Ventilatory Drive Withdrawal Rather Than Reduced Genioglossus Compensation as a Mechanism of Obstructive Sleep Apnea in REM Sleep

RATIONALE: REM sleep is associated with reduced ventilation and greater obstructive sleep apnea (OSA) severity than non-REM (nREM) sleep for reasons that have not been fully elucidated. OBJECTIVES: Here, we use direct physiological measurements to determine whether the pharyngeal compromise in REM sleep OSA is most consistent with 1) withdrawal of neural ventilatory drive or 2) deficits in pharyngeal pathophysiology per se (i.e., increased collapsibility and decreased muscle responsiveness). METHODS: Sixty-three participants with OSA completed sleep studies with gold standard measurements of ventilatory “drive” (calibrated intraesophageal diaphragm EMG), ventilation (oronasal “ventilation”), and genioglossus EMG activity. Drive withdrawal was assessed by examining these measurements at nadir drive (first decile of drive within a stage). Pharyngeal physiology was assessed by examining collapsibility (lowered ventilation at eupneic drive) and responsiveness (ventilation–drive slope). Mixed-model analysis compared REM sleep with nREM sleep; sensitivity analysis examined phasic REM sleep. MEASUREMENTS AND MAIN RESULTS: REM sleep (⩾10 min) was obtained in 25 patients. Compared with drive in nREM sleep, drive in REM sleep dipped to markedly lower nadir values (first decile, estimate [95% confidence interval], −21.8% [−31.2% to −12.4%] of eupnea; P