The hematocrit (Hct) determines the oxygen carrying capacity of blood, but also increases blood viscosity and thus flow resistance. From this dual role the concept of an optimum Hct for tissue oxygenation has been derived. Viscometric studies using the ratio Hct/blood viscosity at high shear rate showed an optimum Hct of 50–60% for red blood cell (RBC) suspensions in plasma. For the perfusion of an artificial microvascular network with 5–70 μ m channels the optimum Hct was 60–70% for high driving pressures. With lower shear rates or driving pressures the optimum Hct shifted towards lower values. In healthy, well trained athletes an increase of the Hct to supra-normal levels can increase exercise performance. These data with healthy individuals suggest that the optimum Hct for oxygen transport may be higher than the physiological range (35–40% in women, 39–50% in men). This is in contrast to clinical observations. Large clinical studies have repeatedly shown that a correction of anemia in a variety of disorders such as chronic kidney disease, heart failure, coronary syndrome, oncology, acute gastrointestinal bleeding, critical care, or surgery have better clinical outcomes when restrictive transfusion strategies are applied. Actual guidelines, therefore, recommend a transfusion threshold of 7–8 g/dL hemoglobin (Hct 20–24%) in stable, hospitalized patients. The discrepancy between the optimum Hct in health and disease may be due to factors such as decreased perfusion pressures (low cardiac output, vascular stenoses, change in vascular tone), endothelial cell dysfunction, leukocyte adhesion and others.