033 The effect of sleep disturbance on cognition in late-life depression

Abstract

Abstract Introduction Late-life depression is the most common psychiatric disorder in older adults and is associated with cognitive deficits, however, the role of sleep disturbance in cognitive deficits is poorly defined. In the current study we aimed to examine sleep macro and micro-architecture differences between those with late-life depression and controls. Secondly, we sought to determine how sleep changes relate to clinical memory and executive function measures in those with late-life depression and controls. Methods Using prior clinical data, this retrospective study assessed adults >50 years who had completed an overnight PSG study and comprehensive psychiatric, neuropsychological, and medical assessment. Memory performance was measured using the Weschler Memory Scale logical Memory 1 and 2 components, Rey Auditory Verbal Learning Test (Senior) 30-minute recall and Rey Complex Figure 3-minute recall. Executive function was defined by z scores from Trail Making Test, D-KEFS Stroop Test and Controlled Oral Word Association Test. The sample comprised of 71 depressed participants, defined by a Geriatric Depression Scale score ≥6, and 101 non-depressed participants (GDS <6 and no lifetime history of depression using DSM-IV criteria). Results Contrary to our hypothesis no significant macroarchitectural differences were observed between the groups. Less time spent in slow-wave sleep (SWS) was associated with worse delayed memory recall scores in the depression group (z=.342, p=0.008) although this was not seen in the control group. SWS and slow wave activity (SWA) were not related to measures of executive function performance. Depressed participants demonstrated a reduced level of sleep spindles (Dep= 159 ±142.8, con= 213±163, p=.03) although there were no associations with memory outcomes. Conclusion Compared to younger adults with depression, macroarchitectural differences in those with late-life depression are not as pronounced, due to a reduction of SWS and SWA power as a function of ageing. The efficiency of SWS hippocampal dependent memory processes in depression may be reduced, therefore, more time spent in SWS is related to better memory performance. This study assessed the density of sleep spindles but not spindle and slow wave oscillation coupling which may be more important for hippocampal dependent memory. Support (if any):