemic patients. In our model, Hct affects O2 delivery to the microcirculation by changing blood O2 carrying capacity and blood viscosity, which in turn affects blood flow velocity and, therefore, O2 delivery. Changing blood velocity impacts the O2 delivery by affecting the oxygen diffusive losses as blood transits through the arteriolar vasculature. An increase in Hct has two opposite effects: it increases the blood O2 carrying capacity and decreases the flow velocity. This suggests the existence of an optimal Hct that maximizes O2 delivery. Our results show that maximal O2 delivery occurs in the anemic range, where Hct is less than 39%. Optimal blood management is associated with transfusing enough units up to reaching maximal O2 delivery. Although somewhat complex to implement, this practice would result in both substantial blood savings and improved O2 delivery.