Arterial oxygen desaturation kinetics during apnea

Purpose: To quantify the rate of arterial oxygen desaturation during apnea in freedivers. Methods: Ten freedivers and ten controls undertook five maximal face immersion apneas in 10°C water separated by 2 min of recovery. Electrocardiogram (ECG), blood pressure, and pulse oximetry were recorded continuously. Peripheral blood flow was measured by calf plethysmography every 30 s, and venous blood samples were collected at rest and after apneas 1, 3, and 5. The blood was analyzed for hematocrit (Hct), lactate, and hemoglobin (Hb) concentration. The arterial oxygen saturation (SaO2) data were curve fitted with both a sigmoid and two-slope continuous function. Results: Apnea duration increased with successive attempts, with freedivers achieving significantly longer maximal apneas (trained 246 ± 44 s, untrained 129 ± 39 s, P < 0.001). Compared with controls, freedivers displayed a significant change from baseline in heart rate (trained -27.2 ± 9.5 bpm, untrained -19.7 ± 9.3 bpm, P < 0.001) and mean arterial pressure (MAP) (trained 48 ± 20.7 mm Hg, untrained 37 ± 10.0 mm Hg, P = 0.002), but no difference existed in peripheral blood flow, Hct, lactate, or Hb. The maximal slope of the SaO2 sigmoid curve was not significantly different between the groups (trained -0.16 ± 0.05%·s -1, untrained -0.15 ± 0.06%·s-1, P = 0.26), but the ÄSaO2/Ät obtained from the two-slope continuous model indicated that 85% of the variance in the freedivers ÄSaO 2/At could be explained by the apnea-induced bradycardia, preapnea vital capacity, and Hb concentration. Conclusions: The sigmoidal function provided no quantifiable difference in the rate of oxygen desaturation. The two-slope continuous method, however, indicated that freedivers who had larger oxygen stores and produced the largest bradycardia were able to slow the ÄSaO2/Ät to two to three times that of the least marked response.