Tobacco Documents Online

SEMINARS IN RESPIRATORY MEDICINE-VOLUME 10, NUMBER 4 OCTOBER 1989 Physical, Chemical, and Biologic Properties of Tobacco, Cigarette Smoke, and Other Tobacco Products Gary L. Huber, M.D. . The commercial tobacco cigarette of today is a blend of different tobaccos with a number of addi- tives; as such, it is a continually changing product. Since this cigarette is very different from the ciga- rette that was smoked when the epidemiologic studies linking smoking to lung and cardiovascular disease were first reported, comparative assessments of potential health effects from the currently con- sumed commercial product are difficult to develop. It is important, however, to understand the changes that have occurred in cigarette manufacturing, and their potential health implications in the present and the future for the pathogenesis of pulmonary dis- eases and cardiovascular diseases. The primary purpose of this contribution therefore is to review some of the physical, chemical, and biologic prop- erties of tobacco cigarettes and other tobacco prod- ucts. Within the scope allowable, this review cannot be comprehensive. More detailed analyses are pre- sented in several of the reports by the Surgeon General on smoking and health," as well as in other specific contributions that will be cited within this text. Any attempt to review the properties of the tobacco cigarette and other tobacco products is a formidabk undertaking. The literature available is voluminous, is at times very technical, and is often beyond the grasp of all but the experienced tobacco cfiemist. Is it important for the practicing physician or pulmonologist to understand something about the physical, chemical, and biologic properties of the tobacco cigarette? The answer is "yes," for tobacco and the tobacco cigarette are much misunderstood and often misrepresented. Although the literature that is available is extensive, many of the answers sought still remain elusive. The information that now exists is too often inconsistent or, worse yet, contra- dictoty. Yet, better knowledge is the crucial key to clarifying still existing controversies on tobacco and health, to understanding the potential relationship of tobacco to certain diseases, and, within the appropriate framework, to developing smoking ces- sation strategies. This is particularly true for those components of tobacco products that influence hu- man behavior. An understanding of the physical, chemical, and biological properties of the tobacco cigarette also is essential to assimilating and under- standing the continually appearing publications on tobacco and health and on smoking cessation, as well as to designing and implementing research and educational programs in related endeavors. An additional fcxus of this review, then, will be directed toward clarifying general misrepresentations of the tobacco product and improving our understanding of the important related issues. Finally, it must be emphasized that the lay public is frequently exposed through a multitude of media to information about tobacco use and cigarette smoking. From this exposure, a variety of attitudes, opinions, awareness, and the like have developed. Public beliefs are based on public knowledge, and from these eventually come public policies. Un- fortunately, not all of the information on tobacco and health that has been conveyed over these past 35 years of intense interest in and concern for the health consequences of smoking has been accurate. There are, after all, two sides to this issue that have been developed in a continued and consistent manner for now well over a third of a century. Our knowledge about many important issues is far from compkte, some of the inaccuracies probably will never be fully clarified, and controversy is still sustained because Univenity of Texas Health Center at Tyler. Tyler. Texas Reprint requests: Dr. Huber, Department of Medicine. Univernity of rexas, t3mc 4003. Tyler. TX 75710 C19R9 by Thieme McAioi Pubinhcn, tnr., VI Park Avenue Smith. New York. NY 10016. AO riRhts rsser.tid. t97

SEMINARS IN RESPIRATORY MEDICINE-VOLUME 10. NO. 4 OCTOBER 1989 some of the disseminated information is not correct. Even though it was a habit of growing popularity, the general smoking public had some sense that ciga- rette smoking was smoething less than a completely healthy pastime well over half a-century ago. The saying that tobacco smoking would "impair your wind" probably dates back, for example, to a study in school boy track athletes near the turn of the century. The description of cigarettes as "coffin nails" and the phrase that tobacco smoking will "stunt your growth" are at least as old. In the 1950s, according to the Surgeon General's latest assessment (1989), the majority of smokers and even more of the lay public were not aware, however, of many of the specific health risks associated with smoking; today, most smokers and most of the general public have a different perception 6 Relative to the conclusions recently issued by the office of the Surgeon General, however, the public still significantly ntisconceivrs or inadequately understands many of the potential effects of prolonged tobacco use.' This contribution is extensively referenced for several reasons. Many questions about the physical properties and chemicaF nature of the tobacco product still remain unanswered or are controver- sial; references are provided therefore either in support of the conclusions that are presented or as a "road map," of sorts, for those who wish to pursue these considerations in greater depth. Perhaps of equal importance, this is a time, more than ever before, in the history of tobacco and health when public policies on smoking and tobacco use are being established widely in our society. These changes are occurring both within the world of medicine and, to a greater extent, within the broader arena of general public health. This text and its citations, it is hoped, will provide an objective review and balanced frame- work for an understanding of tobacco products that will be useful and helpful to the establishment of improved and more responsible policies. TOBACCO 7daACQO AND 7n8ACCO PRODUCTS Tobacco is a member of the nightshade family, it is an important agricultural crop in many parts of the world, and it is used by httmans in almost every country. There are as many as 65 or more varying types of tobacco. All are members of the genus Nicotiana; variants most commonly grown for com- mercial use include primarily Nicoeiana fabncum and, in China, India, and the USSR. Necotierea rustica. Bright or Virginia (flue cured), burley (air cured), 298 Maryland, oriental (Turkish or aromatic), and cigar tobaccos are the main variants used in tobacco products manufactured in the United States. Most oriental tobaccos are imported. The chemical con- stituents of these tobacco variants differ from each other from year to year and are influenced by growing conditinns, weather, genetic plant variety, agriculture additives or alterations, regional cultur- ing practices, curing, and commercial processing. In- general, however, there are no qualitative differ- ences as a function of these variables and specific tobacco components cannot be completely delted from the plant by selection of agronomic conditions. The individual chemicals that compose the tobacco leaf include carbohydrates (glucose, pectins, cellulose, and several others), lignin, nonfatty or- ganic acids (malic, oxalic, citric, and others), nitro- gen-containing -compounds, resins, and several thousand other ingredients present at extremely low levels."''I-he carbohydrates and related compounds are the primary component of the leaf (about 50% of the weight), along with the tobacco proteins. The major nitrogen-containing compounds are nicotine and other alkaloids (about 0.5 to 5% of the leaf weight), and the tobacco proteins. Other significant constituents of the unburned tobacco include, by weighi, the terpens (0.1 to 3.046), polyphenols (0.5 to 5.4%), phytosterols (0.1-2.5%.), carboxylic acids (0.1- 0.79F), alkancs, aromatic hydrexarbons (in extremely low cencentratiuns), aldehydes, lactones, amines, and other compounds." How the tobacco leaves are treated following harvest, in part, influences their composition, particularly their final carbohydrate and nitrate content. Flue curing Virginia or bright tobacco in steam-heated barns results in tobacco leaf with high sugar concentrations and low levels (less than 0.1%) of nitrate content. The kss complex process of air curing burley tobacco results in a tobacco of low sugar and relatively high nitrate content (5% or so). Sun curing oriental tobacco produces a product of intermediate sugar content that is still low in nitrates (kss than 0.5%). Tobacco, at one period during its growth, is relatively high in total protein content, but this level decreases substantially by the time the tobacco is suitable for harvesting and curing for its eventual use as a smoking product. Because of the ease with which this protein-rich plant can be grrnvn, ongoing agri- cultural research for years has included considera- tions for converting the tobacco crop for cigarettes to one of tobacco prc*ein for f xxl pmdttction." When the protein kvel is high in the tobacco plant, the protein-free tobacco residue is, according to con- sumer and other test panels, totally unsuitable for smoking; by the time the protein-free tobacco resi- due is acceptable as a smoking product, the level of protein is so low that its isolation is economically unattractive.

PROPERTIES OF TOBACCO- Huse. The curing of the harvested tobacco occurs either in dry air or in a Oue-heated environment. Much of the desired taste and aroma are imparted to the tobacco during the essential process of curing. The associated chemical changes predominantly involve hydrolytic enzymes. The curing process also incorporates reactions that create compounds not present, as such, in the growing plant. Most tobacco is cured in small tobacco barns by the farmer who grew it. As an alternative, the Department of Agri- culture has developed a curing process utilizing a homogenized tobacco slurry instead ofthe harvested whole kaf. Once fully cured, tobacco is still not in a form acceptable for use by_ the consumer. -Ilie cureci tobacco must be aged. Aging improves aroma and other properties desirable to smokers. The chemical processes among the many other_changes of aging decrease alpha-aminonitrogen and increase volatile acids. The aging process may take 1, 2, or more years and it is usually carried out in sealcd barrcls in warehouses maintained by the cigarette manulac- turing industry. In addition to curing and aging, cigar tobaccos are fermented. In this fermentation process, organic acids are reduced or eliminated by oxidation and decarboxylation, making the smoke from cigar tobacco more alkaline relative to the more acidic smoke generated-from cigarette tobaccos. The final commercial cigarette product sold in the United States, as well as in Japan and most European countries, is usually a blend of burley, Virginia, Maryland, and oriental leaf tobaccos, as well as some reconstituted tobacco sheet. Tobacco cigarettes in the United Kingdom and in Finland are almost exclusively blends of bright tobaccos. In France, in parts of Italy, in South America, and in North Africa, some of the cigarettes marketed com- prise burley tobaccos. The common practice today of including ribs and stems of the tobacco plant, as well as more burley tobacco in the final commercial bknds has increased the nitrate content of the American cigarettes, with a concurrent reduction in deliveries of "ur," including phenols and polynu- dear aromatic hydrocarbons. An important rise in nitrate content, with the associated increase in nitrogen oxides on pyrolytic degradation, has in- creased the potential for N-nitrosamine formation.'s The tobacco blends are treated with casing solutions. These usually include sugars, syrups, balsams, licorice, and other substances. Retention of moisture and the associated qualities that contribute to flavor are maintained by the addition of hu- mectants to the blend. flavor-enhancing additives, the subject of recent controversy related to their potential health effects, are also a part of the final product; exempting menthol, these in total con- stitute less than 0.3% of the weight of the tobacco filler. As "tar" and nicotine deliveries have been reduced, additional amounts of flavorant have been added in order to keep the products consumer- acceptable. The nature and quantity of flavorants added to cigarettes are trade secrets and, as such, have not been studied much outside of the tobacco industry. Flavorants are derived from extracts of tobacco plants and from other sources, whichinclude methold oils, fruit extracts, and other flavorful substances.16 The concern about tlavorings, casings, and humectants is that, when heated during the smoking process, they may contribute potentially harmful by-products to the inhaled materials, or so-called strangers in the smoke that might be harmful to human consumers. At the time of the first report on the health consequences of smoking, about 16% of the domestic cigarette market consisted of menthol-flavored products." Today, that has nearly doubled and for tlle past decade has leveled off at about 28 to 29% of the total commercial markct."' Up to 15 mg of methol are added to each menthol-flavored ciga- rette.'° Many smokers perceive menthol as providing a sense of "cooling" and it has been postulated that this promotes deeper inhalation. Three of every four black smokers use menthol-ilavored cigarettes, com- pared with only one of four white smokers,'""" prompting some to offer this as an explanation for one of potential multiple reasons why some black populations appear to have higher mortality rates for certain smoking-associated diseases;!" a far more reasonable explanation is that blacks do not have as many "environmentally clean" jobs as do whites and thus are exposed to an additional burden of po- tentially toxic or harmful substances. Tobacco cigarettes are physically wrapped with paper. These cigarette papers can be manufactured with additives, as well, and can be developed with varying porosity. The porosity of the paper in- Ouences the burning characteristics of the cigarette and the nature of the delivered smoke. Increasing the porosity of the wrapping paper reduces the "tar" delivered, as well as the levels of such potential cytotoxins as acetaldehyde, acrolein, furmaldehyde, and hydrogen cyanide. Increasing the porosity also reduces the amount of carbon monoxide delivered. Decreasing the porosity of the wrapping paper has just the opposite effect. Filters were introduced on tobacco cigarettes in the early 1950s. I n 1951, kss than 196 of commercial cigarettes purchased were filtered. Most, but not all, cigarettes sold in this country today are filtered; only a very few percent of the market comprise non- filtered products. The filters are of many types. Most of these are cellulose acetate filters, but they also include charcoal filters, specially bafiled filters, men- tholated filters, and other kinds. Filtering the to- 299 -

SEMINARS IN RFSI'IRA'IY)RV MF.t)It:INF.-V()LUMt: Itl, N(). 4 (:7Y)BF,R 1989 bacco smoke reduces the delivery of mainstream total particulate matter and, ofcourse, its component parts, "tar" and nicotine. This can be achieved mechanically by reducing the total particulate deliv- ery by direct filtration. Adding charcoal to the filters may reduce certain potentially toxic gas-phase com- ponents, including volatile aldehydes and hydrogen cyanide, at least as evaluated by some experimental bioassays. Adsorptive additives incorporated in cel- lulose acetate filters, which are the most commonly used filters, potentially can reduce the volatile nitrosamines, as well as the acidic components of smoke, including the volatile phenols; magnesium silicates potentially can diminish the delivery of organic vapors, stxh as the aldehydes. In addition to direct mechanical filtration, the filter, or for that matter the cigarette wrapping paper as well, can be perforated to introduce air by a Venture-like dilu- tion: this can also be achieved, at least theoretically, with longitudinal grcoves in the filter. Air dilution reduces the delivery of both particulate- phase and gas-phase components."" A burning cigarette has been described by many as "a miniature chemical factory," which produces multitudes of new components from its basic raw materials. The first analysis of tohacco smoke was performed in the mid-1 R(IOs, a few were undertaken in the late 1920s, and amodest number completed in the 1930s. By 1954, about 100 tobacco smoke components had been identified." From that date, additional components were identified-about 450 by 1959," approximately 950 by 196R,'S and nearly 4000 by 1982.M By some estimates. it has been suggested that as many as 40,000 or so chemical entities are present in tobacco smoke. Most of these are present in only very small trace amounts, and some are reactive products of the unstable combus- tion product, making an ultimate quantification very difficult. In fact, most of the components as yet unidentified in mainstream cigarette smoke are present at levels of subnanogram quantities per cigarette, that is, at levels of picogram per cigarette or femtogram per cigarette. The levels of most of the components in mainstream cigarette smoke of con- cern are in the microgram per cigarette or nano- gram per cigarette range. Only a few of the compo- nents, such as prbon monoxide and nicotine, are present in quantities of milligrams per cigarette. The processed but unpyrolized tobacco product contains about 2500 or mote known constituents:" and varies as a function of many parameten, includ- ing local growing conditions and what the tobacco farmer might add to the soil or apply to the growing plant- Pesticides, fertilizers, and other agricultural agents used in tobacco farming become incorpo- rated into the tobacco kaf.='=" l.ess than 20% of the 300 pesticides and other agricultural products incor- porated into the growing tobacco plant are trans- ferred to mainstream smoke during the smoking process. For tobacco harvested from some areas of the country, government approved agricultural sub- stances may add at least another 400 or more chemical species to the already complex cigarette smoke. In most commercial cigarettes, a few parts per million of DDT, DDD, and maleic hydrazide may appear in the mainstream smoke. In general, how- ever, there has been a trend toward overall reduc- tion in the pesticides and agricultural chemicals incorporated into tobacco and delivered in tobacco smoke. THE IGNITED CIGARETTE When a cigarette is smoked, the hydrogen- supported, oxygen-free burning cone pyrolyzes dur- ing puffing at a temperature of about 860 to 900°C;!° between pu(Ts, the smoldering temperatue is about 500 to fi00°(;.z" Approximately 50% of the pyrolyzed product is generated as neutral to alkaline side- stream smoke, which for the most part is released to the environment during smoldering between the ten or more pu(Ts of a full cigarette; a small amount of this sidestream smoke is drawn into each inhaled pull by the smoker. 'llre remainder of the smoke, exclusive of an amount that is lost to the condensate that forms downstream along the tobacco cigarette rod, is generated when air is drawn across the fire cone by puffing; this smoke traverses through the butt end of the ciqarette to be inhaled by the smoker as weakly acidic mainstream smoke. The tempe- rature of mainstream smoke leaving the cigarette usually is about 30 to 50'G or even slightly higher. Much of the condensate produced along the ~ tobacco rod is rcpyrrrlyzcd and/or evaporatcs into the mainstream smoke as burning progresses down the length of the cigarette, delivering what some have considered to be progressively increasing amounts of potentially bioactive components with each subsequent pufi; there are other data in the literature, however, to indicate that the specific activity of the smoke in a given bioassay is essential the same for early and late puffs'0 Some compounds are eluted from the cigarette rod as the smoke is drawn through it. The concept of relative tobacco smoke toxicity, appropriately or not, has been based, in part, on specific tumorigenicity in mouse skin- painting binassays. In the National Cancer Institute Tobacco Working Group, for instance, a cigarette whose delivery was 2 mg of "tar" per cigarette was considered no "safer" than a cigarette whose "tar" delivery was 15 mg per cigarette if the "tars" from the two cigarettes gave the same specific tumorigcnicity in mouse skin-painting studies. As a cigarette is

PROPER7'IES OF 7-OBACCO- Hvera smoked, the amount of "tar" does increase in successive puffs, but the benzoa pyrene to "tar" ratio, for example, remains the same. Regardless of its distribution (sidestream or mainstream), tobacco smoke is an aerosol compris- ing liquid particles (the so-called particulate phase) dispersed in a gas (the so-called gas or vapor phase). There are about 2 to 5 billion panicles in the aerosol per millileter of tobacco smoke.s' The particles are electrically charged by a process termed "chemioni- zation," which in itself may have potential biologic significance." About 10% or less of the total smoke wieght is in the monodispersed aerosol particles, and 90% or greater is in the gas vapor. Some of the tobacco leaf components, such as nicotine, phytosterols, and long-chained paraffins, are transferred to some degree to mainstream and sidestream smoke structurally intact. 1-1te extreme temperatures and reducing atmosphere provided by the fire cone result in pyrogcncsis ol a large mtntlx•r of components not originally present in the tobacco kaf. Most of the nontobacco components in tobacco smoke, such as the polynuclear aromatic hydro- carbons, phenols, volatile nitrosamines, and others, are not actually generated at the fire cone but are generated in a"cylinder" a lew millimeters in length situated in the tobaccri rod a few millimeters in fronl of the first cone and tobacco rod interface, where the temperature range within the "cylindcr" is 40(1° to 700°C. The major components of smoke that are generated within the fire cone itself are products of almost complete comhustion and include water, ammonia, carbon monoxide, carbon dioxide, nitric oxide, and the like." Other constituents of main- stream smoke that are entirely pyrnsynthesved include benzene and benzo a pyrene. Certain volatile aldehydes and nitrosonornicotines are partially py- rosynthesized and partially transferred intact with- out pyrosynthesization.s`''s If nicotine delivery is the primary key to con- sumer gratification and acceptance of the commer- cial cigarette, there are probably absolute limits on reduction of "tar" delivery. In an acidic mainstream smoke, nicotine is delivered for the most part as protonated nicotine within the aerosol particles, and the size of that aerosol particle and the solubility of nicotine limit the quantity of nicotine containable in individual aerosol particles; in acidic smoke, there are only very small amounts of free nicotine in the gas phase. This principle of "solubility" is applicable to any smoke components found primarily in the particulate phase. Within the aerosol particle, pro- tonated nicotine is the singk most plentiful organic components in the particulate, indeed to the degree that it may be asked if protonated nicotine is the solvent or the solute? The optimal ratio of nicotine to Table t. Oistribvtion of Mainstnant timoke• Total mainstream smoke SOp Wet total particulate mstler 22 Nicotine 1 3 WaIer 3.7 "Tar" 17 Aerosol gas phase 478 Water 50 Air components 350 Carbon monoxide 20 Carbon dioxide SO Other components 8 •Atl date expressed in mittiprams tor . 500 mp deliverz ciqareee, es determined by Federal Trade Commission criteria. r "tar" is about 1:10 or slightly less. As "tar" delivery from the commercial cigarette was reduced in a prrtgressivc, stepwise manncr after the early 1 950s, it was paralleled by an equivalent reduction in the nicotine delivery. 'lhis reduction was implemented very gradually, in part because consumers of a given brand usually do no( tolerate significant detectable alterations without changing to a different cigarette. Because nicrrtine-dependent smr>kers appear to re- quite a finite minimal amount ol nicotine per day, or per cigarette for that matter, the relative ratio of nicritine to "tar" has drifted slightl) upward in recent years. l'his has been brnught about either as a result of altered tobacco blending or by other manipula- tions of the processed cigareue, especially as "tar" delivery has been reduced to ultralow levels in some commercial cigarettes. ..T"., lhe particles in the particulate phase of the 400 to 500 mg of aerosol comprising the mainstream smoke have a relative uniform size distribution of about 0.2 to 0.4 m mass median diameter, or 0.1 to 1.0 m absolute diameter n This size is fully respir- able, with most such particles being able to reach the alveolar spaces. Tbe aerosol is both chemically and physically unstable, and the particle size grows by agglomeration as individual aerosol nuclei combine. About 80% of the smoke of the total mainstream smoke from a full-flavor cigarette consists of the components of air drawn into the cigarette during the putT.'I'he gross distribution ofentities within the mainstream of smoke are summarized in '1-able I. The "tar" in this representative example, as mea- sured by the Federal Trade Commission (FTC) criteria, contains about 4000 identified smoke com- ponents. The bulk of the water delivered in main- stream smoke is that which was drawn through the I S01

SEMINARS IN RESPIRATORY MEDICINE-VOLUME 10, NO. 4 OCTOBER 1989 cigarette from the surrounding environment during the puffs. The "air components" include nitrogen, oxygen, argon, and the other components normally present in environmental air. The "other compo- nents" in the gas phase, summarized in Table 1, are usually called the "cigarette mainstream vapor or gas phase," and it contains about 500 now accepted identified smoke components. The FTC utilizes the term "tar" in referring, indirectly, to the total particulate matter generated by smoking an appropriately preconditioned ciga- rette in a standard mechanical way, usually to a given butt kngth or, in some studies, in reference to a predetermined specific number of standard puffs." The standard -FTC procedure requires that the cigarettes be smoked by analysis in a mechanical device that generates 35 ml puffs of 2 seconds' duration, at the rate of one puff per minute at 25*C and 60% relative humidity to a specified butt length. Eight or ten such puffs of 35 ml at 60% relative humidity contain substantial levels of water, com-- pared with the weight of FTC "tar" and gas phase components. The total particulate matter, as a gravimetric expression, is obtained by passing the whole mainstream smoke through an absolute Cam- bridge-type or glass fiber filter (trapping all particles more than 0.1 m in size) and weighing the collected particulate matter. The FTC requires cigarette man- ufacturers to list "tar" delivery (as well as nicotine content) in advertising their commercial products. The measurement of "tar" has been variable and, at times, fraught with controversy. ln 1982, the FTC introduced modifications for mrasiirement in hopes of standardizing testing pnxcdures." appearance. There are "tars" that humans consume that are derived from many substances other than the burning of tobacco leaf. A number of in- vestigators have reported that components present in "tar," regardless of source, generated under a variety of laboratory conditions have the potential to act as tumor initiators, as complete carcinogens, as cocarcinogens, or as pncer-promoting agents, as well as inhibitors of cancer or anticarcinogens.s'-" Eventually, criteria for potential carcinogenicity were established by international standards." Che- mical analyses of tobacco smoke reported in the literature are derived from smoke generated under standardized laboratory conditions and may not reflect what the smoker generates under physiologic conditions."" Some of the more important compo- nents in the particulate matter of tobacco smoke are listed in Table 2." Much -of the early biohazard research on to- bacco focused primarily on these biologically active - agents in the "tar" or cigarette smoke condensates, to the exclusive of potentially equally bioactive gas- phase components. As a result, the primary emphasis of the past three decades or so, both in the United States and abroad, has been to reduce the delivery of "tar" in the smoked product. and thus theoretically to reduce the potential carcinogenicity of tobacco smoke."" The average sales-weighted "tar" deliv- ery at the end of World War II or thereabouts was approximately 45 mg per cigarette and 2.5 mg of nicotine.'" With the introduction of consumer- ac- ceptable filters, "tar" delivery had fallen to about 35 mg per cigarette in the late 1950s. This has been reduced in a progressively stepwise manner since the earliest skin-painting studies linking "tar" carcino- DEFINITION OF "TAR" The word "tar" is often written in quotations because there are differing definitions for its use. The FTC, as well as most of the scientific community, use the word "tar" to express the dry weight of the collected condensate (total particulate matter minus all water) minus the weight of any nicotine present. Tobacco chemists, on the other hand, sometimes refer to "tar" as a laboratory product obtained by passing cigarette smoke through a cold trap at an extremely low temperature. Thus, the assertion that "tar" is inhaled by human smokers is somewhat misleading; smokers inhale tobacco smoke actually an an aerosol of particulate matter suspended in a gas or vapor phase, and not as "tar" in the true senses of the word. The laboratory condensate of tobacco smoke is Table t=ome Intportant Cotnponents In ParftAa/e Mdte+ el =tnoke• Nicdine Neophyladfriss Nornioolinw t3lyoera An.fabine Lknprwrw An.basine PanNic Wld eivrr+dya Sts.rtc .do n-Hentri.corN.ne 9w+c aad NaptNMlene tJnotefc acid Phsnuwnrens LMroNnic acid Annxaoenes Lacfic tteid Fkarenes tndote Prerms Skatae QD Ftuwamnnes Ouinolinss ~ Olher polynuckar arornNic 8snzoturans ~ ~~~ a sUoaera CA C.lechol C.mpeqerol rA 8aopaetin Cho4stera - rA Cyctotsr,es AflNifN A IP o{/tf10wlf Tauidinet so4nssa N-Nilros~min" tieet+t.w referred to as "tar" because of its dark, viscous 'Ad.p.d trom r+oem.nn and

PROPERTIES OF TOBACCO- Huset gens to cancer in skin-painting experimental ani- mals; the present "tar" delivery, based on sales- weighted averages, is about 13 mg per cigarette. "TAR REDUCTION" 'Tar" delivery from a cigarette is influenced by the tobacco blend, kaf thickness, cut width, rate of leaf burn, moisture content, packing density, use of reconstituted tobacco sheets or expanded tobacco, and the properties of the filter tip. Burning rate, as well as dilution of the smoke generated, are influ- enced, in turn, by the porosity and characteristics of the cigarette paper, as well as by other factors. Reductions in "tar" delivery much below 15 mg per cigarette requires, in addition to mechanical filtra- tion, Venturi air dilution, either through the ciga- rette paper or via entry of air into the f lter. Absolute mechanical filtration reduces particulates without necessarily reducing the levels of gas-phase compo- nents; air dilution, on the other hand, reduces the levels of both particulate-phase and gas-phase com- ponents. - The "tar" reductions in consumer-acceptable filter-tip cigarettes were accomplished only in part by the filter tip. The first successful tlter-tip ciga- rette, introduced in late 1953, incorporated not only the cellulose acetate filter tip but also substantial levels of a proprietary reconstructed tobacco sheet. By 1960, all of the other cigarette manufacturers in the United States had-introduced filter-tip brands and were including in their blends their own pro- prietary reconstituted tobacco sheets. By the late 1950s, higher porosity cigarette papers were also in use. Thus, filtration alone was not the sole design factor in lowering "tar" delivery. Because of the introduction in reconstituted tobacco leaf and po- rous paper, "tar" delivery from nonfiltered ciga- rettes was also reduced during this time period. Although the definition of "tar" is somewhat arbitrary, The FTC has accepted it for classifying and reporting cigarette total dry nicotine-free particulate delivery. The definition of "tar," that is, the weight of "tar" equals the weight of the total wet particulate water minus the weight of the water and nicotine present, is the one now issued by the FTC and used in the United States. Thus, high "tar" cigarettes are designated as those delivering greater than 15 mg of "tar" under standard smoking conditions. [.ow "tar" cigarettes are those that deliver 15 mg or kss, but greater than 5 mg, and ultralow "tar" cigarettes are those delivering 5 mg or less of"tar." The delivery of smoke or total particulate matter under "standard smoking conditions" with a mechanical smoking machine may or may not be representative of how much "tar" or total particulate matter is actually delivered to the lungs of a human smoker, depend- ing on a large number of variables. Although it has been widely accepted that reducing "tar" delivery will reduce eventually the rate of lung cancer and other discases associated with smoking, the validity of that assumption can be questioned. If, for instance, people smoke primarily for the nicotine in tobacco cigarettes, and if nicotine is reduced in direct relationship to "tar," there is a suggestion in some studies that smokers will smoke each low or ultralow "tar" cigarette differently, or smoke more low or ultralow "tar" cigarettes in an effort to meet their nicotine needs. Research on smoking behavior, however, is extremely complex, and complicated by many uncontrollable variables. Thus, the "evidence" that individuals "titrate" or "accommodate" their smoking behavior as a func- tion of nicotine delivery is weak, inconsistent, and controverted. If they were to do so, such changes in smoking behavior might result in a higher exposure to gas-phase components than would have occurred with a higher nicotine to "tar" delivery ratio. By using carbon monoxide as an indicator of gas-phase exposure, some studies indicate higher carboxyhe- moglobin levels in smokers using low "tar" ciga- rettes; other studies do not, however, confirm this. l f, relative to exposure tn "tar," the gas phase is independently bioactive, it has been hypothesized that a person may simply be exchanging one disease risk relationship (cardiovascular disease and carbon monoxide, as a theoretical example) for another (lung cancer and "tar") as net or total "tar" reduction occurs;'° this hypothesis, however, has not been supported or borne out by clinical or epidemiologic data. NICOTINE The primary ingredient in tobacco that makes it uniquely desired by the consumer is nicotine. By chemical structure, nicotine is an alkaloid tertiary amine that is composed of a pyridine and pyrolidine ring. There are two possible stereoisomers of nico- tine. Nearly all of the nicotine in tobacco is the pharmacologically active isomer, (S)-nicotine (or levo-nicotine, by some notations). During the smok- ing process, some of the (S)-nicotine is converted to (R)-nicotine (or dextro•nicotine), which is much kss pharmacologically active than the (S)-isomer.'OAbout 90% of the nicotine in smoke is (S)-nicotine.'Onere are additional alkaloids in tobacco other than nico- tine, most of which are 3-pyridyl derivatives. These include nornicotine, myosmine, anabasine, nicoty- rine, and anatabine, as well as others, accounting in total for about 10%, more or less, of total alkaloids.w w'lhese minor tobacco alkaloids have less phar- macologic and biologic importance. 303

Sf:MINARS IN RFSI'IRA'I()RV MF:I11(;INN-V(1L.tlMF: 1(1, N(1. 4(N;7OR/i!t 19R9 In the tobacco plant, the bulk (75 to 95%) of the nicotine is found primarily in the laminae; the remainder is found in the leaf midrib stems or main plant swlk.'-" For flue-cured tobacco, the stem nic- otine to leaf nicotine ratio is about 0.24. For burley tobacco, the ratio is lower, at atx)ut 0. 17. ne tobacco plant stalk represents about 25% of the total tobacco plant weight. The-weight of the main stems in the leaves is also about 25% of the weight of the tobacco plant. Leaves high on the plant generally contain more nicotine than lower leaves.s` Ahead of the burning cone of the tobacco cigarette, nicotine is vaporized during the puff and is transported toward the mcwth end of the ciRarctuc. Vaporized components of tobacco smoke, such as nicotine, are cooled en route through the tobacco rod and participate in the formation of the liquid droplets in the smoke aerosol. Some nicotine con- denses downstream on the cooler tobacco rod and is involved in the same process (vaporization, aerosol formation, condensation) during each of the sub- sequent puffs. Some aerosol particles, containing nicotine, are also filtered from the smoke stream by the tobacco rod. In most commercial cigarettes, only about 1-5% of the total nicotine appears in the mainstream smoke; 25 to 40% is released to side- stream smoke, and 15 to 25% is deposited within the butt (and filter tip). The remainder is pyrolyzed to nicotine decomposition products. Although it is commonly stated that nicotine is vaporized," that is not strictly true. Nicotine is present in tobacco in the protonated form, as salts of a variety of acids. Dnring exposure to the elevated temperatures present in the smoking process, the nicotine salts decompose to nicotine and acid; this nicotine appears in the smoke stream. Some nicotine salts at elevated temperature decompose completely to nicotine plus acid; sonic decompose to nicotine, acid, nicotine degradation products, and acid de- gradation products; and, finally, some nicotine salts decompose completely to nicotine degradation - products and acid degradation products. The major pyrolysis pnxlucts of nicotine itsclf are carbon ntonoxide, cartxon dinxidc, g-vinyl- pyridine, 3-methylpyridine, pyridine, mysomine, and 4,3-dipyridine. Some of the characteristics of smoking, such as the frequency of puffing, may influence the delivery of nicotine and other alkaloids to the smoker.s''The presence of the non-nicotine alkaloids may inQuence the pharmacokgic attiivity and metabolism of nicrnine `R "` Most of tlte nicotine in tobacco smoke generated from American ciga- rettes, as well as most of the cigarettes marketed in F'inland and the United Kingdom, is carried on or within the "tar" microdroplets, with only a very small 304 amount of nicotine in the vapor phaseas ABSORPTION OF NICOTINE Nicotine is a weak base and the absorption of nicotine across biologic membrancs depends on its pH, or state of acid-base dissociation."'The pH of burning tobacco, in turn, is determined primarily by its type and composition. A mixture of flue-cured tobaccos constitutes the major component (30 to 40%) of theblend in most cigarettes manufactured in tlte United States. When burned, the pH of the mainstream smoke from these cigarettes is acidic, ..ith the first puffs having a pH of about 6.0 and subsequent puffs progressively decreasing slightly to a pH of alxiut 5.0.' In an aqueows milicu, nicotine has an itxlcxof iomic dissiriatiom or pKa nf ahnut 8.0, which means that when it is delivered by most tobacco cigarette smoking. nicotine is almost com- pletely protonated. Prutonated nicotine does nut cross biologic membranes in any appreciable amounts. Thus, almost no nicotine from cigarette smoke is absnrbed in the mouth, pharynx, or upper respiratory passaRea•ays.R' Indeed, in its nonproto- nated form, inhaled nicotine has an acute toxicity for man. The unique "advantage" of tobacco cigarette smoke, then, is that in an acidic state it can deliver the nicotine-containing microdroplets in a nontoxic form that will bypass the otherwise operative de- fenses of the upper respiratnry system. When the cigarette smoke microdroplets reach the alveolar spaces and deposit on the vast internal surface of the IunR. they are immediately buffered to a physiologic pH of near 7.4. l'his is close enough to the index of ionic dissociation so that a significant portion (more than 30% of the total amount) of the nicotine is in a nonprotonated state and, as such, is rapidly and efficiently absorbed across the air-bkxxl barriers at the alveolar surfaces."i Concentrations of nicotine then increase rapidly in the blood and peak near the end of the smoking of a cigarette. Depending on inhalation patterns, retention times, and related factors, smokers may retain anywhere from 30% or less up to 9096 or more of the total nicotine generated and dclivered via the inltaled smokc." Bratusc the pl I of cigarette ttAr.tcno smoke is less than 7.0, very little nicotine is absorbed across the oral mucous membnnes, even if the smoke is held in the mouth for relatively long peri- ods."Through a selection of different tobaccos, however, oral snuffand chewing tobacco are basic, or have a p1I greater than 7.0, thus rcndcrina their nicotine mure readily absctrbcd trotn the mouth!"'a Nicotine polacrilex gum is also buffered by the manufacture to a pH greater than 7.0 to facilitate absorption!" Nicotine absorption across the oral mucous membranes is slower than across the much more extensive internal surface of the lung.

PROPERTIES OF TOBACCO- Huitr. Fine ground nasal snuff has nicotine absorption rates comparable to those in inhaled cigarette smoke. Nicotine that is swallowed is usually absorbed very poorly in the stomach because of the gastric acidity 67 In the alkaline small intestine, however, nicotine is nonprotonated and readily ab- sorbed." Nicotine that enters the body via the gas- trointestinal tract has a diminished net psychotropic influence because it must first pass via the portal circulation into the liver, where it is degraded to less biologically active breakdown products. Nicotine that enters via the nasal or oral mucous membranes, or via the lung, passes directly into the systemic circulation and reaches the central nervous system relatively intact. Uncommonly, nicotine can be ab- sorbed across the skin, especially in tc>fmcco field workers, and potentially reach the systemic circula- tion in sometimes very high concentrations.'r Commercial smoking products vary signif- icantly in the amount ornicotine they potentially can deliver to the smoker. )n addition, the act of stnokinR is itself a very ccnnplex behavioral act. '11te smoker can control the exposure and dosimetry to the inhaled smoke, including the delivery of nicotine, by the manner in which the cigarette is smoked. Usually, this is not a conscious manipulation, and attempts to quantify delivery by the smoker's own assessments or perceptions have been generally grossly inaccurate when evaluated against reliably quantifiable chemical parameters. Ilie amount of nicotine that a smoker will extract from a cigarette is determined by the number of puffs generated, the volume of each pufT, the rate and depth of smoke inhalation, the degree that the smoke is diluted by environmental air, the rate of pufling, the intensity of pufiing, and the smoke inhalation and exhalation patterns, as well as other factors"""' Regardless of the puff volume generated, some smokers barely inhale at all, often with an amotmt kss than a nortnal tidal volume. Some smokers inhale larger vulnmes, by taking a forced expiration just before inhalation, but stop their inhalations near their functional residual capacity; others inhale deeply, sometimes to near residual volume. Some smokers hold their inhaled smoke only transicntly or rnN at all, whereas others may hold the inhaled smoke wi:thin thcir lungs for several seconds. Some exhak only through their mouth, others only through their nose, and still others through botlt their mouth and nose. Some smokers may seal the ventilation ports in the ciga- rette filters with their lips, and others may seal the filter parts or paper porosity with their 6ngers.'s All of these facton change the dosimetry for each individual smoker and for each cigarette. The way in which a person smokes, however, generally is re- markably consistent from cigarette to cigarette, providing the constituents of the cigarette remain the same or are altered only slowly and in small increments. Regardless, the dose of smoke delivered can be reliably determined only by measuring one or more of its constituents in the blood." The range of individual intake of nicotine per day in smokers can be quite wide. In a limited number of smokers, several studies have reported comparable results." """ The intake of nicotine from each cigarette averaged about 1 mg in one study, but with a range from well less than 0.5 mg to greater than 1.5 mg with an average total daily intake of near 40 mg and with wide ranges on each side of the mean." FATE OF NICOTINE In the bkxwdstream, at a pli of approximately 7.4, about two thirds of the nicotine is nonproto- nated and less than 5% is bound to plasma pro- teins.'"'° llte lungs, brain, spleen, and liver have a high aflinity for niccttine; in these tissues it reaches a cnncentratiom more than twice greater than that which would be predicted on the basis of measurable blood concentrations and body weight alone. Adi= pose and muscle tissue have an apparent very low affinity for nicotine, and it is not stored within the body in appreciable amounts or for long periods of tinte. Nicotine freely crosses the placenta and can be aspirated from amniotic Ouid.' Nicotine is also found in saliva," in cervical mucous secretions" and other tissue fluids, and in body hair."s Nicotine is also found in nonlactating breast fluids and in breast milk, but only in very low concentrations.' Once in the blood, nicotine peak levels are reduced fairly rapidly. Depending on tlhe urinary pN, some nicotine is lost via the kidneys, up to 30% or more in some reFxorts, but more commonly in the range of 5 to 10% of the total absorbed dosage;41" a stahlc urinary mctabolue is trans-3'-hydroxy- cotinine."'lltere is some metabolism of nicotine within the lung ," bttt for the most part nicotine is metabolically deactivated by oxidation in the liver."`- "The major primary stable degradation product is crninine, which is formed primarily in the liver in two stcps."s"0 t:cttininc has an average half-life in humans of alxmt 16 to 20 hours, and much of it is excreted in the urine. Because of this long half-life, 72-hour collections of urine are necessary for quantitative measurements; many studies using shorter collec- tions have resulted in erroneous conclusions. Coti- nine levels persist in the bkNid at severalfold higher levels than nicotine. The other major metabolic is nicotine-I'-N-0xide. which can be converted by bacteria (both in vivo and in vitro) to nicotine."•" The half-life of nicotine in the circulation is of significant interest to studies on smoking behavior and smoking cessation. The plasma levels of nicotine,

SEMINARS IN RFSPIRAIORti' MF:111(:INF:-VOLUME 1(l, NU. 4 fX;l'f)l31iN 1989 measured in the afierncxon in smokers, arc in the range of 10 to 50 fig/ml, and generally toward the lower end of that range."" The incremental in- crease from smoking one tobacco cigarette is in the range of 5 to 30 Ng/ml, and the peak and trough levels oscillate as a function of individual smoking patterns."'•"" Similar blorxl levels are found in pipe smokers and in the users of snuff, chewing tnhacco, and other noncigarette products, although the rate of increase of nicotine into the blood by these delivery systems is generally slow'µ-6RO°'"= Early re- ports suggested a-nicotine half-life of fairly short duration, in the range of 20 to 40 minutes't`•" More went studies more accurately indicate a half-life in man closer to 2 hours, but with a range of between I and 4 hours." These longer half-lives imply that in human smokers the psychoactive component of tobacco accumulates at probably a slightly progres- sively higher level throughout the awake petiod and persists at lower- but still significant levels even during sleep. I nvohtntary inhalers of environmental smoke have extremely low levels of nicotine coni- pared with nicotine levels in voluntary cigarette smokers and have a slower rate of nicotine elimina- tion.10' ACUTE NICOTINE TOXICITY Acute nicotine intoxication is a well-known entity. The most common manifestations include pallor, generalized weakness, nausea, emcsis, light- headedness or dizziness, headache, and diaphoresis. Less frequently, chills and vigors, abdominal cramps or pain, and hypersalivation occur. In severe cases, hypotension, seizure activity, respiratory arrest, and death can develop. Novice smokers generally experience the less severe of these symptoms, if any, on their firsl exposure to tobacco. This was perhaps more true in the past than it is today, however, because of the now ready availability of the ultralow nicotine delivery cigarettes, compared with the relatively very high nicotine delivery cigarettes that were common be- fore the introduction of filters and air dilution. Accidental poisoning with nicotine-containing pesti- cides or suicide attempts with nicotine products have been reported. "Green tobacco sickness" is an entity that occurs in farm workers exposed to aqueous nicotine solutions during the harvesting of rain- soaked or dew-cnated tobacco leaves. MEASUREMENTS OF NiCOTJNE AND COTIN/NE LEVELS Reliably quantifying nicotine and cotinine levels in human smokers is not easy. Methods of analy- sis1O"'• and selected references for analysis are S06 presented in Table 3. The use of gas chromato- T.bte 3. Methods of Analysis for NIoothte In dtologk Fluids Method M/ennoe Radiodmmunosssay (RIA) 1a-106 Enzyme-Iinked immunoabsorbence (ELISA) 107 Gas chrometopraphy (GC) 108-113 Hiph-perlofinence liquid chromatoprsphy (HPLC) 114.115 Gas chromsloqraphy-msss spectromeky (GC-MS) 116-118 graphy and mass spectrometry provides the most sensitive, speciGc, and reliable analyses, but the cost of these methodologies prohibits widespread appli- cation. High-performance liquid chromatography also has a reasonable sensitivity, but only on such relatively large amounts of biologic samples that its use is not readily applicable. Gas chromatography and radioimmunoassay are probably used most cOmmrntly hnt are (raught with error in thc hands crf the inexperienced analyst. Radioimmunoassays are simple to use, relatively inexpensive, and usually require small sample volumes; their main shortcom- ing is lack of specificity, due primarily to antibody cross-reaction with cotinine and with endogenous prrxlucts common to human fluids. Gas chromatog- raphy is more accnrate and reasonably accessible. but more expensive; use of nitrogen-phosphorus detectors and high-resolution capillary columns en- hances accuracy. Chromatographic analyses are also vulnerable to nicotine contamination from the en- vironment, including from within the research lab- oratory."° lhe higher levels of cotinine in the to- bacco user, its stability as a breakdown product of nicotine, its longer half-life, and its ease ofquantifica- tion make it in most ways a more reliable and desirable product to assay than nicotine. The long half-life of cotinine, however, does not-make it as reliable an indicator as nicotine in assessing relatively acute fluctuations, such as those that occur during smoking.'="" People smoke tobacco or use other tobacco products for the effect that nicotine has on the central nervous system, where specific nicotine re- ceptor sites have been identified. The other biologic effects of nicotine that have been addressed herein are of a very secondary nature. Through the central Co nervotts system, and to a lesser extent indirectly ~ thnxtgh peripheral neuroendocrine interactions, ~ the effects of nicotine are perceived generally by the Cn consumer in a number of different ways that are ,p considered beneficial. These effects of nicotine on .A the central nervous system and human behavior are N reviewed at length elsewhere within this series of cmntributinns in Snniruns, as are the perceivable