SESSION TITLE: Sleep Disorders
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Sunday, October 25, 2015 at 07:30 AM - 08:30 AM
PURPOSE: Zolpidem and eszopiclone are non-benzodiazepine sedative hypnotics (NBSH) that improve sleep onset latency and efficiency. Several studies have shown that neither drug affects the apnea-hypopnea index (AHI) in unselected patients undergoing polysomnography (PSG). In patients with the ‘low arousal threshold phenotype’ of obstructive sleep apnea (OSA), NBSHs may actually improve the AHI.
METHODS: We abstracted demographic and PSG data for patients undergoing studies at our sleep center. It is common practice for physicians to prescribe a NBSH to be used during PSG, and administration of drug is recorded by our technicians. We used a recently defined clinical definition, > 58.3% hypopneas, O2 nadir > 82.5% and mild-to-moderate OSA on PSG, to identify the low arousal phenotype of OSA. Patients who met all three criteria were labeled as having this phenotype.
RESULTS: A total of 583 patients had data available for analysis. Mean age was 42.2±10.1 and the median AHI was 13.2 (6.5-26.0). There were 437 (76.9%) patients who received a NBSH (100 eszopiclone and 328 zolpidem) and 131 (23.1%) who took no medication prior to PSG. Patients who took a NBSH had significantly more time in stage I (p<0.001) and II sleep (p<0.001) and significant less time in stage III (p<0.001) and REM (p<0.001). There were 383 (88.5%) patients who met the low arousal threshold phenotype. After adjusting for pre-test probability using a score that incorporated snoring, hypertension, age and neck circumference, administration of a NBSH for the night of the study was not associated with the low arousal phenotype (OR 0.59, 95% CI: 0.23-1.25; p=0.15).
CONCLUSIONS: Administration of a NBSH did not decrease the likelihood of having a low arousal phenotype for OSA. Prescribing a NBSH for the night of PSG should not reduce the sensitivity for detecting OSA. Our study was limited in that the patients had an average AHI that was very low which likely hurt the discriminatory capacity of the variables used to define the low arousal threshold.
CLINICAL IMPLICATIONS: NBSHs should not reduce the PSG sensitivty for detecting OSA. Because NBSHs did not reduce the frequency of the low arousal phenotype, our data suggests they may not be effective treatment for patients with this OSA phenotype.
DISCLOSURE: The following authors have nothing to disclose: Camille Costan-Toth, Derek Forsthoefel, Patrick Smith, William Londeree, Karen Sheikh, Edward Bridges, Aaron Holley
No Product/Research Disclosure Information