Inhibitory effects of hypocapnia on rapid eye movement (REM) sleep
Sleep disturbances at altitude are accompanied by reductions in end tidal CO2 that are caused by the ventilatory response to hypoxia. To determine whether hypocapnia could contribute to the sleep disturbance, the effects of different levels of CO2 on REM sleep both during apnea caused by mechanical hyperventilation and during hypoxia in spontaneous breathing were studied. Adult cats were prepared for chronic recordings of sleep and respiratory parameters (airflow, tidal CO2, intratracheal pressure and diaphragmatic EMG). The animals were intubated through a tracheal fistula, allowing a connection to a ventilator or allowing for the adjustment of the FIO2. Computerized control of CO2 was achieved by injecting CO2 into the inspired air and thus maintaining a selected end tidal or inspired CO2 level. Four different protocols in five intact, unanaesthetized adult cats were used. In the first two protocols the animals were mechanically hyperventilated to apnea and in the last two protocols they were allowed to breathe spontaneously during three-hour recording periods. In the first protocol, CO2 was cycled continuously from extreme hypocapnia (end tidal level ~ 3.5%) to just above the eupneic level (~ 5%). In the second protocol, CO2 was held constant at 65%, 75%, 85% and 95%) of eupneic levels (hypocapnic normoxia). In the third protocol, two levels of hypoxia (FIO2 = 0.10 & 0.15) were studied both individually (hypocapnic hypoxia) and with CO2 added back to maintain isocapnia (isocapnic hypoxia). The fourth protocol was similar to the third except the inspired CO2 was maintained constant instead of endtidal CO2 levels. Spontaneously breathing animals were used as the control. The results obtained from observations of more than 300 REM periods revealed that, irrespective of the fraction of inspired oxygen, REM sleep was least at lower levels of CO2. In both hypocapnic normoxia and hypocapnic hypoxia, REM sleep was reduced both in time and episodes as the CO2 levels decreased. But in isocapnic hypoxia both REM time and episodes were not significantly affected. NREM sleep was significantly affected at 10% hypoxia but not at 15% hypoxia. Wakefiilness was not significantly affected by hypocapnia in our studies. Hence the current results suggest that the sleep disruption at high altitude may be fully or in part caused by hypocapnia associated with hypoxia.