Obstructive nonapneic respiratory events (ONAREs, i.e., obstructive hypopneas [OHs] and respiratory effort related arousals [RERAs]) are clinically important as producing sleep fragmentation but are much more difficult to detect and classify than obstructive apneas. We characterized 1,061 ONAREs in 15 unselected patients with moderate sleep apnea-hypopnea syndrome (OSAHS) according to the 1999 American Academy of Sleep Medicine (AASM) Task Force recommendations concerning accurate skills for respiratory measurements (i.e., pneumotachograph and esophageal pressure [Pes]) and definitions of respiratory events. OHs were defined as > or = 50% decrease in flow or < 50% but > or = 30% decrease in flow associated with either a desaturation of > or = 3% or a cortical arousal. RERAs corresponded to a less than 30% decrease in flow associated with an arousal. OHs represented a large majority of the events (79.9%, n = 848). Among the events demonstrating a 30 to 50% reduction in flow (n = 392), the occurrence of a cortical arousal permitted the classification of 246 events as OHs (62.8%). RERAs represented only 5.3% of the events. Finally 14.8% of the events were classified as indeterminate owing to a < 50% and > or = 30% airflow decrease without arousal or desaturation or an airflow decrease less than 30% without arousal. The same level of DeltaPes was observed at the end of OHs and RERAs (21.9 +/- 5.5 versus 18.9 +/- 5.7 cm H(2)O respectively [NS]) whereas the reduction in flow was as expected, higher for OH (57.9 +/- 10.7 versus 21.3 +/- 4.9%). In a population of moderate OSAHS, OH represented the dominant type of ONAREs when RERAs should be considered as specific but relatively rare respiratory events. The occurrence and the recognition of a microarousal represented the key associated factor for classifying respiratory events as hypopneas. The high rate (15%) of unclassified events demonstrated some limitations in the AASM Task Force recommendations for definitions of respiratory events. Similar levels of DeltaPes found at the end of RERAs and OHs, although the reduction in flow was higher for OHs, suggest that different levels of collapsibility can exist throughout the night in a given patient.