0970 Sleep Spindle Count In Subsyndromal Depressed vs Normal Elderly: A Protective Effect Of Sleep Spindles?

• Published in: Sleep (New York, N.Y.) Vol. 41; no. suppl_1; p. A360
• Main Authors: Sharma, R A, Miller, M D, Kam, K, Parekh, A, Rivas, J, Bubu, O M, Varga, A W, Iosifescu, D V, Osorio, R S
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 27.04.2018
• Subjects: Neuroplasticity; Older people; Sleep
• ISSN: 0161-8105; 1550-9109; 1550-9109
• DOI: 10.1093/sleep/zsy061.969

Abstract Introduction Sleep disturbances are common early symptoms of Major Depressive Disorder (MDD).Studies suggest that sleep spindles are associated with neuroplasticity, and that having more spindles in NREM stage of sleep can be protective for not only neurodegenerative diseases, but also mood disorders. Moreover, low spindle activity has been considered a risk factor for MDD in young adults. However, it is still not well understood whether sleep disturbances, including changes in sleep architecture, are a symptom of MDD or a risk factor for MDD in the elderly population. This study aims to compare spindle activity in subsyndromal depressive group vs. non-depressed elderly. Methods 24 elderly subjects (mean age 72.2 ± 6.8 years, 67% female) with normal cognitive function (CDR=0, MMSE=29.1 ± 0.9) completed a nocturnal polysomnography (NPSG). Sleep spindles were analyzed using our published automated optimization algorithm DETOKS. Depressive symptoms were measured using the 17 item Hamilton Depression Scale (HAM-D-17). Subjects with subsyndromal depressive symptoms (n=8, with HAM-D-17 scores ≥ 8) were compared with age and gender matched non-depressed individuals (n=16, with HAM-D-17 scores < 8) using the non-parametric Mann Whitney U test. Results No significant difference was found for the in-lab total sleep time between the groups. However, insomnia domain of the HAM-D-17 was different between the groups (p=.027). Compared to non-depressed elderly, subsyndromal depressed group had a statistically significant decrease in spindle count (160.8 ± 87.7 vs 344.7 ± 176.4), translating in an effect size (Cohen’s d) of 1.32; p =0.009. Additional analyses suggested that these associations were driven by a decrease in fast spindles (13.5–15.5 Hz). Conclusion Our data suggest that elderly individuals with subsyndromal depressive symptoms may have a significantly lower spindle count compared to age and gender matched non-depressed elderly. Interestingly, this relationship was driven by fast spindles. Limitations of our study are its cross-sectional design, which limits the understanding of the directionality of these associations, and the mild severity of the depressive symptoms in our sample. It is unclear if our results can be extrapolated to individuals with clinical depression. Future longitudinal studies are warranted. Support (If Any):