from around the world ● focus on China Combined Effects of Tobacco Smoke Exposure and Metabolic Syndrome on Cardiovascular Risk in Older Residents of China Yao He, MD, PHD,* Tai Hing Lam, MD,Bin Jiang, MD, PHD,† Jie Wang, MD, PHD,‡ Xiaoyong Sai, MD, PHD,* Li Fan, MD,§ Xiaoying Li, MD,§ Yinhe Qin, MD,* Frank B. Hu, MD, PHD¶ * Institute of Geriatrics, Chinese PLA General Hospital, Beijing, China; † Department of Acupuncture, Chinese PLA General Hospital, Beijing, China; ‡ Clinic of PLA Communication Department, Beijing, China; § Department of Geriatric Cardiology, Chinese PLA General Hospital, Beijing, China; School of Public Health and Department of Community Medicine, The University of Hong Kong, Hong Kong, China; and the ¶ Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts I n the 1996 National Prevalence Survey of Smok- ing Pattern in mainland China, the prevalence of smoking was 66.9% in men and 4.2% in women; overall, 53.5% reported passive smoke expo- sure. More than 72% of all Chinese (600 million people) are regularly exposed to either active smoking or secondhand smoke (SHS) (1). As the world’s larg- est producer and consumer of tobacco products, China bears a substantial burden of smoking-related diseases and deaths (2). It is estimated that tobacco- related deaths will reach 7 million worldwide in 2025, with 2 million of these deaths predicted to occur in China (1,3). Tobacco smoking is a major risk factor for cardiovascular disease (CVD) (4–7) and type 2 diabetes (8,9). Moreover, there is strong evidence that passive smoking or SHS is causally associated with coronary heart disease (CHD) (10 –13) and may be linked to stroke (14 –16). The metabolic syndrome (MetS) is characterized by a clustering of cardiovascular risk factors, including abdominal obesity, high blood pressure, increased glucose level, and dyslipidemia. The MetS is associ- ated with the development of diabetes and CVD (17,18) as well as an increased risk of CVD and all- cause mortality (19,20). In China, it has become an important public health problem with the prevalence of MetS being 13.7% in the middle-aged (21) and 46.3% in the elderly (22) populations. There is also evidence that SHS is associated with both MetS and insulin resistance (8,23–28). Although SHS has been examined as a risk factor for individual CVD (CHD or stroke), it has not been investigated together with MetS in the Chinese population. In this study, we examined in detail the individual and combined ef- fects of active smoking, SHS, and MetS on cardiovas- cular risk in a population-based survey of elderly Chi- nese in Beijing, China. Methods Study population. We conducted a population- based cross-sectional survey of individuals age 60 years or older in the Wanshoulu Community within the Haidian District, a metropolitan area representative of the geographic and economic characteristics of Bei- jing, China. A 2-stage stratified sampling method was used. First, 9 residential communities (totaling approximately 300 to 600 households each) were se- lected randomly from a total of 94 residential com- munities in the Wanshoulu Area. Second, all individ- uals were chosen from selected streets within these communities, but only 1 participant from each house- hold was invited to enroll in the study. From April 2001 to March 2002, 2,680 people 60 years of age or older were selected and invited for screening. A total
Transcript
from● f
CSo
YXF*
†
‡
§
�
H¶
IwspoeCdrwCfdpcl
baga(cio
around the worldocus on China
ombined Effects of Tobaccomoke Exposure and Metabolic Syndromen Cardiovascular Risk in Older Residents of China
ao He, MD, PHD,* Tai Hing Lam, MD,� Bin Jiang, MD, PHD,† Jie Wang, MD, PHD,‡iaoyong Sai, MD, PHD,* Li Fan, MD,§ Xiaoying Li, MD,§ Yinhe Qin, MD,*rank B. Hu, MD, PHD¶Institute of Geriatrics, Chinese PLA General Hospital, Beijing, China;Department of Acupuncture, Chinese PLA General Hospital, Beijing, China;Clinic of PLA Communication Department, Beijing, China;Department of Geriatric Cardiology, Chinese PLA General Hospital, Beijing, China;School of Public Health and Department of Community Medicine, The University of Hong Kong,ong Kong, China; and the
Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts
4eieoMefcn
M
SboHtjwalmuch2
n the 1996 National Prevalence Survey of Smok-ing Pattern in mainland China, the prevalence ofsmoking was 66.9% in men and 4.2% in
omen; overall, 53.5% reported passive smoke expo-ure. More than 72% of all Chinese (�600 millioneople) are regularly exposed to either active smokingr secondhand smoke (SHS) (1). As the world’s larg-st producer and consumer of tobacco products,hina bears a substantial burden of smoking-relatediseases and deaths (2). It is estimated that tobacco-elated deaths will reach 7 million worldwide in 2025,ith 2 million of these deaths predicted to occur inhina (1,3). Tobacco smoking is a major risk factor
or cardiovascular disease (CVD) (4–7) and type 2iabetes (8,9). Moreover, there is strong evidence thatassive smoking or SHS is causally associated withoronary heart disease (CHD) (10–13) and may beinked to stroke (14–16).
The metabolic syndrome (MetS) is characterizedy a clustering of cardiovascular risk factors, includingbdominal obesity, high blood pressure, increasedlucose level, and dyslipidemia. The MetS is associ-ted with the development of diabetes and CVD17,18) as well as an increased risk of CVD and all-ause mortality (19,20). In China, it has become anmportant public health problem with the prevalence
f MetS being 13.7% in the middle-aged (21) and o
6.3% in the elderly (22) populations. There is alsovidence that SHS is associated with both MetS andnsulin resistance (8,23–28). Although SHS has beenxamined as a risk factor for individual CVD (CHDr stroke), it has not been investigated together withetS in the Chinese population. In this study, we
xamined in detail the individual and combined ef-ects of active smoking, SHS, and MetS on cardiovas-ular risk in a population-based survey of elderly Chi-ese in Beijing, China.
ethods
tudy population. We conducted a population-ased cross-sectional survey of individuals age 60 yearsr older in the Wanshoulu Community within theaidian District, a metropolitan area representative of
he geographic and economic characteristics of Bei-ing, China. A 2-stage stratified sampling methodas used. First, 9 residential communities (totaling
pproximately 300 to 600 households each) were se-ected randomly from a total of 94 residential com-
unities in the Wanshoulu Area. Second, all individ-als were chosen from selected streets within theseommunities, but only 1 participant from each house-old was invited to enroll in the study. From April001 to March 2002, 2,680 people 60 years of age or
lder were selected and invited for screening. A total
otfDt
GptAaecetfssbsbodDDstIcnbofd�4(
�
cenaMea
gcfilCSmttsrwatBacIsblTS
R
Aiacw6c
twmS13susu
M
364 He et al. JACC Vol. 53, No. 4, 2009From Around the World January 27, 2009:363–71
f 2,334 subjects (943 men and 1,391 women) completedhe survey, yielding a response rate of 87.1%, accountingor 11.4% of the older residents in the Wanshoulu Area.etails of data collection, physical examinations, and other
ests have been reported elsewhere (22).The Committee for Medical Ethics of the Chinese PLA
eneral Hospital approved the study in 2001. Each partici-ant signed an informed consent form before completinghe questionnaire.
ctive smoking and SHS exposure. The categories ofctive smoking were: ever, former, and current smoker. Anver smoker was defined as one who had smoked at least 1igarette daily for 1 year or more. Current smokers werever smokers who were still smoking at the time of the in-erview, and former smokers were those who had stoppedor at least 2 years. Individuals with no history of tobaccomoking were considered never smokers. Pack-years ofmoking were calculated by multiplying the average num-er of cigarettes smoked per day by the number of years ofmoking and dividing by 20. SHS exposure was defined aseing exposed to another person’s tobacco smoke at homer in the workplace for at least 15 min daily more than 1ay per week in the past 10 years (1).efinitions. We defined MetS by using the Internationaliabetes Foundation (IDF) criteria of 2005 (29), which are
imilar to the modified U.S. National Cholesterol Educa-ion Program’s Adult Treatment Panel III criteria (30). TheDF definition of MetS includes central obesity (waist cir-umference �90 cm in Chinese men and �80 cm in Chi-ese women) plus any 2 of the following 4 factors: 1) highlood pressure: systolic �130 mm Hg, diastolic �85 mm Hg,r known treatment for hypertension; 2) hypertriglyceridemia:asting plasma triglycerides �1.7 mmol/l; 3) low high-ensity lipoprotein cholesterol (HDL-C): fasting HDL-C1.0 mmol/l in men and �1.3 mmol/l in women; and
) hyperglycemia: fasting glucose level of �5.6 mmol/l�100 mg/dl) or known treatment for diabetes.
Hypertension was defined as diastolic blood pressure of90 mm Hg, systolic blood pressure of �140 mm Hg, or
urrent medication for hypertension. Coronary heart dis-ase and stroke were defined using the World Health Orga-ization’s MONICA (Multinational Monitoring of Trendsnd Determinants in Cardiovascular Disease) criteria (31).
yocardial infarction was defined by a representative set oflectrocardiogram (ECG) changes, cardiac enzyme values,
nd typical symptoms. Angina was defined as use of nitro- (
lycerin, experience of typical chest pain, or ECG changesompatible with ischemic heart disease. Strokes were de-ned as events requiring hospitalization and confirmed by
ocal hospital records. CVD was defined by the presence ofHD and/or stroke.tatistical analysis. Data were entered (double entry) andanaged by Access (Microsoft Corp., Redmond, Washing-
on). The Wald chi-square analysis was used to determinehe sex-specific prevalence of MetS by tobacco exposuretatus. Logistic regression was used to estimate the oddsatios (ORs) and 95% confidence interval (CI) of CVDsith adjustment for age, marital status, years of education,
lcohol consumption, hours per day of physical exercise,otal cholesterol, and family history of CHD or stroke.oth stratified analyses and multiple logistic regressionnalyses were used to examine the independent andombined effects of the tobacco exposure and MetS.nteraction between tobacco exposure groups and MetStatus was investigated by adding interaction terms (to-acco exposure status multiplied by MetS status) to the
ogistic regression model using the likelihood ratio test.he statistical package used was SPSS (version 14.0,PSS Inc., Chicago, Illinois).
esults
mong the study participants, there were 784 CHD patients,ncluding 68 with myocardial infarction and 716 with stablengina. Among participants with CHD, 58% of the cases wereonfirmed by exercise test or B ultra-sound heart image. Thereere 378 cases of stroke (248 ischemic, 70 hemorrhagic, and0 other types), and 83% of these events were confirmed byomputed tomography and magnetic resonance imaging.
On the basis of the general characteristics of study par-icipants, we found significant differences between men andomen (Table 1). The prevalence of ever smoking wasuch greater in men (58.1%) than in women (13.1%), but
HS exposure was more likely in women (42.3% vs.9.9%). On the basis of IDF criteria, MetS was present in4.8% of men and 54.1% of women. Compared with nevermokers without exposure to SHS, the prevalence of individ-al MetS components and MetS was greater in male activemokers as well as in female active and passive smokers, partic-larly, for low HDL-C and central obesity (Table 2).
Compared with never smokers without MetS, men withetS had significantly greater risk of CVD among never
OR: 1.83; 95% CI: 1.17 to 2.86), former (OR: 2.24; 95%
CCMOseaoht
sia0tws
s
esterol
365JACC Vol. 53, No. 4, 2009 He et al.January 27, 2009:363–71 From Around the World
I: 1.39 to 3.63), and current smokers (OR: 3.54; 95%I: 2.03 to 6.18) (Table 3). Among the subjects withetS, if current smokers were used as the reference, theRs of CVD were 0.63 (95% CI: 0.34 to 0.99) for former
mokers and 0.53 (95% CI: 0.28 to 0.94) for never smok-rs, suggesting that smoking cessation reduces CVD risk bypproximately 37% in male smokers with MetS. The ORsf CHD and stroke in current smokers with MetS were theighest among the 6 groups and were 4.57 (95% CI: 2.57
o 8.12) for CHD and 2.41 (95% CI: 1.24 to 4.38) for o
troke. Multivariate logistic models showed no significantnteraction between active smoking (never/former/current)nd MetS on CHD, stroke, and CVD (p � 0.69, p �
.74, and p � 0.57, respectively). In contrast, no associa-ion between active smoking and CVD risk was found inomen, probably the result of the small number of active
moking women.There was a clear dose-response relationship between
moking consumption (pack-years) and the adjusted ORs
ticipants
Women (n � 1,391) p Value
66.6 (6.8) �0.001
25.8 (3.7) 0.002
137.6 (21.9) 0.073
76.8 (10.5) 0.077
5.5 (1.0) �0.001
1.7 (1.1) �0.001
1.4 (0.5) �0.001
3.4 (1.1) �0.001
6.2 (2.0) 0.11
86.1 (9.2) �0.001
708 (50.9) �0.001
483 (34.7)
200 (14.4)
1,094 (78.8) �0.001
7 (0.5)
290 (20.9)
389 (28.0) 0.04
510 (36.7)
491 (35.3)
1,209 (86.9) �0.001
62 (4.5)
120 (8.6)
589 (42.3) �0.001
71 (5.1) �0.001
752 (54.1) �0.001
700 (50.3) 0.13
214 (15.4) 0.86
492 (35.4) 0.03
208 (15.0) 0.05
741 (53.3) 0.001
; LDL-C � low-density lipoprotein cholesterol.
y Par
943)
5.6)
3.2)
20.0)
10.2)
2.3)
0.8)
0.3)
0.8)
1.7)
9.2)
27.4)
37.0)
35.6)
92.6)
0.4)
7.0)
24.0)
41.3)
34.8)
42.1)
33.2)
24.7)
19.9)
30.0)
34.8)
47.1)
15.7)
31.0)
18.0)
46.6)
f CHD and CVD in all 943 men with or without MetS
(C
tnHspo1t2sC(
CsrwtbC(CMfwmM
p � 0.0
*o
366 He et al. JACC Vol. 53, No. 4, 2009From Around the World January 27, 2009:363–71
Table 4, Fig. 1). The test for trend was significant forHD and CVD but not for stroke.
Among never smokers (397 men and 1,209 women),he ORs of CHD, stroke, and CVD with SHS in men wereot significantly different based on MetS status (Table 5).owever, in women, SHS was associated with CHD,
troke, and CVD in subjects with or without MetS. Com-ared with those without passive smoke exposure and with-ut MetS, the adjusted OR of CVD was 1.74 (95% CI:.27 to 2.38) in those with MetS only, 1.72 (95% CI: 1.20o 2.47) in those exposed to SHS only, and 2.91 (95% CI:.03 to 4.16) for those with MetS and exposure to passivemoke. The interaction terms between SHS and MetS forHD, stroke, and CVD were not statistically significant
p � 0.66, p � 0.99, and p � 0.66, respectively).
Table 2 Prevalence of MetS and its Components by Tobacco
Men
Non-SHSin Never Smokers
(n � 333)
SHSin Never Smokers
(n � 64)
MetS components (%)
High blood pressure 71.0 67.3
Hyperglycemia 51.7 49.4
Hypertriglyceridemia 22.0 19.9
Low HDL-C 14.6 13.7
Central obesity 47.7 43.3
MetS 31.6 30.2
value for comparing with the group of non-secondhand smoke (SHS) in never smokers: �pMetS � metabolic syndrome; other abbreviations as in Table 1.
Table 3 ORs of CHD, Stroke, and CVD by Smoking and MetS
OR* (95% CI)
Never
MetS(�)(M � 268, F � 554)
MetS(�)(M � 129, F � 655)
MetS(�(M � 185, F
CHD
Male (n � 292) 1.00 (Ref) 1.87 (1.14–3.06) 1.95 (1.24–
Adjusted for age (years), marital status, education (years: �6, 7–12, �13), exercise (h/day
f coronary heart disease (CHD) or stroke.CI � confidence interval; CVD � cardiovascular disease; F � female; M � male; OR � odds rat
Odds ratios were determined for CHD, stroke, andVD by tobacco exposure status (non-SHS in the never
mokers, SHS in the never smokers, and active smokingegardless of SHS) and MetS status (Table 6). Comparedith never smokers without MetS and SHS, the exposure
o SHS in women, active smoking in men, and MetS inoth men and women were associated with increased risk ofVD. Those with MetS and exposure to tobacco smoke
passive or active) had greater risks of CHD, stroke, andVD than those exposed to tobacco smoke but withoutetS. For example, the ORs of CHD were 2.94 in the
emale passive smokers with MetS versus 1.73 in femalesith exposure to tobacco smoke but without MetS. Amongen, the OR of CHD was 3.18 for active smokers withetS versus 1.57 for men exposed to SHS without MetS.
–3, �4), alcohol drinking (current drinkers vs. current nondrinkers), SHS, and family history
Exp
ASm
(n
7
5
2
1
5
3
and
by Act
Fo
)� 25)
3.05)
2.77)
2.00)
3.69)
2.40)
2.23)
: �1, 1
io; Ref � reference group; other abbreviations as in Tables 1 and 2.
Aat1
Ttstswp
D
TpbbwvcrmfecHc
s
FwdbajhwSr
snmspittp
stbaspca
*
367JACC Vol. 53, No. 4, 2009 He et al.January 27, 2009:363–71 From Around the World
dditionally, the ORs of CHD, stroke, and CVD in thective female smokers with MetS were 1.75 (95% CI: 1.07o 2.88), 3.11 (95% CI: 1.78 to 5.32), and 2.12 (95% CI:.34 to 3.48), respectively.
The p values for the interaction analysis can be seen inables 3 and 5. There was no significant interaction be-
ween active smoking and MetS among the men or of pas-ive smoking and MetS in men and women. However,here was a significant positive interaction between activemoking and MetS on CHD, stroke, and CVD among theomen in this study (p � 0.005, p � 0.002, and� 0.001, respectively) (Table 3).
iscussion
o the best of our knowledge, this is the first study to re-ort the joint effects of tobacco smoke exposure and meta-olic syndrome on cardiovascular risk in a population-ased study in China. We focused on Chinese men andomen age 60 years or older, a group at higher risk of de-eloping CVD. The sample was randomly selected from aluster sample of similar residential communities and theesponse fraction was relatively high, with only approxi-ately 13% of eligible subjects having moved or otherwise
ailed to complete the interviews or examinations. How-ver, there were no statistically significant differences in theharacteristics between the responders and nonresponders.ence, the prevalence of CVD and associated risk factors
an be generalized to similar populations in urban Beijing.In the present study, MetS was associated with CHD and
Table 4 Smoking Consumption and OR (95% CI) of CHD, Str
MetabolicStatus
SmokingConsumption(Pack-Yrs)
CHD
OR* (95% CI) OR† (95% CI)
No MetS 0 1.00 —
1–20 1.18 (0.74–1.89) 1.19 (0.73–1.92)
21–40 1.52 (0.93–2.47) 1.60 (0.97–2.62)
�40 1.89 (1.09–3.28) 2.08 (1.18–3.67)
p value for trend 0.012 0.003
MetS 0 1.64 (1.03–2.63) 1.68 (1.04–2.70)
1–20 2.21 (1.30–3.76) 2.27 (1.33–3.89)
21–40 2.46 (1.42–4.26) 2.57 (1.46–4.52)
�40 3.19 (1.62–6.32) 3.60 (1.79–7.23)
p value for trend �0.001 �0.001
Adjusted for age (years). †Adjusted for variables as defined in Table 3.Abbreviations as in Tables 1 and 3.
troke, which is consistent with previous findings (17–20,32). s
urthermore, active smoking in men and exposure to SHS inomen were strongly associated with increased CVD risk in-ependent of MetS. There was a dose-response relationshipetween pack-years of active smoking and CVD among mennd this difference was evident regardless of MetS status. Sub-ects with MetS exposed to tobacco smoke (active or passive)ad greatly increased CVD risk by 48% to 193% comparedith never smokers without MetS and without exposure toHS. It is likely that smoking cessation could substantiallyeduce the risk of CVD in smokers with MetS.
We found no significant interaction between activemoking and MetS in relation to CVD in men and no sig-ificant interaction between passive smoking and MetS inen and women. Thus, the effects of tobacco smoke expo-
ure and MetS on the risk of cardiovascular outcomes ap-ear to be additive in the present study. The significantnteraction between active smoking and MetS in relation tohe risk of CVD in women suggests a synergistic effect, buthis could be a chance finding due to a very low smokingrevalence and fewer CVD events among women.
Our results are consistent with several recent studieshowing that smoking chronically induces insulin resis-ance, with a dose-response relationship between the num-er of cigarettes smoked and the degree of insulin resistancend metabolic abnormalities (28,33). There is evidence thatmoking cessation improves insulin sensitivity (34) whereasassive smoking reduces insulin sensitivity (35) and in-reases risk of MetS (27). Because both tobacco smokend MetS are independently associated with insulin re-
istance and because insulin resistance may contribute to
alc
tt
p2oIlolistsf
lhpscwfStesadftd
wolrepetwcp
wys
368 He et al. JACC Vol. 53, No. 4, 2009From Around the World January 27, 2009:363–71
ccelerated atherosclerosis, these 2 risk factors may beinked through this common pathophysiology to in-reased risk of CVD.
In the past 30 years, the prevalence of traditional risk fac-ors for CVD (i.e., cigarette smoking, dyslipidemia, and hyper-
Figure 1Dose-Response Relation Between SmokingConsumption and Adjusted ORs of CHD, Stroke,and CVD in Men With or Without MetS
Blue diamonds � without metabolic syndrome (MetS); open squares � withMetS. CHD � coronary heart disease; CVD � cardiovascular disease; OR �
odds ratio.
ension) has gradually declined in the U.S., with smoking s
revalence decreasing from 50% in 1960 to about 23% in000. However, during this same period, the prevalence ofbesity and MetS has increased dramatically in the U.S. (36).n China, smoking rates remain high (1) and cessation ratesow (9.5% of all ever smokers have quit) (37), whereas beingverweight and MetS are rapidly increasing with changes inifestyle (e.g., excessive caloric intake and lack of physical activ-ty) (21,22). It is likely that the growing double epidemics ofmoking and MetS in China will greatly and rapidly increasehe burden of CVD from traditional risk factors (such asmoking, SHS, hypertension) acting together with new riskactors (obesity, MetS, and diabetes).
Considering that both tobacco use and MetS are the 2eading causes of preventable death in China, our findingsave important public health implications. Our results em-hasize that public health policies are needed to controlmoking, SHS, and MetS. In terms of the implications forlinical practice, smokers or those exposed to passive smokeith MetS have very high CVD risk and should be targeted
or intensive interventions.tudy limitations. The present study has several limita-
ions. The cross-sectional nature of the data limits causal infer-nces. The underreporting of smoking, including history ofelf-reported active and passive smoking, could also bias thessociation between smoking and CVD towards null. Theata on smoking status should not have been affected by dif-erential reporting bias in CVD cases versus noncases becausehey were collected before physical examination or the diseasesefined by hospitalization.
The relatively low prevalence of SHS exposure (42.3%) inomen could be attributable to 2 reasons. First, the prevalencef current smoking in elderly men in our study was muchower (24.7%) than that in the general population (1), thuseducing the exposure to SHS in women in close contact tolderly men. Second, exposure to SHS was restricted toassive smoking in the past 10 years to minimize recallrror, but this means fewer opportunities to be exposedo SHS during this limited time period compared withomen in the general population who may have had
onsiderable exposure to SHS earlier than the 10-yeareriod analyzed in this study.
To reduce the bias of “ill quitter effect,” former smokersere defined as those who had stopped smoking for at least 2ears. Furthermore, the pattern of smoking cessation in ourubjects is consistent with other Chinese data (37,38). Some
ubjects could have died young from smoking- or other CVD-
rtoueofpsmStpw
trSosdow
C
Oe
*
*
369JACC Vol. 53, No. 4, 2009 He et al.January 27, 2009:363–71 From Around the World
elated risk factors before the survey, making our participantshe survivors. It is likely that the effects of smoking and MetSn CVD are stronger in younger people than in older individ-als. Additionally, limiting SHS to a 10-year period mightxplain in part the lower prevalence (19.9%) of SHS in theselder men; they are likely retired and have few opportunitiesor work-related SHS exposure. Similarly, because of the lowrevalence of current smoking among women, the men in ourtudy were less likely to be exposed to SHS in the home. Thisay explain why we didn’t find a significant association of
HS and CVD in men. We observed no significant associa-ion between active smoking and risks of CVD in womenrobably due to their low prevalence of active smoking amongomen.
Table 5 OR (95% CI) of CHD, Stroke, and CVD by SHS, MetS
OR (95% CI) by SH
Non-SHS
MetS(�)(M � 220, F � 327)
MetS(�)(M � 113, F � 406)
CHD
Male (n � 90) 1.00 (Ref) 1.80 (1.06–3.06)
Female (n � 431) 1.00 (Ref) 1.76 (1.27–2.45)
Stroke
Male (n � 61) 1.00 (Ref) 1.23 (0.66–2.29)
Female (n � 172) 1.00 (Ref) 1.91 (1.18–3.09)
CVD (CHD/stroke)
Male (n � 137) 1.00 (Ref) 1.76 (1.08–2.86)
Female (n � 516) 1.00 (Ref) 1.74 (1.27–2.38)
Adjusted for variables as defined in Table 3 but not including SHS.Abbreviations as in Tables 1 to 3.
Table 6 OR (95% CI) of CHD, Stroke, and CVD by Tobacco E
OR (95
Non-SHS in Never Smokers
MetS(�)(M � 223, F � 326)
MetS(�)(M � 110, F � 405) (M �
CHD
Male (n � 292) 1.00 (Ref) 1.91 (1.13–3.24) 1.57
Female (n � 492) 1.00 (Ref) 1.76 (1.27–2.45) 1.73
Stroke
Male (n � 170) 1.00 (Ref) 1.30 (0.72–2.35) 0.70
Female (n � 208) 1.00 (Ref) 1.91 (1.18–3.09) 1.85
CVD (CHD/stroke)
Male (n � 390) 1.00 (Ref) 1.84 (1.13–2.97) 1.04
Female (n � 596) 1.00 (Ref) 1.74 (1.27–2.38) 1.73
Adjusted for variables as in Table 3, but not including SHS.Abbreviations as in Tables 1 to 3.
When considering all these factors as well as the poten-ial for misclassification in ascertaining SHS exposure, theelationship between tobacco smoke exposure (especiallyHS), MetS, and CVD is likely to be underestimated inur study. Finally, because of small numbers of formermokers and the lack of detailed information about theuration of SHS, a dose-response relationship of durationf quitting in men and the quantity of SHS exposure inomen could not be analyzed.
onclusions
ur study shows that cigarette smoking in men and SHSxposure in women are highly prevalent in elderly Chinese.
370 He et al. JACC Vol. 53, No. 4, 2009From Around the World January 27, 2009:363–71
obacco smoke exposure (including active and passivemoking) increased the prevalence of CHD, stroke, andVD in the subjects with or without MetS after adjusting
or other CVD risk factors. These findings collectively sug-est that cigarette smoking and SHS, 2 modifiable risk fac-ors, are strongly associated with increased CVD indepen-ent of MetS. The growing double epidemics of MetS andobacco smoke exposure (both active smoking and SHS)ill dramatically increase the risk of CVD in the Chineseopulation.
lease note: This study is supported by research grants from the National Naturalcience Foundation of China (30671809, 30771859), Beijing Natural Scienceoundation (7062063) and the Ministry of Health of China (06H050), Beijingedical Scientific Development Foundation (2007-2039), and the Ministry of
cience and Technology of China (2006BAI01A01). Dr. He was partly supportedy the Cheng Yu Tung and Gordon Wu Exchange Professorship in the Faculty ofedicine at the University of Hong Kong.
eprint requests and correspondence: Dr. Yao He, Institutef Geriatrics, Chinese PLA General Hospital, 28 Fuxing Road,eijing 100853, China. E-mail: [email protected].
EFERENCES
1. Yang G, Fan L, Tan J, et al. Smoking in China: findings of the1996 National Prevalence Survey. JAMA 1999;282:1247–53.
2. Murray CJ, Lopez AD, Jamison DT. The global burden of diseasein 1990: summary results, sensitivity analysis and future directions.Bull World Health Organ 1994;72:495–509.
3. Peto R, Lopez AD, Boreham J, Thun M, Heath C Jr., Doll R.Mortality from smoking worldwide. Br Med Bull 1996;52:12–21.
4. Yuan JM, Ross RK, Wang XL, Gao YT, Henderson BE, Yu MC.Morbidity and mortality in relation to cigarette smoking in Shang-hai, China. JAMA 1996;275:1646–50.
5. Lam TH, He Y, Li LS, He SF, Liang BQ. Mortality attributableto cigarette smoking in China. JAMA 1997;278:1505–8.
6. Chen Z-M, Xu Z, Collins R, Li W-X, Peto R. Early health effectsof the emerging tobacco epidemic in China. JAMA 1997;278:1500–4.
7. Liu B-Q, Peto R, Chen Z-M, et al. Emerging tobacco hazards inChina: retrospective proportional mortality study of one milliondeaths. BMJ 1998;317:1411–22.
8. Kong C, Nimmo L, Elatrozy T, et al. Smoking is associated withincreased hepatic lipase activity, insulin resistance, dyslipidaemiaand early atherosclerosis in type 2 diabetes. Atherosclerosis 2001;156:373–8.
9. Will JC, Galuska DA, Ford ES, Mokdad A, Calle EE. Cigarettesmoking and diabetes mellitus: evidence of a positive association froma large prospective cohort study. Int J Epidemiol 2001;30:540–6.
0. Steenland K, Sieber K, Etzel RA, Pechacek T, Maurer K. Exposureto environmental tobacco smoke and risk factors for heart diseaseamong never smokers in the Third National Health and NutritionExamination Survey. Am J Epidemiol 1998;147:932–9.
1. He J, Vupputuri S, Allen K, Prerost MR, Hughes J, Whelton PK.Passive smoking and the risk of coronary heart disease—a meta-
analysis of epidemiologic studies. N Engl J Med 1999;340:920–6.
2. He Y, Lam TH, Li LS, et al. Passive smoking at work as a riskfactor for coronary heart disease in Chinese women who have neversmoked. BMJ 1994;308:380–4.
3. Whincup PH, Gilg JA, Emberson JR, et al. Passive smoking andrisk of coronary heart disease and stroke: prospective study withcotinine measurement. BMJ 2004;329:200–5.
4. Zhang X, Shu XO, Yang G, et al. Association of passive smokingby husbands with prevalence of stroke among Chinese womennonsmokers. Am J Epidemiol 2005;161:213–8.
5. Qureshi AI, Suri MF, Kirmani JF, Divani AA. Cigarette smokingamong spouses: another risk factor for stroke in women. Stroke2005;36:e74–6.
6. McGhee SM, Ho SY, Schooling M, et al. Mortality associated withpassive smoking in Hong Kong. BMJ 2005;330:287–8.
8. Isomaa B, Almgren P, Tuomi T, et al. Cardiovascular morbidityand mortality associated with the metabolic syndrome. DiabetesCare 2001;24:683–9.
9. Trevisan M, Liu J, Bahsas FB, Menotti A. Syndrome X andmortality: a population-based study. Risk factor and life expectancyresearch group. Am J Epidemiol 1998;148:958–66.
0. Lakka HM, Laaksonen DE, Lakka TA. The metabolic syndromeand total and cardiovascular disease mortality in middle-aged men.JAMA 2002;288:2709–16.
1. Gu D, Reynolds K, Wu X, et al. Prevalence of the metabolic syndromeand overweight among adults in China. Lancet 2005;365:1398–405.
2. He Y, Jiang B, Wang J, et al. Prevalence of the metabolic syndromeand Its relation to the risk of cardiovascular disease in an elderlyChinese population. J Am Coll Cardiol 2006;47:1588–94.
3. Reaven G, Tsao PS. Insulin resistance and compensatory hyperin-sulinemia: the key player between cigarette smoking and cardio-vascular disease. J Am Coll Cardiol 2003;41:1044–7.
4. Ishizaka N, Ishizaka Y, Toda E, Hashimoto H, Nagai R, Ya-makado M. Association between cigarette smoking, metabolicsyndrome, and carotid arteriosclerosis in Japanese individuals. Ath-erosclerosis 2005;181:381–8.
5. Masulli M, Riccardi G, Galasso R, Vaccaro O. Relationship betweensmoking habits and the features of the metabolic syndrome in a non-diabetic population. Nutr Metab Cardiovasc Dis 2006; 16:364–70.
6. Dzien A, Dzien-Bischinger C, Hoppichler F, Lechleitner M. Themetabolic syndrome as a link between smoking and cardiovasculardisease. Diabetes Obes Metab 2004;6:127–32.
7. Weitzman M, Cook S, Auinger P, et al. Tobacco smoke exposureis associated with the metabolic syndrome in adolescents. Circula-tion 2005;112:862–9.
8. Zavaroni I, Bonini L, Gasparini P, Dall’Aglio E, Passeri M, ReavenGM. Cigarette smokers are relatively glucose intolerant, hyperin-sulinemic and dyslipidemic. Am J Cardiol 1994;73:904–5.
9. Alberti KG, Zimmet P, Shaw J. The metabolic syndrome—a newworldwide definition. Lancet 2005;306:1059–62.
0. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis andmanagement of the metabolic syndrome: an American HeartAssociation/National Heart, Lung, and Blood Institute ScientificStatement. Circulation 2005;112:2735–52.
1. WHO/MONICA-Project. Multinational Monitoring of Trendsand Determinants in Cardiovascular Diseases (MONICA Project)and Manual of Operation. Geneva, Switzerland: World HealthOrganization, Cardiovascular Disease Unit, 1983.
2. Ninomiya JK, L’Italien G, Criqui MH, Whyte JL, Gamst A, ChenRS. Association of the metabolic syndrome with history of myo-
cardial infarction and stroke in the Third National Health andNutrition Examination Survey. Circulation 2004;109:42–6.
371JACC Vol. 53, No. 4, 2009 He et al.January 27, 2009:363–71 From Around the World
3. Eliasson B, Attvall S, Taskinen MR, Smith U. The insulinresistance syndrome in smokers is related to smoking habits.Arterioscler Thromb 1994;14:1946–50.
4. Eliasson B, Attvall S, Taskinen MR, Smith U. Smoking cessationimproves insulin sensitivity in healthy middle-aged men. Eur J ClinInvest 1997;27:450–6.
5. Henkin L, Zaccaro D, Haffner S, et al. Cigarette smoking, environ-mental tobacco smoke exposure and insulin sensitivity. The InsulinResistance Atherosclerosis Study. Ann Epidemiol 1999;9:290–6.
6. Flegal KM, Graubard BI, Williamson DF, Gail MH. Excess deathsassociated with underweight, overweight, and obesity. JAMA 2005;
293:1861–7.
7. Yang G, Ma J, Chen A, et al. Smoking cessation in China: findingsof the 1996 National Prevalence Survey. Tobacco control 2001;10:170–4.
8. Lam TH, He Y, Shi QL, et al. Smoking, quitting, and mortalityin a Chinese cohort of retired men. Ann Epidemiol 2002;12:316 –20.
ey Words: metabolic syndrome y smoking y second-hand smoke yardiovascular risk.
