suggest hypotheses for mechanisms of sudden death in man, they do
not, of course, deal directly with cases of sudden death.

The Effect of Smoking on Sudden Cardiac Death in Animals

The smoking and health report of 1976 (138) has tabulated in Table
A20 (pp. 108-108) papers concerned with the effect of smoke or nicotine
on the cardiovascular system of animals. In the presence of myocardial
ischemia, exposure to tobacco smoke or nicotine may precipitate
conditions of increased cardiac demand, relative ischemia, and, in one
experiment, arrhythmias. Bellet and colleagues (20) found that the
ventricular fibrillation threshold was reduced in dogs exposed by
intubation to cigarette smoke both in the presence and in the absence
of acute myocardial infarction.

Malinow and colleagues failed to induce infarction or sudden death
in cholesterol-fed cynomolgus monkeys by chronic exposure to CO (80).
There are, however, no animal experiments in which animals have been
brought chronically to a state of incipient myocardial ischemia by
atherogenesis and then exposed to whole smoke by inhalation in a
nonstressful setting. ,

Research Needs

There are fewer data on sudden cardiac death than on myocardial
infarction in general. Smoking is clearly a strong risk factor for sudden
death, but present indications are that it is not unique among the mix
of risk factors for coronary heart disease and that it is not highly
predictive. However, there are theoretical reasons to speculate that
smoking might have a relationship to sudden death, not only through
its effects on the circulation, but also through a myocardial one. It
should be considered whether present epidemiological and clinical
research data are adequate to exclude in smokers a myocardial element
in sudden cardiac death, in relation to either first or multiple heart
attacks, or whether additional research is warranted.

The mechanisms of sudden cardiac death, its precursor states, and
preventive therapy require further elucidation. These should be
clarified where possible in man and in experimental animal models
with close analogy to man. The study of smoking or of smoke
constituents as variables in such studies may be informative both about
sudden death and the role of smoking in its occurrence.

Conclusions

Smoking is a powerful risk factor for sudden cardiac death. It is,
however, only one of the general group of risk factors that contribute
to coronary heart disease and sudden death. The mechanisms by which
smoking might induce sudden death, in addition to an exacerbation of
coronary artery arteriosclerosis, can be hypothesized from experiments

4—45
that indicate that an exacerbation of regional ischemia may promote
electrical instability of the heart, fibrillation, or asystole. Further
research will be required if these mechanisms are to be well understood
and if they are to be shown to be actual mechanisms in man in relation
to smoking and sudden death.

Angina Pectoris
The Nature of Angina Pectoris in Humans

Pain in the thorax may have several different origins and can create a
difficult problem of differential diagnosis. Angina pectoris arises
typically in the face of exercise and increased demand for work and
oxygen on the part of the heart which cannot be met immediately in
the presence of ischemia imposed by coronary atheroscleosis. The
origin of the pain is thought to be the ischemic myocardium. It can
occur in individuals with or free from preexisting myocardial
infarction. Since the common use of angiographic diagnostic methods,
it has become apparent that angina also occurs occasionally in persons
with little or no evidence of coronary arteriosclerosis.

Angina pectoris is associated with an increased death rate from
heart attack. Women survive better than men. Among the risk factors
associated with a poorer prognosis are hypertension, cardiac hypertro-
phy, congestive heart failure, and electrocardiographic abnormalities
(149). Recent studies employing angiography have shown a close
relationship between the extent of coronary arteriosclerosis and
prognosis in angina pectoris. Reeves and associates (108) have
summarized these reports to indicate that if only one of the three
major coronary artery branches is significantly stenosed, an annual
mortality rate of about 2 percent results; if two major branches are
stenosed, the resulting annual mortality rate is about 7 percent a year;
with three-vessel disease, it is about 11 percent a year.

Summary of Epidemiological Data

The major studies on smoking in relation to the incidence of angina
pectoris in the United States are not consistent in their conclusions.
The 1976 report on smoking and health (138) has tabulated four major
reports in Table 5 on page 38. (Table 5 is reproduced below as Table 6.)
Doyle and colleagues (38) report no association in a 10-year follow-up
of men from the Albany civil servant study, together with men from
the Framingham Heart Study. Jenkins, et al. (63) reported a slight
positive association, but not a statistically significant one. Similarly,
Kannel and Castelli (70) reported on both men and women from the
Framingham Heart Study and found a positive risk association among
men and a negative one among women. In a large study of 110,000 men
and women enrolled in a health insurance medical care plan in New
York City and followed for 3 years, Shapiro, et al. (122) reported a

4—46
TABLE 6. — Coronary heart disease morbidity as related to smoking — angina pectoris — prospective studies
(Risk ratioe—actual number of CHD manifestations shown in parentheses). [SM ~ Smokera NS =

 

 

 

  

 

 

 

   

Ne kers}
Auther, Number aad Number
yeu, type of cole an of Cigareties/day ante Age variation Comsnents
omustry population incidents

Doyle, 2a mass, Detailed 10 a NS include ex-
aa, Pramisgham, medical smokers and
14, 30-82 years azamination pipe and
URA of age aad oar

1915 males, foliow-up a emoker:
Albany,
30-55 years
af age.

Jenkine, «9,182 mason Initial “ BONG eects ea eee 1.00(9) NB inchade
al, aged 90-50 edical All current, formar pipe
1908, at entry, examination CIGREWMOE cece 14416) and cigar
USA. and follow. DIB cece eee ee eee 1.0614) smokers,

up by
repent
oe
tion.

Kanes, 5137 mateo Medical 12
wal, and females examination
USA years of age aad [ollow-

wo-08 ip.

‘Shapiro, 110,000 male Baseline 3 Males p< 0.01}
ot at, and female medical Od 86464 Hp< 0.06)
10m, enrollees of interview 1.00 1.00 NB include
UBA. New York and 1ST 2.08 ex-emokere

City examine- 140 1B

HIP 96-04 ion and een 258 4b

years of regular Females

age. follow-up. Dt 1001.00 1.00
Current cigaretion 154 18?) «(097
<M... LOT 158.08
> an

 

"Unless otherwiee specified, disparities between the total number of manufestations and the eum of the individual amaking categories are due to the exclusion of either cocasional, miansllaneous, mixed, or
er-amokers.
SOURCE: U. 9. Public Healta: Service (138).

Lv—t
significantly increased incidence rate for smokers among men who
were current users of cigarettes. Among females, the trend was
positive but not significant. A study of the incidence over 5 years of
angina among 10,000 Israeli men found that there was a higher
incidence rate among men who smoked over 20 cigarettes a day than in
those who smoked less, but the difference did not reach the 0.01 level
of significance (91). In addition, a questionnaire survey (45) of about
70,000 persons has found that more smokers than nonsmokers admitted
to chest pain. Some nine different kinds of angina-like and nonanginal
pains were included as chest pain. Reid and associates have reported a

significant association between angina and current cigarette smoking
among British civil servants (110).

The Effect of Smoking on Angina Pectoris

As noted above, the predictive risk factor association of smoking with
the incidence of angina pectoris is not clear. However, there is evidence
among persons with angina that smoking lessens the threshold of
exercise for the onset of pain. Aronow (7, 8, 9, 10, 12) has reported
clinical studies in which smoking cigarettes with high, low, or no
nicotine content aggravated angina. In these studies, high nicotine
cigarettes aggravated exercise-induced angina more than low nicotine
cigarettes, and low nicotine cigarettes more than cigarettes without
nicotine. He has also reported in patients with angina pectoris and
coronary artery stenosis documented by angiography that when 50
parts per million of CO were inhaled until the mean COHb level of
venous blood was raised to 2.68 percent, it was accompanied by a
significant decrease in exercise time before anginal pain. There was
also a decrease in the amount of cardiac work represented by the
product of systolic blood pressure and heart rate needed before the
onset of angina compared to when air was breathed. S-T segment
depression of 1.0 mm or greater in the electrocardiogram occurred
earlier, after less exercise and at lower cardiac work levels among
these patients when they breathed CO rather than air. Although it is
uncommon, there are patients in whom the act of smoking a cigarette
will itself precipitate an attack of angina (26, 143).

An interpretation of such data is that, in the patient with a
compromised regional myocardial blood supply who can provide little
or no compensatory increase in circulation to meet an increased cardiac
demand, smoking enhances both hypoxia and cardiac demand,
resulting in a more severe ischemia and an earlier onset of angina.

Research Needs

Epidemiological data with respect to the predictive or risk factor .
association of smoking and angina pectoris tend to show an inconsis-
tent positive association. Despite this unsatisfactory state of affairs,

4—48
there would seem relatively little reason to attempt to study the issue
further at this time.

Conclusions

Studies of the possible role of smoking as a risk factor for the incidence
of angina pectoris suggest a positive association, but the findings are
inconsistent.

In patients with angina pectoris, smoking lowers the threshold for
the onset of angina. Both nicotine and CO aggravate exercise-induced
angina.

Cerebrovascular Disease
The Nature of Cerebrovascular Disease in Man

The underlying circumstances of stroke are varied. They include
tumors and bleeding dyscrasias leading to intracerebral hemorrhage or
infarction, unusual diseases of blood vessels in the brain, aneurysms of
intracranial vessels, embolism, thrombosis, vascular rupture, and
atherosclerosis of the vessels of the neck and their distributing vessels
in the brain.

The great majority of strokes, perhaps more than 90 percent, may be
classified either as intracerebral hemorrhage associated primarily with
hypertension, or ischemic cerebral infarction associated with athero-
thrombotic disease of the vessels of the neck and their main
distributing branches in the brain. Infarction is more common than
hemorrhage. The clinical diagnostic subclassification or separation of
hemorrhagic stroke and ischemic stroke contains an appreciable
margin for misclassification. It is these conditions that are under
consideration here, rather than the rare disorders.

The risk factor data for stroke have been considered recently by two
panels (31, 40). They are less clearly defined than those for coronary
heart disease. The strongest gradients of risk are associated with age,
blood pressure, preexisting cardiovascular disease, and diabetes
mellitus. Prospective studies have not found a clear and direct
relationship with serum cholesterol concentration. It has been of
interest that a Japanese study has recently reported that among a
population with a high incidence of stroke but low levels of blood
cholesterol by Western standards, there was no evidence that
hypercholesterolemia defined as levels above 200 mgm/100 mi
increased the incidence of stroke. Cerebral infarct developed in 11
percent of those with hypertension and hypercholesterolemia and 21
percent of those with hypertension alone (101).

_ Models of cerebrovascular disease in animals have largely been
limited to acute occlusive manipulations. Only recently have experi-
mental dietary and hypertensive sclerosis of cerebral vessels with
cerebral hemorrhage (58) been reported in nonhuman primates. A

4—49
genetic strain of stroke-prone, spontaneously-hypertensive rats has
been developed.

Summary of Epidemiological Data

The epidemiological data on cerebrovascular disease (stroke) and
smoking were summarized in the 1976 reference edition of the report
on The Health Consequences of Smoking (138), Table 137 (pp. 64-66).
Kannel reviewed the subject for the Third World Conference on
Smoking and Health (68).

The results of various studies have not been congruent and no
conclusion can be stated with confidence. Kannel has noted that the
prospectively collected data have been difficult to interpret because of
deficiencies, such as small sample numbers, failure to consider
separately cerebral hemorrhage and ischemic infarction, failure to
consider separately men and women, and inadequate classification by
age.

The 1976 report on The Health Consequences of Smoking (138)
comments (on page 152 and in light of its data in Table 7 on page 158,
reproduced below as Table 7) on the possible role of age dependency in
the various studies, noting that cigarette smoking may be a risk factor
for stroke at all ages, but that other causes of stroke may be
proportionately so important in older ages that the smoking risk is
masked by strokes due to other causes in studies that do not involve
very large populations. Although two very large studies, involving
about 250,000 and 1,000,000 respondents, found relative risks of about
1.52 and 1.41 for cigarette smokers (41), no certain conclusion can be
offered at the present time because of apparently conflicting data. A
recent study of a large cohort of women has reported that the risk of
subarachnoid hemorrhage is significantly associated both with ciga-
rette smoking and with the use of oral contraceptives. The risk to
cigarette smokers was 5.7 times that of nonsmokers while it was
increased 6.5 times for users of oral contraceptives. The risk was
increased 22 times among women who both smoked and used oral
contraceptives compared to nonsmokers and nonusers (106).

The Effect of Smoking on Cerebrovascular Disease

It has been noted that risk factor data are inconclusive on the relation
of smoking to the incidence of stroke. Carbon dioxide causes —
cerebrovascular dilatation. Both nicotine and CO increase cerebral
blood flow (123). Unlike the case of cardiac metabolism, there is no
evidence that nicotine affects cerebral oxidative metabolism in a dose
equivalent to smoking. It is uncertain that these effects relate in any
way to stroke. It may be speculated that pathogenetic mechanisms
could operate through effects on blood platelets, oxygen transfer,
emboli from the heart, or through vessel wall toxicity and enhanced
atherogenesis of large and small vessels to the brain. There are no data

4—50
ISP

TABLE 7.—Age-standardized death rates and mortality ratios for cerebral “ascular lesions for men and women,

by type of smoking (lifetime history) and age at start of study

 

 

 

 

 

 

Type of smoking Age grour
45-4 56-64 66-74 1584
CVL death ratea per 100,000 person-years
Men
Never smoked regularly B 92 9 1,858
Pipe, cigar 9 100 Bi} 1371
Cigarette and other 23 129 361 990
Cigarette only a2 1390 477 1,168
Total 35 116 31 1272
Women
Never smoker regularly 18 57 228 1,082
Cigarette 38 88 315 1,277
Total 238 1,01
CVL mortality ratios
Men
Never smoked regularly 1.00 1.00 1.00 1.00
Pipe, cigar 0.39 1.09 1.06 101
Cigarette and other 1.00 140 1.08 0.73
Cigarette only 1 141 137 0.86
Women

Never smoked regularly 1.00 1.00 1.00 1.00
Cigarette 211 154 1.38 118

 

NOTE. - CVL = Cerebral vascular lesions.
SOURCE: U. S. Public Health Service (133).
dealing directly with experimental cerebrovascular disease in animals
and smoking that examine such pathogenetic hypotheses.

Research Needs

Clarification of the existing conflicting epidemiological data may be
sought. It has been suggested by Kannel (68) that a retrospective study
of brain infarctions under the age of 55 years might help to resolve
some uncertainties.

Chronic experimental cerebrovascular disease of hypertensive or
atherosclerotic types in animals has received little attention. Such
disease has recently been produced in nonhuman primates (58). While
its characterization is incomplete, it may possibly offer an opportunity
to study the effects of smoking or of smoke constituents. The effect of
smoke constituents on the stroke-prone rat is also an area for study.

Conclusions

The relationship of smoking to the incidence of stroke is not
established. An association with subarachnoid hemorrhage has been
reported in women.

Peripheral Vascular Disease
The Nature of Peripheral Vascular Disease in Man

Atherosclerotic peripheral vascular disease (PVD) is primarily a
stenosing or occlusive disorder of the arteries of the legs. Other
branches of the aorta such as the subclavian, celiac, or renal arteries
may be diseased similarly, but use applies the term to the arteries that
supply the leg unless noted otherwise. Atherosclerotic involvement
resembles that of the coronary arteries or aorta, but the plaques are
more fibrous and cellular and contain less fat. Involvement includes
not only the large iliac and femoral arteries, but extends to branches in
the anastomotic connections around the knee and to the lesser
branches of the lower leg and foot. Thrombosis is common, and
embolism from ulcerated plaques in the aorta or iliac arteries occurs.
The effect is to create distal circulatory ischemia of a chronic nature
that can be complicated by acute occlusive events. The circulation to
the leg may become inadequate to the needs of the muscles during
exercise. Pain in the calf or thigh is precipitated by exercise, relieved
by rest, and is designated intermittent claudication. It resembles
angina pectoris in these respects and it is often a changeable and
unstable symptom. Severe ischemia will result over time, in some
individuals, in tissue atrophy and necrosis or ischemic gangrene.

The risk factors for atherosclerotic peripheral vascular disease are
generally similar to those for coronary heart disease, but an elevated
blood pressure may be only a minor contributor to risk of PVD (68).

4—52
Peripheral vascular disease has been reported in experimental
dietary atherosclerosis in the nonhuman primate, but the subject has
only recently received systematic study (144).

Summary of Epidemiological Data

Kannel has recently reviewed the data pertaining to occlusive
peripheral vascular disease (68). Several clinical reports find that about
90 percent of individuals with arteriosclerotic obstructive peripheral
vascular disease (PVD) are cigarette smokers. This is a marked excess
of smokers compared to the general or age- and sex-matched
population. Moreover, clinical experience finds that continuation of
smoking worsens prognosis after surgical therapy (157). In one clinical
study of 187 consecutive patients who underwent surgical vascular
grafting with synthetic grafts for arterial occlusive disease of the
lower abdominal aorta and iliac arteries, the patients who continued to
smoke more than a pack a day had three times the graft occlusion rate
of nonsmokers, both in absolute terms and in month-patency time
(113). Koch (75) has reported that cessation of smoking will lead to a
reversion of risk to that of nonsmokers over 5 years. Diabetes is a
strong risk factor for PVD; it acts synergistically with smoking. A
diabetic who smokes is reported to have a 50 percent greater risk of
PVD than one who does not (151). Lawton has reported from a small
series examined by angiography that smoking is associated with
atherosclerotic distortion of the distal aorta and common iliac arteries
in a dose-dependent manner, but not with lesions in the external iliac
or femoral arteries (77).

Epidemiological studies have also demonstrated an association of
PVD with smoking. In one, it was concluded that cigarette smoking
was More common than expected for both sexes among those with
PVD, that it was an independent risk factor, and that 70 percent of
nondiabetic PVD was related to smoking (152). The prospective
Framingham Heart Study reports a strong association between
smoking and obstructive peripheral vascular disease including inter-
mittent claudication (68). At all ages and in both sexes a higher
incidence of claudication was found in smokers. Heavy smokers had a
three times greater incidence and the risk tended to relate directly to
the number of cigarettes smoked. The effect was independent by
multivariate analysis. At any level of other risk factors the smoker is
at greater risk than the nonsmoker. Smoking was found to contribute
as strongly to PVD in women as in men. Data for pipe and cigar
Smoking do not appear to be available.

The Effect of Smoking on Peripheral Vascular Disease

The epidemiological and clinical evidence for smoking as a risk factor
has been noted above. The Framingham data on multiple risk factors
allow the identification of a top decile of risk from which 40 percent of

4—53
cases will emerge (68). Wald, et al. (146) have reported a closer
association between blood COHb in smokers and myocardial infarction,
angina, or intermittent claudication (considered together) than with
smoking history in a survey of Copenhagen workers.

An acute effect of CO on intermittent claudication has been noted by
Aronow, et al. (11). They have reported that patients manifesting
intermittent claudication of the calf or thigh muscles, and angiograph-
ic evidence of iliofemoral arteriosclerosis, who breathed CO to increase
mean venous COHb levels from 1.08 to 2.77 percent, experienced a
decreased exercise threshold to produce leg pain.

Table A30 (pp. 129-130) of the 1976 report on The Health
Consequences of Smoking (138) lists a number of experiments in man in
which the effect of smoking or of nicotine was assessed on some aspect
of the peripheral circulation of the arm or leg. The data are not
consistent, although the tabulated data in normal individuals tend to
show a decrease in skin temperature and a decrease in blood flow. In
another study, calf-blood flow was measured plethysmographically in
51 men, aged 59, who were heavy smokers, but who ceased to smoke
for about 2 months. They showed an increase in blood flow during
reactive hyperemia (62) after the cessation period. No experiments on
animal models of chronic peripheral vascular disease and smoking have
been found.

Research Needs

In general, epidemiological data are adequate. It is likely that current
epidemiological research will provide additional data to furnish more
exact figures than are currently available. New studies appear to be
unnecessary except to establish levels of risk for different “less
hazardous” cigarettes. The possible association of postmenopausal
estrogen treatment, smoking, and PVD in older women may warrant
attention.

However, it is not clear what roles atherogenesis, nicotine, CO, and
perhaps tobacco allergy may play in the development and expression of
PVD in smokers or in its responsiveness to smoking withdrawal.
Studies of the mechanisms responsible for these aspects of smoking
and PVD are warranted and may also have interest for the study of
the pathogenesis of atherosclerosis in general.

Animal studies involving chronic or acute smoking, hypertension,
atherogenesis, and PVD are possible, particularly in nonhuman
primates conditioned to smoke. These may offer a direct, if difficult,
experimental approach to understanding the circulatory effects of
smoking and smoke components on PVD.

Conclusions

Cigarette smoking is a major risk factor for ischemic peripheral
vascular disease of arteriosclerotic type. It increases appreciably the

4-54
risk of peripheral vascular disease in diabetes mellitus. Clinical
experience and case series studies find that cessation of smoking
benefits the prognosis in peripheral vascular disease and is advanta-
geous to its surgical treatment.

Aortic Aneurysm of Atherosclerotic Type
The Nature of Atherosclerotic Aortic Aneurysm

Atherosclerosis involves the abdominal aorta early in life about equally
in males and females. Progression of tne disease in some individuals is
such that large plaques rich in lipid and pultaceous with necrosis
become confluent and encroach upon the media of the vessel, causing
necrosis of its cells and attenuation of the wall. Dilatation of the vessel
and aneurysm formation follows. Thrombosis on the lumenal surface is
common. Eventually the wall may become so thin that leakage and
rupture occur.

Fatal outcome is more common in men than women. The condition
usually becomes clinically apparent after the age of 50 and its
incidence increases with age. It is not known why some individuals
develop this form of progressive disease in the abdominal aorta. An
association with smoking is noted below. The morphological features of
the process are exaggerated but similar to those of atheroma in other
arteries, and it is generally considered that aortic aneurysms of this
type are variants of the general process of atherogenesis. There is a
high concordance with coronary heart disease.

Equivalent atheromatous lesions have not been produced in experi-
mental animals.

Summary of Epidemiological Data

Atherosclerotic aneurysm of the aorta (nonsyphilitic aneurysm) may
cause death by rupture or, occasionally, by thrombotic occlusion. It is
an uncommon cause of death, less than 1 percent of cardiovascular
deaths being attributed to it. Table 29 (p. 67) of the 1976 report on The
Health Consequences of Smoking (138) lists four population studies in
which a total of 947 such deaths are recorded. The two largest
studies—that of Kahn involving more than 248,000 U.S. male veterans,
and that of Hammond and Garfinkel involving approximately 358,000
males—find a dose-dependent mortality ratio such that pack-a-day
male smokers have a ratio of about 4 or 5, while smokers of more than
39 (Kahn) or 40 (Hammond and Garfinkel) cigarettes per day have a
Mortality ratio between 7 and 8 when compared with nonsmokers.
These are unusually large ratios relative to other atherosclerotic
disease. Data permitting multivariate analysis in terms of other
Conventional risk factors are unavailable.

4—55
The Effect of Smoking on Aortic Aneurysm

Aside from the strong risk factor association noted above, nothing
more is known about smoking and aneurysm formation in man. It may
be speculated that CO exposure enhances the circumstances that
promote plaque growth and medial hypoxia, which leads to attenuation
and necrosis of the aorta. It may also be speculated that smoking may
lead to excessive thrombosis, which leads to excessive plaque develop-
ment and aneurysm formation. However, there are no data in men
with aneurysm formation that allow comment on these speculations.

Spontaneous medial calcific arterioslcerosis occurs in the rabbit,
particularly along the thoracic aorta, leading to mild localized
aneurysmal dilatations (55). It has generally not been specifically
reported in relation to smoking or smoke products, although it may
possibly have been observed incidentally in various experiments.
Wanstrup and associates (147) reported the enhancement of such
change with CO exposure. Schievelbein (120) studied the chronic effect
of nicotine in animals (rabbits) liable to develop spontaneous arterio-
sclerosis in the absence of an atherogenic diet. There was no
enhancement of morphological arteriosclerosis by nicotine, but the
aortas of the experimentally treated group contained more calcium,
more free fatty acids, and more lipoprotein lipase. Aneurysmal
differences were not noted.

Research Needs

Atherosclerotic aneurysms of the aorta are uncommon. Study of their
pathogenesis is not likely to be promising in the absence of convenient
animal model analogues. A study of experimental poststenotic
dilatation might illuminate atherogenic processes in relation to
smoking. Research initiatives in this area show little promise at
present.

Conclusions

Cigarette smoking is a strong risk factor for atherosclerotic aortic
aneurysm. The association provides a mortality ratio of about eight
among males who smoke more than about 40 cigarettes a day and a
dose relationship is evident. ‘

High Blood Pressure

The Nature of Hypertension

Many factors are known to be involved in and affect the control of
arterial blood pressure. It is directly dependent on cardiac output and
total peripheral resistance. Some of the factors influencing pressure
include the renin-angiotensin system, aldosterone, catecholamines,

central and peripheral nervous activity, plasma volume, changes in

4—56
vessel elasticity, red cell mass and blood viscosity, sodium metabolism,
obesity, and genetic predisposition. The manner or means by which
most cases of hypertension—essential hypertension—develop is not
understood. The effect, however, is to enhance atherogenesis and
atherosclerotic diseases, particularly heart disease and stroke, and to
shorten life.

Experimental models of hypertension in animals are available for
research. There are both genetic models and those induced by
hormonal and surgical procedures. However, smoke or smoke constitu-
ents have not been assessed in such models.

Summary of Epidemiological Data .

Arterial hypertension is a very common disorder constituting a risk
factor for atherogenesis, stroke, heart attack, heart failure, renal
failure, and retinal damage. Hypertension is a continuous variable and
an independent risk factor.

Although smoking can raise blood pressure acutely, there is no
evidence that smoking induces hypertension. On the contrary, smokers
appear to have, on the average, a slightly lower blood pressure than
nonsmokers. Table A8 (pp. 99-100) of the 1976 report on smoking and
health (138) tabulates several studies; recent reports repeat such data
trends or show little relationship (23, 129).

An exception to these data is the finding of Kahn and associates (67)
in their study of 10,000 Israeli male civil servants. In a period of 5
years, they found that the incidence of hypertension adjusted for age
was about two times greater in smokers than nonsmokers. However,
the conclusion can be considered in additional ways. Since weight gain
is associated with an increase in blood pressure and weight loss is
associated with a decrease in blood pressure and, moreover, since
smokers tend not to gain as much weight as nonsmokers, this complex
relationship has attracted attention. Seltzer (121) has offered data in
which men who stopped smoking gained about 8 pounds and showed an
increase of about 4 mm Hg in systolic blood pressure. In examining the
data for weight change, it was found that continuing smokers who lost
weight had a decrease in systolic blood pressure of about 3 mm Hg,
while quitters who also lost weight had an increase in blood pressure of
about 2 mm Hg. The gradient between these two groups was about 5
mm Hg in systolic blood pressure. The reference report of 1976 on The
Health Consequences of Smoking (138) comments critically on this
report (p. 138ff.), and notes a marginal sample size.

Available data indicate that smoking is not a major risk factor for
hypertension, and in practice, the association is slightly negative. In
this sense, it should be balanced against the other strong positive risk
factor associations of smoking for various expressions of heart attack,
for PVD, aortic aneurysm, lung disease, and cancers.

4—57
Data from several epidemiological studies indicate that, when
hypertension is present, its combination with another risk factor, such
as elevated blood lipids or smoking, is synergistic.

The Effect of Smoking on Blood Pressure

The chronic epidemiological effects of cigarette smoke on the incidence
and level of hypertension and in conjunction with hypertension as an
additional risk factor for cardiovascular disease have been noted above.

The acute and transient effect of smoking in man is to increase heart
rate and blood pressure to a minor degree. These effects are thought to
be due primarily to the action of nicotine releasing cathecholamines. In
the 1976 report on The Health Consequences of Smoking (138), Table
A20 (pp. 103-108) and Table A21 (pp. 109-114) summarize a series of
acute effects of smoking and nicotine on the blood pressure of animals
and humans. Table A22 (p. 115), notes the effects on catecholamines in
humans and animals. Beaumont and colleagues (17) have recently
reported on a paroxysmal arterial hypertension as a reaction to
cigarette smoking in which, under clinical diagnostic testing, a single
high nicotine cigarette induced a rise in blood pressure of about 50 mm
Hg systolic and 20 mm Hg diastolic over about 20 minutes. The
reaction was accompanied by headache, palpatations, and sweating.
The reaction was elicited in 13 of 178 persons tested, all of whom were
moderate to heavy smokers.

Research Needs

It would be of some interest for an understanding of chronic
hypertension to elucidate the pathogenesis of what appears to be a
very mild hypotensive chronic effect of smoking. Since genetic and
induced animal models of hypertension and hypertensive vasculopathy
exist, including stroke-prone spontaneously hypertensive rats, it may
be informative to assess the acute and chronic effects of smoke and
smoke constituents in them.

Conclusions

Cigarette smoking does not induce chronic hypertension. Indeed,
present evidence indicates that it is associated with a mild chronic
hypotensive effect. However, in the presence of hypertension as a risk
factor for coronary heart disease, smoking acts synergistically to
increase the effective risk by joining the risks attributable to
hypertension and to smoking alone.

Other Conditions

Among other conditions of interest are arterial and venous thrombosis,
the synergism of smoking with oral contraceptives in relation to

4—58
myocardial infarction, thromboangiitis obliterans, the effect of
smoking on blood lipids and lipoproteins, and tobacco constituents
other than CO and nicotine.

Venous Thrombosis

Pathological studies in human autopsies that address the question of a
difference in the presence of venous thrombi in relation to smoking
habits have not been reported. On the other hand, epidemiological
studies have clearly shown that conditions such as myocardial
infarction or peripheral vascular disease that are commonly induced or
accompanied pathogenetically by arterial thrombosis are more common
in smokers than nonsmokers. Vessey and Doll (140) reported in a case
control study among 84 women with venous thromboembolism (deep
vein thrombosis or pulmonary embolism) that there were no apprecia-
ble differences in smoking habits of subjects with or without venous
thromboembolism. In the same paper, the authors mention a mortality
study conducted among British doctors and report that among 31 male
deaths from venous thromboembolism over 15 years of observation, the
age-standardized mortality rates per 100,000 were 96 among nonsmok-
ers, 57 among cigarette smokers, and 71 among pipe and cigar smokers.
Lawson and coworkers (76) report the absence of an effect of smoking
on venous thromboembolism among premenopausal women who were
users of oral contraceptives. It has been reported that smokers suffer
less thrombosis of the deep veins of the leg after myocardial infarction
(89, 88). The failure to confirm such a finding has also been published
(57). There have been a number of studies of various aspects of blood
coagulation and platelet pathophysiology in relation to smoking. In
general, these have been acute experimental investigations. Table A27
(pp. 126-1138) of the 1976 report on smoking and health (138) recorded
a number of such studies, including a review by Murphy. The data tend
in the direction of phenomena that might be expected to promote
thrombosis. However, confounding variables are uncertain and the
meaning of in vitro tests for in vivo phenomena of thrombosis is not
established.

From the limited data available, smoking does not appear to enhance
venous thrombotic disease.

The interest in venous thrombosis and smoking lies not only in the
question of the presence or absence of an association but in its possible
meaning for arterial thrombosis. Arterial thrombosis is involved to an
important degree in atherogenesis, and in the precipitation and
complication of heart attack, ischemic stroke, and peripheral vascular
disease. There are research opportunities to learn more about
thrombosis in general and, in particular, in relation to possible
pathogenetic associations with smoking.

4—59
Thromboangiitis Obliterans (Buerger’s Disease)

Buerger’s disease is a relatively rare vascular disease that severely
affects the legs and sometimes affects the arms and other vessels. It is
usually present as a painful ischemic disease of progressive and
subacute type in young male adults. Pathologically, there is a focal
subacute inflammatory phase involving the artery, nerve, and vein
coursing in the limb. The vascular inflammation is accompanied by
arterial and venous thrombosis and local obstruction to the circulation.
A migrating thrombophlebitis is often prominent. Lesions may heal
with vascular sclerosis and new lesions may appear at other sites. The
ultimate outcome is ischemic loss of the limb(s) and when the lesion
extends to other vessels, loss of life. While the disease has been
regarded as a fulminant form of atherosclerosis (153), the more
common view with stronger evidence is that it is a separate disease (87)
and a vasculitis. An infectious etiology (24) has been proposed, as has a
hypersensitivity cause (54). Risk factors such as hypercholesterolemia
or diabetes are not present and coronary heart disease occurs only very
late in the course of the disease.

Smoking has been noted clinically to be strongly associated with
Buerger’s disease (68). Retrospective studies indicate that its occur-
rence among nonsmokers must be very rare. The lesions are compatible
with an angiitis of hypersensitive or immunologic pathogenesis.
Therefore, it has been speculated that hypersensitivity to tobacco
components may be the basis of thromboangiitis obliterans (54). The
evidence for this theory is suggestive but inadequate at present.
Adequate investigations will probably require the use of much purer
tobacco antigens than have been available in the past (19). There is
conceptual interest for the pathogenesis of atherosclerosis in such
investigations that extends beyond thromboangiitis itself since athero-
sclerotic lesions commonly show evidence of a slight inflammatory
component and since a form of coronary atherosclerosis bearing a
remarkable resemblance to advanced plaques in man has been
produced in fat-fed rabbits by immunologic means (93), and also
because a glycoprotein isolated from tobacco leaves has been shown to
activate Factor XII in samples of human plasma, resulting in the
generation of clotting activity, fibrinolytic activity, and kinin activity
(18).

Oral Contraceptives, Smoking, Myocardial Infarction, and
Subarachnoid Hemorrhage Among Women

Extensive population studies have determined that the risk of non-
fatal myocardial infarction among women during child bearing ages is
increased by a factor of about two times by the use of estrogen-
containing oral contraceptives, and that it is increased to about 10
times the expected value when users also smoke (61, 81, 82, 102). A -
recent study reports that oral contraceptive use increases the risk of

4—60
subarachnoid hemorrhage about six times and that the additional use
of cigarettes increases the risk to about 20 times (106).

The mechanisms that may underlie these phenomena in women are
considered elsewhere, but estrogen and estrogen analogue administra-
tion to men with cancer of the prostate or with preexisting myocardial
infarction have been shown to increase the risk of heart attack (30,
141). These reports did not contain information on smoking, however.
While the associations between smoking, oral contraceptive use, and
enhanced risk of cardiovascular disease are not in doubt, research
opportunities exist in seeking explanations for the effect.

The Effect of Smoking on Blood Lipids

The report, The Health Consequences of Smoking of 1976 (138), dealt
with the question of a possible effect of smoking on blood or serum
cholesterol. Acute effects in man and animals were tabulated in Tables
A25 and A25a (pp. 119-124). Case control and population studies are
listed in Table A7 (pp. 94-98). The data are not very uniform, but there
is a preponderance of results in man in which smokers have a
somewhat higher blood cholesterol level than nonsmokers. Paul (103)
has recently presented additional data with this same finding. Dawber
has analyzed the Framingham Heart Study data in terms of pipe,
cigar, and cigarette smoking (33). Since these forms of smoking deliver
different amounts of tar, nicotine, and CO to the smoker, such an
analysis might reflect specific responses on the part of the serum lipids.
No major differences were found. Pipesmokers had average cholesterol
levels of about 216.25 mg, cigar smokers of 220.95 mg, and cigarette
smokers of 224.34 mg (nonsmokers 223.83 mg). These differences are
too small to account for the observed differences in risk associated
with type of smoking habit. There may indeed be a minor tendency for
cigarette smokers to have slightly elevated blood cholesterol levels for
whatever reason, but smoking and cholesterol are clearly established
independent risk factors.

Experimental data based on acute manipulation of smoke exposure
or nicotine appear to show a consistent elevation of free fatty acids in
the blood. Animals exposed to CO and high cholesterol diets have been
reported to develop more hypercholesterolemia than expected, but
confirmation has not been established with whole smoke (14, 136).

Other recent reports have found HDL levels to be a strong and
independent risk factor for coronary heart disease that has an inverse
relationship (49, 92, 94); high levels are protective and low levels are
associated with increased risk. Both in a subset of the Tromso study
(94) and in the Framingham study (49), almost identical HDL
cholesterol levels among smokers and nonsmokers were found; there
Was no significant association between them.

Observations on 10,000 males in Israel show that alpha cholesterol is
depressed among smokers of cigarettes compared to nonsmokers and

4—61
ex-smokers, with the trend persisting in different age groups. The
concentration of alpha cholesterol decreased according to increased
amounts smoked daily when the smokers were grouped as never
having smoked, and having smoked 0 to 10, 11 to 20, and more than 20
cigarettes smoked per day. Total serum cholesterol, and hence beta
cholesterol, were increased in direct relationship to the amount smoked
(48). HDL cholesterol has also been measured among approximately
4,000 men and women who are the adult offspring of the original
Framingham Heart Study cohort. After control for reported alcohol
consumption, subscapular skinfold thickness, and age in multiple
regression analysis, cigarette smoking was found to be associated with
significantly lower HDL levels in both men and women. There was no
evidence of lower HDL cholesterol among former cigarette smokers
(47). In an examination of 447 women and 471 men aged 40 or 41 in
Holland, it has been found that HDL cholesterol is (as expected) higher
in women than in men. Cigarette smoking was associated with a
reduced serum HDL-cholesterol in both men and women. Among the
women there was also a strong negative association with the use of
oral contraceptives that was independent of smoking (4).

Hulley and colleagues (59) have recently reported in a multiple-risk-
factor intervention trial group that over a period of a year the change
in serum thiocyanate (an indirect measure of smoking activity) showed
a univariate regression coefficient, with an HDL cholesterol of -.12
that was significant at less than the 0.05 level. The multivariate
regression coefficient was -.15 and significant at less than 0.01. While
more data should be gathered to ascertain the effect of smoking on
HDL levels, present indications are that, when other factors that also
affect HDL levels are controlled in statistical analysis, cigarette
smoking displays an independent inverse relationship with HDL levels.
Moreover, since total cholesterol levels appear to be slightly elevated
among smokers, lipoprotein cholesterol that is positively atherogenic
will also be increased. Consequently, it can be hypothesized that the
effect of smoking on CHD morbidity and mortality may be to some
degree a reflection of altered lipoprotein metabolism.

Other Constituents of Smoke

Smoke is a remarkably complex mixture of chemical substances and
physical chemical states. Our understanding of the relationships of
nicotine and CO and of whole smoke to cardiovascular disease have
been noted above. Other substances have attracted some investigation
also. Those of possible cardiovascular interest include cadmium, zinc,
chromium, carbon disulphide, carbon dioxide, hydrogen cyanide, oxides
of nitrogen, and polonium-210. McMillan (90) concluded that, while
these substances provide interesting grounds for speculation as to their
possible role in cardiovascular disease, only nicotine and CO offer both
data and rational concepts for a role in smoking and cardiovascular

4—62
disease that command serious attention at the present time. As noted
very briefly above in the section on thromboangiitis and considered in a
separate chapter, hypersensitivity to tobacco protein does offer
reasonable concepts in relation to the pathogenesis of arteriosclerosis,
thrombosis, and angiitis. Its investigation will require more systematic
study and the use of immunologic methods superior to those employed
in the past.

Discussion and Conclusions

The present report on cardiovascular disease and smoking is able to
summarize and to comment on far more extensive and detailed data
than were available 15 years ago. It draws heavily on the 1976
reference report on smoking and health (138) and adds recent
references.

Systematic observations on the associations between smoking and
cardiovascular diseases have been made on considerably more than a
million individuals in the United States alone and have involved many
millions of person-years of experience. The majority of these have been
gathered on men.

Sample sizes are now extensive in both retrospective and prospective
studies. The variables observed in retrospective studies have been
relatively limited; in some prospective studies, they have been more
numerous and have allowed for complex analyses in which the
independence of smoking as a risk factor among other risk factors has
been defined.

The data collected from western countries, particularly the United
States, but also the United Kingdom, Canada, and others, show that
smoking is one of three major independent risk factors for heart attack
manifest as fatal and nonfatal myocardial infarction and sudden
cardiac death in adult men and women. Moreover, the effect is dose
related, synergistic with other risk factors for heart attack, and of
stronger association at younger ages. Based on smaller but still
extensive samples, smoking cigarettes is strongly associated with
increased morbidity from arteriosclerotic peripheral vascular disease
and with death from arteriosclerotic aneurysm of the aorta.

There is no reasonable doubt that cigarette smoking as a risk factor
for these cardiovascular diseases has been proven. Its dimensions as a
risk factor for them have been established for the American public.

Atherosclerosis, the basic lesion of ischemic disease studied at
autopsy, has been observed in restricted samples and limited numbers
of cases. Nevertheless, the data establish adequately that cigarette
smoking is associated with more severe and extensive atherosclerosis
of the aorta and coronary arteries than is found among nonsmokers.
The effect is related to the amount smoked. Existing autopsy data
have not allowed adequate multivariate analysis, but several prospec-

4—63
tive studies have now collected sufficient standard risk factor data,
including smoking information and autopsy findings, to report
preliminary multivariate analyses. While more data might be desirable
in order to establish better the dimensions of effect as seen at autopsy,
and more data are needed to extend multivariate analyses, there is no
reasonable doubt that cigarette smoking enhances atherogenesis. This
knowledge establishes a fundamental rationale for the findings on the
incidence of heart attack, including sudden cardiac death, aortic
aneurysm, and peripheral vascular disease in relation to smoking. It is
somewhat uncertain, but likely, that smoking has an adverse effect on
the recurrence of heart attack among survivors of a prior myocardial
infarction.

On the other hand, epidemiologic data on the association between
cigarette smoking and angina pectoris and cerebrovascular disease
manifested as stroke are not conclusive. There are major and
unresolved inconsistencies between existing reports. While certain
reports on these diseases may have more technical strength than others
and thus provide more credible conclusions, a basis for drawing final
conclusions is not established in these two conditions. It is of interest
that, in acute experiments on atherosclerotic patients with angina
pectoris or with the intermittent claudication of peripheral vascular
disease, smoking or exposure to carbon monoxide reduces the patients’
established threshold for the precipitation of angina or claudication.

There is no apparent relationship between smoking and the
incidence of hypertension. Available evidence indicates a neutral or
slight hypotensive effect. Nevertheless, in the presence of hyperten-
sion, smoking joins with hypertension to affect the patient with the
cardiovascular burden of both risk factors.

There are opportunities for further epidemiological research into
smoking as a risk factor for cardiovascular disease; these have been
detailed in each of the foregoing sections. The need and priority of such
research should be debated in specific cases. It can be argued that little
public health or medical therapeutic advantage would arise from a
clarification of the relationship of smoking to angina or cerebrovascu-
lar disease in the face of the existing conclusive evidence of its adverse
effect on the incidence of heart attack and lung diseases and the
benefits of smoking avoidance or cessation. On the other hand, it could
be of some medical value to learn more accurately what the association
may be for second heart attacks. It would be of great interest for
preventive medicine to know whether smoking affects the severity of
atherosclerosis of the aorta and coronary arteries in childhood and
adolescence and the premature development of adult forms of lesions
in youth. It would also be of great interest to learn whether present-
day cigarettes modified to deliver less tar and nicotine are less
hazardous for cardiovascular health. Earlier data, which no longer
represent current products, found that low tar and nicotine cigarettes

4—64
carried less risk than high tar and nicotine ones but that they also bore
a considerably greater risk than not smoking.

Relatively little is known about the mechanisms by which smoking
enhances atherogenesis or increases the risk of heart attack. This
ignorance in no way weakens the force of the information noted above;
nevertheless, better insight into the pathogenesis of these effects
would be of potential value in designing less hazardous cigarettes or in
attempting otherwise to limit the hazard of smoking. Moreover, it is
likely that there would be an appreciable gain of information about
basic processes of atherogenesis, thrombosis, cardiac metabolism and
ischemia, and cardiac rhythmicity and ectopic electrical activity. Some
experiments can be done acutely in man; many can be done in animal
models with smoke constituents. Chronic or acute experiments in
nonhuman primates with natural or modified whole smoke taken by
inhalation in a humanlike nonaversive manner of smoking now appear
possible. It should be emphasized that a number of strong concepts
exist in atherogenesis, thrombosis, and cardiac structure and function
within which to mount appropriate experiments.

Data on the epidemiological relationships between smoking and
heart attack, peripheral vascular disease, aortic aneurysm, and
arteriosclerosis noted above have been assembled in a manner to allow
a statistical statement of the nature of the correlations between
cigarette smoking and cardiovascular disease. Correlation is not
synonymous with causation. It is important for the public to
understand the nature or character of the associations that have been
found. The characteristics are fully established for heart attack and
include the fact that the correlations are strong ones, generally having
a relative risk of two or more. They are consistent, reappearing in
different population samples over and over, and they are independent
of other major risk factors. There is also a graded relationship;
smoking is an antecedent event in time and the cessation of smoking is
followed by a reduction in risk over time; the association has strong
Predictive capacity in the same population sample and also when
applied to other samples. Within the limits of the research that has
been done, the findings of epidemiology, clinical investigation, and
pathology are generally congruent. The results from the various
disciplines and techniques of study tend to support each other.
Although there are reports which do not confirm the statements made
above, they constitute a minor part of the data and fail to cast
reasonable doubt. Animal experimentation is not yet well developed in
smoking research in relation to cardiovascular disease.

Smoking is not a necessary condition for atherosclerosis and heart
attack since these occur in nonsmokers. Repeated and very extensive
€xperience has found, however, that it is a sufficient condition to
increase the mortality from heart attack among the category of people
who smoke and that it does so in a predictable way.

4—65
Given the characteristics of its associations with heart attack (such
as strength, graded relationship, independence, consistency, antece-
dence, loss of relationship on withdrawal, predictive capability, and a
degree of coherence), it can be concluded that smoking is causally
related to coronary heart disease in the common sense of that idea and
for the purposes of preventive medicine. It may be argued that the
characteristics of the associations noted above would occur if people
who were constitutionally liable to heart attack were also constitution-
ally liable to smoke; that is, that smoking activity and susceptibility to
atherosclerotic heart disease were both due to some underlying
constitutional condition of the individual. An attempt has been made to
study this point by observing large numbers of monozygotic and
dizygotic twins. The result has been inconclusive. A discussion of
references will be found in the 1976 report on The Health Consequences
of Smoking (p. 44ff.) (188). It should be noted, however, that the fact
that risk in smokers reverts to normal or nonsmokers’ levels after they
cease to smoke is contrary to the constitutional concept as expressed
above, unless further complex assumptions are made and it is assumed
that large numbers of individuals underwent a change in their
underlying constitutional factor in midlife, acquired low risk, and
ceased to smoke because of that new constitution. This is not to say
that genetic susceptibility or resistance may not also be a risk factor
that plays a role in the individual expression of or resistance to disease
along with other risk factors, or that people who stop smoking may not
also adopt additional health-oriented behaviors when they stop; but the
constitutional hypothesis as expressed above does not provide a
credible basis to doubt that cigarette smoking is a cause of coronary
heart disease.

From the point of view of cardiovascular disease, reséarch on the
mechanisms whereby smoking causes its adverse effects and a more
precise quantification of certain risk factors through epidemiological
studies are significant topics of medical science. The major goal in

smoking and cardiovascular disease research is, however, the develop- ~ -

ment of long-term effective methods of smoking avoidance and
cessation.
Cardiovascular Diseases: References

@)

(2)

(3)

(4)

(5)

(6)

(7)

(8)
(9)
(10)

(i)
(12)

(13)

(14)

(15)

(16)

ALBERT, R.E., VANDERLAAN, M., BURNS, F.J., NISHIZUMI, M. Effect of
carcinogens on chicken atherosclerosis. Cancer Research 37(7): 2232-2235, July
1977.

ARMITAGE, A.K., DAVIES RF., TURNER D.M. The effects of carbon
monoxide on the development of atherosclerosis in the White Carneau pigeon.
Atherosclerosis 23(2): 833-844, March/April 1976.

ARMSTRONG, M.L. Regression of Atherosclerosis. In: Paoletti, R., Gotto, A.M.
Jr. (Editors). Atherosclerosis Reviews, Volume 1, New York, Raven Press,
1976, pp. 137-182.

ARNTZENIUS, A.C., VAN GENT, C.M., VAN DER VOORT, H., STEGER-
HOEK, C.1., STYBLAO, K. Reduced high-density lipoprotein in women aged 40-
41 using oral contraceptives. Consultation Bureau Heart Project. Lancet: 1221-
1223, June 10, 1978.

ARONOW, W.S. Carbon monoxide and cardiovascular disease. In: Wynder, E.L.,
Hoffmann, D., Gori, G.B. (Editors). Proceedings of the Third World Confer-
ence on Smoking and Health, New York, June 2-5, 1975. Volume I. Modifying
the Risk for the Smoker. U.S. Department of Health, Education, and Welfare,
Public Health Service, National Institutes of Health, National Cancer
Institute, DHEW Publication No. (NIH) 76-1221, 1976, pp. 321-328.

ARONOW, W.S. Introduction to smoking and cardiovascular disease. In:
Wynder, E.L., Hoffmann, D., Gori, G.B. (Editors). Proceedings of the Third
World Conference on Smoking and Health, New York, June 2-5, 1975. Volume
I, Modifying the Risk for the Smoker. U.S. Department of Health, Education,
and Welfare, Public Health Service, National Institutes of Health, National
Cancer Institute, DHEW Publication No. (NIH) 76-1221, 1976, pp. 231-236.

ARONOW, W.S., CASSIDY, J., VANGROW, J.S. MARCH, H., KERN, J.C.,
GOLDSMITH, J.R., KHEMKA, M., PAGANO, J., VAWTER, M. Effect of
cigarette smoking and breathing carbon monoxide on cardiovascular hemody-
namics in anginal patients. Circulation 50(2): 340-347, August 1974.

ARONOW, W.S., ISBELL, M.W. Carbon monoxide effect on exercise-induced
angina pectoris. Annals of Internal Medicine 73): 392-895, September 1973.

ARONOW, W.S., KAPLAN, M.A., J ACOB, D. Tobacco: A precipitating factor in
angina pectoris. Annals of Internal Medicine 6X3): 529-536, September 1968.

ARONOW, W.S., ROKAW, S.N. Carboxyhemoglobin caused by smoking non-
nicotine cigarettes. Effects in angina pectoris. Circulation 44: 782-788,
November 1971.

ARONOW, W.S., STEMMER, E.A., ISBELL, M.W. Effect of carbon monoxide
exposure on intermittent claudication. Circulation 49: 415-417, March 1974.

ARONOW, W.S., SWANSON, AJ. The effect of low-nicotine cigarettes on
angina pectoris. Annals of Internal Medicine 71(3): 599-601, September 1969.

ASTRUP, P. Some physiological and pathological effects of moderate carbon
monoxide exposure. British Medical Journal 4(5838): 447-452, November 25,
1972.

ASTRUP, P., KJELDSEN, K., WANSTRUP, J. Enhancing influence of carbon
monoxide on the development of atheromatosis in cholesterol-fed rabbits.
Journal of Atherosclerosis Research 7: 343-354, 1967.

AUERBACH, 0O., CARTER, H.W., GARFINKEL, L., HAMMOND, EC.
Cigarette smoking and coronary artery disease: A macroscopic and microscopic
study. Chest 70(6): 697-705, December, 1976.

AUERBACH, 0., HAMMOND, E.C., GARFINKEL, L., KIRMAN, D. Thickness
of walls of myocardial arterioles in relation to smoking and age: Findings in
men and dogs. Archives of Environmental Health 22: 20-27, January 1971.

4—67
(17)

(18)

(19)

(20)

(21)

(22)

(28)

(24)

(25)

(26)

(27)

(28)

(29)

($0)

(81)

(82)

BEAUMONT, J.-L., BUXTORF, J.-C., JACOTOT, B., JAN, F. Les accidents
hypertensifs paroxystiques observes chez les fumeurs (Incidence of paroxys-
mal hypertension in smokers). Concours Medical 98(17): 2611-2612, 2615-2617,
April 24, 1976.

BECKER, C.G., DUBIN, T. Activation of factor XII by tobacco glycoprotein.
Journal of Experimental, Medicine 146: 457-467, 1977.

BECKER, C.G., DUBIN, T., WIEDEMANN, H.P. Hypersensitivity to tobacco
antigen. Proceedings of the National Academy of Sciences of the U.S.A. 735):
1712-1716, May 1976.

BELLET, S., DEGUZMAN, N.T., KOSTIS, J.B., ROMAN, L., FLEISCHMANN,
D. The effect of inhalation of cigarette smoke on ventricular fibrillation
threshold in normal dogs and dogs with acute myocardial infarction. American
Heart Journal 83(1): 67-76, January 1972.

BENDITT, E.P. The monoclonal hypothesis, which holds that the proliferating
cells of an atherosclerotic plaque all stem from one mutated cell, suggests new
lines of research on the causes of coronary disease. Scientific American: 74-85,
February 1977.

BENDITT, E.P., BENDITT, J ‘M. Evidence for a monoclonal origin of human
atherosclerotic plaques. Proceedings of the National Academy of Sciences of
the U.S.A 70(6): 1753-1756, June 1978.

BERGLUND, G., WILHELMSEN, L. Factors related to blood pressure in a
general population sample of Swedish men. Acta Medica Scandinavica 198(4):
291-298, October 1975.

BERNARD, J.G., MOREL, E. Peripheral arteritis—infectious aetiology. In:
Shimamoto, T., Numano, F., Addison, G.M. (Editors). Atherogenesis, Volume
Il. Proceedings of the Second International Symposium on Atherogenesis,
Thrombogenesis and Pyridinolcarbamate Treatment, Tokyo, May 18-20, 1972.
Amsterdam, Excerpta Medica, 1973, pp. 325-329.

BIRNSTINGL, M.A., BRINSON, K., CHAKRABARTI, B.K. The effect of
short-term exposure to carbon monoxide on platelet stickiness. British Journal
of Surgery 58(11): 887-839, November 1971.

BRITISH MEDICAL JOURNAL. Cigarette smoking and chest pain. British
Medical Journal 4(5993): 368, November 15, 1975.

CASTLEDEN, C.M., COLE, P.V. Carboxyhaemoglobin levels of smokers and
nonsmokers working in the City of London. British Journal of Industrial
Medicine 32: 115-118, 1975.

CORDAY, E., DODGE, H.T., (Guest Editors). Symposium on Identification and
Management of the Candidate for Sudden Cardiac Death. The American
Journal of Cardiology 39(6): 813-815, May 26, 1977.

THE CORONARY DRUG PROJECT RESEARCH GROUP. Cigarette smoking
as a risk factor in men with a prior history of myocardial infarction. Journal of
Chronic Diseases, 15 pp. (to be published)

THE CORONARY DRUG PROJECT RESEARCH GROUP. The Coronary Drug
Project: Initial findings leading to modifications of its research protocol.
Journal of the American Medical Assocation 214(7): 1803-1313, November 16,
1970.

COUNCIL ON CEREBROVASCULAR DISEASE. Risk factors in stroke due to
cerebral infarction. A statement for physicians prepared by a subcommittee
and approved by the executive committee of the Council on Cerebrovascular
Diseases of the American Heart Association. Stroke 2(5): 423-428, September-
October 1971.

DAVIES, R.F., TOPPING, D.L., TURNER, D.M. The effect of intermittent
carbon monoxide exposure on experimental atherosclerosis in the rabbit.
Atherosclerosis 24: 527-536, September 1976.
(33)

(34)

(35)

(36)

(87)

(38)

(39)
(40)

(41)

(42)
(48)
(44)

(45)

DAWBER, T.R. The interrelationship of tobacco smoke components to
hyperlipidemia and other risk factors. In: Wynder, E.L., Hoffmann, D., Gori,
G.B., (Editors). Proceedings of the Third World Conference on Smoking and
Health, New York, June 2-5, 1975. Volume I. Modifying the Risk for the
Smoker. U.S. Department of Health, Education, and Welfare, Public Health
Service, National Institutes of Health, National Cancer Institute, DHEW
Publication No. (NIH) 76-1221, 1976, pp. 285-292.

DEBIAS, D.A., BANERJEE, C.M., BIRKHEAD, N.C., GREENE, C.H., SCOTT,
8.D., HARRER, W.V. Effects of carbon monoxide inhalation on ventricular
fibrillation. Archives of Environmental Health 31(1): 42-46, January-February
1976.

DOLL, R., PETO, R. Mortality in relation to smoking: 20 years’ observations on
male British doctors. British Medical Journal (6051): 1525-1536, December 25,
1976.

DOWNEY, H.F., BASHOUR, C.A., BOUTROS, LS., BASHOUR, F.A., PAR-
KER, P.E. Regional myocardial blood flow during nicotine infusion: Effects of
beta adrenergic blockade and acute coronary artery occlusion. Journal of
Pharmacology and Experimental Therapeutics 202(1): 55-68, July 1977.

DOYLE, J.T., DAWBER, T.R., KANNEL, W.B., KINCH, S.H., KAHN, H.A.
The relationship of cigarette smoking to coronary heart disease. The second
report of the combined experience of the Albany, New York, and Framing-
ham, Massachusetts, studies. Journal of the American Medical Association
190(10): 886-890, December 7, 1964.

DOYLE, J.T., KANNEL, W.B., MCNAMARA, P.M., QUICKENTON, P.,
GORDON, T. Factors related to suddenness of death from coronary disease:
combined Albany-Framingham studies. American Journal of Cardiology 37(6):
1073-1078, June 1976.

EMERSON, P.A., MARKS, P. Preventing thromboembolism after myocardial
infarction: Effect of low-dose heparin or smoking. British Medical Journal
1(6052): 18-20, January 1, 1977.

THE EPIDEMIOLOGY STUDY GROUP. Epidemiology for stroke facilities
planning. Report of the Joint Committee for Stroke Facilities. Stroke 3: 360-
371, May-June 1972.

FEINLEIB, M., WILLIAMS, R.R. Relative risks of myocardial infarction,
cardiovascular disease and peripheral vascular disease by type of smoking. In:
Wynder, E.C., Hoffmann, D., Gori, G.B., (Editors). Proceedings of the Third
World Conference on Smoking and Health, New York, June 2-5, 1975. Volume
I. Modifying the Risk for the Smoker. U.S. Department of Health, Education,
and Welfare, Public Health Service, National Institutes of Health, National
Cancer Institute, DHEW Publication No. (NIH) 76-1221, 1976, pp. 243-256.

FISHER, E.R., ROTHSTEIN, R, WHOLEY, M.H., NELSON, R. Influence of
nicotine on experimental atherosclerosis and its determinants. Archives of
Pathology 96: 298-304, November 1973.

FIVE-YEAR STUDY BY A GROUP OF PHYSICIANS OF THE NEWCAS-
TLE UPON TYNE REGION. Trial of clofibrate in the treatment of ischaemic
heart disease. British Medical Journal 4: 767-776, December 25, 1971.

FRIEDMAN, G.D., KLATSKY, A.L., SIEGELAUB, A.B. Predictors of sudden
cardiac death. Circulation (Supplement III to Volumes 51 and 52): IIJ-164—
TII-169, December 1975.

FRIEDMAN, G.D., SIEGELAUB, A.B., DALES, L.G. Cigarette smoking and
chest pain. Annals of Internal Medicine 83(1): 1-7, July 1975.

4—69