Control of segmental upper airway resistance in patients with obstructive sleep apnea.

We sought to determine if the upper airway response to an added inspiratory resistive load (IRL) during wakefulness could be used to predict the site of upper airway collapse in patients with obstructive sleep apnea (OSA). In 10 awake patients with OSA, we investigated the relationship between resistance in three segments of the upper airway (nasal, nasopharyngeal, and oropharyngeal) and three muscles known to influence these segments (alae nasi, tensor veli palatini, and genioglossus) while the patient breathed with or without a small IRL (2 cmH2O.l-1.s). During IRL, patients with OSA exhibited increased nasopharyngeal resistance and no significant increase in either the genioglossus or tensor veli palatini activities. Neither nasal resistance nor alae nasi EMG activity was affected by IRL. We contrasted this to the response of five normal subjects, in whom we found no change in the resistance of either segment of the airway and no change in the genioglossus EMG but a significant activation of the tensor palatini. In six patients with OSA, we used the waking data to predict the site of upper airway collapse during sleep and we had limited success. The most successful index (correct in 4 of 6 patients) incorporated the greatest relative change in segmental resistance during IRL at the lowest electromyographic activity. We conclude, in patients with OSA, IRL narrows the more collapsible segment of the upper airway, in part due to inadequate activation of upper airway muscles. However, it is difficult to predict the site of upper airway collapse based on the waking measurements where upper airway muscle activity masks the passive airway characteristics.