OXYGEN DESATURATION FOLLOWING VOLUNTARY HYPERVENTILATION IN NORMAL SUBJECTS

To investigate the severity of oxygen desaturation following voluntary hyperventilation (VHV) in normal subjects and its possible relation t o chemoresponsiveness, we examined respiration following VHV in 16 nor mal male subjects. Monitoring was performed according to the standard polysomnography protocol including measurements of arterial oxygen sat uration (Sa(O2)) and transcutaneous PCO2 (PtcCO(2)). This subjects hyp erventilated voluntarily for 3 min, and were then observed for more th an 15 min. They hyperventilated again for another 3 min, and were foll owed again for more than 15 min. Eleven subjects fell into non-REM sle ep after VHV, and their mean lowest Sa(O2) was 67.6 +/- 13.0% (n = 15 trials in 11 subjects, mean +/- SD). Falling asleep during hypocapnia caused desaturation, and periodic breathing was invariably observed so on after. The difference between the Ptc(CO2) during non-REM sleep wit h stable breathing and the Ptc(CO2) when the Sa(O2) was 90% following VHV was defined as the Delta Ptc(CO2) (90). The Delta Ptc(CO2) (90) an d hypoxic ventilatory response (HVR) were positively and significantly correlated (r = 0.73, p < 0.01). While the subjects were awake, the m ean lowest Sa(O2) was 73.5 +/- 17.4% (17 trials in 12 subjects). Remai ning awake induced oxygen desaturation in some subjects but not in oth ers. In one subject, desaturation during the waking state was caused b y hypoventilation, not by central apnea. In the seven subjects whose r espiration following VHV was monitored during the waking state in one trial and during the sleeping state in another trial, plots of the Ptc (CO2)-Sa(O2) relationship for the waking state were generally position ed above those made for the sleeping state. We conclude that during hy pocapnia after hyperventilation, (1) desaturation invariably occurs du ring sleep, and may even occur in the waking state, (2) the hypoxic ve ntilatory drive contributes to an earlier recovery from posthyperventi lation desaturation during the sleeping state, and (3) behavioral cont rol of respiration may not fully compensate for the loss of chemical r espiratory control in some instances in the waking state.