Sleep-disordered breathing and chronic opioid therapy.

OBJECTIVE To assess the relation between medications prescribed for chronic pain and sleep apnea. DESIGN An observational study of chronic pain patients on opioid therapy who received overnight polysomnographies. Generalized linear models determined whether a dose relation exists between methadone, nonmethadone opioids, and benzodiazepines and the indices measuring sleep apnea. SETTING A private clinic specializing in the treatment of chronic pain. PATIENTS Polysomnography was sought for all consecutive (392) patients on around-the-clock opioid therapy for at least 6 months with a stable dose for at least 4 weeks. Of these, 147 polysomnographies were completed (189 patients declined, 56 were directed to other sleep laboratories by insurance companies, and data were incomplete for seven patients). Available data were analyzed on 140 patients. OUTCOME MEASURES The apnea-hypopnea index to assess overall severity of sleep apnea and the central apnea index to assess central sleep apnea. RESULTS The apnea-hypopnea index was abnormal (> or =5 per hour) in 75% of patients (39% had obstructive sleep apnea, 4% had sleep apnea of indeterminate type, 24% had central sleep apnea, and 8% had both central and obstructive sleep apnea); 25% had no sleep apnea. We found a direct relation between the apnea-hypopnea index and the daily dosage of methadone (P = 0.002) but not to other around-the-clock opioids. We found a direct relation between the central apnea index and the daily dosage of methadone (P = 0.008) and also with benzodiazepines (P = 0.004). CONCLUSIONS Sleep-disordered breathing was common in chronic pain patients on opioids. The dose-response relation of sleep apnea to methadone and benzodiazepines calls for increased vigilance.