Palatal muscle electromyogram activity in obstructive sleep apnea

Eight subjects (5 men, 3 women, ages 27 to 55) with obstructive sleep apnea syndrome (OSAS) were studied to quantify and compare electromyographic (EMG) activity of levator veli palatini (LVP) and palatoglossus (PG), two velopharyngeal muscles, and genioglossus (GG) during obstructive apnea cycles in non-rapid eye movement (NREM) sleep. EMG activity of three successive preapneic breaths, first and last apneic efforts, and three successive postapneic breaths was quantified for each muscle as peak phasic inspiratory EMG normalized as percent activity of the last preapneic breath. In all subjects, apnea onset coincided with simultaneous inspiratory EMG nadir of all three muscles (LVP = 63 +/- 40%, PG = 74 +/- 53%. GG = 83 +/- 48%. mean +/- SD activity of last preapneic breath). Apnea resolution did not occur until inspiratory EMG of all three muscles simultaneously reached maximal activity, at levels significantly greater than preapneic activity as well as activity of the last preapneic effort (LVP = 215 +/- 205%, PG = 227 +/- 240+, GG = 235 +/- 202%, mean +/- SD activity of last preapneic breath, p < 0.05, Fisher's partial least-squares difference [PLSD] test for each muscle). The presence or absence of electroencephalographic arousal at apnea resolution did not influence these patterns of EMG activity. Inspiratory recruitment of velopharyngeal as well as oropharyngeal muscles appears to be associated with upper airway patency during sleep in patients with OSAS.