Exercise-induced changes in hematocrit and hematocrit/viscosity ratio in male rugby players

We investigated whether the concept of hematocrit/viscosity (h/η) ratio explains the "paradox of hematocrit in athletes", by calculating a "theoretical optimal hematocrit" (i.e., associated with the higher h/η value predicted with Quemada's equation from plasma viscosity, and erythrocyte rigidity index) before and after exercise. 14 rugby players (19-31 yr; weight 65.8-109.2 kg; height 1.7-1.96 m; BMI 21.7-33.1 kg/m2) underwent a standardized submaximal exercise session on cycloergometer corresponding to 225 kjoules over 30 min. The rheologic response to exercise was measured with the MT90 viscometer and the Myrenne aggregometer. After exercise there was an increase in whole blood viscosity (p < 0.05) and hematocrit (p < 0.005) and a decrease in h/η ratio (from 14.7±0.34 to 12.9±0.37, p < 0.005). There was an increase in viscometric RBC rigidity indexes "Tk" and "k" in 9/14 subjects. Predicted and actual h/η are fairly well correlated (preexercise r = 0.998, p < 0.001; postexercise r = 0.985 p < 0.001) but actual h/η was lower than predicted (preexercise p = 0.005; postexercise p = 0.02). This discrepancy between predicted and measured hematocrit was not correlated to dehydration or plasma viscosity but was correlated to red cell rigidity (r = 0.774, p < 0.01) and its exercise-induced change (r = 0.858, p < 0.01). This study suggests that h/η, although it is not directly correlated to parameters of exercise performance, is precisely regulated during exercise according to the classic concept of "viscoregulation", and that the prediction of the theoretical optimal values of h/η and hematocrit by models may help to interpret the actual values of these parameters. However, these models need to be more extendedly tested and improved.