TABLE 25.-Relative risk of cancer of the esophagus for men, comparing cigar, pipe, and cigarette smokers with nonsmokers.
                  A summary of retrospective studies

Author, reference

              Relative risk ratlo and percentage of cases and controls by type of smoking
Number I---__---               ____~-        ----_-_
                       Nonsmoker Cigar only  Pipe only  Total pipe  cigarette    Mixed
                                           and cigar    OIllY

Ssdowsky, et al. (77):                Relative risk _______ ___ __
  Cases__-__________-______.._- 104    Percent cases- - _ _ ___ _ ___
  Controls--__------ ___________ 615   Percent controls ____ _ _ _ _ _

Wynder, et al. (213):
  Cases______________________- 39
  Controls _____ ____ ____________  115

Pernu (73) :
  Cases- - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ __ _  202
  Controls _____________________ 713

Relative risk _____ __ ____ _
Percent cases- - _ ______ -_
Percent controls _____ _ __ _

Relative risk ______ ____. _
Percent cases- _ _ _ _ _ _ _ _ _ _
Percent controls _________

Schwartz, et al. (84) :
  csses_______________________ 249
  Controls ________ - ____________ 249

Relative risk ____ _ _ _ _ _ _ _ _
Percent cases- _ _ _ _ _ _ _ _ __
Percent controls _________

Wynder and Bross (107):
  Cases___-__________--------- 150
  Controls _____ _ _ _ _ __ __ ________  150

Relative risk __________ __
Percent cases.. _ _ ________
Percent controls _________

1. 0     4. 8
4       5
13      3


1. 0     3. 1
13      15
24      9


1.0  _________
17   _ _ _ _ _ _ _ _ _
39   _ _ _ _ _ _ _ _ _


1.0  -__---__-
2   .`________
18   _ - _ _ _ _ _ __


1. 0     3. 6
5      19
15      16

3. 8     5. 1     3. 8      3. 3
8       6      60       18
7      4      53       19

2. 1  _ _ _ _ _ - - _ _
18   _ _ _ _ _ _ _ _ _
16   _ _ _ _ _ _ _ _ _

2. 6
51
36

3' 4


13

3.0 _________
7   - _ _ _ _ _ _ _ _
5   _ _ _ _ _ _ _ _ _

2. 7
59
50

5. 9
18
7

2.6 _________
2   _ _ _ _ _ _ - _ _
7   _ - - _ _ - - _ _

11. 7      8. 6
38       7
67       7

9. 0     6. 0     2. 8     3. 7
9       4      51       11
3       2      55       9


TABLE 25--l!elative risk of cancer of the esophagus for men, comparing cigar, pipe, and cigarette smokers with nonsmokers.
               A summary of retrospective studies.-Continued

Relative risk ratio and percentage of cases and controls by type of smoking
      Nonsmoker Cigar only  Pipe only  Total pipe   Cigarette
                             and cigar     O"lY

Mixed

Relative risk-_- - _ . _ . _ . _
Percent cases- _ _ _. _ _ _ _ - -
Percent controls _______ -_

1.0 __----___    4.8 ______ -_
15 _________ 41    _ _ _ _ _ _ _ _
32 _______ -_ 18    _ _ _ _ _ _ _ _

2. 3  - - _ - _ _ _ _ _
63 ._----___
58   . _ _ _ _ _ _ _ _

Martinez (62) :
  Cases---~_~___._---_---~-~-- 120
  Controls---- _______ -_-__-__-_ 360

Martinez 1 (63) :

Cases_____-___-_-_-_________ 346
Controls- _ _ _ _ _____ _________ _ 346

Relative risk------- _____
Percent cases----------_
Percent controls _________

Relative risk _____ _ ______
Percent cases- _ _ ______ __
Percent controls ____ _____

1. 0     2.0  __----_-   __- _____
8      9   --------   ____-_-_
14       8   ________ ________

1. 5      2. 2
31       43
34       34

1. 0     2. 0     2.8   ____ -_-_    1. 7      2. 5
21      10     15    ______ -_   34      34
22      9       1    - - _ _ _ _ _ _   36       25

1 Thfs study comblnea data for oral cancer and cancer of the esophagus.


Lung ca.ncer

Abundant evidence has accumulated from epidemiological, experi-
mental, and autopsy studies establishing that cigarette smoking is the
major cause of lung cancer. Several prospective epidemiological
studies have demonstrated higher lung cancer mortality ratios for pipe
and cigar smokers than for nonsmokers, but the risk of developing lung
cancer for pipe and cigar smokers is less than for cigarette smokers.
Table 26 presents a summary of these prospective studies. Dose-
response relationships such as those that helped demonstrate the nature
of the association between cigarette use and lung cancer could not be
as thoroughly studied for pipe and cigar smokers because of the rela-
tively few smokers in these categories. Although the number of deaths
mere few, Doll and Hill (96') reported increased death rates from lung
cancer for pipe and cigar smokers with increasing tobacco CO~SU~P-
tion (table 27). Kahn (50) also demonstrated a dose-response relation-
ship for lung cancer by the amount smoked (table 28).
A few of the retrospective studies contained enough smokers to allow
an examination of dose-response relationships for pipe and cigar smok-
ing and lung cancer (I: 61, 74, 77). An increased risk of developing
lung cancer was demonstrated with the increased use of pipes and
cigars as measured by amount smoked and inhalation. The retrospec-
tive investigation of Abelin and Gsell (1) is of particular interest. The
smoking habits of 118 male patients with cancer of the lung from a
rural area of Switzerland were compared with those reported in a sur-
vey of all male inhabitants of a town in the same region. About 20
percent of the population of this area were regular cigar smokers, the
most popular cigar being the Stiimpen, a small Swiss-made machine-
manufactured cigar cut at both ends with an average weight of 4.5 g.
In this investigation, cigar smokers experienced a risk of developing
lung cancer that was similar to the risk of cigarette smokers. A dose-
response relationship was demonstrated for inhalation and amount
smoked. These data suggest that the heavy smoking of certain cigars
lnay result in a risk of lung cancer that is similar to that experienced
by cigarette smokers.
Several pathologists have reported histologic changes in the
bronchial epithelium in relation to smoking in various forms. Knudt-
Son (57) examined the bronchial mucosa of 150 lungs removed at au-
'&PSy and correlated the histologic changes noted with the history
of smoking, age, occupation, and residence. Specimens obtained from
the six cigar and pipe smokers demonstrated basal cell hyperplasia;
however, there was no squamous or atypical proliferative metaplasia
as is frequently seen in the heavy cigarette smokers.
Sanderud (78) examined histologic sections from the bronchial tree
of 100 male autopsy cases for the presence of squamous epithelial

203


metaplasia. In this study, 39 percent of the population were non-
smokers, 20 percent were pipe. smokers, and 38 percent smoked cig-
aret,tes. A total of 80 percent of the pipe smokers and cigaret,te smokers
demonstrated squamous metaplasia of the bronchial tree, whereas only
54 percent of the nonsmokers had this abnormality.

Auerbach, et al. (6) examined 36,340 histologic sections obtained
from 1,522 white adults for various epithelial lesions including:
presence or absence of ciliated cells, thickness or number of cell rows,
atypical nuclei, and the proportion of cells of various types. The
pathologic findings in the bronchial epithelium of pipe and cigar
smokers are compared to those found in nonsmokers and cigarette
smokers (table 25). Pipe and cigar smokers had abnormalities that
were intermediate between those of nonsmokers and cigarette smokers,
although cigar smokers had pathologic changes that in some categories
approached t.he changes seen in cigaret.te smokers.

TABLE 26.-Mortality ratios for lung cancer deaths in male cigar and
     pipe smokers. A summary of prospective studies

Author, reference

NIXI-
smoker

Hammond and Horn (40) _   1. 00   3. 35   8. 50  - ____ -__  23. 12   19. 71
Doll and Hill ($6, 67') _ - _ _   1.00  ______-_   ------   6. 14   13. 29    7. 43
Best (9) ________________   1. 00    2. 94    4.35  ______ -_  14.91  ________
Hammond (38) ____ _ __ _ _ _   1. 00    1. 85    2. 24    1. 97    9. 20    7. 39
Kahn (60) _ _ _ _ _ _ _ _ _ . _ _ _
       _             1. 00    1. 59    1. 84    1. 67   12.14 __- ___._

TABLE 27 .-Lung cancer death rates jor cigar and pipe smokers by amount
            smoked-Doll and HilJ


            Smoking type               Death rate per 100  Number of deaths

Nonsmoker___-_______----------------------
Cigar and pipe:
  1 to 14 g. per day ______ -- _______________
  15t024g.perdayw---S------ ___________
  >24 g.perday___-___--_..__-__--------
Cigaretteonly_--- ____ -__-_--_---___- _______

0. 07           3


.42           12
.45           6-
.96           3-
.96          143

Source: Doll, R., Hill, A. B. (I).

204


TABLE 28.-Lung cancer mortality ratios for cigar and pipe smokers by
            amount smoked-Kahn

Smoking type

Mortality ratio   Number of deaths

<5 pipefuls per day_ _ _- ________ _. _. _ ._._
5 to 19 pipefuls per day_ _______. _______._
>19 pipefuls per day- _ _ ___ ____. _ __. ____ _
Cigar and pipe:

8 or less cigars, 19 or less pipefuls- - - - - - - - -
>S cigars, >19 pipefuls__- _ _ _. _ _ _. _ _ _. __ _

1. 00          78

1. 14           12
2. 64           11
2. 07            2

.77
2. 20
2. 47

1. 62
2. 19

2
12
3

18
2

Source: Kahn, H. A. (60).

205


i   TABLE %.--Relative risk qf lung cancer-for men, comparing cigar, pipe, and cigarette smokers with ncnwnokers. A sum-
                      mary of retrospective studies

               Relative risk ratlo and percentage of cases and controls by type of smolrlng
Number -------. -___                             ---~----
                       Nonsmoker Cigar only  Pipe only  Total pipe %-P    MLrad
                                           and olger

Levin, et al. (60) :

  Cases _____ --.._--- ____ - _.__ -   236
  Controls ._____. .__- ._________   481

Schrek, et al. (81):

  Cases __________. --._-_- _____    82
  Controls_-- _____ -__-- _______   522

Wynder and Graham (1 II) :
  Cases _____ -- __________ - ____ -   605
  Controls-_-..-- ______ -___--__   780

Doll and Hill (86) :

  Cases ___________ -- __________ 1,357
  Controls _____.__ -__- _______ -_ 1,357

Koulumies (66) :

  Cases- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _- _ _ _ _   812
  Controls- _ _ _ _ _________ __ _- __   300

Sadowsky, et, al. (77) :
  Cases-__---- ____ -___-___-___   477
   Controls~~--.-~--..~- _____ -..   615

Relative risk ____________
Percent cases.. _ _________
Percent controls _________

Relative risk ______ - _ ____
Percent cases--- -_- _____
Percent controls-F _ _ _ -_ __

Relative risk ______ ______
Percent cases- _ - ________
Percent controls---- - _ _ _ _

Relative risk ____ _ _ _ _ _ _ _ _
Percent cases- _ _ _ _ _ _ __ _ _
Percent controls- _ _ - - _ _ _ _

Relative risk ____ _ _ - - _ _ _ _
Percent cases- _ - L__ _ __ _ _
Percent controls ____ _____

Relative risk ____________
Percent cases.. _ - ________
Percent controls ____ _ __- _

1. 0     0. 7
15      11
22      23


1.0      .6
15       4
22      23


1. 0     5. 1
1       4
15      8


1.0 __-__ -___
5' 5  _--______
    - - - - - -
     -   _  -


1.0  ___- _____
.6  _____ -___
18   ______ -__


1. 0     2. 4
4       2
13       3

0.8 _________
14   - - _ _ _ - - _ -
25   _ _ _ _ _ _ _ _ _

2. 1  ^ _ _ _ - _ _ _ _
66   ---_---_ -
44   _- __-____

7 _____--__
5' __-__--_-
11 -____----

1.7  ____---_-
61   _ __ __ __ _ _
59   _ - - - - - - -
      _

3.6 ___- _____
4   - - - _ _ - - - -
12   __-_ _ ___ _

15.7  ---_-----
91   ______ -_-
65   _- _______

5. 1  - - _ - _ - _ _ _
4   ---_ - --_-
7   - - - - - - - - -

9.6  _________
74   -______ --
69   _______ -_

9.6 _________
2   ---- ---__
6   - - - - - - - _ -

29.3  ---_--- --
77   ------- --
76   _______ -_

1.4 -________    3. 7      5. 6
3   -- --_----    57       31
7   ----__-__    53       19


Relative risk _____ - ______
Percent cases- _ _ _ ____ ___
Percent controls--- _ _ ____

1. 0     2. 5     4.0 _________
4      13       6   _ _ _ _ _ - - - -
21      27      8   _ _ _ _ _ _ _ _ _

8. 5  _ _ - - _ _ _ _ _
77   _ - - - _ _ _ _ _
45   _ _ _ _ _ _ _ _ _

Randig (74) :

Cases---...-.--------------...   415
Controls~~ _- _ - - _ _ _ - - - - - _ _ _ -. _   381

Relative risk--_- _ _ _ _ _ _ _ _
Percent cases- _ _ _ _ _ _ _ _ _-
Percent controls-_- ______

1. 0     5. 3     5.0 _________
1      21      11   ______-__
6     19      11   _-___--__

5.0 __- ______
67   _ _ _ _ _ _ _ _ _
64   _---__ -__

Mills and Porter (66):
  Cases--..-- _______r____ -_-___   444
  Controls_ - - ._ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _   430

Relative risk ____ _ _ _ _ _ _ _ _
Percent cases- _ _ _ _ _ _ _ _ _ _
Percent controls--- ____ __

1.0  ----____-  __-__-__    0. 0
7   ----__---  _-___--_    37
31   _________  ________    26

5.4 _---_____
55   _ - - - _ _ _ _ _
43   _ - - - _ _ - _ _

Mills and Porter (66) :
  Cases----_~~~-~--~..---___---   484
  Controls_-- ______ - ____ - ______ 1,588

Relative risk---- - - _ _ _ _ _ _
Percent cases- _ ________ _
Percent controls _______ __

1.0  _____ -___  _____ -_-    2.8
8   -------_-  _-___-__    13
28   _______-_  ________    16

4.5 _________
78 _____ -___
57   _ - - - - - - _ _

Schwartz and Denoix (88) :
  Cases_____------_-.---___-_-   430
  Controls-- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _   430

Relative risk ____ __ _ _ _ _ _ _
Percent cases- _ _ _ _ ______
Percent controls-- _ ____ __

1.0 ---_____-    4. 7  _ - _ _ _ _ _ _ _
1   - - - - - _ _ _ -    6   - - _ _ _ - - _ _
11 _________ 14 ____ --___

13. 5 _-_-_____
96   _ _ _ _ _ _ _ _ _
78   _ _ _ _ _ _ _ _ _

Stocks (89) :

Cases_------------.-----..--- 2,101
Controls--- _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ 5,960

Relat,ive risk ____ _ _ _ _ _ _ _ _
Percent cases- _ _ _ _ __ ____
Percent controls--_ ____ -_

            3. 1
LO --------- 9   _ - _ _ _ _ _ _ _
2   -----__ --        - - _ - - _ - _ _
9   _ - _ _ _ _ _ _ _    13 ________ -

5.0 -----____
89   _ _ _ _ _ _ _ _ _
78   _ _ _ _ _ _ _ _ _

Lombard and Snegireff (61) :
  Cases--~--~.--_-_------.._-~-   500
  Controls-- ____ ._- ______ __-.__ 1,839

Relative risk ________ __ __
Percent cases- _ ___ __ _ ___
Percent controls- _ _ _ - _ _ _ _

1.0  -----___-  _____-__     1. 7
2   -------_-  _____---     4
10   _________  ______--    15

8. 1  _ _ _ _ _ _ _ _ -
95   _ _ _ _ _ _ _ _ _
75   _ _ _ - - _ _ _ _

Pernu (75) :                     Relative risk ____ _ _-_ _ __ _
  Cases--~-_-~-~_-_~--_-___--- 1,477   Percent cases- _ _ _ _ _ _ _ _ _ _
  Controls ___.____ -__-_-_-- ____   713   Percent controls _________

1.0 _______ --    4.2 _________
7   _ - _ _ _ _ _ _ -     4   - - _ _ _ - - - -
39 _-______- 5 -._- _--__

9. 2     11. 1
77      13
50       7


TABLE 29.--R&& risk of lung cancer for men, comparing cigar, pipe, and cigarette smokers with nowmokers. A sum-
               mary of retrospective st&i.es-Continued

Author. reterence          Number

Relative risk ratio and pm?.%ntage of ~a868 and controls by type of smoking
       Nonsmoker Cl@ only  Pipe only  Total pipe Ck;leytte
                            and cigar

  --
Mixed

Wicken (106) :                    Relative risk ____ ___ _____   `1.0 - _______- _-_- __--    2. 2     4. 3      4. 2
  Cases____..____._____-_--..~--- 803   Percent cases-- _______ -_    4 _________ ____ -___    10      78       7
  Controls--_. _ ___ _. -_ -_ __ _ _ __ _ __ 803  Percentcontrols--------- 14 - ____ -___ -_-___-_ 16      64       6

Abelin and Gsell (1):                Relative risk ____ _ _ - _ _ _ _ _    1. 0    30. 7    21. 8    39. 9    31. 0     24. 7
  Cases----.--_----....---.-------  118   Percent cases- _ _ _ _ _ _ _ _ _ _    2      28       7      58      25       24
  Controls_.- _ _ _ _ _ _ _ - - -                      __ _-    35      19       6      31      17       10
           _ _ _ _ _ _ _ . _  _        524   Percent controls-- ___

Wynder, et al. (216):               Relative risk ____ _ _ __ _ __ -    1.0 _-------- _-_-_-_-    2. 0    12. 4 --_-___-_
  cssas-----__---~-------------- 210   Percent cases- _ _ _ _ _ _ _ _ _ -    3   _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _    5      92   - ______ --
  Controls..- ____ _______ - ______ __ 420   Percent controls _________   21 - ________ _____ -_-    15      47   - _ _ _ _ _ _ - -


TABLE 30.~Changes in bronchial epithelium of male cigar, pipe, and cigarette smokers as compared to nonsmokers

Group

               Percant Bectloas  Percent a plus    Percant
Number ot Bectlons with                                         Pfxcent
                          call rows with             Total
subjects   epltheuum  wit;; ~illal
           P         clua     atypkd ceus
                                  present     sections

1st set (none vs. pipe vs. cigarette-matched
on 1:l basis):

  Nonsmoker- _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
  Pipeonly ____ --__--__-- _________ - ______
  Cigaretteonly--_-_----- ________ - ______
2d set (none vs. pipe vs. cigarette-matched
on frequency basis) :

  Nonsmoker- ____ _ ___ _ ____ -__ ______ ____ _
  Pipe only ____ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _-
  Cigarette only- _ _ ___ _ ______________ ___ _
3d set (none vs. cigar vs. cigarette) :

Nonsmoker- _- __ _____ _ ___ - __ _ _________ _
Cigaronly~._-__-___-__________________
Cigarette only- - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

20      985      21. 7       11. 2      2. 6    1,031       10. 3
20      924      65. 5      38. 1     37. 0      979       35. 9
20      914      96. 8      88. 6     95. 2      982       72. 1

25    1,246      22. 9
25    1, 164      68. 7
25    1, 126      96. 3

35    1,706      27. 4
35    1,733      90. 8
35    1,526      99. 0

13. 4       7
38. 7   38: 2   1,277       11. 5
           1,247       37. 9
88. 7     89. 5    1,237       75. 5

12. 7       8
   73: 6   1,748
40. 0         1,828
92. 7     97. 8    1,693

15. 3
52. 5
80. 2

Source: Auerbach et al. (6).


Tumorigenic Activity

The tumorigenic activity of tobacco smoke can be modified in both a
quantitative and qualitative sense. Physical or chemical changes in
tobacco that result in a reduction of total particuhte matter upon
combusion of a given quantity of tobacco may result in a reduction
of c.arcinogenic potential. Such factors as tobacco selection, tEat.ment,
blending, cut,, and additives may quantitatively alter tar production.
Wrapper porosity and filtration may also affect tar production.

Quantitative changes in the tumorigenic activity of tobacco tar on
a gram-for-gram basis can be produced by the selection and treatment
of tobacco, the use of additives or tobacco sheets, or adjustments in t,he
cut and packing de,nsity.

Combustion temperature can also produce quantitative changes ia
the particulate matter of tobacco smoke. Although high-temperature
burning produces less part,iculate matter in the smoke, it appears that
tumorigenic components occur in higher concentration when tobacco is
pyrolized at temperatures higher than 700" centigrade (34).

Cigars, pipes, and cigarettes `are similar in that they are smoked
orally and have a common site of introduction to the body. The tissues
of the mouth, larynx, pharynx, and esophagus appear to receive ap-
proximately equal exposure to the smoke of these products. Inhalation
causes smoke to be drawn deeply into the lungs and also allows for
systemic absorption of certain const.ituents of tobacco smoke which
then can be carried further to other organs.

Pipe tobacco and cigars vary from cigarettes in a number of charac-
teristics that can produce bot.h quantitative and qualitative changes in
the t,otal particulate matter produced by their combustion. Experi-
mental evidence suggests that although there is some difference in the
amount. and quality of tar produced by cigars, this cannot account for
the reduced mort.ality observed in cigar smokers compared to cigarette
smoker-s.

Experimental Studies

Several experimental investigations have been conducted to examine
the relative tumorigenic activity of tobacco smokn condensates obtained
from cipare.ttes. cigars, and pipes. Most of these studies were standard-
ized in an attempt to make the results of the cigar and pipe experiments
more directly comparable with the cigarette data and most used t,hr
shared skin of mice for the applicat.ion of tar. Tars from cigars, pipes.
and cigarettes were usually applied on an equal weight basis so that
qualitative differences in the tars could be determined. In several ex-
periments. the nicotine was extracted from the pipe and cigar conden-
sates in an attempt to reduce the acute toxic effects that resulted ia
animals from the high concentrations of nicot,ine frequently found in
these products.

210


Wynder and Wright (117) examined the differences in tumorigenic
nct,ivity of pipe and cigarette condensates. Tars were obt.ained by the
sllloking of a popular brand of king-size cigarettes and the same ciga-
rette tobacco smoked in I', standard-gracle briar bowl pipes. Both the
cigarettes and pipes were pufird three times a minute with a 2-second
luff and a 35ml. volume. Both the cigareltes and pipes attained similar
nlasinium combustion zone temperatures ; however. the use of cigarette
tohac.co in the pipe resulted in a combustion chamber temperature that
:treraged about 150" centigrade higher than temperatures achieved
when pipe tobacco was used. Chemical fractionation was accomplished
and equal concentrations of t,hc neutral fraction were applied in three
I\-eekly applications to the shawd skin of C-IF, and Swiss mice. The
results indicate that neut.ral tar obtained from cigarette tobacco smoked
in pipes is more active than that obtained in the usual manner from
cigarettes. About t,wice as many cancers were obtained in both the CAF,
and the Swiss mice, and t11e latent period was about. 2 months shorter.
Extending these data. Croning~~r. et, al. (-00) r3aiiiined the biologic
activity of tars obtained from cip;lrs. piles. ant1 cigarettes. Each form
of tobacco was smoked as it was manufactured in a manntr to simulate
human smoking or to maintain tobacco combustion. The whole tar was
applied in dilutions of one-to-one and one-to-two with acetone to the
shaved backs of female CAF, and female Swiss mice using three
:lpplications each week for the life-span of the animal. The nicotine was
Wractecl f ram the pipe and cigar condensates to reduce the acut,e
toxicity of the, solutions. The Swiss mice. pipe. cigar. and cigarette tars
produced both benign and malignant t.mmors. The incidence rates of
malignant tumors given as percents were : 44. 41: and 37. respectively.
These results suggested a somc~what higher degree of carcinogenic
clctivity for cigar and pile tars tllilll for c.igar-ette tar.
Similar results were reported by Kcnsler ($9) who applied conden-
Mes obtained from cigars and cigarettes to the shaved skin of mice.
The incidence of papiliomas produced by cigar smoke concentrate was
110 ditf'erent from that of the cigare.tte smoke condensate. Similarly,
there was no differc~nw betwwn cigar and rigarchtte smoke condensates
when carcinoma incidences were compared.
Hornburger, et. al. (4;;) pre.pare(l tars from cigar. pipe. and cigarette
tobaccos that were smoked in the form of cigarettes. In this way. all
tobaccos were smoked in an ident ical manner and uniform combustion
tempcmtuws were acliirved. Because of tllis ~t:lnclal.tlizatioii, cliffer-
(`IWS in tumor vieltl could be attributrd to tobacco blend ilntl not. the
~~iattnrr in w-hi& the tar3 were pwlwrecl. The wl&le tars were diluted
one-to-oncb with acetolw and a1~l~lied to the shawtl skin of C-IF, mice
three times a week for the lifespan of tlw test animal. Skin cancers
Wre produced more quicklv with l'ilw :ml cigar mlokc condensates
than with cigarette smoke cintlcnsates. This suggests that. the smoking

   495-028 0--7:$--l;,                                      211


of pipe and cigar tobaccos in the form of cigarettes does not alter the
condensates to any significant degree.
Davies and Day (.S?) prepared tars from small cigars especially
manufactured from a composite blend of cigar tobacco representing
small cigar brands smoked in the United Kingdom, cigarettes espe-
cially manufactured from the same tobacco used for the cigars de-
scribed above, and plain cigarettes especially manufactured from a
composite blend of flue-cured tobacco representing the major plain
cigaret.te brands smoked in the United Kingdom. The whole tar was
diluted to four concentration levels and applied to the shaved backs
of female albino mice for their lifespan using four dosing regimens.
A statistically significant. increase in mouse skin carcinogenicity was
shown with the cigar smoke condensate compared with the tars
obtained from either flue-cured or cigar t,obacco cigarettes. These
results are consistent with those of the previously reported
investigations.
The effect of curing on carcinogenicit,y was examined by Roe, et al.
(76). Bright tobacco grown in Mexico was either flue-cured or air-
cured and bulk fermented. Both flue-cured and air-cured tobaccos were
made into cigarettes standardized for draw resistance and were smoked
under similar conditions. Condensates from these cigarettes were ap-
plied to mouse skin three times each week in an acetone solution. The
development of skin tumors was higher in mice treated with the flue-
cured condensate than in mice treated wit,h the air-cured condensate
(P<O.Ol). The ditference may have been due to the use of equal
weights of condensate rather than the use of extracts from an equal
number of cigaret.tes. The air-cured cigarettes produced a greater
weight, of condensate than did the flue-cured cigarettes. A chemical
analysis of the two tobaccos and two condensates revealed only small
differences in composition. Evidently air curing of Bright tobacco
in the method used is not associated with a loss of reducing sugars.
A more detailed analysis of these experimental studies is present,ed
in table 31.
These experimental data suggest that cigar and pipe tobacco con-
densates have a carcinogenic potential that is comparable to cigarette
condensates. This is sul)l)orted by human epidemiological data for
those sites exposed equally to the smoke of cigars. pipes, and cigaret,tes.
The partially alkaline smoke derived from pipes and cigars is gen-
erally not inhaled, and as a result there appears to be a lower level
of exposure of the lungs :ni(l other systems to the harmful properties of
pipe and cigar smoke than occurs with cigarette smoking. It is antic-
ipated that modifications in pipe tobacco or cigars which would result
in a product that was more readily inhalable would eventually result
in elevated mortality from cancer of the lung, bronchitis and emphy-
sema, arteriosclerotic cardiovascular diseases, am1 the other condi-
tions which have been clearly associated with cigarette smoking.

212


TABLE~I .-Tumvrigenic activity of cigar, pipe, and cigarette smoke codemates in skin painting experiments on animals

[Key: A=Method. B=Fmquency. C=Dumtion. D=Msterial.l

Author, reference

Aetivlty

Treatment

            Percent
Number
      Paplllomas   Carcinomas

Wynder and
Wright
(117).

Croninger, et
al. (HI).

CAFr and
Swiss mice.

A. Painting shaved skin.
B. 3 times a week.
C. Lifespan (24 months).
D. Neutral fraction tar from
  cigarettes and cigarette
  tobacco smoked in pipes,

Female Swiss
mice.

A. Painting shaved skin.
B. 3 times a week.
C. Lifespan.
D. Whole tar diluted in
    acetone.

CAF, :

Pipe (cigarette tobacco) _ _ _ _ _ _
Cigarette- ____ -_-__--__--___
Swiss:

30     60       20
30     30        3

Pipe (cigarette tobacco) - _ _ _ - _    30     63
Cigarette-_--___-----_.- ____    30     63

50
33

Cigar, nicotine free (1 :I) - _ _ _ - _ _    46     65       41
Pipe, nicotine free (l:l)_--.----    45     71        44
Cigar (1:2) _____ - ___________ -_    78     33       18
Pipe, nicotine free (1:2)------__    89     30       16
Cigarette (l:l))-.- _____ - ___._    86     47       37
Acetonecontrols-_--- _______ --    23      0        0

Kensler (65) _ __ Swiss mice ______ A. Painting shaved skin.
                     B. 3 times a week.
                 C. Lifespan.
                   D. Whole tar diluted in
                             acetone.

Cigar tar (J) 100 mg. per week-..
Cigarette tar (G) 100 mg. per
week.
Cigarette tar (E) 100 mg. per
week.

100     42        41
100     40       28

100     34

34


TABLE 31 .-Tumorigenic actiwity of cigar, pipe, and cigarette smoke condensates in skin painting
            experimenta on animals--Continued
                [Key: A=Method. B=Freguency. C=Durstion. D=Material.]

-

Author. reference      Adllal              Activity                    Treatment

            Percent
Number  ------------_
      Papillomas  Carcinomas

Hornburger, et  CAFl mice__-- - _ A. Painting shaved skin.
al. (46).                B. 2 to 3 times a week.
                C. Lifespan (2 years).
                 D. Whole tar diluted 50 per-
                         cent in acetone

Davies and
Day (ES).

Female albino   A. Painting shaved skin.
mice.       B. Varied.
          C. 116 weeks.
         D. Whole tar in 150 mg.
                acetone.

Roe, et al.     Female Swiss
(76).         mice.

A. Painting shaved skin.
B. 3 times a week.
C. Lifespan.

D. Whole tar diluted in
    acetone.

Cigar tobacco cigarettes 165 mg.
per week.

Pipe tobacco cigarettes 1 64 mg.
per week.

Cigarettes * 62 mg. per week- _ _ _
Acetone controls--_-----_- ____

Cigars, small 83 mm. long 150
per week.

Cigar tobacco cigarettes 150
per week.
Cigarettes 150 per week--_-----

Flue-cured Bright tobacco 180
mg. per week.
Air-cured Bright tobacco 180
mg. per week.
Acetone controls 0.75 cc. per
week.

100     37. 5

100     23


100     I5
100      0

144     44


72     32

144     28       13

400     52


400     68

400      1. 3

19

20

23
0

27

14

30

23

0. 5

1 Cigar. pipe, and cigarette tobacco smoked as cigarettes at similar combustion temperatures.


cARLnovAscoLAR DIsEAsF8

The majority of deat.hs in the United States each year are due to
cardiovascular diseases. Cigarette smoking has been identified as a
major risk factor for the development of coronary heart disease
(CHD) . However, pipe and cigar smokers experience only a small
increase in mortality from coronary heart disease above the rates of
nonsmokers. Cigarette smokers have higher death rates from cerebro-
vascular disease than nonsmokers, whereas pipe and cigar smokers have
cerebrovascular death rates that are only slightly above the rates of
nonsmokers. Table 32 summarizes the major prospective epidemiologi-
ea.1 investigations that examined the association of smoking in various
forms and total cardiovascular diseases, coronary heart disease. and
cerebrovascular disease. Doll and Hill. (28)) Best (9)) and Kahn (50)
examined dose-response relationships for pipe and cigar smokers and
reported a slight increase in mortality from coronary heart disease
with an increase in the number of cigars or pipefuls smoked.

Other prospective epidemiological studies have also examined the
relationship of smoking in various forms to coronary heart disease and
related risk factors. Jenkins, et al. (4.9) in the Western Collaborative
Group Study of coronary heart disease, reported an incidence of coro-
nary heart disease in men aged 50 to 59 who were pipe and cigar smok-
ers that was intermediate between the rates seen in cigarette smokers
and nonsmokers. No increase in incidence of coronary heart disease was
seen among the pipe and cigar smokers in the younger age groups.
Shapiro, et al. (85)) in a study of the health insurance plan (HIP)
population, reported incidence rates for myocardial infarct,ion, angina
pectoris, and possible MI, in pipe and cigar smokers that were similar
to the incidence rates seen in cigarette smokers. These rates were con-
siderably higher than those of nonsmokers. Data from the pooling
project (47) suggested that the incidence of CHD deaths, sudden
death, and the first major coronary event in pipe and cigar smokers
was intermediate between the incidence experienced by cigapette smok-
ers and nonsmokers. In contrast to these. studies, Doyle, et al. (3'0)
reported no increase in CHD deaths, myocardial infarction, or angina
pectoris in pipe and cigar smokers over the rates of nonsmokers in t.he
Framingham study.
The retrospective studies of Mills and Porter (64), Villiger and
Heyden-Stucky (204), Schimmler, et al. (80), and Hood, et al. (46)
contained data suggesting that pipe and cigar smokers experience
mortality rates from coronary heart disease that are essent,ially similar
to t.hose experienced by cigarette smokers. The retrospective study of
Spain and Nathan (86) reported lower rates of coronary heart dis-
ease in all smoking categories than were found in nonsmokers.

Van Buchem (103) and Dawber, et al. (23) examined serum choles-
terol levels in groups of individuals classified according to smoking

215


habits. In these two studies, pipe and cigar smokers had serum choles-
terol levels t.hat were nearly identical with the levels found in
nonsmokers.

Tibblin (91) and Dawber, et al. (23) investigated t.he effect of smok-
ing on blood pressure. The proport,ion of smoker-s decreased in groups
mlth higher blood pressures, although this was not as dramatic for
pipe and cigar smokers as it was for cigarette smokers.

In an experimental study using anesthetized dogs, Kershbaum and
Bellet (54, 55) examined the effects of inhaled and noninhaled ciga-
rette, cigar, and pipe smoke on serum free fatty acid levels and urinary
catecholamine and nicotine excret,ion. In this study, inhalation of to-
bacco smoke from all t,hese sources resulted in similar increases in
serum free fatty acids and in catecholamine and nicotine excretion.

TABLE 32.-Mortality ratios for cardiovascular deaths in male cigar and
pipe smokers. A summary of prospective epidemiological studies

Author, reference       Category

        Type of smoking  ---_
Non-  cigar
smoker  OdY :g pg;d ,:%,ly Mixed

Hammond and   Cardiovascular      1.00  1.26  1.07 ______  1.57  ______
Horn (40).      total.
         Coronary-----------  1.00  1.28  1.03 ______  1.70 _--_._
           Cerebrovascular _____  1. 00  1. 31  1. 23 ______  1. 30 ______
Doll and Hill    Cardiovascular      1.00  -_-__  -----w0.99  1.26   1. 13
(86, 87).       total.
         Coronary ._____ -_-_-  1.00  ---__  ---_-_ .94  1.23   1. 18
           Cerebrovascular-- _-_  1. 00  _..__  . . ____ . 95  1. 13   .97
Best (9)--_---_-  Cardiovascular      1.00  1. 14   95 -_-_-_  1.52  -_-_._
             total.
         Coronary-.._--------  1.00   .99  1.00 ______  1.60  ._._ -_
            Cerebrovascular-----  1. 00  1.28   .85 ._____  .88  ._._ --
Hammond L     Cardiovascular      1.00  ---__  -_-_-_ 1.06  1.90  -----_
(38).          total.
         Coronary_----------  1.00  1.35  1.19 __.___  1.84   1. 58
           Cerebrovascular _.__ -  1.00  -----  ---_-_ 1.09  1.41   1. 40
Kahn (X?)  -  -  - Cardiovascular      1.00  1.05  1. 06 1. 05  1. 75  ----__
             total.
         Coronary.----------  1.00  1.04  1.08 1.05  1.74  ____._
           Cerebrovascular___--  1. 00  1. 08  1. 09 1. 06  1. 52  __._..

1Yortality ratios for ages.56 to 64only are presented.

Chronic bronchitis and pulmonary emphysema account for most of
the morbidity and mortality from chronic respiratory disease in the
I'nited Stattbs. Cigawtt(~ smokers ha\-e higher death rates from these

216


diseases and have more pulmonary symptoms and impaired pul-
monary function than nonsmokers. Cigarette smokers also have more
frequent and more severe respiratory infections than nonsmokers.
The relat,ionship between smoking pipes and cigars and t.hese diseases
is summarized in this chapter. The major prospective epidemiological
studies are summarized in table 33.

In a retrospective study of 1.189 males and matched comrols in
Northern Ireland, Wicken (106) investigated smoking in various
forms and mortality from bronchitis. The relative risk ratios com-
pared to nonsmokers for mortality from chronic bronchitis were 1.98
for all smokers, 1.S for pipe and cigar smokersY 2.25 for cigarette
smokers, and 1.29 for mixed smokers.

From a review of these prospective and retrospective studies, it
appears that pipe and cigar smokers experience mortality rates from
bronchitis and emphysema that are higher than the rates of non-
smokers. Although these morality rates approach those of cigarette
smokers, in most instances they are intermediate bet.ween the rates
of cigarette smokers and nonsmokers.

Pipe and cigar smokers have signific.antly more respiratory symp-
toms and illnesses t.han nonsmokers. Those studies which contain data
011 pipe and cigar smoking as related to respiratory symptoms are
summarized in table 33.

Only a few studies have examined pulmonary function in pipe and
cigar smokers. There appears to be little ditf'erence. in pulmonary func-
tion values for pipe and cigar smokers as compared to nonsmokers
(table 35).
Naeye (67) conducted an autopsy study on 322 Appalachian coal
workers who were classified according to the type of coal mined and
tobacco usage. Emphysema was slight.ly greater in cigarette smokers,
as were anatomic evidences of chronic bronchitis and bronchiolitis.
Those changes found in pipe and cigar smokers were intermediate
between those of cigarette smoking miners and nonsmoking miners.

Changes in pulmonary histology in relation to smoking habits and
age were examined by Auerbach, et al. (8). Fibrosis, alveolar rupture,
thickening of the walls of small arteries, and thickening of the walls
of the pulmonary arterioles were found to be highly related to the
smoking habits of the 1,340 male subjects examined. The 91 pipe and
cigar smokers over the. age of 60 were found to have somewhat more
alveolar rupture than the men of the same age distribution who never
smoked regularly. However, pipe and cigar smokers as a group had
far less rupture than cigarette smokers. The same relations as described
above were found for fibrosis, thickening of the walls of the arterioles
and small arteries, and padlike attachments to the alveolar septums.

Tobacco smoke has been sl~ow~ experiment.ally to 11nvc a ciliostatic
lffect. on the reslnratory elntl~eliunl. The interval between puffs, the

217


amount of volat,ile and particulate compounds in the smoke, and the
exposure volume have been shown to influence the toxic effect of
tobacco smoke. Dalhamn and Rylander (22) exposed the upper trachea
of anesthetized cats to the smoke of cigarettes and cigars, observing
the effect on ciliary activity through an incident-light microscope.
A chemical analysis of the gas and particulate phases revealed that
the cigar smoke. was more alkaline and, in general, contained higher
concentrations of isoprene. ace.tone, acetonitrile, toluene, and total
particulate matter compared to c.igtlrette smoke. The average number
of puffs required to arrest ciliary activity was found to be 73 for the
cigaret.te smoke and 114 for the cigar smoke. The difference is statisti-
cally significant (P <O.Ol). Of the two smokes, the smoke with the
highest concentration of volatile compounds was found to be the least
ciliostatic. This suggests that the degree of ciliotoxicity of a smoke is
not necessarily correlated to the level of one or several of the substances
found in the smoke.

Passey, et' al. (70. 71, 79) studied t,he effect of smoke from flue-cured
cigarette tobacco cigarettes and air-cured cigar tobacco cigaret.tes on
the respiratory system of rats. In two separate but similar experi-
ments, a total of 48 animals were exposed to English cigarette tobacco
smoke, 48 were exposed to air-cured cigar tobacco smoke, and 12 were
exposed to an air-cured Hurley tobacco smoke. The rats in groups were
exposed to the specific smoke in a. smoke-filled cabinet. Animals ex-
posed to t.he smoke from air-cured tobaccos remained healthy through-
out, the experiments. even at high levels of smoke exposure. The three
deat.hs that occurred within this group were from nonrespiratory
causes. In both experiments, the rats exposed to cigarette tobacco smoke
began to die within 1 or 2 months. and in each experiment. most of the
animals died within a week or two of the first deaths. At autopsy the
rats exposed to flue-emed tobacco smoke on gross examination were
fount1 to have greatly enlarged 11mgs. the trachea was often full of
mucus. and there was evidence of pneumonia. On microscopic examina-
tion it was follnd that the trachea and bronchi contained purulent
cellular csudatcs. evidence of metaplastic changes, an absence of cilia,
ant1 goblet cell l~l~yerl~lnsi:~. Typically. the cause of death was a lobar
or bl~oll~llop~lcrlxllonia. T11c allthor concluded that, "the smokes of flne-
cured tobaccos are nlore clangorous to man and to animals t.han those
of air-cured tobaccos."

218


Unfortunately, few details were published concerning the method
used to expose the animals to the different t,ypes of smoke. The fre-
quency and duration of exposure were not specified, and the extent of
act,ual inhalation of smoke by t.he different groups of rats was either
not determined or not reported. It is also difficult to determine t.he
effect of smoke exposure on the frequency and severity of respiratory
infections when animals are exposed to smoke in groups where common
exposure occurs. The rat strain used was not identified. but it. was
noted that animals appeared to suffer from a.n endemic rat bron-
chiectasis. It is not known to what extent epidemics of respiratory
infections occurred among these animals. Because of these difficulties,
no firm conclusion can be drawn concerning the effect of smoking flue-
cured or air-cured tobaccos on the incidence of respiratory infections
in rats.

TABLE 33.-Mortality ratios for chronic obstructive pulmon.ary deaths
in male cigar and pipe smokers. A summary of prospective epidemio-
logical studies

Author, reference       catt?goly             Type of smoking

Hammond and   COPDtotal________ 1.00 1.29 1.77 _.___ 2.85 __.._-
Horn (40).   Emphysema--_----- --__ -___- ___-- ---__ --__--____-.
          Bronchitis~~~~~~~--~ --__ ____- __--- --___ -__--____-__

Doll and Hill   COPD total-_- .___. ____ --___ _____ _____ .__.________
(26, 87).    Emphysema--_-_--- -___ -__-_ _____ --___ --__-_____-_
           Bronchitis---- _.__ -_ 1.00 _-__- _____ 4.00 7.00   6. 67

Best (9) ______ -_ COPD total- _ ____ ._ ____ _____ _____ _____ ________ -___
         Emphysema-..-_----  1.00  3.33   .75  --__-  5.85  -___-_
          Bronchitis-------_--  1. 00  3. 57  2. 11  __-__  11. 42  ______

Hammond (38)-- COPD total- _ _ - _ _ _ _  _ _ _ -  _ _ _ _ _  . _ _ _ _  _ _ _ _ _  _ _ _ _ _ _ _ _ _ _ _ _
         Emphysema-_- _____ 1. 00 - ____ ___-_ 1. 37 1 6. 55 ___.__
          Bronchitis_--_-----_ -__- __.__ _-___ ____- ___---____-_

Kahn (&I- - __- _ COPD total- _ _- _. _ _  1.00    79  2. 36    99  10. 08 _----_
        Emphysema ________  1. 00 1: 24  2. 13 1: 31  14. 17  - _____
          Bronchitis---- ______  1. 00  1. 17  1. 28  1. 17  4. 49  -_----

1 Only mortality ratios for ages 65 to 64 are presented.

495-028 O-73-16

219


TABLE M.-Prevalence of respiratory symptoms and illness by type of
              smoking

-

                                         Percant prevalence
Author, reference  Number and type of       Illness
            population                    NOtI-   Total  Cig*
                                  smoker pipe and rette
                                        c&m   OflY   Mixed

Boske (fO)- _ _ Parents of 59
           families.

Edwards, et   1,737 male
8l. (33).     outpatients.

Ashford, et
8l. (4).

4,014 male
workers in 3
Scottish
collieries.

Bower (11)--- 95 male bank
         employees.

Wynder, et al.
(114).

Densen, et al.
(94).

Cederlof, et
al. (18).

Rimington
(76).

220

315 male pa-
tients in
New York
and 315 male
patients in
California.

5,287 male    Persistent cough- _
postal and    Persistent
7,213 male     sputum
transit      production.
workers in    Dyspnea..- _ _ _ _ _ _ _
New York    Wheeze ______ --___
City.        Chest illness~~-~~~

4,379 twin pairs,  Cough--------_--    4
all U.S.     Prolonged cough-.    2
veterans.     Bronchitis--------    2

41,729 male
volunteers.

Cough . ..___ _____ _   32
Sputum          24
production.
Chest illness------ 5

Chronic bronchitis- 17

Bronchitis--- _ __- _   10   1 35    21    37
Pneumoconiosis---- 11   1 34    14     2

Cough ____ __ ______    0

Sputum           8

production.
Wheeze--- _ _ _ _ _ _ _ _
Chest illness------

8
15

Cough (New
York).
Cough
(California).
Influenza (New
York).
lnfluenza
(California).
Chest illness
(New York).
Chest illness
(California).

Chronic bronchitis-    5

14

22

11

28

9

7

7    11
11    16

16    19
14    21
13    16

32
15


4


`19

0
15


31
54


33


30


21


24


10


6

7
4
3


`9

48 ---___
20  ____._


5  --__-_


31     14

29  -_____
33  ______


33  --____
40  ---_-_


56    51


67    66


24  ____-_


31  --____


12  _-____


11  _-_-._

25  ______
26  _____.

26 -_____
32  -_-__.
18 -__-.-

17 __.---
11  __.---
10  ___---


17  _____-


TABLE 34.-Prevalence of respiratory symptoms and illness by type of
           smoking--Continued

Author, reference  Number and type of       Illness           Percent prevalence

            population                    Non-   Total  C&a-
                                  smoker pipe and rette   Mired
                                         cigar   OdY

Comstock, et 670 male tele-   Persistent cough _ _
81. (19).     phone       Persistent
         employees.     sputum.
                Dyspnea- - _ _ _ - _ _ _
                    Chest illness in
                  past 3 years.

Lefcoe snd   310 male phy-  Chronic respira-
Wonnscott    sicians in     tory disease.
(69).       London,     Chronic bronchitis..
           Ontario.     Obstructive lung
                      disease.
                     Asthma----__----
                     Rhonchi-.- _ _-_ _ - _

10    16    41  __-___
13    20    42  _--___

33    39    44  ____-_
14    18    20  ___--_

9    18    44 -___-_

1    12    34  __-_.-
1    3     4  --__--

7    3    6  -__--_
0    3    9 _-__-_

1 Ngures for pipe only.

TABLE 35.-Pulmonary function dues for cigar and pipe smokers as
         compared to nonsmokers

Author, reierence  Number and type     Function          Type of smoking
          Of population                Non-
                                  smoker To;all gig Cigarette Mixed
                         P    OdY

Ashford, et
81. (4).

4,014 male    FEV,.,,        3. 39   '
                _____ -_         2.59   3. 14   2. 62
workers in
3 Scottish
collieries.

Goldsmith,   3,311 ective   Puffmeter---_ 313. 63  299.26  303. 44  ______
et 81. (57).    or retired   FEV,.o--d---m   2. 99    2. 80   2.91  -___-_
        longshore-   TVC- _ . _ _ _ _ _ _   3. 87    3. 68    3.88  ._____
             men.

Cornstock,    670 msle    FEVl.o _____ -_   3. 12    3. 26   2.82  __--__
et 81. (19).   telephone
        employees.

Lefcoe and    310 msle    FEV,.,, _____ --   3. 39   3. 17   3.11  ______
Wonnacott   physicians   MMFR liters    4. 09    4. 17   3.64  ______
(69).      in London,   per second.
           Ontario.

1 Figures for pipe only.

221


GAGTROINTESTISAL DISORDERS

Cigarette smokers have. an increased prevalence of peptic ulcer
disease and a greater peptic ulcer mortality ratio than is found in
nonsmokers. These relationships are st.ronger for gastric ulcer t.han
for duodenal ulcer. Cigarette smoking appears to reduce the effective-
ness of standard peptic ulcer treatment regimens and slows the rate
of ulcer healing. Cigar and pipe smokers experience higher death
rates from peptic ulcer disease than nonsmokers. These rates are higher
for gastric ulcers than for duodenal uIcers but are somewhat less than
those rates experienced by cigarette smokers. Table 31 presents the
mortality ratios for ulcer disease in cigar and pipe smokers as reported
in the prospective epidemiological studies.

Retrospective or cross-sectional studies by Trowel1 (95), All&one
and Flint (2)) Doll, et al. (.29), and Edwards, et al. (33) contain
data on ulcer disease in pipe smokers as well as cigarette smokers. No
association was found between pipe smoking and ulcer disease in these
investigations.

TABLE 36.-Mortality ratios for peptic ulcer disease in male cigar and
     pipe smokers. Summary of prospective stuo?ies

Author, reference        IlllIeSS

         Type of smoking
-               Total
Non-  cigar   Pipe
smoker  OdY   O&Y  %i W& -
                         Mixed
                   O&Y
               cigar

Hammond and   Duodenalulcer ___. -_  1.00  0.25  1.67  ______  2. 16  __-___
Horn (40).
Doll and Hill   .Gastric ulcer__--_---  1.00  _____   ___-_ 4.00  7.00   5. 30
(66, 27).
Hammond (58)-- Gastric ulcer ______._  1. 00  ____-  -_-__ 2.04  2.95  __-_-.
           Duodenal ulcer ___.__  1.00  - _.__   _____  .92  2.86  ___._ -
Kahn (50) _._._ -  Gastric ulcer ______ -_  1. 00  2. 90  2. 84  2.48  4. 13  ____. -
           Duodenalulcer--_---  1.00  1.58  1.59  1.39  2.98 ____. -

Little Cigars

In the past year, several new brands of little cigars (weighing 3
pounds or less per 1,000) hare appeared on the national market. These
cigarette-sized prodUcts are manufactured, packaged, advertised, and
sold in a manner similar to cigarettes. Little cigars enjoy several legal
advantages over cigaret'tes : They have access to television advertising;
they are taxed by the Federal Government and by most States, at much
lower rates than cigarettes, result,ing in a significant price advantage;

222


and they do not carry the warning label required on cigarette pack-
ages and in cigarette advertising. A market appears to be developing
for these products, as there has recently been a sharp increase in the
shipment of little cigars destined for domestic consumption (table 37).

It is important to estimate the potential public health impact of
these little cigars. An adequate epidemiological evaluation of the ef-
fect of little cigar smoking on health could take 10 or 15 years and is
probably an impractical consideration ; however, a review of the epide-
miological, autopsy, and experimental data concerning the health con-
sequences of cigarette, pipe, and cigar smoking summarized in this and
previous reports is helpful in considering the potential impact on
health of smoking little cigars. An analysis of the chemical c.onstit,-
uents suggests that both cigarettes and cigars contain similar com-
pounds in similar concentrations. Two exceptions are reducing sugars,
which are not found in quantity in the fermented tobaccos commonly
used in cigars, and the pH of the inhaled smoke. The pH of the smoke
from U.S. commercial cigarettes is below 6.2 from the first to the last
puff, whereas the smoke from the last half of a cigar may reach as high
as pH 8 to 9. With increasing pH, nicotine is increasingly present in
the smoke as the free base. Skin painting experiments in mice indicate
that tumor yields with cigar or pipe "tars" are nearly identical wit,h
those obtained with cigarettes "tars". In addition, the epidemiological
data suggest. that depth of inhalation probably accounts for the fact
t.hat cigarettes are so much more harmful than cigars and pipes in con-
tributing to the development of lung cancer, coronary heart disease,
and nonneoplastic respiratory disease. For such diseases as cancer of
the oral cavity, larynx, and esophagus, where smoke from cigars, pipes,
and cigarettes is available to the target organ at comparable levels, the
mortality ratios are very similar for all three forms of tobacco use.
Several factors, including "tar," nicotine, and the pH of the smoke,
probably operate to influence inhalation patterns of smokers. The
relative contribution of individual factors to the inhalability of a
tobacco product has not been determined.

Smoking those brands of little cigars which can be inhaled by a
significant portion of the populat,ion in a manner similar to the pres-
ent use of cigarett.es would probably result in an increased risk of de-
veloping those pulmonary and cardiovascular diseases which have
been associated with cigarette smoking. On the .other hand, smoking
those little cigars which are used like most large ciga.rs whereby the
slnoke is rarely inhaled would probably result in lower rates of those
pulmonary and cardiovascular diseases than would be found among
cigarette smokers.

Only a limited analysis is available comparing the chemical com-
pounds found in little cigars, cigarettes, and large cigars. The FTC
analyzed the tar and nicotine content of all t.he litt.le cigars (34) and
cigar&es (97) currently arailable on the market. Little cigars have

223


generally a higher "tar" and nicotine level than cigarettes, although
considerable overlap results in some little cigar brands having "tar"
and nicotine levels comparable to those of some brands of cigarettes
(figs. 4 and 5). Hoffmann and Wynder (66) recently compared three
brands of little cigars with an unfiltered cigarette, a filtered cigarette,
and a large cigar. They measured a number of smoke constituents, in-
cluding : "tar," nicot,ine, carbon monoxide? carbon dioxide, reducing
sugars, hydrogen cyanide? acetaldehyde, ncrolein, pyridines, phenols,
benz (a) anthracene, and benzo (a) pyrene (table 32). Cigarette A was
the Kentucky reference cigarette, c.igarette n was a popular brand of
filter cigarette. Cigar A was an 85 mm. little cigar, cigar B was an
85 mm. little cigar, cigar C was a 95 mm. small cigar, and ciga.r D was
a 112 mm. popular brand of medium sized cigar.

The smoke pH was analyzed puff by puff (table 39). Cigarette
smoke was found to be acidic (pH less than 7) for %he entire cigarette.
The smoke from little cigars became alkaline only in the last puff or
two, whereas about the last 40 percent of the puffs from the larger
cigar were alkaline. Although t,he pH of the total condensate obtained
from cigarettes is usually acidic and the total condensate obtained
from cigars is usually alkaline, the above data indicate that smoke
pH of tobacco produc.ts changes during the combustion process. Smoke
from large cigars may be acidic during the first portion of the smoke
and not become alkaline unt,il the last half of the cigar is smoked.

Brunnemann and Hoffmann (15) ) using the same techniques de-
scribed above. examined the effect of 60 leaf constituents on smoke pH.
For several varieties of cigarette tobacco. they found a high correlation
between the total aklaloid and nitrogen content and smoke pH. Stalk
position also affected smoke pH. Tobacco leaves near the top of the
plant, which cont.ain high levels of tar and nicotine, yielded a smoke
with a much higher pH than leaves lower on the plant. At present it is
not known to what extent these factors influence the pH of the smoke
of tobaccos commonly used in c.igars or how t.hrsc kinds of pH changes
influence the inhalability of tobacco smoke.

The inha.lation of smoke, however, appears to be the most important
factor determining the impact a cigar will have on overall health.
Those physical and chrmical characteristics of a tobacco product
which most influence inhalation of tobacco smoke have not been
accurately drterminftd. Sc~vertheless, it appears likely that the smoke
of some brands of cigars may be compatible with inhalation by a sig-
nificant. portion of the smoking population, since: (a) Little cigars
have tar and nicotine levels which. in some brands, are similar to the
levels found in cigarettes, and (6) the pH of the smoke of some little
cigar brands is acidic for the major portion of the little cigar and
becomes alkaline only in the last puff or t,wo.

224


It is reasonable to conclude that smoking little cigars may result in
health effects similar to those associated with smoking cigarettes if
little cigars are smoked in amounts and with patterus of inhalation
similar to those used by cigare.tte smokers, for the reasons cited above,
and these additional reasons: (u.) In those little cigars for whic,h pre-
liminary data are available, the concentrations of carbon monoxide,
hydrogen cyanide, acetnldehyde, acrolein, pgridine, phenol, and poly-
cyclic hydrocarbon levels are comparable to those found in cigaret,tes :
(6) cigarette smokers who switch to cigars appear to be more likely
to inhale c.igar smoke than cigar smokers who have always smoked
cigars (24) ; and (c) cigarette smokers who switch to little cigars may
be inclined to use them as they did cigarettes because of the physical
similarit,ies bet.ween the little cigars and cigarettes, including their
size and shape, t.he number in a package, the burning rate, and t.he
time it. takes to smoke them.

Figure 4.-Percent distribution of 130 brands of cigarettes and 25 brands of little cigars by
                        "tar" content.

Little Cinars  I

w. "tar.'     0     0      0   16.0    8.0    32.0   32.0    0     8.0    4.0
Cigarettes   o-4   5-9   lo-14  15-19  20-24  25-29  30-34 35-39   40-44  45-49
Little Cigars  3.1    3.1    10.0   46.2   23.1   10.0    3.9   0.8      0     0
SOURCE: U.S. Department of Health. Education, and Welfare (971 and Federal Trade Commission (34).

225