Nut consumption and incidence of hypertension: The SUN prospective cohort *

Transcription

1 Nutrition, Metabolism & Cardiovascular Diseases (2009) xx, 1e7 available at journal homepage: Nut consumption and incidence of hypertension: The SUN prospective cohort * E.H. Martínez-Lapiscina a, A.M. Pimenta b, J.J. Beunza a, M. Bes-Rastrollo a, J.A. Martínez a, M.A. Martínez-González a, * a Department of Preventive Medicine and Public Health, Medical School, Clínica Universidad de Navarra, Irunlarrea 1, E Pamplona, Navarra, Spain b Department of Maternal and Child Nursing and Public Health, Nursing School, Universidade Federal de Minas Gerais, Avenida Alfredo Balena, 190. Belo Horizonte, Minas Gerais, CEP Brazil Received 5 November 2008; received in revised form 7 March 2009; accepted 23 April 2009 KEYWORDS Hypertension; Nuts; Nutrition; Epidemiology; Cohort studies Abstract Background and aims: The consumption of tree nuts could reduce the risk of hypertension, but scarce research has been done to evaluate this potential association. We assessed the association between nut consumption and the incidence of hypertension among 9919 Spanish university graduates followed-up biennially for a median of 4.3 years. Methods and results: Food habits were assessed with a validated 136-item food-frequency questionnaire. Nut consumption was classified into four categories: rarely/never, 1e3/month, 1/week, and 2þ/week. A participant was classified as an incident case of hypertension when, being free of hypertension at baseline, he/she subsequently reported a physician-made diagnosis of hypertension in at least one of the follow-up questionnaires. The incidence of hypertension was 12.4 per 1000 person-years. We found no association between nut consumption and incidence of hypertension after adjusting for sex, age and other dietary and non-dietary potential confounders (hazard ratio [HR] for those in the highest vs. lowest nut consumption category Z 0.77 [IC 95%: 0.46e1.30] p Z 0.795). Results were not modified when we stratified them analyses according to sex or to body mass index. Conclusion: Our data do not provide evidence for an inverse association between nut consumption and incident hypertension in our cohort. Further results from cohorts and trials with a higher baseline risk of hypertension should be obtained to test this relationship. ª 2009 Elsevier B.V. All rights reserved. Abbreviations: CVD, cardiovascular disease; IHD, ischemic heart disease; BP, blood pressure; PHS, Physicians Health Study; SUN Project, Seguimiento Universidad de Navarra (University of Navarra Follow-up Study); BMI, body mass index; HR, hazard ratio; MET, metabolic equivalent index; DASH, Dietary Approaches to Stop Hypertension; MDP, Mediterranean dietary pattern. * The SUN Study has received funding from the Spanish Government, Instituto de Salud Carlos III (Grants PI030678, PI040233, PI070240, RD06/0045, and G03/140), the Navarra Regional Government (PI141/2005) and the University of Navarra. * Corresponding author. Tel.: þ ; fax: þ address: mamartinez@unav.es (M.A. Martínez-González) /$ - see front matter ª 2009 Elsevier B.V. All rights reserved. doi: /j.numecd

2 2 E.H. Martínez-Lapiscina et al. Introduction Hypertension is a major public health problem, affecting approximately 1 billion individuals,[1] causing 7.6 million premature deaths and 6% of all cases of disability-adjusted life years worldwide [2]. In European countries hypertension prevalence is increasing [3]. Ischemic heart disease (IHD) and stroke are the two main cardiovascular causes of mortality around the world, and hypertension explains 47% and 54% of deaths by IHD and stroke, respectively [2]. Lifestyle modification may lower blood pressure (BP) [4,5] and reduce the incidence of hypertension. Lifestyle modification includes weight lost among overweight/obese individuals [4,5]; increased physical activity [4]; reduced alcohol consumption [4,5]; adoption of a diet rich in fresh fruit, vegetables,[5,6] potassium [5] and a reduced sodium content [4,5,7]. For instance, the DASH diet (Dietary Approaches to Stop Hypertension) has similar BP lowering effects to a single drug therapy [7]. In addition, a DASHtype diet has been shown to be associated with a lower cardiovascular risk in the Nurses Health Study [8]. As is known, nuts are an integral part of the Mediterranean Dietary Pattern (MDP) [9]. MDP has been shown to be protective against age-related changes in blood pressure [10]. Results from the PREDIMED randomized trial showed that a Mediterranean diet enriched with mixed nuts was able to reduce BP levels to a larger extent than a low-fat diet during a 3-month follow-up [11]. Moreover, during a 1- year follow-up of the PREDIMED trial the arm of the trial allocated to the Mediterranean diet plus free provision of nuts exhibited a significant reduction in the prevalence of the metabolic syndrome, with a small reduction in hypertension prevalence after 1 year [12]. However, in that trial most participants initially had hypertension at baseline. Also, recently the Physicians Health Study (PHS) reported a lower risk of hypertension for those who consumed nuts on a daily basis [13]. This is plausible since raw nuts have a high content of magnesium, potassium and unsaturated fatty acids, and only small amounts of sodium. However, with the exception of the PREDIMED trial [11,12] and the PHS cohort [13] there is scarce information about the ability of a nut-rich diet to prevent the incidence of hypertension. The aim of our study was to assess the association between the consumption of tree nuts and the incidence of hypertension in a large prospective cohort of university graduates in Spain. Methods Subjects The SUN project (Seguimiento Universidad de Navarra) is a dynamic cohort comprised exclusively of Spanish university graduates. The recruitment of participants started in December 1999 and it is permanently open. Information is collected using self-administered questionnaires mailed every 2 years. The objectives and methods of this project have been detailed elsewhere [14]. By April 2008, 19,519 participants were recruited and had answered the baseline questionnaire. Those participants who completed the baseline assessment before July 2005 were eligible for longitudinal analyses for either 2-, 4- or 6-year follow-up (n Z 13,740). The overall followup rate was approximately 88%. Participants who had one or more of the following characteristics were excluded from the analysis (n Z 3821): prevalent hypertension (n Z 1495); those who reported extremely low or high values for total energy intake (less than 800 kcal/day in men and 500 kcal/day in women or more than 4000 kcal/day in men and 3500 kcal/day in women) (n Z 1220); pregnant women at baseline (n Z 41); those without information about nut consumption (n Z 274); those who reported at baseline a history of CVD (n Z 360), diabetes mellitus (n Z 114), or cancer (n Z 317). Finally, data from 9919 participants remained available for the analysis. The Human Research Ethical Committee from the University of Navarra approved the study. Voluntary completion of the first questionnaire was considered as informed consent. Dietary assessment The baseline questionnaire included a 136-item semiquantitative food-frequency questionnaire that was previously validated in Spain, it also included open-label questions for information about use of dietary supplements [15]. The questionnaire was adapted from the Willett s questionnaire and it was based on typical portion sizes. Participants selected one of 9 options for the frequency of intake in the previous year for each food item (rarely/never, 1e3/ month, 1/week, 2e4/week, 5e6/week, daily, 2e3/day, 4e6/day, 6þ/day). We asked the participants how often, on average, they had consumed nuts during the previous year. Information on nut consumption was collected in a single aggregated question, which included walnuts, almonds, hazelnuts or peanuts. These varieties represent more than 95% of the total nut consumption in the Spanish population [16]. Due to the small number of subjects in the categories corresponding to a higher frequency of nut consumption, we merged the upper categories to obtain more stable estimates as has been previously reported in similar studies [17,18]. Thus, in all our analyses, we classified nut consumption into four categories: rarely/never, 1e3/month, 1/week, and 2þ/week. The food-frequency questionnaire was also used to obtain information on other nutrients that could act as potential confounders. Nutrient intake scores were computed using an ad hoc computer program. A trained dietitian updated the nutrient data bank using the most upto-date food composition tables for Spain [19,20]. We calculated the amount of each of these nutrients without taking into account the share allocated to nut consumption, and adjusted the estimates for these new variables. The intakes of foods and nutrients (other than nut consumption) were adjusted for total energy intake applying the residual method, and separate regression models were run to obtain the residuals for men and women [21]. Assessment of other covariates The baseline questionnaire included questions about a wide array of characteristics: socio-demographic (e.g. sex, age,

3 Nut consumption and hypertension in Spain 3 educational attainment), anthropometric (e.g. weight, height), lifestyle and health-related habits (e.g. smoking status, physical activity), family history (e.g. diagnosis of hypertension among parents or siblings), obstetric history for women (e.g. pregnancy) and medical history (e.g. prevalence of chronic diseases, medication use, blood cholesterol levels). The section of the questionnaire about physical activity included information about 17 activities such as walking, running, cycling, swimming, judo, soccer, skiing or sailing. To quantify the volume of physical activity during leisure time, an activity metabolic equivalent (MET) index was computed by assigning a multiple of resting metabolic rate (MET score) to each activity [22]. The time spent in each of the activities was multiplied by the MET score specific to each activity, and then summed for the all activities, obtaining a value of overall weekly MET-hours. Our group, using a tri-axial accelerometer as the gold standard, previously validated leisure time physical activity estimated with the questionnaire. Physical activity during leisure time (estimated as MET-h/week) derived from the questionnaire moderately correlated with the objective measurements assessed through the accelerometer (Spearman s rho Z 0.51, p < 0.001) [23]. Body mass index (BMI), defined as weight (kg) divided by height (m) 2, was computed using the self-reported information on weight and height from the baseline questionnaire. The validity of self-reported weight was assessed in a subsample of the cohort (n Z 144). The mean relative error in self-reported weight was 1.5%. The correlation coefficient (r) between measured and self-reported weight was 0.99 (95% CI: 0.98e1.00) [24]. Participants with cardiovascular disease, cancer or diabetes at baseline were excluded from all analyses. Participants were classified as having cardiovascular disease at baseline if they reported at least one of the following conditions: myocardial infarction, stroke, atrial fibrillation, paroxysmal tachycardia, coronary artery bypass grafting or other coronary revascularization procedures, heart failure, aortic aneurism, pulmonary embolism, peripheral arterial disease or peripheral venous thrombosis. Information regarding cancer and diabetes was also collected at baseline. Hypertension ascertainment The baseline questionnaire inquired about the most recent systolic and diastolic blood pressure, use of antihypertensive drugs and whether the participant had received a medical diagnosis of hypertension. A participant was considered to have hypertension at baseline if he/she reported a medical diagnosis of hypertension, was under antihypertensive medication, or reported a systolic blood pressure 140 mmhg and/or a diastolic blood pressure 90 mmhg [1]. New cases of hypertension were defined as those individuals reporting a physician-made diagnosis of hypertension in the follow-up questionnaire and who did not have hypertension at baseline. A validation study showed an adequate validity for the self-reported diagnosis of hypertension in this highly educated cohort [25]. Statistical analysis For each participant we computed person-time of follow-up from the date of return of the baseline questionnaire to the date of return of the last follow-up questionnaire or to the date of the physician-made diagnosis of hypertension, whichever occurred first. Hazard ratios (HRs) for the three upper categories of nut consumption and their 95% CIs were estimated with Cox proportional hazard models, adjusting for potential confounders and taking as the reference category those subjects who consumed nuts never or rarely. The initial model included only age and sex as covariates (model 1). We fitted an additional model adding universally accepted risk factors for hypertension (family history of hypertension [yes, no] and hypercholesterolemia [yes, no]) and variables closely associated with lifestyle and healthrelated habits (smoking [never, past, current, not specified], educational level [degree 1: 3 years of university study, degree 2: 4e5 years of university study, degree 3: 6e9 years of university study], and physical activity [MET hour/week, continuous]) (model 2). To assess the possibility of confounding by other dietary variables, we fitted an additional model adding several dietary factors that have been related to the risk of hypertension in some studies (total energy intake, alcohol intake, the intake of polyunsaturated fatty acids, monounsaturated fatty acids, saturated fatty acids, proteins, carbohydrates, sodium, magnesium, potassium, calcium, folic acid, vitamin E, mineral supplements, caffeine intake, and fiber intake) (model 3). Finally, we fitted a last model adding BMI (body mass index) at baseline, and weight change during followup [lost, gain, equal] (model 4). We also included a separate model adjusting for other food groups instead of macro and micronutrients to avoid the possibility of over-adjustment bias (model 5). To assess whether a multiplicative interaction existed between nut consumption and sex, we introduced a product-term (nut sex) in the fully adjusted model. We also assessed interaction between nut consumption and overweight status (nut BMI), dichotomizing BMI (<25 or 25 kg/m 2 ) and nut consumption (never or else). To assess multiplicative interactions, we introduced interaction terms and then compared both models, the one with and the one without the product-term using the likelihood ratio test. Analyses were performed with SPSS version 15.0 (SPSS Inc, Chicago, IL). All p values are 2-tailed. Statistical significance was set at p < Results We observed five hundred and forty two new cases of hypertension (341 men and 201 women) during a median follow-up of 4.3 years (mean Z 4.5 years). The incidence of hypertension in our population was 12.4 per 1000 person-years. Table 1 shows the non-dietary and dietary characteristics of participants according to their categories of nut consumption. Participants belonging to the highest category of nut consumption were more likely to be men, older, non-smokers, physically more active and reported a higher

4 4 E.H. Martínez-Lapiscina et al. Table 1 Characteristics a of the participants according to categories of nut consumption. Nut consumption Never/rarely 1e3/month 1/week 2þ/week N Z 2089 N Z 4572 N Z 1972 N Z 1286 Sex [n (%)] Female 1463 (70.0) 2859 (62.5) 1121 (56.8) 698 (54.3) Age Smoking status [n (%)] Former smokers 515 (24.7) 1205 (26.4) 504 (25.6) 415 (32.3) Current smokers 528 (25.3) 1093 (23.9) 487 (24.7) 246 (19.1) Physical activity during leisure time (MET s-h/w) Family history of hypertension [n (%)] 800 (38.3) 1710 (37.4) 739 (37.5) 493 (38.3) Hypercholesterolemia [n (%)] 373 (17.9) 918 (20.1) 390 (19.8) 341 (26.5) BMI (kg/m 2 ) Weight change during follow-up [n (%)] No change 351 (16.8) 712 (15.6) 323 (16.4) 241 (18.7) Lost weight 590 (28.2) 1296 (28.3) 479 (24.3) 344 (26.7) Gained weight 1148 (55.0) 2564 (56.1) 1170 (59.3) 701 (54.5) Total energy intake (kcal/day) b Alcohol intake c (g/day) Caffeine intake c (mg/day) Fatty acids intake c (g/day) Saturated fatty acids Monounsaturated fatty acids Polyunsaturated fatty acids Carbohydrates intake c (g/day) Protein intake c (g/day) Fiber intake c (g/day) Sodium intake c (mg/day) Potassium intake c (mg/day) Magnesium intake c (mg/day) Calcium intake c (mg/day) Folic acid intake c (mg/day) Vitamin E intake c (mg/day) Fruit consumption c (g/day) Vegetable consumption c (g/day) Breakfast cereal consumption c (g/day) Whole-fat dairy consumption c (g/day) Low-fat dairy consumption c (g/day) Red and cold meat consumption c (g/day) White meat consumption c (g/day) Fish intake c (g/day) a Mean standard deviation unless otherwise stated. b Macronutrients sum was not equal to total energy intake because the macronutrients were adjusted for total energy intake. The share allocated to nut consumption in the different nutrients was not included. c Nutrient intakes were adjusted for total energy intake using the residuals method proposed by Willett [21]. The share allocated to nut consumption in the different nutrients was not included. frequency of hypercholesterolemia. Among participants in the lowest category of nut consumption, we identified more women and a higher proportion of current smokers. Participants belonging to the highest category of nut consumption had a higher intake of total energy, alcohol, monounsaturated fatty acids, polyunsaturated fatty acids, fiber, magnesium, potassium, folic acid, and vitamin E but lower intakes of sodium, protein, carbohydrate, saturated fats, and caffeine. Nut consumption was positively related to fruit, vegetable and fish intakes and inversely related to dairy and meat intake. Table 2 shows the association between nut consumption and the incidence of hypertension. We found no association between nut consumption and incidence of hypertension after adjusting for sex, age and other dietary and nondietary potential confounders.

5 Nut consumption and hypertension in Spain 5 Table 2 Hazard ratios (HRs) and 95% CIs of hypertension according to nut consumption. Nut consumption Never/rarely 1e3/Month 1/Week 2þ/Week P for trend New cases of hypertension Person-years Rate (cases per person-years) Model 1 a 1 (ref.) 0.98 (0.71e1.35) 1.17 (0.83e1.66) 0.84 (0.57e1.24) Model 2 b 1 (ref.) 0.98 (0.71e1.35) 1.19 (0.84e1.68) 0.87 (0.59e1.29) Model 3 c 1 (ref.) 0.98 (0.71e1.36) 1.16 (0.81e1.67) 0.73 (0.44e1.21) Model 4 d 1 (ref.) 0.99 (0.72e1.38) 1.17 (0.81e1.68) 0.77 (0.46e1.30) Model 5 e 1 (ref.) 1.08 (0.85e1.38) 1.23 (0.93e1.63) 0.91 (0.66e1.26) a Cox regression model adjusted for age and sex. b Model 1 plus additional adjustment for smoking, education, hypercholesterolemia, family history of hypertension and physical activity. c Model 2 plus additional adjustment for total energy intake, alcohol intake, polyunsaturated fatty acids, monounsaturated fatty acids, saturated fatty acids, protein intake, carbohydrates intake, magnesium intake, sodium intake, potassium intake, calcium intake, folic acid intake, vitamin E intake, mineral supplements (calcium, magnesium, potassium, folic acid, vitamin E), caffeine intake, and fiber intake. The share allocated to nut consumption in the different nutrients was not included. d Model 3 plus additional adjustment for BMI baseline and weight change during follow-up. e Model 2 plus additional adjustment for total energy intake, alcohol intake, caffeine intake, fruit and vegetable consumption, wholefat dairy and low-fat dairy consumption, breakfast cereal consumption, red, cold and white meat consumption, fish intake, BMI baseline and weight change during follow-up. No statistically significant interaction was found either for sex (p for interaction Z 0.82) or BMI (p for interaction Z 0.75). Discussion In this prospective cohort study, we found no evidence for an inverse association between nut consumption and incident hypertension. These results are in agreement with a randomized clinical trial that evaluated the effects of nut consumption on selected markers of the metabolic syndrome, and found that mean blood-pressure values only slightly decreased, but the observed changes were not statistically significant [26]. On the other hand, in the PREDIMED randomized trial, a reduction in mean BP was observed in both interventional groups using either a nut-enriched or an olive oilenriched Mediterranean diet [11]. In the group allocated to a Mediterranean diet plus free provision of tree nuts, the PREDIMED trial participants received 30 g/d of raw tree nuts (15 g/d of walnuts, 7.5 g/d of hazelnuts and 7.5 g/d of almonds). In addition to its experimental design (different from the observational nature of the present cohort), the PREDIMED trial differs methodologically from the SUN cohort. First of all, subjects in the PREDIMED trial consumed more nuts than those belonging to the highest category of nut consumption in the SUN cohort; in addition, the PREDIMED trial only included participants at high cardiovascular risk and many participants were taking antihypertensive drug treatment. Any potential synergism between lifestyle factors and pharmacological agents for management of hypertension would have contributed to a stronger effect [27]. The PREDIMED trial did not evaluate nut consumption separately from the overall intervention with a Mediterranean-type dietary pattern. Moreover, the tree nuts provided to the corresponding intervention group in the PREDIMED trial were raw nuts, i.e. non-salted nuts. However, nuts are usually consumed with salt in Spain. In fact, according to Datamonitor, nuts and seeds represent the first source (38.5%) of the savory snacks market in Spain [28]. Nuts in Spain are seldom consumed as raw products, except for walnuts and pine nuts. They are more frequently consumed toasted, grilled, fried, or salted [16]. Salted nuts can contain important amounts of sodium. The effect of sodium to increase BP is well known.[7] The Physicians Health Study (PHS) has recently shown that nut consumption is associated with a lower risk of hypertension [13]. The protective effect on hypertension (adjusted hazard ratio Z 0.82 [0.71e0.94]) reported in that study is similar in magnitude to the point estimates we have found. However, the statistical power of the PHS with more than 8000 new cases of hypertension is considerably higher, rendering narrower confidence intervals than our estimates. However, the protective effect reported in the PHS was restricted only to lean participants, limiting the generalisability of their findings. A potential protection against hypertension for raw nuts is plausible because they contain a high proportion of mono and polyunsaturated fatty acids, but are low in saturated fatty acids. They are also good sources of fiber, protein (arginine), and minerals (magnesium, potassium, and calcium) and all these nutrients are potentially beneficial against hypertension [16]. Our results demonstrated that participants belonging to the highest category of nut consumption had a higher intake of monounsaturated fatty acids, polyunsaturated fatty acids, fiber, magnesium, potassium, folic acid and

6 6 E.H. Martínez-Lapiscina et al. vitamin E, but a lower intake of carbohydrate and saturated fatty acids. In Spain the varieties of nuts most widely consumed are peanuts, walnuts, hazelnuts and almonds [16]. The last two are rich in oleic acid that offers a lowering effect on BP as has already been reported [11]. Walnuts are rich in a- linolenic acid, and its anti-inflammatory properties are being evaluated. This feature could prevent the progression of atherosclerosis, which is an important risk factor for hypertension [29]. Fiber intake could protect against hypertension through weight reduction due to the influence on satiety and energy intake [30]. Potassium [5] also offers a lowering effect on BP. Other explanations make it also likely that a high nut consumption might be able to reduce BP, but this BP lowering effect may be insufficient to prevent the development of hypertension in our study. This fact could be occurring because: (1) nut consumption might have been underestimated since we did not compute the amount included in sauces, cheeses, cereals, and desserts[16]; (2) information about nut consumption was collected only in the baseline questionnaire and we were not able to assess changes in nut consumption during follow-up; (3) we were not able to isolate the role of each specific type of nut; so if only some kind of nut were protective against the development of hypertension, the aggregation of consumption data in a single question would very probably underestimate its specific effect; (4) we could not examine the influence of the preparation method (salted, roasted or raw nuts) on the risk of hypertension; (5) the case ascertainment for hypertension was based on self-reported data and it is well known that about one third of hypertensive individuals are not aware of their status [1]; however, we have shown an adequate validity of this self-report of a physician-made diagnosis of hypertension in this highly educated cohort [25]; (6) and, especially, because the incidence of hypertension was relatively low in our cohort and this scarcity in the number of observed new cases could hinder the finding of a statistically significant association between nut consumption and hypertension. We acknowledge that our statistical power might be too low to detect an association of small magnitude. We observed 542 cases of hypertension over 9919 subjects. Assuming an alfa error (bilateral) equal 0.05, a power of 0.80 and a relative risk of 0.75 for a dichotomized exposure, we would need 3947 exposed and 3947 unexposed subjects to detect a protective effect at that level. The strengths of this study are the large sample size, long duration of follow-up (4.3 years median), its prospective design that could avoid inverse causation bias, and the control for a wide variety of potential confounders. Additionally, both dietary exposure and outcome were ascertained through standardized and validated questionnaires [15,25]. In conclusion, though our estimated confidence intervals are compatible with some degree of protection of nut consumption against hypertension, our data do not support an inverse association between nut consumption and incidence of hypertension in this cohort of Spanish university graduates. Larger cohorts and trials including participants initially free of hypertension and with greater statistical power and ability to differentiate between raw nuts and salted nuts would contribute to clarifying this association. Acknowledgements We thank all members of the SUN Study Group for administrative, technical and material support. We thank participants of the SUN Study for their continued cooperation and participation. Adriano Marçal Pimenta would also like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior e CAPES, Brazil for their scholarship for studies at the University of Navarra, Spain. References [1] Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo Jr JL, et al. The seventh report of the Joint National Committee on detection, evaluation, and treatment of high blood pressure: the JNC 7 report. JAMA 2003;289:2560e72. [2] Lawes CM, Vander Hoorn S, Rodgers A. Global burden of bloodpressure-related disease, Lancet 2008;371:1513e8. [3] Basterra-Gortari FJ, Bes-Rastrollo M, Seguí-Gómez M, Forga L, Martínez JA, Martínez-González MA. Trends in obesity, diabetes mellitus, hypertension and hypercholesterolemia in Spain (1997e2003). Med Clin (Barc) 2007;129:405e8. [4] Dickinson HO, Mason JM, Nicolson DJ, Campbell F, Beyer FR, Cook JV, et al. Lifestyle interventions to reduce raised blood pressure: a systematic review of randomized controlled trials. J Hypertens 2006;24:215e33. [5] Appel LJ, Brands MW, Daniels SR, Karanja N, Elmer PJ, Sacks FM. Dietary approaches to prevent and treat hypertension: a scientific statement from the American Heart Association. Hypertension 2006;47:296e308. [6] Alonso A, de la Fuente C, Martín-Arnau AM, de Irala J, Martínez JA, Martínez-González MA. Fruit and vegetable consumption is inversely associated with blood pressure in a Mediterranean population: the Seguimiento Universidad de Navarra (SUN) study. Br J Nutr 2004;92:311e9. [7] Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, et al. Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet. DASH-Sodium Collaborative Research Group. N Engl J Med 2001;344:3e10. [8] Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women. Arch Intern Med 2008;168:713e20. [9] Willett WC, Sacks F, Trichopoulou A, Drescher G, Ferro-Luzzi A, Helsing E, et al. Mediterranean diet pyramid: a cultural model for healthy eating. Am J Clin Nutr 1995;61(6 Suppl.):1402Se6S. [10] Nuñez-Córdoba JM, Valencia-Serrano F, Toledo E, Alonso A, Martínez-González MA. The Mediterranean diet and incidence of hypertension: the Seguimiento Universidad de Navarra (SUN) Study. Am J Epidemiol 2009;169:339e46. [11] Estruch R, Martínez-González MA, Corella D, Salas-Salvadó J, Ruiz-Gutiérrez V, Covas MI, et al. Effects of a Mediterraneanstyle diet on cardiovascular risk factors: a randomized trial. Ann Intern Med 2006;145:1e11. [12] Salas-Salvadó J, Fernández-Ballart J, Ros E, Martínez- González E, Fitó M, Estruch R, et al. Effect of a Mediterranean diet supplemented with nuts on metabolic syndrome status: one-year results of the PREDIMED randomized trial. Arch Intern Med 2008;168:2449e58. [13] Djoussé L, Rudich T, Gaziano JM. Nut consumption and risk of hypertension in US male physicians. Clin Nutr 2008;xx:1e6. doi: /j.clnu [14] Seguí-Gómez M, de la Fuente C, Vázquez Z, de Irala J, Martínez- González MA. Cohort profile: the Seguimiento Universidad de Navarra (SUN) study. Int J Epidemiol 2006;35:1417e22.

7 Nut consumption and hypertension in Spain 7 [15] Martín-Moreno JM, Boyle P, Gorgojo L, Maisonneuve P, Fernandez-Rodriguez JC, Salvini S, et al. Development and validation of a food frequency questionnaire in Spain. Int J Epidemiol 1993;22:512e9. [16] Aranceta J, Pérez Rodrigo C, Naska A, Vadillo VR, Trichopoulou A. Nut consumption in Spain and other countries. Br J Nutr 2006;96(Suppl. 2):S3e11. [17] Albert CM, Gaziano JM, Willett WC, Manson JE. Nut consumption and decreased risk of sudden cardiac death in the Physicians Health Study. Arch Intern Med 2002;162: 1382e7. [18] Djoussé L, Rudich T, Gaziano JM. Nut consumption and risk of heart failure in the Physicians Health Study I. Am J Clin Nutr 2008;88:930e3. [19] Mataix J. Tabla de Composición de Alimentos. [Food composition tables]. 4th ed. Granada: Universidad de Granada; [20] Moreiras O. Tablas de Composición de Alimentos. [Food composition tables]. 9th ed. Madrid, Spain: Ediciones Pirámide; [21] Willet WC. Nutritional epidemiology. 2nd ed. New York: Oxford University Press; [22] Ainsworth BE, Haskell WL, Whitt MC, Irwin ML, Swartz AM, Strath SJ, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc 2000;32(Suppl. 9):S498e504. [23] Martínez-González MA, López-Fontana C, Varo JJ, Sánchez- Villegas A, Martínez JA. Validation of the Spanish version of the physical activity questionnaire used in the Nurses Health Study and Health Professionals Follow-up Study. Public Health Nutr 2005;8:920e7. [24] Bes-Rastrollo M, Pérez JR, Sánchez-Villegas A, Alonso A, Martínez-González MA. Validación del peso e índice de masa corporal auto-declarados de los participantes de una cohorte de graduados universitarios. Rev Esp Obes 2005;3:352e8. [25] Alonso A, Beunza JJ, Delgado-Rodríguez M, Martínez- González MA. Validation of self reported diagnosis of hypertension in a cohort of university graduates in Spain. BMC Public Health 2005;5:94. [26] Mukuddem-Petersen J, Stonehouse Oosthuizen W, Jerling JC, Hanekom SM, White Z. Effects of a high walnut and high cashew nut diet on selected markers of the metabolic syndrome: a controlled feeding trial. Br J Nutr 2007;97: 1144e53. [27] Ferrara LA, Raimondi AS, d Episcopo L, Guida L, Dello Russo A, Marotta T. Olive oil and reduced need for antihypertensive medications. Arch Intern Med 2000;160:837e42. [28] DATAMONITOR savory snacks in Spain: industry profile. London: DATAMONITOR; [29] Jiang R, Jacobs Jr DR, Mayer-Davis E, Szklo M, Herrington D, Jenny NS, et al. Nut and seed consumption and inflammatory markers in the multi-ethnic study of atherosclerosis. Am J Epidemiol 2006;163:222e31. [30] Alonso A, Beunza JJ, Bes-Rastrollo M, Pajares RM, Martínez- González MA. Vegetable protein and fiber from cereal are inversely associated with the risk of hypertension in a Spanish cohort. Arch Med Res 2006;37:778e86. doi: /j.arcmed