Baik I, Kim J, Abbott RD, Joo S, Jung K, Lee S, Shim J, In K, Kang K, Yoo S, Shin C. Association of Snoring With Chronic Bronchitis. Arch Intern Med. 2008;168(2):167-173. doi:10.1001/archinternmed.2007.8
Snoring is more prevalent in patients with chronic bronchitis than in persons without it. Few studies have examined the effect of snoring on chronic bronchitis. We prospectively investigated the association between snoring and the incidence of chronic bronchitis.
The baseline study was conducted from June 25, 2001, to January 29, 2003. Members of the study cohort consisted of 5015 male and female Korean citizens aged 40 to 69 years at baseline who participated in a comprehensive health examination and on-site interviews at Korea University Ansan Hospital. Of these, 4270 participants (52% men and 48% women) entered the analysis for the first 2-year follow-up from April 17, 2003, to February 20, 2005, and those who met the same inclusion criteria remained in the analysis for a second 2-year follow-up period from February 21, 2005, to November 17, 2006. We collected information on snoring at baseline and identified incident cases of chronic bronchitis during a 4-year follow-up period. On the baseline questionnaire, we excluded participants who reported the presence of cough and sputum production on most days for at least 3 months a year.
During 4 years of follow-up, we documented 314 cases of new-onset chronic bronchitis (27.1 cases per 1000 person-years). After taking into account age, smoking, and other risk factors for chronic bronchitis, the multivariate relative risks of chronic bronchitis were 1.25 (95% confidence interval [CI], 0.95-1.64) for persons snoring 5 times per week or less and 1.68 (95% CI, 1.17-2.42) for those snoring 6 to 7 times per week compared with never snorers (P for trend = .049). The analyses stratified by risk factors, including smoking, occupation, and body mass index, showed a stronger association among never smokers, house workers, and overweight persons. In analysis for the joint effect of smoking and snoring, the relative risks of chronic bronchitis were 1.39 (95% CI, 1.01-1.90) for nonsmoking and snoring, 2.31 (95% CI, 1.38-3.87) for smoking and never snoring, and 2.86 (95% CI, 1.91-4.27) for smoking and snoring compared with nonsmoking and never snoring.
This prospective study observed that snoring is associated with chronic bronchitis. Our findings provide support for the hypothesis that snoring influences the development of chronic bronchitis.
Chronic bronchitis is an inflammation of the lower airway accompanied by the long-standing presence of cough and sputum production.1 Some observational studies2- 4 have investigated the association between snoring and chronic bronchitis. Patients with chronic bronchitis were more likely to snore during sleep,2 and nonsmokers with chronic bronchitis were 2.2-fold more likely to snore habitually compared with nonsmokers without chronic bronchitis.3 However, the causal relationship between snoring and chronic bronchitis remains unclear, and the mechanisms underlying the association are largely unknown. It has been suggested that structural or functional changes in the airway due to inflammation may cause snoring3 and obstructive sleep apnea syndrome (OSAS).4 Conversely, repeated snoring vibrations may act as mechanical stresses, leading to increased inflammatory response in the upper airway.5 Supporting the latter hypothesis, 2 experimental studies using in vitro6 and in vivo models7 found that frequent mechanical vibrations that simulate snoring stimulate inflammatory processes in the airway. Given the association of the severity of lower airway inflammation with upper airway inflammation and with systemic inflammation,8 snoring may be linked to lower airway inflammation. In addition, airway pressure gradients, which are triggered by intermittent airway obstruction occurring in snoring, may be transmitted to the respiratory system, causing mechanical stress and resulting in inflammation.9 To evaluate the effect of snoring on the development of chronic bronchitis, we prospectively investigated the association between snoring and the incidence of chronic bronchitis.
The study cohort, part of an ongoing prospective investigation, is one of the population-based cohorts included in the Korean Genome Epidemiology Study, which in the past has been called the Korean Health and Genome Study.10 The baseline study was conducted from June 25, 2001, to January 29, 2003. Members of the study cohort consisted of 5015 male and female Korean citizens aged 40 to 69 years in 2001 who participated in a comprehensive health examination and on-site interviews at Korea University Ansan Hospital. Members of the study cohort completed an interviewer-administered questionnaire, which included questions on demographic information, medical history and health conditions, family disease history, and lifestyle. On the basis of a standardized protocol, all interviewers are trained every 2 years. Cohort members have been followed up biennially with a scheduled site visit.
Participants for the study were identified from residents of Ansan, which is located 25 miles (40 km) southwest of Seoul. According to the 2000 census, Ansan is an urban community with a population of 555 000. To enroll participants, we used a list of telephone numbers obtained from local telephone companies because the community has a high penetration rate of telephone subscribership. We conducted a 2-stage cluster sampling based on information from the governing district (termed dong) from the telephone directory and information on age and sex distribution from the 2000 census. We identified 10 957 eligible study participants through telephone contact and received an oral agreement for participation from 5792 individuals. Among 5020 participants visiting for the baseline health examination, those who did not specify a birth date or who reported a biological age of 39 years (n = 5) on the questionnaire were excluded. Thus, 5015 participants (2521 men and 2494 women) were included as eligible cohort members. The distribution of age and sex, percentage of alcohol drinking, and prevalence of hypertension, diabetes mellitus, and cerebrovascular disease were similar between participants and nonparticipants, but smokers were more likely to refuse participation in the study. Each participant signed a Human Subjects Committee–approved informed consent form before the baseline health examination.
For the analysis, we excluded participants who did not complete questions related to snoring (n = 53), whose anthropometric measurements were not completed on the health examination (n = 8), whose smoking status was not reported (n = 14), or who reported pregnancy during the follow-up (n = 1). Since the outcome measure for this study was new cases of chronic bronchitis, we excluded participants who reported physician-diagnosed chronic obstructive pulmonary disease before 2001 or who reported on the questionnaire the presence of cough and sputum production on most days for at least 3 months a year. In addition, we further excluded from the analysis participants who reported physician-diagnosed cancer, cardiovascular disease, tuberculosis, or asthma. Thus, 4270 participants (52% men and 48% women) entered the analysis for the first 2-year follow-up from April 17, 2003, to February 20, 2005, and those who met the same inclusion criteria remained in the analysis for a second 2-year follow-up period from February 21, 2005, to November 17, 2006. A follow-up rate of 71% was achieved, resulting in 11 586 person-years accrued for the analysis.
Chronic bronchitis was defined as the presence of cough and sputum production on most days for at least 3 months a year for at least 2 successive years.1,11 The 2001, 2003, and 2005 questionnaires included questions, which were used in the European Community Respiratory Health Survey,12 about respiratory symptoms and chronic bronchitis. On the basis of the response to the questions, cases of chronic bronchitis included only individuals who reported chronic bronchitis in the absence of a physician diagnosis of asthma.
Participants completed an interviewer-administered questionnaire. On the 2001 questionnaire, we collected information on snoring; demographic characteristics, including age, sex, income, occupation, history of occupational exposure to dust or chemicals, marital status, and educational level; and lifestyle factors, including smoking status, alcohol consumption, and physical activity. On the 2003 questionnaire, we collected data on rhinitis diagnosed in 2001 or before by physicians.
In detail, participants were asked if they snored. If so, they were further asked how frequently they snored (infrequently, 1 to 3 times per week, 4 to 5 times per week, or 6 to 7 times per week). A test-retest reliability study10 found a substantial agreement for the question on snoring. Habitual snoring has been described as snoring every night or almost every night in some studies13,14 but has been defined differently in others.3,15 Self-reported snoring status is prone to misclassification, resulting in a reduction in the strength of association; in particular, the categorization of intermediate levels between extreme groups of snoring frequency has been complicated.15 Moreover, repeated mechanical vibrations that simulate snoring, which induced inflammatory processes in experimental studies,6,7 may be comparable to snoring every night in humans. Thus, we created 3 categories: never, snoring 5 times per week or less (snoring infrequently, 1 to 3 times per week, or 4 to 5 times per week), and snoring every night or almost every night (6 to 7 times per week).
Daily alcohol consumption (grams per day) was calculated based on the questionnaire information on alcohol consumption in the past 30 days, average frequency of drinking occasion, amount of alcoholic beverages consumed for a typical occasion, and the volume of 1 standard drink for a specific beverage.
To obtain information on physical activity, participants were asked to report hours spent during a typical day in sleep and 5 categories of activity intensity (sedentary, very light, light, moderate, or vigorous) after interviewers gave examples of activities corresponding to each category. A total metabolic equivalent score was calculated by multiplying hours spent by metabolic equivalent values (1.0 for sleep or sedentary, 1.5 for very light, 2.4 for light, 5.0 for moderate, and 7.5 for vigorous activity), which have been determined based on the activities given for each category.16
A comprehensive health examination was conducted by health care professionals, who were trained with a standardized protocol. Height and body weight were measured to the nearest 0.1 cm or 0.1 kg without shoes, and body mass index (calculated as weight in kilograms divided by height in meters squared) was determined.
We calculated descriptive statistics for the baseline characteristics of study participants by the categories of snoring frequency. For continuous data, statistical differences of the means were evaluated using a Tukey adjustment for multiple comparisons. For categorical data, linear trends in proportions were assessed using a χ2 test for trend.
We calculated person-years from the date of the baseline health examination until diagnosis of chronic bronchitis on follow-up, death, or the date of the last health examination and interview during the 4-year period, whichever came first. We conducted pooled logistic regression analysis to estimate an odds ratio of chronic bronchitis with its 95% confidence interval (CI) and considered odds ratios estimates of relative risks (RRs).17 We presented RRs associated with potential confounding variables for snoring 5 times per week or less and snoring 6 to 7 times per week using never snorers as a comparison group (reference). We also conducted tests for a linear trend in the log odds of chronic bronchitis with increasing snoring frequency, for which snoring frequency was coded as 0 for never, 1 for infrequent, 2 for 1 to 3 times per week, 3 for 4 to 5 times per week, and 4 for 6 to 7 times per week. Potential confounding variables included in multivariate models were age, sex, income, occupation, history of occupational exposure to dust or chemicals, marital status, educational level, smoking status, passive smoking among never smokers, physician-diagnosed rhinitis, body mass index, alcohol consumption, physical activity, and average daily sleep. In the models, age and body mass index were fit as continuous variables, and other variables were entered as categorical variables. We also examined the association stratified by smoking status (never, former, or current smokers), occupation (white collar, housework, or blue collar), and body mass index (<25 or ≥25). In further analyses, we focused on nonsmokers (never smokers or former smokers) vs current smokers, white-collar job or housework vs blue-collar job, and body mass index less than 25 vs 25 or greater and examined the joint effects of these risk factors with snoring on chronic bronchitis. To reduce the number of subgroups created for the analyses of the joint effects, we combined 2 snoring groups, such as snoring 5 times per week or less and snoring 6 to 7 times per week, and compared snorers with never snorers. We used SAS procedures (SAS 9.1; SAS Institute Inc, Cary, North Carolina) to conduct the analyses.
During 4 years of follow-up, we documented 314 cases of new-onset chronic bronchitis (27.1 cases per 1000 person-years). The characteristics of the study participants are presented in Table 1. Persons snoring frequently were more likely to be older, men, working, ever exposed to chemicals, smokers, and heavier and to have consumed alcohol (P < .001).
Table 2 gives the relationship between snoring and chronic bronchitis. Snoring frequency has a positive linear association with the risk of chronic bronchitis (P for trend = .049). The multivariate RRs of chronic bronchitis were 1.25 (95% CI, 0.95-1.64) for snoring 5 times per week or less and 1.68 (95% CI, 1.17-2.42) for snoring 6 to 7 times per week compared with never snorers (Table 2).
Analyses stratified by smoking, occupation, or body mass index were conducted for the association between snoring and chronic bronchitis. In the analysis for smoking, a strong association between snoring and chronic bronchitis was observed for never smokers (P for trend = .006), whereas no significant association was found for former or current smokers (Table 3). In the analysis for occupation, a significant association between snoring and chronic bronchitis was observed among house workers (P for trend = .002). A linear trend was found for white-collar (P = .49) or blue-collar workers (P = .67), but the association was statistically insignificant (Table 3). In the analysis for body mass index, a stronger association was observed among overweight persons compared with persons of healthy weight (Table 3).
Table 4 presents the joint effects of snoring and other risk factors, including smoking, occupation, and body mass index, on chronic bronchitis. In the analysis for smoking and snoring, the multivariate RRs of chronic bronchitis were 1.39 (95% CI, 1.01-1.90) for nonsmoking and snoring, 2.31 (95% CI, 1.38-3.87) for smoking and never snoring, and 2.86 (95% CI, 1.91-4.27) for smoking and snoring compared with nonsmoking and never snoring. However, no significant association was observed in the analysis for the joint effect of occupation and snoring and for body mass index and snoring (Table 4).
This prospective cohort study found that snoring is associated with risk of chronic bronchitis. After taking into account age, smoking, occupation, and other potential risk factors for chronic bronchitis, persons who snore every night or almost every night had a 68% (95% CI, 17%-142%) higher risk of developing chronic bronchitis compared with never snorers.
The worldwide prevalence of chronic bronchitis ranges from 1% to 18% based on data reported in the last 30 years.18 Several factors are suggested to be associated with this disease. Epidemiologic findings consistently support the relationship between cigarette smoking and chronic bronchitis19- 21 or chronic obstructive pulmonary disease.22 Symptoms of chronic bronchitis are more prevalent among smokers,19,20 and the risk of physician-diagnosed chronic bronchitis increases about 3-fold among smokers compared with never smokers.21 It has also been reported that individuals exposed to dust, fumes, or gas are more likely to have symptoms of chronic bronchitis, and thus the population attributable risk of chronic bronchitis is estimated to range from 4% to 29%.23 Other potential risk factors for chronic bronchitis, such as outdoor air pollution,24 obesity,25 and respiratory infection,26 have been explored, but more data are needed.
To our knowledge, the present investigation is the sole prospective cohort study to evaluate the association between snoring and chronic bronchitis. Previous epidemiologic studies2,3 for chronic bronchitis and snoring used a cross-sectional design, and thus the causality of the association has remained ambiguous. Most studies2 have provided only descriptive information that snoring is a prevalent sleep problem in patients with chronic bronchitis. Franklin et al3 examined the joint effects of smoking and chronic bronchitis on snoring. After accounting for age, sex, and body mass index, the adjusted odds ratios for habitual snoring were 1.8 (95% CI, 1.6-1.9) for smokers without chronic bronchitis, 2.2 (95% CI, 1.7-2.9) for nonsmokers with chronic bronchitis, and 3.9 (95% CI, 3.0-5.0) for smokers with chronic bronchitis compared with nonsmokers without chronic bronchitis.3 In our study, the analysis stratified by smoking showed that the association between snoring and chronic bronchitis was more noticeable among never smokers compared with smokers, whereas the joint effect of smoking and snoring was more detrimental for chronic bronchitis than the sole effect of smoking or snoring. The stratified analysis also showed that house workers had a stronger association between snoring and chronic bronchitis than white-collar or blue-collar workers. The elevated risk among house workers might reflect the association among never smokers because most house workers are women who have never smoked. The association between snoring and chronic bronchitis was stronger among overweight persons compared with persons of healthy weight, but a synergistic effect of overweight and snoring on chronic bronchitis was unclear. Obesity is associated with elevated levels of C-reactive protein, which is a marker of systemic inflammation,27 and with habitual snoring.28 However, because a U-shaped curve was observed in our data for the association between body mass index and chronic bronchitis (data available from authors on request) and because obesity is less prevalent in Korea compared with Western countries, greater body mass index might have shown minor effects on chronic bronchitis in this study.
The mechanisms underlying the association between snoring and chronic bronchitis are largely unknown. Recent experimental studies6,7 have evaluated the hypothesis that repeated snoring vibrations may trigger airway inflammation. An in vitro study6 cultured human bronchial epithelial cells on a platform under which a loudspeaker system generated sound pressure vibrations that simulated snoring. During the 24-hour culture, similar proliferation was observed in cells subjected to vibration stimuli compared with control cells, but interleukin 8 levels were significantly elevated in the supernatant of cell cultures subjected to 12 hours and 24 hours of vibration.6 Interleukin 8 is known as a proinflammatory biomarker and has reportedly been involved in the association between smoking and chronic bronchitis.29 A subsequent study7 evaluated the hypothesis with an in vivo rat model and found that snoringlike vibrations applied to the soft-palate tissue of the airway induced messenger RNA overexpression in inflammatory proteins. Thus, biological mechanisms underlying the association between snoring and chronic bronchitis may be postulated given the association between snoring and upper airway inflammation6,7 and between upper airway inflammation and lower airway inflammation.8 In addition, systemic inflammation related to OSAS needs to be considered as another potential cause.30 Elevated measures of systemic inflammation observed in patients with obstructive sleep apnea may be due not only to obesity but also to airway inflammation.31 Using the induced sputum technique, Salerno et al9 observed bronchial inflammation among obese patients affected by OSAS. The investigators suggested that OSAS-related airway inflammation may be caused by obesity, hypoxia, and mechanical stress.9 During an obstructive apnea event, respiratory efforts against a closed airway occur and induce fluctuations of intrathoracic pressure, which may act as mechanical stresses and play a role in enhancing inflammation.9
Strengths of our study include its population-based design, the interviewer administration of a questionnaire, and prospective follow-up of cases of chronic bronchitis. In the interpretation of our findings, however, the fact that the study used self-reported information for exposure and outcome data needs to be taken into account. Cases of chronic bronchitis were documented based on the questionnaire data, without confirmation from a physician diagnosis. Thus, the incidence of chronic bronchitis might be overestimated, although potential overreporting of this outcome is unlikely to be important. The information on snoring was not verified. Individuals without bed partners might be more likely to underreport snoring, but the differential classification of exposure is unlikely. The presence of unknown and uncontrolled confounding factors cannot be ruled out as a further limitation, whereas their effect on the study findings is likely minimal, and our analyses have taken into account a broad range of potential confounding factors.
This study included at baseline residents of Ansan, which is an urban community built in the middle of an industrialized region. The levels of overall air pollution in this city are as high as in Seoul or other urban cities of developed countries.32 The male smoking prevalence in Korea is among the highest in the world,33 whereas habitual snoring is less prevalent in our cohort compared with data on Western populations.10,34 Thus, our data on the incidence rate and the estimated risks of chronic bronchitis may not be generalizable to other populations, but our findings may provide novel information that snorers are at greater risk of developing chronic bronchitis than persons who do not snore during sleep.
In summary, our findings provide support for the hypothesis that snoring is associated with chronic bronchitis. Further investigations are needed to confirm the association between snoring and chronic bronchitis and to explore the mechanisms underlying the association.
Correspondence: Chol Shin, MD, PhD, Department of Internal Medicine, Korea University Ansan Hospital, 516 Gojandong Danwongu, Ansan 425-707, Korea (firstname.lastname@example.org).
Accepted for Publication: August 31, 2007.
Author Contributions: Dr Shin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Baik, Kim, Abbott, and In. Acquisition of data: Joo, Jung, Lee, Shim, Kang, and Shin. Analysis and interpretation of data: Baik, Abbott, and Yoo. Drafting of the manuscript: Baik and Abbott. Critical revision of the manuscript for important intellectual content: Baik, Kim, Abbott, Joo, Jung, Lee, Shim, In, Kang, Yoo, and Shin. Statistical analysis: Baik and Abbott. Obtained funding: Baik and Shin. Administrative, technical, and material support: Joo, Shim, In, Kang, Yoo, and Shin. Study supervision: Kim, Jung, Lee, Shim, In, Kang, Yoo, and Shin.
Financial Disclosure: None reported.
Funding/Support: This study was supported by a grant from the Korea Centers for Disease Control and Prevention (budgets 2001-347-6111-221, 2002-347-6111-221, 2003-347-6111-221, 2004-347-6111-213, 2005-347-2400-2440-215, and 2006-347-2400-2440-215) and by the Japan Society for the Promotion of Science.
Role of the Sponsor: The sponsor of the study had no role in study design, data collection, data analysis, interpretation of data, or writing of the report.
Additional Contributions: We thank the reviewers for helpful comments and also thank members of our research team, Misook Bae, BS, Minjung Cha, BA, Sukyung Jang, BS, Kyungsuk Kim, BA, Hyoyung Lee, BS, Sunmi Lee, BA, Hyunjoo Ryu, BS, and Aeyeon Yun, BS, for their dedication to the Korean Genome Epidemiology Study.