&EFA
United States
Environmental Protection
Agency
R«Marcft and Dvwloomcnt
Health Effects
Research Laboratory
Research Triangle Park
EPA/600/1-90/008
March 1989
•
Indoor Air — Health
Human Exposure
and Dosimetry
of Environmental
Tobacco Smoke
-------
EPA/600/1-90/008
March 1989
Indoor Air — Health
Human Exposure and Dosimetry
of Environmental Tobacco Smoke
by
Joellen Lewtos
Genetic Toxicology Division
Health Effects Research Laboratory
Contributing Authors
Carl G. Hayes, George Goldstein, Judy Mumford,
Randall Watts, Larry Claxton
Health Effects Research Laboratory
Albert Collier, Frederick W. Henderson
University of North Carolina, Dept of Pediatrics
Goran Lofroth
Distinguished Visiting Scientist to EPA
from Nordic School of Public Health
S. Katherine Hammond
University of Massachusetts Medical School
John F. McCarthy, John D. Spengler
Harvard School of Public Health
David B. Coultas, Johnathan M. Samet
University of New Mexico Medical Center
U.S. EnYironmental Protection Agency
Office of Research and Development
Health Effects Research Laboratory
Research Triangle Park, NC 27711
-------
Human Exposure and Dosimetry of Environmental Tobacco Smoke
Abstract
Environmental tobacco smoke (ETS) is the largest source of indoor air
pollution and is also the major combustion source contributing to total human
exposure to nutagens and carcinogens. This report provides specific data on
human exposure to ETS using biochemical and mutagenesis bioassay measures as
well as physical/chemical markers of exposure. Controlled laboratory chamber
studies were used to determine ETS emission factors for mutagenic activity
using three bioassays. The emission factors for alkenes (e.g., 1,3-butadiene)
and aldehydes (e.g., formaldehyde), which are either known or potential
carcinogens, are also reported for the first, time in these papers. Human
exposure concentrations and dosimetry are reported and compared under both
controlled chamber and actual indoor environmental conditions. Nicotine
exposure and its major metabolite, cotinine, have been found to be useful
quantitative and semiquantitative measures of human exposure and dosimetry.
The relationship between nicotine exposure and urinary cotinine excretion has
been studied in pre-school children exposed in their homes and in adults
exposed on commercial airline flights. Air monitoring studies in residences
and public indoor areas using both nicotine and mutagenic activity have
demonstrated that separation of smokers into separate areas does not achieve
an ETS-free or genuine nonsmoking area unless there is both physical
separation and separate ventilation.
-------
TABLE OF CONTENTS
Page
1
Executive Summary
2
Background and Rationale
Approach
Emissions and Exposure Characterization
Biological Markers of Exposure and Dose 6
Results 10
Characterization of ETS
Assessment of Public Exposure to Environmental Tobacco
Smoke Using a Mutagenicity Bioassay of Airborne
Particulate Matter ''
Evaluation of Urinary Cotinine as a Biomarker
of ETS Exposure in Children ''
Variability of Measures of Exposure to Environmental
Tobacco Smoke in the Home 12
Application of the Nicotine/Cotinine Exposure/Dosimetry
Method for Assessment of ETS on Airline Flights 13
Questionnaire Assessment of Lifetime and Recent Exposure
to Environmental Tobacco Smoke Using Urinary Cotinine
Summary of Major Findings
Exposure ' **
Dosimetry '"
Disclaimer '"
List of Publications Summarized 19
ill
-------
EXECUTIVE SUMMARY
Environmental tobacco smoke (ETS) is the largest source of indoor air
pollution and is also the major combustion source contributing to total human
exposure to mutagens and carcinogens. This report provides specific data on
human exposure to ETS using biochemical and mutagenesis bioassay measures as
well as physical/chemical markers of exposure. Controlled laboratory chamber
studies were used to determine ETS emission factors for mutagenic activity
using three bioassays. The emission factors for alkenes (e.g., 1,3-butadiene)
and aldehydes (e.g., formaldehyde), which are either known or potential
carcinogens, are also reported for the first time in these papers. Human
exposure concentrations and dosiraetry are reported and compared under both
controlled chamber and actual indoor environmental conditions. Nicotine
exposure and its major metabolite, cotinine, have been found to be useful
quantitative and semiquantitative measures of human exposure and dosimetry.
The relationship between nicotine exposure and urinary cotinine excretion has
been studied in pre-school children exposed in their homes and in adults
exposed on commercial airline flights. Air monitoring studies in residences
and public indoor areas using both nicotine and mutagenic activity have
demonstrated that separation of smokers into separate areas does not achieve
an ETS free or genuine nonsmoking area unless there is both physical
separation and separate ventilation. This data provides information to
Congress as mandated by the Indoor Air Program and to OAR for information
needed in developing policy and guidance information which will be provided to
the public, local, state, and national governmental agencies and other
specific groups (e.g., ventilation engineers, etc.) on how to achieve indoor
air quality.
1
-------
BACKGROUND AND RATIONALE
The Environmental Tobacco Smoke (ETS) Research Program it a part of EPA's
Indoor Air Research Program. The overall goal of Che ETS research program is
to conduce the research necessary to provide information to the public,
governmental agencies, the building industry, employers, and others in order
to allow them to make well informed choices regarding the control of exposure
to ETS. In order to accurately assess the relationship between human exposure
and the resulting risk from that exposure, one of the goals of this program is
to determine the relationship between ETS exposure, uptake, and ultimate
tissue dose in order to assess the relationship between exposure and risk,
Exposures to ETS have been assessed by questionnaires, air monitoring,
modeling, and biological markers. The simplest and yet the least precise and
reliable method of exposure assessment has been the use of simple
questionnaires (e.g., "If you are a nonsrooker, do you live with, work with, or
have regular contact with persons who are smokers?"), Such questions have
been the basis for classifying individuals into broad categories of exposure,
however there are serious difficulties in developing uniform questions that
elicit unambiguous and correct replies, and even more difficult problems in
using these replies to make quantitative estimates of exposure,
Questionnaires are particularly difficult to use to estimate an integrated
exposure over many years, yet this is the primary method which has been used
to approximate such long*term exposures. The NAS Committee on ETS (1986)
recommended that future epidemlologic studies should Incorporate into their
design several different exposure assessment methods in order to assess
exposure to ETS more accurately and to estimate dose.
-------
Personal exposure and doiimetry of ETS is dependent upon so many factors
that the optimal assessment should thaoratloally b« measured diractly through
tha usa of biological markara that accurately indicate uptake and/or doae in
physiological fluids, tiaauaa or calls. Several chemicals found in body
fluids of active smokers have been evaluated as biological markers of exposure
to ETS.
Nicotine and its metabolite, cotlnina, measured in saliva, blood or urine
have been the most useful biological markers of recent ETS exposure since they
are derived virtually exclusively from tobacco products. Urinary cotlnine
levels have been shown to increase in non-smokers with increasing number of
smokers in the home for all age groups (infants, children and adults).
Currently there is difficulty in interpreting the relative cotlnine levels In
nonsmokers compared to smokers because of the reported slower clearance of
cotlnine In nonsmokers and the lack of good uptake and clearance data for
nonsmokers of different ages, sax and genetic background. The NAS Committee
on ETS (1986) recommended that absorption, metabolism, and excretion of ETS
constituents, including nicotine or cotinine, be carefully studied in order to
evaluate whether there are differences between smokers and nonsmokers in these
factors, Further epidemiologie studies using biological markers are needed to
quantify expoaure-dose relationships in nonsmokers.
Several other potential biological markers which have been evaluated as
Indicators of ETS exposure Including thioeyanate, carboxyheraoglobln and
exhaled CO are not sufficiently sensitive to moderate or low levela of ETS
exposure to be generally useful, Since there are several other sources of CO
-------
in the environment that equal or exceed the contribution made by ETS, this
marker is even less useful. The use of nicotine and cotinine measurements in
urine and possibly saliva are recommended as the best methods now available
for quantifying human exposure to ETS, however, research is urgently needed to
improve the quantitation and interpretation of these markers. Nicotine is not
an ideal marker for all constituents of ETS. In ETS polluted environments,
nicotine is present in the vapor phase as a free base, thus its uptake by the
passive smoker may not be representative of the uptake of acidic and neutral
smoke components from the vapor phase nor of any component in the particulate
phase. Other suggested biological markers of exposure include N-
nitrosoproline, nitrosothioproline, and some of the aromatic amines that are
present in high concentrations in side stream smoke as well as
3-vinylpyridine, solanesol and other tobacco specific constituents. Thus,
future studies should be concerned with developing techniques to measure the
uptake by nonsmokers of various other types of tobacco-specific ETS components
which would be representative of the particulate organic phase of ETS and the
volatile acidic and neutral phases. Studies are needed to develop and apply
highly sensitive methods (e.g., immunoassays or postlabelling) for measuring
DNA and protein adducts of tobacco-specific chemicals.
-------
APPROACH
Emissions and Exposure Characterization
Research is being conducted in chamber studies to determine the emission
factors and reactivities of specific ETS constituents. The components
currently being measured in these studies include RSP (as well as the size
distribution of the aerosol), nicotine, particulate organic mass,
mutagenicity, semi-volatile organics, aldehydes, alkenes (and other
hydrocarbons), CO and nitrogen oxides. Future studies will expand the organic
analysis to include nitrosamines and other components being considered for
tracer compounds. The results from these studies are critical to efforts to
model human exposure to ETS. Effects of various factors such as room size,
temperature, humidity, air-exchange rate, numbers of cigarettes smoked and
surface materials will be determined.
Exposure models developed as a part of EFAs indoor air research program
will be employed together with data from the emission factors determined in
the chamber studies to predict indoor air exposures. These models and
emission factors will be evaluated in test home studies under realistic indoor
exposure conditions prior to field validation studies.
Source receptor modeling procedures will be developed for apportioning
RSP, mutagens and specific organics to ETS in indoor exposure conditions using
the tracers nicotine, solenosol, C-32 (anteiso) HC and other candidate tracers
selected as described above. RSP, mutagens and PAHs originating from other
-------
sources (e.g., kerosene heaters, wood & coal burning, etc) would be
apportioned using inorganic tracer methods previously developed as a part of
other EPA air pollution research programs.
Biological Markers of Exposure and Dose
Although cotinine is currently the best biological marker of human
exposure to ETS, research is needed to improve the interpretation of cotinine
measurements in body fluids. Specifically, needed research is being conducted
to determine the absorbtion, metabolism and excretion of nicotine and its
metabolites. Specific studies either in progress or in planning stages
include: (1) Determination of the dose of nicotine absorbed from ETS by
simultaneous chamber exposure to ETS and infusion of dueterated-nicotine in
adults. (2) Continuation of studies of adults, and children of various ages,
including infants, from homes where ETS is present to determine cotinine
clearance rates and to compare exposure, uptake and dosimetry using nicotine
and its metabolites. (3) Establish relationships between personal air
exposure to RSP, mutagens and nicotine to measured nicotine intake and
nicotine metabolites in body fluids for different exposure conditions and
population groups.
A workshop was convened to evaluate the biological and analytical methods
for cotinine and the recommendations will be implemented in collaboration with
other governmental agencies using cotinine as a biological marker for ETS
exposure. A standard reference materials laboratory is being established to
provide reference chemicals and standardized reference body fluids applicable
-------
to all cotinlne assays. Guidelines for a quality assurance program for
cotinine analysis have been recommended and an intercollaboratlve comparison
study designed and recommended specifically for quantitating cotinine at
levels appropriate to ETS exposures (rather than smoking exposures). EPA will
collaborate with CDC, NCI, and the developer of a new monoclonal immunoassay
for cotinine (Langone, et al., 1986) to implement further improvements in this
assay and apply the assay in collaborative and comparative studies.
Research will be conducted to evaluate candidate marker compounds for the
particulate phase of ETS and possibly the acidic or neutral gaseous phase.
Highest priority will be given to marker compounds which are preferably unique
to ETS and if possible, genotoxic components which would form either protein
(e.g., hemoglobin) or DNA adducts. Candidate compounds under consideration
are the tobacco-specific nitrosamines, solanesol, polyphenols (e.g.,
chlorogenic acid, rutin) and other compounds for which DNA-adduct and/or
hemoglobin adduct methods are already available.
Research to develop and evaluate ETS specific biological markers of
exposure and dose through the use of protein and DNA adducts is being
conducted. Initially, postlabelling methods developed by Randerath, et al.
(1986) and Gupta et. al. (1986) are being applied to blood cells, placental
tissue, buccal cells, and lung cells from smokers and ETS exposed individuals.
Although standard adducts [e.g.. B(a)P-DNA adducts and tobacco specific NNK-
DNA adducts] are used as reference standards, the postlabelling method allows
the detection and quantification of adducts formed in DNA after exposures to
complex mixtures such as ETS without knowledge of the specific adduct formed
-------
and without human exposures to radiolabelled materials. These methods are now
able to detect 1 adduct in 1010 nucleotides, the sensitivity needed to detect
environmental exposures to indoor air pollutants such as ETS.
In collaboration with the UNC-EPA clinical group, we have completed a
pilot study to determine the relative adduct levels for human lavaged
macrophages isolated from smokers and nonsmokers. DNA adducts were detected
in both groups, however, heavy smokers were shown to have higher levels when
compared to nonsmokers. It is not clear whether DNA isolated from light to
moderate smokers has DNA adduct levels above control nonsmokers. Due to the
small number of cells isolated from nonsmokers, we are currently collecting
additional samples to establish a base line level for the control group. We
plan to modify the solvents used in the assay to resolve any nonpolar adducts
that might otherwise go undetected using the standard procedures described in
the original ^P-pos^Iabeiing method.
Exposure-dosimetry studies in progress at the Frank Porter Graham Child
Development Center on 40 pre-school children are designed to understand the
exposure-dose-effect relationships in infants and children exposed in the home
to ETS. Studies on school age children have been conducted in collaboration
with Harvard and the Univ. of New Mexico to assess exposure to ETS using
biological markers will be continued with the addition of improvements in
cotinine and cotinine QA methods.
Studies to evaluate urinary mutagens as markers of ETS exposure under
controlled exposure conditions have been completed. Although the sensitivity
8
-------
of the methods used have been improved more than tenfold, the urinary concent-
rations are near the detection limits. It appears that an ETS specific
mutagen or mutagenic fraction of the urine would have to be isolated and
quantitated to provide the sensitivity and specificity required in such a
biological marker.
-------
RESULTS
Characterization of Environmental Tobacco Smoke
EPA conducted an ETS characterization study in both a human exposure
chamber and an indoor environment. This study characterized both exposure
concentrations and airborne yields for particulate matter and its mutagenic
activity as well as nicotine, aldehydes and alkenes. Environmental tobacco
smoke (ETS) was analyzed with respect to several components following smoking
of research cigarettes in the chamber. Parameters analyzed and their airborne
yield per cigarette (shown in parentheses) included: particulate matter (10
mg) and its mutagenic activity in a Salmonella bioassay, carbon monoxide (67
mg), nitrogen oxides (2 mg). nicotine (0.8-3.3 mg), formaldehyde (2 mg),
acetaldehyde (2.4 mg), acrolein (0.56 mg), benzene (0.5 mg) and several
unsaturated aliphatic hydrocarbons (e.g., 1,3-butadiene) of which isoprene
(3.1 mg) had the highest yield. ETS from commercial cigarettes were likewise
analyzed in the experimental chamber and at a public location. The relative
component composition for ETS is similar when generated from either research
or commercial cigarettes. All components analyzed were present at
concentrations above the background concentrations. Isoprene might be
utilized as a tobacco smoke tracer for unsaturated aliphatic hydrocarbons.
This study provided documentation that the chamber ETS exposure was comparable
to that which people would encounter in indoor environments where tobacco is
being smoked. Additional chemical analyses and subsequent studies were
conducted to relate the chamber ETS components to the analysis of ETS in an
indoor environment.
10
-------
Assessment of Public Exposure to Environmental Tobacco Smoke Using a
Mutaeenicitv Bloassay of Airborne Particulate Matter
Airborne particulate matter has been collected by personal samplers in
public indoor areas and travel situations with environmental tobacco smoke
pollution. Following extraction, the samples were assayed for mutagenicity in
the presence of S9 with a sensitive microsuspension test using Salmonella
TA98. The mutagenic responses of indoor air from public areas were much
higher than those of ambient outdoor air. Depending on the circumstances, the
mutagenic response varied in trains and airplanes but the results show that
physical separation of non-smoking sections from smoking sections is necessary
in order to achieve genuine non-smoking areas. Chemical fractionation and
mutagenicity assay of the basic fraction show that Salmonella mutagenicity of
airborne particulate matter might be used as a tobacco smoke-specific
indicator, as the basic fraction of environmental tobacco smoke contains a
large part of the mutagenic activity. This is not the case for outdoor
ambient airborne particulate matter and many other combustion emissions.
Evaluation of Urinary Cotinine as a Biomarker of ETS Exposure in Children
The extent of correlation between urine cotinine and home air nicotine
levels was examined in 27 children who attended a research day care program
where they were not exposed to ETS during the daytime hours. Average
concentrations of nicotine in home air were determined by active air sampling
during the evening and night hours on two consecutive days. Urine samples for
cotinine and creatinine determinations were collected before, during, and
after the two sampling periods. In addition, four sequential weekly urine
11
-------
samples for cotinine were obtained from study children to determine the extent
to which single determinations of cotinine were representative for individual
children. Fifteen children resided in homes with smokers and 12 in homes
without smokers. Urine cotinine consistently distinguished most exposed and
unexposed children. However, three exposed children had urine cotinines which
clustered routinely around the criterion (30 ng cotinine/mg creatinine) which
best distinguished exposed and unexposed children. In children exposed to ETS
in the home, there was a significant correlation between average home air
nicotine levels and the average logarithm of urine cotinine the two mornings
following the home air monitoring periods (r - 0.67; p - 0.008). In study
children, urine cotinines were remarkably stable over the one month
observation period. Rank correlation coefficients for sequential weekly
determinations of cotinine were consistently greater than r - 0.88; p <
0.0001.
Variability of Measures of Exposure to Environmental Tobacco Smoke in the Home
The variability of four markers of environmental tobacco smoke exposure
was assessed in 10 homes with 20 nonsmoking and 11 smoking household members.
The study included obtaining exposure questionnaires, saliva and urine for
cotinine, and air particle samples for respirable particles and nicotine on 10
sampling days: every other day over 10 days, and then one day every other
week over 10 weeks. The mean concentrations of respirable particles in the 10
homes ranged from 32.4 ng/u? to 76.9 pg/nP and concentrations of nicotine
ranges from 0.59 pg/w? to 6.85 jig/m3. A linear regression model that included
indicator variables for the number of smokers exposed to in the home and the
season, and the number of hours of exposure as a continuous variable explained
12
-------
nine and 6% of the variability of the respirable particle and the nicotine
concentrations, respectively. The individual mean cotinine levels
standardized to urinary creatinine concentration, ranged from 3.89 ng/mg Cr to
55.77 ng/mg Cr. A linear regression model that included the number of smokers
exposed to in the home, the season, the age group, and the number of hours of
exposure explained 8% of the variability of the urinary cotinine levels.
Because of the marked variability of these measures, multiple measurements are
needed to establish a stable profile of exposure to environmental tobacco
smoke in a particular home or individual. Furthermore, detailed questions to
quantitate exposure offered little additional information beyond whether the
subject was exposed or not.
Application of the Nicotine/Cotinine Exposure-Dosimetry Method for Assessment
of ETS on Airline Flights
The National Cancer Institute and EPA jointly conducted a pilot research
study to measure environmental tobacco smoke exposure on typical commercial
flights at the request of The Surgeon General of the Public Health Service.
This study was undertaken (1) to measure nicotine levels in ambient air during
flights of approximately four hours' duration and urinary cotinine levels at
various points during the three days after the flights, and (2) to determine
if these exposure and excretion measurements correlate with each other and
with acute symptoms experienced during the flights.
In-flight exposure to nicotine, urinary cotinine levels, and symptom
self-reports were assessed in a study of nine subjects (five passengers and
four attendants) on four routine commercial flights each of approximately four
13
-------
hours' duration. Urine samples were collected for 72 hours following each
flight. Exposures to nicotine measured during the flights using personal
exposure monitors were found to be variable, with some nonsmoking areas
attaining levels comparable to those in smoking sections. Attendants assigned
to work in nonsmoking areas were not protected from smoke exposure. The type
of aircraft ventilation was important in determining the levels of in-flight
nicotine exposure.
The levels of ETS that occurred during the four-hour flights led to
increased levels of cotinine in the urine of both passengers and attendants.
Subjects who experienced the greatest in-flight nicotine exposure generally
had the highest levels of urinary cotinine and continued to excrete cotinine
for 72 hours after the flight. The shape and time course of the decay pattern
are consistent with a first-order pharmacokinetic decay process following an
initial exposure to nicotine. The peak level of cotinine excreted is related
to the dosage of nicotine received over the range of exposures encountered.
Reports on dose-response data under conditions of exposure are sparse,
especially for the nicotine concentration range typically encountered by
nonsmokers under free-living conditions. This analysis provides estimates of
the response to a bolus of ETS, delivered over a four-hour period, shown by a
subsequent increase in urinary cotinine excretion over time synchronized
across subjects. This study expands upon previous studies employing single-
point estimates of cotinine or self-reported smoke exposure levels and
provides information on the shape of the excretion curve, delay to peak,
amplitude to the peak, approximate functional form, and decay time of cotinine
excretion after ETS exposure.
14
-------
NCI measured changes in eye and nose symptoms between the beginning and
end of the flights. These symptoms were significantly related both to
nicotine exposure during the flight and to the subsequent urinary excretion of
cotinine. In addition, subjects' perceptions of annoyance and smokiness in
the airplane cabin were also related to in-flight nicotine exposure and
urinary excretion measures.
Questionnaire Assessment of Lifetime and Recent Exposure to Environmental
Tobacco Smoke Using Urinary Cotinine
In a sample of 149 adult volunteers recruited in New Mexico in 1986, a
study was designed to assess the reliability of questions on lifetime exposure
to tobacco smoke in the home and also compared urinary cotinine level with
questionnaire report of environmental tobacco smoke exposure during the
previous 24 hours. The agreement of responses obtained on two occasions
within six months was high for parental smoking during childhood: 94% for the
mother and 93% for the father. For smoking by the spouse during adulthood,
agreement was 100%. However, responses concerning amount smoked and hours
smoked in the home were less reliable. For the amounts smoked by the mother
and father during the index subject's childhood, the agreement between the two
interviews was 52% and 39% respectively. For hours smoked in the home, the
Spearman correlation coefficients also indicated only moderate reliability
(r- 0.18 for maternal smoking and r - 0.54 for paternal smoking). For each
set of interviews, responses concerning recent tobacco smoke exposure and
urinary cotinine level were correlated to only a modest degree. It was
concluded that adults can reliably report whether household members smoked
during their childhood, but information on quantitative aspects of smoking is
reported less reliably.
15
-------
SUMMARY OF MAJOR FINDINGS
Exposure
These studies confirm that ETS is a major source of indoor air pollution.
Methods were developed and applied to indoor environments to quantitate human
exposure to the following pollutants from ETS:
• nicotine
• mutagenicity
• alkenes (e.g., 1,3-butadiene)
• aldehydes (e.g., formaldehyde)
Exposure concentrations of these pollutants, as well as emission rates, have
been determined for use in assessment and guidance documents as well as future
exposure assessment and modeling studies.
Dosimetry
Urinary cotinine, a nicotine metabolite, was demonstrated to be a useful
biological marker of human exposure to nicotine from ETS. The results of an
interagency workshop sponsored by EPA recommend steps that should be taken to
standardize and assure the quality of cotinine data. Research in progress
will significantly improve our ability to determine nicotine dosimetry from
ETS exposure.
Exposure-dosimetry studies were conducted both in the normal indoor
exposure environments (day care and home) and controlled chamber environments
16
-------
of preschool children. A similar study was conducted on a commercial airline,
These studies show that urinary cotinine is a reliable and semi-quantitative
biological marker of ETS exposure in young children and adults.
17
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the
United States Environmental Protection Agency. It has been subjected to the
Agency's peer and administrative review, and it has been approved for
publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
18
-------
LIST OF PUBLICATIONS SUMMARIZED AND INCLUDED
1. Human Exposure to Mutagens from Indoor Combustion Sources. J. Lewtas,
L. Claxton, and J. L. Mumford. Indoor Air '87. Vol. 1, B. Seifert, H.
Esdorn, M. Fischer, H. Ruden, J. Wegner, Eds, Oraniendruck GmbH, Berlin,
pp. 473-477 (1987).
2. A Genotoxic Assessment of Environmental Tobacco Smoke Using Bacterial
Bioassays. L.D. Claxton, R.S. Morin, T.J. Hughes, and J. Lewtas.
Mutation Research. 222:81-99 (1989).
3. Characterization of Environmental Tobacco Smoke. G. Lofroth, R. Burton,
L. Forehand, K. Hammond, R. Seila, R. Zweidinger and J. Lewtas.
Environmental Science and Technology. 23:610-614 (1989).
4. Environmental Tobacco Smoke: Mutagenic Emission Rates and Their
Relationship to Other Emission Factors. J. Lewtas, K. Williams, G.
Loforth, K. Hammond, B. Leaderer. Indoor Air '87. Vol. 2, B. Seifert, H.
Esdorn, M. Fischer, H. Ruden, J. Wegner, Eds, Oraniendruck GmbH, Berlin,
pp. 8-12 (1987).
5. Mutagenic Determination of Passive Smoking. P.I. Ling, G. Lofroth, and
J. Lewtas. Toxicology Letters. 15:147-151 (1987).
19
-------
6. The Effect of Solvent and Extraction Methods on the Bacterial
Mutagenicity of Sidestrearn Cigarette Smoke. R.S. Morin, J.J. Tulis, L.D.
Claxton. Toxicology Letters. 18:279-290 (1987).
7. Public Exposure to Environmental Tobacco Smoke. 6. Lofroth, F.I. Ling,
and E. Agurell. Mutation Research. 202:103-110 (1988).
8. Variability of Measures of Exposure to Environmental Tobacco Smoke in the
Home. D.B. Coultas, J.M. Samet, J.F. McCarthy, and J.D. Spengler.
Proceedings of the APCA Conference on Indoor Air, Niagra Falls, 1988.
9. Questionnaire Assessment of Lifetime and Recent Exposure to Environmental
Tobacco Smoke. D.B. Coultas, G.T. Peake, and J.M. Samet. American
Journal of Epidemiology. 130:338-347 (1989).
10. Cotinine Analytical Workshop Report: Determination of Cotinine in Human
Body Fluids as a Measure of Passive Exposure to Tobacco Smoke. R.R.
Watts, J.J. Langone, G.J. Knight and J. Lewtas. Environmental Health
Perspectives. 84:173-182 (1990).
11. Elimination of Urinary Cotinine in Children Exposed to Known Levels of
Side-Stream Cigarette Smoke. George M. Goldstein, Albert Collier, Ruth
Etzel, J. Lewtas, N. Haley. Indoor Air '87. Vol. 2, B. Seifert, H.
Esdorn, M. Fischer, H. Ruden, J. Wegner, Eds, Oraniendruck GmbH, Berlin,
pp. 61-67 (1987).
20
-------
12. Home Air Nicotine Levels and Urine Cotinine-Greatinine Ratios in
Preschool Children. F.W. Henderson, H.F. Reid, R. Morris, 0-L. Wang.,
P.C. Hu, R.W. Helms, L. Forehand, J. Mumford, J. Lewtas, N.J. Haley, and
S.K. Hammond. American Review of Respiratory Diseases. 140:197-201
(1989).
13. Passive Smoking on Commercial Airline Flights. H.H. Mattson, G. Boyd, D.
Byar, C. Brown, J.F. Callahan, D. Corle, J.W. Cullen, J. Greenblatt, N.J.
Haley, S.K. Hammond, J. Lewtas, and W. Reeves. Journal of the American
Medical Association. 261:867-872 (1989).
21
-------
HITMAN EXPOSURE TO MUTAGKNS FROM INDOOR COMBUSTION SOURCES
Joelien Lewtas, Larry D. Claxton and Judy L. Mumford
USEPA, Research Triangle Park, NC 27711, U.S.A.
Abstract
We have measured human exposure to rautagens, using indoor medium-
volume samplers and personal samplers, in targeted field studies of homes
in the U.S. The combustion sources included in these studies were wood-
stoves, fireplaces, gas appliances, cooking, and tobacco smoking. These
studies demonstrate that the presence of environmental tobacco smoke (ETS)
consistently results in human exposure to mutagens which are significantly
higher than outdoor air or non-smoking indoor spaces. The mutagenic emis-
sion rates from the other indoor combustion sources (e.g., kerosene heat-
ers) as determined in chamber studies are more variable than ETS and are
dependent on the combustion source design and operation. Woodstoves and
fireplaces result in higher concentrations of mutagens outdoors, which may
indirectly influence the concentration of mutagens indoors.
Introduction
Combustion sources are known to emit carcinogenic chemicals and ac-
count for most of the mutagenicity associated with particulate matter in
the ambient outside air (3). Apportionment studies of the mutagenicity for
these sources show that automotive emissions and residential heating are
the major sources of mutagens associated with ambient outdoor particulate
matter (4). Time-activity studies show that we spend more than 80% of our
time indoors in primarily two types of indoor environments, home and work.
In order to evaluate human exposure to mutagens, it is necessary to evalu-
ate indoor as well as outdoor sources of mutagens. The sample mass requir-
ed for standard mutagenesis bioassays exceeds that amount conveniently col-
lected in either indoor air or personal exposure studies. Recently the ap-
plication of mutagenesis microassay methods has facilitated the evaluation
of human exposure to mutagens. The purpose of these studies is to assess
the potential human exposure to mutagens from different indoor combustion
sources.
Methods .
Sampling and Analysis
Two medium volume indoor air samplers have been used in these studies.
The 0.22 m3/min (8 CFM) sampling system used in the Columbus, Ohio study
was equipped with a filter to collect total suspended particulate (TSP)
followed by a polyurethane foam (PUF) adsorbent cartridge to collect semi-
volatile organic (SVO) matter. The details of the sampling system as used
-------
in this residential indoor air study in Columbus, Ohio are reported else-
where (5). The 0.11 nr/min (4 CFM) sampling system used in the woodstove
study in Raleigh, NC has a PM-JO size selective inlet (to exclude particles
>10 um) followed by a filter and an organic adsorbant cartridge usually
containing XAD-2 to collect the semi-volatile organics. The details of the
development of this sampling system are described by Mumford et al. (7).
Studies are in progress to compare PUF and XAD-2 as adsorbants for semi-
volatile mutagens indoors. The filter and PUF cartridge samples were
Soxhlet extracted with methylene chloride and 5% ether/hexane respectively
prior to solvent exchange into dimethylsulfoxide for bioassay analysis.
Microsuspension Mutagenicity Assay
The bioassays used in these studies both employed microsuspension
modifications of established bioassay protocols in Salmonella typhimurium
TA98 (6) and TM677 (9). The TM677 forward mutation assay modification has
been described elsewhere (5). In this assay both 8-azaquanine resistant
(mutant) colonies and surviving colonies are counted and the mutant fre-
quency (MF) is the number of mutants per 10^ surviving cells. The TA98
reverse mutation assay is a microsuspension assay described by Kado et al.
(2). This assay also utilizes (8) the nitroreductase deficient strains
TA98NR and TA98DNP6 which are diagnostic for nitroaromatics.
The bioassays were performed with or without the addition of *roclor
1254-induced male Sprague-Dawley rat liver S9 mixture prepared as described
in the above-referenced protocols. All assays were performed at 3-5 doses
and the linear slope of the dose-response curve was used to determine the
•mutagenicity, expressed as either revertants per m^ air, or per ug organ-
ic mass. In all assays both solvent controls (negative) and positive con-
trols were assayed simultaneously as described in the protocols (2, 6, 9).
Results and Discussion
The use of microsuspension bioassays has made it possible to measure
the mutagenicity of indoor air as shown in Table 1. In the absence of any
combustion sources, the mutagenicity associated with particulate-phase
organic matter (PPOM) was similar inside and outside the homes. Across all
of these homes there were elevated concentrations of semi-volatile organics
(SVO) which were mutagenic, resulting in mutagenic exposures indoors which
were twice as high as outdoors. Fireplaces in operation did not result in
higher concentrations of PPOM mutagenicity; however, one of these three
homes had higher concentrations of SVO mutagenicity. The concentration of
PPOM mutagenicity per m^ was greatest in homes with cigarette smoking and
was highly correlated to the number of cigarettes smoked as we have recent-
ly reported (5).
The SVO mass and mutagenicity concentration averaged 3.5 times higher
indoors when compared to outdoors. Neither the SVO mass nor mutagenicity
were correlated with the number of cigarettes smoked and did not appear to
be related to any of the major combustion sources. Although certain cook-
ing activities (e.g., frying bacon) significantly increased the concentra-
tion of PPOMs, the kitchens did not show substantially higher concentra-
tions of SVO when compared to other rooms in the house.
-------
The influence of outdoor PPOM mutagen concentrations on indoor air is
seen in Table 1. The outdoor rautagenicity of the two electric homes was
significantly (14 times) less than outside the gas homes. This may account
for the 10 times lower PPOM rautagenicity observed in these two electric
homes. Table 2 shows the same effect in a neighborhood where woodstove
emissions were shown to increase the outdoor mutagenicity at least 10
times. Under these conditions the concentration of mutagens indoors also
increased but only up to five times. Alfheim and Ramdahl (1) have also
shown that woodstoves and fireplaces result in increased mutagenicity
outdoors.
Table 1: Influence of combustion sources on indoor air mutagenicity in a
winter residential study in Columbus, Ohio
Major
No.
Particulate-Phase
Organics (PPOM)
Combustion Homes Ia
Source ( s )
Moned
Gas3
Gas /Fireplace'
Tobacco Smoke
Average
High*
2
2
3
5
1
(MF/m3)
.13
1.33
.68
5.64
26.70
0°
(MF/m3)
.18
2.53
2.84
2.13
2.73
I/O0
0.7
0.5
0.2
2.6
9.8
Semi-Volatile
Organics (SVO)
I
(MF/m3)
28.2
16.3
32.4
15.2
9.3
O
(MF/m3)
11.1
10.3
2.4
4.9
5.6
I/O
2.5
1.5
13.5
3.1
1.7
Total
I/O
2.3
0.8
1.3
2.7
8.4
Average
12
2.76
2.05 1.3
21.9
6.2
3.5
1.9
alndoor sample; ^outdoor sample; cratio of indoor/outdoor mutant frequency
(mutants/10^ survivors)/m3; "all electric homes; enatural gas appliances,
both central heat and cooking; *fireplace was burning wood; ''kitchen sample
taken from 7:00 a.m. to 3:00 p.m. in a home where 58 cigarettes were smoked
in a 24 hr. day.
Table 2: Exposure to PPOM mutagenicity in a woodstove impacted neighbor-
hood in Raleigh, NC
Date
2/27
2/26
2/13
2/14
2/20
2/21
Community Site
1.0
1.7
3.6
4.7
5.7
11.1
Relative Mutagenicity*
Outside Home
1.0
3.1
5.1
11.2
7.5
16.3
Inside Home
1.0
1.0
2.9
3.0
4.6
5.1
aMutant frequency (MF) per m3 for each location was divided by the MF/m3
for 2/27.
-------
Zn order to examine the mutaganlc emission rates from appliances, we
have collaborated on studies to examine the organic pollutant emissions
from kerosene heater* (11) • Table 3 summarize* the mutagenicity of theee
emissions from initial studies of two heater type*, radiant and convective.
Kerosene heater* have previously been reported to emit highly rmitagenic
nitrated polyeyelie aromatic hydrocarbon* (NC^-PAH*) and dinitropyrene* in
particular (10).
Table 3» Mutaganicity of kerosene heater emi**ion*
Micro*u*pen* ion
Bioa«»ay 89 Radiant Maltuned Conveetive
TA98 (rev/ug)
TA«»8WR (rev/ug)
TA98/1,8DNP6 (rev/Mg)
TM677 (MF/Ug/ml) +
Erai**ion Rate*
TA98 (rev/hr)
TM677 (MF/hr)
Filter
5.1
2.4
0.8
1.9
2,500
930
XAD-2
0.33
neg
neg
neg
17,490
mm
Filter
0.«3
0.42
0.22
4,300
1,500
XAD-2
neg
neg
0.10
--
3,800
•Determined from GRAV emi**ion rate* given in Traynor et al., 1986 (11).
The 84% decreace in mutagenicity observed in TA98/1,8DMPg *ugge*t*
that dinitropyrene* may account for much of the PPOM mutagenicity observed
in the radiant heater*' emi**ion*. The decreased activity observed in
TA98MR *ugge*t* that other H02-PAH* (e.g., 1-nitropyrene) are al*o impor-
tant. The nitro-compounda, including 1-nitropyrene, 3-nitrofluoranthene,
9-nitroanthracene, and 1-nitronapthalene, were detected from theae heater*
(11). We have al*o found an increase of PPOM mutagenicity in aeveral home*
with kerocene heater* (unpublished data).
Further studies are needed to meaaure mutagenic emi**ien rate* and
human exposure* to PPOM and SVO mutagan* emitted from indoor combustion
sourcee and to determine the contribution of combustion sources vented
outdoors on the resulting Indoor air exposures.
Disclaimer
The research described in this paper has been reviewed by the Health
Effects Research Laboratory, U.S. Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Ageney nor does mention
of trade names or commercial products constitute endorsement or recommenda-
tion for use.
-------
Raferencea
1. Alfheim I., and Ramdahl T. Contribution of wood combustion to indoor
air pollution ae meaaurad by mutagenicity in Salmonella and polycyclic
aromatic hydrocarbon concentration. Environmental Mutaganeeia 6
(1984), 121-130.
2. Kado, K.Y., Langley, D., Eisenetadt, E. A simple modification of the
Salmonella liquid incubation assayt Xncreaeed sensitivity for detect-
ing mutagena in human urine. Mutation Reaearch 121 (1983), 25-32.
3. Lewtae, J. Combuation emiaaionat Characterization and comparlaon of
their mutagenic and carcinogenic activity. In H.F. Stich (Ed.)/ Car-
cinogens and Mutagena in the Environment, Volume V, The Workplace:
Sourcea of Carcinogens. Vancouver! CRC Press, (1985) pp. 59-74.
4. Lewtas, J., and Williams, K. A retrospective view of the value of
short-term genetic bioassays in predicting the chronic effects of
dieael, aoot. In W. Zshinishi, A. Koizumi, R.O. McClellan and W.
fttober (Eds.), Carcinogenicity and Mutagenicity of Dieael Engine
Exhaust. Amsterdam! Elsevier, (1986), pp. 119*140.
5. Lewtas, J., Goto, I., Williams, K., Chuang, J.C., Peteraen, B.A., and
Wilson, N.K. The mutagenicity of indoor air particles in a residen-
tial pilot field studyi Application and evaluation of new methodolo-
gies. Atmoapheric Environment 21 (1987), 443-449.
6. Maron, D.M., and Ames, B.M. Revised methods for the Salmonella
mutagenicity teat. Mutation Reaearch 113 (1983), 173-215.
7. Mumford, J.L., Harris, D.B., Williams, K., Chuang, J.C., and Cooke, M.
Development of a medium-volume sampler for indoor air sampling and
mutagenicity studiea. Environmental Science and Technology, in preaa.
8. Rosenkranz, H.8., and Mermelatein, R. Mutagenieity and genotoxicity
of nitroarenest All nitro-contalning chemicala were not created
equal. Mutation Reaearch 114 (1983), 217-267.
9. Skopek, T.R., Liber, H.L., Krolewski, J.J., and Thilly, w.G. Quanti-
tative forward mutation aaaay in Salmonella typhimurium uaing fl-aza-
guanine reaiatance aa a genetic marker. Proceedings National Academy
Science 75 (1974), 410-414.
10. Tokiwa, T., Makagawa, R., and Horikawa, K. Mutagenie/carcinogenlc
agenta in indoor pollutantai The dinitropyrenes generated by keroaene
heaters and fuel gaa and liquid petroleum gaa burners. Mutation Re-
aearch 157 (1985), 39-47.
11. Traynor, G.W., Apte, M.O., Sokol, H.A., Chuang, J.C., and Mumford,
J.L. Selected organic pollutant emlaaiona from unvented keroaene
heatera. Proceedings of the 79th Annual Meeting of the Air Pollution
Control Association 52.5 (1986).
-------
Mutation Research. 222 (1989) 81-99
Elsevier
81
MTR 02102
A genotoxic assessment of environmental tobacco smoke
using bacterial bioassays-
Larry D. Claxton \ Randall S. Morin2, Thomas J. Hughes 3 and Joellen Lewtas '
' MD-68, Genetic Toxicology Division, Health Effects Research Laboratory, U.S. Environmental Protection Agency.
Research Triangle Park, NC 27711 (U.S.A.), ' U.S. Army Health Services Command, Ft. Sam Houston, TX 78234 (U.S.A.),
and 3 Research Triangle Institute, Research Triangle Park, NC 27709 (U.S.A.)
(Received 25 June 1987)
(Revision received 12 October 1987)
(Accepted 16 October 1987)
Keywords: Environmental tobacco smoke; Salmonella typhimurium; Indoor environments; Chemistry, Semi-volatiles; Monitoring
Summary
Recently, the National Research Council in the U.S.A. stated that laboratory studies of environmental
tobacco smoke (ETS) should be important in identifying ETS carcinogens, their concentrations in typical
daily environments, and in understanding how these compounds contribute to ETS dose-response
relationships. This paper demonstrates that integrated chemical and bacterial mutagenicity information
can be used to identify ETS genotoxicants, monitor human exposure, and make comparative assessments.
Approximately 1/3 of the ETS constituents for which there is quantitative analytical chemistry informa-
tion also have associated genotoxicity information. For example, 11 of the quantitated compounds are
animal carcinogens. Work presented in this paper demonstrates that both the nonparticle-bound semi-
volatile and the particulate-bound organic material contain bacterial mutagens. These ETS organics give
an equivalent of - 86000 revertants per cigarette. In addition, this article summarized efforts to estimate
ETS bacterial mutagenicity, to use bacterial tests for the monitoring of ETS-impacted indoor environ-
ments, and to use bacterial assays for the direct monitoring of human exposure.
Environmental tobacco smoke (ETS) is the total
tobacco smoke found in an environment and in-
cludes both sides tream cigarette smoke and the
exhaled tobacco smoke of the smokers within the
environment. The first suggestion that environ-
mental tobacco smoke (ETS) could have detrimen-
tal health effects was a medical case report
published by Rosen and Levy in 1950. This report
Correspondence: Dr. L.D. Claxton, MD-68, Genetic Toxicol-
ogy Division, Health Effects Research Laboratory, U.S. En-
vironmental Protection Agency, Research Triangle Park, NC
27711 (U.S.A.).
concluded that an infant's severe astmatic symp-
toms were directly related to the mother's smoking
of tobacco products. It was not until 31 years later
that the results from epidemiological studies of
passive smoking and lung cancer were available.
Three studies were published in 1981. Based on a
population of 91 540 nonsmoking Japanese
housewives, Hirayama (1981) reported that the
wives of heavy smokers (> 20 cigarettes/day) had
2.4 times the risk of developing lung cancer as the
nonsmoking wives of nonsmokers. Trichopoulos et
al. (1981) reported a slightly larger risk for non-
smoking Greek women whose husbands smoked
0165-1218/89/S03.50
-------
82
more than 20 cigarettes per day. Although
Garfinkel (1981) found a small (10-30%) increase
in mortality associated with passive smoking, his
study of 469000 nonsmokers did not detect a
statistically significant increase in risk of develop-
ing lung cancer. Since that time, a number of
other epidemiologic studies have emerged (Na-
tional Research Council, 1986). The National Re-
search Council (1986) concluded that the summary
estimate of increased risk of lung cancer ranges
from 10% to 34%. They also concluded that al-
though bias may contribute to the results, the best
estimate at present for increased adjusted risk of
lung cancer to nonsmokers due to passive ex-
posure is approximately 25%. In identifying needed
scientific information, the National Research
Council (1986) stated: 'Laboratory studies should
be important in determining the carcinogenic con-
stituents of ETS and their concentrations in typi-
cal daily environments and in facilitating under-
standing of possible dose-responsive relation-
ships.' The purpose of this paper is to partially
fulfill this need by integrating chemical and
bacterial genetic bioassay information concerning
sidestream and indoor air (IA) tobacco smoke.
Materials and methods
Integrating known chemical and genotoxicity infor-
mation for sidestream cigarette smoke
Although it is expected that most of the com-
pounds found in mainstream tobacco smoke (MS)
would also be found in sidestream smoke (SS), the
purpose of this effort was to extract from the
literature compounds identified specifically in SS
and to associate this information with presently
available genotoxicity information. As part of an
earlier effort to identify airborne compounds and
genotoxicity information (Graedel et al., 1986),
confirmed compounds found in tobacco smoke
and indoor air were cataloged. This summary pro-
vided the baseline information for this tabulation.
In addition, an on-line Chemical Abstracts search
was used to locate relevant papers published since
January, 1983. The compounds identified in these
papers were summarized, and the MS, SS, and/or
indoor air (IA) concentrations were tabulated.
Where available, the bioassay summary informa-
tion as reported by Graedel et al. (1986) was
paired with the identified chemicals. All computer
literature searches were kindly supplied by the
Resource Information Center, U.S. Environmental
Protection Agency, Research Triangle Park, NC
(U.S.A.) via Ms. Libby Smith. Although the search
periods covered a time period extending to De-
cember 1986, the information is not meant to be
exhaustive but representative of the available liter-
ature.
Methods for determining the Salmonella mutagenic-
ity of organic extracts from sidestream tobacco
smoke particles
After being lit, a generic U.S. brand of filter
cigarettes was allowed to burn within a 0.04-nr1
Plexiglass* chamber into which filtered air was
allowed to enter. The generated SS was continu-
ously exhausted at a rate of 0.03 m3/min and
collected on a 142-mm Teflon*-coated glass-fiber
filter. Sample preparation was done in the manner
reported by Morin et al. (1987). The method can
be summarized as follows. All filter samples were
extracted using 2 IS-min sonications (Constant
Temperature, Sonicor** waterbath sonicator. Bay
Shore, NY) using either dichloromethane, metha-
nol, or acetone as the solvent. Extracted samples
were concentrated to 5-10 ml using rotatory
evaporation. Samples were then transferred into
15-ml volumetric tubes and concentrated to 1-2
ml using dry nitrogen. After adding 15 ml of
dimethyl sulfoxide (DMSO), the remaining solvent
was removed using the nitrogen purge method.
Negative controls were prepared in the same
manner using filters not containing ETS particles.
The solvent-exchanged samples were tested using
the Salmonella typhimurium plate incorporation
assay as previously described (Ames et al., 1975;
Claxton et al., 1987).
Direct comparisons of mainstream and sidestream
tobacco smoke bacterial mutagenicity
In order to compare the total genotoxic poten-
tial of MS and SS it was necessary to bioassay the
particle-bound, semi-volatile, and volatile com-
pounds emitted. In order to accomplish this, sep-
arate trapping trains for MS and SS were used.
Each train consisted of an ethanol bubbler solvent
trap, a sand trap, and a liquid-nitrogen cold trap
in sequence (Monteith et al., 1986). Cigarette sam-
-------
83
pies were generated using a smoking machine. The
standard conditions were 1 puff/min for each of
the 30 cigarettes (1R3 Kentucky Reference), a
puff duration of 2 s, and a puff volume of 35 ml.
Mainstream smoke was pulled directly into the
MS trapping train. In order to collect the SS, the
cigarette smoking machine was enclosed within a
Tedlar* bag which had a total volume of 0.12 m3.
Air from the bag was continuously drawn into the
SS trapping trains. The ethanol bubbler collected
primarily semi-volatile compounds, the sand trap
collected primarily particulate matter, and the cold
trap collected any remaining organic compounds.
The sand trap and semi-volatile concentrates were
bioassayed in sealed dram vials using a modified
version of the preincubation protocol of Yahagi et
al. (1975). Preincubation was for 15 min at 37 °C
while vials were shaken by a rotary shaker at 60
rpm. Tests were done using 500 /xI/plate of 5%
Aroclor-1254-induced male Syrian hamster liver
homogenate. Volatiles from the cold trap were
transferred as a gas to Tedlar* bags containing
standard bioassay plates. All other aspects of the
test were conducted as described by Ames et al.
(1975).
Results and discussion
When one evaluates the hazard of a complex
organic emission such as ETS, there are multiple
and varied approaches that one may use. As stated
in the introduction, several researchers have as-
sessed ETS through epidemiological efforts
(Hirayama, 1981; Trichopoulos et al., 1981;
Garfinkel, 1981; National Research Council,
1986). Although such studies can demonstrate the
statistical likelihood of increased risk associated
with ETS, these studies were not designed to
answer more specific questions concerning ETS
and human health. For example, these studies do
not identify which constituents within ETS emis-
sions are carcinogens, mutagens, and other types
of toxicants. Epidemiology, therefore, is unlikely
to determine whether or not changes in tobacco
type used, cigarette design, tar values, room ven-
tilation rates, etc. will produce major changes in
the health impact of ETS. When human epidemio-
logical data are unavailable and unlikely to be
available to answer such relevant issues, one can
sometimes use whole animal studies (Stara and
Kello, 1979). However, since whole animal studies
for carcinogenesis and heritable mutations are ex-
pensive, lengthy processes, they too are unlikely to
be useful for answering many relevant questions.
Short-term test data, especially when associated
with quantitative data on analytical chemistry, can
be used to make comparative assessments, to iden-
tify genotoxicants, and to monitor human ex-
posure.
After identification and quantification of the
individual constituents of the mixture, the known
lexicological properties of each constituent can be
related to the source, for example ETS. The total
potential impact of a source can then be estimated
by summing the activity of the known constituents
or by using some form of lexicological modeling
or scaling.
Approximately 10% of the more than 3800
compounds found in MS have been identified in
ETS. Table 1 lists over 100 constituents found in
ETS and in MS and IA samples for which there is
quantitative analytical information. The relative
concentrations of the constituents vary according
to whether they are measured as MS, SS. or IA
components. The reasons (e.g., combustion tem-
perature, oxygen levels, etc.) for differences in
concentrations of organic components in MS and
SS have been discussed previously (Sakuma et al..
1983, 1984; Baker, 1981. 1982; KJus and Kutin.
1982; IARC, 1968; National Research Council.
1986).
Table 2 lists the genotoxicity information
(Graedel et al., 1986; Nesnow et al., 1987) for the
compounds in Table 1. Approximately 1/3 of the
compounds in Table 1 have associated genotoxic-
ity information. Eleven of these compounds are
animal carcinogens (Nesnow et al., 1987). Nine-
teen of the compounds are positive in at least 1
bioassay, and 10 are positive in the Salmonella
bioassay (Graedel et al., 1986). This type of infor-
mation provides a qualitative evaluation of poten-
tial human health impacts. For comparative risk
analysis methods (Lewtas, 1985). the amount of
each constituent can be associated with the bioas-
say activity of each constituent in order to provide
a crude means of comparison between sources or
components of interest. For example, using con-
centrations of constituents in SS and their known
-------
84
TABLE 1
CONCENTRATIONS OF COMPOUNDS ASSOCIATED WITH MAINSTREAM AND SIDESTREAM TOBACCO SMOKE
AND INDOOR AIR POLLUTED WITH TOBACCO SMOKE "
Compound
Acetamidc
Acetamide
Acetic acid
Acetic acid
Acetic acid
Acetic acid
Acrolein
Acrolein
Acrolein
Acrolein
Acrolein
Acrolein (gas only + people)
Acrolein (people absent)
Acrolein (people present)
Acrolein control air
Aldehydes (gas only + people)
Aldehydes (generic)
Aldehydes (people absent)
Aldehydes (people present)
Alkoxyl radicals
Alkoxyl radicals
Ammonia
Ammonia
Ammonia (cigars)
Ammonia (cigars)
Ammonia
Ammonia
Ammonia (cigars)
Ammonia (cigars)
Anatabine
Anatabine
Anthanthrene
Anthanthrene
Anthanthrene
Anthanthrene
Anthracene
Anthracene
Anthracene
Anthracene
Anthracene
Benz{a]anthracene
Ben2(fl]anthracene
Benz(a]anihracene
Benz|a]anthracene
Benz(o]anthracene
Sample
type
MS
SS
MS
MS
SS
SS
IA
IA
IA
IA
SS
IA
IA
IA
IA
IA
IA
IA
IA
MS
SS
MS
MS
MS
MS
SS
SS
SS
SS
MS
SS
IA
IA
MS
SS
IA
MS, P
MS, V
SS, P
SS. V
IA
MS, P
MS. V
SS, P
SS. V
Concentration
Low
70.00
86.00
333.00
272.00
1241.00
695.00
0.90
0.02
6.00
0.01
50.00
130.00
119.00
10.00
0.00
1290.00
0.39
1100.00
391.00
8.00 X1015
6.00x10"
79.40
95.30
30.50
148.00
5.14
6.11
6.98
9.34
2.40
0.00
3.00
22.00
39.00
23.60
0.10
670.00
40.00
13.30
0.09
201.00
2.50
range
High
111.00
156.00
809.00
475.00
2187.00
1 148.00
1.30
0.12
10.00
0.19
70.00
190.00
133.00
48.00
5.00
1 350.00
1.37
1 370.00
622.00
spins/c
spins/c
131.00
163.00
322.00
288.00
5.77
7.18
106.00
20.50
20.10
2.40
Units
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
ppm
mg/m3
ppm
mg/m3
ppm
Mg/m3
Mg/m3
Mg/m3
ppm
Mg/m3
mg/m3
Mg/m3
Mg/m3
Mg/cig
Mg/g smoked
Mg/g smoked
/ig/product
mg/cig.
mg/g smoked
mg/cig
mg/g smoked
Mg/cig
Mg/cig
Qua! (ng/m3)
ng/m3
ng/cig
ng/cig
Qual (ng/m3)
ng/cig
ng/cig
ng/cig
ng/cig
Qual (ng/m3)
ng/cig
ng/cig
ng/cig
ng/cig
Ref. h
25
25
23
25
23
25
2
3
12
19
2
19
19
19
12
19
19
19
19
22
22
9
9
9
9
9
9
9
9
24
24
1
15
14
14
1
16
16
16
16
1
16
16
16
16
-------
85
TABLE 1 (continued)
Compound
Benz( e Jacenaphthylene
Benzene
Benzene (breath, non-smokers)
Benzene (breath, smokers)
Benzene (homes, non-smokers)
Benzene (homes, smokers)
Benzo[ a jfluorene
Benzo[ a jfluorene
Benzo[ a Jfluorene
Benzol a ]fluorene
Benzo[ a jpyrene
Benzol a Ipyrene
Benzo[ a jpyrene
Benzo[a]pyrene
Benzol a Jpyrene
Benzo[ a Jpyrene
Benzo[a]pyrene
Benzo[a]pyrene
Benzo[ a Ipyrene
Benzol a Jpyrene
Benzo[a]pyrene control
Benzo[ b ]naphtho(2.1 -djthiophene
Benzo{ b Jnaph tho[2,l -d]thiophene
Benzo[ 6]naphtho(2,l-d]thiophene
Benzo{ b ]naphtho[2,l-d]thiophene
Benzo[ 2>/c]fluorene
Benzo[ 6/c]fluorene
Benzo( b/j/k ]fluoran thene
Benzo[ b/j/k jfluoranthene
Benza[ b/j/k ]fluoranthene
Benzo{ ejfluorene
Benzo(e Ipyrene
Benzo[ ejpyrene
Benzo[ « Ipyrene
Benzo[? Jpyrene
Benzo[ e ]pyrene
Benzo[e Ipyrene
Benzo[e Ipyrene
Benzo(e Ipyrene
Benzo[ e Jpyrene
Benzo( e Ipyrene control
Benzo( 8hi jfluoranthene
Benzo[ £Ai]perylene
Benzo[ g/i/Jperylene
Benzo[g/»']perylene
Benzo[g/i/]perylene
Benzo[ £*' Iperylene
Benzol ggi'lperylene
Benzo( £/»' Iperylene
Benzol g>i/]perylene
Sample
type
[A
IA
IA
IA
IA
IA
IA
IA
MS
SS
IA
IA
IA
IA
MS
MS. P
MS, V
SS
SS, P
SS, V
IA
MS, P
MS. V
SS, P
SS, V
MS
SS
IA
MS
SS
IA
IA
1A
IA
MS
MS, P
MS. V
SS
SS, P
SS.V
IA
IA
IA
IA
MS
MS. P
MS,V
SS
SS, P
SS, V
Concentration range
Low High
0.05 0.15
2.50
16.00
4.40 9.20
4.80 16.00
39.00
184.00
751.00
7.10 21.70
6.20 144.00
22.00
44.00
10.90
0.08
199.00
103.00
0.48
0.00 0.69
2.80
0.21
50.00
1.10
69.00
251.00
35.00
49.00
260.00
18.00
3.30 23.40
25.00
6.70
0.13
135.00
75.00
0.74
3.00 5.10
17.00
39.00
7.10
0.09
98.00
41.00
0.62
Units
Qual(ng/m3)
mg/m3
Hg/m3
Mg/m3
Mg/cm3
Mg/m3
Qual(ng/m3)
ng/m3
ng/cig
ng/cig
Qual(ng/m3)
ng/m3
ng/m3
ng/m3
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/m3
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/m3
ng/cig
ng/cig
Qual(ng/m3)
Qual(ng/m3)
ng/m3
ng/m3
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/m3
QuaUng/m;1)
Qual(ng/m3)
ng/m3
ng/cig
ng/cig
ng/ciiz
Of *O
ng/cig
ng/cig
Of O
ng/cig
Ref. h
1
3
30
30
30
30
1
15
14
14
1
11
13
15
14
16
16
14
16
16
11
16
16
16
16
14
14
15
14
14
1
1
15
20
14
16
16
14
16
16
20
I
1
15
14
16
16
14
16
16
-------
86
TABLE 1 (continued)
Compound
BenzoCluoramhenes (b + j + k)
Benzofluoranthenes (b + j + k)
Benzofluoranthenes (b+j •+• k)
Benzofluoranthenes (b+j + k)
Benzoic acid
Benzoic acid
Benzoic acid, m-hydroxy-
Benzoic acid, /n-hydroxy-
Benzonitrile
Benzonitrile
Bipyridyl. 2. 3'-
Bipyridyl. 2, 3'-
Bipyridyl. 5-methyl-2. 3'-
Bipyridyl. 5-methyl-2. 3'-
Butyrolactone, gamma-
Butyrolactone. gamma-
Carbon monoxide
Carbon monoxide
Carbon monoxide (gas only •+• people)
Carbon monoxide (people absent)
Carbon monoxide (people present)
Carbon monoxide (people present)
Carbon monoxide control
Carbon monoxide control
Carbon monoxide (artificial cond.)
Carbon monoxide (natural conditions)
Carbon, total
Carbon, elemental
Carbon, organic
Carboxyhemoglobin (blood, passive)
Carhoxyhemoglobin (blood, smoker)
Carboxyhemoglobin
(blood, no smoking)
Catechol
Catechol
Catechol. 2-methyl-
Catechol, 2-methyl-
Catechol, 3-methyl-
Catechol. 3-methyl-
Catechol. 4-ethyl-
Catechol. 4-ethyl-
Catechol, 4-methyl-
Catechol. 4-methyl-
Catechol. 4-vinyl-
Catechol, 4-vinyl-
Catechols (all catechols)
Catechols (all catechols)
Coronene
Coronene control
Sample
type
MS, P
MS. V
SS. P
SS. V
MS
SS
MS
SS
MS
SS
MS
SS
MS
SS
SS
MS
1A
IA
1A
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
MS
SS
MS
SS
MS
SS
MS
SS
SS
MS
MS
SS
MS
SS
IA
IA
Concentration range
Low
20.50
0.22
196.00
1.38
14.00
12.00
8.00
3.00
5.00
33.00
9.90
20.00
6.60
6.00
40.00
11.00
2.00
0.00
23.00
21.00
18.00
3.70
0.00
0.00
8.00
9.00
207.00
11.90
195.00
0.55
3.38
0.57
148.00
138.00
6.00
8.00
31.00
24.00
27.00
19.00
25.00
29.00
23.00
7.00
25.00
88.00
0.50
1.00
High
28.00
23.00
64.00
15.00
6.00
57.00
27.40
73.00
14.70
14.00
103.00
22.00
23.00
1.20
26.00
25.00
22.00
4.20
15.00
0.50
16.00
362.00
292.00
13.00
21.00
62.00
47.00
102.00
68.00
55.00
80.00
113.00
40.00
328.00
212.00
1.20
2.80
Units
ng/cig
ng/cig
ng/cig
ng/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
ppm
ppm
ppm
ppm
ppm
ppm
ppm
ppm
ppm
ppm
Mg/m3
Mg/rn3
Mg/m3
%
%
%
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/m3
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
ng/m3
ng/m3
Ref. "
16
16
16
16
23
23
23
23
25
25
24
24
24
24
25
25
3
31
19
19
19
28
3
31
29
29
27
27
27
28
28
28
23
23
23
23
23
23
23
23
23
23
23
23
4
4
20
20
-------
TABLE 1 (continued)
87
Compound
Cotinine (plasma, non-smoker)
Cotinine (plasma, passive smoker)
Cotinine (plasma, smoker)
Cresoi, m-
Cresol, m-
Cresol, m-
Cresol, m-
Cresol, o-
Cresol, o-
Cresol. p-
Cresol, p-
Cresol, p-
Cresol, p-
Cyclopentenone, 2, 3-dimethyl-2-
Cyclopentenone, 2, 3-dimethyl-2-
Cyclopentenone, 2-
Cyclopentenone, 2-
Cyclopentenone, 2-O//-3-methyl-2-
Cyclopentenone, 2-0/f-3-methyl-2-
Cyclopentenone, 2-methyl-2-
Cyclopentenone, 2-methyl-2-
Dibenzfa, y]anthracene
Dibenz[a, y'Janthracene
Dibenz(
-------
88
TABLE 1 (continued)
Compound
Furoic acid. 2-
Furoic acid. 2-
Glutaric acid
Glutaric acid
Glycolic acid
Glycolic acid
Guaiacol. 4-vinyl-
Guaiacol, 4-vinyl-
Guaiacol, 4-vinyl-
Guaiacol. 4-vinyl-
HCN
HCN (gas only + people)
HCN (people absent)
HCN (people present)
Hydrazine
Hydrazine
Hydroquinone
Hydroquinone
Hydroquinone, methyl-
Hydroqumonc. methyl-
Hydroxypropionic acid, 3-
Hydroxypropionic acid, 3-
Indenofl, 2, 3-«/]pyrene
Indenofl. 2, 3-o/]pyrene
Indeno(l, 2. 3-o/]pyrene
Indenofl, 2. 3-«/]pyrene
Indeno[l, 2. 3-o/]pyrene
Isoquinoline
Isoquinoline
Lactic acid
Lactic acid
Levulinic acid
Levulinic acid
Limonene
Limonene
Lutidine, 2, 4-
Lutidine, 2, 6-
Lutidine, 2, 6-
Lutidine, 3, 5-
Lutidine, 3, 5-
Sample
type
MS
SS
MS
SS
MS
SS
SS
MS
MS
SS
IA
IA
IA
IA
MS
SS
MS
SS
MS
SS
MS
SS
IA
MS, V
SS. P
SS.V
MS, P
MS
SS
MS
SS
MS
SS
MS
SS
SS
MS
SS
MS
SS
Concentration
Low
44.00
25.00
10.00
6.00
37.00
35.00
24.00
23.00
16.00
15.00
0.01
82.00
50.00
10.00
31.50
94.20
114.00
91.00
23.00
21.00
2.00
1.00
0.17
51.00
0.36
8.10
1.60
5.00
63.00
45.00
29.00
25.00
15.00
63.00
35.00
1.40
1.40
0.00
22.00
range
High
107.00
60.00
58.00
18.00
126.00
77.00
32.00
36.00
30.00
37.00
0.08
86.00
14.00
300.00
285.00
39.00
41.00
31.00
29.00
2.00
8.00
174.00
123.00
56.00
49.00
49.00
397.00
315.00
33.00
33.00
17.00
251.00
Units
Mg/cig
MgAig
Mg/cig
Mg/cig
Mg/cig
Mg/c'g
MgA'g
Mg/cig
Mg/cig
Mg/cig
mg/m3
Mg/m3
jig/m3
Mg/m3
ng/cig
ng/cig
Mg/cig
Mg/cig
Mg/cig
MgAig
Mg/cig
Mg/cig
Qual (ng/m3)
ng/cig
ng/cig
ng/cig
ng/cig
Mg/cig
MgAig
Mg/cig
Mg/cig
Mg/cig
Mg/c'g
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Ref."
23
23
23
23
23
23
23
23
25
25
19
19
19
19
21
21
23
23
23
23
23
23
1
16
16
16
16
24
24
23
23
23
23
25
25
6
6
6
6
6
Methylenephenanthrene, 4, 5-
IA
Qual (ng/m3)
Methylnaphthalene, 1-
Methylnaphthalene, 1-
Methylnaphthalene, 2-
Methylnaphthalene, 2-
SS
MS
MS
SS
30.00
1.02
1.21
31.60
Mg/cig
Mg/cig
MgAig
Mg/cig
26
26
26
26
-------
89
TABLE 1 (continued)
Compound
Methylnitrosoamino-pyridyl-butanone
Methylnitrosoamino-pyridyl-butanone
Methylphenanthrene. 1-
Methylphenanthrene, 2-
Methylphenanthrene, 3-
Methylphenanthrene. 4/9-
Myosmine
Myosmine
Naphthalene
Naphthalene
Neophytadiene
Neophytadiene
Nicotine
Nicotine
Nicotine
Nicotine
Nicotine
Nicotine
Nicotine
Nicotine
Nicotine (gas only + people)
Nicotine (people absent)
Nicotine (people present)
Nicotine, office buildings
Nicotyrine
Nicotyrine
Nicotyrine
Nicotyrine
Nitrogen dioxide
Nitrogen dioxide
Nitrogen dioxide (gas only + people)
Nitrogen dioxide (people absent)
Nitrogen dioxide (people present)
Nitrogen dioxide control
Nitrogen oxide
Nitrogen oxide
Nitrogen oxide (gas only + people)
Nitrogen oxide (people absent)
Nitrogen oxide (people present)
Nitrogen oxide control
Nitrogen oxides (combined)
Nitrosoamine, methylethyl-
Nitrosoamine, methylethyl-
Nitrosoamine, methylethyl-
Nitrosoamine, methylethyl-
Nitrosoanabasine, N'-
Nitrosoanatidine, N'-
Nitrosoanatidine, N'-
Sample
type
MS
SS
IA
1A
IA
IA
MS
SS
MS
SS
MS
SS
IA
IA
IA
IA
MS
MS
SS
SS
IA
IA
IA
IA
MS
MS
SS
SS
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
IA
MS
MS
SS
SS
SS
MS
SS
Concentration
Low
46.00
201.00
13.10
73.00
2.76
45.50
66.00
70.00
25.00
0.70
1.00
1.70
1720.00
1483.00
3210.00
2987.00
130.00
102.00
1.70
4.20
17.00
49.00
93.00
0.00
58.00
0.01
0.00
0.00
27.00
0.30
0.00
0.31
0.48
0.30
5.00
59.00
0.10
0.00
9.00
0.00
15.00
82.00
61.00
range
High
240.00
540.00
33.00
224.00
232.00
421.00
1010.00
3.10
10.30
180.00
3330.00
3149.00
5830.00
6588.00
180.00
20.20
41.00
211.00
263.00
0.03
0.03
0.60
9.00
0.40
0.59
0.60
218.00
9.10
1.80
75.00
27.00
40.00
167.00
220.00
Units
ng/cig
ng/cig
Qual(ng/mJ)
Qual(ng/rrr')
Qual(ng/nr )
Qual(ng/mJ)
Mg/cig
Mg/cig
Mg/eig
Mg/cig
Mg/cig
Mg/cig
Mg/mJ
Mg/m3
Mg/nv
pg/m2 min
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Traces only
Mg/m3
Mg/m3
pg/m2 min
Mg/cig
Mg/cig
Mg/cig
Mg/cig
ppm
ppb
Ppm
ppm
ppm
ppb
ppm
ppb
ppm
ppm
ppm
ppb
ppb
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
Ref. h
8
8
1
1
1
1
24
24
26
26
25
25
3
17
18
32
24
25
24
25
19
19
19
32
24
25
24
25
19
31
19
19
19
31
19
31
19
19
19
31
12
5
7
5
7
8
8
8
-------
90
TABLE 1 (continued)
Compound
Nitrosodiethylamine
Nilrosodiethylamine
Nitrosodiethylamine, N-
Nitrosodiethylamine, N-
Nitrosodiethylamine (artificial)
Nitrosodiethylamine (natural cond.)
Nitrosodimethylamine
N itrosodimethylamine
Nitrosodimethylamine
Nitrosodimethylamine
N i trosodi meihylamine
Nitrosodimethylamine. N-
Nitrosodimethylamine. N-
Nitrosodimethylamine (artificial)
Nitrosodimethylamine (natural cond.)
NUrosoethylmethylamine, A/-
Nitrosoethylmethylamine, N-
Nitrosonomicotine
Nitrosonomicotine
Nitrosopyrrolidine
N i irosopy rrolidi ne
N i t rosopy rrolidine
Nitrosopyrrolidine
Nitrosopyrrolidine
Nitrosopyrrolidine, N-
Nitrosopyrrolidine, N-
Octane (breath smokers)
Octane (breath, non-smokers)
Octane (homes, non-smokers)
Octane (homes, smokers)
Parvoline
Parvoline
Pentadien-4-olide, 2, 4-
Pentadien-4-olide, 2, 4-
Perylene
Perylene
Perylene
Perylene
Perylene
Perylene control
Phenanthrene
Phenamhrene
Phenanthrene
Phenanthrene
Phenanthrene
Phenol
Phenol
Phenol
Phenol
Sample
type
MS
SS
MS
SS
IA
IA
MS
MS
SS
SS
SS
MS
SS
IA
IA
MS
SS
MS
SS
MS
MS
SS
SS
SS
MS
SS
IA
IA
IA
IA
MS
SS
MS
SS
IA
IA
IA
MS
SS
IA
IA
MS, P
MS,V
SS, P
SS, V
MS
MS
SS
SS
Concentration range
Low
0.00
8.00
1.80
8.20
0.00
0.00
1.70
0.00
680.00
143.00
460.00
1.70
680.00
0.02
0.00
0.00
9.40
81.00
110.00
2.60
1.50
204.00
28.00
80.00
2.60
204.00
1.10
0.10
1.70
1.50
0.00
10.00
8.00
71.00
11.00
0.70
9.00
39.00
2.80
74.80
2.10
2149.00
248.00
79.00
77.00
69.00
157.00
High
4.80
73.00
4.80
73.00
0.01
0.20
97.00
27.00
1 770.00
415.00
1 880.00
97.00
1040.00
0.15
0.70
9.10
30.00
390.00
390.00
52.00
29.00
612.00
143.00
500.00
51.70
387.00
3.10
4.70
4.30
145.00
41.00
256.00
1.30
1.70
136.00
139.00
241.00
289.00
Units
ng/cig
ng/cig
ng/cig
ng/cig
ng/1
ng/l
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
"8/1
ng/I
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/ciq
/ig/m3
Mg/m3
Mg/m3
/ig/m3
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Qual (ng/m3)
ng/m3
ng/m3
ng/cig
ng/cig
ng/m3
Qual (ng/m3)
ng/cig
ng/cig
ng/cig
ng/cig
MgA'g
Mg/cig
Mg/cig
Mg/cig
Ref. h
5
5
10
10
29
29
5
7
5
7
8
10
10
29
29
10
10
8
8
5
7
5
7
8
10
10
30
30
30
30
6
6
25
25
1
15
20
14
14
20
1
16
16
16
16
23
25
23
25
-------
91
TABLE 1 (continued)
Compound
Phenol. 4-vinyI-
Phenol, 4-vinyl-
Phenols
Phenylacetic acid
Phenylacetic acid
Picoline, 3-
Picoline. 3-
Picoline, alpha-
Picoline, alpha-
Pyran-4-one, 5, 6-diOH-3, 5-diOH-2-ME
Pyran-4-one, 5. 6-diOH-3, 5-diOH-2-ME
Pyrazine, 2, 3-dimethyl-
Pyrazine, 2-methyl
Pyrazine, 2-methyl-
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene
Pyrene control
Pyrene, 1-methyl-
Pyrene, 2-raethyl-
Pyrene, 4-methyI-
Pyridine
Pyridine
Pyridine
Pyridine
Pyridine, 2-(3-pentyl)-
Pyridine, 2-(3-pentyl)-
Pyridine, 2-ethyl-
Pyridine, 2-ethyl-
Pyridine, 3-acetyl-
Pyridine, 3-acetyl-
Pyridine, 3-cyano-
Pyridine, 3-cyanc-
Pyridine, 3-ethyl-
Pyridine, 3-ethyl-
Pyridine, 3-ethyl-
Pyridine, 3-etbyl-4-methyl-
Pyridine, 3-ethyl-4-methyl-
Pyridine, 3-hydroxy-
Pyridine, 3-hydroxy-
Pyridine, 3-hydroxy-
Pyridine, 3-hydroxy-
Pyridine, 4-ethyl-
Pyridine, 4-r-butyl-
Pyridine, 4-r-butyl-
Pyridine, methylvinyl-
Pyridine, methylvinyl-
Sample
type
MS
SS
IA
MS
SS
MS
SS
MS
SS
MS
SS
SS
MS
SS
IA
IA
IA
MS
MS, P
MS,V
SS
SS. P
SS, V
IA
IA
IA
IA
MS
MS
SS
SS
MS
SS
MS
SS
MS
SS
SS
MS
MS
SS
SS
SS
MS
MS
MS
SS
SS
SS
MS
SS
MS
SS
Concentration
Low
18.00
25.00
7.40
18.00
11.00
12.00
90.00
12.30
128.00
13.00
0.00
0.00
0.00
0.00
66.00
4.10
270.00
43.00
1.90
1011.00
466.00
10.30
2.80
32.40
16.00
336.00
187.00
0.00
0.00
2.60
2.60
3.80
9.00
24.00
2.40
4.00
71.00
21.00
6.40
0.00
125.10
90.00
152.00
157.00
27.00
0.00
17.00
2.20
12.00
range
High
45.00
57.00
11.50
38.00
30.00
22.00
166.00
189.00
1 090.00
153.00
143.00
50.00
8.60
8.60
9.40
7.00
648.00
20.00
3420.00
262.00
1.50
143.00
35.00
35.00
6.40
11.00
64.00
4.20
6.00
960.00
36.00
34.00
1.50
211.40
119.00
167.00
191.00
379.00
4.50
287.00
4.10
19.00
Units
Mg/cig
Mg/cig
Mg/m2
Mg/cig
Mg/cig
MS/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Qual(ng/mJ)
ng/mj
ng/m3
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/cig
ng/nr1
Qual (ng/m3)
Qual (ng/m3)
Qua! (ng/m3)
Mg/cig
1" Of ™*O
Mg/cig
Mg/cig
• Of O
Mg/cig
Mg/cig
ue/cie
f~o/ *e
US./ CIS.
~o/ O
ue/cie
f^Of "6
ue/ciz
I^Of "O
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Ref. h
25
25
20
23
23
25
25
6
6
25
25
6
6
6
1
15
20
14
16
16
14
16
16
20
1
1
1
6
25
6
25
6
6
5
6
24
24
24
24
25
6
25
6
6
24
25
24
25
6
6
26
24
23
-------
92
TABLE 1 (continued)
Compound
Pyrrole
Pyrrole
Styrene (breath, non-smokers)
Styrene (breath, smokers)
Styrene (homes, non-smokers)
Styrene (homes, smokers)
Succmic acid
Succinic acid
Succmic acid, metnyl-
Succimc acid, methyl-
Tar radical sol in r-butylbenzene
Tar radical sol in f-butylbenzene
Thiocyanate (plasma, non-smoker)
Thiocyanate (plasma, passive smoker)
Thiocyanate (plasma, smoker)
Thioethers (urine, non-smoker)
Thioethers (urine, passive)
Thioethers (urine, smoker)
Sample
type
MS
SS
IA
IA
IA
IA
MS
SS
MS
SS
MS
SS
IA
IA
IA
IA
IA
IA
Concentration
Low
16.00
140.00
0.30
1.10
0.80
1.10
112.00
65.00
4.00
1.00
70.80
71.80
70.70
6.00
6.40
6.30
range
High
23.00
272.00
1.10
2.20
163.00
70.00
31.00
13.00
Unit*
Mg/cig
Mg/cig
Mg/m3
Mg/m3
fig/Hi3
Mg/m3
Mg/c»g
Mg/cig
Mg/cig
Mg/cig
Qualitative
Qualitative
Mmol/l
Mmol/l
Mmol/l
mmol/ml
mmol/ml
mmol/ml
Ref. "
25
25
30
30
30
30
23
23
23
23
22
22
28
28
28
28
28
28
Toluene
IA
0.04
1.04
mg/m3
Valeric acid, 3-methyl-
Valenc acid, 3-meihyl
Vinylphenol, p-
Vinylphenol. p-
Xylene, m + p- (breath, non-smokers)
Xylene. m + p- (breath, smokers)
Xylene, o- (breath, non-smokers)
Xylene. m + p- (homes, non-smokers)
Xylene. m -*• p- (homes, smokers)
Xylene, o- (breath, smokers)
Xylene. o- (homes, non-smokers)
Xylene. o- (homes, smokers)
Xylenol, 2, 6-
Xylenol. 2. 6-
MS
SS
MS
SS
IA
IA
IA
IA
IA
IA
IA
IA
MS
SS
20.00
20.00
21.00
21.00
2.10
5.50
0.80
10.00
10.00
1.60
4.00
3.20
8.00
8.00
261.00
384.00
51.00
45.00
13.00
20.00
5.20
7.10
16.00
20.00
Mg/cig
Mg/cig
Mg/cig
Mg/cig
Mg/mJ
Mg/m3
Mg/m3
Mg/m3
Mg/m3
Mg/m3
Mg/m3
Mg/m3
Mg/cig
Mg/cig
23
23
23
23
30
30
30
30
30
30
30
30
23
23
' Listings are given in alphabetical order by compound and special conditions are noted in parentheses within the column labeled
'Compound'. The sample type is categorized in the second column as I A, indoor air; SS, sidettream smoke: MS, mainstream
smoke; P. associated with paniculate matter; and V, associated with volatile compounds.
b Reference numbers are as follows: 1, Alfheim and Ramdahl, 1984; 2. Ayer and Yeager, 1982; 3, Badre et al.. 1978; 4, Brunnemann
et al., 1976; 5, Brunnemann et al., 1977; 6, Brunnemann et al., 1978; 7, Brunnemann et al., 1980; 8, Brunnemann et al.. 1983, 9,
Brunnemann and Hoffmann, 1975; 10, Brunnemann and Hoffmann, 1978: 11, Elliott and Rowe, 1975; 12, Fischer et al., 1978; 13,
Galuskinova, 1964; 14, Grimmer et al.. 1977a; 15, Grimmer et al., 1977b; 16, Grimmer et al., 1987; 17, Harmsen and Effenberger,
1957; 18, Hinds and First, 1975; 19, Hugod et al. 1978; 20, Just et al., 1972; 21, Liu et al., 1974; 22. Pryor el al., 1983; 23, Sakuma
et al.. 1983; 24. Sakuma et al.. 1984a; 25. Sakuma et al.. 1984b; 26. Schmettz et al.. 1976; 27. Sexton et al., 1984; 28, Sorsa et al..
1985. 29. Stehlik et al.. 1982; 30. Wallace and Pellizzari. 1986; and 31. Williams el al.. 1985.
-------
93
genotoxicity, one may calculate, in an additive
fashion, the 'total genotoxic potential* of a
cigarette. Comparative human exposure could then
be calculated by knowing the total genotoxic
potential of a cigarette, room air volume, and the
air exchange rate. If one were to make a crude
estimate of the bacterial mutagenicity of 'a
cigarette' by using the upper concentration ranges
from Table 1 and Salmonella potency range for
the 10 bacterial mutagens (1 rev/100 mg to 1
rev/0.10 /ig), the calculated mutagenic potency
[/tg of compound/cigarette X revertants//tg com-
pound] of a cigarette would be approximately
0.5-45 revertants per cigarette. If all of the com-
pounds in Table 2 except the known negatives
were to be shown to be Salmonella mutagens, one
could estimate the mutagenic activity level to be
as high as 1500 revertants/cigarette. Obviously,
this method of summing the total genotoxic activ-
ity for a ' typical cigarette' is most useful when the
chemical and bioassay values are quantitatively
accurate, when all genotoxicants have been identi-
fied and bioassayed properly, and when synergis-
tic and/or antagonistic interactions do not inter-
fere. Because one would not expect this list of
bacterial mutagens to be inclusive of all the
mutagens and because one cannot presently rule
out all possible types of interactions, this type of
calculation would estimate only the lower limits of
mutagenicity as will be clearly demonstrated be-
low.
Upon testing an acetone extract of SS particles,
Lofroth et al. (1983) found the extract prefer-
entially mutagenic in S, typhimurium TA98 in the
presence of S9. The observed response using TA98
corresponded to 15000 revertants per cigarette.
Upon using a cigarette-smoking machine in a 15-
m3 room, Ldfroth and Lazaridis (1986) calculated
that their results represent the equivalent of 30000
revertants per cigarette for MS and 10000-20000
revertants per cigarette for SS when using S.
typhimurium TA98 with S9 activation. In 1987,
Ling et al., using the same methods, observed for
strain TA98 a range of 17200-31300 revertants
per cigarette for SS organics tested using the plate
incorporation protocol and a range of 36000-
118300 revertants per cigarette using a micro-as-
say preincubation protocol. In addition, they ob-
served significant activity using TA100 without
TABLE 2
THE GENOTOXICITY OF COMPOUNDS ASSOCIATED
WITH ENVIRONMENTAL TOBACCO SMOKE
Compound
Acetamide
Acetic acid
Acrolein
Anthracene
Benz{ a (anthracene
Benzene
Benzo(
-------
94
TABLE 2 (continued)
TABLE 2 (continued)
Compound
Hydroquinone
Indenofl, 2. 3-o/]pyrene
Isoquinoline
Limonene
Naphthalene
Nicotine
Nitrosodiethylamine. N'-
Nitrosodimethylamine, N'-
Nitrosonornicotine
Nilrosopyrrolidine
Perylene
Phenanthrene
Phenol
Pyrene
Pyridine
CAS Number
123-31-9
193-39-5
119-65-3
5989-27-5
91-20-3
54-11-5
55-18-5
62-75-9
16543-55-8
930-55-2
198-55-0
85-01-8
108-95-2
129-00-0
110-86-1
Bioassay results '
ALC
ST
CCC
ST
ST
ST
NEU
ST
CCC
CT
CYC
L5
MDR
MST
REC
SCE
ST
V79
ARA
CCC
CT
CYG
CYG
L5
MNT
MST
NEU
SCE
SRL
SRL
ST
V79
YEA
CCC
CCC
SCE
ALC
CCC
CT
CYC
ST
NEU
ST
CCC
CT
CYC
ST
V79
SCE
ST
+
NEG
+
NEG
NEG
NEG
NEG
NEG
+
+
+
+
+
NEG
+
+
+
+
+
+
+
+
NEG
+
+ /-
NEG
+
+
+
+
+
+
+
+
+
NEG
+
I
NEG
NEG
NEG
NEG
NEG
I
NEG
NEG
NEG
NEG
+
NEG
Compound
CAS Number
Bioassay results'
Styrene
Toluene
100-42-5
108-88-3
CCC
SCE
ST
+
NEG
NEG
* Bioassay information is extracted from Graedel et al., 1986.
Abbreviations used for bioassay results are as follows: ALC.
Allium cytogenetics assays; ARA, Arabidopsis mutagen as-
say; ASPH, Aspergillus mutagen assay; CCC, whole animal
carcinogen assays; CT, cell transformation bioassays; CY,
mammalian cytogenetic bioassays; L5, L5178Y mouse
lymphoma assay; MDR, mammalian cell DNA repair as-
says; MNT, micronucleus assays; MST, mouse spot test;
NEU. Neurospora assays; REC, DNA repair-deficient
bacterial assays; SCE. sistcr-chromatid exchange assays;
SRL, sex-linked recessive lethal assays in Drosophila; ST,
Salmonella assays; TRM, Tradescantia mutagen assays; V79,
V79 Chinese hamster mutation assays; and YEA, Yeast
mutation tests. Results are recorded as +, positive; NEG.
negative; and I, Indefinite.
activation. Ong et al. (1984) examined the muta-
genicity of SS using 5. typhimurium TA98W (an
antibiotic-resistant strain of TA98) in an in situ
impinger system. They stated that '... the con-
centration of cigarette smoke that could be de-
tected for mutagenic activity was 0.0065 cigarette
per ml.' Because the concentration of cigarette
smoke in these experiments was equivalent to
approximately 0.036 cigarettes/ml and the ob-
served response was approximately 40 induced
mutants/ml, they (Ong et al., 1984) observed 1111
[(40 rev/ml) • (1 ml/0.036 cig)] revertants per
cigarette. However, if one assumes from their
calculations that 0.0065 cigarettes would give a
doubling (- 20 induced revertants) of the sponta-
neous revertant number, the activity of SS is ap-
proximately 3080 [(20 rev/doubling) • (1 dou-
bling/0.0065 cigarettes)] revertants per cigarette.
The 3- to 10-fold difference between the results of
the L8froth et al. (1983) and Ong et al. (1984) is
most likely due to differences in the collection and
exposure systems used, although, once again, one
cannot rule out the role of differences in chemical
interactions. LSfroth et al. (1983) collected only
paniculate matter and tested the extracted organic
material, whereas Ong et al. (1984), while attempt-
ing to test total SS, probably assayed primarily the
-------
95
nonparticulate, semi-volatile organic material that
would dissolve in the fluid medium. Although the
methods of Lofroth et al. (1983) and Ong et al.
(1984) did not identify the specific genotoxicants
responsible for the biological activity, the direct
bioassay of the emissions did provide a more
accurate determination of the total mutagenic
potential of a cigarette.
Knowing the amounts and bioassay activity of
SS constituents or the total mutagenic activity of a
'typical cigarette' also does not necessarily indi-
cate that one can accurately estimate human ex-
posure to genotoxic (ETS) emissions. For both MS
and SS, the amount of a specific material that is
emitted from a cigarette ranges from sub-nano-
gram per cigarette levels to milligram per cigarette
levels. The concentrations of constituents in
smoke-impacted spaces range from ng/m3 to
mg/m3; however, the calculation of ambient con-
centrations from cigarette emission rates is not
always simple because ETS levels are functions of
smoking rate, ventilation, sink rate, mixing, and
volume of the space (National Research Council,
1986). Also, each individual compound may be
removed from the ambient air at different rates
due to these functions. For example, genotoxicant
exposure levels can be approximated by measur-
ing respirable paniculate (RSP) levels and measur-
ing the level of specific genotoxicants associated
with the particles. RSP levels in a one-smoker
residence can vary by 3 orders of magnitude from
approximately 17 to 5000 Mg/m3 (National Re-
search Council, 1986). Similarly, most IA con-
centrations (Table 1) for specific compounds span
at least 2 orders of magnitude; however, the rela-
tive amounts of components with SS are often
different from the relative amounts found in IA
samples. It is also unlikely that all SS carcinogens
have been identified. By collecting indoor air par-
ticulate matter and bioassaying the extracted
organic matter using S. typhimurium TM677 in
modified bioassay, Lewtas et al. (1987) dem-
onstrated that particle-associated mutagenicity per
cubic volume of air was greatest in the homes with
cigarette smoking and correlated with the number
of cigarettes smoked. The mutagenic activity per
cigarette, however, could not be calculated in this
study due to the presence of other potential sources
of mutagens such as woodstoves, gas appliances,
cooking, etc. Recently. Husgafvel-Pursiainen et al.
(1986) demonstrated that indoor airborne par-
ticulate matter collected in 3 restaurants where
smoking occurred gave up to 2370 revertants/m3
air. On the basis of optical particle counting, they
attributed the majority of the airborne paniculate
matter to cigarette smoking; however, they also
did not estimate mutagenicity on a per cigarette
basis. Husgafvel-Pursiainen et al. (1986), however.
did show that the levels of polynuclear aromatic
compounds roughly correlated with the mutagenic
activity. Human exposure to tobacco-smoke
genotoxicants, therefore, is highly variable and
difficult to assess by evaluating individual compo-
nents and/or equivalent bioassay activity levels
emitted into ambient air.
In order to explore the use of bacterial bioas-
says in evaluating ETS, our laboratory has ex-
amined some of the alternative methods that can
be used for evaluating ETS. The approaches used
are briefly described in the Materials and methods
section. As part of one study (Morin et al., 1987),
filters used to collect SS particles were extracted
with various solvents, solvent-exchanged to
DMSO, and bioassayed. Depending upon the ex-
traction conditions, tester strain, and activation
conditions the revertants per cigarette ranged from
400 to 19000 (Fig. 1). Results using TA98 with S9
of ~ 19000 revertants/cigarette resemble data
obtained by Lofroth et al. (1983).
Because the work of Ong et al. (1984) dem-
onstrated that the semi-volatile and volatile com-
ponents of SS may be mutagenic. we decided to
assess the total mutagenic potential of both MS
and SS. Using the sequential trapping train method
described, we collected volatile, semi-volatile, and
paniculate-bound organics simultaneously from
both MS and SS using separate trapping trains.
Each of the fractions from both trains was bioas-
sayed separately in the Salmonella mutagenicity
assay. No mutagenicity was associated with the
cold trap (volatile fraction) of either the MS or SS
when evaluated as gases in the Tedlar dessicator
system (Hughes et al., 1987). Table 3 summarizes
the results for the other components of the trap-
ping trains. Both the bubbler and sand trap sam-
ples were mutagenic for MS and SS; therefore,
both the particulate and the semi-volatile com-
ponents were mutagenic. The revertants/cigarette
-------
96
TEST CONDITION (Solvent/Strain/Actlvauonl
OC*/T» IOO/-S9
4C£/UIOO/«S9
ICE/UW/-S9
KOH/TMOO/-S9
7777/\^
3°-s
P"
0 3 10 15 20
SEVERTANTS UN THOUSANDS) PER CIGARETTE
Fig. 1. Salmonella typhimurium mutagenicity of sidestream
cigarette smoke when tested using various solvent extraction
systems, strains of 5. typhimurium, and activation conditions.
Abbreviations are as follows: ACE, Acetone; DCM, Dichloro
methane; MEOH, Methanol; TA98 and TA100 for respective
strains of 5. typhimurium; + S9, with exogenous activation;
and — S9 without exogenous activation.
value of 35200 is approximately twice the 19000
and 15 000 values previously seen with particle-as-
sociated fractions; however, when the cigarette
machine was not used to generate a puff mode
sample and the cigarettes were allowed to smolder,
this value was reduced to 23 250 revertants per
cigarette. It has been demonstrated previously that
the products produced during puffing and natural
smoldering are somewhat different due to gen-
erated temperatures, the rate of mass transfer of
oxygen to the tobacco source, properties of the
cigarette paper, etc. (Baker, 1981, 1982). There-
fore, it is reasonable that the mutagenicity of both
MS and SS varies with the degree of puffing. The
effect of burn conditions can also be noted by
examining the revertants/^ g of organic material
values in Table 3. Compared on a per mass of
organic material basis, the MS organics are 2-3
times as mutagenic as the SS organics. This dem-
onstrates that the mainstream and sidestream
combustion processes produce different relative
amounts of bacterial mutagens. One also could
speculate - from knowledge of the 2 types of
combustion processes and differences in MS and
SS chemistry - that the quantitative distribution
of mutagens in MS and SS is different. Due to the
mass of organic material produced, however, SS
organics make a larger contribution to the total
mutagenic activity of a cigarette. It is interesting
to note that within our laboratory the response of
benzo[a]pyrene (B[a]P) in Salmonella typhimurium
TA98 with exogenous activation is approximately
20 revertants//ig; therefore, one can calculate that
total cigarette bacterial mutagenicity is equivalent
to ~ 2.6 mg of B[a]P. Approximately 70%
(61600/86 300) of the total mutagenicity is associ-
ated with the sand trap (paniculate) samples. This
proportion also is approximately the same for the
MS and SS samples. Overall, the SS sample
accounted for approximately 60% of the total
mutagenicity on a per cigarette basis.
The best way presently available to assess hu-
man exposure to genotoxic ETS compounds is by
the analysis of the fluids and tissues of exposed
individuals. Exposure of target tissues and sites to
genotoxic constituents of ETS depends on several
factors, including the number of cigarettes smoked
TABLE 3
THE SALMONELLA TYPHIMURIUM TA98 MUTAGENICITY OF MAINSTREAM AND SIDESTREAM TOBACCO SMOKE
COLLECTED IN A PUFF MODE
Fraction
Main
Side
Total
Ethanol bubbler
Rev/Mg •
8.1
2.7
Rev/cig "
7557
17099
24656
Sand trap
Rev/Mg
4.0
1.6
Rev/cig
26400
35 200 c
61600
Bubbler + sand
Rev/cig
33957
52299
86256
a S. typhimurium TA98 mean revertants per plate per /ig of organic material collected using a preincubation bioassay. Determined
using a linear regression model (Myers et al., 1981)
h S. typhimurium TA98 revertants per cigarette calculated from (Rev//jg)x(;jg of fraction/cigarette).
•• When puff mode was not used and the cigarettes were allowed to smolder, this value was 23 253 rev/cig.
-------
97
in an enclosed area, the size and nature of the
area, the degree of room ventilation, breathing
rates and volume, absorption of the genotoxicants,
body distribution and excretion, and metabolism.
The uptake of individual agents from ETS can be
determined using chemical methods. For example,
one can measure products such as thiocyanate,
carboxyhemoglobin, nicotine and cotinine in phys-
iological fluids. However, some of these (e.g.,
thiocyanate due to HCN exposure and carboxy-
hemoglobin from CO exposure) do not originate
exclusively from SS (National Research Council,
1986). The quantitative aspects of other chemical
and biochemical markers such as cotinine,
hydroxyproline, /V-nitrosoproline, and aromatic
amines are still somewhat questionable (National
Research Council, 1986). In addition, when one
bioassays an agent individually, it may not be
known whether or not different genotoxic com-
pounds are absorbed, distributed, metabolized, and
excreted at the same rate as each other. The use of
a noncompound-specific measurement of exposure
to genotoxicants, therefore, would be very
advantageous. The monitoring of nonsmokers'
urine for mutagenic potential using bacterial as-
says provides a possible means of evaluating in
vivo exposure to ETS genotoxicants. In 1977.
Yamasaki and Ames reported the presence of
bacterial mutagens in the urine of smokers. As
noted by IARC (1986), a number of studies have
confirmed the finding of Yamasaki and Ames;
however, there is a wide variation in the results of
these studies. Much of this variation may be due
to dietary factors (Sasson et al., 1985). The studies
that have examined the urine mutagenicity of pas-
sive smokers (Putzrath et al., 1981; Bos et al.,
1983; Sorsa et al., 1985) demonstrated increased
mutagenicity in adult passive smokers. These stud-
ies generally examined small numbers of people
and did not control for dietary factors; therefore,
the results might be considered somewhat ambigu-
ous. In spite of the shortcomings, these studies did
support the use of bacterial bioassays as a screen-
ing tool for human exposure to ETS genotoxi-
cants; however, more effort is needed in impro-
ving and standardizing the methods and in creat-
ing the proper controls for other environmental
factors such as diet.
In summary, our studies support previous stud-
ies that demonstrated that ETS particle-bound
organic material is mutagenic. In addition, our
studies demonstrated that some semi-volatile and
volatile components were mutagenic; however, the
highly volatile compounds (for both MS and SS)
collected in the third-stage liquid nitrogen cold
trap were not mutagenic. Within these studies, the
total mutagenicity was divided among 4 fractions
approximately as follows: SS sand trap (par-
ticulate) fraction, 40%; MS sand trap (paniculate)
fraction, 30%; SS solvent (semi-volatile) fraction.
20%; and MS solvent (semi-volatile) trap fraction,
10%. Results also gave an indication that the
frequency with which a cigarette is puffed affects
the total amount of mutagenic material produced.
Although these studies illustrate the usefulness of
bacterial mutagenicity bioassays for characterizing
ETS, there are also other uses (e.g.. identifying
specific genotoxicants) for which bacterial assays
will find great utility.
Acknowledgements
The technical assistance of Linda Monteith,
Debra Simmons, Jack Callahan. Ken Davis, and
Jeff Keever is gratefully appreciated. The authors
wish to thank Dr. Goran LSfroth and Ms Virginia
Houk for their careful review and suggestions for
the manuscript. The research described in this
paper has been reviewed by the Health Effects
Research Laboratory, U.S. Environmental Protec-
tion Agency, and approved for publication. Ap-
proval does not signify that the contents neces-
sarily reflect the views and policies of the Agency
nor does the mention of trade names or commer-
cial products constitute endorsement or recom-
mendation for use.
References
Alfheim, I., and T. Ramdahl (1984) Contribution of wood
combustion to indoor air pollution as measured by muta-
genicity in Salmonella and polycyclic hydrocarbon con-
centration. Environ. Mutagen., 6. 121-130.
Ames. B.N., J. McCann and E. Yamasaki (1975) Methods for
detecting carcinogens and mutagens with the Salmonella/
mammalian-microsome mutagenicity test. Mutation Res
31. 347-364.
Ayer. H.E.. and D.W. Yeager (1982) Irritants in cigarette
smoke plumes. Am. J. Publ. Health. 72. 1283-1285.
-------
98
Badre. R., R. Guillerme, N. Abram, M. Bourdin and C. Dumas
(1978) Pollution atmospherique par la fumee de (abac,
Ann. Pharm. Fr., 36, 443-452.
Baker, R.R. (1981) Product formation mechanisms inside a
burning cigarette. Progr. Energy Combustion Sci.. 7,
135-153.
Baker, R.R. (1982) Variation of sidestrcam gas formation
during the smoking cycle, Beitr. Tabakforsch., 11, 181-193.
Bos. R.P.. J.L. Theuws and P.T. Henderson (1983) Excretion of
mutagens in human urine after passive smoking. Cancer
Leu., 19. 85-190.
Brunnemann, K.D.. and D. Hoffman (1975) Chemical studies
on tobacco smoke. XXXIV. Gas chromatographic de-
termination of ammonia in cigarette and cigar smoke. J.
Chromatogr. Sci., 13. 159-163.
Brunnemann, K.D.. and D. Hoffmann (1978) Analysis of
volatile nitrosamines in tobacco smoke and polluted indoor
environments, IARC, 19, 343-356.
Brunnemann. K..D., H.-C. Lee and D. Hoffmann (1976) Chem-
ical studies on tobacco smoke. XLVII. On the quantitative
analysis of catechols and their reduction. Anal. Lett., 9.
939-955.
Brunnemann, K.D.. L. Yu and D. Hoffmann (1977) Assess-
ment of carcinogenic volatile /V-nitrosamines in tobacco
and in mainstream and sidestream smoke for cigarettes.
Cancer Res.. 37. 3218-3222.
Brunnemann. K.D., G. Stahnke and D. Hoffman (1978) Chem-
ical studies on tobacco smoke. LXI. Volatile pyridines:
quantitative analysis in mainstream and sidestream smoke
of cigarettes and cigars, Anal. Lett.. All (7), 545-560.
Brunnemann. K.D., W. Fink and F. Moser (1980) Analysis of
volatile ,V-niirosamines in mainstream and sidestream
smoke from cigarettes by GLC-TEA, Oncology, 37,
217-222.
Brunnemann. K.D., J. Masaryk and D. Hoffman (1983) Role
of tobacco stems in the formation of AT-nitrosamines in
tobacco and cigaret mainstream and sidestream smoke, J.
Agric. Food Chem.. 31. 1221-1224.
Claxton. L.D.. J. Allen. A. Auletta, K. Mortelmans. E. Nest-
mann and E. Zeiger (1987) Guide for the Salmonella
typhimurium /mammalian rrucrosome tests for bacterial
mutagenicity. Mutation Res., 189, 83-91.
Elliott. L.P., and D.R. Rowe (1975) Air quality during public
meetings. J. Air Pollut. Contr. Ass., 25, 356-363.
Fischer, T., A. Weber and E. Grandjean (1978) Air pollution
due to tobacco smoke in restaurants, Int. Arch. Occup.
Environ. Health. 41. 267-280.
Galuskinova, V. (1964) 3. 4-Bcnzpyrene determination in the
smoky atmosphere of social meeting rooms and re-
staurants: a contribution to the problem of the noxiousness
of so-called passive smoking, Neoplasma, 11, 465-468.
Garfinkle. L. (1981) Time trends in lung cancer mortality
among nonsmokers and a note on passive smoking. J. Natl.
Cancer Inst.. 66, 1061-1066.
Graedel. T.E, D.T. Hawkins and L.D. Claxton (1986) Atmo-
spheric Chemical Compounds: Sources, Occurrence, and
Bioassay, and Bioassay, Academic Press. Orlando. FL. 732
PP-
Grimmer. G.. H. Boehnke and H.P. Harke (1977a) Passive
smoking: measuring of concentrations of polycyclic
aromatic hydrocarbons in rooms after machine smoking
cigarettes. Int. Arch. Occup. Environ. Health, 40, 83-92.
Grimmer, G., H. Boehnke and H.P. Harke (1977b) Passive
smoking: intake of polycyclic aromatic hydrocarbons by
breathing air containing cigarette smoke. Int. Arch. Occup.
Environ. Health, 40, 93-99.
Grimmer, G., K..-W. Naujack and G. Dettbam (1987) Gas
chromatographic determination of polycyclic hydro-
carbons, aza-arenes. aromatic amines in the particle and
vapor phase of mainstream and sidestream smoke of
cigarettes, Toxicol. Lett., 35, 117-124.
Harmsen. H., and E. Effenberger (1957) Tabakrauch in
Verkehrsmitteln, Wohn- und Arbeitsraumen, Arch. Hyg.
Bacteriol., 141, 383-400.
Hinds, W.C., and M.W. First (1975) Concentrations of nico-
tine and tobacco smoke in public places. New Engl. J.
Med., 292, 844-845.
Hirayama, T. (1981) Non-smoking wives of heavy smokers
have a higher risk of lung cancer: a study from Japan. Br.
Med. J., 282, 183-185.
Hughes, T.J., D.M. Simmons, L.G. Monteith and L.O. Claxton
(1987) Vaporization techniques to measure mutagenic activ-
ity of volatile organic chemicals in the Ames/Salmonella
assay. Environ. Mutagen.. 9 (4), in press).
Hugod, D., L.H. Hawkins and P. Astrup (1978) Exposure of
passive smokers to tobacco smoke constituents. Int. Arch.
Occup. Environ. Health. 42, 21-29.
Husgafvel-Pursiainen. K., M. Sorsa. M. Moller and C. Be-
nestad (1986) Genotoxicity and polynuclear aromatic hy-
drocarbon analysis of environmental tobacco smoke sam-
ples from restaurants, Mutagenesis, 1, 287-292.
IARC (International Agency for Research on Cancer) (1986)
IARC Monographs on the Evaluation of the Carcinogenic
Risk of Chemicals to Humans, Vol. 38: Tobacco Smoking.
IARC, Lyon. 421 pp.
Just, J., M. Borkowska and S. Maziarka (1972) Tobacco smoke
in the air of Warsaw coffee rooms, Rocz. Panstw. Zakl.
Hig., 23. 129-135.
Klus. H., and H. Kutin (1982) Distribution of different tobacco
smoke constituents in mainstream and sidestream smoke: a
review. Beitr. Tabakforsch., 11, 229-265.
Lewtas, J. (1985) Development of a comparative potency
method for cancer risk assessment of complex mixtures
using short-term in vivo and in vitro bioassays, Toxicol.
Indust. Health. 1. 193-203.
Lewtas, J., S. Goto. K. Williams, J.C. Chuang, B.A. Peterson
and N.K. Wilson (1987) The mutagenicity of indoor air
particles in a residential pilot field study: applications and
evaluation of new methodologies, Atmosph. Environ., 21.
443-449.
Ling, P.I.. G. LBfroth and J. Lewtas (1987) Mutagenic
determination of passive smoking, Toxicol. Lett.. 35,
147-151.
Liu. Y.Y.. I. Schmeltz and D. Hoffmann (1974) Chemical
studies on tobacco smoke. Quantitative analysis of hy-
drazine in tobacco and cigarette smoke. Anal. Chem.. 46,
885-889.
Lofroth, G.. and G. Lazaridis (1986) Environmental tobacco
-------
99
smoke: comparative characterization by mutagenicity as-
says of sidestream and mainstream cigarette smoke. En-
viron. Mutagen.. 8. 693-704.
LoTroth. G.. L. Nilsson and I. Alfheim (1983) Passive smoking
and urban air pollution: Salmonella/microsome mutagen-
icity assay of simultaneously collected indoor and outdoor
paniculate matter, in: M.D. Waters, S.S. Sandhu. J. Lewtas,
L. Claxton, N. Chernoff and S. Nesnow (Eds.), Short-Term
Bioassays in the Analysis of Complex Environmental Mix-
tures, Vol. III. Plenum, New York, pp. 515-525.
Monteith, L.G., D.M. Simmons, K.H. Davis, J.T. Keever, J.C.
Callahan. T.J. Hughes and L.D. Claxton (1986) Mutagenic
activity of sidestream and mainstream cigarette smoke in
the Ames/Salmonella assay. Environ. Mutagen., 8, 56.
Morin, R.S.. J.J. Tulis and L.D. Claxton (1987) The effect of
solvent and extraction methods on the bacterial mutagenic-
ity of sidestream cigarette smoke, Toxicol. Lett., 38,
279-290.
Myers. L.E., N.H. Sexton. L.J. Southerland and T.J. Wolff
(1981) Regression analysis of Ames test data. Environ.
Mutagen.. 3, 575-586.
National Research Council (1986) Environmental Tobacco
Smoke: Measuring Exposures and Assessing Health Ef-
fects. National Academy Press, Washington, DC, 337 pp.
Nesnow, S.. M. Argus, H. Bergman, K. Chu. C. Frith. T.
Helmes. R. McCaughy, V. Ray, T.J. Slaga, R. Tennant and
E. Weisberger (1986) Chemical carcinogens: a review and
analysis of the literature of selected chemicals and the
establishment of the Gene-Tox Carcinogen Data Base. A
report of the U.S. Environmental Protection Agency-Gene-
Tox Program. Mutation Res.. 185, 1-195.
Ong. T.-M.. J. Stewart and W.-Z. Whong (1984) A simple in
situ mutagenicity lest system for detection of mutagenic air
pollutants. Mutation Res., 139,177-181.
Pryor, W.A., D.G. Prier and D.F. Church (1983) Electron-spin
resonance study of mainstream and sidestream cigarette:
nature of the free radicals in gas-phase smoke and in
cigarette tar. Environ. Health Perspect.. 47, 345-355.
Putzrath. R.M.. L. Langley and E. Eisenstadt (1981) Analysis
of mutagenic activity in cigarette smokers' urine by high-
performance liquid chromatography. Mutation Res.. 85.
97-108.
Rosen, F.L., and A. Levy (1950) Bronchial asthma due to
allergy to cigarette smoke in an infant: case report. J. Am.
Med. Ass.. 144. 620-621.
Sakuma. H., M. Kusama, S. Munakata, T. Ohsumi and S.
Sugawara (1983) The distribution of cigarette smoke com-
ponents between mainstream and sidestream smoke. I.
Acidic components. Beitr. Tabakforsch. Int.. 12. 63-71.
Sakuma. H., M. Kusama. K. Yamaguchi and S. Sugawara
(1984a) The distribution of cigarette smoke components
between mainstream and sidestream smoke. II. Buses. Beitr.
Tabakforsch. Int.. 22. 199-209.
Sakuma, H.. M. Kusama. K. Yamaguchi and S. Sugawara
(1984b) The distribution of cigarette smoke components
between mainstream and sidestream smoke. HI. Middle
and higher boiling components. Beitr. Tabakforsch. Int..
12. 251-258.
Sasson, I.M., D.T. Coleman, E.J. LaVoie. D. Hoffman and
E.L. Wynder (1985) Mutagens in human urine: effects of
cigarette smoking and diet. Mutation Res.. 158. 149-157.
Schmeltz. I.. J. Tosk and D. Hoffmann (1976) Formation and
determination of naphthalenes in cigarette smoke. Anal.
Chem., 48. 645-650.
Sexton, K., L.M. Webber, S.B. Hayward. R.G. Sextro and F.J.
Offerman (1984) Characterization of paniculate and organic
emissions from major indoor sources. Indoor Air. Proc. Int.
Conf. Indoor Air Qual.. 4 (PB85-104214). 163-168.
Sorsa, M., P. Einisto, K. Husgafvel-Pursiainen. H. Jarventaus.
H. Kivisto. Y. Peltonen. T. Tuomi. S. Valkonen and O.
Pelkonen (1985) Passive and active exposure to cigarette
smoke in a smoking experiment. J. Toxicol. Environ. Health.
16. 523-534.
Stara. J.F., and D. Kello (1979) Relationship of long-term
animal studies to human disease, in: S.S. Lee and J.B.
Mudd (Eds.), Assessing Toxic Effects of Environmental
Pollutants, Ann Arbor Science, Ann Arbor. Ml, pp. 43-76.
Stehlik. G.. O. Richter and H. Altman (1982) Concentration of
dimethylnitrosamine in the air of smoke-filled rooms.
Ecotoxicol Environ. Saf.. 6, 495-500.
Trichopoulos. D.. A. Kalandidi. L. Sparros and B. Maemahon
(1981) Lung cancer and passive smoking. Int. J. Cancer. 27.
1-4.
Wallace, L.A.. and E.D. Pellizzari (1986) Personal air ex-
posures and breath concentrations of benzene and other
volatile hydrocarbons for smokers and nonsmokers. Toxi-
col. Lett.. 35. 113-116.
Williams. D.C.. J.R. Whitaker and W.G. Jennings (19K5) Mea-
surements of nicotine in building air as an indicator of
tobacco smoke levels. Environ. Health Perspect.. 60.
405-410.
Yahagi. T.. M. Degawa. Y. Seine. J. Matsushima. M. Nagao.
T. Sugimura and Y. Hashimoto (1975) Mutagenicity of
carcinogenic dyes and their derivatives. Cancer Lett.. 1.
91-97.
Yamasaki, E.. and B.N. Ames (1977) Concentration of muta-
gens from urine by adsorption with the nonpolar resin
XAD-2. cigarette smokers have mutagenic urine. Proc. Nail.
Acad. Sci. (U.S.A.). 74. 3555-3559.
-------
Reprinted from ENVIRONMENTAL SCIENCE & TECHNOLOGY. VoL 23, Page 610, May 1989
lit © 1989 by the American Chemical Society and reprinted by permiMion of the copyright owner.
Characterization of Environmental Tobacco Smoke
Goran Lofroth,r Robert M. Burton,* Unda Forehand,5 S. Katharine Hammond,11 Robert L. Seila,*
Roy B. ZweJdinger,* and Joellen Lewtas*-1-
Heatth Effects Research Laboratory and Atmospheric Research and Exposure Assessment Laboratory, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina 27711, Environmental Health Research & Testing Inc., Research
Triangle Park. North Carolina, and Department of Family and Community Medicine, University of Massachusetts Medical
School, Worcester, Massachusetts
• Environmental tobacco smoke (ETS) has been analyzed
with respect to several components following smoking of
research cigarettes in an experimental chamber. Param-
eters analyzed and their airborne yield per cigarette in-
cluded particulate matter (10 mg) and its mutagenic ac-
tivity in a Salmonella bioassay, carbon monoxide (67 mg),
nitrogen oxides (2 mg), nicotine (0.8-3.3 mg), formaldehyde
(2 mg), acetaldehyde (2.4 mg), acrolein (0.56 mg), benzene
(0.5 mg), and several unsaturated aliphatic hydrocarbons
(e.g., 1,3-butadiene) of which isoprene (3.1 mg) had the
highest yield. ETS from commercial cigarettes was like-
wise analyzed in the experimental chamber and at a public
location. The relative component composition for ETS is
similar when generated from either research or commercial
cigarettes. All components analyzed were present at
concentrations above the background concentrations.
Isoprene might be utilized as a tobacco smoke tracer for
unsaturated aliphatic hydrocarbons.
Introduction
Environmental tobacco smoke (ETS), which is derived
primarily from sidestream smoke emitted between puffs,
is a major contributor to indoor air pollution wherever
smoking occurs (1, 2). ETS differs both chemically and
physically from the precursor sidestream smoke, presum-
ably due to chemical and physical transformations that
occur as the mixture is diluted and aged. Chemical
characterization studies have focused on mainstream and
sidestream smoke (1). Data are lacking, however, on the
presence and concentration of potentially toxic and car-
cinogenic components in tobacco-smoke-polluted indoor
environments. An ideal ETS tracer air contaminant is not
available for total ETS exposure (2), although nicotine is
the best available tracer.
In this study we investigated the concentration of a
number of genotoxic components as well as potential
tracers of ETS under controlled and environmental con-
ditions. Some of the components measured are routinely
monitored air pollutants including carbon monoxide, ni-
trogen oxides, and particulate matter. A series of aldeh-
ydes and alkenes were measured in these studies, including
several that are carcinogenic. The mutagenicity of the
particulate phase was assayed in Salmonella typhimurium,
Nicotine was measurd as an ETS tracer. Indoor chamber
experiments were performed at the EPA facility at the
University of North Carolina, Chapel Hill, partly in con-
junction with studies on the urinary cotinine (nicotine
metabolite) concentration and excretion rate in young
children following exposure to sidestream cigarette smoke
'Visiting Scientist at U.S. EPA from Nordic School of Public
Health, P.O. Box 12133, S-402 42 Gothenburg, Sweden.
1 Atmospheric Research and Exposure Assessment Laboratory,
U.S. EPA.
'Environmental Health Research & Testing Inc.
1 University of Massachusetts.
x Health Effects Research Laboratory, Genetic Bioassay Branch,
MD68, U.S. EPA.
(3). Indoor measurements were also made in a tavern.
Experimental Section
Chamber and Smoking. The tests were performed in
a 13.6-m3 Plexiglas chamber (4) set at a ventilation rate
of 3.55 air changes h~l; in addition, air removed by the
sampling added ~0.50 air changes h~l. The air in the
chamber was circulated by a fan at 1.35 m3 h"1. The tem-
perature and the relative humidity are given in Table I.
Research cigarettes of the type 2R1 (5), which had been
equilibrated at 22 °C at 60% relativity humidity for 48 h,
were smoked by machine (RM30, Heinr. Borgwalt, Ham-
burg, FRG). One cigarette was lighted every 30 min and
was smoked with a 35-mL puff of 2 s every minute until
extinguished after ~12 min. Mainstream smoke was
vented to the outside of the chamber. The cigarettes
weighed ~ 1.2 g, of which 0.9-1.0 g was consumed.
One adult and one child were present in the chamber
during the 4-h tests in the first series of nine experiments.
Six additional experiments were performed with the re-
search cigarettes smoked by machine later in a second
series, including two tests with no smoking, two tests (13
and 14) similar to the first series, (one cigarette every 30
min), and two tests (15 and 16) with one cigarette every
15 min. In tests 15 and 16, decay of components in the
chamber was measured. Subsequently, in a third series
of chamber tests, the emissions from two different com-
mercial cigarette brands (A and B, both low-tar and •
nicotine brands) were analyzed in the chamber with regular
smoking by one person without any applied ventilation.
Sampling and Analysis. Particle Sampling and
Analysis. Total suspended particles (TSP) were collected
in duplicate on preweighed Teflon-coated glass fiber filters
(Pallflex) at 1.7 m3 h'1 by modified Anderson samplers
consisting of the 10-mm preseparator and the backup filter.
TSP was also measured continuously by an Electric Aer-
osol Analyzer, EAA (Thermo-System, Inc., Model 3030),
with measurements taken every 9 or 10 rnm. Particles were
also collected in triplicate with personal sampling pumps
(Model P4000, Du Pont, Kennett Square, PA) at 1.7 and
3 L min"1.
Nicotine. Nicotine was collected on bisulfate-impreg-
nated filters (6) placed downstream from the particle filters
on the personal samplers (first series) or on both Anderson
and personal samplers (second series). Extraction and gas
chromatography analysis of nicotine was performed as
described by Hammond et al. (6).
Particle Mutagenicity. The filters were extracted by
sonication in dichloromethane, and the extract was brought
to a fixed volume. Aliquots of the solution were distributed
into 4-mL vials together with 5 nL of dimethyl sulfoxide
(DMSO) and then evaporated by nitrogen gas at 35 °C.
The vials were kept capped at -20 °C until bioassayed.
The mutagenicity was determined by a microsuspension
assay developed by Kado et aL (7) and modified by De-
Marini et al. (in preparation) using Salmonella TA98 in
the presence of S9 (8). The microsuspension was modified
by using a bacterial suspension concentrated 5 times in-
610 Environ. Set Techno!., Vol. 23, No. 5. 1989
0013-936X/89/0923-0610$01.50/0 © 1989 American Chemical Society
-------
Table I. Average Chamber Concentrations (±SD) and Average Airborne Yield* for Carbon Monoxide, Nitrogen Oxides. Total
Suspended Particles, Particle Mutagenicity, and Nicotine daring Smoking of One Cigarette Every 30 Minutes (Tests 1-9,13,
and 14) and Every 15 Minutes (Tests 15 and 16) in a 13.6-m1 Chamber with an Air Exchange of 3.55 h'1.
component series 11-9
relative humidity, % 53 ± 4
temperature, °C 24 ± 0.5
carbon monoxide, mg/m3 2.48 ± 0.29
nitric oxide, Mg/m3 68.0 ± 6.0*
nitrogen oxides,'' Mg/m3 72.8 ± 6.9*
total suspended particles, Mg/m3
by mas* 349 ± 39
by EAA« 349 ± 45?
mutagenitity, revertants/m3
personal sampler 628 ± 49
Anderson sampler 494 ± 58
nicotine, Mg/m3
personal sampler 1-9 29 ± 7
personal sampler 13-16 nd
Anderson sampler 13,14 nd
series II 13 and 14
29± 1
23 ±0
1.79 ± 0.81
61.0 ± 2.8
65.0 ± 8.5
321 ±25
513 ± 47
nd*
517 ±40
nd
127 ±23
150 ± 14
15 and 16 airborne yield per cigarette*
34 ±0
23 ±0
4.76 ± 0.21
139 ±6
139 ±9
934 ±46
1223 ±84
nd
837 ±76
nd
228 ±49
nd
6
6
67 mg
6
1950 ng
10 mg
b
17300 revertants
13400 revertants
800 Mg
3300 Mg
6
• Unconnected for surface removal * Not applicable. * Based on eight tests. d Expressed as NO from total concentration of NO and NO*.
* Assuming unit density. 'Based on four tests, 'nd, not determined.
stead of 10 times, 0.015 M phosphate-buffered saline in-
stead of 0.15 M, no shaking during the 90-min incubation
of the vials at 37 °C, and addition of histidine and biotin
to the plate bottom agar instead of to the top agar.
The combined sample from the duplicate Anderson
filters from each experiment was tested with six doses
corresponding to 25-300 L of air in duplicate tests with
duplicate vials for each dose and test. The combined
sample from the personal filters from each experiment was
tested with three doses corresponding to 50-200 L of air
in one test, with triplicate vials for each dose. The response
was calculated by linear regression using doses on the linear
or almost linear part of the dose-response curve.
Carbon Monoxide and Nitrogen Oxides. Carbon
monoxide (CO) was measured continuously by nondis-
persive infrared absorption (Bendix 8501-5), and nitrogen
oxides NO, (Le., NO plus NOj) were measured indirectly
by chemiluminescence (Bendix 8101-B). Data points were
recorded every 3 min.
Hydrocarbons. Air was collected in evacuated stainless
steel canisters (9), and the sample was then subjected to
speciated gas chromatographic analysis by the method
described by McElroy et al. (10). Samples in the first
experimental series were collected as grab samples at a
peak concentration of carbon monoxide in the chamber,
whereas samples in the second series were collected over
the entire smoking period (4 h).
Aldehydes. Aldehydes were collected in the second
series at a rate of 1.0 L min"1 using 2,4-dinitrophenyl-
hydrazine-coated silica gel cartridges for collection and
high-performance liquid chromatography for analysis of
the hydrazone derivatives as described by Tejada (11).
Calculations. The average chamber concentrations
were calculated as the average value between 1 h after start
until the end of the experiments. When sampling included
the first hour, the average concentration was calculated
by normalizing to the continuous CO concentration; this
correction was approximately 5%. Likewise, grab samples
of hydrocarbons were normalized to the average concen-
tration from the peak concentration, when the sample had
been collected. The airborne yield, expressed as amount
per cigarette, was calculated from the average concentra-
tion by using the known smoking frequency, the chamber
volume, and the total air exchange rate.
Environmental Sampling. The impact of tobacco
smoke was determined in two studies in a local tavern.
The main room in which sampling took place had a volume
of ~ 180 m3 (I = 15 m, w = 4 m, and h — 3 m) and was
variously occuplied by 5-25 persons, many of whom were
smoking.
Indoor TSP and nicotine were collected on a Teflon-
coated glass fiber filter and a second bisulfate-impregnated
filter, respectively, at 20 L/min by an Anderson sampler.
Particulate matter was measured by taking 120-s readings
each 1/2 h over the 3- or 4-h study with a piezobalance
Model 3500 (TSI Inc., St Paul, MN) both indoors and
outdoors with at least two cleaning cycles per hour. Indoor
and outdoor carbon monoxide was determined with two
General Electric Model 15ECS3CO3 carbon monoxide
detectors (Wilmington, MA) that had been calibrated at
zero and 60 ppm CO. Indoor aldehydes and indoor and
outdoor hydrocarbons were collected and analyzed as de-
scribed for the chamber studies. The hydrocarbon sam-
pling was performed during only 2 h in each of the two
studies.
Results
The concentrations and calculated yields are given in
Table I for components that were analyzed in all chamber
tests in the first and second series. Carbon monoxide and
nitrogen oxides were determined continuously every 3 min,
and their concentrations varied in a saw-toothed form with
the smoking cycle of one cigarette every 30 min. The ratio
of the average maTimnin to the minimum concentration
was ~3. The average concentration of carbon monoxide
was about 65-70% of the mummum concentration; similar
ratios were found for nitrogen oxides.
Particle concentrations measured by EAA had the same
type of variation, but the resolution was less because the
analyses were performed less frequently. The average
concentration of particles as measured by EAA (assuming
unit density) was in good agreement with the concentration
obtained by filter collection in the first series and over-
estimated the particle concentration in the second series
under lower relative humidity. Due to the organic char-
acter of ETS, however, the density would be expected to
be somewhat less than 1.0.
The nicotine concentrations and yields were lower
during the first series than during the second series, with
yields of 800 Mg/cigarette and 3300 Mg/cigarette, respec-
tively. There were several differences in the two series.
In the first series, the chamber contained more adsorbant
surfaces: two persons, mother and child, television set,
crib, chair, and a curtain, all of which were absent in the
Environ. Set Tectmot. Vot 23 No. 5, 1989 611
-------
Table II. Selected Hydrocarbon Concentrations
-------
Finally, available portable equipment was used to sam-
ple and analyze the air in a tavern during normal smoking
conditions. All ETS components that were analyzed in-
doors and outdoors were highly elevated in the indoor
environment (Table V). Furthermore, none of the ana-
lyzed components was conspicuously much higher or much
lower relative to each other than what would have been
expected from the studies of the airborne yield of research
cigarettes.
Discussion
This study characterizes both exposure concentrations
and airborne yields for particulate matter and its muta-
genic activity, as well as nicotine, aldehydes, and alkenes.
This study provides documentation that the chamber ETS
exposure was comparable to that which people would en-
counter in indoor environments where tobacco is being
smoked. Additional chemical analyses and subsequent
studies were conducted to relate the chamber ETS com-
ponents to the analysis of ETS in an indoor environment
ETS and other air pollutants emitted into an indoor
environment can disappear by three routes: ventilation,
surface deposition, and chemical reactions while airborne.
The ventilation rate in most indoor environments is 0.5
air change/h or higher, which means that the time frame
of interest is a few hours or less. Whereas few, if any,
studies have dealt with chemical reactions of ETS com-
ponents, there is good evidence that smoke particles can
be removed by surface deposition (2,12), the process being
dependent on surface characteristics and mixing ratio.
Thus, the airborne yield of particulate tar can vary de-
pending on the experimental conditions.
Among the gases studied, carbon monoxide is considered
to be sufficiently stable to be removed only by ventilation.
This is also probably the case for the low molecular weight
hydrocarbons, ethene to isoprene, analyzed in the present
investigation, whereas nicotine as well as the aldehydes
may decay by surface adsorption or reaction. This phe-
nomenon may account for the divergent nicotine concen-
trations found in the first and second series (Table I),
because the chamber was different with respect to surface
characteristics in these two series.
Nitric oxide (NO) is the primary nitrogen oxide formed
in tobacco smoke (13), but it can slowly be oxidized to
nitrogen dioxide, NO2 (14), or species detected as NO2.
The low contribution of NO2 to the total concentration of
nitrogen oxides found in the present study (Table I) most
likely reflects the high ventilation rate, which would not
give sufficient time for the formation of NO2 from NO in
the chamber. In contrast, about 15-25% of the nitrogen
oxides detected in the smoking of commercial cigarettes
(Table IV) was in the form of NO2, indicating that both
the higher concentration and the lower ventilation rate in
these tests resulted in a significant conversion.
The determined ETS airborne yields of carbon mon-
oxide, nitrogen oxides, and nicotine (Table I) are about
the same as those reported for sidestream smoke from
commercial cigarettes (13, 15, 16). The ETS yield of
particles is, however, lower by a factor of 2-3 than those
reported for sidestream smoke from commercial cigarettes
(16). One study of 2R1 cigarettes by Ueno and Peters (17)
found only 6-9 mg of particulate matter/cigarette based
on a sample collected with an Anderson cascade impactor
and 1-2 mg/ cigarette based on EAA measurements.
The mutagenic yield of particulate matter, 13400-17 300
revertants/tigarette, is lower than the mutagenic emission,
36500-118000 revertants/cigarette, for sidestream ciga-
rette smoke collected in a small hood (18). These differ-
ences may be due to the differences between the sample
collection methods for sidestream smoke and ETS and
between the emission rate and airborne yield measure-
ments. In both cases, the loss of mutagens associated with
particles is likely due to loss of the particles to surfaces.
The mutagenicity concentrations and yields determined
on particle extracts from the personal sampler were con-
sistently higher than those from the modified Anderson
sampler. The personal samplers have a 2-fold lower face
velocity (1.3 cm s~3) compared to the Anderson sampler
(2.7 cm s~l). The higher face velocity of the Anderson
sampler may result in the loss of the more volatile organics
and mutagens from the filter. Recent studies comparing
face velocities substantiate this hypothisis (K. Hammond
et al., unpublished data). The mutagenic response of
400-900 revertants/m3 of air is a range that may be en-
countered in moderately smoky environments and is higher
than that found for ambient outdoor air (79).
Airborne yield is a direct measure of the components
present in a particular indoor environment and will vary
with the surface area and characteristics. The advantage
of this measurement is that it can be directly used to
estimate indoor ETS compound concentrations based on
the number of cigarettes smoked and the ventilation.
Emission factors for sidestream cigarette smoke have
classically been determined by using a small-volume col-
lecting device surrounding the cigarette tip (16), and
emission factors for ETS particles and nicotine have been
determined in chambers with correction of surface removal
(6). Thus, it can be expected that airborne yields are less
than emission factors for components that are significantly
removed by processes other than ventilation.
The emission of aliphatic hydrocarbons in sidesteam
smoke has not been assessed quantitatively previously,
although there are several earlier studies on mainstream
smoke that have been summarized by Elmenhorst and
Schultz (20). The airborne yields for sidestream smoke
found in the present study (Table U) are generally higher
than those found for mainstream smoke. Isoprene is the
predominant unsaturated hydrocarbon in sidestream
smoke (Tables II, IV, and V), and the concentrations
measured were well above the background concentrations.
There are several other sources for isoprene: It is exhaled
by man (21, 22) and rodents (23) and thus possibly by
other animals. It is emitted from vegetation, with ambient
concentrations generally below 15 mg/m3 (24, 25). It is
also produced during combustion, but the most likely
combustion source, wood combustion, gives much less
isoprene than ethene (26), indicating that isoprene is a
minor constituent of the hydrocarbon emission. The
carbonyl compounds studied (Table III) had airborne
yields of a magnitude reported earlier (15).
The results presented in Table IV, obtained in the
smoking of commercial cigarettes in the chamber, show
that there is a similar relative distribution of major com-
ponents from such cigarettes when compared to the air-
borne yields from the research cigarettes 2R1. This is also
the fact for components measured in a tavern (Table V)
in which a mixture of commercial cigarettes was being
smoked. These comparisons between research cigarettes
and commercial cigarettes show that both types of ciga-
rettes give rise to ETS with very similar composition.
The ratio of nicotine to particles is ~ 160 Mg/mg for the
tavern samples (Table V), which is intermediate between
the ratios that can be calculated for the chamber study,
80 and 330 Mg/mg. This may indicate that the tavern has
surface characteristics with respect to nicotine removal
intermediate between the occupied and unoccupied Plex-
iglas chamber.
-------
It is well-known that tobacco smoking causes cancer, and
recently a series of epidemiological studies reviewed by the
National Research Council (2) and others (27,28) have
reported excess lung cancer deaths in individuals exposed
to ETS. Cancer from passive smoking at sites other than
the lung is also a possibility (29).
It is not known which of the many components present
in tobacco smoke and ETS are the most hazardous. It is
therefore important to analyze ETS for a variety of com-
ponents comprising both participate matter and gas-phase
constituents. We have in the present study determined
particulate matter and used a Salmonella mutagenicity
bioassay to measure genotoxicity (50). Among the nu-
merous gas-phase compounds in ETS, aldehydes (1) and
unsaturated aliphatic hydrocarbons (31) are known or
potential animal carcinogens. Although these compounds
have a relatively low carcinogenic potency, they might be
of importance in the total evaluation because they are
present in relatively high concentration. Among the un-
saturated hydrocarbons, isoprene might be used as a to-
bacco smoke tracer considering the low background con-
centration of this compound. Studies are needed to ex-
amine other potential indoor sources of isoprene that could
interfere with its use as a tracer.
Acknowledgments
We thank A. A. Strong for operation of the chamber, M
M. Dallas for technical bioassay support, and D. M. De-
Marini for advice and review comments.
Registry No. Carbon monoxide, 630-08-0; nitric oxide,
10102-43-9; nitrogen oxides, 11104-93-1; nicotine, 54-11-5; ethene,
74-85-1; ethane, 74-84-0; propane, 115-07-1; propane, 74-98-6;
1,3-butadiene, 106-99-0; isoprene, 78-79-5; benzene, 71-43-2;
formaldehyde, 50-00-0; acetaldehyde, 75-07-0; acrolein, 107-02-8.
Literature Cited
(1) Tobacco Smoking, IARC Monographs on the Evaluation
of the Carcinogenic Risk of Chemicals; International Agency
for Research on Cancer Lyon, France, 1986.
(2) Environmental Tobacco Smoke, N&tioaaLReteuch Council;
National Academy Press: Washington, DC, 1986.
(3) Goldstein, G. M.; Collier, A.; Etzel, R.; Lewtas, J.; Haley,
N. In Indoor Air Quality and Climate; Seifert, B. et aL,
Eds.; Institute for Water, Soil and Air Hygiene: Berlin,
PRO, 1986; VoL 2, pp 61-67.
(4) Strong, A. A.; Penley, R.; Knelson, J. R Human exposure
system for controlled ozone atmospheres. U.S. Environ-
mental Protection Agency, Research Triangle Park, NC,
1977; EPA-600/1-77-048.
(5) Tobacco and Health Research Institute The Reference and
Research Cigarette Series; University of Kentucky: Lex-
ington, KY, 1984.
(6) Hammond, S. K.; Leaderer, B. P.; Roche, A C.; Schenker,
M. Atmos. Environ. 1987,21, 457-462.
(7) Kado, N. Y.; Langley, D.; Eisenstadt, E. Mutat. Ret. 1983,
121, 25-32.
(8) Ames, B. N.; McCann, J.; YamasaJri, E. Mutat. Ret. 1975,
31, 347-364.
(9) Oliver, K. D.; PleS, J. D^ McClenny, W. A Atmot. Environ.
198«, 20,1403-1411.
(10) McElroy, F. P.; Thompson, V. L.; Holland, D. M.; Lonne-
man, W. A; Seila, R.L.J.Air Pollut. Control Auoc. 1986,
36, 710-714.
(11) Tejada, S. B. Int. J. Environ. Anal. Chem. 1986, 26,
167-186.
(12) Repace, J. L.; Lowrey, A. H. Science 1980,208, 464-472.
(13) Norman, V.; Ihrig, A M.; Larson, T. M.; Moss, B. L. Beitr.
Tabaktfonch. Int. 1983,12, 55-62.
(14) Vilcins, G.; Lephardt, J. O. Chem. Ind. (London) 1975,
974-975.
(15) Klus, R; Kuhn, R Beitr. Tabaktfonch. Int. 1982, 11,
229-265.
(16) Riekert, W. S.; Robinson, J. O; Collishaw, N. Am. J. Public
Health 1984, 74, 228-231.
(17) Ueno, Y.; Peters, L. K. Aerosol Sci. Technol. 1986, 5,
469-476.
(18) Ling, P. L; Lofroth, G.; Lewtas, J. ToxicoL Lett. 1987,35,
147-151.
(19) Lofiroth, G.; Ling, P. L; Agurell, E. Mutat. Ret. 1988,202,
103-110.
(20) Elmenhorat, H.; Schultz, C. Beitr. Tabaktfonch. 1968, 4,
90-123.
(21) Conkle, J. P.; Camp, B. J.; Welch, B. E. Arch. Environ.
Health 1975, JO, 290-295.
(22) Gelmont, O.; Stein, R. A.; Mead, J. F. Biochem. Biophys.
Res. Common. 1981, 99, 1456-1460.
(23) Peter, H.; Wiegand, H. J.; Bolt, R M.; Greim, H.; Walter,
G.; Berg, M.; Fiber, J. G. ToxicoL Lett. 1987, 36, 9-14.
(24) Lamb, R; Westberg, R; Allwine, G. Atmot. Environ. 1986,
20,1-8.
(25) Trainer, M.; Williams, E. J.; Parrish, D. D.; Buhr, M. P.;
Allwine, E. J.; Westberg, H. R; Fehsenfeld, F. C.; Liu, S.
C. Nature 1987. 329, 705-707.
(26) Kleindienst, T. E.; Shepson, P. B.; Edney, E. O.; Claxton,
L. D.; Cupftt, L. T. Environ. Sci. Technol. 1986,20,493-501.
(27) Wald, N. J.; Nanchahal, K.; Thompson, S. G.; Cuckle, H.
S. Br. Med. J. 1986, 299,1217-1222.
(28) Saracci, R.; Riboli, E. Mutat. Ret. 1989, 222,117-127.
(29) Perahagen, G. Mutat. Ret. 1989, 222,129-135.
(30) Claxton, L.; Morin, R. S.; Hughes, T. J.; Lewtas, J. Mutat.
Ret. 1989.222, 81-99.
(31) Lorroth, G. Mutat. Ret. 1989, 222, 73-80.
Received for review May 11,1988. Revised manuscript received
November 4,1988. Accepted December IS, 1988. The research
described in this article hat been reviewed by the Health Effects
Research Laboratory, U.S. Environmental Protection Agency,
and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the
Agency nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
Corrections:
mm should be tun as follows:
p. 610, col. 2, para. "Sampling and Analysis. Particle Sampling and
Analysis.", line 4: consisting of the IQ-fim preseparator and the backup
filter.
p. 613, col. 2, para. 3, line 14: [esoprene 1s emitted from vegetation, with
ambient] ... concentrations generally below 15 MQ/m3 (24, 25). It Is
•14 Environ. 3d. Technol.. Vol. 23. No. 5, 1989
-------
ENVIRONMENTAL TOBACCO SMOKE: MUTAGENIC EMISSION RATES AND
THEIR RELATIONSHIP TO OTHER EMISSION FACTORS
Joellen Lewtas and Katharine Williams
USEPA, Research Triangle Park, NC 27711, U.S.A.
Goran Lofroth
Nordic School of Public Health, Gothenburg, Sweden
Katharine Hammond
University of Massachusetts Medical School
Worcester, MA 01605, U.S.A.
Brian Leaderer
John B. Pierce, Yale University School of Medicine
New Haven, CT 06519, U.S.A.
Abstract
The objective of this study was to evaluate the emission rates and ex-
posure concentrations of mutagens, nicotine, and particles from cigarettes.
Studies were conducted under controlled laboratory and chamber conditions
as well as in personal residences. The mutagenicity of environmental
tobacco smoke (ETS) was evaluated in three bioassays using two strains of
Salmonella typhimurium. Strain TA98 was used in the standard plate-incor-
poration and microsuspension histidine reversion assays; and strain TM677
in a microsuspension forward mutation assay. The mutagenicity, expressed
either per ug particle mass or per ug nicotine, appeared to be a rela-
tively constant factor that did not vary significantly between various cig-
arette brands. These data are being used to model the emissions of muta-
gens to predict mutagenic exposure concentrations under various conditions.
Introduction
The concentration of mutagens associated with particulate matter in
indoor air is substantially increased in the presence of tobacco smoke (1,
6, 7). In a residential field study of 10 homes, we found the mutagenicity
per cubic meter of air sampled to be highly correlated with the number of
cigarettes smoked (4). Recent studies of the mutagenicity of ETS from
cigarettes either smoked or smoldering, suggest that the mutagenicity
emitted per cigarette is relatively constant (5, 7). Controlled chamber
studies showed that the emission rate of nicotine and particles per cigar-
ette from sidestream smoke is also relatively constant (2). The purpose of
this study was to determine the mutagenic emission rate and its relation-
ship to respirable particulate matter and nicotine in environmental tobacco
smoke (ETS).
-------
Methods
Environmental Tobacco Smoke (ETS) Generation, Sampling and Analysis
The smoking experiments, test chamber/ and sample collection methods
used are similar to those described in detail by Hammond et al. (2) and are
summarized here. KTS was generated by 4 smokers smoking at the rate of
R cigarettes per hour in a 34-m3 chamber. In these experiments, the cham-
ber fresh air supply rate was typically 2.6 air changes per hour with a
temperature of 19°C and relative humidity of 45%. Particulate matter was
collected on 37-mm Teflon-coated glass fiber filters or on 0.5-um milli-
pore type FH. Vapor-phase nicotine was collected downstream from the fil-
ters with a second filter impregnated with sodium bisulfate. The particu-
late filters were extracted with acetone by sonication and solvent exchanged
into dimethyl sulfoxide for bioassay by using a stream of dry nitrogen to
evaporate the acetone. The sodium bisulfate-treated filters were extracted
and analyzed for nicotine as described by Hammond et al. (2). Equilibrium
sampling and calculation of emission rates per gram of tobacco consumed
were determined as described by Hammond et al. (2) for nicotine.
Mutagenicity Assays
The principal bioassay employed in this study was a forward mutation
assay described originally by Skopek et al. (9) that uses S_. typhimurium
strain TM677. A microsuspension modification of this assay, described by
Lewtas et al. (4), was used to measure mutant frequency (MF) (mutants per
106 surviving cells) by plating and counting both 8-azaquanine-resistant
(mutant) colonies and surviving colonies. The mutagenicity in this bioas-
say was compared to that obtained in Salmonella typhimurium strain TA99
that was conducted by either (a) the plate-incorporation assay of protocol
of Maron and Ames (8), which will be referred to here as the standard Ames
assay, or (b) a microsuspension assay described originally by Kado et al.
(3) and modified in our laboratory by using 5x cell concentration; 0.015M
phosphate buffer, pH 7.4; and a reduced assay volume that will be referred
to here as the microsuspension reverse assay.
All of the bioassays were performed with the addition of Aroclor 1254-
induced male Sprague-Dawley rat liver S9 mixture prepared as described in
the above-referenced protocols. The initial linear portion of the dose-
response curve was used to determine the mutagenicity expressed as either
revertants per rag of particles, per m3 of air, or per cigarette. In all
assays both solvent controls (negative) and positive controls were assayed
simultaneously as described in the protocols (9, B, 3).
Results and Discussion
The mutagenicity of ETS was compared across six brands including
brands both high and low in tar and nicotine, filtered and unfiltered as
shown in Table 1. Despite the significant differences in the cigarette
brands, very little difference in the mutagenicity of ETS was observed when
measured either per ug of total suspended particulate (TSP) or per m3.
-------
The mutagenic emission rates shown in Table 1 were also very similar across
all six brands and ranged from 22,000-37,000 MF/g tobacco consumed. The
emission rate per cigarette would range from 11,000-18,500 MF/cig., assuming
that 0.5 g of tobacco is consumed per cigarette smoked.
The ratio of mutagenicity to TSP ranged from 0.7 to 1.4 MF/ug TSP.
The TSP and nicotine emission rates for these six brands was determined in
a similar series of experiments. The ETS nicotine emission rates (1.6-2.4
mg nicotine/g tobacco consumed) are very similar to that reported earlier by
Hammond et al. (2) for four brands. Only Brand E (2.2 mg nicotine/g tobac-
co) is identical to Brand B (1.8 mg nicotine/g tobacco) in this earlier
study. Using this nicotine data from separate experiments, the ratio of MF
to nicotine in ETS ranged from 8-16 MF/ug nicotine. This agrees with
subsequent experiments not reported here where we have measured both muta-
genicity and nicotine on the same samples.
The blank chamber experiment was conducted without smokers present.
The TSP and mutagenicity concentrations in this nonsmoking experiment are
typical of outside urban air (4).
Table 1: Comparison of mutagenicity of ETS across several brands and
mutagenic emission rates compared to nicotine and particles
Cigarette
Brand
A*
B
C*
D*
E
pt
TSP
(ug/m3)
1177
1200
936
905
1318
901
Mutant
Frequency*
per m3
1178
1259
1186
1280
907
1025
Emission
Mutagen-
icity
(MF/g)
35,400
25,900
34,600
37,100
21,900
30,400
Rates
TSPb
(mg/g)
35.4
24.7
27.3
26.2
32.0
26.5
Ratio
of MF
to TSP
(MF/ug)
1.0
1.0
1.3
1.4
0.7
1.1
Ratio of
MF to
Nicotineb
(MF/ug)
10
16
15
12
8
Blank
Chamber
25
91
aMicrosuspension forward mutation assay in TM677. Determined from five
separate experiments from the linear portion of the dose-response curves.
^TSP and nicotine emission rates from these brands were determined in
separate chamber studies using the same protocol reported here. The nico-
tine emission rates in mg nicotine/g tobacco consumed were 2.4 (A), 1.6
(B), 1.6 (C), 2.1 (D), and 2.2 (E).
*Unfiltered cigarette, all others are filtered.
*Low tar and nicotine
-------
Table 2: Comparison of ETS mutagenicity per ug of TSP in different bio-
assays
Bioassays
ETS Sample
Micro
TM677
(MF/ug/ml)
Standard
TA98
(rev/ug)
Micro
TA98
(rev/ug/ml)
Chamber Brand Ea 0.69 2.9
Chamber 2R1b 0.79 1.6 7.0
Research Cigarette
aETS from studies in 34 m3 chamber at Yale TJ.; other data shown in Table 1.
bETS from studies in 13.6 m3 chamber at EPA using cigarette smoking ma-
chine. Ratio of nicotine to TSP was 87 ug nicotine/mg TSP.
ETS is present in indoor air in sufficiently high concentrations to
measure the mutagenicity in the standard Ames plate-incorporation assay,
which requires approximately 10 times the sample mass of the two microsus-
pension assays. Table 2 provides a comparison of the mutagenicity of ETS
per ug of TSP for several different cigarettes. Although it is possible
to use the standard Ames assay to measure the mutagenicity of either side-
stream cigarette smoke or ETS, there is one important advantage to using a
microsuspension assay. The microsuspension assays require 3-10 times•less
sample, thereby permitting mutagenicity measurements on samples of less
than 1 m3 of air. These methods will facilitate evaluation of the mutagen-
icity of personal air samples in human exposure assessment studies.
Disclaimer
The research described in this paper has been reviewed by the Health
Effects Research Laboratory, U.S. Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency nor does mention
of trade names or commercial products constitute endorsement or recommenda-
tion for use.
References
1. Alfheim I., and Ramdahl T. Contribution of wood combustion to indoor
air pollution as measured by mutagenicity in Salmonella and polycyclic
aromatic hydrocarbon concentration. Environmental Mutagenesis 6
(1984). 121-130.
-------
2. Hammond, S.K., Leaderer, B.P., Roche, A.C., and Schenker, M. Collec-
tion and analysis of nicotine as a marker for environmental tobacco
smoke. Atmospheric Environment 21 (1987), 457-462.
3. Kado, N.Y., Langley, 0., Eisenstadt, E. A simple modification of the
Salmonella liquid incubation assay: Increased sensitivity for detect-
ing mutagens in human urine. Mutation Research 121 (1983), 25-32.
4. Lewtas, J., Goto, S., Williams, K., Chuang, J.C., Petersen, B.A., and
Wilson, U.K. The mutagenicity of indoor air particles in a residential
pilot field study: Application and evaluation of new methodologies.
Atmospheric Environment 21 (1987), 443-449.
5. Ling, P.I., Lofroth, G., and Lewtas, J. Mutagenie determination of
passive smoking. Toxicology Letters 35 (1987), 147-151.
6. Lofroth, G., Nilsson, L., and Alfheim I. Passive smoking and urban air
pollution: Salmonella/microsome mutagenicity assay of simultaneously
collected indoor and outdoor particulate matter. In M.D. Waters, S.S.
Sandhu, J. Lewtas, L. Claxton, N. Chernoff and S. Nesnow (Eds.), Short-
Term Bioassays in the Analysis of Complex Environmental Mixtures III.
New York: Plenum Press, 1983, pp. 515-525.
7. Lofroth, G., and Lazaridis, G. Environmental tobacco smoke: Compara-
tive characterization by mutagenicity assays of sidestream and main-
stream cigarette smoke. Environmental Mutageneais 8 (1986), 693-704.
8. Maron, D.M., and Ames, B.M. Revised methods for the Salmonella
mutagenicity test. Mutation Research 113 (1983), 173-215.
9. Skopek, T.R., Liber, H.L., Krolewski, J.J., and Thilly, W.G. Quantita-
tive forward mutation assay in Salmonella typhimurium using 8-aza-
guanine resistance as a genetic marker. Proceedings National Academy
Science 75 (1978), 410-414.
-------
Toxicology Letters. 35 (1987) 147-151 147
Elsevier
TXL 01707
MUTAGENIC DETERMINATION OF PASSIVE SMOKING*
(Environmental tobacco smoke; ETS; Salmonella/microsome mutagenicity test)
PER IVAR LING*, GORAN LOFROTH'-"" and JOELLEN LEWTAS*
* Department of Radiobiology. University of Stockholm. Stockholm; " Nordic School of Public Health.
Gothenburg (Sweden), and ' Genetic Bioassay Branch, EPA, Research Triangle Park, NC (U.S.A.)
(Received 29 August 1986)
(Revision received 22 September 1986)
(Accepted 24 September 1986)
SUMMARY
The mutagenic activity of tobacco smoke has been further investigated with the plate-incorporation
method and a microsuspension technique of the Ames Salmonella assay. The microsuspension test gives
a higher response than the conventional plate incorporation test. It is possible to detect environmental
tobacco smoke (ETS) in moderately smoky indoor environments by collection of paniculate matter with
personal low volume samplers followed by particle extraction and mutagenicity testing with the micro-
suspension assay.
INTRODUCTION
Environmental tobacco smoke (ETS) is a very complex mixture of compounds
formed in the pyrolysis and combustion of tobacco products. Many of the com-
ponents are also formed in other anthropogenic combustion processes making it dif-
ficult or impossible to discriminate between ETS and other sources if these
components are used as indicators. Tobacco specific components are nicotine, some
N-nitrosamines and probably also some N-heterocyclic hydrocarbons and aromatic
amines [1] of which nicotine is being used for the specific determination of ETS
12,3].
• Presented at the International Experimental Toxicology Symposium on Passive Smoking. October
23-25, 1986. Essen (F.R.G.)
•• To whom correspondence and reprint requests should be addressed.
Abbreviations: DMSO, dimethyl sulfoxide; ETS, environmental tobacco smoke.
0378-4274/87/S 03.50 © Elsevier Science Publishers B.V. (Biomedical Division)
-------
148
The Ames Salmonella mutagenicity test has earlier been employed to show that
the mutagenic response of indoor airborne paniculate matter is high in the presence
of habitual smoking and that the mutagenic activity mainly is associated with basic
components making it possible to discriminate the activity of ETS from other com-
bustion sources [4,5].
Although the Salmonella plate incorporation assay is highly sensitive, the use of
low or medium volume air samplers in indoor environments may require assay
methods which give higher responses. We are therefore studying the mutagenic ac-
tivity of tobacco smoke in a microsuspension assay described by Kado and co-
workers [6] as giving responses several times those obtained with the conventional
plateincorporation test.
METHODS
Sampling
Mainstream cigarette smoke was collected on glass fiber filter with a smoking
machine which was operated under normal conditions as described earlier [5]. The
machine was placed in a 0.2 mj hood and the emitted sidestream smoke was col-
lected on glass fiber filter with a flow rate of 10 mVh [5].
Environmental samples were collected with a battery-operated system (AFC 123
Personal Air Sampler, Casella London Ltd., England) on glass fiber filter (diam.
2.5 cm) with a flow rate of 2 1/min.
For the present study, personal sampling was performed in an apartment during
a party representing a highly smoke-polluted environment and in an office with one
smoker representing more common indoor smoke conditions.
The apartment, having a volume of about 100 m', was sampled for 8 h during
which it was well ventilated with open windows; 82 butts were found at the end.
The office, having a volume of about 50 m1 and a ventilation rate exceeding 5
changes per hour, was sampled for 8 h on two separate days during which 20 and
25 cigarettes were smoked, respectively.
Extraction and sample preparation
Filters were extracted by sonication with acetone.
For the preparation of samples for the plate incorporation mutagenicity assay, an
aliquot of the extract was reduced to a small volume and then diluted with an ap-
propriate amount of DMSO [4,5]. The DMSO solution was stored at -20°C prior
to and between mutagenicity tests.
For the preparation of samples for the microsuspension mutagenicity test, ali-
quots of the acetone extract were dispensed into 8 cm x 1.1 cm sample tubes
together with 5 /»! DMSO. The samples were then evaporated under a stream of
nitrogen gas in a heating block at a temperature <40°C and the tubes sealed with
silicon rubber stoppers and stored at - 20°C until they were assayed.
-------
149
Mutagenicity assays
Mutagenicity was determined with the Ames Salmonella tester strains TA98 and
TA100 using the plate incorporation method [7] and the microsuspension method
described by Kado et al. [6]. Slight differences exist between the performance of the
mutagenicity assays as described by Lofroth and Lazaridis [5] and by Claxton [8]
and Austin et al. [9]. The liver-S9 was obtained from Aroclor 1254-induced male
Sprague-Dawley rats. Addition of S9 was generally with 10% S9 in the S9-mix, i.e.
SO id S9 per plate for the plate-incorporation assay and 10 n\ S9 per tube for the
microsuspension assay.
RESULTS
The mutagenic activity of the cigarette smoke samples is given in Table I and is
expressed as revertants per cigarette. The results obtained with the plate incorpora-
tion assay agree qualitatively and quantitatively with the previous study [5].
Mainstream cigarette smoke is mainly mutagenic in the presence of S9. The
sidestream smoke is also active in the presence of S9, but has, in addition, mutagenic
activity in the absence of S9 with the TA100 strain.
The mutagenic response in the microsuspension assay is higher and the enhance-
TABLE I
THE MUTAGENIC RESPONSE OF SIDESTREAM AND MAINSTREAM CIGARETTE SMOKE
USING THE SALMONELLA PLATE INCORPORATION AND MICROSUSPENSION ASSAYS
Sample
Stream Collected Revertants per cigarette1
type paniculate Plate incorporation assay
tar
mg/cig.
Microsuspension assay Ratio micro-
susp./plate
TA98
TA100
TA98
TAIOO TA98 TA100
-S9 + S9 -S9 +S9 -S9 + S9 -S9 +S9 -S9
Swedish
brand
Research
brand 2R1
side- 21.8
main- 17.6
side- 17.0
main- 33.3
side-c 10.7
American
generic
American side-d 10.5
generic
2300 31300 23800 34200 16300 81900 118200 2.6 5.0
-------
150
mem over the plate incorporation assay is about three-fold with TA98 in the
presence of S9. The response with TA100 in the absence of S9 is also enhanced with
the microsuspension assay.
The difference in the response obtained with the same type of sample from
American Generic cigarettes extracted and assayed in two different laboratories is
real; similar differences have been obtained with other tobacco smoke samples in-
dicating the need for normalizing the results between laboratories. -
The results of two representative indoor samples assayed by the microsuspension
method are given in Table II. The sample from the highly polluted apartment was
judged to contain sufficient activity for four dose series; two of these were used for
assays with TA98 in the presence of S9 and two with TA100 in the absence of S9.
The samples from the moderately smoky office were each only subdivided into one
dose series which were assayed with TA98 in the presence of S9.
The mutagenic responses detected in these indoor environments agree with
magnitudes which can be estimated from the air volume, the air exchange rate, •
number of cigarettes smoked and their emission factor and considering that the con-
centrations are modulated by surface deposition [5,10].
TABLE II
THE MUTAGENIC ACTIVITY OF INDOOR PART1CULATE MATTER COLLECTED WITH A
LOW VOLUME SAMPLER IN ONE HIGHLY AND ONE" MODERATELY SMOKY ENVIRON-
MENT AS ASSAYED WITH THE SALMONELLA MICROSUSPENSION TECHNIQUE
Apartment during party
Dose
Liter air per plate
0*
16
32
64
128
(rev./mj air)'
Positive controls
0.5 Mg benzo(a)pyrene
1 .0 «ig sodium azide
Office with one smoker*
Revertams
TA98 + S9
41
89
187
271
451
(3200)
435
-
per plate*
TA100-S9
251
356
483
785
1091
(6800)
_(
1020
Dose
Liter air per plate
0"
64
128
256
512
(rev./m1 air)'
0.25 pg 2-amino-
anthracene
Revertants per
plate"
TA98 + S9
65
81
114
175
263
(400)
1660
' These samples were collected, extracted and assayed at the Genetic Bioassay Branch, EPA.
" Average of duplicate plates.
• Average of two samples from the two different days.
* Spontaneous control determined with three or more plates.
' Estimated from the linear part of the dose-response curve.
1 Not tested.
-------
151
DISCUSSION
The Salmonella microsuspension mutagenicity assay is sufficiently sensitive that
it is possible to detect the smoking of a few cigarettes in an average apartment or
house with a normal ventilation of 0.5-1 air changes per hour during a sampling
period of 8 h with a personal sampler having a flow rate of 2-10 1/min.
The major part of the mutagenic activity of environmental tobacco smoke is caus-
ed by basic components [4,5], and if deemed necessary, simple fractionation and
mutagenicity test of the basic fraction can show whether a sample comprises tobacco
smoke. The Salmonella/nucrosome mutagenicity assay may thus be an alternative
or an adjunct to nicotine as a specific indicator of environmental tobacco smoke.
Further studies are, however, required to characterize indoor sources of mutagenic
activity as there may exist other sources of basic mutagenic components.
ACKNOWLEDGEMENT
This study was partly supported by the Cooperative Agreement CR812935-01-0
between the U.S. Environmental Protection Agency and the Nordic School of
Public Health covering a OVS program for GL.
REFERENCES
I Chemistry and analysis of tobacco smoke, in Tobacco Smoking, Vol. 38, IARC Monographs on the
Evaluation of the Carcinogenic Risk of Chemicals to Humans, IARC. Lyon, 1986, pp. 83-126.
2 M. Muramatus, S. Umemura, T. Okada and H. Tomita, Estimation of personal exposure to tobacco
smoke with a newly developed nicotine personal monitor. Environ. Res., 35 (1984) 218-227.
3 S.K. Hammond. B.P. Leaderer, A.C. Roche and M. Schenker. Collection and analysis of nicotine
as a marker for environmental tobacco smoke, Atmos. Environ., 20 (1986) in press.
4 C. Lofroth, L. Nilsson and I. Alfheim, Passive smoking and urban air pollution: Salmonella/m\-
crosome mutagenicity assay of simultaneously collected indoor and outdoor paniculate matter, in
M.D. Waters et al. (Eds.), Short-Term Bioassays in the Analysis of Complex Environmental Mixtures
III, Plenum Press, New York, 1983. pp. 515-525.
5 G. Lofroth and G. Lazaridis, Environmental tobacco smoke: comparative characterization by
mutagenicity assays of sidestream and mainstream cigarette smoke. Environ. Mutag., 8 (1986)
693-704.
6 N.Y. Kado, D. Langley and E. Eisenstadt, A simple modification of the Salmonella liquid-incubation
assay. Mutation Res., 121 (1983) 25-32.
7 D.M. Maron and B.N. Ames, Revised methods for the Salmonella mutagenicity test. Mutation Res..
113(1983) 173-215.
8 L.D. Claxton, Mutagenic and carcinogenic potency of diesel and related environmental emissions:
Salmonella bioassay, Environ. Int., 5 (1981) 389-391.
9 A.C. Austin, L.D. Claxton and J. Lewtas, Mutagenicity of fractionated organic emissions from
diesel, cigarette smoke condensate, coke oven, and roofing tar in the Ames assay. Environ. Mutag.,
7(1985)471-487.
10 J.L. Repace and A.H. Lowrey, Indoor air pollution, tobacco smoke, and public health. Science, 208
(1980) 464-472.
-------
Toxicology Letters, 38 (1987) 279-290 279
Elsevier
TXL 01853
THE EFFECT OF SOLVENT AND EXTRACTION METHODS ON THE
BACTERIAL MUTAGENICITY OF SIDESTREAM CIGARETTE SMOKE
(Soxhlet extraction; sonication; Salmonella typhimurium; dichloromethane;
methanoi; acetone; storage)
RANDALL S. MORIN"-'. JERRY J. TULIS" and LARRY D. CLAXTON"
'School of Public Health, University of North Carolina, Chapel Hill, NC275I4. and "Genetic Bioassay
Branch, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711 (U.S.A.)
(Received 16 March 1987)
(Revision received 10 May 1987)
(Accepted 6 June 1987)
SUMMARY
The 'mutagenic activity of sidestream cigarette smoke particles was estimated by testing sidestream
cigarette smoke particles which had been collected under controlled burning conditions in the laboratory.
Two different extraction methods (Soxhlet and ultrasonic agitation) and 3 different solvents
(dichloromethane, methanoi, and acetone) were compared for their efficiencies in the extraction of com-
pounds from sidestream cigarette smoke particles which are mutagenic in the Ames test. The mutagenic
activity of the sidestream smoke panicles was estimated to be 15 000-20 000 revertants per cigarette in
TA98 with metabolic activation and 12 000-17 000 revertants per cigarette in TA100 without metabolic
activation. Only weak mutagenic activity was detected in TA98 without activation and hi TA100 with
activation. Under test conditions used, ultrasonic agitation produced the most consistent results and
acetone extraction produced the highest levels of mutagenic activity.
Address for correspondence: Dr. Larry Claxton, (MD-68), Genetic Bioassay Branch, Health Effects
Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711,
U.S.A.
Portions of this research were done in partial fulfillment of a doctoral degree (R.S.M.), School of Public
Health, University of North Carolina, Chapel Hill, NC. U.S.A.
* Currently, Occupational Health Staff Officer. U.S. Army Health Services Command, Ft. Sam
Houston, TX 78234, U.S.A.
Abbreviations: ETS. environmental tobacco smoke.
0378-4274/87/S 03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
-------
280
INTRODUCTION
Environmental tobacco smoke (ETS) is composed of sidestream tobacco smoke
and exhaled tobacco smoke. Sidestream smoke enters the ambient air directly from
the burning tobacco source. Exhaled tobacco smoke is any combination of tobacco
smoke products (mainstream and sidestream) that have been inhaled and then exhal-
ed by an individual. The percentage of each mainstream smoke component exhaled
into the environment has been reported to range from as low as 1 % to as high as
70% [1,2]. The ratios of the various smoke components in sidestream smoke com-
pared to mainstream smoke have been reported to range from 1.2- to 46-fold [3].
A typical cigarette smoker inhales mainstream smoke for approximately 3 s, 8-10
times per cigarette. This is equivalent to a total of 24-30 s out of a total average
burn time of 12 min for each cigarette [4]. Sidestream cigarette smoke, therefore,
is produced during 96% of the total time a cigarette is burning.
It has been estimated that sidestream smoke forms approximately 85% of the
total tobacco smoke in the ambient indoor air [5]. Therefore, both smokers and
non-smokers could be exposed, especially in indoor spaces with low levels of air ex-
change, to significant levels of airborne pollutants.
Although a great deal of effort has been expended characterizing the harmful con-
stituents of mainstream tobacco smoke, much less is known about the toxic com-
ponents of sidestream tobacco smoke. Rickert et al. [6] tested 15 different brands
of cigarettes and reported that in. most cases considerably more tar, nicotine, and
carbon monoxide were present in the sidestream than in the mainstream smoke. The
findings from a 1983 study demonstrated that indoor, air contaminated with
sidestream cigarette smoke can be more mutagenic than ambient outdoor air in an
urban area [7]. It was reported that the smoke from one cigarette which is diluted
in a structure with a volume of 300 m1 would result in mutagenic activity 2.5 times
that found in urban ambient air.
More than 2000 compounds have been identified in mainstream tobacco smoke
[8]. Since sidestream smoke would be expected to be similarly complex, it is difficult
to identify the specific mutagenic constituents. A recent report by the National
Research Council [9] says 'Research is needed to standardize both the collection and
evaluation of ETS so that the effects of ETS can be studied in the laboratories'.
They also recommend that 'Further in vitro assays of ETS are needed'. The purpose
of this paper is to examine and compare appropriate methods for preparing
sidestream smoke samples for mutagenicity studies.
MATERIALS AND METHODS
Collection of samples
Sidestream cigarette smoke particles were collected by allowing a generic brand
(Price Breaker, Winn-Dixie Stores, Jacksonville, FL) of king-size filter cigarettes
-------
281
(738.5 mg tobacco/cigarette, average of 10 cigarettes), to burn in a Plexiglas
chamber measuring 32 cm x 38 cm x 29 cm (volume = 0.0353 m3). The air inside
the chamber was exhausted by a constant flow vacuum pump at a rate of 0.028
mVmin (1 cu. ft. of air per min). The air was pulled through a 142 mm diameter
Teflon-coated glass fiber filter (Pallflex Products Corp., Putnam, CT) supported in
a stainless steel filter stand. Filtered air was drawn into the chamber through three
1 cm diameter openings in the bottom of the chamber.
Before sample collection, all filters were heated for 24 h at 175°C and then desic-
cated for 24 h. After sampling, each filter was folded to enclose the sample and
rewrapped in aluminum foil. All samples were returned to the laboratory after col-
lection, usually within 2 h. To minimize sample degradation or reaction, samples
were stored on the filters in the dark at - 70°C.
Sample preparation
During preparation of the samples for mutagenicity testing, all solvents were glass
distilled, high purity, spectroscopic grade (Burdick and Jackson, Muskegon, MI).
To reduce artifact contamination only glass, stainless steel, and Teflon equipment
was used in the preparation of samples. All materials were washed with Alkonox
and consequently rinsed with distilled water, 50% nitric acid, 50% sulfuric acid,
distilled water, methanol, and extraction solvent.
Ultrasonic extraction. Each 142 mm diameter filter was divided into 8 replicate
pieces using stainless steel scissors. Each replicate was placed in a 50 ml glass vial
with a Teflon-lined screw cap along with 25 ml of solvent. One of 3 different
solvents (dichloromethane, methanol, or acetone) was used to extract the organic
material from the filter pieces. The vials were held in a plastic rack and placed in
a water bath sonicator (Sonicor, Model 401, Clean Room Products, Bay Shore, NY)
for 15 min at 25°C. Solvent was then decanted into a clean vial. The samples were
reextracted for an additional 15 min with 25 ml of fresh solvent. After the second
sonication, the extracts were combined to yield a sample volume of approximately
50ml.
Soxhlet extraction. Each replicate was placed in a 500 ml extraction flask with
250 ml of solvent and extracted for 24 h at a rate of 2-3 cycles/h using a standard
Soxhlet apparatus (Kontes Glassware, Vineland, NJ).
Refluxing of cigarette smoke extracts. Additional filters containing sidestream
cigarette smoke particles were extracted with acetone using the sonication procedure
above. The extracts were then split into 4 replicates. Three of the 4 replicates were
placed in individual 500 ml extraction flasks with 100 ml of acetone. The flasks were
attached to a Soxhlet extractor and refluxed for 8, 16, and 24 h respectively.
Sample concentration and solvent exchange. After each of the above pro-
cedures, the extraction solvent was removed by rotary evaporation in individual 250
ml round-bottom flasks. The samples were placed in a 40°C water bath during the
evaporation procedure. Each sample was concentrated to a volume of approximate-
-------
282
ly 5-10 ml and then quantitatively transferred into 15 ml volumetric tubes. The
samples were concentrated to a volume of 1-2 ml by immersing the tubes in a 45°C
water bath while a stream of dry nitrogen was bubbled through the sample using
stainless steel needles. A known volume of DMSO was added to each sample and
the remaining extraction solvent was evaporated using the nitrogen purge method.
Representative samples were analyzed by gas chromatography prior to bioassay to
check for adequate removal of the extraction solvent. In all cases, an unused filter
from each lot was extracted to determine background levels of mutagenic activity
due to the filter.
Salmonella mutagenicity testing
The Salmonella typhimurium plate incorporation assay was performed as describ-
ed by Ames et al. [10], with 2 modifications consisting of: (1) adding a trace amount
of histidine to the agar base layer instead of the overlay agar and (2) incubating the
plates for 72 h instead of 48 h. All samples were tested using Ames Salmonella
typhimurium tester strains TA98 and TA100 with and without exogenous metabolic
activation (S9) unless otherwise indicated. The S9 was prepared in the manner
described by Ames et al. [10] from Aroclor-induced, male CD-I rats. Tester strains
were obtained from Dr. Bruce Ames, University of California, Berkeley, CA and
maintained by the Genetic Bioassay Branch, Genetic Toxicology Division, Health
Effects Research Laboratory, U.S. EPA Environmental Research Center, Research
Triangle Park, NC. All dose levels were run in triplicate, unless otherwise noted.
Negative controls for spontaneous reversion and positive controls for each strain
were performed in triplicate for each experiment. The control chemicals used were
2-aminoanthracene for both strains with metabolic activation. Without metabolic
activation, 2-nitrofluorene and sodium azide were used as positive controls for
TA98 and TA100 respectively. Each sample was tested at 4-5 doses (5, 10, 50, 100,
300 ^I/plate). In most cases, these doses corresponded to 0.008, 0.016, 0.08, 0.16
and 0.48 cigarettes respectively.
All plates were counted using an automatic plate counter (Artek Model 880, Artek
Systems Corp., Farmingdale, NY). Each plate was counted 3 times and a mean for
the 3 counts was recorded. Data analysis was accomplished using the Ames test
curve fitting program of Stead et al. [11]. This program is a FORTRAN program which
fits a non-linear dose-response curve to the plate count data. Using the model slope
obtained from the dose-response curve, an estimate of mutagenicity was calculated
and expressed as the number of revertants per cigarette.
RESULTS
Effect of extraction method and solvent
Tables I and II present the bioassay results of the experiments designed to com-
pare the efficiencies of 2 different extraction methods and 3 different extraction
-------
283
TABLE I
THE EFFECT OF EXTRACTION METHOD AND SOLVENT ON THE MUTAGENIC ACTIVITY
OF SIDESTREAM CIGARETTE SMOKE PARTICLES TESTED IN SALMONELLA
TYPHIMUR1UM TA98 WITH METABOLIC ACTIVATION
Average of 4 experiment*. 5 doses, 3 replicates per dose, analysis by the method of Scead et al. [11|.
Extraction Number of revertants (thousands) per cigarette: as a function of extraction
method solvent
PCM* MEOH* ACET*
Soxhlet 12.2 ± 10.2 10.9 ±7.5 11.4 ± 5.1
Ultrasonic 8.0 ± 3.2 11.3 ± 3.7 14.6 ± 1.8
*DCM, dichloromethane; MEOH, methanol; ACET, acetone.
TABLE II
THE EFFECT OF EXTRACTION METHOD AND SOLVENT ON THE MUTAGENIC ACTIVITY
OF SIDESTREAM CIGARETTE SMOKE PARTICLES WHEN TESTED IN SALMONELLA
TYPH/MURIUM TA100 WITHOUT METABOLIC ACTIVATION
Average of 4 experiments.
Extraction Number of revertants (thousands) per cigarette: as a function of extraction
method solvent
PCM* MEOH* ACET'
Soxhlet 0.7 ±1.0 0.7 ± 1.2 1.9 ± 3.6
Ultrasonic 4.9 ± 3.7 0.4 ± 0.2 • 14.6 ± 6.5
'DCM, dichloromethane; MEOH; methanol; ACET, acetone.
TABLE 111
THE EFFECT OF REFLUXING ON THE MUTAGENIC ACTIVITY OF SIDESTREAM
CIGARETTE SMOKE PARTICULATE EXTRACT
Number of
hours
refluxed
0*
8
16
24
Number of revertants (thousands) per cigarette: S. typhimurium strain/activation
conditions
TA98/ •»- S9
19.6
15.3
17.2
16.2
TA98/- S9
0.5
1.9
3.8
l.i
TA100/ + S9
3.4
3.8
4.4
3.6
TAIOO/-
14.6
1.9
3.6
3.6
S9
'Extract tested for activity without refluxing.
-------
284
TABLE IV
THE EFFECT OF STORAGE ON THE MUTAGENICITY OF SIDESTREAM CIGARETTE SMOKE
PARTICLES
Storage conditions and time Number of revenants
(thousands) per cigarette
TA98 -i- S9 TAIOO - 59
Collected, extracted, bioassayed in less than 6 days
Extract stored frozen at - 70° C in DMSO for 23 days
Extract stored frozen at - 70° C in DMSO for 60 days
Panicles stored frozen at - 70° C in DMSO for 60 days
Panicles stored frozen at -70° C in DMSO for 180 days
17.1
15.9
5.1
9.3
11.5
14.6
3.3
1.7
2.1
0.4
solvents. For TA98 results with metabolic activation (Table I), the differences be-
tween the 2 methods and 3 solvents were small. The activity in TAIOO without
metabolic activation (Table II) was almost equal to that in TA98 with activation,
but only when ultrasonic agitation and acetone were used to extract the particles.
Effect of reflwdng cigarette smoke extracts
Table III contains the mutagenicity data from the experiments conducted to deter-
mine whether or not the refluxing of sidestream cigarette smoke extracts affects
their mutagenic activity. The mutagenic activity in TA98 with metabolic activation
was not significantly affected by refluxing the extracts for up to 24 h. An increase
in activity in TA98 without metabolic activation was seen after refluxing. Significant
reductions in activity, however, were observed in TAIOO without metabolic activa-
tion after the extracts were refluxed.
Effect of storage time and conditions
Sidestream cigarette smoke panicles and extracts were stored for different lengths
of time as either particles or as extracts in DMSO. Following storage, the samples
were tested for mutagenic activity in the Ames test. The results of this experiment
are contained in Table IV.
Substantial reductions in activity were observed in both TA98 with activation and
in TAIOO without activation. The largest reductions in activity were seen in TAIOO
without metabolic activation after the extracts had been frozen for 180 days in
DMSO. The reduction in activity in TA98 was much less than the reduction in
TAIOO especially when the samples were stored as particles rather than as extracts
in DMSO.
DISCUSSION
As early as 1974, Kier et al. [12] demonstrated the mutagenicity of mainstream
-------
285
cigarette smoke condensate in Salmonella typhimurium strain TA1538 with a
metabolic activation system prepared from poiychlorinated biphenyl-induced rat
liver tissue. Since 1974, numerous investigators have reported mutagenic activity for
mainstream cigarette smoke and cigarette smoke condensates in a variety of
mutagenicity test systems [14-18]. More recently, extracts of sidestream cigarette
smoke particles have also been shown to be mutagenic in the Ames test [7,20]. As
in previous studies [7], the highest mutagenic response in this study (15 000-20 000,
revertants per cigarette) was observed in TA98 with metabolic activation (Table III).
Method of extraction
Although Soxhlet extraction is the most widely used procedure for the removal
of organic compounds from airborne paniculate samples, it has potential for ar-
tifact formation and is a lengthy procedure usually requiring a minimum of 24-36
h. During this extraction process there is opportunity for sample loss due to
volatilization. In addition, since this method requires a substantial amount of
glassware and laboratory hood space, it is often not practical when numerous
samples are extracted in a short period of time. Soxhlet extraction also has been
reported to result in the decomposition of some known mutagenic compounds
[21-23].
Ultrasonic agitation, an alternate method for the extraction of airborne particles,
has been reported to yield recoveries which are comparable to Soxhlet extraction
[23-26). Although this method requires less time, glassware, and supplies, the most
important advantage of sonication may be the reduction in the opportunity for sam-
ple loss. A number of extraction solvents have been used in both methods including
benzene, methanol, acetone, dichloromethane, acetonitrile, ethanol, and mixtures
of 2 or more of these solvents [27].
A number of studies have compared these extraction methods for ambient air par-
ticles [24,25,28-31]. In a recent investigation, Rives [32] compared the efficiencies
of sonication and Soxhlet extraction of woodsmoke particles. Using both mutagenic
activity and extractable mass as the criteria for his evaluation, he concluded that
either method can be used for the extraction of mutagenic compounds from
woodsmoke. Despite individual preferences for one method over another, the recent
literature recommends both techniques equally [26,33].
In this study (Tables I and II), the mutagenicity data indicate that the ultrasonic
extraction method is as effective or more effective than the Soxhlet procedure in ex-
tracting those compounds from sidestream cigarette smoke particles which are
mutagenic in the Ames test. The use of sonication also resulted in more consistent
data than that obtained when Soxhlet extraction was used. These findings are in
agreement with Sawicki et al. [28] who reported that ultrasonic agitation was a more
reproducible extraction method than Soxhlet extraction.
We conclude that both Soxhlet extraction and ultrasonic agitation are effective
in the extraction of mutagens from sidestream cigarette smoke particles. Sonication
-------
286
requires less time and equipment and results in less opportunity for loss of sample
during extraction. Thus, sonication would be the favored method in cases where
large numbers of samples require extraction in a short period of time.
The choice of a solvent for use in any extraction procedure is usually influenced
by the class or classes of compounds for which extraction is desired. Numerous
solvents have been used by different investigators for the extraction of organic com-
pounds from air paniculate samples for mutagenicity testing in the Ames test
[27,34]. Acetone has been widely used by several investigators for air particle extrac-
tion; however, some do not recommend its use because of its reactivity with other
chemicals and its unavailability in pure form [35]. Other solvents and combinations
of solvents, such as cyclohexane [36], benzene/dichloromethane mixture [37],
cyclohexane, dichloromethane and acetone serially [38], methanol [39], and benzene
[40] have also been used for the extraction of rautagenic material from air particles.
Jungers and Lewtas [30] concluded that the preferred solvent for those studies in
which the non-mutagenic and sometimes bactericidal mass is to be minimized is
dichloromethane. Krishna et al. [25] compared 7 solvents or combinations of
solvents and reported that sequential extraction with acetone followed by
dichloromethane gave a higher response than acetone alone or acetone in combina-
tion with dichloromethane. Of the solvents tested by Krishna et al. [25], cyclohexane
resulted in the lowest mutagenic response. Studies published by Jungers et al. [29]
and Talcott and Wei [41] report similar findings. In a recently published study,
acetone was the extraction solvent chosen by Alfheim and Ramdahl [20] in their
study which investigated the mutagenicity of wood and sidestream cigarette smoke.
In a comparison of 4 solvent systems, Talcott and Wei [41] using both Soxhlet and
ultrasonic reported that acetone extracted more mutagens than benzene,
chloroform, and methanol. Using both Soxhlet and ultrasonic extraction methods,
Goto et al. [42] concluded that benzene, ethanol or methanol were the solvents of
choice for the detection of ambient air paniculate mutagenicity in the Ames test.
These studies (Tables I and II) suggest that acetone was the solvent-of-choice
among those used for removal of organic compounds from sidestream cigarette par-
ticles. Use of methanol or dichloromethane as an extraction solvent greatly reduced
the ability to detect mutagens with TA100.
Effect of refluxing
Goto et al. [42] cited the results of reflux experiments in their choice of extraction
solvent. Increases or decreases in mutagenic activity were seen when several
mutagenic pure compounds were refluxed in various extraction solvents. Several of
the solvents evaluated by these investigators resulted in significant changes in the
mutagenic activity of pure compounds when these compounds were refluxed in the
extraction solvent, including dichloromethane and acetone. The results obtained in
this study indicate that refluxing extracts of sidestream cigarette smoke panicles
does result in an apparent decrease in mutagenic activity in TA98 with metabolic
-------
287
activation (Table III). Statistical analysis, however, does not demonstrate that this
difference is significant. Activity in TA100 without metabolic activation, however,
was significantly reduced after refluxing (Table III). These results suggest that the
compounds responsible for the mutagenicity in TA100 may be compounds which
are volatile or labile and, therefore, are lost during the refluxing process. Another
possibility is that a chemical reaction occurs during the refluxing process which
results in the transformation of the mutagenic compounds into compounds which
are not mutagenic in the Ames test.
While differences are seen in the mutagenicity of replicate samples of sidestream
cigarette smoke particles extracted with the 3 solvents compared in this study, these
differences are generally not significant. The one significant difference is in the
detection of direct-acting mutagens from sidestream cigarette smoke particles when
acetone was used (Table II). Extraction using acetone resulted in higher levels of ac-
tivity than seen with either dichloromethane or methanol.
Effect of storage
It has been reported that cigarette smoke condensate stored at refrigerator
temperatures showed no loss of activity after 40 days; however, a 50% drop in ac-
tivity in TA1538 was reported after the same length of storage time at room
temperature [17]. The effect of storage of cigarette smoke particles or extracts on
the mutagenicity of these substances was investigated in this study. Substantial
reductions in the mutagenic activity of sidestream cigarette smoke particle extracts
were observed after storage as either particles or as extracts in DMSO (Table IV).
After storage for 60 days at -70°C in DMSO, the activity in TA98 with metabolic
activation was reduced to 30% of the 'fresh* sample (less than 6 days storage).
Greater reductions in activity were observed in TA100 without metabolic activation,
as the samples contained only 12% of the activity of the 'fresh' samples. Storage
as particles for 180 days at -70°C resulted in a 33% reduction of activity in TA98
with metabolic activation compared to the 'fresh* sample. These findings suggest
that the effect of sample storage should be considered when attempting to evaluate
the mutagenic activity of tobacco smoke particles. In general, it would seem prudent
to conduct bioassay of the samples as quickly as possible after collection.
CONCLUSIONS
This study clearly demonstrates the mutagenicity of sidestream cigarette smoke
particle extracts in the Ames Salmonella mutagenicity assay. The level of mutagenic
activity of sidestream cigarette smoke particles was estimated from the slope of a
dose-response curve calculated using the method of Stead et al. [11]. The number
of revertants in the Ames test was estimated to be IS 000-20 000 per cigarette in
TA98 with metabolic activation and 12 000-17 000 per cigarette in TA100 without
metabolic activation. After refluxing acetone extracts of sidestream cigarette smoke
-------
288
particles, the mutagenic activity in TA100 without metabolic activation was
significantly reduced compared to extracts which were not refluxed.
Sidestream cigarette smoke particles were collected under experimental conditions
by burning cigarettes in a chamber and collecting the smoke particles on glass fiber
filters. The filters were extracted using either Soxhlet extraction or ultrasonic agita-
tion and one of 3 solvents (dichloromethane, methanol, or acetone). When the com-
binations of extraction method and solvent were compared for their efficiency in
detecting mutagenic activity in the Ames test, the differences in mutagenic activity
were slight; however, ultrasonic agitation produced results which were more inter-
nally consistent than results with Soxhlet extraction. Of the 3 solvents which were
compared, acetone yielded statistically higher levels of activity than the other 2
solvents (dichloromethane and methanol) in TA100 without exogenous activation.
Differences in mutagenic activity in TA98 with metabolic activation among the 3
solvents, however, were small.
The effect of storing sidestream cigarette smoke particles and particle extracts on
the mutagenic activity was also investigated. Activity in both TA98 with metabolic
activation and TA100 without metabolic activation was found to be reduced after
storage either as particles or as extracts in DMSO. Storage in DMSO at -70°C for
60 days resulted in a reduction in mutagenic activity of 70% in TA98 with activa-
tion. Activity in TA100 without activation was reduced by 88% after the same
storage. When stored as particles, the reduction in activity in TA98 was less (33%);
however, the decrease in activity in TA100 was over 90%.
ACKNOWLEDGEMENTS
We want to thank the faculty of the School of Public Health, University of North
Carolina, Chapel Hill, NC and the Health Effects Research Laboratory (HERL),
U.S. Environmental Protection Agency, Research Triangle Park, NC for providing
the opportunity for this research. The research described in this article has been
reviewed by the HERL, U.S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily reflect the views
and policies of the Agency nor does the mention of trade names or commercial pro-
ducts constitute endorsement or recommendation for use.
REFERENCES
1 T.M. Oalhamn, B. Edfors and R. Rylander, Retention of cigarette smoke components in human
lungs. Arch. Environ. Health. 17 (1968) 746-751.
2 G. Holzer, J. Oro and W. Bertsch, Gas chromatographic-mass spectrometric evaluation of exhaled
tobacco smoke, J. Chromatogr., 126 (1976) 771-775.
3 U.R. Hoegg, Cigarette smoke in closed spaces. Environ. Health Perspect., 1 (1972) 117-121.
4 G. Neurath and H. Horstmann, Einfluss des Feuchtigkeitsgehaltes Glutzonentemperature, Beitr.
Tabak Forsch., 2 (1973) 93-97.
-------
289
5 U.S. Department of Health, Education and Welfare, The Health Consequences of Smoking. Report
of the Surgeon General, OHEW Publ. No. (HSM) 72-7516, Washington. DC. 1972.
6 W.S. Rickert, J.C. Robinson and N. Collishaw, Parameters affecting (he mutagenic activity of
cigarette smokers' urine, J. Appl. Toxicol., 2 (1984) 241-246.
7 G. Lofroth, L. Nilsson and I. Alfheim, Passive smoking and urban air pollution:
Salmonella/microtome mutagenicity assay of simultaneously collected indoor and outdoor par-
ticulate matter, in M.O. Waters, S.S. Sandhu, J.L. Huisingh. L. Claxton and S. Nesnow (Eds.),
Short-term Bioassays in the Fractionation and Analysis of Complex Environmental Mixtures, Vol.
3, Plenum Press, New York. 1983, pp. 515-526.
8 C.R. Green, Neutral (oxygenated) compounds in cigarette smoke and their possible precursors, Re-
cent Adv. Tobacco Sci., 3 (1977) 94-98.
9 National Research Council. Committee on Passive Smoking, Environmental Tobacco Smoke:
Measuring Exposures and Assessing Health Effects, National Academy Press, Washington, DC,
1986, p. 337.
10 B.N. Ames, J. McCann and E. Yamasaki, Methods for detecting carcinogens and mutagens with the
5a/
-------
290
25 G. Krishna. J. Natz. W.Z. Whong and T. Ong, Mutagenicity studies of ambient airborne particles.
I. Comparison of solvent systems, Muiat. Res.. 124 (1983) 113-120.
26 C. Krishna, J. Natz, W.Z. Whong and T. Ong, Mutagenicity studies of ambient airborne panicles.
II. Comparison of extraction methods, Mutat. Res.. 124 (1983) 121-128.
27 U.S. Environmental Protection Agency, IERL-RTP Procedure Manual: Level 1 Environmental
Assessment. EPA Publ. No. 600/7-78-201, U.S. Environmental Protection Agency, Research
Triangle Park, NC, 1978..
28 E. Sawicki. T. Belsky, R.A. Froede, D.L. Hyde. J.L. Monkman, R.A. Rasmussen, L.A. Rilppenon
and L.O. White, Total paniculate aromatic hydrocarbons in air: ultrasonic extraction method.
Health Lab. Sci., 12 (1975) 407-414.
29 R. Jungers, R. Burton, L. Claxton and J.H. Lewtas, Evaluation of collection and extraction methods
for mutagenesis studies on ambient air paniculate, in M.D. Waters, S.S. Sandhu. J.L. Lewtas, L.
Claxton and S. Neswnow (Eds.), Short-term Bioassay.in the Fractionation and Analysis of Complex
Environmental Mixtures. Vol. 2. Plenum Press, New York. 1982. pp. 45-65.
30 R. Jungers and J. Lewtas. Airborne panicle collection and extraction methods applicable to genetic
bioassays. in R.R. Tice, D.L. Costa and K.M. Schaich (Eds.), Genotoxic Effects of Airborne Agents.
Plenum Press. New York. 1982, pp. 35-47.
31 J.S. Mumford and J. Lewtas, Sample collection and preparation factors which affect mutagenicity
and cytotoxicity of coal fly ash, in M.D. Waters, S.S. Sandhu, J.L. Huisingh, L. Claxton and S.
Nesnow (Eds.), Short-term Bioassays in the Fractionation and Analysis of Complex Environmental
Mixtures. Vol. 3. Plenum Press, New York, 1982, pp. 39-58.
32 G.D. Rives,'Genotoxic transformation of woodstove emissions in simulated atmospheres. Master's
Thesis. University of Nonh Carolina. Chapel Hill, NC, 1983.
33 1. Alfheim and A. Lindskog. A comparison between different high volume sampling systems for col-
lecting ambient airborne panicles for mutagenicity testing and for analysis of organic compounds,
Sci. Total Environ., 34 (1984) 203-222.
34 K. Duke. M. David and A. Dennis. IERL-RTP Procedures Manual, EPA Publ. 600/7-77-143.
Washington, DC, 1977.
35 DJ. Brusick and R.R. Young, IERL-RTP Manual: Level I Environment Assessment. Biological
Tests, EPA Publ. No. 600/8-81-024. U.S. Environmental Protection Agency, Office of Research and
Development, Washington, DC. 1981.
36 1. Alfheim and T. Moller, Mutagenicity of long-range transported atmospheric aerosols, Sci. Total
Environ.. 13 (1979) 257-278.
37 J. Pitts, D. Grosjean, T. Mischke. V. Simon and D. Poole, Mutagenic activity of airborne paniculate
organic pollutants. Toxicot. Leu., 1 (1977) 65-70.
38 J.M. Daisey, 1. Hawrlux, T.U. Kneip and F. Mukai, Mutagenic activity in organic fractions of air-
borne paniculate matter, in T. Novakov (Ed.), Proceedings of Conference on Carbonaceous Par-
ticles in the Atmosphere, Natl. Tech. Inform. Service, U.S. Depanment of Commerce, Washington.
DC. 1979. pp. 27-38.
39 W. Dehnen. N. Pitz and R. Tomingas, The mutagenicity of airborne paniculate pollutants. Cancer
Lett., 4 (1977) 5-12.
40 H. Fukino, S. Mumura, K. Inque and Y. Yamane, Mutagenicity of air panicles, Muiat. Res., 102
(1982) 237-247.
41 R. Talcott and E. Wei. Airborne mutagens bioassayed in Salmonella typhimurium, J. Natl. Cancer
Inst.. 58 (1977) 449-451.
42 S. Goto, A. Kawai, T. Yonekawa. Y. Hisamatsu and H. Matsushita, Effect of extracting solvents
on the mutagenicity of airborne panicles, J. Jpn. Soc. Air Pollut., 16 (1981) 18-25.
-------
Mutanon Research. 202 (1988) 103-110
Elsevier
103
MTR 04643
Public exposure to environmental tobacco smoke
Goran Lofroth1<2, Per Ivar Ling 2 and Eva Agurell 2-3
' Nordic School of Public Health. Box 12133. S-402 42 Gothenburg, * Department of Radiobiology, University of Stockholm.
S-106 91 Stockholm, and J Department of Genetic and Cellular Toxicology, University of Stockholm. S-106 91 Stockholm (Sweden)
(Received 25 November 1987)
(Revision received 28 March 1988)
(Accepted 29 March 1988)
Keywords: Environmental tobacco smoke (ETS); Passive smoking; Personal sampling; Salmonella mutagenicity assay; Smoking
Summary
Airborne particulate matter has been collected by personal samplers in public indoor areas and travel
situations with environmental tobacco smoke pollution. Following extraction, the samples were assayed for
mutagenicity in the presence of S9 with a sensitive microsuspension test using Salmonella TA98. The
mutagenic responses of indoor air from public areas were much higher than those of ambient outdoor air.
Depending on the circumstances, the mutagenic response varied in trains and airplanes but the results
show that physical separation of non-smoking sections from smoking sections is necessary in order to
achieve genuine non-smoking areas. Chemical fractionation and mutagenicity assay of the basic fraction
show that Salmonella mutagenicity of airborne particulate matter might be used as a tobacco smoke-specific
indicator, as the basic fraction of environmental tobacco smoke contains a large part of the mutagenic
activity whereas this is not the case for outdoor ambient airborne particulate matter.
Environmental tobacco smoke (ETS) is a com-
plex mixture of gases and particulate tar matter
comprising numerous compounds. ETS is one of
the most common air pollutants in industrialized
and urban societies as 25-40% of the adult popu-
lation are smokers and much smoking takes place
indoors causing pollution of the air breathed by
everyone.
Among several types of pollution indices.
Salmonella mutagenicity of airborne particulate
matter has been used to study the contribution
from ETS in office buildings (Lofroth et al., 1983),
restaurants (Husgafvel-Pursiainen et al., 1986) and
Correspondence: Gdran Ldfroth, Nordic School of Public
Health. Box 12133. S-402 42 Gothenburg (Sweden).
homes (van Houdt et al.. 1984; Alfheim and
Ramdahl. 1984; Lofroth and Lazaridis, 1986:
•Lewtas et al., 1987). In some studies (Lofroth et
al., 1983; Lofroth and Lazaridis. 1986) the origin
of the mutagenic activity was ascertained by frac-
tionation in which a major part of the activity 01
ETS was recovered in the basic fraction.
The use of personal samplers for the collection
of particulate matter coupled with a more sensi-
tive mutagenicity test, a Salmonella microsuspen-
sion assay, was recently explored and found feasi-
ble (Ling et al., 1987). These studies have now
been extended with measurements of the muta
genie response of airborne particulate matter col-
lected during some typical situations outside honu
and work where involuntary exposure to ETS car
occur. The chemical behavior of ETS as comparec
0027-5107/88/$03.50 © 1988 Elsevier Science Publishers B.V. (Biomedical Division)
-------
104
to outdoor ambient paniculate matter with respect
to the contribution from the basic fraction has
also been further studied.
Materials and methods
Sampling
Indoor airborne paniculate matter was col-
lected on glass fiber filters with 2 battery-operated
personal samplers (Casella AFC 123, Casella
London Inc., Great Britain) using a flow rate of 2
1/min for each sampler and 25-mm sampling
heads. The samplers were carried in a small bag
keeping the sampling heads near the breathing
zone, i.e. no more than about 25 cm below the
mouth/nose region. Sampling time varied from
less than 1 h to more than 6 h depending on
location. The filters were stored in Al-foil at
- 20 ° C within 2 h of the end of the sampling.
Parallel sampling of outdoor airborne par-
ticulate matter was performed with a battery-oper-
ated portable sampler (Caseila HFS 800) using a
flow rate of 10 1/min and a 35-mm sampling
head. The sampling was made from a car parked
at the nearby parking lot with the air intake of the
sampler placed outside the car.
Experimental sidestream smoke was collected
as described previously (Ling et al.. 1987) from
machine-smoked cigarettes in a small 0.3-m3 hood.
A high-volume sample of glass fiber filter-col-
lected urban airborne paniculate matter was also
used for comparison.
Sample preparation
The filters were extracted within 10 days after
sampling. Extraction was performed with acetone
using a bath-type sonicator. The solution was
filtered through a No. 4 glass filter to remove glass
fiber debris.
Fractionation with respect to polarity into basic
and non-basic fractions was performed by
liquid-liquid extraction with diethyl ether and
sulfuric acid and sodium hydroxide aqueous solu-
tions as described earlier (Lofroth, 1981).
For the microsuspension assay, aliquots of the
extract corresponding to known air volumes were
transferred to 11 mm X 75 mm sterile glass tubes
containing 5 /*! dimethyl sulfoxide (DMSO). The
acetone was evaporated by a stream of nitrogen
gas with the tubes inserted in a heating block and
the evaporating solution was kept at < 40 ° C. The
tubes were finally stoppered using sterile silicon
stoppers and stored at - 20 ° C until they were
used for the bioassay.
For the plate incorporation assay, the extract
was evaporated to a small volume and then di-
luted with DMSO. The samples were stored at
-20°C.
Bioassay
The microsuspension technique (Kado et al..
1983) was used with minor modifications for all
environmental samples in the present study.
Salmonella TA98 grown for about 13 h with rapid
shaking was centrifuged and then resuspended in
1/10 of the original volume in Vogel and Bonner
(1956) medium E (instead of phosphate-buffered
saline) giving approximately 10'° cells/ml of which
0.1 ml was added to the sample tubes.
The S9 was obtained from livers of Aroclor
1254-induced male Sprague-Dawley rats. Its pro-
tein content was 33 rug/mi as determined by the
method of Lowry et al. (1951). The S9 mix was
prepared as described by Ames et al. (1975) but
with Vogel and Bonner medium E instead of 0.2
M phosphate buffer for the microsuspension as-
say. Each sample tube received 0.1 ml S9 mix
containing 10% S9 or buffer without S9, NADP
and glucose 6-phosphate.
The sample tubes, covered with sterile caps.
were then immediately incubated for 90 min at
37 °C with about 175 rpm shaking after which 2
ml top agar containing histidine and biotin were
added. Following Vortex mixing, the samples were
poured on minimal glucose agar plates. The plates
were incubated for 48 h at 37 ° C. Revertant col-
onies were then counted manually.
The plate incorporation assay was performed as
described by Maron and Ames (1983). The S9 mix
contained 4 or 10% S9 and 0.5 ml was added to
each plate. The S9 amounts employed were those
which are routinely used for other outdoor am-
bient and indoor tobacco smoke samples.
Depending on the amount of sample available,
each sample was tested repeatedly on several occa-
sions with 1 plate per dose and 3-5 plates for the
spontaneous control. Some samples were only
available for 1 or 2 independent tests but larger
-------
105
samples were always tested 3 or 4 times. Each test
comprised positive control compounds. Blank filter
samples have been assayed and not found to give
any detectable mutagenic activity.
The dose response was evaluated with least
square linear regression using all plate counts in
the linear or approximately linear part of the
dose-response curve.
Results
Environmental samples
During this study a number of samples of
airborne paniculate matter have been collected at
locations or travel situations which people may
experience in their daily life.
Table 1 relates the results from 2. visits to a
shopping center in the northern part of the
Stockholm area and 2 visits to the Stockholm
Central (railway) station. The shopping center is
about 12 m high and has a central 25 m X 45 m
indoor plaza with four 45-65-m-long extending
alleys. The Central station main hall has an indoor
area of 28 m x 119 m and a height of about 15 m.
Despite these spacious designs, the indoor pollu-
tion, measured as mutagenic response of par-
ticulate matter, is high and is higher than the
response of simultaneously collected ambient out-
door air.
Samples collected during train travel (Table 2}
were obtained in the common type of passenger
cars containing 2 compartments separated by a
TABLE 1
MUTAGENIC ACTIVITY AND RESPONSE IN TA98 + S9 OF AIRBORNE PARTICULATE MATTER COLLECTED IN
INDOOR PUBLIC AREAS AND SIMULTANEOUSLY AT NEARBY OUTDOOR LOCATIONS
Sample
location, date
and duration
Shopping center
861222
175 min
Benzofajpyrene "
Shopping center
861230
175 min
Benzo[a]pyrene
Central station
870116
235 min
Benzo(a)pyrene
Central station
870130
235 min
Benzo(a]pyrene
Indoors
Dose
(1 air/plate)
0
25
50
100
150
0.5 fig
0
25
50
100
150
0.5 fig
0
25
50
100
200
0.5 fig
0
25
50
100
200
0.5 fig
Counts
(rev./plate)
49
62
112
177
231
301
51
52
94
165
192
345
53
101
163
330
552
445
62
107
178
335
449
477
Outdoors
Response Dose
(rev./m3) (lair/plate)
0
75
1200 150
300
0
75
1000 150
300
0
200
2500 400
600
0
200
2200 400
600
Counts
(rev./plate)
49
46
58
55
51
58
82
85
53
71
149
236
62
59
75
77
Response
(rev./mj)
<50
140
260
<50
* Concurrent positive control.
-------
106
sliding door; one for smoking with 20 seats and
one for non-smoking with 60 seats. The high re-
sponses in the smoking compartment are not un-
expected and a dependence on the number of
cigarettes smoked is evident. Most samples col-
lected in the non-smoking compartment have a
relatively low response but there is one exception
in the sample collected 870215. This train was
congested and there was much passenger move-
ment with frequent openings of the door; the
smell of tobacco smoke in the non-smoking com-
partment was apparent.
Samples have been collected during 2 short air
nights and 2 transatlantic flights (Table 3). The
short flights gave only samples sufficient for 1
assay with 1 plate per dose but even with this
limitation a dose-dependent increase can be ob-
served resulting in relatively high responses. There
is a substantial difference between the two trans-
atlantic flights which may be explained by the fact
that the sample of 861029 was collected in a
non-smoking seat only 2 rows apart from smoking
seats whereas the sample of 870516 was collected
in an entire non-smoking section separated from
smokers by the stewardess' areas.
Experimental tobacco smoke
The experimental cigarette sidestream smoke
(Table 4) was generated from a common Swedish
filter brand which in this experiment gave 29.6 mg
tar particles/cigarette. This sample was used for
comparing the response in the regular assay and
TABLE 2
MUTAGENIC ACTIVITY AND RESPONSE IN TA98 + S9 OF AIRBORNE PARTICULATE MATTER COLLECTED DUR-
ING TRAIN TRAVEL IN SMOKING AND NON-SMOKING COMPARTMENTS
Sample, date and duration
Smoking
860921
250 min
5 cig. smoked
Benzo[a]pyrene J
Smoking
870208
275 min
25 cig. smoked
Benzofajpyrene
Non-smoking
861002
250 min
Benzo[a]pyrene
Non-smoking
870121
270 min
Benzo[a]pyrene
Non-smoking b
870215
240 min
Benzo( a jpyrene
Dose
(1 air/plate)
0
71
143
286
0.5 fig
0
75
150
300
0.5 jug
0
71
143
286
0.5 Mg
0
75
150
300
0.5 Mg
0
50
100
167
0.5 Mg
Counts
(rev./plate)
48
78
126
226
478
62
372
705
929
477
48
49
60
92
478
53
61
75
83
444
53
79
117
147
427
Response
(rev./m^)
600
3500
100
100
600
" Concurrent positive control.
h 2 additional non-smoking samples collected 870210 and 870219 gave responses of 100 rev./m3.
-------
107
the presently employed microsuspension assay and
for studies of the response of the basic and non-
basic fractions of the smoke. For comparison, a
sample of urban airborne paniculate matter was
also investigated and fractionated simultaneously.
The total response of the sidestream smoke
obtained in the present study in the microsuspen-
sion assay is higher than previously reported val-
ues (Ling et al., 1987). This is mainly due to the
use of the modified assay technique. Two other
experimental cigarette sidestream samples were
analyzed in the course of this study and the assays
gave 220000 and 290000 revertants/cigarette with
TA98 in the presence of S9.
The much higher response of the microsuspen-
sion assay (Table 4), known from previous inves-
tigations (Kado et al., 1983, 1986; Ling et al..
1987), is evident for both the sidestream smoke
and the urban paniculate sample.
In the plate incorporation assay, the response
in TA98 + S9 of the basic fraction of sidestream
smoke is about 67% of the total response. This is
in agreement with earlier studies in which 67 and
70% were obtained (Lofroth et al.. 1983; Lofroth
and Lazaridis, 1986). In the microsuspension as-
say, the response of the basic fraction is about
45% of the total response in TA98 + S9. This
lesser relative response of the basic fraction in the
microsuspension assay is further supported by the
results obtained with 2 environmental samples.
The first sample was collected in an appartment
during a party and gave a response of 3200 rev./m3
(Ling et al.. 1987) and showed after fractionation
that 45% was present in the basic fraction (data
not shown). The second sample, collected in the
non-smoking section (corner) of a coffee shop in
downtown Stockholm (data not shown), had about
36% of the total response of 2200 rev./m3 in the
basic fraction. A contribution of about 400 rev./mj
from ambient outdoor paniculate matter would in
this case explain the fractionation result.
Urban airborne paniculate matter has very lit-
TABLE 3
MUTAGENIC ACTIVITY AND RESPONSE IN TA98 + S9 OF AIRBORNE ^ARTICULATE MATTER COLLECTED DUR-
ING AIR TRAVEL IN NON-SMOKING SECTIONS
Sample, date, flight and duration
Gothenburg-Oslo
860919
SK886
38 mm
Oslo-Stockholm
860919
SK708
38 nun
Benzofajpyrene "
New York-Oslo
861029
SK902
360 min
Benzo[ a ]pyrene
New York-Stockholm
870516
SK904
390 min
Benzofojpyrene
Dose
(1 air/plate)
0
25
50
75
0
25
50
75
0.5 ,ig
0
24
48
96
192
0.5 Mg
0
50
100
200
0.5 Mg
Counts
(rev./plate)
52
66
78
91
52
70 .
96
143
467
46
65
82
117
193
403
68
87
96
111
634
Response
(rev./m3)
500
1000
800
200
" Concurrent positive control.
-------
108
TABLE 4
MUTAGENIC RESPONSE IN TA98 OF UNFRACTIONATED AND FRACTIONATED EXTRACTS OF CIGARETTE
SIDESTREAM PARTICULATE TAR AND AN URBAN PARTICULATE MATTER SAMPLE IN THE REGULAR PLATE
INCORPORATION ASSAY AND THE MICROSUSPENSION ASSAY AND THE ACTIVITY OF CONCURRENT POSITIVE
CONTROL COMPOUNDS
Sample
Unit
Plate incorporation
Microsuspension
-S9
•S9-10%
-S9
•59-10%
Sidestream smoke * Rev./cig.
Crude extract
Basic fraction
Non-basic fraction
2700
1000
1500
37000
25000
5400
70000
26000
19000
240000
110000
75000
Urban paniculate* b
Crude fraction
Basic fraction
Non-basic fraction
Positive controls
5.0 /tg quercetin
25.0 /ig quercetin
0.5 /ig benzoffljpyrene
2.5 fig benzofajpyrene
Spontaneous control
Rev./mg
Rev./plate
Rev./plate
450
20
460
318
26
520
30
430
385
33
207
36
3000
100
2700
257
26
4200
120
2200
267
36
' The responses have been evaluated with 6 doses in the range 0.001-0.1 cig./plate in the plate incorporation and in the range
0.00025-0.024 cig./plate in the microsuspension assay.
" The responses have been evaluated with 6 doses in the range 25-1600 jug/plate for both assays.
f -, not tested.
tie activity in the basic fraction (Table 4) which is
in agreement with earlier results (LSfroth, 1981;
Lofroth et ah, 1983).
Discussion
Public indoor locations
This exploratory study shows that typical pub-
lic indoor locations, a shopping center plaza and a
railway station waiting room, are much more pol-
luted than the ambient outdoor air by mutagenic
compounds present in airborne paniculate matter.
Although the sampling and analysis cannot prove
the origin of the increased mutagenic activity,
smoking is the only conceivable source. Concom-
itant with an increase of the mutagenic response
there is consequently also an increase of other
pollutants, such as nitrogen oxides and volatile
hydrocarbons which are not detected by the muta-
genicity test. The level of the indoor mutagenic
response of 1000-2500 revertants/m3 (Table 1)
can be compared with an average response of 45
revertants/m3 (range 9-162) for seventy-six 24-h
samples collected at street level at various lo-
cations in Gothenburg (Sweden) and analyzed with
the microsuspension assay with TA98 + S9
(Lofroth et ah. unpublished results). The ambient
outdoor response of < 50-260 revertants/mj mea-
sured simultaneously in the present study (Table
1) is of a reasonable magnitude considering the
fact that these samples were collected during a few
hours in the afternoon when the traffic is of more
than the average 24-h intensity.
Train travel
The mutagenic responses of the air of smoking
train compartments (Table 2) are of an expected
magnitude. The type of compartment sampled has
a volume of about 40 m3 and assuming an effi-
cient air mixing, the concentration of 600 and
3500 revertants/m3 following smoking of 5 and 25
cigarettes, respectively, during 4-4.5 h may be
obtained with a combined ventilation and surface
removal rate (Repace and Lowrey, 1980) corre-
sponding to about 9-12 air changes/h.
-------
109
The result from the non-smoking compart-
ments (Table 3) shows that smoke from the smok-
ing compartment can penetrate into the non-
smoking section. Although most samples only gave
a small response, an activity of 100 revertants/nr1
must be judged to be above the background as the
electrically powered train mostly travels through
rural areas. Separation of non-smoking and smok-
ing compartments in a train is most easily done by
having an entire smoking car (or several) at the
rear end of the train.
Air travel
Tobacco smoke in airplanes has recently been
studied with nicotine analysis by Oldaker and
Conrad (1987) who found that the average nico-
tine concentration in the non-smoking sections
was about half of that in the smoking sections; 5.5
vs. 9.2 /ig/m3. The concentration in the non-
smoking sections correponds to about 0.0013
cigarettes/m3 using the sidestream emission of
nicotine (4.1 mg/cigarette) given by Rickert et al.
(1984). The mutagenic response of 200-1000 re-
vertants/m3 (Table 3) obtained in the present
study corresponds to 0.001-0.005 cigarettes/m3 of
which the higher value best relates to the situa-
tions investigated by Oldaker and Conrad, i.e.
non-smoking seats near smoking sections.
The results of this study and of the investiga-
tion by Oldaker and Conrad (1987) indicate that if
smoking is permitted on airplanes, the smoking
section ought to be physically separated from the
non-smoking sections, e.g. by stewardess' areas.
Chemical fractionation
Using the plate incorporation assay, previous
fractionation studies have shown that mutagens in
tobacco tar particulates to a large extent are basic
compounds responsible for more than 65% of the
response. The relative activity of the basic fraction
in the microsuspension assay is smaller with
slightly less than 50% of the total response being
recovered in this fraction (Table 4). This is, how-
ever, still significantly more than the correspond-
ing relative response of compounds in ambient
paniculate matter with a very small contribution
from the basic fraction. There are no published
reports indicating that some commonly occurring
processes generate airborne paniculate matter with
a high portion of the mutagenic activity in the
basic fraction. A conceivable source of basic
mutagenic compounds is cooking but it has only
given a weak correlation to the total indoor muta-
genic activity (van Houdt et al., 1984: Lewtas et
al., 1987). The sensitive microsuspension assay
might thus be used as a tobacco-specific analysis
if part of an air paniculate sample is subjected to
fractionation, as has been explored in this study
with analysis of a sample from a restaurant and a
sample from an apartment (see Results). Such
differential analyses are deemed possible for mod-
erately tobacco smoke-polluted air with sample
sizes 2-3 times larger than those used in the
present study of public indoor locations.
Mutagenic activity and other tobacco smoke indica-
tors
Nicotine has so far been the only tobacco-
specific air pollutant. Its value as an indicator
may, however, be limited as nicotine may not be a
health issue and as it may be prone to rapid
adsorption to surface materials causing an under-
estimate of the air pollution of other smoke com-
ponents. Among other compounds emitted in the
sidestream. several unsaturated hydrocarbons have
high emission factors (L6froth et al.. 1987. and
unpublished data). The emission of isoprene is
about 2-3 mg/cigarette and this alkadiene may be
utilized as a semi-specific tobacco smoke indicator
although it is present at low background con-
centrations originating from natural sources
(Gelmont et al., 1981; Lamb et al.. 1986). An
advantage with isoprene. which it shares with
mutagens in the tar particulates. is that it is a
potential mutagen and carcinogen following mam-
malian metabolism (Longo et al.. 1985). Ulti-
mately, a combination of Salmonella mutagenicity
of particulates. isoprene and nicotine may be used
for a better estimate of environmental tobacco
smoke.
Acknowledgement
This study was partly made feasible by the
Cooperative Agreement CR812935-01 between the
U.S. Environmental Protection Agency and the
Nordic School of Public Health covering a DVS
program for GL.
-------
110
References
Alfheim. I., and T. Ramdahl (1984) Contribution of wood
combustion to indoor air pollution as measured by muta-
genicity in Salmonella and •polycyclic aromatic hydro-
carbon concentration. Environ. Mutagen., 6. 121-130.
Ames. B.N.. J. McCann and E Yamasaki (1915) Methods for
detecting carcinogens and mutagens with the SalmoneJ-
la/mammalian-microsome mutagenicity test. Mutation
Res.. 31. 347-364.
Gelmont. D.. R.A. Stein and J.F. Mead (1981) Isoprene —
The main hydrocarbon in human breath. Biochem. Bio-
phys. Res. Conunun., 99. 1456-1460.
Husgafvel-Pursiainen. K., M. Sorsa, M. Mailer and C. Be-
nestad (1986) Genotoxicity and polynuclear aromatic hy-
drocarbon analysis of environmental tobacco smoke sam-
ples from restaurants, Mutagenesis. 1. 287-292.
Kado. N.Y.. D. Langley and E Eisenstadt (1983) A simple
modification of the Salmonella liquid-incubation assay. In-
creased sensitivity for detecting mutagens in human urine.
Mutation Res.. 121. 25-32.
ICado. N.Y.. G.N. Guirguis. C.P. Flessel. R.C. Chan. K.I.
Chang and JJ. Wesolowski (1986) Mutagenicity of fine
( < 2.5 fim) airborne panicles: Diurnal variation in com-
munity air determined by a Salmonella micro preincuba-
tion (microsuspension) procedure. Environ. Mutagen.. 8,
53-66.
Lamb. B.. H. Westberg and G. Allwine (1986) Isoprene emis-
sion fluxes determined by an atmospheric tracer technique.
Atmos. Environ.. 20, 1-8.
Lewtas. J.. S. Goto. K. Williams. J.C. Chuang, B.A. Petersen
and N.K. Wilson (1987) The mutagenicity of indoor air
particles in a residential pilot field study: Application and
evaluation of new methodologies. Atmos. Environ.. 21.
443-449.
Ling, P.I.. G. Ldfroth and J. Lewtas (1987) Mutagenic de-
termination of passive smoking, Toxicol. Lett.. 35.147-151.
Lflfroth. G. (1981) Salmonella/microsome mutagenicity assays
of exhaust from diesel and gasoline powered motor vehicles.
Environ. Int.. 5, 255-261.
Ldfroth. G.. and G. Lazaridis (1986) Environmental tobacco
smoke (ETS): Comparative characterization by mutagenic-
ity assays of sidestream and mainstream cigarette smoke.
Environ. Mutagen.. 8, 693-704.
Ldfroth. G.. L. Nilsson and I. Alfheim (1983) Passive smoking
and urban air pollution: Salmonella/microsome mutagen-
icity assay of simultaneously collected indoor and outdoor
paniculate matter, in: M.O. Waters. S.S. Sandhu. J. Lewtas.
L. Claxton. N. Chemoff and S. Nesnow (Eds.). Short-Term
Bioassays in the Analysis of Complex Environmental Mix-
tures III. Plenum. New York. pp. 515-525.
Ldfroth. G.. R. Bunon. G. Goldstein. L. Forehand. K.. Ham-
mond. J. Mumford. R. Seila and J. Lewtas (1987) Geno-
toxic emission factors for sidestream cigarette smoke com-
ponents (Abstr.). Environ. Mutagen.. 9. Suppl. S. 61.
Longo, V.. L. Citti and P.G. Gervasi (1985) Hepatic micro-
somal metabolism of isoprene in various rodents. Toxicol.
Lett.. 29. 33-37.
Lowry, O.H.. N.J. Rosebrough. A.L. Fair and R.J. Randall
(1951) Protein measurement with the Folin phenol reagent.
J. Biol. Chem.. 193. 265-275.
Maron. D.M.. and B.N. Ames (1983) Revised methods for the
Salmonella mutagenicity test. Mutation Res.. 113. 173-215.
Oldaker III. G.B.. and F.C. Conrad Jr. (1987) Estimation of
effect of environmental tobacco smoke on air quality within
passenger cabins of commercial aircraft. Environ. Sci.
Technol.. 21. 994-999.
Repace. J.L.. and A.H. Lowrey (1980) Indoor air pollution.
tobacco smoke, and public health. Science. 208. 464-472.
Rickert. W.S.. J.C. Robinson and N. Collishaw (1984) Yields
of tar. nicotine, and carbon monoxide in the sidestream
smoke from 15 brands of Canadian cigarettes. Am. J. Publ.
Health. 74. 228-231.
van Houdt. J.J.. W.M.F. Jongen. G.M. Alink and J.S.M. Boleij
(1984) Mutagenic activity of airborne particles inside and
outside homes. Environ. Mutagen.. 6. 861-869.
Vogel. HJ.. and D.M. Bonner (1956) Acetylomithinase of
Escherichia coli: Partial purification and some properties. J.
Biol. Chem., 218. 97-106.
-------
VARIABILITY OF MEASURES OF EXPOSURE TO
ENVIRONMENTAL TOBACCO SMOKE IN THE HOME
David B. Coultas, Jonathan M. Samet
Department of Medicine and
New Mexico Tumor Registry
University of New Mexico Medical Center
Albuquerque, New Mexico 87131
John F. McCarthy, John D. Spengler
Department of Environmental Science Engineering
Harvard School of Public Health
Boston, Massachusetts 02115
We have assessed the variability of four markers of environmental
tobacco smoke exposure in 10 homes with 20 nonsmoking and 11 smoking
household members. We obtained exposure questionnaires, saliva and urine
for cotinine, and air particle samples for respirable particles and
nicotine on 10 sampling days: every other day over 10 days, and then one
day every other week over 10 weeks. The mean concentrations of respirable
particles in the 10 homes ranged from 32.4 ug/m^ to 76.9 ug/m^. and
concentrations of nicotine ranged from 0.59 ug/m^ to 6.85 ug/nn. A linear
regression model that included indicator variables for the number of
smokers exposed to in the home and the season, and the number of hours of
exposure as a continuous variable explained nine and 6% of the variability
of the respirable particle and the nicotine concentrations, respectively.
The individual mean cotinine levels standardized to urinary creatinine
concentration, ranged from 3.89 ng/mg Cr to 55.77 ng/mg Cr. A linear
regression model that included the number of smokers exposed to in the
home, the season, the age group, and the number of hours of exposure
explained 8% of the variability of the urinary cotinine levels.
We conclude that because of the marked variability of these measures,
multiple measurements are needed to establish a stable profile of exposure
to environmental tobacco smoke in a particular home or individual.
Furthermore, detailed questions to quantitate exposure offered little
additional information beyond whether the subject was exposed or not.
Proceedings.of the APCA Conference
'A •;i.;230 USA on Indoor Air, In Press
Niagra Falls, 1988
-------
Introduction
Passive smoking refers to the involuntary exposure of nonsmokers to
the combination of tobacco combustion products released by the burning
cigarette and smoke components exhaled by the active smoker*»2. The
adverse health effects of passive smoking on children and adults have been
described in numerous epidemiologic investigations^.
Although some health effects of passive smoking have been
convincingly demonstrated, many questions on the health effects of passive
smoking remain unanswered. More precise description of exposure-response
relationships is needed for the adverse effects on children and for lung
cancer and nonmalignant effects on adults. In most epidemiologic studies
on involuntary smoking published to date, exposure has been assessed with
questionnaires. However, other indicators of exposure are available,
including atmospheric and biologic markers, which have received increasing
attention.
We have assessed the variability of four markers of environmental
tobacco smoke exposure in 10 homes with 20 nonsmoking and 11 smoking
household members. The markers of exposure included questionnaires,
respirable particles and nicotine from air samples, and urinary cotinine.
Methods
Sample Selection
Between February and December, 1986, 149 nonsmoking volunteers, 18
years of age and older, were recruited from Albuquerque and surrounding
communities to participate in a study of the accuracy of questionnaire
assessment of exposure to environmental tobacco smoke3. From this sample,
10 subjects volunteered their households for this investigation.
Data Collection
We obtained exposure questionnaires, saliva and urine, and air
particle samples on 10 sampling days: every other day over 10 days, and
then one day every other week over 10 weeks. The questionnaires, and
saliva and urine specimens were obtained at the end of a 24-hour air
monitoring period (described below). From the questionnaires we determined
the reported number of smokers and number of hours that the subjects were
exposed, during the previous 24 hours, to cigarettes, cigars, and pipes at
home, at work or school, in a vehicle, and in other places. Questionnaires
were self-completed by the adults, and by a parent for children 13 years of
age and younger. The saliva and urine specimens were frozen at -20°C until
the cotinine assays were performed.
Cotinine Assay
Cotinine was quantitated by a double antibody radioimmunoassay, as
described by Langone et al. . A specific antiserum produced in rabbits was
supplied by Dr. Helen Van Vunakis, Brandeis University. Urine samples were
diluted 1:4 for the assay. The sensitivity of the assay in our hands was
36 pg/tube or 0.78 ng/ml of urine (4204 pmol/L). Urine creatinine
concentrations were determined by the Jaffe reaction^, and the cotinine
concentrations were standardized to the creatinine concentrations. Assays
were performed without knowledge of questionnaire responses.
Particle Measurements
In the major activity room of each home, Harvard School of Public
Health impactors6 operating at a flow rate of 4 liters/minute were used to
collect respirable particles and gaseous nicotine samples. Through a timed
solenoid switching valve, two impactors used a common, mass flow controlled
-------
pump, and each impactor operated on alternate 15 minute collection cycles.
Respirable particle samples were collected on teflon filters (Membrana,
Inc.) and nicotine was collected on sodium bisulfate treated glass fiber
filters (Millipore Corp.) to minimize its volatilization. After extraction
from the filter, analysis for nicotine was done on a Shimadzu GC7A gas
chromatograph with a flame ionization detector. The nicotine collection
and extraction procedure is a modification of that described by Hammond et
al.7. The recovery of nicotine by this procedure has been shown to be 98%
efficient.
Data Analysis
Variability of questionnaire responses, respirable particle and
nicotine concentrations, and urinary cotinine levels were assessed with
univariate analyses. From the questionnaires the measures of exposure in
the home were the total number of smokers, including cigarettes, cigars,
and pipes, and the total number of hours exposed. During the entire
sampling period there were only four days that any subject reported
exposure to a cigar smoker, the predominant exposure was to cigarette
smoke.
To examine .determinants of the variability in the measurements, we
used multiple linear regression. The dependent variables, respirable
particles, nicotine, and urinary cotinine, were analyzed as continuous
variables. For the independent variables, indicators were defined for
house (HOUSE = 1-10), individual (INDIVIDUAL = 1-20), age group (AGE GROUP
<18 years or .>18 years), season (SEASON = March-April or May-October), and
number of smokers per day (NUMBER = 0 or >!)• The other independent
variable, number of hours (HOURS) exposed per day was continuous.
Data analyses were performed with standard programs of the
Statistical Analysis System8.
Results
The 10 households included 11 cigarette smokers and 20 nonsmokers
aged 1.5 to 74 years (Table I). The types of homes included, eight
unattached, single family houses, one mobile home, and one apartment.
Reports on exposure to tobacco smoke in the home were obtained for
all 10 sampling days from 17 subjects, and for nine days from three
subjects. The reported number of cigarette smokers in the home per day did
not vary widely. The median number (range) of smokers per day was one for
18 subjects (0-10), zero for one subject (0-1), and four for one subject
(2-25). Greater variability was reported for the number of hours exposed
to cigarette smoke in the home (Table I).
Respirable particle and nicotine concentrations were obtained for 99%
of the sampling days (Figures 1 and 2). The mean concentrations of
respirable particles in the 10 homes ranged from 32.4 ug/m3 (SD = 13.1) to
76.9 ug/m3 (SD = 32.9) and concentrations of nicotine ranged from 0.59ug/m3
(SD = 0.69) to 6.85 ug/m3 (SD = 8.21). The degree of variability with
sampling every other day or every other week was similar (data not shown).
For the particle and nicotine measurements we used linear regression
to examine determinants of variability and of level. A model that included
indicator variables for the 10 houses explained the greatest variability
(Table II). Compared to the model with the house variables, other models
that included specific exposure variables explained markedly lower
percentages of the variability of levels. Although not statistically
Pitts bur gn. PA 1 &;•„;. ,.;,•/*.
4^2-232- *&ta
-------
significant, increases in respirable particles were associated with
exposure to one or more cigarette smokers in the home and with the colder
months, March and April (Table II). There was no association with the
number of hours of exposure. Nicotine levels also increased with exposure
to smokers in the home, but were not predicted by the season (Table II).
Urinary cotinine levels were obtained on 187 specimens from the 20
nonsmokers. The individual mean cotinine levels standardized to urinary
creatinine concentration, ranged from 3.89 ng/mg Cr (SD = 6.54) to 55.77
ng/mg Cr (SD = 32.02). The mean levels and variability tended to be
greater in the children compared to the adults (Figures 3 and 4). As with
the atmospheric measures, the variability was comparable with sampling
every other day or every other week (data not shown).
For the urinary cotinine levels, we also examined determinants of
variability and determinants of concentration with linear regression. A
model that included indicator variables for the 20 nonsmoking individuals
explained 47% of the variability in cotinine levels (Table III). Compared
to this model, other models that included specific exposure variables and
age group explained much lower proportions of the variability. Cotinine
levels were significantly (p<0.05) higher among children compared to adults
(Table III). Although not significant, exposure to one or more smokers
resulted in higher cotinine levels compared to no exposure. The number of
hours of reported exposure and the season were not significant predictors
of cotinine level.
Conclusions
In a group of volunteers, from 10 homes, we found that concentrations
of respirable particles and nicotine, and urinary cotinine levels varied
widely within homes and individuals, respectively. Sampling every other
day or every other week did not offer any advantage for establishing a
stable profile of exposure with these markers. A moderate degree of the
variability of these measures was explained by variables representing
houses and individuals, but more specific measures of exposure contributed
little to explaining the variability.
Many factors, which we were unable to measure, may contribute to the
variability of these measures of environmental tobacco smoke in the home.
For the atmospheric measurements, concentrations depend on the intensity
and duration of smoking, room size, ventilation, adsorption of smoke
components, methods of collection, and methods for measurement. Particle
concentrations are also affected by sources other than tobacco smoking.
Furthermore, at a given level of nicotine exposure, urinary cotinine level
is also influenced by uptake, metabolism, and excretion, which are likely
to vary among individuals. Finally, the ability to predict levels of these
markers is also limited by subjects' inability to comprehensively and
accurately describe the extent of exposure.
The results of this investigation have several implications for the
measurement of environmental tobacco smoke in epidemiologic investigations.
Because of the marked variability of these measures, multiple measurements
are needed to establish a stable profile of exposure in a particular home
or for a particular individual. Detailed questions quantitating exposure
to environmental tobacco smoke seem to offer little information beyond
determining whether the subject was exposed or not. Future investigations
of methods for measurement of environmental tobacco smoke must determine
the importance of other factors that may contribute to the variability of
-------
atmospheric and biologic markers, the role of personal monitoring, and the
relevance of these markers to acute and chronic health effects of passive
smoking.
Acknowledgments
Supported by a grant, EPA CR 811650 from the Environmental Protection
Agency. Dr. Coultas is recipient of an Edward Livingston Trudeau Scholar
Award from the American Lung Association.
The authors thank Dr. Helen Van Vunakis for providing the reagents
for the radioimmunoassay and Irene Walkiw for technical assistance in
performing the assays.
References
1. "The Health Consequences of Smoking, a Report of the Surgeon General,"
U.S. Department of Health and Human Services, U.S. Department of Health
and Human Services, Rockville, MD, 1986. (DHHS (CDC) publication no.
87-8398).
2. "Environmental Tobacco Smoke. Measuring Exposures and Assessing Health
Effects," National Research Council, National Academy Press, Washington,
DC, 1986. (ISBN 0-309-03730-1).
3. D. B. Coultas, G. T. Peake, J. M. Samet, "Questionnaire assessment of
lifetime and recent exposure to environmental tobacco smoke," Submitted
for publication.
4. J. J. Langone, H. B. Gjika, H. Van Vunakis, "Nicotine and its
metabolites. Radioimmunoassays for nicotine and cotinine," Biochemistry
12: 5025. (1973).
5. W. R. Faulkner, J. W. King, "Renal function," in: N. W. Tietz, ed.,
Fundamentals of Clinical Chemistry, W. B. Saunders Co., Philadelphia,
P/C1976, pp. 975-1014.
6. W. A. Turner, V. A. Marple, J. D. Spengler, "Indoor aerosol impactor,"
in: B. Y. H. Liu, D. Pui, H. Fissan, eds., Aerosols. Elsevier Science
Publishing Co., Inc. 1984.
7- S. K. Hammond, B. P. Leaderer, A. C. Roche, M. Schenker, "Collection
and analysis of nicotine as a marker for environmental tobacco smoke,"
Atmospheric Environment 21: 457. (1987).
8. SAS Institute Inc., SAS User's Guide: Statistics. Version 5 edition.
SAS Institute Inc., Gary NC.13557
-------
Table I
Description of Houses* and Subjects, and Reported Number of
Hours Exposed to Cigarette Smoke 1n the Home per Day,
New Mexico, 1986
House/Subject
One
1
2
Two
3
4
5
Three
6
7
Four
8
9
10
Five
11
12
Six
13
14
Seven
IS
16
Eight
17
18
Nine
19
Ten
20
Number of Rooms
7
6
7
7
•
8
6
4
5
6
4
Age (yrs)
28
1.5
29
9
5
37
4.5
35
13
4
32
2
49
22
41
14
46
13
74
63
Sex
F
M
F
F
M
M
M
F
F
M
F
F
F
M
F
M
F
M
M
F
Hours/day Exposed
Median (range)
4.5 (0-11)
3.0 (0-7)
5.5 (0-10)
6.3 (2-13)
6.5 (0-11)
3.5 (2-8)
5.0 (2-10)
4.5 (4-9)
7.0 (3-12)
8.5 (7-14)
1.0 (0-19)
3.0 (0-19)
4.0 (0-10)
0.0 (0-3)
11.0 (0-16)
5.5 (0-14)
6.0 (3-13)
6.0 (0-13)
5.0 (4-7)
5.5 (2-16)
*Houses 1-6, 8-9 are unattached, single family homes, 7 Is a mobile home,
and 10 Is an apartment.
-------
Table II
Linear Regression Models* Predicting Resplrable Particle and
Nicotine Concentrations In Air Samples from 10 Homes,
New Mexico, 1986
Dependent Variable
Model 1
Model 2
Model 3
Resplrable particles
(mg/m3)
0.34
0.08
0.09
Nicotine
(mg/m3)
0.28
0.04
0.06
Regression Coefficients'1' Model 3
One or More
Smokers
HOURS
Cold Months
Resplrable particles
(mg/m3)
+17.3
(7.4 - 27.2)
+0.4
(-1.0 - 1.8)
+8.9
(-1.1 - 18.9)
Nicotine
(mg/m3)
+2.1
(-1.7 - 5.9)
+0.2
(-0.1 - 0.5)
-0.7
(-2.5 - 1.1)
independent Variables: Model 1 • HOUSE, Model 2 » NUMBER (0 or >1) +
SEASON (March - April or May - October), Model 3 * NUMBER (0 or II) +
HOURS (continuous) + SEASON (March - April or May - October).
confidence Intervals 1n parentheses.
-------
Table III
Linear Regression Models* Predicting
Urinary Cotinine Concentrations from 20 Nonsmokers
Exposed to Tobacco Smoke,
New Mexico, 1986
Dependent Variable Model 1 Model 2 Model 3
Urinary cotinine
(ng/mg Cr) 0.47 0.05 0.08
Regression Coefficients''' Model 3
One or More
Smokers HOURS Cold Months Children
Urinary cotinine +5.4 +0.8 -0.2 +5.4
(ng/mg Cr) (-4.8 - 15.6) (0.0 - 1.6) (-5.8 - 5.4) (0.0 - 10.8)
Independent Variables: Model 1 = INDIVIDUAL, Model 2 = NUMBER (0 or >1) +
SEASON (March - April or May - October), Model 3 = NUMBER (0 or >1) +
HOURS (continuous) + SEASON (March - April or May - October) + AGE GROUP
(<18 or >18 years).
confidence intervals in parentheses.
-------
o>
Figure 1
Resplrable particle concentrations and means In 10 homes.
200
RSP (ug/cubic m)
150
100
50
t
* *
+- *
* +
I
+
*
-t-
T
4 5 6 7 8 9 10
House
-------
25
20
15
10
5
Figure 2
Nicotine concentrations and means in 10 homes.
nicotine (ug/cubic m)
I * _£_ $ *
f * * ± -4- + ±
1 23456789 10
House
-------
120
100
80
60
40
20
Figure 3
Urinary cotinlne levels and means for children from 10 homes
cotinine (ng/mg Cr)
*± + *"F+l +
+ -±- * * -4- +
7 9 10 12 16 18
Subject
child
-------
Figure 4
Urinary cotinine levels and means for adults from 10 homes.
120
100
80
60
40
20
0
cotinlne (ng/mg Cr)
_i_
. T . ,
s
~r
.L
T
*
+
*^
8
11 13 14 15 17 19
Subject
adult
20
-------
AMERICAN JOURNAL OF EPIDEMIOLOGY Vol. 130. No. 2
Copyright ® 1989 by The Johns Hopkins University School of Hygiene and Public Health Printed in U.S.A.
All rights reserved
QUESTIONNAIRE ASSESSMENT OF LIFETIME AND RECENT
EXPOSURE TO ENVIRONMENTAL TOBACCO SMOKE
DAVID B. COULTAS,'-2 GLENN T. PEAKE,3* AND JONATHAN M. SAMET1-2
Coultas, D. B. (New Mexico Tumor Registry, Cancer Center, U. of New Mexico
Medical Center, Albuquerque, NM 87131), G. T. Peake, and J. M. Samet Ques-
tionnaire assessment of lifetime and recent exposure to environmental tobacco
smoke. Am J Epidemic! 1989:130:338-47.
In a sample of 149 adult nonsmokers recruited in New Mexico in 1986, the
authors assessed the reliability of questionnaire responses on lifetime exposure
to tobacco smoke in the home. They also compared urinary cotinine levels with
questionnaire reports of environmental tobacco smoke exposure during the
previous 24 hours. The agreement of responses obtained on two occasions within
six months was high for parental smoking during childhood: 94% for the mother
and 93% for the father. For the amounts smoked by the mother and the father
during the subject's childhood, the agreement between the two interviews was
moderate: 52% and 39%, respectively. For the number of hours per day that each
parent smoked in the home during the subject's childhood, the Spearman corre-
lation coefficients also indicated only moderate reliability (r = 0.18 for maternal
smoking and r = 0.54 for paternal smoking). For each set of interviews, responses
concerning recent tobacco smoke exposure and urinary cotinine levels were
correlated to only a modest degree. The authors conclude that adults can reliably
report whether household members smoked during their childhood, but informa-
tion on quantitative aspects of smoking is reported less reliably.
pyrrolidinones; questionnaires; tobacco smoke pollution
The term "passive smoking" refers to the ologic investigations (1, 2). However, de-
involuntary exposure of nonsmokers to the spite the evidence linking malignant and
combination of tobacco combustion prod- nonmalignant diseases with active and pas-
ucts released by the burning cigarette and sive smoking, tobacco smoking remains
smoke components exhaled by the active highly prevalent worldwide (1). In the
smoker (1, 2). The adverse health effects of United States at present, about 30 per cent
passive smoking on children and adults of adults are active cigarette smokers (3),
have been described in numerous epidemi- so that a large proportion of nonsmokers
Received for publication March 28, 1988, and in New Mexico Medical Center, 900 Camino de Salud
final form October 11, 1988. NE, Albuquerque, NM 87131.
1 New Mexico Tumor Registry, Cancer Center, Uni- Supported by Grant EPA CR811650 from the En-
versity of New Mexico Medical Center, Albuquerque, vironmental Protection Agency.
NM. Dr. Coultas is a recipient of an Edward Livingston
2 Departments of Medicine and of Family, Com- Trudeau Scholar Award from the American Lung As-
munity, and Emergency Medicine, and the Interde- sociation.
partmental Program in Epidemiology, University of The authors thank Dr. Helen Van Vunakis for
New Mexico, Albuquerque, NM. providing the reagents for the radioimmunoassay and
3 Department of Medicine, University of New Mex- Irene Walkiw for technical assistance in performing
ico, Albuquerque, NM. the assays. Special thanks to the interviewers and to
* Deceased. Lee Fernando, Rita Elliott, and Rebecca Mosher for
Reprint requests to Dr. David B. Coultas, New their help in preparing the manuscript
Mexico Tumor Registry, Cancer Center, University of
338
-------
QUESTIONNAIRE ASSESSMENT OF TOBACCO SMOKE EXPOSURE
339
in this country are involuntarily exposed to
environmental tobacco smoke (1, 2).
Although some health effects of passive
smoking have been convincingly demon-
strated, many questions on the health
effects of passive smoking remain un-
answered. More precise description of
exposure-response relations is needed for
assessment of the adverse effects on chil-
dren and the development of lung cancer.
Additionally, further studies on exposure
to environmental tobacco smoke in the
workplace are warranted because of the
high prevalence of smoking among adults
and public concern about this source of
exposure. In most epidemiologic studies on
involuntary smoking published to date, ex-
posure has been assessed with question-
naires; for the purposes of .some inves-
tigations, the questionnaires have spanned
the entire lifespans of the subjects. Ques-
tionnaires will remain the most feasible
method for assessing exposure to environ-
mental tobacco smoke in new studies. How-
ever, the reliability and validity of question-
naire measures of involuntary smoking
have not been adequately characterized.
In this study, we have assessed the reli-
ability of a comprehensive questionnaire
on lifetime exposure to environmental to-
bacco smoke in 149 adult nonsmokers.
While validity is also of interest, no appro-
priate standard for comparison is available
for a lifetime history. Questionnaire re-
sponses with poor reliability are also likely
to have poor validity. In this sample, we
also examined the relation between reports
of recent exposure to environmental to-
bacco smoke and urinary cotinine levels.
MATERIALS AND METHODS
Sample selection
Between February and December of
1986, nonsmokers aged 18 years and older
were recruited from Albuquerque, New
Mexico, and the surrounding communities.
Recruitment was accomplished by two
methods: advertisements and direct contact
with subjects from a population survey (4).
In both approaches, we asked for volunteers
to participate in a study of indoor air qual-
ity that involved completing a question-
naire on two occasions and providing saliva
and urine samples. The subjects were not
informed that the study was directed spe-
cifically at exposure to environmental to-
bacco smoke. We attempted to stratify the
sample uniformly by age and by sex but
were not completely successful (table 1). Of
our sample, 62 per cent were female, and
only five males were aged 60 years and
older.
Data collection
A structured questionnaire on lifetime
and recent exposure to environmental to-
bacco smoke was administered by a trained
interviewer to each subject on two occa-
sions separated by approximately four to
six months. Training involved familiariza-
tion and practice with the questionnaire
and review of probing techniques, which
were standardized. The interviews were
conducted by four interviewers who com-
pleted 89.2, 5.4, 2.7, and 2.7 per cent of the
first interviews and 38.2, 6.7, 54.4, and 0.7
per cent of the second interviews, respec-
tively. We asked whether the subject's
mother had smoked while pregnant with
the subject, and we determined the smoking
status of parents, spouses, and others from
questions on whether these persons had
smoked in the subject's home on a daily
basis for six months or more. These ques-
tions referred to two time periods: birth to
age 18 years and age 19 years to the time
of the interview. These time periods were
chosen to correspond to the usual ages for
TABLE l
Age and sex distribution of 149 participants in a study
of involuntary exposure to tobacco smoke.
New Mexico. 1986
Age
(years)
20-29
30-39
40-49
50-59
>60
Males
No.
12
20
9
10
5
%
21.4
35.7
16.1
17.9
8.9
Females
No.
17
27
15
15
19
%
18.3
29.0
16.1
16.1
20.4
-------
340
COULTAN IT AL.
living in the parents' horn* and subse-
quently living outside the parents' horn*.
In addition, for each smoker, we asked
about the type(e) of tobacco smoked (ciga-
rette, pipe, or cigar), the amount of each
type umokod in the home, the number of
yean aach type wai •moked, and the num-
ber of hours of exposure per day to each
type in the home. Another net of question!
asked about the amount of exposure during
the previous 24 hours. The questions cov-
ered the number of smokers to which the
subject was exposed, the type(s) of tobacco
•moked (cigarette, pipe, or cigar), and the
number of hours of exposure. These ques-
tions were asked separately for exposures
at home, at work, in vehicles, and at social
gatherings. At the time of the interview, a
urine specimen was collected and frozen at
-'20 C until the cotinine assays were per-
formed.
Catinint auay
Cotinine was quantltated by a double
antibody radioimmunoassay as described
by Langone et al. (6). A specific antiserum
produced in rabbits was supplied by Dr.
Helen Van Vunakis of Brandeis University
(Waltham, MA). Urine samples were di-
luted 1:4 for the assay. The sensitivity of
the assay in our hands was 38 pg/tube or
0.78 ng/ml of urine (4,204 pmol/liter). Uri-
nary creatinine concentrations were deter-
mined by the Jaffa reaction (6), and the
cotinine concentrations were standardised
iu the creatinine concentrations. Assays
were performed without knowledge of ques-
tionnaire responses.
Data analytic
Reliability was assessed by comparison
of the two lifetime histories for the expo-
iure variables during the two time periods,
birth to age 18 years and age 19 years to
the time of the interview. Because of the
•mall number of pipe and cigar smokers
among parents (n • 24) and spouses (n «
4), we restricted our analysis to cigarette
smokers. We summarized the per cent
agreement between the first and second
Interviews for categorical variables, which
Included mother's smoking during preg-
nancy; mother's, father's, and spouse's cig-
arette smoking status; amount smoked, cat-
egorized as less than one pack per day, one
pack per day, and more than one pack per
day; and number of other cigarette smokers
in the household, categorized aa none, one,
and two or more. To discount chance agree-
ments between the first and second Inter-
views, Cohen's kappa was calculated for ail
categorical items and tested for significance
(7, 8). Spearman rank order correlation
coefficients (9) were calculated for contin-
uous variables, which included both the
number of years and the number of hours
per day that the subject's mother, father,
spouse, and others had smoked.
For questions on exposure to tobacco
smoke during the previous 24 hours, we
created summary variables for cigarette
smoke exposure only, because exposure to
pipe and cigar smokers was infrequent. The
summary variables for cigarette smoke ex-
posure included the total number of hours
of exposure and the total number of ciga-
rette smokers in all locations. To examine
the relation between measures of short
term exposure to environmental tobacco
smoke within and between interviews, we
calculated Spearman rank order correla-
tions (9).
Data analyses were performed with stan-
dard programs of the Statistical Analysis
System (10).
RBSULTI
Of the 158 subjects enrolled for the first
interview, 149 (94 per cent) also completed
the second interview. Of the nine subjects
who were not reinterviewed, there were
seven males and two females, with mean
ages of 43.6 years and 43.0 years, respec-
tively. This report is based on responses of
those 149 subjects who were reinterviewed.
The age range of the 149 subjects was 21-
79 years (mean « 43 years); 37.6 per cent
were males and 62.4 per cent were females
(table 1). The median duration between
-------
QUESTIONNAIRE ANHENIMBNT OP TOBACCO HMOKB EXPOBURB 341
Intarviawa WM 17 waaka, with a rangt of 6- and kappa atatiitic for tht numbtr of othtr
39 waaka. dgiratta amokara in tha horn* during child-
For tht pariod birth to aga 18 yaara, hood wara 77.0 par cant and 0.47 (p <
agraamant batwaan tha flnt and lacond 0,0001), raipactivaly,
intarviawa waa high for parantai smoking In contraat to tha high reliability of rt-
•tatua during childhood (tabla 2). Tha par iponaaa about parantai smoking itatua dur-
cant agraamant waa ilmilar for mothar'i ing childhood, ooncordanca waa low for ro-
and fathar'i smoking during childhood and sponaaa about tha uaual amount imokad in
waa lowaat for matarnai amoking during tha horaa by tha paranta during childhood
pragnancy, Tha parcantaga of unknown ra- (tabla 3), Tha concordanca waa highaat for
•ponaaa waa highaat for matarnai smoking tha amount amokad by tha mothar and
during pragnancy, Tha par cant agraamant lowaat for tha amount amokad by tha fa*
TABLI 3
Ptntniajt of nantmokin rtporting txpoiun to partntat oigaHttt making during childhood, Ntw Mttloo, ItMM
Miumil imnlilni Miumil imoklni H»wrnil imoliini
RtiponM durini pnanaiuy during childhood clurlni childhood
(n • I4BI (n» 149) (n • UP)
Y.I
Pint inwrvltw 30,1 ,16,0 08,7
8«oond Inwrvttw 30,1 33,8 Dfl,4
No
Pint lnt»rvi«w 87,1 63,4 43.6
8*oond Interview 04.4 07,1 43.0
Unknown
Flrat Inurvltw 13J 0,7 0.7
Stcond InMrvliW Ifl.fl 0,0 0,7
A|rt«m«nt
Conoordano* BA.O 94.0 03.0
Kappa Q.73* 0.87* o.aa*
•p< 0,0001.
TAIL* 1
Ptntntatt of nanimokun nportinf ttpotun to variaut amount* of ctianttti imoktd by tht partnti during
childhood and by tht ipotut during adulthood, Ntui Mtxieo,
Amount imokcd
Leu than on* pick/day
Flm Inwrvitw
SMond Intt rvl«w
OM pack/day
Flnt Intarvlcw
8»oond Intarvlaw
Mora than ona pack/day
Flm Intarviaw
Stoond inwrviaw
Unknown
Flnt Intarviaw
Saoond Intarviaw
Mturnal imokini
durlni childhood
In - 48)
A3.5
ao.o
30,«
33,0
0.3
16.7
10.4
10.4
Pittrnal imoklni
dunni flhlldhood
(n • 7B)
70,9
30.4
11.4
33,0
10,1
32.ft
7,6
8,0
IpouMl imokini
during adulthood
I/I-H4I
S4.4
40,0
7,8
31,3
0.3
38,1
1,6
0,0
Agreement
Concordance 53.1 30,3 43.8
Kappa 0.82* 0.04* -0.04*
•p>0.08. ——
-------
342
COULTAS ET AL.
ther. Compared with the first interview, the
percentage of subjects reporting parental
smoking of one pack per day or more was
higher at the second interview.
We also examined the reliability of re-
sponses on smoking status and amount
smoked by sex and by age. The findings
were similar to the overall analysis within
strata defined by either sex or age, above
and below age 40 years.
Spearman correlations were used to de-
scribe the agreement between the first and
second interviews on the reported number
of years and hours per day of exposure to
environmental tobacco smoke during child-
hood. The correlation coefficients were
high for responses on the number of years
the parents and other smokers in the
household had smoked (table 4). However,
for responses on the number of hours per
day of smoke exposure in the home, the
correlation coefficients were much lower
(table 4).
We next examined the reliability of re-
ported smoke exposure during adulthood
(tables 3 and 5). After age 18 years, the
numbers of subjects living with either their
mother (n = 8) or their father (n = 9) were
small. For this small group of subjects, the
concordance of responses on parental
smoking status was 100 per cent. Similarly,
the per cent agreement on spouse's smoking
status, as obtained at the two interviews,
was 100 per cent (n = 67). For the amount
currently smoked by the spouse, the con-
cordance was lower (table 3). Agreement
between responses about the number of
other cigarette smokers in the household
TABLE 4
Mean years and hours per day of childhood cigarette smoke exposure reported by nonsmokers.
New Mexico, 1986
Exposure variable
Maternal smoking
Years*
Hours/dayt
Paternal smoking
Years
Hours/day
Other household members'
smoking
Years
Hours/day
No.
33
31
57
55
26
20
First
interview
15^4
5.0
16.1
4.8
13.9
9.2
Second
interview
15.7
6.4
15.4
4.8
13.2
8.4
Spearman's
r
0.76
0.18
0.75
0.54
0.63
0.51
* MT")nrinir fha narinrl fWim Kirth tn acra 1A iraaN fnv Vinttr manv uaara Aid Via/aha cm/lira **iaarafrtaa9'*
t "On average, during the period from birth to age 18 years, for how many hours per day were you exposed
to individuals' cigarette smoke?"
TABLE 5
Mean years and hours per day of adulthood cigarette smoke exposure reported by nonsmokers,
New Mexico, 1986
Exposure variable
Spouse's smoking
Years*
Hours/dayt
Other household members' smoking
Years
Hours/day
No.
40
39
67
58
First
interview
16.2
5.9
8.3
12.7
Second
interview
16.4
5.5
8.2
10.3
Spearman's
r
0.95
0.25
0.78
0.54
* "For how many years did he/she smoke cigarettes while you were sharing your home?"
t "On average, how many hours per day were you exposed to their cigarette smoke?"
-------
QUESTIONNAIRE ASSESSMENT OF TOBACCO SMOKE EXPOSURE
343
was 74.0 per cent (n = 66), with a kappa
value of 0.50 (p < 0.0001).
Correlations between responses at the
two interviews were high for the number of
years the spouse and other smokers in the
household had smoked during the subject's
adulthood, but much lower for the number
of hours per day of exposure during adult-
hood (table 5). Because of the small number
of subjects living with their parents after
age 18 years, we did not calculate correla-
tion coefficients for these variables.
Urine specimens were obtained from 98
per cent of the 149 subjects at the first
interview and 95 per cent at the second
interview. The median urinary cotinine lev-
els were zero at both interviews, with mean
levels of 9.2 ng/mg of creatinine at the first
interview and 7.3 ng/mg of creatinine at
the second interview. Cotinine levels varied
widely with the total number of smokers
and the total number of hours of exposure
to tobacco smoke (in various situations)
during the 24 hours prior to urine collection
at both the first interview (figures 1 and 2)
and the second interview (data not shown).
The cotinine levels correlated only mod-
estly with the questionnaire measures of
exposure (table 6).
We also assessed the stability of data on
exposure, as measured by questionnaire
and by cotinine level (table 6). At the first
100
90
O 80
O)
* 70
O)
* 60
LU
| 50
O 40
O
30
20
10
0
1
NUMBER
>3
FIGURE 1. Urinary cotinine levels, standardized to urinary creatinine (Cr) concentration, among nonsmokers
interviewed about tobacco smoke exposure, by the total number of cigarette smokers the subject reported being
exposed to during the 24 hours prior to the first interview. Bars show the mean cotinine level for each group.
Values in parentheses indicate the number of subjects with nondetectable levels of cotinine. New Mexico, 1986.
-------
344
COULTAS ET AL.
0
O)
E
O)
c
LU
1
5
O
f(n-23) . (n-21) , (n-3)
0.0 0.5-4.9 5.0-14.9 15.0-24.0
HOURS
FIGURE 2. Urinary cotinine levels, standardized to urinary creatinine (Cr) concentration, among nonsmoken
interviewed about tobacco smoke exposure, by the self-reported total number of hours that the subject was
exposed to cigarette smoke during the 24 hours prior to the first interview. Ban show the mean cotinine level
for each group. Values in parentheses indicate the number of subjects with nondetectable levels of cotinine.
New Mexico, 1986.
and second interviews, the mean responses
for the reported number of cigarette smok-
ers that the subjects had been exposed to
during the previous 24 hours were 2.1 and
1.8, respectively, with 20 subjects at the
first interview and 22 subjects at the second
interview reporting exposures in "crowds."
For the total number of hours of exposure
during the previous 24 hours, the mean
responses at the first and second interviews
were 5.1 and 4.6, respectively. Both the
questionnaire variables and the cotinine
data indicated a relatively stable pattern of
exposure. The Spearman correlation coef-
ficients were somewhat higher for the
questionnaire-based indexes than for uri-
nary cotinine levels.
DISCUSSION
In a group of adult nonsmokers, we found
high reliability for reports on parental
smoking and on smoking by others in the
home (table 2) but lower reliability for
semiquantitative exposure measures (ta-
bles 3-5). Mean levels of urinary cotinine
increased with exposure to cigarette smoke
compared with no exposure (n =» 37) (fig-
ures 1 and 2). However, within specific
levels of exposure, the cotinine levels varied
widely. Across the follow-up period of sev-
-------
QUESTIONNAIRE ASSESSMENT OF TOBACCO SMOKE EXPOSURE
345
TABLE 6
Spearman correlations between measures of exposure
to environmental tobacco smoke during the 24 hours
prior to interview. New Mexico, 1986
Exposure variable
No.
Total no. of smokers to which subject
was exposed
Responses at the first and second
interviews
Response at the first interview and
cotinine level
Response at the second interview
and continine level
Total no. of hours that subject was
exposed to cigarette smoke
Responses at the first and second
interviews
Response at the first interview and
cotinine level
Response at the second interview
and cotinine level
Cotinine level
Levels at the first and second
interviews
143 0.50
143 0.24
139 0.21
144 0.62
145 0.32
I
138 0.29
140 0.45
eral months, exposures to environmental
tobacco smoke were relatively stable, as
were urinary cotinine levels (table 6). Most
subjects were able to provide responses to
the questions on maternal smoking during
pregnancy, parental smoking during child-
hood, and smoking by a spouse during
adulthood (tables 2 and 3).
Several limitations of these data must be
considered. Because a standard for validat-
ing a lifetime history of exposure to envi-
ronmental tobacco smoke is unavailable,
we used repeatability as an index of the
quality of questionnaire responses. We ad-
dressed the reliability of questions on life-
time exposure at home, but not in the work-
place, an important source of exposure for
a substantial proportion of adults (11). In-
terview with a volunteer subject does not
replicate the usual setting of a case-control
study, the design most often used to ex-
amine lung cancer and passive smoking (1).
In that setting, recall bias by ill subjects
may affect reliability of questionnaire re-
sponses in comparison with a volunteer
population.
Similar observations on the reliability of
questionnaire data on passive smoking
were recently reported by Pron et al. (12).
These investigators interviewed 117 sub-
jects, controls in a case-control study of
lung cancer, on two occasions separated by
an average of six months. Smoking by
spouses was reported with high reliability
(kappa = 0.89 for both wife and husband).
Repeatability was somewhat lower for
smoking by the mother (kappa = 0.76) and
by the father (kappa = 0.44). As in the
present study, repeatability of quantitative
estimates of duration of exposure was lower
than for the categorical descriptions of
smoking by household members.
Although neither the investigation of
Pron et al* (12) nor the present study di-
rectly addresses validity of questionnaires
on lifetime passive smoking, the validity of
subjects' reports on smoking by parents and
spouses has been described. Sandier and
Shore (13) compared responses on parents'
smoking given by cases and controls with
responses given by the parents or siblings
of the index cases. Concordance was high
for whether the parents had ever smoked,
although the agreement was somewhat bet-
ter for smoking by the mother than for
smoking by the father. Responses concern-
ing numbers of cigarettes smoked did not
agree as highly. In a follow-up study of a
nationwide sample, children's responses on
smoking by their deceased parents closely
agreed with the information given 10 years
previously by the parents themselves (14).
Other studies have shown that people gen-
erally report the smoking habits of their
spouses correctly (14-19). However, peo-
ple's reporting of quantitative aspects of
the smoking behavior of their spouses tends
to be less valid (16,18,19).
Smoking by parents during childhood
and by a spouse during adulthood represent
the most important sources of household
exposure to environmental tobacco smoke.
The studies of subject reports for parents
and spouses indicate good validity of re-
sponses on smoking by these household
-------
346
COULTAS ET AL.
members; the study of Pron et al. (12) and
the present study show that these reports
are also highly reliable. Thus, exposure
measures based on cigarette smoking status
of parents and of spouses, as reported by
an index subject, are reported with a high
degree of validity and reliability, although
these measures may only crudely quanti-
tate the dose of biologically relevant to-
bacco smoke components. In contrast, the
accuracy of more quantitative measures of
smoking by these household members is
lower. The resulting misclassification may
explain the failure to find exposure-
response relations for passive smoking and
lung cancer in some studies (1, 20).
We also compared responses to questions
on exposure during the previous 24 hours
with urinary cotinine level. The time period
for the questionnaire was limited to the
previous 24 hours to reduce bias from faulty
recall. However, since this period is approx-
imately the half-life of cotinine in non-
smokers (21, 22), the cotinine level repre-
sents not only exposure during the 24 hours
covered by the questionnaire but prior ex-
posure as well.
We found modest correlations between
the questionnaire-based measures of expo-
sure and urinary cotinine levels (table 6).
The level of correlation must be interpreted
in the context of the different lengths of
time of exposure assessed by the question-
naire and by the urinary cotinine level.
Furthermore, at a given level of nicotine
exposure, urinary cotinine level is also in-
fluenced by uptake, metabolism, and excre-
tion, which are likely to vary among indi-
viduals.
Coultas et al. (23) found that question-
naire measures of household exposure were
not strong predictors of salivary cotinine
level. In 247 adult nonsmokers with a de-
tectable cotinine level, the subject's age, the
number of cigarettes smoked per day by the
spouse, and the number of cigarettes
smoked per day by other smokers in the
household explained only 2 per cent of the
variance in cotinine levels for females and
16 per cent of the variance for males. Even
in active smokers, questionnaire responses
on smoking behavior do not tightly predict
cotinine concentrations in body fluids (24-
27). Higher correlations between urinary
cotinine levels and reported exposure to
cigarette smoke have been reported for
young children (28). The higher correla-
tions in the studies of young children prob-
ably reflect the time-activity patterns in
this age group (29); parental smoking in the
household is generally the dominant source
of exposure.
In adults, the weak relation between co-
tinine level and reported smoke exposure
implies that a single cotinine measurement
should not be used to estimate exposure for
individuals (23). However, in our subjects,
cotinine levels varied among exposure
groups (figures 1 and 2), suggesting that
cotinine measurements might be used as an
index of mean exposure for members of a
particular exposure group.
Nonsmokers are exposed to environmen-
tal tobacco smoke in many different envi-
ronments, including the home, the work-
place, and other private and public loca-
tions. Since subjects in an epidemiologic
investigation cannot be expected to com-
prehensively describe the extent of expo-
sure in each of these environments, mis-
classification of the amount of exposure to
environmental tobacco smoke must be an-
ticipated from the use of questionnaires.
However, subjects can provide valid and
reliable reports concerning the smoking
status of household members. The combi-
nation of questionnaires and biologic mark-
ers offers a feasible approach for assessing
recent exposure to environmental tobacco
smoke.
REFERENCES
1. US Department of Health and Human Services.
The health consequences of smoking: a report of
the Surgeon General. Rockville, MD: US Depart-
ment of Health and Human Services, 1986.
(DHHS (CDC) publication no. 87-8398).
2. National Research Council Environmental to-
bacco smoke: measuring exposures and assessing
health effects. Washington, DC: National Acad-
emy Press, 1986. (ISBN 0-309-03730-1).
3. Centers for Disease Control. Cigarette smoking in
-------
QUESTIONNAIRE ASSESSMENT OF TOBACCO SMOKE EXPOSURE
347
the United States, 1986. MMWR 1987:36:581-5.
4. Samet JM, Coultas DB, Howard CA, et aL Respi-
ratory diseases and cigarette smoking in an His-
panic population in New Mexico. Am Rev Respir
Dis 1988:137:815-19.
5. Langone JJ, Gjika HB, Van Vunakis V. Nicotine
and its metabolites: radioimmunoaasays for nic-
otine and cotinine. Biochemistry 1973:12:5025-
30.
6. Faulkner WR, King JW. Renal function. In: Tietz
NW, ed. Fundamentals of clinical chemistry. Phil-
adelphia, PA: WB Saunders Co, 1976:975-1014.
7. Fleiss JL. Measuring nominal scale agreement
among many raters. Psychol Bull 1971:76:378-82.
8. Maclure M, Willett WC. Misinterpretation and
misuse of the Kappa statistic. Am J Epidemiol
1987:126:161-9.
9. Colton T. Statistics in medicine. Boston, MA:
Little, Brown and Company, 1974.
10. SAS Institute, Inc. SAS user's guide: statistics.
Version 5 edition, 1985. Cary, NC: SAS Institute,
Inc. 1985.
11. Friedman GD, Petitti DB, Bawol RD. Prevalence
and correlates of passive smoking. Am J Public
Health 1983:73:401-5.
12. Pron GE, Burch JD, Howe GR, et al. The reli-
ability of passive smoking histories reported in a
case-control study of lung cancer. Am J Epidemiol
1988:127:267-73.
13. Sandier DP, Shore DL. Quality of data on parents'
smoking and drinking provided by adult offspring.
Am J Epidemiol 1986:124:768-78.
14. McLaughlin JK, Dietz MS, Mehl ES, et al. Reli-
ability of surrogate information on cigarette
smoking by type of informant. Am J Epidemiol
1987:126:144-6.
15. Rogot E, Reid D. The validity of data from next-
of-kin in studies of mortality among immigrants.
Int J Epidemiol 1975:4:51-4.
16. Kolonel LN, Hirohata T. Nomura AMY. Ade-
quacy of survey data collected from substitute
respondents. Am J Epidemiol 1977:106:476-84.
17. Pershagen G. Validity of questionnaire data on
smoking and other exposures with special refer-
ence to environmental tobacco smoke. Ear J Re-
spir Dis 1984;133(suppl):76-80.
18. Humble CG, Samet JM, Skipper BE. Comparison
of self- and surrogate-reported dietary informa-
tion. Am J Epidemiol 1984;! 19:86-98.
19. Lerchen ML, Samet JM. An assessment of the
validity of questionnaire responses provided by a
surviving spouse. Am J Epidemiol 1986;123:
481-9.
20. Humble CG, Samet JM. Pathak DR. Marriage to
a smoker and lung cancer risk. Am J Public Health
1987:77:598-602.
21. Kyerematen GA, Damiano MD, Dvorchik BH, et
aL Smoking-induced changes in nicotine disposi-
tion: application of a new HPLC assay for nicotine
and its metabolites. Clin Pharmacol Ther 1982;
32:769-80.
22. Sepkovic DW, Haley NJ, Hoffmann D. Elimina-
tion from the body of tobacco smoke products by
smokers and passive smokers. (Letter). JAMA
1986:256:863.
23. Coultas DB, Howard CA, Peake GT, et al. Salivary
cotinine levels and involuntary tobacco smoke
exposure in children and adults in New Mexico.
Am Rev Respir Dis 1987; 136:305-9.
24. Benowitz NL, Hall SM, Herning RI, et al. Smok-
ers of low-yield cigarettes do not consume less
nicotine. N Engl J Med 1983:309:139-42.
25. Abrams DB, Follick MJ, Biener L, et al. Saliva
cotinine as a measure of smoking status in field
settings. Am J Public Health 1987;77:846-8.
26. Lee PN. Lung cancer and passive smoking: asso-
ciation an artefact due to misclassification of
smoking habits? Toxicol Lett 1987:35:157-62.
27. Pierce JP, Dwyer T, DiGuisto E, et al. Cotinine
validation of self-reported smoking in commer-
cially run community surveys. J Chronic Dis
1987:40:689-95.
28. Greenberg RA, Haley NJ, Etzel RA, et al. Mea-
suring the exposure of infants to tobacco smoke.
N Engl J Med 1984:310:1075-8.
29. Harlos DP, Marbury M, Samet J, et al. Relating
indoor NO? levels to infant personal exposures.
Atmosph Environ 1987:21:369-76.
-------
Environmental Health Perspectives
VoL 81,, pp. 173-182, 1990
Cotinine Analytical Workshop Report:
Consideration of Analytical Methods for
Determining Cotinine In Human Body Fluids
as a Measure of Passive Exposure to
Tobacco Smoke*
by Randall R. Watts,f John J. Langone,* George J. Knight,5
and Joellen Lewtas*
A two-day technical workshop was convened November 10-11, 1986, to discuss analytical
approaches for determining trace amounts of cotinine in human body fluids resulting from pas-
sive exposure to environmental tobacco smoke (ETS). The workshop, jointly sponsored by the U.S.
Environmental Protection Agency and Centers for Disease Control, was attended by scientists
with expertise in cotinine analytical methodology and/or conduct of human monitoring studies
related to ETS. The workshop format included technical presentations, separate panel discussions
on chromatography and immunoassay analytical approaches, and group discussions related to the
quality assurance/quality control aspects of future monitoring programs. This report presents a
consensus of opinion on general issues before the workshop panel participants and also a detailed
comparison of several analytical approaches being used by the various represented laboratories.
The salient features of the chromatography and immunoassay analytical methods are discussed
separately.
Introduction
Environmental tobacco smoke (ETS) has increas-
ingly become a health concern since a series of epide-
miological studies between 1981 and 1986 (1-6)
•Chairperson: J. Lewtas, US EPA, Research Triangle Park, NC;
Session chairpersons: F. Sperto, CDC. Atlanta, GA, R. Watts, US
EPA, Research Triangle Park, NC; invited speaker/panel partici-
pants included: Neal Benowitz and Peyton Jacob, San Francisco
General Medical Center; Colin Feyerabend, New Cross Hospital,
London, England; Nancy Haley, American Health Foundation;
George Knight, Foundation for Blood Research; Richard Kornfeld,
Battelle Columbus Laboratories; John Langone, Baylor College of
Medicine; Peter McElroy, Rosewell Park Memorial Institute; M. A.
H. Russell, Maudsley Hospital, London, England; Karl Verebey,
New York State Division of Substance Abuse Services; and Helen
Van Vunakis, Brandeis University.
tU.S. Environmental Protection Agency, Health Effects
Research Laboratory, Research Triangle Park, NC 27711.
tBaylor College of Medicine, Department of Medicine, One Bay-
lor Plaza, Houston, TX 77030.
§Foundation for Blood Research, P. 0. Box 190, Scarborough,
ME 04074.
Address reprint requests to R. R. Watts, U.S. Environmental
Protection Agency, Health Effects Research Laboratory, MD-68,
Research Triangle Park, NC 27711.
reported an association between tobacco smoke expo-
sure and increased risk of human lung cancer. Hum-
ble and co-workers (7) recently confirmed the health
risk conclusions of earlier researchers and reported
that people who never smoked and were married to
smokers had about a 2-fold increased risk of lung
cancer.
Methods for determining the degree of exposure of
individuals has received much attention in recent
years, and various biological markers have been stud-
ied as surrogate analytes for determining exposures.
A general consensus is that the nicotine metabolite,
cotinine, has the prerequisites of specificity, retention
time in the body, and detectable concentration levels
that make it the analyte of choice for quantifying
exposures. In recent years a number of procedures
have been reported for determining cotinine in
human body fluids. The majority of these procedures
use either a chromatographic technique or some form
of immunoassay analysis.
This paper is a report from the two-day Cotinine
Analytical Workshop, which was attended by invited
health scientists and analytical chemists recognized
for their expertise in studies of population exposure
-------
174
WATTS ETAL
to ETS and/or analytical methodology related to
these studies. The workshop was jointly sponsored by
the U.S. Environmental Protection Agency (EPA) and
Centers for Disease Control (CDC) and was attended
by 32 scientists who shared their expertise in immu-
noassay or chromatography methods for cotinine and
provided guidance for developing and establishing
related programs for determining passive exposures
to tobacco smoke. The meeting objective was to com-
pare the various analytical approaches to cotinine
analysis and to make recommendations regarding the
general aspects of establishing and conducting moni-
toring programs. Discussions included quality assur-
ance/quality control (QA/QC) programs to support
cotinine monitoring studies and also the possibility of
conducting a future interlaboratory methods compar-
ison study. The diverse analytical approaches repre-
sented by chromatography and immunoassay
methods for cotinine were separately discussed and
reported by respective work groups. The purpose of
this communication is to summarize discussions from
the immunoassay and chromatography work groups
relevant to the aforementioned topics and to convey
the workshop general consensus on other joint issues
including QA/QC aspects of ETS studies.
Chromatography Group Report
The workshop participants with expertise in devel-
oping and applying chromatography methods for
determining cotinine in biological fluids met in a one-
day session. The goal of this session was to develop a
group consensus on several key issues including a)
general method considerations and approaches, b)
QA/QC programs to support cotinine monitoring
studies, and c) considerations related to conducting
an interlaboratory methods comparison study. The
following is a summary of the chromatography group
discussions and a draft of their recommendations
related to topics a and c. The QA/QC recommenda-
tions are contained in a separate section.
General Method Considerations
Sample Type. The body fluids discussed for moni-
toring tobacco smoke exposure included blood serum,
saliva, and urine. Group consensus was that all three
are generally acceptable; however, the choice of a body
fluid to analyze should be predicated on the goals of
the specific monitoring program. For studies that
require a quantitative assessment of exposure, blood
was recommended by the group as the fluid of choice
(8). Saliva was also considered acceptable, and good
correlations were reported between saliva and blood
for results from the same subject (9). Sample collection
considerations, however, resulted in the selection of
blood as the sample medium of first choice. Analysis of
either blood or saliva for cotinine permits an estimate
of the degree of exposure to tobacco smoke in persons
passively exposed at home or in the work place. While
cotinine determination in urine was also recommended
for estimating exposure, it was generally felt that esti-
mation based on urinary cotinine excretion would1 be
less reliable than estimation based on plasma or sali-
vary levels. Cotinine excretion is variable across and
within individuals depending on renal function, urine
flow rate, and urine pH (10). Urine results may be
expressed as micrograms of cotinine per milligram of
creatinine in order to correct, in part, for the variable
dilution effects. This correction or normalization, how-
ever, introduces additional variability since this
requires another analytical determination (and oppor-
tunity for experimental error), and creatinine excre-
tion rates for individuals are also variable. Horstmann
(11) reported creatinine excretion rate for 56 subjects
to be 1.11 ± 0.68 g/day (mean ± SD). Hoffman and
Brunneman (12) also found 13 nonsmokers on a con-
trolled diet to have creatinine values of 1.65 ± 0.5 g per
24 hr urine (mean ± SD). The coefficients of variation
between subjects for these two studies were 61 and
30%, respectively.
Sample Collection and Handling. Chromatogra-
phy procedures for cotinine generally require analysis
of a 1 mL sample with an additional 1 mL volume
needed for reanalysis. A total sample volume of 2.5 to
3.0 mL was therefore recommended. Glass and/or
polypropylene sample tubes with screw cap closures
were recommended. The polypropylene tubes were pre-
ferred to avoid breakage during shipment. Minimum
size sample tubes were suggested to reduce volume
losses from freeze drying during long-term storage.
Blood should be centrifuged at the field site and the
serum samples frozen prior to shipment to the labora-
tory. Urine should be frozen soon after collection to
prevent bacterial degradation of the sample. Saliva
may be collected by expectoration into a sample tube;
however, an alternative saliva collection procedure
that uses highly adsorbent dental rolls is recom-
mended (IS). The subject is asked to place a dental
roll in the mouth for approximately 15 min. The sam-
ple is then placed in a tube and frozen prior to ship-
ment to the laboratory. The thawed sample is
regenerated at the laboratory by placing the dental
roll in a glass syringe and compressing with a glass
plunger. The resultant clear liquid may then be all-
quoted for analysis.
Shipment in a frozen condition with dry ice was
recommended for all three sample types to prevent
bacterial degradation of the sample matrix. Loss or
degradation of the cotinine analyte was not consid-
ered to be a problem since participants had found this
compound to be stable.
Upon receipt at the laboratory, samples should be
placed in a freezer (approximately — 20° C) until ana-
lyzed. Samples that will be held in excess of one year
should be stored at -80°C. No cotinine degradation
problems were reported for frozen samples. Precau-
tions were recommended, however, to prevent concen-
tration errors resulting from freeze-drying of
samples stored over one year in a frost-free freezer.
-------
COTININE ANALYTICAL WORKSHOP REPORT
175
Analytical Method Considerations
The group consensus was that the analytical
method should permit the determination of nicotine
and cotinine in a single analysis and should allow a
clear separation and distinction between these and
other analytes that may be present. The method
should be sufficiently sensitive to give good definition
of passive exposure and thereby yield analytical
results which will show a distinction, for example,
between a child or other nonsmoker that is exposed in
the home and one that is not. Tables 1 and 2 list the
range of detection limits for both chromatography
and immunoassay methods.
The importance of this sensitivity consideration
was supported by the 1981 report of Hirayama (1) and
the 1987 report of Humble et al. (7) which showed an
increased risk of lung cancer for a spouse exposed to a
smoker in the home. Russell reported that cotinine
levels in children's saliva averaged 0.44 ± 0.68 ng/mL
where no parents are smokers, 1.31 ± 1.21 ng/mL
where only the father smoked, 1.95 ± 1.71 where only
the mother smoked, and 3.38 ± 2.45 ng/mL where
both parents are smokers (IS). This study used an
analytical method with a detection and quantification
limit of 0.1 ng/mL, which permitted classification of
the lowest exposures into exposure distributions dif-
fering by only 0.1 ng/mL. Over 30% of the children
from nonsmoking homes had cotinine concentrations
below the 0.1 ng/mL detection limit. In the groups
where one or more parents smoked, the cotinines were
significantly (p < 0.01) elevated, and 50% of the chil-
dren of the lowest exposed group had less than 1 ng/
mL (when only the father smoked). Table 1 shows
that several available chromatography methods have
detection limits ranging from 0.1 to 0.2 ng/mL while
the most sensitive immunoassay method in Table 2
reports a 0.3 ng/mL detection limit.
The question of analyte volatility losses during
analysis was discussed, and it was generally agreed
that if nicotine were included as an analyte, precau-
tions would need to be taken to prevent loss during
concentration steps. Acidification to convert nicotine
to a salt form prevents losses during concentration.
Cotinine primary standards are used in the free
base form by some analysts; however, a salt form was
preferred by meeting participants, since the free base
form is hygroscopic and difficult to maintain at a
well-defined purity. A perchlorate salt of cotinine was
recommended for preparation of 1 mg/mL stock solu-
tions in 0.01 N HC1 (8). This standard solution could
be frozen and kept indefinitely. The group consensus
was that a salt form of cotinine should be made avail-
able as a primary standard.
Chemical analysis is usually accomplished by gas
chromatography with nitrogen/phosphorus thermi-
onic detection (GC-NPD) or GC-mass spectrometry
(GC-MS) using either electron impact ionization or
chemical ionization. Packed columns for GC were suc-
cessfully used; however, fused silica capillary columns
containing a methyl silicone or methyl phenyl silicone
liquid phase were recommended (see Table 1 HRGC
references).
A high-performance liquid chromatography
(HPLC) method using a Ci8 reversed phase column
with paired ion chromatography and UV detection (at
257 nm) was also reported by McElroy where the
HPLC method of Machacek and Jiang (14) was modi-
fied for analyzing urine samples at passive exposure
levels. Further improvement in HPLC sensitivity and
detection limit is required before application to the
more limited sample volumes generally available for
blood or saliva. HPLC was considered a very promis-
ing approach due to the highly efficient columns now
available and the stability and reproducibility of
response commonly obtainable by UV detection.
The final quantitation of residues in all methods
was accomplished with internal standards and stan-
dard curves developed from fortified blank samples.
It was recommended that standard curves be pre-
pared daily or with each batch of samples. A variety
of internal standards were used ranging from deuter-
ated cotinine and nicotine for GC-MS to chemically
similar compounds for other GC or LC detectors.
Table 1 lists the chromatography methods (14-17)
presented at the workshop and summarizes the sali-
ent features of each. Information for this table was
derived from questionnaire responses submitted by
each author/participant.
Chromatography Group
Recommendations
The chromatography group recommended that an
interlaboratory methods comparison study be con-
ducted prior to any large-scale monitoring efforts
aimed at determining population exposure to tobacco
smoke. Specific suggestions and recommendations
relating to method comparison studies were as
follows:
•Separate studies should be conducted for passive
exposure levels and active smoker levels.
•Statisticians should be used in planning study
samples.
• Blood, urine, and saliva should be included in each
study.
• Immunoassay and chromatography methods should
be included in each method comparison study.
•Samples should be fluids from exposed individuals
and also from fortified blanks in order to look for
bias from chromatography or immunoassay methods
through measurement of artifacts or metabolites
related to nicotine/cotinine.
•The study coordinator should supply standard refer-
ence material(s) to each participating laboratory.
-------
176
WATTS ETAL
Table 1. Summary of passive exposure chromatography methods.
Machacek and
Jiang (14)
Jacob, et al. (15) Verebey et al. (16)
Feyerabend et al.
(17)
Kornfeld (pers6nal
communication)
Sample type
Vol. analyzed, mL
Concentration step
Extraction method
Isolation step
Determination
Quantitation
Linear range
Detection limit
Quantification limit
CV,%
recovery
Urine
N2 evaporated to
dryness
SPE column
Chloroform elution
Acid/base partition
HPLC reverse-
phase
Paired ion
chromatography
Internal standard
Calibration curve
0-500 ng
<1 ng/mL
1 ng/mL
13
90
Blood, plasma,
urine
1
N2 evaporated to
dryness
Solvent extraction
Acid/base partition
HRGC-NPD
HRGC-MS
Internal standard
Calibration curve
.0-4000 ng/mL
0.2 ng/mL
0.5-5 ng/mL
6.8 (3.0 ng/mL)
12 (1.0 ng/mL)
13 (0.5 ng/mL)
107
Serum
0.5
None
Solvent extraction
Acid/base partition
HRGC-NPD
Internal standard
Calibration curve
40-400 ng/mL
5 ng/mL
5 ng/mL
5.6
90
All biological fluids Urine
N2 evaporated to
dryness
Solvent extraction
None: plasma and
saliva
back extract
urine
GC-NPD
Internal standard
Calibration curve
0-15,000 ng/mL
0.1 ng/mL
0.1 ng/mL
7.7
90
N2 evaporated
solvent exchange
Solvent extraction
None
HRGC-MS
Internal standard
Calibration curve
1-500 ng/mL
0.13 ng/mL
0.25 ng/mL
6
85
Immunoassay Group Report
Participants in the workshop with expertise in the
development and use of immunoassays for detecting
cotinine in biological fluids met independently of the
chromatography group to discuss and make recom-
mendations regarding methodology and applications
of immunoassay in monitoring passive as well as
active exposure to tobacco smoke, QA/QC programs,
and interlaboratory methods comparison. The follow-
ing discussion presents an overview of the available
immunoassay techniques for cotinine analysis, their
applications with advantages and disadvantages, and
the views and recommendations of the immunoassay
panel members. There is notable agreement between
this group and the chromatography group on most
common issues outside the technical aspects specific
to each methodology.
General Method Considerations
Introduction. The first radioimmunoassay (RIA)
for cotinine was reported in 1973 (18,19). Antisera
were raised in rabbits and goats immunized with a
covalent conjugate prepared by linking cotinine 4'-
carboxylic acid to immunogenic carrier proteins, such
as bovine serum albumin and keyhole limpet hemocy-
anin. The radioactive tracer was prepared by labeling
a tyramine derivative of cotinine 4'-carboxylic acid
with "*!; since then, [^Jcotinine has been prepared
enzymatically (19) from [3H]nicotine and is now
widely used. Another approach uses cotinine deriva-
tized at the 1-position in the pyridine ring for prepar-
ing the immunogen and as a precursor of an 125I-
labeled tracer (20). The original assay has been used
to measure cotinine levels in physiological fluids, e.g.,
urine, blood, saliva, amniotic fluid, and spinal fluid
Table 2. Summary of immunoassay methods.
Langone et al. (18)
Langone and Van
Vunakis (19)
Haley et al. (22)
Knight et al. (29)
Bjercke et al. (-27,28)
Sample type and volume
analyzed, mL
Assay type
Quantitation
Detection limit, ng/mL
Quantitation limit, ng/mL
CV, %
Urine (0.02-0.05)
Serum (up to 0.5)
Saliva (0.02)
RIA (""I, 3H)
Internal standard
Calibration curve
2
2
6-10
Urine, plasma, saliva
(0.005-0.025)
RIA (JH)
Internal standard
Calibration curve
0.37
1
10
Urine (0.01)
Serum (0.1)
RIA (ml)
Internal standard
Calibration curve
0.3
1
10-15
Urine, serum, saliva
(0.1 in RIA; 0.01 in
microtiter plate
assays)
RIA (""I, »H), ELISA,1
FIA
Internal standard
Calibration curve
0.5-1.5
0.5-1.5
9-14
i assay concentrated serum.
-------
COTININE ANALYTICAL WORKSHOP REPORT
177
(18,19,21-28) of active smokers and serum, urine, and
saliva (24-26), of passive smokers.
More recently, monoclonal antibodies specific for
cotinine have been prepared and used to develop fluid
phase RIAs with the 125I- and 3H-labeled tracers as
well as enzyme-linked immunosorbent assays
(ELISA) and a fluorescence immunoassay (FIA) in a
microtiter plate format (27,28). These assays also
have been used to measure cotinine levels in fluids of
active (27,28) and passive smokers (Langone et al.,
unpublished results).
Test Samples and Standards. Because the origi-
nal RIAs and the monoclonal antibody-based
nonisotopic assays have been developed for analysis
of unextracted physiological fluids, careful attention
must be paid to possible nonspecific inhibition of
antigen-antibody binding resulting from effects of pH
or high concentrations of salts or urea, e.g., in urine.
In this regard the immunoassay group agreed and
strongly recommended that pooled standard samples
of serum, saliva, and urine containing known amounts
of added or endogenous cotinine should be made
available through an agency such as the National
Institute of Standards and Technology. There was
general agreement that GC/MS would be the best
method to establish the cotinine concentration for
purposes of methods comparison and that levels
should cover the range from cotinine-free through
concentrations found in passive and active smokers.
Essentially cotinine-free samples might be collected
from a population that would represent a group with
minimal exposure to tobacco smoke (e.g., Mormons in
Utah). The suggestion also was made that low-level or
essentially cotinine-free fluids might be treated (e.g.,
by absorption with XAD-2 resin or charcoal) to
remove possible traces of cotinine. However, because
absorption could remove other constituents that
might affect the assays, it was not considered to be a
firm recommendation.
Although it was suggested that urine may be the
fluid of choice for RIA analysis, there was no strong
consensus for priority over serum or saliva. In this
regard, one participant pointed out the advantage
that salivary cotinine levels determined by RIA are
independent of saliva flow (Van Vunakis and Regas,
unpublished results). The monoclonal antibody assays
also have been used to detect cotinine in saliva and
urine of passively exposed children (Langone et al.,
unpublished observations), and these investigators
tended to favor the use of saliva. In addition to the
use of dental rolls as discussed by the chromatogra-
phy group, one member of the immunoassay group
suggested that subjects chew a piece of Teflon tape to
stimulate the flow of saliva that is then collected in a
glass vial. It was pointed out that Teflon will not
contaminate the sample. Regardless of which fluid is
tested, it was recommended that samples be cen-
trifuged (e.g., 20000 for 10-20 min or 100000 for 1-2
min) to sediment particulate matter before analysis.
Immunoassay group participants concurred with the
sample handling recommendations given in the Chro-
matography Group Report.
Comparison of the Assays
The original RIA and variations of it are used by
the immunoassay group participants. Therefore, the
discussion focused on this method and the monoclonal
antibody based assays, the salient features of which
are summarized in Table 2.
Reagents. The same immunogen was used to pro-
duce the rabbit, goat, or sheep antisera and the
monoclonal antibodies. However, it was emphasized
that cotinine 4'-carboxylic acid (and the ^I-labeled
derivative) is a mixture of stereoisomers giving rise
to a heterogeneous population of polyclonal antibo-
dies recognizing both natural (—)-cotinine and the
(H-)-enantiomer. Also, conventional antisera contain a
population of antibodies that bind specifically to the
linkage group that joins cotinine to the imraunogenic
carrier protein and to the tyramine group in the 1251-
labeled derivative. The practical consequences are a
relatively shallow standard inhibition curve and the
failure to achieve 100% inhibition of immune binding
with (—)-cotinine. Although these problems are cir-
cumvented by using (—)-pH] cotinine, this assay is
somewhat less sensitive, owing to the lower specific
activity and counting efficiency achieved with tri-
tium. Also, disposing of large volumes of radioactive
scintillation fluid is a major concern.
Two approaches have been used with some success
to improve the quality of the 12SI-RIA with rabbit
antisera. They involve removing antibridge group
antibodies by absorption with a nicotine-hemocyanin
conjugate (29) and preparing an 12SI-labeled cotinine
derivative with a bridging group different from that
present in the immunogen (30). In contrast,
monoclonal antibodies to cotinine were produced in a
way that avoided the problems inherent in the use of
polyclonal antisera (27,28). Although the immunogen
contained a mixture of isomers, the hybridomas were
screened using (—)-prI]cotinine to optimize chances of
detecting antibodies that preferentially recognize the
naturally occurring isomer, but not the bridging
group in the immunogen. Furthermore, it was pointed
out that monoclonal antibodies are preferred stan-
dard reagents for immunoassay because they are con-
tinuously available and are homogeneous in terms of
binding affinity and specificity.
The specificity of any newly produced antiserum
must be fully characterized one time with a battery of
compounds that would include at least cotinine, nico-
tine, and metabolites such as nicotine N'-oxide, nor-
nicotine, and trans-3'-hydroxycotinine. This
recommendation holds even for new antisera pre-
pared with a proven immunogen, since the response of
individual immunized animals cannot be predicted.
However, all agreed that when the properties of the
antiserum had been established, it was unnecessary
for each laboratory that received that antiserum to
-------
178
WATTS ETAL
complete a thorough reexamination of specificity,
although it would be good laboratory practice to rou-
tinely compare the cotinine and nicotine inhibition
curves.
Immunoassay methods have often reported using
standard cotinine as the free base. However, because
cotinine is hygroscopic and difficult to weigh accu-
rately, all participants agreed that a nonhygroscopic
salt of cotinine, such as the perchlorate or fumarate,
would be the preferred standard.
Assay Performance. In the original RIA,
antibody-bound and free-labeled cotinine were sepa-
rated by the double-antibody method in which a het-
erologous antibody directed against the species of
anticotinine was used to precipitate antigen-antibody
complexes (18). Other techniques can be used includ-
ing precipitation with ammonium sulfate or polyeth-
ylene glycol (29). Although the latter methods are
faster and less expensive, there was some concern
expressed that background radioactivity precipitated
by ammonium sulfate, when normal serum is used in
place of anti-cotinine, can exceed 10% of the added
amount of cotinine tracer.
In contrast to the conventional fluid phase RIAs,
the monoclonal antibody-based assays are carried out
in a solid-phase system in which a cotinine-polylysine
conjugate is passively adsorbed to the surface of 96-
well plastic microtiter plates (27,28). Immobilized
cotinine and fluid phase cotinine in the test sample
compete for monoclonal anticotinine, which is
detected with a variety of enzyme-labeled antiimmu-
noglobulin reagents including the bacterial product,
protein A. Assay sensitivity can be enhanced by using
a sandwich procedure in which rabbit anti-mouse
immunoglobulin is added before (or along with)
labeled protein A. It was emphasized that protein A
reagents cannot be used to detect low levels of serum
cotinine, because host IgG will bind nonspecifically to
the microtiter wells giving high background binding
of the enzyme-labeled protein A tracer.
Compared to times when rabbit antisera were used,
assays with monoclonal antibodies were more sensi-
tive, the standard curves were steeper, and the anti-
gen-antibody reaction was completely inhibited by
(—)-cotinine, even when the I25l-labeled tyramine
derivative was used in RIA (28). There was good
agreement between the levels of cotinine found in
saliva and serum of smokers determined by conven-
tional RIA, the monoclonal antibody ELISA and GC
(27,28). It was pointed out that high quality rabbit
antisera also can be used in the solid phase
nonisotopic immunoassays with liters that can be
100- to 1000-fold higher than in RIA (27,28).
Specificity and Sensitivity. Both polyclonal and
monoclonal antibodies are specific for cotinine
(18,19,27). Approximately 50 to 100 compounds that
have been tested in the immunoassays including sev-
eral nicotine metabolites, related tobacco alkaloids,
and other compounds that retain structural features
of either or both ring systems found in nicotine or
cotinine fail to inhibit the antigen-antibody reactions
at levels that would interfere in the assays.
One participant emphasized that literally
thousands of serum and urine samples from both
active smokers and nonsmokers had been analyzed
over a period of several years and that few, if any,
false positives had been reported. Although the sub-
jects studied are mainly from the U.S. and England,
these data support the view that diet or other factors
such as prescription or other drugs do not interfere in
the assays and are consistent with high specificity of
anticotinine. It was pointed out that differences in
diet or drug use must be considered when other popu-
lations are studied, or at least be aware that interfer-
ence in the assays might arise from factors which
have not appeared to date.
The immunoassays generally can detect cotinine
down to the ng/mL level or less (Table 2), although it
was emphasized that sensitivity can be affected by
the need to dilute samples (e.g., urine) that may give
spurious results when higher concentrations are
tested. This point was discussed at some length with
the participants in agreement that a sensitivity for
cotinine of 0.1 ng/mL of physiological fluids could not
generally be achieved with confidence using the avail-
able immunoassays. In this regard, it was pointed out
that differences in sensitivity limits between chroma-
tography and immunoassay likely reflect fundamen-
tal differences in methodology and are not strictly
comparable. GC methods, for example, might extract
and analyze a considerably larger portion of sample
than would be analyzed by immunoassay.
Analytical Results. There was general agreement
that cotinine concentration should be expressed as
nanogram per milliliter. However, urine values also
should be given as nanogram per milligram creati-
nine, as this ratio is used conventionally in the medi-
cal literature to account for differences in urine
volume. Because low levels of creatinine in infants
relative to adults may result in misleading values
that fall into the range reported for active smokers,
the need to include primary data for urine was
stressed. Furthermore, experience has shown that
urinary cotinine levels determined by conventional
RIA generally are 30 to 50% higher than values
obtained for the same samples by GC. Discussion cen-
tered on the possibility that the higher RIA values
may reflect cross-reactivity of anti-cotinine with
£rcms-3'-hydroxycotinine, which recently has been
reported to be a major nicotine metabolite found in
smokers' urine at levels up to three times higher than
cotinine (31).
Since this meeting, synthetic £rarw-3'-hydroxyco-
tinine (supplied by Dr. Peyton Jacob, San Francisco
General Medical Center) has been shown to cross-
react only 1 to 2% compared to cotinine in the
monoclonal antibody based ELISA; one participant
stated that he found only about 5% cross-reactivity
with his rabbit antiserum in RIA. This degree of
cross-reaction would not account for the discrepancy
-------
COTININE ANALYTICAL WORKSHOP REPORT
179
Table 3. Summary comparison of cbromatograpby and RIA methods.
Sample type
Vol. analyzed, mL
Extraction and concentration
Quantification
Detection limit, ng/mL
Quantification limit, ng/mL
CV,%
Chromatography methods
Blood, saliva, urine
0.5-6
Yes
Internal standard
Calibration curve
0.1-5
0.1-5
5.6-13
RIA methods
Blood, saliva, urine
0.005-0.5
No
Internal standard
Calibration curve
0.3-2
0.5-2
6-15
in the urine values, and it was agreed that further
research was needed to clarify the basis for the
differences.
Considerations in Selecting an
Analytical Technique
Table 3 shows some comparisons for RIA and chro-
matography methods. Apparent differences are in
sample volumes used, sample work-up requirements,
and limits of detection. RIA methods use less than
10% of the sample volumes required for chromatogra-
phy methods, and this is a major reason that RIA
detection limits are not as low as those for Chroma-
tography methods. Because RIA methods do not
require sample manipulations such as extraction and
concentration, they are faster, simpler, and presuma-
bly less expensive. Chromatography procedures not
only have the advantage of increased sensitivities, but
also are more specific and can provide quantification
of both nicotine and cotinine in a single analysis.
Workshop participants agreed that the choice
between these two approaches would depend on the
goals of a particular study. Both approaches have
been found to be 100% effective in discriminating
smokers from nonsmokers (82}. This particular goal
would favor the use of an RIA method. At least one
participant suggested that the more sensitive chro-
matography methods are recommended to character-
ize ETS exposures for plasma or saliva concentrations
when levels are less than 1 ng/mL.
A compilation of literature values for cotinine con-
centrations in body fluids of nonsmokers before and
after documented ETS exposures is shown in Table 4.
This comparison indicates a similarity between
plasma and saliva concentrations, while urine values
are often a factor of two or more higher. This is a
primary reason that urine is often the fluid of choice
when RIA methods are used in passive smoking
studies.
Quality Assurance for Laboratories
Assaying Cotinine
Participants in the cotinine workshop discussed the
need for developing a quality assurance (QA) pro-
gram for monitoring performance of laboratories
assaying cotinine for the purpose of assessing expo-
sure to ETS. When assuming many subjects, such a
QA program would be essential to ensure that the
conclusions reached are based on reliable data. A one-
time exercise where the ability of laboratories to
measure cotinine levels found in both active smoking
and for passive exposure to ETS was considered as an
alternative possibility. This suggestion was prompted
by the realization that although published data on
cotinine levels found in body fluids for active smokers
show reasonable agreement, levels of cotinine
reported for subjects exposed to ETS show considera-
ble variation. Such differences might not be unex-
pected when measuring the low levels of cotinine
found in ETS exposure, given that the detection limits
for existing analytical methods approximate these
cotinine levels.
To evaluate the between-laboratory variation in
cotinine analyses, an international study was initi-
ated by the Forschungsellschaft Rauchen and
Gesundheit mBH in Hamburg (32). Eleven laborato-
ries experienced in measuring nicotine and cotinine
by RIA and/or GC participated. Serum and urine
specimens from eight nonsmokers and eight smokers,
and from two nonexposed nonsmokers spiked with
nicotine and cotinine were distributed on dry ice to
each laboratory. Results were returned and analyzed
by method and by laboratory. Recoveries on both the
urine and serum specimens spiked with cotinine cor-
responding to levels found in smokers ranged from 79
to 119%, with the exception of one laboratory with a
20% recovery (the data from this laboratory were
excluded from further analysis). The interlaboratory
coefficient of variation on these same samples was
excellent (9-13%). The coefficient of variation on
samples from smokers was fairly large, however,
ranging from 18 to 45% for serum and 21 to 59% for
urine. Further, cotinine levels reported for urine were
about 60% higher than from those using RIA as com-
pared to GO, suggesting a possible interfering sub-
stance in the immunoassay system. Cotinine levels
reported for nonsmokers were extremely variable,
and a number of laboratories could not detect cotinine
in serum from exposed nonsmokers. In addition,
cotinine values reported by some laboratories bore no
relationship to estimated ETS exposure, or they were
so high as to be unrealistic. In spite of this variability,
all laboratories were able to discriminate smokers
from nonsmokers with 100% effectiveness.
-------
180
WATTS ETAL
Table 4. Mean or median concentrations of cotinine in nonsmokers before and after exposure to environmental tobaccp
smoke.
Plasma cotinine,
Study reference ng/mL
(33)
(31>)
(9)
(IS)
(35)
(36)
(25)
(37)
Before
0.82
1.1
0.8
0.9-1.7
After
2.04
7.3
1.8-2.5
2.6-3.3
Saliva cotinine,
ng/mL
Before
0.73
1.5
0.7
0.4
1.0-1.7
1.3-1.7
After
2.48
8.0
2.2-2.8
1.3-3.4
1.4-2.5
2.4-5.6
Urine cotinine,
ng/mL
Before
1.55
4.8
1.5
14
8.5
After
7.71
12.9
6.5-9.4
21-55
25.2
2.8-29.6
The cotinine results reported for ETS exposure
should be viewed with caution, however. A number of
the participants at this conference workshop, who
also were in the study, indicated that the volumes
provided were insufficient for repeat analysis using
GC or an assay was used which had not been opti-
mized for measuring passive levels of cotinine. A fur-
ther limitation of the study was that recovery of
spiked cotinine was only assessed for smoking levels.
Finally, immunoassays based on monoclonal antibo-
dies were not included, nor were HPLC methods
evaluated.
This interlaboratory study indicates the need for
further information on the reliability of data pro-
vided by laboratories for study subjects exposed to
ETS. A quality assurance program could provide such
information, as well as an ongoing assessment of
quality and a mechanism for improving performance.
QA Recommendations
Interlaboratory QA. The need for an
interlaboratory quality assurance program was
endorsed by most of the session participants, with
some concern being expressed that the number of
samples evaluated be kept to reasonable limits to
minimize unnecessary assays. It was recommended
that such a program should be administered by a QA
coordinator laboratory. The coordinating laboratory
would be responsible for monitoring the performance
of participating laboratories and for providing speci-
fied samples as standards and/or controls. This labo-
ratory should have in-house expertise or have access
to laboratories having expertise in both immunoassay
and high resolution gas chromatography/mass spec-
trometry (HRGC-MS).
Suggested objectives of the QA coordinator labora-
tory include:
• To provide an objective measure of the precision and
accuracy of analytical methods used routinely by
laboratories assaying cotinine.
• To identify preferred method(s) for measuring
cotinine.
• To assess the reliability of results provided by dif-
ferent laboratories.
• To provide a mechanism for improving performance
through knowledge of the performance of others.
- To serve as a resource center for communication and
exchange of information among participants.
Recommended mechanisms for accomplishing the
foregoing objectives are as follows:
Interlaboratory Quality Assurance Studies:
Quality Assurance Samples. The coordinator labo-
ratory should periodically conduct a blind or check
sample study consisting of authentic biological fluids
(serum, urine, or saliva) with actual or spiked levels
of cotinine. Samples should be selected to represent
cotinine levels typical of those found in passive and
active smoking. Authentic biological samples with
actual levels of cotinine are strongly recommended
because only they will contain nicotine metabolites or
other substances that may interfere in assays. In
addition, blank samples spiked with known levels of
cotinine should be distributed to evaluate recovery.
Finally, samples with high levels of cotinine should be
diluted with negative specimens to check for linearity.
Samples should ideally have target values assigned by
the QA coordinating laboratory through use of refer-
ence methods. Data returned by participants would be
analyzed and reports containing results and a critique
distributed.
Field Study Samples. The QC coordinating labo-
ratory may assist organizations carrying out field
studies in assessing the performance of the study lab-
oratory on actual study subjects. The workshop con-
sidered that this could be accomplished by submission
at intervals of blind duplicates: duplicates of the same
study subject submitted at intervals to assess preci-
sion; split samples: sample is split with one portion
being sent for analysis to the study laboratory and
one portion to the QC coordinating laboratory for
comparison purposes; blanks: samples that are con-
sidered free of analyte to serve as a check on environ-
mental contamination.
Ancillary Activities of the QA Coordinating
Laboratory: Primary Reference Standard(s). A
strong consensus was reached that a well-character-
ized, pure, primary reference standard be made avail-
able. This material should be aliquoted into
quantities sufficient to allow any laboratory to use
-------
COTININE ANALYTICAL WORKSHOP REPORT
181
the standard for assay calibration. Handling and
storage information should also be provided along
with suggested methods for preparing secondary
standards. It was generally agreed that cotinine
should be in the form of a salt, since cotinine freebase
is hygroscopic and, therefore, likely to vary in compo-
sition dependent on handling conditions. The perchlo-
rate salt was suggested as one possibility (see
chromatography group report). It was further recom-
mended that the standard be supplied in solution to
preclude errors due to dilution. The GC group made
the suggestion that the National Institute of Stan-
dards and Technology might be the appropriate
agency to prepare such a standard. The QC coordinat-
ing laboratory could then distribute the standard.
Biological Reference Samples. The suggestion
was made that, in addition to providing a primary
reference standard, the QA coordinating laboratory
make available authentic biological samples from
actual smokers and subjects exposed to ETS. Cotinine
concentrations would be established by the QA coor-
dinating laboratory using a reference method(s) and
declared on each reference sample. Such samples
would be important because the cotinine would be
present in the matrix (urine, serum, saliva) actually
used by analysts, thus allowing evaluation of possible
matrix interference. In addition, such specimens
would contain nicotine, nicotine metabolites other
than cotinine, and other substances which might
interfere in the assay.
The GC group also felt that blank samples, i.e.,
those essentially free of cotinine would be desirable.
Suggested sources were bovine serum or human sam-
ples with very low exposure to ETS. Pooled specimens
might be necessary because obtaining sufficient vol-
ume of biological reference samples could prove diffi-
cult. Samples would therefore be provided in
restricted quantities only to allow laboratories to
periodically evaluate their own method.
Reference Method. Workshop participants also
discussed a reference method for establishing
cotinine levels in biological samples. The consensus of
the group was that GC/MS would be the ultimate
reference method because of its extreme specificity.
However, in the group discussion, the GC group
pointed out that although the method is highly spe-
cific, it ultimately is no better than the reliability of
the extraction and evaporation methods chosen to
prepare samples for analysis. A further concern was
that differences in assigned values may result from
differences attributable to the detection method
(chemical ionization or electron impact). Conse-
quently, it appears unlikely that a gold standard will
be available and acceptance of a reference method
will depend ultimately on judgment of its reliability.
Representatives of the National Institute of Stan-
dards and Technology indicated that their practice is
to evaluate a variety of independent methods, and if
sufficient agreement is reached, a certified value is
provided, albeit with the understanding that confi-
dence limits are somewhat uncertain. In the absence
of agreement between various methods, NIST pro-
vides a consensus value(s) for informational purposes.
In the event that GC/MS is adopted as a reference
method, the implication for immunoassayists is that
their performance would be judged against this stan-
dard. Judging immunoassay results against GC/MS is
not without precedent, since other immunoassays,
such as those for steroids, are already compared to
this method.
Postscript
A cotinine spiked, freeze-dried human urine refer-
ence material is being prepared by the National Insti-
tute of Standards and Technology (formerly the
National Bureau of Standards). Three lots with dif-
ferent cotinine concentration levels are being pre-
pared: a) an unspiked blank level (< 1 ppb), 6) an
approximately 50 ppb low level, and c) an approxi-
mately 500 ppb high level. This material (EPA/NIST
Reference Material 8444) is planned for issue during
the first quarter of 1989. The material may be ordered
from: Office of Standard Reference Materials, Build-
ing 222 Room B-311, National Institute of Standards
and Technology, Gaithersburg, MD 20899. Telephone
301-975-6776. Technical information may be obtained
from Dr. Lane Sander, Organic Analytical Research
Division, Center for Analytical Chemistry, NIST,
Gaithersburg, MD 20899. Telephone 301-975-3117.
A cotinine perchlorate salt reference material is
also being planned for development by NIST. A date
has not been determined, however, for release of this
standard.
The research described in this report has been reviewed by the
Health Effects Research Laboratory, U.S. Environmental Protec-
tion Agency, and approved for publication. Approval does not sig-
nify that the contents necessarily reflect the views and policies of
the Agency nor does mention of trade names or commercial prod-
ucts constitute endorsement or recommendation for use.
The previously listed speaker/panel participants are acknowl-
edged for their workshop participation and vital contributions to
this report The authors are also appreciative of the attendance and
active participation of the following university and government
agency representatives: Larry Claxton, Ruth Etzel, Linda Fore-
hand, George M. Goldstein, Elaine Gunter, Katharine Hammond,
Ed Hu, Henry S. Kahn, Kevin J. Kimboll, Dave Mage. Ronald K.
Mitchum, Judy Mumford, Todd C. Pasley, Terry Pechacek, George
Provenzano, D. W. Sepkovic, Wanda Whitfield, Ron Williams,
Deborah Winn, Stephen A. Wise, and Ou-Li Wong.
REFERENCES
1. Hirayama, T. Non-smoking wives of heavy smokers have a
higher risk of lung cancer a study from Japan. Br. Med. J. 282:
183-185 (1981).
2. Trichopouloa, D., Kalandidi, A., Sparros, L, and McMahon, B.
Lung cancer and passive smoking. Int J. Cancer 27:1-4 (1981).
3. Garfinkel, L, Time trends in lung cancer mortality among
non-smokers and a note on passive smoking. J. Natl. Cancer
Inst. 66:1061-1066 (1981).
-------
182
WATTS ETAL
4. Correa, P., Williams, P. L., Fontham, E., Lin, Y., and Haenzel,
W. Passive smoking and lung cancer. Lancet ii: 595-597 (1983).
5. Akiba, S., Kato, H., and Blot, W. J. Passive smoking and lung
cancer among Japanese women. Cancer Res. 46: 4804-4807
(1986).
6. Dalager, N. A., Pickle, L. W., Mason, T. J., Correa, P.,
Fontham, E., Stemhagen, A., Buffler, P. A., Ziegler, R. G., and
Fraumeni, J. F., Jr. The relation of passive smoking to lung
cancer. Cancer Res. 46: 4808-4811 (1986).
7. Humble, C. G., Samet, J. M., and Pathak, D. R. Marriage to a
smoker and lung cancer risk. Am. J. Pub. Health 77: 598-602
(1987).
8. Benowitz, N. L. The use of biologic fluid samples in assessing
tobacco smoke consumption. Measurement in the analysis and
treatment of smoking behavior. In: NIDA Research Mono-
graph 48 (J. Grabowski and C. S. Bell, Eds.), U. S. Government
Printing Office, Washington, D. C., 1983.
9. Jarvis, M., Tunstall-Pedoe, H., Feyerabend, C., Vesey, C., and
Salloojee, Y. Biochemical markers of smoke absorption and
self reported exposure to passive smoking. J. Epidemiol.
Comm. Health 38: 335-339 (1984).
10. Benowitz, N. L, Florence, K., Jacob , P., Ill, Jones, R. T., and
Abdel-Latif, 0. Cotinine disposition and effects. Clin.
Pharmacol. Ther. 34: 604-611 (1983).
11. Horstmann, M. Simple high-performance liquid chromato-
graphic method for rapid determination of nicotine and
cotinine in urine. J. Chromatogr. 344: 391-396 (1985).
12. Hoffmann, D., and Brunnemann, K. 0. Endogenous formation
of Af-nitrosoproline in cigarette smokers. Cancer Res. 43:
5570-5574 (1983).
13. Jarvis, M. J., Russell, M. A. H., Feyerabend, C., Eiser, J. R.,
Morgan, M., Gammage, P., and Gray, E. M. Passive exposure
to tobacco smoke: saliva cotinine concentration in a represen-
tative population sample of non-smoking school children. Br.
Med. J. 291: 927-929 (1985).
14. Machacek, D. A., and Jiang, N. Quantification of cotinine in
plasma and saliva by liquid chromatography. Clin. Chem. 32:
979-982 (1986).
15. Jacob, P., Ill, Wilson, M., and Benowitz, N. L. Improved gas
chromatographic method for the determination of nicotine
and cotinine in biologic fluids. J. Chromatogr. 222: 61-70
(1981).
16. Verebey, K. G., DePace, A., Mule, S. J., Kanzler, M., and Jaffe,
J. H. A rapid, quantitative GLC method for the simultaneous
determination of nicotine and cotinine. J. Anal. Tox. 6:294-296
(1982).
17. Feyerabend, C., Bryant, A., Jarvis, M. J., Russell, M. A. H.
Determination of cotinine in biological fluids of non-smokers
by packed column gas-liquid chromatography. J. Pharm.
Pharmacol. 38: 917-919 (1986).
18. Langone, J. J., Gjika, H. B., and Van Vunakis, H. Nicotine and
its metabolites. Radioimmunoassays for nicotine and cotinine.
Biochemistry 12: 5025-5030 (1973).
19. Langone, J. J., and Van Vunakis, J. J. Radioimmunoassay of
nicotine, cotine, and y (3-pyridyl)-Y-oxo-Jv*-methylbutyramide.
Methods Enzymol. 84: 628-640 (1982).
20. Matsukura, S., Sakamoto, N., Seino, Y., Tamada, T., Mat-
suyama, H., and Muranaka, H. Cotinine excretion and daily
cigarette smoking in habituated smokers. Clin. Pharmacol.
Ther. 25: 555-561 (1979).
21. Langone, J. J., and Van Vunakis, H. Radioimmunoassay ot
Drugs and Hormones in Cardiovascular Medicine (A. Alber-
tini, M. Da Prado, and B. A. Peskar, Eds.), Elsevier/North
Holland Biomedical Press, Amsterdam, 1979 pp. 55-70.
22. Haley, N. J., Axelrad, C. M., and Tilton, K. A. Validation of
self-reported smoking behavior: biochemical analyses of
cotinine and thiocyanate. Am. J. Public Health 73: 1204-1207
(1983).
23. Sepkovic, D., and Haley, N. J. Biomedical applications of
cotinine quantitation in smoking related research. Am. J. Pub-
lic Health 75: 663-65 (1985).
24. Greenberg, R. A., Haley, N. J., Etzel, R. A., and Loda, F. A.
Measuring the exposure of infants to tobacco smoke. Nicotine
and cotinine in urine and saliva. N. Engl. J. Med. 310:
1075-1078 (1984).
25. Wald, N. J., Boreham, J., Bailey, A., Ritchie, C., Haddow, J. E.,
and Knight, G. Urinary cotinine as marker of breathing other
people's tobacco smoke. Lancet i: 230-231 (1984).
26. Pattishall, E. N., Strope, G. L, Etzel, R. A., Helms. R. W.,
Haley, N. J., and Denny, F. W. Serum cotinine as a measure of
tobacco smoke exposure in children. Am. J. Dis. Child 139:
1101-1104 (1985).
27. Bjercke, R. J., Cook, G., Rychlik, N., Gjika, H. B., Van Vunakis
H., and Langone, J. J. Stereospecific monoclonal antibodies to
nicotine and cotine and their use in enzyme-linked immu-
nosorbent assays. J. Immunol. Methods 90: 203-213 (1986).
28. Bjercke, R. J., Cook, G., and Langone, J. J. Comparison of
monoclonal and polyclonal antibodies to cotinine in
nonisotopic and isotopic immunoassays. J. Immunol. Methods
96: 239-246 (1987).
29. Knight, G. J., Wylie, P., Holman, M. S., and Haddow, J. E.
Improved IBI radioimmunoassay for cotinine by selective
removal of bridge antibodies. Clin. Chem. 31: 118-121 (1985).
30. Jones, S. R. and Amatayakul, S. Improved 1KI radioimmunoas-
say for cotinine. Clin. Chem. 31:1076-1077 (1985).
31. Neurath, R. B., and Pein, F. G. Gas chromatographic determi-
nation of trans-3'-hydroxycotinine, a major metabolite of nico-
tine in smokers. J. Chromatogr. 415: 400-406 (1987).
32. Biber, A., Scherer, G., Hoepfner, L, Adlkofer, F., Heller, W.,
Haddow, J. E., and Knight, G. J. Determination of nicotine in
human serum and urine: an interlaboratory study. Toxicol.
Lett. 35: 45-52 (1987).
33. Russell, M. A. H. Estimation of smoke dosage and mortality of
nonsmokers from environmental tobacco smoke. Toxicol. Lett.
35: 9-18 (1987).
34. Jarvis, M. J., Russell, M. A. H., Feyerabend, C. Absorption of
nicotine and carbon monoxide from passive smoking under
natural conditions of exposure. Thorax 38: 829-833 (1983).
35. Hoffmann, D., Haley, N. J., Adams, J. D., Brunnemann. K. D.
Tobacco sidestream smoke: uptake by nonsmokers. Prev. Med.
13: 608-617 (1984).
36. Wald, N., and Ritchie, C. Validation of studies on lung cancer
in nonsmokers married to smokers. Lancet i(8385): 1067 (1984).
37. Coultas, D. B., Samet, J. M., Howard, C. A., Peake, G. T., and
Skipper, B. J. Salivary cotinine levels and passive tobacco
smoke exposure in the home. Am. Rev. Respir. Dis. 133
A157-A158 (1976).
-------
ELIMINATION OF URINARY COTININE IN CHILDREN EXPOSED
TO KNOWN LEVELS OF SIDE-STREAM CIGARETTE SMOKE
George M. Goldstein
U.S. Environmental Protection Agency
Research Triangle Park, NC
Albert Collier
University of North Carolina
Chapel Hill, NC
Ruth Etzel
Centers for Disease Control
Atlanta, Georgia
Joelien Lewtas
U.S. Environmental Protection Agency
Research Triangle Park, NC
Nancy Haley
American Health Foundation
Valhalla, New York
Abstract
The establishment of a quantitative personal marker of side-stream
smoke exposure in children is important in the study of potential health
effects in this group. Cotinlne, a metabolite of nicotine, has been
shown to exhibit a dose-response relationship to side-stream smoke
exposure in adults, and has been used to quantitate prior exposure.
This study was undertaken to determine the dose of nicotine, the peak
level of urinary cotinine, the time to peak cotinine levels, and the
elimination half-life of cotinine in young children exposed to a con-
trolled amount of side-stream smoke. With an exposure to 26.4 ug/rn^
of nicotine, nine children, averaging 2.03 years old had peak cotinine/
creatinine levels of 818 ng/mg. The time to peak cotinine was 4.1
hours, with an elimination half-life of 28.7 hours.
Disclaimer
Although the research described in this paper has been funded
wholly or in part by the Health Effects Research Laboratory, U.S.
Environmental Protection Agency through cooperative agreement 0CR812738-
01 to the Center for Environmental Medicine, University of North Caro-
lina, it has not been subjected to the Agency's peer and policy review
and therefore does not necessarily reflect the views of the Agency and
no official endorsement should be inferred.
-------
Introduction
Increased concern has been expressed about the potential health
risks associated with the exposure to side-stream smoke (21). Recent
studies implicate exposure to side-stream smoke as a particular health
risk in infants and young children (3,8,17,19). Research into the
health effects of exposure to side-stream smoke in children would be
greatly aided if a chemical marker could be used to predict the level
of exposure to side-stream smoke. Several substances, isolated from
tobacco smoke, or their metabolic products, have been measured in
biological fluids to estimate this exposure to side-stream smoke.
These substances include carboxyhemoglobin, thiocyanate, nicotine and
cotinine (4,6,12,13,14,15). Cotinine, a metabolite of nicotine, has
been shown to be a good indicator of the exposure to side-stream smoke
(14). Studies in adults have shown that there is a dose-response
relationship between the number of cigarettes smoked and the level of
cotinine in the urine (9,13). The elimination half-life of cotinine in
the urine and in the blood has also been reported in adults (1,2,12).
The use of cotinine as an indicator of side-stream smoke exposure in
children has been studied by Greenberg et al. (6). They found a high
correlation between the exposure of children at home to side-stream
smoke and their levels of urinary cotinine. These results suggested
that uninary cotinine may be a useful indicator of side-stream smoke
exposure in infants and young children. Etzel et al. (4) provided data
on the elimination half-life of cotinine in the urine of newborn in-
fants exposed to side-stream smoke in utero. The level of exposure for
both these studies came from the self-reported smoking behavior of the
mother.
Information on the uptake of nicotine and the elimination of its
metabolite, cotinine in young children, age 1 to 3 years of age, ex-
posed to side-stream smoke is unavailable. This information is con-
si derd critical because it will allow one to estimate prior exposure,
rather than rely on questionnaire data. This study was undertaken to
determine the exposure dose of nicotine, the peak level of urinary
cotinine, the time to peak levels of cotinine, and the elimination
half-life of urinary cotinine when children are exposed to a controlled
amount of side-stream smoke.
Methodology
Subjects: Nine children (5 males, 4 females; mean age 2.03 yrs.; age
range 9 months to 3.5 yrs.), accompanied by their parents or guardians
were exposed to side-stream smoke generated by a smoking machine at the
rate of two cigarettes per hour for a total of four hours. All children
were Individuals who had been exposed to side-stream cigarette smoke in
their home environments. The parents were asked to refrain from ex-
posing the children to side-stream smoke for three days prior to the
chamber exposure, during the day of exposure, and for three days fol-
lowing the exposure. Urine samples were collected for cotinine analyses
each morning prior to the exposure, during the exposure, the evening
-------
following the exposure and for three days following the exposure. Once
collected, the urine samples were coded, frozen, and shipped to the
American Health Foundation in Valhalla, New York, for analysis. Cotinine
was measured by radioimmunoassay (6,18). The concentration of cotinlne in
the urine was standardized by expressing it as the ratio of cotlnine/
creatinine, ng/mg (18).
Prior to each study, the children received a physical examination
during which respiration rate and body weight were determined. This was
repeated at the end of the exposure period. This study was approved by
the Committee on the Protection of the Rights of Human Subjects of the
School of Medicine of the University of North Carolina at Chapel Hill.
Chamber: The experiments were performed in a chamber, 13.6 np, with
3.55 air changes per hour. Temperature was maintained at 2410.3 C,
with a relative humidity of 55±5 Z. The air within the chamber was
mixed by a small boxer fan at 50CFM.
Environmental Measurements: Air nicotine levels were measured on
bisulfice-impregnated filters (7). Each exposure concentration was the
average of 3-6 filters, collected over the four hour exposure period.
Total Suspended Particulate values for the study were averaged from
duplicate Anderson filter samples collected during each 4 hour exposure
at a flow rate of 1.7 m^ per hour. A Thermo-Systems, Inc. Model 3030
Electronic Aerosol Analyzer made measurements every 10 minutes. Carbon
monoxide (CO), NO and NOx were measured continuously and reported every
three minutes. CO was measured by nondispersive infrared, (NDIR) while NO
and NOx were measured by chemiluminescenee.
Exposure: Research grade cigarettes, 2R1, were smoked by a smoking
machine, [RM30, Heinr. Borgwaldt, Hamburg,FRG]. Prior to the study, the
cigarettes were conditioned at 22°C at 60% RH for 48 hours. One cigarette
was ignited every 30 minutes and burned for about 12 minutes. The
cigarette was puffed once a minute, each puff lasting 2 seconds with a
draw of 35 ml (10,16). All mainstream smoke was vented to the outside of
the chamber.
Data Analyses: The urine half-life elimination rate of cotlnine was
calculated by an interactive pharmacokinetic model'(11), starting from
the peak urine level of cotinine.
Results
Concentrations of gases in the side-stream smoke peaked 15 minutes
after ignition and cycled at 30 minute intervals. Peak levels for CO
were 2.2±0.8 ppm; NO, 54.814.3 ppb; and NOx, 58.815.3 ppb. The mean
value for the air concentration of nicotine was 26.4l6.6yg/m^. The
mean concentration of particulates, measured as TSP was 321.81 20.7ug/m^.
and by electronic aerosol analyzer was 345.4±51.2yg/m^.
The dose of nicotine that each child received was calculated from the
mean concentration of air nicotine for that exposure, the average of the
child's respiratory rate, and the child's tidal volume calculated from
-------
Its body weight (cable 1) using the formula of Gautller ee_ al. (5).
The diacribueion of baseline levels of urinary cot1nine appeared Co
be blmodal, Indlcaclng Che possibility of exposure of several children
Co side-scream smoke prior Co encering Che study. The mean cocinlne/
creaclnlne ratio for one group was 36.5±17.3 ng/ag, while Che other
group vaa 190118 og/mg.
Ac 4.1±1.4 hours after exposure Co the first cigarette, peak
levels of urine cocinlne/creacinine averaged 393.7±209.8 ng/mg in Che
first group, and 1635±489 ng/mg in Che second group (cable 1). The
average urine elimination half-life of cotlnine was 28.7±13.1 hours
wich a range of 14.6 Co 55.1 hours.
Table 1: Summary of Nicotine and CoClnine Levels for Each Exposure
Air
Nicocine
Subject Ug/m3
1 20
2 23
3 28
'4 35
5 23
6 23
7 18
8 35
9 33
Mean 26*7
±S.D.
Nlcocine
dose ug
26.9
20.3
21.8
23.8 •
24.9
23.9
12.9
28.0
29.5
23.6
±4.9
Baseline
co ci nine
ng/mg
43*
190b
208 b
21*
44*
37*
13*
172b
61*
37±17*
190±18b
T Peak
coCinlne
ng/mg
175*
1095&
2048 b
429*
429*
667*
238*
1762b
524*
394*
±210
1635b±
489
Time to
peak
(hours )
3.8*
6.3b
4. Ob
5.3*
5.8*
3.9*
2.9*
3.2b
1.9*
3.9*
±1.5
4.5b
±1.6
Tl/2
hours
38.9
24.2
14.6
15.1
55.1
20.3
33.1
22.1
35.3
28.7+13.1
tCotinine/CreaCinine ratio
aNon-exposed subjects
^Subjects suspected of having prior exposure Co smoke in the home environment
-------
Discussion
This study was undertaken to describe the uptake of nicotine by young
children in a controlled exposure to side-stream smoke and the events
associated with cotinine elimination. The gaseous and particulate
fraction of side-stream smoke reported in this study compare very
favorably with the values reported by Hoffman et^ al. (9) when their
data are adjusted for the differences in chamber size, exhaust rate,
and the number of cigarettes consumed per hour.
In this study population, the distribution of baseline levels of
urinary cotinine was bimodal. One group had an average value of 36.5
ng/mg which compares to reported values of 14.5 (4) and 19 ng/mg (9).
We believe that the other group, with an average value of 190 ng/mg,
had been exposed recently to side-stream smoke. This is supported by
values of cotlnine/creatlnine of 459 ng/mg (9) and 117 to 780 ng/mg
(12) reported in exposed children. For each child, a dose of inhaled
nicotine was calculated from the average nicotine concentration for
that child's exposure.
The distribution of the peak levels of cotinine also exhibited a
bimodal distribution (table 1). The group of subjects with the initial
lower baseline level of cotinine had a mean peak level of cotinine/cre-
atine of 393.7 ng/mg compared to 1635 ng/mg value for the group that
may have had prior exposure. The time that it took to reach peak
cotinine levels was 4.1 hours for the entire group. This compares with
a reported value of 2 hours by Wilcox et al. (20) and 2 to 4 hours
reported by Hoffmann et al. (9). In the paper by Hoffmann et al. (9),
it is Interesting to note that as the exposure to side-stream smoke
increases, not only does the peak level of cotinine increase, but the
time to peak also increases. Our data suggest that this may have also
occurred, but our data base is too limited to compare the group with
the presumed prior exposure with the group that had abstained for three
days. The time course of cotinine elimination as measured in the urine
appears to follow exponential kinetics. The average urine half-life was
28.7 hours. This rate appears to be shorter when compared to a half-life
in neonates of 68 hours reported by Etzel e£ al. (4) but within the 7
to 40 hours reported in the adult population (1,2,12).
Conclusion
Although the study population was small, the data suggest that the
elimination half-life of urinary cotinine in very young children is
similar to that reported in the literature for adults. Future work
should replicate this study, while exercising control of the children's
exposure to side-stream smoke during both the pre and post chamber
exposure phase.
-------
References
1. Beeket, A.H., Gorrod, J.W., and Jenner, P. The analysis of nocotine-
l'-N oxide in urine, in the presence of nicotine and cotinine, and
its application to the study of in_ vivo nicotine metabolism In man.
J. Pharm. Pharmac. 23 (1971), 55s-61s.
2. Benowltz, N.L., Kuyt, P., Jacob, 111, P., Jones, R.T., and Osman,
A-L. Cotinine disposition and effects. Clin. Pharnacol. Ther. 34
(1983), 604-611.
3. Colley, J.R.T., Holland, W.W., and Corkhill, R.T. Influence of pas-
sive smoking and parental phlegm on pneumonia and bronchitis in
early childhood. Lancet 11 (1974), 1031-1034.
4. Etzel, R.A., Greenberg, R.A., Haley, N.J., and Loda, F.A. Urine cotin-
ine excretion in neonates exposed to tobacco smoke products in
utero. J. Pediatrics 107 (1985), 146-148.
5. Gaultier, Cl., Boule, M., Allaire, A., Clement, A., and Glrard, F.
Growth of lung volumes during the first three years of life. Bull.
europ. Physlopath. reap. 15 (1979), 1103-1116.
6. Greenberg, R.A., Haley, N.J., Etzel, R.A., and Loda, F.A. Measuring
the exposure of infants to tobacco smoke. New England Journal of
Medicine 310 (1984), 1075-1078.
7. Hammond, S.K., Leaderer, B.P., Roche, A.C., and Schenker, M. Collec-
tion and analysis of nicotine as a marker for environmental tobacco
smoke. Atmospheric Environment (In Press).
8. Harlap, S., and Davis, A.M. Infant admissions to hospital and maternal
smoking. Lancet 11 (1974), 529-532.
9. Hoffmann, D., Haley, N.J., Adams, J.D., and Brunnemann. Tobacco side-
stream smoke: uptake by nonsmokers. Preventive Medicine 13 (1984),
608-617.
10. Hoffmann, D., Adams, J.D., and Haley, N.J. Reported Cigarette smoke
values: a closer look. Am. J. of Public Health 73 (1983), 1050-1053.
11. Johnston, A., and WooHard, R.C. STRIPE: An interactive computer
program for the analysis of drug pharmacokinetics. J. Pharmacologi-
cal Methods 9 (1983), 193.
12. Luck, W., and Nau,H. Nicotine and cotinine concentrations in serum
and urine of infants exposed via passive smoking or milk from smoking
mothers. J. of Pediatrics 107 (1985), 816-820.
13. Matsukura, S., Taminato, T., Kitano, N., Seino, Y., Hamada, H., Uchi-
hashi, M., Nakajlma, H., and Hirata, Y. Effects of environmental
tobacco smoke on urinary cotinine excretion in nonsmokers. The New
England Journal of Medicine 311 (1984), 828-832.
-------
14. Matsukura, S., Sakamoto, N. Seino, Y., Tamada, T., Matauyama, H., and
Muranaka, H. Cotinine excretion and dally cigarette smoking in habi-
tuated smokers. Clin. Pharmacol. Ther. 25 (1979), 555-561.
15. Pojer, R., WhitfieId, J.B., Poulos, V., Eckhard, I.F., Richmond, R.,
and Hensley, W.J. Carboxyhemoglobin, cotinine, and thiocyanate assay
compared for distinguishing smokers from non-smokers. Clin. Chem. 30
(1984), 1377-1380.
16. Rickert, W., and Robinson, J.C. Estimating the hazards of less hazard-
ous cigarettes. 11. Study of cigarette yields of nicotine, carbon
monoxide, and hydrogen cyanide in relation to levels of cotinine,
carboxyhemoglobin, and thiocyanate in smokers. J. of Toxicology and
Environmental Health 7 (1981), 391-403.
17. Samet, J.M., Tager, I.B., and Spizer, F.E. The relationship between
respiratory illness in childhood and chronic air-flow observation in
adulthood. Am. Rev. Respir. Dis. 127 (1983), 508-523.
18. Sepkovic, D.W., and Haley, N.J. Biomedical applications of cotinine
quantitation in smoking related research. Am. J. Public Health 75
(1985), 663-664.
19. Weiss, S.T., Tager, I.B., Schenker, M., and Spelzer, F.E. The health
effects of involuntary smoking. Am. Rev. Respir. Dis. 128 (1983),
933-942.
20. Wilcox, R.G., Hughes, J., and Roland, J. Verification of smoking
history in patients after infarction using urinary nicotine and
cotinine measurements. British Medical Journal 2 (1979), 1026-1028.
21. The health concequences of Involuntary smoking, a report of the Sur-
geon General. U.S. Dept. of Health and Human Services (1986).
-------
Home Air Nicotine Levels and Urinary Cotinine Excretion in
Preschool Children1'4
FREDERICK W. HENDERSON, HOLLY F. REID, ROBIN MORRIS,
OU-LI WANG, PING C. HU, RONALD W. HELMS, LINDA FOREHAND, JUDY MUMFORD,
JOELLEN LEWTAS, NANCY J. HALEY, and S. KATHARINE HAMMOND
Introduction
Measurement of cotinine concentra-
tions in body fluids has received recent
scrutiny for its usefulness as an indica-
tor of tobacco smoke exposure in adults
and children (1-13). Average levels of coti-
nine in serum, saliva, and urine differ
significantly for unexposed individuals,
those exposed passively to environmen-
tal tobacco smoke (ETC), and active smok-
ers. There remain questions, however,
regarding the extent to which cotinine lev-
els in body fluids are quantitatively in-
dicative of the intensity of ETS exposure
among passive smokers. Lack of this
knowledge and the absence of informa-
tion regarding the extent to which single
cotinine determinations are representa-
tive for individuals has limited the use-
fulness of cotinine measurements in epi-
demiologic studies of the health effects
of passive smoking.
In the present study, we examined the
relationship between levels of nicotine in
the home air and levels of cotinine in
urine of 27 children between 11 months
and 5 yr of age. In addition, we obtained
a midweek urine sample from study chil-
dren for cotinine determination each of
4 consecutive weeks to examine the sta-
bility over time of this biologic marker
of ETS exposure.
NWtnOQS
The children studied attended the research day
care program of the Frank Porter Graham
Child Development Center in Chapel Hill,
North Carolina. The respiratory disease re-
search conducted in this population has been
described previously (14-16). Children start
attending the Center between 6 wk and 3
months of age and are followed continuous-
ly to the age of kindergarten entry. They at-
tend the day care program 8 h/day, 5 days/
wk ani return home each ewning and on week-
ends. Smoking is not permitted in child care
areas of the day care center or in vans that
are used to transport children to and from
the day care center. Twenty-seven children par-
ticipated in the study: 15 from homes with
SUMMARY Waanmtnadtna extant of <
nlna/cmattnlne ratio* (CCH) hi 27 cMMn
landurinecott-
OOt eXDOead to (
canualloi»a of ntaMloamr»maakwafadatan
fiMjM noufs on 2 cofUMCuow
landed a maaaren day cam program wham they
• (TTS) during tha daytime hour*. Average can-
id by active elf aampHng during tha evening and
Ufino MMHDJOS toe oottoww MM
COUOCtOO D0VOfV| OUftn0v MM flftOC tflO tWO •MllpllflQ pOfMOB* III MJOmlOllf ROUT OaMjUOfluM
urinaiamplaa lor CC« warn obtstfnadm^ study chlidmn to datatm^
datarmlnattona o* OCR «am rapraaanHUita tor IndMdual chsMrnn. Fmaan cnMmn maided m homaa
wtth amofcim. and 13 dM not Urine CCH conamleiHty dmttngulanad meat wpoeid and unawpoead
CnlMIVfli HOWMMOTf tfMM OKpOOMJ OfNUPMI nMJ UnRO CCnB tlMt CIIJa9lM9Q fOUVIIOiy MOUOO tnO CntOTIOft
OCR (30 ng/mg eottnloaoaattHna) that baat dtotinguwhad Mpmid and unaapoiad eMUnjn. In
children eapoeid to ET8 In tha homo, than •** • Hgnmeant cormiaUon between average honw
air nteotlna lava* and tha mitng* lugnlthm of urina OCR tha BJO roomlnga «nar tha homa ik
monitoring parted* (r « OM; p • OttM). In study cMdran, urin* CCRa worn lomailably ctaMo
0¥ar tha 1-month oeaarvotlon ported. Rank eorioloiloiicoaMfclanli«Dr*aquomiol»aaMyd»>innlno.
tlona of CCfl warn constantly amour than r • OM; p < O0001.
smokers and 12 from homes without smok-
ers. Three pairs of siblings were studied: two
pairs from smoking homes and one pair from
a nonsmoking home. The study population
consisted of 15 males and 12 females; 14 black
and 13 white children. Their ages ranged from
11 months to 5 yr.
Measurement of the Extent of Tobacco
Smoking in the Home
Three methods were used to estimate the con-
centration of tobacco smoke present in home
air. (/) A questionnaire was administered
with questions that identified the persons
who smoked in the home and how many cig-
arettes were smoked. (2) Parents were asked
to save the unsmoked portions of all cigarettes
smoked in the home over 2 consecutive mon-
itoring days. Parents were also asked to save
the remnants of cigarettes smoked in their cars
while riding with their children. Separate cig-
arette collections were made for each study
day. (J) Active home air sampling for nico-
tine was performed simultaneously with cig-
arette remnant collections. Air sampling pumps
(Model 520; Gilian Co., Wayne, NJ) operat-
ing at 15 L/min were employed with Ander-
sen sampling heads (Particle Fractionating
Sampler, Andersen Co., Atlanta, GA). Teflon-
coated glass fiber filters (75 mm, T60A20;
Pallflex, Putnam, CT) impregnated with
aqueous 57t sodium bisulfate were used for
nicotine trapping (17). The pumps operated
from 5:00 r.u. to 7:00 A.M. for each of the
2 days of monitoring. Separate air nicotine
determinations were made for each study day.
Sampling heads were kept in the child's main
activity room during the evening hours, then
moved to the child's bedroom at bedtime.
(Received in original form Junt 22, 1988 and in
rtvised form January 12, 7989)
1 From the Department of Pediatrics, the Frank
Porter Graham Child Development Center, the
Department of Biostatistics, University of North
Carolina, Environmental Health Research and Test-
ing Inc. and the Health Effects Research Labora-
tory United States Environmental Protection Agen-
cy. Research Triangle Park, North Carolina; the
American Health Foundation. Valhalla. New York;
and the Department of Family and Community
Medicine, University of Massachusetts, Worcester,
Massachusetts.
' Supported by Co-operative Agreement CR
80739202 from the U.S. Environmental Protection
Agency and the University of North Carolina Center
for Environmental Medicine
1 Correspondence and requests for reprints
should be addressed to Frederick W. Henderson.
MJX, CB 7220. University of North Carolina,
Department of Pediatrics, Chapel Hill NC 2759*-
7220.
* This paper has been subjected to internal re-
view by staff of the Health Effects Research Labo-
ratory of the UiEnviionmenial Protection Agency
and has been approved for publication. Approval
does not signify that the contents necessarily re-
flect the views and policies of the agency, nor does
mention of trade names of commercial products
coogituieendonemeaorrecomtnfnriarionforuae.
197
-------
196
HtMOCmOM, ROD. HOMW, WANO. HO, HfUM, FOMMANO, UUMFONO, UWTM, MALtY. AND HAMMOMO
Tin* line
1200 1500* 1700 2100*
1
2400 700* 930* 1200
Home air nicotin* monitoring
* Indicates scheduled urina collection
Fig. 1. Horn* air monitoring and urine collection scfwdula: time line tor both study days 1 and 2.
Collection of Urine Specimens for
Cotinine Determinations
Urine specimens were obtained before, dur-
ing, and at the end of the period of home
air nicotine monitoring (figure 1). Children
who were not toilet-trained (n = 9) did not
have urine specimens collected at home. These
children had the 3:00 P.M. and midmorning
urine specimens specified by the protocol col-
lected while they were at the day care center.
In addition, urine specimens for cotinine de-
termination were obtained on four consecu-
tive Thursdays after the home air monitor-
ing period.
Measurement of Nicotine and Cotinine
Nicotine was eluted from sodium bisulfate-
impregnated filters into ethanol and subse-
quently quantitated by gas chromatography
using described methods (17). Concentrations
of cotinine in urine samples were determined
by competitive-inhibition radioimmunoassay
using rabbit cotinine antiserum and initiated
cotinine (purchased from Helen VanVunakis,
Brandeis University) (18). To adjust for urine
dilution, urine cotinine concentrations were
standardized to creatinine concentration and
expressed as cotinine/creatinine ratios (CCR).
Remits
Comparability of Estimates of
Quantity of Tobacco Smoked in
Home and Classification of
Exposure Status
Except for one instance, there was agree-
ment among the three measures of ex-
posure employed: questionnaire, cigarette
remnant collection, and active home air
nicotine monitoring. Figure 2 shows the
distribution of two consecutive home air
nicotine determinations in homes where
parents did or did not save cigarette rem-
nants. Average air nicotine concentra-
tions in homes where cigarette remnants
were saved were consistently greater than
0.5 ug/m1; all children from these homes
were considered exposed to ETS in the
home. One family that reported smok-
ing in the home in questionnaire respons-
es saved no cigarette remnants. Average
concentrations of nicotine in home air
were greater than 0.5 ug/mj on both
monitoring days; therefore, this child was
classified as exposed to ETS in the home.
Otherwise, children who lived in homes
where no cigarette remnants were saved
were classified as not exposed to ETS in
the home. The average concentration of
air nicotine in the homes of unexposed
children was 0.34 ug/m1 (SEM, 0.07
ug/nr1) and in the homes of exposed chil-
dren was 3.74 ug/mj (SEM, 0.52 ug/m3).
Accuracy of Exposure Assignment
Using Midmorning Urine Cotinine
Determinations
The study protocol specified that mid-
morning urine specimens for cotinine be
collected from study participants 6 times
over a 5-wk period (two consecutive
mornings after active home air nicotine
monitoring and four consecutive Thurs-
days thereafter). A total of 153 of 162
(94%) of these urine specimens were suc-
cessfully obtained and assayed for coti-
nine and creatinine concentrations. The
distribution of CCR for these specimens
is plotted in figure 3 in relation to ex-
posure status. A urine CCR criterion level
of 30 ng/mg resulted in the highest rate
of agreement in exposure assignment.
Sixty-six of 69 (95%) midmorning urine
specimens obtained from 12 unexposed
children had CCRs < 30 ng/mg (first
column, figure 3). Urine specimens from
12 of 15 exposed children consistently (63
of 67; 94%) contained greater than 30
ng cotinine/mg creatinine (second col-
umn, figure 3). However, three exposed
children consistently had urine CCRs
that clustered near the 30 ng/mg break-
point (13 of 18 = 72% of urine speci-
mens tested had CCR < 30 ng/mg) (third
column, figure 3). The sensitivity of urine
CCR was 80%; 12 of IS exposed chil-
dren were consistently designated as ex-
posed to ETS using urine CCR at the 30
ng/mg criterion level. Thus, for 24 chil-
dren (12 exposed and 12 unexposed), ac-
curacy of exposure categorization using
urine cotinine was excellent at the 30 ng/
mg criterion level. Only seven of 136 (5%)
urine specimens tested from these chil-
dren resulted in misctassification of ex-
posure status. The remaining three ex-
posed children were misclassified as not-
exposed by 13 of 18 (72%) urine cotinine
determinations using the 30 ng/mg CCR
criterion. Two-day average home air nico-
tine levels in the homes of these three chil-
dren were 1.45, 3.85, and 3.40 ug/m1.
Home air nicotine averaged 2.90 ug/mj
for the three children not classified as ex-
posed by urine cotinine compared to 3.95
ug/mj for the 12 children whose exposure
status was consistently correctly identi-
fied by urine cotinine determinations.
The analyses that follow were per-
formed with and without the data from
the three exposed children with consis-
tently low urine CCRs. Exclusion of da-
ta of exposed children with low CCRs
did not alter the pattern of any analytic
result; data shown are for all exposed
children.
Correlation of Home Air Nicotine
and Urine Cotinine Levels
Among ETS-exposed children, there was
a significant positive correlation between
the average concentration of nicotine in
home air over the 2 days of monitoring
and the average logarithm of urine CCR
on the mornings after the 2 evenings of
home air monitoring. In figure 4, aver-
age CCRs are plotted in relation to aver-
age home air nicotine levels for each of
the 15 exposed children. The Pearson
correlation for this set of data was r =
0.68; p = 0.006. The regression equation
for the linear correlation is given in the
figure legend.
CIGARETTE REMNANTS SAVED
Fig. 2. DtatrtbuUon o> ivataga homa air nicotine con-
cantrattona tor horn** wftara ogartfl* ramnanti want
and want not «avad; maaauratninn obtained on two
conaacirtva daya paf home.
-------
KOMI MM MCOTWI LCVILS AND UMNAMV COT»«H« CKfWTKNI IN PMOCMOOL CMLMBI
199
10001-
1000 r-
I
?
UjKJO
Z
S
5
UJ
O 10
cr
3
100
12 NOT EXPOSED 12 EXPOSED
EXPOSURE STATUS
3EXPOSED
10
1234567
AVERAGE HOME AIR NICOTINE (mcq/m3)
Fig. 3 (left). Distribution of al midmoming urine cotinine dalarminattona from: column 1,12 unexpoaad ciiildnn (geometric mean CCR, U7 ngfmg; SEM. 23 ng/mg);
column 2,12 exposed children consistently classified u exposed with the CCR (g*wn«ricm^nCCB,8«ng/mg;SEM. 7Bng/mg); and column 3. mr^Bxpowd children
connrtsntty classified M unsapossd by urine CCR (geometric mean CCR. 245 ng/mg; SEM, 2J ngAng).
Fig. 4 (right). Correlation between average horn* air nicotine concentrations tor two monitoring day* and average natural togailftm of urine CCR on (he mornings after
the two mooitoring days «or 15 ETS-«Kpc*ed children. The eolation hx the rsy - 3.13 (SEM. 0.40) + O315 (SEM. 0.10) -average home air nicotine.
(F - lift p - 000&)
Observations Relevant to the Use of
Urine Cotinine Determinations in
Epidemiologic Studies
The urine collection protocol allowed us
to examine whether time of urine collec-
tion during the day influenced urine coti-
nine levels. The means of logc CCR did
not differ significantly by sampling time.
Furthermore, individual children main-
tained their absolute and relative levels
of cotinine excretion throughout the two
24-h study intervals. In exposed children,
rank correlations of sequential measure-
ments ranged from 0.68 < r < 0.93; 0.008
< p < 0.03. There was no evidence that
routine evening ETS exposures in chil-
dren exposed chronically to ETS had
acute effects on urine cotinine levels. For
unexposed children, rank correlations of
sequential urine CCRs were more vari-
able, but as indicated previously the great
majority of urine specimens (48 of 51 =
94%) had CCRs less than 30 ng/mg.
We then examined the stability of
repeated urine cotinine/creatinine deter-
minations over a 1-month time period.
Urine specimens were collected while the
children were at the day care center on
each of four consecutive Thursdays. The
data of 14 exposed children are shown
in figure 5; one exposed child was absent
frequently from the Center during the 1-
month observation period. In exposed
children, absolute and relative levels of
urine cotinine were very stable over time.
Rank correlations for sequential urine
CCRs were all r > 0.88; p = 0.0001.
Discussion
Measurement of concentrations of coti-
nine in body fluids has been established
as a useful method for identifying active
smokers, persons exposed passively to en-
vironmental tobacco smoke, and neo-
nates exposed to nicotine and cotinine
in utero (1-13). Although active smokers
can be distinguished readily from pas-
sive smokers with cotinine assays, these
tests appear less accurate in distinguish-
ing persons who are routinely exposed
passively to ETS from those who are
sporadically exposed. Furthermore, our
knowledge of the usefulness of cotinine
determinations for ranking the intensity
of passive smoke exposure in populations
of individuals is fragmentary. Most in-
vestigators who have attempted to ex-
amine this question have relied on ques-
tionnaire data to provide estimates of the
intensity of ETS exposure in study popu-
lations. In the present work, we employed
direct assays of home air nicotine concen-
trations to estimate the intensity of home
nicotine exposure for study children.
The results of this study provide a firm-
er foundation for use of the CCR in urine
as a biologic marker of ETS exposure in
epidemiologic studies involving young
children. Twelve of 15 preschool children
routinely exposed passively to tobacco
in the home had urine CCRs that were
consistently > 30 ng/mg, whereas 12 of
12 children without home ETS exposure
consistently had CCRs < 30 ng/mg.
Three exposed children (20%) had urine
-------
200
1000
H0NM9MON. ROD, MONMS, WMtt. HU, HIUM. F
-------
HOMI AM MCOTWC LfVfLS AND UMNAiff COTTMNt GtCMfTlON IN HMCIIOOt CMLOMOt
201
identifying passive tobacco smoke ex-
posure in young children. Because some
exposed children have relatively low lev-
els of cotinine in urine, cotinine deter-
minations should probably be used in
conjunction with questionnaire data re-
garding ETS exposure. Data presented
here indicate that urinary CCRs are relat-
ed quantitatively to the intensity of home
tucoTinevexposure among ETS-exposed
children. This and the relative stability
of urine CCR over time support use of
the assay in epidemiologic studies of ETS
exposure in young children.
References
I. Haley NJ. Axelrad CM. Tilton KA. Validation
of self-reported smoking behavior biochemical
analyses of cotinine and thiocyanate. Am J Public
Health 1983; 73:1204-7.
2. Friedman CD. Petitti Da Bawol RD. Preva-
lence and correlates of passive smoking. Am J Pub-
lic Health 1983; 73:401-5.
3. Jarvis MJ, Tuns tall-Pedoe H. Feyerabend C.
Vesey C. Sailoojee Y. Biochemical markers of smoke
absorption and self reported exposure to passive
smoking. J Epidemic! Community Health 1984;
38:335-59.
4. GreenbergRA, Haley NJ, Etzd RA, Loda FA.
Measuring the exposure of infants to tobacco
smoke: nicotine and cotinine in urine and saliva.
N Engl J Med 1984; 310:1075-8.
5. Matsukura S, Taminato T, Kitano N,« at. Ef-
fects of environmental tobacco smoke on urinary
cotinine excretion in nonsmokers: evidence for pas-
sive smoking. N Engl J Med 1984; 311:828-32.
6. Jarvis MJ. Russell MAH. Feyerabend C, etal.
Passive exposure to tobacco smoke: saliva cotinine
concentrations in a representative population sam-
ple of non-smoking schoolchildren. Br Med J 1985;
291:927-9.
7. EtzelRA. GreenbergRA. Haley NJ. Loda FA.
Urine cotinine excretion in neonates exposed to
tobacco smoke products in utero. J Pediatr 1985;
107:146-8.
8. Sepkovic DW, Haley NJ. Biomedical applica-
tion of cotinine quantitation in smoking related
research. Am J Public Health 1985; 75:663-5.
9. PattishaUEN.StropeGL.EtzdRA.HelmsRW,
Haley NJ, Denny FW. Serum cotinine as a mea-
sure of tobacco smoke exposure in children. Am
J Dis Child 1985: 139:1101-4.
10. Luck W, Nau H. Nicotine and cotinine con-
centrations in serum and urine of infants exposed
via passive smoking or milk from smoking mothers.
J Pediatr 1985; 107:816-20.
II. Luck W, Nau H. Haitsen R, Steldinger R. Ex-
tent of nicotine and cotinine transfer to the human
fetus, placenta and aminiotic fluid of smoking
mothers. Dev Pharmacol Ther 1985; 8:384-95.
12. Giusto ED, Eckhard I. Some properties of sali-
va cotinine measurements in indicating exposure
to tobacco smoking. Am J Public Health 1986;
76:1245-6.
13. Coultas DB, Howard CA, Peake GT, Skipper
BJ, Samet JM. Salivary cotinine levels and involun-
tary tobacco smoke exposure in children and adults
in New Mexico. Am Rev Respir Dis 1987; 136:305-9.
14. Henderson FW, Collier AM. Clyde WA Jr.
Denny FW. Respiratory syncytial virus infections,
reinfections, and immunity. A prospective longitu-
dinal study in young children. N Engl J Med 1979;
300:530-4.
15. Henderson FW. Collier AM. Sanyal MA, et
al. A longitudinal study of respiratory viruses and
bacteria in the etiology of acute otitis media with
effusion. N Engl J Med 1982; 306:1377-81.
16. Pacini DL. Collier AM. Henderson FW.
Adenovimi infections and respiratory illnesses in
children in group day care. J Infect Dis 1987;
156:920-7.
17. Hammond SK.LeadererBP, Roche AC, Schen-
ker M. Collection and analysis of nicotine as a
marker for environmental tobacco smoke. At-
mospheric Environment 1987; 21:457-62.
18. Langone J J. Gjika HB. VanVunakis H. Nico-
tine and its metabolites: radkxnununoassavs for ruco-
tine and cotinine Biochemistry 1973; 12:5025-30.
-------
Original Contributions
Passive Smoking on Commercial
Airline Flights
Margaret E. Mattson, PhD; Gayle Boyd, PhD; David Byar, MD; Charles Brown. PhD;
James F. Callahan, DPA; Donald Corle, MS; Joseph W. Cullen, PhD; Janet Greenblatt, MPH;
Nancy J. Haley, PhD; S. Katharine Hammond, PhD; Joellen Lewtas, PhD; Warren Reeves
In-flight exposure to nicotine, urinary cotinine levels, and symptom self-reports
were assessed in a study of nine subjects (five passengers and four attendants) on
four routine commercial flights each of approximately four hours' duration. Urine
samples were collected for 72 hours following each flight. Exposures to nicotine
measured during the flights using personal exposure monitors were found to be
variable, with some nonsmoking areas attaining levels comparable to those in
smoking sections. Attendants assigned to work in nonsmoking areas were not
protected from smoke exposure. The type of aircraft ventilation was important in
determining the levels of in-flight nicotine exposure. The environmental tobacco
smoke levels that occurred produced measurable levels of cotinine (a major
metabolite of nicotine) in the urine of passengers and attendants. Passengers who
experienced the greatest smoke exposure had the highest levels of urinary cotin-
ine. Changes in eye and nose symptoms between the beginning and end of the
flights were significantly related both to nicotine exposure during the flight and to the
subsequent urinary excretion of cotinine. In addition, subjects' perceptions of
annoyance and smokiness in the airplane cabin were also related to in-flight
nicotine exposure and urinary excretion measures.
(JAMA 1989-^6l«67-872)
THE ADVERSE health effects on non-
smokers of passive, or involuntary,
smoking include lung cancer and respi-
ratory disease, the latter especially
among children, as well as acute irritant
effects. The scientific evidence for these
effects has been reviewed and codified
in the Surgeon General's Report on the
Health Consequences of Involuntary
Smoking published in 1986.'
Attention recently has been focused
See also p 898.
on the extent of exposure to passive
smoking experienced in various indoor
environments where smoking is al-
lowed. The National Academy of Sci-
ences reviewed the data on passive
smoking in relation to the quality of
Fromihe National Cancer Institute. Bethesda. Md (Ore
Manson.Boyd.Byar.8rown.Callahan.and Cullen and Mr
Corle): Prospect Associates. Rockville. Md (Ms Greenb-
latl); the American Health Foundation. Valhalla. NY (Or
Haley); the University of Massachusetts Medical School.
Worcester (Or Hammond): the Environmental Protection
Agency. Research Triangle Park. NC(Drl_ewtas);and Air
Canada. Montreal (Mr Reeves).
This publication does not necessarily relied EPA
policy
Repnni requests to the National Institutes ol Health.
National Cancer Institute. Division o( Cancer Prevention
and Control. 9000 Rockville Pike. Executive Plaza North.
Room 330. Belhesda. MO 20692 (Or Mattson)
indoor air environments in two separate
reports also published in 1986. The first
is a more general statement, Environ-
mental Tobacco Smoke: Measuring Ex-
posures and Assessing Health Effects,1
and the second, specifically addressing
the environment of airline flights, is The
Airliner Cabin Environment: Air
Quality and Safety* The latter re-
commended a ban on smoking in all do-
mestic commercial flights for four major
reasons: to minimize irritation, to re-
duce health risks, to reduce fire haz-
ards, and to bring air cabin quality into
line with standards for other closed
environments.
Public opinion, by smokers and non-
smokers alike, is increasingly in favor of
restrictions on smoking in public areas
and in the workplace.4-J However, there
are presently no regulatory standards
that specify limits on pollutants affect-
ing air quality in airplanes. Recent leg-
islation in the United States, enacted in
April 1988 on a trial basis, has banned
smoking on all US carrier domestic
flights of two hours or less. The ban is
due to terminate in April 1990 unless
additional congressional action occurs.'
In Canada, smoking on flights of two
hours or less was banned by the govern-
ment in December of 1987. Subsequent-
ly, both major Canadian airlines volun-
tarily went beyond the two-hour ruling
and banned smoking on all North Amer-
ican flights. One US airline has volun-
tarily banned smoking on flights of any
length within the United States, except
flights to and from Hawaii.' On Jan 1,
1988, the California legislature enacted
a law that banned smoking on all forms
of public transportation, including all
intrastate airline flights, and limited
smoking in transportation waiting ar-
eas. Reactions to the ban obtained in a
March 1988 survey of 677 passengers
and crew showed strong support among
nonsmokers (95%) and smokers (57%)
(University of San Francisco news re-
lease, April 6,1988).
The Surgeon General of the Public
Health Service requested that the
Smoking, Tobacco, and Cancer Pro-
gram of the National Cancer Institute
conduct a research study to measure
environmental tobacco smoke exposure
on typical commercial flights. This
study was undertaken (1) to measure
nicotine levels in ambient air during
flights of approximately four hours' du-
ration and urinary cotinine levels at var-
ious points during the three days after
the flights, and (2) to determine if these
exposure and excretion measurements
correlate with each other and with acute
symptoms experienced during the
flights.
METHODS
Subjects and Procedures
Nine subjects (four attendants and
five passengers) participated on each of
four flights. Based on smoking chamber
data,1 we determined that this study
design would have 77% power for find-
ing a difference in urinary cotinine ex-
cretion between subjects moderately
exposed to environmental tobacco
smoke and unexposed subjects. All sub-
jects were nonsmokers; were not regu-
larly exposed to smoke; were free of
chronic respiratory disorders such as
asthma, bronchitis, or emphysema;
JAMA. February 10, 1989-Vol 261. No. 6
Passive Smoking on Airlines — Malison et al 867
-------
Table 1.-Inflight Nicotine Exposure*
Flight
Variable
Attendants
t
2
3
4
passengers
&
6
7
a
9
Average No. of
smokers per
count (No. of
observations)
3.0 (St)
0.1 (NS)
9.8 (NS)
0.9 (S)
16.6 (S)
58.6 (S)
0.1 (NS-B)
0.6 (NS-B)
0.2 (NS-B)
*2 (9)
6.8 (S)
9.4 (NS)
0.2 (NS)
10.S (S)
0.8 (NS-8)
0.1 (NS-B)
3.5 (S)
0.3 (S)
12.6 (S)
3.9 (7)
0.7 (NS)
0.5 (S)
1.5 (S)
1.4 (NS)
5.5 (NS-B)
32.1 (S)
2.5 (NS-B)
8.3 (S)
27.5 (NS-B)
4.4 (7)
7.7 (NS)
9.5 (S)
7.5 (S)
5.4 (NS)
12.9 (S)
27.3 (NS-B)
8.4 (S)
71.0 (NS-8)
11.4 (S)
4.4 (S)
"Nicotine values from personal monitoring pump (assigned work or seating area). Values were adjusted for
sampling time during which smoking was allowed. Units are micrograms per cubic meter (|ig/m>).
fS indicates smoking section: NS. nonsmoking section; and NS-B. borderline between smoking and nonsmoking
sections.
were willing to be assigned to the smok-
ing section of aircraft; and volunteered
to participate in the study. The protocol
was approved by the human subjects
institutional review board of the Na-
tional Institutes of Health and by the
participating airline. Eight of the nine
subjects were recruited from Air Cana-
da employees.
Data collection for the entire study
took place over a 19-day period in May of
1988. The flights' schedules were as fol-
lows: flight 1, Toronto to San Francisco;
flight 2, San Francisco to Toronto; flight
3, Toronto to Vancouver; and flight 4,
Vancouver to Toronto. Seventy-two
hours elapsed between all flights except
flights 2 and 3, which were separated by
four days. Flights 1 and 2 were on B727
jets with 100% fresh air. Flights 3 and 4
were on B767 jets with 50% of the air
recirculated.
The same subjects participated in all
four flights. The five passenger subjects
(three men and two women) were
seated in the smoking section or in the
nonsmoking section bordering the
smoking section. The four attendant
subjects (three men and one woman)
were assigned to work in smoking or
nonsmoking sections of the cabin. All
subjects rotated exposure conditions
over the four flights, shown in Table 1.
Each flight had four smoking section
rows, consisting of 24 seats in flights 1
and 2 and 28 seats in flights 3 and 4.
There were 78 seats in nonsmoking in
flights 1 and 2 and 93 in 3 and 4.
Although the smoking section was fully
occupied on the first three flights and
there were only two empty seats on the
fourth flight, some passengers in the
smoking section did not smoke, eg,
study subjects and children. Logistical
considerations (ie, the airline's need for
a rapid turnaround of aircraft) pre-
vented a direct count of all cigarette
butts produced by smokers on the
flight. Smoking activity was estimated
by the field coordinator, who observed
the smoking section at intervals
throughout the flight and counted the
number of persons smoking each time.
The concentration of nicotine in the
cabin air was used as a marker or surro-
gate for the exposure to environmental
tobacco smoke. During the flights, nico-
tine was collected from the air by active
sampling as described elsewhere.' A
personal exposure monitor consisting of
a pump sampling at 3 L/min through a
cassette containing two filters was
used. The first filter collected particles
for a separate study of mutagenicity of
extractable organics to be reported
elsewhere. The second filter was
treated with sodium bisulfate to collect
the nicotine. The nicotine on the treated
filters was desorbed in solvents and ana-
lyzed by gas chromatography with ni-
trogen-selective detection. The sensi-
tivity was 0.07 jjig of nicotine per cubic
meter. Each subject wore an active
sampling system during each flight to
measure his or her actual nicotine expo-
sure ("in-flight exposure").
Exposure to cigarette smoke be-
tween flights was monitored with both a
diary, in which subjects recorded the
extent and duration of exposure outside
of the flights, and with a passive moni-
tor. For 72 hours before and after each
flight, each subject wore a small, light-
weight passive monitor that contained a
bisulfate-treated filter that collected
nicotine by diffusion. These filters were
also analyzed by gas chromatography
with nitrogen-selective detection.'"10
In addition to a preboarding baseline
urine sample, subjects collected all their
urine for each of 12 six-hour periods
after the flight. They recorded the total
volume collected for each six-hour peri-
od and took a sample from the pooled
specimen. All samples were shipped in
dry ice to the testing laboratory and
analyzed for cotinine and creatinine.
The method of cotinine analysis was ra-
dioimmunoassay, as described by Haley
et al."'u All cotinine values were nor-
malized by creatinine excretion.
Before each flight, all subjects were
asked to complete a simple question-
naire about the following symptoms:
eyes (itching, burning, dryness, teari-
ness, or increased blinking); nose (dry-
ness, itching, discharge, obstruction, or
stuffiness); dry mouth; coughing; sneez-
ing; scratchy or sore throat; and head-
ache. These same questions were asked
at the completion of each flight along
with additional questions concerning
annoyance from cigarette smoking dur-
ing the flight ("During this flight, were
you annoyed or irritated by cigarette
smoke?") and an estimate of how smoky
the flight appeared to be ("How smoky
was the area of the plane in which you
spent most of your time?").
Data Analysis
Air nicotine concentrations (in micro-
grams of nicotine per cubic meter of air)
were corrected for the sampling time
during which smoking was permitted
and the pumps activated. The lengths of
time the air sampling equipment was on
were as follows: flight 1,4.8 hours; flight
2,4.5 hours; flight 3, 4.0 hours; flight 4,
3.8 hours. Since all flights had minimal
time during which smoking was cur-
tailed, these corrections were minor.
Levels of air nicotine were classified as
"high" (>12 tJLg/m1), "moderate" (1 to 12
jjig/m1), or "low" (<1 tig/m1). Statistical
significance was assessed by the Wil-
coxon-Mann-Whitney rank test1* and
the Mantel linear trends test." Both
one- and two-tailed P values have been
used, depending on whether the direc-
tion of the statistical comparison could
be anticipated from prior knowledge.
Two types of analyses were done with
the urinary data. In the first analysis,
the relationship between air nicotine
exposure during the flight ("in-flight ex-
posure") and cotinine excretion over the
72-hour period after the flight was ex-
amined. Twenty-four-hour moving av-
erages, ie, an average of the four cotin-
ine values for a consecutive 24-hour
period, were created to smooth out vari-
ability.1* The cotinine moving average
(MA) was computed for the mean cotin-
ine values, normalized for creatinine in
units of nanograms of cotinine per milli-
gram of creatinine as follows (where
UCP indicates urine collection period
868 JAMA. February 10. 1989-Vol 261. No. 6
Passive Smoking on Airlines—Malison et al
-------
30 -
O)
c
'c
o
O
No. of Hours After Right
X High Exposure • Low Exposure
Rg 1. — Cotinine excretion over time. Twenty-four-hour moving average of ootinine excreted after flight. Value
at ( = 0 is average urinary cotinine before boarding flight. High in-flight exposure is defined as nicotine
exposure greater than median value. Units are nanograms of cotinine per milligram of creatinine. See text lor
description of methods.
5.00
3.75
£
5 2.50
O
81
1.25
0.00
+ Subjects Not Reexposed
o Reexposed Subjects
0.00 1.25 2.50 3.75 5.00
Log Nicotine
Rg 2.—In-flight nicotine exposure and cotinine ex-
cretion at 12 hours after flight. Values are natural
logarithms of air nicotine (Ln (nicotine + 1|) and
creatinine-normalized cotinine (Ln [cotinine + 1]).
Regression line through data points for subjects not
reexposed between flights, correlation coefficient =
.74. P = . 0003.
for the stated interval): MA (at 12
hours) = UCP (0 to 6 hours) 4- UCP (6
to 12 hours) + UCP (12 to 18 hours) +
UCP (18 to 24 hours) divided by four,
plotted at t = 12, the midpoint of that24-
hour interval. MA (at 18 hours) = UCP
(6 to 12 hours) -t- UCP (12 to 18 hours) +
UCP (18 to 24 hours) + UCP (24 to 30
hours) divided by four, and similarly for
the remaining moving averages. Com-
puted in this way, the 12 postflight urine
collection periods yielded a single mov-
ing average curve consisting of nine
points. Measured levels of in-flight air
nicotine were partitioned into those
above and below the median value for all
flights. Twenty-four-hour moving av-
erages of cotinine levels were plotted
against time separately for subjects re-
ceiving high (above median) and low
(below median) in-flight exposure.
In the second analysis, the dose-
response relationship between nicotine
exposure and cotinine excretion at 12
hours was examined by scatter plots of
in-flight exposure vs cotinine excretion
using log-transformed values for all nic-
otine and cotinine measures. The total
number of urinary points available for
t = 12 hours was 29 (instead of 36), due to
missing data.
Linear plots of both raw and creati-
nine-normalized measures of urinary
cotinine and plots of the log-trans-
formed urinary cotinine data (original
units were nanograms per milligram of
creatinine) done early in the analysis
revealed marked variation for some in-
dividuals. Some began a flight with un-
expectedly high baseline values. For
some subjects, peak excretion of cotin-
ine occurred at irregular intervals, with
some subjects peaking early, others
late, and some snowing multiple peaks.
The most likely explanation for the pat-
terns observed is that some subjects
were reexposed while not on a flight.
The possibility of reexposure was an-
alyzed by examination of the interflight
nicotine badge worn between flights
and the exposure diaries. Subjects
whose passive nicotine badge values in-
dicated exposure of 0.13 u,g of nicotine
or greater during the 72-hour between-
flight interval were considered to have
received at least moderately high re-
exposure. Data from the diaries docu-
menting the day and time of exposures
were used in combination with known
half-life values for cotinine to determine
which collection intervals were af-
fected. The mean baseline cotinine level
(nanograms of cotinine per milligram of
creatinine) of subjects with interflight
badge values of 0.13 u,g of nicotine or
greater was 34.3, and 12.1 for those
with badge values of less than 0.13 u-g of
nicotine.
Urine samples collected after the fol-
lowing flights were considered unsuit-
able for analysis due to occurrence of
interflight exposures to tobacco smoke:
all data for attendant subjects 1 and 2,
flights 3 and 4 for attendant subject 3,
flight 1 for attendant subject 4, flight 4
for passenger subject 5, and flight 3 for
passenger subject 8. These reexposed
subjects were excluded from some ana-
lyses (Fig 1).
Both questions about symptoms and
the questions concerning annoyance
and smoke levels experienced during
the flight were recorded on a six-point
scale from zero to five. Differences in
the symptom scores before and after the
flight were calculated and categorized
into three groups of roughly equal size,
coded as 0, 1, and 2. For eye and nose
score changes, - 2 to 0 was classified as
none or mild, 1 to 2 as moderate, and 3 to
4 as marked. For "annoyed" and
"smoky," the corresponding categories
were 0 to 2,3, and 4 to 5. A logarithmic
transform for both the nicotine values
from the personal sampling pumps and
the urinary cotinine values was used.
These coded symptom score changes
were related to the nicotine and cotinine
values by linear least squares regres-
sion with the continuous measurements
(air nicotine or urinary cotinine) treated
as the response variables. The analysis
for the self-reported symptoms used all
subjects and was based on those urinary
cotinine values obtained at 12 hours
after the flight, not moving averages.
Twelve hours was chosen as an appro-
priate point where there was minimal
contamination from reexposure to
smoke experienced later during the
data collection period.
RESULTS
In-flight Nicotine Exposure
Subject placement and exposure con-
ditions for the four flights along with in-
JAMA. February 10. 1989-Vol 261. No. 6
Passive Smoking on Airlines — Mattson et ai 869
-------
(light nicotine measurements of air con-
centrations from the active personal
exposure monitors worn in-flight are
shown in Table 1. Measurable exposure
to environmental tobacco smoke oc-
curred for all subjects on all four flights.
Review of the in-flight logs kept by the
study field coordinator of counts of
smokers done at least hourly indicates a
maximum of eight smokers at any one
time during the flight, with the average
number of smokers per count ranging
from 3.9 to 4.4 (Table 1). Table 2 also
shows the frequency distribution of the
in-flight nicotine readings and levels of
statistical significance for various group
comparisons.
The personal exposure monitors of
subjects on flights with 100% fresh air (1
and 2) recorded less exposure than
those on the two flights with 50% fresh
air and 50% recirculating air (3 and 4), a
statistically significant difference by
both tests, with two-tailed P values. On
flights 1 and 2, all passengers seated in
the nonsmoking section (five samples)
were exposed to less than 1 u.g/ms of
nicotine with a median exposure of 0.2
fig/m', while all passengers seated in
the nonsmoking section of flights 3 and 4
(five samples) were exposed to more
than 2.5 M-g/m1 of nicotine, with a medi-
an exposure of 27 (ig/m*. Flight 4 from
Vancouver to Toronto had the highest
levels of air nicotine, with all exposed to
greater than 5 jig/m*.
Passengers in the nonsmoking border
section experienced variability in their
exposure, with some attaining expo-
sures comparable to or higher than
those in the smoking section. The three
highest nonsmoking border readings
(27.3,27.5, and 71 ng/rn') were obtained
on the flights with 50% recirculating and
50% fresh air. There was also a differ-
ence in spatial configuration of the seat-
ing between the two types of aircraft.
All nonsmoking border readings on
flights with recirculated air were great-
er than all nonsmoking border values on
flights with 100% fresh air. The differ-
ences between exposures in the non-
smoking border and smoking sections
were not statistically significant. No
passenger subjects in the study were
placed in the center of the nonsmoking
section far from the border with smok-
ing because of the expectation that ex-
posures there would be very low.
Exposure among attendants was not
statistically different from that of pas-
sengers, although none of the eight high
exposures observed on the flights oc-
curred among attendants (Table 2). Ex-
posure among attendants assigned to
work in the smoking section was not
different from that among those work-
ing in the nonsmoking section.
Table 2.—Frequency Distribution of Air Nicotine Readings and Tests of Significance*
Pt
Moderate
Significantly "
Low Different? WMW
MLT
Flights 1 and 2
Flights 3 and 4
Attendants
Passengers
Passengers in smoking section
Passengers in nonsmoking border
Attendants in smoking section
Attendants in nonsmoking section
3
5
0
8
S
3
0
0
6
11
11
6
4
2
6
5
1}
I)
s}
1}
Yes
No
No
No
.054
.15
.075
.26
.040*
.093*
.058§
.30§
•All readings are from personal pump monitors worn by all subjects during the (lights. Units are mfcrograms per
cubic meter (ng/m1). High • > 12 |ig/m'; moderate » 1 -12 iig/m': and low - < 1 mj/m1.
\P values are from the comparison between the members of each of the pairs indicated. Listed are the P values
for the Wilcoxon-Mann-Whitney test (WMW) and Mantel linear trends test (MLT).
{Two-tailed P value.
§One-tailed P value.
3.5
3.0 -
2.5-
'o
.9
f
-J
2.0
1.5
1.0
0.5
0
r 1S
T
-
Nose Changes
P = .031
Eye Changes
P = .016
Annoyed
P = .002
Smoky
P = .0003
Fig 3.—Relationship of in-flight levels ot air nicotine measured by personal monitoring pump to symptoms and
perceptions during flight. Units are micrograms of nicotine per cubic meter (log transformed: Ln [nicotine +•
1]). Numbers of subjects in each category are shown above bars representing 1 SE. Black bars indicate
marked changes; striped bars, moderate changes; and white bars, no or mild changes. P values are one-
tailed. See text for description of scoring system.
Urinary Cotinine Excretion
Postflight urinary cotinine excretion
over time in subjects without apprecia-
ble reexposure between flights is shown
in Fig 1. A median split was performed
on the in-flight nicotine exposure values
to divide these subjects into "high" and
"low" nicotine exposure groups. The
median nicotine exposure value was 5.5
jig/m1. The urinary cotinine level for
each exposure group was plotted
against time using a moving 24-hour
average. In the high-exposure group,
cotinine excretion increased quickly
from preflight levels, rose to a peak val-
ue, and monotonically decayed back to a
value slightly above the preflight levels
by 72 hours. Similar plots of cotinine
excretion over time for only the reex-
posed subjects showed highly elevated
baselines, irregular excretion time
courses, and no relationship to in-flight
nicotine exposure.
Scatter plots of all data using differ-
ent symbols to distinguish reexposed
subjects were done to investigate dose-
response relationships (Fig 2). Urinary
cotinine excretion (creatinine correct-
ed) at 12 hours after the flight for
subjects not reexposed showed a clear
correlation with nicotine exposure re-
ceived during the flight. The correlation
coefficient was .74 with P = .0003.
There was no significant correlation be-
tween the cotinine and nicotine values
for the reexposed subjects.
Self-reported Symptoms
Dry mouth, coughing, sneezing,
scratchy or sore throat, and headache
were not significantly related to nico-
tine exposure. On the other hand,
870 JAMA. February 10. 1989-Vol 261. No. 6
Passive Smoking on Airlines — Mattson et al
-------
Table 3.—Air Nicotine Levels (vJ3/m>) in Various
Indoor Environments*
Nose Change
P=.004
Eye Change
P=.O22
Rg 4.—Relationship of urinary cotinine excretion at 12 hours after flight to symptoms and perceptions during
(light. Units are nanograms of cotinine per milligram of creaUrune (log transformed: Ln (counine +• 1|).
Numbers of subjects in each category are shown above bars representing 1 SE. Black bars indicate marked
changes; striped bars, moderate changes; and white bars, no or mild changes. P values are one-tailed. See
text for description of scoring system.
changes in eye symptoms, nose symp-
toms, annoyance with smoking, and
perception of a smoky environment
were all related significantly to the nico-
tine exposures. These relationships are
illustrated in Fig 3.
As with the nicotine analysis, no sig-
nificant relationships were seen be-
tween the 12-hour log-transformed uri-
nary cotinine values and dry mouth,
coughing, sneezing, scratchy or sore
throat, or headache. Like the nicotine
data just described, these urinary coti-
nine data also showed significant rela-
tionships to eye and nose symptom
changes as well as to the perception of
the smokiness in the aircraft cabin (Fig
4). The relationship for the annoyance
index was not statistically significant,
but the observed changes were in the
expected direction. Some evidence sug-
gests that changes in eye symptoms
may have been more marked in the two
subjects wearing contact lenses.
COMMENT
Several other studies have measured
air nicotine levels in indoor environ'
ments and report that a wide variety of
factors act in combination to produce a
particular "microenvironment." Differ-
ences in methodology, such as type of
monitoring devices and assays, ventila-
tion, selection of sampling times and
number of smokers present, and their
location relative to sampling devices, all
limit precise comparisons across these
studies. However, the levels of nicotine
found in this study are comparable to
the measurements reported in other
studies shown in Table 3 and support
the conclusion that air nicotine levels in
the nonsmoking areas that border the
smoking area may be at least as high as
in similar indoor environments fre-
quented by smokers.
A small number of studies have been
published in the scientific literature
that assess environmental tobacco
smoke specifically on board aircraft.
These studies have assessed exposure
by measuring concentrations of carbon
monoxide,*" particulates," and nico-
tine,1*" with one also assessing physio-
logical absorption (ie, blood nicotine) re-
sulting from exposure," The results
from the two studies employing in-flight
nicotine measurements are summarized
in Table 3.
In this study, air levels of nicotine
were highly variable, with some non-
smoking areas attaining levels greater
than those in some smoking sections.
Seating section was a less important
predictor of actual nicotine exposure.
This bears out travelers' anecdotal ob-
servations that the section in which one
sits is often not as important in deter-
mining exposure to smoke as is the envi-
ronment generated by one's neighbors.
This environment is determined by sev-
eral factors, including the number of
cigarettes smoked by neighbors, seat-
ing configuration, air flow patterns, and
the percentage of recirculated air. In
these flights, the average number
smoking at any one time was only about
four and was never observed to be
greater than eight. This may represent
relatively low exposure compared with
flights with many more smokers.
The type of ventilation appeared to be
Environment
Airp4anest
Airplanes!
Airplanes§
Trains|
Offices!
Offices^
Offices'.
Pubs, coffee
shops!
Cafetenas§
Lobbies.
waiting
f 00(71 Sj
Open
ventilation
Closed
ventilation
Room]
Submarine!
(NS)
(NSa)
(S)
(Sa)
(NS)
(NS)
(NS)
(S)
(S)
(S)
Mean
14 (median - 2)
4 (median -3)
17 (median -12)
5 (median -S)
7 (geom-3)
13 (geom-8)
8 (geom-4)
11 (geom-7)
30 (geom-7)
26 (geom-22)
IS
09
19
5
to
26
37
7
1
3
65
1010
500
15-35
Range
01-71
0.1-10
0.3-59
0.7-1 1
NO-24
NO-40
NO-17
04-42
NCM12
NO-77
6-29
0.7-50
14 (peak)
9-31
3-28
25-52
12-42
•NS indicates nonsmoking section; S, smoking sec-
tion. NSa. nonsmoking section (attendants); Sa. smok-
ing section (attendants); and NO. not detectable.
tThis report.
{Otdaker nonsmoking values were measured at or
near the border with smoking in three different types of
planes. Arithmetic means were presented originally":
geometric means (geom) were subsequently pub-
Sshed."
fMuramatsu el al."
{Summary across studies taken from National Acad-
emy of Sciences review of studies reported in the
literature between 1957 and 1980.' The number of
studies cited in the report are as follows: trains, three:
offices, one: pubs etc. three: lobbies etc. four: automo-
biles, one: rooms, one; and submarines, one.
IHammond el al.'
an important factor in the levels of air
nicotine attained. Planes with 100%
fresh air (flights 1 and 2) had significant-
ly less ambient nicotine than those with
50% fresh air and 50% recirculating
(flights 3 and 4). Recalculation systems
significantly increase the fuel economy
and so are an integral part of the design
of some newer aircraft. Passengers and
attendants may be exposed to higher
levels of environmental tobacco smoke
in the next decade as the percentage of
seat-hours on airplanes with recircula-
tion systems increases from 30% in 1985
to an estimated 40% in 1990.'
Attendants are not confined to the
section in which they are assigned to
work and move through all areas of the
plane. Although the attendants were
assigned either to the smoking or non-
smoking sections, in fact there were
only about four rows of smokers on each
of the four flights and the attendants
worked in both smoking and nonsmok-
ing areas when they were assigned to
the smoking area. Attendants assigned
to nonsmoking areas may have received
exposure from the first-class smoking
section and from passing through the
JAMA. Feoruary 10. 1989-Vo* 261. No 6
Passive Smoking on Airlines - Malison el al 871
-------
smoking section in coach. This may ex-
plain why there was no significant dif-
ference in exposure between attendants
assigned to the two sections.
Although the levels of exposure of
attendants measured by the personal
exposure pumps were less than those of
passengers (although not statistically
significant), the amount of nicotine and
other cigarette smoke products actually
inhaled and ingested may have been
greater due to the greater physical ac-
tivity and increased respiratory rate of
the attendants. If true, this, together
with the cumulative exposure attained
from long periods of flight duty, could
result in greater total exposure over the
course of an attendant's career.
The levels of environmental tobacco
smoke that occurred during the four-
hour flights led to increased levels of
cotinine (a major metabolite of nicotine)
in the urine of both passengers and at-
tendants. Subjects who experienced the
greatest in-flight nicotine exposure
generally had the highest levels of uri-
nary cotinine and continued to excrete
cotinine for 72 hours after the flight.
The shape and time course of the decay
pattern are consistent with a first-order
pharmacokinetic decay process follow-
ing an initial exposure to nicotine. The
peak level of cotinine excreted is related
to the dosage of nicotine received over
the range of exposures encountered.
Reports on dose-response data under
conditions of environmental tobacco
smoke exposure are sparse, especially
for the nicotine concentration range
typically encountered by nonsmokers
under free-living conditions. This analy-
sis provides estimates of the response to
a bolus of environmental tobacco
smoke, delivered over a four-hour peri-
od, shown by a subsequent increase in
urinary cotinine excretion over time
synchronized across subjects. This
study expands upon previous studies
employing single-point estimates of co-
tinine or self-reported smoke exposure
levels"* and provides information on
the shape of the excretion curve, delay
to peak, amplitude to the peak, approxi-
mate functional form, and decay time of
cotinine excretion after environmental
tobacco smoke exposure.
Changes between the beginning and
end of the flights in eye and nose symp-
toms indicative of acute irritation are
related both to a measure of in-flight
nicotine exposure and to the later uri-
nary excretion of cotinine. In addition,
perceptions of annoyance and smoki-
ness in the airplane cabin were likewise
related to the in-flight nicotine expo-
sure and urinary cotinine excretion
measures. Experimental studies under
controlled conditions indicate that in
smoky environments, eye, nose, and
throat symptoms gradually increase
over time with the duration of exposure
even when smoke concentrations re-
main constant. Annoyance tends to rise
quickly as soon as exposure begins and
then remain constant over time." The
irritant effects of cigarette smoke arc
reflected in the numerous complaints
about smoky conditions by attendants
and passengers alike in records com-
piled by the Association of Flight Atten-
dants and in government and industry
surveys.""70 In these surveys, 60% of
nonsmoking passengers and 15% of
smokers reported being annoyed by in-
flight tobacco smoke.14 Ninety-five per-
cent of cabin attendants reported irrita-
tion and annoyance," a with 69% in one
study" perceiving smoky air to be a
more serious concern than other work
environment conditions such as tem-
perature, odors, dust levels, and noise.
Taken together, data from this study
on in-flight nicotine exposure, subse-
quent cotinine excretion, and acute
symptoms demonstrate that total sepa-
ration of smoking and nonsmoking sec-
tions was not achieved on the flights
studied. The exposures experienced by
passengers and attendants are compa-
rable to those in other closed environ-
ments where smoking is allowed and
represent another contributor to the cu-
mulative health risk, acute irritation,
and annoyance that nonsmoking indi-
viduals receive from passive smoking.
The urine analyses were supported by National
Cancer Institute grant CA 32617-05 to the Ameri-
can Healtb Foundation.
The contributions of the following people to the
study are gratefully acknowledged by the authors:
Caryn Axdrad. MS; Neil Benowitz, MD; Neil Colli-
shaw, MA; John Fitzgerald; Clair Harvey; Thomas
Manuccia; Lorraine Poirier; Byron Rogers; Daniel
W. Sepkovic, PhD; Donald Shopland; Rachel Ten-
nant, RN; Debra Walsh, MS; Ronald Williams; and
Coyla Wosltie. In addition, we are most apprecia-
tive of the time and effort given by the nine partici-
pant volunteers, and Ms Vanessa Hooker provided
expert manuscript preparation assistance.
R«ta<«nc**
1. The Health Con*tqu£net* of Involuntary Smok-
ing: A Report of the Surgeon General. US Dept of
Health and Human Services, 1986.
2. Environmental Tobacco Smoke— Measuring
Exposures and Assessing Health Effect*. Wash-
ington, DC, National Academy Press, 1386.
3. The Airliner Cabin Environment- AIT Quality
and Safety. Washington, DC. National Academy
Press, 1986.
4. Adult Useo/Tobacco Survey. Office on Smoking
and Health, Centers for Disease Control, 1986.
5. Gallup Organization Survey of Attitude* To-
wardt Smoking, conducted for the American Lung
Association. Princeton, NJ, The Gallup Organiza-
tion. Inc. 1985.
6. Tobacco-Free Young America Reporter 1988;
7. Biber A, Scherer G, Hoepfoer I, et al: Determi-
nation of nicotine and cotinine in human serum and
urine -an interiaboratory study. Read before the
Experimental Toxicology Symposium on Passive
Smoking, Essen, Federal Republic of Germany.
Oct 23-25,1986.
8. Hammond SK. [>eadcrer BP, Roche AC. et al:
Collection and analysis of nUroune as a marker for
environmental tobacco smoke. Mmnrplienc Envi-
ron 1987:21:457.462.
9. Hammond SK, Coghlin J. Leadcrer BP. et al:
Field study of passive smoking exposure with pas-
sive sampler, in Indoor Air '«7. Proceeding* of the
Fourth International Conference on Indoor Air
Quality and Climate: Environmental Tobacco
Smoke. Muliuomponeni Sludie*. Radon, Sick
Building*. Odour* and Irritant*. Hyj>er-reactari-
ties and AUergie*. West Berlin, Institute for Wa-
ter, Soil and Air Hygiene, 1987, vol 2, pp8-12.
10. Hammond SK, Leaderer BP: A diffusion moni-
tor to measure exposure to passive smoking. Envi-
ron Sci Technol 198721:494-497.
11. Haley NJ, Axelrad CM, Tilton KA: Validation
of self-reported smoking behavior Biochemical
analysis of cotinine and thiocyanate. Am J Public
Health 1983,73:1204-1207.
12. Sepkovic DW. Haley NJ: Biomedical applica-
tion of cotinine quantification in smoking-related
research. Am J Public Health 1985^75:663-664.
13. Uindgren BW: Statistical Theory. New York,
Macmillan Publishing Co Inc. 1966, pp 325-354
14. Mantel N: Chi-«quare tests with no degree of
freedom: Extensions of the Mantel-Haenzel proce-
dure. J Am Statiftic* Attoc 1963^8*90-700.
IS. Bower-man BL, O'Connell RT: Time Seria
Forecasting, ed 2. Boston, Duxbury Press. 1987, p
239.
16. Health Aspect* of Smoking in Transportation
Aircraft. US Dept of Transportation and US Dept
of Health, Education, and Welfare. December
1971.
17. Foliart D. Benowitz NL, Becker CE: Passive
absorption of nicotine in airline flight attendants. N
EngU Med 1983:308:1105.
18. Oldaker GB, Conrad FW: Estimation of the
effect of environmental tobacco smoke (ETS) on air
quality within aircraft cabins. Read before the In-
ternational Experimental Toxicology Symposium
on Passive Smoking, Essen, Federal Republic of
Germany, Oct 23-25.1986.
19. Oldaker GB, Conrad FW: Estimation of the
effect of environmental tobacco smoke (ETS) on air
quality within aircraft cabins of commercial air-
craft. Environ Sci Technol 198721:994-999.
20. Muramatsu M. Umemura S. Okada T. et al:
Estimation of personal exposure to ambient nico-
tine in daily environment. Environ Re* 1984;35:
218-227.
21. Wall MA, Johnson J, Jacob P. et al: Cotinine in
the serum, saliva, and urine of nonsmokers, passive
smokers, and active smokers. Am J Public Health
1988;78:669-701.
22. Jarvis MJ, Russell MAH, Benowitz NL, et al:
Elimination of cotinine from body fluids: Implica-
tions for noninvasjve measurement of tobacco
smoke exposure. Am J Public Health 1988;78:696-
698.
23. Jarvis MJ, TunstaH-Podoe H. Feyerabend D,
et al: Biochemical markers of smoke absorption and
self-reported exposure to passive smoking. J Ejn
denial Community Health 1984:38:335-339.
24. Hoffman D, Haley NJ, Adams JD. et al: Tobac-
co sidestream smoke: Uptake by nonsmokers. Prev
Med 1984;13:608-617.
25. Sepkovic DW, Haley NJ, Hoffman DW: Elimi-
nation from the body of tobacco products by smok-
ers and passive smokers. JAM A 1986^56:863
26. Weber A: Annoyance and irritation by passive
smoking. Prev Med 1984;13:618-625.
27. Eng WG: Survey on eye comfort in aircraft: 1.
Flight attendants. Aviation Space Eniiron Med
1979;5fr40l-404.
28. Eng WG: Survey on eye comfort in aircraft: (I.
Use of ophthalmic solutions. Aviation Space Entn
ran Med 1979.50:1166-1169.
29. OstbergO. Mills-Orring R: Cabin Attendant*
Working Environment—A Questionnaire Study,
technical report 1980:74 T. Lulea. Sweden. Dept of
Human Work Sciences. University of Lul<-». 1980
872 JAMA. February 10. 1989-Vol 261. No 6
Passive Smoking on Airlines — Walloon v ±\
------- |