not resemble a cigarette (such as many large cigars do not) and has q distinctive cigar taste and aroma is of considerable significance in making this determination” (702). Cigars are also classified by size. “Small cigars” weigh not More than 8 pounds per thousand and “large cigars” weigh more than 3 pounds per thousand. “Large cigars” are further divided into seven classes for tax purposes based on the retail price intended by the manufacturer for such cigars (96). Cigars are made of filler, binder. and wrapper tobaccos. Most cigar tobaccos are air-cured and then fermented. More recently, reconsti- tuted cigar tobaccos have been used as wrapper, binder, or both. Cigars are either hand-rolled or machine made. Some brands of smal] cigars are manufactured on regular cigarette making machines. The aging and fermentation processes used in cigar tobacco production produce chemical catalytic, enzymatic, or bacterial transformations as eyj- denced by increased temperature, oxygen utilization, and carbon dioxide generation within fermenting cigar tobaccos. In this complex process, up to 20 percent of the dry weight of the leaf is lost through decreases in the concentration of the most readily fermentable ma- terials such as carbohydrates, proteins, and alkaloids. The flavor and aroma of cigar tobaccos are in large measure the results of precisely controlled treatment during the fermentation process (35, 36, 112), ‘ Pipe Tobaccos The definition of pipe tobacco used by the U.S. Government was repealed in 1966 and there is no Federal tax on pipe tobaccos. The most popular pipe tobaccos are made of Burley; however. many pipe tobaccos are blends of different types of tobacco. A few contain a significant proportion of midrib parts that are crushed between rollers. “Saucing” material, or casings containing licorice, sweetening agents, sugars, and other flavoring materials are added to improve the flavor. aroma. and smoke taste. These additives modify the characteristics of smoke components (772). Conclusion Because of the unique enring and processing methods used in the production of cigar and pipe tobaccos. significant physical and chemi- cal differences exist between pipe and cigar tobaceos and those used in 176 cigarettes. The extent to which these changes may alter the health consequences of smoking pipes and cigars can best be estimated by an analysis of the potentially harmful chemical constitutents found in the smoke of these tobaccos. the tumorigenic activity of smoke conden- sates In experimental animals, and a review of the epidemiological data which has accumulated on the health effects of pipe and cigar smoking. Chemical Analysis of Cigar Smoke Only a few studies have been conducted that compare the chemical constituents of cigar smoke with those found in cigarette smoke. Hoffmann, et al. (43) compared the yields of several chemical com- ponents in the smoke from a plain 85 mm. cigarette, two types of cigars, and a pipe. The particulate matter. nicotine. benzo(a)pyrene, and phenols were determined quantitatively in the smoke of these tobacco products. One cigar tested was a 135-mm,-long, 7.8-g., U.S.- made cigar. The other was a handmade Havana cigar 147 mm. long weighing 8.6 g. The relative content of nicotine in the particulate matter produced by the cigars was similar to that of the cigarette tars. The benzo(a)pyrene and phenol concentrations in the cigar condensate was two to three times greater than in cigarette “tar” (table 4). Kuhn (58) compared the alkaloid and phenol content in conden- sates from an 80-mm. Bright-blend cigarette sold commercially in Austria with that obtained’ from 103-mm. cigars. These were tested TaBLe 4.—Amounts of several components of 1 g. of particulate material Jrom mainstream smoke of tobacco products Tobacco product ! Standard 85 mm. 85 mm. Compound US. Havana pipe Cigarette plain U.S. plain U.S. cigar A cigar B tobacco tobacco cigarette —_cigarette (b) in pipe in pipe (a) (b) Nicotine (mg.)__ 2-2 46. 2 63. 6 56. 1 61.0 65. 9 77.4 Benzo(a)pyrene (ug.).___ 3.9 3.6 6.0 3.6 1.2 1.3 Phenol (mg.).....______- 8,2 6.7 15.0 7.3 2.9 4.1 9-Cresol (mg.)___-._____- 1.6 1.7 1.9 1.4 6 .8 m+ p-Cresol (mg.)___. 2 4.8 3. 8 5. 6 3.4 1.4 1.9 ™+p-Ethylphenol (mg.)_- L4 1.5 Li 1.3 .7 .7 ' Smoking conditions: (a) 1 puff per minute, duration 2 see., puff volume 35 ml. (b) 2 puffs per minute, duration 2 sec., puff volume 35 ml. Source: Hoffmann, et al. (43). 177 with and without the use of a cellulose acetate filter. The concentra. tions of total alkaloids and phenol in the cigar smoke condensate Were essentially the same as in the cigarette condensate, but pyridine Values were about 214 times higher in the cigar condensate. Campbell and Lindsey (77) measured the polycyclic hydrocarbon, levels in the smoke of a small popular-type cigar 8.8 em. long, weighi 1.9 g. Significant quantities of anthracene, pyrene, fluoranthene, and benzo(a) pyrene were detected in the unsmoked cigar tobacco, in cop. centrations much greater than those found in Virginia cigarettes hy of the same order as those found in some pipe tobaccos. The smoking process contributed considerably to the hydrocarbon content of the smoke. Table 5 compares the concentrations in the mainstream smoke of cigarettes, cigars, and pipes of four hydrocarbons frequently found in condensates. The authors reported that the mainstream smoke from a popular brand of small cigar contained the polycyclic aromatic hydrocarbons; acenaphthylene, phenanthrene, anthracene, pyrene, fluoranthene, and benzo (a) pyrene. The concentrations of these hydro. carbons in the mainstream smoke were greater than those found in Virginia cigarette smoke, Osman, et al. (69) analyzed the volatile phenol content. of Cigar smoke collected from a 7-g. American-made cigar with domestic filler. After quantitative analysis of phenol. cresols, xylenols, and meta and para ethy] phenol, the authors concluded that the levels of these com. pounds were generally similar to those reported for cigarette smoke. Osman and Barson (68) also analyzed cigar smoke for benzene, toluene, ethyl benzene, m-, p-, and o-xylene, m- and p-ethyltoluene, 1,2,4-trimethylbenzene, and dipentene, and generally found levels within the range of those previously reported for cigarette condensates, In summary, available evidence suggests that cigar smoke contains many of the same chemical] constituents, including nicotine and other alkaloids. phenols, and polycyclic aromatic hydrocarbons as are found TABLE 5.—A comparison of several chemical compounds found in the mainstream smoke of cigars, pipes, and cigarettes I Compound oo Cigars Pipes 1 Cigarettes an Acenaphthylene.--. 982228 1.6 29. 1 5.4 Anthracene_--.---2 88 11.9 110.0 10.9 Pyrene._--- 22-22. 17.6 75.5 12.5 3,4-benzpyrene___ 22. 3. 4 8.5 4 ! This is a light pipe tobacco. Source: Campbell, J. M., Lindsey, A.J. (17). 178 in cigarette smoke. Most of these compounds are found in concentra- tions which equal or exceed levels found in cigarette “tar.” A more complete picture of the carcinogenic potential of cigar “tars” is ob- tained from experimental data in animals. Mortality Overall Mortality Several large prospective studies have examined the health conse- quences of various forms of smoking. The results of these investiga- tions have been reviewed in previous reports of the Surgeon General in which the major emphasis has been on cigarette smoking and its effect on overall and specific mortality and morbidity. The following pages present a current review of the health consequences of smoking pipes and cigars. Data from the prospective investigations of Dunn, et al. (37), Buell, et al. (26), Hirayama (42), and Weir and Dunn (105) are not cited, because in these studies a separate category for pipe and cigar smokers was not established. The smoking habits and mortality experience of 187,783 white men between the ages of 50 and 69 who were followed for 44 months were reported by Hammond and Horn (41). The overall mortality rates of men who smoked pipes or cigars were slightly higher than the rates of men who never smoked. The overall mortality rate of cigar smokers was slightly higher than that of pipe smokers. In a study of 41,000 British physicians, Doll and Hill (26, 27) re- ported the overall mortality of pipe and cigar smokers as being only 1 percent greater than that among nonsmokers. Best (9), in a study of 78,000 Canadian veterans, reported overall mortality rates of Pipe and cigar smokers slightly above those of nonsmokers. Kahn (50) exam- ined the death rates and smoking habits of more than 293,000 U.S. veterans and Hammond (38) examined the smoking habits of and mortality rates experienced by 440,559 men. In these studies, pipe smokers experienced mortality rates similar to those of men who never smoked regularly, whereas cigar smokers had death rates somewhat higher than men who never smoked regularly. Table 6 summarizes the results of these five studies. Thus, data from the major prospective epidemiological studies demonstrate that the use of pipes and cigars results in a small but defi- nite increase in overall mortality. Cigar smokers have somewhat higher death rates than pipe smokers, and mixed smokers who use cigarettes in addition to pipes and cigars appear to experience an inter- mediate level of mortality that approaches the mortality experience of cigarette smokers, 495-028 O—73-—___13 179 TABLE 6.—Mortality ratios for total deaths by type of smoking (males only) Smoking type Author, reference Non- Cigar Pipe Cigar Cigarette Cigarette Mixed Cigaretts smoker only only and and cigar and pipe (cigarette only pipe and other) Hammond and Horn? (40)_._ 1.00 1.22 1.12 1.10 1. 36 1. 50 1. 43 1. 68 Doll and Hill (26). 100 _--- Lt) LOL LL. LL 1.11 1, 28 Best (9)________ 1.00 1.06 1.05 .98 1, 22 1. 26 1.13 54 Kahn (60)_____. 1.00 1.10 107 1.08 ___... _..___. 1. 1 1. 84 Hammond ? (38) _ ee 1.00 1.25 1.19 101 _...-8 _. 1 1. 57 1. 86 1 Only mortality ratios for ages 50 to 69 are presented. ? Only mortality ratios for ages 54 to 64 are presented. Mortality and Dose-Response Relationships A consistent association exists between overall mortality and the total dose of smoke a cigarette smoker receives. The methods most frequently used to measure dosage of tobacco products are: Amount smoked, degree of inhalation, duration of smoking experience, age at initiation, and the amount of tar in a given tobacco product. For cigarette smokers, the higher the dose as measured by any of these parameters, the greater the mortality. The significance of the smal] increase in overall mortality that occurs for the entire group of pipe and cigar smokers can be analyzed by examining the mortality of subgroups defined by similar measures of dosage as used in the study of cigarette smokers. Amount SMOKED Hammond and Horn (40) reported an increase in the overall mor- tality of pipe and cigar smokers with an increase in the amount smoked. Individuals who smoked more than four cigars a day or more than 10 pipefuls a day had death rates significantly higher than men who never smoked (P<0.05 for cigar smokers and P<0.05 for pipe smokers) (table 7). Cigar and pipe users who smoked less than this amount experienced an overall mortality similar to men who never 180 smoked. The study of Canadian veterans (9) also contained evidence of a dose-response in mortality by amount smoked for cigar smokers. No dose-response relationship was observed among pipe smokers (table 8). Kahn (50) reported a consistent increase in overall mortality with an increase in the amount smoked for both pipe and cigar smokers (table 9). Hammond (38) found no consistent relationship between overall mortality and the number of elgars or pipefuls smoked (table 10). TaBLE 7.—Mortality ratios Jor total deaths of cigar and pipe smokers by amount smoked— Hammond and Horn Number of deaths Amount smoked oo Observed Expected Mortality ratio Nonsmoker_._-__-_...---__- 1, 664 1, 664 1. 00 Cigar only: Total... 2-2 653 598 1. 09 1 to 4 cigars___._-8 ee 410 400 1. 03 >4 cigars-_______88e 229 185 1, 24 Pipe only: Total_ 22-22 609 560 1. 09 1 to 10 pipefuls___.-.-2 8 391 374 1.05 > 10 pipefuls------ 22 204 172 1.19 Source: Hammond, E. C., Horn, D. (40). TaBLe 8.—Mortality ratios for total deaths of cigar and pipe smokers by amount smoked—Best Number of deaths Amount smoked Observed Expected Mortality ratio Nonsmoker..-.------ 2-22 2-2 1. 00 Cigar only: Total___.-_2- 2. 90 82. 07 1.10 1 to 2 cigars.__._- 64 56. 05 1.14 3 to 10 cigars._______.__- 23 19. 40 1.19 > 10 cigars__..-_-__-____ 8 1 1. 59 . 63 Pipe only: Total. 2-8 eee 570 566. 99 1. 00 1 to 10 pipefuls_..-- 222 374 370. 09 101 10 to 20 pipefuls...--_..- 141 140. 84 1. 00 1. 00 > 20 pipefuls....-2_2 36 35. 90 Source: Best, E. W. R. (9). 181 The above evidence suggests that a dose-response relationship may exist between the number of cigars and pipefuls smoked and overall] mortality. However, because of the high-mortality rate of ex-smokers of cigars and pipes, it is difficult to interpret the data presented with- out including this group with the continuing smokers. Without data which examines patterns of both daily rate of smoking and inhalation at various age levels, no firm conclusions can be drawn as to the nature of this dosage relationship. TaBLEe 9.—Mortality ratios for total deaths of cigar and pipe smokers by age and amount smoked—Kahn Mortality ratio, age Amount smoked 55 to 64 65 to 74 Nonsmoker_______.-__--.2 2 2-2-2 eee 1. 00 1. 00 Cigar only: Total___------ 22 --eeeeee 1.01 1. 08 1 to 4 cigars per day.____..._--_____________. . 89 1. 00 5 to 8 cigars per day_____...___ 2-2 1.14 1. 23 >8 cigars per day_____..__. lee 1. 65 1. 28 Pipe only: Total. _.-.-----2 2-2 eee 1. 08 1. 06 1 to 4 pipefuls per day_____.______-__________. 1. 16 .91 5 to 19 pipefuls per day____..____._.______._- 1. 04 1.10 >19 pipefuls per day._____.____-.-- eee ee 1. 18 Source: Kahn, H. A. (56). TaBLE 10.—Mortality ratios for total deaths of cigar and pipe smokers by amount smoked—Hammond Amount smoked Mortality Amount smoked Mortality tatio ratio Nonsmoker.____..___._____- 1. 00| Current pipe smokers: Current cigar smokers: Total___-.2 22 ee 1. 04 Total___ 2-2 ee 1. 09 1 to 9 pipefuls per day____- 1. 08 1 to 4 cigars per day_______ 1. 03 >9 pipefuls per day______- . 92 >4 cigars per day.____.___ 1.18 Source: Hammond, E. C, ($8). 182 INHALATION Inhalation of tobacco smoke directly exposes the bronchi and the lungs to smoke and results in the absorption of the soluble constituents of the gas and particulate phases. Without inhalation tobacco smoke only reaches the oral cavity and the upper digestive and respiratory tracts and does not reach the lungs where further direct effects and systemic absorption of various chemical compounds can occur. Although the smoker has some voluntary control over the inhalation of smoke, the physical and chemical properties of tobacco smoke to a degree determine its acceptability and “inhalability.” The condensate of pipe and cigar smoke is generally found to be alkaline when the pH is measured by suspending a Cambridge filter in CO,-free water. Cigarette condensate is slightly acidic as measured by this method. Since alkaline smoke is more irritating to the respira- tory tract, it has been assumed that the more alkaline smoke of pipes and cigars was in part responsible for the lower levels of inhalation reported by pipe and cigar smokers. Brunnemann and Hoffmann (15) have analyzed the pH of whole, mainstream smoke of cigarettes and cigars on a puff-by-puff basis using a pH electrode suspended in main- stream smoke. Smoke from several U.S. brands of cigarettes was found to be acidic throughout the entire length of the cigarette. Of interest was the finding that cigar smoke also had an acidic pH for the first two-thirds of the cigar and became alkaline only in the last 20 to 40 percent of the puffs from the cigar. Available epidemiological evidence indicates that most cigar smokers do not inhale the smoke and most cigarette smokers do. The fact that smoke from the first half or more of a cigar is acidic, near the range of pH values commonly found in Cigarette smoke, and becomes alkaline only toward the end of the cigar might suggest that the pH of the smoke of a tobacco product May not be the only factor that influences inhalation patterns. Per- haps “tar” and nicotine levels as well as the concentration of other “irritating” chemicals also affect the degree to which a tobacco smoke will be inhaled. Nicotine is rapidly absorbed into the blood stream from the lungs *hen tobacco smoke is inhaled. The amount of nicotine absorbed from the lungs is primarily a function of the nicotine concentration in the smoke and the depth of inhalation. Some nicotine may also be ab- Sorbed through the mucous membranes of the mouth. This is more likely to occur under alkaline conditions when nicotine is unprotonated (3, 15, 79). This suggests that cigar smokers may be able to absorb Some nicotine through the oral cavity without having to inhale, par- ticularly during the time that the smoke from the cigar is alkaline. 183 With the development of sensitive measures of serum nicotine level, (48) the extent to which nicotine is absorbed through the membraneg of the mouth in pipe and cigar smokers can be more accurately determined. , Inhalation patterns of smokers were determined in several of the large prospective and some of the retrospective epidemiological studies, Inhalation was usually determined by the administration of a ques. tionnaire that required a subjective evaluation of one’s own patterns of inhalation. Although the accuracy of these questionnaires has not been confirmed by an objective measure of inhalation, such as carboxy. hemoglobin or serum nicotine levels, their reliability is supported by mortality data which demonstrate higher overall and specific death rates with self-reported increases in the depth of inhalation. Doll and Hill (26) and Hammond (38) presented information on inhalation patterns of pipe, cigar, and cigarette smokers (figs. 1, 2, 3, and table 12). Some 80 to 90 percent of cigarette smokers reported inhaling, with the majority of individuals inhaling moderately or deeply, whereas most pipe and cigar smokers denied inhaling at all, Pipe smokers reported slightly more inhalation than cigar smokers, For each type of smoking, less inhalation was reported by older smokers. This change may represent less awareness of inhalation, differences in smoking habits of successive cohorts of smokers, or it may reflect the operation of selective factors which favor survival of noninhalers. The Tobacco Research Council of the United Kingdom has, since 1957, periodically reported the use of tobacco products by the British, Figure 1.—Inhalation among pipe smokers by age. No inhalation Some inhalation 34.8 31.2 26.2 23.9 25.5 Age 40 50 60 70 80 SOURCE: Hammond, E. C. (38). 184 Figure 2.—Inhalation among cigar smokers by age—-Hammond. No inhalation Phataton 26.4 22.9 17.1 13.7 18.5 4 1 A ‘ 4 Age 40 50 60 70 80 SOURCE: Hammond, E. C. (38). Figure 3.—Depth of inhalation among cigarette smokers by age.—Hammond. None Slight inhalation Moderate inhalation __ Deep inhalation 12.1 9.2 4 4. L 4 Age 40 50 60 70 80 SOURCE; Hammond, E. C. (38). Recent reports edited by Todd have contained data on the inhalation pattern of cigar, pipe, and cigarette smokers (92, 93, 94). Table 11 shows that most cigarette smokers inhale a “lot” of “fair amount” whereas most pipe and cigar smokers do not inhale at all or “just a little.” Little change is observed in the inhalation patterns of a given product since 1968. Best (9) reported inhalation data among male cigarette smokers by smoking intensity and age group, but did not report the inhalation 185 patterns of pipe and cigar smokers, The overall mortality rates of current pipe smokers who inhaled at least slightly were reported by Hammond (38) as being somewhat higher than for men who never smoked regularly. The overall mortality rates of current cigar smokers who reported inhaling at least slightly were appreciably higher than for men who never smoked regularly (table 13). Available evidence indicates that cigarette smokers inhale smoke to a greater degree than smokers of cigars or pipes. Once a smoker has learned to inhale cigarettes, however, there appears to be a tendency to also inhale the smoke of other tobacco products. For cigars, this is evidently true whether one smokes both cigarettes and cigars or switches from cigarettes to cigars (tables 14, 15, 16). Bross and Tidings (74) examined the inhalation patterns of smokers of large cigars, cigarettes, and those who switched from one tobacco product to another (table 15). Nearly 75 percent of those who were currently smoking only cigarettes reported inhaling “almost every puff” and only 7 percent never inhaled. The opposite was true for per- sons who had always smoked only cigars among whom 4 percent re- Tape 11.—The extent of inhaling pipes, cigars, and cigarettes by British males aged 16 and over in 1968 and 1971 Tobacco product Cigars Pipes Cigarettes Amount of inhalation a 1968 1971 1968 1971 1968 1971 Inhale a lot___-__-------------.---- 23 19 8 8 47 47 Inhale a fair amount___.__-_--_.---- 16 19 10 8 31 30 Inhale just a little.___.._------.---- 27 27 24 26 13 15 Do not inhale at all_._____..--_----- 34 35 59 58 9 8 Total___..__-._----_--------- 100 ©100 100 100 =§=100 100 Source: Todd, G. F. (98, 94). TasLe 12.—IJnhalation among cigar, pipe, and cigarette smokers by age-—Doll and Hill Percentage of inhalers, age Smoking type 25to34 35to44 45to54 S5to064 65 to 74 >74 Cigar and pipe__.------------- 12.00 10.00 7.00 5.00 4.00 4. 00 Mixed (cigarette and other)..... 74.00 60.00 47.00 36. 00 30.00 26. 00 Cigarette only_--_------------ 90.00 85.00 75.00 66.00 58.00 41.00 Source: Doll, R., Hill, A. B. (#6). 186 ported inhaling almost every puff and 89 percent said they never inhaled. Cigar smokers who also smoked cigarettes reported inter- mediate levels of inhalation between the cigar only and cigarette only categories. Inhalation patterns were similar whether the individual continued to smoke both products, stopped smoking cigarettes but continued smoking cigars, or stopped smoking cigarettes and switched to cigars. In all three groups, about 20 percent reported inhaling “almost every puff.” This suggests that once an individual’s inhalation patterns are established on cigarettes, he may be more likely to inhale cigar smoke if he switches to cigars, or uses both cigars and cigarettes, than the cigar smoker who has not smoked cigarettes. Todd (93) reported similar data for a sample of smokers in the United Kingdom (table 16). The prevalence of inhaling a “lot” or “fair amount” of smoke was highest among cigarette smokers who were currently smoking cigarettes (77 percent) and lowest among current, cigar smokers who had previously smoked only cigars or pipes (18 percent). Individuals who switched from cigarettes to cigars main- TaBLe 13.—Mortality ratios for total deaths of cigar and pipe smokers by age and inhalation—Hammond Mortality ratio, age Inhalation 45 to 64 65 to 84 Nonsmoker________.__...--2---__--_.--- ee 1. 00 1. 00 Cigar only: Total__ 2222-2 ee 1. 09 98 No inhalation____....-_.2---__-e 1. 02 91 Some inhalation._..____..--__..__.__-. 1. 28 1. 37 Pipe only: Total. 2-2-2 eee 1. 04 95 No inhalation___..__._.._-__-.2_-o._ . 98 87 Some inhalation.....__...___________........ 1. 21 111 Source: Hammond, E. C. (38). TABLE 14.—Percentage of British male cigar smokers who reported inhaling a lot or a fair amount by type of product smoked 1968 1971 Type of product — Number of Percent Numberof Percent individuals individuals Cigars only__.......___.-.-.-8- 706 =—-:23.0 111 27.0 Cigars and cigarettes... -________ 1, 193 42.0 277 44.0 Cigars and pipes....--.____..__...__- 596 = 35.0 109 32.0 Cigars, cigarettes, and pipes._________ 26 52. 0 15 32.0 Source: Todd, G. F, (98, 94). 187 tained somewhat higher levels of cigar smoke inhalation than those cigar smokers who had never smoked cigarettes (30 percent). Todd (93) examined further the relationship between the inhalation of cigarette and cigar smoke. In general, cigarette smokers who switched to cigars were much less likely to report inhaling cigar smoke than cigarette smoke; however, those who in the past reported inhaling cigarette smoke a “lot” or “fair amount” were much more likely to report inhaling cigar smoke to the same degree than those ex. cigarette smokers who in the past did not inhale the smoke of their cigarettes (table 17). TaBLe 15.—Percentage of individuals reporting inhalation of “almost every puff” of tobacco smoke by current and previous tobacco usage and type of tobacco used Confidence Type of tobacco smoked Number Percen- limits of Type inhaled tage _ Current usage Previous usage patients inhaled Lower Upper Cigarettes only._._ Cigarettes only_... 2,359 Cigarette... 74.8 73.1 76.6 Cigars only. _-_--- Cigars only_____-- 649 Cigars_____ 45 3.0 60 Cigarettes and Cigarettes and 520 _____ do_____ 20.4 10.5 280 cigars. cigars. Cigars_____-_---- Cigarettes and 93 ____- do____- 183 9.0 30.0 cigars. None_____----_-- Cigarettes and 186 ____- do____- 21.5 17.8 242 cigars. Cigars....------- Cigarettes only---- 64 ..__- do_____ 17.2 16.0 280 Source: Bross, I. D.J., Tidings, J. (74). Tas ie 16.—Percentage of British males who reported inhaling a lot or fair amount of cigar smoke by current and previous tobacco usage and type of tobacco previously smoked (1968) Type of tobacco smoked Number of Percentage individuals Type inhaled inhaled Current usage Prev.ous usage Cigarettes only_.._-.__ Cigarettes only______ 2,586 Cigarette._.__ 77,7 Cigars only._...-.---- Nonsmoker-___----. 306 Cigars. .____- 18.0 Cigars only_.-.__------ Cigarettes only_.-.--- 321 ____- do_____-- 30. 0 Source: Todd, Q. F. (94). Taste 17.—Eztent of reported inhalation of cigar smoke by British male cigar smokers who were ex-cigarette smokers in 1968, analyzed by extent of reported inhalation of cigarette smoke when previously smoking cigarettes Extent of inhaling cigarettes Extent of inhaling cigars Inhale a lot Inhale a little or fair amount or not at all Percent Percent Inhale a lot or fair amount__._.______._..._______ 44.0 5. 0 Inhale a little or not at all.------ 56. 0 95. 0 a Total... 2222 ee. 100. 0 100. 0 wee Sample size______-- 222 244 56 Source: Todd, G. F. (93). Specifie Causes of M ortality Cancer Several prospective epidemiological studies have shown a signifi- cantly higher overall cancer mortality among pipe and cigar smokers compared to the cancer mortality of nonsmokers (table 18), Pipe and cigar smokers have much higher rates of cancer at certain Sites than at others. The upper airway and upper digestive tracts appear to be the most likely target organs. The relationship of pipe and cigar smoking to the development of specific cancers is detailed in the following sections. TABLE 18.—Mortality ratios for total cancer deaths in cigar and pipe smokers. A summary of prospective epidemiological studies Type of smoking Author, reference Nonsmoker Cigar only Pipe only Total pipe Cigarette and cigar only Hammond and Horn (40)____ 1. 00 1. 34 1.44 _oll 1. 97 Best (9)....-.-. 1. 00 1.13 1.38 _______. 2. 06 Hammond (88).____..._____ 100 2-0.) LL 1.21 1. 76 ahn (60)__.- 1. 00 1. 22 1. 25 1. 25 2.21 Cancer of the Lip Approximately 1,500 new cases of cancer of the lip are reported each year. Because of the possibility of early detection and surgical accessibility of cancers in this area, there are less than 200 deaths from cancer of the lip each year in the United States, Some of the earliest scientific investigations exploring the association between tobacco use and disease examined the smoking patterns of individuals with cancer of the lip. Broders (13) in 1920 examined the smoking habits of patients jp a retrospective study of 526 cases of epithelioma of the lip and 500 controls. Of the cancer cases, 59 percent smoked pipes, whereas this was true for only 28 percent of the controls. No association was found between cigar or cigarette smoking and cancer of the lip. In a restrospective study of 439 clinic patients with cancer of the lip and 300 controls conducted in Sweden, Ebenius (32) reported a significant association between pipe smoking and cancer of the lip. A total of 61.8 percent of the lip cancer cases smoked pipes, while only 22.9 percent of the controls smoked pipes. No association was found between the use of cigarettes, cigars, or chewing tobacco and cancer of the lip. In other retrospective studies, Levin, et al. (60) reviewed a series of 143 cases of cancer of the lip, and Sadowsky, et al. (77) reviewed 571 cases of cancer of the lip. In both studies, a strong association was found between pipe smoking and cancer of the lip. No significant association was found between the use of tobacco in other forms and cancer at this site. In a study of environmental factors in cancer of the upper alimen- tary tract, Wynder, et al. (173) found an association between Pipe smoking, cigarette smoking, and cancer of the lip. There were only 15 cases of cancer of the lip in this study. Staszewski (87) examined the smoking habits of 394 men with carcinoma or precancerous lesions of the lips. An association was found between the smoking of pipes and cigars and cancer of the lip, but this was only of doubtful significance. A significant association was found between the use of cigarettes and cancer of the lip. Keller (51) conducted a study of lip cancers in which he considered a number of factors including histologic types, survival, race, occupa- tions, habits, and associated diseases, A total of 304 patients with primary basal cell or squamous cell carcinoma of the lip and 304 controls from the same hospital matched for age and race were con- sidered in this series. A significant association was found between smoking in all forms and combinations and carcinoma of the hip. It was also found that increasing age and outdoor occupations with exposure to the sun were equally significant factors in the etiology of lip cancer. 190 In summary, it appears that there are several factors involved in the etiology of cancer of the lip. Among the various forms of tobacco use, pipe smoking either alone or in combination with other forms of smoking seems to be a cause of cancer of the lip. Table 19 summarizes the results of these retrospective studies. Oral Cancer The lips, oral cavity, and pharynx are the first tissues exposed to tobacco smoke drawn in through the mouth. Variations in inhalation during the smoking of various tobacco products result in different pat- terns of distribution of smoke throughout the respiratory tree. How- ever, the oral cavity and adjacent tissues are the sites most consistently exposed to tobacco smoke. For this reason, differences in inhalation should result in less variation in exposure to tobacco smoke for these sites than for the lower trachea and the lung. The inherent carcinogen- icity of pipe, cigar, and cigarette smoke is most reliably compared at those tissue sites where dosage and exposure to tobacco smoke are most nearly equal. Data from the epidemiological studies suggest that little difference exists between the smoking of cigarettes, pipes, or cigars and the risk of developing oral cancer. Hammond and Horn (40) examined the association between smok- ing in various forms and cancer of the combined sites of lip, mouth, pharynx, larynx, and esophagus. The mortality ratios were 5.00 for cigar smokers, 3.50 for pipe smokers, and 5.06 for cigarette smokers compared to nonsmokers. All the deaths from cancer of the hp, oral cav- ity, and pharynx reported by Doll and Hill (26) occurred in smokers. The death rates from cancer at these sites were 0.04 per 1,000 for pipe and cigar smokers, 0.10 per 1,000 for mixed smokers, and 0.05 per 1,000 for cigarette smokers. A fairly detailed analysis of oral cancer was pre- sented by Kahn (50) who differentiated between cancer of the oral cavity and cancer of the pharynx. The mortality ratios for oral cancers Were 1.00 for those who never smoked, 3.89 for all pipe and cigar smokers, and 4.09 for cigarette smokers. A further breakdown of the pipe and cigar smokers demonstrated a mortality ratio of 4.11 for cigar smokers, 3.12 for pipe smokers, and 4.20 for smokers of pipes and cigars. For cancer of the pharynx, the mortality ratios were 1.00 for those who never smoked, 3.06 for all pipe and cigar smokers, and 12.5 for cigarette smokers. No deaths occurred among those who smoked only cigars. The mortality ratio was 1.98 for pipe smokers and 7.76 for smokers of pipes and cigars. Hammond (38) combined cancers of the lip, oral cavity, and pharynx. The pipe and cigar smokers had a mortality ratio of 4.94 and the cigarette smokers a mortality ratio of 9.90 compared to nonsmokers. 191 TéL TABLE 19.—Relative risk of lip cancer for men, comparing cigar, pipe, and cigarette smokers with nonsmokers. A summary of retrospective studies Relative risk ratio and percentage of cases and controls by type of smoking Author, reference Number ~- —-- —— ere Nonsmoker Cigar only Pipe only Total pipe Cigarette Mixed and cigar only Broders (13): Relative risk..-..-_ | 1.0 0. 8 4.3 LL 0 LLL Cases. 537 Percent cases___________ 7 19 410 lle Lo elle Controls_..-. 222-8 500 Percent controls_________ 4 16 6 LLL 2600 LLL Ebenius (32): Relative risk___..-- 208. 1.0 7 4.1 050 222 eee Le Cases__. 2 ee el eee _... 439 Percent cases. .___._____ 49 6 41 re Controls—9 2.2 300 Percent controls__.______ 65 12 13 10, LLL wee Levin, et al. (60) Relative risk... 1.0 19g 2.9 -oeeee 14 oo Cases... 22s 143 Percent cases... | 15 27 48 LL 45 LLL Controls... 554 Percent controls_________ 22 20 2400 LL 46 lll Sadowsky, et al. (77) Relative risk... __ 1.0 11 4.3 2.6 14 0. 4 Cases___ 571 Percent cases.____..____ 8 18 6 44 22 Controls___ 615 Percent controls.________ 13 3 7 4 53 19 Wynder,! et al. (113) Relative risk__..._._____ 0 8 18 21222 LLL L0 2.2 Cases... 2, 14 Percent cases_______.___ 0 7 2900 LLL 36 29 Controls___. 222-2 115 Percent controls_________ 24 9 WL 36 13 Staszewski (87): Relative risk_.._________ |S | 2.1 24 2-1 Cases____. 394 Percent cases__..________ Gwenn Lee 12 73 2 Controls_._.-- 2-2. 912 Percent controls.________ 1 ll 61 _-L LLL. Keller: (61): Relative risk_...__..._._ 1.0 14 4.0 2.6 Cases___ 2. wae nee eee 301 Percent cases___-_ ______ 7 2 6 l 60 6 Controls___-..-2 2 265 Percent controls_________ 17 4 3 0 53 0 ) Percentage based on less than 20 patients. Ratios: relative to cigarette smokers. These studies are summarized in table 20. They demonstrate that smokers experience a large and significant risk of developing cancer of the oral cavity compared to nonsmokers. This risk seems to be about the same for all smokers whether an individual uses a Pipe, cigar, or cigarette, A number of retrospective studies have examined the relationship between smoking in various forms and cancer of the oral cavity. The results of these studies are presented in table 21. Some of the variations in relative risk of developing oral cancer observed in the retrospective studies is probably due to the lack of a uniform definition of oral cancer by anatomical site and the various means used in selecting and defin- ing cases and controls. It appears, however, that a significant risk of developing oral cancer exists for smokers compared to nonsmokers and this risk is similar for smokers of pipes, cigars, and cigarettes. Several epidemiological investigations have demonstrated an asso- ciation between the combined use of alcohol and tobacco and the development of oral cancer. A few of these studies (52, 62, 63, 109) contain data on pipe and cigar smokers. Heavy smoking and heavy drinking are associated with higher rates of oral cancer than are seen with either habit alone. TaBLE 20.—Mortality ratios for oral cancer in cigar and pipe smokers. A summary of prospective epidemiological studies Smoking type Author, reference — Pipe Total pipe Cigarette Mixed Non- Cigar smoker only only and cigar only Hammond and Horn!(40)_ 1. 00 5. 00 3.50 _______- 5.06 ___._ _- Doll and Hill? (26,27)... 0.00 _.__..... ______ 0. 80 1. 00 2. 00 Hammond ($8) _.- 1.00 -2---222 LL LLL 4. 94 39.90 .._-____ Kahn (60): Oral +.. 22-2 1. 00 4.11 3.12 3. 89 4.09 _.._____ Pharynx____.___ 1.00 .. 2 LLL 1. 98 8.06 12.54 __..____ ' Combines data for oral, larynx, and esophagus. ? Ratios: relative to cigarette smokers. 5 Mortality ratios for ages 45 to 64 only are presented. ‘ Excludes pharynx. Cancer of the Larynx The larynx is situated at the upper end of the trachea. Because of Its proximity to the oral cavity, the larynx probably has a similar *xposure to smoke drawn through the mouth as the buccal cavity and pharynx. Tobacco smoke that is not inhaled may still reach as far as the larynx and upper trachea. Pipe and cigar smokers develop cancer of the larynx at rates comparable to those of cigarette smokers. These 193 vél TaBLY 21.—Relative risk of oral cancer for men, comparing cigar, pipe, and cigarette smokers uith nonsmokers. A summary of retrospective studies Relative risk ratio and percentage of cases and controls by type of smoking Author, reference Number — Nonsmoker Cigaronly Pipeonly Total pipe Cigarette Mixed and cigar only Mills and Porter (64): Relative risk_....._---._- 10-22-22. ee Lee 7.0 4,1 0.00 2 L- Cases________. eee 124 Percent cases.__.._.____- WW Lee eee Lee 55 386 Lee Controls_.......-.--__-_- 185 Percent controls.__._._._- 38 _8 le Lee 30 820 ee Sadowsky, et ai. (77): Relative risk._._._-___-_- 1.0 2.0 4.4 20. 1.4 21 Cases.___ 2 eee 1,136 Percent cases_._.. 2... 8 4 1B Lee 42 28 Controls_._______________ 615 Percent controls.._.._.._- 13 3 T woe ee 53 23 Schwartz, et al. (83): Relative risk_.-._....___. 10 --.-.___. L6 _L 222 Le 15 _..2L LL. Cases__._..-_ 22 332 Percent cases__.___.____- WLLL ne 63 LLL ele Controls.........--.--_-- 608 Percent controls..________ 23 77 re 58 LL. ee Wynder, et al. (109): Relative risk___..-.-____- 1.0 3. 6 6.1 ____---_- 3.0 3.3 Cases_______.__-_--._-__ 543 Percent cases________.__- 3 20 Wo LeeLee 57 8 Controls_____._._._----.- 207 +Percent controls___._.____ 10 13 6 _LLLLLlee 63 8 Wynder, et al. (113): Relative risk_--__.._-__--- 1.0 1.7 a 1.2 1.4 Cases_..- 115 Percent cases___.._______ 23 13 W220 LLL 37 16 Controls__._..._..2-_---- 115 Percent controls..___..___ 26 9 WG LLL 36 13 Sét a © on oS we % ° hy as | me nn Wynder, et al. (116): Cases__._-___- 178 Percent cases____________ Controls_..---2. 8 220 =Percent controls__._.____. Pernu (73): Relative risk_.___________ Cases_____-_ ee 1,400 Percent cases____________ Controls___.-- = 713 Percent controls__.._____. Staszewski (87): Relative risk__.._________ Cases... 383 Percent cases____________ Controls_...-__-. ee 912 Percent controls_.__._____ Keller (62): Relative risk....___._____ Cases... 408 Percent cases____________ Controls._.0.222 408 Percent controls..________ Martinez (62): Relative risk...---- Cases___.. 2 170 Percent cases____.________ Controls_..--- 510 Percent controls_.._______ Martinez! (63): Relative risk..._________. Cases. _ 22 346 Percent cases___________. Controls___...--2 2. 346 Percent controls__________ 1.0 12 22 50 72 61 69 39 44 34 36 an ' This study combines data for oral cancer and cancer of the esophagus. rates are several times the rates of nonsmokers. The similarity of the mortality ratios of cancer of the larynx for smoking in various forms suggests that the carcinogenic potentials of the smoke from cigars, pipes. and cigarettes are quite alike at this site. Several of the prospective epidemiological studies include data on deaths from cancer of the larynx for pipe and cigar smokers as well as for cigarette smokers. Hammond and Horn (40) combined data for cancer of the larynx with cancer of the esophagus and oral cavity. The mortality ratios compared to nonsmokers were 5.00 for cigar smokers, 3.50 for pipe smokers, and 5.06 for cigarette smokers. There were no deaths from carcinoma of larynx among nonsmokers in the study of British physicians by Doll and Hill (26); however, the death rate for cancer of the larynx among pipe and cigar smokers was 0.10 per 1,000 while the death rate for cigarette smokers was 0.05 per 1,000. Kahn (50) reported mortality ratios for cancer of the larynx of 10.33 for cigar smokers, 9.44 for pipe and cigar smokers, 7.28 for all pipe and cigar categories combined, and 9.95 for cigarette smokers. No deaths from cancer of the larynx occurred in pipe smokers. Hammond (38) reported a mortality ratio of 3.387 for all pipe and cigar smokers and a mortality ratio of 6.09 for cigarette smokers in the age category 45 to 64, These studies are summarized in table 29. Several retrospective studies have examined the smoking habits of patients with cancer of the larynx and appropriately matched controls. The small number of pipe and cigar smokers in each study results in relative risk ratios that are quite unstable; however, it appears that pipe and cigar smokers experience a risk of developing cancer of the larynx that is similar to the risk observed among cigarette smokers (table 18). TaBLE 22.—Mortality ratios for cancer of the larynz in cigar and pipe smokers. A summary of prospective epidemiological studies Smoking type Author, reference _ Non- Cigar only Pipe only Total pipe Cigarette Mixed smoker and cigar only Hammond and Horn! (40)_----------------- 1. 00 5. 00 3.50 __------ 5.06 __.----- Doll and Hill? (26, 27)__. 0.00 22-22-20 Loe e-e 2. 00 1. 00 0. 60 Hammond (38)_..-..2-.- 1.00 _------2 LL eee 3.387 36.09 _.-____- Kahn (50)_...-.----.--- 100 10.388 _.___- 7. 28 9.95 __...--- ! Combines data for oral, larynx, and esop 1agus. ? Ratios: relative to cigarette smokers. 3 Only mortality ratios for ages 45 to 64 are presented. 196 Wynder, et al. (108, 113) distinguished between intrinsic and ex- trinsic larynx cancers. For smokers the relative risk of developing cancer of the intrinsic larynx was similar to the relative risk of lung cancer whereas the relative risk of developing extrinsic larynx cancer was more like the relative risk of cancer of the upper digestive tract. Histologic changes of the larynx in relation to smoking in various forms were described by Auerbach, et al. (5). Microscopic sections of the larynx fron: 942 subjects were examined for the presence of atypical nuclei and proliferation of cell rows. Sections were taken from four separate areas of the larynx in each case. Among those who smoked cigars and pipes but not cigarettes. only 1 percent had no atypical cells and more than 75 percent of the subjects had lesions with 50 to 69 percent atypical cells. Four of the cigar and pipe smokers had carcinoma in situ and in one of these four cases early invasion was seen in three of the sections. Of those who never smoked regu- larly, 75 percent had no atypical cells. The cigar and pipe smokers had a similar percentage of cells with atypical nuclei as cigarette smokers who smoked one to two packs per day. With respect to the prolifera- tion of cell rows in the basal layer of the true vocal cord, the least proportion of cases with eight or more cell rows was found in men who never smoked, and the greatest. proportion was found in heavy cigarette smokers. Pipe and cigar smokers had a distribution of cell rows that was comparable to that of cigarette smokers who consumed about a pack a day. Several retrospective studies have reported an association between the combined use of tobacco and alcohol and cancer of the larynx. A study by Wynder, et al. (108) included some information on pipe and cigar smoking in relation to drinking habits and the development. of cancer of the larynx, but because of the limited number of pipe and cigar smoking subjects this relationship could not be adequately determined. Cancer of the Esophagus The esophagus is not directly exposed to tobacco smoke drawn into the mouth; however, the esophagus does have contact with that portion of tobacco smoke that is condensed on the mucous membranes of the mouth and pharynx and then swallowed. The esophagus is also ex- posed to a portion of tobacco smoke that is deposited in the mucus cleared from the lung by the ciliary mechanism or by coughing. Varia- tions in inhalation of a tobacco product may not appreciably alter the &xposure the esophagus receives from smoke dissolved in mucus and saliva. This suggestion receives support from the prospective and retrospective epidemiological studies which demonstrate similar mor- tality rates for cancer of the esophagus in smokers of cigars, pipes, and cigarettes. 197 861 TaBLe 23.—Relative risk of cancer of the larynx for men, comparing cigar, pipe, and cigarette smokers with nonsmokers. A summary of retrospective studies Relative risk ratio and percentage of cases and controls by type of smoking Author, reference Number —— - Nonsmoker Cigaronly Pipeonly Total pipe Cigarette Mixed and cigar only Schrek, et al. (81): Relative risk___.________ 1.0 0 Lhd ill 7 Cases___..22 2-2-8 ee 73 Percent cases ___ 14 0 TO lee 8000 LL Controls__.-. 222-222 ee §22 Percent controls___._____ 24 10 Wo LLL 59° Le Sadowsky, et al. (77): Relative risk.___.__ _____ 1.0 2.2 2.3 2-22 Le 3.7 4,1 Cases___..-_ eee. 273 Percent cases_.____..___ 4 2 5 LLL 60 29 Controls... 615 Percent controls__._______ 13 3 , 53 23 Wynder, et al. (108): Relative risk__....______ 1.0 15.5 27.7 11.1 24.6 .-- 2 2 Cases___2 ee 209 Percent cases.____..____ .5 8 5 1 86 LLL. Controls... 2-8. 209 Percent controls_________ 11 10 4 2 74 ok Wynder, et al. (113) Relative risk.._.________ 1.0 9.7 4.5 _-__ Le 6.3 6.3 Cases_.._.2 2-2-8. 60 Percent cases____._____- 5 17 1 Lele 47 17 Controls.._..___-. 22-8. 271 Percent controls__.______ 24 9 WL le 36 13 Wynder, et al. (116): Relative risk. ..__..___. 1.0 14.5 16.0 _--- LL. 22. 0 16. 0 Cases____ 08 142 Percent cases__.________ 1 20 Lo wlll 62 16 Controls_.... 222 220 Percent controls_________ 16 22 To Loe 45 16 661 Pernu (73): Cases___-__ ee 546 Controls._...-222 22 713 Staszewski (87): Cases... eee 207 Controls... 2-2 912 Svoboda (90): Cases___.-.22 22 205 Controls... 22 2-22. 320 Stell (88): 78 50 50. 2 61 95 71 79 50 In the prospective epidemiological studies, cigar, pipe, and cigarette smokers all had similar mortality ratios from cancer of the esophagus, Hammond and Horn (40) combined the categories of carcinoma of the esophagus, larynx, pharynx, oral cavity, and lip and described mortality ratios of 5.00 for cigar smokers, 3.50 for pipe smokers, and 5.06 for cigarette smokers. Doll and Hill (26) reported an esophageal cancer mortality ratio of 2.0 for pipe and cigar smokers, 4.8 for mixed smokers, and 1.5 for cigarette smokers. Kahn (50) reported the fol- lowing mortality ratios for srnoking in various forms compared to non- smokers: cigar only, 5.33; pipe only, 1.99; pipe and cigar, 4.17; all pipes and cigars combined, 4.05; and cigarettes only, 6.17. The results of these prospective studies are summarized in table 24, Several retrospective investigations have also examined the associa- tion between smoking in various forms and cancer of the esophagus. These studies have been summarized in table 25. The evidence sug- gests that cigar, pipe, and cigarette smokers develop cancer of the esophagus at rates substantially higher than those seen in nonsmokers, and that little difference exists between these rates observed in smokers of pipes and cigars and cigarettes. Histologic changes in the esophagus in relation to smoking in vari- ous forms were investigated by Auerbach, et al. (7), who looked for atypical nuclei, disintegrating nuclei, hyperplasia, and hyperactive esophageal glands. A total of 12.598 sections were made from tissues obtained from 1,268 subjects. For each of the parameters investigated, pipe and cigar smokers demonstrated significantly more abnormal histologic changes than nonsmokers; however, these changes were not as severe or as frequent as those seen in cigarette smokers. Several retrospective studies conducted in the United States and other countries have examined the synergistic roles of tobacco use and heavy alcohol intake on the development of cancer of the esophagus. Four of these investigations contain data on Pipe and cigar smoking (12. 62, 63, 107). It appears that smoking in any form in combination with heavy drinking results in especially high rates of cancer of the esophagus. TaBLe 24.—Mortality ratios for cancer of the esophagus in cigar and pipe smokers. A summary of prospective epidemiological studies Smoking type Author, reference Non- Cigar Pipe Total Cigarette omy smoker only pipe and only Mixed cigar Hammond and Horn! (40) 1. 00 5. 00 3.50 222 Le 5. 06 __-__._-- Doll and Hill (26, 27)___. 1.00 _-.-- 222 LLL. 2. 00 1. 50 4, 80 Hammond (88)._________ 1.00 _-.-_-.-. LLL. 3.97 44.17 _______. Kahn (60)______._______ 1. 00 5. 33 1.99 4.05 6.17 ..L_L-- ' Combines data for oral, larynx, and esophagus. 2 Mortality ratio for ages 45 to 64. 200