Interaction between hypoxia and training on NIRS signal during exercise: contribution of a mathematical model.

Acute exposure to hypoxia provokes a decrease in peak oxygen consumption (V˙O2,peak). At and above 4000m, the decrease in V˙O2,peak is greater than expected from the decrease in arterial oxygen content (CaO2) suggesting the participation of other factors. We hypothesized that O2 transfer within the active muscle may play a role. Therefore we used Near Infra Red Spectroscopy (NIRS) to assess oxy (O2Hb) and deoxyhemoglobin (HHb) concentration in the vastus lateralis of trained athletes (TA) and untrained subjects (US) exercising at various inspired oxygen pressure (PIO2=149.4, 131.4, 107.3 and 87.0mmHg). A mathematical model has been developed to compute: (i) the pulmonary (Kp) and muscular (Ktm) O2 diffusion coefficients and (ii) the proportion of arteriolar:capillary:venous blood participating in the NIRS signal at every exercise intensity from rest to peak exercise in the normoxic and various hypoxic conditions. In TA, O2Hb decreased near maximal exercise at 2500 and 4000m, while in US, altitude had no effect. In normoxia O2Hb was higher in TA than in US, the difference disappearing in hypoxia. Ktm increased linearly with workload and altitude and was higher in TA than US while Kp plateaued near maximal exercise, which was consistent with athletes’ greater decrease in CaO2. The greater participation of arterial blood in the NIRS signal in TA at altitudes account for their higher O2Hb values as well as the greater decrease they underwent in hypoxia. At 4000m, athletes loose their advantages of adaptation to training due to a reduced arterial content, and both from NIRS variables and model output, characteristics of O2 transfer of TA converge toward those of US.