Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Increase in saliva cotinine after three hours' exposure to second‐hand smoke in bars

Increase in saliva cotinine after three hours' exposure to second‐hand smoke in bars Abstract Objective: To determine whether measurement of cotinine in saliva is a sensitive measure of exposure to secondhand smoke (SHS) among customers in bars. Alistair Woodward School of Population Health, University of Auckland, New Zealand Jeff Fowles, Stuart Dickson, Dinusha Fernando, Richard Berezowski Institute for Environmental Science Research, Kenepuru Science Centre, New Zealand Design: Before/after comparison of saliva cotinine and subjective assessments of SHS. Papaarangi Reid Department of Public Health, Wellington School of Medicine, University of Otago, New Zealand Setting: Three bars in Wellington, New Zealand, June 2003. Participants: Eleven non-smoking medical students spent three hours in each location. They provided saliva samples before and after the visit, counted numbers of lit cigarettes in each bar, and assessed the smokiness of the venue. Samples were tested for cotinine using liquid chromatography coupled with mass spectrometry. Results: Cotinine levels post-visit were consistently higher than baseline. The mean difference was 1.03 ng/mL with a 95% confidence interval of 0.76-1.30 ng/ mL. Adjustments to post-visit levels for metabolism and clearance of cotinine made very little difference to these results. Males tended to have higher baseline levels than females, and to show smaller increases. The bar with the greatest increase in cotinine was judged to be the smokiest on the basis of averaged cigarette counts and scores for presence of smoke and odour. Conclusion: The cotinine in saliva, when tested with the analytic methods described here, provides a means of assessing relatively short-term exposures to SHS. (Aust N Z J Public Health 2005; 29: 272-5) he first evidence of the harmful effects of second-hand smoke (SHS) on health was published in the early 1 1980s. Since that time, smoking has been prohibited in many indoor public spaces. In New Zealand, these restrictions were extended by legislation passed in December 2004 to cover all workplaces. But in most countries, bars, clubs and other entertainment venues are not yet covered by smoke-free laws, and these are the public settings in which the heaviest exposures to SHS are likely to occur. Most studies of SHS in bars and clubs have focused on workers. 2-5 There are good reasons for this. Bar tenders and other staff are frequently exposed to smoke for the full duration of their working day. However, the exposures received by customers also deserve attention. Customers may spend less time in bars and clubs, per visit, than those who work in these establishments, but they contribute many more person-hours of exposure in total. Also, if there is a need to carry out rapid assessments of the smokiness of bars and clubs and similar venues, it may be more practical to work with customers than with employees. Such assessments may be required to monitor the implementation of smoke-free legislation, or to investigate the effectiveness of measures such as ventilation or voluntary smoking restrictions. It has been shown that concentrations in air of nicotine and other smoke products are raised in workplaces that permit smoking.6 However, as a means of rapid assessment, air monitoring has some disadvantages. It is relatively intrusive and, moreover, does not provide information on the dose of smoke products that is actually taken up by workers or customers. Cotinine, the principal metabolite of nicotine, is a sensitive measure of exposure to SHS,7 is related to incidence of heart disease and stroke,8 and is found in saliva at levels that correlate closely with those measured in blood.9 Previous studies have measured cotinine and tobacco-specific carcinogens in urine of patrons after brief visits to entertainment venues.10,11 However, there are practical advantages to working with saliva samples rather than urine. Therefore, in this study we set out to determine whether cotinine in saliva is a sensitive measure of exposure of customers in bars to SHS. Methods A group of 11 medical students (five males, six females) visited three bars in Wellington during June 2003. The students Submitted: August 2004 Revision requested: November 2004 Accepted: February 2005 Correspondence to: Professor Alistair Woodward, School of Population Health, University of Auckland, Private Bag 92019, Auckland, New Zealand. Fax: +64 9 373 7624; e-mail: a.woodward@auckland.ac.nz AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2005 VOL. 29 NO. 3 Smoking Increase in saliva cotinine after exposure to second-hand smoke were non-smokers, aged 22-26 years, none of whom lived with regular smokers, who agreed to avoid other exposures to SHS for the course of the study. The study sites were selected from a list of public bars in the Wellington central business district. We chose venues in the medium to large category (so that a group of 11 students would be relatively inconspicuous), that were, on the basis of casual observations, likely to contain relatively high levels of SHS. Visits lasted three hours, in the period 8pm to midnight, on a Thursday, Saturday and Wednesday nights, respectively. Staff and management of the bars were not told the purpose of the student visits. We tested three methods of estimating SHS levels. Cotinine was measured in saliva before and after each visit, and students counted the numbers of lit cigarettes in each bar and judged smokiness indoors based on visible smoke and odour. The mass spectrometer was used in the Multiple Reaction Monitoring mode. Transition ions monitored were at m/z 177¡80 for cotinine and 180¡80 for cotinine D3. All injections were 15 µL. A nine-point standard curve over a concentration range of 0 to 40 ng/mL was created by spiking 0.5 ml of Barnstead H2O with cotinine standards. The analyses were carried out over seven separate runs on seven different days. As the lower end of the calibration was considered of greater significance, the calibration was weighted using a 1/χ2 weighting. The mean correlation coefficient was 0.995. The intraday reproducibility (five replicates) of the standard (0.3 ng/mL) had a CV of 9.4%. The interday CV (three days) was 14.5%. The detection limit was 0.1 ng/mL of cotinine in 0.5 mL of saliva. Cigarette counts and perceived smoke levels On entering the bar, each student assessed the smokiness of the venue on a scale of 1-5: 1. No smoke detectable. 2. Smoke not obviously visible but able to be smelt. 3. Areas or clouds of smoke visible, but only in discrete areas of the establishment. 4. Diffuse smoke easily visible. 5. Heavy amounts of smoke leading to reduced visibility. At the same time, the students also counted all lit cigarettes in the bar and the total number of customers. Assessments were made, and recorded, independently. The measures of perceived smoke levels and lit cigarettes were repeated at 30-minute intervals during the visit to the bar. The period between counts was spent circulating in the bar. Information was collected on physical aspects of the bar and the presence or absence of non-smoking areas. Cotinine Saliva was collected prior to entering each bar and 15 minutes after leaving. We did not use salivary stimulants. Salivette tubes were used to collect the samples, as reported previously.2 The tubes, once filled with 1-2 mLs of saliva, were stored in plastic bags, kept on wet ice for several hours, and then placed on ice in the refrigerator over night. Next morning tubes were placed in a 20C freezer, where they were kept for up to two weeks before being extracted for analysis. Non-smokers, living in a non-smoking environment, carried out all extractions and extreme care was taken to ensure that contamination was eliminated. All glassware used in the analysis was pre-rinsed with methanol to ensure that no cotinine was present. Saliva (0.5 mL) was pipetted into a pre-rinsed 7 ml silanised culture tube. Each sample was spiked with 50 mL of cotinine D3 internal standard solution and capped immediately with caps with clean Teflon liners. The samples, along with the extracted standards, were then vortexed briefly and the internal standard was allowed to equilibrate for five minutes. These were basified with 0.5 mL 1.0 M sodium hydroxide, and then 3 mL ethyl acetate was added. The tubes were recapped and mixed for 15 minutes on the vortex mixer. The tubes were centrifuged and the ethyl acetate was transferred to clean culture tubes, which had been pre-rinsed with hexane. Glacial acetic acid (30 mL) was added to each tube, and then the ethyl acetate was evaporated just to dryness in Savant evaporator. The dry residue was reconstituted in 100 mL 30:70 acetonitrile and deionised water. The samples and standards were again capped with clean caps and vortexed briefly and then sonicated for 10 minutes. The samples were placed into micro inserts with rubber feet and immediately put in microvials and capped tightly. Analyses were conducted using a Shimadzu 10AVP HPLC system attached to a PE Sciex API 300 Triple Quadrupole mass spectrometer equipped with a TurboIon spray ionR source. The HPLC column used was a Phenomenex SynergiTM 4mm Polar, 2.0 x 75mm with a 4.0 x 2.0mm C18 Phenomenex SecurityGuardTM cartridge. The mobile phase was a gradient of acetonitrile and 5 millimolar ammonium acetate. 2005 VOL. 29 NO. 3 Statistical analysis The principal outcome variable was the difference between saliva cotinine levels before and after visits to each bar. In the absence of any further exposure to SHS, one would expect cotinine levels to fall due to the passage of time and metabolism and clearance of nicotine already in the body. To allow for this decay, we adjusted post-visit levels, based on cotinine concentrations before each visit, three hours elapsed time and assuming no further exposures to SHS and a first-order elimination half-life of 18 hours. It was also assumed that the students had not just come from a setting in which they were exposed to significant amounts of SHS. Cotinine levels below the detection limit were assigned a value of 0.05 ng/mL, which is half the analytical limit of detection. Initially, univariate analyses were performed to identify significant factors contributing to the change in cotinine levels. Mixed model analysis was used, using repeated measures designs, tests were carried out for multicollinearity, and Pearson correlation coefficients were calculated. These analyses were conducted with and without adjustment for metabolism and clearance of cotinine and were repeated after excluding outliers (individuals with the highest pre-exposure cotinine levels). AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Woodward et al. Article Statistical analyses were performed using the Statistical Analysis Software (SAS) System version 8.2. A p value of <0.05 was taken to be statistically significant. The study was approved by the Wellington Ethics Committee. Figure 1: Boxplots of differences in salivary cotinine levels before and after visits to three bars, with the bars ordered by smoking density (average number of lit cigarettes per maximum number of customers). Results Saliva cotinine levels ranged from not detectable to 4.85 ng/ ml. (By comparison, levels for active smokers are commonly 500 ng/mL or higher.7) Overall, cotinine levels after visiting the bars were 1.03 ng/mL higher than the readings before the visits, with a 95% confidence interval of 0.76-1.30 ng/mL. These figures changed little after adjustments for metabolism and clearance of cotinine, so only unadjusted data are presented. Figure 1 displays the differences in cotinine levels (ng/mL) for each bar. Bars 1 and 2 contained similar numbers of customers (average numbers over three hours were 53 and 44); bar 3 was busier (average 90 customers). The bars are ranked in terms of smoking density calculated as the ratio of cigarettes smoked to the capacity of the bar (the maximum number of customers was treated as a proxy for the volume of the bar) (see Table 1). The bar with the highest smoking density, which was also perceived to be the smokiest, had the greatest increase in salivary cotinine levels. Baseline cotinine levels were higher among males than females at all three visits (average of 0.62 ng/mL and 0.24 ng/mL respectively). The average post-visit cotinine level for males was 1.34 ng/mL and was 1.53 ng/mL for females. Univariate analysis showed that the greater increase for females was statistically significant (p=0.03). When this comparison was repeated using the adjusted values, the difference remained, but was now borderline significant (p=0.053) Half-hourly measures of lit cigarettes and smokiness were averaged. The mixed effects model, which takes into account repeated measures over time, confirmed that average lit cigarettes and smokiness scores were higher in bar 3 than the other venues. These differences were statistically significant (p<0.05). Information on the size of each venue was obtained in the form of maximum number of customers permitted (see Table 1). Univariate analyses suggested a non-signif icant positive relationship between change in cotinine levels and the average lit cigarettes in each bar, adjusted for capacity. Adjustments to metabolism and clearance of cotinine made no difference to these results. To investigate sensitivity of the results to outliers, the analyses were repeated after excluding individuals with the highest preexposure cotinine levels. The findings were essentially unchanged. Discussion The most important finding is that an increase in salivary cotinine is detected after three hours’ exposure to SHS. This has not been reported previously. Earlier studies have found changes in cotinine levels in bar workers over the course of an eight-hour shift.2 But demonstrating an effect after a relatively short period of exposure is new. We note also that levels of smoking in these venues were not high – we estimate no more than 5-10% of patrons were smoking at any one time. The result is consistent – only one individual, on one occasion, did not show an increase in cotinine after the visit. Moreover, the increases tended to be substantial Table 1: Saliva cotinine levels (ng/mL) before and after three bar visits (average and standard error), average smoking prevalence (number of lit cigarettes at any one time), maximum number of customers, perceived smokiness in each venue and smoking density. Bar number Pre-visit cotinine n=11 0.36 (0.10) 0.12 (0.30) 0.76 (0.30) Post-visit cotinine n=11 0.80 (0.09) 1.19 (0.10) 2.34 (0.32) Average lit cigarettes Maximum number of customers Perceived smokiness Cigarettes per max. no. of customers 0.019 (low) 0.024 (medium) 0.026 (high) AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2005 VOL. 29 NO. 3 Smoking Increase in saliva cotinine after exposure to second-hand smoke (more than a doubling of pre-visit levels) and were most unlikely to be due to chance. Our result fits with the findings of Anderson et al.,11 who reported a fourfold increase in cotinine levels in urine in non-smokers after a four-hour visit to a casino in which smoking was unrestricted. The results suggest an association between the average increase in cotinine levels and emissions from cigarettes in the bar, as measured by cigarette count and perceived smokiness. The largest absolute increases in cotinine occurred in bar 3, and this bar was judged to be smokier than the other two, and the greatest numbers of lit cigarettes were also reported at this venue. However, the questionnaire measures did not show as large or consistent a difference between the venues as was apparent in the cotinine levels. Other factors, not measured directly in this study, which may have contributed to the between-bar variation in cotinine levels include ventilation rates and the volume of the space in which smoke was disseminated. Individuals varied greatly in both their initial cotinine levels and the increase resulting from the visit to the bar. The students were asked whether they could recall exposures to SHS during the study, other than those experienced in the bars. None could recall such exposures. However, males tended to have higher levels pre-visit, and smaller increases following the visit to the bars, than females – differences between the genders were statistically significant and not explained by individual outliers. Reasons may include greater frequency of incidental exposure to SHS among males and more rapid clearance of nicotine. The rate at which cotinine is produced depends on factors that influence either the non-renal clearance of nicotine, or the proportion of nicotine cleared by the liver that is oxidised to cotinine. These metabolic processes are influenced by gender, age, ethnicity, previous smoking behaviours, and medications that induce hepatic enzymes.9 Age was not a relevant factor in this study; and we did not obtain information on other possible causes of betweenindividual variation. The fact that cotinine was detected in all samples is testament to the sensitivity of the assay (0.1 ng/mL is a conservative assessment of the limit of detection – in many samples readings were possible to much lower levels), but it demonstrates also how widespread is exposure to SHS in public places. The participants in this study were non-smokers, studying in a setting that is nominally smoke-free, and living with other non-smokers. The interval between visits was relatively short (72 hours between bars 1 and 2, 48 hours between bars 2 and 3). The half-life of cotinine in saliva is approximately 18 hours, so a longer break between measures would be ideal. However, there was no sign that the results in this study were materially affected – pre-visit levels were lower, on average, on the second occasion than the first. On the basis of these findings, we suggest that saliva cotinine is a suitable tool for rapid assessment of SHS in bars and similar locations. The method is unobtrusive (bar staff and other patrons were not disturbed by the study, and appeared unaware that it was being conducted) and provides a direct measure of uptake of SHS. It may be applied to studies of workers or exposures in the general 2005 VOL. 29 NO. 3 population, and, in our experience, the costs compare favourably with those of air monitoring. It should be feasible to use this method to study large-scale interventions such as national smokefree legislation. Further work might explore the optimal ‘washout’ period between bars, the effect of shorter periods of observation than three hours, studies in a wider range of bars to test whether there is indeed a relation between smoking density and the size of the increase in salivary cotinine, and personal factors that influence rates of production and clearance of cotinine. What this paper adds What is already known. Cotinine, a metabolite of nicotine, can be detected in the saliva following exposure to SHS. Levels are raised in restaurant and bar staff after a full shift working in spaces where smoking is permitted. What does this study add. With the analytic techniques described here, it is possible to detect increases in cotinine levels in saliva of customers after only three hours in a bar containing smokers. There is considerable variability between individuals in their response to a common environment. Nevertheless, these results suggest that saliva cotinine is a suitable tool for rapid assessment of SHS in bars and similar locations. Acknowledgements The students who participated in the study were: Kate Bartlett, Alicia Blaikie, Fiona Christian, Hannah Donaldson, Ben Griffiths, Richard Holland, Kate James, Lawrence Kim, Jo Knight, Tim Panckhurst, and Mayada Zaki. Gabrielle Davie assisted with the statistical analyses. Matthew Hosking ran the cotinine extractions. Funding was provided by the New Zealand Ministry of Health. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

Increase in saliva cotinine after three hours' exposure to second‐hand smoke in bars

Loading next page...
 
/lp/wiley/increase-in-saliva-cotinine-after-three-hours-exposure-to-second-hand-Cc1GNOncyo

References (11)

Publisher
Wiley
Copyright
Copyright © 2005 Wiley Subscription Services, Inc., A Wiley Company
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/j.1467-842X.2005.tb00767.x
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective: To determine whether measurement of cotinine in saliva is a sensitive measure of exposure to secondhand smoke (SHS) among customers in bars. Alistair Woodward School of Population Health, University of Auckland, New Zealand Jeff Fowles, Stuart Dickson, Dinusha Fernando, Richard Berezowski Institute for Environmental Science Research, Kenepuru Science Centre, New Zealand Design: Before/after comparison of saliva cotinine and subjective assessments of SHS. Papaarangi Reid Department of Public Health, Wellington School of Medicine, University of Otago, New Zealand Setting: Three bars in Wellington, New Zealand, June 2003. Participants: Eleven non-smoking medical students spent three hours in each location. They provided saliva samples before and after the visit, counted numbers of lit cigarettes in each bar, and assessed the smokiness of the venue. Samples were tested for cotinine using liquid chromatography coupled with mass spectrometry. Results: Cotinine levels post-visit were consistently higher than baseline. The mean difference was 1.03 ng/mL with a 95% confidence interval of 0.76-1.30 ng/ mL. Adjustments to post-visit levels for metabolism and clearance of cotinine made very little difference to these results. Males tended to have higher baseline levels than females, and to show smaller increases. The bar with the greatest increase in cotinine was judged to be the smokiest on the basis of averaged cigarette counts and scores for presence of smoke and odour. Conclusion: The cotinine in saliva, when tested with the analytic methods described here, provides a means of assessing relatively short-term exposures to SHS. (Aust N Z J Public Health 2005; 29: 272-5) he first evidence of the harmful effects of second-hand smoke (SHS) on health was published in the early 1 1980s. Since that time, smoking has been prohibited in many indoor public spaces. In New Zealand, these restrictions were extended by legislation passed in December 2004 to cover all workplaces. But in most countries, bars, clubs and other entertainment venues are not yet covered by smoke-free laws, and these are the public settings in which the heaviest exposures to SHS are likely to occur. Most studies of SHS in bars and clubs have focused on workers. 2-5 There are good reasons for this. Bar tenders and other staff are frequently exposed to smoke for the full duration of their working day. However, the exposures received by customers also deserve attention. Customers may spend less time in bars and clubs, per visit, than those who work in these establishments, but they contribute many more person-hours of exposure in total. Also, if there is a need to carry out rapid assessments of the smokiness of bars and clubs and similar venues, it may be more practical to work with customers than with employees. Such assessments may be required to monitor the implementation of smoke-free legislation, or to investigate the effectiveness of measures such as ventilation or voluntary smoking restrictions. It has been shown that concentrations in air of nicotine and other smoke products are raised in workplaces that permit smoking.6 However, as a means of rapid assessment, air monitoring has some disadvantages. It is relatively intrusive and, moreover, does not provide information on the dose of smoke products that is actually taken up by workers or customers. Cotinine, the principal metabolite of nicotine, is a sensitive measure of exposure to SHS,7 is related to incidence of heart disease and stroke,8 and is found in saliva at levels that correlate closely with those measured in blood.9 Previous studies have measured cotinine and tobacco-specific carcinogens in urine of patrons after brief visits to entertainment venues.10,11 However, there are practical advantages to working with saliva samples rather than urine. Therefore, in this study we set out to determine whether cotinine in saliva is a sensitive measure of exposure of customers in bars to SHS. Methods A group of 11 medical students (five males, six females) visited three bars in Wellington during June 2003. The students Submitted: August 2004 Revision requested: November 2004 Accepted: February 2005 Correspondence to: Professor Alistair Woodward, School of Population Health, University of Auckland, Private Bag 92019, Auckland, New Zealand. Fax: +64 9 373 7624; e-mail: a.woodward@auckland.ac.nz AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2005 VOL. 29 NO. 3 Smoking Increase in saliva cotinine after exposure to second-hand smoke were non-smokers, aged 22-26 years, none of whom lived with regular smokers, who agreed to avoid other exposures to SHS for the course of the study. The study sites were selected from a list of public bars in the Wellington central business district. We chose venues in the medium to large category (so that a group of 11 students would be relatively inconspicuous), that were, on the basis of casual observations, likely to contain relatively high levels of SHS. Visits lasted three hours, in the period 8pm to midnight, on a Thursday, Saturday and Wednesday nights, respectively. Staff and management of the bars were not told the purpose of the student visits. We tested three methods of estimating SHS levels. Cotinine was measured in saliva before and after each visit, and students counted the numbers of lit cigarettes in each bar and judged smokiness indoors based on visible smoke and odour. The mass spectrometer was used in the Multiple Reaction Monitoring mode. Transition ions monitored were at m/z 177¡80 for cotinine and 180¡80 for cotinine D3. All injections were 15 µL. A nine-point standard curve over a concentration range of 0 to 40 ng/mL was created by spiking 0.5 ml of Barnstead H2O with cotinine standards. The analyses were carried out over seven separate runs on seven different days. As the lower end of the calibration was considered of greater significance, the calibration was weighted using a 1/χ2 weighting. The mean correlation coefficient was 0.995. The intraday reproducibility (five replicates) of the standard (0.3 ng/mL) had a CV of 9.4%. The interday CV (three days) was 14.5%. The detection limit was 0.1 ng/mL of cotinine in 0.5 mL of saliva. Cigarette counts and perceived smoke levels On entering the bar, each student assessed the smokiness of the venue on a scale of 1-5: 1. No smoke detectable. 2. Smoke not obviously visible but able to be smelt. 3. Areas or clouds of smoke visible, but only in discrete areas of the establishment. 4. Diffuse smoke easily visible. 5. Heavy amounts of smoke leading to reduced visibility. At the same time, the students also counted all lit cigarettes in the bar and the total number of customers. Assessments were made, and recorded, independently. The measures of perceived smoke levels and lit cigarettes were repeated at 30-minute intervals during the visit to the bar. The period between counts was spent circulating in the bar. Information was collected on physical aspects of the bar and the presence or absence of non-smoking areas. Cotinine Saliva was collected prior to entering each bar and 15 minutes after leaving. We did not use salivary stimulants. Salivette tubes were used to collect the samples, as reported previously.2 The tubes, once filled with 1-2 mLs of saliva, were stored in plastic bags, kept on wet ice for several hours, and then placed on ice in the refrigerator over night. Next morning tubes were placed in a 20C freezer, where they were kept for up to two weeks before being extracted for analysis. Non-smokers, living in a non-smoking environment, carried out all extractions and extreme care was taken to ensure that contamination was eliminated. All glassware used in the analysis was pre-rinsed with methanol to ensure that no cotinine was present. Saliva (0.5 mL) was pipetted into a pre-rinsed 7 ml silanised culture tube. Each sample was spiked with 50 mL of cotinine D3 internal standard solution and capped immediately with caps with clean Teflon liners. The samples, along with the extracted standards, were then vortexed briefly and the internal standard was allowed to equilibrate for five minutes. These were basified with 0.5 mL 1.0 M sodium hydroxide, and then 3 mL ethyl acetate was added. The tubes were recapped and mixed for 15 minutes on the vortex mixer. The tubes were centrifuged and the ethyl acetate was transferred to clean culture tubes, which had been pre-rinsed with hexane. Glacial acetic acid (30 mL) was added to each tube, and then the ethyl acetate was evaporated just to dryness in Savant evaporator. The dry residue was reconstituted in 100 mL 30:70 acetonitrile and deionised water. The samples and standards were again capped with clean caps and vortexed briefly and then sonicated for 10 minutes. The samples were placed into micro inserts with rubber feet and immediately put in microvials and capped tightly. Analyses were conducted using a Shimadzu 10AVP HPLC system attached to a PE Sciex API 300 Triple Quadrupole mass spectrometer equipped with a TurboIon spray ionR source. The HPLC column used was a Phenomenex SynergiTM 4mm Polar, 2.0 x 75mm with a 4.0 x 2.0mm C18 Phenomenex SecurityGuardTM cartridge. The mobile phase was a gradient of acetonitrile and 5 millimolar ammonium acetate. 2005 VOL. 29 NO. 3 Statistical analysis The principal outcome variable was the difference between saliva cotinine levels before and after visits to each bar. In the absence of any further exposure to SHS, one would expect cotinine levels to fall due to the passage of time and metabolism and clearance of nicotine already in the body. To allow for this decay, we adjusted post-visit levels, based on cotinine concentrations before each visit, three hours elapsed time and assuming no further exposures to SHS and a first-order elimination half-life of 18 hours. It was also assumed that the students had not just come from a setting in which they were exposed to significant amounts of SHS. Cotinine levels below the detection limit were assigned a value of 0.05 ng/mL, which is half the analytical limit of detection. Initially, univariate analyses were performed to identify significant factors contributing to the change in cotinine levels. Mixed model analysis was used, using repeated measures designs, tests were carried out for multicollinearity, and Pearson correlation coefficients were calculated. These analyses were conducted with and without adjustment for metabolism and clearance of cotinine and were repeated after excluding outliers (individuals with the highest pre-exposure cotinine levels). AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Woodward et al. Article Statistical analyses were performed using the Statistical Analysis Software (SAS) System version 8.2. A p value of <0.05 was taken to be statistically significant. The study was approved by the Wellington Ethics Committee. Figure 1: Boxplots of differences in salivary cotinine levels before and after visits to three bars, with the bars ordered by smoking density (average number of lit cigarettes per maximum number of customers). Results Saliva cotinine levels ranged from not detectable to 4.85 ng/ ml. (By comparison, levels for active smokers are commonly 500 ng/mL or higher.7) Overall, cotinine levels after visiting the bars were 1.03 ng/mL higher than the readings before the visits, with a 95% confidence interval of 0.76-1.30 ng/mL. These figures changed little after adjustments for metabolism and clearance of cotinine, so only unadjusted data are presented. Figure 1 displays the differences in cotinine levels (ng/mL) for each bar. Bars 1 and 2 contained similar numbers of customers (average numbers over three hours were 53 and 44); bar 3 was busier (average 90 customers). The bars are ranked in terms of smoking density calculated as the ratio of cigarettes smoked to the capacity of the bar (the maximum number of customers was treated as a proxy for the volume of the bar) (see Table 1). The bar with the highest smoking density, which was also perceived to be the smokiest, had the greatest increase in salivary cotinine levels. Baseline cotinine levels were higher among males than females at all three visits (average of 0.62 ng/mL and 0.24 ng/mL respectively). The average post-visit cotinine level for males was 1.34 ng/mL and was 1.53 ng/mL for females. Univariate analysis showed that the greater increase for females was statistically significant (p=0.03). When this comparison was repeated using the adjusted values, the difference remained, but was now borderline significant (p=0.053) Half-hourly measures of lit cigarettes and smokiness were averaged. The mixed effects model, which takes into account repeated measures over time, confirmed that average lit cigarettes and smokiness scores were higher in bar 3 than the other venues. These differences were statistically significant (p<0.05). Information on the size of each venue was obtained in the form of maximum number of customers permitted (see Table 1). Univariate analyses suggested a non-signif icant positive relationship between change in cotinine levels and the average lit cigarettes in each bar, adjusted for capacity. Adjustments to metabolism and clearance of cotinine made no difference to these results. To investigate sensitivity of the results to outliers, the analyses were repeated after excluding individuals with the highest preexposure cotinine levels. The findings were essentially unchanged. Discussion The most important finding is that an increase in salivary cotinine is detected after three hours’ exposure to SHS. This has not been reported previously. Earlier studies have found changes in cotinine levels in bar workers over the course of an eight-hour shift.2 But demonstrating an effect after a relatively short period of exposure is new. We note also that levels of smoking in these venues were not high – we estimate no more than 5-10% of patrons were smoking at any one time. The result is consistent – only one individual, on one occasion, did not show an increase in cotinine after the visit. Moreover, the increases tended to be substantial Table 1: Saliva cotinine levels (ng/mL) before and after three bar visits (average and standard error), average smoking prevalence (number of lit cigarettes at any one time), maximum number of customers, perceived smokiness in each venue and smoking density. Bar number Pre-visit cotinine n=11 0.36 (0.10) 0.12 (0.30) 0.76 (0.30) Post-visit cotinine n=11 0.80 (0.09) 1.19 (0.10) 2.34 (0.32) Average lit cigarettes Maximum number of customers Perceived smokiness Cigarettes per max. no. of customers 0.019 (low) 0.024 (medium) 0.026 (high) AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2005 VOL. 29 NO. 3 Smoking Increase in saliva cotinine after exposure to second-hand smoke (more than a doubling of pre-visit levels) and were most unlikely to be due to chance. Our result fits with the findings of Anderson et al.,11 who reported a fourfold increase in cotinine levels in urine in non-smokers after a four-hour visit to a casino in which smoking was unrestricted. The results suggest an association between the average increase in cotinine levels and emissions from cigarettes in the bar, as measured by cigarette count and perceived smokiness. The largest absolute increases in cotinine occurred in bar 3, and this bar was judged to be smokier than the other two, and the greatest numbers of lit cigarettes were also reported at this venue. However, the questionnaire measures did not show as large or consistent a difference between the venues as was apparent in the cotinine levels. Other factors, not measured directly in this study, which may have contributed to the between-bar variation in cotinine levels include ventilation rates and the volume of the space in which smoke was disseminated. Individuals varied greatly in both their initial cotinine levels and the increase resulting from the visit to the bar. The students were asked whether they could recall exposures to SHS during the study, other than those experienced in the bars. None could recall such exposures. However, males tended to have higher levels pre-visit, and smaller increases following the visit to the bars, than females – differences between the genders were statistically significant and not explained by individual outliers. Reasons may include greater frequency of incidental exposure to SHS among males and more rapid clearance of nicotine. The rate at which cotinine is produced depends on factors that influence either the non-renal clearance of nicotine, or the proportion of nicotine cleared by the liver that is oxidised to cotinine. These metabolic processes are influenced by gender, age, ethnicity, previous smoking behaviours, and medications that induce hepatic enzymes.9 Age was not a relevant factor in this study; and we did not obtain information on other possible causes of betweenindividual variation. The fact that cotinine was detected in all samples is testament to the sensitivity of the assay (0.1 ng/mL is a conservative assessment of the limit of detection – in many samples readings were possible to much lower levels), but it demonstrates also how widespread is exposure to SHS in public places. The participants in this study were non-smokers, studying in a setting that is nominally smoke-free, and living with other non-smokers. The interval between visits was relatively short (72 hours between bars 1 and 2, 48 hours between bars 2 and 3). The half-life of cotinine in saliva is approximately 18 hours, so a longer break between measures would be ideal. However, there was no sign that the results in this study were materially affected – pre-visit levels were lower, on average, on the second occasion than the first. On the basis of these findings, we suggest that saliva cotinine is a suitable tool for rapid assessment of SHS in bars and similar locations. The method is unobtrusive (bar staff and other patrons were not disturbed by the study, and appeared unaware that it was being conducted) and provides a direct measure of uptake of SHS. It may be applied to studies of workers or exposures in the general 2005 VOL. 29 NO. 3 population, and, in our experience, the costs compare favourably with those of air monitoring. It should be feasible to use this method to study large-scale interventions such as national smokefree legislation. Further work might explore the optimal ‘washout’ period between bars, the effect of shorter periods of observation than three hours, studies in a wider range of bars to test whether there is indeed a relation between smoking density and the size of the increase in salivary cotinine, and personal factors that influence rates of production and clearance of cotinine. What this paper adds What is already known. Cotinine, a metabolite of nicotine, can be detected in the saliva following exposure to SHS. Levels are raised in restaurant and bar staff after a full shift working in spaces where smoking is permitted. What does this study add. With the analytic techniques described here, it is possible to detect increases in cotinine levels in saliva of customers after only three hours in a bar containing smokers. There is considerable variability between individuals in their response to a common environment. Nevertheless, these results suggest that saliva cotinine is a suitable tool for rapid assessment of SHS in bars and similar locations. Acknowledgements The students who participated in the study were: Kate Bartlett, Alicia Blaikie, Fiona Christian, Hannah Donaldson, Ben Griffiths, Richard Holland, Kate James, Lawrence Kim, Jo Knight, Tim Panckhurst, and Mayada Zaki. Gabrielle Davie assisted with the statistical analyses. Matthew Hosking ran the cotinine extractions. Funding was provided by the New Zealand Ministry of Health.

Journal

Australian and New Zealand Journal of Public HealthWiley

Published: Jun 1, 2005

There are no references for this article.