Abstract
The purpose of this study was to clarify the relationships of cardiovascular risk factors with prehypertension and hypertension in women. Japanese women (35–54 years old, n = 9939) were divided by blood pressure into normotensive, prehypertensive and hypertensive groups. Odds ratios (ORs) for prehypertension and hypertension were calculated after adjustment for confounding factors such as age, smoking, alcohol intake, habitual exercise, body mass index (BMI), and/or therapy for dyslipidemia or diabetes. ORs of light drinkers and heavy drinkers vs non-drinkers for prehypertension and hypertension were significantly higher than a reference level of 1.00, whereas ORs for prehypertension and hypertension of smokers vs non-smokers were significantly lower than the reference level. ORs of the obese group (high BMI or waist-to-height ratio) vs the non-obese group for prehypertension and hypertension were significantly higher than the reference level. ORs for prehypertension and hypertension of the subject group with vs without high triglycerides or low-density lipoprotein (LDL)-cholesterol were significantly higher than the reference level. Thus, obesity, alcohol drinking and dyslipidemias such as high triglycerides and high LDL-cholesterol are associated with prehypertension and hypertension in women. Interventional studies on whether correction of obesity and dyslipidemias and restriction of alcohol consumption are beneficial for prevention of prehypertension and hypertension in women need to be carried out.
Introduction
Hypertension is a major risk factor for cardiovascular disease. There are various risk factors for hypertension, including age, gender, obesity, physical activity and alcohol consumption, and, in addition, hypertension clusters with other risk factors for cardiovascular disease such as dyslipidemia and glucose intolerance (Citation1). The concept of prehypertension has been proposed (Citation2), and correction of prehypertension by lifestyle modification is recommended in order to prevent cardiovascular disease as well as hypertension in the future (Citation3). However, it remains an issue of debate whether antihypertensive drug therapy is needed for persons showing prehypertension (Citation4,Citation5). Non-pharmacological interventions for prehypertension include weight loss, reduced sodium intake, regular physical activity and moderation of alcohol intake (Citation6). There is limited information on associations between prehypertension and its risk factors in women. Obesity has been shown in previous studies to be a major risk factor for prehypertension in women (Citation7–9). Prehypertension has also been shown to be associated with dyslipidemia and glucose intolerance in women (Citation7–9). However, a relationship between alcohol drinking and prehypertension in women has not been determined. The purpose of this study was therefore to clarify the relationships between known risk factors for hypertension and the risk of prehypertension in middle-aged Japanese women. For this purpose, odds ratios of each risk factor for prehypertension and hypertension were calculated, and mean levels of each risk factor were compared among normotensive, prehypertensive and hypertensive subjects.
Methods
Subjects
The subjects were 9939 Japanese women aged ≥ 35 and < 54 years who had received periodic health examinations at workplaces in Yamagata Prefecture in Japan. A cross-sectional study was performed using a local population-based database for the above subjects. This study was approved by the Ethics Committee of Yamagata University School of Medicine.
In a questionnaire at the health checkup, subjects who were receiving treatment for any illnesses were requested to state the names of diseases. Histories of alcohol consumption, cigarette smoking and habitual exercise (almost every day with exercise for 30 min or longer per day) were also surveyed by questionnaires. Average alcohol consumption of each subject per week was reported on questionnaires during health examinations at each workplace. Since it is difficult to know the correct average alcohol consumption of occasional drinkers, only regular drinkers who drank almost every day were used as drinkers for analysis in this study. The subjects were divided into three groups according to usual ethanol consumption per day (non-drinkers; light drinkers: < 22 g of ethanol per day; heavy drinkers: ≥ 22 g of ethanol per day). Since the population of heavy smokers (≥ 20 cigarettes per day) was small, subjects were classified into non-smokers and smokers, and no further detailed classification was done for the smoker group.
Measurements
Height and body weight were measured with light clothes at the health checkup. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Waist circumference was measured at the navel level according to the recommendation of the definition of the Japanese Committee for the Diagnostic Criteria of Metabolic Syndrome (Citation10). Blood pressure was measured by trained nurses, who were part of the local health-checkup company, with a mercury sphygmomanometer once on the day of the health checkup after each subject had rested quietly in a sitting position. Korotkoff phase V was used to define diastolic pressure. Fasted blood was sampled from each subject, and serum triglycerides, high-density lipoprotein (HDL)-cholesterol and low-density lipoprotein (LDL)-cholesterol were measured by enzymatic methods using commercial kits, and hemoglobin A1C was determined by the latex cohesion method using a commercial kit.
Criteria for blood pressure status and risk factors
Normal blood pressure was defined as systolic blood pressure of < 120 mmHg and diastolic blood pressure of < 80 mmHg. Prehypertension was defined as systolic blood pressure of ≥ 120 and < 140 mmHg and/or diastolic blood pressure of ≥ 80 and < 90 mmHg. Hypertension was defined as systolic blood pressure of ≥ 140 mmHg and/or diastolic blood pressure of ≥ 90 mmHg. In addition, subjects who were receiving drug therapy for hypertension were included in the hypertensive group regardless of blood pressure levels.
The criterion for each risk factor was defined as follows: high BMI, BMI ≥ 25 kg/m2; high waist-to height ratio (WHtR); WHtR ≥ 0.5; high triglycerides, triglycerides ≥ 150 mg/dl; low HDL-cholesterol, HDL-cholesterol < 40 mg/dl; high LDL-cholesterol, LDL-cholesterol ≥ 140 mg/dl; hyperglycemia, hemoglobin A1c > 5.8%. The percentages of subjects showing BMI ≥ 30 kg/m2 and BMI ≥ 25 kg/m2 were 4.6% and 20.1%, respectively. The cut-off value of high BMI, 25 kg/m2, proposed by the Japan Society for the Study of Obesity (JASSO), is generally used as criteria for obesity in Japanese men and women (Citation11). Subjects receiving drug therapy for diabetes were also included in the above definition of hyperglycemia.
Statistical analysis
Statistical analyses were performed using a computer software program (SPSS version 16.0 J for Windows, Chicago, IL, USA). Since there is a possibility that serum triglyceride levels do not fit a normal distribution, they were used after log-conversion. In univariate analysis, mean values of each variable in the normotensive, prehypertensive and hypertensive groups were compared using analysis of variance and subsequent Scheffé F-test. In multivariate analysis, mean values of each variable, calculated after adjustment for additional variables, including age, BMI, history of smoking, history of alcohol intake, history of habitual exercise and/or history of therapy for dyslipidemia or diabetes, were compared among the normotensive, prehypertensive and hypertensive groups using analysis of covariance and then Student's t-test after Bonferroni correction. The percentages of drinkers, smokers, subjects with habitual exercise and subjects receiving therapy for hypertension, dyslipidemia or diabetes were compared in each group pair using the chi-square test for independence. In logistic regression analysis, odds ratios for prehypertension or hypertension were calculated after adjustment for confounding factors including age, BMI, history of smoking, history of alcohol intake, history of habitual exercise and/or history of therapy for dyslipidemia or diabetes. Probability (p) values less than 0.05 were defined as significant.
Results
Profiles of subjects
shows characteristics of the subjects. Subjects in the prehypertensive and hypertensive groups were significantly older than those in the normotensive group, and subjects in the hypertensive group were significantly older than those in the prehypertensive group. The percentage of smokers was significantly lower in the prehypertensive and hypertensive groups than in the normotensive group and was significantly lower in the hypertensive group than in the prehypertensive group. The percentage of drinkers was significantly higher in the prehypertensive group than in the normotensive group. The percentages of subjects receiving therapy for dyslipidemia and diabetes were significantly higher in the prehypertensive and hypertensive groups than in the normotensive group and were significantly higher in the hypertensive group than in the prehypertensive group. Subjects in the normotensive and prehypertensive groups were significantly taller than those in the hypertensive group. Weight and waist circumference were significantly larger in the prehypertensive and hypertensive groups than in the normotensive group and were significantly larger in the hypertensive group than in the prehypertensive group.
Table I. Profiles of the normotensive, prehypertensive and hypertensive groups.
Relationships of alcohol drinking and smoking with prehypertension and hypertension
Odds ratios of drinkers vs non-drinkers and odds ratios of smokers vs non-smokers for prehypertension or hypertension are shown in . Odds ratios were calculated after adjustment for age, BMI, habitual exercise, and alcohol intake or smoking. Significantly higher odds ratios than the reference level of 1.00 in heavy drinkers and light drinkers vs non-drinkers were found for both prehypertension and hypertension. Odds ratios for prehypertension and hypertension in smokers vs non-smokers were significantly lower than the reference level of 1.00.
Table II. Relationships of alcohol intake and smoking with prehypertension and hypertension.
Interaction of smoking with the relationships between alcohol intake and blood pressure status
shows relationships of alcohol intake with prehypertension and hypertension in smokers and non-smokers. In non-smokers, odds ratios of heavy drinkers vs non-drinkers for prehypertension and hypertension were significantly higher than the reference level of 1.00, whereas odds ratios of light drinkers vs non-drinkers were not significantly different from the reference level. In smokers, odds rations of light vs non-drinkers and odds rations of heavy vs non-drinkers were significantly higher than the reference level. Odds ratios of the interaction term consisting of alcohol drinking (heavy drinkers/non-drinkers) and smoking (smokers/non-smokers) for prehypertension and hypertension were significantly higher than the reference level.
Table III. Relationships of alcohol intake with prehypertension and hypertension in smokers and non-smokers.
Relationships of obesity with prehypertension and hypertension
Odds ratios of the obese group vs the non-obese group for prehypertension or hypertension are shown in . Odds ratios were calculated after adjustment for age, alcohol intake, smoking and habitual exercise. Odds ratios of the subject group with BMI of ≥ 25 kg/m2 vs the subject group with BMI of < 25 kg/m2 for prehypertension and hypertension were significantly higher than the reference level of 1.00. Similarly, significantly higher odds ratios than the reference level of 1.00 were found in the subject group with WHtR of ≥ 0.5 vs the subject group with WHtR of < 0.5 for prehypertension and hypertension.
Table IV. Relationships of adiposity with prehypertension and hypertension.
Relationships of dyslipidemias and hyperglycemia with prehypertension and hypertension
Odds ratios of the subject group showing dyslipidemia or hyperglycemia vs the subject group not showing dyslipidemia or hyperglycemia are shown in . Odds ratios were calculated after adjustment for age, BMI, alcohol intake, smoking and habitual exercise. Therapy for dyslipidemia was also adjusted for calculation of odds ratios in subject groups with vs without high triglycerides, low HDL-cholesterol and high LDL-cholesterol. Odds ratios for prehypertension and hypertension in the subject group with high triglycerides vs the subject group without high triglycerides were significantly higher than the reference level of 1.00. Odds ratios for prehypertension and hypertension in the subject group with high LDL-cholesterol vs the subject group without high LDL-cholesterol were slightly but significantly higher than the reference level of 1.00, whereas odds ratios for prehypertension and hypertension of the subject group with low HDL-cholesterol vs the subject group without low HDL-cholesterol were not significantly higher than the reference level of 1.00. Odds ratios of the subject group with hyperglycemia vs the subjects group without hyperglycemia were significantly higher than the reference level of 1.00 for hypertension but not for prehypertension.
Table V. Relationships of blood lipid levels and glycemic status with prehypertension and hypertension.
Comparison of mean levels of blood lipids and glucose among the normotensive, prehypertensive and hypertensive groups
shows the results of multivariate analysis for the relationship between the mean level of each risk factor and blood pressure status. BMI and WHtR were significantly higher in the prehypertensive and hypertensive groups than in the normotensive group and were significantly higher in the hypertensive group than in the prehypertensive group. Log-converted triglycerides was significantly higher in the prehypertensive and hypertensive groups than in the normotensive group and was significantly higher in the hypertensive group than in the prehypertensive group. LDL-cholesterol was slightly but significantly higher in the prehypertensive and hypertensive groups than in the normotensive group, whereas there were no significant differences in HDL-cholesterol among the normotensive, prehypertensive and hypertensive groups. Hemoglobin A1c was significantly higher in the hypertensive group than in the normotensive and prehypertensive groups.
Table VI. Multivariate analysis for comparison of means of each variable among the normotensive, prehypertensive and hypertensive groups.
Relationships of HDL-cholesterol and blood glucose status with prehypertension and hypertension in multivariate analyses excluding BMI from explanatory variables
In the above multivariate analyses, no significant relationship was found between low HDL-cholesterol and prehypertension and hypertension or between hyperglycemia and prehypertension (). Since there was a possibility of potent confounding of obesity for the relationships of dyslipidemias and hyperglycemia with blood pressure status, relationships of HDL-cholesterol and blood glucose with blood pressure status were tested in multivariate analyses excluding BMI from explanatory variables and including other risk factors such as age, smoking, alcohol drinking, habitual exercise and therapy for dyslipidemia. Odds ratios of subjects with vs without low HDL-cholesterol or hyperglycemia for prehypertension and hypertension were significantly higher than a reference level of 1.00 [odds ratios for prehypertension: ± low HDL-cholesterol, 1.56 (1.37–1.77, p < 0.01); ± hyperglycemia, 2.30 (1.57–3.37, p < 0.01); odds ratios for hypertension: ± low HDL-cholesterol, 2.20 (1.87–2.59, p < 0.01); ± hyperglycemia, 10.14 (7.00–14.69, p < 0.01)]. Means levels of HDL-cholesterol and hemoglobin A1c calculated after adjustment for age, smoking, alcohol drinking, habitual exercise and therapy for dyslipidemia or diabetes were significantly lower (p < 0.01) and higher (p < 0.01), respectively, in the hypertensive group than in the normotensive and prehypertensive groups and were significantly lower (p < 0.01) and higher (p < 0.01), respectively, in the prehypertensive group than in the normotensive group [HDL-cholesterol (mg/dl): 66.6 ± 0.2 (normotensive) vs 64.1 ± 0.2 (prehypertensive) vs 61.3 ± 0.4 (hypertensive); hemoglobin A1c (%): 4.94 ± 0.01 (normotensive) vs 4.99 ± 0.01 (prehypertensive) vs 5.12 ± 0.01 (hypertensive)].
Discussion
Multivariate analyses using odds ratios and means showed that obesity, alcohol intake, LDL-cholesterol and triglycerides were positively related to prehypertension and hypertension in women. There is limited knowledge on risk factors for prehypertension in women. The results of the present study suggest that obesity, alcohol intake and dyslipidemia are risk factors for prehypertension as well as for hypertension. Among these variables, obesity is the strongest factor influencing blood pressure status. Obesity is known to be also associated with dyslipidemias such as high triglycerides, high LDL-cholesterol and low HDL-cholesterol (Citation12). However, significant associations of obesity with high triglycerides and high LDL-cholesterol were found after adjustment for BMI in multivariate logistic regression analysis, and thus these associations are independent of obesity.
Alcohol drinking is an important risk factor for essential hypertension, and sensitivity to alcohol is stronger in Asians than in other ethnic people due to polymorphism of alcohol-metabolizing enzymes, especially acetoaldehyde dehydrogenase 2 (Citation13). It is controversial whether there is a gender difference in relationship between alcohol drinking and blood pressure (Citation14–16). Our recent study using Japanese subjects has shown that odds ratios for hypertension in men were significantly higher compared with a reference level of 1.00 both in light and heavy drinkers, whereas in women, this association was found in heavy vs non-drinkers but not in light vs non-drinkers (Citation17). This suggests that light alcohol intake is a risk factor for hypertension in men but not in women. Moreover, women who took two or fewer drinks of alcohol per day reportedly had slightly lower blood pressure than that of women who did not drink (Citation14). The reason for the gender difference can be explained as follows: increase in sympathetic activity is involved in elevation of blood pressure due to habitual alcohol drinking (Citation18), and sympathetic activity is known to be decreased by estrogen (Citation19). Therefore, gender-related hormonal difference is speculated to be a reason for lower sensitivity of blood pressure to alcohol drinking in women than in men. In the present study, significant positive relationships were found between alcohol drinking and prehypertension as well as hypertension. In addition, the odds ratios for prehypertension and hypertension were significantly higher compared with a reference level of 1.00 both in light drinkers and in heavy drinkers, and there was a dose-dependent relationship between alcohol intake and odds ratios for prehypertension and hypertension. Therefore, light and heavy drinking is associated with prehypertension and hypertension. This is, to the best of our knowledge, the first study showing a positive relationship between alcohol drinking and prehypertension in women.
Significant associations of light alcohol drinking with prehypertension and hypertension were found in smokers but not in non-smokers. Heavy drinking was associated with prehypertension and hypertension both in smokers and nonsmokers, and the odds ratios of heavy drinkers vs non-drinkers for prehypertension and hypertension were higher in smokers than in non-smokers. Moreover, significant interaction was found in smoking with the relationship of heavy alcohol drinking with prehypertension and hypertension. Therefore, the associations of alcohol drinking with prehypertension and hypertension were stronger in smokers than in non-smokers. This agrees with the results of previous studies by the German, Dutch and Japanese groups showing a stronger association between alcohol consumption and blood pressure in smokers than in non-smokers (Citation20–24). Thus, smokers are suggested to be more prone to become prehypertensive and hypertensive by habitual alcohol drinking compared with non-smokers.
In this study, smoking was inversely associated with prehypertension and hypertension in Japanese women. This agrees with the findings of a recent study using Mongolians (Citation9). Smokers are known to show lower blood pressure than non-smokers possibly due to reduction in body weight by smoking and an adaptation or rebound effect following repeated sympathetic stimulation by nicotine (Citation25). There is no doubt that smoking is an important risk factor for atherosclerotic disease and should be forbidden for persons with high blood pressure, although, as mentioned above, there is evidence that blood pressure is slightly lower in smokers than in non-smokers. Previous studies have shown no dose–response relationships between smoking and its effects on blood pressure (Citation23,Citation26). However, the relationship between smoking and hypertension was investigated using only two subject groups of smokers and non-smokers in the present study because of the small number of female heavy smokers.
In multivariate logistic regression analysis and multivariate analysis of means, hyperglycemia was significantly associated with hypertension but not with prehypertension, and low HDL-cholesterol was associated with neither hypertension nor prehypertension. When BMI was excluded from explanatory variables in these analyses, both hyperglycemia and low HDL-cholesterol were significantly related to prehypertension and hypertension. Thus, the relationships of hyperglycemia and low HDL-cholesterol with prehypertension and hypertension were strongly confounded by obesity. These findings of confounding by obesity are plausible since obesity is a central risk factor comprising metabolic syndrome, which represents a cluster of cardiometabolic risk factors, including glucose intolerance, dyslipidemia and hypertension (Citation27).
There are limitations in this study. Multivariate analyses with adjustment for various confounding factors have been performed in this study. However, there are other confounding factors for relationships between atherosclerotic risk factor and prehypertension or hypertension not used in multivariate analyses in this study, including diet, socioeconomic factors and menopausal status. Menopausal status is a particularly important confounding factor, since prevalences of hypertension and other hypertension-related metabolic factors such as obesity and dyslipidemia increase after menopause (Citation28–30). Unfortunately, information on menopausal status of the subjects was not available in the database used in this study. Salt consumption and kind of anti-hypertensive drug for subjects receiving therapy for hypertension are also possible confounding factors for relationships between blood pressure and risk factors for hypertension. However, information on salt consumption and kind of anti-hypertensive drug was not available in the database used. Second, the subjects of this study were all Japanese and there is a possibility of ethnic and/or racial differences in the relationships of cardiovascular risk factors with prehypertension and hypertension. Third, the design of this study is cross-sectional, and thus further prospective studies are needed to clarify causality of the findings of this study.
In conclusion, in women, prehypertension and hypertension are positively related to obesity, alcohol drinking and dyslipidemias, such as high triglycerides and LDL-cholesterol, and are inversely related to smoking. Correction of obesity and dyslipidemias and restriction of alcohol consumption are important for prevention of prehypertension and hypertension in women.
Acknowledgements
This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (No. 21390211).
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Related Research Data
References
- Messerli FH, Williams B, Ritz E. Essential hypertension. Lancet. 2007;370:591–603.
- Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr ; the National High Blood Pressure Education Program Coordinating Committee. The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: The JNC 7 report. JAMA. 2003;289:2560–2572.
- Elliott WJ, Black HR. Prehypertension. Nat Clin Pract Cardiovasc Med. 2007;4:538–548
- Kjeldsen SE, Oparil S, Narkiewicz K, Hedner T. Should we treat prehypertension? Blood Press. 2009;18:298–299.
- Kiely AE, Kwatra SG, Kwatra MM. Treating prehypertension: Medically sound and economically viable. Blood Press. 2009;18:300–303.
- Svetkey LP. Management of prehypertension. Hypertension. 2005;45:1056–1061.
- Agyemang C, van Valkengoed I, van den Born BJ, Stronks K. Prevalence and determinants of prehypertension among African Surinamese, Hindustani Surinamese, and White Dutch in Amsterdam, the Netherlands: The SUNSET study. Eur J Cardiovasc Prev Rehabil. 2007;14:775–781.
- Ishikawa Y, Ishikawa J, Ishikawa S, Kayaba K, Nakamura Y, Shimada K, .; Prevalence and determinants of prehypertension in a Japanese general population: The Jichi Medical School Cohort Study. Hypertens Res. 2008;31:1323–1330.
- Li H, Xu T, Tong W, Liu Y, Zhao L, Zhang Y. Comparison of cardiovascular risk factors between prehypertension and hypertension in a Mongolian population, Inner Mongolia, China. Circ J. 2008;72:1666–1673.
- Anonymous. Metabolic syndrome – Definition and diagnostic criteria in Japan. J Jpn Soc Int Med. 2005;94:794–809 (in Japanese).
- The Examination Committee of Criteria for “Obesity Disease” in Japan; Japan Society for the Study of Obesity. New criteria for “obesity disease” in Japan. Circ J. 2002;66:987–992.
- Bamba V, Rader DJ. Obesity and atherogenic dyslipidemia. Gastroenterology. 2007;132:2181–2190.
- Hui P, Nakayama T, Morita A, Sato N, Hishiki M, Saito K, . Common single nucleotide polymorphisms in Japanese patients with essential hypertension: Aldehyde dehydrogenase 2 gene as a risk factor independent of alcohol consumption. Hypertens Res. 2007;30:585–592.
- Klatsky AL, Friedman GD, Siegelaub AB, Gérard MJ. Alcohol consumption and blood pressure: Kaiser–Permanente Multiphasic Health Examination data. N Engl J Med. 1977;296: 1194–1200.
- Paulin JM, Simpson FO, Waal-Manning HJ. Alcohol consumption and blood pressure in a New Zealand community study. N Z Med J. 1985;98:425–8.
- Yoon YS, Oh SW, Baik HW, Park HS, Kim WY. Alcohol consumption and the metabolic syndrome in Korean adults: The 1998 Korean National Health and Nutrition Examination Survey. Am J Clin Nutr. 2004;80:217–224.
- Wakabayashi I. Influence of gender on the association of alcohol drinking with blood pressure. Am J Hypertens. 2008;21:1310–1317.
- Potter JF, Bannan LT, Saunders JB, Ingram MC, Beevers DG. Blood pressure and pressor mechanisms during alcohol withdrawal. J Hypertens Suppl. 1983;1:97–99.
- He XR, Wang W, Crofton J, Share L. Effects of 17β-estradiol on sympathetic activity and pressor response to phenylephrine in ovariectomized rats. Am J Physiol. 1998;275:R1202–R1208.
- Keil U, Chambless L, Remmers A. Alcohol and blood pressure: Results from the Luebeck Blood Pressure Study. Prev Med. 1989;18:1–10.
- Keil U, Chambless L, Filipiak B, Härtel U. Alcohol and blood pressure and its interaction with smoking and other behavioural variables: Results from the MONICA Augsburg Survey 1984–1985. J Hypertens. 1991;9:491–498.
- van Leer EM, Seidell JC, Kromhout D. Differences in the association between alcohol consumption and blood pressure by age, gender, and smoking. Epidemiology. 1994, 5:576–582.
- Wakabayashi I. Modification of the association of alcohol drinking with blood pressure by cigarette smoking. Blood Press. 2008;17:87–93.
- Wakabayashi I, Araki Y. Associations of alcohol consumption with blood pressure and serum lipids in Japanese female smokers and nonsmokers. Gend Med. 2009;6:290–299.
- Omvik P. How smoking affects blood pressure. Blood Press. 1996;5:71–77
- Okubo Y, Suwazono Y, Kobayashi E, Nogawa K. An association between smoking habits and blood pressure in normotensive Japanese men: A 5-year follow-up study. Drug Alcohol Depend. 2004;73:167–174.
- Després JP, Lemieux I, Bergeron J, Pibarot P, Mathieu P, Larose E, . Abdominal obesity and the metabolic syndrome: Contribution to global cardiometabolic risk. Arterioscler Thromb Vasc Biol. 2008;28:1039–1049.
- Welty FK. Cardiovascular disease and dyslipidemia in women. Arch Intern Med. 2001;161:514–522.
- Lovejoy JC, Sainsbury A; The Stock Conference 2008 Working Group. Sex differences in obesity and the regulation of energy homeostasis. Obes Rev. 2009;10:154–167.
- Yanes LL, Reckelhoff JF. Postmenopausal hypertension. Am J Hypertens. 2011;24:740–749.