1089 Roles of Depression, Rapid Eye Movement Percent, and Weight on Cognition in Older Adults with Obstructive Sleep Apnea: Earlier Diagnosis May Help

Abstract Introduction Rapid-eye movement (REM) sleep (~20% of total sleep time in adults aged ≥65) plays an important role in memory consolidation. Patients with Mild Cognitive Impairment (MCI) have reduced REM sleep and increased REM latency. Although major depressive disorder (MDD) is also associated with MCI, few studies explore interactions between affective disorders, REM-sleep disorders, and neurocognitive deficits (NCDs). This study addresses said literature gap by sampling from adults ≥65 populations to characterize interactions between NCDs, depressive disorders (Dep), obstructive sleep apnea (OSA), and weight. Methods In a retrospective chart review, patients with OSA who received polysomnography (PSG) were identified using ICD-10 codes. Inclusion criteria were ≥65, alive, and without history of cerebrovascular events. Primary outcome was NCD presence. Results N=70; NCD+=35. Female=41. No REM sleep=15. Mean age=75.0. Mean Weight=87.9 kg. Dep+=16. Taking psychotropic medications (PsyMeds+)=33. On univariable logistic regression, Dep+ shows strongest association with NCDs (OR 6.30 [1.78-29.90], p=0.0083), and weight is inversely associated with NCDs (OR per kg=0.97 [0.95-0.99], p=0.0144). Age at first PSG (OR per year=1.07 [1.00-1.16], Wilcoxon p=0.051) and REM% (OR=0.96 [0.91-1.01], Wilcoxon p=0.106) both approach significance. Age at Dep diagnosis and PsyMed+ are not significantly associated with NCDs. In multivariable models, depression remained the dominant NCD predictor, increasing probabilities from ~30–60% to ~75–90% across age and REM% strata; age and REM% exerted modest effects. After adjustments of REM% and depressive disorders, lower weight continued to trend toward higher NCD odds (OR per kg=0.97 [0.94–1.00], p=0.054). Conclusion Depression in general ≥65 populations has reported ORs between 1.2-2.5 for NCDs. While more data are necessary to narrow 95% CIs, our higher OR for depression in older patients with OSA suggests complex interactions between depression and REM% for NCD risks. Weight loss, a recognized depressive symptom, has also been linked to NCDs. Our results suggest lower weight could be predictive of NCDs independently of depression in patients with OSA. Earlier OSA diagnosis and CPAP treatment could potentially reduce risks for NCDs. Prospective studies are needed. Finally, even with OSA (potentially memory-impairing), our data do not support concerns that antidepressants independently contribute to memory loss. Support (if any)