Clin Nutr. 2013 - Universidad de Navarra

Clinical Nutrition 32 (2013) 172e178

Contents lists available at SciVerse ScienceDirect

Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu

Randomized control trials

The Mediterranean diet improves the systemic lipid and DNA oxidative damage in metabolic syndrome individuals. A randomized, controlled, trial Maria Teresa Mitjavila a, l, Marta Fandos b, l, Jordi Salas-Salvadó c, l, María-Isabel Covas d, l, Silvia Borrego b, l, Ramón Estruch e, l, Rosa Lamuela-Raventós f, l, Dolores Corella g, l, Miguel Ángel Martínez-Gonzalez h, l, Julia M. Sánchez b, l, Mónica Bulló c, l, Montserrat Fitó g, l, Carmen Tormos b, l, Concha Cerdá i, l, Rosario Casillas a, l, Juan José Moreno a, l, Antonio Iradi j, l, Cristóbal Zaragoza b, l, Javier Chaves k, l, Guillermo T. Sáez b, i, *, l a

Department of Physiology, INSA, University of Barcelona, Spain Department of Biochemistry and Molecular Biology, Faculty of Medicine University of Valencia, INCLIVA, Valencia, Spain c Human Nutrition Unit, IISPV and CIBER of Physiopathology of Obesity and Nutrition (CIBEROBN), Rovira i Virgili University, Reus, Spain d Cardiovascular Risk and Nutrition Research Group, CIBER of Physiopathology of Obesity and Nutrition (CIBEROBN), IMIM-Institut de Recerca Hospital del Mar, Barcelona, Spain e Department of Internal Medicine, IDIBABS, Hospital Clinic, University of Barcelona, Spain f Nutrition and Food Science Department-XaRTA, INSA, University of Barcelona, Spain g Department of Preventive Medicine, University of Valencia, Spain h Department of Preventive Medicine and Public Health, University of Navarra, Pamplona, Spain i Service of Clinical Analysis, CDB-General University Hospital, Valencia, Spain j Department of Physiology, Faculty of Medicine, University of Valencia, Spain k Laboratory of Genetic Studies, Valencia Clinical Hospital Research Foundation (INCLIVA), Valencia, Spain b

a r t i c l e i n f o

s u m m a r y

Article history: Received 1 March 2012 Accepted 6 August 2012

Background & aims: Metabolic syndrome (MetS), in which a non-classic feature is an increase in systemic oxidative biomarkers, presents a high risk of diabetes and cardiovascular disease (CVD). Adherence to the Mediterranean Diet (MedDiet) is associated with a reduced risk of MetS. However, the effect of the MedDiet on biomarkers for oxidative damage has not been assessed in MetS individuals. We have investigated the effect of the MedDiet on systemic oxidative biomarkers in MetS individuals. Methods: Randomized, controlled, parallel clinical trial in which 110 female with MetS, aged 55e80, were recruited into a large trial (PREDIMED Study) to test the efficacy of the traditional MedDiet on the primary prevention of CVD. Participants were assigned to a low-fat diet or two traditional MedDiets (MedDiet þ virgin olive oil or MedDiet þ nuts). Both MedDiet group participants received nutritional education and either free extra virgin olive oil for all the family (1 L/week), or free nuts (30 g/day). Diets were ad libitum. Changes in urine levels of F2-Isoprostane (F2-IP) and the DNA damage base 8-oxo-7,8dihydro-20 -deoxyguanosine (8-oxo-dG) were evaluated at 1-year trial. Results: After 1-year urinary F2-IP decreased in all groups, the decrease in MedDiet groups reaching a borderline significance versus that of the Control group. Urinary 8-oxo-dG was also reduced in all groups, with a higher decrease in both MedDiet groups versus the Control one (P < 0.001). Conclusions: MedDiet reduces oxidative damage to lipids and DNA in MetS individuals. Data from this study provide evidence to recommend the traditional MedDiet as a useful tool in the MetS management. Registered under Clinical Trials.gov Identifier no. NCT00123456. Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Keywords: Mediterranean diet Oxidative stress Metabolic syndrome DNA damage F2-isoprostanes PREDIMED

Non standard abbreviations: ATP III, Adult Treatment Panel; BMI, body mass index; CVD, cardiovascular disease; MedDiet, Mediterranean diet; MetS, metabolic syndrome; F2-IP, F2-isoprostanes; PUFA, polyunsaturated fatty acids; TG, triglycerides; TC, total cholesterol; VOO, virgin olive oil; 8-oxo-dG, 8-oxo-7,8-dihydro-20 -deoxyguanosine. * Corresponding author. Department of Biochemistry and Molecular Biology, Faculty of Medicine, University of Valencia, Avda. Blasco Ibañez 15, 46010 Valencia, Spain. Tel.: þ34 96 3864160; fax: þ34 96 3864101. E-mail address: [email protected] (G.T. Sáez). l For the PREDIMED study investigators. 0261-5614/$ e see front matter Ó 2012 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. http://dx.doi.org/10.1016/j.clnu.2012.08.002

M.T. Mitjavila et al. / Clinical Nutrition 32 (2013) 172e178

1. Introduction The metabolic syndrome (MetS) is a constellation of metabolic abnormalities that includes central obesity, dyslipidemia, elevated blood pressure, and hyperglycemia, all well documented risk factors for cardiovascular disease (CVD).1 The prevalence of the MetS is increasing, affecting almost one-fourth of the adult population, in a direct relationship with the global epidemic of obesity and diabetes.2 MetS components separately increase the risk of diabetes, CVD and all-cause mortality, but the full syndrome is associated with risk increases over and above those incurred by each isolated feature.3 Low-grade inflammation and high oxidative stress has been shown to be associated to this risk. MetS promotes an increase in the degree of low density lipoprotein (LDL) oxidation in a feedback loop.4 Systemic oxidative markers, such as F2isoprostanes (F2-IP), and markers of oxidative damage to DNA have been also reported to be increased in MetS patients.5,6 The Mediterranean diet (MedDiet), a reputedly healthy food pattern, has been associated with reduced prevalence7 and incident8 MetS. A recent meta-analysis with a large sample size showed that adherence to the MedDiet was associated with reduced risk of MetS, as well as the protective role of the MedDiet on all individual MetS components.9 However, no randomized, controlled intervention studies have assessed the efficacy of the MedDiet on the reversion of the systemic oxidative damage in MetS individuals. Olive oil is the main fat component of the MedDiet. Among olive oils, extra virgin olive oil (VOO) has the highest antioxidant phenolic content in front of other olive oils as ordinary or pomace.10 Nuts, which are also typical Mediterranean foods, are a rich source of nutrients and antioxidant phytochemicals.11 We designed a large-scale feeding trial in a population at high-risk for coronary heart disease in order to assess the effects on cardiovascular outcomes of two MedDiets, one supplemented with VOO and the other with mixed nuts, compared with a low-fat diet (the PREDIMED Study). We report, herein, the results of a 1-year intervention on two systemic markers of oxidative damage, the F2-IP as a marker of lipid oxidation and the 8-oxo-7,8-dihydro-20 -deoxyguanosine (8-oxo-dG) as a marker of DNA damage, in MetS participants recruited into the trial. 2. Subjects and methods 2.1. Subjects One hundred and ten non-smoking female participants from the PREDIMED study with the diagnosis of MetS randomly selected from the Primary Care Centers of Reus and Barcelona areas (Spain) were included in the present sub-study aimed to evaluate the effect of MedDiet on systemic oxidative stress biomarkers during the first year of intervention. MetS was defined by the updated Adult Treatment Panel (ATP III) criteria, when individuals met at least three of the following features: (a): waist circumference 88 cm, (b) triglycerides (TG) 150 mg/dl or drug treatment for elevated TG, (c) HDL-cholesterol concentrations < 42 mg/dl or on drug treatment for reduced HDL-cholesterol, (d) blood pressure 130/85 mmHg or on pharmacological treatment for hypertension, and (e) fasting serum glucose 100 mg/dL or drug treatment for elevated glucose. Subjects with acute or chronic infection, inflammatory disease, cancer, or leukocytosis before the randomization were excluded from the present study. Other causes of patient exclusion were body mass index (BMI) > 35 kg/m2, a weight change 5 kg in the previous three months, use of non-steroidal anti-inflammatory drugs, corticosteroid, hormonal or antibiotic treatment, alcoholism or active drug dependence or smoking. The protocol was

173

approved by the institutional review boards of the implicated University Hospitals from Reus and Barcelona. All participants provided written informed consent, and all procedures were conducted following the Helsinki Declaration Principles. 2.2. Study design The PREDIMED study is a large, parallel group, multicenter, controlled 4-year clinical trial that aims to assess the effects of the traditional MedDiet on the primary prevention of CVD. Full details of the protocol have been published elsewhere.12 Briefly, participants were randomly assigned to 3 intervention groups: a) MedDiet with extra VOO (MedDiet þ VOO), b) MedDiet with mixed nuts (MedDiet þ Nuts), and c) advice on low-fat diet (Control group). Both MedDiet groups receive intensive education to follow the MedDiet and VOO or mixed nuts that were free provided. In the Control group, participants were given advice to follow a low-fat diet. At the pre-inclusion visit, a medical clinical history was taken, a physical examination performed and biochemical analyses made to verify health status and compliance with the inclusion criteria. 2.3. Dietary interventions On the basis of the initial assessment of individual scores of adherence to the Mediterranean diet using a validated 14-item questionnaire,13 dieticians gave personalized dietary advice to participants randomized to both MedDiets, with instructions directed to upscale the score, including, among others: 1) abundant use of olive oil for cooking and dressing; 2) increased consumption of fruits, vegetables, legumes and fish; 3) reduction in total meat consumption, recommending white meat instead of red or processed meat; 4) preparation of home-made sauce with tomato, garlic, onion, and spices with olive oil to dress vegetables, pasta, rice, and other dishes; 5) avoidance of butter, cream, fast-food, sweets, pastries and sugar-sweetened beverages; and 6) in alcohol drinkers, moderate consumption of red wine. Dietary interventions have been described in detail elsewhere.12 Periodical visits (at baseline and every 3-month) were programmed with the dietician to assess and provide dietetic support so that participants could comply with dietetic recommendations. At inclusion and quarterly thereafter, dieticians administered both individual interviews and group sessions, separately for each group. Sessions consisted of informative talks and delivery of written material with elaborated descriptions of typical foods for each dietary pattern, seasonal shopping lists, meal plans, and recipes. Participants assigned to MedDiet groups were given free allotments of either VOO (1 L/week) or mixed nuts (15, 7.5 and 7.5 g/d of walnuts, almonds and hazelnuts, respectively). Participants allocated the low-fat diet received recommendations to reduce all types of fat, from both animal and vegetable sources, following the American Heart Association dietary guidelines.14 Instead, to encourage adherence, at quarterly visits they were given small gifts, such as oil dispensers, aprons, shopping bags, or cookbooks. In the three intervention groups, energy restriction was neither advised, nor physical activity promotion during the study. 2.4. Measurements At baseline (before randomization) and after 1-year follow-up, information was collected on subjects’ medical history and use of medication. Food intake was evaluated using a semiquantitative food frequency questionnaire of 146 item validated for our population.15 Energy and nutrient consumption were calculated using

174


Spanish food composition tables. Subjects had their height and weight measured wearing light clothing and no shoes by trained staff. Their BMI was calculated as the weight (kg) divided by the square of the height (m). Waist circumference was measured midway between the lowest rib and the iliac crest. Blood pressure was measured, using a validated semi-automatic oscillometer (Omron HEM-705CP, Hoofddorp, Netherlands), in duplicate with a 5-min interval between each measurement, and the mean of these values was recorded. Centralized laboratory biochemical analyses were performed on blood samples obtained in fasting conditions. Plasma glucose, serum cholesterol, HDL-c and TG and 24 h urinary creatinine levels were measured using standard enzymatic automated methods. In patients whose TG levels were less than 400 mg/dL, LDL-c concentrations were estimated using the Friedewald formula.16 2.4.1. Assay of urinary F2-isoprostanes F2-IP were extracted from the urine samples using an immunoaffinity column (Cayman Chemical Corp). Aliquots of 0.5 mL urine were directly applied to the columns, and the columns were washed with 2 mL of 0.1 mol/L PBS followed by 2 washes with 2 mL of water and the levels of F2-IP determined by immunoassay commercial EIA kits from Cayman Chemical. Readings were done at 405 nm in a Multiskan EX (Thermo Labsystems). The results were expressed as ng of F2-IP/mmol creatinine. 2.4.2. Assays of urinary 8-oxo-7,8-dihydro-20 -deoxyguanosine Aliquots (2 10 mL) of premixed 24 h urine samples collected in polyethylene bottles were kept at 20 C until analyses. The method of detection of 8-oxo-dG was based on that of Brown et al.17 For the separation of 8-oxo-dG a Waters 515 HPLC pump model was used. This separation was carried out using a 5 mm Spherisorb ODS2 column (4.6 mm 250 mm) with a flow rate of 1 mL/min. The buffer used was 50 mmol/L potassium phosphate pH 5.1 in 5% of acetonitrile, and the retention time was 7.5 min. To assess the optimization and accuracy of the HPLC-EC assay for the isolation and detection of 8-oxo-dG, appropriate chromatograms of samples and standards were recorded at the beginning of each working day. The 8-oxo-dG values were expressed as the ratio to creatinine urine concentration given in mmol/mL.18 2.5. Statistical analysis Quantitative variables are presented as mean values and standard deviation (SD) or standard error (SE). Categorical variables are presented as percentages (%). Data were first tested for normal distribution using the KolmogoroveSmirnov statistical test, and all of them were normally distributed. Baseline means were compared using the one-way unpaired Student’s t-test. To test differences at baseline between groups of intervention, the Chi-squared Pearson test for categorical variables, and the analysis of variance (ANOVA) for quantitative variables were used. In case of body weight, serum glucose and lipid profile, changes between baseline and 1-year follow-up were analyzed using the Student’s-t test for paired data in the same group and between interventional groups were analyzed using ANOVA. The differences in changes between groups were tested in the case of oxidative stress biomarkers measurements by ANCOVA using a general linear model, with baseline values of each biomarker, intervention group and blood pressure as covariates. Differences were considered statistically significant at P 0.05. All statistical analyses were carried out with the Statistical Package for the Social Sciences, SPSS software for Windows version 17.0.

3. Results A total of 110 female participants with MetS were included in the present study (38 were randomized to the MedDiet þ VOO group, 35 to the MedDiet þ Nuts group and 37 to the Control diet group). Table 1 shows the general baseline characteristics of the subjects. Most of them had abdominal obesity and hypertension. Diabetes mellitus was present in approximately 55% of the population. No differences in the prevalence of the MetS components (except for the blood pressure component that was less prevalent in the MedDiet þ VOO group), the medication use for controlling these metabolic abnormalities, lifestyle variables (physical activity or score of MedDiet adherence) and glucose level and lipid profile were shown between groups. Baseline urine concentrations of F2-IP/mmol creatinine were significantly lower in patients randomized to the MedDiet þ VOO group than the other two groups (Table 1). Urinary levels of 8-oxodG/mmol creatinine were significantly higher in the Control group in comparison to the MedDiet group supplemented with mixed nuts and in MedDiet þ VOO group (Table 1). Table 2 shows baseline and 1-year follow-up changes in energy and nutrient intake. The main dietary changes recorded at 1 year were large increases in VOO and nut consumption in the corresponding MedDiet groups indicating a good compliance of the participants with the supplemental foods. Participants assigned to both MedDiet groups had non-significantly increased energy and energy from total fat. A significant increase in the intake of monounsaturated and polyunsaturated fatty acids was shown in the MedDiet þ VOO and MedDiet þ nuts participants, respectively. Participants in the MedDiet þ nuts reduced saturated fat intake. Participants in the Med þ VOO group increased fish and cereal intake. The global pattern of adherence to the MedDiets increased in all groups, the mean increase in the MedDiet groups being 1.7e 1.8-fold versus that of the Control group. No significant differences in changes between groups were observed in physical activity, body weight, blood pressure, serum glucose concentrations and lipid profile during the first year of follow-up (data not shown). Urinary levels of F2-IP/mmol creatinine and 8-oxo-dG/mmol creatinine were significantly decreased in all groups (Table 3) after 1-year follow-up. The decrease in the urinary levels of F2-IP reached a borderline significance (P ¼ 0.051) in both MedDiet groups compared to the Control diet group. The differences in changes of urinary levels of 8-oxo-dG during the follow-up between groups were significantly higher in both MedDiet groups compared to the Control group subjects, after adjusting for baseline levels and the prevalence of the blood pressure component of MetS (Table 3). 4. Discussion The present study is, to our knowledge, the first randomized, controlled clinical trial focused on the effect of a Mediterranean type diet on systemic oxidative damage in MetS individuals. In all intervention groups systemic F2-IP and 8-oxo-dG were reduced after 1-year intervention. However, the reduction of the oxidative damage was greater in MetS participants who improved their diet toward a MedDiet pattern in comparison with individuals assigned to a low-fat diet. DNA oxidative damage was reduced 6.6 and 8.1fold in MedDiet þ VOO and MedDiet þ nuts, respectively, versus the Control group. These are important observations suggesting that reduction of ROS-induced oxidative modifications may be an essential step toward cardiovascular prevention. However, the effect is higher in term of DNA damage compared with the lipid peroxidation product. This difference may be related with the fact


175

Table 1 Baseline characteristics of participants completing 1-year follow-up. Variables

MedDiet þ VOO (n ¼ 38)

MedDiet þ Nuts (n ¼ 35)

Control diet (n ¼ 37)

Pb

Age in years, mean SD Body weight in kg, mean SD Body mass index in kg/m2, mean SD Waist circumference in cm, mean SD Metabolic syndrome components, % (n)a Abdominal obesity Low level of HDL cholesterol High triglyceride levels or receiving treatment for same High fasting serum glucose or drug treatment for diabetes High blood pressure (130/85 mm Hg) or antihypertensive treatment Medications, % (n) Antihypertensive agents Statins Fibrates Insulin Oral hypoglycemic agents Antiplatelet agents Lifestyle Mediterranean diet score, mean SD Physical activity in METS, mean SD Current smokers, % (n) Glucose and lipid profile, mean SE Glucose in mg/dL LDL-cholesterol in mg/dL HDL-cholesterol in mg/dL Triglycerides in mg/dL Oxidation markers, mean SE F2-IP in ng/mmol creatinine 8-oxo-dG in mmol/mmol creatinine

69.2 5.7 74.6 10.5 31.4 3.6 101.2 8.7

68.6 5.2 76.4 10.7 31.9 3.5 102.0 8.2

68.2 5.3 74.7 9.2 31.7 3.7 102.5 8.8

0.722 0.689 0.864 0.789

97.3 21.0 15.8 57.9 71.0

97.1 (34) 34.3 (12) 34.3 (12) 68.6 (24) 100.0 (35)

94.6 18.9 18.9 48.6 89.2

(35) (7) (7) (18) (33)

0.779 0.262 0.135 0.171 0.001

71.0 (27) 34.2 (13) 0.0 (0) 10.5 (4) 36.8 (14) 23.7 (9)

91.4 (32) 34.3 (12) 5.7 (2) 11.4 (4) 45.7 (16) 25.7 (9)

78.4 (29) 32.4 (12) 2.7 (1) 8.1 (3) 29.7 (11) 8.1 (3)

0.690 0.982 0.326 0.888 0.373 0.105

7.53 3.32 154.0 103.9 0 (0)

7.97 2.25 179.7 135.0 0 (0)

7.78 2.50 157.1 127.8 0 (0)

0.785 0.396

129.0 48.2 131.9 30.8 57.2 16.4 128.8 56.0

129.3 50.7 124.9 33.8 55.3 15.2 150.2 68.8

117.4 30.9 129.4 29.8 56.1 11.5 136.2 68.8

0.419 0.644 0.861 0.333

76.8 12.7 20.3 3.7

96.8 11.5 19.9 3.7

98.2 19.2 22.8 4.4

0.000 0.012

(37) (8) (6) (22) (27)

Abbreviations: MedDiet, Mediterranean diet; VOO, virgin olive oil; HDL, high-density lipoprotein. a The metabolic syndrome is defined by using the National Cholesterol Education Program’s Adult Treatment Panel III criteria. b P value for comparisons across groups with Pearson’s chi-square test for categorical variables or one-way ANOVA for continuous variables.

that VOO and nuts are rich in unsaturated fatty acids which may increase lipid peroxidation threshold. This observation supports the use of more than a single oxidative stress marker to validate the antioxidant efficiency of specific dietary components. We used urinary F2-IP as marker of systemic lipid peroxidation and urinary 8-oxo-dG as marker for DNA oxidation. F2-IP are produced in vivo by a non-enzymatic free radical-catalyzed peroxidation of the arachidonic acid and its measure in the urine has emerged as one of the reliable approaches to assess systemic lipid peroxidation in vivo.19 High F2-IP levels have been shown to be predictors for atherosclerosis20 and also to be independently associated with angiographic evidence of coronary artery disease.21 8oxo-dG is the final product of guanine oxidation. This modified base has an important mutagenic potential and its presence in the DNA is responsible of its molecular instability giving to an accumulation to gene mutations and cancer initiation.22 In this sense, MetS, in which a non-classical feature is an increased DNA oxidation,6 has been associated with a higher risk of malignant tumors such as hepatocellular,23 breast,24 prostate,25 and colorectal.26 Somatic DNA mutations, promoted by DNA oxidation, are also considered to be involved in the atherosclerotic processes.27 Increased levels of 8-oxo-dG has been found both in nuclear and mitochondrial DNA of hypertensive subjects28 as well as other cardiovascular-related alterations such as combined hyperlipemia.29 The MedDiet is a dietary pattern rich in antioxidants and it has been proved to be successful in cardiovascular risk prevention.30 Olive oil, the main fat component of the diet, has been shown to reduce the levels of 8-oxo-dG in healthy individuals,31 and polyphenols from VOO have shown to have an added value in reducing urinary and mitochondrial 8-oxo-dG in increased oxidative stress situations.32,33 In our experience, nuts (30 g/day, 3 months) have been shown to reduce the levels of 8-oxo-dG in individuals with MetS.34 In the frame of the PREDIMED Study, we have previously

reported a decrease in the circulating oxidized LDL35 after 3-month of both MedDiet interventions in the PREDIMED cardiovascular risk population. In this population, we have also observed an improvement of some MetS components such as an increase in HDL cholesterol, and reductions in both systolic and diastolic blood pressure and TG36 after 3-month interventions with MedDiets. We have also reported a significant reversion of the MetS as a whole after the MedDiet þ nuts intervention in individuals at high cardiovascular risk.37 Few intervention studies have reported decreases in F2-IP after healthy nutritional interventions. Due to this, the biological relevance of direct antioxidant effects of polyphenols for cardiovascular health in humans has been questioned.38 However, the most part of the intervention trials with antioxidants has been performed with a single food component, instead of a whole dietary pattern.38 In this study, extra VOO and nuts together with the antioxidants present in other main components of the MedDiet such as fruits, vegetables, and spices could explain the decrease in the systemic lipid and DNA oxidation observed in our MetS participants. In the present study, 110 women were randomly selected from different primary care centers of a Catalonian area in Spain. We decided to include only women in order to avoid variability due to gender differences. Sex-dependent differences have been reported in terms of, postprandial fat processing, mitochondrial function and oxidative stress39,40 as well as in plasma lipid response to dietary fat41 or the impact of MedDiet on cardiovascular risk profile.42 In our study, also a significant reduction from baseline of both oxidative stress markers in the urine of the low-fat control subjects was observed. Although this effect is less evident compared with that observed in both MedDiet groups, the decrease in the F2-IP urinary levels may have significant importance. This effect can be probably attributed to the low fat content of the control diet. Diets reduced in fat have been related to low levels of oxidative stress and

176


Table 2 Baseline and 1-year follow-up changes in energy and nutrient intake.a MedDiet þ VOO (n ¼ 38) Energy, (kcal/day) Baseline 2166 504 Change 203 606 Energy from total protein, (%) Baseline 16.6 3.2 Change 0.08 2.8 Energy from total carbohydrates, (%) Baseline 40.6 8.4 Change 1.2 8.1 Energy from total fat, (%) Baseline 42.0 7.6 Change 1.5 8.1 Saturated fatty acids, (%) Baseline 11.2 2.0 Change 0.03 2.7 Monounsaturated fatty acids, (%) Baseline 21.8 2.8 Change 1.0 3.2d Polyunsaturated fatty acids, (%) Baseline 7.3 2.7 Change 0.6 2.8e Alcohol (g/day) Baseline 2.1 4.4 Change 0.5 4.2 Virgin olive oil (g/day) Baseline 41.5 19.5 Change 11.3 23.2 Nuts (g/day) Baseline 11.5 14.6 Change 1.0 14.7e Vegetables (g/day) Baseline 321.4 177.5 Change 46.8 183.4 Fruits (g/day) Baseline 356.3 366.7 Change 9.2 439.9 Legums (g/day) Baseline 19.1 13.0 Change 0.7 11.0 Meat and meat products (g/day) Baseline 113.3 60.0 Change 13.3 66.5 Fish (g/day) Baseline 95.0 46.1 Change 24.9 49.5 Cereals (g/day) Baseline 194.2 96.1 Change 36.8 90.1 P14-item questionnaire Baseline 7.53 3.32 Change 1.7 3.9 a b c d e f

P valueb

MedDiet þ nuts (n ¼ 35)

P valueb

Control diet (n ¼ 37) 2227 490 10 577

P valueb

P valuec

0.898

0.877 0.234

0.078

2199 555 240 662

0.083

0.733

16.7 2.6 0.7 2.5

0.085

16.4 2.1 0.08 2.7

0.958

0.917 0.475

0.482

39.3 5.3 2.5 7.9

0.064

42.1 5.1 0.6 5.4

0.700

0.196 0.530

0.446

42.4 5.3 3.5 7.6

0.013

40.0 5.8 0.8 1.0

0.645

0.229 0.280

0.913

10.9 1.9 1.2 2.0

0.001

11.6 1.9 0.5 2.1

0.071

0.326 0.080

0.106

22.6 2.3 0.6 2.6

0.132

21.8 2.3 0.7 2.4

0.084

0.255 0.033

0.204

6.7 1.3 1.9 1.7f