Abstract P280: The Association of Sleep Apnea with Psychiatric History: A Veteran Twin Study

Abstract

Introduction: Sleep apnea is exceedingly common in elderly men and is associated with higher risk of cardiovascular disease. Post-traumatic stress disorder (PTSD) and depression have been shown to be associated with sleep apnea, however results have not been consistent, and it is unclear whether this association is confounded by cardiovascular and behavioral risk factors. In this study, our objective was to explore the risk factors for sleep apnea with a focus on psychiatric history, and we hypothesized that the association of sleep apnea with psychiatric history can be largely influenced by cardiovascular and behavioral factors. Methods: This was a cross-sectional analysis of 100 members of the Vietnam Era Twin (VET) Registry (50 twin pairs, mean age=68; range: 61-71 years). All twins underwent a one night in-lab polysomnography (PSG) to assess the apnea/hypopnea index (AHI). Clinical diagnoses of lifetime history of major depression and PTSD were obtained with the Structured Clinical Interview for DSM V (SCID); we also measured current depressive symptoms with the Beck Depression Inventory-II (BDI-II). To assess associations of study variables with AHI, within-pair differences in multivariable mixed-effects regression models were examined and β coefficients were calculated. In addition to lifetime history of depression and PTSD, we included in the model the following variables previously reported in the literature to be associated with AHI: body mass index (BMI), current smoking, history of alcohol abuse, and physical activity assessed with the Baecke score. Other variables included years of education, sedative/hypnotic use, and antidepressant use. Results: The mean AHI among these men was 15.5 [SD=16.1]. A total of 18 and 28 twins had diagnoses of lifetime history of depression and PTSD, respectively. In bivariate analysis without adjustment for covariates, psychiatric history (depression or PTSD) was not significantly associated with higher AHI. In mixed-effects multivariable regression analysis, only higher BMI (β=2.3, 95% CI=1.5, 3.1) and less education (β=-1.3, 95% CI=-2.4, -0.1) were independently associated with higher AHI. Psychiatric history of depression (β=3.9, 95% CI=-3.5, 11.3) or PTSD (β=6.0, 95% CI=-0.9, 12.9) were not significantly associated with AHI. Additional analyses examining number of depressive symptoms (BDI-II) showed similar results. Conclusion: As expected, higher BMI was associated with higher AHI, but several other variables thought to be associated with higher AHI were not confirmed, such as smoking status, alcohol abuse, sedative/hypnotic use, and lack of physical activity. In contrast to the prevailing literature, neither lifetime history of depression nor PTSD were associated with sleep apnea. However, the role for poor sleep quality other than sleep-disordered breathing still remains to be investigated.