Adherence to the mediterranean diet and risk of metabolic syndrome and its components

Nutrition, Metabolism & Cardiovascular Diseases (2009) 19, 563e570 Adherence to the Mediterranean diet and riskof metabolic syndrome and its components ´ ,, on behalf of the Nureta-PREDIMED investigators a Human Nutrition Unit, Department of Biochemistry and Biotechnology, Faculty of Medicine and Health Sciences,Rovira i Virgili University, Spainb CIBER Fisiopatologı´a de la Obesidad y Nutricio´n CB06/03. Instituto de Salud Carlos III, Madrid, Spainc Direccio´ d’Atencio´ Prima`ria Tarragona-Reus. Institut Catala` de la Salut, Reus, Spaind Department of Preventive Medicine and Public Health, Medical School-Clı´nica Universitaria,University of Navarra, Pamplona, Spaine Preventive Medicine and Public Health Unit, Faculty of Medicine and Health Sciences,Rovira i Virgili University, Spain Received 28 July 2008; received in revised form 24 October 2008; accepted 29 October 2008 Background and aims: The role of diet in the aetiology of metabolic syndrome (MetS) is not well understood. The aim of the present study was to evaluate the relationship between adherence to the Mediterranean diet (MedDiet) and MetS.
Methods and results: A cross-sectional study was conducted with 808 high cardiovascular risk participants of the Reus PREDIMED Centre. MetS was defined by the updated National Choles- terol and Education Program Adult Treatment Panel III criteria.
An inverse association between quartiles of adherence to the MedDiet (14-point score) and the prevalence of MetS (P for trend < 0.001) was observed. After adjusting for age, sex, totalenergy intake, smoking status and physical activity, participants with the highest score ofadherence to the MedDiet (!9 points) had the lowest odds ratio of having MetS (OR [95% CI]of 0.44 [0.27e0.70]) compared to those in the lowest quartile.
Participants with the highest MedDiet adherence had 47 and 54% lower odds of having low HDL-c and hypertriglyceridemia MetS criteria, respectively, than those in the lowest quartile. Some componentsof the MedDiet, such as olive oil, legumes and red wine were associated with lower prevalence of MetS.
Conclusion: Higher adherence to a Mediterranean diet is associated with a significantly lower oddsratio of having MetS in a population with a high risk of cardiovascular disease.
ª 2008 Elsevier B.V. All rights reserved.
* Corresponding author. Human Nutrition Unit, Department of Biochemistry and Biotechnology, Faculty of Medicine and Health Sciences, Rovira i Virgili University, C/Sant Llorenc 0939-4753/$ - see front matter ª 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.numecd.2008.10.007 The MetS is a cluster of common cardiovascular risk factors, including central obesity, hyperglycaemia, low HDL-cholesterol A cross-sectional analysis on baseline data from the PRE- triglyceridemia. The association of MetS with the risk of DIMED trial centre in Reus was conducted. The PREDIMED is developing diabetes and cardiovascular disease (CVD) is a 5-year clinical trial and constitutes a randomized, well documented, implying a greater risk of mortality .
parallel-group multicentre that aims to assess the effects of In developed countries the MetS seems to affect around the MedDiet on the primary prevention of CVD .
25% of the population , and there is evidence that MetS Participants were men and women, without prior CVD and has become more prevalent over the last decade who fulfilled at least one of the following two criteria: (1) probably influenced in part by increases in obesity.
type-2 diabetes or (2) !3 major cardiovascular risk factors The aetiology of MetS, although largely unknown, is (current smoking, hypertension [blood pressure ! 140/ considered to reside in a complex interaction between 90 mmHg or treatment with antihypertensive drugs], LDL- genetic, metabolic, and environmental factors .
cholesterol level ! 160 mg/dL [or treatment with hypolipi- Weight loss or gain can be considered to be one of the main demic drugs], HDL-cholesterol level 40 mg/dL, body mass environmental determinants of the MetS . Physical index [BMI] ! 25 kg/m2, or a family history of premature activity and diet have also been reported to be important CVD). Other details have been reported elsewhere Of the first 847 participants recruited in the Reus node, Adherence to a healthy dietary pattern has been shown 39 were excluded because some of the MetS variables were to be inversely associated with MetS some of its missing. Our centre’s institutional review boards approved components and type-2 diabetes[13]. In contrast, the study protocol, and all the participants signed an prospective finding have suggested that a Western dietary pattern is directly associated with the risk of MetS .
Clinical trials have also shown that behavioural programs can be used to implement healthy dietary patterns anddecrease the prevalence of MetS .
The baseline examination included assessment of cardio- In recent decades the Mediterranean Diet (MedDiet) vascular risk factors, medical conditions and medication has emerged as a healthy dietary pattern that protects use. Height and weight were measured with light clothing against CVD and other chronic conditions . The and no shoes. Waist circumference was measured midway MedDiet is characterized by a high consumption of between the lowest rib and the iliac crest. Blood pressure legumes, fruit and vegetables, grains and olive oil, was measured (using a validated oscillometer [Omron HEM- a moderate consumption of wine and dairy products and 705CP, Hoofddorp, Netherlands]) in triplicate with a 5-min a low consumption of red and processed meat, cream and interval between each measurement, and the mean of these values was recorded. The validated Spanish version of Few studies have analysed the relationship between the Minnesota Leisure-Time Physical Activity Questionnaire MedDiet adherence and the prevalence or incidence of was used to estimate energy expenditure.
MetS and its components . With the exception ofthe cross-sectional study of Alvarez-Leo who did not find any relationship between the adher- ence to the MedDiet and MetS prevalence, other studies,conducted in samples of healthy participants do support A previously validated 136-item food frequency question- the beneficial effect of the MedDiet pattern on MetS naire was used to appraise dietary habits. Spanish food . Recent data from the SUN prospective study composition tables were used to derive nutrient intake.
also show an inverse relationship between MedDiet adher- A 14-point food-items questionnaire, an extension of a previ- ence and the incidence of MetS . Finally, Esposito et al.
ously validated questionnaire , was used to assess adherence by means of a feeding trial showed a decrease in the to the traditional MedDiet (see Appendix 1).
prevalence of MetS in patients losing weight by adhering to Previously applied scores were also used to assess adherence to the MedDiet in secondary analyses. In To the best of our knowledge, the association between the Trichopoulou score, subjects whose consumption MedDiet or its components and MetS has never been was at or above the sex-specific median (monounsaturated- explored in an elderly population with a high risk of CVD.
to-saturated fat ratio, legumes, cereals, vegetables, fruits Likewise, no studies have analysed which of the compo- and fish) or below the median (meat and dairy products) nents of the MedDiet are most associated with MetS.
were assigned 1 point. Moderate alcohol consumption was So, the aim of the present study was to assess the also assigned 1 point. Therefore, the total Trichopoulou relationship between adherence to the MedDiet and the score had a potential range from 0 to 9 points.
prevalence of the MetS in participants of the PREDIMEDstudy, and to identify specific components of the MedDiet associated with this syndrome that might support thebeneficial effects attributed to this overall healthy dietary Centralized laboratory analyses were performed on frozen serum samples obtained in fasting conditions. Serum Mediterranean diet and metabolic syndrome glucose, cholesterol, and triglyceride levels were measured abnormal waist circumference and low level of HDL- using standard enzymatic automated methods. HDL- cholesterol features were significantly higher in women. No cholesterol was measured by enzymatic procedure after significant differences between women and men were observed in relation to the MedDiet scores.
Association between Mediterranean diet andmetabolic syndrome The updated criteria from the National Cholesterol Educa-tion Program’s Adult Treatment Panel III were used to Multiple logistic regression models with polynomial contrast define MetS. That is to say, subjects had to have !3 of the showed an inverse association between adherence following: (1) waist circumference > 102 cm in men and to the MedDiet (14-point food-item score) and the preva- >88 cm in women; (2) serum triglyceride ! 150 mg/dL; (3) lence of the MetS (P for trend < 0.001). This association HDL-cholesterol < 40 mg/dL in men and <50 mg/dL in persisted after adjusting for age, sex, energy intake, women; (4) blood pressure ! 130/85 mmHg; and (5) fasting smoking status and physical activity, and even after plasma glucose level ! 100 mg/dL. Participants who were adjusting for BMI (P for trend Z 0.002). The category with being treated with antidiabetic, antihypertensive or the highest score of adherence to the MedDiet exhibited triglyceride-lowering medications were considered as dia- the lowest odds ratio of having MetS, (OR (95% CI) 0.44 betic, hypertensive or hypertriglyceridemic, respectively.
(0.27e0.70) for subjects on quartile 4 compared to those inquartile 1). No interaction between gender or age category(<65 and 65e80 years old) and the MedDiet score was observed (P Z 0.537 for gender and P Z 0.321 for agecategory).
We built categories (approximately quartiles) of the Significant inverse associations between adherence to 14-point food-item MedDiet or Trichopoulou score. A MedDiet and MetS were observed only in men, even when multiple logistic regression model was used to evaluate the adjusted for all potential confounders. This association was odds ratio for each of the three upper quartiles of adher- ence to the MedDiet (compared with the lowest quartile) trend Z 0.056). Men at the top quartile of adherence to the and the prevalence of MetS or its components. Multiple MedDiet had a statistically significant lower odds ratio of logistic regression models with polynomial contrast were having MetS in comparison with the lowest quartile. Higher used to generate the P for trend. After the univariate adherence to the MedDiet in men between 65 and 80 years model, we built first a sex- and age-adjusted model (model old was significantly associated to a lower prevalence of 1) and another model additionally adjusted for energy intake, smoking and physical activity (model 2). Further An inverse association between adherence to the Tri- adjustments were made for BMI (model 3). We conducted chopoulou score and MetS was observed in the unadjusted the analyses first for the whole sample and then separately model (Appendix 2). However, when we adjusted for potential confounders the models were not statistically In addition, a multiple logistic regression model was significant. No significant association between MedDiet fitted to evaluate which of the 14-point food items in the score and MetS prevalence was observed when the Pan- MedDiet were most strongly associated with the preva- agiotakos MedDiet score was used (Appendix 3).
lence of the MetS, having previously adjusted for the shows the results of the multiple logistic regres- above-mentioned potential confounders and mutually sion analysis evaluating which of the 14 elements used to adjusted for the rest of the items of the score. Also, define the MedDiet score were more strongly associated with a multiple logistic regression model was fitted to evaluate the prevalence of MetS. Even when mutually adjusted, no which of the 14-point food MedDiet items were most item in the score was significantly associated to MetS. When strongly associated with the risk of having MetS, forcing we used the forward method, the items ‘olive oil as culinary the adjustment for non-dietary variables (sex, age, energy fat’, ‘legumes consumption’ and ‘wine consumption’ were intake, smoking [current smoker or not] and physical selected in this order into an overall model as the highest activity), but using the stepwise algorithm (forward odds ratio of having MetS. There was no evidence of multi- method) to allow the additional inclusion of each of the colinearity when the items were simultaneously adjusted for 14 items included in the MedDiet adherence score each other. No colinearity between the food item included in the score was observed (VIF < 2). In separated models The level of significance for all statistical tests was stratified by sex, we found that ‘legumes consumption’ and P < 0.05 for bilateral contrasts.
‘red meat consumption’, were significantly associated withthe MetS risk only among men.
Mediterranean diet adherence and metabolic MetS was significantly more prevalent in women than in The associations between the adherence to the MedDiet men The prevalence of each ATP III component of and each of the MetS criteria were considered separately the MetS is summarized in The prevalence of ) using multiple logistic regression models with 4) was associated with a lower prevalence of hyper- General characteristics of the subjects studied.
triglyceridemia criteria (P for trend Z 0.011).
When we stratified by sex, marginally significant differ- ences across successively increasing quartiles of the score were found for the high triglyceride criterion both among Z 0.055) and men (P for trend Z 0.059).
This was also observed in men for the central obesity criteria Although there is some evidence that the Mediterranean dietary pattern protects against CVD few epidemio- logic studies have assessed the relationship between The present study is unique in assessing this relationship in an elderly population with a high risk of CVD and also a high baseline level of MetS risk factors (including diet).
Our main finding is that the highest adherence to MedDiet is associated with a lower prevalence of MetS defined by the Metabolic syndromeand its components, % (n) ATP III criteria. It is important to note that this association remained significant when the MetS criteria given by the International Diabetes Federation (IDF) were used (data not shown) and adjustments had been made for BMI.
To the best of our knowledge, only one longitudinal study , two cross-sectional studies and one clinical trial[15] have been conducted to assess the asso- ciation between MedDiet and MetS. The longitudinal study, including only university graduates at low cardiovascular risk, reported an inverse association with the MetS (defined by the IDF criteria) but only weak associations with most of its defining criteria The two cross-sectional studies found contradictory results. In both studies, in contrast to our high risk participants, this association was analysed inpopulation samples with low cardiovascular risk. In a sample of the general Greek population, individuals with higher adherence to a MedDiet had a 20% lower odds ratio of having MetS In contrast, in a cross-sectional study performed on subjects from the general population of the Canary Islands no significant relationship between MedDiet In a Spanish cohort , after 6 years follow-up, participants with the highest Trichopoulou score at base- line presented the lowest cumulative incidence of MetS. In our study we observed a better relationship between MedDiet adherence and MetS when we used the 14-point food-item score compared with the Trichopoulou score.
For this reason, the 14-item MetDiet score should be evaluated in the future as a possible tool for analysing the a The metabolic syndrome was defined by using the updated relationship between MedDiet adherence and chronic National Cholesterol Education program’s Adult Treatment diseases. The advantage of the 14-item tool is that it is assessed using only a short questionnaire and, therefore,the application of a full-length food frequency question-naire may not be needed to assess adherence to the polynomial contrast and all the MetS components as MedDiet. However, no significant relationship between dependent variables. A higher adherence to the MedDiet MedDiet adherence and MetS prevalence was observed (quartile 4) was associated with a lower prevalence of the when the Panagiotakos MedDiet score was used The low HDL-cholesterol and high triglycerides criteria (47.0 cross-sectional design of our study and the space limits of and 54%; P for trend Z 0.034 and 0.006 respectively) in the manuscript precluded any analysis of specificity and relation to the first quartile. After adjusting for BMI (data sensitivity estimates such as that done by Panagiotakos in not shown), only higher adherence to the MedDiet (quartile Mediterranean diet and metabolic syndrome Metabolic syndrome risk (odds ratio and their 95% confidence intervals) across quartiles of adherence to the Medi- Multiple logistic regression was used to assess the association between MedDiet quartiles (dependent variables) and MetS (dependentvariable), and multiple logistic regression with polynominal contrast was used to generate the P for trend.
a Quartiles of Mediterranean diet score (0e14).
b Adjusted for sex and age.
c Adjusted for age.
d Adjusted for sex, age, energy intake, smoking and physical activity.
e Adjusted for age, energy intake, smoking and physical activity.
f Adjusted for sex, age, energy intake, smoking, physical activity and body mass index.
g Adjusted for age, energy intake, smoking, physical activity and body mass index.
Only one clinical trial has analysed the effect of the pressure has been reported . A clinical trial did not traditional MedDiet on the MetS. Esposito et al. showed show any blood pressure effect after consumption of red wine that an intervention using a Mediterranean-style diet seems or de-alcoholized red wine. Moderate alcohol consumption effective in reducing the prevalence of the MetS.
raises HDL-cholesterol levels and decreases plasma Although Alvarez-Leon et al. found no relationship concentrations of some inflammatory molecules. A meta- between adherence to the MedDiet and lower MetS preva- analysis indicated that the consumption of 30 g/day of alcohol lence, they did observe that some components of the tradi- increased HDL-cholesterol levels by a mean of 4 mg/dL, irre- tional MedDiet (wine, fruit, vegetables and cereals) showed spective of the alcoholic beverage consumed an inverse association with MetS and its components. In fact, High legume and low red meat consumption were it has been suggested that not all the components of the inversely associated in our study with MetS, but only among MedDiet are likely to provide the same level of protection men. Red meat is rich in saturated fat, which increases In this sense, in our study, the use of olive oil as culinary inflammation, LDL and TG-levels. Individuals adopting fat, the high consumption of legumes and the moderate a Western dietary pattern (characterized mainly by high red consumption of wine were the principal components of the or processed meat consumption) have an increased risk of MedDiet associated with lower risk of MetS.
developing diabetes , whereas a MedDiet, low in meat The inverse relationship observed between those typical or meat products but rich in legumes was associated with foods present in the MedDiet and MetS prevalence could be a substantially lower diabetes risk Red meat intake explained by many of the constituents of these foods. For was also positively associated with the risk of hypertension example olive oil is very rich in oxidation-resistant MUFA in middle-aged and older men Furthermore, dietary especially if the olive oil consumed is as the virgin pattern analyses have shown that a dietary pattern char- variety retaining all its polyphenol content Replacing acterized by high red meat consumption is associated with SFA by MUFA beneficially affects the plasma lipid profile and improves insulin sensitivity . Olive oil consumption Legumes are rich in fibre, magnesium and other was reported to be associated with lower risk of developing components that are associated with a better lipid profile, hypertension, but only among men Virgin olive oil and with improved glucose and inflammatory responses or a diet rich in MUFA has been shown to reduce blood . Subjects consuming legumes four or more times per pressure more than an SFA-rich diet and, unlike a low- week registered a lower risk of CVD Whole grain fat, carbohydrate-rich diet, has no deleterious effect on products and legumes protect against the development of blood pressure in normotensive subjects .
diabetes . The viscous fibre content of legumes tends to Some polyphenols present in red wine have beneficial slow down carbohydrate digestion, decreasing the glycae- endothelial and anti-inflammatory properties. An inverse mic index of the diet, thus contributing to an increase in association between red wine consumption and blood the HDL-c levels and insulin sensitivity .
Metabolic syndrome risk (odds ratio and their 95% confidence intervals) in all subjects and Mediterranean diet food Consumption ! 2 servings/day of vegetables Consumption ! 3 fruit units/day (including natural fruit juices) Consumption < 1 serving/day of red meat, hamburger or meat products Consumption < 1 serving/day of butter, margarine or cream Consumption < 1 serving/day of sweetened or carbonated beverages Consumption ! 3 servings/week of fish or shellfish Consumption < 3 times/week of commercial sweets or pastries (not Consumption ! 1 servings/week of nuts (including peanuts) Consumption preferentially of chicken, turkey, or rabbit meat instead of Consumption ! 2 servings/week of vegetables, pasta, rice or other dishes seasoned with sofrito (sauce made with tomato and onion, leek, or garlicand simmered with olive oil) Multiple logistic regression models (enter method or forward method) including the presence or not of metabolic syndrome (dependentvariable) and the diet food items (independent variables), adjusted for sex, age, energy intake, smoking status and physical activity.
a Specific questions about these items are shown in Supplementary Appendix 1.
Metabolic syndrome components risk (odds ratio 95% confidence intervals) by quartiles of adherence to Mediterra- nean diet in women (n Z 446) and men (n Z 362).
Waist circumference (>102 cm in men, >88 cm in women)All subjects HDL-cholesterol (<40 mg/dL in men, <50 mg/dL in women)All subjects Triglycerides (!150 mg/dL or hypotriglyceridemic treatment)All subjects Glucose (>100 mg/dL or antidiabetic medication)All subjects Blood pressure (systolic !130 mmHg, diastolic !85 or antihypertensive medication)All subjects The multivariable logistic regression was adjusted for age, energy intake, smoking and physical activity. Multiple logistic regression wasused to assess the association between MedDiet quartiles (dependent variables) and each of the MetS components (dependent variable).
Multiple logistic regression with polynominal contrast was used to generate the P for trend.
a Quartiles of Mediterranean diet score (0e14).
Mediterranean diet and metabolic syndrome Finally, consistent with previous evidence regarding outcomes other than MetS , a beneficial association of theoverall MedDiet pattern contrasts with the lack of evidence of [1] Gami AS, Witt BJ, Howard DE, Erwin PJ, Gami LA, Somers VK, an important association of each of its individual components.
et al. Metabolic syndrome and risk of incident cardiovascular We highlight that the associations we found for the individual events and death: a systematic review and meta-analysis of components of the MedDiet with MetS were mostly non- longitudinal studies. J Am Coll Cardiol 2007;49:403e14. doi: significant. This may be because individual components are associated only when they are integrated into an overall score.
The main limitation of the present analysis is its cross- Kumpusalo E, Tuomilehto J. The metabolic syndrome and totaland cardiovascular disease mortality in middle-aged men.
sectional nature, which does not allow any causal rela- tionship between MedDiet and MetS to be established.
[3] Athyros VG, Ganotakis ES, Elisaf M, Mikhailidis DP. The prev- Moreover, since the sample was composed of older partici- alence of the metabolic syndrome using the National Choles- pants from Spain with a high risk of CVD, our findings cannot be extrapolated to younger lower risk populations from Federation definitions. Curr Med Res Opin 2005;21:1157e9.
other Mediterranean countries. Furthermore, studying high [4] Ford ES, Giles WH, Mokdad AH. Increasing prevalence of the cardiovascular risk individuals is a limitation rather than an metabolic syndrome among US adults. Diabetes Care 2004;27: advantage for testing our hypothesis. These individuals may have additional confounders because of their dietary [5] Park YW, Zhu S, Palaniappan L, Heshka S, Carnethon MR, recommendations or medication, which probably makes it Heymsfield SB. The metabolic syndrome: prevalence andassociated risk factor findings in the US population from the more difficult to detect associations.
Third National Health and Nutrition Examination Survey, In summary, a higher adherence to a MedDiet was found to 1988e1994. Arch Intern Med 2003;163:427e36.
be associated with significantly lower odds of having MetS in [6] Mirmiran P, Noori N, Azizi F. A prospective study of determi- a population with a high risk of CVD. Subjects in the fourth nants of the metabolic syndrome in adults. Nutr Metab Car- quartile of the MedDiet adherence have a 56, 47, and 54% lower odds ratios of having MetS, low HDL-c and hyper- [7] Branth S, Ronquist G, Stridsberg M, Hambraeus L, Kindgren E, triglyceridemia MetS criteria respectively than those in the Olsson R, et al. Development of abdominal fat and incipient lowest quartile. Some components of the MedDiet, such as metabolic syndrome in young healthy men exposed to long- olive oil, legumes and red wine were associated with lower term stress. Nutr Metab Cardiovasc Dis 2007;7:427e35.
odds of MetS. Further longitudinal study designs, including [8] Phelan S, Wadden TA, Berkowitz RI, Sarwer DB, Womble LG, Cato RK, et al. Impact of weight loss on the metabolic trials, are needed in order to clarify these relationships.
syndrome. Int J Obes (Lond) 2007;31:1442e8.
Sociodemographic risk factors associated with metabolicsyndrome in a Mediterranean population. Public Health Nutr2008;11:1372 The authors thank the participants for their enthusiastic [10] Williams DE, Prevost AT, Whichelow MJ, Cox BD, Day NE, collaboration, the PREDIMED personnel for excellent assis- Wareham NJ. A cross-sectional study of dietary patterns with tance and the personnel of all primary care centres affili- glucose intolerance and other features of the metabolic URV) for the administrative and technical support provided.
[11] Esmaillzadeh A, Kimiagar M, Mehrabi Y, Azadbakht L, Hu FB, We gratefully acknowledge the Spanish Ministry of Health Willett WC. Dietary patterns and markers of systemic inflam- (Instituto de Salud Carlos III, Thematic Network G03/140 mation among Iranian women. J Nutr 2007;137:992e8.
and RTIC RD06/0045, Fondo de Investigaciones Sanitarias, [12] Newby PK, Muller D, Tucker KL. Associations of empirically PI04/1828 and PI05/1839, PI07/0240), CYCYT AGL2005- derived eating patterns with plasma lipid biomarkers: 0365, the Public Health Division of the Department of a comparison of factor and cluster analysis methods. Am J ClinNutr 2004;80:759 Health of the Autonomous Government of Catalonia, and [13] van Dam RM, Rimm EB, Willett WC, Stampfer MJ, Hu FB.
Dietary patterns and risk for type 2 diabetes mellitus in U.S.
Catalans. None of the funding sources played a role in the design, collection, analysis or interpretation of the data or [14] Lutsey PL, Steffen LM, Stevens J. Dietary intake and the in the decision to submit the manuscript for publication.
development of the metabolic syndrome: the atherosclerosis None of the authors have any conflict of interest. CIBER is risk in communities study. Circulation 2008;117:754e61.
an initiative of the Instituto Carlos III, Spain.
[15] Esposito K, Marfella R, Ciotola M, Di Palo C, Giugliano F, Nureta-PREDIMED Investigators not listed as authors: Giugliano G, et al. Effect of a Mediterranean-style diet on ´ J, Isach-Subirana A, Tort Vernet R, Marti E, endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA 2004;292:1440 Martı´n-Lujan F, Sagarra Alomo R, Cabre [16] Azadbakht L, Mirmiran P, Esmaillzadeh A, Azizi T, Azizi F.
Beneficial effects of a Dietary Approaches to Stop Hyperten-sion eating plan on features of the metabolic syndrome. Dia- [17] Orchard TJ, Temprosa M, Goldberg R, Haffner S, Ratner R, Supplementary data associated with this article can be Marcovina S, et al. Diabetes Prevention Program Research Group. The effect of metformin and intensive lifestyle intervention on the metabolic syndrome: the Diabetes Prevention Program randomized trial. Ann Intern Med 2005; [31] Aparicio R, Roda L, Albi MS, Gutierrez F. Effect of various compounds on virgin olive oil stability measured by rancimat.
of Mediterranean diets in cardiovascular epidemiology: [32] Kecely T, Gordon MH. The antioxidant activity and stability of monounsaturated fats, olive oil, red wine or the whole the phenolic fraction of green olives and extra virgin olive oil.
pattern? Eur J Epidemiol 2004;19:9e13.
´n EE, Henrı´quez P, Serra-Majem L. Mediterranean diet and metabolic syndrome: a cross-sectional study in the Rivellese AA, Tapsell LC, et al. Substituting dietary satu- Canary Islands. Public Health Nutr 2006;9:1089e98.
rated for monounsaturated fat impairs insulin sensitivity in [20] Panagiotakos DB, Pitsavos C, Chrysohoou C, Skoumas J, healthy men and women: The KANWU Study. Diabetologia Tousoulis D, Toutouza M, et al. Impact of lifestyle habits on the prevalence of the metabolic syndrome among Greek adults from the ATTICA study. Am Heart J 2004;147:106e12.
reduced incidence of hypertension: the SUN study. Lipids [21] Tortosa A, Bes-Rastrollo M, Sanchez-Villegas A, Basterra- [35] Ferrara LA, Raimondi AS. d’Episcopo L, Guida L, Dello Russo A, ˜ez-Cordoba JM, Martinez-Gonzalez MA. Medi- Marotta T. Olive oil and reduced need for antihypertensive terranean diet inversely associated with the incidence of medications. Arch Intern Med 2000;160:837e42.
metabolic syndrome: the SUN prospective cohort. Diabetes [36] Rasmussen BM, Vessby B, Uusitupa M, Berglund L, Pedersen E, Riccardi G, et al. Effects of dietary saturated, mono- unsaturated, and n-3 fatty acids on blood pressure in healthy ´rrez V, Covas MI, et al. Effects of a Mediterranean- subjects. Am J Clin Nutr 2006;83:221e6.
style diet on cardiovascular risk factors: a randomized trial.
[37] Mensink RP, Janssen MC, Katan MB. Effect on blood pressure of two diets differing in total fat but not in saturated and poly- [23] Elosua R, Garcia M, Aguilar A, Molina L, Covas MI, Marrugat J.
unsaturated fatty acids in healthy volunteers. Am J Clin Nutr Validation of the Minnesota leisure time physical activity questionnaire in Spanish women. Investigators of the MAR- [38] Brenn T. The Tromsø heart study: alcoholic beverages and coro- ATDON group. Med Sci Sports Exerc 2000;32:1431e7.
nary risk factors. J Epidemiol Community Health 1986;40:249e56.
[24] Martin-Moreno JM, Boyle P, Gorgojo L, Maisonneuve P, Fer- [39] Zilkens RR, Burke V, Hodgson JM, Barden A, Beilin LJ, nandez-Rodriguez JC, Salvini S, et al. Development and Puddey IB. Red wine and beer elevate blood pressure in validation of a food frequency questionnaire in Spain. Int normotensive men. Hypertension 2005;45:874e87.
[40] Ellison RC, Zhang Y, Qureshi MM, Knox S, Arnett DK, Province MA, et al. Lifestyle determinants of high-density Martı´nez M, Wright M, Gomez-Gracia E. Development of lipoprotein cholesterol: The National Heart, Lung, and Blood a short dietary intake questionnaire for the quantitative Institute Family Heart Study. Am Heart J 2004;147:529e35.
estimation of adherence to a cardioprotective Mediterranean diet. Eur J Clin Nutr 2004;58:1550e2.
[41] Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ.
[26] Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adher- Moderate alcohol intake and lower risk of coronary heart ence to a Mediterranean diet and survival in a Greek disease: meta-analysis of effects on lipids and haemostatic population. N Engl J Med 2003;348:2599e608.
[27] Panagiotakos DB, Pitsavos C, Stefanadis C. Dietary patterns: ´lez MA, de la Fuente-Arrillaga C, Nunez- a Mediterranean diet score and its relation to clinical and Cordoba JM, Basterra-Gortari FJ, Beunza JJ, Vazquez Z, et al. Adherence to Mediterranean diet and risk of devel- oping diabetes: prospective cohort study. BMJ 2008;336: [28] Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the meta- [43] Wang L, Manson JE, Buring JE, Sesso HD. Meat intake and the bolic syndrome. An American Heart Association/National risk of hypertension in middle-aged and older women.
Heart, Lung, and Blood Institute Scientific Statement. Circu- [44] Bazzano LA, He J, Ogden LG, Loria C, Vupputuri S, Myers L, ´lez MA, Tortosa A, Basterra-Gortari FJ, Bes- et al. Legume consumption and risk of coronary heart disease Rastrollo M. Mediterranean diet inversely associated with the in US men and women: NHANES I epidemiologic follow-up incidence of metabolic syndrome: the SUN prospective study. Arch Intern Med 2001;161:2573e8.
cohort: response to Giugliano, Ceriello, and Esposito.
[45] Mellen PB, Walsh TF, Herrington DM. Whole grain intake and cardiovascular disease: a meta-analysis. Nutr Metab Cardiovasc [30] Panagiotakos DB, Pitsavos C, Das UN, Skoumas Y, Stefanadis C.
Dis 2008;18:283e90. doi:10.1016/j.numecd.2006.12.008.
The implications of anthropometric, inflammatory and gly- [46] Frost G, Leeds AA, Dore CJ, Madeiros S, Brading S, caemic control indices in the epidemiology of the metabolic Dornhorst A. Glycaemic index as a determinant of serum HDL- syndrome given by different definitions: a classification cholesterol concentration. Lancet 1999;353:1045e8. doi: analysis. Diabetes Obes Metab 2007;9:660e8.

Source: http://predimed.onmedic.es/Portals/0/adherence_11.pdf