Gender differences in association between expiratory dynamic airway collapse and severity of obstructive sleep apnea

Objectives Repetitive unbalances and tensions generated by inspiratory efforts against an obstructive upper airway during sleep predispose the development of expiratory central airway collapse. In addition, structures of the upper airway between men and women have differences and could be the reasons for differences in obstructive sleep apnea (OSA) prevalence between genders. The present study aimed to evaluate the association between parameters of expiratory dynamic tracheal collapse measured using chest multidetector CT and objectively measured OSA severity between men and women. Materials and methods A total of 901 participants who underwent chest CT and overnight in-home polysomnography from the Korean Genome and Epidemiology Study were cross-sectionally analyzed (women: 46.2%). The participants were divided into three groups based on OSA severity by apnea–hypopnea index (AHI). Multivariate linear regression analysis was performed to determine the effects of central airway collapse after adjustment for cardiovascular-related covariates. Results In a multivariate analysis, percentages of expiratory lumen structure reductions involving area, diameter, and perimeter were associated with AHI (all p values < 0.05) and with OSA severity (moderate-to-severe OSA than no OSA: β = 3.30%, p = 0.03; β = 2.05%, p = 0.02; β = 1.97%, p = 0.02, respectively) in women, whereas men had only a greater percentage of expiratory wall thickness reduction in moderate-to-severe OSA than no OSA (β = 0.72%, p = 0.003). In addition, women with both mild OSA and moderate-to-severe OSA had higher expiratory tracheal collapse than men without OSA, and a moderate effect of sex was observed (p for interaction = 0.007). Conclusion The expiratory dynamic tracheal collapse was independently associated with severity of OSA in women than in men. Clinical relevance statement Differences of pharyngeal structures and inherent features of airways by genders may affect the dissimilarities in vulnerability to sleep apnea between men and women. Key Points • The expiratory dynamic tracheal collapse was independently associated with severity of OSA in women than in men. • Women with over mild OSA had higher expiratory tracheal collapse than men without OSA, and moderate effect of sex was observed. • Structural differences of airway may affect differences in susceptibility of sleep apnea between genders. Supplementary Information The online version contains supplementary material available at 10.1007/s00330-023-10322-x.


Study design and population
All study participants were part of the 2001-2002 Korean Genome and Epidemiology Study, which is an ongoing prospective investigation.Detailed information on participant recruitment is available elsewhere [1][2][3].Briefly, a total of 5012 participants from Ansan, South Korea, were examined at baseline between 2001 and 2002.The cohort participants had a questionnaire-based interview and health examination, and biospecimens were collected by health professionals.Blood samples were collected after a fasting period of at least 8h.The questionnaire included in demographic characteristics, medical history, lifestyle, and sleeprelated parameters.The health examination comprised anthropometric and clinical evaluations, including chest CT.Follow-up examinations were performed biennially during scheduled site visits.

Polysomnography
An oxygen desaturation event was detected when oxygen saturation dropped by at least 4%.All saturation values <50% were excluded as artifact values and were not considered as part of the desaturation event.An apnea event was detected if both of the following criteria were met: (ⅰ) a drop in the peak signal excursion by ≥90% of the pre-event baseline (reference amplitude) was observed and (ⅱ) the duration of the ≥90% drop in the sensor signal was ≥10 s.In addition, a hypopnea event was detected if all of the following criteria were met: (ⅰ) the peak signal excursion dropped by ≥30% of the reference amplitude; (ⅱ) the duration of the ≥30% drop in signal excursion was ≥10 s; and (ⅲ) a ≥3% oxygen desaturation from the reference amplitude occurred or the event was associated with an arousal.The reference amplitude was calculated as the mean value of the peak amplitudes in the period of 100 s preceding the event.

Assessment for sleep-related parameters by questionnaires
Insomnia severity was measured using the Insomnia Severity Index (ISI).The ISI is a seven-item self-administered questionnaire assessing the nature, severity and impact of insomnia in adults over the past 2 weeks.A 5-point Likert scale on a 0-4 point is used to rate each item, and the total score ranges from 0 to 28.Severity levels is categorized as no clinically significant insomnia (0 to 7 points), subthreshold insomnia (8 to 14 points), and moderate to severe insomnia (15 points or more) [4].In addition, we measured depressive symptoms and daytime sleepiness.Depressive symptoms were measured using the Beck Depression Inventory (BDI).The BDI was developed to assess the type and degree of depression, based on symptoms of depression.The questionnaire contains 21 questions, and each item consists of four statements describing increasing intensities of symptoms of depression (total score ranges: 0-63); higher scores reflect more severe of depressive symptoms [5].Daytime sleepiness was measured using the Epworth Sleepiness Scale, a validated eight-item questionnaire that assessed excessive daytime sleepiness.A sum score is calculated (range: 0-24), with lower scores indicating less sleepiness [6].Sleep quality was assessed via the Pittsburgh Sleep Quality Index (PSQI), which is a widely used and wellvalidated 19-item questionnaire that measures sleep quality in adults [7].It provides a global PSQI score (rages: 0-21), which consists of seven sub scores, and higher PSQI global score indicates poorer sleep.

Figure E1 .
Figure E1.Methods for analysis of trachea.(A) Cross-section images were acquired from 10 locations.(B) Method for 120 profiles from each sample to measure the wall thickness of the tracheal wall.(C) Method of measuring the diameter of the airway.(A)

Table E1 . Comparison of tracheal computed tomography parameters between men and women
*P-values for one-way analysis of covariance including age, body mass index, hypertension, type 2 diabetes, pack-years of smoking, and inspiratory whole lung volume.

Table E3 . Multivariate linear regression analysis of the relationship between OSA severity and percentage of expiratory reduction in tracheal CT parameters Models Relationship to AHI Comparison of OSA group
AHI, apnoea-hypopnea index; SE, standard error; exp, expiration; insp, inspiration; Ref. reference.Model 1: adjusted for age and sex.Model 2: adjusted for age, sex, and body mass index.Model 3: adjusted for age, sex, body mass index, hypertension, type 2 diabetes, pack-years of smoking, and inspiratory whole lung volume.