A carbohydrate (CHO) solution consumed before exposure to hypoxia has been reported to reduce arterial oxygen desaturation at rest. The purpose of this study was to determine whether this effect occurred during exercise and when the CHO load is part of a meal. Eleven male subjects (mean age 20.1 ± 1.8 years, BMI 24.3 ± 2.4 kg m(2)) consumed either a high-CHO (2,340 kJ, 70 % CHO, 12 % protein) or an isoenergetic high-protein (35 % CHO, 48 % protein) breakfast meal 60 min before being exposed to 15 min of hypoxia (F(I)O(2) = 13.5 %) followed by 30 min of exercise in hypoxia (60 % of VO₂max). Saturation of oxygen via a pulse oxymeter (SpO(2)), ventilatory parameters, substrate oxidation, interstitial glucose concentrations, and heart rate variability (HRV) were monitored continuously during the whole session. Results showed no effect from the type of meal on SpO(2) at rest but a 3.1 ± 0.4 % reduction of desaturation during exercise (P < 0.005) compared to the high-protein version. This was associated with higher levels of ventilation (P < 0.05) and CO(2) production (P < 0.01). Glucose oxidation was higher after the high-CHO than the high-protein breakfast over the whole session (+19.4 + 4.0 %, P < 0.0001), whereas the interstitial glucose levels were increased only at rest (P < 0.001). HRV indices were not different between conditions. In conclusion, a high-CHO meal consumed prior to moderate exercise in hypoxia condition reduced oxygen desaturation compared to a high-protein meal.