Though red blood cells are essential to deliver oxygen to tissue, abnormally high levels can have deadly consequences. High hematocrit (Hct)-the percentage of total blood volume occupied by red blood cells (RBCs)-can occur in arterial flow due to a variety of causes. Examples include: dehydration, polycythemia vera, and exogenous use of recombinant human erythropoietin. In this phenomenological study, we present a simplified quantitative analysis of the physiological effects of increased hematocrit, especially for small arteries. In particular, we are primarily interested in the potential relation between the exogenous use of recombinant human erythropoietin by athletes (blood 'doping') and the not insignificant numbers of sudden heart attacks suffered by these otherwise superbly conditioned contestants. We proceed as follows: An expression for the variation of the viscosity of blood with both shear rate and hematocrit is obtained using the low and high shear experimental measurements of Chien and Cinar, respectively. This is done in a two-step process: (i) curve fits of shear rate dependent viscosity for hematocrits ranging from 40% to 80% are derived with resulting R-squared values greater than 0.99; (ii) viscosity is shown to be well-represented as an exponential function of shear rate for various hematocrit levels. The resulting relationship is used, in a simplified analysis, to calculate velocity profiles in a generic coronary artery as a non-linear function of viscosity for various hematocrit levels. Using these relationships, increasing hematocrit is shown to dramatically decrease and flatten velocity profiles in a small artery. The resulting velocity profiles of slower flow with greater area of low shear rate due to increased hematocrit makes likely a corresponding increased risk of thrombus formation. (C) 2009 Elsevier Ltd. All rights reserved.