Impact of Gender on Symptoms and Comorbidities in Obstructive Sleep Apnea

Objective: Obstructive sleep apnea (OSA) is more common in men than in women. In this study, we aimed to address the impact of gender on symptoms and comorbidities in patients with OSA. Materials and Methods: This cross-sectional study was conducted among 1,317 consecutive patients, who were admitted to the Sleep Apnea Clinic of the Marmara University Hospital between November 2015 and October 2018, and who completed questionnaires and a sleep study with cardiorespiratory polygraphy. OSA was defined as Apnea Hypopnea Index (AHI) ≥15/hour. Results: In all, 1,042 patients (334 women) fulfilled the inclusion criteria. OSA was observed in 589 patients (56.5%). Women were older than men (50.2±12.5 versus 45.6±15.1 years) and had lower AHI (22.1±20.1 versus 26.8±21.9 events/h). In the OSA group, women were older (53.7±11.5 versus 47.8±12.8 years) and more obese (BMI 34.6 versus 31.8 kg/m 2 ). Symptoms were categorized as frequent/very frequent, and women with OSA complained more about daytime fatigue (74.6% versus 63.7%), nocturia (69.7% versus 51.8%), headache in the morning (50.0% versus 28.4%), depressive mood (49.0% versus 19.5%), and restless legs symptoms (43.1% versus 17.2%), than did men (all p values <0.05). Comorbidities were observed more in women than in men (lung disease [25.4% versus 13.7%], hypertension [29.6% versus 15.0%], diabetes [20.3% versus 11.3%], and hypothyroidism [14.0 % versus 4.1%]). In regression analysis, age (OR 1.03, p<0.001), BMI (OR 1.13, p<0.001), and male sex (OR 2.08, p<0.001) were significantly predictive for OSA while history of tonsillectomy was protective (OR 0.48, p=0.033). Conclusion: Fatigue, nocturia, headache, depressive mood, restless leg, and comorbidities were observed more in women. OSA-related symptoms develop late and/or the referral of women for diagnostic evaluation of OSA is delayed. Symptoms and comorbidities in women should be evaluated more attentively for earlier referral and diagnosis of OSA.


Introduction
Obstructive sleep apnea (OSA) is characterized by frequent collapse of the upper airway during sleep, resulting in repetitive desaturations in arterial oxygen and arousal from sleep, as well as daytime symptoms, such as headache in the morning, fatigue, disturbed quality of life, depressive mood, and excessive daytime sleepiness (EDS) [1]. Estimated prevalence of OSA is approximately 22% in men and 17% in women [2]. Witnessed apnea, EDS, and snoring are the cardinal symptoms of OSA; however, symptoms related to OSA may show differences according to gender [3]. For females who do not present with the cardinal symptoms of OSA, depressive mood, fatigue, morning headache, and restless leg syndrome may be the symptoms of OSA [4]. Diagnosis of sleep disordered breathing (SBD) in women is usually delayed or missed due to different presentation of the symptoms, rather than classic exhibition, and women tend to underreport their symptoms [5]. In the previous literature, it was reported that moderate-to-severe OSA was undiagnosed in more than 90 % of women, and it was also suggested that mortality among undiagnosed women was significantly higher compared with diagnosed OSA patients [6]. Prevalence of OSA is higher in males when compared with females due to an increase in upper airway resistance and collapsibility; moreover, women's hormones have a protective function in premenopausal period [7,8]. Women with OSA have higher levels of body mass index (BMI), and they are older compared with men with OSA. In addition, co-morbid conditions such as hypertension, diabetes mellitus, and thyroid disease are observed more in women [9,10]. In this study, our aim was to evaluate the symptoms and comorbidities in men and women and to identify the important sexrelated differences in symptoms, thereby aiding physicians to diagnose OSA and refer patients to sleep clinics.

Materials and Methods
Participants This cross-sectional study was conducted among 1,315 consecutive patients, who were admitted to the Sleep Apnea Outpatient Clinic of the Department of Pulmonary Medicine, Marmara University Hospital between November 2015 and October 2018, and who had complete questionnaires and a sleep study with cardiorespiratory polygraphy. Data of the participants were collected retrospectively, and ethical approval was taken from Marmara University ethical committee (approval number: 09.2017.650).

Cardiorespiratory polygraphy
Sleep studies were performed with portable cardiorespiratory polygraphy (CRPG) (NOX T3, Nox Medical Inc., Reykjavik, Iceland) device. Nasal cannula or pressure transducer system was used for detection of nasal pressure, thoracic and abdominal plethysmography belts were used to detect thoraco-abdominal movement and body position, and heart rate as well as oxyhemoglobin saturation (SpO 2 ) were measured by a finger pulse oximeter. Snoring was recorded using a microphone in built in the CRPG device. Estimated sleep time was calculated via self-report by the participants and the pattern of the sleep recordings. Participants whose estimated sleep time was less than four hours were excluded from the protocol. Apnea was defined as a complete (≥90%) cessation of airflow, and hypopnea was defined as a reduction in nasal pressure amplitude of ≥30% and/or thoraco-abdominal movement ≥30% for ≥10 seconds if there was a significant oxyhemoglobin desaturation (decrease by at least 3% from the immediate preceding baseline value) according to the latest recommendations of the American Academy of Sleep Medicine [1]. Moderateto-severe OSA was defined as an apneahypopnea index (AHI) ≥15 events/h of the total estimated sleep time. In order to avoid misdiagnosis, mild OSA participants were not included in the analysis. Healthy participants with AHI <5 might be classified as mild OSA when sleep study is done with a polygraphy because real sleep time cannot be calculated with polygraphy, and estimated sleep time usually exceeds total sleep time.

Epworth sleepiness scale
Epworth sleepiness scale (ESS) questionnaire was used to evaluate daytime sleepiness of the participants [11,12]. The ESS consists of eight questions for assessing the chance of dozing off under eight situations in the past month. Each item is scored from 0 to 3 (0, would never doze; 1, slight chance of dozing; 2, moderate change of dozing; 3, high chance of dozing). The ESS score ranges from 0 to 24. Excessive daytime sleepiness (EDS) was defined as the ESS score of at least 11 [13].

Comorbidities and symptoms
Baseline anthropometrics, smoking habits, medical histories, and co-morbid conditions such as hypertension, cardiac disease, and diabetes mellitus were recorded, and symptoms of the participants were obtained from a questionnaire, which was filled when the patients first presented to the outpatient clinic. Symptoms of OSA were classified as never, rarely, sometimes, frequent, and very frequent; symptoms were accepted as positive when reported as frequent and very frequent.

Statistical Analysis
Data were demonstrated as mean ± standard deviation or standard error of mean for continuous variables, and categorical variables were represented as numbers and percent-• This is a single center study that evaluates cardiorespiratory findings of more than 1000 patients.
• We evaluated the impact of gender on symptoms of obstructive sleep apnea, co-morbid diseases, and associates of moderate-to-severe obstructive sleep apnea was evaluated.
• This study suggest that Symptoms and comorbidities in women should be evaluated more attentively for earlier referral and adequate diagnosis and treatment of OSA. The study was planned with 1297 patients who completed the questionnaire data.

Main Points
Data of 18 patients not available 255 were excluded. 154 were not interested in the study. 101 have TST <4 hours The study was completed with 1042 patients (334 women; 32.1%) with complete questionnaire data and cardiorespiratory polygraphy.
ages. An independent sampled t-test was used for differences in means between groups, and the chi-squared test (or when appropriate, Fisher's exact test) was used to compare categorical variables. A logistic regression analysis was applied for the associates of OSA. Variables that were significant in univariable regression were analyzed in the multivariable model and odds ratios (ORs) with 95% confidence intervals (CI) were reported. All statistical tests were two-sided, and p <0.05 was considered statistically significant. Statistical analysis was performed using the Statistical Package for Social Sciences, version 22.0 for Windows ® system (IBM Corp.; Armonk, NY, USA)

Results
As illustrated in Figure 1, a total of 1,315 participants were screened in the main study. Eighteen of them who did not fill the questionnaires completely (ESS and sleep symptoms) were not included in the main protocol. In addition, 154 patients who did not want to perform CRPG and 101 whose total sleep time was less than 4 hours/night were excluded from the study. The protocol was completed with 1,042 (334 female, 32.1%) patients. OSA was observed in 589 (56.5%) patients. As shown in Table 1

Discussion
In this cross-sectional study, we observed that women with moderate-to-severe OSA were older, more obese, had more co-morbid diseases than men with OSA. When considering cardinal symptoms of OSA (loud snoring, witnessed apnea, and excessive daytime sleepiness [EDS]), there was no significant difference among the men and women; however, more unusual symptoms of OSA, such as morning headache, fatigue, and depressive mood were observed more in women.
Among the patients referred to our clinic, we observed OSA more in men, and average AHI levels were higher in men. We also observed that the proportion of moderate-to-severe OSA among the participants was observed more in men when compared with women. Previous studies have reported that the prevalence of OSA has been known to be more in men when compared with women, and the severity of OSA was worse in men [14]. These findings could be attributed to the fact that milder events are observed in women with sleep apnea when compared with men with sleep apnea. The difference in the architecture of the upper airway anatomy between the sexes (women having a shorter and thus less collapsible pharynx) [15] may contribute to the lack of cardinal symptoms of OSA in women, thus delaying the diagnosis of OSA in women. Undiagnosed OSA in women due to lack of cardinal symptoms may be another reason for the low prevalence of OSA in women [16].
When the symptoms were evaluated in participants with moderate-to-severe OSA, cardinal symptoms, such as loud snoring, EDS, and witnessed apnea had shown no significant difference between sexes; however, most of the studies reported that cardinal symptoms were the indicators of OSA, particularly in men [3,17]. Morning headache, nocturia, and depressive mood were observed significantly more in women with moderate-to-severe OSA in this study. These underestimated and unusual symptoms might be the only sign of OSA, especially in women [18].
The average age and BMI were higher among women in moderate-to-severe OSA participants. It is known that increased age and BMI are the risk factors for OSA [19,20]. Women had less incidences of OSA before menopause when compared with men; however, after menopause there was no difference in the prevalence of OSA between genders [20]; therefore, incidence of OSA increases with advanced age in women [16].
In this study, as expected, advanced age and increased BMI were reported to be predictors of moderate-to-severe OSA. Previously, significant association between OSA and age as well as BMI were reported, and it was also described that  the prevalence of OSA was higher in men, and the male gender was found to be a risk factor for OSA [21,22]. Besides these known associates, we found an inverse relationship between OSA and history of tonsillectomy. Previously, Kang et al. reported tonsillectomy as an effective management option for OSA in children; however, they also reported that residual OSA is frequently observed after tonsillectomy [23]. In management of OSA in children, tonsillectomy is an effective method, and the history of tonsillectomy in childhood can also be protective against occurrence of OSA in advanced age.
The most important limitation of this study was the diagnostic method of OSA as diagnosis was based on cardiorespiratory sleep studies. Therefore, the total sleep time and sleep stages could not be recorded exactly. However, the cut-off value for AHI (15/h) chosen for OSA diagnosis was previously shown to be reliable for the polygraph system [24] used in the current protocol.
In this sleep clinic cohort, there was no gender difference regarding the cardinal symptoms, such as loud snoring, witnessed apnea, and daytime sleepiness while women presented with fatigue, nocturia, headache, depressive mood, and restless legs symptoms more frequently.
In addition, comorbidities were more prevalent in women. Our findings hypothesize that OSA-related symptoms develop late and/or the referral of women for diagnostic evaluation of OSA is delayed. Symptoms and comorbidities in women should be evaluated more attentively for earlier referral and adequate diagnosis and treatment of OSA.
Ethics Committee Approval: Ethics committee approval was received for this study from the Ethics Committee of Marmara University (approval number: 09.2017.650).

Informed Consent: N/A
Peer-review: Externally peer-reviewed.