topeesp.gif (5672 bytes)

Scientific Activities - Actividades Científicas

Honorary Committee Lecture

Passive smoking - cardiovascular disease
Published in ProCOR

Bernard Lown, MD
Awarded with Nobel Prize

Passive smoking: cardiovascular disease Experimental human studies
The epidemiological data Final comments
Effect of ETS on the arterial wall


Passive smoking: cardiovascular disease

An advertisement against smoking phrases the message starkly, "There's a warning label on cigarette packs for people who smoke. Where should the warning go for people who breathe?"(22) In fact breathing in certain surrounds may be hazardous to one's health.. In the United States around 53,000 annual fatalities are ascribed to environmental tobacco smoke. A majority of these deaths are due to ischemic heart disease. The evidence is largely epidemiological and therefore susceptible to confounding by some unappreciated variable. As the association is weak, extraordinary large sample size is essential to achieve adequate statistical power. However the plausibility that second hand smoke contributes to the toll of ischemic heart disease gains credibility and confounding appears more remote as converging results are recorded in studies with different populations, in diverse geographic locations and disparate historical periods. (23) Reinforcing this association is the indirect evidence from animal models and clinical investigations relating to causal pathways.

The epidemiological data

Glantz and Parmley (24) reviewed 14 epidemiologic studies and found a consistent increased risk of heart disease among those exposed to second hand smoke. The risk ratio for both men and women was 1.2.  In a recently reported meta-analysis, Law et al (25) examined the impact of ETS (ETS: environmental tobacco smoke)on the incidence of ischemic heart disease among lifelong non-smokers who lived with smokers.  In these 19 studies the end points examined were non-fatal myocardial infarction and death from ischemic heart disease. The estimated relative risk of ischemic heart disease in never smokers whose spouses smoked relative to the risk of those whose spouses never smoked was 1.3 (95% confidence interval 1.22 to 1.38 P<0.001). The findings were similar irrespective of gender where all ischemic events among passive smokers manifested as death; in contrast, in the control population, the majority of endpoints were non fatal infarcts. The risk of ischemic heart disease in the non-smoker increased continuously with the daily cigarette consumption by the smoking spouse. In these studies the relative risk for heart disease was similar to that for lung cancer. Because of the far higher prevalence of CHD, ETS, however, accounts for many more clinical events and deaths than its health effects on other organs.

Law et al (25) also examined various confounders especially the eating of fruit and vegetables, since their intake is reduced among  people exposed to environmental tobacco smoke.  Fruit and vegetables contain folic acid, potassium, linoleic acid that protect against CHD.  In these populations, the estimated excess risk for ischemic heart disease in relation to dietary differences did not exceed 3%.  There was no substantial difference in the plasma cholesterol between smokers and nonsmokers, similarly there was no substantial difference in blood pressure, though smokers had a smaller body mass index.   The fact that passive smoking also exerts deleterious cardiac effects in the work place tends to confirm that diet is not a significant confounder.  A meta analysis of studies of occupational exposure to environmental tobacco smoke found a nearly similar increased relative risk of 1.36 for heart disease among passive smokers..(26) People in the work place do not share  identical food intakes, therefore a dietary explanation for ETS predisposing to CHD is untenable.

The Harvard Nurses' Health Study recently provided important data relating passive smoking to heart disease. (27) The design and large size of this prospective investigation circumvents the problem of detecting small ETS effects on cardiovascular disease risk. In this same population, the Harvard group reported earlier, that after 12 year follow-up, those who were smoking at baseline had a 3.7-fold increase in risk of cardiovascular mortality (95% CI, 2.9 to 4.9). A striking dose response effect was observed in these women, with a relative risk of cardiovascular death of 5.6 (3.5 to 9.0) for heavy smokers and 2.7 (1.8 to 4.0) for those smoking 1 to 14 cigarettes daily.  Relevant to the issue of passive smoking is the fact that even light smokers experienced at least a doubling of risk.

After statistical adjustment for a number of potential confounding factors in the nurses study, a relative risk of 1.9 (1.1 to 3.3) was observed for nonfatal and fatal CHD events among nonsmoking women with regular exposure to ETS at home or at work. This estimate of increased risk due to ETS is far larger than reported in most other studies of women.  The difference may be due to chance, but more likely it is due to the selection of a cohort in which the background determinants of CHD (other than tobacco smoke exposure) were limited thus resulting in a stronger relative risk and reduced likelihood of spuriousness due to confounding.  The observation that the relative risk increased across levels of ETS exposure (occasional and regular) lends support to a causal interpretation of the association.

Effect of ETS on the arterial wall

The validity of deleterious cardiovascular effects of ETS is reinforced by the changes in the arterial wall.  Cigarette smoking has been shown to be one of the strongest predictors of plaque progression.(28)  Such progression has also been demonstrated among passive smokers.

A significant breakthrough in the study of atherosclerosis has been the use of B-mode ultrasound for measuring arterial intimal-medial thickness (IMT) as a surrogate endpoint for the degree of atherosclerosis and its progression.  In one of the largest such studies reported by Howard and coworkers, carotid artery IMT was followed over a three year period in black and white men and women,  aged 45 to 65 years recruited for the Atherosclerosis Risk in Communities Study (ARIC).(29)  The  10914 participants were enrolled between 1987-1989.  Included in this population were 4298 exposed to ETS either who had been past smokers or who had never  smoked.  The results were striking. Compared to non-smokers never exposed to second hand smoking, cigarette smokers had a 50% increase in IMT, while the increase was  20% for those exposed to ETS. (See Table)


   N N+E  P-E P+E C
Numbers 2316 2449 1344 1849 2956
increase (um) 26.9 32.7 33.4 39.8 41.2
N: Non-smoker without ETS exposure
N+E: Non-smoker with ETS exposure
P-E: Past smoker without ETS exposure
P+E: Past smoker with ETS exposure
C: Current smoker

There was a gradient in IMT from never smok-ers not exposed to ETS, never smokers exposed to ETS, past smokers, and current smokers. The ETS effect on IMT persisted after control for diet, physical activity, body mass index, al-cohol intake, education, and major cardiovascular risk factors. Among nonsmoking men exposed to ETS, there was a significant increase in IMT with increasing number of hours per week of ETS exposure. Astounding is the degree of progression of atherosclerosis due to ETS with an increase in IMT of 5.9um, representing 34% of the change of 17.1 um, recorded for current smokers. A still greater impact from smoking and passive smoking was observed among diabetics and those with hypertension. 

A critical question is whether these ultrasound recorded IMT changes are valid surrogate markers for progression of atherosclerosis.(30)   Existing evidence substantiates that carotid artery IMT measurement correlates with prevalence of atherosclerosis in other arteries.  Two recent studies address the relation between increases in IMT and the progression of cardiovascular disease.  In over 1000 Finnish men, Salonen and Salonen (31) observed that for each millimeter increase in carotid artery IMT, increased the risk of an acute coronary events 2.14-fold.   A similar finding was reported by Bots et al. (32) For every 0.163mm increase in carotid artery IMT, the risk of acute myocardial infarction rose by 43%. One may therefore infer that changes in IMT induced by ETS indicate progression of arterial atherosclerosis including the coronary arteries.

The actual mechanism respon-sible for the arterial damage is not known, but may be related to the effects of tobacco smoke on interactions between platelets and the vessel wall or as yet undefined oxidation products or atherogenic lipid components that change with long- term exposure to smoke.  Indeed tobacco smoke enhances platelet aggregability. This is supported by the observation that aspirin abolishes the effect of tobacco smoke on platelet aggregation and reduces the risk of ischemic heart disease.   The toxic substance or substances in tobacco appear to be present in both environmental as well as inhaled cigarette smoke.

Experimental human studies

As endothelial dysfunction is considered to be an early event in the pathogenesis of the arterial plaque, the role of ETS in affecting this progenitor change has been studied in both animals and humans. Celermayer and coworkers (33) investigated this issue in 78 subjects (39 male and 39 female) aged 15 to 30 years. Us-ing ultrasonography, they measured the brachial-artery diameter under base-line conditions, during reactive hyperemia, which promotes increased flow through endothelium mediated dilatation and after adminis-tering sublingual nitroglycerin which induces dilatation independent of endothelium.  The study population consisted of 26 subject in each of three groups: control subjects who had never smoked and had no regular exposure to ETS, those who had never smoked but had been exposed to ETS for at least one hour dai-ly for three or more years, and a group of active smokers.  Passive smoking markedly impaired flow mediated brachial artery dilatation without affecting the changes induced by nitroglycerin. In non-smoking controls hyperemia increased brachial flow by 8.2%. However, during passive smoking the flow increase was attenuated to only 3.1% (P<0.001).  The impaired endothelium-dependent dila-tation was related to the dose of passive smoking  (r = -0.67, P<0.001).  The authors suggest that ETS, by affecting endothelial function, may damage arterial intima even in young healthy adults.

Adverse cardiac effects of ETS may also be brought about by curtailing aortic distensibility.  Stefanidis and coworkers (34) investigated aortic elasticity in  48 male patients most of whom had coronary artery disease. By means of a special sonometric catheter they measured the diameter of the aorta while simultaneously determining arterial pressure at the very same locus.  Passive smoking significantly reduced aortic elasticity. Such stiffening of the aorta augments left ventricular afterload thereby adversely affecting ventricular function.   As tobacco induces coronary vasoconstriction, (35,36), a stiffer aorta would necessarily worsen myocardial ischemia in those with coronary disease and among those with hypertension  whose aortic  distensibility is already compromised.

ETS exposure may also affect the athrosclerotic process by altering blood lipids.  Highly relevant in this regard is the investigation by  Neufield and colleagues(37) of the effects of passive smoking  in children with dislipedemia  who had low  HDL fractions. Children of parents who smoked evidenced a 11.2% (4.9 mg/dL) lower HDL cholesterol level compared to children of nonsmokers. Adjustment for potential confounders attenuated this difference between groups to 8.5% (3.7 mg/dL).   This effect of ETS is greater than previously reported in children. The large difference in HDL cholesterol observed in the Neufeld study may, therefore, be a chance finding.   On the other hand, it may indicate that ETS is more harmful for children with pre-existing lipid abnormalities.

In addition to lowering HDL cholesterol, damaging arterial endothelium, and increasing platelet aggregation, ETS raises blood concentration of carboxyhemoglobin (COHb). According to Glantz and Parmley (24) elevated COHb levels reduces exercise performance in coronary patients as well as in normal individuals, adversely affect platelet function and damage the arterial endothelium.  Thus ETS exerts both short-term effects by inducing coronary vasoconstriction and platelet aggregation and long term effects via endothelial damage and atherosclerotic plaque formation. 

Additional support indicting ETS also derive from animal experiments. Passive smoking accelerates atherosclerosis in rabbits and cockerels.(38) Expos-ing cockerels to levels of environmental tobacco smoke routinely encountered by people in smoke-filled envi-ronments increases  the size of atheromatous  plaques in the aorta.(39)   Passive smoking worsens endothelial function in lipid fed rabbits leading to inappropriate vasoconstriction. (40) Dietary supplementation of L-arginine, precursor of nitric oxide or endothelium-derived relaxing fac-tor, protects cholesterol-fed rabbits from the endotheli-al dysfunction associated with exposure to ETS. (41)  This finding  strengthens the argument for a critical role for impaired endothelial production of nitric oxide in augmenting the effect of ETS on the progression of atherosclerosis.

One problem with the epidemiologic data on which the essential concept of passive smoking is largely based relates to the seemingly large effect of ETS on heart disease compared to the effect of active smoking. The risk for heart disease of smokers compared to non-smokers is about 1.7 (12) Most studies on passive smoking provide a relative risk of the order of 1.2 to 1-3 (11) This is far more than anticipated on the basis of urinary conitine excretion by passive smokers, which is only 0.7% of that found in smokers.   However, sidestream and mainstream smoke differ. Many toxic constituents, such as carbon monoxide and benzopyrene, are found in higher concentrations in sidestream than in inhaled smoke.

Final comments

Hitherto smoking has been regarded as an individual evil, this is no longer tenable in view of the persuasive body of evidence that it injures all those within reach of the wafting of the smoke.  In the United States it is estimated that of annual death 480,000 are smoking related, 11% of these fatalities are due to ETS and more than two thirds die from heart disease. (24) Passive smoking is the third leading cause of preventable death after active smoking and alcohol use. (24)

Physicians need to incorporate in their history gathering possible exposures to ETS, either at home or in the work place, as an important risk factor for cardiovascular disease.  As Werner and Pearson have commented "not much is passive about 'passive smoke.'"(42) The response of the tobacco industry to this newly discovered and horrendous disease burden has been denial and a barrage of misleading double talk.  They even went as far as comparing the hazard of second hand smoke with the risks of drinking milk and eating biscuits. (43)



22.  Advertisement by the Public Health Commission, Boston.  Boston Globe 1998; March 22.
23.  Dwyer JH. Exposure to environmental tobacco smoke and coronary risk. Circulation Editorial 1997;96:1367.
24.  Glantz SA, Parmley WW Passive smoking and heart disease epidemiology, physiology and biochemistry. Circulation 1991;83:1-12.
25.  Law MR, Morris JK, Wald NJ. Environmental tobacco smoke exposure and ischemic heart disease; an evaluation of the evidence. BMJ 1997;313:973.
26.  Wells AJ Post-bearing comment re OSHA hearing meta-analysis of relative risk of heart disease from ETS exposure in the workplace. Washington DC: USDepartment of Labor. 1995 (Docket H-122. Docket Office Room N-2625.
27.  Kawachi I, Codlitz GA, Speizer FE, et al.  A prospective study of passive smoking and coronary heart disease Circulation 1997;95:2374-2379. 28.  Howard G, Burke GL, Szklo M, et al.  Active and passive smoking are associated with increased carotid wall thickness Arch Intern Med 1994;154:1277-1282.
29.  Howard G, Wagenknecht LE, BurkeGL et al Cigarette smoking and progressio of atherosclerosis; The atherosclerosis risk in communities (ARIC) study JAMA 1998;279:119-124.
30.  Werner RM, Pearson TA. What is so passive about passive smoking? Secondhand smoke as a cause of atherosclerotic disease. JAMA  Editorial 1998;279:157-58.
31.  Salonen JT, Salonen R. Ultrasonography assessed carotid morphology and risk of coronary heart disease. Arterioscler Thromb Vasc Biol 1991;11:1245-49
32.  Bots ML, Hoes AW, Koudstaal PJ et al.  Common carotid intima-media thickness and risk of stroke and myocardial infarction: the Rotterdam Study. Circulation 1997;96:1432-37.
33.   Celermajer DS, Adams MR, Clarkson P, et al.  Passive smoking and impaired endothelium-dependent arterial dilatation in healthy young adults.  N EngI J Med 1996;334:150-4.
34.   Stefanidis C, Vlachopoulos C, Tsiamis E et al.  Unfavorable effects of passive smoking in aortic function in men. Ann Intern Med 1998;128:426-434.
35.   Moreyra  AE, Lacy CR, Wilson AC et al. Arterial blood nicotine concentration and and coronary vasocosntrivtive effect of low-nicotine cigerette smoking. Am Ht J   1992;124:392-7.
36.  Quillen  JE, Rosen JD, Oskarson HJ, et al. Acute effect of cigarette smoking on the coronary circulation constriction of epicardial and resistance vessels. J Am Coll Cardiol 1993;22:642-7.
37.  Neufeld EJ, Mietus-Snyder MBeiser AS, et al.  Passive cigarette smoking and reduced HDL cholesterol levels in children with high-risk lipid profiles. Circulation 1997;96:1403-07.
38.  Zhu B-Q,  et al. Passive smoking increases experimental atherosclerosis in cholesterol fed rabbits. J Am Coll Cardiol 1993;21:225.
39.  Penn A et al Butadiene, a vapor phasecomponent of environmental tobacco smoke, accelerates atherosclerotic plaque development. Circulation 1996;93:552P (Abstract).
40.  Hutchenson S. et al L-arginine restores normal endothelium: mediated relaxation in hypercholesterolemic rabbits exposed to environmental tobacco smoke J. Am Coll Cardiol 1996;39A. (Abstract).
41.  Sun Y-P,Zhu B-Q,Sievers RE, et al. L-arginine preserves endothelial dependent relaxation during environmental tobacco smoke in lipid fed rabbits  Circulation 1994-Suppl:I-459 Abstract.
42.  Werner RM and Pearson TA. What is so passive about passive smoking? Second hand smoke as a cause of atherosclerotic disease JAMA 1998;279:157-8.
43.  Davey Smith G. et al Passive smoking and health  should we believe Phillip Morris 's experts? BMJ 1996;313:929.