Diet and stroke: recent evidence supporting a Mediterranean style diet and food in the primary prevention of stroke

Introduction

Every year there are approximately 795,000 incident strokes, a leading cause of long-term disability in the United States¹. The cost of stroke in 2010 was $36.5 billion and is projected to increase, with lost wages being the most substantial cost¹. Identifying targets for primordial prevention of stroke is critical to public health as the population in the US is aging. Diet is often suggested as a stroke prevention option, as diet quality has effects on weight maintenance and blood pressure control beyond directly reducing the risk of stroke².

Based on the impact of diet on blood pressure and cholesterol, a recent Cochrane Review estimated that dietary interventions may decrease stroke risk by 19%³. In addition to blood pressure and cholesterol, diet may increase risk of stroke through other mechanisms, including insulin resistance, inflammation, thrombosis, endothelial function, and oxidation⁴. When describing diet and stroke risk, one can consider nutrients, foods, food groups, and dietary patterns^4-8. The multiple of methods used to quantify diet can lead to confusion in terms of dietary recommendations, since many studies appear to conflict with one another. Recently, in an effort to simplify and clarify nutritional recommendations, dietary patterns approaches have been the focus of national recommendations for dietary change⁹. In fact many are simply recommending a Mediterranean style diet¹⁰.

The Mediterranean diet refers to the dietary patterns which were found in the olive-growing areas of the Mediterranean region in the early 1950s and 1960s and has long been associated with better cardiovascular health^11,12. In the Seven Countries Study, Ancel Keys observed that the Greek island of Crete, had the lowest rates of coronary heart disease of the seven countries, sparking interest in the Mediterranean diet¹¹. While details of this pattern may vary, the Mediterranean diet is characterized by high amounts of plant-based food, olive oil, and moderate amounts of meat, dairy, and wine ¹¹.

This review will focus on the evidence for a Mediterranean style diet and the foods that characterize this diet and stroke.

Mediterranean Diet and Stroke Prevention

A recent meta-analysis of 12 studies also found that high adherence to the Mediterranean diet pattern was associated with reduced stroke risk (Risk Ratio (RR): 0.71, 95% CI: 0.57, 0.89)¹³ which was confirmed in a second systematic re-meta-analysis in 2014 that added an additional three studies (RR: 0.68, 95% CI: 0.58, 0.79), which included the PREDIMED trial¹⁴. This was the first large, multi-center, randomized control trial (RCT), to examine primary prevention of cardiovascular disease through a Mediterranean diet ¹⁵. Although the study was not designed to specifically examine stroke, the effect of the dietary intervention in reducing stroke risk was greater than the effect on myocardial infarction. For stroke, the hazard ratio (HR) was 0.61; 95% CI: 0.44-0.86 and for myocardial infarction HR= 0.77; 95% CI: 0.52-1.15. The Mediterranean diet in this study recommended consumption of tree nuts and peanuts, fresh fruits, vegetables, fish, legumes, white meat, and wine¹⁵. Consumption of soda drinks, commercial bakery goods, spread fats, red and processed meats was discouraged¹⁵ .

To better understand and expand upon these findings, we will examine the evidence for the individual components of the Mediterranean diet in terms of stroke risk (Table 1). If specific components of the Mediterranean diet drive the association between high adherence to the pattern and reduced stroke risk, more targeted dietary recommendations for stroke risk reduction can be provided. Diet and stroke prevention has been the subject of several reviews, which have been conducted on studies published through 2012^{4, 6-8}. Since 2012 however, a number of meta-analyses and systematic reviews on the associations between a wide variety of dietary factors and stroke have been conducted. Our review examines the most current comprehensive studies on diet and stroke prevention, through the lens of recommended or discouraged foods in the Mediterranean diet.

Table 1.

Components of the Mediterranean Diet

Food	Meta-Analysis: First Author, Year reference	+ (Inverse association with stroke risk) / − (Association with stroke risk)
Fruits and Vegetables	Hu et al, 2014 16	+
Fish	Chowdhury et al, 2012 17	+
	Xun et al, 2012 18	+
	Larsson et al, 2011 19	+
Olive Oil	Martinez-Gonzalez et al, 2014 20	+
Tree Nuts and Peanuts	Afshin et al, 2014 21	Not statistically significant
Legumes	Afshin et al, 2014 21	Not statistically significant
Red and Processed Meat	Kaluza et al, 201222	−
White Meat	Bernstein et al, 2012 23 *	+
Alcohol	Zhang et al, 2014 24	−
Dairy	Soedamah-Muthu et al, 201125 †	Not statistically significant
	Hu et al, 201426 ‡	+

^*Pooled analysis found substituting red meat for white meat was associated with decreased stroke risk

^†Examined the association between milk consumption and stroke

^‡Total dairy consumption was inversely associated with stroke, but associations between different types of dairy products and stroke varied.

Fruits and vegetables

Five or more servings of fruits and vegetables should be consumed a day in a Mediterranean diet¹⁵. Previous reviews of diet and stroke have found strong evidence supporting an association between fruit and vegetable consumption and stroke risk^{4, 6, 8}. This was confirmed by a meta-analysis of 20 prospective cohort studies published in 2014, which found that compared to the lowest quantile of fruit and vegetable consumption, the highest quantile of consumption was associated with a 21% lower risk of stroke (RR: 0.79; 95% CI: 0.75-0.84)¹⁶. When examined separately, there was a linear association between fruit and vegetable consumption and stroke risk. Further, several studies studying diet quality have consistently demonstrated that dietary patterns that include large amounts of fruits and vegetables are associated with lower rates of stroke^{5, 27}.

Fruits and vegetables are a rich source of micronutrients and antioxidants that decrease lipid peroxidation; however, a recent meta-analysis of RCTs of antioxidant supplements (folic acid, β-carotene, selenium, vitamin B6, vitamin B12, vitamin C, vitamin D, and vitamin E) found that antioxidant supplements did not significantly reduce the risk of stroke²⁸. Similarly, another meta-analysis of five prospective studies found that each 20 μmol/L of circulating vitamin C was associated with a 19% reduction in stroke risk; however only dietary vitamin C intake, not supplemental intake, had a statistically significant inverse association with stroke risk^{28, 29}. This suggests that some micronutrients have stronger associations with reduced stroke risk when obtained from dietary, rather than supplemental sources. The interpretation of meta-analyses of supplementation is complicated by several factors. Many modifiable stroke risk factors (coronary heart disease, diabetes) take years to develop; it is possible that micronutrient supplementation at the time of the RCT does not reduce stroke risk. Supplements administered at different stages of stroke risk factor development may have different effects²⁸. Additionally, the effect of micronutrient supplementation on stroke risk among micronutrient deficient populations has not been well examined in meta-analyses and it is possible that correcting overt micronutrient deficiency would have a differential association with stroke risk compared than what has been observed in populations that are micronutrient replete ^{28, 29}. Additional studies are needed to further elucidate mechanisms through which micronutrient supplementation may reduce stroke risk in all populations before recommendations on specific micronutrients are warranted.

Fish

Those with strong adherence to a Mediterranean diet should eat fish regularly. Earlier reviews of diet and stroke, strongly suggest an association between fish intake and stroke prevention ^{4, 8}; findings that are supported by three recent meta-analyses ^17-19. One meta-analysis of seven prospective cohort studies, published in 2012 found that compared to the lowest tertile of fish consumption, participants in the highest tertile had a statistically significant reduced risk of ischemic (RR: 0.93; 95% CI: 0.87-0.99) and hemorrhagic stroke (RR: 0.81; 95% CI: 0.70-0.94)¹⁷. Other meta-analyses examined fish intake by servings per week and found similar statistically significant inverse associations between fish consumption and ischemic stroke, but not hemorrhagic stroke risk ^{18, 19}. Fatty fish are a source of omega-3 fatty acids, which may play a role in reducing triglyceride level¹⁷. Interestingly, a meta-analysis of nine RCTs of omega-3 fatty acid supplements found that the supplements did not significantly decrease the risk of stroke³⁰. This could indicate that it is the fish itself providing the benefit observed when examining fish intake and stroke risk. Additionally, high consumption of fish and other foods recommended in the Mediterranean diet may also be representative of healthier dietary patterns or higher socioeconomic status, which are both associated with better health¹⁷.

Olive oil

The consumption of olive oil, which is high in monounsaturated fatty acids, is recommended as part of a Mediterranean diet¹⁵. Previous reviews of diet and stroke found limited evidence to suggest that monounsaturated fats reduce stroke risk; however, these reviews did not examine olive oil separately^{4, 6}. A 2014 meta-analysis of cohort studies of 38,673 participants found that a 25g increase in olive oil consumption is associated with an 18% reduced risk of stroke (RR: 0.82; 95% CI: 0.70-0.96)²⁰.

Tree nuts and peanuts

In a Mediterranean diet, three or more servings of nuts are recommended per week¹⁵. Although previous reviews of diet and stroke did not examine the association between nuts and stroke risk, they did examine the association between nutritional components of nuts and stroke^4,6. Nuts contain high levels of polyunsaturated fatty acids, for which there was insufficient evidence to suggest an association with stroke risk in previous reviews^{4, 6}. These findings are in agreement with a recent meta-analysis of 3 cohort studies and 1 RCT which found that consumption of four servings (28.4g each) of nuts a week had no statistically significant association with stroke risk (RR: 0.89; 95% CI: 0.74-1.05)²¹.

Legumes

In accordance with a Mediterranean diet, three or more servings of legumes should be consumed every week¹⁵. Legumes are a rich source of fiber and have a lipid lowering effect³¹. Although a recent meta-analysis of seven studies reported that a 7g increase in fiber intake was associated with decreased risk of incident stroke (RR: 0.93; 95% CI 0.88-0.98), pooled effect estimates of fiber from sources other than fruits and vegetables were not reported due to heterogeneity between studies ³². One previous review of diet and stroke examined soy alone, and found insufficient evidence to suggest an association with stroke⁸. Similarly, a recent meta-analysis of six prospective cohorts found that consumption of 4 servings (100g each) of legumes a week was not significantly associated with stroke risk (RR: 0.98; 95% CI 0.84-1.14)²¹.

Red and processed meat

Daily consumption of red and processed meats is not recommended in the Mediterranean diet¹⁵. Prior reviews of diet and stroke have found limited evidence supporting an association between red meat consumption and stroke risk^{4, 6, 8}. Stronger evidence was provided by meta-analysis of 6 prospective studies published in 2012, which examined the association between fresh red meat, processed meat, and total red meat intake with stroke risk²². A one serving per day increase in processed meat had a stronger association with stroke risk (RR: 1.13; 95% CI 1.03-1.24) than fresh red meat (RR: 1.11; 95% CI 1.03-1.20) and total red meat (RR: 1.11; 95% CI 1.06-1.16)²². When examined by stroke subtype, a statistically significant association was observed with ischemic but not hemorrhagic stroke²². Components of red meat, which may contribute to stroke risk, include saturated fat and sodium in processed meat²². Previous reviews of diet and stroke found very strong evidence to support the association between sodium and stroke^{4, 6, 7}, but there was insufficient evidence to support an association between saturated fat and stroke risk^{4, 6, 7}.

White meat

In the Mediterranean diet, consumption of white meat is preferred over red meat¹⁵. In support of this recommendation, a pooled analysis of the Nurse’s Health Study (NHS) and the Health Professionals Follow-Up Study (HPFS), observed that substituting one serving of red meat per day for one serving of poultry was associated with decreased risk of stroke (RR: 0.73; 95% CI: 0.61–0.88)²³. When poultry consumption was examined by stroke subtype however, no statistically significant association was observed between poultry intake and incidence of hemorrhagic or ischemic stroke, although poultry intake was associated with decreased stroke risk when the subtypes were pooled (RR: 0.87; 95% CI: 0.78–0.97) ²³. Similarly, a prospective study of 134,290 Chinese participants, found no statistically significant associations between poultry intake and hemorrhagic or ischemic stroke mortality³³.

Wine

Although red wine is considered a component of the Mediterranean diet, there are very few studies that have considered red wine and stroke. Examining alcohol consumption and diet has been controversial when creating recommendations for disease prevention. Although a number of studies have shown that low or moderate alcohol consumption is associated with decreased risk of cardiovascular disease, the evidence for alcohol and stroke is mixed²⁴. A 2014 meta-analysis of 15 studies observed a non-linear, J-shaped association between alcohol intake and stroke risk²⁴. Compared to non-drinkers, low levels of alcohol consumption (<15g/day) was associated with reduced risk of stroke (RR, 0.85; 95% CI: 0.75–0.95). Moderate levels (15-30g/day) of alcohol consumption were not associated with risk of stroke (RR, 1.01; 95% CI: 0.93–1.09); however, high levels (>30g/day) of daily consumption were associated with increased risk of stroke (RR: 1.20; 95% CI: 1.01–1.43)²⁴. Due to the complicated and differential association by amount of alcohol consumed^{24, 34}, we conclude there is not strong evidence to place a strong focus on this component of Mediterranean diet. Further, there is very little evidence examining red wine specifically for stroke prevention therefore consumption of this component should be minimized.

Sweetened beverages and foods

Reducing the amount of sweetened foods in the diet has become a matter of intense scrutiny in recent decades. Much of the recent evidence has focused on the role of high sugar diets in leading to obesity and diabetes while little has been done in stroke specifically. However, a recent study of Swedish women and men found that consuming 2 or more servings of sugar sweetened beverages was associated with increased risk of ischemic but not hemorrhagic stroke³⁵. Similar findings were observed an American cohort study in which consuming soda at least once per day was associated with a 16% increased risk of stroke³⁶. Although little evidence was available at the time of this review to consider all sweetened foods rather than just sugar sweetened beverages, a meta-analysis did examine the association between glycemic index, glycemic load, and stroke risk³⁷. The association between glycemic load and incident stroke risk was borderline statistically significant (RR: 1.19; 95% CI 1.00–1.43), but no statistically significant association was observed between glycemic index and stroke risk³⁷. One of the limitations of this meta-analysis was that only 3 studies were available for inclusion.

Added Fats

The evidence examining added fats and risk of stroke is a bit more limited. A recent study which examined dietary patterns and risk of stroke and found that people who adhered to a dietary pattern that was high in fatty meats and fried foods were at increased risk of stroke (HR: 1.39; 95% CI 1.05-1.84)⁵. However, if we consider individual fats (saturated, trans, polyunsaturated) there is a large body of evidence, not always consistent, on each type of fat. The polyunsaturated fats in fish oil, omega-3 fatty acids have been described above in the fish paragraphs. Trans-fats were associated with a 13% increased risk of stroke (HR=1.13; 95% CI 1.00, 1.28) in American men³⁸. This finding has not always been consistent across studies, possibly due to the changing amounts of trans-fats in the American diet³⁸.

Dairy

Prior reviews have found mixed results on the association between dairy consumption and stroke^{4, 8}. A 2011 meta-analysis of six studies observed no association between a 200 mL/day increase in milk consumption and stroke risk²⁵. A more recent meta-analysis examined the associations between different types of dairy foods and found that different dairy products had distinct associations with stroke. Compared to those who consumed low levels, those who consumed high levels of low fat dairy had a lower risk of stroke (RR: 0.91; 95% CI 0.85-0.97)²⁶. Consumption of fermented milk was also inversely associated with stroke risk (RR: 0.80; 95% CI 0.71-0.89), although only a few studies have examined this²⁶. Other dairy products such as cheese, butter, and cream did not have a statistically significant association with stroke risk²⁶. Based on the evidence, the association between dairy and stroke risk remains unclear; consumption of different types dairy products may have different associations with stroke risk, and further studies should be conducted.

Conclusion

We have reviewed the evidence supporting the use of a Mediterranean style diet in stroke prevention. Although it is clear there is a role of diet quality in stroke prevention, it is also clear that there are few trials specifically designed to test the effect on diet on stroke as the primary outcome. This complicates the interpretation of any findings we have presented above for two reasons. First, the bulk of the evidence presented above is obtained from prospective studies in which the role of residual confounding cannot be ignored. People who choose to adhere to healthier dietary patterns may also engage in other health promoting behaviors such as physical activity and may be more likely to maintain a healthy weight. It therefore becomes challenging to determine the direct effect of diet. Second, many of the studies presented in this review were selected for their excellent examination of diet. In many of the studies, stroke sub-type was not described. Therefore, we cannot know if the observed findings are consistent across both hemorrhagic and ischemic strokes. Third, there was a lack of racial and ethnic diversity in many of the epidemiological studies and many racial and ethnic minorities were not well represented. Although there is not evidence to suggest that the association of Mediterranean diet with stroke would differ across races, lack of studies in diverse racial populations limit the generalizability of the evidence presented.

Examining diet quality as a whole rather than focusing on the individual components of diet is becoming increasingly popular. The American Diabetes Association recently modified dietary recommendations for people with diabetes and stated that there is no one dietary pattern that is most efficacious for all people⁹. The authors further conclude that personal and cultural preferences need to be considered when designing a nutrition management plan since a variety of dietary patterns, including the Mediterranean diet, have demonstrated health benefits. They went on to describe the beneficial aspects of maintaining an optimal mix of fats, carbohydrates and protein through whole grains, fruits, vegetables and protein while minimizing sodium, sweets and added fats. This applies to what we have described in this article. When discussing dietary choice with patients, recommendations should focus the whole diet and not individual sub-components and also focus on maintaining balance based on personal and/or cultural preferences.

Few studies have been conducted to rigorously examine diet quality following stroke as a means of improved recovery. One RCT, The Healthy Eating and Lifestyle After Stroke (HEALS) trial was designed to better understand the effects of lifestyle interventions on stroke survivors³⁹. We look forward to learning the results of this intervention. Future studies are also needed to examine how the dietary components described above may be used for secondary stroke prevention.

A Mediterranean diet pattern is similar to many other recommendations for improved whole diet quality including the Dietary Approaches to Stop Hypertension (DASH)-like eating plan⁴⁰. Most recommendations emphasize consumption of legumes, nuts, fruits, vegetables, whole grains and, lean proteins, and suggestion minimizing processed meats, and refined high sugar foods and we have laid out evidence for each component above. There is evidence supporting an association between the Mediterranean diet pattern (both individual foods and full pattern) and decreased stroke risk. Therefore, dietary recommendations for stroke risk reduction should focus on increased consumption of a Mediterranean style diet.