Arousal threshold to respiratory stimuli in OSA patients: evidence for a sleep-dependent temporal rhythm.

It has recently been described that the maximal respiratory effort developed at the end of an apnea (Pesmax)--which is regarded as an index of arousal threshold in patients with obstructive sleep apnea syndrome (OSA)--increases progressively during the night, probably as a consequence of associated sleep fragmentation. In order to find out whether the nocturnal trend of Pesmax may be more influenced by a sleep-dependent circadian rhythm than by sleep fragmentation, we revised the polygraphic recordings of 37 patients in whom obstructive apneas were recorded for at least 7 hours. In 15 of these patients, analysis was made for eight hours of the night. During each hour we analyzed at least 7 obstructive apneas, in which we measured the minimal esophageal pressure at the start of the apnea, the maximum value recorded at the end of the apnea (Pesmax), the difference from the minimum to the maximum (delta Pes), and the rate of increase in esophageal pressure (RPes). As indices of sleep fragmentation, we defined the number of arousals, awakenings and sleep state transitions. In the group of patients as a whole, we found a trend toward a gradual increase for apnea duration (F = 98.8, p < 0.001) and Pesmax F = 31.6, p < 0.001) which was significant from the first to the last hour. The time-dependent evolution of apnea duration and Pesmax showed that the rise in these two variables peaked during the first 3 hours of sleep, followed by a plateau and a decrease in the last hour of the night. This temporal profile was more evident when the analysis was available for 8 hours. No significant changes across the night were found for nocturnal hypoxemia and number of arousals. Considering the slope of Pesmax changes across the night, we saw that neither the apnea+hypopnea index nor the indices of sleep fragmentation affected the nocturnal trend. The present data demonstrate the presence of a nocturnal trend in arousal threshold in OSA patients independent of sleep fragmentation. The biphasic evolution of the arousal threshold may be caused by factors that influence the circadian and homeostatic processes.