Association Between Cotinine-Verified Smoking Status and Metabolic Syndrome: Analyses of Korean National Health and Nutrition Examination Surveys 2008

Abstract

Background:

The association between smoking and metabolic syndrome has not been clarified, especially for women, probably because of the inaccurate self-reported smoking status. This study aimed to investigate the association between cotinine-verified smoking status and metabolic syndrome.

Methods:

A total of 11,559 participants from the Korean National Health and Nutrition Examination Surveys were included in this cross-sectional study. Metabolic syndrome was determined according to revised National Cholesterol Education Program Adult Treatment Panel III criteria. Smokers were distinguished from nonsmokers by a urinary cotinine level above 50 ng/mL. Multivariable adjusted logistic regression analysis was used to evaluate the association between cotinine-verified smoking status and metabolic syndrome.

Results:

Prevalence of metabolic syndrome was 28.2% in men and 24.6% in women. Self-reported smoking status was much less consistent with cotinine-verified smoking status in women (kappa values=43.0%) compared with men (kappa value=88.6%). Risk of metabolic syndrome was significantly higher in cotinine-verified smokers than in nonsmokers for both men and women. Among the components of metabolic syndrome, smokers had an increased risk of high triglycerides (TGs), low high-density lipoprotein cholesterol, and decreased risk of high blood pressure compared with nonsmokers in men. In women, smokers had a higher risk of abdominal obesity and high TGs compared with nonsmokers.

Conclusions:

This population-based study showed that smoking was associated with increased risk for metabolic syndrome in men as well as in women and this association is mainly due to the association between smoking and dyslipidemia.

Introduction

Metabolic syndrome is a cluster of risk factors for cardiovascular disease (CVD) and type 2 diabetes mellitus (T2DM)^1
–3 and is associated with an increased risk of various types of cancers.⁴ Metabolic syndrome is composed of abdominal obesity, dyslipidemia, high blood pressure, and impaired fasting glucose.⁵ Although the unique pathophysiology has not been clarified, insulin resistance and obesity are thought to be major clinical features of metabolic syndrome, and previous studies have revealed certain inherited factors, geographical area of residence, low socioeconomic status, low physical activity, and Western-style eating behaviors as risk factors for metabolic syndrome.^6

–10

Tobacco smoking is a well-known independent risk factor for CVD and T2DM.^11
–13 Findings from previous studies have suggested that smoking is also associated with an increased risk of insulin resistance and metabolic syndrome.^14
–16 However, the metabolic effects of cigarette smoking on metabolic syndrome seem complex, with inconsistent findings between studies. Some cross-sectional studies found that abdominal obesity and dyslipidemia were the components of metabolic syndrome that mainly contributed to the association between metabolic syndrome and smoking.^14,15 Longitudinal studies showed that occurrence rates of dyslipidemia, glucose intolerance, and diabetes were higher among smokers than among nonsmokers.^17
–19 A meta-analysis also revealed that smoking is associated with increased levels of serum triglycerides (TGs) and decreased levels of serum high-density lipoprotein cholesterol (HDL-C).²⁰ On the other hand, some evidence negates the positive association between smoking and metabolic syndrome. A cohort study conducted in Italy found that insulin resistance was higher in nonsmokers than in smokers.¹⁹ Additionally, a prospective study of young Dutch participants showed that high blood pressure and abdominal obesity were more commonly developed among subjects who quit or cut down on smoking than those who continued to smoke.²¹ Thus, it is still unclear whether smoking is actually associated with the risk of insulin resistance and metabolic syndrome.

Although several other factors may exist, information bias caused by data collection through self-administered questionnaires could explain the discrepancies in the findings between the studies. The extent of smoking is likely to be underestimated by self-reporters because smoking is often considered a socially undesirable behavior, especially in adolescents and women in Asian countries.²² Thus, to estimate a participant's smoking status accurately, it is necessary to use biochemical verification of smoking, such as measuring cotinine levels. In the present study of a Korean population, we evaluated an association between smoking and metabolic syndrome using cotinine-verified smoking data.

Methods

Study subjects

This study used datasets from the second (2008) and third years (2009) of the fourth wave of the Korea National Health and Nutrition Examination Surveys and the dataset from the first year (2010) of the fifth-wave the survey. The Korea National Health and Nutrition Examination Survey is a population-based, nationwide, cross-sectional survey conducted by the Korea Center for Disease Control. This survey randomly selected study participants using stratified multistage clustered probability sampling from the noninstitutionalized general Korean population. Participation rate in the survey was 77.8% (9744 of 12,528) in 2008, 82.8% (10,533 of 12,722) in 2009, and 81.9% (8958 of 10,533) in 2010, respectively. Of the total 29,235 eligible subjects, we identified 12,226 adults (≥19 years old) with data related to smoking status and urinary cotinine level. We excluded 667 subjects because of the following reasons: Current pregnancy (n=60); missing anthropometric data for waist circumference (WC) (n=72) or blood pressure (n=37); or missing laboratory data for overnight serum fasting glucose (n=429), serum TGs, or HDL-C (n=29). Thus, a total of 11,559 (5,358 men and 6,201 women) subjects were included in the present study. All participants provided written, informed consent.

Study measurements

The Korea National Health and Nutrition Examination Survey is composed of three sections—a health interview survey, a health examination survey, and a nutrition survey. The health interview survey and nutrition survey were conducted by well-trained research assistants. Anthropometric measurements were taken, and fasting blood samples were drawn for laboratory testing after overnight fast from 7 p.m. of the day before survey until next morning.

Sociodemographic and health behavioral characteristics of survey participants were measured by face-to-face interview or self-administered questionnaires. Sociodemographic characteristics included age, sex, marital status (married or unmarried), and years of educational achievement (≤12 years or >12 years). Smoking status, alcohol consumption, and physical activity were measured as health behaviors. Smoking status was categorized into three groups, current smoker, ex-smoker, and never smoker on the basis of responses to the following two questions: “Have you smoked 100 or more cigarettes in your lifetime?” and “Do you smoke cigarettes now?” Participants who reported smoking less than 100 cigarettes in their lifetime were defined as never smokers. Participants who had smoked more than 100 cigarettes in their lifetime but did not smoke at the time of survey were classified as ex-smokers, and those who smoked more than 100 cigarettes in their lifetime and reported smoking either every day or occasionally at the time of survey were classified as current smokers. We defined the “high-risk alcohol intake” group as those who consumed more than 60 grams/day of alcohol for men and 40 grams/day for women with a frequency of more than once a week.²³ The regular physical activity group was composed of participants who took part in either moderate (swimming leisurely; playing doubles tennis, volleyball, badminton, table tennis; or carrying light loads) or high-intensity (running; climbing; fast-cycling; fast swimming; playing soccer, basketball, jump rope, squash, singles tennis, or carrying heavy loads) physical activity for more than 20 min at a time at a frequency of more than three times per week. Daily dietary intake of fat and fiber was assessed using the 24-hr dietary recall method.²⁴

Well-trained research assistants carried out physical examinations and anthropometric measurements. Body height and body weight were measured to the nearest 0.1 cm and 0.1 kg, while participants wore light indoor clothing without shoes. Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared. WC was measured to the nearest 0.1 cm at the narrowest point between the lower borders of the rib cage and the iliac crest after a normal expiration. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) was measured three times at 5-min intervals after resting for at least 15 min; an average of the second and third records was calculated for analysis.

Serum fasting glucose, TGs, and HDL-C were measured in fresh serum. Blood tests were performed using an ADIVIA 1650 Chemistry Analyzer System (Siemens, Elangen, Germany) between January, 2007, and February, 2008, and a Hitachi Automatic Analyzer 7600 (Hitachi, Tokyo, Japan) after 2008.

Urinary cotinine levels were measured by tandem mass spectrometry using a Tandem Mass Spectrometer API 4000 (Applied Biosystems, Foster City, CA) between January, 2007, and December, 2009, and by gas chromatography mass spectrometry using a PerkinElmer Clarus 600T (PerkinElmer, Turku, Finland). We classified participants with urinary cotinine levels above 50 ng/mL as cotinine-verified smokers.^16,22

This study was approved by the Institutional Review Board of Samsung Medical Center (IRB file no. SMC 2013-04-083). The data used for this study are open to the public for health research and do not include any identifiable personal information. As such, informed consent was waived by the by institutional review board.

Definition of metabolic syndrome

We determined metabolic syndrome in participants according to the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) criteria. However, abdominal obesity was ascertained abiding by the cutoff level proposed by the Korean Society for the Study of Obesity, which was determined on the basis of ethnic waist normative data.²⁵ If a person had at least three of the following conditions, he/she was regarded as having metabolic syndrome: Abdominal obesity, defined as a WC ≥90 cm in men and ≥85 cm in women, serum TGs ≥150 mg/dL (1.7 mmol/L) or drug treatment for elevated TGs, serum HDL-C <40 mg/dL (1 mmol/L) in men and <50 mg/dL (1.3 mmol/L) in women or drug treatment for low HDL-C, blood pressure ≥130/85 mmHg or drug treatment for elevated blood pressure, and fasting plasma glucose (FPG) ≥100 mg/dL (5.6 mmol/L) or drug treatment for elevated blood glucose.

Statistical analysis

Sensitivity, specificity, and kappa value (95% confidence intervals [CIs]) for agreements were calculated to validate comparisons between smoking status self-reports and smoking status verified by urinary cotinine levels. For this analysis, self-reported “never smokers” and “past smokers” were classified as nonsmokers, and a cutoff value of 50 ng/mL urine cotinine was used to discriminate current smokers from nonsmokers according to the recommendation by the Society for Research on Nicotine and Tobacco.²⁶ In addition, we performed a sensitivity analysis by changing the cotinine cutoff value from 50 ng/mL to 100 ng/mL because there was an argument about the cutoff value for the Korean population to differentiate smokers from nonsmokers,²⁷

We compared general characteristics between men and women and between cotinine-verified smokers and nonsmokers in each sex using the Student t-test for continuous variables (age, BMI, WC) and a chi-squared test for categorical variables. Because of the differences in age distribution between groups, age-adjusted mean values were calculated using analysis of covariance (ANCOVA) for continuous variables. For categorical variables and prevalence of metabolic syndrome, direct age-standardized percentage was calculated using all of the participants of the fourth and fifth wave of the Korea National Health and Nutrition Examination Survey between 2008 and 2010 as a reference population. We performed a multivariable adjusted logistic regression analysis to determine the independent effect of smoking on metabolic syndrome with an adjustment for age, BMI, marital status, level of education, alcohol intake, physical activity, total fat intake, and total fiber intake.

To examine the dose-response relationship between smoking and metabolic syndrome, we categorized cotinine-verified smokers into four groups according to the quartile distribution of urinary cotinine level. Logistic regression analysis was used to estimate odds ratios (OR) and 95% CIs that compared the odds of metabolic syndrome and the metabolic component of the cotinine-verified smoking group to the odds of the cotinine-verified nonsmoking group after adjustment for covariates.

All the statistical analyses were performed with subpopulation analysis of PASW Statistics version 21.0 (SPSS Inc., Chicago, IL) with consideration of sampling weights and the complex survey design. A two-tailed P value less than 0.05 was considered statistically significant.

Results

Table 1 shows the demographics and clinical characteristics of study subjects. The age-standardized prevalence of self-reported smoking was 43.9% in men and 5.5% in women, whereas the prevalence of cotinine-verified smoking was 48.6% in men and 12.5% in women. The age-standardized prevalence of metabolic syndrome was 28.2% in men and 24.6% in women. Compared with men, women were more likely to be unmarried and more educated, consume less alcohol, and have better profiles in all components of metabolic syndrome.

Table 1.

Characteristics ^a of Study Subjects According to Gender, Korean National Health and Nutrition Examination Survey between 2008 and 2010

	Men (n=5358)	Women (n=6201)	P value ^b
Age, years	42.5 (41.8–43.1)	45.2 (44.6–45.9)	<0.001
Current smoker by self-report	2354 (43.9%)	344 (5.5%)	<0.001
Cotinine-verified current smokers	2602 (48.6%)	774 (12.5%)	<0.001
Urine cotinine, nmol/L	3804.5 (3587.5–4021.4)	595.3 (524.8–665.7)	<0.001
Unmarried, %	17.4 (15.8–19.0)	12.7 (11.2–14.2)	<0.001
Living in rural area, %	20.8 (16.0–25.6)	19.7 (15.4–24.1)	0.958
Duration of education (>12 years), %	68.9 (65.6–72.1)	56.3 (53.6–59.1)	<0.001
High-risk alcohol intake, %^c	34.4 (30.8–38.1)	8.3 (6.5–10.1)	<0.001
Fat intake, grams/day	48.1 (46.9–49.4)	32.6 (31.7–33.4)	<0.001
Fiber intake, grams/day	8.2 (8.0–85)	6.8 (6.6–6.9)	<0.001
Physical activity, %^d	29.1 (27.3–30.9)	23.8 (22.2–25.4)	<0.001
Waist circumference, cm	84.3 (83.9–84.6)	77.5 (77.1–77.9)	<0.001
Body mass index, kg/m²	24.1 (24.0–24.2)	23.1 (23.0–23.2)	<0.001
Systolic blood pressure, mmHg	118.0 (117.4–118.6)	112.6 (112.1–113.2)	<0.001
Diastolic blood pressure, mmHg	77.8 (77.3–78.2)	72.8 (72.4–73.2)	<0.001
High-density lipoprotein cholesterol, mmol/L	1.27 (1.26–1.28)	1.44 (1.43–1.45)	<0.001
Triglycerides, mmol/L	1.77 (1.72–1.82)	1.22 (1.20–1.25)	<0.001
Fasting glucose, mmol/L	5.46 (5.42–5.49)	5.26 (5.22–5.29)	<0.001
Insulin, nmol/L	59.4 (57.6–60.6)	60.0 (58.8–61.2)	0.494
Homeostasis model assessment of insulin resistance (HOMA-IR)^e	2.4 (2.4–2.5)	2.4 (2.3–2.4)	0.197
Metabolic syndrome, %	28.2 (26.7–29.7)	24.6 (22.8–26.4)	0.006

Direct age standardization of all participants in the fourth and fifth wave of the Korea National Health and Nutrition Examination Survey as a reference population was done for categorical variables. Age was adjusted for continuous variables using an analysis of covariance (ANCOVA) test. Proportion or mean level (95% confidence intervals) were presented.

P value was obtained using the Student t-test or chi-squared test.

Alcohol intake ≥moderate amount (60 grams/day for men, 40 grams/day for women), more frequently than once a week.

Moderate- or high-intensity physical activity for ≥20 min, ≥3 times/week.

HOMA-IR calculated by fasting glucose (mg/dL)×fasting insulin (μIU/mL)/405.

Table 2 shows the validity of smoking status assessment by comparing self-reports and urinary cotinine levels. Of the subjects who claimed to nonsmokers in the questionnaire, 737 (296 men, 441 women) were revealed to be current smokers by their cotinine levels. In men, self-reported smoking status was highly consistent with cotinine-verified smoking status—sensitivity, specificity, and kappa values of self-reported smoking status were 88.6%, 98.3%, and 0.87, respectively. However, self-reported smoking status was much less consistent with cotinine-verified smoking status in women—sensitivity, specificity, and kappa values of self-reported smoking status were 43.0%, 99.8%, and 0.56, respectively.

Table 2.

Validity of Smoking Status Assessment by Self–Report Compared with Smoking Status Verified by Urinary Cotinine Level in Korean Adults (>19 Years Old), Korea National Health and Nutrition Examination Survey Between 2008 and 2010

Self-reported smoking status	Number (%)	Cotinine-verified nonsmokers number (%)	Cotinine-verified current smokers number (%)	Kappa (95% CI)	Sensitivity (95% CI)	Specificity (95% CI)	Area under ROC curve (95% CI)
Men
Nonsmokers	3004 (56.1)	2708 (98.3)	296 (11.4)	0.87 (0.86–0.89)	88.6 (87.3,89.8)	98.3 (97.7–98.7)	0.93 (0.93–0.94)
Current smoker	2354 (43.9)	48 (1.7)	2306 (88.6)
Total	5358 (100.0)	2756 (100.0)	2602 (100.0)
Women
Nonsmokers	5857 (94.5)	5416 (99.8)	441 (57.0)	0.56 (0.53–0.60)	43.0 (39.5–46.6)	99.8 (99.6–99.9)	0.71 (0.70–0.73)
Current smoker	344 (5.5)	11 (0.2)	333 (43.0)
Total	6201 (100.0)	5427(100.0)	774 (100.0)

CI, confidence interval; ROC, receiver operating characteristic.

Table 3 shows characteristics of study subjects according to cotinine-verified smoking status. Current smokers tended to be younger than nonsmokers. High-risk alcohol intake was more prevalent among smokers than nonsmokers, and serum TGs levels were significantly higher in smokers than in nonsmokers. In men, HDL-C levels were lower in smokers than in nonsmokers; however, this difference was not observed in women.

Table 3.

Characteristics ^a of Study Subjects According to Cotinine-Verified Smoking Status, ^b Korean National Health And Nutrition Examination Survey Between 2008 and 2010

	Men (n=5358)		Women (n=6201)
	Nonsmokers	Current smokers	Nonsmokers	Current smokers
Number of subjects	2756	2602	5427	774
Urine cotinine, nmol/L	60.8 (45.4–76.1)^c	7480.0 (7144.9–7814.5)^c	46.0 (42.6–50.0)^c	3845.9 (3505.7–4186.0)^c
Age	44.3 (43.5–45.2)^c	40.7 (40.0–41.5)^c	45.9 (45.3–46.6)^c	40.8 (39.2–42.4)^c
Unmarried, %	17.4 (15.2–19.6)	17.7 (15.6–19.8)	12.8 (11.3–14.3)	13.9 (10.5–17.3)
Living in rural area, %	19.7 (14.0–25.4)	22.3 (15.4–29.3)	19.7 (15.1–24.4)	20.4 (10.2–30.6)
Duration of education (>12 years), %	71.9 (67.8–75.9)	65.1 (60.3–69.8)	56.4 (53.5–59.3)	55.5 (47.4–63.6)
High-risk alcohol intake, %^d	28.4 (23.6–33.2)^c	40.1 (33.8–46.4)^c	8.3 (6.4–10.3)^c	21.8 (14.0–29.6)^c
Fat intake, grams/day	49.2 (47.3–51.2)	48.6 (46.9–50.4)	31.6 (30.7–32.4)	32.3 (29.7–34.9)
Fiber intake, grams/day	8.3 (8.0–8.6)	8.1 (7.8–8.4)	6.9 (6.7–7.0)	6.4 (0.7–12.1)
Physical activity, %^e	29.3 (23.9–34.7)	26.6 (20.7–32.4)	23.2 (19.7–26.6)	22.1 (12.9–31.2)
Waist circumference, cm	83.8 (83.3–84.3)	84.3 (83.8–84.7)	77.7 (77.4–78.1)	78.2 (77.2–79.2)
Body mass index, kg/m²	24.1 (23.9–24.3)	24.1 (23.9–24.2)	23.2 (23.0–23.3)	23.0 (22.7–23.4)
Systolic blood pressure, mmHg	117.9 (117.2–118.6)	116.9 (116.1–117.6)	113.4 (112.8–114.0)	112.9 (111.6–114.2)
Diastolic blood pressure, mmHg	78 (77.5–78.5)	77.1 (76.5–77.7)	73.2 (72.8–73.5)	72.5 (71.6–73.4)
High-density lipoprotein cholesterol, mmol/L	1.30 (1.28–1.31)^c	1.26 (1.25–1.28)^c	1.44 (1.42–1.45)	1.44 (1.41–1.47)
Triglycerides, mmol/L	1.64 (1.56–1.72)^c	1.86 (1.79–1.93)^c	1.22 (1.19–1.25)^c	1.40 (1.31–1.49)^c
Fasting glucose, mmol/L	5.41 (5.36–5.46)	5.44 (5.38–5.51)	5.36 (5.32–5.39)	5.27 (5.19–5.35)
Insulin, nmol/L	60.6 (58.8–62.4)	58.2 (56.4–59.4)	59.4 (58.2–61.2)	60.0 (57.0–62.4)
Homestasis model assessment of insulin resistance (HOMA–IR)^f	2.5 (2.4–2.6)	2.4 (2.3–2.4)	2.4 (2.3–2.5)	2.4 (2.3–2.5)
Metabolic syndrome, %	27.5 (25.6–29.5)	28.5 (26.4–30.7)	24.3 (22.8–25.9)	26.5 (23.3–29.6)

Proportion or mean levels (95% confidence intervals) are presented.

Direct age standardization of all participants in the fourth and fifth wave of the Korea National Health and Nutrition Examination Survey reference population was carried out for categorical variables. Age was adjusted for continuous variables using analysis of covariance.

Urinary cotinine level of 50 ng/mL or more was defined as cotinine-verified current smoking.

Significant difference between current smokers and nonsmokers assessed by the Student t-test or chi-squared test.

Alcohol intake ≥moderate amount (60 grams/day for men, 40 grams/day for women), more frequently than once a week.

Moderate- or high-intensity physical activity for ≥20 minutes, ≥3 times/week.

HOMA-IR calculated by fasting glucose (mg/dL)×fasting insulin (μIU/mL)/405.

Table 4 shows an association between urinary cotinine-verified smoking status and metabolic syndrome components. Smoking was found to increase the risk of metabolic syndrome significantly for both men (OR=1.26; 95% CI 1.04–1.53) and women (OR=1.32; 95% CI 1.01–1.73), even after adjusting for covariates. Associations were also evaluated between smoking and metabolic components, with an adjustment for all covariates. The risk of having high TGs levels was significantly increased in smokers compared with nonsmokers for both men (OR=1.30, 95% CI 1.12–1.51) and women (OR=1.45, 95% CI 1.13–1.85). The risk of having low HDL-C levels was higher in smokers than nonsmokers for men (OR=1.40, 95% CI 1.16–1.68). Compared with nonsmokers, the smokers had a lower risk of high blood pressure in men (OR=0.81, 95% CI 0.68–0.96) and an increased risk of abdominal obesity (OR=1.52, 95% CI 1.03–2.23) in women (OR=1.25, 95% CI 0.99–1.58). In sensitivity analyses, a higher cutoff value of urinary cotinine did not alter any relationships, but the association between smoking and abdominal obesity in women was attenuated (see Table S1; Supplementary Data are available at www.liebertpub.com/met/).

Table 4.

Odds Ratios (95% Confidence Intervals) ^a of Cotinine-Verified Current Smoking ^b for Components of Metabolic Syndrome ^c Compared to Nonsmoking

	Men OR (95% CI)	P value	Women OR (95% CI)	P value
Metabolic syndrome^c	1.26 (1.04, 1.53)	0.02	1.32 (1.01, 1.73)	0.05
Metabolic component
Abdominal obesity	1.25 (0.99, 1.58)	0.07	1.52 (1.03, 2.23)	0.03
Impaired fasting glucose	1.00 (0.84, 1.20)	0.97	1.00 (0.77, 1.31)	0.98
High triglycerides	1.30 (1.12, 1.51)	<0.01	1.45 (1.13, 1.85)	<0.01
Low high-density lipoprotein cholesterol	1.40 (1.16, 1.68)	<0.01	1.16 (0.95, 1.42)	0.15
High blood pressure	0.81 (0.68, 0.96)	0.01	0.83 (0.63, 1.09)	0.18

Estimated by logistic regression analysis with an adjustment for age, body mass index, marital state, education, alcohol intake, physical activity, total fat intake, and total fiber intake.

Urinary cotinine level of 50 ng/mL or more was defined as cotinine-verified current smoking.

Metabolic syndrome is defined by any three or more following criteria: Abdominal obesity, defined as a waist circumference in men ≥90 cm and in women ≥85 cm; serum triglycerides ≥150 mg/dL or drug treatment for elevated triglycerides; serum high-density lipoprotein cholesterol (HDL–C) <40 mg/dL in men and <50 mg/dL in women or drug treatment for low HDL-C; blood pressure ≥130/85 mmHg or drug treatment for elevated blood pressure, fasting plasma glucose (FPG) ≥100 mg/dL or drug treatment for elevated blood glucose.

OR, odds ratio; CI, confidence interval.

Table 5 shows a dose-response relationship between smoking status and metabolic syndrome according to urinary cotinine level. In this analysis, we found that the risk of metabolic syndrome tended to increase with increasing levels of urinary cotinine (P for trend in men <0.01, P for trend in women=0.02). There were significant dose-dependent associations between urinary cotinine levels and high TGs levels (P for trend <0.01 for both men and women) and between urinary cotinine levels and low HDL-C (P for trend <0.01 for men, 0.02 for women). A significant dose-response relationship between urinary cotinine level and abdominal obesity was only observed in women (P for trend=0.02).

Table 5.

Association ^a Between Urinary Cotinine Level (ng/mL) and Metabolic Syndrome and Its Components

	Nonsmoker ^b	Current smoker ^b
Men	<50 (n=2756)	50–538.49 (n=650)	538.40–1148.17 (n=651)	1148.18–1843.08 (n=651)	>1843.09 (n=651)	P for trend^c
Metabolic syndrome	1	0.86 (0.54, 1.36)	1.50 (1.04, 2.18)	1.84 (1.28, 2.65)	1.01 (0.66, 1.53)	0.01
Metabolic components
Abdominal obesity	1	1.42 (0.85, 2.38)	1.15 (0.69, 1.93)	1.36 (0.88, 2.12)	1.05 (0.64, 1.75)	0.33
Fasting glucose	1	1.04 (0.72, 1.50)	1.09 (0.76, 1.56)	1.19 (0.86, 1.66)	0.72 (0.51, 1.04)	0.36
Triglycerides	1	0.99 (0.73, 1.35)	1.57 (1.15, 2.04)	1.48 (1.07, 2.04)	1.24 (0.92, 167)	<0.01
Low high-density lipoprotein cholesterol	1	1.12 (0.79, 1.61)	1.45 (1.03, 2.04)	1.51(1.04, 2.18)	1.55 (1.07, 2.23)	<0.01
Blood pressure	1	0.72 (0.50, 1.02)	0.72 (0.52, 0.98)	0.99 (0.70, 1.42)	0.82 (0.58, 1.17)	0.21

	Nonsmoker ^b	Current smoker ^b
Women	<50 (n=5427)	50–99.39 (n=193)	99.40–465.03 (n=194)	465.04–992.96 (n=194)	>992.97 (n=193)	P for trend
Metabolic syndrome	1	1.05 (0.58, 1.93)	1.18 (0.54, 2.56)	1.53 (0.76, 3.08)	1.59 (0.81, 3.15)	0.02
Metabolic components
Abdominal obesity	1	1.47 (0.69, 3.17)	0.90 (0.36, 2.25)	1.92 (0.97, 3.80)	1.95 (0.90, 4.22)	0.02
Fasting glucose	1	0.86 (0.48, 1.55)	1.06 (0.58, 1.93)	0.84 (0.43, 1.63)	1.28 (0.64, 2.56)	0.69
Triglycerides	1	1.02 (0.59, 1.76)	1.13 (0.59, 2.18)	1.71 (0.94, 3.08)	2.09 (1.18, 3.69)	<0.01
Low high-density lipoprotein cholesterol	1	0.90 (0.56, 1.47)	1.02 (0.58, 1.80)	1.29 (0.79, 2.09)	1.50 (0.94, 2.40)	0.02
Blood pressure	1	1.05 (0.60, 1.83)	0.80 (0.39, 1.63)	0.76 (0.36, 1.61)	0.74(0.40, 1.38)	0.11

Estimated by logistic regression analysis with an adjustment for age, body mass index, marital state, residence area, education, alcohol intake, physical activity, total fat intake, and total fiber intake.

Urinary cotinine level less than 50 ng/mL was defined as cotinine-verified nonsmoking. Cotinine-verified current smokers were categorized into four groups on the basis of their quartile distribution of urinary cotinine levels.

Assessed by linear by linear association in which urinary cotinine-verified smoking status was put as a continuous variable.

Discussion

In this cross-sectional study of men and women representing the Korean adult population, we ascertained that smoking, as verified by urinary cotinine levels, is significantly associated with an increased risk of metabolic syndrome. When smoking status is assessed on the basis of a self-administered questionnaire, the smoking rate is likely to be underestimated, especially in studies targeting the female population. In the present study, we found that the self-reported smoking rate (5.5%) was much lower than the urinary cotinine-verified smoking rate (12.5 %) in women, suggesting that many women in the study reported false information. Although the present study does not address the reason why women tend to conceal their smoking status, this finding suggests that information bias could be a significant problem in smoking-related studies of women when self-reported smoking status is relied upon. However, the high kappa value (0.87) for the accuracy of self-reported smoking status in men suggests that smoking status assessment by self-report may not result in significant bias in studies of men.

The positive association between cotinine-confirmed smoking and metabolic syndrome is compatible with the findings of previous studies.^9,14,15 Some studies suggested that the higher prevalence of metabolic syndrome in smoking groups was attributed to low physical activity, excessive alcohol consumption, and the Western style diet of smokers.^10,28,29 However, the present study found that the positive association between smoking and metabolic syndrome was maintained after adjusting for these sociodemographic and lifestyle factors. In addition, the present study revealed a dose-dependent relationship between urinary cotinine level and metabolic syndrome. These findings indicate that smoking is independently associated with the development of metabolic syndrome and therefore may be a causal risk factor for metabolic syndrome.

Most previous studies from cross-sectional to meta-analysis consistently reported that smoking increases serum TGs and decreases HDL-C levels.^14,15,19,20 In the present study, high serum TGs and low HDL-C were also positively associated with smoking, and this relationship was thought to contribute to a higher risk of metabolic syndrome in smokers. Although smokers were more likely to be high-risk alcohol drinkers in both men and women, the increased risk of high TGs in smokers does not seem to be caused by concomitant alcohol intake because the association between high TGs and smoking persisted, even after adjusting for alcohol intake.

There was no significant association between smoking and impaired fasting glucose in the present study. This finding is consistent with the findings of other cross-sectional studies.^14,15 However, longitudinal studies and meta-analysis have shown that smoking increases fasting glucose levels and, thus, the risk of T2DM.^13,17,18 Differences in the study design might have caused this discrepancy, and further studies are needed to clarify the association between smoking and glucose. However, when we analyzed diabetes instead of glucose impairment, smokers showed a 1.3-fold higher prevalence of diabetes compared with nonsmokers in men (Table S2).

Smoking influences insulin sensitivity by increasing insulin-antagonizing hormones, such as catecholamine, cortisol, and growth hormone, and increasing lipolysis, subsequently causing an increase in serum free fatty acid levels.³⁰ Furthermore, nicotine absorbed from smoking activates the cholinergic nicotinic receptor, which promotes lipolysis and reduces plasma adiponectin concentration derived from adipocytes.³¹ As a consequence, an increased serum free fatty acid level causes insulin resistance in muscle and liver and reduced hepatic extraction of insulin, which results in fasting glucose impairment.³²

In the present study, smokers showed a significantly higher risk of abdominal obesity in women. In men, smokers exhibited only a modest increase in the risk of abdominal obesity, with marginal statistical significance, although there was no significant difference in the BMI values between smokers and nonsmokers. This result indicates that cigarette smoking changes body fat distribution and leads to adverse metabolic outcomes, such as abdominal obesity.^33,34 Additionally, the dose-response relationship between urinary cotinine level, which quantifies smoking, and abdominal obesity in women smokers supports an association between smoking and insulin resistance, especially in women.^34,35

The effects of smoking on blood pressure differed by sex. In men, smoking was inversely associated with blood pressure, whereas this association did not exist in women. Although some previous studies could not identify the effect of smoking on blood pressure,^14,15 smoking has been thought to contribute to higher blood pressure in other studies.^36
–38 The acute effect of tobacco smoking has been reported to increase blood pressure, leading to hypertension,³⁸ whereas cotinine, a metabolite of tobacco, is known to cause vasodilation and reduces blood pressure in chronic smokers.³⁹ Conversely, in a study of 24-hr ambulatory blood pressure monitoring, smokers were found to have a higher mean daytime ambulatory SBP than nonsmokers, whereas office blood pressure was similar between smokers and nonsmokers.⁴⁰ This finding suggests that office blood pressure may not accurately represent usual blood pressure, which may result in underestimation of the undesirable effect of smoking on blood pressure. In the present study, data on 24-hr ambulatory blood pressure were not available and we could not evaluate this issue sufficiently. Thus, further studies on the association between smoking and blood pressure and mechanisms underlying the association seem to be needed.

Most of the tobacco users in Korea are cigarette smokers, which may limit generalization of the findings from our study to other populations who mainly use other types of tobacco products. Cigarette smoking is known as the most efficient form of nicotine supply because it escapes first-pass intestinal and hepatic metabolism and moves quickly to the brain. Given that cigarette smoking supplies nicotine in a repeated-bolus dosing pattern, there could be considerable fluctuation in blood levels of nicotine from cigarette to cigarette compared to chewing tobacco, gum, snuff, or patch. However, nicotine has a half-life of 2 hr and accumulates in the body over 6–9 hr of regular smoking and, thus, chronic smoking results in an chronic exposure to nicotine that lasts 24 hr per day rather than intermittent and transient exposure to nicotine.⁴¹ Therefore, disease risks associated with cigarette smoking in chronic smokers are less likely to significantly differ from the risk associated with other types of tobacco products.

The present study has several limitations. First, this study was conducted in a cross-sectional design that cannot establish a time relationship between smoking behavior and the development of metabolic syndrome. However, biologically, it is more likely that smoking causes metabolic syndrome rather than metabolic syndrome causing smoking. Second, nicotine replacement therapy and second-hand smoke can affect urinary cotinine levels, which may result in misclassification of some nonsmokers or ex-smokers as smokers and underestimation of the association between smoking and metabolic syndrome. However, a Korean study using the same data source as our study reported that mean urinary cotinine level was 11.1 [standard deviation (SD)=0.7] ng/mL in nonsmoking women exposed to second-hand smoke.⁴² Therefore, given that the cutoff level for identifying smokers was 50 ng/mL, a significant bias by the misclassification of smokers is less likely to have occurred in the present study.

Despite these limitations, the findings of the present study can be generalized because this study was conducted using The Korea National Health and Nutrition Examination Survey data, which represents the entire Korean population. Furthermore, we used smoking information verified by urinary cotinine levels, which allowed us to investigate more precisely the association between smoking and metabolic syndrome by eliminating significant information bias.

In conclusion, this population-based study found that smoking was associated with an increased risk of metabolic syndrome, and the increased risks for high TGs levels and low HDL-C levels in smokers seem to mainly explain the association between smoking and metabolic syndrome.

Footnotes

Acknowledgments

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (#2014R1A2A2A01002705). The funding sources had no role in conducting this study.

Author contributions: J. Kang and Y.M. Song, planning, data analysis, and data interpretation; J. Kang, data acquisition. Both authors have read and approved the final version submitted.

Author Disclosure Statement

No competing financial interests exist.

References

Hanson

, Imperatore

, Bennett

, et al. Components of the “metabolic syndrome” and incidence of type 2 diabetes. Diabetes, 2002; 51:3120–3127.

Isomaa

, Almgren

, Tuomi

, et al. Cardiovascular morbidity and mortality associated with the metabolic syndrome. Diabetes Care, 2001; 24:683–689.

Lakka

, Laaksonen

, Lakka

, et al. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA, 2002; 288:2709–2716.

Esposito

, Chiodini

, Colao

, et al. Metabolic syndrome and risk of cancer: A systematic review and meta-analysis. Diabetes Care, 2012; 35:2402–2411.

Alberti

, Eckel

, Grundy

, et al. Harmonizing the metabolic syndrome: A joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation, 2009; 120:1640–1645.

Cossrow

, Falkner

. Race/ethnic issues in obesity and obesity-related comorbidities. J Clin Endocrinol Metab, 2004; 89:2590–2594.

Freedman

, Sigurdson

, Rajaraman

, et al. The mortality risk of smoking and obesity combined. Am J Prev Med, 2006; 31:355–362.

Peeters

, Barendregt

, Willekens

, et al. Obesity in adulthood and its consequences for life expectancy: A life-table analysis. Ann Intern Med, 2003; 138:24–32.

Onat

, Ugur

, Hergenc

, et al. Lifestyle and metabolic determinants of incident hypertension, with special reference to cigarette smoking: A longitudinal population-based study. Am J Hypertens, 2009; 22:156–162.

10.

Lee

, Jung

, Park

, et al. Effects of smoking, alcohol, exercise, education, and family history on the metabolic syndrome as defined by the ATP III. Diabetes Res Clin Pract, 2005; 67:70–77.

11.

Huxley

, Woodward

. Cigarette smoking as a risk factor for coronary heart disease in women compared with men: A systematic review and meta-analysis of prospective cohort studies. Lancet, 2011; 378:1297–1305.

12.

Jee

, Suh

, Kim

, et al. Smoking and atherosclerotic cardiovascular disease in men with low levels of serum cholesterol: The Korea Medical Insurance Corporation Study. JAMA, 1999; 282:2149–2155.

13.

Willi

, Bodenmann

, Ghali

, et al. Active smoking and the risk of type 2 diabetes: A systematic review and meta-analysis. JAMA, 2007; 298:2654–2664.

14.

, Yoon

, Lee

, et al. Association between cigarette smoking and metabolic syndrome: The Korea National Health and Nutrition Examination Survey. Diabetes Care, 2005; 28:2064–2066.

15.

Chen

, Li

, Chang

, et al. Association among cigarette smoking, metabolic syndrome, and its individual components: The metabolic syndrome study in Taiwan. Metabolism, 2008; 57:544–548.

16.

SRNT Subcommittee on Biochemical Verification. Biochemical verification of tobacco use and cessation. Nicotine Tob Res, 2002; 4:149–159.

17.

Kowall

, Rathmann

, Strassburger

, et al. Association of passive and active smoking with incident type 2 diabetes mellitus in the elderly population: The KORA S4/F4 cohort study. Eur J Epidemiol, 2010; 25:393–402.

18.

Houston

, Person

, Pletcher

, et al. Active and passive smoking and development of glucose intolerance among young adults in a prospective cohort: CARDIA study. BMJ, 2006; 332:1064–1069.

19.

Masulli

, Riccardi

, Galasso

, et al. Relationship between smoking habits and the features of the metabolic syndrome in a non-diabetic population. Nutr Metab Cardiovasc Dis, 2006; 16:364–370.

20.

Craig

, Palomaki

, Haddow

. Cigarette smoking and serum lipid and lipoprotein concentrations: An analysis of published data. BMJ, 1989; 298:784–788.

21.

Bernaards

, Twisk

, Snel

, et al. In a prospective study in young people, associations between changes in smoking behavior and risk factors for cardiovascular disease were complex. J Clin Epidemiol, 2005; 58:1165–1171.

22.

Jung-Choi

, Khang

, Cho

. Hidden female smokers in Asia: A comparison of self-reported with cotinine-verified smoking prevalence rates in representative national data from an Asian population. Tob Control, 2012; 21:536–542.

23.

Gaziano

, Buring

, Breslow

, et al. Moderate alcohol intake, increased levels of high-density lipoprotein and its subfractions, and decreased risk of myocardial infarction. N Engl J Med, 1993; 329:1829–1834.

24.

Willett

. Nutritional epidemiology, 2nd ed. New York: Oxford University Press, 2012.

25.

Lee

, Park

, Kim

, et al. Appropriate waist circumference cutoff points for central obesity in Korean adults. Diabetes Res Clin Pract, 2007:75:72–80.

26.

Verification SSoB. Biochemical verification of tobacco use and cessation. Nicotine Tob Res, 2002; 4:149–159.

27.

Jung

, Lee

, Kim

, et al. Urine cotinine for assessing tobacco smoke exposure in Korea: Analysis of the Korea National Health and Nutrition Examination Survey (KNHANES). Tuberc Respir Dis (Seoul), 2012; 73:210–218.

28.

Shin

, Kweon

, Choi

, et al. Average volume of alcohol consumed, drinking patterns, and metabolic syndrome in older Korean adults. J Epidemiol, 2013; 23:122–131.

29.

Park

, Zhu

, Palaniappan

, et al. The metabolic syndrome: Prevalence and associated risk factor findings in the US population from the Third National Health and Nutrition Examination Survey, 1988–1994. Arch Intern Med, 2003; 163:427–436.

30.

Cena

, Fonte

, Turconi

. Relationship between smoking and metabolic syndrome. Nutr Rev, 2011; 69:745–753.

31.

Iwashima

, Katsuya

, Ishikawa

, et al. Association of hypoadiponectinemia with smoking habit in men. Hypertension, 2005; 45:1094–1100.

32.

Bergman

, Ader

. Free fatty acids and pathogenesis of type 2 diabetes mellitus. Trends Endocrinol Metab, 2000; 11:351–356.

33.

Jee

, Lee

, Nam

, et al. Effect of smoking on the paradox of high waist-to-hip ratio and low body mass index. Obes Res, 2002; 10:891–895.

34.

Canoy

, Wareham

, Luben

, et al. Cigarette smoking and fat distribution in 21,828 British men and women: A population-based study. Obes Res, 2005; 13:1466–1475.

35.

Rosmond

, Bjorntorp

. Psychosocial and socio-economic factors in women and their relationship to obesity and regional body fat distribution. Int J Obes Relat Metab Disord, 1999; 23:138–145.

36.

Green

, Jucha

, Luz

. Blood pressure in smokers and nonsmokers: Epidemiologic findings. Am Heart J, 1986; 111:932–940.

37.

Mikkelsen

, Wiinberg

, Hoegholm

, et al. Smoking related to 24-h ambulatory blood pressure and heart rate: A study in 352 normotensive Danish subjects. Am J Hypertens, 1997; 10:483–491.

38.

Rhee

, Na

, Kim

, et al. Acute effects of cigarette smoking on arterial stiffness and blood pressure in male smokers with hypertension. Am J Hypertens, 2007; 20:637–641.

39.

Benowitz

, Sharp

. Inverse relation between serum cotinine concentration and blood pressure in cigarette smokers. Circulation, 1989; 80:1309–1312.

40.

Mann

, James

, Wang

, et al. Elevation of ambulatory systolic blood pressure in hypertensive smokers. A case-control study. JAMA, 1991; 265:2226–2228.

41.

Benowitz

. Pharmacology of nicotine: Addiction, smoking-induced disease, and therapeutics. Annu Rev Pharmacol Toxicol, 2009; 49:57–71.

42.

Kim

, Cho

, Evangelista

. Effect of second-hand smoke exposure on lung function among non-smoking Korean women. Iranian J Public Health, 2013; 42:1363–1373.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

Association Between Cotinine-Verified Smoking Status and Metabolic Syndrome: Analyses of Korean National Health and Nutrition Examination Surveys 2008–2010

Abstract

Background:

Methods:

Results:

Conclusions:

Introduction

Methods

Study subjects

Study measurements

Definition of metabolic syndrome

Statistical analysis

Results

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References

Supplementary Material