Cotinine as a Biomarker of Environmental Tobacco Smoke Exposure

Transcription

1 Epidemiologic Reviews Copyright 1996 by The Johns Hopkins University School of Hygiene and Public Health All rights reserved Vol. 18, No. 2 Printed in U.S.A. Cotinine as a Biomarker of Environmental Tobacco Smoke Exposure Neal L. Benowitz INTRODUCTION A biomarker is desirable in quantitating systemic exposure both in smokers and nonsmokers to constituents of tobacco smoke. Self-report measures in smokers, such as cigarettes smoked per day, are highly imprecise owing to individual differences in how cigarettes are smoked, with ranges of nicotine intake per cigarette from 0.3 to 3.0 mg (1, 2). Self-report measures, such as hours per day exposed to environmental tobacco smoke (ETS) by nonsmokers, are also likely to be imprecise indicators of intake of tobacco smoke owing to variations in the number of cigarettes smoked, proximity of nonsmokers to smokers, room ventilation and other environmental characteristics, as well as individual differences in sensitivity to and/or concern about adverse effects of ETS. The optimal assessment of exposure to tobacco smoke would be by analysis of the concentrations of a component of smoke in the body fluids of an exposed individual i.e., a biologic marker or biomarker. There are two broad questions that need to be considered in assessing the validity of a biomarker of tobacco smoke exposure. The first is how well does the concentration of a marker chemical in the air reflect exposure to toxic constituents of smoke that are of concern? The second is how well does a concentration of a particular chemical in a biologic fluid reflect an individual's intake of that chemical (or a related chemical) from tobacco smoke? The National Research Council (3) has proposed criteria for a valid marker of ETS in the air as follows: The marker 1) should be unique or nearly unique for ETS so that other sources are minor in comparison; 2) should be easily detectable; 3) should be emitted at similar rates for a variety of tobacco products; and 4) Received for publication February 7, 1995, and accepted for publication July 16, Abbreviations: COT, cotinine; ETS, environmental tobacco smoke; NIC, nicotine; RSP, respiratory suspended particles. From the Division of Clinical Pharmacology and Experimental Therapeutics, University of California, San Francisco, San Francisco, CA. Reprint requests to Dr. Neal L. Benowitz, San Francisco General Hospital, Building 30, Room 3316, 1001 Potrero Avenue, San Francisco, CA should have a fairly constant ratio to other ETS components of interest under a range of environmental conditions encountered. Furthermore, the validity of a biomarker depends on the accuracy of the biologic fluid measurement in quantitating the intake of the marker chemical, which in turn may be influenced by individual differences in rates or patterns of metabolism or excretion, the presence of other sources (such as diet) of the chemical, and the sensitivity and specificity of the analytical methods used to measure the chemical. Cotinine, the major proximate metabolite of nicotine, has been widely used as a biomarker of tobacco exposure (4, 5). Plasma cotinine concentrations correlate better to various measures of biologic effects of cigarette smoking than does self-reported cigarettes per day (6, 7). Cotinine concentrations in plasma, urine, and saliva of nonsmokers have been used in assessing population exposure to ETS for the purpose of developing risk estimates for lung cancer related to ETS exposure (8, 9). Based on cotinine measurements, the prevalence of significant ETS exposure in control (reportedly unexposed) groups has been estimated, which, in turn, has been used to adjust upwards lung cancer risk estimates for comparison to truly unexposed controls (3, 10, 11). The validity of the use of cotinine as a biomarker for ETS exposure has recently been questioned (12, 13). Concerns include: 1) The concentration of nicotine in the air is not a good marker of other constituents of ETS because the ratio of nicotine to other ETS components is highly variable and depends on such factors within a space as surfaces, ventilation rate, sampling duration, time since smoking, and air distribution patterns; 2) the ratio of nicotine emission to respirable suspended particles (RSP) emission is not constant across a wide range of cigarettes; 3) exposure to nicotine vapor in the absence of other ETS components can occur; 4) there is no standard method for determining nicotine or its metabolites in biologic fluids; 5) interindividual differences in rates and patterns of nicotine and cotinine metabolism render the use of nicotine or cotinine as biomarkers of limited utility; and 6) dietary nicotine exposure may confound low- 188

2 Validity of Cotinine as ETS Marker 189 level determinations of nicotine and cotinine in biologic fluids. This review will provide an overview on the suitability of concentrations of nicotine in the air as a marker of ETS, and will present the pharmacokinetic basis for the quantitative use of cotinine levels in various biologic fluids to estimate daily intake of nicotine from ETS. I will conclude that potential problems in the use of cotinine are of limited magnitude and that cotinine is a valid quantitative predictor of the level of ETS exposure for population or epidemiology studies. EXPOSURE TO NICOTINE FROM ETS Nicotine is a chemical found in all tobacco products. The tobacco in manufactured cigarettes contains between 6 and 12 mg of nicotine (14). On average, a cigarette smoker absorbs into the body about 1 mg of nicotine per cigarette smoked (2, 15). The average intake of 1 mg is generally independent of the brand or nominal Federal Trade Commission nicotine rating. The latter rating is determined by standardized smoking machine tests. However, smokers smoke their cigarettes differently than do the machines and can adjust their smoking behavior to take in the desired dose of nicotine, even from the lowest nicotine-rated cigarettes (2, 4, 16). Seventy-five percent or more of the nicotine that is emitted from a cigarette is emitted into the air as sidestream smoke, which contributes substantially to ETS (3, 10, 11). The amount of nicotine in sidestream smoke, when normalized for the generation of tar, is similar for different brands of cigarettes, independent of nominal nicotine yield (10, 11, 17). For example, Rickert et al. (17) found an average tar to nicotine ratio in the sidestream smoke of 15 different brands of cigarettes of 5.9, with a relatively low degree of variability (coefficient of variation, 12.2 percent). They found no difference in the sidestream smoke tar to nicotine ratio comparing ventilated versus nonventilated cigarettes. As sidestream smoke is not identical to ETS, it is desirable to examine the composition of ETS generated by different brands of cigarettes. Leaderer and Hammond (18) have confirmed in chamber studies that the air RSP/nicotine ratio generated by smokers smoking 10 different brands of US cigarettes with widely differing machine-determined yields is similar. The average RSP/nicotine ratio was 14.1 (coefficient of variation, 13.4 percent). The ratio was similar for filtered and nonfiltered cigarettes. It may be argued that sampling a relatively small number of available cigarette brands is insufficient for extrapolation to cigarettes in general. While the number of brands that have been tested is small, they represent the brands most commonly consumed in the United States and Canada. The results of the studies of Rickert et al. (17) and Leaderer and Hammond (18) are consistent with one another, and there is little reason to believe that the composition of sidestream smoke generated by other currently marketed cigarettes (which are of similar nicotine content) should be different. However, if in the future cigarettes are produced that contain less nicotine but generate similar amounts of RSP and other combustion products as currently marketed cigarettes, as has been proposed as a way to make cigarettes nonaddictive (19), then ETS nicotine would be expected to underestimate exposure to other tobacco combustion products. In mainstream smoke that is, what is taken in by the smoker nicotine is contained in particles (composed of tar, water, and other nicotine-like alkaloids). In ETS, most of the nicotine leaves the paniculate phase and becomes part of the gaseous or vapor phase (18, 20, 21). Nicotine in ETS is breathed into the nose and throat and is inhaled into the lungs by nonsmokers. Nicotine is extremely soluble in water and is highly extracted from ETS within the respiratory tree (22). Levels of nicotine and other chemicals in ETS decay at different rates over time so that the ratio of nicotine to other constituents of ETS may differ at different points in time after generation of the ETS (23-25). Another source of variability in the RSP/nicotine ratio is due to a background level of particles arising from sources other than ETS. Thus, when nicotine and particle levels from ETS decline to low levels, the background particle concentrations substantially influence the ratio of RSP/nicotine. At low concentrations of ETS, this ratio becomes very large. For this reason, measuring the slope of the regression line between air/nicotine concentration and particle concentration is the best way to assess the degree of correlation between air RSP and nicotine (18). In addition, when air samples are collected over the time interval of a typical human exposure, that is, over hours or days, RSP to nicotine ratios are much less variable compared with spot or brief measurements. Leaderer and Hammond (18) have shown an RSP/nicotine slope of 9.8 for 47 home-air samples that were sampled over several days. The correlation coefficient between RSP and nicotine was 0.8. Similar ratios for RSP/nicotine have been reported in workplace samples by Miesner et al. (26). Thus, time-averaged ratios of nicotine to RSP, and presumably other ETS constituents, are relatively consistent (9). When a person is exposed to ETS over time, the intake of nicotine reflects exposure to other constituents of ETS.

3 190 Benowitz PHARMACOKINETIC BASIS FOR THE USE OF COTININE AS A QUANTITATIVE BIOMARKER OF DAILY NICOTINE INTAKE Absorption and metabolism of nicotine When nicotine is taken into the body through the lungs, it enters the bloodstream and is circulated to various body organs, including the liver and kidneys. The liver converts nicotine to several metabolites. A small percentage, usually 5-10 percent, of nicotine is excreted unchanged into the urine (27). The metabolic fate of nicotine in the body is shown in figure 1 (27). The proximate metabolite to which most nicotine is converted is cotinine. On average, percent of nicotine is converted to cotinine (27, 15). Cotinine itself is excreted in the urine to a small degree (10-15 percent). The remainder of the cotinine is converted to other metabolites, particularly cotinine glucuronide, trans-3'-hydroxycotinine and trans-3'- hydroxycotinine glucuronide. Levels of cotinine in saliva and blood are highly correlated with saliva to blood ratios of (correlation coefficients = 0.82, 0.95) (28, 29). Levels of cotinine in the urine and blood are also highly correlated, with the typical urine to blood ratio of 5 (correlation coefficient = 0.81) (28). As will be discussed later, the absorbed dose of nicotine is best indicated by the concentration of cotinine in the blood. However, because the values in various biologic fluids are highly correlated, the level of cotinine in the blood can be estimated reasonably well by measuring cotinine levels in saliva or urine. Nicotine and cotinine can also be measured in hair, which provides an index of cumulative exposure over time (30, 31). Pharmacokinetic terminology Pharmacokinetics refers to how the body handles a drug. Some mathematical terms have been developed to quantitatively describe the processes by which the body handles a drug. These terms, as well as the equations presented below, are well described in standard textbooks (32). These terms include half-life, volume of distribution, and clearance. Half-life (VA) refers to the time it takes for the body to eliminate 50 percent of a drug from the body. Volume of distribution (Vd, generally expressed in liters) refers to the extent to which a drug is taken up by various body organs in comparison with the concentration in the bloodstream. Clearance (CL, generally expressed in ml/min or liters/hour) refers to how much blood is cleared of a drug by eliminating organs (such as the liver and kidneys) per unit of time. The rate of elimination of a drug from the body ( ), that is, the amount of drug eliminated per unit of time, can be calculated as the product of the clearance and the concentration of drug in the blood (CB) as follows: E = CL X CB. (1) Steady state refers to a situation in which the concentration of a drug in the blood (reflecting the amount of a drug in the body) is stable, because the rate of absorption or intake of a drug (D) is equal to the rate NICOTINE-1-N-OXIDE NORNICOTINE NICOTINE GLUCURONIDE COTININE 13.0% TRANS HYDROXYCOTININE HYDROXYCOTININE COTININE GLUCURONIDE COTININE-N-OXIDE NORCOTININE 12.6% 2.4% TRANS HYDROXYCOTININE GLUCURONIDE FIGURE 1. Quantitative scheme of nicotine metabolism. Circled compounds indicate excretion in urine and associated numbers indicate percent of systemic dose of nicotine. (From Benowitz et al. (27). Reprinted with the permission of the publisher and authors.) 7.4%

4 Validity of Cotinine as ETS Marker 191 at which the drug is being eliminated. Thus, at steady state: D = E = CL X C Bll (2) where C B refers to the concentration of drug in the blood at steady state. Rearranging this equation, it can be seen that at steady state the concentration of a drug in the body is determined solely by the dosing rate and the clearance of the drug: D clr (3) When a drug is dosed (or absorbed) at a constant rate, steady state will develop in about five half-lives. Pharmacokinetics of nicotine and cotinine in humans The pharmacokinetics of nicotine and cotinine have been extensively studied in adult humans in experiments in which known amounts of nicotine and/or cotinine have been administered (15, 33-39). Once it enters the bloodstream, the pharmacokinetics of nicotine that is inhaled, as from ETS, are expected to be similar to those observed after intravenous infusion. Similarity between inhaled and infused nicotine has been confirmed for half-life data; other parameters are difficult to estimate after the inhalation of nicotine because the exact absorbed dose is unknown. Average pharmacokinetic parameters for nicotine are shown in table 1. The half-life of nicotine averages 2-3 hours. With intermittent exposure, nicotine levels in the body rise and fall throughout the day. The half-life of cotinine averages about 17 hours. Because of the longer half-life, cotinine levels tend to build up throughout the day, and cotinine is eliminated over a much longer period of time compared with nicotine. With intermittent nicotine exposure such as occurs with cigarette smoking, cotinine levels remain relatively constant throughout the day and remain at near steady-state values (figure 2). The same is expected to be true for cotinine levels during ETS exposure. Because of the relative stability over time of cotinine levels in blood, cotinine has been the preferred measure used to estimate nicotine exposure from tobacco. Data from my laboratory indicate that the pharmacokinetics of nicotine and cotinine are similar in smokers and nonsmokers (table 1) (35). Using brief intravenous infusions of deuterium-labeled nicotine, the average clearance of nicotine was slightly faster in 11 nonsmokers compared with 11 smokers, each group comprised of nine men and two women, with the groups being age-matched. In a similar study with infusion of deuterium-labeled cotinine, cotinine clearances were found to be the same in 20 smokers compared with six nonsmokers. Studies using deuteriumlabeled as well as unlabeled -nicotine and ^-cotinine have shown that people eliminate deuterium-labeled nicotine and cotinine at the same rate as unlabeled or natural nicotine and cotinine (15,40). In support of our findings, other investigators have shown that the pharmacokinetics of unlabeled ^-cotinine in nonsmokers and smokers are similar (36, 37, 39). It should be noted that there are papers published by other researchers suggesting that smokers metabolize nicotine and cotinine faster than nonsmokers (41-43). It is known that smokers metabolize some drugs more rapidly than nonsmokers, so it is logical to consider the possibility that smokers metabolize nicotine and cotinine more rapidly than do nonsmokers. However, the studies of Kyerematen et al. that report that nicotine and cotinine are metabolized faster in smokers than in nonsmokers have some significant methdological problems (41, 42). These studies were performed using low doses of l4 C-radiolabeled racemic nicotine. Racemic refers to a mixture of ^-nicotine and d- nicotine, two forms of nicotine with different three- TABLE 1. Human pharmacokinetics of nicotine and cotinine Half-life (minutes) Volume of distribution (liters) Total clearance (ml/min) Average valued 157 Smokers (n = : 11) 196 1,085 Standard deviation Nicotine* (" = 11) Average value ,319 Standard deviation Cotininef Smokers (n=20) (" = 6) Average Standard Average Standard value deviation value deviation 1, , * From Benowitz and Jacob (35). t From Benowitz and Jacob (15). t Average values determined by intravenous infusion of deuterium-labeled /-nicotine (d 2 ) and deuteriumlabeled l-cotinine (d 4 ).

5 192 Benowitz A 12N 4P 8P 12M 4A 8A Time FIGURE 2. Average plasma cotinine concentrations at various times of day in 31 smokers, smoking an average of 22 cigarettes per day. Dashed lines represent 95% confidence intervals. (From Benowitz and Jacob (15). Reprinted with the permission of the publisher and authors.) dimensional structures, as well as different biologic activities. Only -nicotine is present in tobacco, and only ^-cotinine is produced in the body from - nicotine. Kyerematen et al. used a radiometric assay, measuring the total radioactivity of nicotine and metabolites after column separation. Thus, the assay does not distinguish the stereoisomers of nicotine. Different stereoisomers may be metabolized at different rates (44, 45). In addition, at low doses of nicotine and cotinine, analytical sensitivity may be such that the terminal half-life is underestimated. In fact, the values reported by Kyerematen et al. for clearance of nicotine were much greater and the half-life much shorter than we and others have found using either natural or labeled nicotine in a number of studies (15, 33-35). Finally, the particular form of the drug metabolizing enzyme CYP450, forms that are induced by cigarette smoking, are not the same ones that are believed to be involved in the metabolism of nicotine or cotinine (35). Thus, it is likely that the administration of racemic nicotine and/or the very low doses of nicotine administered with associated analytical sensitivity problems explain the discrepant results of Kyerematen et al. compared with the results from my laboratory and others. Based on our data and those of most other investigators, it appears that the pharmacokinetics of ^-nicotine and ^-cotinine are similar in smokers and nonsmokers. There are reports that the rate of elimination of nicotine and cotinine after low level exposure to nicotine, such as from ETS, is slower than that following exposure to high levels of nicotine such as in primary tobacco users (43). This difference probably results from the slow release of nicotine from certain body tissues in which it is stored and not from differences in metabolic rates (clearance). Slow release from tissues would result in a slow rate of elimination because nicotine is only slowly made available to the bloodstream for metabolism and excretion. A slow elimination rate or a long half-life in this situation does not mean that the clearance rate, that is, the capacity of the liver and kidneys to eliminate the drug, is slow. In fact, as noted above, clearance rates of nicotine and cotinine are similar in smokers and nonsmokers. Further, it has been suggested that slow elimination of cotinine may result in an overestimation of ETS exposure in nonsmokers (43, 46). The concern about delayed elimination is valid only if measurements are made in non-steady-state conditions. When there is steady state, the effects of tissue distribution are not apparent, and only the daily exposure level and the rate of clearance of nicotine or cotinine in an individual influence the level of the nicotine or cotinine in the body of that individual (see equation 3). Although no empirical data are available, assumption of steady state for cotinine levels is reasonable in considering daily exposure to ETS in the workplace and/or at home. With such exposures, cotinine levels are expected to rise slowly and progressively during the period of exposure, peaking 4-6 hours after the end of exposure and then falling slowly during periods of nonexposure. For population studies, a random cotinine measurement would be a reasonable indicator of daily nicotine exposure. Cotinine as a biomarker for intake of nicotine The presence of cotinine in a biologic fluid indicates exposure to nicotine. There is some individual variation in the quantitative relation between cotinine levels in the blood (or saliva or urine) and the intake of nicotine. This is because different people convert different percentages of nicotine to cotinine (usual range percent) and because different people metabolize cotinine at different rates (usual range of cotinine clearance, ml/min) (15). The relation between nicotine and cotinine can be expressed mathematically as follows (based on steady-state exposure): Generation rate of COT = Intake rate of NIC X % conversion NIC to COT. (4) As noted previously in equation 2, at steady state, Generation rate of COT = Elimination rate of COT = CL COT X C Btt. (5)

6 Validity of Cotinine as ETS Marker 193 Therefore, combining equations 4 and 5: Intake rate of NIC = CL COT X C B,,% conversion NIC to COT. (6) Or rearranging: Intake rate of NIC X C BB. (7) % conversion NIC to COT In adult smokers, the conversion factor (K) that on average converts a blood level of cotinine to a daily intake of nicotine has been estimated to be 0.08 (mg/24h/ng/ml) (range ; coefficient of variation, 21.9 percent) (21). Thus, a cotinine level of 300 ng/ml (a typical value for a smoker) will correspond to a daily intake of 24 mg of nicotine. Since clearance data are similar for smokers and nonsmokers, the K factor is expected to be similar in nonsmokers. Because there is variability in the conversion factor discussed above, cotinine levels only approximate nicotine intake. However, the degree of variability in the conversion factor (coefficient of variation, 21.9 percent) is not particularly great compared with variability in the clearance of most other drugs, and is much less than the degree of variability typically observed for pharmacodynamic parameters (47). Even with this (inevitable) degree of imprecision, cotinine levels in large groups of subjects would be expected to accurately reflect average group exposure to nicotine from ETS. Because nicotine is absorbed into the bloodstream, and because cotinine is generated from nicotine in the liver and released into the bloodstream, blood levels of cotinine most closely reflect the dose of nicotine absorbed from ETS. However, most studies of ETS exposure have used saliva or urine concentrations of cotinine because these samples are easier to obtain. As noted previously, saliva and blood cotinine levels are highly correlated, with a saliva to blood ratio of Thus, saliva and blood cotinine levels are used interchangeably. The interpretation of urine levels, and potential sources of variability in using urine levels to estimate blood levels of cotinine, requires some discussion. As noted above, the intake of nicotine can be estimated from a steady-state blood cotinine concentration (see equation 7). The steady-state blood cotinine concentration can be estimated from the urine concentration using the known pharmacokinetic values for renal (kidney) clearance of cotinine as follows: 'I/COT X V (8) where CL Rcm is renal clearance of cotinine, Q, cot is urine concentration of cotinine, V is urine flow rate, and C fi is the average or steady-state blood cotinine concentration. The renal clearance of cotinine in nonsmokers has been determined to average 6 ml/min (34, 36). On average, urine output is 1 ml/min or 1,440 ml/24 hours. Rearranging equation 8, 'B,, CL Rc0T 6 ml/min V = 6. 1 ml/min (9) This predicted ratio of urine to blood cotinine concentration of 6 is similar to that which has been measured experimentally (28). Individual variability in this ratio can be contributed to by variation in renal clearance of cotinine and in urine flow rate. Idle (12) has argued that individual differences in nicotine and/or cotinine metabolism limit the utility of cotinine measurements. This argument is based, in part, on the study of Cholerton et al. (48) of nonsmokers who received a dose of 2 mg nicotine orally, and had nicotine and cotinine measured in the urine collected over the next 24 hours. That study reported 35-fold variability in cotinine excretion. However, that study is not directly relevant to ETS exposure for two reasons: First, nicotine was dosed orally, which results in extensive first pass metabolism of nicotine. Orally administered nicotine is absorbed from the intestines into the portal circulation and then into the liver where it is extensively metabolized before reaching the systemic circulation. This phenomenon is called first-pass metabolism. When nicotine is dosed orally, percent is metabolized by first-pass metabolism; that is, before ever reaching the systemic circulation (1). In this situation, the ratio of concentrations of nicotine/ cotinine in the blood stream, which is reflected by the ratio in the urine, is influenced much more by individual differences in the rate of metabolism by the liver compared to when nicotine is dosed systemically. Absorption of nicotine from ETS occurs via the alveoli in the lungs. In this case, there is very little first-pass metabolism, and nicotine is all, or nearly all, absorbed into the systemic circulation. Second, urine sampling was performed for only 24 hours after the dose, which is insufficient time for cotinine excretion to have been complete. With a half-life averaging 17 hours, at least hours are needed to collect all of the cotinine. With a single oral dose of nicotine, the nicotine and cotinine levels in the urine will be more affected by the rate of absorption of nicotine, and the rate of nicotine and cotinine elimination, than would be the case at steady state. Notwithstanding these problems, data on population variability (i.e., coefficient of variation)

7 E 2. Cotinine concentrations in nonsmokers and smokers (selected studies) Study (rel. no.) Year Number of subjects Smoking status Exposure level Analytical method Plasma or serum cotinine (ng/ml) Urine cotinine (ng/ml) Saliva cotini (ng/m rvls et al. (28) ld and Ritchie (81) Smokers No exposure Exposed Wife nonsmoker Wife smoker * * , (SE* ±1.3, median 5.0) 25.2 (SE ±14.8, median 9.0) ld et al. (82) Smokers hours ETS* exposure/week hours ETS exposure/week hours ETS exposure/week hours ETS exposure/week hours ETS exposure/week ,645(537-3,326) rvis et al. (83) , children, children, children, children Neither parent smoked Father smoked Mother smoked Both parents smoked 0.4 (me 1.3 (me 2.0 (me 3.4 (me ultas et al. (cited from ref. 10, p. 212) aged <5 years aged <5 years aged <5 years aged 5-17 years aged 5-17 years aged 5-17 years aged >17 years aged >17 years aged >17 years No smoker in home 1 smoker in home 2 or more smokers in home No smoker in home 1 smoker in home 2 or more smokers in home No smoker in home 1 smoker in home 2 or more smokers in home 1.7 (me 4.1 (me 5.6 (me 1.3 (me 2.4 (me 5.6 (me 1.5 (me 2.8 (me 3.7 (me achan et al. (76) , age 7 years, age 7 years, age 7 years No smokers in home 1 smokers in home 2 or more smokers in home 1.1 nm/ 10.2 nm/ 37.5 nm/l ompson et al. (8) Smokers Lives alone or with nonsmoker Lives with smoker 4.4 (geometric mean) (95% Cl* ) 11.4 (geometric mean) (95% Cl ) 1,691 (arithmetic mean) mmings et al. (84) No exposure past 4 days 1-2 exposures past 4 days 3-5 exposures past 4 days 6 or more exposures HPLC HPLC HPLC HPLC 6.2 (mean) 7.8 (mean) 9.8 (mean) 12.5 (mean) nstall-pedoe et al. (59) ,873 1,940 2,271 1,386, male Smokers, male, female Smokers, female 0.68 (median) 240 (median) 0.10 (median) 243 (median)

8 et al. (85) 1994 et al. (86) 1990 et al. (87) ,260, aged 5-7 years 293, aged 5-7 years 521, aged 5-7 years 553 Nonmokers, aged 5-7 years 629, females from 10 countries 210, females from 10 countries 359, females from 10 countries 124, females from 10 countries 1,071, aged 4-11 years 713, aged 4-11 years 379, aged years 268, aged years 3,154, aged &17 years 1,332, workers aged 17 years 779, workers aged 17 years 315, workers aged 17 years 246, workers aged 17 years No smokers in home Mother smoker Father smoker Mother and father smokers No home or work ETS exposure Exposure at work but not at home Exposure at home but not at work Exposure at home and at work No home ETS exposure Home ETS exposure only No home ETS exposure Home ETS exposure only No home or work ETS exposure No home or work ETS exposure Work ETS exposure only Home ETS exposure only Home and work ETS exposure eviations:, gas chromatography;, radioimmunoassay; ETS, environmental tobacco smoke; SE, standard error; matography-mass spectrometry. 2.7 ng/mg creatinine 4.8 ng/mg creatinine 9.0 ng/mg creatinine 10.0 ng/mg creatinine 0.29 (geom mean) ( (geome (95% Cl 1.2 (geome (95% Cl 4.0 (geome (95% Cl LC-MS 0.12 (geometric ) LC-MS 1.14 (geometric ) LC-MS 0.11 (geometric ) LC-MS 0.81 (geometric ) LC-MS 0.12 (geometric ) LC-MS 0 13 (geometric ) LC-MS 0.32 (geometric ) LC-MS 0.65 (geometric ) LC-MS 0.93 (geometric ) Cl, confidence interval; HPLC, high performance liquid chromatograph

9 196 Benowitz such as we present for the conversion factor, K, are more relevant to assessing the impact of interindividual variability of nicotine and cotinine metabolism on the utility of cotinine as a marker of ETS exposure than data on the range of urinary cotinine/nicotine ratios as presented by Cholerton et al. (48). TYPICAL LEVELS OF COTININE FROM ETS EXPOSURE AND ESTIMATION OF CORRESPONDING DOSE OF NICOTINE Cotinine levels in people exposed to ETS have been studied by many research groups (summarized in references 3, 10, and 11). Most studies have found increasing cotinine levels with increasing levels of selfreported ETS exposure. Values of cotinine in saliva, plasma, and urine from selected large studies are shown in table 2. The nonsmoking subjects in these studies represent several different populations. The data of Jarvis et al. (28) from adults attending cardiovascular disease clinics at a London, England, hospital can be used as an example for estimation of daily nicotine intake from ETS. Using urine concentrations of 7.7 and 1.6 ng/ml and the equations described previously, the estimated daily intake of nicotine by nonsmokers was 100 /Leg and 20 /xg for those reporting exposure and no exposure to ETS, respectively. The extreme of ETS exposure is likely to occur in pubs and bars where smoking is common and ventilation is often poor. In 42 nonsmoking bar staff in London and Birmingham, England, the median saliva cotinine concentration was 7.95 ng/ml (standard deviation, 6.1), with a range from 2.2 to 31.3 ng/ml (49). Using equation 7, the median nicotine intake can be estimated to be 630 /xg/day. The maximal nicotine intake, corresponding to a saliva cotinine concentration of 31.3 ng/ml, is estimated to be 2.5 mg/day (the equivalent nicotine intake to actively smoking two and a half cigarettes). The distribution of serum cotinine levels in a large US population between 1988 and 1991, as measured in the Third National Health and Nutrition Examination Survey, is shown in figure 3. AIR LEVELS OF NICOTINE FROM ETS AND PREDICTED COTININE LEVELS IN BIOLOGIC FLUIDS The theoretic relation between air levels of nicotine and cotinine levels in the urine of nonsmokers can be described, and is useful in understanding potential sources of variability in that relation. Assume a workplace level of nicotine in the air due to ETS of 20 /xg/m 3 (50). The dose of nicotine inhaled is equal to the product of air concentration and ventilation rate. A typical ventilation rate for an adult during light activity is 1 m 3 /hour. Thus, the intake of nicotine would be about 20 /xg/hour. About 71 percent of nicotine that is inhaled is absorbed (22), so the systemic dose of nicotine is estimated to be about 14 p,g/hour. Assuming an 8-hour workplace exposure, this would be equivalent to 112 /xg/day. Using equations described previously, this level of intake would produce an average urine cotinine concentration of 8.6 ng/ml (which is a value consistent with that measured in nonsmokers exposed to ETS). Air nicotine levels measured by Hammond over 9 hours in 11 Massachusetts office worksites that allowed smoking indicated a median level of 8.6 ju-g/m 3 (51). The absorption of nicotine from this level of exposure over 9 hours would be predicted to be 55 /xg, resulting in an average urine cotinine concentration of 4.0 ng/ml. For perspective, in office workplaces that banned smoking the median air nicotine level was 0.3 /xg/m 3. Individual variability may exist in the factors that determine the relationship between air levels of nicotine and urine cotinine concentrations. Potential sources of variability include respiratory ventilation rate (for example, higher minute ventilation with higher work levels), extent of pulmonary retention of nicotine by the lung, timing of the sample collection versus time of exposure, sources of exposure other than that under study, percent metabolic conversion of nicotine to cotinine, total and renal clearance of cotinine, and urine flow rate. It is expected, therefore, that urine cotinine would only approximate air nicotine levels. The relation between ambient air nicotine levels and cotinine levels in the urine or saliva of nonsmokers has been reported in three recent studies. Studies by Marbury et al. (52), and Henderson et al. (53) involving children in the home, and one study by Coultas et al. (50) of adults in the workplace, found a reasonably strong correlation between ambient air nicotine and urine cotinine concentrations (correlation coefficients, r = 0.81, Marbury et al.; r = 0.68, Henderson et al.; and r = 0.60, Coultas et al.). These correlations are probably as high as can be expected given the sources of variability in nicotine uptake and metabolism. In view of the multiple potential sources of individual variability, the studies of Marbury et al., Henderson et al., and Coultas et al. support the predictive value of urine cotinine concentration as a biomarker of ETSderived nicotine exposure. NICOTINE IN FOOD AS A SOURCE OF COTININE IN PEOPLE Several foods contain small amounts of nicotine (54-57). It has been suggested that nicotine from food

10 Validity of Cotinine as ETS Marker 197 No Reported Home or Work ETS Exposure Reported Home or Work ETS Exposure Serum Cotinine. ng/ml 1000 FIGURE 3. Distribution of cotinine levels in the US population aged 4 years and older by reported ETS exposure and tobacco use. Data from the Third National Health and Nutrition Examination Survey, (From Pirkle et al. (87). Reprinted with the permission of the publisher and authors.) might falsely indicate exposure to ETS (12, 13). Foods that contain nicotine, and the levels of nicotine measured in those foods, are shown in table 3. Davis et al. (55) have estimated that an average daily consumption of tomatoes, potatoes, cauliflower, and black tea (i.e., consumption of all of these foods together) might result in a daily intake of 8.8 jltg nicotine. They estimated, based on a maximum consumption of all of these particular foods (on the same day), that a person could take in as much as 99.9 /xg of nicotine per day from food. It should be noted that more than 50 percent of intake in the Davis study was based on drinking black tea, and this and other studies have shown that some black teas contain no nicotine. It is unknown how much nicotine the typical tea consumed by most Americans contains. Also of note, the nicotine intake from tea reported by Davis et al. (55) necessitates the consumption of about four quarts (3,840 ml) of fluid per day. Consumption of such large volumes of fluid would result in a urine output much greater than the 1,000 ml these authors assume in their prediction of urine cotinine concentration. A larger urine volume would substantially reduce the concentrations of cotinine in the urine below the Davis et al. estimate of 6.2 ng/ml. To place exposure to nicotine from food in perspective, one needs to compare the average intake of nicotine from food with that from ETS. This can be done by using typical urine cotinine values in significantly ETS-exposed individuals (about 6 ng/ml) along with pharmacokinetic calculations.

11 198 Benowitz TABLE 3. Cauliflower Eggplant Potatoes Green tomatoes Pureed tomatoes Ripe tomatoes Tomatoes Tea leaves Tea, instant Average nicotine content of vegetables and tea Mean nicotine content (ng/g)* 3.8,t 16.8* 100, 0* 4.8,$ 15.3,* , 4.1*9.6* 10.7D 0-109* * * Per gram wet weight. t From Domino et al. (54). * From Davis et al. (55). From Castro and Monji (56). Tl From Sheen (57). Amount (g) required to provide 13 ng nicotine, equivalent to 1 ng/ml cotinine in urine 774-3, , ,354-3,170 1,214 As shown in the previous section, the daily intake of nicotine (in mg/24 hours) can be estimated as: Intake of nicotine (mg/24hours) = 0.08 and on average C, 'tvcot = 6. X C Bss (in ng/ml); Therefore, using urine concentration of cotinine, one can establish daily intake of nicotine (mg) as: Intake of nicotine (mg/24hours) = g- X C ycot (ng/ml) = X C Ucor (ng/ml). A typical value for cotinine in the urine in a person exposed to ETS is 6 ng/ml. This would correspond to a daily nicotine intake of 80 /xg. This estimate is similar to that predicted from measurement of concentrations of nicotine in the air in the presence of ETS and typical ventilation rates as described previously. Thus, an average nicotine intake (i.e., absorbed dose) from significant ETS plus dietary exposure is about 80 /Jig. Repace (58) used average American vegetable consumption data, which included 27 grams of tomatoes and 75 grams of potatoes, to estimate a daily nicotine intake from the diet of 0.7 /tg/day. As noted previously, Davis et al. (55) calculated that daily consumption of tomatoes, potatoes, cauliflower, and black tea might result in a daily intake of 8.8 /xg of nicotine. Even in the latter case, the expected intake from a diet rich in nicotine-containing food is only 10 percent of total nicotine exposure seen in a person with significant ETS exposure. Conversely, an intake of 8.8 /xg of nicotine per day from food would be expected to yield a steady-state urine cotinine level of less than 0.7 ng/ml, which is well below the level taken to indicate significant ETS exposure. Table 3 lists the amounts of various vegetables needed to provide 13 /xg of nicotine, which would produce a urine cotinine of 1 ng/ml (15 percent of that seen with significant ETS exposure). These quantities of vegetables are substantially greater than most people consume on a daily basis. Based on the data in table 3, to produce a level of cotinine seen with a typical level of ETS exposure, i.e., 80 ju-g of nicotine/day, one would have to consume a minimum of 4.6 kg (10.2 pounds) of cauliflower, 0.8 kg (1.7 pounds) of eggplant, 5.5 kg (11.2 pounds) of potatoes, or 7.3 kg (16.0 pounds) of tomatoes daily. The impact of tea drinking on serum cotinine levels of nonsmokers in Scotland has been studied explicitly (59). No effect on plasma cotinine levels was seen with consumption levels of up to 10 cups or more per day of tea. In contrast, the same study showed a robust relation between self-reported ETS exposure and plasma cotinine levels. Thus, nicotine in tea appears to contribute little to cotinine levels in most people and would be insignificant compared with nicotine exposure from ETS. I conclude that while food is a source of low level nicotine exposure, for most people it represents an insignificant exposure compared with ETS, and is likely to inflate population estimates of ETS nicotine exposure by very little. NICOTINE EMISSIONS FROM THE ENVIRONMENT AS A SOURCE OF HUMAN EXPOSURE Nicotine from ETS deposits on room surfaces, such as walls and carpets, and may contaminate house dust. Nicotine emissions from surfaces or dust in the air may result in measurable levels of nicotine in the air that persist after the last cigarette was smoked in the room (60). Likewise, nicotine can emit from the clothes of smokers even when they are not smoking in the room. Thus, it has been suggested that nicotine exposure by nonsmokers reflects not only direct exposure to ETS but also exposure to room air where smoking has occurred in the past or where the room is shared by people whose clothes have been contaminated by tobacco smoke (60). While concentrations of nicotine have been measured in the air in these conditions, levels of nicotine are quite low compared with those in ETS. For example, house dust in the home of a nonsmoker in which cigarettes were smoked on one occasion was reported to result in nicotine levels in the air of /xg/m 3 over the succeeding few days

12 Validity of Cotinine as ETS Marker 199 (60). Using calculations described previously and assuming a ventilation rate of 1 m 3 /hour, and assuming 8-hour exposure, this would produce a daily nicotine intake of /i,g, which would result in a urine cotinine concentration of ng/ml. These values are trivial compared with those derived from ETS exposure. ANALYTICAL CHEMISTRY ISSUES The levels of cotinine in biologic fluids produced by exposure to ETS are quite low and measurement is challenging (61, 62). Several analytical techniques have been used, including gas chromatography with nitrogen phosphorus detection, gas chromatography with mass spectrometry, radioimmunoassay, and high performance liquid chromatography with ultraviolet or mass spectrometric detection (table 4). Radioimmunoassay involves the use of antibodies to cotinine. Radioimmunoassay is sensitive but is nonspecific. Nonspecificity means that radioimmunoassay detects chemicals that are structurally similar to as well as the particular chemical that is to be measured. Thus, radioimmunoassay could measure metabolites of cotinine and detect them as if they were cotinine. Gas chromatography and high performance liquid chromatography techniques involve separating specific chemicals using a column, and then measuring the separated drugs using a detector. Such techniques are highly specific. When gas chromatography and radioimmunoassay techniques have been compared in measuring cotinine concentrations on the same urine samples, radioimmunoassay levels tend to be higher, reflecting detection of cotinine metabolites (61). This difference is reflected in the data in table 2. Differences in analytical technology presumably explain the considerable differences in cotinine levels reported in various published studies of ETS exposure. However, regardless of the assay used, most researchers find significant within-study differences in cotinine levels between ETS exposed and nonexposed populations of nonsmokers. COMPARISON OF COTININE AND OTHER MARKERS OF ETS Cotinine concentrations in biologic fluids have been used by many scientists to evaluate ETS exposure because cotinine reflects exposure to nicotine, which is nearly specific to tobacco. Chemicals in tobacco smoke, such as carbon monoxide or cyanide (the latter metabolized in the body to thiocyanate), can be measured in blood (table 5). However, the levels of these chemicals are nonspecific, meaning there are significant sources of carbon monoxide and cyanide, including the body's own metabolism, other than ETS. Thus, these markers are both nonspecific and insensitive markers of ETS exposure. Other markers that have been proposed to quantitate tobacco exposure include adducts of 4-aminobiphenyl to hemoglobin in red blood cells (63-65), adducts of benzo[a]pyrene and other potential carcinogens to DNA in white blood cells (66-69), adducts of polycyclic aromatic hydrocarbons to plasma albumin (70), urinary excretion of nicotine-derived nitrosoamines (71), urinary hydroxyproline or N-nitrosoproline excretion (72), and urinary mutagenicity (63, 73). Also, solanesol has recently been proposed as a marker of particle exposure (23, 74). Unfortunately, solanesol is extensively metabolized in people and levels are quite low, making quantitation difficult. Overall, it appears that while these markers may reflect exposure to particular components of tobacco smoke in active smokers, the measures are too nonspecific (i.e., have high baseline values even in nonexposed nonsmokers or have environmental sources other than tobacco smoke) and/or insensitive (i.e., the increment due to ETS exposure is small compared with baseline values) for use in quan- TABLE 4. Analytical methods for measurement of cotinine in nonsmokers Method Radioimmunoassay High performance liquid chromatography Gas chromatography Gas chromatography-mass spectrometry Liquid chromatography atmospheric pressure chemical ionization tandem mass spectrometry Study (ref. no.) Langone et al. (88) Haley et al. (89) Knight et al. (90) Hariharan and VanNoord (91) Jacob et al. (92) Feyerabend et al. (93) Jacob et al. (40) Bernert et al. (94) Sensitivity Specificity Cost 1-2 ng/ml -1 ng/ml ng/ml ng/ml <0.05 ng/ml Variable (poorest in urine) Good Good Excellent Excellent Moderate Moderate High Extremely high

13 200 Benowitz TABLE 5. Comparison of possible biomarkers of environmental tobacco smoke (ETS) exposure Cotinine Nicotine Carbon monoxide Thiocyanate Biomarker Specificity Sensitivity 4-aminobiphenyl hemoglobin adduct Benzo[a]pyrene DNA adduct Polycyclic aromatic hydrocarbon albumin adduct Urinary tobacco-specific nitrosamines Urine hydroxyproline Urine thioethers Urine mutagenicity High High Moderate Moderate High High High Moderate Moderate Moderate Duration after exposure reflected 3-4 days Hours Hours Weeks Months Probably months 21 days Probably hours Probably hours Probably hours to day Hours to day Comments Can be measured in urine, plasma, saliva, or hair Short half-life means results are very dependent on time of sampling; saliva nicotine can be elevated by local deposition of ETS without indicating systemic absorption. Plasma levels are very low. Urine levels are highly influenced by urine volume and ph. Hair measurement is promising as a long-term marker of exposure Many environmental sources, carbon monoxide also produced by endogenous metabolism. Only small changes in carbon monoxide levels seen after ETS exposure Many dietary sources. Most studies show no difference between nonsmokers who are or are not exposed to ETS Levels in nonsmokers may be 10-20% those of smokers. Analytical technique technically difficult Analysis is technically difficult. Difference between smokers and nonsmokers not found in all studies Analysis is technically difficult Analysis is technically difficult No difference in nonsmokers who are or are not exposed to ETS in most recent study Environmental sources, including diet Influenced by dietary and other factors. Inconsistent results in comparing nonsmokers with and without exposure to ETS titation of levels of smoke exposure to which most nonsmokers are exposed (table 5). At the present time, cotinine appears to be the most specific and most sensitive biomarker for exposure to nicotine from ETS. HEALTH AND OTHER BIOLOGIC EFFECTS AS VALIDATING THE USE OF COTININE AS A BIOMARKER OF ETS EXPOSURE A significant relation between biologic effects of ETS and cotinine levels of biologic fluids would further support the idea that cotinine is a quantitative marker of ETS exposure. Several studies support this concept. Matsunga et al. (75) studied the effects of ETS exposure on the metabolic clearance of theophylline, a drug whose metabolism is known to be increased by cigarette smoking in nonsmokers. In 14 nonsmokers, significant correlations were found between plasma cotinine (r = 0.72) or urinary cotinine (r = 0.79) and the clearance of theophylline. Strachan et al. (76) studied year-old school children and found a positive correlation between the quintile of salivary cotinine levels and the risk of middle ear effusion. Strachan et al. (77) reported in a large group of 770 children a significant inverse correlation between salivary cotinine and various tests of lung function. Similar results were reported in another group of 2,511 children by Cook et al. (78). Likewise, the risk of wheezing bronchitis in children 18 months of age or younger increased as urinary cotinine excretion increased (79). Finally, Tunstall-Pedoe et al. (80) found a gradient of risk of diagnosed coronary heart disease

14 Validity of Cotinine as ETS Marker 201 that increased with increasing serum cotinine in nonsmokers. Thus, several different biologic effects of ETS have been shown to be quantitatively related to cotinine levels, supporting the idea that cotinine levels do reflect ETS exposure and effects. SUMMARY RESPONSES TO CONCERNS ABOUT THE USE OF COTININE AS A BIOMARKER OF ETS EXPOSURE Based on the analysis presented above, the following comments are offered in response to various concerns about the use of cotinine as a valid biomarker of ETS exposure. Concern 1 The ratio of nicotine to other ETS components is highly variable in relation to other ETS components, and depends on such factors within a space as surfaces, ventilation rate, sampling duration, time since smoking, and air distribution patterns. Comment. While nicotine to paniculate ratios change over time as ETS decays, and such ratios may be influenced by the characteristics of the environment, time-weighted average measurements in a variety of home and work environments produce reasonably consistent ratios. Since most nonsmokers are exposed to ETS over time as well, nicotine intake reasonably reflects exposure to particles and other components of ETS. Nicotine to particle ratios may not be accurate at low ETS concentrations because background levels of RSP can have a major influence on that ratio, but this does not invalidate the utility of nicotine as a marker of tobacco-derived particles. Concern 2 The ratio of nicotine emission to RSP emission is not constant across a wide range of cigarettes. Comment. While different brands of cigarettes emit different amounts of nicotine and tar, the nicotine to particulate ratio in sidestream smoke or in ETS in popular US and Canadian brands is reasonably constant. There is no relation between the tar or nicotine yield or ventilation characteristics of the cigarette and the tar/nicotine ratio in sidestream smoke. Concern 3 Exposure to nicotine vapor in the absence of other ETS components can occur. Comment. While nicotine vapor may be present due to emission from room surfaces, people's clothing, or house dust, the magnitude of this exposure in most environments appears to be insignificant compared with that derived from moderate ETS exposure. Concern 4 There is no standard method for determining nicotine or its metabolites in biologic fluids. Comment. The major problem to date has been the nonspecificity of radioimmunoassay measurements, which detect other nicotine metabolites as well as cotinine in urine. Recent studies indicate that radioimmunoassay and gas chromatography with mass spectrometry measurements in urine are proportional. Most researchers have found positive relations between airborne nicotine exposures or self-reported exposures to ETS and cotinine concentrations regardless of analytical techniques, indicating that differences between techniques are systematic, but that cotinine levels are consistent for one analytical technique. Concern 5 Interindividual differences in rates and pattern of nicotine and cotinine metabolism render the use of nicotine or cotinine as biomarkers of limited utility. Comment. There is considerable individual variability in the extent and metabolism of nicotine and cotinine and in the rate of cotinine metabolism and elimination. Such variation is seen for all drugs and chemicals. Using pharmacokinetic data derived from cigarette smokers, a proportionality constant between blood levels and nicotine intake can be estimated. While there are rare outliers, variability for this constant in most smokers is reasonable (coefficient of variation, 22 percent). Recent studies indicate that the pharmacokinetics of nicotine and cotinine are similar in smokers and nonsmokers, suggesting that this degree of variability is likely to be similar for nonsmokers. Concern 6 Dietary nicotine exposure may confound low-level determination of nicotine and cotinine in biologic fluids. Comment. While nicotine is contained in some foods, at usual food consumption levels dietary nicotine is trivial compared with moderate ETS exposure. CONCLUSION As described previously, the National Research Council criteria for a valid marker of ETS exposure include: 1) should be unique or nearly unique for ETS so that other sources are minor in comparison; 2) should be easily detectable; 3) should be emitted at

15 202 Benowitz similar rates for a variety of tobacco products; and 4) should have a fairly constant ratio to other ETS components of interest under a range of environmental conditions encountered (3). Nicotine in the air and measurement of its metabolite cotinine in biologic fluids meet these criteria reasonably well. There is interindividual variability in any biologic measurements. While such variability may limit the value of prediction based on a measurement in an individual, variability is compensated for in studies of large numbers of subjects, as in epidemiologic studies. In support of this conclusion is the observation that cotinine levels in nonsmokers have positively correlated to the risks of some ETS-related health complications in children. The evidence presented in this review indicates that cotinine levels provide a valid and quantitative measure of average human ETS exposure over time. Cotinine is clearly the best available biomarker of ETS exposure at present. ACKNOWLEDGMENTS Supported in part by US Public Health Service grants DA02277 and DA This manuscript is based in part on testimony prepared for the US Occupational Safety and Health Administration hearing on the proposed standard for indoor air quality, September 26, The author wishes to thank Drs. Peyton Jacob, Lewis Sheiner, James Repace, and Surender Ahir for their helpful suggestions, and Kaye Welch for manuscript preparation. REFERENCES 1. Benowitz NL, Jacob P III, Denaro C, et al. Stable isotope studies of nicotine kinetics and bioavailability. Clin Pharmacol Ther 1991;49: Benowitz NL, Jacob P III. Daily intake of nicotine during cigarette smoking. Clin Pharmacol Ther 1984;35: Committee on Passive Smoking, Board of Environmental Studies and Toxicology, National Research Council. Environmental tobacco smoke: measuring exposures and assessing health effects. Washington, DC: National Academy Press, Benowitz NL. The use of biologic fluid samples in assessing smoke consumption. In: Grabowski J, Bell CS, eds. Measurement in the analysis and treatment of smoking behavior. NIDA research monograph no. 48. Washington, DC: US GPO, 1983:6-26. (DHHS publication no. (ADM) ). 5. Etzel RA. A review of the use of saliva cotinine as a marker of tobacco smoke exposure. Prev Med 1990; 19: Perez-Stable EJ, Benowitz NL, Marin G. Is serum cotinine a better measure of cigarette smoking than self-report? Prev Med 1995;24: Benowitz NL, Sharp DS. Inverse relation between serum cotinine concentration and blood pressure in cigarette smokers. Circulation 1989;80: Thompson SG, Stone R, Nanchahal K, et al. Relation of urinary cotinine concentrations to cigarette smoking and to exposure to other people's smoke. Thorax 1990;45: Repace JL, rey AH. An enforceable indoor air quality standard for environmental tobacco smoke in the workplace. Risk Anal 1993; 13: The health consequences of involuntary smoking: a report of the Surgeon General. Washington, DC: US GPO, (DHHS publication no. (CDQ ). 11. Respiratory health effects of passive smoking: lung cancer and other disorders. Washington, DC: Office of Health and Environmental Assessment, Office of Research and Development, US Environmental Protection Agency, Idle JR. Titrating exposure to tobacco smoke using cotinine a minefield of misunderstandings. J Clin Epidemiol 1990;43: Nelson PR. Submission to US Occupational Safety and Health Administration Hearing on the Proposed Standard on Indoor Air Quality (Docket No. H-122), Benowitz NL, Hall SM, Herning Rl, et al. Smokers of lowyield cigarettes do not consume less nicotine. N Engl J Med 1983;309: Benowitz NL, Jacob P III. Metabolism of nicotine to cotinine studied by a dual stable isotope method. Clin Pharmacol Ther 1994;56: Coultas DB, Stidley CA, Samet JM. Cigarette yields of tar and nicotine and markers of exposure to tobacco smoke. Am Rev RespirDis 1993; 148: Rickert WS, Robinson JC, Collishaw N. Yields of tar, nicotine, and carbon monoxide in the sidestream smoke from 15 brands of Canadian cigarettes. Am J Public Health 1984;74: Leaderer BP, Hammond SK. Evaluation of vapor-phase nicotine and respirable particle mass as markers for environmental tobacco smoke. Environ Sci Technol 1991;25: Benowitz NL, Henningfield JE. Establishing a nicotine threshold for addiction: the implications for tobacco regulation. N Engl J Med 1994;331: Eatough DJ, Benner CL, Bayona JM, et al. Chemical composition of environmental tobacco smoke. I, Gas-phase acids and bases. Environ Sci Technol 1989;23: Benner CL, Bayona JM, Caka FM, et al. Chemical composition of environmental tobacco smoke. 2. Particulate-phase compounds. Environ Sci Technol 1989;23: Iwase A, Aiba M, Kira S. Respiratory nicotine absorption in non-smoking females during passive smoking. Int Arch Occup Environ Health 1991 ;63: Nelson PR, Heavner DL, Collie BB, et al. Effect of ventilation and sampling time on environmental tobacco smoke component ratios. Environ Sci Technol 1992;26: Eatough DJ, Benner CL, Tang H, et al. The chemical composition of environmental tobacco smoke. III. Identification of conservative tracers of environmental tobacco smoke. Environ Int 1989;15: Rando RJ, Menon PK, Poovey HG, et al. Assessment of multiple markers of environmental tobacco smoke (ETS) in controlled, steady-state atmospheres in a dynamic test chamber. Am Ind Hyg Assoc J 1992;53: Miesner EA, Rudnick SN, Hu FC, et al. Paniculate and nicotine sampling in public facilities and offices. JAPCA 1989;39: Benowitz NL, Jacob P III, Fong I, et al. Nicotine metabolic profile in man: comparison of cigarette smoking and transdermal nicotine. J Pharmacol Exp Ther 1994;268: Jarvis M, Tunstall-Pedoe H, Feyerabend C, et al. Biochemical markers of smoke absorption and self-reported exposure to passive smoking. J Epidemiol Community Health 1984;38: Curvall M, Elwin CE, Kazemi-Vala E, et al. The pharmacokinetics of cotinine in plasma and saliva from non-smoking healthy volunteers. Eur J Clin Pharmacol 1990;38: Eliopoulos C, Klein J, Phan MK, et al. Hair concentrations of nicotine and cotinine in women and their newborn infants.

16 Validity of Cotinine as ETS Marker 203 JAMA 1994;271: Nafstad P, Botten G. Hagen JA, et al. Comparison of three methods for estimating environmental tobacco smoke exposure among children aged between 12 and 36 months. Int J Epidemiol 1995;24: Rowland M, Tozer TN. Clinical pharmacokinetics: concepts and applications. 2nd ed. Philadelphia, PA: Lea and Febiger, 1989: Benowitz NL, Jacob P III, Jones RT, et al. Interindividual variability in the metabolism and cardiovascular effects of nicotine in man. J Pharmacol Exp Ther 1982;221: Feyerabend C, Ings RMJ, Russell MAH. Nicotine pharmacokinetics and its application to intake from smoking. Br J Clin Pharmacol 1985;19: Benowitz NL, Jacob P HI. Nicotine and cotinine elimination pharmacokinetics in smokers and nonsmokers. Clin Pharmacol Ther 1993;53: De Schepper PJ, Van Hecken A, Daenens P, et al. Kinetics of cotinine after oral and intravenous administration to man. Eur J Clin Pharmacol 1987;31: Jarvis MJ, Russell MAH, Benowitz NL, et al. Elimination of cotinine from body fluids: implications for noninvasive measurement of tobacco smoke exposure. Am J Public Health 1988;78: Benowitz NL, Kuyt F, Jacob P III, et al. Cotinine disposition and effects. Clin Pharmacol Ther 1983;4:604-l Curvall M, Vala EK, Enzell CR, et al. Simulation and evaluation of nicotine intake during passive smoking: cotinine measurements in body fluids of nonsmokers given intravenous infusions of nicotine. Clin Pharmacol Ther 1990,47: Jacob P III, Yu L, Wilson M, et al. Selected ion monitoring method for determination of nicotine, cotinine, and deuteriumlabeled analogs: absence of an isotope effect in the clearance of (S)-nicotine-3'-3'-d 2 in humans. Biol Mass Spectrom 1991; 20: Kyerematen GA, Damiano MD, Dvorchik BH, et al. Smokinginduced changes in nicotine disposition: application of a new HPLC assay for nicotine and its metabolites. Clin Pharmacol Ther 1982;32: Kyerematen GA, Morgan ML, Chattopadhyay B, et al. Disposition of nicotine and eight metabolites in smokers and nonsmokers: identification in smokers of two metabolites that are longer lived than cotinine. Clin Pharmacol Ther 1990;48: Haley NJ, Sepkovic DW, Hoffmann D. Elimination of cotinine from body fluids: disposition in smokers and nonsmokers. Am J Public Health 1989;79: Jacob P III, Benowitz NL, Copeland JR, et al. Disposition kinetics of nicotine and cotinine enantiomers in rabbits and beagle dogs. J Pharm Sci 1988;77: Nwosu CG, Godin CS, Houdi AA, et al. Enantioselective metabolism during continuous administration of S-(minus)- and R-(plus)-nicotine isomers to guinea-pigs. J Pharm Pharmacol 1988;40: Gori GB, Mantel N. Mainstream and environmental tobacco smoke. Regul Toxicol Pharmacol 1991 ;14: Levy G, Ebling WF, Forrest A. Concentration- or effectcontrolled clinical trials with sparse data. Clin Pharmacol Ther 1994;56:l Cholerton S, Arpanahi A, McCracken N, et al. Poor metabolisers of nicotine and CYP2D6 polymorphism. (Letter). Lancet 1994,343: Jarvis MJ, Foulds J, Feyerabend C. Exposure to passive smoking among bar staff. Br J Addict 1992;87: Coultas DB, Samet JM, McCarthy JF, et al. A personal monitoring study to assess workplace exposure to environmental tobacco smoke. Am J Public Health 1990;80: Hammond SK, Sorensen G, Youngstrom R, et al. Occupational exposure to environmental tobacco smoke. JAMA 1995;274: Marbury MC, Hammond SK, Haley NJ. Measuring exposure to environmental tobacco smoke in studies of acute health effects. Am J Epidemiol 1993;137: Henderson FW, Reid HF, Morris R, et al. Home air nicotine levels and urinary cotinine excretion in preschool children. Am Rev Respir Dis 1989;140: Domino EF, Hombach E, Demana T. The nicotine content of common vegetables. (Letter). N Engl J Med 1993,329: Davis RA, Stiles MF, debethizy JD, et al. Dietary nicotine: a source of urinary cotinine. Food Chem Toxicol 1991;29: Castro A, Monji N. Dietary nicotine and its significance in studies on tobacco smoking. Biochem Arch 1986;2: Sheen SJ. Detection of nicotine in foods and plant materials. J Food Sci 1988;53: Repace JL. Dietary nicotine: won't mislead on passive smoking... (Letter). BMJ 1994;308: Tunstall-Pedoe H, Woodward M, Brown CA. Tea drinking, passive smoking, smoking deception, and serum cotinine in the Scottish Heart Health Study. J Clin Epidemiol 1991 ;44: Nelson PR, debethizy JD, Davis RA, et al. Where there's smoke...? Biases in the use of nicotine and cotinine as environmental tobacco smoke biomarkers. In: Proceedings of the 1991 Environmental Protection Agency/Air and Waste Management Association International Symposium: Measurement of toxic and related air pollutants. Pittsburgh, PA: Air and Waste Management Association, 1991: Biber A, Scherer G, Hoepfner I, et al. Determination of nicotine and cotinine in human serum and urine: an interlaboratory study. Toxicol Lett 1987;35: Watts RR, Langone JJ, Knight GJ, et al. Cotinine analytical workshop report: consideration of analytical methods for determining cotinine in human body fluids as a measure of passive exposure to tobacco smoke. Environ Health Perspect 1990;84: Bartsch H, Caporaso N, Coda M, et al. Carcinogen hemoglobin adducts, urinary mutagenicity, and metabolic phenotype in active and passive cigarette smokers. J Natl Cancer Inst 1990; 82: Hammond SK, Coghlin J, Gann PH, et al. Relationship between environmental tobacco smoke exposure and carcinogen-hemoglobin adduct levels in nonsmokers. J Natl Cancer Inst 1993;85: Maclure M, Katz RB, Bryant MS, et al. Elevated blood levels of carcinogens in passive smokers. Am J Public Health 1989; 79: Binkova B, Lewtas J, Miskova I, et al. DNA adducts and personal air monitoring of carcinogenic polycyclic aromatic hydrocarbons in an environmentally exposed population. Carcinogenesis 1995; 16: Jahnke GD, Thompson CL, Walker MP, et al. Multiple DNA adducts in lymphocytes of smokers and nonsmokers determined by 32 P-postlabeling analysis. Carcinogenesis 1990; 11: Savela K, Hemminki K. DNA adducts in lymphocytes and granulocytes of smokers and nonsmokers detected by 32 P- postlabeling assay. Carcinogenesis 1991; 12: van Maanen JMS, Maas LM, Hageman G, et al. DNA adduct and mutation analysis in white blood cells of smokers and nonsmokers. Environ Mol Mutagen 1994;24: Crawford FG, Mayer J, Santella RM, et al. Biomarkers of environmental tobacco smoke in preschool children and their mothers. J Natl Cancer Inst 1994;18: Hecht SS, Carmella SG, Murphy SE, et al. A tobacco-specific lung carcinogen in the urine of men exposed to cigarette smoke. N Engl J Med 1993,329: Adlkofer F, Scherer G, Heller WD. Hydroxyproline excretion in urine of smokers and passive smokers. Prev Med 1984; 13: Bos RP, Theuws JLG, Henderson PT. Excretion of mutagens in human urine after passive smoking. Cancer Leu 1983;19: Ogden MW, Maiolo KC. Collection and determination of