Austin Health

Title
Association of Obstructive Sleep Apnea with Post-Acute Sequelae of SARS-CoV-2 infection (PASC).
Publication Date
2023-12-31
Author(s)
Quan, Stuart F
Weaver, Matthew D
Czeisler, Mark É
Barger, Laura K
Booker, Lauren A
Howard, Mark E
Jackson, Melinda L
Lane, Rashon I
McDonald, Christine F
Ridgers, Anna
Robbins, Rebecca
Varma, Prerna
Wiley, Joshua F
Rajaratnam, Shantha M W
Czeisler, Charles A
Subject
COVID-19
Long COVID
Obstructive Sleep Apnea
PASC
Post-Acute Sequelae of SARS-CoV-2 infection
Type of document
Journal Article
OrcId
0000-0002-9474-7679
0000-0003-3578-336X
0000-0003-3100-7347
0000-0001-8547-7331
#PLACEHOLDER_PARENT_METADATA_VALUE#
0000-0001-7262-2632
0000-0003-4976-8101
0000-0002-0612-1466
0000-0001-6481-3391
0000-0003-1360-9387
0000-0003-0288-2505
0000-0001-5408-1625
0000-0002-0271-6702
0000-0001-7527-8558
0000-0002-7408-1849
DOI
10.1101/2023.12.30.23300666
Abstract
Obstructive sleep apnea (OSA) is associated with COVID-19 infection. Fewer investigations have assessed OSA as a possible risk for the development of Post-Acute Sequelae of SARS-CoV-2 infection (PASC). In a general population, is OSA associated with increased odds of PASC-related symptoms and with an overall definition of PASC? Cross-sectional survey of a general population of 24,803 U.S. adults. COVID-19 infection occurred in 10,324 (41.6%) participants. Prevalence rates for a wide variety of persistent (> 3 months post infection) putative PASC-related physical and mental health symptoms ranged from 6.5% (peripheral edema) to 19.6% (nervous/anxious). In logistic regression models adjusted for demographic, anthropometric, comorbid medical and socioeconomic factors, OSA was associated with all putative PASC-related symptoms with the highest adjusted odds ratios (aOR) being fever (2.053) and nervous/anxious (1.939) respectively. Elastic net regression identified the 13 of 37 symptoms most strongly associated with COVID-19 infection. Four definitions of PASC were developed using these symptoms either weighted equally or proportionally by their regression coefficients. In all 4 logistic regression models using these definitions, OSA was associated with PASC (range of aORs: 1.934-2.071); this association was mitigated in those with treated OSA. In the best fitting overall model requiring ≥3 symptoms, PASC prevalence was 21.9%. In a general population sample, OSA is associated with the development of PASC-related symptoms and a global definition of PASC. A PASC definition requiring the presence of 3 or more symptoms may be useful in identifying cases and for future research.
Link
Citation
MedRxiv : the Preprint Server for Health Sciences 2023-12-31
Jornal Title
MedRxiv : the Preprint Server for Health Sciences

Files:

NameSizeformatDescriptionLink