0620 Effect Of Antidepressants On Sleep Architecture

Introduction Antidepressants are presumed to have a strong effect on sleep stage. Current standards for performance of MSLT recommend that selective serotonin reuptake inhibitor (SSRI) type antidepressants are held prior to administration, to avoid potentially spurious result based on the presumption they suppress REM sleep during the MSLT. Due to the consequences of such medications change, including risk of worsening depression or even higher suicide risk, often the test is omitted altogether. Most prior studies are small, including 12-30 subjects, and most examined healthy volunteers. Furthermore, most examine the effects of these medications on the overnight polysomnogram, with very few reporting data on MSLT. While in some circumstances stimulants are prescribed symptomatically without an established diagnosis, for most patients we see, the firm diagnosis is needed to justify services or medication coverage. Methods To determine the extent of the antidepressant/REM suppressant effect on the MSLT from a large volume retrospective analysis, we compared MSLT results among 3 groups of patients - on antidepressants, on no antidepressants and on held antidepressants. Our data source is over 850 PSG/MSLT studies gathered at 3 Boston-area medical centers from 2014-2018. The MSLT consists of 5 naps and we will compare REM latency and sleep latency on those 5 naps among these 3 groups. To avoiding censoring data in the event of no nap, we will perform survival analysis. Results Preliminary data from approximately 200 patients demonstrates mean sleep latency in SSRI group of 9.9+/-6.3 minutes. The mean sleep latency of the non-SSRI group is 9.7+/-7.6 minutes. The mean REM latency of the SSRI group is 19.3+/-3.9 minutes, and then mean REM latency of the non-SSRI group is 17.4+/-5.9 minutes. Conclusion Preliminary analysis demonstrates similar sleep-onset latencies between SSRI and non-SSRI groups, with a possibly shorter REM latency, however, with further data accrual and survival analysis, this effect may become more apparent. The preliminary data may lend support to extending the criteria for SOREM in an MSLT in a patient taking SSRI. Support (If Any) Jazz