RESUMEN
BACKGROUND: In mid-later life adults, early-onset and late-onset (i.e., onset ≥50 years) depression appear to be underpinned by different pathophysiology yet have not been examined in relation to autonomic function. Sleep provides an opportunity to examine the autonomic nervous system as the physiology changes across the night. Hence, we aimed to explore if autonomic profile is altered in mid-later life adults with remitted early-onset, late-onset and no history of lifetime depression. METHODS: Participants aged 50-90 years (n = 188) from a specialised clinic underwent a comprehensive clinical assessment and completed an overnight polysomnography study. General Linear Models were used to examine the heart rate variability differences among the three groups for four distinct sleep stages and the wake after sleep onset. All analyses controlled for potential confounders - age, sex, current depressive symptoms and antidepressant usage. RESULTS: For the wake after sleep onset, mid-later life adults with remitted early-onset depression had reduced standard deviation of Normal to Normal intervals (SDNN; p = .014, d = -0.64) and Shannon Entropy (p = .004, d = -0.46,) than those with no history of lifetime depression. Further, the late-onset group showed a reduction in high-frequency heart rate variability (HFn.u.) during non-rapid eye movement sleep stage 2 (N2; p = .005, d = -0.53) and non-rapid eye movement sleep stage 3 (N3; p = .009, d = -0.55) when compared to those with no lifetime history. LIMITATIONS: Causality between heart rate variability and depression cannot be derived in this cross-sectional study. Longitudinal studies are needed to examine the effects remitted depressive episodes on autonomic function. CONCLUSION: The findings suggest differential autonomic profile for remitted early-onset and late-onset mid-later life adults during sleep stages and wake periods. The differences could potentially serve as peripheral biomarkers in conjunction with more disease-specific markers of depression to improve diagnosis and prognosis.
Asunto(s)
Edad de Inicio , Sistema Nervioso Autónomo , Frecuencia Cardíaca , Polisomnografía , Humanos , Frecuencia Cardíaca/fisiología , Femenino , Masculino , Persona de Mediana Edad , Anciano , Anciano de 80 o más Años , Sistema Nervioso Autónomo/fisiopatología , Fases del Sueño/fisiología , Sueño/fisiología , Depresión/fisiopatologíaRESUMEN
BACKGROUND: While depression is intrinsically and bidirectionally linked with both sleep disturbance and cognition, the inter-relationships between sleep, cognition, and brain integrity in older people with depression, especially those with late-onset depression are undefined. METHODS: One hundred and seventy-two older adults (mean age 64.3 ± 6.9 years, Depression: n = 66, Control: n = 106) attending a memory clinic underwent a neuropsychological battery of declarative memory, executive function tasks, cerebral magnetic resonance imaging and overnight polysomnography with quantitative electroencephalography. RESULTS: The time spent in slow-wave sleep (SWS) and rapid eye movement (REM) sleep, slow-wave activity, sleep spindles, hippocampal volume and prefrontal cortex thickness did not differ between depression and control and depression onset groups. However, sleep onset latency (p = 0.005) and REM onset latency (p = 0.02) were later in the Depression group compared to controls. Less SWS was associated with poorer memory (r = 0.31, p = 0.023) in the depression group, and less SWS was related to better memory in the control group (r = -0.20, p = 0.043; Fishers r-to-z = -3.19). LIMITATIONS: Longitudinal studies are needed to determine if changes in sleep in those with depressive symptoms predict cognitive decline and illness trajectory. CONCLUSION: Older participants with depressive symptoms had delayed sleep initiation, suggestive of delayed sleep phase. The association between SWS and memory suggests SWS may be a useful target for cognitive intervention in older adults with depression symptoms. Reduced hippocampal volumes did not mediate this relationship, indicating a broader distributed neural network may underpin these associations.