Otsuka R, Watanabe H, Hirata K, Tokai K, Muro T, Yoshiyama M, Takeuchi K, Yoshikawa J. Acute Effects of Passive Smoking on the Coronary Circulation in Healthy Young Adults. JAMA. 2001;286(4):436-441. doi:10.1001/jama.286.4.436
Author Affiliations: Department of Internal Medicine and Cardiology, Osaka City University Medical School, Osaka, Japan.
Context Recent studies have shown that passive smoking is a risk factor for
ischemic heart disease and may be associated with vascular endothelial dysfunction.
The acute effects of passive smoking on coronary circulation in nonsmokers
are not known.
Objective To determine the acute effects of passive smoking on coronary circulation
using coronary flow velocity reserve (CFVR), assessed by noninvasive transthoracic
Design, Setting, and Participants Cross-sectional study conducted from September 2000 to November 2000
among 30 Japanese men (mean age, 27 years; 15 healthy nonsmokers and 15 asymptomatic
active smokers) without history of hypertension, diabetes mellitus, or hyperlipidemia.
Main Outcome Measures Coronary flow velocity reserve, calculated as the ratio of hyperemic
to basal coronary flow velocity induced by intravenous infusion of adenosine
triphosphate and measured in each participant before and after a 30-minute
exposure to environmental tobacco smoke.
Results Heart rate and blood pressure responses to adenosine triphosphate infusion
were not affected by passive smoking exposure in either group. Passive smoking
exposure had no effect on basal coronary flow velocity in either group. Mean
(SD) CFVR in nonsmokers was significantly higher than that in active smokers
before passive smoking exposure (4.4 [0.91] vs 3.6 [0.88], respectively; P = .02), while CFVR after passive smoking exposure did
not differ between groups (P = .83). Passive smoking
exposure significantly reduced mean (SD) CFVR in nonsmokers (4.4 [0.91] vs
3.4 [0.73], respectively; P<.001).
Conclusions Passive smoking substantially reduced CFVR in healthy nonsmokers. This
finding provides direct evidence that passive smoking may cause endothelial
dysfunction of the coronary circulation in nonsmokers.
Passive smoking has been identified as an important risk factor for
cardiovascular disease.1- 6
In 1992, the American Heart Association concluded that the risk of death due
to heart disease is increased by about 30% among those exposed to environmental
tobacco smoke at home, and could be much higher in those exposed at the workplace,
where higher levels of environmental tobacco smoke may be present.1 There is evidence that exposure of nonsmokers to environmental
tobacco smoke breaks down the serum antioxidant defenses7
and is associated with impairment of endothelium-dependent function of arterial
walls.8 However, the acute effects of passive
smoking on the coronary circulation in nonsmokers have not been evaluated.
Coronary flow velocity reserve (CFVR), a measure of endothelial function
in the coronary circulation, can be noninvasively measured in the left anterior
descending coronary artery (LAD) using transthoracic Doppler echocardiography
The purpose of this study was to determine the acute effects of passive smoking
on coronary circulation using measurement of CFVR by TTDE.
We studied 30 healthy Japanese men (mean [SD] age, 27  years) including
15 nonsmokers and 15 asymptomatic active smokers from September 2000 to November
2000. These subjects were recruited from the students of Osaka City University
Medical School. All were clinically well and had no history of hypertension,
diabetes mellitus, hyperlipidemia, or left ventricular hypertrophy (septal
or posterior wall thickness at diastole >12 mm). Nonsmokers lived in smoke-free
homes, worked in smoke-free environments, and had no exposure to environmental
tobacco smoke for more than 12 hours. Subjects were considered active smokers
if they regularly smoked at least 20 cigarettes per day; the average (SD)
duration of smoking was 6.8 (2.7) years. Active smokers had refrained from
smoking for more than 12 hours before this study in order to minimize effects
of acute smoking on study measurements. All subjects gave informed consent
to the protocol, which was approved by the Committee for the Protection of
Human Subjects in Research at Osaka City University Medical School.
From all subjects, blood samples were taken into a heparinized syringe
by venipuncture for determination of plasma carboxyhemoglobin level (HbCO),
total cholesterol, triglycerides, and high-density lipoprotein (HDL) cholesterol
levels. Plasma HbCO level was determined by spectrophotometry as a parameter
of exposure to passive smoking.
After baseline hemodynamic and echocardiographic recording, all subjects
spent 30 minutes in the smoking room (450 cm × 300 cm with a 250-cm
ceiling) in our hospital. When this room was used for the study, some individuals
who were not among the study participants visited to smoke on their own accord.
The air concentrations of carbon monoxide in the echocardiographic laboratory
and smoking room were determined by averaging values measured every 5 minutes
in each room using Indoor Pollution Evaluating System Model IES-1000 (constant-potential
electrolysis, Sibata Scientific Technology, Ltd, Tokyo, Japan).
All subjects underwent heart rate and electrocardiographic monitoring
continuously and blood pressure measurement every 1 minute during echocardiographic
examinations. We calculated mean arterial pressure and heart rate–blood
pressure product as indices of cardiac work.14,15
Before and after passive smoking, we measured echocardiographic parameters
with a digital ultrasound system (Acuson Sequoia 512, Acuson Corporation,
Mountain View, Calif) using a frequency of 5 to 12 MHz (Doppler frequency,
3.5 MHz). For color Doppler flow mapping, the velocity range was set at ±
12 to ± 25 cm/s. The color gain was adjusted to provide optimal imaging.
The acoustic window was around the midclavicular line in the fourth and fifth
intercostal spaces in the left lateral decubitus position. The left ventricle
was imaged in the long-axis cross-section and the ultrasound beam was inclined
laterally. Next, coronary blood flow in the distal portion of the LAD was
searched for under color Doppler flow mapping guidance. With a sample volume
(1.5 or 2.0 mm wide) positioned on the color signal in the LAD, we recorded
Doppler spectral tracings of the flow velocity by fast Fourier transformation
analysis. Adenosine triphosphate16 was administered
(140 µg/kg per minute) for 2 minutes to record spectral Doppler signals
during hyperemic conditions. All studies were continuously recorded on videotape
and clips of stopped frames were also stored digitally on magneto-optical
disks (230 MB) for subsequent off-line analysis. Coronary flow velocity was
measured at baseline and at peak hyperemic conditions by tracing contours
of spectral Doppler signals using the software incorporated in the ultrasound
system. These measurements were made by the investigators who were blinded
to the subjects' smoking status. Each parameter was averaged over 3 consecutive
cycles. Coronary flow velocity reserve was calculated as the ratio of hyperemic
to basal coronary flow velocity.
Baseline characteristics including age, total cholesterol, triglycerides,
and HDL cholesterol in the 2 groups at baseline were compared with the unpaired t test; P<.05 was considered
significant. To compare effects of adenosine triphosphate administration and
passive smoking, we used repeated measures analysis of variance (ANOVA) for
hemodynamic parameters, the air concentration of carbon monoxide, HbCO level,
coronary flow velocity, and CFVR over adenosine triphosphate administration
before and after passive smoking. Where appropriate, directed pairwise comparisons
of individual groups were conducted using the unpaired t test. We used a paired t test for directed
comparisons of passive smoking effect in each group. For all analyses, we
used SAS software version 6.12 (SAS Institute, Cary, NC). Lipid values are
reported in conventional units. To convert total and HDL cholesterol from
mg/dL to mmol/L, multiply by 0.0259. To convert triglycerides from mg/dL to
mmol/L, multiple 0.0113.
Patient age did not significantly differ in nonsmokers and active smokers
(mean [SD], 27  years for both groups; P = .82).
Other baseline characteristics including heart rate, blood pressure, mean
arterial pressure, and heart rate–blood pressure product were also similar
in nonsmokers and active smokers (Table
1). Total cholesterol, triglycerides, and HDL levels did not significantly
differ in nonsmokers and active smokers (167  mg/dL vs 163  mg/dL, P = .78; 102  mg/dL vs 90  mg/dL, P = .53; and 56.1 [7.8] mg/dL vs 55.0 [13.5] mg/dL, P = .19, respectively).
None of the subjects experienced any symptoms or had any electrocardiogram
change during either passive smoking or adenosine triphosphate administration.
Passive smoking had no effect on hemodynamic parameters including heart rate,
blood pressure, mean arterial pressure, and heart rate–blood pressure
product in each group (Table 1).
The results of repeated measures ANOVA analysis for carbon monoxide
level in air and HbCO level in blood are presented in Table 2. Carbon monoxide level in the smoking room was higher than
that in the echocardiographic laboratory for both nonsmokers and active smokers.
There were significant group, passive smoking, and interaction effects on
HbCO level over passive smoking between both groups. Before passive smoking,
the HbCO level in the blood was significantly lower in nonsmokers than in
active smokers. Passive smoking significantly increased HbCO level in nonsmokers
but did not significantly increase HbCO level in active smokers.
Coronary flow velocity could be observed at baseline and during hyperemia
in all subjects. There was a significant interaction effect between the 2
groups over adenosine triphosphate administration before and after passive
smoking (Table 3 and Figure 1). Coronary flow velocity during
hyperemia in nonsmokers was significantly higher than that in active smokers
before passive smoking. This parameter was quite similar in the 2 groups after
passive smoking. Thus, CFVR in nonsmokers was significantly higher than that
in active smokers before passive smoking (P = .02),
whereas CFVR did not differ between the 2 groups after passive smoking (P = .83). Coronary flow velocity reserve in nonsmokers was significantly
reduced by passive smoking (P<.001) (Table 4 and
Our data revealed that temporary passive smoking abruptly reduced CFVR
in nonsmokers but did not affect CFVR in active smokers. This provides direct
evidence of a harmful effect of passive smoking on the coronary circulation
Cigarette smoking is one of the major risk factors for cardiovascular
disease.17,18 This may be the
result of structural19 or functional changes18,20 in the coronary artery produced by
smoking. Some epidemiological studies have linked passive smoking to excess
risk for atherosclerotic heart disease.1,21- 25
It is thought that some premature deaths of nonsmokers may be related to passive
smoking, with the majority of such deaths due to cardiac ischemia.1,21- 25
Celermajer et al8 have shown that passive smoking
is associated with dose-related impairment of endothelium-dependent dilatation
of the brachial artery in healthy young adults. Dilatation mediated by brachial
artery flow is endothelium-dependent, mediated by the release of nitric oxide.
Although endothelial dysfunction in the brachial artery appears to be well
correlated with both coronary endothelial physiological function and the degree
of coronary atherosclerosis, flow-mediated dilatation of brachial artery does
not evaluate response of the coronary circulation directly.
The predictive association of coronary endothelial function with clinical
outcome of patients with coronary artery disease supports the concept that
endothelial function may serve as an integrating index of overall coronary
risk factor stress. Thus, assessment of coronary endothelial vasoreactivity
may be an important diagnostic and prognostic tool.26
Coronary flow reserve has been proposed as a parameter of physiological changes
in the coronary circulation reflecting the function of large epicardial arteries
and microcirculation.27,28 Impaired
coronary flow reserve has been suggested as a surrogate measure of subclinical
coronary atherosclerosis, providing an integrated measure of both vascular
endothelial function and smooth muscle relaxation.29
Recent studies have found a good agreement between CFVR as assessed with Doppler
guide wire and the results of perfusion scintigraphy and positron emission
Tanaka et al32 found that smoking a cigarette
with a high nicotine content abruptly reduced CFVR. Sumida et al33
found that long-term exposure to environmental tobacco smoke impaired acetylcholine-induced
coronary artery dilatation, indicating coronary endothelial dysfunction. However,
CFVR has previously been measured only by invasive or semi-invasive procedures18,27,28,34- 36
and few findings have been reported on the direct impact of passive smoking
on coronary circulation in healthy nonsmokers. CFVR can now be measured noninvasively
by TTDE,9- 13
and good agreement has been found between CFVR as assessed with TTDE and the
results of Doppler guide wire examination.37
Thus, CFVR measurement by TTDE has become a clinical tool for noninvasive
and physiological assessment of coronary circulation.
In this study, CFVR before passive smoking was lower in active smokers
than in nonsmokers. This difference was also found in recent studies of CFVR
in active smokers.32,38 Kaufmann
et al38 found that mean (SD) CFVR values in
nonsmokers and active smokers were 4.55 (0.84) and 3.79 (0.60), respectively
(P<.05). In the present article, CFVR in nonsmokers
was reduced to the same level as in active smokers after passive smoking.
On the other hand, CFVR in active smokers was not significantly reduced by
passive smoking. The present study is the first to demonstrate that passive
smoking may have a stronger adverse effect on CFVR in nonsmokers than in active
Environmental tobacco smoke includes many toxic constituents, such as
carbon monoxide, benzopyrene, and more than 4000 chemicals.1,39
One or some of these toxic constituents may injure the arterial wall.21- 23 Allred et al40 found that increased carbon monoxide level induced
by short-term exposure to environmental tobacco smoke resulted in more rapid
onset of angina in patients with coronary artery disease as a result of endothelial
dysfunction. In the present article, short-term exposure to environmental
tobacco smoke increased the level of HbCO in nonsmokers, but in active smokers
no difference in HbCO was found before and after passive smoking. This may
be one of the reasons why passive smoking had a stronger adverse effect on
CFVR in nonsmokers than in active smokers.
We measured changes in coronary flow velocity, not changes in coronary
blood flow. However, it has been reported that changes in coronary flow velocities
induced by coronary vasodilatation closely reflect changes in coronary blood
flow.41 Furthermore, we cannot exclude the
possibility that some of the volunteers in this study had epicardial coronary
artery disease. This may have been ruled out only with coronary angiography,
the performance of which seemed unjustified in these asymptomatic volunteers.
However, none of the subjects had hypertension, diabetes, hyperlipidemia,
or a history of coronary artery disease. Thus, their clinical risk for coronary
artery disease was considered low.
A limitation in our study was that our design did not allow us to comment
on long-term effects of passive smoking or the duration of the CFVR reduction
after passive smoking; these effects may be worth testing in a large-scale
In healthy individuals without coronary artery disease, reduction of
CFVR can result from dysfunction of the coronary microcirculation.27,28 The present findings suggest that
reduction of CFVR after passive smoking may be caused by endothelial dysfunction
of the coronary circulation, an early process of atherosclerosis, and that
this change may be one reason why passive smoking is a risk factor for cardiac
disease morbidity and mortality in nonsmokers.