Adherence to the Mediterranean Diet in Relation to Acute Coronary Syndrome or Stroke Nonfatal Events

A Comparative Analysis of a Case/Case-Control Study

Christina-Maria Kastorini, MSc; Haralampos J. Milionis, MD, PhD; Aggeliki Ioannidi, MSc; Kallirroi Kalantzi, MD; Vassilios Nikolaou, MD; Konstantinos N. Vemmos, MD; John A. Goudevenos, MD, PhD; Demosthenes B. Panagiotakos, PhD

Am Heart J. 2011;162(4):717-724. 

Abstract and Introduction

Abstract

Background Although the role of Mediterranean diet on cardiovascular disease prevention has long been evaluated and understood, its association with the development of stroke has been rarely examined. The aim of the present work was to comparatively evaluate the association between adherence to the Mediterranean diet and the development of an acute coronary syndrome (ACS) or ischemic stroke.
Methods During the period from 2009 to 2010, 1,000 participants were enrolled; 250 were consecutive patients with a first ACS, 250 were consecutive patients with a first ischemic stroke, and 500 population-based, control subjects, 1-for-1 matched to the patients by age and sex. Sociodemographic, clinical, psychological, dietary, and other lifestyle characteristics were measured. Adherence to the Mediterranean diet was assessed by the validated MedDietScore (theoretical range 0–55).
Results After various adjustments were made, it was observed that for each 1-of-55-unit increase of the MedDietScore, the corresponding odds ratio for having an ACS was 0.91 (95% CI 0.87–0.96), whereas regarding stroke, it was 0.88 (95% CI 0.82–0.94).
Conclusions The present work extended the current knowledge about the cardioprotective benefits from the adoption of the Mediterranean diet by showing an additional protective effect on ischemic stroke development.

Introduction

According to the World Health Organization, 17.1 million people died from cardiovascular disease (CVD) in 2004; 7.2 million, from coronary heart disease (CHD); and 5.7 million, from stroke. Also, there is an increasing trend globally.[1] Thus, prevention of CVD is considered of major public health importance, not only because of its increasing prevalence but also because of its negative impact on quality of life.[2]

Lifestyle changes and, especially, dietary interventions are cost-effective means for the prevention of CVD and the reduction of the social and economic burden of this condition.[1, 3, 4] The Mediterranean dietary pattern is one of the most well-known dietary patterns regarding its cardioprotective effects.[5] This pattern is characterized by high consumption of monounsaturated fatty acids, primarily from olives and olive oil. It encourages daily consumption of fruits, vegetables, whole grain cereals, and low-fat dairy products; weekly consumption of fish, poultry, tree nuts, and legumes; monthly consumption of red meat; and a moderate daily consumption of alcohol, normally with meals. The beneficial role of the Mediterranean diet regarding mortality from all causes, CHD, some types of cancer, and metabolic disorders, has already been reported.[6–9] In a recent meta-analysis of 530,000 participants from 50 studies, greater adherence to the Mediterranean diet was associated with a lower risk for metabolic syndrome and its constituents.[10] The main mechanism for this cardiometabolic protection is the antioxidant and anti-inflammatory effects that the aforementioned pattern has.[11, 12] However, despite the fact that many studies have underlined the beneficial role of the Mediterranean diet regarding the development of CHD, the role of this diet on the development of stroke has not been thoroughly examined.[13, 14] To the best of our knowledge, only one study has evaluated and attempted to compare the effect of the Mediterranean diet on the development of both acute coronary syndromes (ACS) and stroke.[13]

Thus, the aims of the present work were to evaluate the role of adherence to the Mediterranean dietary pattern on the likelihood of developing an ACS or stroke and to perform a comparative analysis between these 2 outcomes as regards the effect size of this potential cardioprotection. The significance of this work is of major importance for public health care planners, since it would test the Mediterranean diet–stroke hypothesis at population basis and may provide an additional, non pharmacologic means for the prevention of ischemic stroke.

Methods

Design and Sampling Procedure

A multicenter, case-control study, with individual (1-for-1) matching by age (within ±3 years) and sex, was conducted.[15] From October 2009 to December 2010, 500 of the 615 consecutive patients with a first ACS event (n = 250; 209 acute myocardial infarction, 41 unstable angina) or ischemic stroke (n = 250) and without any suspicion of previous CVD who entered in the cardiology and pathology clinics or the emergency units of 3 major general hospitals in Greece agreed to participate (participation rate 81%). For the patients with stroke who were unable to communicate (speech disorders, aphasia, memory problems), the information was obtained by a valid surrogate respondent (first-degree relative living in the same home with the patient and being aware of the participant's dietary habits and medical history). Patients with chronic neoplasmatic disease or chronic inflammatory disease, as well as individuals with recent changes in their dietary habits, were not enrolled in the study. Five hundred control subjects (250 matched 1-for-1 with patients with ACS and another 250 matched 1-for-1 with patients with stroke) were selected concurrently with the patients (to eliminate residual confounding) on a volunteer, population basis and from the same region of the patients. Controls were without any clinical symptoms or suspicions of CVD in their medical history, as this was assessed by a physician.

Based on a priori statistical power analysis, a sample size of 500 patients (250 ACS, 250 stroke) and 500 age- and sex-matched healthy subjects was adequate to evaluate 2-sided odds ratios equal to 1.20, achieving statistical power greater than 0.80 at .05 probability level (P value). However, to achieve more robust estimates of the effect-sized measures, bootstrap resampling method was also applied (see details in the "Statistical analysis" section).

Bioethics

The study was approved by the Ethics Committee of Cardiology Clinic, University of Ioannina Medical School, and was carried out in accordance to the Declaration of Helsinki (1989) of the World Medical Association. Before the collection of any information, participants (or valid surrogate respondents) were informed about the aims and procedures of the study and provided their written signed consent.

Diagnosis of ACS or Stroke

Regarding the patients with ACS, clinical symptoms were evaluated at hospital entry, and a 12-lead electrocardiogram was performed. Evidence of myocardial cell death was assessed with blood tests and measurement of the levels of troponin I and the MB fraction of total creatinine phosphokinase (according to the Universal Definition of Myocardial Infarction [Joint European Society of Cardiology/American College of Cardiology Foundation/American Heart Association/World Heart Federation Task Force]);[16] unstable angina was defined by the occurrence of 1 or more angina episodes, at rest, within the preceding 48 hours, corresponding to class III of the Braunwald classification.[17]Ischemic strokes were defined through symptoms of neurologic dysfunction of acute onset of any severity, consistent with focal brain ischemia and imaging/laboratory confirmation of an acute vascular ischemic pathology.[18]

Dietary Assessment and Evaluation of Mediterranean Diet

Dietary habits of the past year were assessed through a 90-item, validated semiquantitative food frequency questionnaire (FFQ) that has been previously described.[15] In brief, the validation of the FFQ was performed using, as a reference method, a 3-day food record; both tools were administered in 136 subjects (40 ± 14 years, 59 men) over the same time span, in 2010. Moderate-to-good validity was observed as regards all food groups and beverages studied (Kendall τ coefficients varied between 0.22 and 0.56, Ps < .05).[15] Level of adherence to the Mediterranean diet was evaluated using an 11-item, large-scale, composite index, the MedDietScore.[19] The score was calculated for each participant based on the information retrieved through the FFQ. For the consumption of foods presumed to be part of the Mediterranean pattern (ie, those suggested on a daily basis or >4 servings/week, such as nonrefined cereals, fruits, vegetables, legumes, olive oil, fish, and potatoes), lower scores were assigned when participants reported no, rare, or moderate consumption, whereas higher scores were assigned when the consumption was according to the rationale of the Mediterranean pattern. For the consumption of foods presumed not to be part of the Mediterranean pattern (ie, consumption of meat and meat products, poultry, and full-fat dairy products), scores were assigned on a reverse scale. For alcohol, score 5 was assigned for consumption of less than 3 wineglasses per day; score 0, for consumption of more than 7 wineglasses per day; and scores from 4 to 1, for consumption of 3, 4 to 5, 6, and 7 or 0 wineglasses per day, respectively. The theoretical range of the MedDietScore was between 0 and 55. Higher values of this diet score indicate greater adherence to the Mediterranean diet, whereas no specific thresholds have been proposed. The validation properties of the MedDietScore have been presented elsewhere in the literature.[19]

Sociodemographic, Clinical, Anthropometric, and Lifestyle Characteristics

Sociodemographic variables recorded were age and sex (for the matching procedure), educational level measured by years of school, and financial status evaluated indirectly using an index measuring how satisfied the participant was from his/her income (ie, from value 1, which means not at all satisfied, to value 9, which means very satisfied). Current smokers were defined as those who smoked at least 1 cigarette/day; former smokers, as those who had stopped smoking more than 1 year previously; and the rest of the participants, as noncurrent smokers. Physical activity was assessed using the International Physical Activity Questionnaire index,[20] which has been validated for the Greek population.[21] According to their physical activity levels, participants were classified as inactive or physically active (moderate or health-enhancing physical activity). Body mass index (BMI) was calculated as weight (in kilograms) divided by standing height (in meters squared);[22]overweight and obesity were defined as BMI 25.0 to 29.9 and >29.9 kg/m,[2] respectively.

In all participants, detailed medical history was recorded, including family history of CVD, as well as personal and family history of hypertension, hypercholesterolemia, hypertriglyceridemia, and diabetes. Patients whose average blood pressure levels were ≥140/90 mm Hg or were under antihypertensive medication were classified as having hypertension. Hypercholesterolemia was defined as total serum cholesterol levels >200 mg/dL or the use of lipid-lowering agents, and diabetes mellitus was defined as fasting blood glucose >126 mg/dL or the use of antidiabetic medication. A previously translated and validated version of the Zung Depression Rating Scale (ZUNG-DRS; range 20–80) was used for the assessment of depressive symptoms.[23, 24]

Statistical Analysis

Normally distributed continuous variables (age, BMI, education years, ZUNG-DRS and MedDietScore) are presented as mean values ± SD and categorical variables (sex, smoking habits, medical history, BMI categories, physical activity, financial status, MedDietScore categories, and ZUNG-DRS categories) as frequencies. Associations between categorical variables were tested by the calculation of the χ[2] test. Comparisons of mean values of normally distributed continuous variables by clinical outcome were performed using the Student t test. Correlations between continuous variables were evaluated using the Pearson r or Spearman ρ coefficients. Normality of the variables was tested using P-P plots. Estimations of the relative odds of having ACS or stroke according to the level of adherence to the Mediterranean diet and other covariates were performed through conditional logistic regression analysis; results are presented as odds ratios and the corresponding 95% CIs. Hosmer-Lemeshow statistic was calculated to evaluate a model's goodness of fit. Comparisons between the effect size measures (ie, odds ratios) of the 2 logistic models (the one for ACS and the other for stroke) were based on the Wald test (ie, log(odds ratio)/Var(log odds ratio), the higher the better). The robustness of the estimated odds ratios was evaluated using bootstrap resampling method of 1,000 data sets. A bias of the estimate less than 0.01 was considered adequate for the performance of the calculated odds ratio. All reported P values were based on 2-sided hypotheses. SPSS 18.0 software (SPSS Inc, Chicago, IL) was used for all the statistical calculations.

The authors are solely responsible for the design and conduct of this study, data analyses, drafting and editing of the paper, and its final contents. No extramural funding was used to support this work.

Results

Basic Characteristics of the Participants

Characteristics of the participants are presented in . Compared with the patients with stroke, the patients with ACS were mostly male and younger. All control subjects reported healthier dietary habits (higher MedDietScore) and were more physically active, satisfied with their financial status, and less frequently smokers than the patients (P < .05). Furthermore, lower prevalence of hypertension, diabetes mellitus, and CVD family history was observed in the controls compared with the patients (P < .05).

Table I.  Sociodemographic, lifestyle, and clinical characteristics of the study's participants

Patients with ACS (n = 250) Controls with ACS (n = 250) Patients with stroke ( n = 250) Controls with stroke (n = 250)
Age (y) 60 ± 12 60 ± 12 77 ± 9 73 ± 9
Male gender 208 (83.2) 208 (83.2) 139 (55.6) 139 (55.6)
Smoking habits
   Never smoker 56 (22.4) 108 (43.2) 151 (60.4) 139 (55.8)
   Current smoker 127 (50.8) 63 (25.2) 49 (19.6) 47 (18.9)
   Former smoker 67 (26.8) 79 (31.6) 50 (20) 63 (25.3)
Physical activity 150 (64.1) 203 (82.5) 99 (47.1) 181 (74.8)
Education years 10 ± 4.3* 11 ± 5.0 8 ± 4.6 8 ± 4.9
Financial status satisfaction
   Not satisfied 50 (22.7) 19 (8.1) 74 (35.9) 35 (14.6)
   Satisfied 118 (53.6) 154 (65.3) 114 (55.3) 157 (65.7)
   Very satisfied 52 (23.6) 63 (26.7) 18 (8.7) 47 (19.7)
Family history of CVD 81 (36.2) 39 (16.7) 51 (31.3) 38 (16.7)
Hypertension 148 (62.2) 90 (37.7) 206 (84.4) 137 (56.8)
Hypercholesterolemia 165 (71.4) 100 (45.5) 159 (68.5)* 119 (54.1)
Diabetes mellitus 58 (26.1) 29 (12.4) 71 (32.9)* 50 (21.5)
BMI (kg/m2) 27.82 ± 4.29 27.23 ± 3.50 26.72 ± 3.57 27.35 ± 4.24
   Normal weight (18.5–24.9) 57 (24.9) 63 (26.3) 79 (33.1) 73 (30)
   Overweight (25–29.9) 116 (50.7) 132 (55) 124 (51.9) 120 (49.4)
   Obese (>30) 56 (24.5) 45 (18.8) 36 (15.0) 50 (20.6)
MedDietScore (range 0–55) 30.67 ± 5.02 32.50 ± 4.41 29.99 ± 3.79 32.03 ± 4.08
   1st tertile (0–29) 86 (41.1) 50 (21.9) 94 (49.5)† 60 (26.8)
   2nd tertile (30–33) 66 (31.6) 79 (34.6) 64 (33.7) 82 (36.6)
   3rd tertile (34–55) 57 (27.3) 99 (43.4) 32 (16.8) 82 (36.6)
ZUNG-DRS (range 20–80) 38.50 ± 8.50 35.07 ± 7.90 46.30 ± 8.23 38.01 ± 8.57
   20–49: normal 203 (89.4) 233 (94.7) 121 (57.1)† 224 (91.8)
   50–59: mild depression 22 (9.7) 12 (4.9) 86 (40.6) 18 (7.4)
   60–69: moderate-marked depression 2 (0.9) 1 (0.4) 5 (2.4) 2 (0.8)
   70–80: severe depression 0 (0) 0 (0) 0 (0) 0 (0)

Data are expressed as mean ± SD or frequencies (n [%]). P values were derived from Student t test or the χ2 test.
* P < .05 compared with the control group with ACS.
P < .001 compared with the control group with stroke.

Compared with the patients with stroke, the patients with ACS were more likely to be physically active (P < .001), smokers (P < .001), and obese (P = .02) and had better education status (P < .001), higher financial satisfaction (P < .001), and lower ZUNG-DRS levels (P < .001); however, the patients with ACS were less likely to have hypertension (P < .001) ().

Table I.  Sociodemographic, lifestyle, and clinical characteristics of the study's participants

Patients with ACS (n = 250) Controls with ACS (n = 250) Patients with stroke ( n = 250) Controls with stroke (n = 250)
Age (y) 60 ± 12 60 ± 12 77 ± 9 73 ± 9
Male gender 208 (83.2) 208 (83.2) 139 (55.6) 139 (55.6)
Smoking habits
   Never smoker 56 (22.4) 108 (43.2) 151 (60.4) 139 (55.8)
   Current smoker 127 (50.8) 63 (25.2) 49 (19.6) 47 (18.9)
   Former smoker 67 (26.8) 79 (31.6) 50 (20) 63 (25.3)
Physical activity 150 (64.1) 203 (82.5) 99 (47.1) 181 (74.8)
Education years 10 ± 4.3* 11 ± 5.0 8 ± 4.6 8 ± 4.9
Financial status satisfaction
   Not satisfied 50 (22.7) 19 (8.1) 74 (35.9) 35 (14.6)
   Satisfied 118 (53.6) 154 (65.3) 114 (55.3) 157 (65.7)
   Very satisfied 52 (23.6) 63 (26.7) 18 (8.7) 47 (19.7)
Family history of CVD 81 (36.2) 39 (16.7) 51 (31.3) 38 (16.7)
Hypertension 148 (62.2) 90 (37.7) 206 (84.4) 137 (56.8)
Hypercholesterolemia 165 (71.4) 100 (45.5) 159 (68.5)* 119 (54.1)
Diabetes mellitus 58 (26.1) 29 (12.4) 71 (32.9)* 50 (21.5)
BMI (kg/m2) 27.82 ± 4.29 27.23 ± 3.50 26.72 ± 3.57 27.35 ± 4.24
   Normal weight (18.5–24.9) 57 (24.9) 63 (26.3) 79 (33.1) 73 (30)
   Overweight (25–29.9) 116 (50.7) 132 (55) 124 (51.9) 120 (49.4)
   Obese (>30) 56 (24.5) 45 (18.8) 36 (15.0) 50 (20.6)
MedDietScore (range 0–55) 30.67 ± 5.02 32.50 ± 4.41 29.99 ± 3.79 32.03 ± 4.08
   1st tertile (0–29) 86 (41.1) 50 (21.9) 94 (49.5)† 60 (26.8)
   2nd tertile (30–33) 66 (31.6) 79 (34.6) 64 (33.7) 82 (36.6)
   3rd tertile (34–55) 57 (27.3) 99 (43.4) 32 (16.8) 82 (36.6)
ZUNG-DRS (range 20–80) 38.50 ± 8.50 35.07 ± 7.90 46.30 ± 8.23 38.01 ± 8.57
   20–49: normal 203 (89.4) 233 (94.7) 121 (57.1)† 224 (91.8)
   50–59: mild depression 22 (9.7) 12 (4.9) 86 (40.6) 18 (7.4)
   60–69: moderate-marked depression 2 (0.9) 1 (0.4) 5 (2.4) 2 (0.8)
   70–80: severe depression 0 (0) 0 (0) 0 (0) 0 (0)

Data are expressed as mean ± SD or frequencies (n [%]). P values were derived from Student t test or the χ2 test.
* P < .05 compared with the control group with ACS.
P < .001 compared with the control group with stroke.

Modeling the Role of Mediterranean Diet on the Presence of ACS

Unadjusted analysis showed that adherence to the Mediterranean diet had a beneficial effect regarding the likelihood of having ACS (model 1) (). Further data analysis revealed that the association of the Mediterranean diet with the development of ACS remained significant (ie, 9% lower likelihood of ACS per 1/55-unit increase in the MedDietScore), even after controlling for several other potential confounders (models 2 and 3) (). The results were slightly mediated after controlling for education years, financial satisfaction, and depression status (models 4 and 5). Furthermore, stratified analyses on treated and nontreated (for hypertension, hypercholesterolemia, and diabetes) participants did not alter the effect size measures of MedDietScore on ACS outcome (data not shown). In addition, the participants in the highest tertile of the MedDietScore had a 68% (95% CI 40%-83%) lower likelihood of having an ACS, compared with those in the first tertile, after adjusting for the factors included in model 3. The bias of the odds ratio as regards the MedDietScore from the bootstrap procedure was very low, that is, −0.005 (with 95% CI for the bootstrap estimates for 1/55-unit increase, ie, log(odds ratio): −0.167, −0.044).

Table II.  Results from the multiple logistic regression analysis that was developed to evaluate the likelihood of having an ACS (outcome), according to the level of adherence to the Mediterranean diet (main effect), among 250 ACS cases and 250 controls

Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
MedDietScore (per 1/55 unit) 0.92 (0.88–0.96) 0.91 (0.86–0.96) 0.91 (0.87–0.96) 0.94 (0.89–1.00) 0.95 (0.90–1.02)
Physical activity (yes/no) 0.36 (0.19–0.66) 0.34 (0.18–0.64) 0.38 (0.18–0.80) 0.40 (0.19–0.85)
Ever smoker vs no smoker 3.24 (1.84–5.69) 3.53 (1.98–6.29) 3.23 (1.69–6.16) 3.50 (1.81–6.80)
Family history of CVD (yes/no) 2.14 (1.18–3.88) 2.11 (1.15–3.87) 2.46 (1.23–4.90) 2.42 (1.20–4.88)
Hypertension (yes/no) 2.58 (1.50–4.41) 2.71 (1.55–4.74) 3.17 (1.69–5.96) 2.90 (1.53–5.49)
Hypercholesterolemia (yes/no) 2.98 (1.81–4.90) 2.83 (1.71–4.69) 2.86 (1.63–5.03) 2.85 (1.60–5.05)
Diabetes mellitus (yes/no) 1.95 (0.98–3.88) 2.08 (1.03–4.23) 2.27 (1.03–4.98) 1.94 (0.86–4.35)
BMI (per 1 kg/m2) 0.99 (0.93–1.06) 0.98 (0.91–1.05) 0.99 (0.92–1.07)
Education years (per 1 y) 0.96 (0.90–1.03) 0.98 (0.91–1.05)
Financial status satisfaction
   Not satisfied 1.00 1.00
   Satisfied 0.30 (0.13–0.71) 0.39 (0.16–0.94)
   Very satisfied 0.35 (0.13–0.89) 0.47 (0.18–1.28)
ZUNG-DRS (per 1/80 unit) 1.06 (1.02–1.11)

Results are presented as odds ratio (95% CI) obtained from multiple conditional logistic regression.

Table II.  Results from the multiple logistic regression analysis that was developed to evaluate the likelihood of having an ACS (outcome), according to the level of adherence to the Mediterranean diet (main effect), among 250 ACS cases and 250 controls

Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
MedDietScore (per 1/55 unit) 0.92 (0.88–0.96) 0.91 (0.86–0.96) 0.91 (0.87–0.96) 0.94 (0.89–1.00) 0.95 (0.90–1.02)
Physical activity (yes/no) 0.36 (0.19–0.66) 0.34 (0.18–0.64) 0.38 (0.18–0.80) 0.40 (0.19–0.85)
Ever smoker vs no smoker 3.24 (1.84–5.69) 3.53 (1.98–6.29) 3.23 (1.69–6.16) 3.50 (1.81–6.80)
Family history of CVD (yes/no) 2.14 (1.18–3.88) 2.11 (1.15–3.87) 2.46 (1.23–4.90) 2.42 (1.20–4.88)
Hypertension (yes/no) 2.58 (1.50–4.41) 2.71 (1.55–4.74) 3.17 (1.69–5.96) 2.90 (1.53–5.49)
Hypercholesterolemia (yes/no) 2.98 (1.81–4.90) 2.83 (1.71–4.69) 2.86 (1.63–5.03) 2.85 (1.60–5.05)
Diabetes mellitus (yes/no) 1.95 (0.98–3.88) 2.08 (1.03–4.23) 2.27 (1.03–4.98) 1.94 (0.86–4.35)
BMI (per 1 kg/m2) 0.99 (0.93–1.06) 0.98 (0.91–1.05) 0.99 (0.92–1.07)
Education years (per 1 y) 0.96 (0.90–1.03) 0.98 (0.91–1.05)
Financial status satisfaction
   Not satisfied 1.00 1.00
   Satisfied 0.30 (0.13–0.71) 0.39 (0.16–0.94)
   Very satisfied 0.35 (0.13–0.89) 0.47 (0.18–1.28)
ZUNG-DRS (per 1/80 unit) 1.06 (1.02–1.11)

Results are presented as odds ratio (95% CI) obtained from multiple conditional logistic regression.

Modeling the Role of Mediterranean Diet on the Presence of Stroke

Regarding stroke, a greater adherence to the Mediterranean diet was also associated with a lower likelihood of having an event, in the unadjusted model and even after controlling for various potential confounding factors (). Specifically, each 1-of-55-unit increase of the MedDietScore was associated with a 12% lower likelihood of stroke. Results were slightly mediated after controlling for additional factors (models 4 and 5). Stratified analyses on treated and nontreated (for hypertension, hypercholesterolemia, and diabetes) participants did not alter the previous effect size measures on stroke outcome (data not shown). Further analysis showed that the participants in the highest tertile of adherence to the Mediterranean dietary pattern had a 78% (95% CI 54%-90%) lower likelihood of developing a stroke, compared with those in the first tertile, after taking into account potential confounders (used in model 3). The bias of the odds ratio as regards the MedDietScore from the bootstrap procedure was very low, −0.005 (with 95% CI for the bootstrap estimates for 1/55-unit increase: −0.224, −0.064).

Table III.  Results from the multiple conditional logistic regression analysis that was developed to evaluate the likelihood of having a stroke (outcome) according to the level of adherence to the Mediterranean diet (main effect), among 250 stroke cases and 250 controls

Independent variables Model 1 Model 2 Model 3 Model 4 Model 5
MedDietScore (per 1/55 unit) 0.87 (0.83–0.92) 0.87 (0.82–0.94) 0.88 (0.82–0.94) 0.91 (0.83–1.00) 0.94 (0.85–1.04)
Physical activity (yes/no) 0.57 (0.31–1.05) 0.54 (0.29–1.02) 0.59 (0.26–1.31) 0.66 (0.26–1.67)
Ever smoker vs no smoker 0.95 (0.51–1.75) 0.84 (0.45–1.58) 0.60 (0.28–1.31) 0.41 (0.17–0.99)
Family history of CVD (yes/no) 3.15 (1.68–5.92) 3.21 (1.69–6.09) 4.16 (1.86–9.32) 3.27 (1.30–8.24)
Hypertension (yes/no) 1.82 (1.00–3.28) 1.85 (1.01–3.38) 1.88 (0.86–4.12) 2.47 (0.95–6.44)
Hypercholesterolemia (yes/no) 1.92 (1.08–3.40) 1.88 (1.05–3.35) 1.70 (0.81–3.53) 1.53 (0.67–3.48)
Diabetes mellitus (yes/no) 1.18 (0.64–2.18) 1.26 (0.67–2.38) 1.68 (0.75–3.75) 1.62 (0.66–3.96)
BMI (per 1 kg/m2) 0.96 (0.89–1.02) 0.89 (0.81–0.99) 0.92 (0.83–1.02)
Education years (per 1 y) 0.99 (0.92–1.08) 1.02 (0.99–1.12)
Financial status satisfaction
   Not satisfied 1.00 1.00
   Satisfied 0.37 (0.16–0.89) 0.28 (0.11–0.74)
   Very satisfied 0.14 (0.04–0.45) 0.16 (0.04–0.59)
ZUNG-DRS (per 1/80 unit) 1.05 (1.00–1.11)

Results are presented as odds ratio (95% CI) obtained from multiple conditional logistic regression.

Based on the previous models, adherence to the Mediterranean diet seems to exert a slightly stronger effect size measure on stroke outcome as compared with ACS (Wald test: 12.93 vs 11.70; comparisons were made for similarly adjusted logistic regression models).

Discussion

Results of the present work support the beneficial effect of the Mediterranean dietary pattern regarding the presence not only of an ACS but also of an ischemic stroke event. It is of interest that the effect of this diet on stroke outcome was at least similar to that on ACS. This is one of the few studies that have examined the relationship between adherence to the Mediterranean diet and stroke and, to the best of our knowledge, one of the first studies attempting to perform a comparative analysis regarding the role of this traditional diet on the development of ACS or stroke. Despite the limitations this study may carry due to the retrospective design of the study, the results were robust and multiadjusted for various potential confounders, revealing an important, nonpharmacologic, public health message for the prevention of stroke.

Since the Seven Countries Study[5] in the 1970s and the randomized clinical trial Lyon Heart Study[25] in the 1990s, many studies have supported the beneficial effect of the Mediterranean diet on the development of CVD and, particularly, CHD.[6] The CARDIO2000 study, a case-control study with 848 patients with ACS and 1,078 age- and sex-matched control subjects, showed that a 10-unit increase of the MedDietScore was associated with a roughly 30% lower likelihood of having an ACS.[26] Trichopoulou et al[27] showed that adherence to the Mediterranean dietary pattern was associated with a 33% (95% CI 0.47–0.94) lower mortality from CHD. In addition, recent results of the large-scale, multinational INTERHEART study, including 27,098 participants from 52 countries, highlighted the important role of unhealthy dietary habits as a risk factor for myocardial infarction. Most importantly, the population attributable risk of an unhealthy diet was approximately 27% in men and 26% in women;[28] suggesting that most CHD evens could have been avoided by adopting a healthier dietary pattern. In the present work, the estimated attributable risk for the lowest tertile of adherence to the Mediterranean pattern was 40% for ACS.

Despite the plethora of studies as regards Mediterranean diet and CHD, few studies have examined the role of the diet on the development of stroke. The Nurses' Health Study, a prospective cohort study of 74,886 female participants, showed that adherence to this pattern exerts a protective effect regarding the development of stroke (relative risk of highest compared with lowest quintile: 0.87, 95% CI 0.73–1.02).[13] Furthermore, a recent case-control study of only 48 patients with stroke and 47 age- and sex-matched controls reported that adherence to the Mediterranean diet was associated with a 91% lower likelihood of ischemic stroke (95% CI 0.02–0.40).[14] In addition, results of the INTERSTROKE case-control study suggested that unhealthy dietary habits were associated with a 34% higher likelihood of ischemic stroke (95% CI 1.09–1.65, highest vs lowest tertile), whereas the population attributable risk was 17.3% (95% CI 9.4–29.6).[29] In the present work, similar to the 2 aforementioned studies, it was observed that a greater adherence to the Mediterranean diet was associated with a lower likelihood of having an ischemic stroke event, whereas the estimated attributable risk for the lowest tertile of adherence to the Mediterranean pattern was 37%.

It is widely known that oxidative stress and chronic inflammation play a crucial role for the development of atherosclerosis, influencing endothelial and vascular function. Not surprisingly, the protective role of the Mediterranean dietary pattern regarding CVD has been mainly attributed to the antioxidant and anti-inflammatory properties of this pattern. The basic components of this diet—olive oil, red wine, fruits and vegetables, and fish—are foods rich in vitamins, antioxidants, polyphenols, phytochemicals, and omega-3 fatty acids. Results of epidemiologic studies and clinical trials have shown that subjects following closer the Mediterranean diet had a higher total antioxidant capacity[11] and lower inflammatory and coagulation markers: C-reactive protein, interleukin-6, homocysteine, white blood cell, and fibrinogen levels.[12, 30] Furthermore, latest studies have shown the beneficial role of this diet on endothelial function.[30] In particular, adherence to the Mediterranean diet has been associated not only with a reduction in endothelial damage and dysfunction but also with improvement in the degenerative activity of the endothelium.[31]

Both CHD and ischemic stroke share many common risk factors[28, 29] but also important dissimilarities in the development and phenotypic expression of atherosclerotic plaques in coronary and cerebral arteries, possibly due to differences in genetic characteristics, anatomy, or different response of coronary and cerebral arteries to risk factors.[32] The presented results suggest that adherence to Mediterranean diet exerts a stronger effect regarding the development of stroke. It is possible that the antioxidant and anti-inflammatory benefits of the Mediterranean dietary pattern could have an apparent effect in cerebral and carotid arteries. In fact, a study by Shai et al[33] has shown significant regression of carotid atherosclerosis, for participants following a low-fat Mediterranean diet or a low-carbohydrate diet.

At this point, it should be noted that although the Mediterranean diet is believed to be followed closely only in countries surrounding the upper Mediterranean basin, population-based studies performed in northern Europe, United States, Chile, or Australia, suggest that the adoption of this type of diet can exert its beneficial effects globally.[13, 34, 35] In addition, globalization and improved transportations have contributed to the consumption of Mediterranean products worldwide. Nevertheless, it should be noted that there are still cultural, religion, and geoclimatologic aspects as regards dietary habits that will be difficult to overcome to spread this dietary pattern around the world.[36] Moreover, it could be speculated that the comparison of this diet with a control but less healthy, dietary pattern adopted elsewhere might provide more prominent results against ACS or stroke risk.

The holistic approach of dietary pattern, instead of food-specific, analysis applied here to evaluate the research hypotheses is now considered a very promising scientific area in dietary assessment and encouraged by several investigators in the field of nutritional epidemiology. Compared with the traditional food-specific analyses, dietary pattern analyses have considerable advantages regarding both concept and design; first of all, people do not eat isolated nutrients but consume a variety of foods and complex meals, whereas several methodological limitations, such as high levels of intercorrelation and synergistic effect of foods, are better controlled.[37, 38]

Strengths and Limitations

In the present work, the existing knowledge regarding adherence to the Mediterranean diet and the development of stroke was expanded; moreover, the effect size measures were even better to that of the role of diet on ACS. These findings state a new hypothesis as regards the role of diet in the primary prevention of both heart disease and stroke and may suggest common pathophysiologic mechanisms for the aforementioned outcomes. However, there are some limitations due to the retrospective, observational design of the study, such as the selection and the recall bias and the lack of causal interpretations. To minimize the selection bias, only cases with a first event were enrolled, and to minimize recall bias, accurate and detailed data from all participants during the first 3 days of hospitalization were obtained. Nevertheless, a controlled clinical trial would be more effective in assessing causal relationships, and it is suggested as a future work. For the dietary evaluation, an FFQ was administered; although these tools may carry measurement error and be less accurate (especially in energy and nutrients assessment) as compared with a diary, an effort was made to reduce these errors and inaccuracies of dietary reporting with its application by trained dietitians through face-to-face interviews. Moreover, although the FFQ used here has shown good validation properties, the information used was based only on food groups and level of adherence to a specific pattern. Overestimation/underestimation in reporting may also exist, especially in the measurement of diet (eg, people with a disease usually tend to overreport unhealthier dietary habits to provide a reason for their condition, or healthy individuals usually report healthier dietary habits when interviewed by specialists), smoking habits, and the onset of CVD risk factors (eg, the best available information is the date of first diagnosis, whereas the initiation of the condition and the burden of damage it may cause to the cardiovascular system remains unknown). However, an effort was given to retrieve accurate information from the participants' medical records, as well as their relatives. Residual confounding and the omitted variables, for example, direct financial status and other social parameters (due to lack of accuracy), separate information for medication (not combined with the health condition), and presence of stress (due to the collinearity with depression status), may have influenced the effect size measures. Regarding the patients with stroke, self-reported information was obtained from 76% of the sample, whereas for 60 patients (24%) unable to answer to the interviewer due to their condition, data were collected from a valid surrogate respondent. Moreover, the patients with coronary disease and stroke who died at hospital entry or the following day were not included in the study (survivor bias); thus, the results should be generalized only to CVD survivors. Finally, the inclusion of patients and controls from only 2 regions may limit the generalization of the findings to the whole country; nevertheless, it should be noted that Athens metropolitan area and Ioannina city in western Greece represent most of the Greek urban and rural population.

Conclusion

The present work is one of the first in the literature that revealed the beneficial effect of Mediterranean diet on ischemic stroke, extending the up-to-now cardioprotective action of this traditional dietary pattern to the brain. Taking into account the estimates regarding further increase of cases of both CHD and stroke in the future years, intense preventive actions should be made by health care authorities, to encourage the adherence of a healthy dietary pattern, like the Mediterranean diet.

References

  1. WHO. World Health Organization, Cardiovascular diseases (CVDs), Fact Sheet No. 317. Available at: http://www.who.int/mediacentre/factsheets/fs317/en/index.html 2011;Last accessed February 16, 2011

  2. Kahn R, Robertson RM, Smith R, et al. The impact of prevention on reducing the burden of cardiovascular disease. Diabetes Care. 2008;31:1686–1696

  3. Nishida C, Uauy R, Kumanyika S, et al. The joint WHO/FAO expert consultation on diet, nutrition and the prevention of chronic diseases: process, product and policy implications. Public Health Nutr. 2004;7:245–250

  4. Wood D. Established and emerging cardiovascular risk factors. Am Heart J. 2001;141(2 Suppl):S49–57

  5. Keys A, Menotti A, Karvonen MJ, et al. The diet and 15-year death rate in the seven countries study. Am J Epidemiol. 1986;124:903–915

  6. Sofi F, Abbate R, Gensini GF, et al. Accruing evidence on benefits of adherence to the Mediterranean diet on health: an updated systematic review and meta-analysis. Am J Clin Nutr. 2010;92:1189–1196

  7. Panagiotakos DB, Pitsavos C, Polychronopoulos E, et al. Can a Mediterranean diet moderate the development and clinical progression of coronary heart disease? A systematic review. Med Sci Monit. 2004;10:RA193–RA198

  8. Esposito K, Kastorini CM, Panagiotakos DB, et al. Mediterranean diet and weight loss: meta-analysis of randomized controlled trials. Metab Syndr Relat Disord. 2011;9:1–12

  9. Giugliano D, Esposito K. Mediterranean diet and metabolic diseases. Curr Opin Lipidol. 2008;19:63–68

  10. Kastorini CM, Milionis HJ, Esposito K, et al. The effect of Mediterranean diet on metabolic syndrome and its components a meta-analysis of 50 studies and 534,906 individuals. J Am Coll Cardiol. 2011;57:1299–1313

  11. Pitsavos C, Panagiotakos DB, Tzima N, et al. Adherence to the Mediterranean diet is associated with total antioxidant capacity in healthy adults: the ATTICA study. Am J Clin Nutr. 2005;82:694–699

  12. Chrysohoou C, Panagiotakos DB, Pitsavos C, et al. Adherence to the Mediterranean diet attenuates inflammation and coagulation process in healthy adults: the ATTICA Study. J Am Coll Cardiol. 2004;44:152–158

  13. Fung TT, Rexrode KM, Mantzoros CS, et al. Mediterranean diet and incidence of and mortality from coronary heart disease and stroke in women. Circulation. 2009;119:1093–1100

  14. Yau WY, Rexrode GJ. Which dietary and lifestyle behaviours may be important in the aetiology (and prevention) of stroke?. J Clin Neurosci. 2011;18:76–80

  15. Kastorini CM, Milionis HJ, Goudevenos JA, et al. Modelling the role of dietary habits and eating behaviours on the development of acute coronary syndrome or stroke: aims, design, and validation properties of a case-control study. Cardiol Res Pract. 2011;

  16. Thygesen K, Alpert JS, White HD. Universal definition of myocardial infarction. J Am Coll Cardiol. 2007;50:2173–2195

  17. Braunwald E. Heart disease. London, UK: 5th Edited by W.B. Saunders Company; 1997;

  18. Kidwell CS, Warach S. Acute ischemic cerebrovascular syndrome: diagnostic criteria. Stroke. 2003;34:2995–2998

  19. Panagiotakos DB, Pitsavos C, Stefanadis C. Dietary patterns: a Mediterranean diet score and its relation to clinical and biological markers of cardiovascular disease risk. Nutr Metab Cardiovasc Dis. 2006;16:559–568

  20. Craig CL, Marshall AL, Sjostrom M, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–1395

  21. Papathanasiou G, Georgoudis G, Papandreou M, et al. Reliability measures of the short International Physical Activity Questionnaire (IPAQ) in Greek young adults. Hellenic J Cardiol. 2009;50:283–294

  22. Eknoyan G. Adolphe Quetelet (1796-1874)—the average man and indices of obesity. Nephrol Dial Transplant. 2008;23:47–51

  23. Zung WW. A self-rating depression scale. Arch Gen Psychiatry. 1965;12:63–70

  24. Fountoulakis KN, lacovides A, Samolis S, et al. Reliability, validity and psychometric properties of the Greek translation of the Zung Depression Rating Scale. BMC Psychiatry. 2001;1:6

  25. de Lorgeril M, Salen P, Martin JL, et al. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation. 1999;99:779–785

  26. Panagiotakos DB, Pitsavos C, Matalas AL, et al. Geographical influences on the association between adherence to the Mediterranean diet and the prevalence of acute coronary syndromes, in Greece: the CARDIO2000 study. Int J Cardiol. 2005;100:135–142

  27. Trichopoulou A, Costacou T, Bamia C, et al. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med. 2003;348:2599–2608

  28. Anand SS, Islam S, Rosengren A, et al. Risk factors for myocardial infarction in women and men: insights from the INTERHEART study. Eur Heart J. 2008;29:932–940

  29. O'Donnell MJ, Xavier D, Liu L, et al. Risk factors for ischaemic and intracerebral haemorrhagic stroke in 22 countries (the INTERSTROKE study): a case-control study. Lancet. 2010;376:112–123

  30. Esposito K, Marfella R, Ciotola M, et al. Effect of a Mediterranean-style diet on endothelial dysfunction and markers of vascular inflammation in the metabolic syndrome: a randomized trial. JAMA. 2004;292:1440–1446

  31. Marin C, Ramirez R, Delgado-Lista J, et al. Mediterranean diet reduces endothelial damage and improves the regenerative capacity of endothelium. Am J Clin Nutr. 2011;93:267–274

  32. Dalager S, Paaske WP, Kristensen IB, et al. Artery-related differences in atherosclerosis expression: implications for atherogenesis and dynamics in intima-media thickness. Stroke. 2007;38:2698–2705

  33. Shai I, Spence JD, Schwarzfuchs D, et al. Dietary intervention to reverse carotid atherosclerosis. Circulation. 2010;121:1200–1208

  34. Kouris-Blazos A, Gnardellis C, Wahlqvist ML, et al. Are the advantages of the Mediterranean diet transferable to other populations? A cohort study in Melbourne, Australia. Br J Nutr. 1999;82:57–61

  35. Trichopoulou A, Bamia C, Norat T, et al. Modified Mediterranean diet and survival after myocardial infarction: the EPIC-Elderly study. Eur J Epidemiol. 2007;22:871–881

  36. Regmi A, Ballenger N, Putnam J. Globalisation and income growth promote the Mediterranean diet. Public Health Nutr. 2004;7:977–983

  37. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13:3–9

  38. Panagiotakos DB, Pitsavos C, Stefanadis C. Alpha-priori and alpha-posterior dietary pattern analyses have similar estimating and discriminating ability in predicting 5-Y incidence of cardiovascular disease: methodological issues in nutrition assessment. J Food Sci. 2009;74:H218–H224