Exercise hemorheology: Moving from old simplistic paradigms to a more complex picture.

Classic studies on exercise hemorheology evidenced that blood fluidity is impaired during exercise (short term exercise-induced hyperviscosity) and is improved as a result of regular exercise practice (hemorheologic fitness). Extensive description of these events led to the concepts of "the triphasic effects of exercise", "the paradox of hematocrit", and "the hemorheological paradox of lactate". However, some results obtained in training studies do not fit with this classical picture and cannot be explained by a simplistic paradigm based on the Hagen-Poiseuille law. Taking into account the non-linearity of the effects of viscosity factors on blood flow and oxygen delivery helps to elaborate another picture. For example, moderately high values of hematocrit and erythrocyte rigidity induced by high intensity exercise are likely to trigger a physiological vasodilation improving circulatory adaptation (rather than limiting performance as was previously assumed). This may apply to the acute rise in red cell rigidity observed during strenuous exercise, and also to the paradoxical rise in hematocrit or red cell rigidity observed after some training protocols and that did not fit with the previous (simplistic) paradigms. The "healthy primitive lifestyle" hypothesis assumes that evolution has selected genetic polymorphisms leading to insulin resistance as an adaptative strategy to cope with continuous low intensity physical activity and a special alimentation based on lean meat and wild herbs (i.e., moderately high in protein, rich in low glycemic index carbohydrates, and poor in saturated fat). We propose here that this model may help to explain on an evolutionary perspective these apparently inconsistent findings. The pivotal explanation is that the true physiological picture would be that of an individual whose exercise and nutritional habits are close from this lifestyle, both sedentary subjects and trained athletes representing situations on the edge of this model.