The effects of nonbenzodiazepine hypnotics on the quality of polysomnography (PSG) in patients with suspected sleep-disordered breathing (SDB) is unclear. We compare the effects of eszopiclone and ramelteon on PSG quality and efficacy of continuous positive airway pressure (CPAP) titration.
Patients (18–65 yrs) with suspected SDB who met inclusion criteria and referred for initial split PSG were enrolled. In a randomized, double-blind fashion, patients received eszopiclone 3mg, ramelteon 8mg, or placebo thirty minutes prior to PSG. Patients completed an Epworth Sleepiness Scale (ESS) the night of the PSG. We compared sleep latency, total sleep time (TST), sleep efficiency, wake after sleep onset (WASO), sleep architecture, and residual apnea-hypopnea index (AHI) in patients initiated on CPAP. Poor quality studies were defined as those with total sleep time < 120 minutes, sleep efficiency < 70%; and in those underwent CPAP titration, either CPAP intolerance or residual AHI > 5/ hr.
26 (4 female) patients (n=9 eszopiclone; n=9 ramelteon; n=8 placebo) completed the study. Average age, body mass index, ESS, and baseline AHI were similar for all groups. Average sleep latency tended to be shorter in eszopiclone and ramelteon groups, 16.8 and 18.9 min, respectively, than with placebo of 47 min. Average total sleep time, sleep efficiency and WASO were no different for all groups. Sleep architecture was also similar with a trend for shorter stage N3 in the ramelteon (1.2 %) than in the eszopiclone (5.0 %) and placebo (7.5 %) groups. The proportion of patients undergoing CPAP titration and the residual AHI at optimal CPAP pressure was similar for all groups.
In patients with suspected SDB, eszopiclone or ramelteon did not improve sleep efficiency, total sleep time, WASO, or residual AHI at optimal CPAP pressure. Sleep latency tends to be shorter with both hypnotics than with placebo. Stage N3 tends to be low with ramelteon.
Eszopiclone or ramelteon does not improve PSG quality or efficacy of CPAP titration in patients with suspected SDB.
Sterling Malish, No Financial Disclosure Information; No Product/Research Disclosure Information