INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tobacco smoking is the single most important source of preventable morbidity and premature mortality in Western society (1). Cigarette smoking has been causally linked to the development of lung, laryngeal, and nasal neoplasia, chronic obstructive pulmonary disease (COPD), increased airway responsiveness, impaired pulmonary immune function, and increased pulmonary infections, notwithstanding its established link with cardiovascular mortality (2). The pathogenetic mechanisms for these smoke-induced respiratory diseases are not, however, fully understood. Recently, the ubiquitous molecule nitric oxide (NO) has been implicated in the pathophysiology of some of such diseases (3).

Nitric oxide (NO) plays a key role in the physiological regulation of many airway functions and can be rapidly and noninvasively measured in the expired air of human subjects (4). NO is formed from L-arginine by the activity of NO synthase (NOS), of which at least three isoforms exist within the human airway. Measurement of exhaled nitric oxide (eNO) has been increasingly utilized by investigators into lung diseases and is now generally accepted as a surrogate measure of airway inflammation in asthma (5). NO is raised in individuals with asthma compared with normal subjects and eNO rises during the late asthmatic response, whereas therapeutic doses of glucocorticosteroids return levels to normal (6). In addition, NO is involved in the control of the pulmonary circulation, is important for resistance to pulmonary pathogens, and has also been implicated in genotoxicity (7).

It has now been confirmed by several investigators that smoking decreases eNO in normal subjects and that levels are low in habitual smokers compared with nonsmokers (3, 8). This effect of cigarette smoke appears to be rapid and dose related (3). eNO levels rise on cessation of smoking (11). Although controversy currently exists about whether eNO is increased or decreased in COPD, there is evidence to suggest a correlation between eNO and the degree of airflow limitation as well as with quantitative measures of cigarette use (12). Currently, there is interest in the use of eNO in the investigation of smoking-related lung disease, as this may yield information about the mechanisms underlying cigarette-induced lung damage (13).

Cigarette smoke not only affects the smoker, but also those environmentally exposed. Environmental tobacco smoke (ETS) is formed from the mixture of sidestream smoke (smoke emitted directly into the air from the smouldering cigarette) and mainstream smoke (the mixture inhaled by the smoker and exhaled after lung filtration). Passive smoking has been associated with persistent middle ear effusion, respiratory illness in children, and lung cancer in nonsmokers (14). However, despite documented adverse effects of ETS on the lungs, objective methods of assessing airway damage have been lacking. eNO provides a sensitive and rapid measure of airway inflammation, and may be useful in this regard. Therefore, to assess the utility of eNO for measuring early effects of ETS exposure, and to quantify any such effect, we studied the effect of ETS exposure on eNO and compared this with the effect of active smoking in normal subjects.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Design and Subjects

The study was crossover in design, and randomized with respect to exposure order. There were three study visits (sham, passive smoke, and active smoke, respectively). Seventeen nonsmoking normal subjects with no history of any respiratory disease, or symptoms suggestive of asthma, were recruited. All had normal methacholine bronchial reactivity, and six were atopic, as defined by skin prick tests to six common aeroallergens (Table 1). None was exposed to passive smoke at home or at work, or had suffered any symptoms suggestive of an upper respiratory tract infection within the previous 6 wk. None was taking inhaled or oral glucocorticosteroids or other medication. Of the 15, only 2 had ever smoked cigarettes, and both of these had stopped smoking more than 20 yr before the study. Two subjects were excluded from the study because of failure to complete both parts of the study (development of an upper respiratory tract infection and dislike of passive smoke exposure). The study was approved by the institutional ethics committee.

Subjects were exposed to tobacco smoke over a period of 1 h in a specially designed challenge chamber, which was externally ventilated and enclosed within the air-conditioned Respiratory Laboratory. Sham inhalation consisted of measurement of background NO levels every 15 min for 60 min. Passive smoke and active smoke inhalations consisted of identical measurements during exposure to the relevant agent.

Spirometry (measured according to the recommendations of the American Thoracic Society) was performed before and after each exposure. It was impossible to "blind" subjects to smoke exposure, but eNO measurements were analyzed without knowledge of exposure details. No subject was exposed to ETS during the day prior to each study day, and each study visit was performed at an identical time of day. Of the 15 subjects, 8 agreed to actively smoke cigarettes, but of these only 7 were able to complete the exposure due to adverse effects including nausea, headache, vomiting, and nasal discomfort.

Measurement of Exhaled NO

NO concentrations were measured using a modified chemiluminescence analyzer (Dasibi Environental Corporation, Glendale, CA) sensitive to NO at 2-4000 parts per billion (ppb, vol/vol), as previously described (3). After flushing the analyzer with NO-free compressed air, NO was sampled continuously at 250 ml/min as subjects performed a slow vital capacity maneuver over 30-45 s into wide-bore Teflon tubing, creating a positive oral pressure to close the soft palate. In addition, nose clips were not used in order to decrease nasal contamination. eNO was measured at baseline, and at 15, 30, 45, and 60 min. Results were displayed on a chart recorder and compared with the signal generated from a calibration mixture of NO. Three successive peaks were recorded and the mean values were analyzed. Ambient air NO levels were also measured.

Smoke Exposure

Subjects were exposed to ETS within an enclosed, specially designed separately ventilated exposure chamber. For passive smoke exposures, smoke was generated by a smoking machine operated manually, initiating a 3-s puff every 30 s from a popular brand of cigarette. A total of seven cigarettes was "smoked" over an hour, generating CO levels of approximately 23 ppm (15). During active smoking, subjects smoked a total of seven cigarettes over 1 h, having been instructed to inhale slowly to vital capacity and to breath hold for 2 s.

Statistical Analysis

All data were expressed as means + standard errors, and eNO data log was transformed to the normal distribution. Data were compared with ANOVA and comparisons were made by corrected Student's t test. Significance was taken as a p value of <0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subject characteristics are shown in Table 1.

Spirometry

Mean forced expiratory volume in 1 s (FEV₁) values are shown in Table 2. No significant effect of any exposure was found on FEV₁.

Exhaled Nitric Oxide

Exhaled nitric oxide was measured in all subjects. Mean values are shown in Table 3. Baseline levels did not differ significantly between passive and sham inhalations (mean [SE] 119 [25] ppb versus 134 [29] ppb, respectively), but were lower in the active smoking group (71 [16] ppb).

With sham inhalation (n = 15), eNO levels did not change significantly from baseline, although a small decrease (possibly related to repeated expiration) occurred. On passive smoke exposure, eNO levels fell significantly within the first 15 min from 134 (29) ppb to 102 (22) ppb, or by 23.6% (p < 0.05), and remained low until 60 min (Figure 1, n = 15). With active smoking (n = 7), levels fell acutely from baseline within the same time interval (71 [16] ppb to 49 [11] ppb, or by 30.3%) and again remained low for 60 min. These changes were significant compared with sham exposure for both passive smoke inhalation (p < 0.05) and active smoke inhalation (p < 0.01). Although the fall in eNO was slightly greater for active than for passive smoke exposure, the difference between the two decreases did not reach significance.

View larger version (12K): [in this window] [in a new window]   Figure 1.   Effect of sham, passive, and active smoke inhalation on exhaled nitric oxide in normal subjects. Diamonds, sham; squares, passive smoke; triangles, active smoke; *p < 0.05; ppb, parts per billion; n = 15 normal subjects with sham and passive smoke inhalation; n = 7 for active smoke inhalation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our study has demonstrated that eNO levels decrease rapidly in normal subjects exposed to smoke both passively and actively, when compared with sham exposure. The effect is rapid, occurring within 15 min, and maintained for the duration of the smoke exposure. The decrease in eNO due to smoke exposure was significantly greater with both passive smoke inhalation (24%) and active smoking (30%) than with the sham exposure (4%). As would be expected, the decrease in eNO was greater with active than passive smoke inhalation, although the difference between the two methods of smoke exposure failed to reach significance. This effect was observed with levels of ETS exposure frequently experienced in the community setting (15). To our knowledge, this is the first time that the effect of passive smoke on eNO has been reported. Thus, eNO may represent a sensitive and objective parameter to assess effects of ETS exposure that cannot be documented by other techniques. Our observation may have implications for research into the health effects of ETS, particularly in children, where eNO can easily be measured and has the advantage of being totally noninvasive.

Cigarette smoke induces an acute inflammatory reaction in the airways and lung parenchyma. Increases in bronchoalveolar lavage total inflammatory cell and neutrophil counts are seen in cigarette smokers, and exposure of nonsmoking humans to ETS for 3 h results in increased circulating neutrophils, neutrophil chemotaxis, and neutrophil release of oxidants upon stimulation (16). Airway epithelial cells are likely to be key cells involved in this reaction, as they line the respiratory tract and interact directly with inhaled cigarette smoke. Airway epithelial cells are an important source of inflammatory mediators and nitric oxide. Human bronchial epithelial cell cultures exposed to smoke extract exhibit significantly greater neutrophil chemotactic activity than control cell cultures. Nitric oxide appears to be involved in this response, as pretreatment with L-N^G-monomethyl-L-arginine (L-NMMA), a nitric oxide synthase inhibitor, abolishes this increase (17). There is evidence of a similar response in humans, where inhaled L-NMMA rapidly reduces eNO (18). Although there is little information regarding the role of NO in cigarette-induced lung damage, it is probable that NO is involved in the normal response to lung injury, as eNO rises in response to viral infection and to other environmental insults such as allergen exposure. NO is also important in nonspecific host defense and has antimicrobial properties against a variety of pathogens (4).

Several studies have now confirmed that eNO levels are reduced in habitual smokers without lung disease (3, 8, 9, 12). In smokers, eNO levels correlate with cigarette consumption per day and with past smoking history. Smoking a single cigarette results in an acute decrease in eNO (3), and this has been confirmed in animal studies (19). One study, however, has reported the reverse (20), although measurements were performed only over a short period. Cessation of smoking is associated with an increase in eNO toward normal levels (11). Exposure to passive smoke has been shown to result in abnormal NO-mediated vascular responses in young adults (21). The inhibitory effect of smoke on eNO can also be demonstrated in vitro, where cigarette smoke downregulates NO production (22), possibly through an effect on NO synthase activity.

Little is known, however, about the effect of smoke inhalation on eNO in patients with lung disease. Some information is available from studies that have examined eNO in patients with chronic obstructive pulmonary disease (COPD), which is closely related to smoking. Maziak and colleagues reported eNO on 13 current smokers with COPD, 12 patients with unstable COPD, and 10 smokers with chronic bronchitis (23), and compared these with 8 patients with COPD who had stopped smoking. eNO levels were no different from normal levels in ex-smokers with COPD, but increased with unstable COPD and showed a negative correlation with lung function. Two more recent studies have shown contradictory findings in the same area (12, 24). In the study of Corradi and coworkers, eNO levels also correlated with the degree of airflow limitation (12).

Several different methods of measurement of eNO exist (3, 25). We measured eNO using exhalation against a resistance and measurement via a side arm, which minimizes nasal contamination, according to European taskforce recommendations (4). In a recent study using argon to measure nasal contamination with this technique, Kharitonov and colleagues have demonstrated that any nasal leakage is negligible (26). Several investigators have now compared the different methods of measurement of eNO, and concluded that although measurements are not interchangeable, all methods can be used to measure differences within groups (25). In our study, interindividual variation cannot account for the changes in eNO that were seen, because the study was crossover in design. Significant differences between individuals were, however, observed in baseline eNO levels. We did not assess direct or surrogate measures of smoke exposure, because these have been measured in a previous study using an identical technique of smoke inhalation (15).

The mechanism for the decrease in eNO with smoking is uncertain. Cigarette smoke is known to contain high concentrations of oxides of nitrogen and the reduction in eNO may be due to downregulation of nitric oxide synthases (NOS) by a negative feedback mechanism (3). It does not appear to be due to carbon monoxide, as inhalation of cigarette smoke in concentrations sufficient to raise exhaled CO levels to 12 ppm results in no change in eNO level (3). Similarly, inhaled CO alone does not increase eNO (3). One recent in vitro study of porcine pulmonary endothelial cells suggests that the decreased NO generation seen in response to cigarette smoke solution may not be related to a direct cytotoxity, and that the mechanism could be related to NOS gene or to NOS activity (27). Other possible mechanisms include accelerated uptake of NO, inactivation of NO by oxidants in cigarette smoke, increased breakdown of NO, or damage to NO producing epithelial cells by toxins. It is possible that NO is reacting very rapidly with superoxide, yielding the harmful oxidant peroxynitrite. Cigarette smoking is associated with induction of lung neutrophils (16), which in turn produce increased concentrations of peroxynitrite. This mechanism would be comparable to that observed in cystic fibrosis, where nitrite levels are elevated in breath condensate of such patients, whereas eNO is not (30). Nonetheless, it is possible that the decrease in eNO is related to some other substance or combination of substances, as cigarette smoke contains over 3000 chemicals.

It is therefore plausible that changes in NO are involved in the early induction of lung injury produced by cigarette smoke, and that impairment of NO production is important in the genesis of tobacco-induced lung disease. Measurement of eNO after passive smoke exposure may thus provide a novel and sensitive method of detection of early lung injury, which could be utilized for investigation of the mechanisms of cigarette-induced lung damage both in the laboratory and in an epidemiological setting.