Correlation between genioglossal and diaphragmatic responses to hypercapnia during sleep.

Oropharyngeal patency during sleep is dependent on the dilating force of the genioglossus, the main tongue protrusor muscle. We measured the ventilatory (Vl), diaphragmatic (EMGdi), and genioglossal (EMGgg) electromyographic responses to CO2 in awake and sleeping goats; delta Vl/delta PETCO2 decreased from awake (0.85 +/- 0.087 L/min/mm Hg) to NREM sleep (0.69 +/- 0.069) to REM sleep (0.57 +/- 0.078, p less than 0.005 versus awake). There were minimal decreases in delta EMGdi/delta PETCO2 and EMGdi at a PETCO2 of 55 mm Hg from awake to NREM, but a significant decrease in EMGdi at a PETCO2 of 55 mm Hg during REM sleep (p less than 0.025 versus NREM). Inspiratory EMGgg was only present above a PCO2 threshold, which was similar for each state (49.3 +/- 2.3 mm Hg PETCO2 awake, 48.8 +/- 2.4 during NREM, 49.5 +/- 2.5 during REM), and delta EMGgg/delta PETCO2 tended to be less during both sleep states compared with that while awake (p = 0.10). At any PCO2, inspiratory EMGgg was markedly inhibited during REM sleep when rapid eye movements were present (phasic REM). We conclude that there is disproportionate inhibition of the genioglossus relative to the diaphragm at low CO2 tensions and at any PCO2 during phasic REM sleep. This imbalance may predispose the upper airway to inspiratory occlusion during sleep.