SESSION TITLE: Consequences of Sleep Disorders
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Tuesday, October 28, 2014 at 02:45 PM - 04:15 PM
PURPOSE: In a well defined cohort of patients referred for the evaluation of sleep-disordered breathing, we examine the association of opioid use, sleep-disordered breathing, and all-cause mortality.
METHODS: A retrospective analysis of patients was performed on patients referred for overnight polysomnography at three Veterans Administration sleep centers from 01/01/00 to 12/31/04 (n = 3,300). Inclusion criteria were referral for suspected sleep-disordered breathing, at least one history and physical documented in the chart, and at least two hours of attended sleep monitoring using full polysomnography. Exclusion criteria were referral for reasons other than the evaluation of sleep-disordered breathing, a study performed with an uncovered tracheostomy, and a study performed while the patient was using airway pressure of any form. A final analytic cohort of 2,000 was obtained and refined to 344 patients who have at least moderate sleep apnea (AHI > 15) and are on opioid medications. Patients were followed until 12/31/10 for a minimum of three years followup and a maximum of 10 years followup. Risk adjustment was done with the Charlson Comorbidity Index. Independent variables were assessed against the rate of sleep-disordered breathing in a dose response fashion, the rate of sleep-disordered breathing in a binary fashion of opioid use, and all-cause mortality.
RESULTS: The is no association between opioid use and severity of sleep-disordered breathing in a dose response fashion (Spearman's correlation coefficient -0.081, p value 0.19). There is an increase in mortality when opioid use is analyzed as an unadjusted variable (OR 1.53, 95% CI 1.12-2.09). The effect is attenuated with adjustment by sleep apnea (OR 1.52, 95% CI 1.07-2.16) and attenuated further with adjustment by sleep apnea and the Charlson Comorbidity Index (OR 1.37, 95% CI 0.95-1.99).
CONCLUSIONS: Opioids are not associated with severity of sleep-disordered breathing. Opioid use is associated with all-cause mortality, and much of this association is explained by coexisting medical comorbidity. Moderate to severe sleep apnea (AHI > 15) is an independent predictor of mortality even after adjustment for opioids and even after adjustment for the Charlson Comorbidity Index.
CLINICAL IMPLICATIONS: Whatever risk opioid use confers for death in patients with sleep apnea cannot be explained by sleep apnea alone. It may be explained by an increased prevalence of known risk factors for morbidity and mortality in patients who take opioids and have sleep apnea.
DISCLOSURE: The following authors have nothing to disclose: Husham Sharifi, Megan Sands, Selim Bernardo, Kingman Strohl, Jared Ferguson, William Jimenez, Li Qin, Adam Bennett, Kervin Doctor, Frederick Struve, Dawn Bravata, Henry Yaggi
No Product/Research Disclosure Information