Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Mar 1.
Published in final edited form as: Stroke. 2015 Jan 27;46(3):780–785. doi: 10.1161/STROKEAHA.114.007894

Adherence to a Mediterranean diet and prediction of incident stroke

Georgios Tsivgoulis 1,2,3, Theodora Psaltopoulou 4, Virginia G Wadley 5, Andrei V Alexandrov 2, George Howard 6, Frederick W Unverzagt 7, Claudia Moy 8, Virginia J Howard 9, Brett Kissela 10, Suzanne E Judd 6
PMCID: PMC4621211  NIHMSID: NIHMS730469  PMID: 25628306

Abstract

Context

There are limited data regarding the potential association of adherence to Mediterranean Diet (MeD) with incident stroke in non-Mediterranean populations.

Objective

We sought to assess the longitudinal association between greater adherence to MeD and lower risk of incident stroke both with and without adjustment for factors that are independently related to stroke burden by capitalizing on the large, sample of the REasons for Geographic And Racial Differences in Stroke (REGARDS) subjects.

Design, Setting and Participants

Prospective, population-based, cohort of 30,239 individuals enrolled in REGARDS Study 2003–2007, excluding participants with stroke history, missing demographic data and Food Frequency Questionnaires and unavailable follow-up information. Adherence to MeD was categorized using dichotomization and trichotomization of MeD-score.

Main Outcome Measures

Incident stroke diagnosed by expert panel review of medical records using clinical and neuroimaging data during a mean follow-up period of 6.5 years.

Results

Incident stroke was identified in 565 participants (2.8%; 497 cases of ischemic and 68 cases of hemorrhagic stroke) out of 20,197 individuals fulfilling the inclusion criteria. Low (HR versus high adherence: 1.28; 95%CI:1.00–1.63) and moderate (HR versus high adherence: 1.32; 95%CI:1.05–1.66) adherence to MeD were associated with higher risk of incident ischemic stroke (IS) in initial univariate analysis. After adjusting for demographics, vascular risk factors, blood pressure levels, and antihypertensive medications use, low adherence to MeD (MeD-score: 0–3) tended to be associated with a higher risk of incident IS [HR versus high adherence (MeD-score: 6–9): 1.29; 95%CI:0.99–1.67]. A similar relationship was documented for moderate adherence to MeD (MeD score: 4–5; HR versus high adherence: 1.25; 95%CI:0.98–1.58). After dichotomization of MeD-score, low adherence to MeD (MeD-score: 0–4) was independently associated with higher incidence of IS in multivariable analyses (HR: 1.27;95%CI:1.04–1.54). We documented no association (p>0.5) of dichotomized or trichotomized MeD-score with incident hemorrhagic stroke. There was no interaction of race (p=0.38) on the association of adherence to MeD with incident IS.

Conclusions

Low adherence to MeD appears to be associated with a higher risk of incident IS independent of numerous potential confounders. Adherence to MeD is not related to the risk of incident hemorrhagic stroke.

Keywords: Mediterranean diet, stroke, ischemic, hemorrhagic, adherence, predictor

Introduction

Recent data from American Heart Association (AHA) indicates that every year, more than 795,000 individuals in the United States (US) have a stroke, with cerebrovascular diseases accounting for almost 1 of every 18 deaths in the US.1–3 On average, every 40 seconds, someone in the US has a stroke and every four minutes one American dies from stroke.1,2 Stroke is a major cause of disability in the US and stroke-related costs are estimated to reach $34 billions each year.1 By 2050, the incidence of stroke is estimated to more than double, with particularly large increases in the elderly and in minority groups (particularly Hispanics).4 Despite recent advances in acute stroke treatment, effective primary stroke prevention, by means of improved control of vascular risk factors, has the greatest potential to reduce its burden.5 Epidemiological studies have identified hypertension, diabetes, atrial fibrillation, transient ischemic attack, hypercholesterolemia, smoking and carotid stenosis as risk factors for stroke.6 However, only up to three quarters of all strokes can be attributed to these recognized risk factors.5,7 In recent years, several dietary factors including salt or high saturated fatty acids intake, dietary fiber, olive oil, fresh fruit/vegetable intake and moderate alcohol consumption have been shown to exhibit harmful or protective effects on the risk and mortality from stroke.6,8–10

The traditional Mediterranean diet (MeD) is a dietary pattern characterized by high consumption of plant foods (vegetables, fruits, legumes and cereals), high intake of olive oil as principal source of monounsaturated fat, low intake of saturated fat with limited consumption of meat and dairy products and moderate intake of fish and alcohol.11 It has recently received increased attention since high adherence to MeD has been associated with longer survival, reduced risk of cardiovascular or cancer mortality and lower likelihood of neurodegenerative disorders including Alzheimer’s Disease.12–15 Moreover, higher adherence to MeD was recently reported to be associated with a lower likelihood of incident cognitive impairment in the the REasons for Geographic And Racial Differences in Stroke (REGARDS) study.16 This association was attenuated by incident stroke and was moderated by the presence of diabetes mellitus. Furthermore, a systematic review ranked the Mediterranean-style diet as the most likely dietary model to provide protection against coronary heart disease.17 There are limited prospective data18–21 investigating the potential relationship of adherence to MeD with incident stroke and these are mainly in Caucasian populations18,19, while blacks have been under-represented in a single North-American study.21

The present study aims to extend previous research by capitalizing on the large, geographically dispersed, black-white race- and sex-balanced sample of REGARDS subjects. Acknowledging the fact that collection of data on numerous potential confounders or moderators of the relationship between adherence to a healthy dietary pattern such as MeD and incident stroke (vascular risk factors, health behaviors including exercise and smoking, demographic characteristics, depressive symptoms and cognitive status) have been included in the design of REGARDS, we sought to determine longitudinally the potential association between higher adherence to MeD and lower risk of stroke before and after controlling for factors that are independently related to cerebrovascular disease burden.22 Moreover, we also investigated potential relationships of MeD patterns with specific stroke subtypes including ischemic and hemorrhagic stroke by capitalizating on the validated REGARDS protocol for verification of incident stroke and specific stroke subtypes.22 Finally, we aimed to evaluate the potential interaction of race (blacks vs. whites) on the association of adherence to MeD with incident stroke (as well as stroke subtypes).

Subjects and Methods

Study Design

REGARDS is a national, population-based, longitudinal cohort study with oversampling from the Stroke Belt region of the US, an area that has stroke mortality rates higher than the rest of the US.22 From January 2003 – October 2007, a total of 30,239 individuals 45 years or older were enrolled; the cohort is followed by telephone twice a year for incident stroke. By design, the sample includes 21% from the “buckle” of the Stroke Belt (coastal plain region of North Carolina, South Carolina, and Georgia), 35% from the Stroke Belt states (remainder of North Carolina, South Carolina, and Georgia, plus Alabama, Mississippi, Tennessee, Arkansas, and Louisiana), and the remaining 44% from the other 40 contiguous states and the District of Columbia. Methodological details are available elsewhere.22–25

Standard Protocol Approvals, Registration and Consents

The study methods have been reviewed and approved by the Institutional Review Boards of all participating institutions. Written informed consent was obtained from all participants involved in the study.

Data collection and definitions

Data variables were gathered at baseline via a computer-assisted telephone interview followed by a home visit 3–4 weeks later during which blood, urine, blood pressure, ECG, medication audit, and anthropometric data were collected. Self-administered questionnaires (including a food frequency questionnaire) were left with the participant to gather information and mail back. Variables included in the present analysis are: age, race, sex, region of residence, height, weight, BMI, waist circumference, income, education, smoking status, alcohol use, sedentary behavior (more than 4 hours of screen time and no physical activity), myocardial infarction, diabetes mellitus, atrial fibrillation (by self report or centrally read electrocardiogram), systolic blood pressure (SBP), diastolic blood pressure (DBP), high cholesterol, antihypertensive regimen (specific drug classes), and perceived general health status. Additional details regarding definitions of data variables are available elsewhere.16,22–25

Dietary assessment and MeD

Average food consumption information at baseline was obtained using the self-administered Block 98 Food Frequency Questionnaire,26 which was left with each participant during the in-person visit with instructions for completion and a stamped envelope in which to return the questionnaire. For construction of MeD-score, we followed the most commonly described method that has been previously used by our group as well as other investigators.14–16 First, we identified the nine food groups considered to be part of the Mediterranean-type Diet Score: (i) vegetables, (ii) fruits, (iii) legumes, (iv) cereals (including bread, pasta and rice), (v) fish; (vi) meat; (vii) dairy products; (viii) fat intake and (ix) alcohol intake. Second, we regressed caloric intake (kilocalories) and calculated the derived residuals of daily gram intake for 7 food categories (vegetables, fruits, legumes, cereals, fish, meat and dairy products). Individuals were assigned a value of 1 (i) for each beneficial component (fruits, vegetables, legumes, cereals and fish) whose consumption was at or above the median and (ii) for each detrimental component (meat and dairy products) whose consumption was below the median. For fat intake (eighth food category) we used the ratio of daily consumption (in grams) of monounsaturated lipids to saturated lipids14–16 and we calculated the median separately for each sex. Individuals with ratios at or above the sex-specific median were assigned a value of 1.

Alcohol intake was analyzed according to the National Institute on Alcohol Abuse and Alcoholism recommendations.14–16 Moderate consumption was defined as between 1 and 7 drinks per week for women and between 1 and 14 drinks per week for men. More-than-moderate consumption was defined as more than 7 drinks per week for women and more than 14 drinks per week for men. Individuals were assigned a score of 1 (lower risk for cardiovascular disorders or dementia) for moderate consumption (different cut-offs for men and women) and a score of 0 for the other two categories (zero and more-than-moderate consumption).14–16 The MeD score was computed as the sum of scores in the nine food categories (range 0–9) with a higher score indicating a higher adherence to MeD.14–16,27–29 Adherence to MeD was categorized as high, moderate and low using MeD-score tertiles (MeD-score: 6–9, 4–5 & 0–3 respectively).14–16 The diet score was also analyzed in a median split (low adherence range: 0–4; high adherence range: 5–9).14–16 Adherence to MeD was assessed using both trichotomization and dichotomization of MeD-score in order to yield comparable results to previous studies evaluating potential associations of adherence to MeD with incident neurological disorders14–16 and to be consistent with the findings of a recent meta-analysis that assessed the relationship of MeD with the risk of cerebrovascular diseases by dichotomizating and trichotomizating MeD-score29.

Stroke Ascertainment

REGARDS participants were contacted via telephone for every 6 months to ascertain vital status and to obtain information on reasons for hospitalization, including stroke, transient ischemic attack, and stroke symptoms. Medical records were pursued if the participant reported seeking medical care for stroke or transient ischemic attack or was hospitalized for stroke symptoms or unknown reason. Once a medical record was received, it was reviewed by a committee of clinicians to verify that a stroke occurred and to provide stroke subtying as haemorrhage versus infarction. Stroke was defined either according to the World Health Organization definition of focal neurological deficit lasting >24 hours or as a nonfocal neurological symptoms with imaging consistent with stroke as previously described.27,28 The National Death Index was also queried annually to identify stroke deaths that might not have been hospitalized.28 For the present analysis, strokes occurring through September 1, 2012, were included.

Statistical analyses

The 2-tailed Pearson 2 test for categorical variables and ANOVA (Analysis of Variance) or Kruskal-Wallis test for continuous variables were used to assess intergroup differences between participants with high, moderate and low adherence to MeD. The relationship between adherence to MeD (trichotomized and dichotomized MeD-score) and incident ischemic as well as hemorrhagic stroke was evaluated separately in a set of incremental Cox proportional hazards models with high adherence to MeD as reference group. First, crude associations were estimated followed by subsequent adjustment for: (1) age, race (whites vs. blacks), age-race interaction, region (Stroke Belt/Stroke Buckle vs. Remainder of Continental US), sex (male vs. female); Model I; (2) age, race, age-race interaction, region, sex, income (categorized as ≥$75.000, $35.000-$74.000, $20.000-$34.000 and <$20.000), education (categorized as less than high school, high school graduate, some college, or college graduate); Model II; (3) variables in Model II plus total energy (calories from food or beverages analyzed by Nutrition Quest), smoking status (current smoking vs. no smoking or prior smoker, sedentary behavior (watching more than 4 hours of television and no physical activity); Model III; (4) variables in Model III plus self-reported history of heart disease, atrial fibrillation, obesity (defined as a body mass index>30kg/m2), waist circumference, diabetes (defined as fasting glucose levels >126 ml/dL, nonfasting glucose >200 ml/dL, or self-reported use of diabetes medications), hypertension, hypertension medication use, systolic and diastolic blood pressure levels; Model IV. We also evaluated the presence of interactions between race, stroke region and vascular risk factors with the association of adherence to MeD and risk of incident stroke. Since there were fewer events for hemorrhagic stroke, the final model included only age, race, age-race interaction, sex, region, systolic blood pressure, and hypertension medication use. In our prespecified analyses adherence to MeD was categorized both by dichotomization and trichotomization of MeD-score as previously described.16 Analyses were conducted using SAS version 9.2 (SAS Institute, Inc., Cary, NC).

Results

A total of 30,239 individuals were included in REGARDS. After excluding participants with history of stroke (n=2,088), incomplete dietary data (n=7,738) and missing follow-up (n=216), the sample of 20,197 (67%) individuals included in the present analyses had a mean age of 65±9 years, 33% were black (n=6,670), 44% were male (n=8,853) and 56% (n=11,368) were from the stroke belt region. The MeD-score ranged from 0 to 9 with a bell-shaped (i.e., approximately normal) distribution with 42% of participants (n=8354) having a score of 4 or 5. The mean MeD-score was 4.4±1.7. A total of 9181 individuals (53%) had a low adherence to MeD (MeD-score 0–4). Demographic characteristics, environmental and vascular risk factors in participants with low, moderate and high adherence to MeD are presented in Table 1. MeD adherence was greater among males, black race, residents of regions other than the stroke belt, and non-smokers, and was lower among those with hypertension, diabetes mellitus, obesity and sedentary life style.

Table 1.

Baseline Characteristics by Tertile of Mediterranean Diet Score in the REasons for Geographic and Racial Differences in Stroke (REGARDS) Study

Variables Mediterranean Diet Score
6–9
N=5211		 4–5
N=8354		 0–3
N=6632		 p
Demographics
Male (%) 2465 (47) 3614 (43) 2774 (42) <0.001
Black (%) 1780 (34) 2820 (34) 2070 (31) <0.001
Age <65 years (%) 4225 (51) 3675 (55) 2462 (47) <0.001
Mean age, years (SD) 66 (9) 65 (9) 64 (9) <0.001
Residence in the Stroke Belt (%) 2721 (52) 4800 (57) 3847 (58) <0.001
Income over $75,000 (%) 1168 (22) 1488 (18) 902 (14) <0.001
College graduate (%) 2517 (48) 3277 (39) 1891 (29) <0.001
Co-morbid Conditions
Diabetes mellitus (%) 771 (15) 1531 (18) 1250 (19) <0.001
Heart Disease (%) 842 (16) 1331 (16) 1057 (16) 0.947
Hypertension (%) 2426 (47) 4075 (49) 3225 (49) 0.019
Atrial Fibrillation (%) 409 (8) 685 (8) 540 (8) 0.733
Obesity (%) 1559 (30) 3071 (37) 2687 (41) <0.001
Lifestyle Factors
Sedentary (%) 376 (7) 994 (12) 1087 (16) <0.001
Current smoking (%) 428 (8) 1034 (12) 1246 (19) <0.001
Blood pressure levels
Mean systolic blood pressure, mmHg (SD) 126 (16) 126 (16) 127 (16) 0.001
Mean diastolic blood pressure, mmHg (SD) 76 (9) 76 (9) 77 (10) 0.006
Open in a new tab

During a mean follow-up period of 6.5±2.2 years, incident stroke was identified in 565 participants (2.8%). Ischemic stroke was documented in 497 participants (88% of all strokes), while hemorrhagic stroke occurred in the remaining 68 participants (12% of all strokes). Results of incremental Cox proportional hazards models estimating the association of adherence to MeD (stratified by MeD tertiles) with incident ischemic stroke appear in Table 2. Low adherence to MeD was associated with higher risk of incident ischemic stroke in initial univariate analyses (HR versus high adherence: 1.28; 95%CI:1.00–1.63; p=0.047). Moderate adherence to MeD was also related to incident IS (HR versus high adherence: 1.32; 95%CI:1.05–1.66; p=0.020). After adjusting for demographics, environmental and vascular risk factors, blood pressure levels, and antihypertensive medications use (Model 4), low adherence to MeD tended to be associated with a higher risk of incident IS (HR versus high adherence: 1.29; 95%CI:0.99–1.67; p=0.057). A similar relationship was documented for moderate adherence to MeD (HR versus high adherence: 1.25; 95%CI:0.98–1.58; p=0.079). When MeD was evaluated in Cox proportional hazards models as a continuous variable, a 1-point increase in MeD-score was independently associated with a 5% reduction in the risk of incident (95%CI: 0%-11%) after adjustment for all potential confounders (Model IV). There was no interaction of age (p=0.46), sex (p=0.76), race (p=0.38), stroke-belt region (p=0.67), atrial fibrillation (p=0.47), history of heart disease (p=0.42), diabetes mellitus (p=0.85), hypertension (p=0.90) and systolic blood pressure levels (p=0.67) on the association of adherence to MeD with incident ischemic stroke.

Table 2.

Association of adherence to Mediterranean Diet (MeD) stratified by tertiles of MeD-score with incident ischemic stroke on incremental Cox proportional hazards models.

Association Mediterranean Diet Score
6–9
N=5211		 4–5
N=8354		 0–3
N=6632
Nstrokes=108 Nstrokes=222 Nstrokes=167
Crude Reference HR=1.32 (95%CI: 1.05, 1.66)
p=0.020		 HR=1.28 (95%CI:1.00,1.63)
p=0.047
Model 1 Reference HR=1.37 (95%CI: 1.09, 1.73)
p=0.007		 HR=1.46 (95%CI:1.14, 1.86)
p=0.003
Model 2 Reference HR=1.33 (95%CI: 1.05, 1.67)
p=0.016		 HR=1.36 (95%CI: 1.07, 1.74)
p=0.013
Model 3 Reference HR=1.31 (95%CI: 1.04, 1.65)
p=0.023		 HR=1.31 (95%CI: 1.02, 1.68)
p=0.034
Model 4 Reference HR=1.24 (95%CI:0.98, 1.58)
p=0.079		 HR=1.29 (95%CI: 0.99, 1.67)
p=0.057
Open in a new tab

Model 1 adjusts for age, race, age-race interaction, region, sex

Model 2 adjusts for age, race, age-race interaction, region, sex, income, education

Model 3 adjusts for age, race, age-race interaction, region, sex, income, education, total energy, smoking status, sedentary behavior

Model 4 adjusts for age, race, age-race interaction, region, sex, income, education, total energy, smoking status, sedentary behavior, history of heart disease, atrial fibrillation, BMI, waist circumference, diabetes, hypertension, hypertension medication use, systolic and diastolic blood pressure levels

Results of incremental Cox proportional hazards models estimating the association of adherence to MeD (stratified by MeD tertiles) with incident hemorrhagic stroke are shown in Table 3. We documented no association of high or moderate adherence to MeD with incident hemorrhagic stroke both in the initial univariate and multivariate analyses adjusting for potential confounders (Model II; p for the association of high adherence to MeD with incident hemorrhagic stroke: 0.16; p for the association of moderate adherence to MeD with incident hemorrhagic stroke 0.53). There was no interaction of age (p=0.28), sex (p=0.21), race (p=0.38), stroke-belt region (p=0.67) and systolic blood pressure levels (p=0.39) on the association of adherence to MeD with incident hemorrhagic stroke.

Table 3.

Association of adherence to Mediterranean Diet (MeD) stratified by tertiles of MeD-score with incident hemorrhagic stroke on incremental Cox proportional hazards models.

Association Mediterranean Diet Score
6–9
N=5211		 4–5
N=8354		 0–3
N=6632
Nstrokes=23 Nstrokes=29 Nstrokes=16
Crude Reference HR=0.80 (95%CI: 0.46, 1.39)
p=0.80		 HR=0.57 (95%CI:0.30,1.08)
p=0.08
Model 1 Reference HR=0.85 (95%CI: 0.49, 1.48)
p=0.57		 HR=0.65 (95%CI:0.34, 1.24)
p=0.19
Model 2 Reference HR=0.84 (95%CI: 0.49, 1.46)
p=0.53		 HR=0.64 (95%CI: 0.34, 1.21)
p=0.16
Open in a new tab

Model 1 adjusts for age, race, age-race interaction, region, sex

Model 2 adjusts for age, race, age-race interaction, region, sex, hypertension medication use, systolic blood pressure levels

All analyses were repeated after dichotomizing MeD-score. Table 4 presents the findings of incremental Cox proportional hazards models evaluating the association of adherence to MeD with incident IS. Low adherence to MeD was independently associated with a higher risk of incident IS (HR: 1.27;95%CI:1.04–1.54; p=0.0164) after adjusting for multiple confounders including demographics, environmental and vascular risk factors, blood pressure levels, and antihypertensive medications use. Table 5 shows the results of incremental Cox proportional hazards models investigating the association of adherence to MeD with incident hemorrhagic stroke. High adherence to MeD was not associated with the risk of hemorrhagic stroke in either the unadjusted or adjusted Cox proportional hazards models (HR= 0.93; 95%CI: 0.57–1.51; p=0.7621).

Table 4.

Association of adherence to Mediterranean Diet (MeD) stratified by dichotomization of MeD-score with incident ischemic stroke on incremental Cox proportional hazards models.

Association Mediterranean Diet Score
5–9
N=9234		 0–4
N=10889
Nstrokes=208 Nstrokes=284
Crude Reference HR= 1.20 (95%CI: 1.00, 1.43)
p=0.0460
Model 1 Reference HR=1.31 (95%CI: 1.10, 1.57)
p=0.0030
Model 2 Reference HR=1.25 (95%CI: 1.05, 1.50)
p=0.0140
Model 3 Reference HR=1.22 (95%CI: 1.02, 1.46)
p=0.0336
Model 4 Reference HR=1.27 (95%CI: 1.04, 1.54)
p=0.0164
Open in a new tab

Model 1 adjusts for age, race, age-race interaction, region, sex

Model 2 adjusts for age, race, age-race interaction, region, sex, income, education

Model 3 adjusts for age, race, age-race interaction, region, sex, income, education, total energy, smoking status, sedentary behavior

Model 4 adjusts for age, race, age-race interaction, region, sex, income, education, total energy, smoking status, sedentary behavior, history of heart disease, atrial fibrillation, BMI, waist circumference, diabetes, hypertension, hypertension medication use, systolic and diastolic blood pressure levels

Table 5.

Association of adherence to Mediterranean Diet (MeD) stratified by dichotomization of MeD-score with incident hemorrhagic stroke on incremental Cox proportional hazards models.

Association Mediterranean Diet Score
5–9
N=9234		 0–4
N=10889
Nstrokes=35 Nstrokes=33
Crude Reference HR= 0.82 (95%CI: 0.51, 1.32)
p=0.4182
Model 1 Reference HR=0.91 (95%CI: 0.56, 1.46)
p=0.6853
Model 2 Reference HR= 0.93 (95%CI: 0.57, 1.51)
p=0.7621
Open in a new tab

Model 1 adjusts for age, race, age-race interaction, region, sex

Model 2 adjusts for age, race, age-race interaction, region, hypertension medication use, systolic blood pressure levels

Comment

Our longitudinal study showed that low adherence to MeD was associated with a higher risk of incident ischemic stroke in a large population-based sample of U.S. black and white adults during a mean-follow-up period of 6.5 years. This relationship persisted after adjustment for numerous potential confounders including demographic characteristics, health behaviors, educational level, income, physical activity, vascular risk factors, blood pressure levels and antihypertensive medications. Moreover, we found no evidence for interaction between race, sex, or region of residence and the relationship of adherence to MeD and incident ischemic stroke. Finally, we failed to document any association between adherence to MeD and risk of hemorrhagic stroke.

Our findings are in line with the results of a recent meta-analysis that quantitatively synthesized all studies evaluating the association between adherence to MeD and risk of stroke, depression, cognitive impairment, and Parkinson disease.29 The protective association of high adherence to MeD with likelihood or risk of incident stroke has been replicated in both case-control (2 studies; OR: 0.20; 95%CI: 0.10–0.41) and longitudinal cohort (5 studies; OR: 0.84; 95%CI: 0.74–0.95) studies that were published till October 31st, 2012. In addition, the former association was detected also in studies conducted in non-Mediterranean countries (OR: 0.80; 95%CI: 0.66–0.97). Our study puts more insight in the examined relation for US populations, notably concerning both races. Interestingly, when stroke subtypes were separately examined, the meta-analysis detected no association between moderate or high adherence to MeD and likelihood of hemorrhagic stroke both in case-control (n=1) and cohort studies (n=2). In contrast, high adherence to MeD was strongly related with a lower likelihood of ischemic stroke in case-control studies (2 studies; OR: 0.19; 95%CI: 0.08–0.46), while a non-significant trend was documented between high adherence to MeD and lower risk of incident ischemic stroke in longitudinal cohort studies (3 studies; OR: 0.88; 95%CI: 0.70–1.12). Notably, the prior three cohort studies18,20,21 that investigated the potential association of dietary patterns with risk of cerebrovascular diseases, had a lower statistical power to identify effects on specific stroke subtypes and the relationship of high or moderate adherence to MeD with the incidence of ischemic stroke did not reach the level of statistical significance. Consequently, our study expands the preliminary findings of the prior cohort studies18,20,21 and comprehensive meta-analysis29 that detected a trend towards lower incidence of ischemic stroke in persons with higher adherence to MeD.

Our observations are consistent with the mounting literature underscoring a potential beneficial role of a healthy dietary pattern such as MeD in cerebrovascular disease protection. More specifically, a detailed analysis within the Nurses' Health Study showed that 70% of incident stroke (both ischemic and hemorrhagic) could be avoided by healthy food choices that are consistent with the traditional MeD in combination with regular physical exercise and smoking abstinence.30 Similarly, a systematic review ranked the MeD as the most likely dietary model to provide protection against cardiovascular diseases.31 Finally, the results of a recent randomized controlled trial provided additional evidence that among persons at high cardiovascular risk, a MeD supplemented with extra-virgin olive oil or nuts reduced the incidence of major cardiovascular events, with comparisons of stroke risk being the only component of the primary endpoint that reached statistical significance in secondary analyses.32 More specifically, the incidence of stroke was reduced by 34% (95%CI: 3%-54%) and 49% (95%CI: 21%-67%) in individuals who were randomized to MeD with extra-virgin olive oil and nuts respectively in comparison to the control group. In contrast, MeD was not associated with a reduction in myocardial infarction, cardiovascular and all-cause mortality. In view of the former considerations, it may be postulated that high adherence to MeD appears to be effective in terms of primary stroke prevention.

The association of high adherence to MeD with lower incidence of ischemic but not hemorrhagic stroke is intriguing. Neuroimaging studies have provided preliminary evidence demonstrating protective relation between MeD and white matter hyperintensities33 as well as infarcts detected in brain Magnetic Resonance Imaging34. However, high adherence to MeD was not cross-sectionally or longitudinally related to a lower risk of cerebral microbleeds or intracerebral hemorrrhages.33,34 These observations are suggestive of a vascular protective component of this diet that may preferentially reduce the risk of cerebral ischemia. Another plausible explanation may be related to the fact that MeD has been shown to exert salutary effects in risk factors that are associated with a high risk of ischemic (but not hemorrhagic) stroke including diabetes mellitus35 and metabolic syndrome36. Interestingly, Esposito et al recently provided randomized trial data indicating that MeD is associated with a higher rate of regression as well as a lower rate of progression of common carotid artery intima-media thickness (CCA-IMT) in patients with diabetes mellitus in comparison to a standard low-fat diet.37,38 CCA-IMT has been repeatedly shown to be strongly associated with the risk of ischemic (but not hemorrhagic) stroke both cross-sectionally39,40 and longitudinally41,42. However, it should be noted that the incidence of hemorrhagic stroke was lower than that of ischemic stroke (consistent with national estimates) and power may have been limited.

The findings of the present study are subject to certain limitations. For one, methodological shortcomings related to the construction of Mediterranean-type Diet Score (equal weighting of underlying food categories and underestimation of total food and caloric intake) need to be acknowledged. Moreover, the assessment of dietary intake was performed only at baseline and thus we were unable to capture potential changes in dietary patterns during the follow-up period that may have affected the reported associations. Additionally, it should be noted that the dietary assessment was self-administered and subject to differential return by participants and this may have resulted in selection or recall biases. Moreover, the observational design of the present study cannot rule out the possibility of residual confounding by unknown risk factors, such as general healthier lifestyle or higher adherence to medication intake of individuals adhering to a MeD. Lastly, it should be kept in mind that REGARDS findings focus on black/white disparities in stroke risk and fail to examine other racial/ethnic groups

In conclusion, greater adherence to MeD was associated with lower risk of incident IS in REGARDS dataset independent of potential confounders. The former association was not reproduced for hemorrhagic stroke, while we detected no evidence for race, sex or region interactions in the relationship of adherence to MeD with incident cerebrovascular events (both ischemic and hemorrhagic). Our findings lend support to the accruing data indicating that dietary habits are critical to effective primary stroke prevention and highlighting that the beneficial effects of a healthy dietary pattern in reducing the risk of stroke appear to apply to different racial subgroups and different regions across the US.

Acknowledgments

This research project is supported by a cooperative agreement U01 NS041588 from the National Institute of Neurological Disorders and Stroke, National Institutes of Health, Department of Health and Human Service. The content is solely the responsibility of the authors and does not necessarily represent the official views and positions of the National Institute of Neurological Disorders and Stroke or the National Institutes of Health. The authors thank the other investigators, the staff, and the participants of the REGARDS study for their valuable contributions. A full list of participating REGARDS investigators and institutions can be found at http://www.regardsstudy.org. Dr Georgios Tsivgoulis has been supported by European Regional Development Fund - Project FNUSA-ICRC (No. CZ.1.05/1.1.00/02.0123).

Footnotes

Conflict of interest disclosure statement

There are no conflicts of interest to disclose.

Contributor’s statement

Georgios Tsivgoulis had the idea for the paper and data analyses and wrote the first version of the manuscript

Theodora Psaltopoulou contributed to the study design, analysis plan and editing of the manuscript

Virginia G. Wadley: contributed to editing of the manuscript

Andrei V. Alexandrov: contributed to editing of the manuscript

George Howard: contributed to editing of the manuscript

Frederick W. Unverzagt: contributed to editing of the manuscript

Claudia Moy: contributed to editing of the manuscript

Virginia J. Howard: contributed to editing of the manuscript

Brett Kissela: contributed to the study design and editing of the manuscript

Suzanne Judd: analyzed the data, contributed to study design, analysis plan and editing of the manuscript

References

  • 1.Roger VL, Go AS, Lloyd-Jones DM, et al. American Heart Association Statistics Committee and Stroke Statistics Subcommittee Heart disease and stroke statistics--2012 update: a report from the American Heart Association. Circulation. 2012;125:e2–e220. doi: 10.1161/CIR.0b013e31823ac046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Miniño AM, Murphy SL, Xu J, Kochanek KD. Deaths: final data for 2008. Natl Vital Stat Rep. 2011;59:1–126. [PubMed] [Google Scholar]
  • 3.Hall MJ, Levant S, DeFrances CJ. Hospitalization for stroke in U.S hospitals, 1989–2009. NCHS Data Brief. 2012;95:1–8. [PubMed] [Google Scholar]
  • 4.Howard G, Goff DC. Population shifts and the future of stroke: Forecasts of the future burden of stroke. Annals of the New York Academy of Sciences. 2012;1268:14–205. doi: 10.1111/j.1749-6632.2012.06665.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Ezzati M, Lopez AD, Rodgers A, Vander Hoorn S, Murray CJ Comparative Risk Assessment Collaborating Group. Selected major risk factors and global and regional burden of disease. Lancet. 2002;360:1347–1360. doi: 10.1016/S0140-6736(02)11403-6. [DOI] [PubMed] [Google Scholar]
  • 6.Goldstein LB, Bushnell CD, Adams RJ American Heart Association Stroke Council; Council on Cardiovascular Nursing; Council on Epidemiology and Prevention; Council for High Blood Pressure Research,; Council on Peripheral Vascular Disease, and Interdisciplinary Council on Quality of Care and Outcomes Research. Guidelines for the primary prevention of stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2011;42:517–584. doi: 10.1161/STR.0b013e3181fcb238. [DOI] [PubMed] [Google Scholar]
  • 7.Whisnant JP. Modeling of risk factors for ischemic stroke. The Willis Lecture. Stroke. 1997;28:1840–1804. doi: 10.1161/01.str.28.9.1840. [DOI] [PubMed] [Google Scholar]
  • 8.Galimanis A, Mono ML, Arnold M, et al. Lifestyle and stroke risk: a review. Curr Opin Neurol. 2009;22:60–68. doi: 10.1097/WCO.0b013e32831fda0e. [DOI] [PubMed] [Google Scholar]
  • 9.Ding EL, Mozaffarian D. Optimal dietary habits for the prevention of stroke. Semin Neurol. 2006;26:11–23. doi: 10.1055/s-2006-933305. [DOI] [PubMed] [Google Scholar]
  • 10.He FJ, Nowson CA, MacGregor GA. Fruit and vegetable consumption and stroke: meta-analysis of cohort studies. Lancet. 2006;367:320–326. doi: 10.1016/S0140-6736(06)68069-0. [DOI] [PubMed] [Google Scholar]
  • 11.Trichopoulos D, Lagiou P. Mediterranean diet and cardiovascular epidemiology. Eur J Epidemiol. 2004;19:7–8. doi: 10.1023/b:ejep.0000013409.70608.d9. [DOI] [PubMed] [Google Scholar]
  • 12.Trichopoulou A, Costacou T, Bamia C, Trichopoulos D. Adherence to a Mediterranean diet and survival in a Greek population. N Engl J Med. 2003;348:2599–2608. doi: 10.1056/NEJMoa025039. [DOI] [PubMed] [Google Scholar]
  • 13.Sofi F, Cesari F, Abbate R, Gensini GF, Casini A. Adherence to Mediterranean diet and health status: meta-analysis. BMJ. 2008;337:a1344. doi: 10.1136/bmj.a1344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Scarmeas N, Stern Y, Tang MX, Mayeux R, Luchsinger JA. Mediterranean diet and risk for Alzheimer's disease. Ann Neurol. 2006;59:912–921. doi: 10.1002/ana.20854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Féart C, Samieri C, Rondeau V, et al. Adherence to a Mediterranean diet, cognitive decline, and risk of dementia. JAMA. 2009;302:638–648. doi: 10.1001/jama.2009.1146. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Tsivgoulis G, Judd S, Letter AJ, et al. Adherence to a Mediterranean diet and risk of incident cognitive impairment. Neurology. 2013;80:1684–1692. doi: 10.1212/WNL.0b013e3182904f69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Mente A, de Koning L, Shannon HS, Anand SS. A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease. Arch Intern Med. 2009;169:659–69. doi: 10.1001/archinternmed.2009.38. [DOI] [PubMed] [Google Scholar]
  • 18.Agnoli C, Krogh V, Grioni S, et al. A priori-defined dietary patterns are associated with reduced risk of stroke in a large Italian cohort. J Nutr. 2011;141:1552–1558. doi: 10.3945/jn.111.140061. [DOI] [PubMed] [Google Scholar]
  • 19.Hoevenaar-Blom MP, Nooyens AC, Kromhout D, et al. Mediterranean style diet and 12-year incidence of cardiovascular diseases: the EPIC-NL cohort study. PLoS One. 2012;7:e45458. doi: 10.1371/journal.pone.0045458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.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. doi: 10.1161/CIRCULATIONAHA.108.816736. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Gardener H, Wright CB, Gu Y, et al. Mediterranean-style diet and risk of ischemic stroke, myocardial infarction, and vascular death: the Northern Manhattan Study. Am J Clin Nutr. 2011;94:1458–1464. doi: 10.3945/ajcn.111.012799. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Howard VJ, Cushman M, Pulley L, et al. The Reasons for Geographic and Racial Differences in Stroke Study: objectives and design. Neuroepidemiology. 2005;25:135–143. doi: 10.1159/000086678. [DOI] [PubMed] [Google Scholar]
  • 23.Wadley VG, McClure LA, Howard VJ, et al. Cognitive status, stroke symptom reports, and modifiable risk factors among individuals with no diagnosis of stroke or transient ischemic attack in the REasons for Geographic and Racial Differences in Stroke (REGARDS) Study. Stroke. 2007;38:1143–1147. doi: 10.1161/01.STR.0000259676.75552.38. [DOI] [PubMed] [Google Scholar]
  • 24.Tsivgoulis G, Alexandrov AV, Wadley VG, et al. Association of higher diastolic blood pressure levels with cognitive impairment. Neurology. 2009;73:589–595. doi: 10.1212/WNL.0b013e3181b38969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wadley VG, Unverzagt FW, McGuire LC, et al. Incident cognitive impairment is elevated in the stroke belt: the REGARDS study. Ann Neurol. 2011;70:229–236. doi: 10.1002/ana.22432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Block G, Hartman AM, Dresser CM, Carroll MD, Gannon J, Gardner L. A data-based approach to diet questionnaire design and testing. Am J Epidemiol. 1986;124:453–469. doi: 10.1093/oxfordjournals.aje.a114416. [DOI] [PubMed] [Google Scholar]
  • 27.Howard VJ, Kleindorfer DO, Judd SE, et al. Disparities in stroke incidence contributing to disparities in stroke mortality. Ann Neurol. 2011;69:619–627. doi: 10.1002/ana.22385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Judd SE, Gutiérrez OM, Newby PK, et al. Dietary patterns are associated with incident stroke and contribute to excess risk of stroke in black americans. Stroke. 2013;44:3305–3311. doi: 10.1161/STROKEAHA.113.002636. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Psaltopoulou T, Sergentanis TN, Panagiotakos DB, Sergentanis IN, Kosti R, Scarmeas N. Mediterranean diet, stroke, cognitive impairment, and depression: A meta-analysis. Ann Neurol. 2013;74:580–591. doi: 10.1002/ana.23944. [DOI] [PubMed] [Google Scholar]
  • 30.Willett WC. The Mediterranean diet: science and practice. Public Health Nutr. 2006;9:105–110. doi: 10.1079/phn2005931. [DOI] [PubMed] [Google Scholar]
  • 31.Mente A, de Koning L, Shannon HS, Anand SS. A systematic review of the evidence supporting a causal link between dietary factors and coronary heart disease. Arch Intern Med. 2009;169:659–669. doi: 10.1001/archinternmed.2009.38. [DOI] [PubMed] [Google Scholar]
  • 32.Estruch R, Ros E, Salas-Salvadó J, et al. PREDIMED Study Investigators. Primary prevention of cardiovascular disease with a Mediterranean diet. N Engl J Med. 2013;368:1279–1290. [Google Scholar]
  • 33.Gardener H, Scarmeas N, Gu Y, et al. Mediterranean diet and white matter hyperintensity volume in the Northern Manhattan Study. Arch Neurol. 2012;69:251–256. doi: 10.1001/archneurol.2011.548. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Scarmeas N, Luchsinger JA, Stern Y, et al. Mediterranean diet and magnetic resonance imaging-assessed cerebrovascular disease. Ann Neurol. 2011;69:257–268. doi: 10.1002/ana.22317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Martínez-González MA, de la Fuente-Arrillaga C, Nunez-Cordoba JM, et al. Adherence to Mediterranean diet and risk of developing diabetes: prospective cohort study. BMJ. 2008;336:1348–1351. doi: 10.1136/bmj.39561.501007.BE. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kastorini CM, Milionis HJ, Esposito K, Giugliano D, Goudevenos JA, Panagiotakos DB. 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. doi: 10.1016/j.jacc.2010.09.073. [DOI] [PubMed] [Google Scholar]
  • 37.Esposito K, Giugliano D. Mediterranean diet for primary prevention of cardiovascular disease. N Engl J Med. 2013;369:674–675. doi: 10.1056/NEJMc1306659. [DOI] [PubMed] [Google Scholar]
  • 38.Esposito K, Maiorino MI, Ciotola M, et al. Effects of a Mediterranean-style diet on the need for antihyperglycemic drug therapy in patients with newly diagnosed type 2 diabetes: a randomized trial. Ann Intern Med. 2009;151:306–314. doi: 10.7326/0003-4819-151-5-200909010-00004. [DOI] [PubMed] [Google Scholar]
  • 39.Vemmos KN, Tsivgoulis G, Spengos K, et al. Common carotid artery intima-media thickness in patients with brain infarction and intracerebral haemorrhage. Cerebrovasc Dis. 2004;17:280–286. doi: 10.1159/000077338. [DOI] [PubMed] [Google Scholar]
  • 40.Tsivgoulis G, Vemmos KN, Spengos K, et al. Common carotid artery intima-media thickness for the risk assessment of lacunar infarction versus intracerebral haemorrhage. J Neurol. 2005;252:1093–100. doi: 10.1007/s00415-005-0821-0. [DOI] [PubMed] [Google Scholar]
  • 41.Tsivgoulis G, Vemmos K, Papamichael C, et al. Common carotid artery intima-media thickness and the risk of stroke recurrence. Stroke. 2006;37:1913–1916. doi: 10.1161/01.STR.0000226399.13528.0a. [DOI] [PubMed] [Google Scholar]
  • 42.Prati P, Tosetto A, Vanuzzo D, et al. Carotid intima media thickness and plaques can predict the occurrence of ischemic cerebrovascular events. Stroke. 2008;39:2470–2476. doi: 10.1161/STROKEAHA.107.511584. [DOI] [PubMed] [Google Scholar]

RESOURCES