Subscriptions: Information about subscribing to Hypertension is online at

Transcription

1 Inflammation, Abdominal Obesity, and Smoking as Predictors of Hypertension Leo Niskanen, David E. Laaksonen, Kristiina Nyyssönen, Kari Punnonen, Veli-Pekka Valkonen, Ricardo Fuentes, Tomi-Pekka Tuomainen, Riitta Salonen and Jukka T. Salonen Hypertension 2004;44; ; originally published online Oct 18, 2004; DOI: /01.HYP Hypertension is published by the American Heart Association Greenville Avenue, Dallas, TX Copyright 2004 American Heart Association. All rights reserved. Print ISSN: X. Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: Subscriptions: Information about subscribing to Hypertension is online at Permissions: Permissions & Rights Desk, Lippincott Williams & Wilkins, 351 West Camden Street, Baltimore, MD Phone Fax: Reprints: Information about reprints can be found online at Downloaded from hyper.ahajournals.org by on April 25, 2006

2 Inflammation, Abdominal Obesity, and Smoking as Predictors of Hypertension Leo Niskanen, David E. Laaksonen, Kristiina Nyyssönen, Kari Punnonen, Veli-Pekka Valkonen, Ricardo Fuentes, Tomi-Pekka Tuomainen, Riitta Salonen, Jukka T. Salonen Abstract Development of hypertension has been linked to chronic low-grade inflammation. However, it is not known whether this connection is mediated by features of the metabolic syndrome or smoking, or their changes, which themselves have been linked to inflammation. We studied the predictive value of highly sensitive C-reactive protein (hs-crp), smoking, and abdominal obesity to the development of hypertension in an 11-year follow-up of a population-based study cohort comprising 379 middle-aged normotensive men. During the follow-up, 124 men (33%) developed hypertension. Men with hs-crp 3.0 mg/l were 2.8 (95% confidence interval, 1.2 to 6.6) more likely to develop hypertension than with hs-crp 1.0 mg/l even after adjustment for features of the metabolic syndrome, lifestyle factors, and their changes. Cigarette smoking was also associated with development of hypertension independently of inflammation and other confounders. Waist girth increased more in men who quit smoking than in other men. An increase in waist girth during follow-up strongly predicted incident hypertension. The decrease in smoking was not associated with a lower risk of hypertension in age-adjusted analyses. Hypertension is preceded by low-grade chronic inflammation in middle-aged white men independently of smoking or features of the metabolic syndrome. Furthermore, smoking may be a risk factor for hypertension. Although stopping smoking is beneficial with respect to health outcomes, the subsequent increase in weight and waist girth associated with smoking cessation may offset the decrease in the risk of hypertension that one may otherwise expect. (Hypertension. 2004;44: ) Key Words: obesity smoking prospective studies insulin resistance Chronic low-grade inflammation seems to be an early feature of many chronic degenerative disorders, including atherosclerosis, abdominal obesity, and type 2 diabetes. 1 3 These disorders are also commonly associated with hypertension, which itself has also been linked recently to inflammation. The most compelling evidence comes from the Women s Health Study, in which C-reactive protein (CRP) as a marker of low-grade inflammation predicted the development of hypertension in a cohort of female US health professionals during a follow-up of 7.8 years. 4 This effect was seen even in those with low baseline blood pressure levels and in those without other conventional cardiovascular risk factors, but no adjustment could be made for the presence of the metabolic syndrome, a possible mediator of this connection. 5 Smoking causes an acute rise in blood pressure, whereas the connection between chronic smoking and development of hypertension is still unclear. Smoking in its own right increases inflammation, 6 but smoking cessation may not reverse it. 7 Furthermore, stopping smoking commonly leads to weight gain. Weight gain is a well-established risk factor for hypertension. Moreover, weight gain and obesity, especially abdominal, appear to not only cause inflammation but may be preceded by inflammation. 8 Complicating the picture, increased alcohol intake is associated with hypertension and smoking but possibly decreased abdominal fat. 9 Observational studies are hampered with problems when describing these intertwined phenomena. If we are to understand chronic low-grade inflammatory processes in relation to the multifactorial origin of hypertension, the ideal study setting should describe various pathways leading to hypertension by taking into account the changes in these factors. In this study, we assessed the predictive value of low-grade inflammation in the development of hypertension in middleaged men during an 11-year follow-up, controlling for baseline blood pressure, various lifestyle factors, features of metabolic syndrome, and also the changes in smoking, alcohol intake, and waist girth during follow-up. Furthermore, we evaluate the role of smoking independently of inflammation and changes in waist girth with the development of hypertension. Received July 1, 2004; first decision July 22, 2004; revision accepted September 22, From the Departments of Medicine (L.N., D.E.L.) and Clinical Chemistry (K.P.), Kuopio University Hospital, Finland; Research Institute of Public Health (K.N., V-P.V., T-P.T., J.T.S.) and Department of Public Health and General Practice (R.S.), University of Kuopio, Finland; and Jurilab Ltd (R.F., J.T.S.), Kuopio, Finland. Reprint requests to Kristiina Nyyssönen, PhD, Research Institute of Public Health, University of Kuopio, PO Box 1627, FIN Kuopio, Finland. Kristiina.Nyyssonen@uku.fi 2004 American Heart Association, Inc. Hypertension is available at DOI: /01.HYP

3 860 Hypertension December 2004 Methods Participants Subjects were participants of the Kuopio Ischemic Heart Disease Risk Factor Study. 10 Participants were a random age-stratified sample of men living in Eastern Finland who were 42, 48, 54, or 60 years old at the baseline examination from 1987 to The recruitment and study design have been described previously in detail. 10 Repeat examinations for those who had undergone carotid ultrasound at baseline were performed from 1991 to 1994 (4-year follow-up) and 1998 to 2001 (11-year follow-up). In all, 854 men participated in the 11-year follow-up (90% of those alive). Men with hypertension or diabetes at baseline were excluded, leaving 379 men for analyses of incident hypertension during the 11-year follow-up. The university ethics committee approved the study. All participants gave their written informed consent. Definition of Hypertension Blood pressure was measured with a random-zero mercury sphygmomanometer (Hawksley and Sons). The protocol included 3 measurements while supine, 1 while standing, and 2 while sitting, with 5-minute intervals between measurements. The mean of the 2 sitting measurements was used as the systolic and diastolic blood pressure. Hypertension was defined at baseline and follow-up as systolic blood pressure 140 mm Hg, diastolic blood pressure 90 mm Hg, or use of antihypertensive medication. 11,12 Anthropometric and Biochemical Measurements Body mass index (BMI) was computed as the ratio of weight to the square of height (kg/m 2 ). Waist circumference was defined as the average of 2 measurements taken at the midpoint between the lowest rib and the iliac crest after inspiration and expiration. Fasting blood glucose was measured using a glucose dehydrogenase method after precipitation of proteins by trichloroacetic acid. Diabetes was defined as fasting blood glucose concentration 6.1 mmol/l (equivalent to plasma glucose 7.0 mmol/l) or a clinical diagnosis of diabetes with dietary, oral, or insulin treatment. 13 Serum insulin was determined with a Novo Biolabs radioimmunoassay kit (Novo Nordisk). HDL fractions were separated from fresh serum by combined ultracentrifugation and precipitation. The cholesterol contents of lipoprotein fractions and serum triglycerides were measured enzymatically. Fibrinogen was measured based on the clotting of diluted plasma with excess thrombin. Measurement of Highly Sensitive CRP Serum CRP was measured with an immunometric assay (Immulite High Sensitivity CR Assay; DPC). 14 We used CRP cut-offs of 1.0 mg/l and 3.0 mg/l, as recommended by the Center for Disease Control and the American Heart Association. 15 To limit confounding with acute infection or occult disease, we excluded men with CRP concentrations 10.0 mg/l. Other Assessments Assessments of medical history and medications, smoking, alcohol consumption, adult socioeconomic status, and moderate-to-vigorous leisure-time physical activity have been described previously. 16,17 Dietary intake of saturated fat, sodium, potassium, and fruits and vegetables was measured with 4-day food records as grams per day and adjusted by regression analysis for energy intake. 18 Highresolution B-mode ultrasonography was used to examine a 1.0- to 1.5-cm section at the distal end of the left and right common carotid artery proximal to the carotid bulb, as explained in detail previously. 19 Statistical Analyses Differences in baseline characteristics between men who developed hypertension and those who did not were assessed with Student t test, and where indicated, the 2 test. Fibrinogen is an acute phase reactant and as such, a marker of inflammation that has a high correlation with CRP levels (r 0.50). To avoid problems with multicollinearity, we adjusted fibrinogen by CRP using linear regression before including it in the logistical regression models. To investigate the associations of CRP concentrations with hypertension, CRP concentrations were categorized using the cut-offs 1.0 mg/l, 1.0 to 2.9 mg/l, and 3.0 to 9.9 mg/l. 15 Cigarette smoking was categorized as not at all, 1 to 19 cigarettes per day, and 20 cigarettes per day. Alcohol intake was converted into grams of ethanol per week and classified as none at all, 1 to 83 g per week, and 84 g per week (84 g of ethanol is 7 drinks). Waist circumference was categorized into thirds (only 17% and 3% of nonhypertensive men had waist girths 94 and 102 cm, values that have been recommended as action levels for prevention of chronic disease). 20 The variables in question were entered into logistic regression models adjusting for age and potential mediating or confounding variables. The association of changes in waist circumference, smoking (as cigarettes per day), and alcohol intake with incident hypertension were similarly analyzed as continuous variables in logistic regression models. The covariates for the logistic regression models were forced into the model. Variables are given as means SD, except for variables with a skewed distribution (CRP, insulin, triglycerides, alcohol intake, and physical activity), which are given as medians and interquartile ranges, and proportions, which are given as percentages. In analyses using continuous variables, skewed variables were log transformed. The mean was substituted for missing values (n 10 to 23) of covariates. Statistical significance was considered to be P All statistical analyses were performed with SPSS 11.0 for Windows. Results Baseline Clinical Characteristics During the 11-year follow-up, men who developed hypertension were heavier, had a larger waist, and were more dyslipidemic at baseline (Table 1). They also had higher blood pressure. The prevalence of the metabolic syndrome was uncommon in these relatively lean men without hypertension or diabetes but was more common in those who developed hypertension during the follow-up. Smokers more commonly developed hypertension, but the change in the number of cigarettes smoked did not differ between men who developed and did not develop hypertension. Alcohol intake was similar in both groups, but increased alcohol intake at the 4-year follow-up was more common in men who developed hypertension by the 11-year follow-up. Baseline CRP concentrations were much higher in men who later developed hypertension. CRP at Baseline and Hypertension Men with CRP concentrations 3 mg/l were 3.6 more likely to develop hypertension than men with levels 1.0 mg/l in age-adjusted analyses (Table 2). Adjustment for baseline blood pressure, lifestyle factors, and cardiovascular disease (model 2) only marginally weakened the association. After additional adjustment for features of the metabolic syndrome at baseline (model 3) or further for changes in waist circumference, number of cigarettes smoked per day, or intake of alcohol (model 4) during the first 11 years of follow-up, men with CRP levels 3 mg/l were still 2.8 more likely to develop hypertension during the 11-year follow-up. Adjustment for changes in waist girth, smoking, or alcohol intake over the 4-year period gave similar results as in model 4. Adjustment for development of cardiovascular disease during follow-up also had little effect on the results. Inclusion of BMI in models 2 through 4 in place of or in

4 Niskanen et al CRP, Smoking, and Hypertension 861 TABLE 1. Baseline Characteristics of Men Who Developed Hypertension During Follow-Up and Those Who Did Not Descriptive Variables Nonhypertensive at Follow-Up Hypertensive at Follow-Up P N Age (years) 50.2 (6.6) 51.8 (6.7) Smokers (%) Change in smoking during follow-up (for smokers) At 4 years (n 133), cigarettes per day 3 (11) 4 (9) 0.88 At 11 years (n 132), cigarettes per day 8 (15) 4 (13) 0.14 Alcohol consumption, g/week 31 (6, 87) 28 (7, 89) 0.83 Change in alcohol intake during follow-up (for nonabstainers) At 4 years (n 314), g/week 5 (119) 22 (74) At 11 years (n 346), g/week 4 (150) 25 (112) 0.19 Cardiovascular disease, % Maximum carotid intima thickness, mm 0.88 (0.78, 0.98) 0.91 (0.80, 1.05) Adult socioeconomic status score 6.9 (3.9) 6.8 (4.0) 0.73 Systolic blood pressure, mm Hg 121 (9) 128 (8) Diastolic blood pressure, mm Hg 79 (6) 84 (6) Body mass index, kg/m (2.7) 26.1 (2.9) Waist circumference, cm 85.3 (7.9) 88.3 (8.0) Change in waist girth during follow-up At 4 years 4.4 (4.1) 5.3 (3.9) At 11 years 7.1 (5.8) 8.7 (5.9) Serum LDL cholesterol, mmol/l 3.8 (0.9) 4.0 (1.1) Serum HDL cholesterol, mmol /L 1.40 (0.29) 1.28 (0.26) Serum triglycerides, mmol/l 0.96 (0.75, 1.38) 1.10 (0.83, 1.64) Fasting blood glucose, mmol /L 4.48 (0.45) 4.52 (0.47) 0.42 Fasting serum insulin, pmol/l 7.9 (6.2, 10.3) 8.9 (7.0, 11.2) Fibrinogen (g/l, unadjusted) 2.81 (0.50) 3.02 (0.50) Metabolic syndrome (NCEP), % Moderate and vigorous LTPA, min/week 143 (62, 256) 128 (56, 241) 0.24 Serum CRP, g/dl 0.81 (0.50, 1.46) 1.24 (0.76, 2.33) Values are means (SD), medians (interquartile ranges), or percentages. Higher adult socioeconomic status score means lower socioeconomic status. NCEP indicates National Cholesterol Education Program 29,30 ; LTPA, leisure-time physical activity. addition to waist girth had little effect on the results. CRPadjusted fibrinogen was not associated with incident hypertension, and inclusion of CRP-adjusted fibrinogen in the models did not affect the association of CRP with incident hypertension. The risk associated with CRP levels was graded, and men with levels between 1.0 and 3.0 mg/l were also at increased risk for developing hypertension. CRP concentrations also predicted incident hypertension at 11 years even after adjustment for the average maximum thickness of the intima carotid for the left and right carotid arteries in addition to technical covarates (odds ratio [OR] for CRP, 3 mg/l versus 1 mg/l 2.4 to 2.9 in models 2 to 4; P to for the trend). CRP levels 3.0 mg/l were quite uncommon (n 39) in these relatively lean men without hypertension or diabetes at baseline, but the results were very similar when using tertiles to categorize CRP (data not shown). CRP concentrations also predicted use of blood pressure medication at 11 years (OR for CRP, 3 mg/l versus 1 mg/l 3.1 to 3.5 in all models; P to for the trend) and incident hypertension when defining hypertension at the 11-year follow-up solely by systolic and diastolic blood pressure rather than including medication in the definition (data not shown). There was no evidence of modification of the association of CRP with hypertension after stratification by median systolic blood pressure, BMI, or smoking status. Smoking at Baseline and Hypertension Men who smoked 20 cigarettes per day were more than twice as likely to develop hypertension (Table 2). Additional adjustment for features of the metabolic syndrome at baseline (model 3) weakened the association. However, after taking

5 862 Hypertension December 2004 TABLE 2. ORs (95% CI) of Developing Hypertension During the 11-Year Follow-Up According to CRP Concentrations, Smoking, Alcohol Intake, and Waist Circumference, Respectively Predictive Variable Model 1 Model 2 Model 3 Model 4 CRP concentrations* mg/l (n 210) 1 (reference) mg/l (n 130) 1.82 ( ) 1.83 ( ) 1.66 ( ) 1.78 ( ) mg/l (n 39) 3.55 ( ) 3.31 ( ) 2.87 ( ) 2.81 ( ) P for trend Cigarettes per day 0(n 260) 1 (reference) (n 41) 1.57 ( ) 1.83 ( ) 1.76 ( ) 2.50 ( ) 20 (n 77) 2.38 ( ) 2.37 ( ) 1.95 ( ) 3.37 ( ) P for trend Alcohol, g/week 0(n 49) 1 (reference) (n 232) 1.25 ( ) 0.89 ( ) 0.90 ( ) 0.91 ( ) 84 (n 98) 1.46 ( ) 0.84 ( ) 0.94 ( ) 0.92 ( ) P for trend Waist girth at baseline 82.5 cm (n 127) 1 (reference) cm (n 121) 2.23 ( ) 1.96 ( ) 1.43 ( ) 1.57 ( ) 88.5 (n 131) 2.32 ( ) 1.83 ( ) 1.10 ( ) 1.29 ( ) P for trend *Model 1, Age-adjusted; model 2, adjusted for age, smoking (none, 1 19, and 20 cigarettes per day), alcohol intake (abstinent, 1 83 g, and 84 g ethanol per week), adult socioeconomic status, leisure-time physical activity, presence of cardiovascular disease, intake of energy-adjusted dietary factors (dietary intake of saturated fat, sodium, potassium, and fruits and vegetables), and baseline systolic blood pressure; model 3, adjusted for variables in model 2 and waist girth ( 82.5 cm, cm, and 88.5 cm), and concentrations of insulin, glucose, and HDL cholesterol; model 4, adjusted for variables in model 3 and changes in waist girth, smoking (cigarettes per day), and alcohol intake (g/day) during the 11-year period. Models as above, except that CRP categories were used as a covariate in place of smoking in models 2 4. Models as above, except that CRP categories were used as a covariate in place of alcohol intake in models 2 4. Models as above, except that CRP categories were used as a covariate in place of waist girth in models 3 4. into account the changes in waist circumference, number of cigarettes smoked per day, or in the intake of alcohol (model 4) during the 11-year follow-up, smoking at baseline was associated with a graded increase in the risk of hypertension during the 11-year follow-up. The increase in risk of smoking in model 4 compared with model 3 was a result of adjustment for the change in smoking, alcohol intake, and waist circumference. Adjustment for the changes in waist girth, smoking, or alcohol intake during the 4-year period gave essentially identical results as in model 4. Abdominal Obesity at Baseline and Hypertension Waist girth at baseline more than doubled the likelihood of hypertension in age-adjusted analyses (Table 2). Further adjustment for baseline systolic blood pressure and lifestyle factors (model 2) attenuated the association, which was mainly attributable to inclusion of CRP and baseline systolic blood pressure into the model. Additional adjustment for variables related to the metabolic syndrome (model 3) further attenuated the association such that it was no longer significant. Adjustment for the changes in waist girth, smoking, or alcohol intake during the 4-year period gave essentially identical results as adjustment for changes during the 11-year follow-up. Alcohol Intake at Baseline and Hypertension Men reporting consumption of 7 drinks per week were 1.5 more likely than nondrinkers to develop hypertension during the 11-year follow-up in age-adjusted analyses, but the association was not significant (Table 2). The ORs attenuated even further in multivariate analyses. Incident Hypertension and Changes in Smoking, Alcohol Intake, and Waist Girth Of the changes in smoking, alcohol intake, and leisure-time physical activity, only the changes in smoking and alcohol intake were associated with incident hypertension. When adjusting for only age and baseline smoking, the change in the number of cigarettes smoked was not associated with development of hypertension (Table 3). However, with further multivariate adjustment, the association strengthened, which was attributable mainly to the inclusion of baseline waist girth and HDL cholesterol levels in model 3 and to the inclusion of the change in waist girth to model 4. The association of the change in number of cigarettes smoked per

6 Niskanen et al CRP, Smoking, and Hypertension 863 TABLE 3. ORs for Developing Hypertension According to the Change in Smoking, Alcohol Intake, and Waist Girth During the 11-Year Follow-Up Predictive Variable Model 1 Model 2 Model 3 Model 4 Change in smoking (for an decrease of 10 cigarettes per day)* 4-year change 0.97 ( ) 0.90 ( ) 0.65 ( ) 0.58 ( ) 11-year change 0.88 ( ) 0.86 ( ) 0.56 ( ) 0.51 ( ) Change in alcohol intake (for an increase of 84 g ethanol per week, equivalent to 1 drink per day) 4-year change 1.34 ( ) 1.37 ( ) 1.40 ( ) 1.34 ( ) 11-year change 1.13 ( ) 1.16 ( ) 1.18 ( ) 1.14 ( ) Change in waist (for a 5-cm increase during follow-up) 4-year change 1.38 ( ) 1.55 ( ) 1.57 ( ) 1.59 ( ) 11-year change 1.32 ( ) 1.27 ( ) 1.27 ( ) 1.31 ( ) *Model 1, Adjusted for age and baseline smoking (none, 1 19, and 20 cigarettes per day); model 2, adjusted for age, baseline CRP categories, smoking, alcohol intake (abstinent, 1 83 g, and 84 g ethanol per week), adult socioeconomic status, leisure-time physical activity, presence of cardiovascular disease, intake of energy-adjusted dietary factors (dietary intake of saturated fat, sodium, potassium, and fruits and vegetables), and baseline systolic blood pressure; model 3, adjusted for variables in model 2 and waist girth ( 82.5 cm, cm, and 88.5 cm), and concentrations of insulin, glucose, and triglycerides; model 4, adjusted for variables in model 3 and changes in waist girth and alcohol intake (g/day) during the 11-year period. Models as above, except that alcohol categories were used as a covariate in place of smoking in model 1, and the change in smoking (cigarettes per day) during follow-up was used as a covariate in place of change in alcohol intake in model 4. Models as above, except that baseline waist categories were used as a covariate in place of baseline smoking in model 1, baseline waist circumference was also included in place of CRP categories in model 2, and adjustment for changes in alcohol intake and smoking was made in model 4. day during the 11-year follow-up with incident hypertension was significant also in analyses confined to smokers (eg, in models 2 to 4; OR, 0.42 to 0.56 for a 10-cigarette decrease; P to 0.006). The relevance of taking into account the change in waist girth in the association of the change in smoking with development of hypertension is also illustrated by the increase in waist circumference with smoking cessation. The waist girth of nonsmokers who remained nonsmokers increased 7.3 cm (95% confidence interval [CI], 6.6 to 8.1), and that of smokers at baseline who remained smokers increased 7.0 cm (95% CI, 5.7 to 8.3) during the 11-year follow-up. In contrast, the waist girth of men who quit during the follow-up increased by 10.4 cm (95% CI, 8.7 to 12.2; P for the difference between groups after adjustment for age and baseline waist girth). An increase in alcohol intake also increased the risk of hypertension during the 11-year follow-up. Of the changes during follow-up in the metabolic variables shown in model 3, only the change in waist girth was associated with development of hypertension (Table 3). Multivariate adjustment had little effect on the association. Discussion This study extends the findings of Women s Health Study 4 of chronic low-grade inflammation as a predictor of hypertension in men. Moreover, this association is not explained by various lifestyle factors or features of the metabolic syndrome. Although hypertension is another member of a long list of disorders preceded by low-grade inflammation, the mechanisms of the mediating factors remain elusive. Based on the results of this study, inflammation as measured by highly sensitive CRP concentrations is clearly more than a marker of abdominal obesity or the metabolic syndrome and is not explained by smoking or presence of cardiovascular disease. There could well be an underlying genetic propensity, although little data currently exist to support this conviction. On the other hand, low-grade inflammation as measured by CRP levels could still be a marker of highly active cytokineproducing adipose tissue infiltrated by macrophages and leading to complement activation and release of vasoactive cytokines and subsequent endothelial dysfunction; 8,21 and any increase in the amount of fat tissue, especially in the abdominal area, could lead to more active function and development of hypertension. This mechanism is unlikely to explain a major part of the association of CRP levels with incident hypertension because adjustment for waist girth and other variables related to the metabolic syndrome only slightly attenuated the association. Elevated CRP concentrations may also be secondary to artery wall thickening and atherosclerotic burden, but adjustment for carotid intima thickness did not alter the predictive value of CRP. CRP itself may cause endothelial dysfunction. 22 Furthermore, CRP may upregulate angiotensin type 1 receptors, leading to proliferation of vascular smooth muscle cells 23 and subsequently to hypertension. In this study, the predictive role of smoking with development of hypertension was somewhat surprising. Middle-aged men who decreased the number of cigarettes smoked daily by 10 were not less likely to develop hypertension in ageadjusted analyses. However, when taking into account waist at baseline and the increase in waist girth in addition to other variables, decreasing the number of cigarettes smoked by 10 halved their adjusted risk of hypertension. Thus, even smok-

7 864 Hypertension December 2004 ing cessation did not seem to protect from hypertension, most likely because of the associated increase in abdominal obesity. There are previous reports suggesting that smoking or smoking cessation may indeed be a harbinger for development of hypertension, but generally, smoking has not been considered a causative factor of hypertension. One of the mechanisms may be smoking-induced renal damage. 28 Alcohol intake was weakly associated with development of hypertension, and after multivariate adjustment, the connection was abolished. On the other hand, an increase in alcohol intake during the follow-up was clearly associated with a higher incidence of hypertension. The overall weak association of alcohol intake with incident hypertension may be attributable to the difficulty in the reliable recording of alcohol intake. The strengths of this study are the population-based prospective design, high participation rate, detailed assessment of variables related to lifestyle, and the metabolic syndrome and use of a uniform and constant definition of hypertension with careful measurement of blood pressure according to a standard protocol. Importantly, the changes of major variables related to lifestyle and the metabolic syndrome during follow-up were determined. The limitations are that of the various inflammatory markers, only a single baseline measurement of CRP was taken into account. However, this, if anything, leads to conservative estimate of the association in question. The study was performed in middle-aged white men, and generalization to other groups should be viewed with caution. Perspectives We have demonstrated that hypertension is preceded by low-grade chronic inflammation as assessed by CRP in middle-aged white men. This connection was not explained by smoking, baseline systolic blood pressure, or features of metabolic syndrome. Furthermore, smoking may be a risk factor for hypertension. Although stopping smoking is extremely beneficial with respect to cardiovascular risk and other health outcomes, the subsequent weight gain and increase in waist girth associated with smoking cessation may offset the decrease in risk of hypertension that one may otherwise expect. Therefore, efforts to curtail smoking should also take more seriously into account the consequences of the subsequent weight gain because the early prevention of abdominal obesity may be easier than later attempts at weight loss and control of hypertension. This study illustrates that multiple pathways seem to lead to hypertension independently of inflammation, including an increase in waist girth and smoking, even when the relevant changes during the follow-up are taken into account. Acknowledgments The Kuopio Ischemic Heart Disease Risk Factor Study was supported by grants from the Academy of Finland (grants 41471, 45155, , and ), the Ministry of Education of Finland (grants 167/722/96, 157/722/97, and 156/722/98), and the National Heart, Lung, and Blood Institute of the United States (grant HL44199). We thank the staff of the Research Institute of Public Health, University of Kuopio, and Kuopio Research Institute of Exercise Medicine for data collection in the Kuopio Ischemic Heart Disease Risk Factor Study. References 1. Ridker PM, Buring JE, Shih J, Matias M, Hennekens CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation. 1998;98: Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein, IL 6, and risk of developing type 2 diabetes mellitus. JAmMed Assoc. 2001;286: Laaksonen DE, Niskanen L, Nyyssönen K, Punnonen K, Tuomainen TP, Valkonen VP, Salonen R, Salonen JT. C-reactive protein and the development of the metabolic syndrome and diabetes in middle-aged men. Diabetologia. 2004;47: Sesso HD, Buring JE, Rifai N, Blake GJ, Gaziano JM, Ridker PM. C-reactive protein and the risk of developing hypertension. JAmMed Assoc. 2003;290: Grundy SM. Inflammation, hypertension, and the metabolic syndrome. J Am Med Assoc. 2003;290: Bazzano LA, He J, Muntner P, Vupputuri S, Whelton PK. Relationship between cigarette smoking and novel risk factors for cardiovascular disease in the United States. Ann Intern Med. 2003;138: Crook MA, Scott DA, Stapleton JA, Palmer RM, Wilson RF, Sutherland G. Circulating concentrations of C-reactive protein and total sialic acid in tobacco smokers remain unchanged following one year of validated smoking cessation. Eur J Clin Invest. 2000;30: Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW Jr. Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest. 2003;112: Greenfield JR, Samaras K, Jenkins AB, Kelly PJ, Spector TD, Campbell LV. Moderate alcohol consumption, dietary fat composition, and abdominal obesity in women: evidence for gene-environment interaction. J Clin Endocrinol Metab. 2003;88: Salonen JT. Is there a continuing need for longitudinal epidemiologic research? The Kuopio Ischaemic Heart Disease Risk Factor Study. Ann Clin Res. 1988;20: Joint National Committee. The sixth report of the Joint National Committee on prevention, detection, evaluation, and treatment of high blood pressure. Arch Intern Med. 1997;157: World Health Organization-International Society of Hypertension World Health Organization-International Society of Hypertension Guidelines for the Management of Hypertension. Guidelines Subcommittee. J Hypertens. 1999;17: Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional report of a WHO consultation. Diabet Med. 1998;15: Laaksonen DE, Niskanen L, Punnonen K, Nyyssönen K, Tuomainen T-P, Salonen R, Rauramaa R, Salonen JT. Sex hormones, inflammation and the metabolic syndrome: a population-based study. Eur J Endocrinol. 2003;149: Pearson TA, Mensah GA, Alexander RW, Anderson JL, Cannon RO III, Criqui M, Fadl YY, Fortmann SP, Hong Y, Myers GL, Rifai N, Smith SC Jr, Taubert K, Tracy RP, Vinicor F. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107: Lynch JW, Kaplan GA, Cohen RD, Kauhanen J, Wilson TW, Smith NL, Salonen JT. Childhood and adult socioeconomic status as predictors of mortality in Finland. Lancet. 1994;343: Laaksonen DE, Lakka HM, Salonen JT, Niskanen LK, Rauramaa R, Lakka TA. Low levels of leisure-time physical activity and cardiorespiratory fitness predict development of the metabolic syndrome. Diabetes Care. 2002;25: Laaksonen DE, Nyyssonen K, Niskanen L, Rissanen TH, Salonen JT. Dietary and serum linoleic and polyunsaturated fatty acids predict cardiovascular mortality in middle-aged men. Arch Intern Med. In press. 19. Salonen JT, Salonen R. Ultrasound B-mode imaging in observational studies of atherosclerotic progression. Circulation. 1993;87:II56 II Lean ME, Han TS, Seidell JC. Impairment of health and quality of life in people with large waist circumference. Lancet. 1998;351:

8 Niskanen et al CRP, Smoking, and Hypertension Wellen KE, Hotamisligil GS. Obesity-induced inflammatory changes in adipose tissue. J Clin Invest. 2003;112: Verma S, Yeh ET. C-reactive protein and atherothrombosis beyond a biomarker: an actual partaker of lesion formation. Am J Physiol Regul Integr Comp Physiol. 2003;285:R1253 R Wang CH, Li SH, Weisel RD, Fedak PW, Dumont AS, Szmitko P, Li RK, Mickle DA, Verma S. C-reactive protein upregulates angiotensin type 1 receptors in vascular smooth muscle. Circulation. 2003;107: Rywik SL, Williams OD, Pajak A, Broda G, Davis CE, Kawalec E, Manolio TA, Piotrowski W, Hutchinson R. Incidence and correlates of hypertension in the Atherosclerosis Risk in Communities (ARIC) Study and the Monitoring Trends and Determinants of Cardiovascular Disease (POL-MONICA) Project. J Hypertens. 2000;18: Lee DH, Ha MH, Kim JR, Jacobs DR Jr. Effects of smoking cessation on changes in blood pressure and incidence of hypertension: a 4-year follow-up study. Hypertension. 2001;37: Janzon E, Hedblad B, Berglund G, Engstrom G. Changes in blood pressure and body weight following smoking cessation in women. J Intern Med. 2004;255: Schut AF, Sayed-Tabatabaei FA, Witteman JC, Avella AM, Vergeer JM, Pols HA, Hofman A, Deinum J, van Duijn CM. Smoking-dependent effects of the angiotensin-converting enzyme gene insertion/deletion polymorphism on blood pressure. J Hypertens. 2004;22: Orth SR. Effects of smoking on systemic and intrarenal hemodynamics: influence on renal function. J Am Soc Nephrol. 2004;15(suppl 1):S58 S National Cholesterol Education Program. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). J Am Med Assoc. 2001;285: Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, Tuomilehto J, Salonen JT. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. J Am Med Assoc. 2002; 288: