|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Departments of Medicine (Drs. M.S.M. Ip, B. Lam, W.K. Lam and Ms. T.Y. Ip) and Statistics and Actuarial Science (Dr. Lauder), The University of Hong Kong; and Department of Obstetrics and Gynecology (Dr. Tang), Kwong Wah Hospital, Hong Kong, SAR, China.
Correspondence to: Mary S. M. Ip, MD, FCCP, Department of Medicine, The University of Hong Kong, 4/F, Professorial Block, Queen Mary Hospital, Pokfulam Rd, Hong Kong, SAR, China; e-mail: msmip{at}hkucc.hku.hk
| Abstract |
|---|
|
|
|---|
Design: A cross-sectional study conducted in Hong Kong from 1998 to 2000.
Setting: Sleep questionnaires were distributed to women (30 to 60 years old) in three offices and two community centers. All were invited to undergo full polysomnography in a sleep laboratory.
Participants: Questionnaires were distributed to 1,532 women, and 854 questionnaires were returned. Polysomnography was conducted in 106 respondents.
Measurements and results: Conservative estimated prevalence of SDB (apnea-hypopnea index [AHI]
5) and OSAS (AHI
5 plus excessive daytime sleepiness [EDS]) were 3.7% and 2.1%, respectively. Age-specific prevalence of OSAS was 0.5%, 2.2%, and 6.1% in the 30- to 39-year-old, 40- to 49-year-old, and 50- to 60-year-old age groups, respectively. Stepwise multiple logistic regression analysis identified body mass index (BMI) and age as predictors of SDB. Compared to Chinese men, the prevalence of SDB and OSAS in women was lower, but the gender difference decreased with age. The AHI of affected women was also significantly lower despite comparable BMI. Compared to men, women with SDB had same degree of self-reported snoring and a similar degree of EDS despite the lower AHI.
Conclusions: This study demonstrated an estimated prevalence of OSAS at 2.1% among middle-aged Chinese women in Hong Kong, with a 12-fold rise from the fourth to the sixth decade of life. BMI and age were significant independent predictors of SDB. Compared to men, women with SDB had lower AHIs, despite similar BMIs.
Key Words: middle-aged Chinese women polysomnography sleep apnea
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
Sample
The target population was women 30 to 60 years old in the community. Questionnaires (modified from National Auxillary Publications Services document No. 050174
), similar to that used in a study of men, with addition of questions on menopause and hormone replacement therapy, were distributed to the office staff of three public institutions, and to visitors of two community centers in Hong Kong. The public institutions were government departments for non-health-related functions, while the community centers were operated by nongovernment organizations that provided a variety of welfare activities for the community.
Collection of Polysomnographic Data
Studies were conducted at the sleep laboratory at Queen Mary Hospital. Polysomnography (Alice 3 System; Respironics, Pittsburgh, PA), consisting of continuous polygraphic recording from surface leads for EEG, electrooculography, electromyography (chin and legs), ECG, thermistors for nasal and oral airflow, thoracic and abdominal impedance belts for respiratory effort, pulse oximetry for oxyhemoglobin level, tracheal microphone for snoring, and sensors for sleep position.
Interpretation of Polysomnographic Data
Polysomnographic records were scored manually. Sleep data were scored according to standard criteria.8
An abnormal breathing event during objectively measured sleep was defined in the same way as the study in men7
for the benefit of comparability, according to the criteria of either a complete cessation of airflow lasting
10 s (apnea) or a discernible reduction in airflow accompanied by a decrease of
4% in oxyhemoglobin saturation (hypopnea). The average number of episodes of apnea and hypopnea per hour of sleep (apnea-hypopnea index [AHI]) was calculated as the summary measurement of SDB. Arousals were identified according to established criteria.9
Assessment of Daytime Sleepiness
Daytime sleepiness was assessed with four subjective questions. Using a 5-point scale (0 to 4), the subjects rated the following: how often they felt excessively sleepy during the daytime; how often they felt unrefreshed or tired during the day, regardless of how long they had slept; how often they fell asleep or dozed off momentarily while watching television, reading, or at meeting/church; and how often they felt sleepy while driving. The answer was considered positive if the score was
2. Subjects were identified as having excessive daytime sleepiness (EDS) if they gave a positive response to three of the four questions.
Definition of SDB and Obstructive Sleep Apnea Syndrome
The minimum criterion for SDB was AHI
5, and data for three AHI cutoff threshold values at 5, 10, and 15, respectively, are presented. Similarly, the minimum criterion for diagnosis of obstructive sleep apnea syndrome (OSAS) was AHI
5 in the presence of EDS (as defined by the questions above), and data for three cutoff AHI values are presented.
Calculation of Prevalence
This was calculated as previously described.7
Briefly, questionnaire respondents were classified as habitual snorers and nonhabitual snorers based on self-reported symptoms. Within each group, the age and body mass index (BMI) of those who underwent polysomnography and those who did not were compared. If significant differences were found, a conservative estimate was adopted, treating the subjects with SDB documented by polysomnography as the only subjects with SDB in the entire corresponding questionnaire group. The overall prevalence of SDB in the entire cohort would be calculated as follows: (estimated No. of subjects with SDB among snorers and nonsnorers)/(total No. of questionnaire respondents) x 100%.
Age-adjusted and BMI-adjusted prevalence rates were calculated by categorizing subjects into three age groups (30 to 39 years, 40 to 49 years, 50 to 60 years) and whether their BMIs were < 23 and
23, where 23 is the recently proposed threshold BMI value for overweight Asians.10
Statistical Analysis
Descriptive statistics were used to summarize subject characteristics and questionnaire data. Comparison between groups was done with the Student t test (two sided) for continuous variables and
2 test for discrete variables; p < 0.05 was considered statistically significant. To adjust the effect of various factors on the likelihood of snoring and SDB developing, multiple logistic regression analysis was employed. This analysis included all variables that were found to be significant in the respective between-group comparisons (Tables 1
, 2
). Stepwise logistic regression was used to determine the principal covariates affecting SDB. Comparison of anthropometric parameters, polysomnography data, and symptoms were compared between subjects in this study and their male counterparts in the previous reported study.7
|
|
| Results |
|---|
|
|
|---|
The percentage of self-reported habitual snorers was 15%, and the mean age, BMI, and features suggestive of sleep apnea were significantly higher among habitual snorers (Table 1) . Both habitual snorers and nonhabitual snorers showed significant difference in BMI between subjects who came for polysomnography and those who did not (snorers, p = 0.046; nonsnorers, p < 0.001). Stepwise logistic regression identified history of hypertension, witnessed breathing irregularity, nasal congestion at night, BMI, and leg movements during sleep as significant correlates of habitual snoring.
Polysomnographic Data
One hundred six questionnaire respondents came for polysomnography. Ten subjects were receiving antihypertensive medications and/or antidiabetic medications, 3 subjects took sedatives/hypnotics more than five times a month, and 6 subjects were receiving miscellaneous medications, including nasal steroids, antihistamines, antiulcer drugs, and thyroxine replacement.
Of the 106 polysomnographic evaluations, 1 was rejected due to a technical fault in the recording. The distribution of sleep stages of the polysomnography-positive (AHI
5) and polysomnography-negative groups were similar, but those with AHI
5 had more snoring (p = 0.001).
Prevalence of SDB and OSAS
Among the 105 patients who underwent polysomnography, 31 subjects (30%), 16 subjects (15%), and 10 subjects (10%) had AHIs
5,
10, and
15, respectively. Due to the significant difference in BMI of the subjects who came for polysomnography and those who did not, the projection of the prevalence of SDB was based on the conservative estimate (ie, assuming no subject in the no-polysomnography group would have SDB). Since the age distribution and BMI of the questionnaire respondents were similar to those of the general middle-aged female population (30 to 60 years old) in Hong Kong, it would be acceptable to use the questionnaire respondents as the denominator of evaluation. Hence, the estimated prevalence of OSAS in the entire female cohort 30 to 60 years old would be 2.1% (Fig 1
), with the age-specific prevalence of OSAS rising from 0.5 to 6.1% (Fig 2
). At various cutoff points of AHI, estimated prevalence of SDB and OSAS were, respectively, 3.7% and 2.1% (AHI
5), 1.9% and 1.4% (AHI
10), and 1.2% and 0.8% (AHI
15) [Fig 1
].
|
|
Figure 2
shows the age-specific prevalence of SDB and OSAS in the target population. An increasing trend for SDB and OSAS with age was present, and this was seen in subjects with BMI < 23 and BMI
23 (Table 3
). The prevalence of both SDB and OSAS in subjects 50 to 60 years old was markedly higher than in subjects 30 to 39 years old, in both the normal weight and overweight subjects (Table 3)
.
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
5, and 2.1% were symptomatic with daytime sleepiness. SDB among women has received growing recognition in recent years. Population studies4
5
have demonstrated a prevalence of symptomatic obstructive sleep apnea of 1 to 2% among middle-aged white women. Our study presents the first data of SDB and OSAS in a Chinese community based on polysomnographic documentation. This study, very similar in design to that of the study7 in men, has a similar limitation of potential bias introduced by the partial questionnaire response rate and restricted polysomnography participation rate. However, the questionnaire respondents had an age distribution within the target age strata and age-specific BMIs similar to those of the local population, allowing the results to be projected to the general population. The proportion of office workers in this study sample was higher than that of the general population in the same age strata. The effect of occupation or educational level on the prevalence of sleep apnea has not been well reported. These factors might affect the occurrence of SDB indirectly, through lifestyle and behavioral patterns that in turn influence body weight control. Since the average BMI of study subjects was similar to that of the general population, these differences in demographics probably would not affect the validity of generalization of the prevalence data to the community at large.
Medications such as sedatives, tricyclic antidepressants, and stimulant drugs may alter sleep architecture and occurrence of SDB. A number of drugs, including sedatives, antidepressants, antihistamines, and lipophilic ß-blockers, may result in daytime sleepiness that confounds the symptoms of sleep apnea. Drug effect was unlikely to have affected the findings in this study, as there was no significant difference in the frequency of the use of these drugs between subjects with and without SDB/OSAS.
Since the polysomnography participants had many more associated features of SDB than those who did not undergo polysomnography, we have adopted the conservative estimate of prevalence, assuming that all those who did not undergo polysomnography did not have SDB. This is likely to result in some underestimation of the prevalence.
The estimated prevalence of OSAS at 2.1% in Chinese women 30 to 60 years old is similar to that of studies4 6 of white subjects utilizing similar diagnostic criteria. It is also comparable to that of a study13 in Asian women in Singapore utilizing a diagnostic triad of self-reported snoring, sleep symptoms of apnea/hypopnea, and/or hypertension/wide neck circumference that arrived at an estimated SDB prevalence of 2.3% in female subjects 40 to 59 years old.
As in men, higher BMI and advancing age were predictive of the presence of SDB in women. The prevalence of OSAS increased nearly 12-fold in the 50- to 60-year-old group, compared to the 30- to 39-year-old group. Although men also showed increasing prevalence with age, the escalation in the sixth decade was much less marked. It also appeared that within this middle-aged population, younger men and women with SDB were more obese than those who were in their fifties. This may reflect an increasing influence of advancing age on the development of SDB as found in other studies14 that focus on elderly population. Since women have a distinct change in hormonal status relating to menopause, which corresponds temporally to age, it is tempting to speculate that this increase in prevalence of SDB, despite a lower BMI, may be related to menopause. However, there were not enough women above or below the age threshold of 50 years with different menopause status in our study population, and analysis of the prevalence of OSAS in relation to menopausal status independent of age was not possible. It has also been reported that prevalence of SDB in postmenopausal women receiving hormonal replacement therapy was similar to that of premenopausal women,15 but again there were too few women receiving hormonal replacement therapy in this study population to make any meaningful analysis. Although the prevalence of SDB increased with age, the group mean AHI across the 3 decades was similar, which differed from the finding of higher AHI in postmenopausal women in a previous study.16
Obesity is a significant risk factor for SDB in white populations.4
6
17
18
In this study in women and the previous study7
in men, higher BMI is a major risk factor for SDB in Chinese subjects as well. Using the definition of BMI
23 in Asians as the overweight threshold,10
our subjects with SDB, whose mean BMI was 27, were definitely obese by peer comparison. In comparison with white subjects in the Wisconsin study,4
Chinese subjects had a lower risk attributable to central obesity, as indicated by the significant difference in the waist:hip ratios, and a pattern of lower risks relating to other obesity indexes such as BMI, neck circumference, and subcutaneous fat. This finding supports the notion that risk factors apart from obesity, such as craniofacial structure,19
may be of greater pathogenic importance in Chinese compared to white populations.
Apart from a lower prevalence of SDB in women, gender differences in features of OSA have been reported in studies20
21
22
of white subjects. Consistent with other data, our study showed that the severity of SDB as indicated by AHI in Chinese women was mostly in the mild range. The overall male:female ratio for OSAS was 2:1, but increased from 1.3:1 in the group with mild OSA (AHI
5 to 15) to 9:1 in the group with severe OSA (AHI
30) [data not shown]. It is also observed that women with SDB had lower AHIs compared to male counterparts despite a highly comparable BMIs between men and women. Similar to a previous Canadian study,20
AHI during REM sleep relative to the total AHI was higher in women, although this may just reflect that women had less severe obstructive sleep apnea.
The Wisconsin Sleep Cohort Study23 reported that white women with SDB had similar symptoms as those of men. In our study, Chinese women with SDB had a similar degree of snoring and daytime sleepiness as men. The gender disparity of SDB in the clinical setting24 and this community study suggests that SDB is underrecognized in women, despite similar symptoms as those in men.
| Conclusion |
|---|
|
|
|---|
| Acknowledgements |
|---|
| Footnotes |
|---|
This study was supported by the Competitive Earmarked Research Grant No. HKU457/96M from the Hong Kong Research Grants Council, Hong Kong.
Received for publication March 19, 2003. Accepted for publication August 20, 2003.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. K. Sharma, S. Kumpawat, A. Banga, and A. Goel Prevalence and risk factors of obstructive sleep apnea syndrome in a population of delhi, India. Chest, July 1, 2006; 130(1): 149 - 156. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. T. Pawlik, E. Hansen, D. Waldhauser, C. Selig, and T. S. Kuehnel Clonidine Premedication in Patients with Sleep Apnea Syndrome: A Randomized, Double-Blind, Placebo-Controlled Study Anesth. Analg., November 1, 2005; 101(5): 1374 - 1380. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Young, P. E. Peppard, and S. Taheri Excess weight and sleep-disordered breathing J Appl Physiol, October 1, 2005; 99(4): 1592 - 1599. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Kim, K. In, J. Kim, S. You, K. Kang, J. Shim, S. Lee, J. Lee, S. Lee, C. Park, et al. Prevalence of Sleep-disordered Breathing in Middle-aged Korean Men and Women Am. J. Respir. Crit. Care Med., November 15, 2004; 170(10): 1108 - 1113. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |