Morgenthaler, Timothy I.

Abstract Introduction Obstructive sleep apnea (OSA) is associated with atrial fibrillation (AF). Several features of OSA are thought to play role in the development of AF. A recently developed artificial intelligence (AI) enabled electrocardiogram (ECG) algorithm predicts the probability of future development of AF from a single ECG. We sought to examine the relationship between polysomnogram (PSG) and clinical features of patients with OSA and the probability of future AF (P-AF). Methods Consecutive adults (age ≥ 18) with OSA and with an ECG obtained within 2 years of their attended polysomnograms (PSG) were examined. We excluded patients with AF, left ventricular ejection fraction < 50%, implanted pacemaker or cardioverter defibrillator. We recorded demographics, apnea-hypopnea index (AHI), respiratory arousal threshold (RAT), percent sleep time with SpO2 < 89%, arousal index, and a comorbidity score composed of presence of coronary artery disease (CAD), hypertension (HTN), diabetes mellitus (DM), stroke, and chronic lung disease. One and two-way ANOVA was used to examine the relationship of OSA severity (mild, moderate, or severe) with P-AF. Multiple linear regression further characterized associations with P-AF. Results Mean AI determined P- AF was 0.052±0.012 (mean±SE) in mild, 0.064±0.014 in moderate, and 0.105 + 0.014 in severe OSA, demonstrating a significant association of OSA severity with P-AF (p = 0.01). Post-hoc pairwise tests showed a significant difference in P-AF between mild and severe (p = 0.008) but not between moderate and severe OSA (p = 0.083). Multiple linear regression with all the variables listed above showed the association between OSA severity and P-AF remains significant (P = 0.044). Age (p<0.001) and comorbidity score (p<0.001) were the only other variables significantly associated with P-AF. Conclusion OSA severity determined by AHI was significantly associated with P-AF as determined by a novel ECG-based AI algorithm independent of other variables. Given the strength of association with age and the comorbidity score, OSA severity may have been associated as it is known to increase with age and comorbid conditions. Future directions include incorporating AI enabled algorithms to identify individuals with the highest risk of AF in whom the role of OSA treatment in reducing risk of AF may be examined. Support (If Any)

Abstract Introduction Cognitive behavior therapy for insomnia (CBT-I), a first-line therapy for patients with chronic insomnia disorder (ChID), is underutilized due to reduced access to trained providers and other barriers. We developed an interactive care plan (ICP) combining individual aspects of CBT-I. This ICP, which integrates with the electronic health record (EHR), is designed to deliver personalized and scalable CBT-I to patients with ChID. We report data from initial implementation and evaluation of a ChID ICP deployed at a tertiary sleep medicine clinic. Methods The ICP was developed following patient and provider interviews and focus groups. Patients diagnosed with ChID were offered enrollment and engaged in the ICP through an app. The ICP has embedded logic that may escalate care or offer an exit from the program, depending upon the patient’s response to questions. Key variables ascertained from the patient included sleep efficiency (SE), sleep timing and the Insomnia Severity Index (ISI), and how satisfied they were with their progress (Likert scale). We also measured In Basket messaging related to use of the ICP. Results A total of 222 patients [57% female, age=56.4±15.9 years (average ±SD), 92.7% white] were enrolled in the initial 120 days of ICP implementation. Most referrals were initiated by physicians or advanced practitioners (77.9%). ISI at initiation was 15±5.18 and SE was 74.55±16.65%. Patients spent 35.9±26.25 days engaging with the ICP. The ISI score at 28 days was 12.89±4.98 and SE was 78.9±13.4% with both showing significant improvement from baseline (p=0.013; p=0.002; respectively). Self-rated satisfaction with progress did not significantly correlate with actual improvement in ISI or SE (p>0.05). The ICP generated 3.89±2.3 In Basket messages per patient. Conclusion Patients who actively engaged with the ICP showed significant improvements in insomnia severity and sleep efficiency. While a majority of the patients were neutral towards the ICP and only a small minority expressed dissatisfaction, these data indicate that the ICP will have clinical utility in busy sleep medicine practices with reduced access to behavioral sleep specialists. Support (If Any)

Abstract Introduction The COVID-19 pandemic accelerated use of telehealth, an expansion of services that required the appropriate technological infrastructure for health care facilities and in patient homes. Telehealth on the one hand has held promise for addressing health disparities perpetuated by inadequate rural access, but delivery requires extensive broadband and technologic access. That creates the possibility for new kinds of healthcare disparities. In addition, COVID-19 forced marked reduction in in-lab polysomnography (PSG), and concurrent expansion of home sleep apnea testing (HSAT). We hypothesized that the pandemic led to an increase in the overall frequency of telehealth and HSAT utilization, particularly in those who were younger, White, more educated, and from a non-local area. Methods We completed a retrospective chart review of all adult patients seen by all provider types across the Sleep Medicine practices in Mayo Clinic Rochester, Arizona, Florida, and the Mayo Clinic Health System between 1) 6/1/18—3/8/20 (Pre-COVID-19), 2) 3/9/20—4/19/20 (Early COVID-19), and 3) 4/20/20—present (Late COVID-19). We recorded the total number of PSGs and HSATs and total number of face-to-face and telehealth visits, along with the age, gender, race, educational level, and locality by zip code for patients served. These data were compared across the 3 timeframes. Results Average monthly visits changed from Pre-COVID-19, Early COVID-19, to Late COVID-19 [2194.7/m, 1416.5/m, 2690.6/m (P<0.001)]. Average monthly sleep test volumes also changed [1004.1/m, 530.5/m, 1123.4/m (P<0.001)], with a proportionate increase in HSATs across the 3 periods [34.71%, 65.37%, 53.59% (P<0.005)]. The increase in Late COVID-19 in telehealth visits occurred proportionately more in those who were younger, female, non-White, college and post-graduate educated, and from a non-local area. The increase in use of HSATs occurred proportionately more in patients who were younger, female, non-White, college and post-graduate educated, and from a local area. Conclusion The COVID-19 pandemic increased the use of telehealth visits and HSATs in Sleep Medicine practices across our enterprise, particularly in those who are younger and more educated, which may be due to ease of use with and access to technology. The reasons for the presence of additional disparities based upon gender, race, and locality needs further exploration. Support (If Any)

Abstract Introduction We have previously shown that older adults with excessive daytime sleepiness (EDS) appear to be more vulnerable to longitudinal amyloid PET accumulation before the onset of the dementia. It remains unclear whether this vulnerability is specific to amyloid or extends to other biomarkers of Alzheimer’s disease pathology, or axonal integrity and inflammation, which can also contribute to neurodegeneration and cognitive changes. Methods For this cross-sectional analysis, we identified 260 cognitively unimpaired adults (>60 years old) without neurologic disorder who underwent CSF quantification of AD biomarkers (CSF Aβ-42, p-tau181, p-tau217) along with at least one other biomarker of interest (neurofilament light chain [NfL], IL-6, IL-10, and TNF-α) from the Mayo Clinic Study of Aging – a longitudinal population-based cohort in Olmsted County, Minnesota. CSF biomarkers were available in 251-260 individuals, depending on the biomarker. We fit linear regression models to assess whether the CSF biomarkers were associated with sleepiness as measured by the Epworth Sleepiness Scale (ESS), after controlling for age, sex, APOE4 genotype, BMI, hypertension, dyslipidemia, and OSA diagnosis (by chart review). Results Higher ESS scores were independently associated with higher CSF IL-6 and NfL, but not with the other biomarkers in the whole sample. For every single-point increase in the ESS score, there was a .008 ([95% CI .001-.016], p=0.033) increase in the log of IL-6 and .01 ([95% CI .002-.018], p=0.016) increase in the log of NfL. A sensitivity analysis showed a correlation between ESS scores and log of p-tau/Aβ-42 ratio only in participants with abnormal ratio (>0.023), after controlling for APOE4 (partial r=.27, p=039). Conclusion Our results corroborate previous literature suggesting that higher inflammatory milieu reflected by increased CSF IL-6 is associated with sleepiness. The association between NfL and sleepiness suggests that sleepiness may be related to disturbed connectivity due to axonal damage. Alternatively, NfL may be a surrogate of active axonal injury associated with more disrupted sleep. A correlation between sleepiness and CSF p-tau/ab-42 ratio was only seen in patients with abnormal ratio, suggesting a stronger association between sleepiness and AD pathology as the disease progresses, possibly because AD pathology worsens sleep quality and/or vice-versa. Support (If Any) NIH/NIA