Table 2.
Obstructive sleep apnea endotypes
Characteristic | Placebo | Pimavanserin | P value |
---|---|---|---|
Loop gain* | 0.59 (0.52–0.65) | 0.57 (0.45–0.69) | 0.570 |
Ventilatory response to arousal, %VEUPNEA* | 17.9 (13.3–22.7) | 16.5 (7.3–25.7) | 0.514 |
Arousal threshold, %VEUPNEA† | 148.7 (115.5–181.8) | 158.6 (102.8–214.5) | 0.379 |
VPASSIVE, %VEUPNEA† | 104.3 (85.9–122.7) | 90.6 (54.8–126.4) | 0.120 |
VMIN, %VEUPNEA† | 82.0 (69.0–94.9) | 89.5 (61.1–118.0) | 0.329 |
VACTIVE, %VEUPNEA† | 114.5 (94.2–134.8) | 93.6 (55.0–132.2) | 0.026 |
Definition of abbreviations: VACTIVE = ventilation when the pharyngeal muscles are relatively active; VEUPNEA = baseline ventilation obtained from periods of resting wakefulness; VMIN = ventilation at the lowest drive levels; VPASSIVE = ventilation when the pharyngeal muscles are passive.
Data are expressed as mean (confidence interval). Data for VPASSIVE, VMIN, and VACTIVE were derived from the sigmoidal transformation function to handle known floor and ceiling effects. Of note, the difference in arousal threshold between the nights was the primary outcome of the study, whereas the difference in the other endotypes between the study nights was classified as secondary outcomes in our prespecified statistical analysis plan. Values are the percent of stable breathing during sleep (VEUPNEA).
Paired data were not available for two patients (i.e., ventilatory response quantification was possible for only one out of the two nights in two different participants).
Paired data were not available for one patient (i.e., pharyngeal physiology and arousal threshold quantification was possible for only one out of the two nights).