Carotid Artery Compliance in Users of Plant Stanol Ester Margarine

ORIGINAL ARTICLE Carotid artery compliance in users of plant stanol ester margarine

OT Raitakari1, P Salo2 and M Ahotupa3

1Department of Clinical Physiology, University of Turku, Turku, Finland; 2Research Centre of Applied and Preventive Cardiovascular Medicine, University of Turku, Turku, Finland and 3Department of Physiology (MCA Research Laboratory), University of Turku, Turku, Finland

Objective: To investigate the effects of stanol ester margarine use in healthy subjects on arterial compliance, endothelial function and intima-media thickness. Design: Case–control study comparing regular stanol ester margarine users to non-users. Setting: Occupational health service clinic. Subjects: We recruited 50 cases and 50 controls (mean age 5178, range 26–65 years). All subjects were non-smokers and the study groups were matched for age and sex. As cases, we invited subjects who had been using regularly (daily) plant stanol ester margarine for a period of 2 years or longer. Non-invasive ultrasound was used to measure carotid artery compliance, carotid intima-media thickness and brachial artery flow-mediated endothelial dependent vasodilatation. Results: The carotid artery compliance was non-significantly higher in cases compared with controls, 1.8471.02 vs 1.5870.76 %/10 mm Hg (P ¼ 0.13). The difference in compliance became statistically significant (P ¼ 0.04) when the unbalance between the groups in family history of coronary artery disease and years of education were taken into account. There was also a significant dose–response relationship between stanol margarine use and carotid compliance, longer use being associated with higher compliance. Serum lipoproteins, blood pressure, flow-mediated dilation and intima-media thickness values did not differ between cases and controls. Conclusion: These data raise the possibility that regular stanol ester margarine use may be associated with beneficial changes in arterial compliance. Intervention studies are needed to test this hypothesis and to reveal possible mechanisms. European Journal of Clinical Nutrition (2008) 62, 218–224; doi:10.1038/sj.ejcn.1602705; published online 14 March 2007

Keywords: stanol ester; ultrasound; vascular; endothelial

Introduction total and LDL cholesterol concentrations in several intervention studies (Law, 2000). It has been estimated that that Plant stanols inhibit the absorption of both exogenous life-long consumption of plant stanol ester spread may lower dietary and endogenous biliary cholesterol from the prox- coronary events by about 20–25% (Law, 2000; Miettinen and imal small intestine. This action leads to reductions in both Gylling, 2004). Therefore, the use of plant stanols have been serum low-density lipoprotein (LDL) cholesterol and total recommended in guidelines for the prevention of coronary cholesterol concentrations, without affecting the concentra- heart disease (Expert Panel, 1996, 2004; Third Report of the tions of high-density lipoprotein (HDL) cholesterol or National Cholesterol Education Program (NCEP), 2002). triglycerides (Miettinen et al., 1995; Tammi et al., 2000; Luoto et al. (2004) assessed plant stanol ester margarine use Katan et al., 2003). The plant stanol ester margarine Benecol in Finland between 1996 and 1999. They reported that Raisio Group, Raisio, Finland has been shown to lower serum regular users of plant stanol ester margarine were more likely to be men and to have a higher household income than inconsistent users. Stanol margarine use was also associated Correspondence: Dr OT Raitakari, Department of Clinical Physiology, with a healthy lifestyle and diet, as well as a history of University of Turku, Kiinamyllynkatu 4-8, FIN-20521 Turku, Finland. cardiovascular disease. The investigators concluded that it E-mail: [email protected] Received 21 April 2006; revised 21 December 2006; accepted 25 January would be important to examine the health effects of stanol 2007; published online 14 March 2007 ester margarines. Arterial compliance and plant stanol margarine use OT Raitakari et al 219 Ultrasonographically assessed changes in arterial function (difference in age o3 years). All study subjects completed a and structure are useful means to study early pathophysio- questionnaire on physical activity, education and family risk of logical changes in the arteries relevant to the development of coronary heart disease. Family history was considered positive atherosclerosis. Carotid arterial compliance is a functional if study subjects’ first-degree relative had been diagnosed with marker of arterial elasticity. Compliance decreases with age premature coronary artery disease (before the age 65 years in and as a consequence of many disease states such as women and 55 years in men). Physical activity index was atherosclerosis and diabetes (Zieman et al., 2005). Risk calculated from the frequency, intensity and duration of factors, such as high LDL-cholesterol, elevated blood pres- reported physical activity (Raitakari et al., 1994). The frequency sure, obesity and smoking, have been associated with was assessed by inquiring how often the participants engaged decreased arterial compliance in cross-sectional studies in physical activities (choices included: 1 ¼ never, 2 ¼ monthly, (Zieman et al., 2005). Impaired carotid artery compliance 3 ¼ weekly, 4 ¼ twotothreetimesaweek,5¼ four to six times a has been implicated as an independent predictor for week and 6 ¼ daily). The intensity was assessed by inquiring cardiovascular events in high-risk individuals (Blacher ‘becoming breathless and sweating during exercise’ (choices et al., 1998). Arterial compliance may be influenced by included: 1 ¼ never, 2 ¼ a little, 3 ¼ much). The duration was life-style interventions, such as weight reduction, physical assessed by inquiring the usual time spend in one physical activity or changes in diet (Zieman et al., 2005). activity session (choices included: 1 ¼ o20 min, 2 ¼ 20–40 min, The thickness of the common carotid intima-media 3 ¼ 40–60 min and 4X60 min). A physical activity index was complex measured by ultrasound represents a structural calculatedfromtheproductofthesethreevariables.Theusers marker of preclinical atherosclerosis. Increased carotid were also asked to provide categorical information about the intima-media thickness correlates with vascular risk factors duration of their stanol product use (2 years, 2–4 years and (Poli et al., 1988; Haapanen et al., 1989) and predicts more than 4 years). The study was conducted according to the cardiovascular events in population groups (Salonen and guidelines of the Helsinki declaration and the study protocol Salonen, 1991; Chambless et al., 1997; Hodis et al., 1998; Bots was approved by the Joint Ethics Committee of Turku et al., 1999; O’Leary et al., 1999). A third commonly used University and Turku University Central Hospital. All subjects marker of arterial health is brachial artery flow-mediated gave their informed consent. dilatation (Celermajer et al., 1992). The dilatation response for increased blood flow is mainly mediated by nitric oxide Biochemical analyses. Fasting blood samples were taken on released from arterial endothelial cells Mullen et al. (2001). the same day as the ultrasound study. Serum total cholesterol, Brachial flow-mediated dilatation response correlates with HDL-cholesterol, triglycerides and plasma glucose concentra- coronary endothelial function tested with invasive methods tions were measured using standard enzymatic methods. LDL- (Anderson et al., 1995) and has been shown to predict cholesterol was calculated (Friedewald et al., 1972). Oxidized cardiovascular events (Chan et al., 2003; Gokce et al., 2003). LDL was measured by determining the level of LDL diene We performed a case–control study to investigate the effects conjugation, as described previously (Toikka et al., 2000). of regular stanol ester margarine use on these ultrasound Serum high-sensitivity C-reactive protein was analyzed using markers of subclinical atherosclerosis. an immunoturbidimetric assay (Orion Diagnostica, Finland).

Ultrasound studies. We used Acuson Sequoia 512 mainframe Materials and methods (Mountain View, CA, USA) with 8.0 MHz linear array transducer. Studies were performed after an overnight fast. Subjects The digitally stored scans were analyzed by one reader The cases (users) were 50 healthy subjects (25 men) recruited blinded to subjects’ details. from an occupational health service clinic in Turku, Finland. Information about the study was sent to several work places Carotid artery intima-media thickness. Left carotid artery was to be distributed among their employees. We invited healthy scanned following a standardized protocol. A moving image 20-to–65-year-old subjects who had been using regularly clip of the beginning of the common carotid artery with (daily) plant stanol margarine for period of 2 years or longer. duration of 5 s was acquired and stored in digital format The study subjects needed to be non-smokers (at least for for subsequent offline analysis. Carotid artery intima-media 5 years), and without antihypertensive and cholesterol thickness was measured in the far wall, as described lowering medications. The exclusion criteria were diabetes previously. We have reported previously 5.2% between and pregnancy. Potential participants were also informed observer and 6.4% between-visit coefficients of variation that alcohol and drug abuse were included in the exclusion in carotid artery intima-media thickness measurements criteria, and at the entry they filled in a form declaring that (Raitakari et al., 2003). they have no past or present history of alcohol or drug abuse. The controls (non-users) were also recruited from the Carotid artery compliance. To assess carotid artery compli- occupational health service clinic. For each user case, we ance, the best quality cardiac cycle was selected from the selected a nonuser control subject matched for sex and age 5-s clip image. The common carotid diameter 10 mm from

European Journal of Clinical Nutrition Arterial compliance and plant stanol margarine use OT Raitakari et al 220 carotid bifurcation was measured from the B-mode images skewed distributions. All analysis was performed using using ultrasonic calipers in end-diastole and end-systole, Statistical Analysis System, SAS (version 8.1), and statistical respectively. Brachial blood pressure was measured during significance was inferred at a two-tailed P-value p0.05. the ultrasound study with an automated sphygmo- manometer (Omron M4, Omron Matsusaka Co., Ltd., Japan). The ultrasound and concomitant brachial blood pressure Results measurements were used to calculate carotid artery compli-

ance, as ([DsÀDd]/Dd)/(PsÀPd), where Dd is the diastolic The characteristics of study subjects are shown in Table 1. diameter; Ds, the systolic diameter; Ps, systolic blood pressure There were equal number of men and women among users and Pd, diastolic blood pressure (Juonala et al., 2005). Thus, and controls, and the study groups were well matched for compliance is a marker of arterial elasticity, which measures age. The mean age was 51 years in both groups. Body mass the ability of the arteries to expand as a response to pulse index, serum lipids, glucose, C-reactive protein and blood pressure caused by cardiac contraction and relaxation. pressure values were similar between the groups. In addition, We have previously reported 2.7% between-visit coefficient the users and non-user groups did not differ in respect of of variation for carotid artery diastolic diameter measure- physical activity index, smoking history and exercise ments, and 19.5% for carotid artery compliance (Juonala capacity, but the users tended to report more often than et al., 2005). the non-users a positive family risk for coronary heart disease (34 vs 20%, P ¼ 0.11). Brachial artery test. Brachial artery diameter was measured The characteristics of users categorized according to the as described previously (Juonala et al., 2004). In brief, a duration of stanol ester margarine use are shown in Table 2. resting scan above the elbow was performed. Thereafter, The duration of stanol ester margarine use was associated hyperemia was induced by inflation of a cuff placed around with higher prevalence of positive family history of coronary the forearm followed by release. Subsequent scans were artery disease, lower body mass index and, borderline taken at 40, 60 and 80 s after the cuff release. Flow-mediated significantly, with higher education. Subjects who had dilatation, as a marker of endothelial function, was calcu- longer history of stanol product use also tended to be older lated as the maximal percent increase in arterial diameter and more often males (Table 2). Lipid profile and blood during hyperemia compared with the resting value. pressure values were similar in each category (data not This dilation response reflects endothelium-dependent shown). vasorelaxation capacity, because it can be blunted by The bivariate correlations coefficients between vascular simultaneous infusion of nitric oxide synthase inhibitor parameters and other continuous study variables in all (Mullen et al., 2001). We did not perform flow measurements subjects are shown in Table 3. Carotid compliance correlated to quantify the hyperemia stimulus after the cuff release. inversely with age, body mass index, total cholesterol, LDL This was justified by our earlier findings showing that the cholesterol, oxidized LDL, triglycerides, glucose, systolic flow stimulus does not correlate with the flow-mediated blood pressure and diastolic blood pressure. Intima-media response (Ja¨rvisalo et al., 2002). Nitrate-mediated, endothe- thickness correlated directly with age, body mass index, total lium-independent, vasodilatation was tested by scanning the cholesterol, LDL cholesterol and glucose. The associations artery 4 min after a sublingual dose of 1.25 mg isosorbide between flow-mediated dilatation and other continuous dinitrate. study variables were generally in the anticipated direction but did not reach statistical significance. The bivariate Exercise test. All subjects underwent maximal bicycle ex- relations between vascular parameters and two-level study ercise testing using a protocol with fixed load increments variables (family risk and sex) we assessed by linear every minute (20 W in men; 15 W in women). Subjects regression. Male sex was inversely associated with carotid exercised the protocol until limited by fatigue. The electro- compliance (b ¼À0.23570.102, P ¼ 0.023), flow-mediated cardiogram was continuously recorded. Blood pressure was dilation (b ¼À3.170.8, P ¼ 0.0004), and directly with in- measured at every load increment stage. Two indicators of tima-media thickness (b ¼ 0.05970.028, P ¼ 0.040). Family exercise capacity are reported: the estimated maximal risk of premature coronary artery disease was inversely oxygen uptake and maximal workload achieved (Arstila related with carotid compliance (b ¼À0.23170.115, et al., 1990). P ¼ 0.049). The comparison of ultrasound variables between users and Statistical methods. The comparisons between study groups non-users are shown in Table 4. The distributions of intima- were performed using t-test, w2 test and analysis of covar- media thickness, flow-mediated dilation and nitrate- iance, as appropriate. Associations between variables were mediated dilatation were similar between the groups. studied by calculating Pearson’s correlation coefficients and Carotid artery compliance was non-significantly higher in using multivariate linear regression analysis. Values for users compared to non-users, and the difference between the carotid artery compliance, triglycerides and C-reactive groups was statistically significant after adjustment for protein were log-transformed before analyses owing to family risk and years of education. We adjusted primarily

European Journal of Clinical Nutrition Arterial compliance and plant stanol margarine use OT Raitakari et al 221 Table 1 Characteristics of study subjects

Users (N ¼ 50) Controls (N ¼ 50) P-value

Age (years) 51785178 0.98 Body mass index (kg/m2) 25.072.7 25.073.2 0.99 Total cholesterol (mmol/l) 5.5470.94 5.5270.97 0.90 HDL-cholesterol (mmol/l) 1.6570.42 1.7970.58 0.16 Triglycerides (mmol/l) 1.2070.60 1.1070.53 0.39 LDL-cholesterol (mmol/l) 3.3670.88 3.2770.91 0.62 Oxidized LDL (mmol/l) 35.7712.0 33.8711.6 0.47 Glucose (mmol/l) 5.3570.54 5.4870.58 0.27 C-reactive protein (mg/l) 0.8070.73 1.1871.70 0.32 Systolic blood pressure (mm Hg) 133716 136719 0.34 Diastolic blood pressure (mmHg) 837985710 0.17 Physical activity index (unitless) 18715 15713 0.36 Ex-smoker (no, %) 13, 25% 15, 30% 0.66 Family risk (no, %)a 17, 34% 10, 20% 0.11 Education (total number of years) 14741575 0.11 Maximal oxygen uptake in stress test (ml/kg) 35.576.8 38.479.1 0.92 Maximal work load in stress test (watts) 224763 218768 0.62

The values are mean and s.d., or the number (percentage) of subjects. aFamily history was considered positive if study subjects’ first-degree relative had been diagnosed with premature coronary artery disease (before the age 65 years in women and 55 years in men).

Table 2 Characteristics of users according to the duration of stanol ester margarine use

2 years 2–4 years More than 4 years P for trend

No. 15 22 13 Age (years) 49.0710.6 51.778.0 52.475.1 0.16 Males (%) 40% 45% 69% 0.14 Family risk (%)a 13% 36% 54% 0.02 Education (years) 1373.2 13.674.4 15.674.1 0.08 Body mass index (kg/m2) 26.373.5 24.771.98 24.272.3 0.04 aFamily history was considered positive if study subjects’ first-degree relative had been diagnosed with premature coronary artery disease (before the age 65 years in women and 55 years in men).

Table 3 Bivariate correlation coefficients between vascular markers and other study variables

Carotid artery compliance Carotid artery intima-media thickness Brachial artery flow-mediated dilation

Age (years) À0.439*** 0.489*** À0.037 Body mass index (kg/m2) À0.237* 0.261** À0.111 Total cholesterol (mmol/l) À0.247* 0.204* À0.160 LDL cholesterol (mmol/l) À0.237* 0.279** À0.184 ox-LDL (mmol/l) À0.278** 0.158 À0.151 HDL cholesterol (mmol/l) 0.044 À0.135 0.090 Triglycerides (mmol/l) À0.219* 0.068 À0.167 C-reactive protein (mg/l) À0.036 0.152 0.061 Glucose (mmol/l) À0.350*** 0.300** À0.037 Systolic blood pressure (mm Hg) À0.327*** 0.143 À0.026 Diastolic blood pressure (mm Hg) À0.268** 0.007 À0.059 Physical activity (unitless) 0.176 À0.029 À0.143 Education (years) 0.124 À0.159 À0.090

***Po0.001, **Po0.01, *Po0.05.

for family risk and years of education because there seemed ox-LDL, triglycerides and blood pressure (Pp0.05 for carotid to be an unbalance in these variables between the groups. compliance between study groups). However, the result remained essentially similar after further In a subgroup analysis, comparing only users and non- adjustments for age, sex, body mass index, LDL cholesterol, users with negative family history, the carotid compliance

European Journal of Clinical Nutrition Arterial compliance and plant stanol margarine use OT Raitakari et al 222 was 2.0371.15 vs 1.6270.79 10%/mm Hg (P ¼ 0.06) in users a case–control design, we were unable to demonstrate (N ¼ 33) and non-users (N ¼ 40), respectively. In subjects significant differences in serum cholesterol concentrations with positive family history, the carotid compliance was between users and non-users. We did not have information 1.4770.56 in users (N ¼ 17) and 1.4370.62 10%mm Hg about the cholesterol values in the cases before they had (P ¼ 0.62) in non-users (N ¼ 10), respectively. started using stanol ester margarine. Therefore, it remains As there was a suggestion towards higher carotid com- unclear whether changes in cholesterol concentration due to pliance among the stanol ester margarine users, we exam- stanol ester margarine use have influenced arterial compli- ined whether a dose–response relation exists between the ance. History of high cholesterol and cardiovascular disease duration of stanol ester margarine use and compliance. have been associated with stanol ester margarine use. Carotid compliance values in the categories of stanol ester Therefore, subjects at increased risk presumably use plant margarine use duration are shown in the Figure 1. There was stanol ester margarine in order to control serum cholesterol a significant relation between the duration of stanol ester levels through dietary practices (Luoto et al., 2004). In line, margarine use and compliance when the effects of age, sex we found that longer use was associated with more prevalent and family risk were taken into account using multiple linear family history of coronary artery disease. regression (b ¼ 0.11270.041, P ¼ 0.008). And this association further remained significant after simultaneously controlling for body mass index and years of education (b ¼ 0.10770.043, P ¼ 0.016). The regression coefficient (b) a 2.2 P=0.12 for trend indicates the increase in log-transformed carotid compliance unadjusted 2.0 across each category of stanol ester margarine use duration. 1.8

1.6

1.4 Discussion 0.2 0.0 Non-users 2 years 2-4 years >4 years The effects of stanol ester margarine use on health, other P=0.029 for trend than on cholesterol metabolism, have not been previously b 2.2 adjusted for age and sex studied. We used a case–control design to test if markers of 2.0 arterial function and structure would differ between stanol 1.8 ester margarine users and non-users. The mean values for brachial artery flow mediated dilation and carotid artery 1.6 1.4 intima-media thickness were similar between users and non- 0.2 users. A significant difference in compliance was observed 0.0 Non-users 2 years 2-4 years >4 years when the imbalance between the groups in family history of coronary artery disease was taken into account. In addition, P=0.008 for trend

Carotid compliance (10% / mmHg) / Carotid compliance (10% c 2.2 there was a significant dose–response relationship between adjusted for age, sex and family risk stanol margarine use and carotid compliance, longer use 2.0 being associated with increased compliance. 1.8 In all subjects, carotid artery compliance correlated 1.6 inversely with total cholesterol and LDL cholesterol con- 1.4 centrations. Cholesterol lowering has been shown to 0.2 0.0 improve arterial compliance in patients with familial Non-users 2 years 2-4 years >4 years hypercholesterolemia (Tomochika et al., 1996; Smilde et al., Figure 1 The mean carotid compliance values are shown in non- 2000) and hypertension (Ferrier et al., 2002). Stanol ester users and in stanol ester margarine users according to the duration of margarine lowers serum total and LDL cholesterol concen- stanol use. (a) unadjusted mean values, (b) mean values adjusted for trations (Law 2000). However, in the present study, by using age and sex, (c) mean values adjusted for age, sex and family risk.

Table 4 Ultrasound variables. Mean, s.d. and 95% confidence intervals

Users Non-users P-valuea

Flow-mediated dilation (%) 8.074.4 (6.7 to 9.2) 8.074.5 (6.7 to 9.2) 0.86 Nitrate-mediated dilation (%) 25.4710.2 (22.4 to 28.4) 23.378.9 (20.6 to 25.9) 0.27 Intima-media thickness (mm) 0.6870.16 (0.63 to 0.72) 0.6670.13 (0.63 to 0.70) 0.92 Carotid compliance (%/10 mm Hg) 1.8471.02 (1.55 to 2.13) 1.5870.76 (1.37 to 1.80) 0.04

aAdjusted for family risk and years of education.

European Journal of Clinical Nutrition Arterial compliance and plant stanol margarine use OT Raitakari et al 223 In addition to cholesterol lowering stanol ester, the seen only a posteriori after adjustments. Therefore, the results Benecol margarine contains n-3 a linolenic fatty acid from should be considered as hypothesis generating. However, the rapeseed oil. Previous observations have suggested that n-3 data raise the possibility that regular stanol ester margarine fatty acids may increase arterial compliance. Systemic use may be associated with beneficial changes in arterial arterial compliance was improved in dyslipidaemic subjects compliance. Intervention studies are needed to test this after 7-week supplementation with n-3 eicosapentaenoic hypothesis and to reveal possible mechanisms. acid and docosahexaenoic acid (Nestel et al., 2002). In addition, dietary fish oil supplementation for 6 weeks improved arterial compliance in patients with non-insulin Acknowledgements dependent diabetes (McVeigh et al., 1994). Furthermore, n-3 rich flaxseed oil supplementation for 4 weeks increased This study was partially supported by Raisio Finland Inc. systemic arterial compliance in obese subjects (Nestel et al., 1997). Thus, the association between stanol ester margarine use and higher arterial compliance may partly be mediated References by a linolenic fatty acid. Our study had limitations. The use of stanol ester Anderson TJ, Uehata A, Gerhard MD, Meredith IT, Knab S, margarine was self-reported, and the exact information of Delagrange D et al. (1995). Close relation of endothelial function J Am Coll the amounts consumed and the regularity of use were in the human coronary and peripheral circulations. Cardiol 26, 1235–1241. lacking. Another limitation of our study is the blood pressure Arnett DK, Chambless LE, Kim H, Evans GW, Riley W (1999). measurement method used in the calculation of carotid Variability in ultrasonic measurements of arterial stiffness in the artery compliance. The pulse pressure used in the equations Atherosclerosis Risk in Communities Study. Ultrasound Med Biol 25, 175–180. to calculate carotid compliance was measured from the Arstila M, Impivaara O, Maki J (1990). New ergometric reference brachial artery. It would be more ideal to study the pulse values for clinical exercise tests. Scand J Clin Lab Invest 50, 747–755. pressure from the artery in question, because the use of Blacher J, Pannier B, Guerin AP, Marchais SJ, Safar ME, London GM brachial pressures may overestimate pulse pressure in central (1998). Carotid arterial stiffness as a predictor of cardiovascular and all-cause mortality in end-stage renal disease. Hypertension 32, arteries (Karamanoglu et al., 1993). This increase is a 570–574. consequence of pulse wave reflection from the periphery, Borow KM, Newburger JW (1982). Noninvasive estimation of central which augments the peak of the pressure wave in peripheral aortic pressure using the oscillometric method for analyzing arteries close to the reflection sites. However, Borow and systemic artery pulsatile blood flow: comparative study of indirect systolic, diastolic, and mean brachial artery pressure with Newburger (1982) have shown an excellent correlation of simultaneous direct ascending aortic pressure measurements. r ¼ 0.98 between systolic blood pressures and r ¼ 0.97 Am Heart J 103, 879–886. between diastolic pressures measured invasively from as- Bots ML, Hoes AW, Hofman A, Witteman JC, Grobbee DE (1999). cending aorta and non-invasively from brachial artery Cross-sectionally assessed carotid intima-media thickness relates supporting the assumption that brachial pulse pressure can to long-term risk of stroke, coronary heart disease and death as estimated by available risk functions. J Intern Med 245, 269–276. be used in deriving carotid compliance. The reported Celermajer DS, Sorensen KE, Gooch VM, Spiegelhalter DJ, Miller OI, variation in arterial compliance measured with the ultra- Sullivan ID et al. (1992). Noninvasive detection of endothelial sound method is rather high (between 15 and 20%) (Arnett dysfunction in children and adults at risk of atherosclerosis. Lancet 340, 1111–1115. et al., 1999; Toikka et al., 2002; Juonala et al., 2005). This is, Chambless LE, Heiss G, Folsom AR, Rosamond W, Szklo M, Sharrett in part, owing to the fact that several variables including AR et al. (1997). Association of coronary heart disease incidence arterial diameter and blood pressure measurements are used with carotid arterial wall thickness and major risk factors: the to derive this index. The large variation in carotid compli- Atherosclerosis Risk in Communities (ARIC) Study, 1987–1993. Am J Epidemiol 146, 483–494. ance is a limitation and the results need to be interpreted Chan SY, Mancini GB, Kuramoto L, Schulzer M, Frohlich J, with caution. The noise in the data is likely to weaken any Ignaszewski A (2003). The prognostic importance of endothelial observed relationship, and the present study may have been dysfunction and carotid atheroma burden in patients with underpowered to detect a significant difference in compli- coronary artery disease. J Am Coll Cardiol 42, 1037–1043. Expert Panel (1996). Prevention of coronary heart disease in clinical ance between the study groups. Although the users and non practice in Finland (in Finnish). Finnish Med J 51, 783–802. users were well matched for age and sex, there was a Expert Panel (2004). Treatment of dyslipidemias (in Finnish). tendency for higher prevalence for family history of Duodecim 120, 1794–1816. coronary artery disease. Finally, we did not measure other Ferrier KE, Muhlmann MH, Baguet JP, Cameron JD, Jennings GL, et al. variables potentially relevant in a study of vascular structure Dart AM (2002). Intensive cholesterol reduction lowers blood pressure and large artery stiffness in isolated systolic hypertension. and function, such alcohol consumption, estrogen use J Am Coll Cardiol 39, 1020–1025. among post-menopausal women subjects, vitamin intake Friedewald WT, Levy RI, Fredrickson DS (1972). Estimation of the and homocysteine. concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 18, 499–502. Our study has restrictions. The study was a case–control Gokce N, Keaney Jr JF, Hunter LM, Watkins MT, Nedeljkovic ZS, study without randomization and double-blind, and overall Menzoian JO et al. (2003). Predictive value of noninvasively the results were essentially negative. Significant results were determined endothelial dysfunction for long-term cardiovascular

European Journal of Clinical Nutrition Arterial compliance and plant stanol margarine use OT Raitakari et al 224 events in patients with peripheral vascular disease. J Am Coll hexaenoic acid increase systemic arterial compliance in humans. Cardiol 41, 1769–1775. Am J Clin Nutr 76, 326–330. Haapanen A, Koskenvuo M, Kaprio J, Kesaniemi YA, Heikkila K O’Leary DH, Polak JF, Kronmal RA, Manolio TA, Burke GL, Wolfson Jr (1989). Carotid arteriosclerosis in identical twins discordant for SK (1999). Carotid-artery intima and media thickness as a risk cigarette smoking. Circulation 80, 10–16. factor for myocardial infarction and stroke in older adults. Hodis HN, Mack WJ, LaBree L, Selzer RH, Liu CR, Liu CH et al. (1998). Cardiovascular Health Study Collaborative Research Group. The role of carotid arterial intima-media thickness in predicting N Engl J Med 340, 14–22. clinical coronary events. Ann Intern Med 128, 262–269. Poli A, Tremoli E, Colombo A, Sirtori M, Pignoli P, Paoletti R (1988). Juonala M, Viikari JS, Laitinen T, Marniemi J, Helenius H, Ro¨nnemaa Ultrasonographic measurement of the common carotid artery wall T et al. (2004). Interrelations between brachial endothelial thickness in hypercholesterolemic patients. A new model for the function and carotid intima-media thickness in young adults: quantitation and follow-up of preclinical atherosclerosis in living the cardiovascular risk in young Finns study. Circulation 110, human subjects. Atherosclerosis 70, 253–261. 2918–2923. Raitakari OT, Porkka KV, Taimela S, Telama R, Rasanen L, Viikari JS Juonala M, Ja¨rvisalo MJ, Ma¨ki-Torkko N, Ka¨ho¨nen M, Viikari JSA, (1994). Effects of persistent physical activity and inactivity on Raitakari OT (2005). Risk factors identified in childhood and coronary risk factors in children and young adults. The Cardio- decreased carotid artery elasticity in adulthood. The Cardiovas- vascular Risk in Young Finns Study. Am J Epidemiol 140, 195–205. cular Risk in Young Finns Study. Circulation 112, 1486–1493. Raitakari OT, Juonala M, Kahonen M, Taittonen L, Laitinen T, Maki- Ja¨rvisalo MJ, Ro¨nnemaa T, Volanen I, Kaitosaari T, Kallio K, Hartiala Torkko N et al. (2003). Cardiovascular risk factors in childhood JJ et al. (2002). Brachial artery dilatation responses in healthy and carotid artery intima-media thickness in adulthood–The children and adolescents. Am J Physiol Heart Circ Physiol 282, Cardiovascular Risk in Young Finns Study. JAMA 290, 2277–2283. H87–H92. Salonen JT, Salonen R (1991). Ultrasonographically assessed carotid Karamanoglu M, O’Rourke MF, Avolio AP, Kelly RP (1993). An morphology and the risk of coronary heart disease. Arterioscler analysis of the relationship between central aortic and peripheral Thromb 11, 1245–1249. upper limb pressure waves in man. Eur Heart J 14, 160–167. Smilde TJ, van den Berkmortel FW, Wollersheim H, van Langen H, Katan MB, Grundy SM, Jones P, Law M, Miettinen T, Paoletti R Kastelein JJ, Stalenhoef AF (2000). The effect of cholesterol (2003). Efficacy and safety of plant stanols and sterols in the lowering on carotid and femoral artery wall stiffness and thickness management of blood cholesterol levels. Mayo Clin Proc 78, in patients with familial hypercholesterolaemia. Eur J Clin Invest 965–978. 30, 473–480. Law M (2000). Plant sterol and stanol margarines and health. BMJ Tammi A, Ronnemaa T, Gylling H, Rask-Nissila L, Viikari J, Tuominen 320, 861–864. J et al. (2000). Plant stanol ester margarine lowers serum total and Luoto R, Simojoki M, Uutela A, Boice Jr JD, McLaughlin JK, Puska P low-density lipoprotein cholesterol concentrations of healthy (2004). Consistency of use of plant stanol ester margarine in children: the STRIP project. Special Turku Coronary Risk Factors Finland. Public Health Nutr 7, 63–68. Intervention Project. J Pediatr 136, 503–510. McVeigh GE, Brennan GM, Cohn JN, Finkelstein SM, Hayes RJ, Third Report of the National Cholesterol Education Program (NCEP) Johnston GD (1994). Fish oil improves arterial compliance in non- (2002) Expert Panel on Detection, Evaluation, and Treatment of insulin-dependent diabetes mellitus. Arterioscler Thromb 14, High Blood Cholesterol in Adults (Adult Treatment Panel III) final 1425–1429. report Circulation 106, 3143–3421. Miettinen TA, Gylling H (2004). Plant stanol and sterol esters in Toikka JO, Laine H, Ahotupa M, Haapanen A, Viikari JS, Hartiala JJ prevention of cardiovascular diseases. Ann Med 36, 126–134. et al. (2000). Increased arterial intima-media thickness and in vivo Miettinen TA, Puska P, Gylling H, Vanhanen H, Vartiainen E (1995). LDL oxidation in young men with borderline hypertension. Reduction of serum cholesterol with sitostanol-ester margarine in Hypertension 36, 929–933. a mildly hypercholesterolemic population. N Engl J Med 333, Toikka JO, Niemi P, Ahotupa M, Niinikoski H, Ronnemaa T, Viikari JS 1308–1312. et al. (2002). Decreased large artery distensibility in borderline Mullen MJ, Kharbanda RK, Cross J, Donald AE, Taylor M, Vallance P hypertension is related to increased in vivo low-density lipoprotein et al. (2001). Heterogenous nature of flow-mediated dilatation in oxidation. Scand J Clin Lab Invest 62, 301–306. human conduit arteries in vivo: relevance to endothelial dysfunc- Tomochika Y, Okuda F, Tanaka N, Wasaki Y, Tokisawa I, Aoyagi S et al. tion in hypercholesterolemia. Circ Res 88, 145–151. (1996). Improvement of atherosclerosis and stiffness of the Nestel PJ, Pomeroy SE, Sasahara T, Yamashita T, Liang YL, Dart AM thoracic descending aorta with cholesterol-lowering therapies in et al. (1997). Arterial compliance in obese subjects is improved familial hypercholesterolemia. Arterioscler Thromb Vasc Biol 16, with dietary plant n-3 fatty acid from flaxseed oil despite increased 955–962. LDL oxidizability. Arterioscler Thromb Vasc Biol 17, 1163–1170. Zieman SJ, Melenovsky V, Kass DA (2005). Mechanisms, patho- Nestel P, Shige H, Pomeroy S, Cehun M, Abbey M, Raederstorff D physiology, and therapy of arterial stiffness. Arterioscler Thromb (2002). The n-3 fatty acids eicosapentaenoic acid and docosa- Vasc Biol 25, 932–943.

European Journal of Clinical Nutrition