SUMMARYAlterations in rapid eye movement sleep have been consistently related to depression in clinical studies. So far, there is limited evidence from population-based studies for this association of rapid eye movement sleep alterations with depressive symptoms. In 489 participants of the Rotterdam Study, we assessed rapid eye movement sleep latency, rapid eye movement sleep duration and rapid eye movement density with ambulant polysomnography, and depressive symptoms with the Center of Epidemiologic Studies-Depression Scale. A longer rapid eye movement sleep latency (B = 0.002, P = 0.025) and higher rapid eye movement density (B = 0.015, P = 0.046) were related to depressive symptoms after age-sex adjustment. When we excluded persons who used sleep medication or medication for the nervous system (n = 124), only rapid eye movement density remained related to depressive symptoms (B = 0.018, P = 0.027). Our results suggest that rapid eye movement density is a marker of depressive symptoms in the general population, and that associations of rapid eye movement sleep with depressive symptoms are modified by the use of medication.
IN TROD UCTI ONClinical research suggests that a decreased rapid eye movement (REM) sleep latency, increased REM sleep duration and increased REM density are prominent in persons with depressive disorders (Palagini et al., 2013). Changes in sleep are found during and before the onset of depressive disorders, suggesting that sleep alterations can be both a trait and state marker (Pillai et al., 2011).Studies on the associations of REM sleep and depression have been conducted in clinical populations with severe depressive episodes mostly. However, in the general population, symptoms of depression are usually lingering and less severe. In two large population-based studies, no associations of REM sleep with depressive symptoms were found (Castro et al., 2013;Kravitz et al., 2011). In a large sample of older men, a lower percentage of time spent in REM sleep was related to more depressive symptoms (Smagula et al., 2013).Clinical studies typically investigate medication-na€ ıve patients, as sleep medication and other medication can impact REM sleep (Wilson and Argyropoulos, 2005). Population-based studies mostly adjust for medication use, such as anti-depressants, which are commonly prescribed. However, medication can both confound and modify the association between REM sleep and depressive symptoms. In this study we examined whether REM sleep latency, REM sleep duration and REM density are related to depressive symptoms in a population-based sample. We also assessed these associations excluding persons who used sleep medication or medication prescribed for the nervous system.
MAT ERIALS AN D METH ODS
Study populationThe current study was embedded in the Rotterdam Study, a population-based cohort of middle-aged and elderly inhabitants of Rotterdam, the Netherlands (Hofman et al.,