|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From INSERM Unit 420 (Drs. Teculescu and Hannhart, and Mr. Michaely), Vand
uvre; the Center of Preventive Medicine (Dr. Aubry and Mr. Gueguen), Vand
uvre; the Department of ENT Diseases (Dr. Montaut-Verient), Nancy University Hospital, Nancy; and the Center of Clinical Investigations (Mr. Virion), Vand
uvre, France.
Correspondence to: Dan Teculescu, MD, INSERM Unité 420, BP 184, 54505 Vand
uvre, France; e-mail: Dan.Teculescu{at}nancy.inserm.fr
| Abstract |
|---|
|
|
|---|
Design: A field survey of a sample of middle-aged men in France.
Participants: Male employees of a local university and subjects from the community attending a preventive medicine center. Participation rate was 93.5%.
Measurements: Anthropometric variables were recorded in 499 subjects aged 23 to 66 years (mean, 44.3 years). The subjects completed a standard sleep questionnaire and were classified according to the snoring frequency as never, rarely, sometimes, occasional, several nights per week, and every night. The subjects who snore occasionally represented 8.6% of the total.
Results: The anthropometric data of subjects who snore occasionally were similar to those of subjects who habitually snore. When compared with subjects who do not snore, older age and a larger neck girth were significant. Subjects who snore occasionally were also significantly more often subjects who snore loudly, and tended more frequently to have breathing stops during sleep.
Conclusions: Our epidemiologic study shows that approximately 9% of a sample of middle-aged men snore occasionally. Subjects who snore occasionally have anthropometric characteristics close to those of subjects who snore habitually. The prevalence of the main sleep-related symptoms is between that of subjects who do not snore and of subjects who snore habitually. In an epidemiologic setting, inclusion of subjects who snore occasionally as subjects who do not snore or subjects who snore habitually will lead to bias. The present results suggest they should be identified and considered as a separate category.
Key Words: breathing sleep disorder male subjects obstructive sleep apnea syndrome occasional snoring snoring frequency
| Introduction |
|---|
|
|
|---|
In laboratory studies, snoring sound has been recorded using microphones placed either on the subjects sternum or over the subjects head,7 and then is digitally processed and analyzed.8 This approach is inapplicable in field surveys, where information on snoring is usually collected using questionnaires. A series of shortcomings and limitations of such questionnaires has been described: a large proportion of "do not know" responses9 10 11 ; underreporting by subjects12 or overreporting by wives of husbands snoring13 ; or lack of standardization of the questionnaires, etc. Some authors recorded snoring frequency dichotomously,14 15 others used a 3-point scale,11 16 17 some used a 4-point scale,18 19 20 21 22 and others used a 5-point scale.5 23 24 25 Some authors report "habitual" snoring without any operational definition,26 whereas in other studies snoring remains ill-defined.27 28
As stressed by Young,13 while self-report of symptoms may be (relatively) valid in patient settings, it is less so in population samples, where "no direct health benefit of accurate reporting is present." This probably partly explains the multiplication of questionnaires, making the comparison of results between different studies very difficult. A significant standardization effort was made in 1988 by the working group of the Scandinavian Society for Sleep Research and recently by the National Institutes of Health Specialized Centers of Research in Cardiopulmonary Disorders of Sleep (Universities of Pennsylvania and Wisconsin). The frequency of the main symptoms was defined on a quantitative and subjective (so-called Likert) scale.13 The Basic Nordic Sleep Questionnaire contained 27 items in 21 questions.29 The question about snoring frequency was labeled, "Do you snore while sleeping?" with a choice of five answers: 1, never or less than once per month; 2, less than once per week; 3, on 1 to 2 nights per week; 4, on 3 to 5 nights per week; and 5, every night or almost every night. To reduce misclassification, the American version introduced a "do not know" category.
Research teams using the Basic Nordic Sleep Questionnaire30 31 32 or the North American version5 33 to evaluate snoring prevalence labeled subjects who answered 4 or 5 as "habitual snorers" and subjects answering "never" or "less than once per week" as "nonsnorers." The latter category served as a reference group for the evaluation of the risk factors for snoring. The classification of the intermediate category, ie, the subjects who snore occasionally, remained problematic, however. The subjects who snored "at least once a week, but pattern may be irregular" were included among the subjects who habitually snore by some authors,34 whereas most authors included them among the subjects who do not snore.5 35 36 37 38 39
Strictly speaking, these subjects are neither subjects who do not snore, nor subjects who habitually snore. We therefore decided to analyze the subgroup of subjects who snore occasionally in a sample of middle-aged male subjects, with two objectives: first, to see whether their anthropometric data and life habits are closer to those of groups of habitual snorers or to the subjects who do not snore; and second, to see whether the prevalence of the other OSAS symptoms are different from those of the extreme groups mentioned above.
| Materials and Methods |
|---|
|
|
|---|
20 year and sleeping with a bedmate were studied. The first group consisted of employees from a local university who attended a yearly preventive medical examination. An explanatory letter was sent to each subject, explaining the purpose of the study, stressing the noninvasive character of the protocol, and noting the right to deny participation. Three hundred thirty-four of the 357 male subjects present on the campus volunteered (93.6%). The second group was made up of male subjects from the community (northeastern France) who attended a free preventive medical check-up in a regional center of preventive medicine, serving approximately 2 million people. Of the 216 subjects who were approached (in consecutive order on the computer file) 202 subjects agreed to participate (93.5%). After a full explanation, written consent was obtained. The study protocol was approved by the Regional Committee on Ethics in Medical Research.
Anthropometry
Height and weight were measured using standard methods, and body mass index (BMI) was calculated. The circumferences of the neck (at the cricothyroid membrane level),38
waist, and hip were measured with a cloth tape, and the waist-to-hip ratio was obtained. All of the measurements were performed by the same observer.
Questionnaire
A French version of the 32-item Wisconsin University Sleep Cohort Study Questionnaire5
was completed by the subjects, with assistance from their spouses. Detailed explanations were given to the subjects; for example, subjects were urged to answer "do not know" rather than leave the question unanswered. The 3-month test-retest repeatability of the French version of the questionnaire was assessed in a subsample of the second group of subjects and was found to be satisfactory.40
For example, the question, "Have you ever been told that you snore?" had a Cohens
(concordance test) of 0.86; "How often do you seem to have momentary periods during sleep when you stop breathing or you breathe abnormally?" had a Cohens
of 0.50; and "How often do you, gasp, choke, or make snorting sounds during sleep?" had a Cohens
of 0.74.
Nose and Throat Examination
The policy of the Regional Ethics in Medical Research Committee was to exclude any invasive diagnostic procedure in epidemiologic research. Accordingly, we limited the instrument to a tongue blade and simplified the score proposed for clinical purposes by Wilms et al,41
assessing semiquantitatively the nasal septum, nasal mucosa, soft palate, uvula, tonsils, tongue, and chin abnormalities. All subjects were examined by the same observer; details of the ear, nose, and throat examination have been given elsewhere.42
Statistical Analysis
Data processing (analysis of variance, univariate analysis using Cochrane-Armitage trend test, and logistic regression)43
was performed using the Statistical Analysis System (version 8.1; SAS Institute; Cary, NC).
| Results |
|---|
|
|
|---|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
A number of weak points of this study should be acknowledged. Snoring was based on self-reporting, as is usual in epidemiologic studies. We only took into account frequency of snoring and did not analyze its duration, effect of position, or disruptiveness44 ; no count of number of snores per hour (snoring index) was obtained. Subjects with complaints of insomnia or excessive daytime sleepiness were not excluded to obtain a group of primary subjects who snore,45 nor did we exclude subjects with sleep apnea, to limit the sample to subjects who snore but do not have apnea. Finally, the number of subjects included was relatively limited, and this influenced the study power, especially when the prevalence of a symptom was low. With higher prevalence, the power was satisfactory: for example, at a 95% confidence level and a relative risk of 3.5, we had 94.9% power to detect a significant difference for loud snoring in the subjects who snore occasionally. The strengths of this study were the inclusion of a sample of subjects homogeneous with respect to age, gender, and race, and the use of a validated standard questionnaire.
The prevalence of occasional snoring in various populations is difficult to compare because of the use of different definitions, as mentioned before. In French youth, "sometimes snoring" was not different between girls (15%) and boys (15.9%),46 data that are comparable to those reported by Lugaresi et al1 in their San Marino study, which indicated a rate of 16.8% for men. Higher figures were reported from Wisconsin (31.0%),5 from the United Kingdom (47.1%),6 from Finland47 for men (60.0%), and from the United States for female nurses (65.%).17
Older age, excess weight, central obesity, and larger neck circumference are known risk factors for habitual snoring48 ; our data in habitual snorers confirm this. The quasicontinuum alteration of demographic data along the various categories of snoring (Table 1) adds a welcome reassuring validity to the self-report as an epidemiologic tool. It would be tempting, in this respect, to consider subjects who snore occasionally as a step in the natural history of snorers disease; prospective studies are necessary to test this hypothesis. Our attention was caught by the fact that the higher BMI in subjects who snore habitually was in part the result of their smaller height. Thirty-five years ago, as Fletcher et al49 described two types of COPD patients, the "pink puffer" and the "blue and bloated," they depicted the latter as stout. We wonder whether a link could exist between habitual snoring and the bronchitic type of airway obstruction in the frame of the overlap syndrome. Subjects who snore occasionally in the present study had mean anthropometric values similar to habitual snorers; the difference for subjects who do not snore was significant only for age and for neck circumference. Older age in subjects who snore occasionally was reported by Lindberg et al.50 Koskenvuo et al34 reported a higher proportion of BMI > 27, less leisure jogging, and a higher prevalence of systemic hypertension in subjects who snore occasionally compared with subjects who do not snore. The subjects who snore occasionally among the French adolescents studied by Delasnerie-Laupretre et al46 had a significantly higher BMI when compared with subjects who do not snore; the effect was present in both sexes. Scandinavian authors34 50 insisted on the role of smoking as a risk factor for snoring; in the present study, current smoking was only present as a (nonsignificant) trend.
The nose and throat examination (Table 3) showed that soft palate and uvula abnormalities are present not only in habitual snorers, but in those who snore occasionally as well. These two ear, nose, and throat findings were previously found to be risk factors for breathing pauses as well.42
We are not aware of any analysis of the prevalence of the main sleep-related symptoms in subjects who snore occasionally compared with subjects who snore habitually or subjects who do not snore. An association with loud snoring was somewhat expected in both habitual and occasional snorers. Because no direct measurement of snoring intensity was obtained in the present study, this effect could in part be caused by a better perception of snoring by this category of subjects or their bedmates. The prevalence of grasping/choking, breathing pauses, excessive daytime sleepiness, and nasal congestion/discharge at night was increased in habitual subjects who snore. When subjects who snore occasionally were taken as reference group, using a logistic regression procedure, we found habitual snorers to snore significantly more often for a long time (> 20 years; OR, 4.80), to snore loudly (OR, 2.70), and to experience difficulty in waking in the morning (OR, 8.80). Systemic hypertension was significantly more prevalent in habitual snorers (16.9% vs 8.2%) compared with subjects who do not snore, but the proportion was not increased in subjects who snore occasionally, a result at variance with the findings of Koskenvuo et al.34 Doctor-diagnosed sleep apnea was twice as frequent among subjects who snore occasionally compared with those who do not snore (2.4% vs 1.2%); however, the numbers were too small for statistical significance.
| Conclusion |
|---|
|
|
|---|
Our results suggest that occasional snoring is at present not satisfactorily defined; inclusion of subjects who snore occasionally among subjects who do not snore or who snore habitually is not justified and represents a source of bias. Until a more precise definition is adopted, it seems advisable to consider subjects who snore occasionally a special, separate group. Subjects who snore occasionally possibly represent a step in the natural history of "heavy snorers disease"51 , but only prospective studies could test this hypothesis.
| Acknowledgements |
|---|
| Footnotes |
|---|
This work was performed at the French National Institutes of Health and Medical Research (INSERM) Unit 420, Vand
uvre, France.
Received for publication August 15, 2001. Accepted for publication January 23, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Svensson, E. Lindberg, T. Naessen, and C. Janson Risk factors associated with snoring in women with special emphasis on body mass index: a population-based study. Chest, April 1, 2006; 129(4): 933 - 941. [Abstract] [Full Text] [PDF] |
||||
![]() |
Minerva BMJ, September 28, 2002; 325(7366): 724 - 724. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |