Mediterranean Diet Reduces Risk of Incident Stroke in a Population With Varying Cardiovascular Disease Risk Profiles.

Background and Purpose- Although some evidence has found that the Mediterranean diet (MD) is protective for stroke risk, few studies have investigated whether this relationship differs by sex or cardiovascular disease risk. Methods- We investigated the relationship between adherence to the MD score, estimated using 7-day dietary diaries and risk of incident stroke in an observational prospective population-based cohort study of 23 232 men and women (54.5% women) aged 40 to 77 years who participated in the European Prospective Investigation into Cancer study in Norfolk, United Kingdom. Risk of incident stroke was calculated using multivariable Cox regression, in the whole population, and also stratified by sex and cardiovascular disease risk profile, using the Framingham risk score. Results- During 17.0 years of follow-up (395 048 total person-years), 2009 incident strokes occurred. Risk of stroke was significantly reduced with greater adherence to the MD score (quartile 4 versus quartile 1 hazard ratio [HR], 0.83; 95% CI, 0.74-0.94; P-trend <0.01) in the whole population and in women (quartile 4 versus quartile 1 HR, 0.78; 95% CI, 0.65, 0.93; P-trend <0.01) but not in men (quartile 4 versus quartile 1 HR, 0.94; 95% CI, 0.79-1.12; P-trend =0.55). There was reduced risk of stroke in those at high risk of cardiovascular disease and across categories of the MD score (quartile 4 versus quartile 1 HR, 0.87; 95% CI, 0.76-0.99; P-trend =0.04). However, this was driven by the associations in women (quartile 4 versus quartile 1 HR, 0.80; 95% CI, 0.65-0.97; P-trend =0.02). Conclusions- Greater adherence to the MD was associated with lower risk of stroke in a UK white population. For the first time in the literature, we also investigated the associations between the MD score in those at both low and high risk of cardiovascular disease. Although the findings in our study were driven by the associations in women, they have implications for the general public and clinicians for prevention of stroke.

The modified Mediterranean diet score (MDS) was developed for use in populations where intake of monounsaturated fatty acids (MUFA) from olive oil is minimal and vegetable oils such as sunflower oils, rich in n-6 polyunsaturates, are the main sources of fat.

Health and Lifestyle information
Participants were asked to complete a Health and Lifestyle Questionnaire which included questions on physical activity, smoking status, educational level, occupation, pre-existing medical conditions (including stroke, diabetes and Myocardial Infarction) and medication use. The physical activity questionnaire asked about work and leisure time physical activity in the last year. Participants were classified as either active, moderately active, moderately inactive or inactive 3 . For smoking status participants were classified as non-, current-or former-smokers. Missing smoking status information was reclassified as 'current smoking' (n=176) to attenuate under reporting bias. Educational status was based on the highest qualification attained and was categorized into 4 groups as follows: degree or equivalent, Alevel or equivalent, O-level or equivalent, and no qualifications. O-level indicates educational attainment to the equivalent of completion of schooling to the age of 15 y, and A-level indicates educational attainment to the equivalent of the completion of schooling to the age of 17 y. Townsend deprivation scores were divided into tertiles with cut off points of 7.0, −1.8, and −3.2 respectively with larger numbers indicating higher level of deprivation. Participants with missing data on aspirin use in the previous three months (n=3548) were recoded as nonusers.

Anthropometric and biological measurements
Uniform procedures using trained personnel at participants' GP surgeries or the study clinic were used for determining weight, height, blood pressure (BP) and serum cholesterol levels 3 . A free standing stadiometer was employed to determine height to the nearest mm. Weight was measured without shoes and with light clothing and recorded with 0.2 kg precision. Body Mass Index was calculated as weight in kg / (height in meters squared).
Two BP readings were taken after participants had been sitting for 3 minutes. An Accutorr Sphygmomanometer (Datascope, UK) was used and the participants' held their arm horizontally at the point of the mid sternum 3 . Serum cholesterol was measured from nonfasting venous blood sample.

Supplemental results
We observed a significant inverse relationship between consumption of vegetables and incident stroke risk in the whole cohort (HR 0.87; 95% CI 0.80, 0.96; P-trend = 0.01) and in men only (HR 0.87; 95% CI 0.76, 0.99; P-trend = 0.03). Moderate alcohol consumption was also associated with significantly reduced stroke risk in the whole cohort (HR 0.90; 95% CI 0.82, 0.99; P-trend = 0.03) which was driven by the associations for women (HR 0.83; 95% CI 0.72, 0.95; P-trend = 0.01). No other significant associations were observed. (Table III)  42 Values are the adjusted hazard ratios (95% CI) comparing participants with intakes above vs. below the median for each component except the Meat and eggs and Dairy components, which compare participants with intakes below vs. above the median. Each component was included in a separate model. P= difference between the two groups calculated using Cox Proportional Hazards Regression. Hazard ratios were adjusted for sex, age, body mass index, physical activity, smoking status, educational attainment, material deprivation, energy intake, alcohol intake, serum total cholesterol, baseline myocardial infarction or diabetes and family history stroke or myocardial infarction, systolic blood pressure, aspirin use and antihypertensive use. Alcohol intake was not included as a covariate in the alcohol model.