N A L A R T I C L E

Acute Effect of Cigarette Smoking on

Glucose Tolerance and Other
Cardiovascular Risk Factors
AL.BF.RTO C . FRATI, MD, FACP an independent cardiovascular risk factor
FEI.IPF. INIESTRA, MD (9). Nevertheless, the pathogenic linkage
C. RAUL ARIZA, MD, FACP between these changes and smoking is
not well understood. The acute effects of
cigarette smoking in smokers include
transient increases in heart rate and blood
pressure (10), a nonconsistent decrease of
OBJECTIVE To investigate the acute effect of cigarette smoking on glucose tolerance, serum HDL cholesterol (11,12), and im-
insulin sensitivity, serum lipids, blood pressure, and heart rate. paired insulin action (13). The acute ef-
fects of smoking in nonsmokers (i.e., pas-
RESEARCH DESIGN A N D METHODS This nonrandomized experimental control sive smoking) have not been studied. The
trial in a tertiary care center included 20 healthy chronic smokers and 20 age-, sex-, and
aim of this study was to investigate the
BMI-matched healthy volunteers. Two oral glucose tolerance tests (OGTTs) were performed on
each subject. Three cigarettes were smoked during the first 30 min in one of the tests. Serum acute effect of cigarette smoking on glu-
glucose, insulin, and C-peptide levels were measured every 30 min; the area under the curve cose tolerance, insulinemia, insulin sensi-
(AUC) and the insulin sensitivity index (ISI) were calculated; serum total cholesterol, LDL tivity, serum lipids, heart rate, and blood
cholesterol, HDL cholesterol, and triglyceride levels were measured at 0 and 180 min; and blood pressure in nonsmokers as well as in
pressure and heart rate were recorded every 5 min throughout 180 min. chronic smokers.

RESULTS Smoking acutely impaired glucose tolerance: the AUC for glucose in smokers
was 2*5.5 1.03 mmol/1 (mean SE) (95% CI 22.9-28) during the smoking OGTT and 21.8 RESEARCH DESIGN A N D
0.85 mmol/1 (CI 19.2-24.3) in the control OGTT (P < 0.01); in nonsmokers, it was 19.7 METHODS Two groups of healthy
0.3 mmol/1 (CI 18.8-20.5) in the smoking OGTT and 18.7 0.35 mmol/1 (CI 17.8-19.5) in the volunteers were studied. They were re-
control OGTT (P < 0.05). Smoking acutely increased serum insulin and C-peptide levels and cruited from relatives of patients or hos-
decreased ISI only in smokers: ISI in smokers was 55 2.8 (CI 47.4-62.6) in the control OGTT pital personnel. At the beginning of the
and 43 2.7 (CI 35.4-50.6) in the smoking OGTT (P < 0.05). Smoking acutely caused a rise study, they were consecutively admitted,
of serum total cholesterol levels in both groups and increased LDL cholesterol and triglyceride divided into one of two groups according
serum levels significantly only in smokers (P < 0.05). A significant rise of blood pressure and to their smoking habits, and matched by
heart rate while smoking was present in all the subjects. sex, age, and BMI. Group I (nonsmokers)
included 10 men and 10 women who had
CONCLUSIONS Smoking acutely impaired glucose tolerance and insulin sensitivity,
enhanced serum cholesterol and triglyceride levels, and raised blood pressure and heart rate.
never smoked. Their ages ranged from 20
These findings support the pathogenetic role of cigarette smoking on cardiovascular risk factors. to 55 years (29 7 [mean SD]); their
BMI (Quetelet index) was 24.9 0.5 kg/
m2. Group II (smokers) included 10 men
and 10 women ranging from 20 to 53

S
moking is a well-established major risk factors such as hypercholesterolemia years (29 7) of age with BMI of 24.8
risk factor for coronary and periph- and hypertension (3). 0.3 kg/m2. They had a history of smoking
eral vascular disease. Several epide- Cigarette smoking may increase at least 10 cigarettes per day (21.2 7
miological studies have shown that ciga- the incidence of NIDDM (4,5) and is ep- [range 10-40] during the last year) with a
rette smoking increases the incidence of idemiologically related to at least two mean duration of smoking of 7.5 6
myocardial infarction and sudden coro- other risk factors such as serum lipid years (range 2-20). Waist-to-hip ratio
nary death, and it adversely affects the changes and insulin resistance (6-8). The was not measured. Subjects with diabetes
prognosis in patients with previous myo- latter, which has recently been demon- or severe obesity, hypertension, or endo-
cardial infarction or angina pectoris (1,2). strated in a group of heavy smokers (8), crine, liver, or kidney diseases or those
Smoking also enhances the effects of other has also increasingly been mentioned as taking any medication were not included.
Two subjects in the group of nonsmokers
had a first-degree relative with diabetes,
From the Department of Internal Medicine, Hospital de Especialidades del Centra Medico La Raza, Instituto
while three subjects in the group of smok-
Mexicano del Seguro Social, Mexico City, Mexico. ers had a first-degree relative with hyper-
Address correspondence and reprint requests to Alberto C. Frati, MD, FACP, Division de Medicina tension. No other hereditary factor for
Interna, Hospital de Fspecialidades del Centra Medico La Raza, Instituto Mexicano del Seguro Social, Seris glucose intolerance or atherosclerosis was
y Zaachila, Col. La Raza, 02990, Mexico DF, Mexico. clinically detected.
Received for publication 1 May 1995 and accepted in revised form 28 September 1995.
AUC, area under the curve; FFA, free fatty acid; ISI, insulin sensitivity index; OGTT, oral glucose tolerance All the subjects were ambulatory.
Before the test, they did not consume a

112 DIABETES CARE, VOLUME 19, NUMBER 2, FEBRUARY 1996


special diet, but they were not allowed to
drink alcohol or to smoke for 24 h before
the tests. Two oral glucose tolerance tests
(OGTTs) in random order were per-
formed in each subject with an interval of
3 days. In one test, three tobacco ciga-
rettes were smoked (at 0,15, and 30 min),
while the other test, without smoking,
was the control. The smoking velocity of
each cigarette was not monitored. The
nicotine content of each cigarette was 1
mg (information supplied by the manu-
facturer). Each test was performed after a
night of normal sleep with subjects in the
fasting condition. Tests were started at
8:00 A.M. while the subject was resting
seated. A venous blood sample for labo-
ratory measurements was drawn, and an
oral glucose load of 75 g in 200 ml water
was given. The subject started to smoke
the first cigarette immediately after the
glucose load. Serum glucose, insulin, and
C-peptide levels were measured every 30
min throughout 180 min. Serum total
cholesterol, HDL cholesterol, LDL choles-
terol, and triglyceride levels were deter-
mined at 0 and 180 min. Blood pressure
and pulse rate were recorded every 5 min
throughout the test.
Serum glucose was measured im-
mediately after the blood samples were
obtained by the glucose oxidase auto-
mated method (570 Alliance, Ciba Corn-
ing Shimadzu, Oberlin, OH) (coefficient
of variance, 4%). Aliquots of serum were
kept frozen at 20C until insulin and
C-peptide were determined by a radioim-
munoassay (CIS Biointernational, Oris
Industrie SA, Gif-Sur-Ivette, France) that
measures human insulin (100%) and hu-
man proinsulin (40%). Total cholesterol,
LDL cholesterol, HDL cholesterol, and
triglyceride levels were measured by en-
zymatic methods (PAP, BIO Merieux,
France). The coefficient of intra-assay er-
ror for each method was <6%. Blood
pressure was measured with a standard
mercury sphygmomanometer. The same
sphygmomanometer was used in every
instance, and the 12-cm-wide cuff was
placed on the same arm in both tests. Di-
astolic blood pressure was defined at Ko-
rotkoff 5.
The areas under serum concentra-
tion curves (AUCs) of glucose, insulin,
and C-peptide were estimated by the lin-
ear trapezoidal method. The insulin sen-
sitivity index (ISI) was calculated accord-
ing to the formula of Cederholm and
Wibell (14), adjusting the glucose uptake
DIABETES CARE, VOLUME 19, NUMBER 2, FEBRUARY
rate for different plasma glucose levels
(15) as has been used by Lindhal et al.
(16): ISI = M/MSG/log MSI. The glucose
uptake rate (M) was estimated as the dif-
ference between the glucose load (75,000
mg) and the glucose left in the glucose
space. M was calculated as follows: M =
glucose load (120 + [0-h glucose 2-h
glucose] X 180 X 0.19 X body weight/
120). MSG is the mean of serum glucose
at 0 and 120 min. MSI is the mean of
serum insulin concentration at 0 and 120
mm.
Data are means SE. Analysis of
variance was used for statistical calcula-
tions, but the Mann-Whitney U test was
used to compare the time of the maxi-
mum peak of serum insulin. P < 0.05
(two-tailed) was considered significant.
Every volunteer was aware of the
purpose of the investigation and possible
hazards and side effects and signed an in-
formed consent. To know if smoking (i.e.,
passive) could also affect nonsmokers,
this group of volunteers had to be in-
cluded. The long-term side effects of
smoking three cigarettes were thought to
be negligible. The study was approved by
the Investigation Committee and was per-
formed according to principles of the Hel-
sinki Declaration.
RESULTS In the group of non-
smokers, the control OGTT was normal
in 18 subjects and was classified as glu-
cose intolerance (2-h glucose >7.7 and
<11.1 mmol/l [>140 and <200 mg/dl])
in 2 subjects. Cigarette smoking was fol-
lowed by a further rise in serum glucose
levels that was significant at 90 and 120
min (P < 0.01) (Table 1). The AUC for
glucose was 18.7 0.355 in the control
test and 19.7 0.305 mmol/l in the
smoking test (P < 0.05).
The control OGTT in the group of
smokers showed significantly higher se-
rum glucose values than in nonsmokers at
90, 120, 150, and 180 min (P < 0.01).
The AUC for glucose in the control OGTT
was also greater in smokers (21.8 0.854
mmol/l) than in nonsmokers (P < 0.01).
In the group of smokers, the OGTT was
normal in 11 subjects and showed glu-
cose intolerance in 9 subjects. In the
group of smokers, the OGTT during cig-
arette smoking was clearly different from
the test without smoking. In the former,
serum glucose levels were significantly
higher than in the control test at 90, 120,
and 150 min (P < 0.01). The AUC for
1996
Frati, Iniestra, and Arizci
glucose in the smoking test was 25.5 :+:
1.038 mmol/l, which was significantly
greater than in the test without smoking
(21.8 0.854 mmol/l) (P < 0.01). The
2-h serum glucose after smoking was
>11.1 mmol/l in 8 individuals, >7.7 and
<11.1 in 11 subjects, and <7.7 mmol/l
only in 1 smoker subject.
The effect of cigarette smoking on
the OGTT was greater in usual smokers
than in nonsmokers (Fig. 1). In smokers,
serum glucose values at 90, 120, and 150
min in smoking test were means of 3,2.1,
and 2.4 mmol/l above the control test, re-
spectively, compared with 0.9, 0.84, and
0.2 mmol/l above control test in non-
smokers (P < 0.01).
Serum insulin levels in nonsmok-
ers were similar in both OGTTs. The AUC
values for insulin were also similar in both
tests (smoking 932 66.7 pmol/1; con-
trol 889 63.8 pmol/1) (P > 0.05). In the
group of smokers, the AUC for insulin in
the control OGTT was 832 66.7
pmol/1, which was similar to that in the
nonsmokers. However, a late peak of se-
rum insulin concentrations was noticed
in chronic smokers. In the control OGTT,
the maximal peak of serum insulin levels
in nonsmokers appeared at close to 90
min (range 30-150), while in smokers it
appeared at close to 150 min; mode and
median also were 150 min (range 60-
180). The difference was highly signifi-
cant (P < 0.01, Mann-Whitney 1/ test). In
smokers, cigarette smoking had an en-
hancing effect on serum insulin levels;
these were significantly higher in the
smoking than in the control test at 60 and
90 min (P < 0.05). The AUC for insulin
was greater in the smoking test (1,112
48 pmol/1) than in the control test (832 :+:
66.7 pmol/1) ( P < 0.01).
Similarly to serum insulin, a sig-
nificant rise of C-peptide levels after
smoking was present only in the group of
smokers. The AUC for C-peptide in the
control OGTT was similar in both groups
(4.1 0.362 and 4.1 0.218 pmol/1);
after smoking, it was significantly greater
in the group of smokers (4.9 0.218
pmol/1) (P < 0.05).
The ISI in the control test was sig-
nificantly higher in nonsmokers (67 :+:
4.1) than in smokers (55 2.8) (P <
0.05). After smoking, a minor and non-
significant decrease of the index was ob-
served in the group of nonsmokers (60
2.9) (P > 0.05), while a significant dimi-
nution of the ISI was present in the group
113
Smoking and glucose tolerance

Table 1Effects of smoking on serum glucose, insulin, and C-peptide levels during OGTT

Nonsmokers Smokers
Glucose Insulin C-peptide Glucose Insulin C-peptide
Test (mmol/1 [mg/dl]) (pmol/1) (pmol/1) (mmol/1 [mg/dl]) (pmol/1) (pmol/1)
0 min
Control 5.6 0.26 [101.1 4.7] 111 4 0.42 0.03 5.5 0.19 [100.6 3.5] 90 7 0.55 0.06
Smoking 5.6 0.13 [100.5 2.4] 113 10 0.46 0.06 4.9 0.18 [88.5 3.3] 137 7 0.62 0.05
30 min
Control 6.0 0.14 [109.5 2.6] 128 22 0.60 0.09 5.8 0.23 [105.2 4.2] 96 12 0.55 0.05
Smoking 6.0 0.11 [109.7 2.0] 124 11 0.42 0.05 5.5 0.26 [100.2 4.7] 210 53 0.81 0.11*1
60 min
Control 6.8 0.23 [122.5 4.2] 320 46 1.36 0.27 7.2 0.27 [131.3 4.9] 262 51 1.28 0.13
Smoking 6.5 0.18 [119.9 3.3]* 220 41 1.30 0.15 8.0 0.55 [144.6 10.0]*t 426 62*T 1.46 0.15
90 min
Control 7.0 0.18 [127.3 3.3] 428 68 2.09 0.23 8.0 0.28 [144.0 5.1]*t 322 39 1.80 0.16
Smoking 7.9 0.18 [143.7 3.4]* 457 33 1.72 0.12 10.9 0.60 [146.7 10.4]*t 459 41* 1.89 0.13
120 min
Control 6.3 0.24 [113.8 4.4] 403 54 2.01 0.20 8.2 0.34 [148.5 6.2]t 363 35 1.99 0.11
Smoking 7.1 0.18 [128.4 3.3] 365 38 1.93 0.17 10.3 0.48 [187.0 8.7]*t 408 25t 2.20 0.17
150 min
Control 6.0 0.24 [108.2 4.4] 370 54 1.81 0.14 8.1 0.39 [146.7 7.1]t 426 45t 2.12 0.17
Smoking 6.2 0.22 [112.8 4.0] 393 44 1.93 0.15 10.5 0.65 [190.1 11.8]t 497 53t 2.51 0.18*t
180 min
Control 5.0 0.27 [90.2 4.9] 154 17 1.57 0.25 7.4 0.42 [134.5 7.1] 320 38t 1.78 0.14
Smoking 5.2 0.16 [93.8 2.9] 250 41 1.55 0.08 7.5 0.59 [135.1 10.7] 363 22t 2.43 0.12*t
Data arc means SH. Statistically significant (P < 0.05) *smoking vs. control test; +smokers vs. nonsmokers.

of smokers (43 2.7) (P < 0.05 vs. the
control test).
No differences between groups in
basal serum levels of total cholesterol,
LDL cholesterol, HDL cholesterol, and
triglycerides were noticed. However, in
both groups, total cholesterol levels rose
after smoking (P < 0.05). A significant
increase in serum LDL cholesterol and tri-
glyceride concentrations after smoking
was present in the group of smokers,
while similar, although not statistically
significant, changes were observed in the
group of nonsmokers (Table 2).
Basal systolic and diastolic blood
pressure values were marginally lower in
smokers than in nonsmokers: 103 1/63
1 mml Ig (systolic/diastolic) in smokers
and 106 1/65 0.8 mmHg in non-
smokers. The difference was not statisti-
cally significant (P > 0.05). Basal heart
rate was similar in both groups (64 0 . 8
beats/min). In both groups, cigarette
smoking caused a rise of blood pressure
and heart rate that started at 5 min, and it
was sustained throughout the 1st hour
(Fig. 2). In nonsmokers, systolic/diastolic
blood pressure from 5 to 60 min was 106
1.2/66 1 . 3 mmHg in the control test
and 110 2.5/71 1.77 mmHg in the
smoking test (P > 0.05 for systolic and P
< 0.05 for diastolic). Values of heart rate
from 5 to 60 min were 63 1 and 73
1.9 beats/min in the control and smoking
tests, respectively (P < 0.01). In the
group of smokers, these values were sys-
tolic/diastolic blood pressure of 103
2.1/64 1.1 in the control test and 110
1.7/69 1 . 5 mmHg in the smoking
test (P < 0.05 for systolic and <0.01 for
diastolic). Heart rates from 5 to 60 min
were 65 1.2 and 74 2.2 beats/min in
control and smoking tests, respectively (P
< 0.01). No effect on blood pressure and
heart rate was seen after the 1st hour.
C O N C L U S I O N S This study
shows that cigarette smoking acutely de-
teriorates glucose tolerance in healthy
nonsmokers as well as in usual tobacco
smokers. This impairment was more ma-
jor in chronic smokers than in nonsmok-
ers. This effect is probably due to dimin-
ished sensitivity to insulin since the ISI
decreased significantly after smoking
compared with the test without smoking.
ISI has been used as a practical measure of
insulin resistance (14,16). Good agree-
ment was obtained between calculated ISI
values at the OGTT and metabolic clear-
ance rate and steady-state plasma glucose
values at the clamp (the gold standard)
and insulin suppression tests (14). The
mechanism of insulin resistance in this
case cannot be determined by this study.
However, an enhancement of catechol-
amines may be at least partially responsi-
ble. Nicotine absorbed during smoking
increases the discharge of catecholamines
from the adrenal medulla and from extra-
adrenal chromaffin tissue, which causes
heart rate and blood pressure to rise
(17,18). In our study, an early rise in heart
rate and blood pressure induced by
smoking was seen in both smokers and
nonsmokers.
On the other hand, hyperinsu-
linemia, which is a feature of insulin re-
sistance, may cause adrenergic activation
(19). Nevertheless, sympathetic overac-
tivity (tachycardia, high blood pressure)
due to smoking appeared before signifi-
cant hyperinsulinemia was present. Thus,
it is more likely that adrenergic activity
induced by smoking was responsible for
insulin resistance than that hyperinsu-
linemia caused adrenergic activation.

114 DIABETES CARE, VOLUME 19, NUMBER 2, FEBRUARY 1996

 Frati, hiiestra, and Arizci

cigarette smokers and in snuff users was

NON SMOKERS SMOKERS reported by Eliasson et al. (21), while nor-
mal plasma insulin concentrations in
o I '\ smokers were found by others (22). Se-
rum insulin and C-peptide basal levels
and AUCs were similar in smokers and in
nonsmokers in the present study, but
chronic cigarette smokers had impaired
glucose tolerance compared with non-
smokers. Impaired glucose tolerance in
smokers was probably caused by insulin
resistance, since the LSI was lower in
smokers than in nonsmokers. Both glu-
cose tolerance and ISI were further im-
INSULIN paired immediately after subjects smoked
three cigarettes. Chronic heavy cigarette
smokers have been recently found to be
hyperinsulinemic and insulin resistant
compared with a matched group of non-
smokers (6). Janzon et al. {22) also
showed that smokers had a delect in glu-
cose removal in response to an intrave-
C PEPTIDE nous glucose challenge. Recently it has
been shown (2.3) in a group of healthy
pmol/L male smokers that the amount of nicotine
smoked per day is related to the degree of
OJ insulin resistance measured by an eugly-
60 120 180 0 60 120 180 cemic hyperinsulinemic clamp. As far as
we know there is only one report con-
MINUTES MINUTES
cerning the acute effect of smoking on in-
#<0.05 sulin resistance (13). These authors stud-
0<0.01 ied seven smokers with the euglycemic
Figure 1Serum glucose, insulin, and C-peptide levels during an OGTT performed while subjects clamp technique and reported that smok-
were smoking three cigarettes ( ) or without smoking ( ). The OGTT while subjects smoked ing impaired insulin action mainly be-
.showed significantly higher values of glucose, insulin, and C-peptide than the control test in the group of cause of lower peripheral glucose uptake.
smokers. A higher serum glucose level in the smoking test was also present in nonsmokers.
Nicotine also affects gastric emp-
tying, which in turn can modify glucose
Moreover, enhancement of serum insulin plasma glucose concentrations before and tolerance tests (24). However, if rapid
levels by glucose challenge alone, as in the after an oral glucose challenge in smokers gastric emptying were the cause of the
control test, did not affect heart rate and and nonsmokers and failed to find signif- high glycemic levels, early high serum in-
blood pressure. Artificially induced high icant differences between the groups. sulin levels would also be expected in the
hyperinsulinemia does not acutely raise However, higher serum insulin levels af- smoking test.
blood pressure (20). ter glucose intake were present in the A surprising finding was that sub-
Facchini et al. (8) measured group of smokers. Hyperinsulinemia in jects who smoked had a retarded peak of

Table 2Acute effect of cigarette smoking on serum lipids

Total cholesterol LDL cholesterol HDL cholesterol Triglyeerides

Basal values (mmol/1)
Nonsmokers 3.96 0.151 2.6 0.151 1.08 0.05 1.25 + 0.070
Smokers 3.36 0.379 2.8 0.151 1.08 0.05 1.40 0.158
Changes after OGTT (mmol/1)
Nonsmokers -0.07 0.151 -0.22 0.126 -0.05 0.025 -0.08 + 0.045
Smokers -0.02 0.151 -0.05 0.126 -0.05 0.025 -0.02 + 0.00
Changes after OGTT plus smoking (mmol/1)
Nonsmokers +0.45 0.151* +0.12 + 0.126 -0.10 0.025 +0.08 0.112
Smokers +0.43 0.101* +0.68 0.126;:; -0.05 0.025 +0..S2 0.0lV>(v
Data arc means SE. *P < 0.05 vs. basal values and control OGTT.

DIABETES CARE, VOLUME 19, NUMBER 2, FEBRUARY 1996 115

Smoking and glucose tolerance

NON SMOKERS SMOKERS

BP 120.
SYSTOLIC
110.
mmHg /' " " ' * ' v
^\.i

100^

70 >*
DIASTOLIC
60
smoking test
80_
control test
HEART
RATE 70
beot/min
60J
CIGARETTE
SMOKED 6 60 120 180 0 60 120 180
MINUTES MINUTES
<0.05*
<0.01 O
Figure 2Blood pressure and heart rate during an OGTT while subjects were smoking ( ) or without smoking ( ). Cigarettes were smoked at 0,
15, and 30 min (arrows). A significant rise ofdiastolic blood pressure and ojheart rate during the 1st hour in the smoking test was seen in both groups of
subjects. Systolic blood pressure also rose, but it was significant only in the group oj smokers.

serum insulin in the OGTT, as occurs in
NIDDM patients (25,26). This rinding
would be easily explained if individuals
with NIDDM were inadvertently included
in the group of smokers. However, none
had an OGTT indicating diabetes accord-
ing to American Diabetes Association and
World Health Organization criteria
(27,28) in the test without smoking. Fur-
thermore, age, sex, BMI, and family his-
tory of insulin-resistant related condi-
tions (diabetes, obesity, or hypertension)
were similar in both groups. Moreover,
despite that, an OGTT indicating glucose
intolerance was more frequent in the
group of smokers: the 11 smoker subjects
who had a normal control OGTT had a
similarly retarded peak of insulin (median
150 min, range 90-180) than the 9 sub-
jects with glucose intolerance (median
150 min, range 60-180). To our knowl-
edge, this finding has not previously been
reported. A role of retarded insulin releas-
ing in the overall impairment of glucose
tolerance and insulin-resistant condition
cannot be ruled out.
Measurements of serum choles-
terol, triglyceride, and lipoprotein con-
centrations in smokers have disclosed
conflicting results (6-8,21,29-31). How-
ever, the general conclusion is that smok-
ers have high serum concentrations of
total cholesterol, triglycerides, LDL cho-
lesterol, and VLDL cholesterol and have
low serum HDL cholesterol values (7). No
differences in serum total cholesterol,
LDL cholesterol, HDL cholesterol, and
triglyceride concentrations were found
between smokers and nonsmokers in our
study, perhaps because of the small size of
the sample herein studied.
In this study, smoking acutely
caused a rise in serum levels of total cho-
lesterol, LDL cholesterol, and triglycer-
ides. Reports about the acute effect of
smoking upon serum lipid concentra-
tions are also controversial, ranging from
no effect to a decrease in plasma total cho-
lesterol, triglycerides, VLDLs, and HDL
cholesterol (11,12,32,33). The acute ef-
fect of smoking on serum lipids has been
studied in the fasting condition, but only
in one instance (11). The test in this study
was performed after an oral glucose load.
The combined effect of glucose intake and
tobacco smoking was a rise of serum total
cholesterol, LDL cholesterol, and triglyc-
eride concentrations mainly in usual
smokers. Postprandial smoking is a fre-
quent habit. The explanation of these
changes is far from clear. However, we
know that nicotine stimulates the release
of adrenaline, which in turn raises plasma
concentrations of free fatty acids (FFAs)
through enhanced lipolysis and FFA mo-
bilization from adipose tissue (34). FFAs
stimulate the hepatic secretion of VLDLs
and triglycerides and also cholesterol (7).
An inhibition of lipoprotein lipase activity
by smoking has also been proposed as a
mechanism of enhanced levels of circulat-
ing triglycerides and VLDLs (35). Never-
theless, adipose tissue lipoprotein lipase
activity is increased in smokers (36). This
enzyme promotes the storage of triglycer-
ides in adipose tissue and perhaps has a
counterregulatory role in the mainte-
nance of body weight in smokers (36).
The data from this investigation
agree with other reports showing that
smoking acutely raises heart rate and
blood pressure (37-40). Despite the tran-
sient rise of blood pressure caused by nic-
otine, data from epidemiological studies
have generally shown smokers to have
lower blood pressure than nonsmokers
(41). The importance of the effect of to-

116 DIABETES CARE, VOLUME 19, M M M R 2, FIBRIARY 1996

 Frati, Inicstra, and Arizu

bacco on blood pressure might be over-
looked because the transient elevation of
blood pressure may occur often in heavy
smokers. The effect of repeated nicotine
stimuli on blood pressure was assessed by
Mann et al. (42) in untreated hypertensive
patients monitoring 24-h ambulatory
blood pressure; smokers maintained a
higher daytime ambulatory systolic blood
pressure than nonsmokers even though
blood pressure in the office was similar.
In this study, the changes acutely
caused by smoking were in general more
pronounced in chronic smokers than in
nonsmokers, and in the latter, even the
nonstatistically significant changes
seemed to behave similarly to those in the
group of smokers. The differences in the
magnitude of response might suggest
some kind of sensitization or memory
elicited by cigarette smoking.
The influence of long-term low-
concentration tobacco smoke exposure
(as chronic passive smokers) or of acute
passive smoking cannot be entirely in-
ferred from this investigation. However,
since metabolic and hemodynamic
changes appeared also in healthy non-
smokers and the magnitude of these
changes depends on both the duration
(group of chronic smokers) and the inten-
sity (acute smoking) of the exposure, it is
highly probable that at least small changes
can occur in passive smokers.
Smoking causes metabolic and
hemodynamic changes that are similar to
those described in syndrome X (insulin
resistance syndrome) (43). This syn-
drome is thought to be a primary condi-
tion; however, smoking tobacco caused
an acute syndrome X. Smoking might be
an acquired and easily avoidable cause of
this syndrome.
The overall epidemiological, clin-
ical, and experimental data support a role
lor several concurrent pathogenic mech-
anisms for cigarette smoking-induced
vascular damage: changes in serum lipid
concentrations, enhanced peroxidation of
I.DI. and its metabolism by macrophages
(44), insulin resistance, elevated blood
pressure, and altered hemostatic function
(45). This study shows that some of these
(actors such as hyperglycemia, hyperin-
sulinemia, insulin resistance, hypercho-
lesterolemia, hypertriglyceridemia, and
high blood pressure are acutely provoked
by cigarette smoking. Repeated stimuli
may lead to sustained changes and con-
tribute to atherogenesis.
References
1. Wilhclmsen L: Coronary heart disease:
epidemiology of smoking and interven-
tion studies of smoking. Am Heart] 115:
242-249, 1988
2. McGill HC: The cardiovascular pathology
of smoking. Am Heart] 115:250-257,
1988
3. Mancia G: Opening remarks: the need to
manage risk factors of coronary heart dis-
ease. Am Heart] 115:240-242, 1988
4. Feskens EJ, Kromhout D: Cardiovascular
risk factors and the 25-years incidence of
diabetes mellitus in middle-aged men. Am
] Epidemiol 130:1101-1108, 1989
5. Rimm EB, MansonJE, Stampfer MJ, Cold-
itz GA, Willet WL, Rosner B, Hennekens
CH, Speizer FE: Cigarette smoking and
the risk of diabetes in women. Am] Public
Health 83:211-214, 1993
6. Willet W, Hennekens CH, Castelli W,
Rosner B, Evans D, Taylor J, Kass EH: Ef-
fects of cigarette smoking on fasting tri-
glycerides, total cholesterol, and HDL-
cholesterol in women. Am Heart ] 105:
417-421, 1983
7. Craig WY, Palomaki GE, HaddowJE: Cig-
arette smoking and serum lipid and li-
poprotein concentrations: an analysis of
published data. Br Med ] 298:784-788,
1989
8. Facchini FS, Hollembeck CB, JeppesenJ,
Ida Chen YD, Reaven GM: Insulin resis-
tance and cigarette smoking. Lancet 1992:
339:1128-1130
9. Reaven GM: Treatment of hypertension:
focus on prevention of coronary heart dis-
ease. J CHn Endocnnol Metab 76:537-540,
1993
10. Freestone S, Ramsay LE: Effect of coffee
and cigarette smoking on the blood pres-
sure of untreated and diuretic-treated hy-
pertensive patients. Am J Med 73:348-
353,1982
11. Stamford BA, Matter S, Fell RD, Sady S,
Papanek P, Cresanta M: Cigarette smok-
ing, exercise and high density lipoprotein
cholesterol. Atherosclerosis 52:73-83,
1984
12. Swank AM, Fell RD: Effect of acute smok-
ing and exercise on high-density lipopro-
tein-cholesterol and subfractions in black
female smokers. Metabolism 39:343-348,
1990
13. Attvall S, FowelinJ, Lager I, Von Schenck
H, Smith U: Smoking induces insulin
resistance: a potential link with the in-
sulin resistance syndrome. J Intern Med
233:327-332, 1993
14. Cederholm J, Wibell L: Insulin release
and peripheral sensitivity at the oral glu-
cose tolerance test. Diabetes Res Clin Pract
10:167-175, 1990
15. Doberne L, Greenfield MS, Rosenthal M,
Widstrom A, Reaven GM: Effect of varia-
tions in basal plasma glucose concentra-
tion on glucose utilization 0 and meta-
bolic clearance (.MCR") rates during
insulin clamp studies in patients with
non-insulin-dependent diabetes mellitus.
Diabetes 31:396-400, 1982
16. Lindhal B, Asplund K, Hallmans G: High
scrum insulin, insulin resistance and their
associations with cardiovascular risk lac-
tors: the Northern Sweden MONICA
population study. / Intern Med ?. W:.'.6 V
270,1993
17. Cryer PE, Haymond M\V, Santiago )V,
Shah SD: Norepinephrine and epincph
rine release and adrenergic mediation of
smoking-associated hemodynamic and
metabolic events. N t:.n$l J Med 295:'->7 V
577,1976
18. Trap-Jensen J: Effect of smoking on heart
and peripheral circulation. Am Hcait /
115:263-256,1988
19. Minaker KL, Rowe J\V, Young JB, Spar-
row D, Pallotta JA, Landsberg I.: Meet o\
age on insulin stimulation of sympathetic
nervous system activity in man. Metabo
Item 21:1181-1184, 1982
20. Frati AC, Ariza CR, Martine.-.-Cedillo J,
Altamirano P: Efecto de la hiperinsuline
mia aguda en la tension arterial de indi
viduos sanos. Anh lnst Cardiol Mex 60:
571-576, 1990
21. Eliasson M, l.undblad D, Hagg F: Cardio-
vascular risk factors in young snuff users
and cigarette smokers. / Intern Med ?A0:
17-22, 1991
22. Janzon L, Berntorp K, Jansson M, l.indell
SE, Trell E: Glucose tolerance and smok-
ing: a population study oi oral and in-
travenous glucose tolerance test in mid-
dle-aged men. Diabetologia 25:86-88,
1983
23. Eliasson B, Attvall S, Taskinen MR, Smith
U: The insulin resistant syndrome is re-
lated to smoking habits. Arfniosi KT
Thromb 14:1946-1950, 1994
24. Horowitz M, Eraser R: Disordered gastric
motor function in diabetes mellitus. Phi-
betologia 37:543-551, 1994
25. Polonsky KS, Given BD, Hirsch l.J, lillil
H, Shapiro ET, Beebe C, Frank RH, Gal
lowayJA, Van Canter H: Abnormal pat-
terns of insulin secretion in non insulin
dependent diabetes mellitus. N' lingl J Med
318:1231-1239, 1988
26. Frati-Munari AC, Roca-Vidcs RA, Lope;:-
Perez RJ, De Vivero 1, Rui;:-Vela;:co M:
Indices glucemicos de algunos alimentos
comunes en Mexico. Cac Med Mex \?.Y:
163-170, 1991
27. World Heath Organization: Piabetes Mi1!-
litus: Report of a WHO Study Croup. Ge
neva, World Health Org., 1985 OVch.
Rep. Ser., no. 727)
28. Rull JA: Definicion, elasilicaeion y diat;
nostieo de la diabetes melliius: analisis de
los criterios del Grupo Nacional de Datos

DlABMHS CAR!-:, VOLUME 1 9 , NUMBER 2 , FEBRUARY 1 9 9 6 117

Smoking and glucose tolerance
dc Diabetes y de la Organization Mundial
de la Salud. In Diabetes Mellitus. Ariza CR,
Frati AC, Eds. interamericana-McGraw-
Hill, Mexico City, Mexico, 1993, p. 621
641
29. Howell RW: Smoking habits and labora-
tory tests. Ixincet ii: 152, 1970
30. Tucker LA: Use of smokeless tobacco, cig-
arette smoking, and hypercholesterol-
emia. Am] Public Health 79:1048-1050,
1989
31. Wynder EL, Harris RE, Haley NJ: Popula-
tion screening for plasma cholesterol:
community-based results from Connecti-
cut. Am Heart J 117:649-656, 1989
32. Gnasso A, Haberbosch W, Schettler G,
Schmitz G, AugustinJ: Acute influence of
smoking on plasma lipoproteins. Klin
Wochenschr 1984:62 (Suppl. 2):36-42
33. Quensel M, Agardh CD, Nilsson-Ehle
P: Nicotine does not affect plasma li-
poprotein concentrations in healthy men.
.Saint/J Clin Lab Invest 49:149-153,
1989
34. Mjos OD: Lipid effects of smoking. Am
118
Heart] 115:272-275, 1988
35 Elkeles RS, Khan SR, Cowdhury S,
Swallow MB: Effects of smoking on oral
fat tolerance and high density lipopro-
tein cholesterol. Clin Sci 65:669-672,
1983
36. Carney RM, Goldberg AP: Weight gain af-
ter cessation of cigarette smoking. N Engl]
Med 310:614-616, 1984
37. Aronow WS, Cassidy J, Vangrow JS,
March H, KernJC, Goldsmith JR, Khemka
M, Pagano J, Vawter M: Effect of cigarette
smoking and breathing carbon monoxide
on cardiovascular hemodynamic in angi-
nal patients. Circulation 50:340-347,
1974
38. Freestone S, Ramsay LE: Effect of coffee
and cigarette smoking on the blood pres-
sure of untreated and diuretic-treated hy-
pertensive patients. Am J Med 73:348-
353,1982
39. Robertson D, Tseng GJ, Appalsamy M:
Smoking and mechanisms of cardiovas-
cular control. Am Heart J 115:258-262,
1988
DIABETES CARE, VOLUME 19,
40. Benowitz NL: Pharmacologic aspects of
cigarette smoking and nicotine addiction.
NEnglJ Med 319:1318-1330, 1988
41. Green MS, Jucha E, Luz Y: Blood pressure
in smokers and non smokers: epidemio-
logic findings. Am Heart] 111:932-940,
1986
42. Mann SJ, James GD, Wang RS, Pickering
TG: Elevation of ambulatory systolic
blood pressure in hypertensive smokers: a
case-control study. JAMA 265:2226-
2228,1991
43. Reaven GM: Role of insulin resistance in
human disease. Diabetes 37:1595-1607,
1988
44. Harats D, Dabach Y, Hollander G, Ben-
Nairn M, Schwartz R, Berry EM, Stein O,
Stein Y: Fish oil ingestion in smokers and
non smokers enhances peroxidation of
plasma lipoproteins. Atherosclerosis 90:
127-139, 1991
45. Fitzgerald GA, OatesJA, NowakJ: Ciga-
rette smoking and hemostatic function.
Am Heart] 115:267-271, 1988
NUMBER 2, FEBRUARY 1996