OSA Severity Measures Independently Associated With Spindle Abnormalities in Men

Gary Wittert, MD, FRCP, FRACP, MBBch

Obstructive sleep apnea (OSA) severity measures were independently associated with spindle abnormalities in a large population-based sample of middle-aged and older men. 

Urban, community-dwelling men who were aged 35 years or older at baseline were included from the Men Androgen Inflammation Lifestyle Environment and Stress (MAILES) study. Sleep studies were completed for 837 men, of whom 751 (89.7%) had spindle data of adequate quality. Investigators, including Gary Wittert, MD, FRCP, FRACP, MBBch, Mortlock Professor of Medicine, the University of Adelaide; senior consultant endocrinologist, Royal Adelaide Hospital; and director, Freemasons Foundation Centre for Men’s Health, performed analyses for 675 men after 76 (10.1%) were excluded due to their use of 1 or more sleep-disrupting medications. Participants were primarily of Australian and European descent (96%) and had no previous OSA diagnosis. 

Analyses of participants after undergoing home-based polysomnography (PSG) showed that higher apnea-hypopnea index (AHI) per hour and percentage total sleep time with oxygen saturation less than 90% (TST90) was associated with decreased slow spindle density (AHI: B = –0.003 [P = .032]; TST90: B = –0.004 [P = .047]) but increased frequency (AHI: B = 0.002 [P = .009]; TST90: B = 0.002 [P =.043]). Investigators also found that higher TST90 was associated with greater spindle amplitude (N2 sleep: B = 0.04 [P = .011]; N3 sleep: B = 0.11 [P <.001]). 

Higher arousal index was also associated with greater spindle amplitude during N2 sleep (B = 0.31 [P <.001]), but during N3 sleep it was associated with decreased overall density (B = –1.27, [P = .030]) and fast density (B = –4.36 [P = .028]). Investigators also performed analyses of subjective sleep quality, as measured by the Pittsburgh Sleep Quality Index (PSQI), and sleepiness, as measured by the Epworth Sleepiness Scale (ESS), noting that unadjusted analyses showed no associations between subjective sleep quality or sleepiness with spindle metrics during N2 or N3. In comparison, adjusted analyses showed an association of higher PSQI with decreased spindle frequency during N3 sleep. 

READ MORE: Evaluating a Sleep Medicine ‘Boot Camp’ to Improve Knowledge, Core Curriculum 

“This study is one of the first to examine cross-sectional associations of OSA severity measures (AHI and TST90) and sleep fragmentation (arousal index) with spindle metrics among a large population-based sample of men while accounting for potential confounders,” Wittert et al wrote. “Measures of OSA severity and sleep fragmentation were independently associated with spindle abnormalities during N2 and N3 sleep. Poorer subjective sleep quality as assessed by the PSQI appeared to be independently associated with lower spindle frequency but only during N3 sleep.”

Participants underwent 8-channel, home-based ambulatory PSG that recorded all-night electrical brain activity and left electrooculography. Study equipment was set up by trained staff who also administered the PSQI and ESS and obtained anthropometric measures. All PSG measures were scored by a sleep technician, with OSA defined as an AHI of 10 events or greater per hour, then categorized as mild (10-19/hour), moderate (20-29/hour), or severe (≥30/hour).

The MAILES study, which evaluated the effect of sex steroids, inflammation, environmental, and biopsychosocial factors on cardiometabolic risk disease in men, included 2569 participants between 2002 and 2006. During follow-up, 1629 participants were interviewed and asked about OSA, with a total of 184 men excluded due to previous OSA diagnosis. A total of 1445 participants were asked to undergo the 8-channel PSG, with 1087 participants (75.2%) initially agreeing to participate. 

The study authors acknowledged that the work was limited as it only collected data from men and was unable to infer causality due to the cross-sectional design. Topographical differences in spindle metrics may have been missed as spindle data were collected using only the frontal derivation. Investigators also noted the inability to adjust for sleep-associated movement disorders that may modify sleep microarchitecture characteristics, and the “relatively weak” associations between OSA severity measures and spindle metrics that render clinical significance uncertain. 

“The utility of respiratory and hypoxemia-related decrease in slow spindles to identify patients with OSA at risk of potential cognitive decline and adverse functional health outcomes warrants further prospective investigation,” Wittert et al wrote. 

REFERENCE
Parker JL, Melaku YA, D’Rozario AL, et al. The association between obstructive sleep apnea and sleep spindles in middle-aged and older men: a community-based cohort study. Sleep. 2021;zsab282. doi:10.1093/sleep/zsab282