(78) TOPPING, D.L., DWYER, T., WELLER, R.A. Peripheral vascular disease in
   cigarette smokers and impaired hepatic metabolism of lipoprotein rem-
   nants. Land 2@052/3053): 1327-1328, December 24 and 31,1977.
(79) TRAPJENSEN, J., CARLSEN, J.E., SVENDSEN, T.L., CHRISTENSEN, N.J.
   Cardiovascular and adrenergic effects of cigarette smoking during immedi-
   ate non-selective and selective beta adrenoceptor blockade in humans.
   European Journnl of Clinical Investigation 9(3): 181-183,1979.
(80) TRAYNER, I.M., MANNARINO. E., CLYNE, C.A.C., THOMPSON, G.R.
   Serum lipids and high density lipoprotein cholesterol in peripheral vascular
   disease. British Journal of Surgery 67(7): 497-499, July 1980.
(81) US. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE. Smokillg
   and Health: A Report of the Surgeon GencmI. U.S. Department of Health,
   Education, and Welfare, Public Health Service, Office of the Assistant
   Secretary for Health, Office on Smoking and Health, DHEW Publication No.
   m3s)79-50066,1979,1136 pp.
(82) U.S. PUBLIC HEALTH SERVICE. The Health Consequences of Smoking. A
   Report of the Surgeon Geneml: 1971. U.S. Department of Health, Education,
   and Welfare, Public Health Service, Health Services and Mental Health
   Administration. DHEW Publication No. (HSM)71-7513,1971,453 pp.
(8.3) U.S. PUBLIC HEALTH SERVICE. The Health Consequences of Smoking. A
   Report of the Surgeon Geneml: 1974. U.S. Department of Health, Education,
   and Welfare, Public Health Service, Health Services and Mental Health
   Administration. DHEW Publication No. (CDC)74-3704,1974,124 pp.
(84) WEINROTH, L.A., HERZSTEIN, J. Relation of tobacco smoking to art&o
   sclerosis obliterans in diabetes mellitus. Journal of the American Medical
   Association 131(3): 205-209, May 1946.
(85) WEIR, J.M., DUNN, J.E., Jr. Smoking and mortality: A prospective study.
   Cancer 25(l): 105-112, January 1970.
(86) WEISS, N.S. Cigarette smoking and arteriosclerosis obliterans: An epidemic
   logic approach. American Journal of Epidemiology 9X1): 17-251972.
(87) WIDMER, L.K., HARTMANN, G., DUCHOSAL, F., PLECHL, S.-Ch. Risk
   factors in arterial occlusion of the limbs. German Medical Monthly 14(10):
   476479, October 1969.
(88) WILBERT, L. Nikotinkonsum und arterielle verschlusskrankheit. [Consump
   tion of nicotine and occlusive disease]. Medizinische Klinik 66@6): 1190-
   1192, September 3,197l.
(8s) WILENS, S.L., PWR, C.M. Cigarette smoking and arteriosclerosis. Science
   X%(3544): 975977, November 30,1932.
(90) WRAY, R., DePALMA, R.G., HUBAY, C.H. Late occlusion of aortofemoral
   bypass grafts: Influence of cigarette smoking. Surge0 70(6): 969-973,
   December 1971.
(91) YAMORI, Y., KIHARA, M., HORIE, R., NOTE, S., TSUNEMATSU, T.,
   FUKASE, M. Epidemiological analysis on the mechanism of cigarette
   smoking as a risk factor in cardiovascular diseases. Japanese Circulation
   Journd 44t12): 966-970, December 1980.
(92) ZELIS, R., MASON, D.T., BRAUNWALD, E., LEVY, R.I. Effects of hyperlipo
   proteinemias and their treatment on the peripheral circulation. Journal of
   Clinical Investigation 49(5): 1007-1015,197O.


SECTION 6. PHARMACOLOGICAL
        AND TOXICOLOGICAL
         IMPLICATIONS OF
        SMOKE CONSTITUENTS
        ON CARDIOVASCULAR
         DISEASE

203


Introduction
Cardiovascular diseases are the leading causes of death in most of
the technologically advanced countries of the Western Hemisphere,
accounting for approximately half of all deaths annually in the
United States (see Appendix A). The most common among these
diseases are atherosclerosis and coronary heart disease; their
ischemic complications result in increased morbidity and mortality.
Coronary heart disease is the leading cause of death in the United
States, accounting for two-thirds of all cardiovascular deaths (9s).
It is generally acknowledged that coronary heart disease is a
multifactorial process; that is, a variety of factors are involved in the
development and clinical manifestations of this disease. Therefore, it
is not a simple task to determine the etiology and time course of
atherogenic development. In addition, the study of atherosclerosis is
singularly difficult because no model in the experimental animal
exactly replicates the human disease in physiological, morphological,
and clinical detail. Investigations in human subjects are further
limited by the inability to diagnose the disease in preischemic phases
(44). Most studies of the pathology of cardiovascular diseases (CVD)
have been based on autopsies by coroners or on hospital populations
in which only a limited fraction of decedents have been examined.
Individuals may show considerable variance in the degree of
atherosclerosis identified at autopsy, limiting the value of retrospec-
tive analysis (137).
In 1971, the U.S. Government established the Task Force on
Arteriosclerosis to assess research needs and to make recommenda-
tions on priorities for future program plans in this area. Most of the
recommendations of this task force have been implemented during
the past decade, and important advances have been made in basic
and clinical research (96'). Most important, major epidemiological
associations of cardiovascular disease risk not only have been
established, but also have been supported by examinations of the
arterial wall itself, enabling an increased understanding of the basic
mechanisms of disease processes.
Research in cell and molecular biology has provided new informa-
tion about the interaction of blood-borne components, such as
cholesterol, with the arterial wall. Basic research regarding this risk
will help to increase our understanding of the effects of other
circulating components, such as inhaled cigarette smoke constitu-
ents, and will elucidate the susceptibility of arterial cells to these
effects.
The most firmly established modifiable risk factors for atheroscle-
rotic CVD are hypercholesterolemia, hypertension, and cigarette
smoking. In addition to these, diabetes mellitus, lack of exercise,
obesity, and type A behavior have all been suggested as contributors
to the multifactorial process known as atherogenesis (82). The


assessment of any risk factor, such as cigarette smoking, must be
made within the constellation of other risks, i.e., susceptibility to
disease that is predicted by multifactorial analysis (53, 82).

In the case of cigarette smoking, we are faced with an extremely
difficult effort in determining direct cause and effect phenomena
that are attributable to single factors. Over 4,000 different com-
pounds have been identified in tobacco smoke (45), and the determi-
nation of the direct or indirect actions of each upon the arterial wall
seems an impossible task.

We will attempt, however, to examine the major components
believed to be associated with increased risk for CVD and to
remember the multiple risk factors that might be associated with the
development of cardiovascular dysfunctions in cigarette smokers.
The variety of possible pharmacological and toxicological implica-
tions of smoke and its constituents-and the absence of firm proof of
what mechanisms are precisely involved in the unequivocal cause
and effect relationship between smoking and cardiovascular dis-
ease--should not detract from our confidence in the epidemiological-
ly and clinically irrefutable evidence of the cause and effect role of
cigarette smoking in contributing importantly toward heart disease.

Tobacco Smoke: Physical Nature and Chemical Composition

Inside the burning cone of a cigarette, a variety of physical and
chemical processes occur in an oxygendeficient, hydrogen-rich
environment at temperatures up to 900oC. Two major regions for the
smoke formation are primarily observed-the heat-producing com-
bustion zone and the pyrolysisdistillation zone (II). The mainstream
smoke (MS) is formed during puff drawing; the sidestream smoke
(SS) is generated largely by the smoldering of the cigarette between
puffs. Throughout this review, data are discussed relating to
cigarette smoke generated by smoking machines, unless otherwise
noted. The standard machine smoking parameters for cigarettes
were primarily developed for comparing smoke yields obtained
under identical conditions. Today, these smoking parameters do not
reflect the smoking behavior of many of the cigarette smokers and
especially not that of smokers of low-yield cigarettes who tend to
draw puffs of greater volume more frequently (64, 68,145).

The mainstream smoke of tobacco products represents a very
dense aerosol. In the case of a cigarette without a filter tip, it
contains about 5 x 10' spherical particles per milliliter. The size of
the particles varies between 0.1 and 1.0 pm, with an average
diameter of 0.4 pm (12). Three to eight percent of the weight of the
total mainstream smoke of a cigarette without a filter tip is
attributable to the particulate matter. The remainder consists of
vapor phase components with nitrogen (So to 70 percent), oxygen (10


TABLE l.-Approximate number of smoke compounds
     identified in some major compound classes

Compound class           Number identified

          Amiden, imidea, lactonea
         Carboxylic acids
            LWtOneS
            Ektem
          Aldehydea
            KetOIlCS
            Alcohols
            Phenols
            Aminm
         N-Hetemcyclics
         Hydrocarbons
            Nit&n
         Anhydridee
         Carbohydridea
           Ethers

               Total


SOURCE: Dube and Groeo W'J.

237
227
150
414
108
521
379
292
196
921
755
106
11
42
311

4,720

to 15 percent), carbon dioxide (10 to 15 percent), and carbon
monoxide (3 to 6 percent) as major constituents (27,106). Of the more
than 4,060 components identified in cigarette smoke, 400 to 500 are
present in the vapor phase (27, 45). Table 1 lists some major classes
of smoke components as recently recorded by Dube and Green (45).
The total number of 4,720 in the table exceeds by far the total
number of identified compounds because of repeated listing of the
compounds that contain multifunctional groups. The acute toxicity
of tobacco smoke is influenced not only by the chemical composition,
aerosol concentrations, and particle sizes of the smoke, but also by
the smoke pH. With a pH greater than 6.2, the smoke contains
increasing amounts of unprotonated nicotine, which is the most toxic
form of this habituating agent (Figure 1) (26). The unprotonated
nicotine is at least partially present in the vapor phase and thus is
likely to be more rapidly absorbed by the smoker (5).

The U.S. cigarette is filled with a blend of tobaccos consisting of
Bright, Burley, and Turkish types. Its mainstream smoke pH lies
between 5.5 and 6.1. The smoke of cigarettes and cigars made up
entirely of Burley or dark tobacco varieties has pH values of about
6.5 for the first puffs and up to 8.0 for the last puffs (2s).

Sidestream smoke, which is formed between puff drawings, is
freely emitted into the air from the smoldering tobacco products. The
peak temperature in the burning cone of a cigarette during puff
drawing is about 900oC and between puffs it is about 600oC (162).
This is an important factor for the divergence of specific toxic agents
generated in mainstream and sidestream smoke. Another major

207


FIGURE l.-Protonation of nicotine
SNJRCJC Srunnemann and Hofimann LX9.

difference is that the side&ream smoke leaving the site of formation
is subjected to greater air dilution and faster temperature decline
than is the mainstream smoke, which travels through the tobacco
column and is then inhaled as a concentrated aerosol. These
conditions for sidestream smoke generation favor formation of
aerosol particles of smaller size (0.01 to 0.1 pm) than those occurring
in the mainstream smoke (0.1 to 1.0 urn) (12). The pH of sidestream
smoke of a U.S. blended cigarette varies between 6.7 and 7.5,
compared with pH values of less than 6.2 for the mainstream smoke
of the same cigarette (26). This sidestream smoke thus contains free
nicotine, which is essentially absent in mainstream smoke. Further-
more, during smoldering, sidestream smoke is generated in a zone
that is even more oxygen deficient than the zones involved in
mainstream smoke generation during puff drawing. Consequently,
components that are primarily formed in a reducing atmosphere are
released into the environment to a greater extent than those formed
in mainstream smoke that is inhaled by the smoker. Table 2 lists the
amounts of some selected toxic compounds in mainstream smoke and
the ratios of undiluted sidestream smoke components to mainstream
smoke components (691.
The air dilution of sidestream smoke emitted into the atmosphere
is, of course, a determining factor for any assessment of human
exposure. Nonetheless, in risk assessment, consideration must also


TABLE t.-Distribution of selected toxic compounds in
      cigarette mainstream smoke (MS) and
     sidestream smoke (SS) of nonfilter cigarettea

Compound                   MS           ss/Ms

Gas phase

Carbon monoxide
Carbon dioxide
Formaldehyde
Acmlein
Acetone
Pyridine
SVinylpyridine
Hydrogen cyanide
Nitrogen oxidea (NW
Ammonia
N-NitrcsdimethyLemine
N-Nitroeopyrrolidine
Particulate phase

lo-23 mg         2.M.l
=mg         a11
70-100 jig         0.1150
-100 )lg         a15
lo&250 pg         2-5
20-4ow          lo-20
1-w          20-40
-I@        0.1-0.25
l~lG3         4-10
50-130 pg         4U-130
10-4oW          20-100
ma3          6-30

Particulate matter
Nicotine
Phenol
Catechol
Aniline
2-Toluidine
2Naphthylamine
EZen$aJanthracene
B-Mw=ne
Quinoline
N'-Nitmeonomicotine
N-Nitmecdiethanolamine
Nickel
Polonium-210

1mmg         1.3-1.9
l-2.3 mg          2.63.3
ML120 pg         2.&3Jl
100-280 pg        0.M.9
3@Jng           30
l@J w           19
1.7 ng           30
2.0-7.0 ng         2-t
mng          2.5-3.5
5f3w@30 ng       8-11
2@J-w@J w       0.53
2&70 ng          1.2
-ng          13-30
0.0345 pci        ?

SOURCE Hoffmann et al. ( 70).

be given to the fact that nitrogen oxide (NO), emitted into the
environment as a sidestream smoke component, is rapidly oxidized
to the more toxic nitrogen dioxide (NOz) (27).

Nicotine

Chemistry

A number of observations have supported the concept that
nicotine is the major habituating agent in tobacco and tobacco smoke
(90). In addition to nicotine, tobacco contains a large variety of other
alkaloids, most of which are 3-pyridyl derivatives (Figure 2). In the
blended U.S. cigarette, nicotine constitutes 85 to 95 percent of the
total alkaloids. Its concentration in the leaf depends primarily on the
tobacco type and variety, stalk position, and cultivating practices
(140). A study on the fate of Wlabeled nicotine, added in the form of

209


a salt solution to the tobacco rod of a filter cigarette, revealed that
14.9 percent of labeled nicotine emerged in the mainstream smoke
and 37 percent appeared in the sidestream smoke; 18.5 percent of
Wnicotine was deposited in the butt, and the remainder (= 30
percent) was broken down into pyrolysis products (Table 3) (73). The
major pyrolysis products of nicotine in MS and SS of cigarettes are
carbon dioxide, carbon monoxide, 3-vinylpyridine, 3-methylpyridine,
pyridine, myosmine, and 2,3'dipyridyl(130).

In most countries, cigarettes have shown a gradual and significant
reduction over the last three decades in the sales-weighted average
delivery of nicotine. In the United States the sales-weighted average
nicotine yields decreased from 2.7 mg in 1955 to 2 1.0 mg in 1982
(146). These nicotine reductions have been achieved primarily by
technological modifications and perhaps some agricultural changes.
The technological methods encompass extraction, oxidation or
transformation of nicotine into less toxic compounds (91), formula-
tion, and whole leaf curing. Reduction of nicotine delivery may be
achieved by lowering the transfer of the alkaloid from tobacco into
the smoke. This is accomplished by use of expanded tobacco laminae,
adding leaf mid-veins and stems in the form of tobacco sheets
(reconstituted tobacco), and by modifications of cigarette paper and
by filtration (air dilution). From an agricultural standpoint, breeding
lines have been developed with low levels of nicotine; however, these
are not being used in commercial varieties at present (35).

In 1982, about 90 percent of the U.S. cigarettes sold had filter tips
made of cellulose or cellulose acetate or combinations of these with
charcoal. Twenty-five percent of these filter cigarettes were perforat-
ed to allow greater air dilution of the drawn smoke puffs. More
recent filter construction utilizes longitudinal air channels in
addition to perforation for maximal smoke dilution by air (70, 146).

From the machine smoking of cigarettes, using standardized
parameters of taking one puff per minute of 2 seconds' duration with
a volume of 35 ml, the U.S. Federal Trade Commission reported in
March 1983 that the nicotine values of 208 commercial brands
ranged from ~0.05 to 2.0 mg per cigarette (146). However, many
people who smoke these cigarettes derive very different levels of
smoke components from them, primarily because nicotine delivery
in the mainstream smoke influences human smoking behavior and
causes many smokers of low-yield products to draw puffs more
frequently, take larger puff volumes, and inhale more deeply. This
phenomenon has been observed by determining the smoking profiles
of individuals or by assaying nicotine and cotinine in the sera of
smokers (64,65,127). Cigarette filter construction that allows partial
occlusion of the perforations or air channels of the filter tip may also
lead to delivery of higher concentrations of mainstream smoke (89).
Nicotine in mainstream and sidestream smoke of tobacco products is

210


Nrcotine
(loo-3,000 "g)l

?a' -Formylnornicotine
  (40-200 uq)

Nornicotine
(20-100 ug)

CP
0  I

  N

Anabasine
(3-20 ug)

Cotinine          Nicotine-Xl-Oxide
(9-57 ug)        (n.d.)'

N'-Acetylnornicotine
     (n-d.1

Nrcotyrine
(<SO ug)

Hyosmine
(S-23 ug)

Nornicotyrin
  (n.d.1

O-9 QP
  N

Anatabine
(3-40 2'1)

2,3'-Diayridyl
  (7-20 ug)

FIGURE 2.-Major alkaloids in tobacco and tobacco smoke
' Numbersri~parrntheseedenoteIrgint~wokeofoaecigarette.
' n.d.:Nc&determined.

today primarily quantitated by gas chromatography (151). In physio-
logic fluids such as saliva, serum, or urine, smokers' exposure to
nicotine is assessed by measuring the alkaloid itself or its major
metabolite, either by gas chromatography with a special detector (63)

211


TABLE t.-Distribution of nicotine and its pyrolysis
      products in cigarette smoke

Smoke products

Nicotine     Nicotine pyrolyaia
kmcent)      producta (percent)

     Maid- particulstea
     - phase
              vapor
     Sidestream particulatea
     Side&ream phase
             vapor
     Butt (deposition)
      Ash

       Total


SOURCE: Hourman (73).

14.9             0.6
0.0             4.1
37.0            4.1
0.0            16.3
16.5             1.7
0.0             0.0

70.4       26.8

or, for large volumes of analyses, preferably by radioimmunoassay
(RW (92).

Metabolism

The quantity of nicotine absorbed by the smoker depends on a
number of factors, such as its concentration in the smoke, the
individual's smoking pattern, and the smoke pH. As discussed
earlier, pH values greater than 6.2 increase the amount of unproton-
ated nicotine in the smoke (Figure 1) (26). In the oral cavity, nicotine
absorption varies between 4 and 45 percent (5). In the case of cigar
smoke, the presence of free nicotine in the vapor phase, which is
readily absorbed, and the harsh nature of alkaline smoke appear to
be the major reasons for the cigar smoker's tendency to avoid
inhalation of this smoke (144). That levels of cotinine in the serum of
cigar smokers approximate those of cigarette smokers supports the
concept that nicotine from cigar smoke is absorbed through the
mucous membranes of the oral cavity. Smokers of blended cigarettes
and of British types of cigarettes (pH < 6.2) tend to inhale the smoke.
In general, between 15 and 25 percent of the nicotine in tobacco
appears in cigarette mainstream smoke, from which up to 90 percent
is absorbed (162). The extraction of nicotine from the lungs occurs
quite efficiently (143). Nicotine enters the pulmonary capillary blood
and reaches the brain via the arterial systems (4).

The metabolism of nicotine occurs principally in the liver. Its main
metabolite, cotinine, appears in the blood within a few minutes after
inhalation, and significant amounts of other metabolites appear in
the tissues after about 5 minutes. The kidneys and lungs may, to a
minor extent, also be involved in the metabolism of nicotine (24,143).

Figure 3 illustrates the pathways of the metabolism of nicotine
(54). Cotinine is the major metabolite of nicotine. It appears within a
few minutes in the blood of smokers and has a half-life time of
between 20 and 30 hours (92). Cotinine has been detected in

212


concentrations up to 10 pg per milliliter in the urine of smokers and
up to 40 ng in the urine of nonsmokers who remained in a heavily
polluted indoor environment for a minimum of 1 hour (65, 69).
Nicotine-N'oxide has been detected in the urine of smokers (55). In
the oral cavity of man, nicotine-N'oxide is reduced to nicotine (78). It
has also been found that ingested nicotine-N'-oxide is reduced to
nicotine by the gut flora or by intestinal enzymes (54). Nornicotine
has been observed to be formed by Ndemethylation of nicotine (54).
y-(3-Pyridyl)-y-methylamino butyric acid has been identified in
human urine as 3-pyridylacetic acid, the major end product of
nicotine metabolism (54).

Association With Cardiovascular Diseases

The pharmacological effects of nicotine absorbed by inhalers
might be considered small and transient, but they are repeated
many times each day and act directly on the sympathetic and
parasympathetic cells of the central nervous system (CNS). Nicotine
exerts a direct effect on ganglion cells, producing transient excita-
tion followed by depression or transmission blockade. At the level of
the central nervous system, nicotine causes CNS stimulation fol-
lowed by CNS depression (133).

Nicotine, like acetylcholine, discharges adrenaline from the adre-
nal glands and other chromaffin tissue, and releases noradrenaline
from the hypothalamus. It also releases antidiuretic hormone from
the pituitary (331, and by excitation of chemoreceptors in the carotid
body, augments various reflexes (36).

The cardiovascular responses to nicotine, in general, parallel those
that follow stimulation of the sympathicoadrenal system. Because
nicotine has both stimulant and depressant effects, the responses of
the cardiovascular system represent the sum of several different
modes of action of this compound.

Recently, progress has been made in identifying nicotine receptor
sites in the brains of animals (81). When microinjected into the third
cerebral ventricle, nicotine increased cardiac and respiratory rates,
but it did not alter these parameters when microinjected into the
periaqueductal gray.

Most studies have shown that in people with known coronary
heart disease, cigarette smoking increases the incidence of angina
pectoris, although angina directly precipitated by smoking is rare.
With additive risk factors such as hypertension, the threshold for
attacks of angina can be significantly lowered by daily cigarette
smoking. A long-term followup study of part of the Framingham
cohort demonstrated a lack of association between smoking at
diagnosis and subsequent cardiovascular events, however (74). This
appeared to be related to changes in habits following diagnosis,
implying that improved prognosis could be achieved by withdrawal

213


FIGURE 3.-Pathways of nicotine metabolism
SOURCE: US. Department of Health. Education. and Welfare (144~.

214


TABLE 4.-Effects of nicotine on the cardiovascular system

Increases in blood pressure
Increases in heart rate
Electorcardicgraphic changes
  Nonapecifk ST and T-wave changes
  Increased conduction velocity, propensity toward arrhythmias
Exacerbation of angina in coronary patients
Diminished left ventricular performance in coronary patients

SOURCE: Stimmel(136).

of daily cigarette smoking. The relationship here cannot be attrib-
uted solely to removal of nicotine; it can also be related to
improvements in cardiovascular status such as increased oxygen
saturation.

The exact mechanism whereby nicotine could trigger a cardiovas-
cular event is unknown. However, a common initial feature of
several suggested mechanisms is a sympathetic discharge (Table 4)
(14). This stimulation of sympathetic nerves, with consequent release
of the neurotransmitter norepinephrine within the myocardium, has
been shown to lower the ventricular fibrillation threshold in animals
(38). Responses to sympathetic activity such as hemodynamic
parameters of heart rate and systolic blood pressure generally reflect
plasma nicotine concentrations (88, 112, 135). Increase in pulse rate
and systolic pressure, decrease of pressure pulse transit time, and
digital blood flow are well correlated with nicotine levels of the
cigarettes smoked (88); the same effects can be noted with intrave-
nous injection of varying doses of Gnicotine (150).

In a recent study on the cardiovascular effects of infused nicotine,
Benowitz and colleagues (18) showed that nicotine infusions could
achieve plasma concentrations and cardiovascular effects similar to
those induced by cigarette smoking. Heart rate increased after low
concentrations of nicotine were administered and reached a plateau,
beyond which increasing blood concentrations of nicotine had no
effect. In these studies, as in those of Cryer et al. (38), elevations in
pulse rate and blood pressure occurred promptly after the start of
nicotine exposure and preceded the investigators' measurement of
increased increments of plasma epinephrine and norepinephrine
concentrations. These hemodynamic effects of nicotine are probably
not mediated by circulating catecholamine levels, but are due to
local release of the sympathetic neurotransmitter norepinephrine
from adrenergic axon terminals.

The increase in circulating catecholamines has been demonstrated
to be nicotine dose dependent by Hill and Wynder (66), who found an
increase in epinephrine levels in plasma to be proportional to the
nicotine content of the cigarette smoked. Although elevated with
smoking, norepinephrine levels were not correlated with plasma

215


NICOTINE

SYMPATHETIC NERVE GANGLION,
ADRENAL MEDULLA, NERVE ENDINGS
I

RELEASE OF EPINEPHRINE AND NOREPINEPHRINE

PHYSIOLOGICAL HEART CHANGES
    PERIPHERAL VESSELS
CARBOHYDRATE METABOLISM
      FAT METABOLISM
        PLATELETS
       COAGULATION

FIGURE I.-Net results of catecholamine elevations as
      mediated through nicotine stimulation
SOURCE:Schievelb& andE%erhardtWZ?J.

nicotine. The release of epinephrine is probably mediated through
the adrenal medulla and chromaffm tissue. A more recent study
showed that when the pH of the smoke is altered to favor more
effective absorption of the alkaloid, smokers compensate to adjust
nicotine intake, resulting in modulation of the epinephrine response
(5%

The net results of catecholamine elevations as mediated through
nicotine stimulation are summarized in Figure 4. In general, the
cardiovascular responses of increased heart rate and augmented
contractility create a variable oxygen need; nicotine also dilates the
coronary arteries, which exerts a measurable effect only in healthy
individuals without coronary heart disease (129).

216


Available evidence suggests that the release of adrenergic trans-
mitters by nicotine and other cholinergic agonists is mediated by
nicotine receptors (159). Norepinephrine outflow from adrenergically
innervated tissues is evoked by nicotine and the nicotinic agonist
dimethylphenylpiperainium, but not by the muscarinic agonists
methacholine and oxytremorine (20). Studies using a variety of
inhibitors of nicotinic receptors have resulted in a number of
possible and related mechanisms of action of nicotine on the central
nervous system and the cardiovascular system: (a) prevention of
inactivation of the adrenergic transmitter after its release, (b)
facilitation of adrenergic transmitter release, Cc) involvement of
histamine (311, (d) mediation of nicotine by neuronal uptake, (e)
induction of nerve action potentials, (f) depolarization of nerve
terminals, and (e;) exocytotic release of adrenergic transmitter. These
mechanisms are not necessarily mutually exclusive.

Through these mechanisms, nicotine might exert powerful effects
on the sympathicoadrenal control of cardiovascular function. Many
issues remain to be elucidated and understood, however, such as the
effects of drug interactions with nicotine (421, the inter-individual
variability in metabolism of this alkaloid (18), and tolerance (1321, as
well as the habituating mechanisms of the cigarette smoking habit
itself (77).

In addition to hemodynamic perturbations, nicotine intake has
been implicated in atherosclerotic changes in the arterial wall itself.
Nicotine has been associated with acceleration of atherosclerotic
disease, the initiation of which is becoming more clearly understood.
Experimental evidence is emerging that links atherogenesis to
tobacco smoke inhalation, possibly through mechanically induced
intimal damage as well as through hypoxia. The pathogenesis of
vascular injury might occur via two mechanisms. The first mecha-
nism is accelerated by high levels of blood fat, particularly cholester-
ol in the low density lipoprotein fraction, and it is the abnormal
accumulation of this lipid in the lining of large arteries that leads to
changes in cells of the arterial wall, causing a thickened intima rich
in fat and smooth muscle cells.

The other mechanism of atherogenesis resides in vessel injury and
thrombosis, with the thrombus persisting and becoming organized,
leading to vessel wall thickening (103). Plaque development may
therefore involve damage to the intima and a complex interaction
between intima and the circulating blood, resulting in (a) platelet
release of mitogenic factors, (b) increased uptake of serum lipopro-
teins, and (c) intracellular deposition of lipid (123). It has been shown
that platelets and cells, such as macrophages, can make available at
sites of injury a protein that will cause smooth muscle cells to
proliferate and another that will make them migrate into the
arterial intima (122).

217


Unfortunately, we have relatively little evidence about how the
endothelium can be injured. Blood flow, particularly at a high shear
rate, can damage the endothelium, and high serum lipid `levels can
be associated with vessel injury (125). Immune reactions have been
shown to damage vessels, and products from cigarette smoke
components are suspected of causing injury to the cellular lining of
vessels (87).

Nicotine can be implicated in the processes of atherogenesis
through several of its known actions on the cardiovascular system
(47). Via catecholamine release and the subsequent increase in free
fatty acids, nicotine affects different steps in blood coagulation
pathways, including platelet aggregation (48, 129) and increased
fibrinolytic activity, with possible injurious effects to the cells lining
the arterial wall. Blood pressure and heart rate can influence the
blood shear rate, and transitory increases in blood sugar can
influence basic metabolic rates (42).

A desquamating effect of nicotine on the vascular endothelium has
been demonstrated in rabbits when nicotine was given in relatively
low doses, as compared with those received by cigarette smoking (23).
In a clinical assay, smoking two cigarettes increased the number of
circulating endothelial cells by 50 percent. The interaction of
nicotine and other cigarette smoke constituents could influence this
increase in endothelial cells released into the bloodstream (137).

The possible role of prostaglandins in the promotion of atherogene
sis is tied in with the nicotine induction of altered prostaglandin
activity (155, 157). EXdence suggests that smoking causes a throm-
boxane (TxAz):prostacyclin (PG12) imbalance. Nicotine increases
platelet activity in vitro, and cigarette smoking in general has been
shown to increase this activity in humans (95) as well as to
potentiate platelet aggregability in the presence of hyperlipidemia
(67). Nicotine inhibits the ability of coronary arteries to synthesize
prostacyclin-like substances in vitro, an effect that is more pro-
nounced in vascular cultures derived from patients who smoke than
from nonsmoking control subjects (156). This phenomenon and
alterations in endothelial barrier functions have also been investi-
gated in umbilical arteries derived from smoking and nonsmoking
mothers (7). Pronounced changes were seen in the endothelium of
the vessels derived from smoking mothers. These consisted of
swelling of the cells with numerous blebs on the luminal membrane.
In addition, extensive edema of the subendothelial spaces was a
regular feature,  as was an increase in basement membrane
thickness. These changes are consistent with an increase in endothe
lial permeability (7).

Cigarette smoking and nicotine in particular may therefore alter
platelet function and prostaglandin production in potentially delete-
rious ways, suggesting that smoking may interfere with vascular

218


defense against platelet deposition, a factor in atherogenesis (124).
Specifically, nicotine might alter the occupancy of endothelial
receptors for &adrenergic catecholamines, increasing the intracellu-
lar concentration of cyclic AMP and inhibiting the release of
arachidonic acid from endothelial cell phospholipids, and thereby
reducing prostacyclin synthesis (I).

Nicotine has been implicated at other levels of arterial cell
function. This alkaloid has been found to produce an exposure-time-
and concentrationdependent effect on the lysosomes of endothelial
cells through increases in lysosome fragility and formation of large
acid phosphatase positive vacuoles, presumably lysosomes, in rat
endothelial cells (158). These findings suggested that the lysosomal
vacuolar system should be examined as a possible target for cellular
dysfunction following chronic exposure to nicotine.

Alterations in the levels of lysosomal enzymes in atherosclerotic
tissue have been demonstrated in the presence of single risk factors
including hypercholesterolemia (601, hypertension, and diabetes
mellitus (160) in animals, as well as in diseased areas of human
vessels (49). Inhalation experiments have shown that aortic tissue
from animals chronically exposed to cigarette smoke exhibited
increased levels of several lysosomal enzymes as well as cholesterol
and cholesteryl esters (57). Studies are currently being carried out to
determine the specific cigarette smoke constituents responsible for
these changes.

Cigarette smoking has been implicated in elevation of serum lipids
and changes in lipoprotein distribution (127). Reports in the litera-
ture run from no alteration (118) to significant changes in the
direction of those levels promoting atherogenesis (25, 71). By
stimulating sympathetic nervous activity and catecholamine release,
nicotine can cause an elevation in plasma free fatty acids and
increased secretion of very low density lipoprotein triglycerides (21).
Kirchbaum and colleagues (84) found that the rise of free fatty acids
in plasma produced by smoking was twice as high in patients who
had suffered a myocardial infarction as in controls. This rise in free
fatty acids is probably a result of catecholamine-mediated lipolysis
and may be an important mediator in the production of endothelial
cell injury (161). In addition, cigarette smoking in general is
associated with a reduction in high density lipoprotein apoprotein
components (19) and may attenuate this lipoprotein's antiatherogen-
ic properties by altering surface phospholipid constituents (62).

Carbon Monoxide
Chemistry

The formation of carbon monoxide (CO) occurs in and close to the
burning cone of a cigarette by thermal decomposition, by reaction of

219


tobacco with atmospheric oxygen, and by secondary reactions of
tobacco with carbon dioxide, water, and other primary pyrolysis
products (13, 69). Studies with labeled precursors have shown that
CO is formed at temperatures above 460oC and that about 30 percent
of it derives from thermal decomposition of tobacco, 36 percent from
combustion of tobacco, and 23 percent from reduction of carbon
dioxide (13). The yield of CO in the mainstream smoke of a cigarette
depends on the amount of tobacco and the type of paper burned
during puff drawing, the concentration of pyrolytic precursors, the
temperature profile of the tobacco during puff drawing, and the
permeability of the wrapper for the outward diffusion of CO (111).
The CO concentration in the inhaled smoke is also a function of the
permeability of the cigarette paper and of the physical and chemical
properties of the filter (86,153).

The importance of the air velocity created during puff drawing is
demonstrated by the significant increase in the smoke with increas-
ing puff volume (Figure 5) (69). The somewhat higher CO yields in
the smoke of cigarettes with conventional filter tips compared with
those from nonfilter cigarettes have been attributed to the loss of air
dilution and to the increased smoke velocity that occurs when the
last 15 to 25 mm of the tobacco column are replaced by a filter tip
with a nonporous wrapper (69, 153). This concept is also reflected in
the higher CO yield for a filter cigarette smoked with puffs of
different volumes (Figure 5).

In recent years, cigarettes with perforated filter tips have gained
in market share. The air entering through the perforations in these
filter tips dilutes the mainstream smoke and thus reduces the
concentration of CO in the smoke (Figure 5). In addition, carbon
monoxide is selectively reduced, most likely because of the reduced
velocity of air entering the burning cone (86). Finally, the newly
introduced filter cigarettes with longitudinal air channels reduce CO
in the smoke even further (68). However, as discussed before, because
of the compensatory changes in smoking behavior that smokers
make in order to satisfy their need for a certain amount of nicotine
and because of the possible obstruction of perforated filter tips or
their air channels during puff drawing, the smokers may not fully
benefit from the intended air dilution of the smoke (64, 65, 68, 89,
127).

Association With Cardiovascular Diseases

The health effects of exposure to carbon monoxide are not fully
known. However, research findings in selected population groups
indicate that carbon monoxide acts as an added stress factor to
precipitate cardiac symptomatology or ischemic episodes in individu-
als already compromised by coronary disease (3). Additionally,
excessive levels of carbon monoxide in the blood have been found by

220


A
30-

25-

20-

5-

TAR

  30 35 40 45 50
PUFF VOLUME - ML

FIGURE 5.-Carbon monoxide, tar, and nicotine in the
      smoke of cigarettes
NOTE: NF = Nonfilter; F = Filter; PF = Perforated Filter.
SOURCE: Hoffmann et al. (69).

221


some investigators to impair certain perceptual and motor functions
(104).

Carbon monoxide is a common industrial pollutant generated by
any burning process, It is odorless and tasteless and gives no warning
of its presence in most circumstances, thus allowing for chronic
exposure over extended periods of time. The early symptoms of
carbon monoxide poisoning often resemble those of a variety of
diseases; thus, tissue hypoxia might occur in healthy people without
forewarning (163).
Carbon monoxide combines with hemoglobin in large quantities.
Its chemical affinity for hemoglobin is over 200 times greater than
that of oxygen. The ability of carbon monoxide to cause tissue
hypoxia stems from two effects on circulating processes: (a) a
reduction in the total amount of oxygen carried by red cells in blood,
which reduces delivery of oxygen to tissues, and (b) a shift to the left
of the oxyhemoglobin dissociation curve of the blood that contains
both oxyhemoglobin and carboxyhemoglobin. The leftward shift of
the oxyhemoglobin dissociation curve decreases the tension at which
oxygen molecules are dissociated from hemoglobin, and therefore
decreases the driving pressure for diffusion of oxygen into tissues
and cells (142). The concentration of carboxyhemoglobin reached in
blood will depend upon the concentration of carbon monoxide in the
inhaled gas mixture as well as on the alveolar concentration, and on
arterial oxygen tension, duration of exposure, and ventilation
volume (142).

It is noteworthy that carbon monoxide binds to muscle myoglobin
as well as to hemoglobin and may exert tissue hypoxia by interfering
with oxygen transport to muscle mitochondria. It can also have an
effect on other hemoproteins such as those present in the cyto-
chrome oxidase system (35). Cytochrome P-450, a mixed-function
oxidase, probably does not bind sufficient carbon monoxide to cause
inhibition of drug hydroxylation, even at a carboxyhemoglobin
saturation of 15 to 20 percent (126). However, smokers and nonsmok-
ers differ in their clinical reactivity to drugs such as phenacetin and
diazapam through acceleration of biotransformation processes. This
increased drug metabolism is believed to result from an induction of
microsomal drug-metabolizing enzymes after chronic exposure to
tobacco smoke (148). Although smoking in general results in
accelerated enzyme induction, carbon monoxide has been implicated
in decreased hepatic protein synthesis, a phenomenon that can be
duplicated by chronic cigarette smoke exposure (52). This effect
could be due to a deleterious reaction of hepatic tissue because of
binding of carbon monoxide to intracellular hemoproteins.

It is possible that carbon monoxide may bind to hemoproteins
other than cytochrome oxidase, hemoglobin, or myoglobin in suffi-
cient amounts to inhibit their function. Tryptophan dioxygenase and

222


cat&se have high affinities for carbon monoxide, which in the first
case, could result in increased serotonin levels in several tissues, and
through the latter enzyme, could cause the cellular accumulation of
hydrogen peroxide, a toxic oxidant (126).

Cigarette smoking causes increased carboxyhemoglobin levels,
thereby reducing the oxygencarrying capacity of the blood and
subsequent tissue oxygenation and, possibly, cellular metabolism.
Interpretation of the effects of carbon monoxide absorption by
healthy people is a subject of much controversy because effective
threshold levels have not been established. Environmental exposure
to carbon monoxide may be without chronic consequences in healthy
people who appear able to compensate for any acute effects at levels
of industrial exposure. However, in patients with angina pectoris,
exposure to carbon monoxide has reduced exercise time until the
onset of chest pain (3).

Atherosclerosis is a multifactorial disorder in which cigarette
smoking and carboxyhemoglobin levels may exert varying effects,
depending upon the other risk factors present. Carbon monoxide is
believed to be a contributing factor to the acceleration of the disease
process. Wald and associates (152, 153) have suggested that carboxy-
hemoglobin levels in tobacco smokers correlate better than a
smoking history with the development of myocardial infarction,
angina pectoris, and intermittent claudication. However, it should be
pointed out that in such studies, elevation in carboxyhemoglobin also
reflects the absorption of other constituents of tobacco smoke.

In his earlier studies, Astrup (8) found that carbon monoxide or
decreased oxygen tension enhanced coronary atherosclerosis in
cholesterol-fed rabbits and suggested that high carboxyhemoglobin
levels resulting from tobacco smoking were associated with the
development of occlusive arterial vascular disease. More recent
experiments, conducted in a blind fashion, confirmed the presence of
myocardial lesions, but failed to produce aortic atherosclerosis (91,
although other researchers had noted morphological alterations in
coronary arteries similar to those seen in the earlier experiments (8,
40).

Animal experiments on accelerated atherogenesis with carbon
monoxide must be considered to be unsatisfactory, and although
clinical trials have linked elevated carboxyhemoglobin levels with
cardiovascular symptomatology, a cause and effect phenomenon for
carbon monoxide and disease of the arterial wall has not been
elucidated.

Carbon monoxide exposure has been related to increases in
circulating blood lipids. This was first demonstrated in rabbits in
which carboxyhemoglobin levels were 15 to 25 percent, resulting in
elevations of serum cholesterol for a transient 2 to 3 weeks following
exposure (10). In an epidemiological study by Van Houte and

223


Kesteloot (147) involving 42,000 subjects, significantly higher serum
cholesterol levels were demonstrated in male smokers than in
nonsmokers, particularly in the 30- to 39-year age group. The early
Framingham results also reported slightly higher serum cholesterol
levels among smokers (41), whereas Blackbum et al. (22) report
nonsignificant differences between smokers and nonsmokers. Many
other epidemiological investigations have analyzed the correlation
between smoking status, serum cholesterol levels, and lipoprotein
distributions, but the majority of results suggest a trend rather than
significance in support of a connection between smoking and
elevated cholesterol (100). This relationship does not offer an
explanation for an effect of cigarette smoking on serum lipid levels,
but experimental evidence suggests that carbon monoxide might
exert an effect on the metabolism of chylomicron remnants (51).
Ascorbic acid levels could limit the conversion of cholesterol to bile
acids (85), or carbon monoxide could diminish hepatic degradation of
lipoprotein constituents (40).

Carbon monoxide has also been implicated in the noted decrease in
patency following vascular surgery and arterial reconstruction
among continuing cigarette smokers (138). Cigarette smoking as
measured by extrapolation from carboxyhemoglobin values had a
definite adverse effect on the healing process and success rate of
vascular surgery (20,56).

Additional Contributors

Hydrogen Cyanide

Nitrates in tobacco serve as precursors for hydrogen cyanide
(HCN) in the smoke (61, 79). This concept has been supported by
studies in which `5N-labeled nitrates of potassium, sodium, or
calcium, respectively, had been added to cigarette tobacco and in
which the isolated HCN from the smoke was found to contain l5N-
HCN (79). However, tobacco proteins appear to be the major group of
precursors for HCN in the smoke (27, 80). This was demonstrated in
pyrolysis studies with tobacco protein and amino acids and by
recovery in the smoke of 15N-HCN from cigarettes spiked with l5N-
glycine (27, 80). It has been shown that HCN is formed via N-
heterocyclic intermediates and from pyrolidine, the decarboxylation
product of proline (80). From these intermediates, hydrogen cyanide
is split off in the pyrolysis-distillation zone of the burning cigarette
under the opening of the N-containing ring (27, 80).
Several methods have been explored for the selective reduction of
hydrogen cyanide in cigarette smoke. Charcoal-containing filter tips
can remove 70 to 80 percent of the HCN from cigarette mainstream
smoke (139). With increasing smoke dilution through filter perfora-
tion, HCN can be selectively reduced up to 80 percent, with a 70

224


percent ventilation rate (105). The extraction of protein fractions
from cigarette tobacco leads also to a selective reduction of HCN
(141). The smoke of one cigarette contains from 20 to 480 pg of HCN,
the lowest values being measured in the smoke of cigarettes with
charcoalcontaining filter tips (120).

Although only nicotine and carbon monoxide have been incrimi-
nated as contributors to the increased risk of cigarette smokers for
cardiovascular disease, other gas phase constituents might also play
roles in the pathogenic processes. Hydrogen cyanide is an inhibitor
of several respiratory enzymes and as such can influence cellular
metabolism in the myocardium and arterial wall. It is also a
powerful ciliatoxic agent allowing for decreased efficiency in remov-
al of tar constitutents from the respiratory system (16).

The arterial effects of hydrogen cyanide, nitric oxide, and carbonyl
sulfide were investigated by Hugod and Astrup (76) in experiments
in which rabbits were exposed to these compounds alone or in
combination with carbon monoxide. The duration of exposure was
from 5 days to 12 weeks, and aortas were assessed morphologically
for intimal damage suggestive of early atherosclerotic changes. No
hi&toxic effect on intimal or subintimal morphology was noted, and
a parallel experiment demonstrated no effect on the coronary
arteries (75). The duration of exposure to these compounds must be
considered short, however, with the possibility that prolonged
exposure might result in morphological or enzymatic alterations.

Nitrogen Oxides

A number of studies have shown that nitrate is a major precursor
for nitric oxide (NO) and traces of nitrous oxide (NzO) and nitrogen
dioxide (NO2) found in cigarette mainstream and sidestream smoke
(79, 134). The mainstream smoke of a cigarette contains from 6 to
600 pg of NO per cigarette, depending primarily on the nitrate
content of the tobacco blend and the nature of the filter tip (105). The
correlation of nitrate content of tobacco with nitric oxide yield in the
mainstream smoke appears to be linear (27). The smoke of cigarettes
made with Burley tobacco and particularly those made with Burley
stems, which are rich in nitrate, is especially high in nitric oxide (29,
107). Tobacco proteins appear also to contribute to the NO yield in
cigarette smoke (79).

The concentration of nitrogen dioxide and that of methyl nitrite in
freshly generated cigarette smoke is very minute ( < 5 w/cigarette);
however, these agents increase rapidly as a function of the aging of
mainstream and sidestream smoke. Within about 3 minutes, 50
percent of NO is oxidized to NO2 (149). The best approaches toward
reducing NO in cigarette smoke are by reduction of nitrate in
tobacco and by dilution of smoke by air entering through the holes of
perforated filter tips (27,105).