© 2004 American Thoracic Society doi: 10.1164/rccm.2408001
Snoring and Passive SmokingA Counterblaste?Woolcock Institute of Medical Research Sydney, Australia James I, king of England (James VI of Scotland), in his essay "A Counterblaste to Tobacco" was probably the first writer to draw attention to potential harmful effects of tobacco on sleep (1). Apart from noting that smoking was an "incurable stink" and "dangerous to the lungs," the royal polemicist also theorized that the heating of the body induced by tobacco was harmful as "the coldness and moisture of our brains being the only ordinary means that procure our sleep and rest." It is a matter of record that royal distaste for tobacco dwindled as the taxes levied on it increased. Only in second half of last century was it firmly established that smoking was the most damaging reversible risk factor for respiratory health. More recently, medical authorities have drawn attention to health effects of environmental tobacco exposure or passive smoking. Exposure to second-hand smoke increases the risk of lung cancer and asthma and increases the risk of fatal and nonfatal myocardial infarction by approximately 30% (2). In this issue of the Journal (pp. 799–803), Franklin and colleagues extend the link between tobacco and impaired cardiorespiratory health by showing a strong association between passive smoking and habitual snoring (3). The authors undertook a postal survey of 21,802 randomly selected subjects from six cities (five countries) in Northern Europe and obtained a 71% response rate. All measures were by self-report and the established association between habitual snoring and "active" smoking was observed. Passive smoking was defined by identifying individuals who reported never having smoked but were exposed to tobacco smoke in their home every day. Compared with a 13.3% prevalence of snoring in never-smokers, 19.8% of passive smokers reported habitual snoring. Importantly, this effect was independent of obesity and sex. In fact, one of the more controversial issues raised by the study was the proportion of snoring that could be explained by different measured risk factors (population attributable risk). Smoking accounted for an attributable risk for snoring four times higher than that of obesity (17% versus 4.3%), and even passive smoking had a similar risk for snoring as did obesity. The implication of this data for the primary management of snoring and mild sleep apnea is potentially huge. An unknown humorist once said that "Smoking is one of the leading causes of statistics." How much reliance do we place on this new data on passive smoking and snoring in adults? First, there is epidemiologic data in children supporting the association between passive smoking and snoring. The authors cite two older studies (4, 5) in children showing this link, but there are also recent data from over 5,000 children showing a potential dose-dependent effect of parental smoking on childhood snoring (6). Second, the study by Franklin and coworkers involved a large population and self-reports regarding habitual snoring, smoking, and passive smoking. Self-reporting has been shown to be reasonably reliable compared with objective data (7, 8). However, results from the smaller Wisconsin Sleep Cohort study, which used full polysomnography to measure sleep-disordered breathing, found an effect of current but not former smoking on the prevalence of sleep apnea (9). Similarly, the data showing a much higher attributable risk from smoking than obesity needs to be interpreted with caution. As the authors point out, attributable risk varies with the prevalence of risk factor exposure and is limited by the use of self-report data or incomplete response. Lack of information on alcohol consumption and the known inaccuracy of obesity self-reporting are problematic. We have no information on fat distribution, which is a stronger correlate of sleep-disordered breathing than body mass index (10). The dangers of cross-sectional data interpretation are well known—are we missing some confounding variable or is there hidden misclassification bias ? Do passive smokers report snoring more? Are their smoking partners poorer sleepers and more aware of the acoustic challenge? Others have taken an alternate view about smoking and snoring (11). This study in the Journal raises other conundrums. For example, although the surveyed population (n = 1,647) in Tartu, Estonia had the highest rate of passive smoking (13.9% of never-smokers) and second highest rate of smoking (35.3%), it had the lowest prevalence of habitual snoring (12.0%). This geographical curiosity would be contrary to the author's conclusions about smoking or snoring. It may reflect the younger age of the population (and less exposure time) or indicate some local "snoring protective factor" worthy of further research. In addition, although we may have plausible explanations for the role of obesity and sex in the pathogenesis of snoring and sleep apnea, we really do not have a clue to explain the smoking association. Smoking-induced inflammatory damage to mucosal neural protective mechanisms against snoring is one possibility (9, 12). It is also possible that there may be developmental biological aspects to the smoking–snoring relationship. The passive smoking–snoring association is present in childhood (4–6) and maternal smoking influences airway development (13). It is entirely speculative, but possible, that the predisposition for snoring is preprogrammed during fetal development. However, this would not readily explain how current environmental tobacco exposure in adults would increase the risk of sleep-disordered breathing. For the clinician, the data from Franklin and colleagues are timely and important. Too often, we pay little attention to lifestyle modification to treat snoring and sleep apnea and focus on mechanical therapies or surgery. In particular, as the authors suggest, smoking is often not acknowledged as a risk factor for sleep-disordered breathing. Smoking cessation will clearly be of general health benefit, and probably of specific benefit, in reducing snoring and sleep apnea, but objective data verifying this is needed. The dose–response relationship between smoking and snoring suggests this will be likely. Moreover, consideration needs to be given to urging the partner of a snorer to stop smoking. Snoring and sleep apnea have established adverse health effects, so the new data in adults presented in the current issue of the Journal add strength and rationale to government efforts aimed at reducing environmental tobacco exposure in the community. A "blast" of snoring should be matched by a "counterblaste" of prevention of royal proportions. FOOTNOTES Conflict of Interest Statement: R.G.'s department received $30,000 for company-initiated investigative studies for new therapeutic technology for sleep-disordered breathing from ResMed and in 2003 received a travel grant of $1,500 from Respironics to speak at a course organized by the company. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
