Efficacy of Antidepressants in the Treatment of Obstructive Sleep Apnea Compared to Placebo. A Systematic Review with Meta-Analyses.

To establish the efficacy of oral antidepressants compared to placebo in improving obstructive sleep apnea (OSA) as measured on a polysomnography study. Secondary outcomes included self-reported sleepiness. Authors identified prospective randomized placebo-controlled studies from MEDLINE through PubMed, Web of Science, the Cochrane Library and EMBASE up to February 2019 in English language. Antidepressants included tricyclic antidepressants (TCA), tetracyclic antidepressants (TeCA), selective serotonin reuptake inhibitors (SSRI), and serotonin receptor modulators (SRM). Studies were assessed for inclusion and exclusion criteria, as well as risk of bias based on the Cochrane handbook. The initial search yielded 254 unduplicated references ultimately reduced to 8 relevant studies, in which 198 OSA participants were included. Patients with an average baseline AHI of 26.7 events/hour taking 15-45mg mirtazapine had a statistically significant reduction in apnea-hypopnea index compared to placebo by -10.5 events/hour (p<0.001), apnea index by -3.6 events/hour (p=0.001) and hypopnea index by -5.9 events/hour (p=0.037). In one study, patients taking 100mg trazodone 1 night improved significantly in AHI compared to placebo group (p<0.001). Arousal index, sleepiness, and sleep efficiency were not statistically significantly reduced with any antidepressant medication compared to placebo (p>0.05). Of the five antidepressant medications studied, only mirtazapine and trazadone showed a statistically significant reduction in AHI in the treated groups but not in sleepiness scale nor an increase in sleep efficiency. In this review, the total sample sizes were small, adverse side effects of some of the antidepressant medications were clinically significant, overall risk of bias of the studies was high or unclear, and overall quality of the evidence was low. Based on the evidence available at this time, we cannot recommend the antidepressants studied in the treatment of OSA.