Tobacco Documents Online

. jc1'9Lb,WC,,- (-, N~,b . Resiew Artick <.linical Pharmatokmetics 9_ 13SSI9,(It9641 0317-596319+10900-035407,3010' O ADIS Pfess Limiud, A'U rots reservcd. Cigarette 'Smoking Pharmacakinetics: and, its Relationship to Smoking Behaviaour Thomras D: Darby, James E. McNamee and' Jaques 14f van' Rossucm Departments of'Pdarmacologp and Pfiysiology, School of Medicine. Uniiversity of South Catolina: Columbia:'and Depattment of' Pharmarobgy. School of' Medicine. Uniwasity of Nijtncgcn. I+Iijmegen Sammary The )ield'of a cigarette is determined b,r the tobacco blend. th'e length' of the cigarette. the cigareu'te'paper the filter and air dilutiorr: Cigarette yield has bcen defined b}- tradiieon and, b;r lax^ to be the yldd of nicotine. tar and carbon monoside obtained from a 33m1 ' putl'rolume of 2-second duration taken e+•env minute during the burning time of the cigarette. \armall•i• smakers' draw a puff into their mouth and then inhale. Sfouth deltrer!t• is : largeh- determined'b3 ptrrsonalantolGingbehatiour. The'puff volume. numberalpullstaken per c+garette: and number of cigarettes smol:ed'per dor determine both'the volume and, the mass: qf' daila mouth, delitmr. There are marlced i dllferences in smoking behariour. , and the delivery ts substantiallv altered'from the }•icld ratues obtained with the standar- dised trsr procedure_ Bodr uptake o•(smoke tngre dients is determinud by smoke ch'crrrtcall parameters: smoker inhalation behariour, lung morpho%n•: and phtstoiogica/ para, merers: ThephnYtological parametersiruludetidal volume. rualicapaeut: rateo•fbreath< ing; and rate of clearancc.for the lung; Given these'behartoural, and phtsiolagiraldifferrnces in indiiidual'drafiren•and'uprakr it'is not surprising that dijTeronces in measured parameters occur'wtthtn smokers ot ctR- aretres wyth a particular .rield: Biological dilTivenees among tndirrduais. such as metaf+olic : and size di/_farcncus, cause additronal'aartattons in thrse values. Thcmtitrc. the estimates ot'mcottne and'tar deGven-can ran• wtdch• in ,rutitcs of tndnvdual uptake N'hcn :he estimates are hasnd, upon sample populattonidara: The variables in both' smoking beharuour and in chemical and'phYstologual't•actars which ahrrr uptake'mak'c it essential to h'are a crossover dtstgn for ant, studr.. Thc'largc' standard error ror the riasraa conc•enrratrttn uLitrrurttnc• ta matur rm+tahulue of ntaottnct Kvrhtny sarrrTle,nanulutaon. .tnd'thc log linear nature tq thn rlasntu mnntnc uonucntratron .unc. mqutresa vern'laren'samplcsrrofiir ant, sruvnc!f,-itarcrtcddtrcn•nrr.7ro>•c 11-:ecn comparisons or hrands arc rnadc. avaragr ra/ucs are rnrsli•adinc in thut the si.a-" fi, the high values rhscures 1'nt-yuenct' dr!ti•rcnics amrrnr the !owt+ %aiucs «tthtn thc sarnpirs l:' is important tn remember that smrucr cttrnnlrant'r -uh svu.it- Ja•sttwa is rtrn• . sscr.rwl: h' 11 rtnuld hh• rrnn,rauNc rn knrtr,- that tndnrtduai srn.ri,t•- , rddt d':.• rcrrtsrn •rn the rre..'r:h.rd rPCtOnc!t idr:nC' S:uJi :hut duemt'rrd th nate .m.•wr•• ,:at1' • rrdlJ ••r'lnti• .w rt.'r: :., lOV If4li••r.r r.r t..u.~ .7•mr hn 1rJnd..: J11'Jntrrrlr<• "rl/11 /l 1 a'.•rriPJL'tj i•t. r.• tn .'IIJ. n:a• •reJt W ~

Cigarette Smoking Pharmacokinetia 4366 the redttcrion in sales-weighred atrrage tar yield oJ30i+tg to 15mg has been acooetpanibd by a decrease in the tncidence of those diseases reported to be increaseadi in the srnoking yopulatton over non-smokers. Serrcal studies have sf+o.m a ddse-nsponse relationship for the number of cfgaretres smol;eid'and lung eancer.7Jte dose-nsponu relationship for cardiorascular disease is less dear: A major part of the reduction in these disease states coould,be related,to reduced numbera ofs+nokers per 100.000 population. As agarene yieJds decrease to tar values near lrttg. measurements or tar and nicotine uptakr miest be tntprored P; The interaction between a person and a cigar- ette is'a complicated event. Wluie exposttre is dif- ficult to determine in many instances, the smoker~'s exposure dose of nicotine, carbon monoxide and the 300 to 500 compounds contained in the 'tar' component of smoke has eluded measurement: Smoking behaviour is obviously an imporrtantde- tertninant of smoke uptake: Biowever, cigarette yield andiother faetors aiso appear to be important 1: Cigarette Yields and Smoke Uptake Cigarette yields of ' nicotute, tar and carbon monoxide included in advertisements are. either by law orcustom, measured'by a~standard smoking machine procedure (Guerin. 1'980; Surgeon Gen- eral, 1979). Over the last 30 yeats this rnethodol'. ogy has been used to demonstrate a substantialre- duction in the yield of cigarettes sold on the United States market, and this reduction, in yield has oc- curred in other countries (Hjern, 1976:' Surgeon GeneraL 1979; Wald et aL 19811). Concomitant with use of 'thexc lower yield' cigarettes there has been a rpeduction, in the incidence of those disease states reported to be higher among smokers than among non-smokers. A reduction in disease incidence has occurred in, spite of a reported smoker compen, sation for the reduction in cigarette rield (Doll and Peto. 1978: Garfinkel. 1980: Kannell l'981k Kunze and Vutucz 1980: Lynch and Gorii 1980] W'ynder. 1980). The US data substantiates a reduction in the number of both fcmalf* and malt smokers oser the last 30 %cars (Harris. 19831. Thereforc. a major pan of' the reduction in those disease states couldl be related to reduced numbers of smokers per 100.000 population in the C'nrted States O1".ndi-r: 1!98(l)j The reported teduction in cigarette yields is largely the result of f ltration of the smoke (Keith. 1975, 198% Smoke is composed of a gaseous phase and a hydrophilic liquid particulate phase (Guetin, 198Q;Surgeon General, l'979)~. The part of smoke that passesahrough, a glass-fibre filter that removes 99.996 of all particles i= 0.1 fun in diameter is de• 5ned as the gaseous phase'(C,uerinL 1980):17te ma- terial thatis trapped by the filter is defined as 'tar', total particulate matter (TPM). or 'condensate'. Conventional cigarette filters reduce the mass of the particulate phase which enters the mouth. but do not change the size of'thr particles generated by the.burning cigarette (Keith: 1975, 1980). The gas gqnerated' b.• but4ting', is minimally affected by most: cigarette filtration systems. More recentliy, air dilution of cigarette smoke has been used to reduce the potential yield of both phases (Selke and Mathews, 1978). Air is iatro. duced! into the smoke tHrough, the papersur, ound- ing the tobacco of the vod, as well as through the filter paper wrap. Air introduction redtrces the puff volume obtained from thrburning,tobacco and thus substantially reduces the volume andi the concen- tration of the tobacco smoke which enters the mouth: Standard I smoking, machine yields of high air dilution brands (80 to 1 8596) 1 are reported to be Img,of tar or less. Brands with yields of 6mg, tar or less are referred to as 'ultralow' yieldl brands (Guerin. 19802 Keith. 1975. 1'9801 Surgeon Gen- eral. 1979); While the cigarette filter and air dilution has re- N duccd ctgarette %. telds as measured by the standard' O smokmg,machinc test. the question remains: -whati N reduction occurs tn rndisidualluptake of~smohe tn- ~ grcdicnts as a resui; of'swttchtng to a brand thauw, ~ ~ w ~

Ciganetu Smoking Ptuumaeokinetia' N tap. 1. Struetw and ibeisstion a I ereoon.. N'aotiiw is aDII :,se.y to+: a tn" para«daoI w.gnt a c;gar.e.,snrok.. II dmu.noes chany. wsip«oaltsp.. n ts monoarotonat.b aamost pnyswio~ DM ivatws. The dtprotonaced ion woukl,ewist at pH vawes rouna in ma stI Mweolism a tarq.y'dw t,o ox- idaI coa,iwn ama;Qr rmta11011. I,ow.w.rI ttw» ,.R at Mast 4 pr.nuY maa0olit•s of eioaoit.. pro.rides sucbz measured reduction in yield?' There ane a large number ofIpharmacttitinetic parameters and' other' factots that must be considered 1 in re- search protocol design and in the interpretation of data generated by studies intended to measure or estimate human intake of any chemical I ([Jarby„ 1978: Goldstein and Kapman, 1974; Rowland and Tozer 1980):1ndiMdual uptake of smoke. and thus its ingredients. is determined!by'a surprisingly large number of factots. Many of these factors can affect both the particle dose and the dose of any ingre- dient contained' in the smoke (Brain and Walberg, 1979: Guerin, 1980). The following sections dis- cuss the speeific', relationships between factors of eithera chemical orbiological'nature, and those of smoking behaviour on uptake of nicotine„tarand carbon monoxide. 2. The Standard Smoking SYrzch'ine Test The standard smoking machine test'does not and is not, intcndl+d to measure human uptake of ni- cotmc, tar, or carbon monoxide. It is intended to 437 give a relative rating of'difterent cigarette brandi yields. The test uses ai standard puH•consisting of' a 35m1 volume drawn in over a 2-second period.. The frequency of thepuffs is one each minute (Snr- geon General. 1979}. On average, thi's resutts in 10' putTt per cigarette, but numbers of puffs can vary from about 6 to 12 (Surgeon General, 1979): A glass-fibre filter is used to trap smoke particles above 0.lwn in size (Criteria; 1'980.): The weight of the particulate retained on the trap is reprcsettta• ti>re of the weight of the particulate that wotdd be delivered to the mouth and, with inhalation, to the respiratory system~of the smoken The particulate material represents approximately 8% of the total mass of cigarette smoke. During, burning of the cigarette, smoke is pro- duced by distillation and pyrolysis (Schmeltz,1!972) j The gaseous phase is composed of oxygen, carbon dioxide, nitrogen and its oxides, carbon monoxide, hydrogen, cy,attide, and very volatile hydrocarbons and carbmnyl-0otttaining compounds (Gtterin,19sa Surgeon General. 1979), Water and canbon dioxide generated duting burning are ttappcd Ion the glass- fibre filter, and the pH of the filter pad becomes acidic (pHI5:8): ?' 1 I Measurement of Nicotine and Tar Yields The structure and ionisation of nicotine ane shown in figure 1_ Tliciutine„ which is one of thee few liquid alkaloids. is composed of a pyndine ring and! a pyrrolidine ring with a methyl group at- tached to the nitrogen atom. Protonation of the pyrrolidine nitrogen is' important to the under- standing of nicotine chemistry. P}•rRolldinc has the basic strength of an ordinarv secondary amine and salts are fornned with carbonic acid'and other sub- stances contained in tobacco smoke. In the tobacco plant: nicotine is found as a salt of'either citric orr malic acid. The boiling point for nicotine is'_8!9'C. Thus, as the tcmperature rises in~ the area just be- hind the fire cone, nicotine base is dtstilltd. Iwi- cotinc in smoke reacts with, carbonic acid in the smoke or on the filter trap: As nicottne dissolves, the pH of the water solution, increases. Thus. op% of the nicotine in the maensnroam smol.c - that

Cigarette Smoking Phanmacokinetia smoke which ~ would' be taken into the mouth - is trapped in the acid water, partides or droplets on the filter. In the standard smoking machine test; condensate or 'tar' is determined by weight, after nicotine and water are extracted. Nicotine is de- termined by chemical analytical procedures. Thus;, variations in the ratio of nicotine-free condensate to nicotine can be determined. 2.2 Limiiations of Derived Data. With the ulttalow, tar deliivuy products. consid'- etable difficulty is encountered in accurate meas- urement of nicotine and tar yield. The methods used to determine both nicotine and tar contribute to this difficulty. Repot•ts of'results from several intra- and'intetdaboratory, comparisons have shown the marked differences in results obtained'for tltesee product& [These data are availablrfrom repotxs'of Working, Group 3, TC-126,Internationai Stand- ards Organiradon(ISO).] The deviations which are acceptable when the standard methods are used ( i mg tar and 0.5mg nicotine) show the lack of sen- sitiivity of the measurements (Wynder and Hoff- manm 1967). Given these dill'et•enees, cigarette brands ranked as 0L8mg nicotine and $mg tar yield could be identical to cigarette brands ranked as. 1.0mg nicotine and U0mg tar. Further. 1.0mg ni- cotinrand 10mg tar products could be identical to 1.2mg nicotine and 12mg,tar products. Thus, shelf products over the range 0.8 to lL2mg nicotine and 8 to 113mg tar, could have smoker deliveries that are very similar. In addition, changes' in pufi"vol: ume alter the tar to nicotine ratio of the trapped particles as.well as changing the amount of smoke deliNet•ed to the trap (Schmelu, 1972). Tliese con- siderations are important in interpretation of data derived'from qiellls provided in.advcrtisements or stated oni the package. In comparisons of standard smoking machine tesr yields with irrt rit•o tests for deliiven•4 a carefully controlled smoking machine test should be used to determine the yield of'the particular lot or batch of the product used in the study. Howe.•er. it must be realised that the multiple factors that affect up- taCc ofntcouncandlvhe other tngreditnts of smoke 438' would be expected to cause considerably more variation in daily nicotine intake than that whic7i could be expected from multiplication of standard smoking machine yields by number of cigarettes smoked:. 3, Fltarmacokvinetic Considerations in Uptake of Smoke Iitgredicnts 3.1 Factors Influencing Absorption: There are chemical and physical aspects of smoke as well as.lttng morphology and physiolog- ieru Pactors that contribute to the uptake of the in- _. gredients of smoke (Albett et' al„ 1974; Austin et al.. 1979; Brain and Vytlbet+g, 1979;, Gttetin, 1980; Surgeon Genetal.1979; Yeh et a1.1976; Yu. 197$): 3:1:1' Smoke CompoiiuonlTvpe of Cigarertr Cigarette smoke is a complex and ever-changing mixture of gases and panides proditced! by distil- lation and pyrolysis within and adjacent to the burning coal of the Ht cigarette (SchmdM 1972; Wyndtr and Hotfmann: 19b7). Brand differences in,tobacco blend, additiives, wrapping paper,lengtli of the tobacco mod and filter design all substantiallv modify the contents of smoke (Guerin. 1'980; Sur- geon Genc7aL 1979; Wynder and Hoffmann. 1967)j As a smoker draws a partial vacuum: on the cig- anette, the materials produced at the fire cone are successively filtered first bv unburned tobacco, and then bv various filter materials before the smoke reaches the lips. Rapid cooling of smoke takes place within the first few centimetres behind the fire eone; thus distilled products are deposited on the to- bacco. The initial deposition and tobacco filtratiow process causes a continued, change in the ingnedi- ents and their concentrations in smoke that passess into the cigarette filter_ The smoke which enters the mouth contains particles that are between 0:01 andl0•4µm in size and the pH is slightly acidic (pH 5 to 6), [Guernn. 1980: Sut•geon! General. 1979; w'ynder and Hoffmann. 1967E j. t: ' llorph ological ond Ph rsiologtcal Factors Even though smoke passes the lips and enters the mouth. the contents eannor, be constderedl to

Cigarette Smoking Pttiarrttacokinetics have entered the body until an epithelial surface has been crossed. Some small amount of matter from cigarette smoke will contact the mucosal sur- faces of the otai and nasal cavities (Surgeon Gen- ecal. 1979): This surface represents a small fraction of a square metre of'potential absorptive area and accounts for little measurable uptake of substances from cigarette smoke (Guerinl l9'B& Schievdbein; 1976: SitrgeonGeneral, 1979). Moreover, the acidic nature of cigarette smoke causes most of' the ni- cotine mass to be ionised and' these charged mo- lecules do not cross epitheliat! surfaces. It is not until I smoke particles and': gases enter thrremainderoftlirrespiratorye tract that'the bulk of substance exchange oceurs: Conducting airways di vide as they lead inspii•ed, smoke into the lung. With each new division of the respicatory traet, the potential surface area for exchange of smoke par- ticles increases substantially until at, the alveolus approximately 70 square metres of moist single- layered epitheliall surface~ is available for contact. The ingredients of these small liquid' particles of smoke and the gases pass through the epithelium of the alveolus and thus enter the body. A review article by Brain and Valbetg (1'979) calls attention to the fact that' the current state of knowledge prevents calculation of the particulate dose in man. Some of the factors which affect dis- tribution and1 deposition include: the tidal volume;, inspiratory and expiratory flow tate, vital capacity, litng, morphology, and particle size and! density. While the number of puffs taken, puff'volumeff andd depth of inhalation are imponartt, it is obvious that many other parameters aiso deterrnine the fate of particulate matter taken in through a smoker s mouth. Distnibution and deposition of particulate matter will affect the site and rate oflabsorption of the ingredients of smoke. Particles of the size found in cigarette smoke (0:011 to 0_4k m) are initially deposited only mini+ mally in large and small conducting airways Molmberg, 1979): According to Austin~et al. (1979) less than 1% of administered particles below -34m in diameter are deposited in an airwav '_mm, or larger. Duration of inhalhtton. breath holdiiog, and other pro:csscs that sloN gas and aerosull movc- •39 ment will increase deposition and certainly affect distribution of particulate deposition (Goldberg and' Sinith; 1981). Since a variable amount of smoke is exhaled with, each putL and distribution and deposition of ' the retained smoke particles varies. it follows that uptake of smoke ingredients will vary. Lung clear- anceaates for deposited materials; and the amount of' matter expt:ctotated! or swallowed, and! ulti- mately absorbed from thegastrointesti.nal'ttact will change bioavailability: Uptake of the ingredients of smoke willldepend upon solubiility; however„most ingredients of smoke are water soluble (Guerin, 198% Naturally, the degree of ionisation, is also important to rate of absorption. Uptake at the lung would'be inhibited by the relatively acidipH arthe alveolus kvel'. Thc lttrge amount of carbon dioxide which is metabolicatly produced, and produced byy burning of the cigarette, would result in an acid pH: Giiren: these: considentions: it is easy to sne why calculation: of smoke aerosol dose, even when aerosol physics and chemistry are known, is gen- eralliv considered l impossible (Brain and, Valbetg; 1979; Surgeon Generall 1979). "ile there is Ctttle doubtithat voluntary control of puff volume and depth of inhalation provide a means of compensation for changes in cigarette yield as measured by the standard smoking ma- chine procedure, it is also obvious that other fac- tors control uptake of nicotine and'. the other in- gredicnu of particulate generated by cigarette smoking. 3.2 Nicotine Pharmacokinetics Measurement of nicotine plasma concentration has been usedl to, determine delivery of one con- stituent of'smoke (Benowitz et al.. 1982ai Russell et al:, 1'96'_:Surgeon General. 1979). Interpretation of the data dene•ed'from these ecperiments is com, plicated by both analxtical problemsassociated with deterrninatimn of the low concentrations of nico- N tinr present in the plasma of' smokers-and by the (~ ~ complzs pharmacokineuc considerations which must be taken into account. Table 11 pron tdrs phar- ~ maiokanetic parameters for nicotine. Improvc-w W ~,

Cigarette Smoking, Pharmacokinetia ments in measurement procedures and in our knowledge of data analysis has provided changes in estiinates of nicotine half-life . fnom vahtes near 30 minutes to the better estimates shown in. the table. 3.2.1 Varrnbility in Rate ofdbsorption ~ and Plasntn Nicati.re Concentrations The peak plasma nieotine eoncentration is re- lated'more to: thesptxd of'input or absorptionnte of the substanoe .than it is to the dose administered (.(',oldsteinand''Kapman,1974; Rowland and Tozer, 1980). Under steady.state conditions, where input is equal to output, plasma concentration would ret llect inpul Distribution and' dearancae rates also have a profound'eSect on the plastna'conotntration over time. Basic amines, like niootine, are rapidly dis- tributed to tissue compartments. The area under the plasma concentration-time curve (AUC) for a given input~ is used to detetmine the dose admin- isterrod for a given individuaL The wide variation in apparent volume of distribution and clearance reported by Benowitz et ~ aL (1982b) ~ and the fact that there was no correlation between volume of distribution and clbarance (inithat individuals with a larger apparent volume of distribution did not have a higher clearantx and those with a small ap- parent volume of distribution did not have a lowerr ciearance, as,would be ezpected'if half-life did not, var+y.videly)'provide evidence that for a given dose. plasma concentrations of nicotine at steady-state in, individuals can vary substantially. According to Cohen.and Row (''1981)j nicotinee is completely monoprolonated at pH 6.1 and di- protonation takes place at lower pH values such as those that: ocnir in the stomach. In a review by Schievelbein (1976), plasma, concentrations of ni- cotine were reported to be considerably higher when nicotine plus smoke condensate is basic (pH 10.8)~ than under conditions near normal plasma pH l val- ues (ipH 7:4): At pH 6.5. nicotine absorption iss nearly zero in the mouth. Thus. it is not surprising that very little absorption of nicotine takes place in the mouth with cigarette smol;ing: Attempts to correlate ntcottnc content of'smoke writh, lung ab- sorption is complicated by differenees, in the aqueous part of the smoke and by changes in tes- p'uadon attributable to the : pulmonary irritation properties of some tobaccoa~: During,ciganette smoki,ng, transfer of nicotine from : the alveolus to blood is not fiully understoodl It appcars that nicotine can be taken up by either absorption of'pattides or by leaching of nicotine from the deposited hydrophilic liquid particulate (Guerin, 1980; Surgeon Generalj 1979)J Given thee acid pH' for the alweolarenviuonment, either thee small size of the particles allows them to be trans- fetred by molecular diffusion across the alveolal membrane, or an associated salt form of nicotine is transferred as an uncharged molecule. Those particles containing nicotine adhering to the mu- cous layer of ciliary airways will be cleared from the healthy lung in a matter of minutes before ap- preciable exchange can occur (Yeh et aL, 1976). . Particles carried beyond conducting airways wt7l be cleared more slowly by marrophages with movement into bronchopulmonary lymphatics or eventually, movement towards ciliated airways. Thus. systemic availability of nicotine will1depend on the rate of ]eaching and rate of particulate ab- sorption. Naturally, nicotine concentration, pH'~ Table L Phamucakir>.oc p.r.rtrtus of nwotxv tmear s SEFq in ,14 suDjects• VC CL t.a, _ tl1' tL/-l /hl 163 _ 49 109 z 02E'. t i98 x OJ73 a mata7rom Benowttt of aLi (1982b). The su0jects~.wre twrmai i healthy young maros. Assuming an average Uoajnwergnt ot' 7okq woua ! pWOrde a; volume ot' aste4wtton of 2.611 l./rtq and a.dearanos of 0.0161akq/rmn~ tl'.snoutd be noteo tnat, tne SEM /er tMso wdues a.bout 25% of the mean value. n the.case.olitM hatl,Wte vaAte.tM ~SEM n 3T'4 0l tM mean vatw: Ths vQ arb'CL do not cctnreute. Th.reior., tMre is asubstandat EWfterenaa - .W-Cuar steaoy-state apparent Mdume oudisnabueOn and cdearance of Mcotrrle:. A'oYrerat.onr.Vtl - apparent ~vowme of d.sttitxAwn: CL - ciear- anca, t ~ - eism/nation natt-atr-

Cigarette Smoloin` Pturtnacokinetics factors, and other favourablt physicochemical as- pects of the environment will'contiibute to rate of nicotine absorption. Once the particles impact in: the respiratory tract they lose~ their identity„ and, the chemicals which make up the liquid particles. induding nicotine, are taken into the body as in- dividual chemical substances with varying rates of absorption (Guerin, 1980): Only a few solid parr ticles or poorly water-solubk particles are found in cigarette smoke. Lessr soluble vr poorly permeable molecules may penetrate tbeepitheli.al junction and also gain access to the body: 3.22 Blood Clearance and Tissue Distribution Organ and ltissue distribution of nicotine is more dependent upon differences in pH, than upon lipid solubility. At physiological pH values, small changes in pH cause substantial changes in the un- ionised nicotine 5raction. Ljpidisoluble basic amines such as nicotine are known to produce higher tis, sue concentrations than blood'concentrations: The lipid solubility promotes passage through mem- branes (Goldstein and' Kapman, 1974). This fact most likely aocounts in part' for the large apparent distribution volume found in studies conducted'by Benowitz et a!, (1982b). Armitage and' co-workers (Armitage. 1976; AAr- mitage et aL 1975) reported i marked differences in arterial blood concentrations of nicotine between inhalers and non-inhaltrs. In addition, these in- vestigators also reported that t"C-nicotine given intravenously (antecubital vein) disappeared'.more slowly from arterial blood than when the nicotine was absorbedi after smoking. Although Armitage (1976) suggested that variability in metabolism (see alsosection 3.2.3) might'explain this difference, he concluded that itiwas more likely to be related too differences in mixing with circulating, blood vol- ume which occurred following adtninistr-rtion by the ?' routesj Ntitutnlly; the transit time from, the lungs to aorta is shorter than from the intravenous injection site to the aorna: With a variable input. as would occur.with smoking, or with slow mixing. one would expect tluctuations in blood concentra- tion. Higher peak eoncenuamons m lhe brain and other tissuos %%oultl also be expcctcd with smoking 441 than following intravenous injections. With intravenous administration of t"C-nico- tine, the proportions of' t"Ccotinine. a major me- tabolite of nicotine, differed widely between spe- cies (Turner, •1'97b): Ditfert:noes in blood fraction disttibutionbetweett species might account for these differences. Intragastric administration produced low blood levels of nicotine and higher levels of coxinine- This could be due to differences in firat- pass metabolism, ini the liver (Vesdl'W and Penno, 1983). Nicotitx contained in the blood lipid frac- tion would be deared by the liver. 3.2:3' Metabolism and Excretion McKennis (1976) reported' that nicotine is ex- eneted partiallyunchanged by the kidney, but'largelyy in the form of'20 or more distinctive metabolites that contain an intact pyridine ring. Progressive metabolic oxidation of'the pyimlidine ring of ni'- cotine leads to the formation of cotinine (3-pyri- dylaeetic acid) and a: variety of other compounds. Since nicotine metabolism leads to at least 4' priim, ary metabolites, and I since at, least 2 of these path, ways are reversible, variability in plasma nicotine concentrations between individuals at steady-state could also be due to differences in metabolic clear- ance (Vesell and Penno; 1983)j Urinary pH is known to be a major determinant of the rate of nicotine clearance: by the kidney (Cohen and Row, l98'lk Surgeon GcneraL 1979):. During the night, when sleep acidosis causes acidic urine: clearance of unchanged nicotine would be greater than during,the day when diet, exercise and other factors cause tlitctuations of urinary pH. Ni- cotine excreted in the bile would be reabsorbed. Reabsorbed nicotine is more rapidly converted'to cotinine: as found with ingested: nicotine in the studies of Turner (1'97tn): While some of the more recent studies of intra- v.enouslh- and orally administered nicotine have cited the chemical form, several of the iudies re- pomed here did not include this information. ;`la- turall%, salt forms could affect, distribution, andd clearance, and calculations which do not, state 'as nicotine base' or 'as the salt formi prevent inde- p%:ndent assessment o1 the darta~: •

Cigarette Smokint. Pharmacokinetias Trae t< Ptwmaooka+.x wramet.r: of ooavr ~Im.an _ San :n t{ wDj.as4 vd <x t„, (LAO) ""W%"ml 1_1NI c 0.20 0.00094 s oi0o01s' .15.! 1<lo a oan from Baoowia et at.' (1993by: TM SIiAMS n ttbese da4 are tss vanatlN than tAat saen with nicotin.: TM apparent votuma of dsaieuuon was :rqtwy yreater ttvn bodyweiptK fAanat atearanoa was inareas.d by 50% by, urinsry, aaditlr aocn. CaorwN was adnwnntereaas tn. itumarat. isatk how- ew.r.,un0er oaWitions a stWOy-state maasurea»nt of tt» pharmaookirwtie prameters; this ftot sAaikt not atter tlw newlts trom itAosa expected iwhh totinin produced ifrom ni- ccoi». 3:3 Cotinine Pharmacokinetia Recentlyn plasma eotinine determinations have been used to estimate daily nicotine intake from smoking (Gori and Lynch, 1983; Hetning et al., 1983). As stated above, cotinine!is a major metab- olite of nicotine. Cotinine excretion via the kidney is somewhat variablc(Cohen~and Row, P981'), and its major route of elimination is by hepatic metab- olism. The half=life of this compound is at least 6 times longer than: that of nicotine (Herning et al., 1983). Table II shows the reported phartnacokin- etic values for cotinine. The apparent volume of distribution appears to ibr slightly greater than total body water volume. The extent of protein binding is reported to be less than 5%. Variability in individual conversions of' nico- tine to cotinine andl differences in individual clearances of the metabolite certainly complicate interpretation of nicotine uptake dose from meas- urcments of steady-state plasma cotininc concen= trations. TNe information given in table I'lI is im- potnant to these considerations. Mch:ennis(1976) reponed studies in the dmg_and the rabbit'w•hich indicated thac.reactions leading to cottntne-h-oxide are at least panly re+°ersible: This changes the size of the available cotininc pool. While under reasonable saeady-state conditions this factor woulii nrnoralTcct the individual relationship ui couninr plhsma concentratton to Jad% nicozin: 142' intake. individual differenoes in pool size would be important to studies which attetnpt to make inter- individual loomparisons (Benowitz et al':, 1983a). It is known that enzyme induction occurs in humans. Gorrod (1976) reported that femalr smokers ex- crcted! less nicotine but more cotinine than 1 female non-smokers. Male smokcrs'eould be divided into 2 groups: those who gave high total recovery of nicotine and cotinine- and those who pve lower total recovery. The first group excreted'much moae cotinine than their non-smoking cottnterparts; whereas the latter group exaeted rather less. I!sow- ever. the statistical significance of the data gained from.these small popuiation~stttdles must be ques- tioned. Variations in site and raate of absorption: of ni- eotine can affect first-pass metabolism in the liver and' thus' atfect both the rate and amount of ' ni- cotine convated to cotinine and'nicotine clearance (Breimer; 1983; Veseil and Penno, 1983). There- fone, it is suggested, that studies which use plasma cotinine concentrations as an approximation of'ni, cotine dose should use a paired self-matching de- sign and the relative particulate delivery of the cig- arette brands should be reasonably uniform. Studies conducted over several weeks with weekly com- parisons of plasma cotinine concentrations have the advantage of greater assurance of steadly-state conditions for comparisons. Where the investigas tors attempt to predict. tar intake based upon ni- cotine uptake; using plasrna cotinine determina- tions as a marker! for nicotine, the multiplicity of factors which affect accuracy oftho estimate of ni, cotine intake; individual differences in smoking pattern bchaviour, physiological parameters: and finally the physical and chemical considerations re- lated' to cigarette ield prevent the attainment of reasonable accuracy. 4+ 1bf11re+tce of Smoking Behnziotrr on Uptake of Smoke ingndients Several rt•.•ie.• articles claim smokc uptake by the tndiM1•,idual!smokcr is primarily related to smok- ing behaviour (iR'usstll et al.. 1982; Surgeon Gen. cral: 19'0: Tubin and Sackner. 19$').

Cigarette Smoking Pharmacokineucs Tab1.IN. StrWy-stat. plasma carc.nwatans of nrootirN an0, oo6rrw "nrq to0trorxjous nioomn ntaka by ampkitq fnowieq aM irwbonsApo t>•twMan Qlasma 6o0nirw eonCNWaOon atW up- taw of niooerr. aonv.rsm of MO+a to eooniiw. ano iauar- anoe of cean«e Mrroooa» pasma rrw X r.cwn .paoew. ealo,ntr,0,n - Mmen. wawo. cofinin., ptaama Yraw : tacson.o.ae.o !x tsc.on avwaw0 aonc.naation - Ca..r,wraa rw/d - ttws dos, of 'nioo0na n tlw smoka (mp/1k1. Titis factor is dapMWMlton (tN.typa OfGqanatta. Fr.etion a0soabwl- Ithe fraction ot'n» niootanrtnaW iabunfi.d~ Ttiis factor is dap.n0ant on smokrg bMaviow antl pfq%iob9- ecu and morptrotogM,M consklsrations. ficwtN. alNrane. - aNaranas af nroodiw. This taetlx is sutt >«aa•v«b«w f=~.ction convKt.d - ftw kacnon ~a nqoena itonwrtW in0o co- onn.. nrs factor is suej}cxOap.rw«n: Gunnin. Waranor - daaranoe of cofirw+a. TAes factor is sub iect-depeoa.nt. q is obvious fhf+l >wrDj.adliparb.nt t.ctlxs:anl vary impartant fn~Caparmeninp oo0nirrplasmaoOnGanVation at a yi,wn,0osa. tYna.r s+a.ayTata» condieons, tAe pasma oonoantraeon of eo- m,.,, for rgiv«+ .,aMa" aaa w procoroonilliw nw t>:o- availaDle doas a niootx+.: Russell et al. (1982) measured! the number off cigarettes smoked, puffing pattertts, and mouth de- livery of nicotine, which was determined by ana- lysis of nicotine content of cigarette butts collected: from the individual smokers. It is generally ac- cepted that as mouth delivery of nicotine is in- creasedi so is the nicotine content of cigarette butts (Surgeon General, 1979). Plasma nicounc'and co- tinine concentrations were also measured. The parameters were measured before and after switch- ing from cigarettes which have average standard smoking machine yields of 1'7amg tar and 1.33mgg nicotine, to low, tar (10.0mg) and low nicotine (0:7mg) cigarettes for l2 weeks. 12 subjects were srudiedl While most of: the measured' parametcrs did not change: there was a measured reduction in mouth delincr%- of nicotine (;1S% reduction in cig- arcttc butt content of nicotine) equal to the re- duccd nominal smoCrn¢ machinc tiIc,kllof thr ncw 443 cigarettes. However, plasma nioatitte'and cotinine concentrations were reduceid less than would be ex- pected from the reduced nominal'yielli It was eon- duded'from this small sample of smokers that sub- jects compensated to some extent! by increasing inhalation, On the other hand, Tobin and Sackner (11982) concluded that puff volume change was the majorr method of compensation by smokers when they switched from high tar, to low tar cigzrettes Data gathered from: pkthy,smogtaptiy testing where smokers acutely switched from high tar content to low tar content cigarettes showed that smokers consistently draw latger puffs, thereby circttm- venting, the preconceived benefits of smoking loww tar content cigarettes. These investigators also ob- served that cigarette holders modify smoking pat- terns. Hence, if' such holders arr used with flow meters to monitor the pattern of smoking, spurious data can be obtained. An increase in the numberr of cigarettes consumed is genetally accepted as a factor that increases smoke intake. These investigations indicate that number of putfs taken per cigarette, puff volume, and depth of inhalation arc also factors that increase individ- ual uptake. However, the conclusions rcached about the method, of compensation (puff behaviour change .ersus inha'lation change) are different from those reached!by Russell et al. (1'982). In the Tobin and Sackner study, mean puff volume measure- ments obtained from the 10' subjects were 52 ± 15m1 while they smoked'the low tar content prod- uct andl39 ± 1om'1 while they smoked the high tarr content product The volume of air inhaled with the lowv tar product was 12% greater than the 748ml volume obtained with the high tar product. The standard deviation ofl inhaled volume was ex- tremelw• large: 517 and 323m11 respectivel,v, for the 2lproducts: These and other studies place emphasis on the effecnof behavioural factors as determinants of human uptake of'tar, nicotine and carbon mon- ocradr as well as the other constituents c.: smoke in studies designed to detertnine the relationship be- t-ren a reduction ini smoke rntzke and the inci- dence of diseases associated with smoking tSur- grun Grnrral. 19"0l .

Cigarette Smoking Pharmacokinetics Hbwever, the large variations in puff volume and inhalation volume and the limited number of observations cast doubt on these and other studies designed to determine the relationship between a reduced standard smoking machine yield and hu, ntaann uptake. Where puff volume variatipns otanu, use of smoking machines capable of being pro- grammed to mimic human mouth delivery mightt provide more comparable data' for deternninations of actual smoker uptake. However,,that'is not the intention of the standard test procedure. Hofacker (1983);, using analysis of random data, found that'noise' can!easily lead to statistically sig- nifieant rtsults: This author states: "In particular, performing all possible 't' tests among numerous groups. calculating many correlations and circiing' the ones with 'p' valhes less than 0.03, and per- forming stepwise regression all can easily lead to. Table IV. lsnar regression i andysis of'tM data of Gorn and Lyr,a, tnse3r Depsna«>t va,taa. 1 e.atw C.Mr,sw tlrand A txand n Number of'observations 106 96 stope ose aas tnterayspt 90 24, t.orretaoon caeffiasnt 0.74 0:79 Slope wM zero,iMereept 1:34 0.93 a Since aoorrne is only present as a result of nicotine metab- olism: , tne zero pont a very vaWd. NTiN C+ere is a oonst0- erabk range of plasma ootrMne conoentration. a matcnetl- parred desgt,nwneraeach smoker smoked.all 3 brands for atilaastl2 weeks allows tniscomparrson: ft is interestyng that tne , 1.34 srope for brand 'C'vs brand Wis altqhtly. Mss tttan the . raooo of the standard smoking madtw» . test results 1or tnese 2'1 nranas ~(brana 'A' 0:11Smq n~otme and brand 'C' 0.11mg nooimel: tMiratq is 1.63. A,sufWart oompanSpn for macnine testing of brand 'B" (0.10irg maonnet provided a ratio of 0 91 '. TMretore; reqaratess of tne Smoking benav- iOur,efllQ3of rWmOVais wMtetMy sRqkeld tMse 3 brands :. tnepnK'Anne intake as measUred by,determinatlOnof steady-state tHas+na corxurue concentration was saniar to tne sianaaro smoKiny macnine irannqs of tnsse oranos spurious statistical conclusions.° The difficulty in dealing with, the multiple factors involved in up- take of nicotine andttubythesmoker~ is obvious in the differl:nces in the conclusions reached! by these investigatots. While in each case the condil- sions were statistically correct, differences in study design altered the data. In one case, the subjects were allowed to smoke in their natural environ- ment without laboratory constraints and in the other the measurements were made in i the labora, tory environment In addttion, Rhusell'et aL (1982) allowed some time for compensation to the neww brand yields In studies of health status amonp smokers it seems advisable to collect the data from indiwiduals who ane frae to smoke in their normal manner influenced only by those factors: which generally affect their cigarette consumption. Benowitz et al. (1987a) reported AUC values for nicotine from smokers who smoked cigarettes over the course of several hours. These subjects were given cigarettes of dilTerent standardismoking ma- chine yields. l'he data indicated individital com- pensation in smoking behaviour for changes in the yields. However, the AUC values were reduced as cigarette yields were reduad, despite the fact the smokers were required to consume 12 cigarettes at each yield value. Under these cirrtlmstanees,,com- pensation was accomplished! by changes in inspir- ation of smoke.. S. Fredictioa of Nicotiru aRd Tar Uptakes from Plasma Corinine Concentrations Gori and Lynch ('1983) reported plasma cotin- ine concentration measurements obtained in 300 smokers of ultralow tar delivery brands (1mg tar, vield or Iess): These investigators found that on av- erage the change in plasma cotinine,concentration was proportional to the measured change in ni- cotine yield. Hioweverl the standard error of the measurement was very large. This wile variatdon in plasma couintne concentration among smokers oflultralow tar yield brands was generally accepted as bcing due tachangcs in smoking behaviour. This stuA% is iinportant, tn that it attempts to measure atr•ragc datl% ntcc*ttnr uptake in aJargr pupulattun ~