Reportedly, snoring is associated with chronic bronchitis. This association warrants further studies including polysomnographic evaluations because of few epidemiologic studies on the association. Via a polysomnography study, we evaluated the associations of snoring, obstructive sleep apnea, and systemic inflammation with chronic bronchitis among 442 participants from a population-based cohort.
At baseline, we assessed participants’ serum levels of C-reactive protein, a biomarker of systemic inflammation. Over a 5-year period, we conducted overnight polysomnography and identified any new cases of chronic bronchitis.
After taking into account age, smoking, and other potential risk factors, the multivariate odds ratio (95% CI) for chronic bronchitis was 2.9 (95% CI, 1.3–6.4) for snorers with cumulative duration of snoring episodes ≥ 1 hour as compared with those snoring < 1 hour. This association did not change after further adjustment for the presence of apnea. Obstructive sleep apnea had no association with chronic bronchitis. A higher level of serum C-reactive protein was associated with chronic bronchitis (p value for trend < 0.05). In a joint analysis of snoring and C-reactive protein, longer cumulative duration of snoring episodes accompanied by systemic inflammation was associated with a 10-fold (95% CI, 2.9 to 37.4) increase in the multivariate odds of chronic bronchitis.
This polysomnography study provides additional data supporting the hypothesis that snoring is associated with chronic bronchitis implying that snoring-related local and systemic inflammation may play roles in the development of chronic bronchitis.
We first reported that snorers are at an increased risk of developing chronic bronchitis.
As stated in our previous study,
We carried out a cross-sectional study, embedded in a population-based cohort study within the Korean Genome Epidemiology Study (KoGES). Information regarding the KoGES’s ongoing, prospective cohort study is available elsewhere.
In our Ansan Sleep Cohort Study, which is an investigation via PSG, we first enrolled the 5,015 cohort members, using disproportional stratified sampling to allow efficient analyses. Based on the questions about presence and frequency of snoring, we stratified the cohort members into habitual snoring (snoring 4 nights per week or more) and non-habitual snoring (no snoring or snoring less than 4 nights).
Then, we randomly selected 50% of the habitual snorers (habitual snorers were 17% of all cohort members) and 10% of the non-habitual snorers, selecting a group with similar proportions to the snoring group regarding sex and number of members in each of the 10-year age range groups. Ultimately, we included 637 study participants with a participation rate of 75 percent. All participants underwent overnight PSG between August 27, 2001 and March 16, 2005. When comparing their incidences of chronic bronchitis over the 5-year period, we observed no difference between our participants and the other cohort members, who did not participate in the PSG.
For this current study, we excluded participants who reported on the baseline questionnaires that they had a physician’s diagnosis of a major disease, such as chronic obstructive pulmonary disease, cancer, cardiovascular disease, tuberculosis, and asthma (n = 85); or had experienced cough and sputum production on most days for at least 3 months in the preceding year (n = 6); who did not complete the follow-up interviews (n = 47); or whose serum C-reactive protein (CRP) levels were 4 mg/L or higher, based on assay results from the baseline health examination (n = 57). This left 442 individuals for analysis.
Chronic bronchitis was defined as the presence of cough and sputum production on most days for at least 3 months each year, for at least 2 successive years.
All participants fasted for at least 8 hours before blood collection began. Using the turbidimetric immunoassay method, we assessed participants’ serum concentrations of CRP, a systemic inflammation biomarker, at baseline. We defined the top tertile of serum CRP levels (≥ 1.8 mg/L) as a high level of systemic inflammation and the bottom 2 tertiles as a low level of systemic inflammation.
To determine the presence and severity of snoring and OSA, we conducted PSG evaluations of the participants over a 5-year period. They underwent overnight PSG at Korea University Ansan Hospital’s sleep laboratory, which has a computerized PSG device (Alice 4; Respironics, Atlanta, GA, USA). The evaluations measured the following parameters: electroencephalogram, electrooculogram, submental and leg electromyogram, electrocardiogram, airflow through the nose and mouth, chest and abdominal respiratory movement, pulse oximetry, microphone recording for any snoring sound, and body position. According to the standard criteria,
For our purposes, apnea was defined as the absence of airflow for 10 seconds, and hypopnea was defined as a discernible reduction in airflow associated with a reduction in oxygen saturation by at least 4% from the baseline. We obtained apnea-hypopnea indices (AHI, the average number of apnea-hypopnea events per sleep hour) and the cumulative duration of each participant’s snoring episodes. In this study, we defined OSA patients as those with AHIs of 5 events/hour or greater.
Trained interviewers administered the questionnaires to participants. On the 2001 questionnaire, we collected information on demographic characteristics, including age; sex; income; occupation; history of occupational exposure to dust or chemicals; marital status; education; and lifestyle factors, including smoking status, alcohol consumption, and physical activity. A previous study shows how we calculated daily alcohol consumption (g/day) and total metabolic equivalent of task per hour (MET/hour) scores.
We calculated descriptive statistics for study participants’ baseline characteristics, dividing them into 2 categories based on cumulative duration of snoring episodes (< 1 hour or ≥ 1 hour). To evaluate the statistical difference between these categories, we performed unpaired t-tests for continuous data and Kruskal-Wallis tests for categorical data. To evaluate the associations of snoring and OSA with chronic bronchitis and systemic inflammation, we conducted logistic regression analyses and estimated odds ratios (ORs) of chronic bronchitis or of a high level of systemic inflammation with a 95% CI. Using the < 1 hour cumulative snoring group as a comparison (reference), we calculated ORs for the ≥ 1 hour cumulative snoring group. Likewise, we calculated ORs for the OSA group, using the non-OSA group as a reference. Potential confounding variables in the multivariate models were age, BMI, sex, income, occupation, history of occupational exposure to dust or chemicals, marital status, education, smoking status, passive smoking (among participants who had never smoked), alcohol consumption, physical activity, average daily sleep time, and rhinitis diagnosis. Furthermore, we took serum levels of CRP into account in the multivariate models. In the models, we fitted age, BMI, and CRP as continuous variables and entered the other variables as categorical variables. We also examined the association between snoring and chronic bronchitis according to smoking status. Because approximately 77% of study participants reported having never smoked or being ex-smokers, however, we present here only data for non-smokers, not for current smokers. In the analyses, we did not categorize current smokers by smoking intensity, because only 14 persons reported smoking more than 20 cigarettes per day.
We documented 41 cases of new-onset chronic bronchitis (9.3%) during the 5-year period.
In a population-based, cross-sectional study, we evaluated snoring and OSA via overnight PSG and found a positive association between snoring and newly-identified chronic bronchitis over a 5-year period. After taking into account age, smoking, BMI, and other potential risk factors of chronic bronchitis, we observed that snorers with a longer cumulative duration of snoring episodes had a 2.9-fold (95% CI, 1.3 to 6.6) higher chance of having chronic bronchitis as compared to those snoring less than 1 hour. We also observed that higher levels of normal CRP values are associated with the development of chronic bronchitis. Furthermore, we observed that snoring had a stronger association with chronic bronchitis when combined with systemic inflammation.
In a previous study, we reported the first data on the association between self-reported snoring and chronic bronchitis, as developed over a 4-year follow-up period.
As we suggested in our previous report, the underlying mechanisms for the association between snoring and chronic bronchitis may involve local airway inflammation and systemic inflammation.
Our present study produced findings consistent with our previous report supporting the hypothesis that snoring is associated with chronic bronchitis
In summary, our PSG study provides additional data supporting the hypothesis that snoring is associated with chronic bronchitis implying that snoring-related local and systemic inflammation may play roles in the development of chronic bronchitis. Further investigations assessing lung function, as linked to snoring-related local and systemic inflammation, may be useful for un derstanding the etiology of chronic bronchitis.
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 research program 2010, of Kookmin University in Korea.
The authors have no financial conflicts of interest.
Comparison of characteristics according to cumulative duration of snoring episodes, as recorded during polysomnographic evaluation
Variables | Cumulative duration of snoring episodes
|
p-value | |
---|---|---|---|
< 1 hour | ≥ 1 hour | ||
No. (%) of participants | 318 (72.0) | 124 (28.0) | |
Polysomnographic measures (2001–2006) | |||
Cumulative duration of snoring episodes, mean ± SD (min) | 13.8 ± 17.5 | 122.3 ± 52.0 | < 0.001 |
Duration of sleep, mean ± SD (hours) | 6.2 ± 1.0 | 6.6 ± 0.8 | < 0.001 |
Apnea-hypopnea index, mean ± SD | 5.4 ± 9.5 | 12.7 ± 14.8 | < 0.001 |
Diagnosis of obstructive sleep apnea (%) |
29.3 | 54.8 | < 0.001 |
Baseline measurement (2001–2002) | |||
Age, mean ± SD (years) | 49.1 ± 7.2 | 48.0 ± 6.6 | < 0.13 |
Male (%) | 55.0 | 83.1 | < 0.001 |
Monthly wage of < 106 won (%) |
12.6 | 3.2 | < 0.01 |
Occupation (%) | |||
White-collar job | 17.0 | 19.4 | < 0.56 |
Blue-collar job | 55.4 | 67.7 | < 0.05 |
Others | 27.7 | 12.9 | < 0.01 |
History of occupational exposure (%) | |||
To dust | 20.8 | 17.7 | < 0.48 |
To chemicals | 9.4 | 10.5 | < 0.74 |
Living with spouse (%) | 90.6 | 96.8 | < 0.05 |
Educational level of > 9 years (%) | 73.9 | 81.5 | < 0.10 |
Smoking status (%) | |||
Current smokers | 18.6 | 35.5 | < 0.001 |
Former smokers | 22.6 | 28.2 | < 0.22 |
Passive smoking among never-smokers (%) | 32.6 | 42.2 | < 0.22 |
Body mass index, mean ± SD (kg/m2) | 24.9 ± 2.7 | 26.4 ± 3.0 | < 0.001 |
Alcohol consumption, mean ± SD (g/day) | 10.1 ± 22.6 | 23.7 ± 36.4 | < 0.001 |
Physical activity, mean ± SD |
24.1 ± 9.4 | 23.3 ± 8.6 | < 0.43 |
Average daily sleep, mean ± SD (hours) | 6.6 ± 1.5 | 6.4 ± 1.3 | < 0.37 |
Serum C-reactive protein, mean ± SD (mg/L) | 1.3 ± 1.0 | 1.4 ± 1.0 | < 0.23 |
Data compared between groups, using the Kruskal-Wallis test for categorical variables and the t-test for continuous variables.
During this period, we collected or assessed information on these measures.
Defined as apnea-hypopnea index ≥ 5 events/hour.
This wage corresponds approximately to the government-set minimum wage for a family of 3 persons.
Average daily metabolic equivalents per hour.
SD: standard deviation.
Association of snoring and obstructive sleep apnea with chronic bronchitis
Models for chronic bronchitis | Cumulative duration of snoring episodes
|
Apnea-hypopnea index
| ||
---|---|---|---|---|
<1 hour | ≥ 1 hour | < 5 events/hr | ≥ 5 events/hr | |
Case/Noncase | 23/295 | 18/106 | 25/256 | 16/145 |
Age-adjusted OR (95% CI) | Reference | 2.18 (1.13–4.21) | Reference | 1.15 (0.58–2.26) |
Age and BMI-adjusted OR (95% CI) | Reference | 2.45 (1.24–4.85) | Reference | 1.21 (0.61–2.42) |
Model 1: Multivariate OR (95% CI) |
Reference | 2.77 (1.26–6.10) | Reference | 1.21 (0.57–2.58) |
Model 2: Multivariate OR (95% CI) |
Reference | 2.86 (1.28–6.38) | Reference | 1.15 (0.54–2.46) |
Model 3: Multivariate OR (95% CI) |
Reference | 2.90 (1.27–6.59) | Reference | 0.94 (0.43–2.06) |
Boldface type indicates a p-value < 0.05.
In Model 1, we adjusted data for age (continuous), BMI (continuous), sex, income (monthly wage of < 106 or ≥ 106 won), occupation (white-collar, blue-collar, or other), occupational dust exposure (yes or no), occupational chemical exposure (yes or no), marital status (married or other status), education (< 9 or ≥ 9 years), smoking status (never smoked, former smoker, or current smoker), passive smoking among the “never smoked” (yes or no), alcohol consumption (non-drinkers or currently drinking alcohol, at < 5.1 g, 5.1–15 g, 15.1–30 g, or > 30 g/day), quartiles of physical activity (metabolic equivalents per hour daily), quartiles of sleep duration during polysomnography, and a physician’s diagnosis of rhinitis (yes or no).
In Model 2, we further adjusted data for serum levels of C-reactive protein (continuous) with the covariates from Model 1.
In Model 3, we further adjusted data for categories of cumulative duration of snoring episodes and the presence of obstructive sleep apnea with the covariates from Model 2.
OR: odds ratio, CI: confidence interval, BMI: body mass index.
Association between C-reactive protein and chronic bronchitis
Models for chronic bronchitis | C-reactive protein
|
p-value for trend | ||
---|---|---|---|---|
Tertile 1 | Tertile 2 | Tertile 3 | ||
Median, mg/L (range) | 0.2 (0.1–0.8) | 1.3 (0.9–1.7) | 2.3 (1.8–3.9) | |
Case/Noncase | 11/144 | 9/134 | 21/123 | |
Age-adjusted OR (95% CI) | Reference | 0.88 (0.35–2.19) | 2.24 (1.04–4.83) | < 0.06 |
Age and BMI-adjusted OR (95% CI) | Reference | 0.92 (0.37–2.29) | 2.42 (1.10–5.30) | < 0.05 |
Model 1: Multivariate OR (95% CI) |
Reference | 0.96 (0.36–2.59) | 3.01 (1.28–7.04) | < 0.05 |
Model 2: Multivariate OR (95% CI) |
Reference | 0.95 (0.35–2.62) | 3.36 (1.40–8.08) | < 0.05 |
Boldface type indicates a p-value < 0.05.
We fitted C-reactive protein as a continuous variable in the model.
In Model 1, we adjusted data for age (continuous), BMI (continuous), sex, income (monthly wage of < 106 or ≥ 106 won), occupation (white-collar, blue-collar, or other), occupational dust exposure (yes or no), occupational chemical exposure (yes or no), marital status (married or other status), education (< 9 or ≥ 9 years), smoking status (never smoked, former smoker, or current smoker), passive smoking among “never smoked” (yes or no), alcohol consumption (non-drinkers or currently drinking alcohol, at < 5.1 g, 5.1–15 g, 15.1–30 g, or > 30 g/day), quartiles of physical activity (metabolic equivalents per hour daily), quartiles of sleep duration during polysomnography, and a physician’s diagnosis of rhinitis (yes or no).
In Model 2, we further adjusted data for the categories of cumulative duration of snoring episodes and the presence of obstructive sleep apnea with the covariates from Model 1.
OR: odds ratio, CI: confidence interval, BMI: body mass index.
Joint analyses of snoring and C-reactive protein in relation to chronic bronchitis
Snoring | CRP |
Case/Noncase | Multivariate OR (95% CI)
| |
---|---|---|---|---|
Model 1 |
Model 2 | |||
< 1 hour | Low level | 10/203 | Reference | Reference |
≥ 1 hour | Low level | 10/75 | 3.62 (1.29, 10.15) | 3.65 (1.28, 10.44) |
< 1 hour | High level | 13/92 | 3.83 (1.49, 9.82) | 3.84 (1.49, 9.85) |
≥ 1 hour | High level | 8/31 | 10.34 (2.91, 36.73) | 10.43 (2.91, 37.43) |
p-value for trend | < 0.001 | < 0.001 |
Boldface type indicates a p-value < 0.05.
A low CRP level is defined as < 1.8 mg/L and a high CRP level is defined as ≥ 1.8 mg/L.
In Model 1, we adjusted data for age (continuous), BMI (continuous), sex, income (monthly wage of < 106 or ≥ 106 won), occupation (white-collar, blue-collar, or other), occupational dust exposure (yes or no), occupational chemical exposure (yes or no), marital status (married or other status), education (< 9 or ≥ 9 years), smoking status (never smoked, former smoker, or current smoker), passive smoking among “never-smoked” (yes or no), alcohol consumption (non-drinkers or currently drinking alcohol, at < 5.1 g, 5.1–15 g, 15.1–30 g, or > 30 g/day), quartiles of physical activity (metabolic equivalents per hour daily), quartiles of sleep duration during polysomnography, and a physician’s diagnosis of rhinitis (yes or no).
In Model 2, we further adjusted data for the presence of obstructive sleep apnea with the covariates from Model 1.
OR: odds ratio, CI: confidence interval, CRP: C-reactive protein.