Tobacco Documents Online

1 I I I I 1 I I I I I I I I I I I I P VIACOEPIDEVIIOLOGY AND DRUG UTILIZATION How the steady-state cotinine concentration in cigarette smokers is directly related to nicotine intake The relationship between nicotine intake and steady-state cotinine concentntion was studied in a sample of 125 subjects who smoked their usual brands of cigarettes. Nicotine and tar yield of cigarettes was determined with a smoking machine, under standardized conditions. Blood was drawn about 8 hours after the last cigarette was smoked and serum cotanine was measured by high performance liquid chro- matography. Cotinine levels ranged from 11 to 400 ng/ml, and nicotine daily intake ranged from 1 to 33 mgidav. Regression anaivsis and the correlation coefficient, r = 0.919, significant at p < 0.000 1, showed that steady-state cotinine level was linearly and directly related to daily available nicotinea with an in- crease in correlation coei$cient directly related to the increase in tar and nicotine yieid. From the find- ings we also conclude that smokers of low-tar cigarettes do not tend to compensate for lower yields of nicotine. (Ct.LV PttAL.Kacot. THEx 1992;52:3249.) Mirella Rosa, BSc, Roberta Pacifici, BSe, Ilaria Altieri, BSc, Simona Pichini. ChB, Giorgio Ottaviani, MD, and Piergiorgio Zuccaro, PhD Rvmc, IralY The inhalation of cisarette smoke is the most im- portant cause of death from cancer in the world`. More than :800 components of tobacco smoke have been identified. with nicotine bein¢ the most exten- iivelv studied."3 The content of nicotine. tar tor total particulate mat- ter). and carbon monoxide can be conveniently mea- sured by a smokins machine under standardized labo- ratorv conditions.:*s Si¢nificant changes have been made in the past few decades by cigarette manufactur- ers to obtain ci¢arettes with lower tar and nicotine vtelds. The major changes in cigarette design include more specific blend selection. addition of filters, the From thr Clinical Biuchcmistrv IJepanment. Istttuto Supenorr di S;uvta. Lnd the Air Force Medical Centre ".AIdo di Loreto." Supponed in ?art by grant No. ?61 from Ministero delle Finanze. `tonopult dt Stato. Kcctis•ed tur oubltcauon March 4. 1992: accepted May 11. 1992. Repnnt requests: Piereturgto Zuccaro. PhD. Clinical Biochemistry Dtpanment. tbtituto Supenure di S:utua. V le Regtna Elena :99. 00461 Rome. Italy. 13/1/39187 use of reconstituted tobacco sheet. and the develop- ment of ventilation techniques.°-' The effect of these changes is reflected in a trend of declining sales- weighted average tar and nicotine levels in the ctga- rettes brands in many countries. ' Epidemioio¢ic studies have reported that the type of cigarette smoked (filter versus nontilter) and tar content were associated with a lower luns cancer risk.'o't t Because lun¢ cancer is believed to be related to disposition of the particulate traction (tar and nicotine of cigarette smoke in the lun_). it has been suggested that reduction in tar %itld of cigarettes would diminish the incidence of cancer of the lunst. `-'''` Three constituents of tobacco smoke-carbon mon- oxide. nicotine. and hvdro=en cyanide-are used rou- tinely to measure the extent to which tobacco smoke is inhaled and taken into the human body.'s"7 Cotinine. which is a maior metabolite of nicotine. offers several sdvanta_es over other hiochemical marKers: it is a specitic indicator of nicotine intake and is stable in bodv tluids. it has a long half-life. and its concentra- 324 I

I I I I I I I I I I I I I I I I I l'() LC\tE 5_ \•L•Nit3tP. ? Table 1. Subiect characteristics of 125 smokers tiiroune ~•ield tmgictgarette) Tar tnicoune rreef tmcicicarettei No. of cigarettes preceding 24 hours bio. ot years subjects smoked Serum cottmne t ngimh tion is not influenced bv confounding factors such as diet or environment. 1s''o Because the tar-to-nicotine ratio is quite constant in each brand of cigarette. determination of nicotine in- take could provide the main data necessary for epide- miologic studies related to tobacco smoking. The pur- pose of the study was to determine how steady-state cotinine reiatts to an index of nicotine intake derived from both cigarettes smoked per day and nicotine vield information in a sample of cigarette smokers. MATERIAL AND METHODS Nicotine and tar yield of cigarettes were measured with a type SM 350 smoking machine (Filtrona Instru- ments. Blechiey, En¢land) under standardized condi- tions.; S The machine generates a puff of smoke of standard volutne each minute until the ciQarette is smoked to a specific length. The particulate matter in the smoke is removed by a Cambridge filter. The nic- otine content of the particulate is reported as nicotine yield. The dry weight of the particles minus the nico- tine content is reported as the tar yield. Subjects. The study included military and civilian employees and their relatives, who attended the Air Force Medical Centre (Rome. Italy) for routine analy- ses. Only subjects who had not smoked during the night and in the early motning and who had not changed brand of cigarettes smoked in the last 24 weeks wete recrttited. One hundred twenty-five healthy voltmteets (30 women and 95 men) were stud- ied. Their mean age was 34.99 = 10.58 years (age range. 18 to 62 years), and mean body weight was 70.55 = 11.67 kg (weight range, 50 to 98 kg). Each volunteer gave informed consent to the inves- tigation. Smokers provided data concerning medical histones, drug usage. alcohol and coffee consump- tion. cigarette brand. and the number of cigarettes smoked in the last 24 hours. Blood was drawn at 9Am for determination of se- rum cotinine. Samples were immediately centrifuged. and serum was collected. frozen without preserva- tives, and sxorai at -20° C until assayed. Standard statistical analysis were used: the Student \'icotine intakc ancti catinirie fcrels 3 2 5 Mean = SD Rmrer 0.88 - 0,24 0.38-I.: ; 11.03-4.15 4.0-199 10.88 - 6.50 2-40 14.3 -, 9.3 1-40 134.9 - 85.2 11-400 r test for unpaired data and Pearson's correlates for re- gression analysis. All data were expressed as mean values = SD. Cotinine analvsis. The method of Pichini et al.21 for serum cotinine determination bv HPLC was slightly modified in our study. These modifications consisted of the followina: a reversed-phase µBonda- pak C,K steel column (10 µm particle size. 30 cm X 3.9 mm internal diameter: Waters Chromatography Div.. Millipore. Milford. Mass) ) was used for serum analyses: furthermore. the mobile phase. which was changed to waterimethanol/0.1 mol/L sodium ace- tate/acetonitrile (67:24.5:6.5:2 vol/vol) adjusted to pH 4.3 with acetic acid and used at a flow rate of 1.5 mhmin. N-Ethylnorcotinine was used as internal stan- dard. Cotinine was extracted from serum. as previ- ously described. by an Extrelut I extraction column (Bracco. Milan. Italy) eluting with 5 ml dichlo- romethane/pentane (1 : 2 vol/vol). and it was detected at 254 nm. The sensitivity of this assay was 5 no co- tinine per milliliter of body fluid. ~ RESULTS The smoking histories of the subjects entering the study are summarized in Table 1. There was consider- able variability among smokers in nicotine yield. smoking years. and serum cotinine. The subjects smoked an average of 10 cigarenes per day. The study included some subjects who were heavy smokers. but it also included many light smokers. that is. those who had drastically reduced the number of cigarettes smoked or nicotine and tar content in the last few years. thus the number of cigarettes smoked by the study subjects reflected the Italian mean daily con- sumption of cigarettes (13 cigarettes per day). The most popular European and U.S. cigarette brands were represented within the sample of cigarette smok- ers. Cigarette filters did not present particular charac- teristics but did provide filter ventilation in the case of low-yield cigarettes. The smoking population was arbitrarily divided into four groups on the basis of tar yields of their cigarettes and according to the classification adopted by the International Agency I

I I I I I I I I I I I I I I I I I I 326 Rosa et al. Table II. Characteristics of smokers according to tar yield of their cigarettes cu.` rtiAltas.lcot. iHBR SEI'fE.MEER 1992 Grouo Range of tar (mg/cigarette) n Age (vr) Range of nicotlne yield (mgicigarerre) No. of cigarettes preceding 24 hours No. of years subjects smoked 1 0-4.9 8 35.2 - 7.9 0.38-0.46 8.5-6.1 14.6=7.5 2 5-9.9 45 36.3 = 10.9 0.55-0.84 10.9 = 7.2 14.3 = 8.3 3 10-14.9 38 38.5 - 8.6 0.88-1.06 11.6 = 6.6 17.8 = 8.1 s 15-19.9 34 29.3 - 10.6 1.12-1.38 10.9 - 6.0 10.3 - 10.2 Dau un mean vslues - SD. Table III. Serum concentration of codnine and daily available nicotine in groups of smokers according to nicotine yield Group n Range of nicotine yteld tmgtcigarettet ,Viconne/dav (mg1* Cotinine (ng/m!) Cotmine/nlcottne ratio (m1-, • 10-3) 1 8 0.38-0.46 3.41 t 2.20 63.12 - 22.90 1.85 2 -t5 0.55-0.84 7.46 - 4.95 104.68 = 62.51 1.40 3 38 0.88-1.06 11.17 - 6.26 152.34 = 87.13 1.36 4 34 1.12-1.38 12.25 - 6.99 172.26 = 95.82 1.41 Change ~ -72.2% -63.4% Change: -39.1% -39.2% Dan ue mean values = SD. -tiicounadav was cakulued from nicounernullienurvciaarette and No. of cicuetses smoked per da,v. The persennge chanee in the mean vUues by companson of groups 4 and l. :The perceneage chanze in tbe me.an values by comQanson of groups 4 and 2. for Research on Cancer (IARC) Working Group (Table II).'-= The subjects of different groups did not differ sig- nificantly in age. in number of cigarettes smoked in the preceding 24 hours. or in the number of years of cigarette smoking. Only group 1 was different from the others with regard to the number of subjects. In Table III the daily nicotine intake (calculated from milligrams of nicotine per cigarette measured with the smoking machine and number of cigarettes smoked per day), serum cotinine levels, and cotinine levelsrnicotine intake ratios were reported for the four sroups. Cotinine levels and daily nicotine decreased proportionally to the decrease in the yield of nicotine. For example, comparison of smokers in groups 4 and l showed a 72.2% decrease in daily nicotine and a 63.4% decrease in serum cotinine. whereas compari- son of smokers in groups 4 and 2 showed a decrease of 39.1c and 39.2~c. respectively. The cotinineinico- tine ratio was not significantly different in the groups 2. 3. or -a•. Only the first group showed an higher ra- tio, but in this case the number of the subjects was quite small. The correlation between serum cotinine and daily available nicotine for 125 smokers is shown in Fig. i. Cotinine serum levels ranged from ! 1 to 400 n_vml and nicotine daily intake ranged from 1 to 33 mg/day. The linear correlation coefficient of r = 0.919 was significant at p < 0.0001. A significant correlation was also found in each of the four Qroups, with an in- crease in the Pearson's correlation coefficient directly related to the increase in nicotine yield (Table IV). The relationships between serum cotinine and ciga- rettes per day or nicotine yield exhibited correlation coefficients of r = 0.650 (significant at p< 0.01) and r = 0.300 (difference not significant), respectively. DISCUSSION A larse number of studies have established that many smokers will alter their smoking behavior if they smoke a new brand of cigarette with a lower tar concentration, changing the number of cigarettes smoked. the method of puffing, and depth of inhaling the smoke. to maintain a desired level of nicotine in- take.- "23 It is necessary to consider that several studies never take into consideration the time of accommodation to fit a new level of nicotine.=6'=$ Moreover, if compari- son of brand yields are to be made. a matched-paired design should be used. and the study should be carried out over several weeks to allow the subject to get used to his smoking habits.=D Recentiy Bridges et ai.~' studied 108 subjects divided into four groups accord- ing to relatively narrow ranges of nicotine yield of their cigarettes: they reported that the number of ciga- rettes smoked dailv and the volume inhaled were the

I I I I I I I I I I I I I 11•oL~aa s_ \'L'\iBER : 0 5 tiTicotzne zrttlike RfZd cotl11i31e 1CPd,t 32i 10 15 20 Nicotine !mg/dayJ 30 Fig. 1. Relationship between serum steady-state concentration of cotinine and daily available nic- otine (n = 125. }• = 16.33 T 12.3QY, r = 0.919. p < 0.0001). Table IV. Correlation between daily nicotine intake and serum cotinine levels 35 Group n r p Regression equanon l 8 0.755 <0.03 }• = 36.20 + 7.86x 2 45 0.863 <0.0001 Y = 23.41 - 10.88x 3 38 0.900 <0.0001 ~• = 12.38 - 1'_.52x 4 34 0.936 <0.0001 ~ = 15.04 + 12.82x TOTAL 125 0.919 <0.0001 c = 16.33 = 12.30x same in all the groups, except for five subjects who smoked cigarettes with super low nicotine yield (mean. 0.34 mg/cigarette). In our study we have con- firmed that smokers of low nicotine and tar cigarettes (groups 1 and 2) do not smoke a larger number of cig- arettes (Table II). Benowitz et a1..31 studying 272 smokers. reported that when blood was drawn in the afternoon around 5 PM (group 1) cotinine concentration was significantly correlated with the number of cigarettes smoked (r = 0.40, p < 0.01) and with the daily intake of nicotine (r = 0.42). There was no cotrelation. instead, be- tween machine-determined nicotine yield in one ciga- rette and blood cotinine (r = 0.15. difference not sig- nificant). The same results occurred when blood samples were drawn in the morning at 8 AM after 8 to 12 hours of abstinence (group 2). Therefore the author argued that smokers of low nicotine cigarettes did not consume less nicotine. These findings cannot be gen- eralized because we do not know if a correlation exists between the daily intake of nicotine and blood coti- nine levels. Our study used conditions of steady state for serum cotinine. The findings revealed a linear correlation be- tween daily nicotine intake and serum levels of coti- nine (Fig. 1), which is in agreement with the linear re-

~ 3 O Rosa Ct Rl. ~v I'~E~COf~ R I I I I I I I I I I I I I I lationship between steady-state cotinine and different infusion doses of nicotine found by Galeazzi et a1.3' In particular. a correlation exists for all the cigarettes with tar yield >5 mg (groups 2. 3. and 4), and it is supposed to be a small partial compensation for nico- tine only for cigarettes with tar yield <5 mg, but the findings are not definitive. In fact, in this case the number of the subjects is rather small because the population that smokes the ultralow tar delivery ciga- rettes is limited (Table III). However, when tar yields of cigarettes are reduced to these very low values, it is obvious that the measurement techniques used to as- sess human uptake need to become more sensitive. These findings agree with previous studies of Gori and Lynch33 reporting plasma codnine concentration measurements obtained in 300 smokers of ultralow tar delivery brands (1 mg tar yield or less). These inves- tigators found that. on average. the change in plasma cotinine concentration was proportional to the mea- sured change in nicotine yield for the cigarettes smoked. suggesting no behavioral compensation upon switching. Conversely, our findings do not agree with the stud- ies of Fagerstrom34 who suggested that compensation for nicotine is a real phenomenon for highly nicotine- dependent subjects. whereas subjects low in nicotine- dependence do not seem to compensate. In that study.34 for example. subjects were asked for a blood sample without any previous warning. In addition. nicotine increase was obtained by spraying nicotine ci- trate in the tobacco: consequently, most of the sub- jects rated the nicotine-enriched cigarette as bad tast- ing. Therefore many smokers who did not taste the usual smoke of the cigarettes reduced the number of puffs so that lower levels of cotinine were found in blood. There is a large number of pharmacokinetic param- eters and other factors that must be considered in a re- search protocol design. in the interpretation of data senerated by studies intended to measure or estimate human intake of tobacco smoke.3S•'b and in the com- parison of data from different reports. With a com- pound such as nicotine. for which tissue distribution is rapid and half-life is short. blood level is more natu- rally dependent on rate of uptake and clearance than it is to amount of nicotine taken up.=o Unfortunately. many of the experiments compare single blood levels of nicotine obtained a few minutes after smoking a cii;arette.'}''7 Cotinine. however, is a major metabolite of nico- tine. has an elimination half-life of about 15 hours. a ' low plasma protein binding and dose-independent dis- position kinetics. It is therefore more useful as a marker of nicotine intake.38 Furthermore, it is neces- sary to point out that blood samples have to be stan- dardized-taken after a period of time (8 to 10 hours) without smoking when this metabolite is at steady state.32 Moreover, the subjects examined should smoke the same brand of cigarettes for at least 3 months to avoid changes in the smoking habits. Daily consumption of cigarettes has to be regis- tered. Subjects in the present study smoked about 10 cigarettes per day, whereas subjects in the previous studies='•3t.33 smoked a higher number of cigarettes per day. However, in these studies the sample of smokers was not chosen from among the population; study subjects were recruited from subjects who had begun a smoking treatment program31 or were hospi- talized in a smokers' clinic=3 and, in one case. sub- jects who smoked fewer than 20 cigarettes per day were not enrolled in the study.'} Our study shows that. for a given individual. serum cotinine concentration is directly related to daily nico- tine intake under reasonably steady-state conditions. The implication of this finding is that subjects who smoke lower tar and nicotine cigarettes do not show a different smoking behavior. In fact. they do not in- crease the number of cigarettes smoked or the number of puffs. A partial compensation could not be ex- cluded only in the case of very-low-yield brands of cigarettes. In conclusion, persons who smoke lower tar and nicotine cigarettes inhale a lower quantity of toxic agents. Health benetits are expected to be coniirmed by epidemioloeic studies. We thank Simonetta Di Carlo. Antonella Bacosi. Anna Rita Passa. and Dr. Carla Balestreri for their contributions. References 1. Koop E. The health consequences of smoking: nicotine addiction. A report of the Surgeon General. Rockviile. Maryland: US Department of Health and Human Ser- vices: 1988: publication no iCDC) 88-8406. :. Chemistry and analysis of tobacco smoke. In: IARC monographs on the evaluation of the carcinogenic risk of chemicals to human. Tobacco smoking. voi. 38. Lyon. France: IARC. 1986:83-126. 3. Dawson GW. Vestal RE. Smokin¢ and drug metabo- lism. Pharrnacol Ther 1983:15::07-21. ~. Routine analytical cizarette•smoking machine. Defini- tions and standard conditions. International organization tor standardization. Geneva. Switzerland: Reference number ISO 3308:15>y1(E). 5. Cisarettes determination of total and nicotine-frce dry particulate matter using a routine analytical smoking machine. International organizanon for standardizatiun. I

I I I I I I I I I I I I I I I o< < .U a~~u,~.x s Gene\a. S~~ itzerland: Reterencc number ISO 4387. Iy911Ei. 6. Srlhe W.A, Recon.tttuted tobaccn sheet. In: Gon GB. Bock FG. ed.. A ,afe ctearcttc? Banbun• report No. 3. Cold Spnn, Harbor Cold Spnn_ Harbor Laboratorn he., l ork. lyKO:2U5-I ~, - Norman \ Chan_es in smoke chemtstrn ot modem da% ctearetteti. Recent Ad% Tob Sci 198'_:8:141-77. b. Baker RR. The eftect of ventilation ol cizarette com- bustion mechantsm. Recent ,1d% Tob Sci 19M-1:10:88- 15(}. 9 Wald N. Doll R. Copeland G. Trends in tar. nicotine and carbon monoxide vields of UK cicarettes manulac- tured since 19 14. Br Med J 1981:282:763-5. ]0. Rtmin_ton J. The effect of filters on the incidence of lune cancer in cisarette smokers. Environ Res 1981: 24:162-6. 11. Hammond EC. Garfinkel L. Seidman H. Leu• EA. Tar and nicotine content of cigarette smoke in relation to death rates. Environ Res 1976:12:263-74. 12. Lee PN. Garfinkel L. Mottality and type of cigarette smoked. J Epidemiol Community Health 1981:35:10- , , 13. The health conseouences of smokina: the chan¢in2 ci:- arette. A report of the Sur¢eon General. Washington DC: L'S Depanment of Health and Human Service: 1981: US Public Health Service. 14. Doll R. Peto R. The causes of cancer: quantitative esti- mates of avoidablc risks of cancer in the United States todav. J Natl Cancer Inst 1981:66:1191-308. 15. Borgers D. Junge B. Thiocyanate as an indicator of to- bacco smokins. Prev Med 1979:8:351-7. 16. Jarvis MJ. Russell MAH. Saloo.iee Y. Expired air car- bon monoxide: a simple breath test of tobacco smoke intake. Br Med J 1980281:484-5• 17. Isaac PF. Rand MJ. Cigarette smoking and plasma lev- els of nicotine. Nature 1972:236:308-10. 18. Matsukura S. Sakamota N. Seino Y. Tamada T. Mat- suvama H. Muranaka H. Cotinine excretion and daily cigarette smoking in habituated smokers. CuN PHARMA- CoL THER 1979:25:555-61. 19. Pojer R. Whitfield JB. Poulos V. Eckard IF. Richmond R. Hensle% WJ. Carboxvhaemo¢lobin. cotinine. and thiocyanate assay compared for distinguishing smokers from nonsmokers. Clin Chem 1984:30:1377-80. 20. Benownz NL. The use of biologic fluid samples in as- sessing tobacco smoke consumption. In: Grabowski J. Bell CS. eds. Measurement in the analysis and treat- ment of smokin_ behavior. NIDA research monograph series no 48. Rockville. Maryland: US Department of Health and Human Services. 1983:6-26. 21. Pichini S. Altien 1. Pacifici R. Rosa M. Zuccaro P. Elimination of caffeine interference in high perfor- mance liquid chromatottraphic determination of cotinine in human plasma. J Chromatogr 1991:568:267-9. 2-1. World wide use of smoking tobacco. In: IARC mono- graphs on the evaluation of the carcinogenic risk of ~160I111C 1T2rLIkC [rJICI 6ori11111C IC7'cIs '029 chemicals to tiuman>. Tobacco smoking. ~vl. 3S. Lvon. F rancc: l A RC . 1986:47-81. 23. Russell MA. Jarvis M, lcer R. Feverabend C. Relation of nicotinc yield of cigarettes to blood nicotine concen- trations in smoi:ers. Br Med J 19kU:'_80:972-6. 24. Herntne R1. Jones RT. Benowttz NL. Mine. AH. Ho%% a ctearette is smoked determtnes blood nicotine levels. CuN PHARAtACUL THER 1983:33:84-90. 25. Sutton SR. Russell MAH. iver R. Feverabend C. Sa- loo.tec Y. Relationship between cigarette vield,, puffing patterns and smoke intake: evidence for tar compensa- tion? Br Med J 1982:285:6W-3. 26. Hill P. Marquardt H. Plasma and urine change, after smoking different brands of cigarettes. Cu. PHARMACOL THEx 1980•?7:652-8. 27. Benowitz NL. Jacob P. Nicotine and carbon monoxide intake from high and low-yield cigarettes. Cux PHARMA- COL THER 1984:36:265-70. 28. Stepney R. Would a medium-nicotine. low tar cigarette be less hazardous to health'' Br Med J 1981:283: 1292-6. 29. Russell MAH. Sutton SR. lver R. Feverabend C. Vesev CJ. Long term switchinR to low-tar low-nicotine ctga- rettes. Br J Addict 1982:77:145-58. 30. Bridges RB. Humble M'. Turbek JA. Reehm Sr. Smoking historti•, cigarette yield and smoking behavior as determinants of smoke exposure. Eur J Respir Dis 1986:69( suppl 146 ):129-33. 31. Benowitz NL. Hall SM. Hernina RI. Jacob P III. Jones RT. Osman AL. Smokers of low yield cigarettes do not consume less nicotine. New Engl J Med 1983:309:139- 4'_'. 3?., Galeazzi RL. Daeneus P. Gugger M. Steadv-state con- centration of cottnine as a measure of nicotine intake bN • smokers. Eur J Clin Pharmacol 1985:28:301-i. 33. Gori GB. Lynch CJ. Smoker intake from cigarettes in the I mg Federal Trade Commission Tar Class. Regul Toxicol Pharmacol 1983:3:110-20. 34. Facerstrom KO. Effect.c of a nicotine-enriched ciaarette on-nicotine titration dailv cigarette consumption and levels of carbon monoxide. cotinine and nicotine. Psy- chopharmacology 1982:77:164-7. 35. Vesell ES. Penno MB. Assessment of methods to iden- tifv sources of intenndividual phatmacokinetic varia- tions. Clin Pharnacokinet 1983:$:378-409. 36. Darby TD. McNffitteaJE. vaat Rossum JM. Cigarette smoking pharmacokinetiFS andits relationship to st•nok- ing behavior. Clin Phatm 1984:9:435-49. 37. Erbert RV. McKendree EM. McCuster KT. Snow SL. Amount of nicotine and carbon monoxide inhaled bv smokers of low-tar, low-nicotine cigarettes. JAMA 1983 •?.50:2840-2. 38. Curvall M. Eiwin CE. Kazemi-Vala E. Warholm C. Enzell CR. The pharmacokinetics of cotinine in plasma and saliva from non-snSoking healthy volunteers. Eur J Clin Pharmaco1 ><990:38:281-7. I